Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Characterization of Free Exopolysaccharides Secreted by Mycoplasma mycoides Subsp. mycoides

  • Clothilde Bertin,

    Current address: Centre International de Recherche en Agronomie pour le Développement, UMR CMAEE, Montpellier, France; Institut National de Recherche Agronomique, UMR1309 CMAEE, Montpellier, France

    Affiliations Agence Nationale de Sécurité Sanitaire, Laboratoire de Lyon, UMR Mycoplasmoses des Ruminants, Lyon, France, Université de Lyon, VetAgro Sup, UMR Mycoplasmoses des Ruminants, Marcy-L’étoile, France

  • Corinne Pau-Roblot,

    Affiliation Unité Biologie des Plantes et Innovation, EA 3900, Université de Picardie Jules Verne, Amiens, France

  • Josiane Courtois,

    Affiliation Unité Biologie des Plantes et Innovation, EA 3900, Université de Picardie Jules Verne, Amiens, France

  • Lucía Manso-Silván,

    Affiliations Centre International de Recherche en Agronomie pour le Développement, UMR CMAEE, Montpellier, France, Institut National de Recherche Agronomique, UMR1309 CMAEE, Montpellier, France

  • François Thiaucourt,

    Affiliations Centre International de Recherche en Agronomie pour le Développement, UMR CMAEE, Montpellier, France, Institut National de Recherche Agronomique, UMR1309 CMAEE, Montpellier, France

  • Florence Tardy,

    Affiliations Agence Nationale de Sécurité Sanitaire, Laboratoire de Lyon, UMR Mycoplasmoses des Ruminants, Lyon, France, Université de Lyon, VetAgro Sup, UMR Mycoplasmoses des Ruminants, Marcy-L’étoile, France

  • Dominique Le Grand,

    Affiliations Université de Lyon, VetAgro Sup, UMR Mycoplasmoses des Ruminants, Marcy-L’étoile, France, Agence Nationale de Sécurité Sanitaire, Laboratoire de Lyon, UMR Mycoplasmoses des Ruminants, Lyon, France

  • François Poumarat,

    Affiliations Agence Nationale de Sécurité Sanitaire, Laboratoire de Lyon, UMR Mycoplasmoses des Ruminants, Lyon, France, Université de Lyon, VetAgro Sup, UMR Mycoplasmoses des Ruminants, Marcy-L’étoile, France

  • Patrice Gaurivaud

    Affiliations Agence Nationale de Sécurité Sanitaire, Laboratoire de Lyon, UMR Mycoplasmoses des Ruminants, Lyon, France, Université de Lyon, VetAgro Sup, UMR Mycoplasmoses des Ruminants, Marcy-L’étoile, France


Contagious bovine pleuropneumonia is a severe respiratory disease of cattle that is caused by a bacterium of the Mycoplasma genus, namely Mycoplasma mycoides subsp. mycoides (Mmm). In the absence of classical virulence determinants, the pathogenicity of Mmm is thought to rely on intrinsic metabolic functions and specific components of the outer cell surface. One of these latter, the capsular polysaccharide galactan has been notably demonstrated to play a role in Mmm persistence and dissemination. The free exopolysaccharides (EPS), also produced by Mmm and shown to circulate in the blood stream of infected cattle, have received little attention so far. Indeed, their characterization has been hindered by the presence of polysaccharide contaminants in the complex mycoplasma culture medium. In this study, we developed a method to produce large quantities of EPS by transfer of mycoplasma cells from their complex broth to a chemically defined medium and subsequent purification. NMR analyses revealed that the purified, free EPS had an identical β(1−>6)-galactofuranosyl structure to that of capsular galactan. We then analyzed intraclonal Mmm variants that produce opaque/translucent colonies on agar. First, we demonstrated that colony opacity was related to the production of a capsule, as observed by electron microscopy. We then compared the EPS extracts and showed that the non-capsulated, translucent colony variants produced higher amounts of free EPS than the capsulated, opaque colony variants. This phenotypic variation was associated with an antigenic variation of a specific glucose phosphotransferase permease. Finally, we conducted in silico analyses of candidate polysaccharide biosynthetic pathways in order to decipher the potential link between glucose phosphotransferase permease activity and attachment/release of galactan. The co-existence of variants producing alternative forms of galactan (capsular versus free extracellular galactan) and associated with an antigenic switch constitutes a finely tuned mechanism that may be involved in virulence.


Bacteria of the Mycoplasma genus have evolved from their Gram-positive ancestors by massive reduction of their genome size. They are considered the smallest and simplest self-replicating organisms, with many missing biosynthetic pathways and the notable absence of a cell wall [1]. Despite this apparent simplicity, mycoplasmas are able to develop metabolic responses and adaptations similar to those of more complex bacteria, due notably to their higher proportion of multifunctional enzymes [2]. Several mycoplasma species are known to cause important diseases in humans and animals. Members of the so-called Mycoplasma mycoides cluster are particularly deleterious for ruminants and are responsible for severe economic losses worldwide. Mycoplasma (M.) mycoides subsp. mycoides (Mmm) is the causal agent of contagious bovine pleuropneumonia (CBPP), a highly contagious respiratory disease notifiable to the World Organization for Animal Health (Office International des Epizooties, OIE). CBPP was formerly one of the major cattle diseases worldwide and has now been eradicated from many countries, although it persists in Africa. Under natural conditions, CBPP affects only the Bos genus, producing pathognomonic clinical lesions confined to the thoracic cavity [3]. Contrary to other mycoplasmas involved in cattle respiratory diseases, such as M. bovis [4], Mmm is not considered an agent of co-infection and is often isolated on its own. This clinical picture contrasts with that observed in infections by M. mycoides subsp. capri, its closest phylogenetic neighbor and one of the agents of contagious agalactia, a small-ruminant syndrome affecting diverse organs (udder, joints, eyes, lungs).

None of the usual virulence determinants described for other bacteria, such as toxins, invasins or cytolysins, has been identified in the currently sequenced Mmm genomes [5], [6]. However, several mechanisms have been proposed to account for Mmm pathogenicity such as adherence to the host tissues, immune evasion, persistence, dissemination, inflammation and cytotoxicity [7]. The polysaccharides have been often implicated in the pathogenesis of bacteria colonizing the lung [8] and have attracted attention for decades as a potential Mmm virulence factor. Mmm has been shown to produce two types of polysaccharides: a capsular one, namely galactan, identified [9] and chemically characterized [10] by Buttery and Plackett in the 1960s, and an extracellular one, hereafter named exopolysaccharide (EPS) as defined by Branda and collaborators [11]. The purification of free EPS has been jeopardized by polysaccharide contaminants in the complex culture medium used to grow mycoplasmas. EPS is immunologically related to the capsular galactan and has been found in large amounts as a free Mmm product unassociated with cells, both in culture supernatants and in the blood of infected cattle [12]. Mmm strains able to produce large amounts of capsular polysaccharide proved less sensitive to growth inhibition by bovine antisera [13] and displayed longer bacteremia than strains with little capsular polysaccharide in a mouse infection model [14]. This clearly pointed towards a role of capsular polysaccharides in Mmm persistence and dissemination. Antigenic cross-reactivity between Mmm capsular galactan and the pneumogalactan isolated from bovine lung was proposed to be at the origin of autoimmune reactions [15] and thus to participate in the inflammation process. In addition, intravenous injection of free, extracellular EPS induced specific effects on the vascular system of the lung and respiratory system of cattle [16] and promoted persistent bacteremia in calves, when it preceded inoculation of an Mmm strain [17], [18].

The present study was conducted to elucidate the composition and structure of the EPS produced by Mmm and to compare it with the capsular galactan. We developed a purification method consisting of transferring mycoplasma cells from their complex growth medium to a chemically defined medium to minimize polysaccharide contamination. Two strains were used: (i) the type strain PG1T, for which the complete genome sequence is available [5], and (ii) the field strain Afadé, highly virulent and already used for biofilm formation [19] and virulence studies [20], as it causes CBPP under natural and experimental conditions [21]. EPS purified from the supernatant of Mmm strain Afadé proved to have the same composition and structure as capsular galactan and was also recognized by CBPP-convalescent bovine serum. Furthermore, the production of capsular versus free galactan was associated with variations in colony opacity related to an antigenic switch of a specific glucose permease. In silico analyses were conducted in order to decipher the metabolic pathways that might account for the production of either capsular galactan or free EPS by the corresponding opaque and translucent variants.

Materials and Methods

1. Strains, Variant Selection and Growth Conditions

Two strains were used in this study: Mmm Afadé (From the CIRAD collection, isolated from cattle in Chad in 1968), used for EPS purification and for selection of opaque (OP) and translucent (TR) colony variants and PG1T (type strain, NCTC 10114), used for electron microscopy and for in silico analysis of galactan biosynthetic pathways. Both strains were grown at 37°C under 5% CO2 in PPLO-based medium (DIFCO) supplemented as described previously [22]. The chemically defined CMRL-1066 medium (Invitrogen), supplemented with amoxicillin 2 g/L (GlaxoSmithKline), was used for exopolysaccharide production. OP and TR colony variants of strain Afadé were selected on PPLO agar medium and sub-cultured to check their homogeneity. They were also grown on 30 µg/ml Congo red plates to visualize any extracellular material. OP and TR colonies were analyzed by colony blotting as described previously [23]. In brief, the monoclonal antibody 3F3 [24] was used to detect TR colonies, while OP colonies were counterstained with Ponceau red. The proportion of each variant was estimated by colony counting after blot staining.

2. EPS Production and Purification

Fifty mL of PPLO-based medium were inoculated with 500 µL of mycoplasma culture in stationary phase and incubated for 2 days at 37°C in 5% CO2. Mycoplasma cell density was estimated by cell enumeration using the most probable number method [25]. Cells were harvested by centrifugation at 12000 g for 45 min at 20°C, washed once with 20 mL sterile PBS (10 mM sodium phosphate, 150 mM NaCl, pH 7.4) and finally suspended in 500 µL sterile PBS. This suspension was used to inoculate PPLO-based broth or CMRL-1066 medium with 108–1010 mycoplasmas/mL. The cultures were further incubated for 72 h to 96 h. Aliquots were sampled at 24 h intervals to measure pH and cell density and to purify EPS.

EPS were purified as described by De Vuyst et al. [26], with a few modifications. Cells were removed by centrifugation (14000 g, 45 min, 4°C). A proteinase K (Promega) treatment (60 µg/mL for 4 hours at 37°C) was applied only as an extraction control from the PPLO-based medium in order to degrade serum proteins associated with EPS. Polypeptides were precipitated with 1/10 volume of cold 100% (w/v) trichloroacetic acid (Sigma-Aldrich) at 4°C for 2 hours followed by centrifugation (14000 g, 60 min, 4°C). Polysaccharides contained in the supernatant were precipitated with 6 to 10 volumes of cold acetone at −20°C for 48 h and collected after centrifugation at 14000 g, 60 min at 4°C. The acetone was carefully removed and the pellet was air-dried and dissolved in ultra-pure sterile water. For NMR and monomer composition analysis, extracts were dialyzed against regularly-renewed ultra-pure sterile water for 48 hours at room temperature in 3.5 kDa cut-off dialysis tubing (Spectrum Laboratories).

The effect of carbon sources on EPS production by strain Afadé was studied by adding to CMRL-1066 medium, which already contains 1 g/L glucose, either (i) 4 g/L glucose, mannitol, galactose, sorbitol or sucrose before seeding or (ii) 1 g/L glucose at 24, 48 and 72 hours after seeding. Carbohydrates were prepared as 200 g/L stock solutions and sterilized by filtration (0.22 µm filter units) before addition to CMRL-1066.

3. EPS Quantification and Characterization

The concentration of total sugars was estimated by phenol/sulfuric acid method [27] using glucose as standard. For each sample, the determination was done in triplicate and results were expressed as µg of glucose/mL of CMRL-1066 mean +/− standard deviation (SD). The potential presence of polypeptides in the EPS extract was assessed by silver staining of Any kD polyacrylamide gel (Bio-Rad) using the proteosilver kit from Sigma-Aldrich.

EPS was detected with a polysaccharide detection kit and immunodetected by dot blotting using a CBPP positive serum. In brief, 2µL of the EPS extract were spotted onto a nitrocellulose membrane that was either stained with the glycoprotein detection kit from Sigma-Aldrich or treated as previously described for dot immunobinding [28].

4. Electron Microscopy

Electron microscopy was performed as previously described [29] using five-day-old colonies of Mmm strain PG1T grown on solid PPLO-based medium. Samples were examined under a Philips CM120 transmission electron microscope operating at an accelerating voltage of 75 kV.

5. Monosaccharide Composition Determination

The monosaccharide components were determined by high-performance anion exchange chromatography (HPAEC) on a CarboPak PA 1 with a pulsed amperometric detector (Dionex ICS 3000 system). After hydrolysis of EPS (1 mg) with 4 M CF3CO2H (100°C, 4 h), aliquots of the extract were passed through a 4×50 mm Propac PA1 pre-column (Dionex) before separation of anionic compounds on a 4×250 mm Propac PA1 column (Dionex) at 30°C. Gradient elution was performed with a multi-step gradient as follows: 0–25 min, 90% H2O and 10% NaOH 160 mM; 25–34 min, 100% NaOH 200 mM; 35–50 min 90% H2O and 10% NaOH 160 mM at a flow rate of 1 mL/min. Peak analysis was performed using Chromeleon software, version 7.0.

6. NMR Spectroscopy

Prior to NMR analysis, samples were exchanged twice with 99.9% D2O (Euriso-top), dried under vacuum, and dissolved in 99.96% D2O (<1 mg/0.5 mL). 1H NMR spectra were recorded, at 80°C, on a Bruker Avance 500 spectrometer equipped with a 5 mm BBI probe and Topspin 1.3 software. 1H NMR spectra were accumulated using a 30° pulse angle, a recycle time of 1 s and an acquisition time of 2 s for a spectral width of 3 000 Hz for 32 K data points with presaturation of the HOD signal using a presaturation sequence provided by Bruker. 13C NMR experiments were conducted on the same spectrometer operating at 125.48 MHz with 2 s as relaxation delay.

The 2D 1H/1H COSY, 1H/1H TOCSY, 1H/1H NOESY, 1H/13C HSQC and 1H/13C HMBC spectra were acquired with standard pulse sequences delivered by Bruker.

7. In silico Analysis of Candidate Galactan Biosynthetic Pathways

Genes potentially involved in EPS biosynthetic pathways were retrieved from the molligen database [30] and from the Carbohydrate-Active EnZymes database (CAZy) [31]. Potential transmembrane regions were predicted by TMHMM2 [32]. Putative functions and conserved domains of proteins encoded by genes were analyzed using the profile HMM method [33], [34] and PSI-BLAST [35].

Results and Discussion

1. Mmm Growth and EPS Secretion in Defined Medium versus Complex Mycoplasma Broth

It had been previously shown that large amounts of polysaccharides not bound to the cells could be recovered from Mmm cultures [17], [36]. However, at the time, these polysaccharides could not be purified from the complex, undefined medium used to grow mycoplasmas. In this work we developed an experimental procedure that limits the potential contamination of newly synthesized polysaccharides by saccharides from the growth medium. Mycoplasma cells were first grown to stationary phase in a classic PPLO-based medium, then washed and transferred to a chemically defined, synthetic medium (CMRL-1066) for further incubation. CMRL-1066 is a constituent of the commonly used mycoplasma SP4 medium [37] and contains all the components needed to support cellular metabolism, but not growth. Following a 72 h-incubation in CMRL-1066, polysaccharide extracts were clearly detected in culture supernatants of Mmm strain Afadé, while no polysaccharides were detected in non-inoculated CMRL-1066 used as negative control (Figure 1A). These polysaccharides were not attached to the cells and were therefore considered as EPS [11]. In contrast, polysaccharide production by Mmm could not be assessed in PPLO-based medium because of detection of a strong background in the negative control corresponding to the saccharides contained in this growth medium (Figure 1A).

Figure 1. Exopolysaccharide (EPS) secretion by Mmm strain Afadé in supplemented PPLO broth and CMRL-1066 medium.

EPS extracted either from Mmm culture (Afadé) or from unseeded medium used as negative control (Neg) were spotted onto a nylon membrane and either stained with a glycoprotein detection reagent (A) or immunodetected using a serum from a CBPP-positive bovine (B).

EPS production in CMRL-1066 was shown to be reproducible, maximum after 72 h of incubation and related to the initial viable mycoplasma titer (Figure 2, Table 1). EPS detection proved to be difficult when CMRL-1066 was inoculated with less than 108 mycoplasmas/mL. CMRL-1066 was not able to sustain mycoplasma growth and a decrease in the number of viable cells was observed during the incubation period (Figure 2, Table 1). This decrease in viability was not related to a high inoculum concentration, as it was also evidenced in several serial ten-fold dilutions of the initial culture (Table 1). The poor cholesterol concentration (0.2 mg/L) and the absence of other fatty acids in CMRL-1066 might result not only in growth inhibition but also in cell lysis, as previously suggested [38]. In turn, it can be hypothesized that this cellular decay may artificially contribute to EPS release into the culture supernatant [39]. However, no correlation was found between loss of viability and EPS production after 72 h of incubation (data not shown), suggesting that the contribution of cell lysis to EPS release was minor.

Figure 2. Kinetics of exopolysaccharides secretion and viability of Mmm strain Afadé following transfer into CMRL-1066 medium with (red) or without (blue) glucose supplementation.

Glucose (1 g/L) was added at 24, 48 and 72 h (arrows).

Table 1. Exopolysaccharide secretion and viability of Mmm strain Afadé in CMRL-1066 medium at the time of inoculation (T0) and after 72 h of incubation (T72).

In 1972 Sutherland suggested that bacterial growth inhibition may result in a rise in EPS production due to the increased availability of EPS precursors, not used for glycolysis [40]. This was explored here by analysis of CMRL-1066 supplementation with different sugars during incubation. Regular addition of glucose (1 g/L) had no effect on mycoplasma viability but resulted in a 50% increase in EPS production (Figure 2). A similar increase of EPS production was measured when 4 g/L of glucose were added to CMRL-1066 at the beginning of incubation (Table 1). This indicated that during incubation in minimal medium unable to sustain growth, glucose was not catabolized by mycoplasmas but was used to produce EPS. Supplementation of CMRL-1066 medium, inoculated with 109 mycoplasmas/ml, with other carbohydrates (mannitol, galactose, sucrose or sorbitol; 4 g/L) was also tested to measure the effect of the sugar source on EPS synthesis. None of these sugars modified the viability of strain Afadé after 72 h of incubation (Table 1). As expected, the addition of sorbitol, which is not used as an energy source [41] or exploitable for EPS biosynthesis (due to absence of the required metabolic pathways) had no effect. Mannitol and sucrose supplements resulted in increased EPS production. This could be due to the more efficient transport and recycling of these two sugars into EPS, in comparison with glucose.

2. Structural Determination and Immunogenic Properties of Free EPS Secreted by Mmm

Mmm EPS was subjected to total acid hydrolysis with 4 M trifluoroacetic acid. HPAEC-PAD analysis of the acid hydrolysate revealed the presence of galactose (96.7±1.0%) and glucose (3.3±1.0%). This composition is similar to that described by Plackett et al. for the galactan Mmm capsule [36]. The structural analysis of Mmm EPS was carried out by 1H and 13C NMR spectroscopy. The 7 resonances observed on the 1H NMR spectrum (Figure 3A) and correlations detected on the 2D NMR COSY spectrum (Figure 3B) led us to attribute the observed chemical shifts to protons of galactosyl residues: 5.046 (H-1), 4.118 (H-2), 4.068 (H-3), 4.013 (H-4), 3.965 (H-5), 3.864 (H-6) and 3.648 (H-6′) ppm. On the 2D NMR HSQC spectrum (Figure 3C), the connectivities observed between H-1 and C-1 (109.0 ppm), H-2/C-2 (82.1 ppm), H-3/C-3 (78.0 ppm), H-4/C-4 (84.5 ppm), H-5/C-5 (70.8 ppm) and H-6,H-6′/C-6 (70.2 ppm) were characteristic of a galactan homopolymer. The 13C signal at 109.0 ppm and 1H signal at 5.046 ppm were typical of D-galactose residues with a β furanoside configuration [42]. Chemical shift values obtained for C-6 indicated the presence of a linkage on the C-6 in Gal residues. The 1H/1H NOESY spectrum presented interesting inter-residue connectivities: H-1 of galactose (δ 5.046) was connected to two signals at δ 3.864 and 3.648 belonging to H-6 and H-6′ of galactose. This correlation was confirmed on the 1H/13C HMBC spectrum. Thus, these connectivities and 1H and 13C chemical shifts are consistent with the literature values for a β(1−>6)-galactofuranose polymer [43]. Again, the structure of the EPS released into the culture medium was identical to that of the capsular galactan [10]. Previous studies had reported galactan to be immunogenic [12]. The EPS produced in our experimental set up was also detected by serum from a CBPP-convalescent bovine (Figure 1B). This did not represent a cross-reaction due to protein contamination of the EPS extract, since no polypeptides were detected by silver staining after SDS PAGE (data not shown).

Figure 3. NMR spectra of EPS produced by Mmm.

A: 1H NMR spectrum, B: 2D COSY NMR spectrum and C: 2D HSQC NMR spectrum.

3. Correlation between EPS Production and Colony Appearance

In this study we showed that the free galactan secreted by Mmm strain Afadé presented the same monomeric composition and structure as the capsular polysaccharide previously described, which was associated with Mmm strain V5 cells [10], [36]. In other bacteria the presence of a polysaccharidic capsule has often been associated with an opaque colony phenotype [29], [44], [45]. Variation in colony opacity has also been described in mycoplasmas (i.e. M. pulmonis [46] and M. hyorhinis [47]) but not in Mmm. However, Minga reported the existence of rough and smooth morphotypes of Mmm colonies and showed that they were related both to the capacity to produce a capsule and to pathogenicity [48]. Here we showed that both Mmm strains PG1T and Afadé produced a mixture of opaque (OP) and translucent (TR) colonies on agar (Figure 4A). TR colonies showed the typical “fried egg” appearance, with a small opaque center surrounded by a large translucent peripheral area. In contrast, OP colonies were darker and their center was not always distinct from the peripheral area. When grown on agar plates containing Congo red, a direct dye known to interact with the extracellular matrix, OP colonies showed an intense and uniform red coloration, while TR colonies were poorly stained, with staining limited mainly to their center (Figure 4A). OP and TR variants of Mmm PG1T were then observed by electron microscopy following staining with Ruthenium red, which binds to polysaccharides. Electron micrographs obtained from TR colonies showed separated cells with scanty extracellular material, while cells from OP colonies were aggregated and coated with a patchy, electron-dense material made of polysaccharides (Figure 4B), which appeared to attach the cells together. This cellular aggregation may account for the colony opacity, as previously suggested for Nesseria gonorrhoeae [49].

Figure 4. PG1T colony opacity is related to the presence of capsular material.

A: Colony morphology of PG1T opaque (OP) and translucent (TR) colony variants grown on PPLO supplemented agar with (right) or without (left) Congo red. Magnification: x30. B: Electron microscopy of Ruthenium red-stained PG1T OP and TR colonies. CM: cytoplasmic membrane.

Interestingly, OP and TR colony variants differed in their capacity to secrete EPS, with the TR yielding three-fold more EPS than the OP variant (Table 1, Figure 5). Viability in CMRL-1066 was comparable for both colony variants, though acidification of the culture medium was more pronounced for the TR variant, indicating that glycolysis was more efficient (Figure 5). Glucose metabolism is largely dependent on the capacity to import sugars and, in a previous study, we had demonstrated that a glucose permease (MSC_0860/0873), belonging to the phosphoenolpyruvate phosphotransferase system (PTS), of Mmm PG1T was regulated by a genetic ON/OFF switch that resulted in premature termination of the PTS permease and epitope variation [23]. The monoclonal antibody targeting this epitope, namely 3F3, was shown to bind to TR but not to OP colonies (Figure 6). This may suggest that truncation of the glucose PTS permease in OP variants results in less efficient glucose transport and, consequently, less efficient glycolysis and EPS synthesis.

Figure 5. Growth, pH and exopolysaccharide (EPS) secretion of translucent (blue) and opaque (red) colony variants in CMRL-1066 medium.

50 ml of CMRL-1066 medium were inoculated with 108–109 CFU/mL and incubated for 96 hours. Viable titer, pH and EPS production were determined every 24 h.

Figure 6. Variation in Mmm colony opacity associated to antigenic variation of the glucose PTS permease.

A: Mmm strain Afadé colonies on PPLO supplemented agar observed under stereomicroscope with indirect light. B: Colony-blotting performed from the same plate using the 3F3 monoclonal antibody. (Magnification: x10).

The frequency of reversion from 3F3-negative (3F3NEG) to 3F3-positive (3F3POS) variants of strain Afadé was shown to vary from 10−6 to 10−2 [23]. In the present study, this reversion resulted in less than 0.2% of 3F3NEG (OP colony variant) in 3F3POS cells (TR colony variant) or in Mmm strain Afadé cultures, and a proportion of 1 to 5% of 3F3POS (TR colony variant) cells in 3F3NEG (OP colony variant) cultures at the time of transfer into CMRL-1066 medium. This may account for the weak EPS production obtained from the OP (3F3NEG) colony variant. We then attempted to minimize the proportion of TR (3F3POS) cells in OP (3F3NEG) cultures in order to assess EPS production by the OP colony (3F3NEG) variant cells. Several cultures were tested and the one with the lowest titer of 3F3POS (TR colony variant) cells (i.e. 0.5%) was used to inoculate CMRL-1066 medium. Under these conditions, EPS production was very low, whatever the sugar source (Table 1). However, addition of different sugars to the CMRL-1066 medium appeared to modulate the level of reversion to 3F3POS (TR colony variant) after 72 hours of incubation. Mannitol was the only carbohydrate which led to an increase of 3F3POS (TR colony variant), from 0.5% to 12%, but the EPS release was not enhanced, due to considerable cell decay. This suggests the existence of complex regulations for transport and catabolism of the different sugars.

4. In silico Analysis of Candidate Galactan Biosynthetic Pathways

We showed that TR colony variants, which were detected by 3F3 monoclonal antibody, produced free, extracellular galactan. In contrast, OP colony variants, which did not display any reaction with the 3F3 antibody, because of a premature termination of their glucose PTS permease, produced capsular galactan. Polysaccharide biosynthesis by OP colony variants was intriguing, since truncation of the glucose PTS permease would be expected to result in an overall less efficient glucose transport and therefore hampering notably of polysaccharide synthesis.

We tried to decipher the biochemical links between glucose intake, polysaccharide chain elongation and attachment/release of the galactan by retrieving in silico all potentially involved Mmm genes. The genomes of the 2 currently sequenced Mmm strains, namely PG1T (accession number NC_005364.2) [5] and Gladysdale (accession number CP002107.1) [6], were used.

As NMR analyses revealed that galactan is composed of galactose residues in the furanose ring form, we first looked for enzymes involved in UDP-galactofuranose synthesis, which is the usual precursor of a galactofuranose polymer. UDP-galactofuranose is usually formed from UDP-galactopyranose after the action of an UDP galactofuranose mutase Glf [50], [51]. One glf gene identified in Mmm strain PG1T [5] is also present in the Gladysdale genome. In bacteria, UDP-galactopyranose can be synthesized via two enzymes of the Leloir pathways [52], either from galactose-1-phosphate after the action of a galactose-1-phosphate uridylyltransferase (GalT) or from UDP-glucose via an UDP-glucose 4-epimerase (GalE). No galT homologue was retrieved from the genomes of Mmm, indicating that galactose is not a source for galactan synthesis. This is concordant with our experiment showing that galactose was unable to enhance galactan synthesis (Table 1). In contrast the glucose-dependent synthetic pathway is highly probable since two genes were identified in both genomes, one of which encodes for UDP-glucose 4-epimerase (GalE) and the other for a glucose-1-phosphate uridylyltransferase (GalU) that uses glucose-1-phosphate to generate UDP-glucose.

Glucose intake in Mmm certainly relies on two similar (47%) glucose PTS permeases (MSC_0054 and MSC_0860/873) that have been identified in strain PG1T [23] and are also present in the Gladysdale strain. The PTS glucose permeases usually allow the concomitant uptake and phosphorylation of glucose into glucose-6-phosphate [53], the typical precursor of glycolysis. The probability that glucose intake might be mediated by other generic permeases, such as ABC transporter, is low since it would require further activation of glucose by phosphorylation that is impaired because of the mutation of the two glucokinase genes in PG1T [23]. Entry into the galactan biosynthetic pathway further requires the isomerization of glucose-6-phosphate to glucose-1-phosphate, which is most probably achieved by the phosphoglucomutase (ManB) predicted in Mmm (MSC_0829, [5]). It has been shown recently in Vibrio cholerae that glucose PTS permeases are general regulators of genes involved in EPS synthesis [54]. In our model, we hypothesize that MSC_0860/873 might play a role in the regulation of the attachment/release of galactan, its impairment in OP variants resulting in a constitutive attachment of the galactan.

Three predicted glycosyltransferases (GT), cps (MSC_0109), EpsG (MSC_0108) and MSC_0771 previously identified in the Mmm PG1T genome [5] are good candidates for elongation of the galactan chain from the UDP-galactofuranose precursor. MSC_0109 (cps) showed 45% and 42% similarity with glycolipid synthases described in M. genitalium [55] and M. pneumoniae [56], respectively. In Mycoplasmas, glycolipid synthases are able to transfer galactosyl and glucosyl residues to membrane-bound diacylglycerol [57]. Therefore, MSC_0109 is a good candidate for galactan attachment to the cytoplasmic membrane. This was supported by preliminary northern blot experiments that showed a differential expression of the cps gene in capsulated versus non-capsulated colony variants (Figure S1).

Functional prediction of EpsG using HHPred [34] and Phyre2 [33] revealed a structure that was highly similar (e-value = 3.4e-38, probability 100%) to that of the cellulose synthase BcsA of Rhodobacter sphaeroides [58]. Synthases are membrane-embedded glycosyltransferases involved in polymer formation and translocation across the cytoplasmic membrane [59]. Four potential transmembrane regions were predicted by TMHMM2 [32] for EpsG, with a large cytoplasmic loop (36–312) (Figure S2) which showed homology with family 2 glycosyltranferases (, [31]), a group of GT known to add sugars in the β-configuration which is in agreement with the galactan structure. Two conserved motifs (DXD and QxxRW), which are common markers of processive enzymes in various bacterial models [60], were found in the cytoplasmic loop (Figure S2). Further structural analysis showed that the cytoplasmic domain was similar (e-value = 2.8e-26, probability 99.93%) to the UDP-galactofuranosyl glycotransferase GlfT of Mycobacterium turberculosis [61]. We therefore hypothesized that EpsG is a putative galactan synthase which could catalyze the polymerization of β-1,6-linked galactofuranose and facilitate its export across the cytoplasmic membrane. Furthermore, synthase-dependent exopolysaccharide secretion was shown to occur in the presence or absence of a lipid acceptor molecule [59]. In our model, synthase-dependent galactan synthesis might be active whether galactan is attached to a glycolipid anchor, or not.

The third predicted GT, MSC_0771, was not related to any of the 90 GT families described so far. Structural analysis revealed two domains of which only the N-terminal one (1–199) was similar to diverse bacterial and eukaryotic glycosyltransferases. Its role in galactan biosynthesis remains hypothetical.

In conclusion, this in silico analysis points toward i) a synthase-dependent pathway for galactan synthesis, ii) a putative role of MSC_0109 in attachment of galactan to the membrane and iii) a potential regulation of this attachment by glucose PTS permease. However detailed functional studies are now needed to unravel the precise roles of each GT and the genetic basis and triggering conditions underlying the switch between capsule production (OP colony variants) and EPS secretion (TR colony variants).


We have developed a simple method for the purification of Mmm EPS. The incubation of washed mycoplasma cells in a completely defined medium, able to support cellular metabolism, allowed EPS synthesis for at least 72 hours, while preventing contamination with polysaccharides from the culture medium. Preliminary results in our laboratory indicate that this method is suitable for use in other mycoplasmas.

This method was applied to characterize the free EPS secreted by Mmm. Purified EPS had an identical β(1−>6)-galactofuranosyl structure to that of Mmm capsular galactan. The presence of OP and TR colony variants in Mmm cultures, which varied greatly in their capacity to produce free EPS, was observed. Both variants appeared to be able to synthesize polysaccharides. However, these polysaccharides either remained attached to the cells, constituting a capsule (opaque colony variants), or were released into the culture medium as free EPS (translucent colony variants). We finally conducted in silico analyses of Mmm genes potentially involved in polysaccharide biosynthesis and proposed candidate pathways that might account for the alternative production of capsular versus free galactan by the corresponding opaque and translucent colony variants.

In Streptococcus pneumoniae spontaneous phase variation between opaque and translucent colony variants has been associated with diverse levels of virulence at different stages of the disease, from invasion to transmission [62]. It remains to be elucidated whether variations in the production of capsular versus free galactan can regulate Mmm pathogenicity and how. Although free galactan has been found in the body fluids and urine of infected animals [12], [63], the occurrence and proportion of EPS versus capsule-producing variants in vivo has yet to be determined. This would constitute a key step in the understanding of Mmm pathogenicity, since the emergence of sub-populations in a given ecological niche is known to lead to bacterial persistence.

Supporting Information

Figure S1.

Northern blot hybridization of total RNA of the opaque (OP) and translucent (TR) colony variants of Mmm strain Afadé with a cps (MSC_0109) or rDNA 16S probe. Total RNA extraction and northern blot hybridization was performed as previously described [1]. The rDNA 16S probe was obtained by PCR [2]. Transcription of the rDNA 16S was used to normalize the hybridization. The cps gene probe was obtained by PCR with specific primers (5′ TGATGGATCAACAGATAACACCA 3′ and 5′ TTTGGGCGTGAGTATCAATAAG 3′).


Figure S2.

Schematic representation of the predicted membrane topology of Mmm EpsG (MSC_0108) glycosyltransferase. Numbers indicate the localization of transmembrane helices. DxD (red) and RxxQW (blue) motifs are showed.



We express our gratitude to A. Paoli and V. Lefriand for technical assistance.

Author Contributions

Conceived and designed the experiments: PG CB. Analyzed the data: PG F. Tardy CPR JC. Project conception: FP F. Thiaucourt LMS. Electron microscopy: PG DLG. Monomer composition and NMR analysis: CPR JC. All other experiments: PG CB. Drafted the manuscript: F. Tardy PG CPR LMS. Critically reviewed the manuscript and approved the final version: CB CPR JC LMS F. Thiaucourt F. Tardy DLG FP PG.


  1. 1. Razin S, Yogev D, Naot Y (1998) Molecular biology and pathogenicity of mycoplasmas. Microbiol Mol Biol Rev 62: 1094–1156.
  2. 2. Yus E, Maier T, Michalodimitrakis K, van Noort V, Yamada T, et al. (2009) Impact of genome reduction on bacterial metabolism and its regulation. Science 326: 1263–1268.
  3. 3. OIE (2008) Manual of diagnostic tests and vaccines for terrestrial animals. Chapter 2.4.9. Contagious bovine PleuroPneumonia. Available: Accessed 2013 Feb 8.
  4. 4. Nicholas RA (2011) Bovine mycoplasmosis: silent and deadly. Vet Rec 168: 459–462.
  5. 5. Westberg J, Persson A, Holmberg A, Goesmann A, Lundeberg J, et al. (2004) The genome sequence of Mycoplasma mycoides subsp. mycoides SC type strain PG1T, the causative agent of contagious bovine pleuropneumonia (CBPP). Genome Res 14: 221–227.
  6. 6. Wise KS, Calcutt MJ, Foecking MF, Madupu R, DeBoy RT, et al. (2012) Complete genome sequences of Mycoplasma leachii strain PG50T and the pathogenic Mycoplasma mycoides subsp. mycoides small colony biotype strain Gladysdale. J Bacteriol 194: 4448–4449.
  7. 7. Pilo P, Frey J, Vilei EM (2007) Molecular mechanisms of pathogenicity of Mycoplasma mycoides subsp. mycoides SC. Vet J 174: 513–521.
  8. 8. Buttery J, Moxon ER (2002) Capsulate bacteria and the lung. Br Med Bull 61: 63–80.
  9. 9. Buttery SH, Plackett P (1960) A specific polysaccharide from Mycoplasma mycoides. J Gen Microbiol 23: 357–368.
  10. 10. Plackett P, Buttery SH (1963) A Galactofuranose Disaccharide from the Galactan of Mycoplasma mycoides. Biochem J 90: 201–205.
  11. 11. Branda SS, Vik S, Friedman L, Kolter R (2005) Biofilms: the matrix revisited. Trends Microbiol 13: 20–26.
  12. 12. Turner AW (1962) Circulating M. mycoides antigen as a cause of loss agglutination and complement fixation reactivity during acute pleuropneumonia. Austral Vet J 38: 401.
  13. 13. March JB, Jones GE, Hitchen P, Morris HR, Dell A (1999) Analysis of the capsular polysaccharide of Mycoplasma mycoides subsp. mycoides SC, the causal agent of CBPP: purification, composition and its role in infection and immunity. COST: 69–72.
  14. 14. March JB, Brodlie M (2000) Comparison of the virulence of European and African isolates of Mycoplasma mycoides subspecies mycoides small colony type. Vet Rec 147: 20–21.
  15. 15. Shifrine M, Gourlay RN (1965) Serological relationship between galactans from normal bovine lung and from Mycoplasma mycoides. Nature 208: 498–499.
  16. 16. Buttery SH, Lloyd LC, Titchen DA (1976) Acute respiratory, circulatory and pathological changes in the calf after intravenous injections of the galactan from Mycoplasma mycoides subsp. mycoides. J Med Microbiol 9: 379–391.
  17. 17. Hudson JR, Buttery S, Cottew GS (1967) Investigations into the influence of the galactan of Mycoplasma mycoides on experimental infection with that organism. J Pathol Bacteriol 94: 257–273.
  18. 18. Lloyd LC, Buttery SH, Hudson JR (1971) The effect of the galactan and other antigens of Mycoplasma mycoides var. Mycoides on experimental infection with that organism in cattle. J Med Microbiol 4: 425–439.
  19. 19. McAuliffe L, Ayling RD, Ellis RJ, Nicholas RA (2008) Biofilm-grown Mycoplasma mycoides subsp. mycoides SC exhibit both phenotypic and genotypic variation compared with planktonic cells. Vet Microbiol 129: 315–324.
  20. 20. Pilo P, Vilei EM, Peterhans E, Bonvin-Klotz L, Stoffel MH, et al. (2005) A metabolic enzyme as a primary virulence factor of Mycoplasma mycoides subsp. mycoides small colony. J Bacteriol 187: 6824–6831.
  21. 21. Abdo el M, Nicolet J, Miserez R, Goncalves R, Regalla J, et al. (1998) Humoral and bronchial immune responses in cattle experimentally infected with Mycoplasma mycoides subsp. mycoides small colony type. Vet Microbiol 59: 109–122.
  22. 22. Poumarat F, Perrin B, Longchambon D (1991) Identification of ruminant mycoplasmas by dot immunobinding on membrane filtration (MF dot). Vet Microbiol 29: 329–338.
  23. 23. Gaurivaud P, Persson A, Grand DL, Westberg J, Solsona M, et al. (2004) Variability of a glucose phosphotransferase system permease in Mycoplasma mycoides subsp. mycoides Small Colony. Microbiology 150: 4009–4022.
  24. 24. Brocchi E, Gamba D, Poumarat F, Martel JL, De Simone F (1993) Improvements in the diagnosis of contagious bovine pleuropneumonia through the use of monoclonal antibodies. Rev Sci Tech 12: 559–570.
  25. 25. Cochran WG (1950) Estimation of bacterial densities by means of the “most probable number”. Biometrics 6: 105–116.
  26. 26. De Vuyst L, Vanderveken F, Van de Ven S, Degeest B (1998) Production by and isolation of exopolysaccharides from Streptococcus thermophilus grown in a milk medium and evidence for their growth-associated biosynthesis. J Appl Microbiol 84: 1059–1068.
  27. 27. Dubois M, Hamilton JK, Rebers PA, Smith F (1956) Colorimetric Method for Determination of Sugars and Related Substances. analytical chemistry 28: 350–356.
  28. 28. Shi F, Harada T, Ogawa Y, Ono H, Ohnishi-Kameyama M, et al. (2012) Capsular Polysaccharide of Erysipelothrix rhusiopathiae, the Causative Agent of Swine Erysipelas, and Its Modification with Phosphorylcholine. Infect Immun 80: 3993–4003.
  29. 29. Enos-Berlage JL, McCarter LL (2000) Relation of capsular polysaccharide production and colonial cell organization to colony morphology in Vibrio parahaemolyticus. J Bacteriol 182: 5513–5520.
  30. 30. Barré A, de Daruvar A, Blanchard A (2004) MolliGen, a database dedicated to the comparative genomics of Mollicutes. Nucleic Acids Res 32: D307–310.
  31. 31. Cantarel BL, Coutinho PM, Rancurel C, Bernard T, Lombard V, et al. (2009) The Carbohydrate-Active EnZymes database (CAZy): an expert resource for Glycogenomics. Nucleic Acids Res 37: D233–238.
  32. 32. Moller S, Croning MD, Apweiler R (2001) Evaluation of methods for the prediction of membrane spanning regions. Bioinformatics 17: 646–653.
  33. 33. Kelley LA, Sternberg MJ (2009) Protein structure prediction on the Web: a case study using the Phyre server. Nat Protoc 4: 363–371.
  34. 34. Söding J, Biegert A, Lupas AN (2005) The HHpred interactive server for protein homology detection and structure prediction. Nucleic Acids Res 33: W244–248.
  35. 35. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, et al. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402.
  36. 36. Plackett P, Buttery SH, Cottew GS (1963) Carbohydrates of some mycoplasma strains. Recent Progress in Microbiology Symposia of the 8th congress of Microbiology, Montreal, Canada: 535–547.
  37. 37. Freundt EA (1983) Culture media for classic mycoplasmas. Methods in Mycoplasmology volume 1: mycoplasma characterization: 127–135.
  38. 38. Rodwell AW, Abbot A (1961) The function of glycerol, cholesterol and long-chain fatty acids in the nutrition of Mycoplasma mycoides. J Gen Microbiol 25: 201–214.
  39. 39. Goncalves VM, Takagi M, Carneiro SM, de Campos Giordano R, Tanizaki MM (2006) Introduction of air in the anaerobic culture of Streptococcus pneumoniae serotype 23F induces the release of capsular polysaccharide from bacterial surface into the cultivation medium. J Appl Microbiol 101: 1009–1014.
  40. 40. Sutherland IW (1972) Bacterial exopolysaccharides. Adv Microb Physiol 8: 143–213.
  41. 41. Smith GR, Oliphant JC (1982) Some in-vitro characters of the subspecies of Mycoplasma mycoides. J Hyg (Lond) 89: 521–527.
  42. 42. Beynon LM, Richards JC, Perry MB (1997) Identification of the common antigenic determinant shared by Streptococcus pneumoniae serotypes 35A and 20 capsular polysaccharides–structural analysis of the Streptococcus pneumoniae serotype 35A capsular polysaccharide. Eur J Biochem 250: 163–167.
  43. 43. Cordeiro LM, Beilke F, Bettim FL, Reinhardt Vde F, Rattmann YD, et al. (2012) (1–>2) and (1–>6)-linked beta-D-galactofuranan of microalga Myrmecia biatorellae, symbiotic partner of Lobaria linita. Carbohydr Polym 90: 1779–1785.
  44. 44. de Rezende CE, Anriany Y, Carr LE, Joseph SW, Weiner RM (2005) Capsular polysaccharide surrounds smooth and rugose types of Salmonella enterica serovar Typhimurium DT104. Appl Environ Microbiol 71: 7345–7351.
  45. 45. Yildiz FH (1999) Schoolnik GK (1999) Vibrio cholerae O1 El Tor: identification of a gene cluster required for the rugose colony type, exopolysaccharide production, chlorine resistance, and biofilm formation. Proc Natl Acad Sci U S A 96: 4028–4033.
  46. 46. Liss A, Heiland RA (1983) Colonial opacity variation in Mycoplasma pulmonis. Infect Immun 41: 1245–1251.
  47. 47. Rosengarten R, Wise KS (1991) The Vlp system of Mycoplasma hyorhinis: combinatorial expression of distinct size variant lipoproteins generating high-frequency surface antigenic variation. J Bacteriol 173: 4782–4793.
  48. 48. Minga UM (1981) Smooth and rough colony types of Mycoplasma mycoides subsp. mycoides: in vitro characterization and pathogenicity in cattle. Tanzania Veterinary Bulletin 3: 53–62.
  49. 49. Swanson J (1978) Studies on gonococcus infection. XII. Colony color and opacity varienats of gonococci. Infect Immun 19: 320–331.
  50. 50. Koplin R, Brisson JR, Whitfield C (1997) UDP-galactofuranose precursor required for formation of the lipopolysaccharide O antigen of Klebsiella pneumoniae serotype O1 is synthesized by the product of the rfbDKPO1 gene. J Biol Chem 272: 4121–4128.
  51. 51. Kremer L, Dover LG, Morehouse C, Hitchin P, Everett M, et al. (2001) Galactan biosynthesis in Mycobacterium tuberculosis. Identification of a bifunctional UDP-galactofuranosyltransferase. J Biol Chem 276: 26430–26440.
  52. 52. Holden HM, Rayment I, Thoden JB (2003) Structure and function of enzymes of the Leloir pathway for galactose metabolism. J Biol Chem 278: 43885–43888.
  53. 53. Saier MH Jr, Reizer J (1992) Proposed uniform nomenclature for the proteins and protein domains of the bacterial phosphoenolpyruvate: sugar phosphotransferase system. J Bacteriol 174: 1433–1438.
  54. 54. Houot L, Chang S, Pickering BS, Absalon C, Watnick PI (2010) The phosphoenolpyruvate phosphotransferase system regulates Vibrio cholerae biofilm formation through multiple independent pathways. J Bacteriol 192: 3055–3067.
  55. 55. Andrés E, Martinez N, Planas A (2011) Expression and characterization of a Mycoplasma genitalium glycosyltransferase in membrane glycolipid biosynthesis: potential target against mycoplasma infections. J Biol Chem 286: 35367–35379.
  56. 56. Klement MLR, Ojemyr L, Tagscherer KE, Widmalm G, Wieslander A (2007) A processive lipid glycosyltransferase in the small human pathogen Mycoplasma pneumoniae: involvement in host immune response. Mol Microbiol 65: 1444–1457.
  57. 57. Andrés E, Biarnés X, Faijes M (2012) AP (2012) Bacterial glycoglycerolipid synthases: processive and non-processive glycosyltransferases in mycoplasma. Biocatalysis and Biotransformation 30: 274–287.
  58. 58. Morgan JL, Strumillo J, Zimmer J (2013) Crystallographic snapshot of cellulose synthesis and membrane translocation. Nature 493: 181–186.
  59. 59. Whitney JC, Howell PL (2013) Synthase-dependent exopolysaccharide secretion in Gram-negative bacteria. Trends Microbiol 21: 63–72.
  60. 60. Saxena IM, Brown RM Jr, Fevre M, Geremia RA, Henrissat B (1995) Multidomain architecture of beta-glycosyl transferases: implications for mechanism of action. J Bacteriol 177: 1419–1424.
  61. 61. Wheatley RW, Zheng RB, Richards MR, Lowary TL, Ng KK (2012) Tetrameric structure of the GlfT2 galactofuranosyltransferase reveals a scaffold for the assembly of mycobacterial Arabinogalactan. J Biol Chem 287: 28132–28143.
  62. 62. Hava DL, LeMieux J, Camilli A (2003) From nose to lung: the regulation behind Streptococcus pneumoniae virulence factors. Mol Microbiol 50: 1103–1110.
  63. 63. Gourlay RN (1962) Polysaccharide haptens from urine of cattle infected with Mycoplasma mycoides. Nature 195: 99.