Glycobiology Advance Access originally published online on October 23, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Glycobiology vol 14 no 2 pp. 187-196, 2004
© Oxford University Press 2004; all rights reserved.
Carbohydrate recognition by enterohemorrhagic Escherichia coli: characterization of a novel glycosphingolipid from cat small intestine
2 Institute of Medical Biochemistry, Göteborg University, P.O. Box 440, SE 405 30 Göteborg, Sweden; and 3 Department of Medical Microbiology, Dermatology and Infection, Lund University, SE 223 62 Lund, Sweden
Received on September 2, 2003; revised on October 1, 2003; accepted on October 2, 2003
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
A key virulence trait of pathogenic bacteria is the ability to bind to receptors on mucosal cells. Here the potential glycosphingolipid receptors of enterohemorrhagic Escherichia coli were examined by binding of 35S-labeled bacteria to glycosphingolipids on thin-layer chromatograms. Thereby a selective interaction with two nonacid glycosphingolipids of cat small intestinal epithelium was found. The binding-active glycosphingolipids were isolated and, on the basis of mass spectrometry, proton NMR spectroscopy, and degradation studies, identified as Gal
3Galß4Glcß1Cer (isoglobotriaosylceramide) and Gal3Gal3Galß4Glcß1Cer. The latter glycosphingolipid has not been described before. The interaction was not based on terminal Gal3 because the bacteria did not recognize the structurally related glycosphingolipids Gal3Gal4Galß4Glcß1Cer and Gal3Galß4GlcNAcß3Galß4Glcß1Cer (B5 glycosphingolipid). However, further binding assays using reference glycosphingolipids showed that the enterohemorrhagic E. coli also bound to lactosylceramide with phytosphingosine and/or hydroxy fatty acids, suggesting that the minimal structural element recognized is a correctly presented lactosyl unit. Further binding of neolactotetraosylceramide, lactotetraosylceramide, the Lea-5 glycosphingolipid, as well as a weak binding to gangliotriaosylceramide and gangliotetraosylceramide, was found in analogy with binding patterns that previously have been described for other bacteria classified as lactosylceramide-binding. Key words: carbohydrate binding / glycosphingolipids / enterohemorrhagic E. coli / mass spectrometry / microbial adhesion
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Infection with enterohemorrhagic Escherichia coli (EHEC) has been responsible for several large outbreaks of nonhemorrhagic colitis, hemorrhagic colitis, and hemolytic uremic syndrome in recent years (reviewed in Kaper, 1998
; Nataro and Kaper, 1998). One major virulence factor of EHEC is the expression one or more toxins of the Shiga toxin (Stx) family. The Stxs are hetero-oligomeric proteins comprised of a single A subunit and five B subunits (Ling et al., 1998). The B subunit pentamer binds to cell surface receptors, and the receptors for most Stxs are glycosphingolipids with a terminal Gal4Gal moiety, that is, galabiosylceramide (Gal4Galß1Cer), globotriaosylceramide (Gal4Galß4Glcß1Cer), and the P1 pentaglycosylceramide (Gal4Galß4GlcNAcß3Galß4Glcß1Cer) (Lindberg et al., 1987; Ling et al., 1998; Lingwood et al., 1987) (see Table I for glycosphingolipid nomenclature).
|
Although a substantial amount of information about the receptors for Stxs are available, less is known about the receptor(s) for EHEC bacterial cells. Binding to specific receptors on the target cells allows microorganisms to colonize and cause infection and leads to an efficient delivery of bacterial toxins (Beachey, 1981
). In the present study the potential carbohydrate recognition by EHEC bacterial cells was investigated by binding of EHEC bacteria to glycosphingolipids from various sources on thin-layer chromatograms. A binding-active glycosphingolipid not previously described was found in the nonacid glycosphingolipid fraction of cat intestinal epithelium. This EHEC-binding compound was isolated and characterized by mass spectrometry (MS), degradation studies, and protonnuclear magnetic resonance (NMR) as Gal3Gal3Galß4Glcß1Cer. In addition, as previously described for several other bacteria (Karlsson, 1989), EHEC interacted with lactosylceramide with phytosphingosine and/or hydroxy fatty acids, neolactotetraosylceramide, lactotetraosylceramide, the Lea-5 glycosphingolipid, and weakly with gangliotriaosylceramide and gangliotetraosylceramide. |
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Detection and isolation of EHEC-binding glycosphingolipids from cat intestinal epithelium
During the initial binding studies, mixtures of glycosphingolipids isolated from various species and organs were used to expose the bacteria to a large number of potentially binding-active carbohydrate structures. Thereby a selective binding of EHEC to compounds migrating in the tri- and tetraglycosylceramide region in the nonacid glycosphingolipid fraction from the epithelial cells of cat small intestine was detected (Figure 1, lane 7).
|
To isolate the EHEC-binding glycosphingolipids, the nonacid fraction of cat small intestinal epithelium was first separated by silicic acid chromatography, followed by high-performance liquid chromatography. The glycosphingolipid-containing fractions were tested for binding of EHEC using the chromatogram binding assay, and binding of P-fimbriated E. coli was tested in parallel to detect Gal
4Gal-containing compounds. The fractions were pooled according to the mobility of the compounds on thin-layer chromatograms and their binding activities. Thereby, 1.8 mg EHEC-binding triglycosylceramides (Fraction C1; Figure 2, lane 2) and 10.2 mg tetraglycosylceramides were obtained. However, initial analysis with negative ion fast atom bombardement (FAB) MS showed that the tetraglycosylceramide fraction contained two compounds (data not shown). This fraction was therefore acetylated and further separated by Iatrobeads column chromatography. An early-eluting compound was recognized by EHEC but not by P-fimbriated E. coli, and after pooling 1.5 mg of this compound was obtained (fraction C2; Figure 2, lane 3), whereas subsequent tubes contained a mixture of the compounds recognized by EHEC and P-fimbriated E. coli. Pooling the contents of these tubes yielded 2.4 mg (fraction C3; Figure 2, lane 4).
|
The EHEC-binding tri- and tetraglycosylceramides (fractions C1 and C2) were characterized by MS, proton NMR spectroscopy, gas chromatography–electron ionization (EI) MS after degradation, as follows.
EI MS of the EHEC-binding triglycosylceramide from cat intestinal epithelium (fraction C1)
The isolated triglycosylceramide (fraction C1) was permethylated, permethylated and LiAlH4-reduced, and analyzed with EI MS. Figure 3 shows the mass spectrum of the permethylated and reduced derivative, with a simplified formula for interpretation, representing the species with phytosphingosine and hydroxy 24:0 fatty acid.
|
A series of intense immonium ions were found at m/z 954–1066 in the spectrum of the permethylated and reduced glycosphingolipid. These ions contain the complete carbohydrate chain together with the fatty acid, and in the present case they gave evidence of a saccharide part composed of three hexoses, in combination with hydroxy 16:0–24:0 fatty acids. In the spectrum of the permethylated sample (not shown), immonium ions corresponding to a trihexosylceramide with hydroxy 22:0 and hydroxy 24:0 fatty acids were found at m/z 1052 and 1080, respectively.
The spectra of both derivatives had carbohydrate sequence ions at m/z 219 and 187 (terminal hexose) and m/z 423 (terminal hexose-hexose). The rearrangement fragment at m/z 700, seen in both spectra, containing the saccharide chain plus part of the fatty acid, confirmed a hexose-hexose-hexose sequence.
The spectrum of the permethylated and reduced glycosphingolipid (Figure 3) had molecular ions of trihexosylceramide with phytosphingosine and hydroxy 22:0 and 24:0 fatty acids at m/z 1323 and 1351, respectively.
The rearrangement ions at m/z 616 and 644 and at m/z 820 and 848 present in the spectrum of the permethylated and reduced derivative (explained below the formula in Figure 3) gave further evidence for the deduced carbohydrate sequence and fatty acid composition.
The conclusion from MS thus was a triglycosylceramide with a hexose-hexose-hexose sequence having phytosphingosine as long-chain base and mainly hydroxy 22:0 and 24:0 fatty acids.
EI MS of the EHEC-binding tetraglycosylceramide from cat intestinal epithelium (fraction C2)
EI mass spectra of the permethylated and the permethylated and LiAlH4-reduced derivatives of the isolated tetraglycosylceramide (fraction C2) are reproduced in Figure 4. Above the spectra are simplified formulae for interpretation, representing the species with phytosphingosine and hydroxy 24:0 fatty acid.
|
Carbohydrate sequence ions were found in the mass spectrum of the permethylated derivative (Figure 4A) at m/z 219 and 187 (terminal hexose), m/z 423 (terminal hexose-hexose), m/z 627 (terminal hexose-hexose-hexose), and m/z 831 (terminal hexose-hexose-hexose-hexose). The ions at m/z 219 and 187 and at m/z 423 were also present in the spectrum of the permethylated and reduced deivative (Figure 4B). The suggested hexose-hexose-hexose-hexose sequence was confirmed by the rearrangement fragment at m/z 904, present in both spectra. In Figure 4B a series of intense immonium ions at m/z 1158 to 1270, corresponding to the complete saccharide chain and the fatty acid, were found, demonstrating four hexoses and a series of hydroxy 16:0–24:0 fatty acids. The corresponding immonium ions were found at m/z 1172–1284 in the spectrum of the permethylated glycosphingolipid, along with ions at m/z 1300 and 1328, indicating phytosphingosine in combination with hydroxy 22:0 (1541 minus 241) and hydroxy 24:0 (1569 minus 241) fatty acids. The ions in the molecular region at m/z 1541 and 1569 (Figure 4A), and m/z 1527 and 1555 (Figure 4B), indicated the same carbohydrate composition and hydroxy 22:0 and 24:0 fatty acids, together with phytosphingosine long chain base.
The rearrangement ions at m/z 616 and 644, m/z 820 and 848, and m/z 1024 and 1052, produced by the permethylated and reduced derivative (explained below the formula in Figure 4B) further corroborated the proposed carbohydrate sequence and fatty acid composition.
The conclusion from MS of the EHEC-binding tetraglycosylceramide from cat intestine was thus a tetrahexosylceramide with phytosphingosine as long chain base and mainly hydroxy 22:0 and 24:0 fatty acids.
Degradation studies
The binding positions between the carbohydrate residues were obtained by degradation of the permethylated tri- and tetraglycosylceramide, that is, the samples were subjected to acid hydrolysis, followed by reduction and acetylation. The resulting partially methylated alditol acetates were analyzed by gas chromatography–EI MS, and the components were identified by comparison of retention times and mass spectra of partially methylated alditol acetates obtained from reference glycosphingolipids. The reconstructed ion chromatograms obtained from the triglycosylceramide (fraction C1; Figure 5A) and the tetraglycosylceramide (fraction C2; Figure 5B) were very similar, and both had three carbohydrate peaks. In both cases the acetate of 2,3,4,6-tetramethyl-galactitol identified a terminal galactose, and the presence of the acetates of 2,4,6-trimethyl-galactitol and 2,3,6-trimethyl-glucitol identified 3-substituted galactose and 4-substituted glucose, respectively.
|
In combination with the data from MS, this suggested that the triglycosylceramide had a carbohydrate chain with the sequence Gal1-3Gal1-4Glc1 and the tetraglycosylceramide had a Gal1-3Gal1-3Gal1-4Glc1 sequence.
Proton NMR spectroscopy
Both the tri- and tetraglycosylceramide structures were analyzed by 1H NMR spectroscopy at 500 MHz and 30°C. Starting with the smaller compound (data not shown), the spectrum revealed three H1 anomeric signals at 4.84 ppm (), 4.29 ppm (ß), and 4.21 ppm (ß), thus identifying this structure as Gal3Galß4Glcß1Cer (isoglobotriaosylceramide) through comparison with earlier published spectra of compounds having a terminal Gal3 (Dabrowski et al., 1988a,b). The identity of the Gal3 H1 signal was further corroborated in the double quantum-filtered correlated spectroscopy (DQF-COSY) spectrum by its connectivity to the H2 resonance at 3.60 ppm and the connectivities between the H5 resonance (3.99 ppm) and the H6,6' resonances (3.43 ppm and 3.52 ppm, respectively) (Dabrowski et al., 1988a). These values differ from those of a terminal Gal4, which reveals H1, H2, and H5 signals at 4.81 ppm, 3.65 ppm, and 4.05 ppm, respectively (Dabrowski et al., 1988b).
The four anomeric signals of the tetraglycosylceramide structure were located at 4.89 ppm (), 4.83 ppm (), 4.31 ppm (ß), and 4.21 ppm (ß), of which the latter two correspond to Galß4 and Glcß1 as previously (Figure 6). Regarding the two former signals, the one at 4.83 ppm is indicative of a terminal Gal3 as also found above (the H2 signal is located at 3.61 ppm), whereas the signal at 4.89 ppm can be indentified as an internal Gal3 (the H2 signal is now found at 3.71 ppm) from the COSY spectrum (not shown) and through comparison with earlier data on related compounds such as isoglobotetraosylceramide (GalNAcß3Gal3Galß4Glcß1Cer) (Hansson et al., 1987). An internal Gal4 as in, for example, globoside can be excluded because the H1 and H2 resonances in this case are found at 4.82 ppm and 3.82 ppm, respectively (Scarsdale et al., 1986). Considered together, these data (including the degradation studies) strongly suggest that the tetraglycosylceramide structure is Gal3Gal3Galß4Glcß1Cer.
|
From all the data combined, the structures of the EHEC-binding glycosphingolipids from cat small intestinal epithelium were established as Gal
3Galß4Glcß1Cer (isoglobotriaosylceramide) and Gal3Gal3Galß4Glcß1Cer. The latter glycosphingolipid has not been described previously.
Comparison of binding preferences
Binding to Gal3Galß4Glcß1Cer and Gal3Gal3Galß4Glcß1Cer was obtained with all 10 EHEC strains (representing 5 different serotypes) listed in Table II.
|
The EHEC binding activity of glycosphingolipids structurally related to Gal
3Gal3Galß4Glcß1Cer was thereafter examined (exemplified in Figures 7 and 8, and summarized in Table I). No binding of EHEC to structurally related compounds as the tetrahexosylceramide of rat intestine (Gal3Gal4Galß4Glcß1Cer (Figure 7, lane 4), the B5 glycosphingolipid (Gal3Galß4GlcNAcß3Galß4Glcß1Cer; Figure 2, lane 7, and Figure 7, lane 5), or isoglobotetraosylceramide (GalNAcß3Gal3Galß4Glcß 1Cer, not shown), was obtained.
|
|
However, further testing of EHEC binding with pure reference glycosphingolipids revealed binding patterns very similar to the patterns obtained with bacteria classified as lactosylceramide-binding (Ångström et al., 1998
; Hugosson et al., 1998; Karlsson, 1989; Strömberg and Karlsson, 1990a,b). Thus the bacteria selectively interacted with lactosylceramide with phytosphingosine and/or hydroxy fatty acid (no. 6 in Table I; Figure 8, lane 1), neolactotetraosylceramide (no. 12; Figure 8, lane 2), lactotetraosylceramide (no. 21; Figure 8, lane 3), the Lea-5 glycosphingolipid (no. 22; Figure 8, lane 4), gangliotriaosylceramide (no. 10), and gangliotetraosylceramide (no. 11). Similar binding patterns have thus been reported for a number of other bacteria, for example, Propionibacterium granulosum (Strömberg and Karlsson, 1990a) and Helicobacter pylori (Ångström et al., 1998). It should be noted, however, that although the binding of EHEC to lactosylceramide, neolactotetraosylceramide, lactotetraosylceramide, and the Lea-5 glycosphingolipid was highly reproducible, binding to gangliotriaosylceramide and gangliotetraosylceramide was only occasionally obtained.
All other glycosphingolipids tested, including those having a Gal4Gal sequence (nos. 8, 14, 24, 26, 27 in Table I), were nonbinding in the chromatogram binding assay.
Binding to reference glycosphingolipids was tested with the EHEC strains Bi 135A, Bi 110, RL 66, H10, H13, and Bi 44/91. All five strains displayed the same glycosphingolipid binding pattern.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Intimate attachment to the host cells, followed by the attaching and effacing (A/E) lesions on the surfaces of intestinal epithelial cells, is a characteristic feature of both EHEC infection and infection with enteropathogenic E. coli (EPEC). The A/E lesion formation involves degeneration of intestinal epithelial microvilli and formation of actin-rich pedestals in the cell beneath the adherent bacteria. A number of recent studies have shown that these effects are mediated primarily by the bacterial outer membrane protein intimin and a second bacterial protein Tir (translocated intimin receptor), which is exported by the bacteria and integrated into the host cell membrane. Subsequent binding of intimin to Tir triggers a signal transduction cascade followed by cytoskeletal rearrangements (reviewed in Campellone and Leong, 2003
).
Based on homologies in the C-terminal parts of the proteins, five antigenically distinct subtypes of intimin (designated , ß, 
, 
, and 
) have been identified (Abu-Bobie et al., 1998; Oswald et al., 2000). Intimin- and intimin- are found in human EPEC strains, intimin-ß and intimin- are produced both by EPEC and EHEC strains, and intimin- was found in EHEC strains.
Although both EPEC and EHEC are capable of forming A/E lesions, they have different target tissues. EPEC infects the small intestine, but the target tissue of EHEC is the large intestine (Levine, 1987). Thus the presence of a host-encoded intimin receptor has been suggested (reviewed in Frankel et al., 2001). A role for carbohydrate recognition in intimin-mediated adhesion of bacterial cells has also been suggested by the identification of a C-type lectin-like moiety at the carboxy-terminus of the intimin polypeptide by an NMR study of the C-terminal 280 amino acids of EPEC intimin- (Kelly et al., 1999).
In the present study, the potential carbohydrate recognition of EHEC bacteria was investigated by binding of radiolabeled bacteria to glycosphingolipids on thin-layer chromatograms. Thereby a specific recognition of lactosylceramide phytosphingosine and/or hydroxy fatty acids, neolactotetraosylceramide, lactotetraosylceramide, the Lea-5 glycosphingolipid, gangliotriaosylceramide, and gangliotetraosylceramide was demonstrated. Similar binding specificities have been previously been described for other bacteria with different target tissues (Ångström et al., 1998; Hugosson et al., 1998; Karlsson, 1989; Strömberg and Karlsson, 1990a,b).
In addition, a specific interaction of EHEC to two nonacid glycosphingolipids of cat small intestinal epithelium was detected during the course of the study. The binding-active glycosphingolipids were isolated, and on the basis of MS, gas chromatography–MS after degradation, and 1H NMR characterized as Gal3Galß4Glcß1Cer (isoglobotriaosylceramide) and Gal3Gal3Galß4Glcß1Cer. The latter glycosphingolipid is a novel structure.
The absence of interaction of EHEC with the rat intestinal tetrahexosylceramide (Gal3Gal4Galß4Glcß1Cer) and the B5 glycosphingolipid (Gal3Galß4GlcNAcß3Galß4Glcß1Cer) shows that a terminal 3-linked Gal is not sufficient for the binding process to occur. More likely the basic element recognized in the novel glycosphingolipid is the Gal3Galß4Glcß sequence of isoglobotriaosylceramide or even the Galß4Glcß sequence of lactosylceramide. The terminal 3-linked Gal of Gal3Gal3Galß4Glcß1Cer is thus a tolerated substitution, whereas the terminal GalNAcß3 of isoglobotetraosylceramide (GalNAcß3Gal3Galß4Glcß1Cer) blocks the binding. However, the latter glycosphingolipid may also be nonbinding due to the absence of the correct ceramide species.
The binding of EHEC and many other bacteria to lactosylceramide is ceramide-dependent, that is, the presence of phytosphingosine and/or hydroxy fatty acids is necessary for binding to occur. Selective binding to liposomes with lactosylceramide having phytosphingosine and hydroxy fatty acids has also been demonstrated (Ångström et al., 1998), suggesting that this selectivity may be present also in vivo. Previous molecular modeling studies have indicated that the selectivity is due to binding of a conformation of lactosylceramide in which the oxygen of the fatty acid hydroxyl group forms a hydrogen bond with the hydroxy methyl group of the glucose (Ångström et al., 1998).
Inspection of the binding-active and nonbinding glycosphingolipids summarized in Table I shows that not all glycosphingolipids with phytosphingosine and hydroxy fatty acids are recognized by EHEC bacteria. In these cases (nos. 16, 22, 23, and 27 in Table I) the extensions of the carbohydrate chains most likely limit the access to the lactosyl element.
Furthermore the requirement of phytosphingosine and/or hydroxy fatty acids is restricted to lactosylceramide/ isoglobotriaosylceramide, and vanishes upon is elongation into ganglio and neolacto series compounds (nos. 10–12). For other lactosylceramide-binding bacteria, such as H. pylori, Neisseria meningitidis, and Haemophilus influenzae, the binding to gangliotria- and gangliotetraosylceramide is abolished after conversion of the acetamido group of GalNAc to an amine (Ångström et al., 1998; Hugosson et al., 1998), suggesting that the binding to these two glycosphingolipids is a specificity separate from lactosylceramide recognition. Our current hypothesis is that the binding to lactosylceramide/isoglobotriaosylceramide, gangliotria-/gangliotetraosylceramide, and N-acetyllactosamine-terminated glycosphingolipids correspond to three different adhesins that are frequently coexpressed by the lactosylceramide-binding bacteria.
The relevance of the Gal3Galß4Glcß1Cer/Gal3Gal3Galß4Glcß1Cer binding capacity for adhesion of EHEC to host cells and colonization has yet to be determined. Isoglobotriasylceramide is present in, for instance, dog intestine (Hansson et al., 1983), but has hitherto not been found in human tissues. EHEC are strict large intestinal pathogens (Levine, 1987). In the epithelial cells of human large intestine the Lea-5 glycosphingolipid is a major compound, and there are also trace amounts of lactosylceramide with phytosphingosine and/or hydroxy fatty acids (Holgersson et al., 1988). These may thus function as targets for EHEC adherence.
Nucleolin was recently identified as the receptor for intimin- on Hep-2 cells (Sinclair and O'Brien, 2002). Obviously it would be of interest to examine the carbohydrate binding of purified intimin, both from EHEC and EPEC, to determine if there is a correlation with the carbohyrate binding specificities found in this study or if the carbohydrate binding is caused by other bacterial adhesins.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Bacterial strains, culture conditions, and labeling
The serotypes and characteristics of the EHEC isolates used are summarized in Table II. The strains were isolated from fecal specimens of patients with hemorrhagic colitis. RL 66 was from Robert Koch Institut, Berlin; B1409C and C984 from Center for Vaccine Development, Baltimore, MD; and the other strains from Clinical Microbiology and Immunology, Lund University Hospital, Lund, Sweden. Most of the binding experiments were done with the strain Bi 135A S; two or three of the other strains were used in parallel.
The Gal4Gal-binding recombinant E. coli strain HB101/pPIL291–15, carrying a plasmid-born pap gene cluster with a class II papG allele, was a kind gift from Dr. Benita Westerlund (University of Helsinki, Finland).
All E. coli strains were cultured on Luria agar with the addition of 10 µl 35S-methionine (400 µCi; Amersham Pharmacia Biotech, Little Chalfont, UK) at 37°C for 12 h. The bacteria were harvested by scraping, washed three times with phosphate-buffered saline (PBS), pH 7.3, and thereafter resuspended in PBS containing 1% mannose (w/v) to 1 x 108 CFU/ml. The specific activities of the suspensions were approximately 1 cpm per 100 bacteria.
Reference glycosphingolipids
Total acid and nonacid glycosphingolipid fractions were obtained by standard procedures (Karlsson, 1987). The individual glycosphingolipids were isolated by repeated chromatography on silicic acid columns of the native glycosphingolipid fractions or acetylated derivatives thereof. The identity of the purified glycosphingolipids was confirmed by MS (Samuelsson et al., 1990), proton NMR spectroscopy (Koerner et al., 1983), and degradation studies (Stellner et al., 1973; Yang and Hakomori, 1971).
Thin-layer chromatography
Thin-layer chromatography of glycosphingolipids was performed on glass- or aluminum-backed silica gel 60 high-performance thin-layer chromatography plates (Merck, Darmstadt, Germany), using chloroform/methanol/water (60:35:8 v/v/v) as solvent system. Chemical detection was done with anisaldehyde (Waldi, 1962).
Chromatogram binding assays
Binding of 35S-labeled bacteria to glycosphingolipids on thin-layer chromatograms was done as reported (Hansson et al., 1985). Dried chromatograms were dipped for 1 min in diethylether/n-hexane (1:5 v/v) containing 0.5% (w/v) polyisobutylmethacrylate (Aldrich, Milwaukee, WI). After drying, the chromatograms were soaked in PBS containing 2% bovine serum albumin (w/v), 0.1% NaN3 (w/v), and 0.1% Tween 20 (v/v) for 2 h at room temperature. The chromatograms were subsequentely covered with radiolabeled bacteria diluted in PBS (2–5 x 106 cpm/ml). Incubation was done for 2 h at room temperature, followed by repeated washings with PBS. The chromatograms were thereafter exposed to XAR-5 X-ray films (Eastman Kodak, Rochester, NY) for 12 h.
Isolation of EHEC-binding nonacid glycosphingolipids from epithelial cells of cat small intestine
A total nonacid glycosphingolipid fraction (64 mg) was obtained from pooled epithelial cell scrapings from the small intestines of 12 cats by standard methods (Karlsson, 1987). Briefly, the tissue was lyophilized, followed by extraction in two steps with chloroform-methanol (2:1 and 9:1, v/v) in a Soxhlet apparatus. The material obtained was pooled and subjected to mild alkaline hydrolysis and dialysis, followed by separation on a silicic acid column. Acid and nonacid glycosphingolipids were separated on a DEAE column.
The isolated nonacid glycosphingolipid fraction (50 mg) was first separated on a silicic acid column stepwise eluted with increasing amounts of methanol in chloroform. The fractions containing triglycosylceramides and larger glycosphingolipids were pooled, giving 45 mg, which were further separated by high-performance liquid chromatography on a Kromasil 5 Silica column (2.12 x 25 cm inner diameter, particle size 5 µm; Skandinaviska Genetec, Kungsbacka, Sweden) eluted with a linear gradient of chloroform/methanol/water (80:20:1 to 40:40:12 v/v/v) over 180 min, with a flow rate of 4 ml/min. Each 4-ml fraction was analyzed by thin-layer chromatography using anisaldehyde for detection.
The glycosphingolipid-containing fractions were tested for binding of EHEC and P-fimbriated E. coli using the chromatogram binding assay. The EHEC-binding compound migrating in triglycosylceramide region eluted in tubes 78–88, whereas tetraglycosylceramides were collected in tubes 89–106. After pooling, 1.8 mg of triglycosylceramides (designated fraction C1) and 10.2 mg of tetraglycosylceramides were obtained. The tetraglycosylceramide fraction was acetylated (Handa, 1963), and further separated on a 10-g Iatrobeads column (Iatron Laboratories, Tokyo) eluted with a linear gradient of methanol in chloroform (0–5%, v/v). Fractions of 1 ml were collected and, after deacetylation, analyzed by thin-layer chromatography and tested for binding activities. The compound collected in tubes 120–140 was recognized by EHEC but not by P-fimbriated E. coli, and after pooling, 1.5 mg of this compound was obtained (designated fraction C2). Tubes 141–195 contained a mixture of compounds recognized by EHEC and P-fimbriated E. coli. Pooling of these tubes yielded 2.4 mg (designated fraction C3).
MS
Before EI MS, aliquots of the isolated glycosphingolipids were permethylated (Larson et al., 1987) or permethylated and reduced with LiAlH4 (Karlsson, 1974). The derivatized samples were analyzed on a JEOL SX-102A mass spectrometer (JEOL, Tokyo), using the in-beam technique (Breimer et al., 1980). The analyses of both derivatives were performed with an electron energy of 70 eV, trap current of 300 µA, and acceleration voltage of 10 kV. The temperature was raised from 150°C to 410°C, by increases of 10°C/min.
Degradation studies
The permethylated glycosphingolipids from cat small intestinal epithelium were hydrolyzed, reduced, and acetylated (Stellner et al., 1973; Yang and Hakomori, 1971), and the partially methylated alditol acetates obtained were analyzed by capillary gas chromatography on a Hewlett-Packard 5890A gas chromatograph using a BPX-35 0.25 µm column (30 m x 0.22 mm ID) (SGE, Perth, Australia), with split injection and with helium as carrier gas. The samples were dissolved in ethylacetate. The temperature was raised from 150°C to 280°C by increases of 10°C/min.
Gas chromatography–EI MS was performed on a Hewlett-Packard 5890-II gas chromatograph coupled to a JEOL SX-102A mass spectrometer. The chromatographic conditions, as well as the capillary column, were the same as for the analyses by gas chromatography, and the conditions for MS were: interface temperature 260°C, ion source temperature 200°C, electron energy 70 eV, trap current 300 µA, acceleration voltage 10 kV, mass range scanned 50–500, total cycle time 1.5 s, resolution 1200 (10% valley definition), and pressure in the ion source region 10–5 Pa.
Proton NMR spectroscopy
1H NMR spectra were acquired on a Varian 500 MHz spectrometer at 30°C. Samples were dissolved in dimethyl sulfoxide/D2O (98/2) after deuterium exchange. Two-dimensional DQF-COSY spectra were recorded by the standard pulse sequence (Marion and Wüthrich, 1983) using 32 scans per t1 increment in 2K * 256 point matrices.
The glycosphingolipid nomenclature follows the recommendations by the IUPAC-IUB Commission on Biochemical Nomenclature (CBN for lipids: Eur. J. Biochem. [1977] 79, 11–21, J. Biol. Chem. [1982] 257, 3347–3351, and J. Biol. Chem. [1987] 262, 13–18). It is assumed that Gal, Glc, GlcNAc, GalNAc, and NeuAc are of the D-configuration, Fuc of the L-configuration, and all sugars present in the pyranose form.
|
Acknowledgements |
|---|
This study was supported by the Swedish Medical Research Council (Grant No. 12628), the Swedish Cancer Foundation, the Wallenberg Foundation and Lund University Hospital, Lund, Sweden. Gal
3Gal4Galß4Glcß1Cer from rat small intestine was a kind gift from Dr. G. C. Hansson at the Institute of Medical Biochemistry, Göteborg University. The use of the Varian 500 MHz machine at the Swedish NMR Centre, Hasselblad Laboratory, Göteborg University, is gratefully acknowledged. |
Footnotes |
|---|
1 To whom correspondence should be addressed; e-mail: susann.teneberg{at}medkem.gu.se
|
Abbreviations |
|---|
A/E, attaching and effacing; DQF-COSY, double quantum filtered correlation spectroscopy; EHEC, enterohemorrhagic Escherichia coli; EI, electron ionization; EPEC, enteropathogenic Escherichia coli; FAB fast atom bombardment; MS, mass spectrometry; NMR, nuclear magnetic resonance; PBS, phosphate buffered saline; Stx, Shiga toxin
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Abu-Bobie, J., Frankel, G., Bain, C., Guedes Concalves, A., Trabulsi, L.R., Douce, G., Knutton, S., and Dougan, G. (1998) Detection of intimins
, ß, and , four intimin derivatives expressed by attaching and effacing microbial pathogens. J. Clin. Microbiol., 36, 662–668.
Ångström, J., Teneberg, S., Abul Milh, M., Larsson, T., Leonardsson, I., Olsson, B.-M., Ölwegård Halvarsson, M., Danielsson, D., Ljungh, Å., Wadström, T., and Karlsson, K.-A. (1998) The lactosylceramide binding specificity of Helicobacter pylori. Glycobiology, 8, 297–309.
Beachey, E.H. (1981) Bacterial adherence: adhesin-receptor interactions mediating the attachment of bacteria to mucosal surfaces. J. Infect. Dis., 143, 325–345.[Web of Science][Medline]
Breimer, M., Hansson, G.C., Karlsson, K.-A., Larson, G., Leffler, H., Pascher, I., Pimlott, W., and Samuelsson, B.E. (1980) Fingerprinting of lipid-linked oligosaccharides by mass spectrometry. In Quayle, A. (Ed.), Advances in mass spectrometry. Heyden and Son, London, vol. 8, pp. 1097–1108.
Campellone, K.G. and Leong, J.M. (2003) Tails of two Tirs: actin pedestal formation by enteropathogenic E. coli and enterohemorrhagic E. coli 0157:H7. Curr. Opin. Microbiol., 6, 82–90.[CrossRef][Web of Science][Medline]
Dabrowski, J., Dabrowski, U., Kordowicz, M., and Hanfland, P. (1988a) Structure elucidation of the blood group B like and blood group I active octaantennary ceramide tetracontasaccharide from rabbit erythrocyte membranes by two-dimensional 1H NMR spectroscopy at 600 MHz. Biochemistry, 27, 5149–5155.[CrossRef][Medline]
Dabrowski, J., Trauner, K., Koike, K., and Ogawa, T. (1988b) Complete 1H-NMR spectral assignments for globotriaosyl-Z- and isoglobotriaosyl-E-ceramide. Chem. Phys. Lipids, 49, 31–37.[CrossRef][Web of Science][Medline]
Frankel, G., Phillips, A.D., Trabulsi, L.R., Knutton, S., Dougan, G., and Matthews, S. (2001) Intimin and the host cell—is it bound to end in Tir(s)? Trends Microbiol., 9, 214–218.[CrossRef][Web of Science][Medline]
Handa, S. (1963) Blood group active glycolipid from human erythrocytes. Jap. J. Exp. Med., 33, 347–360.[Medline]
Hansson, G.C., Karlsson, K.-A., Larson, G., McKibbin, J.M., Strömberg, N., and Thurin, J. (1983) Isoglobotriaosylceramide and the Forssman glycolipid of dog small intestine occupy separate tissue compartments and differ in ceramide composition. Biochim. Biophys. Acta, 750, 214–216.[Medline]
Hansson, G.C., Karlsson, K.-A., Larson, G., Strömberg, N., and Thurin, J. (1985) Carbohydrate-specific adhesion of bacteria to thin-layer chromatograms: a rationalized approach to the study of host cell glycolipid receptors. Anal. Biochem., 146, 158–163.[CrossRef][Web of Science][Medline]
Hansson, G.C., Bouhours, J.-F., and Ångström, J. (1987) Characterization of neutral blood group B-active glycosphingolipids of rat gastric mucosa. A novel type of blood group active glycosphingolipid based on isogloboside. J. Biol. Chem., 262, 13135–13141.
Holgersson, J., Strömberg, N., and Breimer, M.E. (1988) Glycolipids of human large intestine: difference in glycolipid expression related to anatomical localization, epithelial/non-epithelial tissue and the ABO, Le and Se phenotypes of the donors. Biochimie, 70, 1565–1574.[Medline]
Hugosson, S., Ångström, J., Olsson, B.-M., Bergström, J., Fredlund, H., Olcén, P., and Teneberg, S. (1998) Glycosphingolipid binding specificities of Neisseria meningitidis and Haemophilus influenzae: detection, isolation and characterization of a binding-active glycosphingolipid from human oropharyngeal epithelium. J. Biochem., 124, 1138–1152.
Karlsson, K.-A. (1974) Carbohydrate composition and sequence analysis of a derivative of brain disialo ganglioside by mass spectrometry, with molecular weight ions at m/e 2245. Potential use in the specific microanalysis of cell surface components. Biochemistry, 13, 3643–3647.[CrossRef][Medline]
Karlsson, K.-A. (1987) Preparation of total non-acid glycolipids for overlay analysis of receptors for bacteria and viruses and for other studies. Methods Enzymol., 138, 212–220.[Web of Science][Medline]
Karlsson, K.-A. (1989) Animal glycosphingolipids as membrane attachment sites for bacteria. Annu. Rev. Biochem., 58, 309–350.[CrossRef][Web of Science][Medline]
Kaper, J.B. (1998) Enterohemorrhagic Escherichia coli. Curr. Opin. Microbiol., 1, 103–108.[CrossRef][Web of Science][Medline]
Kelly, G., Prasannan, S., Daniell, S., Fleming, K., Frankel, G., Dougan, G., Connerton, I., and Matthews, S. (1999) Structure of the cell-adhesion fragment of intimin from enteropathogenic Escherichia coli. Nature Struct. Biol., 6, 313–318.[CrossRef][Web of Science][Medline]
Koerner, T.A.W. Jr., Prestegard, J.H., Demou, P.C., and Yu, R.K. (1983) High-resolution proton NMR studies of gangliosides. 1. Use of homonuclear spin-echo J-correlated spectroscopy for determination of residue composition and anomeric configurations. Biochemistry, 22, 2676–2687.[CrossRef][Medline]
Larson, G., Karlsson, H., Hansson, G.C., and Pimlott, W. (1987) Application of a simple methylation procedure for the analysis of glycosphingolipids. Carbohydr. Res., 161, 281–290.[CrossRef][Web of Science][Medline]
Levine, M.M. (1987) Escherichia coli that cause diarrhea: enterotoxigenic, enteropathogenic, enteroinvasive, enterohemorrhagic, and enteroadherent. J. Infect. Dis., 155, 377–389.[Web of Science][Medline]
Lindberg, A.A., Brown, J.E., Strömberg, N., Westling-Ryd, M., Schultz, J.E., and Karlsson, K.-A. (1987) Identification of the carbohydrate receptor for Shiga toxin produced by Shigella dysenteriae type 1. J. Biol. Chem., 262, 1779–1785.
Ling, H., Boodhoo, A., Hazes, B., Cummings, M.D., Armstrong, G.D., Brunton, J.L., and Read, R.J. (1998) Structure of the Shiga-like toxin I B-pentamer complexed with an analogue of its receptor Gb3. Biochemistry, 13, 1777–1788.
Lingwood, C.A., Law, H., Richardsson, S., Petric, M., Brunton, J.L., De Grandis, S., and Karmali, M. (1987) Glycolipid binding of purified and recombinant Escherichia coli produced verotoxin in vitro. J. Biol. Chem., 262, 8834–8839.
Marion, D. and Wüthrich, K. (1983) Application of phase sensitive two-dimensional correlated spectroscopy (COSY) for measurements of 1H-1H spin-spin coupling constants in proteins. Biochem. Biophys. Res. Commun., 113, 967–974.[CrossRef][Web of Science][Medline]
Nataro, J.P. and Kaper, J.B. (1998) Diarrheagenic Escherichia coli. Clin. Microbiol. Rev., 11, 142–201.
Oswald, E., Schmidt, H., Morabito, S., Karch, H., Marchès, O., and Capriolo, C. (2000) Typing of intimin genes in human and animal enterhemorrhagic and enteropathogenic Escherichia coli: characterization of a new intimin variant. Infect. Immun., 68, 64–71.
Samuelsson, B.E., Pimlott, W., and Karlsson, K.-A. (1990) Mass spectrometry of mixtures of intact glycosphingolipids. Methods Enzymol., 193, 623–646.[Medline]
Scarsdale, J.N., Yu, R.K., and Prestegard, J.H. (1986) Structural analysis of a glycolipid head group with one- and two-state NMR pseudoenergy approaches. J. Am. Chem. Soc., 108, 6778–6784.[CrossRef]
Sinclair, J.F. and O'Brien, A.D. (2002) Cell surface-localized nucleolin is a eukaryotic receptor for the adhesin intimin- of enterohemorrhagic Escherichia coli 0157:H7. J. Biol. Chem., 277, 2876–2885.
Stellner, K., Saito, H., and Hakomori, S.-i. (1973) Determination of aminosugar linkages in glycolipids by methylation. Aminosugar linkages of ceramide pentasaccharides of rabbit erythrocytes and of Forssman antigen. Arch. Biochem. Biophys., 155, 464–472.[CrossRef][Web of Science][Medline]
Strömberg, N. and Karlsson, K.-A. (1990a) Characterization of the binding of Propionibacterium granulosum to glycosphingolipids adsorbed on surfaces. An apparent recognition of lactose which is dependent on the ceramide structure. J. Biol. Chem., 265, 11244–11250.
Strömberg, N. and Karlsson, K.-A. (1990b) Characterization of the binding of Actinomyces naeslundii (ATCC 12104) and Actinomyces viscosus (ATCC 19246) to glycosphingolipids, using a solid-phase overlay approach. J. Biol. Chem., 265, 11251–11258.
Waldi, D. (1962) Sprühreagentien für die dünnschicht-chromatographie. In Stahl, E. (Ed.), Dünnschicht-Chromatographie. Springer-Verlag, Berlin, pp. 496–515.
Yang, H.-j. and Hakomori, S.-i. (1971) A sphingolipid having a novel ceramide and lacto-N-fucopentose III. J. Biol. Chem., 246, 1192–1200.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Zhou, J. Mattner, C. Cantu III, N. Schrantz, N. Yin, Y. Gao, Y. Sagiv, K. Hudspeth, Y.-P. Wu, T. Yamashita, et al. Lysosomal Glycosphingolipid Recognition by NKT Cells Science, December 3, 2004; 306(5702): 1786 - 1789. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Hellstrom, E. C. Hallberg, J. Sandros, L. Rydberg, and A. E. Backer Carbohydrates act as receptors for the periodontitis-associated bacterium Porphyromonas gingivalis: a study of bacterial binding to glycolipids Glycobiology, June 1, 2004; 14(6): 511 - 519. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||