Glycobiology Advance Access originally published online on April 24, 2008
Glycobiology 2008 18(7):502-508; doi:10.1093/glycob/cwn031
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Comparative structural analyses of the 
-glucan and glycogen from Mycobacterium bovis
Département ‘Mécanismes Moléculaires des Infections Mycobactériennes’, Institut de Pharmacologie et Biologie Structurale (UMR 5089), Université Paul Sabatier (Toulouse III) and Centre National de la Recherche Scientifique, 205, route de Narbonne, 31077 Toulouse cedex, France
1 To whom correspondence should be addressed: Tel: (+33)-561-175-568; Fax: (+33)-561-175-580; e-mail: anne.lemassu{at}ipbs.fr
Received on August 24, 2007; revised on April 16, 2008; accepted on April 18, 2008
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsFundingConflict of interest statementReferences |
|---|
Pathogenic mycobacteria such as Mycobacterium tuberculosis, the causative agent of tuberculosis, are surrounded by a noncovalently bound capsule, whose major carbohydrate constituent is a glycogen-like
-glucan. In the present study we compared the structures of the extracellular polysaccharide to that of the ubiquitous intracellular glycogen. The -glucan was isolated from the culture medium of Mycobacterium bovis Bacille Calmette Guérin, the vaccine strain, in which it is released whereas the intracellular glycogen was obtained after the disruption of cells. The two purified polysaccharides were eluted from permeation gel at a similar position but glycogen was less soluble and gave a more opalescent solution in water than -glucan. Combination of gas chromatography-mass spectrometry analysis of partially O-methylated, partially O-acetylated alditols and NMR analysis confirmed that both polysaccharides were composed of 
4--D-Glcp-1 core, substituted at some six positions with short chains. Degradation of polysaccharides with pullulanase, followed by mass spectrometry analysis of the resulting products, also showed that the two polysaccharides do not differ in terms of lengths of branching. Interestingly, application of analytical ultracentrifugation and dynamic light scattering to the mycobacterial -glucan and glycogen and their enzymatic degradative products indicated that the -glucan possessed a higher molecular mass and was more compact than the glycogen from the same species, allowing the formulation of working structural models for the two polysaccharides. Consistent with the models, the -glucan was found to be less accessible to pullulanase, a debranching enzyme, than glycogen. Key words: capsule / DLS / Mycobacterium tuberculosis / polysaccharide / pullulanase
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingConflict of interest statementReferences |
|---|
Tuberculosis remains one of the most serious health problems worldwide and the most widespread infectious disease due to a single agent, with an estimated one-third of the population infected by Mycobacterium tuberculosis, the causative agent of tuberculosis, and 2–3 million people deaths annually. While the disease has always been a major health problem in developing countries, the recent re-increment in parts of industrialized countries has led to energetic sanitary programs to fight against mycobacterial infections. Although tuberculosis is one of the oldest known diseases, the mechanisms developed by its causative agent during the infection are not well understood. M. tuberculosis and pathogenic mycobacterial species are not only able to survive in their host, but they are also capable to multiply in an hostile environment such as the macrophage, an ability that is believed to represent one of the key steps of the pathogenicity of these microorganisms.
The envelope of the tubercle bacillus, at the interface between the pathogen and its hostile environment, is very complex. Although mycobacteria are classified among the Gram-positive bacteria, they possess an envelope structurally related to, but probably more complex than, that of Gram-negative bacteria. Schematically the mycobacterial envelope is composed of a typical plasma membrane and a thick cell wall that consists of a peptidoglycan covalently linked to the heteropolysaccharide D-arabino-D-galactan, which in turn is esterified by long chain (up to C90) -branched β-hydroxylated fatty acids, called mycolic acids (Daffe and Draper 1998
). The unusual structure of the cell wall confers to mycobacteria their impermeability to small molecules that include nutrients and antibiotics (Nikaido and Jarlier 1991). Additionally, the outermost compartment of the cell envelope of pathogenic mycobacterial species consists in a loosely bound structure called capsule (Daffe and Draper 1998). We have previously shown that the capsular material is composed of carbohydrate and protein, with a tiny amount of species- or type-specific lipids (Ortalo-Magne et al. 1995; Lemassu et al. 1996). Although part of this material is released into the culture medium of the in vitro-grown bacteria (Ortalo-Magne et al. 1995), the capsular components are found around the in vivo bacteria (Schwebach et al. 2002), probably retained by the phagosomal membrane (Daffe and Draper 1998). The major carbohydrate constituent of the capsule of M. tuberculosis, representing up to 80% of the extracellular polysaccharides, is an -glucan that was recently shown to induce monocytes to differentiate into altered dendritic cells, allowing mycobacteria to evade immune host response (Gagliardi et al. 2007). Mycobacterial -glucan is composed of a 4--D-Glc-1 core branched every 5 or 6 residues by mono- and di- glucosides, with a molecular mass estimated to be about 100,000 Da by gel permeation (Lemassu and Daffe 1994; Ortalo-Magne et al. 1995). The mycobacterial glycogen has been described as a highly branched polysaccharide, with chains shorter than those of glycogens isolated from Escherichia coli or human liver (Antoine and Tepper 1969a). In eucaryotic cells the intracellular glycogen is formed by a tree-like structure of about 106–109 Da (Preiss 1984) and is less branched than the -glucan. Thus, mycobacteria, notably M. tuberculosis, produce two structurally related highly branched glucans differing from one another by their cell location. This raises the question of the origin and, as a consequence, the structural differences between the two polysaccharides. To address this latter point, the present study was undertaken by comparatively analyzing the structural features of the capsular -glucan and the intracellular glycogen purified from the vaccine strain Mycobacterium bovis Bacille Calmette Guérin (BCG).
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingConflict of interest statementReferences |
|---|
Purification and determination of the glycosyl-linkage composition and ring forms of the
-glucan and glycogen from M. bovis BCGBecause the structural features of glycogen are known to be age-dependent (Antoine and Tepper 1969b
), the mycobacterial -glucan and intracellular glycogen analyzed herein were purified from the same culture batches. The -glucan present in the culture medium is believed to be derived from the capsular material and shares the same structure with the latter polysaccharide (Lemassu and Daffe 1994; Ortalo-Magne et al. 1995). Accordingly, and for convenience, the -glucan was extracted from the macromolecules of the culture fluid as described (Lemassu and Daffe 1994). The corresponding cells were disrupted and the intracellular glycogen was recovered from the cytosolic fraction by adding chloroform to the acellular extract in a glycine buffer (Bueding and Orrell 1964). The purification process was monitored by determining the carbohydrate content of the fractions and by 1H-NMR analysis. That the polysaccharides were readily pure was ascertained by dynamic light scattering (DLS). As expected, this technique showed the monodispersity of polysaccharides (data not shown). Gel permeation chromatography was performed on polyacrylamide gel, to avoid contamination by dextran originated from the sephadex column (Dinadayala et al. 2004). The two mycobacterial polysaccharides were included in the Bio-Gel P-200 gel and gave a single chromatographic peak (Figure 1).
|
The lyophilized mycobacterial glycogen forms particles, as does the commercial rabbit glycogen, whereas the
-glucan exhibits a downy aspect. Furthermore, when solubilized in water, the intracellular mycobacterial glycogen was obviously more opalescent and less soluble than the -glucan. This difference in solubility between the two polysaccharides was the first indicator of a difference in terms of structure and/or ultrastructure.
Gas chromatography (GC) analysis of the trimethylsilyl derivatives of the acid hydrolysis products of the purified polysaccharides showed that both classes of compounds were exclusively composed of Glc. GC-mass spectrometry (GC-MS) analysis of the partially O-methylated, partially O-acetylated alditols derived from the polysaccharides showed that both consisted of the three expected types of residues, i.e., (i) terminal Glc-(1 residue (characteristic ions fragments observed in MS in the electronic impact mode at m/z 161, 162, 118, and 205), (ii) 4)-Glcp-(1 or 5-linked-Glcf (mass peaks at m/z 233 and 118), and (iii) 4,6)-Glcp-(1 or 5,6-Glcf residue (peaks at m/z 118 and 261). The pyranosyl conformation of the glycosyl residues that composed the polysaccharides was established from the examination of the 13C-NMR spectra (data not shown) that showed a signal at 100.5 ppm corresponding to the resonances of anomeric carbons in -glucopyranosyl units (Zang et al. 1991; Dinadayala et al. 2004). The 1H-NMR spectra of the polysaccharides (data not shown) were similar to those published for the -glucan of M. bovis BCG (Dinadayala et al. 2004) and eukaryotic glycogen (Zang et al. 1991).
Determination of the molecular masses of the intact -glucan and glycogen of M. bovis BCG
The molecular mass of the -glucan of M. tuberculosis was previously estimated to be about 100,000 Da by gel permeation (Lemassu and Daffe 1994) whereas analytical ultracentrifugation has shown that the intracellular mycobacterial glycogen exhibited a mass of 107 Da (Antoine and Tepper 1969a). Accordingly, the molecular masses of the two polysaccharides isolated from M. bovis were comparatively investigated by the two approaches. Unexpectedly, the -glucan from M. bovis and the commercial dextran T110 (estimated MM of 110,000 Da) exhibited a different elution volume on the Bio-Gel P-200 column. Furthermore, both the intracellular mycobacterial glycogen and commercial rabbit glycogen were eluted from the permeation gel at a position similar to that of the -glucan (Figure 1). These data indicated that gel permeation chromatography was not suitable for estimating the molecular masses of polysaccharides such as dextran T110 and glycogens, with apparent masses between 105 and 107 Da, probably because this technique is more related to the sizes and spatial conformations than to the masses of polysaccharides.
Analytical ultracentrifugation, a technique currently used for determining the molecular weights of proteins, has been less frequently used for polysaccharides, with few exceptions (Harding 2005). Among these is the work of Brammer and colleagues who have tentatively correlated the sedimentation coefficient and the molecular masses of a number of saccharides (Brammer et al. 1972). We first validated the technique by determining the molecular masses of commercial samples, used as controls. Using this technique we determined a sediment coefficient of 8.24 S and a molecular mass of 108 000 Da for the commercial 110 000 Da dextran (T110). Similarly, the rabbit glycogen gave a sediment coefficient of 100 S, leading to a calculated molecular mass of 7.0 x 106 Da. Both mass values are in good agreement with those estimated by other techniques (Brammer et al. 1972), validating the approach. Accordingly, analytical ultracentrifugation was used to determine the sedimentation coefficients of the glycogen and glucan purified from M. bovis BCG. The mycobacterial glycogen exhibited a sediment coefficient of 104 S, which corresponds to a calculated molecular mass of 7.5 (±0.2) 106 Da, a value that was in the same order of magnitude than that of the rabbit glycogen. Unexpectedly, however, in view of the molecular mass determined by gel permeation, the -glucan of M. bovis BCG gave a sedimentation coefficient of 154 S, which corresponds to a calculated molecular mass of 13 (±1.6) 106 Da, twice bigger than that determined for the glycogen of M. bovis and 1000-fold larger than that estimated by gel permeation chromatography (Figure 1).
Conformational analyses of the -glucan and glycogen of M. bovis BCG
We hypothesized that the difference observed between the masses of the two polysaccharides by ultracentrifugation should have an impact on other physical parameters of the compounds and, accordingly, investigated additional gross structural features of the polysaccharides by both polarimetry and DLS.
Several centrifugations were necessary to obtain reproducible []D values for mycobacterial glycogen, probably because of its low solubility while no such treatment was required for -glucan. Polarimetry is a powerful technique that is widely used in carbohydrate analysis because it may be very informative. In the case of complex polysaccharides, however, the optical rotations ([]D) are more related to the gross conformation of the compounds than to its sugar composition (Rees 1972). []D values of 170° and 75° were measured for the -glucan and glycogen of M. bovis BCG, respectively, validating our assumption. Furthermore, the dynamic radii determined by DLS for the two polysaccharides also showed that they differed significantly from one another (Figure 2). The molecular radius of the -glucan from M. bovis (27.9 ± 2.4 nm, Pd: 23%) was significantly smaller than that of the intracellular glycogen from M. bovis (35.2 ± 3.1 nm, Pd: 26%) (Figure 2). Applied to the commercial rabbit glycogen and dextran T110, the radii of 23.0 ± 0.1 nm (Pd: 28%) and 8.80 ± 0.02 nm (Pd: 20%), respectively, were obtained (Figure 2), in accordance with their molecular masses.
|
Branching of the polysaccharides
The observed differences in terms of molecular masses and global conformations of the two polysaccharides may be attributable to differences in branching. Accordingly, the lengths of the six-branched chains in the two types of polysaccharides from M. bovis were determined by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) MS of per-O-acetylated fragments released after their digestion with pullulanase, an enzyme that specifically hydrolyses Glc-
(16)-Glc linkages (Dinadayala et al. 2004). The efficiency of the enzymatic hydrolysis was monitored by the disappearance of the signal resonances at 5.00 ppm in the 1H-NMR spectra. The digestion of Pullulan into triglucosides was used as the positive control, whereas maltoheptaose, from which no degradation was observed, was the negative control.
The digestion products of the -glucan and intracellular glycogen from M. bovis exhibited similar patterns, mainly of 2–6 residues; minor oligosaccharides with 7–9 units were also present in the mixtures of per-O-acetylated oligosaccharides analyzed (Figure 3). The presence of per-O-acetylated monoglucosides, undetected by MALDI-TOF analysis (because of the matrix ion peaks recovering the corresponding signal expected at m/z = 413), was established by thin-layer chromatography analysis of the degradation products of the two polysaccharides. No higher oligosaccharide derivatives detectable by MALDI-TOF analysis, whose detection limit is about 9000 Da (30 residues), were observed in the mass spectra. These data indicated that the two polysaccharides contained branched chains of similar lengths.
|
The possible difference in branching was also examined by 1H-NMR spectroscopy. As a difference in branching may explain a difference of solubility between the two polysaccharides, the integration ratios of signals corresponding to the resonances of 4-linked or terminal
-D-Glcp-(16) anomeric protons versus those terminal, branched or linear -D-Glcp-(14) anomeric protons were determined; these values would reflect the branching ratios in the two polysaccharides. The spectra of the glucan and glycogens were recorded and analyzed. The two types of signals were observed at 5.38 ppm (attributable to the resonances of terminal, branched or linear -D-Glcp-(14) anomeric protons) and 5.00 ppm (assignable to the resonances of 4-linked or terminal -D-Glcp-(16) anomeric protons). The ratios determined for the -glucan and glycogen from M. bovis BCG were very similar [1/(9 ± 0.7) and 1/(8.8 ± 0.4)] but significantly different from that of the commercial rabbit glycogen [1/(13.6 ± 3.0)]. Thus, the difference in solubility between the two polysaccharides isolated from M. bovis was not due to a difference in branching.
To determine the lengths of the backbone chains, not detectable by MALDI-TOF due to their sizes, these were analyzed by DLS. The completion of the enzymatic digestion was followed by 1H-NMR analysis. While the backbone of intracellular M. bovis BCG glycogen exhibited a radius similar to that of the commercial bovine, about 15 nm, the radius of the M. bovis BCG -glucan main chain (25 nm) was significantly higher than that of glycogens (Figure 4).
|
Altogether, the data point to structural differences between the capsular
-glucan and the intracellular glycogen from M. bovis BCG: the two compounds exhibit similar branching ratio but the latter polysaccharide exhibits a higher hydrodynamic radius but contains shorter backbone chains than -glucan; furthermore, the intracellular glycogen possesses a molecular mass estimated to be half of that of the former, resulting in two different working structural models (Figure 5).
|
Further conformational analyses of the
-glucan and glycogenBased on the proposed models (Figure 5), we reasoned that a difference between the mycobacterial polysaccharides should exist in terms of access to the branched chains by enzymes. Accordingly, we challenged the models by treating both the
-glucan and intracellular glycogen purified from M. bovis BCG with pullulanase for different time points and monitored the evolution of the enzymatic degradation by following the disappearance of the signal assigned to the resonances of anomeric protons of branched residues at 5.00 ppm (relative to that corresponding to the resonances of the anomeric protons of both terminal and linear units at 5.38 ppm) in the 1H-NMR spectra of the polysaccharides. While a complete digestion of the branched chains of the intracellular M. bovis BCG glycogen by pullulanase was observed after only 1 day, 3 days of enzymatic hydrolysis were required to fully degrade the -16 linkages of the -glucan. The commercial glycogen from bovine source exhibited an intermediate behavior. Consistently, when the sizes of the resulting presumably degraded molecules were measured by DLS, we observed a progressive diminution of the radius of the degradative products. Taking into account the observed similar branching ratios of M. bovis BCG -glucan and glycogen, the faster rate of enzymatic degradation of M. bovis BCG glycogen is in agreement with the suggested expanded glycogen structure compared to the compact structure for the -glucan (Figure 5). |
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingConflict of interest statementReferences |
|---|
The well-established occurrence of a glycogen-like
-glucan in the outermost compartment of the mycobacterial (Lemassu and Daffe 1994; Ortalo-Magne et al. 1995; Dinadayala et al. 2004) raised several important questions. Among these is the structural relationship between the extracellular glucan and intracellular polysaccharide glycogen. To address this question, the present study comparatively investigated the structural features of the two polysaccharides isolated from the same batches of M. bovis, keeping in mind that the structure of glycogen varies according to the bacterial growth phase. We first confirmed that the -glucan of this mycobacterial species exhibited a glycogen-like structure based on the 4--D-Glc-1 structure, with branching at some positions 6 (Dinadayala et al. 2004). Furthermore, the lengths of the branched chains were found very similar to those of the glycogen isolated from the same batches. These data strongly suggest that the two polysaccharides share a common biosynthetic pathway, not only in M. bovis but also in other mycobacteria examined so far in which similar structural features were found for their -glucans (Lemassu and Daffe 1994; Ortalo-Magne et al. 1995; Lemassu et al. 1996). This hypothesis is currently investigated by generating -glucan- and/or glycogen-less mutants.
Although very similar to glycogen in terms of primary structure, analytical studies, which include DLS, NMR, polarimetry, and ultracentrifugation, revealed that the M. bovis -glucan exhibited a higher molecular mass and a smaller radius than glycogen. These data fit well with a condensed structure for the M. bovis -glucan and the expanded structure for the glycogen (Figure 5), which would explain the observed poor accessibility of the -glucan to the degradative enzymes, and consistent with their different cell localization. These conformational differences would imply that the two polysaccharides possess specific routes for their assembly. To reach its final surface location from its synthesis compartment, virtually the cytosol, the -glucan or its precursor has to cross the complex cell envelope that includes the unusual thick peptidoglycan–arabinogalactan–mycolic acid complex. This process would induce many physical and biophysical constraints that may lead to the condensed structure proposed for the -glucan (Figure 5).
Another intriguing question raised by the surface exposure of the glycogen-like -glucan, a feature so far restricted to the best of our knowledge to mycobacteria and related bacterial genera, e.g., corynebacteria (Puech et al. 2001), is the physiological function of the polysaccharide. Indeed, it is tempting to postulate its involvement in the mycobacterial pathogenicity, primarily because of its surface location and its abundance in pathogenic mycobacterial species (Lemassu and Daffé 1994; Ortalo-Magné et al. 1995; Lemassu et al. 1996). In this respect, it was shown that the -glucan mediates the interaction between the host and the pathogen via the complement receptor type 3 (Cywes et al. 1997) and, as such, would be one of the surface compounds that play a key role during the first stages of the infection. Of possible importance is the fact that resembling -glucans have been described in the root of Angelica sinensis (Cao et al. 2006) and in Pseudallescheria boydii (Bittencourt et al. 2006), the former polysaccharide being involved in fungal phagocytosis. This substantiates the idea that such polysaccharides readily interact with the immune system. In that connexion, an antiphagocytic activity has been attributed to the glycan-rich capsule of M. tuberculosis (Stokes et al. 2004). More recently, the mycobacterial -glucan was shown to induce monocytes to differentiate into altered dendritic cells that fail to express CD1 and to upregulate CD80 and that produce IL10, but not IL-12 (Gagliardi et al. 2007). Altogether, these data point to the idea that the mycobacterial extracellular -glucan contributes to divert the immune host response. More generally, the amorphous and weakly soluble major capsular polysaccharide that embeds pathogenic mycobacteria in the hostile environment of the macrophage (Schwebach et al. 2002) may well contribute to protect the bacilli against both chemical and physical aggressions.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingConflict of interest statementReferences |
|---|
Strain and growth conditions
The Pasteur strain of M. bovis BCG was grown on the synthetic Sauton's medium (Sauton 1912
) as surface pellicles at 37°C for 3 weeks. The culture media from the different batches were collected separately for the purification of the -glucans whereas the corresponding cells pellicles were harvested for the extraction of the intracellular glycogens.
Purification of polysaccharides
The -glucan was purified from the culture medium as previously described (Lemassu and Daffe 1994). Briefly, the culture fluid was filtered through a 0.22-µm Nalgen filter in order to remove bacteria, concentrated 10-fold and macromolecules were precipitated by 6 volumes of cold ethanol overnight at 4°C. The precipitate was collected after centrifugation at 14,000 x g for 1 h, dissolved in distilled water, dialyzed for 1 day against water to eliminate traces of salts and glycerol, lyophilized, and weighed. The -glucan content in the mixture of extracellular macromolecules was estimated by sugar compositional analysis by GC. Macromolecules were dissolved in NH4Cl 0.01 M at pH: 8.35 and loaded on a DEAE-trisacryl column (27 cm x 1 cm). Neutral polysaccharides were eluted from the column by the buffer alone whereas proteins were retained. The -glucan was separated from other neutral polysaccharides by gel permeation chromatography on Bio-Gel P-200 (100–200 mesh, BioRad, 80 cm x 1 cm), eluted with a 0.5% acetic acid solution.
Cells were sterilized at 130°C for 2 h, suspended in water, and centrifuged twice at 1100 x g for 15 min to remove the potentially remaining loosely attached -glucan. The pellet was homogenized in distilled water and cells were broken in a French press (140 bars) to liberate the cytosol content. The supernatant was centrifuged first at 1100 x g for 30 min, to remove unbroken cells. The resulting supernatant was re-centrifuged at 27,000 x g for 15 min, to remove cell walls and cell envelopes. The intracellular glycogen was extracted by adapting the procedure of Bueding et al. (Bueding and Orrell 1964). Briefly, 1 volume of 27,000 supernatant was mixed with 3 volumes of glycine buffer (pH 10.5, 0.2 M) and two volumes of chloroform. This mixture was decanted overnight at 4°C. The upper phase was removed and the lower phase was re-extracted three times with 2 volumes of the glycine buffer. The aqueous phases were pooled, concentrated, and ultracentrifuged at 100,000 x g for 4 h at 4°C. The gelatinous residue was homogenized in 15 mL of water, shaken for 15 min with 5 mL of a chloroform/1-octanol 3/1, v/v, and decanted. The upper glycogenic phase was aspired and the treatment with organic solvent mixture was repeated for longer periods (30 min and then 150 min) until the disappearance of the proteineous interphase. The aqueous phase was then precipitated by 1 volume of cold ethanol overnight at 4°C, centrifuged for 1 h at 1500 g and lyophilized.
Determination of the sugar and glycosyl-linkage compositions
To determine the glycosidic composition, 20 µg of erythritol was added to 100 µg of polysaccharides, which were hydrolyzed with 200 µL of 2 M trifluoroacetic acid for 1 h at 110°C. The hydrolysate was dried under nitrogen, and the products were trimethylsilyled according to Sweeley et al. (1963), solubilized in petroleum ether and analyzed by GC using authentic standards for comparison. The sugar derivatives were analyzed on a Hewlet–Packard 4890 gas chromatograph, equipped with a nonpolar OV1 (12 m x 0.22 mm) capillary column and using helium as the gas vector. The temperature program consisted of a temperature increase from 60°C to 100°C at a rate of 20°C/min, followed by an increase from 100°C to 290°C at 5°C/min.
The glycosyl-linkage composition of polysaccharides was determined by analyzing the partially O-methylated, partially O-acetylated alditol acetates by GC-MS. For this purpose, polysaccharides were per-O-methylated according to Blakeney and Stones (1985) and hydrolyzed with 2 M trifluoroacetic acid for 2 h at 110°C. The resulting products were reduced with 200 µL of NaBD4 (10 mg in 250 µL NaOH 5 N in 1 mL absolute ethanol) for 1 h at room temperature. The reaction was stopped by adding three drops of glacial acetic acid and the mixture was dried. The resulting products were per-O-acetylated with a pyridine/acetic anhydride (1/1; v/v) for 20 min at 110°C, solubilized in petroleum ether and analyzed by GC-mass spectrometry (GC-MS).
Enzymatic degradation
Twenty microliters of pullulanase (4.2 mg/mL, EC 3.2.1.4
[EC]
1, Sigma) were added to 2 mg of polysaccharides solubilized in 1 mL of sodium acetate buffer (0.1 M, pH 5.6), and incubated at 40°C for 1, 2, and 3 days. After inactivation of the enzyme by heating at 100°C for 10 min, the products were dried. Analysis of oligosaccharides constituting short chains (<30 sugar residues) were done after per-O-acetylated as described above. An equal volume (2 mL) of dichloromethane and water was added to the mixture acetylated compounds, and then the organic phase was washed twice by water, dried, and analyzed by MALDI-TOF mass spectrometry or by thin-layer chromatography to analyze monosaccharidic content. Analysis of the long oligosaccharide fragments (>30 sugar residues) constituting the backbone of the polysaccharides was achieved by DLS as described above, upon the completion of the enzymatic degradation of branched chains as ascertained by 1H-NMR.
Mass spectrometry analyses
The MALDI-TOF mass spectra were acquired on a Voyager-DE-STR mass spectrometer (PerSeptive Biosystems, Framingham, MA) equipped with a pulsed nitrogen laser emitting at 337 nm, as previously described (Laval et al. 2001; Dinadayala et al. 2002). Pullulan and maltoheptaose were used as positive and negative controls, respectively.
GC-MS analyses were performed on an HP 5889X mass spectrometer coupled with a HP 5890 series II gas chromatograph, equipped by a OV1 column (12 m x 0.3 mm), with a temperature program from 80°C to 290°C at 8°C/min. Mass spectrometry analyses were performed in the electron impact mode (70 eV).
Nuclear magnetic resonance (NMR) analysis
The 1H- and 13C-NMR analyses were performed on a Bruker AMX-500 spectrometer at 500.13 MHz and 125.77 MHz, respectively, using a 5-mm BBI probe, at 343 K in D2O. A basic pulse program was used for the 1H-NMR spectra, without presaturation of the HOD signal, in order to preserve the integration of the signal at 5.00 ppm. A pulse of 45° was applied before a 2 s recycle delay. FID was recorded on 16 k points, and processed by a sine-bell apodization function before Fourier transformation and integration. Bruker DEPT (distortionless enhancement by polarization transfer) sequence was applied for 13C-NMR analyses.
Determination of the molecular masses
The molecular masses were determined by two methods: (i) gel permeation on Bio-Gel P200 (100–200 mesh, Bio-Rad, 80 cm x 1 cm) irrigated with 0.5% acetic acid in 0.1 M NaCl at a flow rate of 6 mL/h. The elution was monitored by refractometry; (ii) ultracentrifugation performed using a Beckman OptimaTM XL-A analytical ultracentrifuge equipped with an analytical rotor An-55. All sedimentation velocity experiments were carried out at 1 mg/mL of water, run at 10,000 rpm during 5 h at 20°C, and analyzed by Origin 4.1 software.
Miscellaneous techniques
The polarimetric analysis was performed on a Perkin–Elmer spectropolarimeter model 241, at 
= 589 nm, in the distilled water solution (1 mg/mL) at 25°C.
The DLS analysis was performed on a DynaPro-MS/X (ProteinSolutions, Charlottesville, VA) using a 12 µL sample cell at 293 K. The experiments were conducted at 1 mg/mL of distilled water, with an acquisition time of 200 s, and analyzed with Dynamic V6 software. Nine to fifteen different batches were examined for mycobacterial polysaccharides.
|
Funding |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingConflict of interest statementReferences |
|---|
The work was supported by the Centre National de la Recherche Scientifique (CNRS), Université Paul Sabatier (Toulouse III), and the European Commission Contract n%B0QLK2-CT-1999-01093. The NMR equipment was funded via European structural funds, CNRS, and the Region Midi-Pyrénées funds as part of the 2000–2006 CPER program.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingConflict of interest statementReferences |
|---|
None declared.
|
Acknowledgements |
|---|
The authors thank Françoise Laval Vizcaïno (IPBS, Toulouse) for MALDI-TOF mass experiments and Valérie Guilhet and Catherine Birck (IPBS, Toulouse) for their technical help in the determination of physical parameters of the polysaccharides.
|
Footnotes |
|---|
2 Present address: Sanofi Pasteur, Département recherche, Plateforme Biochimie, 1541 avenue Marcel Merieux, 69280 Marcy l’Etoile, France.
3 These authors contributed equally to this work.
|
Abbreviations |
|---|
BCG, bacille calmette guérin; DLS, dynamic light scattering; f, furanosyl; GC, gas chromatography; MALDI-TOF, matrix-assisted laser desorption/ionization time-of-flight; MS, mass spectrometry; p, pyranosyl; S, Svedberg
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingConflict of interest statementReferences |
|---|
Antoine A, Tepper B. Characterization of glycogens from mycobacteria. Arch Biochem Biophys (1969) 134:207–213.[CrossRef][Web of Science][Medline]
Antoine A, Tepper B. Environmental control of glycogen and lipid content of Mycobacterium tuberculosis. J Bacteriol (1969) 100:538–539.
Bittencourt VC, Figueiredo RT, et al. An alpha-glucan of Pseudallescheria boydii is involved in fungal phagocytosis and Toll-like receptor activation. J Biol Chem (2006) 281:22614–22623.
Blakeney AB, Stone BA. Methylation of carbohydrates with lithium methylsulphinyl carbanion. Carbohydr Res (1985) 140:319–324.[CrossRef][Web of Science]
Brammer GL, Rougvie MA, et al. Distribution of alpha-amylase-resistant regions in the glycogen molecule. Carbohydr Res (1972) 24:343–354.[CrossRef][Web of Science][Medline]
Bueding E, Orrell S. A mild procedure for the isolation of polydisperse glycogen from animal tissues. J Biol Chem (1964) 239:4018–4020.
Cao W, Li XQ, et al. Structural analysis of water-soluble glucans from the root of Angelica sinensis (Oliv.) Diels. (2006) 341:1870–1877.
Cywes C, Hoppe HC, et al. Nonopsonic binding of Mycobacterium tuberculosis to complement receptor type 3 is mediated by capsular polysaccharides and is strain dependent [In process citation]. Infect Immun (1997) 65:4258–4266.
Daffe M, Draper P. The envelope layers of mycobacteria with reference to their pathogenicity. Adv Microb Physiol (1998) 39:131–203.[Web of Science][Medline]
Dinadayala P, Laval F, et al. Tracking the putative biosynthetic precursors of oxygenated mycolates of Mycobacterium tuberculosis: Structural analysis of fatty acids of a mutant strain devoid of methoxy- and keto-mycolates. J Biol Chem (2002) 6:6.
Dinadayala P, Lemassu A, et al. Revisiting the structure of the anti-neoplastic glucans of Mycobacterium bovis Bacille Calmette-Guerin. Structural analysis of the extracellular and boiling water extract-derived glucans of the vaccine substrains. J Biol Chem (2004) 279:12369–12378.
Gagliardi MC, Lemassu A, et al. Cell wall-associated alpha-glucan is instrumental for Mycobacterium tuberculosis to block CD1 molecule expression and disable the function of dendritic cell derived from infected monocyte. Cell Microbiol. 9(8):2081–92 (2007).
Harding SE. Challenges for the modern analytical ultracentrifuge analysis of polysaccharides. Carbohydr Res (2005) 340:811–826.[CrossRef][Web of Science][Medline]
Laval F, Laneelle MA, et al. Accurate molecular mass determination of mycolic acids by MALDI-TOF mass spectrometry. Anal Chem (2001) 73:4537–4544.[Medline]
Lemassu A, Daffe M. Structural features of the exocellular polysaccharides of Mycobacterium tuberculosis. Biochem J (1994) 297:351–357.[Web of Science][Medline]
Lemassu A, Ortalo-Magne A, et al. Extracellular and surface-exposed polysaccharides of non-tuberculous mycobacteria. Microbiology (1996) 142:1513–1520.
Nikaido H, Jarlier V. Permeability of the mycobacterial cell wall. Res Microbiol (1991) 142:437–443.[Medline]
Ortalo-Magne A, Dupont MA, et al. Molecular composition of the outermost capsular material of the tubercle bacillus. Microbiology (1995) 141:1609–1620.
Preiss J. Bacterial glycogen synthesis and its regulation. Annu Rev Microbiol (1984) 38:419–458.[CrossRef][Medline]
Puech V, Chami M, et al. Structure of the cell envelope of corynebacteria: Importance of the non-covalently bound lipids in the formation of the cell wall permeability barrier and fracture plane. Microbiology (2001) 147:1365–1382.
Rees DA. Shapely polysaccharides. The eighth Colworth medal lecture. Biochem J (1972) 126:257–273.[Web of Science][Medline]
Sauton B. Sur la nutrition minérale du bacille tuberculeux. C R Acad Sci, Ser III, Sci Vie (1912) 155:860–863.
Schwebach JR, Glatman-Freedman A, et al. Glucan is a component of the Mycobacterium tuberculosis surface that is expressed in vitro and in vivo. Infect Immun (2002) 70:2566–2575.
Stokes RW, Norris-Jones R, et al. The glycan-rich outer layer of the cell wall of Mycobacterium tuberculosis acts as an antiphagocytic capsule limiting the association of the bacterium with macrophages. Infect Immun (2004) 72:5676–5686.
Sweeley CC, Bentley R, et al. Gas-liquid chromatography of trimethylsilyl derivatives of sugars and related substances. J Am Chem Soc (1963) 85:2497–2507.[CrossRef][Web of Science]
Zang LH, Howseman AM, et al. Assignment of the 1H chemical shifts of glycogen. Carbohydr Res (1991) 220:1–9.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Geurtsen, S. Chedammi, J. Mesters, M. Cot, N. N. Driessen, T. Sambou, R. Kakutani, R. Ummels, J. Maaskant, H. Takata, et al. Identification of Mycobacterial {alpha}-Glucan As a Novel Ligand for DC-SIGN: Involvement of Mycobacterial Capsular Polysaccharides in Host Immune Modulation J. Immunol., October 15, 2009; 183(8): 5221 - 5231. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||