The initial sequencing of immunoglobulins and -microglobulin implied that these molecules were formed by individual protein structural units that evolved from a single primordial domain of about 100 amino acids (reviewed in (1) ). The immunoglobulin superfamily (IgSF) ()concept arose from the discovery that sequences characteristic of these domains are also present in molecules without an immunological function(1) . Structural studies of immunoglobulins(2) , major histocompatibility complex class I(3) , and class II antigens(4) , CD4(5, 6) , CD8(7) , and CD2 (8, 9, 10) have revealed that the IgSF fold consists of a sandwich of two anti-parallel -sheets stabilized in some instances by a conserved disulfide bond. The conserved patterns of sequence characteristic of IgSF domains are generally limited to -strand residues forming the hydrophobic core of the domain (1) which appears to be responsible for the strict conservation of the three-dimensional structure of these domains(10) .

A recent survey of the leucocyte surface has indictated that 36% of leucocyte antigens belong to the IgSF suggesting that the IgSF forms the largest single family of molecules present on the cell surface (11) . The ligand interactions of the cell-cell recognition molecule CD2 are among the best characterized of those involving IgSF cell surface molecules. In humans and rodents the ligands for CD2 are CD58(12, 13) and CD48(14, 15) , respectively. Along with CD2, CD48 and CD58 form a subset of molecules within the IgSF that also includes the carcinoembryonic antigens(16) , Ly-9 (17) and 2B4(18) . The clustering of the CD2, CD48, and CD58 genes in the genomes of humans and mice implies that CD2, CD48, and CD58 have all evolved from a common precursor involved in homophilic interactions(19) . The crystal structures of rat sCD2 (9) and human sCD2 (hsCD2) (8) have revealed that the extracellular region of CD2 consists of two IgSF domains: an NH-terminal V-set domain and a C2-set domain. Mutational analyses of CD2 (20, 21, 22, 23) established that the ligand binding site is located on the GFCC`C" face of the -sheet of the V-set domain. The highly conserved linker region seen in the sCD2 structures places this relatively flat, highly charged face at the membrane-distal surface of the molecule(8, 9) . The interactions of rat and human CD2 with their respective ligands, CD48 (24) and CD58(25) , are characterized by relatively fast on-rates and very fast off-rates which, together with the structural data, suggest that the binding of CD2 with its ligands is not dependent on large conformational changes.

The role of glycosylation in stabilizing IgSF domains is controversial. While many IgSF molecules are N-glycosylated, the extent of glycosylation varies considerably and glycosylation sites are generally not conserved, even between species homologues. The ligand binding function of rat CD2, which has four N-glycosylation sites, is not glycosylation-dependent(15, 24) . In contrast, it has been suggested that ligand binding by human CD2, which has three N-glycosylation sites, is glycosylation-dependent(26) . This observation has significant implications given that the evolutionary success of the IgSF, and the high level of conservation of the IgSF fold, are both thought to reflect the ability of IgSF protein domains to form stable structural units for the presentation of receptor-ligand recognition motifs(1, 27) . In the present study a systematic analysis of the effect of glycosylation on the ligand binding properties of CD2 has been undertaken. The data indicate that the structural integrity and ligand binding function of human CD2 are not glycosylation dependent.

MATERIALS AND METHODS

Protein Expression

After preclearing the spent tissue culture medium at 10,000 g for 30 min, the hsCD2 was purified by affinity chromatography according to published methods (28) using an antibody affinity column prepared with the anti-CD2 monoclonal antibody, (mAb) X/3. Final purification involved gel filtration on Sephacryl S-200 in 10 mM Hepes, 140 mM NaCl, pH 7.4.

Deglycosylation and Crystallization

For large-scale endo H treatment of hsCD2 for crystallization experiments, 4 mg of the purified glycoprotein were concentrated to 1-2 mg/ml in 0.1 M sodium acetate, pH 5.2, and then digested with 0.1 I.U.B. units/ml endo H overnight at 37 °C. To purify the endo H-treated hsCD2 from the contaminating endo H-resistant fraction, the protein mixture was concentrated to 0.5 ml and then passed through a 5-ml Sephadex G-50 column to remove free oligosaccharides. The eluate was then passed through a 15-ml lectin affinity column consisting of equal parts of lentil lectin, concanavalin A and wheat germ agglutinin, each coupled to Sepharose 4B (Sigma). The homogeneity of the hsCD2 was then confirmed by SDS-PAGE on a 15% acrylamide gel. The lectin purified protein was concentrated to 2 ml and then applied to Sephadex G-75 in 10 mM Hepes, 140 mM NaCl, pH 7.4, to remove free lectin eluting from the lectin-affinity column. The deglycosylated hsCD2, in 10 mM Hepes, 140 mM NaCl, pH 7.4, was concentrated to 17 mg/ml. Crystals were grown by vapor diffusion in sitting drops at room temperature. Initial trials were conducted using Crystal Screen reagents (Hampton Research).

For large scale preparation of fully deglycosylated protein, hsCD2 (at 600 µg/ml) purified from untreated cultures was digested with peptide:N-glycosidase F (PNGase F; New England BioLabs) at 0.085 I.U.B. units/ml in 0.5 M Tris, pH 8. Deglycosylated protein was then purified by gel filtration on a Superdex G-75 fast protein liquid chromatography system (Pharmacia Biotech Inc.).

Carbohydrate Analyses and Mass Spectrometry

Binding Experiments

Figure 6: Crystallization of hsCD2 and diffraction analysis of crystals. Crystals of purified endo H-treated hsCD2 were grown by vapour diffusion against 1.25 M sodium citrate, 0.1 M Hepes, pH 7.4. Typical crystals are shown in Panel A; the largest crystals from 2-4-µl drops grew to 0.6 mm along the longest axis. In Panel B a typical two-pass 1.5° oscillation image taken at Daresbury Synchrotron Radiation Source (line 9.5, = 0.999 Å, T = 20 °C) is shown. The crystal to film distance was 212.2 mm. The edge of the image corresponds to diffraction to Bragg spacings of 2.5 Å. The diffraction pattern was displayed with the program PSIMAGE (R. Esnouf, Oxford).

The equilibrium binding data (see Fig. 3) were analyzed by 1) nonlinear curve fitting of the Langmuir binding isotherm to the primary data and 2) linear curve fitting of the Scatchard plots. The dissociation phases (see Fig. 4) were analyzed by first normalizing them so that the response before dissociation was 100% and the base line response was 0%. Dissociation rate constants (k) for each dissociation phase were then determined by fitting mono- or bi-exponential decay functions to the data (see Fig. 4). All curve fitting was performed using the curve-fitting functions of the program Origin version 2 (MicroCal Software Inc, Northampton, MA) which was run on a Compaq PC. Linear curve fitting was by linear least squares regression. Nonlinear curve fitting employed iterative least squares curve fitting using the Marquardt-Levenberg algorithm.

Figure 3: Measurement of the affinity of sCD58 binding to untreated, endo H-treated and PNGase F-treated hsCD2. Panel A, sCD58 was injected for 6 s at the indicated concentrations over a flow cell with immobilized untreated (open circles) or endo H-treated (closed circles) hsCD2. The levels of untreated and endo H-treated hsCD2 immobilized were 3160 and 4983 response units (RU), respectively. At saturation, 39 and 34% of the immobilized untreated and endo H-treated hsCD2 had bound sCD58, respectively. Panel B, untreated or PNGase F-treated hsCD2 were injected for 6 s at the indicated concentration over a flow cell with sCD58 immobilized. The level of immobilization of sCD58 was 10,863 RU. The equilibrium binding levels shown in Panels A and B were calculated as described elsewhere (24) by subtracting the response obtained when the same sCD58 and hsCD2 samples are injected over a control flowcell with nothing immobilized. The flow rate was 20 µlmin. Insets, Scatchard plots of the binding data. The K values were determined by linear-regression analysis of the Scatchard plots as well as by nonlinear curve fitting of the saturation binding curve. Both methods gave the same K values.

Figure 4: Comparison of the rates of dissociation of sCD58 from untreated hsCD2, endo H-treated hsCD2, and PNGase F-treated hsCD2. Panel A, the dissociation of sCD58 (0.22 mg/ml) from immobilized untreated (open triangles, 3160 RU) or endo H-treated (filled triangles, 4983 RU) hsCD2. The fall in response following injection of sCD58 over a flow cell with nothing immobilized is also shown (dotted line). The equilibrium responses (100%) following injection of sCD58 over nothing, untreated hsCD2 and endo H-treated hsCD2 were 92, 373, and 705 RU, respectively. Panel B, dissociation of untreated hsCD2 (open triangles, 0.9 mg/ml) or PNGase F-treated hsCD2 (closed triangles, 1 mg/ml) from immobilized sCD58 (10,863 RU immobilized). The fall in response following injection of untreated (open triangles) and PNGase F-treated (closed triangles) hsCD2 over a flow cell with nothing (dotted line) immobilized is shown. The equilibrium responses (100%) following injection of untreated and PNGase F-treated hsCD2 over immobilized sCD58 were 850 and 1193 RU, respectively. The same samples injected though a control flow cell gave responses of 161 and 317 RU. The samples were injected at flow rates of 20 µl/min. The apparent dissociation times were obtained by fitting mono-exponential decay curves to the data (see dotted and solid lines). One exception was the dissociation of PNGase F-treated hsCD2 from immmobilized sCD58 which was fitted using a bi-exponential decay function.

RESULTS

Effects of NB-DNJ and Digestion of hsCD2 with Endo H

Figure 1: Effect of NB-DNJ on endo H sensitivity of hsCD2 oligosaccharides. Soluble CD2 was expressed in the presence of 0, 0.5, 1.0, 1.5, or 2 mM NB-DNJ, purified to homogeneity and then digested overnight with endo H at 0.012, 0.06 or 0.3 I.U.B. units/mg of hsCD2. The digestion products (3 µg) were then electrophoresed with undigested hsCD2 on a 15% SDS-PAGE gel alongside equivalent amounts of the starting material for each digestion. The gel was then stained with Coomassie Blue.

Digestion of the 2 mM NB-DNJ-treated hsCD2 with limiting amounts of endo H produced three smaller products each differing by 2-3 kDa (Fig. 1, lane 18). These are likely to correspond to hsCD2 forms bearing zero, one, or two oligosaccharides indicating that all three glycosylation sites can be utilized. There was a concomitant increase in sensitivity to endo H with increasing NB-DNJ concentrations. However, at the two highest NB-DNJ concentrations there was little difference in the endo H sensitivity. Thus, 1.5 mM NB-DNJ appears to completely inhibit glucosidase I. Densitometric analysis of the gel revealed that at the highest NB-DNJ concentration 15% of the hsCD2 is endo H-resistant indicating that a relatively inefficient glucosidase I bypass mechanism exists in these cells (data not shown). Some of the N-linked glycosylation of hsCD2 expressed in the absence of NB-DNJ was also sensitive to endo H. The 2-3-kDa change in SDS-PAGE mobility is consistent with the presence of a single unprocessed site on hsCD2 expressed in CHO cells as observed previously(26) . This contrasts with rat sCD2 expressed in CHO cells which is completely endo H-resistant(31) .

Milligram quantities of hsCD2 were produced in the presence of NB-DNJ, purified and treated with endo H. The digested material was then purified to homogeneity by lectin affinity chromatography and gel filtration chromatography for functional studies and crystallization trials (Fig. 2, lane 2). The electrophoretic mobility of the endo H-treated hsCD2 suggested that the oligosaccharides had been removed from each glycosylation site. Using methods which quantitatively release amino sugars (incubation with 6 N HCl at 100 °C for 4 h), 3.36 ± 0.31 and 14.79 ± 0.32 mol of GlcNAc/mol of protein were detected in endo H-treated and untreated hsCD2, respectively. Mass spectrometric analysis of the endo H-treated hsCD2 gave a mass expected for the polypeptide backbone with three GlcNAc residues (21, 575 Daltons; data not shown). This data indicates that all three glycosylation sites are fully utilized in CHO cells, and that endo H truncates the oligosaccharides at each site to single GlcNAc residues.

Figure 2: Gel electrophoretic analysis of the purified endo H and PNGase F treated hsCD2. Three micrograms of untreated hsCD2 (lane 1), 3 µg of hsCD2 expressed in the presence of 1.5 mM NB-DNJ that had been endo H-treated and lectin purified (lane 2) and 3 µg of PNGase F-treated hsCD2 that had been purified by gel filtration chromatography (lane 3) were electrophoresed on a 15% SDS-PAGE gel. The gel was then stained with Coomassie Blue.

Ligand Binding Properties of Endo H-treated hsCD2

The binding affinities of sCD58, prepared as described elsewhere(25) , for untreated hsCD2 and for endo H-treated hsCD2 prepared after expression in the presence of NB-DNJ, were measured by equilibrium binding analysis on the BIAcore, as described previously (24) . Endo H-treated hsCD2 was compared with untreated hsCD2 by injecting sCD58 at increasing concentrations through flow cells in which either untreated or endo H-treated hsCD2 had been immobilized (Fig. 3A). sCD58 clearly binds endo H-treated hsCD2 and untreated hsCD2 with similar affinities (Fig. 3A). Scatchard plots of the data (Fig. 3A, inset) indicated that the affinities of sCD58 for untreated and endo H-treated hsCD2 were K 7 µM and 9 µM, respectively.

While the affinity of endo H-treated hsCD2 for sCD58 was essentially unchanged, this does not rule out the possibility that the kinetics of binding are altered by glycosidase treatment. Very low affinity interactions, such as those between rat CD2 (24) and human CD2 (25) and their respective ligands, reach equilibrium very rapidly, making kinetic analysis of the association phase impossible on the BIAcore. However, the dissociation phase of such interactions can usually be analyzed, and so the dissociation rates of untreated and endo H-treated hsCD2 were compared (Fig. 4A). It has previously been shown that the apparent dissociation rate constant (k) observed on the BIAcore is slower than the actual k because 1) it takes time for the dissociated protein to wash out of the flow cell and 2) rebinding can occur during this washing phase(24, 25) . As a result, the measured k values shown in this study represent a lower limit for the actual k and are used here for comparative purposes. Under identical experimental conditions sCD58 dissociated from untreated and endo H-treated hsCD2 at about the same rate (with observed values of k 0.8 s and 0.7 s, respectively). In both cases the shape of the dissociation curve was mono-exponential, consistent with the bound sCD58 having a single k (Fig. 4A). A similar result was obtained with endo H-treated hsCD2 binding to immobilized sCD58 (not shown).

Antibody Binding Properties of Endo H-treated hsCD2

Figure 5: Binding of anti-CD2 mAbs to untreated, endo H-treated, and PNGase F-treated hsCD2. Panel A, ascites (1 in 20 dilution) containing the indicated human CD2 mAbs was injected for 6 min through flow cells with immobilized untreated (2678 RU) or endo H-treated (3502 RU) hsCD2. Bound mAbs were eluted with 10 mM HCl (arrow) prior to injection of the subsequent mAb. The flow rate was 5 µlmin. Panel B, ascites (1 in 50 dilution) containing the indicated human CD2 mAb was injected for 10 min through flow cells in which untreated hsCD2 (4689 RU), endo H-treated hsCD2 (5955 RU), and PNGase F-treated hsCD2 (5568 RU) were immobilized. The flow rate was 3 µlmin.

Crystallization and Structural Analysis of Endo H-treated hsCD2

The ligand and antibody-binding data alone do not rule out the possibility that the single GlcNAc left by endo H digestion directly stabilizes the ligand binding face of hsCD2. However, the location of the GlcNAc residue on domain 1 of the hsCD2 crystal structure indicates that the monosaccharide is unlikely to stabilize the protein structure in general and the spatially distant GFCC`C" face in particular (Fig. 7).

Figure 7: Location of the domain 1 GlcNAc residue in the crystal structure of endo H-treated hsCD2. Domain 1 is shown in space-filling format with the line of view parallel to the GFCC`C`` and DEBA faces. The side chains of residues previously shown to form part of the CD58 binding site (left) and the GlcNAc residue left by endo H treatment (right) are shaded grey (the mutagenesis of human CD2 is discussed in detail by Bodian et al.(8) ).

Properties of PNGase F-treated hsCD2

PNGase F-treated hsCD2 could not be coupled at high levels to the dextran matrix of the BIAcore flow cell (not shown) and so its affinity for sCD58 was determined in the opposite orientation, with sCD58 immobilized. When untreated or PNGase F-treated hsCD2 were injected over sCD58 they both bound with a similar affinity (Fig. 3B). The affinities determined from a Scatchard plot were K 3 and 3.2 µM for untreated and PNGase F-treated hsCD2, respectively (Fig. 3B, inset). It should be noted that the affinities measured in this study (3-9 µM) were obtained at 25 °C for comparative purposes and are slightly higher than the affinity at 37 °C (22 µM) obtained previously for the interaction of sCD2 and sCD58(25) .

In contrast to untreated and endo H-treated hsCD2, PNGase F-treated hsCD2 dissociated from immobilized sCD58 in two phases (Fig. 4B), with fast initial dissociation (k 0.8 s) and then some slow dissociation (k 0.16 s), suggesting that a proportion of the PNGase F-treated hsCD2 has a high avidity. This high avidity binding is specific since it is not seen when PNGase F-treated hsCD2 is injected over 1) a control flow-cell with no immobilized protein (Fig. 4B) or over 2) immobilized CD58 which had been pre-saturated with the inhibitory mAb TS2/9 (data not shown). The observation that PNGase F-treated hsCD2 has a tendency to aggregate in solution during purification (see above) suggests that this high avidity binding represents the binding of multimeric aggregates of the PNGase F-treated hsCD2. Although the immobilization of PNGase F-treated hsCD2 was difficult and relatively inefficient, the binding of the regions 1-, 2-, and 3-reactive mAbs, T11/3PT2H9, TS1/8.1.1, and OCH.217 was tested and each was shown to bind to the immobilized PNGase F-treated protein (Fig. 5D).

DISCUSSION

The results of this study indicate that the structural integrity of human CD2 is glycosylation independent. First, the affinity of two-domain hsCD2 for its ligand, CD58, and the kinetics of this interaction are not significantly affected by truncation of the oligosaccharides to single GlcNAc residues with endo H. Second, the binding of ligand-blocking and other mAbs to endo H-treated hsCD2 is essentially indistinguishable from the binding of the same mAbs to untreated hsCD2. The antibody epitopes of protein antigens with significant secondary structure are usually formed by discontinuous polypeptide segments (34) and thus antibody binding can generally be considered to be good evidence for the correct folding and conformation of protein antigen derivatives if the antibodies also recognize the correctly-folded native antigen. Third, since local or global losses in structural integrity would be expected to prevent the crystallization of hsCD2 as this depends on the formation of stable, reproducible lattice contacts, the crystallization of the endo H-treated hsCD2 implies that the protein is not destabilized by oligosaccharide truncation. It is not inconceivable, however, that the endo H-treated hsCD2 adopts a new configuration that is sufficiently stable to crystallize. This possibility seems unlikely given that the crystallographic analysis has shown that the endo H-treated hsCD2 consists of two domains with conventional IgSF folds(8) . The structural analysis also indicates that the apparent stability of the ligand binding face is unlikely to be due to any stabilizing effect of the single GlcNAc residue located on the DE loop of domain 1. Finally, hsCD2 fully deglycosylated with PNGase F bound CD58 with an affinity similar to that of untreated hsCD2 and bound a series of anti-CD2 mAbs. The slightly reduced dissociation rate for the binding of PNGase F-treated hsCD2 to immobilized CD58 is more likely to be due to aggregation of a fraction of the deglycosylated hsCD2 molecules during the experiment than to disruption of the structure per se. The stability of human CD2 in the absence of glycosylation is consistent with the view that the evolutionary success of the IgSF reflects the ability of IgSF domains to form stable structural units for the presentation of protein recognition motifs.

The results of this study contrast with those of Recny et al.(26) who have proposed that the ligand binding function and stability of CD2 are glycosylation-dependent. The conclusions of that study were based on two observations. First, deglycosylated CD2 domain 1, produced by proteolysis of a two domain form of hsCD2 expressed in Chinese hamster ovary cells, fails to bind CD58 and anti-CD2 mAbs. Second, the removal of the domain 1 glycosylation site at Asn-65 by mutagenesis also prohibits ligand and mAb binding to cell-surface expressed CD2. As an explanation for these observations, Withka et al.(35) and Wyss et al.(36) have suggested that the Asn-65 oligosaccharide interacts with and stabilizes residues surrounding the GFCC`C" face involved in ligand binding by filling a cavity between the BC, C`C", and FG loops at the top of domain 1.

The present experiments do not provide an explicit explanation for the contrary data of Recny et al.(26) . The first observation, that in the absence of domain 2 deglycosylated CD2 domain 1 loses its ligand and antibody binding properties, is not consistent with studies of rat CD2 which have shown that domain 1 of the rat homologue retains its ligand and mAb binding properties when expressed in bacteria in an unglycosylated state and in the absence of domain 2(15, 37) . Inspection of the crystal structures of human and rat CD2 fails to reveal any obvious differences between rat and human CD2 domain 1 that could account for the differing stabilities of the isolated unglycosylated domains. The B factors for residues of domain 1 in the human and rat sCD2 crystal structures, which give some insight into the overall stability of the domain, are not significantly different for the two homologues(8) . The PNGase F-treated hsCD2 used in the present studies did tend to aggregate as discussed above, and it is conceivable that this is exacerbated by proteolysis. However, the interface residues located between domains 1 and 2 of human and rat CD2 are very highly conserved(8) , suggesting that exposure of this interface in human CD2 is unlikely to reduce its stability any more than exposure of the same region of rat CD2 domain 1. Finally, the side chain of Asn-65 and the GlcNAc moiety attached to this residue project toward the interface region of the molecule (data not shown); assuming that this reflects the orientation of the intact oligosaccharide, there are no conspicuous hydrophobic residues that are likely to be shielded by the intact oligosaccharide and exposed by deglycosylation. Irrespective of the underlying cause of these differences, the two domain form of sCD2 used in the present study more closely resembles the natural state of CD2 than does the proteolytic fragment studied by Recny et al. It therefore seems reasonable to conclude that the in vitro behavior of two domain sCD2 more faithfully reflects the structural properties of the molecule in vivo.

It is less difficult to explain the second observation that the ligand and antibody binding properties of human CD2 expressed at the cell surface are disrupted by the mutation of Asn-65 Gln(26) . It is now a widely held view that the folding and stability of glycoproteins is, in many cases, glycosylation dependent(38) . Much of the evidence supporting this view has been obtained from experiments in which glycosylation has been blocked with inhibitors such as tunicamycin or by mutation of glycosylation sites. In many instances it is clear that under these conditions the unglycosylated proteins do not leave the rough endoplasmic reticulum and are degraded (see, for example, (39, 40, 41, 42) ). It can also readily be envisaged that bulky, hydrophilic oligosaccharides might influence protein folding pathways by limiting the number of alternative kinetically accessible conformations or by reducing their thermodynamic stability. Work in this laboratory on T cell receptor/CD4 chimeras has also shown that incorrectly folded proteins can in some instances emerge from the endoplasmic reticulum as indicated by the tendency of the chimeras to form disulfide-bonded aggregates outside the cell(43) . Also, major histocompatibility class I and II antigens reach the surface of transfected cells in the absence of -microglobulin and invariant chain, respectively, and these antigens both fail to bind conformationally sensitive antibodies(44, 45) . Consistent with the observations of Recny et al.(26) >90% of human CD2 domain 1 expressed in an unglycosylated state in Escherichia coli forms large aggregates and the remaining material appears to be incorrectly folded according to antibody binding assays(37) . Thus, the inability of the cell-surface expressed Asn-65 mutant of CD2 to mediate antibody and ligand binding (26) may be due to incorrect folding in the endoplasmic reticulum and its subsequent transport to the surface in an inappropriately folded, or unfolded, state.

These studies of deglycosylated CD2 emphasize the difficulty of distinguishing, on the basis of mutational and inhibitor studies, between any effects of glycosylation on protein folding and any role in maintaining protein conformation once folding is complete. The post-folding, structural role of glycosylation has not thus far been explored in great detail presumably because it is relatively difficult to deglycosylate glycoproteins under non-denaturing conditions in vitro. Structural comparisons of naturally occurring glycosylated and unglycosylated forms of bovine pancreatic RNase by crystallographic (46) and solution H NMR analyses (47, 48) indicate that glycosylation has little effect on the overall conformation of the enzyme although studies of hydrogen-deuterium solvent exchange rates have indicated that the oligosaccharide enhances the global dynamic stability of the protein(47, 49) . Conversely, structural studies of glycoproteins and glycopeptides suggest that N-linked oligosaccharides are largely unaffected by the presence of the protein (48, 50, 51) . Molecular dynamic simulations of the interaction of the oligosaccharide of RNase B with the polypeptide backbone suggest that while the di-N-acetylchitobiose core is relatively rigid, flexibility in the linkages to outer arm residues of the oligosaccharide and in the asparagine-GlcNAc linkage allow the oligosaccharide potentially to contact relatively large areas of the protein surface(52) . In toto, these studies suggest that, while oligosaccharides may significantly influence the functions of glycoproteins, this is probably likely to be due in most cases to steric effects, and that oligosaccharides forming intimate contacts with the protein backbone which influence the polypeptide conformation are likely to be rare. CD2 may be unusual in that nuclear Overhauser effects involving a terminal mannose residue(s) of the Asn-65 oligosaccharide and Gly-90 of domain 1 have been tentatively identified (35, 36) . While the current study does not rule out the possibility that the domain 1 oligosaccharide interacts directly with CD2 in this way in vivo, the weight of evidence is against the view that such interactions affect the conformation of the domain or are important for ligand binding.

FOOTNOTES

*

This work was supported by the Human Frontier Science Program. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked ``advertisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§

To whom correspondence should be addressed. Tel.: 0865-275597; Fax: 0865-275591.

¶

P. A. van der Merwe is an Oxford Nuffield Medical Fellow.

()

The abbreviations used are: IgSF, immunoglobulin superfamily; endo H, endoglycosidase H; PNGase F, peptide:N-glycosidase F; NB-DNJ, N-butyldeoxynojirimycin; GlcNAc, N-acetylglucosamine; CHO, Chinese hamster ovary; PAGE, polyacrylamide gel electrophoresis; RU, response unit(s); hsCD2, human soluble CD2; mAb, monoclonal antibody.

ACKNOWLEDGEMENTS

REFERENCES

1. Williams, A. F., and Barclay, A. N. (1988) Annu. Rev. Immunol. 6, 381-405 [Medline] [Order article via Infotrieve]
2. Amzel, L. M., and Poljak, R. J. (1979) Annu. Rev. Biochem. 48, 961-997 [CrossRef][Medline] [Order article via Infotrieve]
3. Bjorkman, P. J., Saper, M. A., Samraoui, B., Bennett, W. S., Strominger, J. L., and Wiley, D. C. (1987) Nature 329, 506-512 [CrossRef][Medline] [Order article via Infotrieve]
4. Brown, J. H., Jardetzky, T. S., Gorga, J. C., Stern, L. J., Urban, R. G., Strominger, J. L., and Wiley, D. C. (1993) Nature 364, 33-39 [CrossRef][Medline] [Order article via Infotrieve]
5. Ryu, S.-E., Kwong, P. D., Truneh, A., Porter, T. G., Arthos, J., Rosenberg, M., Dai, X., Xuong, N.-H., Axel, R., Sweet, R. W., and Hendrickson, W. A. (1990) Nature 348, 419-426 [CrossRef][Medline] [Order article via Infotrieve]
6. Wang, J., Yan, Y., Garrett, T. P. J., Liu, J., Rodgers, D. W., Garlick, R. L., Tarr, G. E., Husain, Y., Reinherz, E. L., and Harrison, S. C. (1990) Nature 348, 411-418 [CrossRef][Medline] [Order article via Infotrieve]
7. Leahy, D. J., Axel, R., and Hendrickson, W. A. (1992) Cell 68, 1145-1162 [CrossRef][Medline] [Order article via Infotrieve]
8. Bodian, D. L., Jones, E. Y., Harlos, K., Stuart, D. I., and Davis, S. J. (1994) Structure 2, 755-766 [Medline] [Order article via Infotrieve]
9. Jones, E. Y., Davis, S. J., Williams, A. F., Harlos, K., and Stuart, D. I. (1992) Nature 360, 232-239 [CrossRef][Medline] [Order article via Infotrieve]
10. Driscoll, P. C., Cyster, J. G., Campbell, I. D., and Williams, A. F. (1991) Nature 353, 762-765 [CrossRef][Medline] [Order article via Infotrieve]
11. Barclay, A. N., Birkeland, M. L., Brown, M. H., Beyers, A. D., Davis, S. J., Somoza, C., and Williams, A. F. (1992) The Leucocyte Antigen Facts Book 1st Ed., pp. 13-37, Academic Press, New York
12. Dustin, M. L., Sanders, M. E., Shaw, S., and Springer, T. A. (1987) J. Exp. Med. 165, 677-692 [Abstract/Free Full Text]
13. Selvaraj, P., Plunkett, M. L., Dustin, M., Sanders, M. E., Shaw, S., and Springer, T. A. (1987) Nature 326, 400-403 [CrossRef][Medline] [Order article via Infotrieve]
14. Kato, K., Koyanagi, M., Okada, H., Takanashi, T., Wong, Y. W., Williams, A. F., Okumura, K., and Yagita, H. (1992) J. Exp. Med. 176, 1241-1249 [Abstract/Free Full Text]
15. van der Merwe, P. A., McPherson, D. C., Brown, M. H., Barclay, A. N., Cyster, J. G., Williams, A. F., and Davis, S. J. (1993) Eur. J. Immunol. 23, 1373-1377 [Medline] [Order article via Infotrieve]
16. Killeen, N., Moessner, R., Arvieux, J., Willis, A., and Williams, A. F. (1988) EMBO J. 7, 3087-3091 [Medline] [Order article via Infotrieve]
17. Sandrin, M. S., Gumley, T. P., Henning, M. M., Vaughan, H. A., Gonez, L. J., Trapani, J. A., and McKenzie, I. F. C. (1992) J. Immunol. 149, 1636-1641 [Abstract]
18. Mathew, P. A., Garni-Wagner, B. A., Land, K., Takashima, A., Stoneman, E., Bennett, M., and Kumar, V. (1993) J. Immunol. 151, 5328-5337 [Abstract]
19. Wong, Y. W., Williams, A. F., Kingsmore, S. F., and Seldin, M. F. (1990) J. Exp. Med. 171, 2115-2130 [Abstract/Free Full Text]
20. Peterson, A., and Seed, B. (1987) Nature 329, 842-846 [CrossRef][Medline] [Order article via Infotrieve]
21. Somoza, C., Driscoll, P. C., Cyster, J. G., and Williams, A. F. (1993) J. Exp. Med. 178, 549-558 [Abstract/Free Full Text]
22. Arulanandam, A. R. N., Withka, J. M., Wyss, D. F., Wagner, G., Kister, A., Pallai, P., Recny, M. A., and Reinherz, E. L. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 11613-11617 [Abstract/Free Full Text]
23. Wolff, H. L., Burakoff, S. J., and Bierer, B. E. (1990) J. Immunol. 144, 1215-1220 [Abstract]
24. van der Merwe, P. A., Brown, M. H., Davis, S. J., and Barclay, A. N. (1993) EMBO J. 12, 4945-4954 [Medline] [Order article via Infotrieve]
25. van der Merwe, P. A., Barclay, A. N., Mason, D. W., Davies, E. A., Morgan, B. P., Tone, M., Krishnam, A. K. C., Ianelli, C., and Davis, S. J. (1994) Biochemistry 33, 10149-10160 [CrossRef][Medline] [Order article via Infotrieve]
26. Recny, M. A., Luther, M. A., Knoppers, M. H., Neidhardt, E. A., Khandekar, S. S., Concino, M. F., Schimke, P. A., Francis, M. A., Moebius, U., Reinhold, B. B., Reinhold, V. N., and Reinherz, E. L. (1992) J. Biol. Chem. 267, 22428-22434 [Abstract/Free Full Text]
27. Williams, A. F. (1987) Immunol. Today 8, 298-303 [CrossRef]
28. Arvieux, J., and Williams, A. F. (1988) Antibodies: A Practical Approach , pp. 113-136, IRL Press, Oxford
29. Seed, B., and Aruffo, A. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 3365-3369 [Abstract/Free Full Text]
30. Karlsson, G. B., Butters, T. D., Dwek, R. A., and Platt, F. M. (1993) J. Biol. Chem. 268, 570-576 [Abstract/Free Full Text]
31. Davis, S. J., Puklavec, M. J., Ashford, D. A., Harlos, K., Jones, E. Y., Stuart, D. I., and Williams, A. F. (1993) Protein. Eng. 6, 229-232 [Free Full Text]
32. Davis, S. J., Jones, E. Y., Bodian, D. L., Barclay, A. N., and van der Merwe, P. A. (1993) Biochem. Soc. Trans. 21, 952-958 [Medline] [Order article via Infotrieve]
33. Jönsson, U., Fagerstam, L., Ivarsson, B., Johnsson, B., Karlsson, R., Lundh, K., Lofas, S., Persson, B., Roos, H., Ronnberg, I., Sjolander, S., Stenberg, E., Stahlberg, R., Urbaniczky, C., Ostlin, H., and Malmqvist, M. (1991) BioTechniques 11, 620-627 [Medline] [Order article via Infotrieve]
34. Laver, W. G., Air, G. M., Webster, R. G., and Smith-Gill, S. J. (1990) Cell 61, 553-556 [CrossRef][Medline] [Order article via Infotrieve]
35. Withka, J. M., Wyss, D. F., Wagner, G., Arulanandam, A. R. N., Reinherz, E. L., and Recny, M. A. (1993) Structure 1, 69-81 [Medline] [Order article via Infotrieve]
36. Wyss, D. F., Withka, J. M., Knoppers, M. H., Sterne, K. A., Recny, M. A., and Wagner, G. (1993) Biochemistry 32, 10995-11006 [CrossRef][Medline] [Order article via Infotrieve]
37. Cyster, J. G. (1992) Ph.D thesis, Leukosialin and CD2: Contrasting Structures at the Leukocyte Surface, Oxford University _
38. Varki, A. (1993) Glycobiology 3, 97-130 [Abstract/Free Full Text]
39. Benjannet, S., Rondeau, N., Paquet, L., Boudreault, A., Lazure, C., Chrétien, M., and Seidah, N. G. (1993) Biochem. J. 294, 735-743
40. Ciccarelli, E., Alonso, M. A., Cresteil, D., Bollen, A., Jacobs, P., and Alvarez, F. (1993) Eur. J. Biochem 213, 271-276 [Medline] [Order article via Infotrieve]
41. Loch, N., Tauber, R., Becker, A., Hartel-Schenk, S., and Reutter, W. (1992) Eur. J. Biochem 210, 161-168 [Medline] [Order article via Infotrieve]
42. Yang, B., Hoe, M. H., Black, P., and Hunt, R. C. (1993) J. Biol. Chem. 268, 7435-7441 [Abstract/Free Full Text]
43. Brown, M. H., and Barclay, A. N. (1994) Protein Eng. 7, 515-521 [Abstract/Free Full Text]
44. Anderson, M. S., and Miller, J. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 2282-2286 [Abstract/Free Full Text]
45. Allen, H., Fraser, J., Flyer, D., Calvin, S., and Flavell, R. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 7447-7451 [Abstract/Free Full Text]
46. Williams, R. L., Greene, S. M., and McPherson, A. (1987) J. Biol. Chem. 262, 16020-16031 [Abstract/Free Full Text]
47. Joao, H. C., Scragg, I. G., and Dwek, R. A. (1992) FEBS Lett. 307, 343-346 [CrossRef][Medline] [Order article via Infotrieve]
48. Berman, E., Walters, D. E., and Allerhand, A. (1981) J. Biol. Chem. 256, 3853-3857 [Abstract/Free Full Text]
49. Rudd, P. M., Joao, H. C., Coghill, E., Fiten, P., Saunders, M. R., Opendakker, G., and Dwek, R. A. (1994) Biochemistry 33, 17-22 [CrossRef][Medline] [Order article via Infotrieve]
50. Wormald, M. R., Wooten, E. W., Bazzo, R., Edge, C. J., Feinstein, A., Rademacher, T. W., and Dwek, R. A. (1991) Eur. J. Biochem 198, 131-139 [Medline] [Order article via Infotrieve]
51. Brockbank, R. L., and Vogel, H. J. (1990) Biochemistry 29, 5574-5583 [CrossRef][Medline] [Order article via Infotrieve]
52. Woods, R. J., Edge, C. J., and Dwek, R. A. (1994) Nature Struct. Biol. 1, 499-501 [CrossRef][Medline] [Order article via Infotrieve]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				H. Yang and E. L. Reinherz CD2BP1 Modulates CD2-Dependent T Cell Activation via Linkage to Protein Tyrosine Phosphatase (PTP)-PEST J. Immunol., May 15, 2006; 176(10): 5898 - 5907. [Abstract] [Full Text] [PDF]

				E. V. Tibaldi and E. L. Reinherz CD2BP3, CIN85 and the structurally related adaptor protein CMS bind to the same CD2 cytoplasmic segment, but elicit divergent functional activities Int. Immunol., March 1, 2003; 15(3): 313 - 329. [Abstract] [Full Text] [PDF]

				E. V. Tibaldi, R. Salgia, and E. L. Reinherz CD2 molecules redistribute to the uropod during T cell scanning: Implications for cellular activation and immune surveillance PNAS, May 28, 2002; 99(11): 7582 - 7587. [Abstract] [Full Text] [PDF]

				X.C. Yu, EdwardD. Sturrock, Z. Wu, K. Biemann, MarioR.W. Ehlers, and JamesF. Riordan Identification of N-Linked Glycosylation Sites in Human Testis Angiotensin-converting Enzyme and Expression of an Active Deglycosylated Form J. Biol. Chem., February 7, 1997; 272(6): 3511 - 3519. [Abstract] [Full Text] [PDF]

				E. L. Reinherz, J. Li, A. Smoylar, D. F. Wyss, M. H. Knoppers, K. J. Willis, A. R. N. Arulanandam, J. S. Choi, and G. Wagner Response: CD2: An Exception to the Immunologlobulin Superfamily Concept? Science, August 30, 1996; 273(5279): 1242 - 1242. [PDF]

				D. Wyss, J. Choi, J Li, M. Knoppers, K. Willis, A. Arulanandam, A Smolyar, E. Reinherz, and G Wagner Conformation and function of the N-linked glycan in the adhesion domain of human CD2 Science, September 1, 1995; 269(5228): 1273 - 1278. [Abstract] [PDF]

				H.-H. Lin, M. Stacey, C. Saxby, V. Knott, Y. Chaudhry, D. Evans, S. Gordon, A. J. McKnight, P. Handford, and S. Lea Molecular Analysis of the Epidermal Growth Factor-like Short Consensus Repeat Domain-mediated Protein-Protein Interactions. DISSECTION OF THE CD97-CD55 COMPLEX J. Biol. Chem., June 22, 2001; 276(26): 24160 - 24169. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Davis, S. J. Articles by van der Merwe, P. A. Search for Related Content PubMed PubMed Citation Articles by Davis, S. J. Articles by van der Merwe, P. A. Social Bookmarking                      What's this?