Glycobiology Advance Access originally published online on August 2, 2007
Glycobiology 2007 17(10):1094-1103; doi:10.1093/glycob/cwm082
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The binding of human betacellulin to heparin, heparan sulfate and related polysaccharides
2 School of Biological Sciences, Royal Holloway University of London, Egham Hill, Egham, Surrey TW20 OEX, UK
3 Laboratory for Molecular Structure, National Institute for Biological Standards and Control, Blanche Lane, South Mimms, Potters Bar, Herts EN6 3QG, UK
1 To whom correspondence should be addressed: Tel: +44-0-1784-443548; Fax: +44-0-1784-414224; e-mail: c.rider{at}rhul.ac.uk
Received on April 26, 2007; revised on June 29, 2007; accepted on July 27, 2007
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
Recombinant human betacellulin binds strongly to heparin, requiring of the order of 0.8 M NaCl for its elution from a heparin affinity matrix. This is in complete contrast to the prototypic member of its cytokine superfamily, epidermal growth factor, which fails to bind to the column at physiological pH and strength. We used a well-established heparin binding ELISA to demonstrate that fucoidan and a highly sulfated variant of heparan sulfate compete strongly for heparin binding. Low sulfated heparan sulfates and also chondroitin sulfates are weaker competitors. Moreover, although competitive activity is reduced by selective desulfation, residual binding to extensively desulfated heparin remains. Even carboxyl reduction followed by extensive desulfation does not completely remove activity. We further demonstrate that both hyaluronic acid and the E. coli capsular polysaccharide K5, both of which are unsulfated polysaccharides with unbranched chains of alternating N-acetylglucosamine linked ß(1–4) to glucuronic acid, are also capable of a limited degree of competition with heparin. Heparin protects betacellulin from proteolysis by LysC, but K5 polysaccharide does not. Betacellulin possesses a prominent cluster of basic residues, which is likely to constitute a binding site for sulfated polysaccharides, but the binding of nonsulfated polysaccharides may take place at a different site.
Key words: betacellulin / glycosaminoglycan / heparan sulfate / heparin / polysaccharides
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
The epidermal growth factor (EGF) family cytokines are characterized by the possession of a sequence of 35–40 amino acids containing six cysteine residues with a conserved spacing. The covalent bond pairing of these cysteines forms three intrapolypeptide chain disulfide bridges which gives rise to a characteristic domain of three short, closely linked loops (reviewed by Dunbar and Goddard 2000
; Normanno et al. 2001). Members of the EGF family show many important biological activities, in particular potent mitogenesis, with proliferation being stimulated in a wide range of cell types of ectodermal, mesodermal, and endodermal origins. In addition, numerous activation and differentiation activities have been described in diverse cell types. The EGF-related cytokines are therefore important in growth, development, and tissue repair. As a consequence of these activities, the EGF signaling pathway is frequently subverted in cancer. EGF-family receptors are over-expressed at high frequency in various human cancers, including gastrointestinal carcinomas (Kopp et al. 2003), nonsmall-cell lung cancer (Sridhar et al. 2003) and primary breast cancer (Kim and Muller 1999). Such over-expression is associated with poor clinical prognosis, tumor progression and metastasis.
Given the pleiotropic mitogenic activities of EGF-related cytokines, molecular mechanisms exist to enable them to act in a juxtacrine manner, by restricting the diffusion of what are, in their mature forms, small soluble glycoproteins (Mrs 
10– 40 kDa). Thus, EGF and other members of the family are initially secreted in the form of domains within the extracellular regions of large cell surface precursor proteins possessing plasma membrane inserted hydrophobic sequences. The intact membrane-associated EGF precursor exhibits EGF bioactivity (Massague and Pandiella 1993). In the case of EGF, the extracellular portion of this membrane-bound protein can be proteolytically released as a soluble, bioactive protein of 160 kDa. This soluble EGF precursor was found to bind to heparin, and furthermore, to retain its bioactivity when heparin-bound (Parries et al. 1995). Notwithstanding the presence of this protein in urine, from which it was initially isolated, such interaction with heparin-like glycosaminoglycans, which are widespread in the extracellular matrix and on cell surfaces, would tend to retain the protein close to its sites of secretion within the tissues.
Once fully processed, the mature form of EGF, a 53 amino acid polypeptide, is not considered to be heparin/heparan sulfate binding cytokine. However, other members of this cytokine family are, including amphiregulin (AR), heparin-binding EGF-like growth factor (HB-EGF), and the neuregulins. In the case of AR, binding to heparin, a highly acidic polysaccharide, involves a 21 residue sequence in which 14 are the basically charged aminoacids, arginine, and lysine. This sequence is located immediately aminoterminal to the EGF homology domain (Johnson and Wang 1994). These workers also demonstrated not only that soluble heparin and heparan sulfate inhibit the in vitro cellular activity of AR, but also that stripping of cell surface heparan sulfate through the use of heparinases or the sulfation inhibitor chlorate, reduced receptor phosphorylation and AR-induced proliferation. Therefore, heparan sulfate appears to play a major role in AR-signaling. HB-EGF binds to glycosaminoglycans by a similar cluster of basic residues, but in this case, the binding sequence is shifted slightly in the nine basic arginines and lysines remain immediately aminoterminal to EGF domain, but three lie within the EGF domain, immediately carboxyterminal to the first of the conserved cysteines (Thompson et al. 1994). By contrast, the neuregulins bind to heparin and heparan sulfate via immunoglobulin homology domains, which are quite distinct from their EGF motifs (Loeb and Fischbach 1995; Meier et al. 1998).
In order to investigate further the extent and diversity of the heparin/heparan sulfate binding properties of EGF family cytokines, we have studied betacellulin (BTC) in this regard. Mature BTC is an 80-residue glycoprotein with a 33 amino acid aminoterminal extension to its EGF homology domain (Dunbar and Goddard 2000). This extension forms a loop due to its possession of a pair of covalently-bridged cysteine residues. The BTC sequence from Lys 30 to Arg 52 is rich in basic residues. BTC is of especial interest because it can bind and activate a broad spectrum of ErbB receptors including all possible heterodimer combinations of the four ErbB polypeptide chain types (Dunbar and Goddard 2000). This promiscuity, apparently unique for an individual member of the EGF family, underlies its potent mitogenicity in a broad range of cell types, both normal and transformed. For instance, by inducing proliferation of endothelial cells, BTC has angiogenic activity (Kim et al. 2003). However, BTC also appears to have a particular role in the development of the pancreas. This property has therapeutic potential in insulin-dependent diabetes, as recombinant BTC, or viral constructs expressing it, promote the development in the liver of insulin secreting cells resembling those of the pancreatic ß islets, resulting in improved control of blood glucose levels in drug-induced murine models of this disease (Yamamoto et al. 2000; Kojima et al. 2003).
Although the binding of BTC to heparin affinity columns has been reported for both naturally occurring and recombinant forms (Shing et al. 1993; Watanabe et al. 1994), the glycosaminoglycan binding properties of this cytokine have awaited characterization. Here we have investigated BTC-glycosaminoglycan interactions using a competitive heparin binding ELISA, and demonstrate that in addition to an important role for sulfate moieties, glycosaminoglycan backbone polysaccharides lacking sulfate groups retain low but significant levels of competitive activity.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
In ELISA experiments, as may be seen in Figure 1A, dose-dependent binding of recombinant human BTC (rBTC) to wells coated with heparin-BSA complex was readily and reproducibly detected. This binding was detectable with as little as 2 ng/well cytokine, and increased with rBTC loading in quasi-linear manner. By comparison, under the same conditions, there was negligible binding to wells coated instead with mock-conjugated BSA (Figure 1A). To confirm that the strong binding to the heparin complex is indeed due to interaction with this glycosaminoglycan, rBTC was preincubated with increasing concentrations of soluble heparin, prior to addition to the wells. As shown in Figure 1B, low concentrations of heparin are able to displace the binding of rBTC to the immobilized complex. Thus the binding of 15 ng/well rBTC is almost completely competed out by 5 µg/mL soluble heparin, with the competition curve giving an IC50 of the order of 0.5 µg/mL. This IC50 value appeared to be independent of the rBTC loading, as it was unchanged in experiments in which 25 ng rather than 15 ng cytokine was added per well (data not shown).
|
In order to confirm the heparin-binding nature of rBTC, and to provide a measure of the affinity of this interaction, we performed heparin affinity chromatography on a mixture of rBTC and the prototypical member of its cytokine family, recombinant epidermal growth factor (rEGF). As may be seen in Figure 2, at physiological pH and ionic strength, rEGF elutes through the column without retention (inset). However, rBTC binds to the heparin matrix, and elutes only on the application of a NaCl gradient, with peak immunoreactivity eluting in the range 0.8–1.0 M NaCl.
|
To investigate the specificity of BTC-glycosaminoglycan interactions, we employed further the competitive format of the heparin binding ELISA by preincubating rBTC with various sulfated polysaccharides prior to addition to heparin complex coated wells. As may be seen in Figure 3A, fucoidan, a sulfated fucose polymer from marine algae, is like heparin an effective competitor of binding to the immobilized complex. Chondroitin sulfates are however markedly less active in this regard, despite being employed here at higher concentrations. Over a series of four similar experiments, little or no difference was observed in competitive activity between the three classes of chondroitin sulfate, A, B, and C. Given the comparative high affinity of BTC for heparin, the specificity of BTC-glycosaminoglycan interactions was further examined using a series of heparan sulfates as soluble competitors. As may be seen in Figure 3B, the heparan sulfates are less effective competitors than heparin, and also vary in their activities. Thus, kidney heparan sulfate and HSA, a relatively low sulfated heparan sulfate from intestinal mucosa, are both weak competitors, whereas the more highly sulfated, HSE, like heparin, provides nearly complete inhibition. However, as shown in Figure 3C, across a range of low concentrations of soluble GAG, the competitive activity of HSE is significantly less than that of heparin.
|
In order to elucidate the structural features of heparin and highly sulfated heparan sulfate required for the binding of BTC, the specificity of this interaction was further examined using a series of derivatives of bovine lung heparin produced by selective chemical modifications, including de-N-sulfation, de-O-sulfation, N-acetylation, and carboxyl reduction. The modified heparins were characterized by 1H NMR spectroscopy in comparison with published data (Mulloy et al. 1994
; Yates et al. 1996) and by molecular weight analysis (Table I). NMR spectra of the carboxyl reduced, N-desulfated, N-acetylated (CRNDESNAC) heparin, and carboxyl-reduced, N- and O-desulfated, N-acetylated (CRDESNAC) heparin, were assigned by TOCSY and COSY 1H NMR spectroscopy. Assignments for CRNDESNAC heparin are listed in Table II. For CRDESNAC heparin, it was impractical to assign the 1H NMR spectrum due to severe overlap. An anomeric signal was identifiable at 5.15 ppm, probably including both N-acetylglucosamine and iduronate anomeric resonances, with the ring protons overlapping in an envelope between 3.7 and 3.9 ppm and a prominent N-acetyl methyl signal at 2.02 ppm. A signal of relatively low intensity at about 4.33 ppm probably indicated some remaining sulfation, but as the sample had depolymerized considerably (see Table I) it was not re-treated.
|
|
As may be seen in Figure 4A, the unmodified bovine lung heparin, like the routinely employed porcine intestinal mucosal heparin, provides almost complete inhibition of rBTC binding to the immobilized heparin complex. However, N-desulfation results in a substantial loss of this activity. Subsequent N-acetylation results in a modest but reproducible increase in competitive activity. Both selective 2-O- and 6-O- desulfations reduce activity to levels similar to that seen with N-desulfation. However, the most striking finding here is that complete desulfation of heparin does not result in a complete loss of competitive activity, but in an activity similar to that obtained by the selective desulfations.
|
As shown in Figure 4B, heparin subjected to carboxyl reduction followed by N-desulfation and N-acetylation still retains competitive activity. However this activity is lower than that of heparin subjected only to N-desulfation and N-acetylation. Thus in the context of N-sulfate removal, carboxyl groups are seen to play a role in betacellulin binding. The carboxyl-reduced and totally N-and O-desulfated, N-acetylated heparin (CRDESNAC) is less active than its counterpart which retains O-sulfates (CRNDESNAC). However, the notable observation here is that a heparin preparation stripped of all its acidic moieties, both sulfate and carboxyl groups, is still able to interact with rBTC, albeit as a weak competitor with intact heparin. The interpretation of the relative activities of the chemically modified heparins is compounded by the likelihood that beyond the simple presence or absence of the various substituents, such as particular sulfate groups, the various modified preparations may well show conformational differences (Mulloy et al. 1994
). However, overall it is clear that rBTC can bind heparin derivatives of both high and very low degrees of sulfation.
In order to examine further the binding of rBTC to totally unsulfated glycosaminoglycan chains, we employed hyaluronic acid and E. coli capsular polysaccharide K5. As may be seen in Figure 4C, both of these polysaccharides show significant competitive activity. By contrast acharan sulfate, a sulfated polysaccharide with the major repeating disaccharide of [
4)-
-L-IdoA2OSO3–-(14)--D-GlcNAc-(1-] (Kim et al. 1996) shows negligible competitive activity.
A high resolution solution structure of residues 32–81 of mature BTC has been determined by NMR spectroscopy (Miura et al. 2002). This structure is revealed as a member of the EGF-type module fold family (SCOP database: http://scop.mrc-lmb.cam.ac.uk/scop/index.html). Inspection of electrostatic potential at the surface of the protein (Figure 5A) reveals a prominent patch of positive charge arising from basic residues in an approximately linear arrangement available for interaction with the negatively charged sulfate groups of heparin. The reverse side of the protein is largely neutral (Figure 5B). To explore the heparin-binding potential of this protein, a validated automated docking protocol (Forster and Mulloy 2006) was carried out, using the coordinates of the average NMR structure (1ipo.pdb) together with the coordinates for heparin endecamer oligosaccharide. The 10 lowest energy docked complexes show the oligosaccharide binding to the basic residues aligned diagonally in the view seen in Figure 5A to cover the long axis of this positive patch (not shown). As may be seen in a view across the plane of the interaction between BTC and heparin oligosaccharides (Figure 5C), two arginines and two lysines near the aminoterminus of the structured EGF homology domain of BTC are important contact residues.
|
To measure the influence of heparin-related glycosaminoglycans on BTC activity we employed the cell line T-47D (Musgrove and Sutherland 1993
) which is widely used to investigate the activity of EGF family cytokines. On exposure to BTC, immunoblotting with a monoclonal antibody specific for tyrosine-phosphorylated Erk MAP kinase, cell extracts showed a large increase in the intensity of a doublet of bands of around 45 kDa. This doublet co-migrated with the immunoreactivity detected with a pan-Erk MAP kinase antibody. The phosphorylation response was dose-dependent on BTC, and maximal at 10–15 min, thereafter declining gradually over 2 h. The preincubation of BTC with exogenous heparin at a final concentration of 10 µg/mL after addition to the cells caused no reproducible effect on the intensity of this doublet as determined by densitometry. Likewise the pre-treatment of the cells with heparinase III was without reproducible effect on Erk-MAP Kinase phosphorylation in response to BTC (data not shown). On exposure to the bacterial endoprotease LysC, BTC is rapidly degraded under the conditions employed, with almost complete lost of BTC immunoreactivity by 15 min of digestion in the absence of polysaccharide (Figure 6). Heparin at low concentrations is able to protect against this degradation. However the K5 polysaccharide shows no effect on BTC digestion. This is despite both polysaccharides being present at concentrations that give around 50% competition for the binding of BTC to the heparin-BSA solid phase in ELISA (Figures 1B and 4).
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
Using an ELISA approach that we developed to investigate the heparin binding properties of cytokines and other proteins (Najjam et al. 1997
), we now establish that human recombinant BTC binds strongly to heparin. In a competitive format of this assay, addition of increasing concentrations of soluble heparin yielded an IC50 value of the order of 0.5 µg/mL. This compares with an IC50 estimate of 0.15 µg/mL obtained for FGF-2 (Najjam et al. 1997), which is a cytokine with well established high affinity for heparin, and values of 5 µg/mL for interleukin 2 (Najjam et al. 1997), 2 µg/mL for interleukin 6 (Mummery and Rider 2000), and 0.1 µg/mL for both interleukin 12 (Hasan et al. 1999) and GDNF (Rickard et al. 2003), all obtained by the same experimental approach. On heparin affinity chromatography, BTC binds at physiological pH and ionic strength, with peak elution occurring at salt concentrations of 0.8–1.0 M. This compares to a salt concentration of around 1.5 M required for the similar elution of FGF-2 (Esch et al. 1985). Taken overall, compared to that of other cytokines, the affinity of BTC for heparin can be considered as moderate rather than high affinity.
Since heparan sulfate is widely distributed on cell surfaces and in the extracellular matrix, compared to the restricted localization of heparin within mast cell granules, the binding of BTC to heparan sulfates, shown here by their activity as competitors in the heparin binding ELISA, is of some physiological significance. All three heparan sulfates examined are active, with the two intestinal mucosa heparan sulfates being more active. We have previously reported the structural characterization of these two preparations, with the smaller HSE being heparin-like in the density and pattern of its sulfation whereas the larger HSA has markedly lower levels of N- and O-sulfation (Rickard et al. 2003). Here both of these preparations are seen to exhibit competitive activity, but with HSE being substantially more active. This is similar to their relative activities found in our previous study of GDNF (Rickard et al. 2003), but is in contrast to their equal activities with IL-6 (Mummery and Rider 2000), and also to the case of IL-12 where HSA showed negligible activity (Hasan et al. 1999). The conclusion here is that the heparin binding of BTC is favored by high levels of sulfation, although binding low sulfated heparan sulfate still occurs.
The biological consequences of the interaction of BTC with heparin-like glycosaminoglycans remains to be established. The binding of a cytokine to heparin/heparan sulfate is often associated with a potentiation of signaling activity. However in a widely used cellular bioassay of BTC activity which measures the phosphorylation of Erk MAP kinase in the T-47D breast carcinoma cell line, we were unable to find either increases or decreases in activity when exogenous heparin was added. Likewise exposure of the cells to heparinase III, to degrade cell surface heparan sulfate, both before and during BTC stimulation, had no reproducible effect on Erk MAP kinase phosphorylation. The absence of readily measurable effects of either exogenous heparin addition or cell surface heparan sulfate degradation in this simple cell culture system, appears to rule a role for these GAGs as co-receptors for BTC, such as is seen with other cytokines, notably FGF-2 (Pellegrini et al. 2000; Sclessinger et al. 2000; Ostrovsky et al. 2002). Our studies do not preclude the possibility that in the more complex extracellular environment within the tissues, heparin-like glycosaminoglycans may have an important role in regulating BTC activity.
The use of selectively modified heparins provides insights into the specificity of protein-heparin interactions. Our current results with BTC show again, as in our previous studies employing the same modified heparin preparations, that each protein studied has a differing profile of binding activities. Here, selective N-desulfation of heparin substantially reduces its competitive activity, with subsequent N-acetylation providing a small restoration. These findings are in strong contrast to those for HIV-1 gp120, where N-desulfation had only a modest effect (Rider et al. 1994), and GDNF, where N-desulfation all but completely removed competitive activity, for this to be partially returned on subsequent N-acetylation (Rickard et al. 2003). With BTC selective 6-O-desulfation and 2-O-desulfation both result in partial activities of a similar degree. Again, this contrasts with other previously studied cytokines. In particular, 2-O-desulfation had no measurable effect on competitive activity with IL-6 (Mummery and Rider 2000), but totally removed activity in the case of GDNF (Rickard et al. 2003). Overall these findings show that the specificity of the interaction between BTC and the structural moieties in heparin differs from that of other cytokines.
The observation that with BTC each selective desulfation caused only partial loss of competitive activity, led us to examine totally N- and O-desulfated heparin. Remarkably, given the generally accepted importance of sulfates in heparin/heparan sulfate—protein interactions (Hileman et al. 1998; Esko et al. 2002), we found that totally desulfated heparin remains an effective competitor, indeed with an activity similar to those of the three selectively desulfated heparins. Use of further novel, extensively modified heparins established that reduction of the carboxyl groups followed by total desulfation of heparin, thus removing all acidic groups, still failed to abolish competitive activity. This was despite the modifications causing some cleavage of the glycosaminoglycan chains. It should be borne in mind that the absence of the bulky and strongly acidic sulfate groups will allow access to different conformational equilibria for the polysaccharide, especially for the conformationally mobile iduronate pyranose ring. Such flexibility may allow optimization of weak interactions between the glycosaminoglycan and an interacting polypeptide. We also found that the unsulfated glycosaminoglycans hyaluronic acid and K5 possess competitive activities. The K5 polysaccharide, a repeat polymer of the disaccharide [4)-ß-D-GlcA-(14) --D-GlcNAc-(1-], has the same structure as heparin prior to its extensive postpolymerization modifications (Esko and Selleck 2002). Hyaluronic acid is also an alternating polymer of glucuronic acid and N-acetyl glucosamine, but with a different linkage pattern: [4)-ß-D-GlcA-(13) -ß-D-GlcNAc-(1-]. Their similar activities here suggest that the differences in their hexose linkages are without consequence on BTC binding. The activities of these two unsulfated polysaccharides are significantly greater than that of acharan sulfate, which is repeat polymer of iduronic acid-N-acetyl glucosamine sulfated at the C2 of iduronic acid.
The binding of heparin to proteins is mediated predominantly by ionic interactions between the sulfate groups of the glycosaminoglycan and the basic side chains of arginines and lysines on the polypeptide. This is well illustrated by the high resolution structures of heparin oligosaccharides bound to FGF-FGF receptor complexes (Pellegrini et al. 2000; Schlessinger et al. 2000). However unsulfated polysaccharides bind to proteins largely by hydrophobic interactions with aromatic amino acid side chains (reviewed by Boraston et al. 2004). In the case of BTC, although we have established here that unsulfated polysaccharides compete with binding to heparin, it remains unclear whether the sulfated and unsulfated polysaccharides bind to the same site. Within mature BTC, a polypeptide of 80 residues, all the five lysines, and four of the six arginines lie within a sequence of 22 residues. A high resolution NMR structure has been solved for residues 31–80 of rBTC, encompassing all but the first of the clustered basic residues (Miura et al. 2002). Inspection of the protein surface shows that most of these basic residues constitute a prominently exposed, large patch of basic charge (Figure 5A), which therefore is a very likely location of the heparin-binding site; our docking calculations also predict that this is the case. It is notable that these residues lie in an approximately comparable position to the heparin binding sites of the BTC homologues AR (Johnson and Wang 1994) and HB-EGF (Thompson et al. 1994). Our finding that heparin protects BTC from degradation by the bacterial endoproteinase LysC, which is specific for lysine residues, is entirely consistent with this prediction for the heparin binding site. It also raises the possibility that within the tissues, interaction of BTC with heparin-related glycosaminoglycans might offer similar protection against proteolysis.
However, there is no reason to suppose that unsulfated or low sulfated polysaccharides will bind at the same site. Indeed, the unsulfated K5 polysaccharide does not provide BTC with protection against LysC proteolysis under the conditions employed. This implies that, unlike heparin, K5 binding does not block exposed lysines. BTC is a small protein and if the less acidic polysaccharides were to bind to a different part of the surface, for instance, in a cleft lined with aromatic residues in the lower centre of the neutral face of the protein (Figure 5B and C). It is possible that occupancy of such a second site might alter the conformation or dynamics of the protein in such a way as to affect heparin binding. BTC is structurally related to the knottins, several of which display a carbohydrate binding site of this type (Smith et al. 1998). There is a precedent for such binding of both sulfated and unsulfated glycosaminoglycans at different sites, in the link module of TSG-6 protein (Mahoney et al. 2005). This is a compact domain of slightly larger size (97 residues) than BTC and there is competition for binding between heparin and hyaluronan even though the two binding sites on TSG-6 do not overlap. We speculate that this model of two separate but competing binding sites within a compact protein domain, one for sulfated polysaccharides, and the other for nonsulfated polysaccharides, may apply in the case of BTC.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
Materials
rBTC and rEGF both expressed in E. coli, goat polyconal anti-rBTC immunoglobulin, and murine anti-rBTC monoclonal antibodies were purchased from R&D Systems Europe Ltd. (Abingdon, UK). Affinity purified rabbit anti-EGF was supplied by PeproTech EC Ltd. (London, UK). Bovine serum albumin (A 2513, Fraction V) and alkaline phosphatase substrate tablets were obtained from Sigma-Aldrich Chemical Co. Ltd. (Dorset, UK). NUNC Maxisorp 96-well ELISA plates were obtained from Life Technologies (Paisley, Scotland). Porcine horseradish peroxidase conjugated antirabbit immunoglobulin was supplied by DakoCytomation, (Ely, UK). Nitrocellulose Protran BA85 0.45 µm membrane was supplied by Schleicher and Schuell, (Dassel, Germany). Porcine intestinal mucosal heparin (sodium salt, grade I-A); chondroitin sulfates A, B, and C; fucoidan; and bovine kidney heparan sulfate were all purchased from Sigma-Aldrich. Chondroitin sulfate-free hyaluronic acid (70 kDa) was supplied by Genzyme Biosurgery (Cambridge, MA). E. coli K5 polysaccharide was purchased from Iduron Ltd. (Manchester, UK). Acharan sulfate was the kind gift from Prof. Robert Linhardt (University of Iowa). Heparan sulfates A and E, respectively low and highly sulfated fractions isolated from porcine intestinal mucosa, and their structural characteristics as determined by 500 mHz 1H NMR have been previously described (Rickard et al. 2003
). Bovine lung heparin (2nd International Standard) was N-desulfated by the method of Nagasawa and Inoue et al. (1977). N-desulfated, N-acetylated heparin, and N- and O-desulfated, N-acetylated heparin were prepared as previously described (Mulloy et al. 1994). Carboxyl reduced heparin (using the method of Taylor et al. 1976) was N-desulfated (Inoue and Nagasawa 1976) and subsequently N-acetylated (Danishevsky et al. 1960). A portion of the resulting compound was further O-desulfated (Nagasawa et al. 1977). 2-O-desulfated heparin and 6-O-desulfated heparin were prepared as previously described (Ostrovsky et al. 2002).
NMR spectroscopy
1H NMR spectra of the carboxyl reduced, N-desulfated, N-acetylated heparin, and the carboxyl reduced, N- and O-desulfated, N-acetylated heparin were recorded in D2O solution at 60°C using a Varian Inova 500 MHz spectrometer. Two-dimensional TOCSY and COSY spectra were recorded using pulse sequences supplied by the manufacturer.
Molecular weight determinations of modified heparins
The molecular weight profiles of the parent and chemically modified heparin was determined by high-performance gel permeation chromatography using a protocol for unfractionated heparin (Mulloy et al. 2000) or low molecular weight heparin (Mulloy et al. 1997).
Heparin binding ELISA
For heparin binding ELISA (Najjam et al. 1997), wells were coated overnight at 4°C with 100 µL 50 mM Tris-HCl buffer, pH 7.4, containing 12.7 mM EDTA, and either 5 ng heparin-BSA complex as determined by protein content, or the same amount of mock-conjugated BSA, in PBS. After washing three times with PBS, wells were blocked with 2% (wt/vol) BSA. Wells were incubated for 2 h with rBTC diluted in PBS containing 2.5 mg/mL BSA. After washing three times with PBS containing 0.05% (vol/vol) Tween 20, 100 µL anti-BTC polyclonal antibody was added at a dilution of 1/200 in blocking buffer for 2 h. Following three further washes in PBS-Tween, alkaline phosphatase-coupled rabbit antigoat IgG second antibody (R&D Systems) was added at a dilution of 1/1000 in blocking buffer for 30 min. After five washes in PBS-Tween, alkaline phosphatase activity was detected by adding 100 µL/well p-nitrophenol phosphate solution. In some experiments, a competitive variant of the ELISA was used in which cytokine diluted in PBS was preincubated with soluble glycosaminoglycan for 30 min prior to the addition of 100 µL aliquots of this mixture to coated and blocked wells. Absorbances were read at 405 nm after an appropriate time of incubation at room temperature, typically around 1 h, in an Emax plate reader (Molecular Devices, Sunnyvale, CA). For each plate, readings were read against replicate blank wells, which were blocked without prior coating, received no rBTC, and were incubated with first and second antibodies, and developed as described above.
Heparin affinity chromatography
A 1 mL heparin HP HiTrap column (Amersham Biosciences, Little Chalfont, Bucks, UK), equilibrated with PBS, was loaded with a mixture of 100 ng each of rBTC and rEGF. After eluting with 5 mL PBS, a linear gradient of 0–1.5 M additional NaCl was applied, followed by a wash of 1.5 M NaCl in PBS. The EGF content of the fractions was determined using dot blotting, by applying 30 µL aliquots to a nitrocellulose membrane held in a 96-well dot blot manifold and pre-wetted with PBS. After 2 h incubation at 4°C, samples were drawn through the membrane under suction, and the membrane was removed from the manifold and allowed to dry. After storage at 4°C overnight, the membrane was washed in PBS, blocked in PBS containing 2% (wt/wol) BSA for 30 min, and incubated for 2 h with anti-EGF diluted 1:500 in blocker solution. Following washing in PBS containing 0.05% (vol/vol) Tween 20, the membrane was incubated for 30 min in horseradish peroxidase-conjugated antirabbit immunoglobulin diluted 1:1000 in blocking solution. Following extensive washing in PBS-Tween, the membrane was developed by chemiluminescence using a Supersignal West Pico kit (Perbio Science UK Ltd., Tattenhall, UK). For BTC ELISA, microtiter plate wells were coated overnight at 4°C with monoclonal anti-BTC diluted to 4 µg/mL in 30 mM sodium bicarbonate buffer, pH 8.5. The plate was washed with PBS, and blocked in PBS-BSA for 30 min at room temperature. After further washing with PBS, wells were incubated for 2 h with 80 µL of column eluate fractions diluted 1:10 in PBS containing 0.25% (wt/vol) BSA. After washing the plate, with PBS-Tween, wells were incubated for 1 h with anti-rBTC polyclonal antibody diluted 1:300 in blocker solution. Further washes in PBS-Tween were followed by a 30 min incubation with alkaline phosphatase-coupled antigoat antibody (Jackson Immunoresearch, supplied by Stratech Scientific, Soham, Cambridgshire, UK) diluted 1:1000 in blocker, three washes in PBS-Tween, and development with p-nitrophenol phosphate.
Cellular assays of BTC activity
T-47D human breast carcinoma cells were grown in RPMI 1640 medium containing 2.0% foetal calf serum. For determination of BTC activity, cells were plated out in medium containing 0.25% foetal calf serum for 48 h, with a change of medium for every 24 h. To some cultures, BTC preincubated with or without heparin was then added. After incubation, cell layers were washed with ice-cold PBS, and extracted with SDS sample buffer at 100°C. Alternatively, the effect of degrading cell surface HS was examined by adding heparinase III to a final concentration of 0.25 units/mL from a stock solution of the enzyme stored at –20°C in 10 mM Tris/HCl buffer, pH 7.4, containing 50% glycerol. Control wells received only the same volume of this buffer. After 90 min, BTC was added without heparinase removal, and cells were washed and extracted after a further 30 min.
Proteolytic experiments
Digestion of BTC by the bacterial endoprotease LysC was carried out essentially as described previously (Rickard et al. 2003). Proteolysis of 1 ng/µL rBTC was carried out with 0.7 ng/mL enzyme, in the presence and absence of heparin or K5 polysaccharide. Samples removed at the indicated times and immediately boiled in SDS-polyacrylamide gel electrophoresis sample buffer. Aliquots corresponding to 8 ng rBTC were resolved on 15% gels.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
None declared.
|
Abbreviations |
|---|
AR, amphiregulin; BTC, betacellulin; CRDESNAC, carboxyl-reduced totally desulfated N-acetylated heparin; CRNDESNAC, carboxyl-reduced N-desulfated N-acetylated heparin; EGF, epidermal growth factor; ErbB, epidermal growth factor receptor bB; Erk MAP kinase, extracellular signal-regulated mitogen-associated protein kinase; FGF, fibroblast growth factor; GDNF, Glial cell line-derived neurotrophic factor; HB-EGF, heparin-binding EGF-like growth factor; rBTC, recombinant betacellulin; rEGF, recombinant epidermal growth factor; TSG-6, tumor necrosis factor-stimulated gene-6
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
Boraston AB, Bolam DN, Gilbert HJ, Davies GJ. Carbohydrate-binding modules: fine-tuning polysaccharide recognition. Biochem J (2004) 382:769–781.[CrossRef][Web of Science][Medline]
Danishefsky I, Eiber HB, Carr JJ. Investigations on the chemistry of heparin. I. Desulfation and acetylation. Arch Biochem Biophys (1960) 90:114–121.[CrossRef][Web of Science][Medline]
Dunbar AJ, Goddard C. Structure-function and biological role of betacellulin. Int J Biochem Cell Biol (2000) 32:805–815.[CrossRef][Web of Science][Medline]
Esch F, Baird A, Ling N, Ueno N, Hill F, Denoroy L, Klepper R, Gospodarowicz D, Bohlen P, Guillemin R. Primary structure of bovine pituitary basic fibroblast growth factor (FGF) and comparison with the amino-terminal sequence of bovine brain acidic FGF. Proc Natl Acad Sci USA (1985) 82:6507–6511.
Esko JD, Selleck SB. Order out of chaos: assembly of ligand binding sites in heparan sulfate. Annu Rev Biochem (2002) 71:435–471.[CrossRef][Web of Science][Medline]
Forster M, Mulloy B. Computational approaches to the identification of heparin-binding sites on the surfaces of proteins. Biochem Soc Trans (2006) 34:431–434.[CrossRef][Web of Science][Medline]
Hasan M, Najjam S, Gordon MY, Gibbs RV, Rider CC. IL-12 is a heparin binding cytokine. J Immunol (1999) 162:1064–1070.
Hileman RE, Fromm JR, Weiler JM, Linhardt RJ. Glycosaminoglycan-protein interactions: definition of consensus sites in glycosaminoglycan binding proteins. BioEssays (1998) 20:156–157.[CrossRef][Web of Science][Medline]
Inoue Y, Nagasawa K. Selective N-desulfation of heparin with dimethyl sulfoxide or methanol. Carbohydrate Res (1976) 46:87–95.[CrossRef][Web of Science][Medline]
Johnson G, Wong L. Heparan sulphate is essential to amphiregulin-induced mitogenic signalling by the epidermal growth factor receptor. J Biol Chem (1994) 269:27149–27154.
Kim H, Muller WJ. The role of epidermal growth factor receptor family in mammary tumorigenesis and metastasis. Exp Cell Res (1999) 253:78–87.[CrossRef][Web of Science][Medline]
Kim HS, Shin HS, Kwak HJ, Cho CH, Lee CO, Koh GY. Betacellulin induces angiogenesis through activation of mitogen-activated protein kinase and phosphatidylinositol 3'-kinase in endothelial cell. FASEB J (2003) 17:318–320.
Kim SY, Jo YY, Chang IM, Toida T, Parks Y, Linhardt RJ. A new glycosaminoglycan from the giant African snail Achatina fulica. J Biol Chem (1996) 271:11750–11755.
Kojima H, Fujimiya M, Matsumura K, Younan P, Imaeda H, Maeda M, Chan L. NeuroD-betacellulin gene therapy induces islet neogenesis in the liver and reverses diabetes in mice. Nat Med (2003) 9:596–603.[CrossRef][Web of Science][Medline]
Kopp R, Rothbauer E, Ruge M, Arnholdt H, Spranger J, Muders M, Pfeiffer DG, Schildberg FW, Pfeiffer A. Clinical implications of the EGF receptor/ligand system for tumor progression and survival in gastrointestinal carcinomas: evidence for new therapeutic options. Recent Results Cancer Res (2003) 162:115–132.[Medline]
Loeb JA, Fischbach GD. ARIA can be released from extracellular matrix through cleavage of a heparin-binding domain. J Cell Biol (1995) 130:127–135.
Mahoney DJ, Mulloy B, Forster MJ, Blundell CD, Fries E, Milner CM, Day AJ. Characterization of the interaction between tumour necrosis factor-stimulated gene 6 and heparin. J Biol Chem (2005) 280:27044–27055.
Massague J, Pandiella A. Membrane-anchored growth factors. Annu Rev Biochem (1993) 62:515–541.[CrossRef][Web of Science][Medline]
Meier T, Masciulli F, Moore C, Schoumacher F, Eppenberger U, Denzer AJ, Jones G, Brenner HR. Agrin can mediate acetylcholine receptor gene expression in muscle by aggregation of muscle-derived neuregulins. J Cell Biol (1998) 141:715–726.
Miura K, Doura H, Aizawa T, Tada H, Seno M, Yamada H, Kawano K. Solution structure of betacellulin, a new member of EGF-family ligands. Biochem Biophys Res Commun (2002) 294:1040–1046.[CrossRef][Web of Science][Medline]
Mulloy B, Forster MJ, Jones C, Drake AF, Johnson EA, Davies DB. The effect of variation of substitution on the solution conformation of heparin: a spectroscopic and molecular modelling study. Carbohydr Res (1994) 255:1–26.[CrossRef][Web of Science][Medline]
Mulloy B, Gee C, Wheeler S, Wait R, Thomas S, Gray E, Barrowcliffe TW. Molecular weight measurements of low molecular weight heparins by gel permeation chromatography. Thromb Haemost (1997) 77:668–674.[Web of Science][Medline]
Mulloy B, Gray E, Barrowcliffe TW. Characterisation of unfractionated heparin samples: comparison of samples from the last 50 years. Thromb Haemost (2000) 84:1052–1056.[Web of Science][Medline]
Mummery RS, Rider CC. Characterization of the heparin binding properties of IL-6. J Immunol (2000) 165:5671–5679.
Musgrove EA, Sutherland RL. Acute effects of growth factors on T-47D breast cancer cell cycle progression. Eur J Cancer (1993) 29A:2273–2279.[CrossRef]
Nagasawa K, Inoue Y, Kamata T. Solvolytic desulfation of glycosaminoglycuronan sulfates with dimethyl sulfoxide containing water or methanol. Carbohydr Res (1977) 58:47–55.[CrossRef][Web of Science][Medline]
Najjam S, Gibbs RV, Gordon MY, Rider CC. Characterisation of human recombinant interleukin 2 binding to heparin and heparan sulphate using an ELISA approach. Cytokine (1997) 9:1013–1022.[CrossRef][Web of Science][Medline]
Nicholls A, Sharp KA, Honig B. Protein folding and association: insights from the interfacial and thermodynamic properties of hydrocarbons. Proteins (1991) 11:281–296.[CrossRef][Web of Science][Medline]
Normanno N, Bianco C, De Luca A, Salomon DS. The role of EGF-related peptides in tumor growth. Front Biosci (2001) 1:D685–D707.
Ostrovsky O, Berman B, Gallagher JT, Mulloy B, Fernig DG, Berman B, Ron D. Differential effects of heparin saccharides on the formation of specific FGF and FGFR receptor complexes. J Biol Chem (2002) 277:2444–2453.
Parries G, Chen K, Misono KS, Cohen S. The human urinary epidermal growth factor (EGF) precursor. J Biol Chem (1995) 270:27954–27960.
Pellegrini L, Burke DF, von Delft F, Mulloy B, Blundell TL. Crystal structure of fibroblast growth factor receptor ectodomain bound to ligand and heparin. Nature (2000) 407:1029–1034.[CrossRef][Medline]
Rickard SM, Mummery RS, Mulloy B, Rider CC. The binding of human glial cell-line derived neurotrophic factor (GDNF) to heparin and heparan sulphate: importance of 2-O-sulphate groups and effects on its interaction with its receptor GFRa1. Glycobiology (2003) 13:419–426.
Rider CC, Coombe DR, Harrop HA, Hounsell EF, Bauer C, Feeney J, Mulloy B, Mahmood N, Hay A, Parish CR. Anti-HIV-1 activity of chemically modified heparins: correlation between binding to the V3 loop of gp120 and inhibition of cellular HIV-1 infection in vitro. Biochemistry (1994) 33:6974–6980.[CrossRef][Medline]
Schlessinger J, Plotnikov AN, Ibrahimi OA, Eliseenkoya AV, Yeh BK, Yayon A, Linhardt RJ, Mohammadi M. Crystal structure of a ternary FGF-FGFR-heparin complex reveals a dual role for heparin in FGFR binding and dimerization. Molecular Cell (2000) 6:743–750.[CrossRef][Web of Science][Medline]
Shing Y, Christofori G, Hanahan D, Ono Y, Sasada R, Igarashi K, Folkman J. Betacellulin: a mitogen from pancreatic ß cell tumors. Science (1993) 259:1604–1607.
Smith GP, Patel SU, Windass JD, Thornton JM, Winter G, Griffiths AD. Small binding proteins selected from a combinatorial repertoire of knottins displayed on phage. J Mol Biol (1998) 277:317–332.[CrossRef][Web of Science][Medline]
Sridhar SS, Seymour L, Shepherd FA. Inhibitors of epidermal-growth-factor receptors: a review of clinical research with a focus on non-small-cell lung cancer. Lancet Oncol (2003) 4:397–406.[CrossRef][Web of Science][Medline]
Taylor RL, Shively JE, Conrad HE. Stoichiometric reduction of uronic acid carboxyl groups in polysaccharides. Methods Carbohydr Chem (1976) 7:149–151.
Thompson SA, Higashiyama S, Wood K, Pollitt NS, Damm D, McEnroe G, Garrick B, Ashton N, Lau K, Hancock N, Klagsburn M, Abraham JA. Characterization of sequences within heparin-binding EGF-like growth factor that mediate interaction with heparin. J Biol Chem (1994) 269:2541–2549.
Watanabe T, Shintani A, Nakata M, Shing Y, Folkman J, Igarashi K, Sasada R. Recombinant human betacellulin. Molecular structure, biological activities, and receptor interactions. J Biol Chem (1994) 269:9966–9973.
Yamamoto K, Miyagawa J, Waguri M, Sasada R, Igarashi K, Li M, Nammo T, Moriwaki M, Imagawa A, Yamagata K, Nakajima H, Namba M, Tochino Y, Hanafusa T, Matsuzawa Y. Recombinant human betacellulin promotes the neogenesis of beta-cells and ameliorates glucose intolerance in mice with diabetes induced by selective alloxan perfusion. Diabetes (2000) 49:2021–2027.
Yates EA, Santini F, Guerrini M, Naggi A, Torri G, Casu B. 1H and 13C NMR spectral assignments of the major sequences of twelve systematically modified heparin derivatives. Carbohydr Res (1996) 294:15–27.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||