Department of Microbiology, Hellenic Pasteur Institute, Vassilissis Sofias 127, Athens 11521, Greece1
Author for correspondence: Evangelia Vretou. Tel: +30 1 64 78 873. Fax: +30 1 64 78 873. e-mail: vretou{at}mail.pasteur.gr
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Keywords: polymorphic outer-membrane protein family, bacterial glycoprotein, concanavalin A, two-dimensional electrophoresis
Abbreviations: ConA, concanavalin A; EB, elementary body; MOMP, major outer-membrane protein; OG, n-octyl ß-D-glucopyranoside; OMC, outer-membrane complex; PMP, polymorphic membrane protein; PNGase F, N-endoglycosidase F from Flavobacterium meningosepticum; POMP, polymorphic outer-membrane protein
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Metabolic labelling of chlamydiae.
Labelling with Tran35S-Label was based on the procedure described for Chlamydia trachomatis by Goswami et al. (1990) . Briefly, confluent McCoy cell cultures were infected with strain ATCC VR 656T in the presence of cycloheximide (1 µg ml-1). At 818 h post-infection, the medium was replaced with labelling medium (MEM with Earles salts; ICN) supplemented with 5% (v/v) of the normal amount of unlabelled methionine and cysteine, and 1 µg emetine ml-1. Tran35S-Label (35S Escherichia coli hydrolysate labelling reagent, containing 70% L-methionine, 1175 Ci mmol-1, 43·5 TBq mmol-1; ICN) was added at a concentration of 500 µCi (18·5 MBq) per 175 cm2 culture flask, and the pH was adjusted with 1 M HEPES. Labelled EBs were isolated 72 h post-infection as usual. Sham-infected McCoy cells were labelled in parallel and used as a control.
Solubilization of outer-membrane complexes (OMCs).
Purified EBs (1·5 mg ml-1) were incubated for 1 h at 4 °C in PBS (0·140 M NaCl, 0·026 M KCl, 0·0014 M KH2PO4, 0·008 Na2HPO4) containing 1% (v/v) N-lauroylsarcosine and then pelleted at 45000 g for 20 min. The procedure was repeated once more and the combined EB pellets were extracted with 2% n-octyl ß-D-glucopyranoside (OG; Sigma) containing 10 mM DTT, for 1 h at 4 °C. The solubilized material was probed on Western blots following gel electrophoresis.
Two-dimensional-PAGE and immunoblotting.
Two-dimensional gel electrophoresis was performed essentially as described by Giannikopoulou et al. (1997) with a modification to the lysis buffer. EBs were lysed by heating (5 min at 95 °C) in a 4% CHAPS, 70 mM DTT and 40 mM Tris-base. Urea was added to a final concentration of 8 M and the sample was left overnight before loading onto a non-linear immobilized pH gradient (pH 310; Pharmacia) and run at 110000 V h, for 24 h. The second-dimension run, silver staining, immunoblotting and detection with 3,3'-diaminobenzidine or chemiluminescence (Super Signal; Pierce) were performed according to published procedures (Giannikopoulou et al., 1997
). Used nitrocellulose membranes were recycled by stripping off the bound ligands in 0·063 M Tris/HCl (pH 6·7), 100 mM ß-mercaptoethanol and 2% SDS for 30 min at 70 °C. Proteins captured on Western blots were visualized with Indian ink (black Pelican Fount Indian drawing ink) before or after immunoenzyme detection.
mAbs.
The anti-POMP mAbs 181 and 8976/073, and mAb 4H9 against the POMP-like protein at 105 kDa, have been described previously, as has mAb 188 against the major outer-membrane protein (MOMP) (Vretou et al., 1996 ; Giannikopoulou et al., 1997
). mAb ABF8, an IgG3, which recognizes EF-Tu (heat-labile elongation factor) in Western blots after two-dimensional electrophoresis, was produced in this laboratory. mAbs EB3G2 and JA6C7, both reacting with the antigen family at 90 kDa, were kindly provided by Dr A. Rodolakis (INRA, Nouzilly, France), and have been described by Souriau et al. (1994)
. The anti-POMP mAbs were selected from a larger panel on the basis of their variant binding pattern to Western blots of whole EBs.
Lectin-binding assay.
Concanavalin A (ConA)-binding assays were performed on nitrocellulose-immobilized proteins after separation by SDS-PAGE or by two-dimensional electrophoresis. The procedure was based on the lectin-binding assays of EBs immobilized on microtitre plates (Goswami et al., 1991 ). After blocking with 0·2% Tween 20 in PBS for 1 h at 37 °C and washing with PBS, biotinylated concanavalin A [Sigma; 1 µg ConA per ml PBS containing 0·025% Tween 20 (PBS-T), 1 mM CaCl2 and 1 mM MgCl2] was added. The haptenic sugar methyl
-D-mannoside (FLUKA) was included as a control during the incubation with the lectin for 1 h at 37 °C. Membranes were washed three times with PBS-T and the bound lectin was detected with streptavidin-conjugated horseradish peroxidase and Super Signal (Pierce).
Treatment with N-endoglycosidase F (PNGase F).
Chlamydial proteins obtained by OG/DTT extraction were treated with the amidase PNGase F (from Flavobacterium meningosepticum; 500000 U ml-1; New England Biolabs), which cleaves between the innermost glucosamine and the asparagine residue on the peptide backbone. The chlamydial proteins were incubated for 1 h at 37 °C either in solution (with 1500 U PNGase F ml-1), or in an immobilized form after transfer onto nitrocellulose paper (with 500 U PNGase F ml-1 in 0·05 M sodium phosphate, pH 7·5). To examine the orientation of the oligosaccharides, intact bacteria (1·5 mg intact bacteria per ml PBS containing 200 mM sucrose and 5 mM MgCl2) were incubated for 2 h at 37 °C with 7500, 15000 or 30000 U PNGase F ml-1. The EBs were pelleted, washed twice with PBS and analysed by SDS-PAGE and Western blotting.
Treatment of EBs with proteases.
To investigate the surface exposure of proteins, purified EBs were incubated for 30 min at 37 °C with increasing amounts of trypsin or proteinase K (both from Sigma) in PBS supplemented with 200 mM sucrose and 5 mM MgCl2.. The enzyme digestion was stopped with trypsin-inhibitor in the case of trypsin, and with 5 mM PMSF in the case of proteinase K. The EBs were pelleted and analysed by SDS-PAGE and Western blotting. For the trypsin control, trypsin was incubated in the presence of trypsin-inhibitor.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
Treatment with PNGase F
To ascertain whether the results observed in the two-dimensional analysis demonstrated that the POMPs actually did contain glycan moieties, we studied the specificity of the lectin binding as well as the response to endoglycosidase treatment of the various POMPs. To this end, OMCs were solubilized with OG/DTT, analysed by SDS-PAGE, transferred to nitrocellulose membranes and stained with Indian ink and mAbs. The MOMP and the 90 kDa triplet were the most prominent bands in the OG extract, as judged by the molecular masses in the blot stained with Indian ink (Fig. 3, lane 1) and the reactivity of specific mAbs 188 (anti-MOMP) and JA6C7 (anti-POMP 91B; Fig. 3
, lane 2) against the OG extracts. Incubation with biotinylated ConA was performed in the absence (lane 3) and in the presence of the haptenic sugar methyl
-D-mannoside (lane 5). The MOMP, and the POMPs at 85, 90 and 105 kDa (lane 3) bound the lectin in a specific manner, since binding was inhibited by the presence of methyl
-D-mannoside (lane 5). ConA reacts widely with mannose and specifically with biantennary high-mannose, hybrid and complex oligosaccharides from N-linked oligosaccharides. Treatment of the proteins in the nitrocellulose membrane with the peptide-glycanase PNGase F, an enzyme that cleaves only N-linked oligosaccharides, abolished binding of the lectin to the POMPs and decreased its reactivity with the MOMP (lane 4). Similar results were obtained when endoglycosidase treatment of the solubilized OMCs was performed in solution, before running the gel (lane 6). The membranes treated with ConA (lanes 3, 4 and 6) were probed with mAb JA6C7 immediately after the lectin-binding assay without the removal of the bound ConA (lanes 79). After stripping off mAb JA6C7, the membranes were subsequently reincubated with mAb 8976/073 (lanes 10 and 11, corresponding to membranes 3 and 6). The binding of both mAbs was unaffected by the treatment with the endoglycosidase whether it was in solution or on the membrane (compare lanes 7, 8 and 9, and lane 10 with 11), suggesting that their respective epitopes lie outside the oligosaccharide chains. Removal of the lectin ligand by the endoglycosidase treatment did not alter the migration of the 90 kDa peptide (lanes 7 and 9, and 10 and 11), suggesting that the carbohydrate chains might be small. In summary, these data indicated that the POMPs at 85, 90 and 105 kDa bear an asparagine-linked oligosaccharide.
|
Treatment with proteolytic enzymes
The strong lectin-binding ability of POMP 91B and the POMP-related protein at 105 kDa prompted us to examine the surface exposure of the polypeptides and the attached glycan moieties. To this end, whole purified EBs suspended in PBS containing sucrose and Mg2+ were incubated at 37 °C with trypsin, pelleted, analysed by SDS-PAGE, transferred to a nitrocellulose membrane and probed with mAb 4H9, which was specific for the 105 kDa protein, and the panel of anti-POMP mAbs shown in Fig. 1. Exposure of purified EBs to trypsin at 10:1 (w/w, i.e. 100 µg intact EBs ml-1 to 10 µg trypsin ml-1) for 30 min resulted in the cleavage of the 105 kDa protein and the generation of an intermediate fragment at 66 kDa and a stable fragment at 32 kDa (both marked by asterisks in Fig. 4a
). To check that trypsin did not penetrate the chlamydial cell envelope, even in the presence of excess Mg2+ ions, which are required for maximal lipopolysaccharide cross-linking (Vretou et al., 1992
), we probed the electro-transferred digest with mAb ABF8, which reacts with EF-Tu. As shown in Fig. 4(b)
, treatment of EBs with increasing concentrations of trypsin (EB:trypsin ratio 20:1 and 10:1, w/w, respectively) resulted in no decrease in the binding pattern displayed by anti-EF-Tu, suggesting that this cytoplasmic protein remained intact after tryptic digestion. Similar results were obtained when the digests were probed with an anti-DnaK mAb (data not shown).
|
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
We have shown in this study that the 85 kDa and 105 kDa POMP-like proteins and the POMP triplet at 90 kDa have short asparagine-linked mannose-containing carbohydrate chains, and have further addressed the issue of the exposure of their polypeptide chains and the orientation of the attached oligosaccharides. All five proteins were released from the OMC by OG and reducing agents (Fig. 3). The results from the protease digestion experiments suggested that the exposure of the POMP at 105 kDa and at 90 kDa was very different (Fig. 4a
, c
). The former is particularly sensitive to trypsin, which cleaves the molecule down to a 32 kDa fragment possibly anchored in the outer membrane (Fig. 4a
). The POMPs at 90 kDa were resistant to trypsin with the exception of POMP 91B and the POMP-like protein at 85 kDa, which were cleaved with a large excess of trypsin (Fig. 4c
). Tanzer et al. (2001)
reported recently that two of the six identified POMPs in Chlamydia psittaci strain 6BC were sensitive to trypsin; however, no relative differences between the POMPs regarding their sensitivity to the protease were noted. Previous experiments using electron microscopy have shown that the epitope of mAb 181 is accessible to the antibody on the surface of EBs. These data have led to the suggestion that at least one of the three POMPs at 90 kDa reacting with this mAb is surface exposed (Longbottom et al., 1998b
). We examined the orientation of the oligosaccharides attached to the POMPs by treating EBs with glycosidase under non-denaturing conditions. The data showed that the ConA ligand in the large POMP-related molecule was not accessible to the enzyme, which was in contrast to the ConA ligands in the POMPs at 90 kDa, which were partially cleaved. These results are consistent with the concept that some of the oligosaccharides in the 90 kDa proteins faced outwards, possibly protecting the polypeptides from proteolytic enzymes, whereas the oligosaccharides in the large POMP were oriented inwards, thus rendering the polypeptide chain accessible to proteases.
The addition of carbohydrates to proteins is the most common post-translational modification in the eukaryotic cell. Oligosaccharide linkage occurs either at the asparagine residue of the consensus tripeptides N-X-S/T, or at a serine or threonine residue. In prokaryotes, glycosylation has been considered uncommon for a long time, being restricted to proteins in archaeal and eubacterial S-layers (Lechner & Wieland, 1989 ; Messner & Sleytr, 1991
). The best-studied prokaryotic glycoproteins remain the glycoproteins of the S-layer. However, an increasing series of surface- or membrane-associated bacterial proteins, bearing unambiguously demonstrated glycosidic linkages, have been reported recently (Schäffer et al., 2001
). Besides the oligosaccharide linkages at asparagine and serine or threonine residues, similar to those encountered in the eukaryotic cell, protein glycosylation in prokaryotes may also occur at tyrosine residues and shows in general greater diversity in glycan composition (Schäffer et al., 2001
). N-Glycosidic linkages are rare in prokaryotes; when we scanned the sequences of POMPs 90, 91A and 91B for potential glycosylation sites (program Scanprosite) 17, 16 and 19 putative N-linked glycosylation sites were detected, respectively. This is a relatively high number compared to the MOMP, where only three potential N-glycosylation sites were found. Nine of these consensus tripeptides (ten for POMP 91B) were located in the N-terminus of the amino acid sequences within a 200 amino acid domain (amino acids 27223 in POMP 91B). It is worth noting that immunoelectronic microscopy has demonstrated that this domain is accessible to antibodies utilized in this study on the chlamydial outer-membrane surface (Longbottom et al., 1998b
). The majority of the remaining sites reside in the second half of the C-terminal part of the molecules. It is interesting that many of the potentially glycosylated asparagine residues coincide with the asparagine in the signature motif FXXN, which is common to the POMPs and the extended PMP family in Chlamydia trachomatis and Chlamydophila pneumoniae. Six out of nine FXXN motifs in POMP90, four out of eight motifs in 91A, and five out of seven in POMP91B are theoretically potential glycosylation sites. Three of these potentially glycosylated FXXN motifs precede the three GGAI signatures in the POMPs. FXXN motifs which are also consensus sites for glycosylation are found in C. pneumoniae orthologues, i.e. two sites in PMP 9. However, the strong ConA binding of POMP 91B compared to POMP 90 and 91A as shown in Fig. 2(d)
, suggests that only a few of the consensus glycosylation sites may bear an asparagine-linked carbohydrate chain. The sensitivity of the lectin binding to treatment with the asparaginyl-N-acetylglucosamine amidase, as shown in Fig. 3
, provides additional support for the existence of covalently N-linked glycosylation sites in the POMPs, since lectin binding alone might be caused by firmly associated non-covalently linked saccharides. Further chemical characterization, such as protease digestion and peptide analysis by mass-spectrometry, is needed to allow the exact identification of the glycosylation sites and the carbohydrate constituents in the POMPs.
Oligosaccharide chains are often involved in cell recognition and regulatory processes because of their diversity and specificity. In Chlamydia spp. carbohydrates have been implicated early on in the interaction of the bacterium with the host cell. Stimulation by the lectin wheatgerm agglutinin (WGA) on the attachment of HeLa, but not McCoy, cell-grown C. trachomatis has been reported, suggesting that enzymes from the host cell may contribute to the glycosylation of chlamydial proteins (Bose et al., 1983 ). WGA-binding proteins were demonstrated in unheated, OG-extracted preparations of the same species (Goswami et al., 1991
). The lectin-binding ligand was isolated as the glycan moiety of the MOMP that interfered with the attachment process (Swanson & Kuo, 1994
). It is conceivable that the surface-exposed glycan moieties in the POMPs at 90 kDa (Fig. 5b
) play a role in the attachment and entry process of Chlamydophila abortus to the host cell. However, the short carbohydrate chains suggested from this study do not support a role in the host-attachment process through the oligosaccharide chains.
The genomic organization of the POMPs in the N- and C-terminal domains (Longbottom et al., 1998a ), the different surface exposure of these domains on the EB surface (Longbottom et al., 1998b
), as well as the sequence similarity of the PMPs with the RompA protein of Rickettsia spp. and the filamentous haemagglutinin of Bordetella pertussis (Grimwood & Stephens, 1999
), have brought the POMPs into association with the class IV secretion molecules, the autotransporters. Christiansen et al. (2000)
recently reported the structural similarity between the C-terminal part of the PMPs and the transmembrane ß-barrel in autotransporters. Moreover, they showed that antibodies against non-denatured OMCs of C. pneumoniae recognized on the surface of E. coli a stably expressed construct which consisted of the N-terminal part of Omp4 and the ß-barrel of the E. coli autotransporter AIDA. Such transport mechanisms require extensive folding and processing, particularly when the molecules are rich in cysteine as is the case with the POMPs. In the eukaryotic cell, N-linked oligosaccharides affect the folding process of polypeptide chains. A great majority of N-linked glycans occur in locations of compact ß-turns, and oligosaccharides may orient peptide segments toward the surface of protein domains, mimicking chaperones (Helenius & Aebi, 2001
). On this basis, we speculate that a possible role for the N-linked oligosaccharides in the POMPs might be the promotion of the proper folding and processing of these proteins as parts of a sophisticated transport system. Alternatively, short oligosaccharide chains on membrane-anchored proteins, when cross-linked, could provide the EB with additional stability and act as analogues of peptidoglycan.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Cevenini, R., Moroni, A., Sambri, V., Perini, S. & La Placa, M. (1989). Serological response to chlamydial infection in sheep, studied by enzyme-linked immunosorbent assay and immunoblotting. FEMS Microbiol Immunol 47, 459-464.
Christiansen, G., Pedersen, A. S., Hjerno, K., Vandahl, B. & Birkelund, S. (2000). Potential relevance of Chlamydia pneumoniae surface proteins to an effective vaccine. J Infect Dis 181, S528-S537.[Medline]
Everett, K. D. E., Bush, R. M. & Andersen, A. A. (1999). Emended description of the order Chlamydiales, proposal of Parachlamydiaceae fam. nov. and Simkaniaceae fam. nov., each containing one monotypic genus, revised taxonomy of the family Chlamydiaceae, including a new genus and five new species, and standards for the identification of organisms. Int J Syst Bacteriol 49, 415-440.[Abstract]
Giannikopoulou, P., Bini, L., Simitsek, P. D., Pallini, V. & Vretou, E. (1997). Two-dimensional electrophoretic analysis of the protein family at 90 kDa of abortifacient Chlamydia psittaci. Electrophoresis 18, 2104-2108.[Medline]
Goswami, P. C., Vretou, E. & Bose, S. K. (1990). Extensive heterogeneity of the protein composition of Chlamydia trachomatis following serial passage in two different cell lines. J Gen Microbiol 136, 1623-1629.[Medline]
Goswami, P. C., Vretou, E. & Bose, S. K. (1991). Host passage-dependent wheat germ agglutinin-binding proteins of Chlamydia trachomatis. FEMS Microbiol Lett 81, 53-56.
Grimwood, J. & Stephens, R. S. (1999). Computational analysis of the polymorphic membrane protein superfamily of Chlamydia trachomatis and Chlamydia pneumoniae. Microb Comp Genomics 4, 187-201.[Medline]
Helenius, A. & Aebi, M. (2001). Intracellular functions of N-linked glycans. Science 291, 2364-2369.
Kalman, S., Mitchell, W., Marathe, R. & 7 other authors (1999). Comparative genomes of Chlamydia pneumoniae and C. trachomatis. Nat Genet 21, 385389.[Medline]
Lechner, J. & Wieland, F. (1989). Structure and biosynthesis of prokaryotic glycoproteins. Annu Rev Biochem 58, 173-194.[Medline]
Longbottom, D., Russell, M., Jones, G. E., Lainson, F. A. & Herring, A. J. (1996). Identification of a multigene family encoding the 90 kDa proteins of the ovine abortion subtype of Chlamydia psittaci. FEMS Microbiol Lett 142, 277-281.[Medline]
Longbottom, D., Russell, M., Dunbar, S. M., Jones, G. E. & Herring, A. J. (1998a). Molecular cloning and characterization of the genes encoding the highly immunogenic cluster at 90-kilodalton envelope proteins from the Chlamydia psittaci subtype that causes abortion in sheep. Infect Immun 66, 1317-1324.
Longbottom, D., Findlay, J., Vretou, E. & Dunbar, S. M. (1998b). Immunoelectron microscopic localisation of the OMP90 family on the outer membrane surface of Chlamydia psittaci. FEMS Microbiol Lett 164, 111-117.[Medline]
Messner, P. & Sleytr, U. B. (1991). Bacterial surface layer glycoproteins. Glycobiology 1, 545-551.[Abstract]
Schäffer, C., Graninger, M. & Messner, P. (2001). Prokaryotic glycosylation. Proteomics 1, 248-251.[Medline]
Souriau, A., Salinas, J., De Sa, C., Layashi, K. & Rodolakis, A. (1994). Identification of subspecies- and serotype 1-specific epitopes on the 80- to 90-kilodalton protein region of Chlamydia psittaci that may be useful for diagnosis of chlamydial induced abortion. Am J Vet Res 55, 510-514.[Medline]
Stephens, R. S. (editor) (1999). Chlamydia: Intracellular Biology, Pathogenesis, and Immunity. Washington, DC: American Society for Microbiology.
Stephens, R. S., Kalman, S., Lammel, C. & 9 other authors (1998). Genome sequence of an obligate intracellular pathogen of humans: Chlamydia trachomatis. Science 282, 754759.
Storz, J. (1971). Chlamydia and Chlamydia-induced Diseases. Springfield, IL: Charles C. Thomas.
Swanson, A. F. & Kuo, C.-C. (1994). Binding of the glycan moiety of the major outer-membrane protein of Chlamydia trachomatis to HeLa cells. Infect Immun 62, 24-28.[Abstract]
Tan, T. W., Herring, A. J., Anderson, I. E. & Jones, G. E. (1990). Protection of sheep against Chlamydia psittaci infection with a subcellular vaccine containing the major outer membrane protein. Infect Immun 58, 3101-3108.[Medline]
Tanzer, R. J., Longbottom, D. & Hatch, T. P. (2001). Identification of polymorphic outer membrane proteins of Chlamydia psittaci 6BC. Infect Immun 69, 2428-2434.
Vretou, E., Psarrou, E. & Spiliopoulou, D. (1992). The role of lipopolysaccharide in the exposure of protective antigenic sites on the major outer membrane protein of Chlamydia trachomatis. J Gen Microbiol 138, 1221-1227.[Medline]
Vretou, E., Loutrari, H., Mariani, L. & 7 other authors (1996). Diversity among abortion strains of Chlamydia psittaci demonstrated by inclusion morphology, polypeptide profiles and monoclonal antibodies. Vet Microbiol 51, 275289.[Medline]
Received 11 April 2001;
revised 14 August 2001;
accepted 20 August 2001.