| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Glycobiology | Pages |
© |
Biotinyl-l-3-(2-naphthyl)-alanine hydrazide derivatives of N-glycans: versatile solid-phase probes for carbohydrate-recognition studies
Introduction
Results
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
Biotinyl-l-3-(2-naphthyl)-alanine hydrazide derivatives of N-glycans: versatile solid-phase probes for carbohydrate-recognition studies
Biotinyl-oligosaccharides are a relatively new generation of saccharide probes that enable immobilization of desired oligosaccharides on streptavidin matrices for studies of carbohydrate-protein interactions. Here we describe the facile preparation of biotinyl-l-3-(2-naphthyl)-alanine hydrazide (BNAH) derivatives of oligosaccharides, containing a strong UV absorbing and fluorescent group, in which the ring of the reducing-end monosaccharide is nonreduced. We evaluate reactivities of immobilized BNAH-N-glycans with plant lectins that recognize aspects of the oligosaccharide core or outer-arms. We make some comparisons with 2-amino-6-amidobiotinyl-pyridine (BAP) derivatives obtained by reductive amination, and 6-(biotinyl)-aminocaproyl-hydrazide (BACH) derivatives which have a longer spacer-arm. N-Glycan-BNAH and-BAP derivatives have, overall, comparable reactivities with lectins which recognize N-glycan outer-arms or the trimannosyl core, but only BNAH and BACH derivatives are bound by lectins which recognize the non-reduced core. Moreover, with Pisum sativum agglutinin (PSA) which additionally requires the fucosyl-N-glycan-asparaginyl core for high affinity binding, the immobilized BNAH derivative (which is an alanine hydrazide [beta]-glycoside) can substitute for the natural [beta]-glycosylasparaginyl core, whereas the BACH derivative (aminocaproyl-hydrazide-[beta]-glycoside) is less effective. BNAH is a derivatization reagent of choice, therefore, for solid phase carbohydrate-binding experiments with immobilized N-glycans. Key words: biotinylated oligosaccharides/carbohydrate recognition/N-linked oligosaccharides/lectins/1H-NMR
Introduction
Techniques for the study of protein-carbohydrate interactions have become a focus of interest in glycobiology. This is due largely to the awareness that interactions of proteins with specific oligosaccharide sequences of glycoproteins and glycolipids are crucial steps in the cascade of events that constitute the inflammatory response (Bevilacqua and Nelson, 1993; Drickamer and Taylor, 1993; Feizi, 1993; Crocker and Feizi, 1996; Rosen and Bertozzi, 1996). Thus, it has become desirable to have microtechniques that enable binding studies to be performed with structurally defined oligosaccharide sequences. It is also clear that there is a need for multiple techniques, as the binding signals may be markedly influenced by different modes of oligosaccharide presentation in in vitro experiments; these differences may have a bearing on the behavior of oligosaccharides as ligands in vivo (Feizi, 1993; Green et al., 1995, and references therein).
Mono- and oligosaccharides linked to proteins or synthetic cluster glycosides mimicking branched oligosaccharide structures have been valuable reagents for probing the carbohydrate binding specificities of proteins, and for assessing the cooperative effects of multivalence in the strength of the binding signal (Lee, 1992). Oligosaccharides linked to aminophospholipids (neoglycolipids) behave as cluster ligands when immobilized on plastic matrices; they have the additional advantage that they can be resolved on chromatograms, and the technology is applicable for the pinpointing of ligand-bearing sequences among mixtures of oligosaccharides derived from biological sources (Feizi et al., 1994). Biotinylated oligosaccharides have emerged as an additional class of promising probes that allow exploitation of the strong affinity of avidin and streptavidin for biotin (Gitlin et al., 1987). As each streptavidin molecule has four biotin binding sites, the streptavidin-biotinyl-glycan complexes may behave as cluster ligands. Applications have included the immobilization of biotinylated glycopeptides on microtiter wells for conventional binding experiments (Shao, 1992; Shao and Chin, 1992; Shao et al., 1990), and on the sensor microchip of the BIAcore surface plasmon resonance instrument for kinetic measurements of lectin-carbohydrate interactions (Shinohara et al., 1995, 1996).
Biotinylated oligosaccharides have been prepared in various ways: for example, by coupling biocytin hydrazide to oxidized sialic acid or to oxidized galactose residues of glycoproteins (Bayer et al., 1988) or by a multistep procedure involving formation of glycosylamines (Manger et al., 1992a,b). Biotinylation of glycosylasparagines has been achieved by coupling activated biotin (N-hydroxy-succinimido-biotin) to the amino group of the asparagine (Shaoand Chin, 1992). Drawbacks of these procedures include the chemical modification of oligosaccharide structure, low yields in the multistep procedure, and difficulties in obtaining homogeneous preparations of glyco-asparagines, respectively. Three papers have described improved methods for biotinylation of oligosaccharides. In one, oligosaccharides were treated with the UV-absorbent/fluorescent BAP, under reductive amination conditions, and biotinylated oligosaccharides were obtained in good yield (Rothenberg et al., 1993; Toomre and Varki, 1994). In another, oligosaccharides were coupled directly, with preservation of the reducing end monosaccharide, to the weakly UV-absorbent but nonfluorescent BACH (Shinohara et al., 1995). In a more recent paper, oligosaccharides were coupled to a hydrazide reagent containing a moderately UV-absorbing, nonfluorescent group: 4-biotinamido-phenylacetylhydrazide, BPH (Shinohara et al., 1996).
Figure
Figure
Here we describe the preparation of novel biotinylated oligosaccharide derivatives using the reagent BNAH, which has both UV-absorbing and fluorescent properties, and allows coupling to oligosaccharide under nonreductive conditions. We evaluate the binding of the immobilized BNAH derivatives by plant lectins that recognize aspects of the core region and outer chains of N-glycans. Comparisons with the binding to oligosaccharide-BAP and -BACH derivatives show considerable advantages of the BNAH derivatives as immobilized ligands for carbohydrate-binding studies with the lectins investigated.
Results
Biotinylated oligosaccharides
The oligosaccharides investigated in the present study, their abbreviations and sequences are given in Table I. In preliminary experiments it was established that the reaction with BNAH could be carried out in either methanol/water or methanol/water/acetic acid. In the acidic solvent the reaction was complete after 5 h at 60°C as revealed by HP-TLC of the reaction mixture followed by orcinol staining, whereas in the neutral solvent, conjugation was incomplete. A 16 h incubation at 60°C was needed to reach complete conjugation. Under the conditions used (methanol/water/acetic acid at 60°C), the oligosaccharides were stable. As an example a chromatogram of Man5-BNAH reaction mixture is shown in Figure 1A. Reaction mixtures for the BAP and BACH derivatives were similarly monitored by HP-TLC and orcinol staining, the BAP reaction mixtures being additionally visualized under UV light at 300 nm (Figure 1B,B[prime]).
Purification of the reaction mixtures were carried out by RP-HPLC (Figure 2). The eluted fractions were evaporated and analyzed by HP-TLC using aliquots of the reaction mixtures as reference materials. Hexose-containing fractions with Rf values appropriate for derivatives (cf. Figure 1) were analyzed by LSIMS and their identities established from their positive-ion spectra, MH+ and/or MNa+ ions which were consistent with the calculated molecular masses (Table II). Spectra for Man5-BAP and Man5-BNAH are shown in Figure 3, and the retention times of the oligosaccharide derivatives prepared are as indicated in Table II. NMR studies of the saccharide-tag linkage were performed with three disaccharide derivatives as model compounds: LacNAc-BNAH, GN2-BNAH and LacNAc-BACH. 1H chemical shifts in Table III show unambiguously the closed ring structure of the reducing monosaccharides, which are found to adopt almost exclusively the [beta]-anomeric configuration (JH1,H2 = 9.3-9.5 Hz). In the BNAH derivatives, strong highfield shifts occur for protons of the reducing-end monosaccharides; in particular H1 and H3 are strongly affected by the ring current effect of the neighboring naphthyl group. H3 of GN2-BNAH appears at 2.073 ppm, a region normally occupied by NAc protons.Table I
| Oligosaccharide derivatives | Retention time (min) | HPLC gradient acetonitrile % | Molecular mass | ||
| Observed | |||||
| Calculateda | MH+ | MNa+ | |||
| Man5GlcNAc2-BNAH | 14.19 | 5 to 100 | 1671.7 (m)a 1672.7 (a)a |
1672.9 | 1694.9 |
| Man5-BNAH | 15.10 | 0 to 100 | 1265.5 (m) 1266.3 (a) |
1266.6 | 1288.6 |
| Man3-BNAH | 15.68 | 0 to 100 | 941.4 (m) 942.0 (a) |
942.4 | 964.4 |
| NA2-BNAH | 14.05 | 5 to 100 | 2077.8 (m) 2079.1 (a) |
2079.8 | 2101.9 |
| NA2F-BNAH | 13.91 | 5 to 100 | 2223.8 (m) 2225.2 (a) |
2225.7 | 2247.5 |
| GN2-BNAH | 20.58 | 5 to 70 | 861.4 (m) 862.0 (a) |
862.3 | 884.3 |
| LacNAc-BNAH | 21.43 | 5 to 70 | 820.3 (m) 820.9 (a) |
821.3 | 843.3 |
| Man5GlcNAc2-BAP | 16.23 | 5 to 70 | 1553.6 (m) 1554.6 (a) |
1554.7 | 1576.7 |
| Man5-BAP | 11.95 | 10 to 100 | 1147.4 (m) 1148.2 (a) |
1148.5 | 1170.5 |
| Man3-BAP | 12.52 | 10 to 70 | 823.3 (m) 823.9 (a) |
824.4 | Not observed |
| Man2-BAP | 15.17 | 10 to70 | 661.3 (m) 661.8 (a) |
662.2 | 684.2 |
| NA2-BAP | 15.74 | 5 to 70 | 1959.7 (m) 1960.9 (a) |
1961.3 | 1983.1 |
| NA2F-BAP | 16.23 | 5 to 70 | 2105.8 (m) 2107.1 (a) |
2106.9 | 2128.6 |
| NA2-BACH | 17.35 | 5 to 85 | 1993.8 (m) 1995.0 (a) |
1995.3 | 2017.1 |
| NA2F-BACH | 17.28 | 5 to 85 | 2139.9 (m) 2141.2 (a) |
2141.8 | 2163.3 |
| LacNAc-BACH | 17.63 | 5 to 85 | 736.3 (m) 736.8 (a) |
737.4 | 759.4 |
Table II
| Protons | LacNAc-BACH | LacNAc-BNAH | GN2-BNAH |
| Naphthyl | 7.95-7.504 | 7.954-7.497 | |
| Alanine | |||
| CH | 4.699 (8.2) | 4.592 (9.7) | |
| CH2 | 3.285, 3.195 | 3.271, 3.166 (13.4) | |
| Biotin | |||
| 2[beta] | 3.342 (4.5) | 2.932 (4.5) | 3.105 (4.6) |
| 3[beta] | 4.428 (7.9) | 4.088 (7.8) | 4.230 (8.0) |
| 4[beta] | 4.614 (4.9) | 4.517 (4.9) | 4.561 (4.9) |
| 5[beta] | 2.998 | 2.914 | 2.952 |
| 5[alpha] | 2.785 (13.1) | 2.711 (13.0) | 2.742 (13.0) |
| (CH2)4 | |||
| [alpha],(C=OCH2) | 2.251 | 2.272 (6.9) | 2.302 (7.0) |
| [beta] | 1.750-1.591 | 1.528-1.431 | 1.600-1.530 |
| [gamma] | 1.750-1.591 | 1.390-1.334 | 1.492-1.417 |
| [delta]a | 1.407 | 1.120-1.058 | 1.243-1.150 |
| [delta]b | 1.012-0.949 | ||
| (CH2)5 | |||
| [alpha],(C=OCH2) | 2.206 | ||
| [beta] | 1.601 | ||
| [gamma] | 1.323 | ||
| [delta] | 1.520 | ||
| [epsis],(NHCH2) | 3.182 | ||
| Gal[prime] or GlcNAc[prime] | |||
| 1[prime] | 4.466 (7.8) | 4.261 (8.0) | 4.364 (8.4) |
| 2[prime] | 3.536 (9.9) | 3.503 (9.9) | 3.726 (9.6) |
| 3[prime] | 3.666 (3.3) | 3.661 (3.5) | 3.473b |
| 4[prime] | 3.919 | 3.928 | 3.473b |
| 5[prime] | 3.783-3.690 | 3.767-3.716 | 3.528 (3.6) |
| 6[prime]a | 3.783-3.690 | 3.767-3.716 | 3.895 |
| 6[prime]b | 3.783-3.690 | 3.767-3.716 | 3.752 (12.1) |
| NAc | 2.166c | ||
| GlcNAc[beta] | |||
| 1 | 4.223 (9.6) | 3.623 (9.3) | 3.249 (9.5) |
| 2 | 3.843 | 3.411 | 3.190 |
| 3 | 3.690b | 2.711 | 2.073 |
| 4 | 3.690b | 3.128 (9.3) | 3.044 (8.4) |
| 5 | 3.554 (5.2) | 3.461 | 3.326b |
| 6a | 3.953 | 3.517 | 3.326b |
| 6b | 3.825 (11.5) | 3.421 | 3.206 (4.6, 12.0) |
| NAc | 2.029 | 2.057 | 2.059c |
Figure
The purified BNAH-, BAP-, and BACH-oligosaccharide derivatives are stable for at least a year when stored at -20°C in water/methanol, 9:1 v/v.
Lectin interactions with biotinylated oligosaccharides
Concanavalin A. Among the immobilized BNAH and BAP derivatives tested for binding by ConA (Figure 4A,B), the minimum structure bound was Man3-BNAH in which the pyranose ring of the branched mannose is retained. No binding was detected to the Man3-BAP in which the pyranose ring of the branch-point mannose is open, nor was there any binding to Man2-BAP. The binding signal with Man3-BNAH was weaker, however, compared with that obtained with the Man5 and the Man5GlcNAc2 derivatives. All four Man5-containing derivatives gave binding signals stronger than those with the complex-type NA2 derivatives (see also results of a binding experiment Figure 5C) that included NA2-BACH. These results are in accordance with previous knowledge on the higher binding affinity of this lectin toward high-mannose type oligosaccharides (Baenziger et al., 1979; Debray et al., 1981; Mandal et al., 1994a,b).
Figure
Figure
RCA120. In the case of RCA120 which recognizes the outer, [beta]-galactose-terminating domain of the NA2 oligosaccharide (Green et al., 1987), binding signals with its BNAH and BAP derivatives were equivalent (Figure 4C) and, as predicted, no binding was detected to Man5-BAP.
LCA and PSA. Initial binding experiments with LCA and PSA were performed with the BNAH and BAP derivatives of NA2 and NA2F (not shown). Both lectins recognize biantennary N-glycans with core fucose on an intact 2-acetamido-glucopyranose (Debray et al., 1981; Kornfeld et al., 1981; Rini et al., 1993; Schwarz et al., 1993). In the case of PSA, an additional requirement for an adjoining asparagine has been documented for high affinity binding (Yamamoto et al., 1982). With these two lectins, as predicted, no binding was observed to the NA2 derivatives nor to the BAP derivative of NA2F. With NA2F-BNAH, however, there was binding not only by LCA but also PSA. This raised the possibility that in the BNAH derivative, the naphthyl alanine 'mimics" the asparagine adjoining the core monosaccharide of N-glycans. In the experiment illustrated (Figure 5A,B), binding of the two lectins to the NA2- and NA2F-BACH derivatives (containing an aminocaproyl group) was examined, as well as to their BNAH derivatives. With both lectins, binding signals were elicited by the NA2F-BACH, but were lower than those with NA2F-BNAH.
We also evaluated the inhibitory activity of the BNAH- and BACH-derivatives of NA2F towards PSA, knowing that the corresponding biantennary asparaginyl-glycopeptide inhibits hemagglutination at low concentration (Debray et al., 1981). As a reference compound we used the glycopeptide A2-Asn, which is ~20-fold less active as an inhibitor of PSA-mediated hemagglutination than the fucosylated analogue (Debray et al., 1981) which was not available for our investigations. Glycopeptide A2-Asn inhibited hemagglutination at 126 µM (Table IV). The free NA2F inhibited hemagglutination at 69 µM, and was 12-fold more active than the nonfucosylated oligosaccharide NA2 (852 µM). However, neither the BNAH nor the BACH derivative of NA2F was active as an inhibitor at the highest concentrations assayed, 158 µM and 206 µM, respectively. Thus, this class of biotinylated oligosaccharides are suitable for use as solid phase rather than liquid phase probes for carbohydrate-binding proteins. Table III
Inhibitor
Concentration giving inhibition (µM)
Expt. 1
Expt. 2
A2-Asn
126
126
NA2
852
NA2F
69
NA2-BNAH
>250
NA2F-BNAH
>158
NA2F-BACH
>206
Discussion
The BNAH derivatives have the advantage of having both UV absorbent/fluorescent properties (a feature of BAP derivatives) and an intact pyranose ring at the reducing-end (a feature of the BACH derivatives). With the preservation of the pyranose ring at the reducing-end, it would be possible, if desired, to regenerate the free oligosaccharides under mild acidic conditions (Lubineau et al., 1995). The BNAH derivatives can be prepared with ease by a one-step, one-pot reaction. Other technical advantages in their preparation are that the solvents used do not need to be anhydrous, and they are compatible with the solubility of oligosaccharides. As only a small excess (5 equivalents) of biotinylating reagent is needed, the purification of the oligosaccharide derivatives is relatively easy.
The solid phase lectin binding experiments with the four plant lectins investigated here indicate that BNAH-derivatives would be probes of choice for exploring the N-glycan binding specificities of novel carbohydrate-binding proteins for which different parts of the glycan sequence, including the unmodified reducing-end monosaccharide may be involved as recognition elements. With RCA120 which recognizes the outer-arms of the biantennary N-glycan investigated, the BNAH and BAP derivatives gave equivalent binding signals, whereas with ConA, LCA and PSA which recognize trimannosyl cores or reducing-ends of the oligosaccharides tested, the BNAH derivatives gave the strongest signals in the binding experiments. The minimum recognition structure, Man3 (Mandal et al., 1994a,b; Naismith and Field, 1996) elicited a binding signal with ConA when it was presented as the BNAH but not the BAP derivative. With LCA, and the more fastidious PSA, stronger binding signals were elicited using NA2F-BNAH than with the corresponding BACH derivative but no binding signals were elicited using NA2F-BAP. The strong interaction of PSA with NA2F-BNAH was unexpected as this lectin has been reported to require an asparaginyl as well as a 2-acetamido-2-deoxy-glucopyranose moiety for detectable binding (Yamamoto et al., 1982). The weaker signals elicited by the BACH derivative suggest that the BNAH derivative, containing a naphthyl alanine moiety, may 'substitute" for the asparaginyl residue on glycoproteins more effectively than the BACH derivative, containing an aminocaproyl group.
Although substantial PSA-binding signals were elicited by the biotinylated derivatives, NA2F-BNAH and NA2F-BACH, when these were immobilized on a streptavidin matrix, the derivatives were inactive as inhibitors of PSA binding at concentrations at which the original free oligosaccharide was inhibitory. The reasons for the negative effect of these aglycons on ligand recognition in solution are not known. The NMR data showed proximity of the naphthyl group of BNAH to the reducing-end monosaccharide. This is unlikely to be the sole explanation for the lack of inhibitory activity of the BNAH derivatives of NA2F as the BACH derivative which lacks the naphthyl group was also inactive as an inhibitor. It should be borne in mind that any impairment in ligand status resulting from the derivatizations would be readily manifest with the monomeric oligosaccharide derivatives free in solution, whereas with the immobilized derivatives impairment of binding affinity may be rendered less apparent by the multivalency.
The findings in the present investigation extend earlier observations (Lee, 1992; Feizi, 1993; Crocker and Feizi, 1996, and references therein) on the marked influence that the mode of presentation of oligosaccharides may have on their behavior as ligands for carbohydrate-binding proteins. A variety of aglycons are included in synthetic strategies aimed at creating oligosaccharide analogs that may mimic natural oligosaccharides, inhibit carbohydrate-binding and thus may serve as therapeutic agents (Liang et al., 1996). Knowledge that negative effects of the aglycon may be more noticeable when an oligosaccharide derivative is in solution, in a monomeric state, may explain differences such as those observed by Liang et al. (1996) in the relative potencies of immobilized and soluble forms of oligosaccharide derivatives in lectin interactions.
Materials and methods
Oligosaccharides and reagents
Chitobiose (GN2), N-acetyllactosamine (LacNAc), and Mannobiose (Man2) were from Sigma, Poole, UK. Mannotriose (Man3) and mannopentaose (Man5) were from Dextra Laboratories, Reading, UK. Man5GlcNAc2 and the biantennary structures NA2 and NA2F were from Oxford GlycoSystems, Abingdon, UK. Biotinyl-l-3-(2-naphthyl)-alanine hydrazide ((BNAH) part of the Glycan Receptor Binding Assay kit (750 nmol per kit)) was provided by Dr. M. Watzele (Boehringer-Mannheim). 2-Amino,6-amidobiotinyl-pyridine (BAP) was prepared as described previously by Toomre and Varki (1994). 6-(Biotinyl)-aminocaproyl-hydrazide (BACH) was purchased from Sigma, Poole, UK. High grade anhydrous pyridine and glacial acetic acid were purchased from Aldrich, Gillingham, UK. Concanavalin A (ConA)-digoxigenin, Ricinus communis agglutinin120 (RCA120)-digoxigenin, anti-digoxigenin-peroxidase-Fab fragments from Boehringer-Mannheim, Lewes, UK. Bovine serum albumin (BSA), 2,2[prime]-azinobis(3-ethylbenzthiazoline-sulfonic acid) (ABTS) and protein G-peroxidase were from Sigma, Poole, UK. Lens culinaris agglutinin (LCA), Pisum sativum agglutinin (PSA), and goat antibodies reactive with both LCA and PSA were from Vector Laboratories, Burlingame, CA.
Conjugation of oligosaccharides to biotinylated tags
For the preparation of BNAH derivatives which involves formation of glycosylhydrazides in a single step reaction, without reduction (Scheme 1), oligosaccharides (100 nmol) were mixed with BNAH (500nmol; 25 µl of 20 mM in MeOH), evaporated to dryness, and then dissolved in 25 µl of methanol/water/acetic acid, 74:8:8 v/v, or methanol/water, 9:1 v/v. Reaction mixtures were heated for 5 or 16 h at 60°C and evaporated to dryness under a nitrogen stream.
Scheme 1.
BAP derivatives (Scheme 1) were prepared according to Toomre and Varki (1994) with modifications as follows. In initial experiments, a range of concentrations of the reducing reagent, BDA (30, 100, 250, 400, and 1000 mM) were tried. The reactions were monitored by HP-TLC using UV light (300 nm) after acidifying the TLC plate to reveal BAP containing components, followed by orcinol staining to reveal hexose containing components. As concentrations of BDA were raised above 100 mM, increasing amounts of carbohydrate by-products were visible at the TLC origin (not shown). Below 100 mM, the reaction rate decreased such that incomplete conjugation resulted as assessed by HP-TLC analysis using UV and orcinol monitoring to identify product and disappearance of starting oligosaccharide. For the preparation of the six oligosaccharide-BAP derivatives, shown in Table II, the oligosaccharides (100 nmol) were mixed with BAP (1.5 µmol) in 10 µl of dry pyridine/acetic acid, 2:1 v/v, and heated for 1 h at 80°C. Oligosaccharide adducts were then reduced by mixing with 10 µl of 200 mM BDA in dry pyridine/acetic acid and heated for 1 h at 80°C and the mixtures evaporated to dryness.
BACH derivatives which involve the formation of glycosylhydrazides (Scheme 1) were prepared according to Shinohara et al. (1995) with the following modifications. Oligosaccharides (100 nmol) were mixed with BACH (500 nmol; 100 µl of 5mM in 30% aq. acetonitrile), evaporated to dryness, and then dissolved in methanol/water/acetic acid, 95:4:1 v/v, (25 µl). The reaction mixtures were heated overnight at 60°C and evaporated to dryness.
HP-TLC for monitoring conjugation reactions
Approximately 1 nmol of oligosaccharide was applied per lane. Reaction mixtures were chromatographed on Silica gel 60, high performance thin layer chromatography (HP-TLC) aluminum-backed plates from Merck developed with butan-1-ol/acetone/water, 6:5:4 v/v (BAP), or 2-butanone/methanol/water, 6:2:2 v/v (BNAH and BACH), to a height of 8.5 cm. Hexose containing components were visualized with orcinol stain. BAP and its derivatives were also visualized as blue fluorescent bands under UV light (300 nm), after exposure of the TLC surface to acetic acid vapor.
Chromatography for purification
To purify the oligosaccharide derivatives, the reaction mixtures were reconstituted in acetonitrile/water, 1:1 v/v, and fractionated on a 5 µ C18-silica reverse phase high performance liquid chromatography (RP-HPLC) column (250 × 4.6 mm) from Nucleosil. Solvent gradients were water to acetonitrile in 30 min as described in Table II at flow rates of 0.75, 1, and 0.5 ml/min. for BAP-, BNAH-, and BACH-derivatives, respectively, and UV detection at 250 nm, 275 nm, and 215 nm, respectively. UV-absorbing peaks were pooled and evaporated, and hexose containing derivatives were identified by orcinol spray after separation by HP-TLC.
Liquid secondary ion-mass spectrometry (LSIMS)
LSIMS analysis was carried out on a VG Analytical ZAB2-E mass spectrometer, fitted with a cesium ion gun operated at 35 keV and an emission current of 0.5 µA. Oligosaccharide-BNAH, -BAP and -BACH derivatives were dissolved in acetonitrile/water, 1:1 v/v, and 1-2 nmol were applied to the target precoated with liquid matrix thioglycerol (~1 µl). Full spectra were acquired in positive ion mode at 20 s decade-1 using the VG analytical VAX-Opus data system in continuum acquisition mode.
1H-NMR spectroscopy
Spectra were recorded on Bruker AM-400, Varian UNITY plus-500, and Varian UNITY-600. The samples were prepared by repeated dissolution/lyophilization from D2O. Finally, ~200 nmol of Lactosamine-BACH (LacNAc-BACH), Lactosamine-BNAH (LacNAc-BNAH), and Chitobiose-BNAH (GN2-BNAH) were dissolved in 0.6 ml of D2O. All data sets were recorded relative to internal acetone (2.225 ppm). 1D-NMR spectra were typically recorded using 6000 Hz sweep widths.
Lectin binding assays
Serial dilutions of biotinylated oligosaccharides were prepared in 10 mM Tris-HCl buffer, pH 8.0, 150mM NaCl (TBS). Two hundred microliters of each dilution were added to duplicate wells of high-capacity streptavidin-coated microtiter plates (Boehringer Mannheim, Germany) and incubated at 20°C for 4 h. In separate experiments (not shown) with the BNAH- and BAP-derivatives, uptake of the biotinyl-oligosaccharides onto the wells under these conditions was estimated to be greater than 90% as measured by fluorescence of the supernatants using a SLM 8000S fluorometer (at [lambda]ex 275 nm, [lambda]em 324 nm for BNAH and at [lambda]ex 345 nm, [lambda]em 400 nm for BAP). The wells were washed four times with TBS and treated at 20°C for 1 h with TBS containing 3% (w/v) BSA, and then washed twice with TBS. ConA-digoxigenin or RCA120-digoxigenin, 1 µg/ml, in TBS containing 0.5% BSA was added to the wells and incubated at 20°C for 1 h. The wells were washed with TBS and 200 µl of anti-digoxigenin-peroxidase-Fab fragments, 50mU/ml, in TBS containing 0.5% BSA were added and incubated at 20°C for 1 h. The wells were washed with TBS. Bound peroxidase was detected using the ABTS substrate and absorbance was read at 405 nm after 30 min. Absorbance values from control wells (i.e., wells to which lectins were added in the absence of biotinylated oligosaccharides) were typically in the range 0.20-0.30, and were subtracted from test readings.
For the binding assays with LCA and PSA, the streptavidin coated wells were incubated with biotinyl-oligosaccharides and quenched with BSA as described above. The lectins, 10 µg/ml, in TBS containing 0.5% BSA were added to the wells and incubated for 1 h at 20°C; wells were washed and goat antibodies to LCA/PSA, 10 µg/ml in TBS containing 0.5% BSA were added at 20°C and incubated for 1 h. Wells were washed and incubated at 20°C for 30 min with protein G-peroxidase, 10 µg/ml, in TBS containing 0.5% BSA, and bound peroxidase was detected as described above.
Hemagglutination assay
Agglutination of human red blood cells, group O, by PSA and inhibition of agglutination by free or biotinylated oligosaccharides or a glycopeptide was carried out in microtiter plates with U-bottom wells (3077 plates from Falcon, NJ). To each 10 µl of 2-fold dilution of sugar solution in PBS was added 50 µl of PSA solution (30 µg/ml) in PBS equivalent to four hemagglutinating doses. After incubation for 1 h at room temperature, 25 µl of 2% erythrocyte suspension in PBS was added. The mixture was incubated for 1 h at room temperature and agglutination was recorded by examining macroscopically the settling pattern. Results were expressed as the minimum concentration (µM) required to completely inhibit hemagglutination.
Acknowledgments
This work was supported by Program Grant E400/622 from the Medical Research Council, the Leukaemia Research Fund, and Boehringer-Mannheim.
Abbreviations
ABTS, 2,2[prime]-azinobis(3-ethylbenzthiazoline-sulfonic acid); BACH, 6-(biotinyl)-aminocaproyl-hydrazide; BAP, 2-amino-6-amidobiotinyl-pyridine; BDA, borane dimethylamine complex; BNAH, biotinyl-l-3-(2-naphthyl)-alanine hydrazide; BPH, 4-biotinamido-phenylacetylhydrazide; BSA, bovine serum albumin; Con A, concanavalin A; COSY, correlated spectroscopy; HP-TLC, high performance thin layer chromatography; LCA, Lens culinaris agglutinin; LSIMS, liquid secondary ion-mass spectrometry; PSA, Pisum sativum agglutinin; RCA120, Ricinus communis agglutinin120; ROESY, rotating-frame Overhauser effect spectroscopy; RP-HPLC, reverse phase high performance liquid chromatography; TBS, Tris buffered saline; TOCSY, total correlation spectroscopy.
References
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 4 Feb 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. J. Martin, T. Feizi, C. Leteux, D. Pavlovic, V. E. Piskarev, and W. Chai An investigation of the interactions of E-selectin with fuco-oligosaccharides of the blood group family Glycobiology, December 1, 2002; 12(12): 829 - 835. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Leteux, W. Chai, R. W. Loveless, C.-T. Yuen, L. Uhlin-Hansen, Y. Combarnous, M. Jankovic, S. C. Maric, Z. Misulovin, M. C. Nussenzweig, et al. The Cysteine-Rich Domain of the Macrophage Mannose Receptor Is a Multispecific Lectin That Recognizes Chondroitin Sulfates a and B and Sulfated Oligosaccharides of Blood Group Lewisa and Lewisx Types in Addition to the Sulfated N-Glycans of Lutropin J. Exp. Med., April 3, 2000; 191(7): 1117 - 1126. [Abstract] [Full Text] [PDF]
|
|