Glycobiology Advance Access originally published online on June 18, 2007
Glycobiology 2007 17(9):983-993; doi:10.1093/glycob/cwm062
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Influence of substitution pattern and cation binding on conformation and activity in heparin derivatives
2 School of Biological Sciences, University of Liverpool, Liverpool L69 7ZB, UK
3 Ronzoni Institute for Chemical and Biochemical Research, Via G. Colombo 81, Milan 20133, Italy
4 CCLRC, Daresbury Laboratory, Warrington, Cheshire WA4 4AD, UK
1 To whom correspondence should be addressed: Tel: +44-151-795-4429; Fax: +44-151-795-4406; e-mail: eayates{at}liv.ac.uk
Received on April 25, 2007; revised on June 4, 2007; accepted on June 5, 2007
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgements/FundingReferences |
|---|
As model compounds for the biologically important heparan sulfate, eight systematically modified heparin derivatives were studied by synchrotron radiation circular dichroism (SRCD), which is sensitive to uronic acid confor- mation. Substitution pattern altered uronic acid conformation, even when structural changes were made in adjacent glucosamine residues (e.g. 6-O-desulfation) and did not involve a chromophore. SRCD spectra of these derivatives following conversion to the Na+, K+, Mg2+, Ca2+, Mn2+, Cu2+ and Fe3+ cation forms revealed that almost all substitution/cation combinations resulted in unique spectra, showing that each was structurally distinct. The detailed effects that binding Na+, K+, Mg2+ and Ca2+ ions had on a 2-de-O-sulfated derivative was also studied by NMR, revealing that subtle changes in conformation (by NOE) and flexibility (by T2 measurements) resulted. Conversion to the K+ and Cu2+ ion forms also drastically modified biological activity, from inactive to active, in a cell-based assay of fibroblast growth factor-receptor (FGF2/FGFR1c) signalling and this effect was not reproduced by free cations. These observations could explain the often-contradictory data concerning structure–activity relationships for these derivatives in the literature and, furthermore, argue strongly against the established trend of considering sequence as a complete structural definition. It also provides additional means of modifying the activity of these polysaccharides and suggests a possible additional level of control in biological systems. There are also obvious potential applications for these findings in the biotechnology sphere.
Key words: cations / fibroblast growth factor / heparin / heparan sulfate / synchrotron radiation circular dichroism
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgements/FundingReferences |
|---|
The evolution of multi-cellular organisms presented nature with the need to develop systems capable of encoding the large amounts of information required for inter-cellular communication and co-ordination. The solution was to use carbohydrates, which are structurally several orders of magnitude more complex than proteins or nucleic acids and, therefore, have much higher potential information content.
Heparan sulfate (HS) refers to a family of linear anionic sulfated polysaccharides containing one of the highest levels of structural diversity of any group of molecules expressed in multi-cellular animals and is found on practically all mammalian cells. It is composed of repeating disaccharide units comprising 1,4 linked uronic acid and glucosamine residues. The uronic acid can be ß-D-glucuronic acid (GlcA) or its C-5 epimer, 
-L-iduronic acid (IdoA). These are found either 2-O-sulfated (IdoA2S, or more rarely as GlcA2S) or non-sulfated, while glucosamine can be N-sulfated (GlcNS), N-acetylated (GlcNAc), 6-O-sulfated (GlcN6S) and, more rarely, unsubstituted (Westling et al. 2002
). In addition, lower levels of 3-O-sulfation are found in glucosamine notably, but not exclusively, in the pentasaccharide sequence that is responsible for binding antithrombin (Grootenhuis et al. 1995).
HS has become the focus of intense research following discovery of its involvement in many critical biological processes including cell growth and division, embryogenesis, development and homeostasis (Delehedde et al. 2001; Turnbull et al. 2001; Parish et al. 2006; Stringer et al. 2006). These activities are mediated by interactions with diverse proteins. Among these are the cell signalling complexes of the fibroblast growth factor ligand receptor–tyrosine kinase system (FGF/FGFR) (Ornitz et al. 1996; Gallagher 2006) and other growth factors such as glial derived neurotrophic growth factor (GDNF) (Davies et al. 2003; Rickard et al. 2003), which is implicated in Parkinson's disease as well as kidney development. HS is also an inhibitor of the Alzheimer's disease ß-secretase (BACE-1) (Scholefield et al. 2003) and serves as a means of attachment for many pathogenic micro-organisms (Wadstrom et al. 1999; Vogt et al. 2004: Tiwari et al. 2006).
There is increasing evidence that the biological activities of HS and its close structural relative heparin and its derivatives, which are usually mediated through interactions with proteins, depend on the substitution pattern of O-, N-sulfates, N-acetyl groups and the uronic acid content (Turnbull et al. 2001; Capila and Linhardt 2002; Powell et al. 2004), but this relationship is not straightforward. Typically, biological activity is observed with several, but not all derivatives, suggesting that there is a degree of degeneracy and/or that some other factors, which are not being accounted for experimentally, are involved in vivo. Furthermore, the relationship between substitution pattern, conformation and activity has usually been hard to establish experimentally. Indeed, the wider structural consequences of substitutions at particular positions are often subtle. The conformation of IdoA2S residues is sensitive to the identity of adjacent residues in heparin and heparin fragments (Ferro et al. 1986; Mulloy et al. 1993; Yates et al. 1996) and the glycosidic linkage geometry is also influenced by altered substitution pattern (Yates et al. 2000). However, despite considerable efforts with heparin and some of its derivatives (Mulloy et al. 1993; Hricovini et al. 1995) and the structurally related K-5 polysaccharide (Hricovini et al. 1997) involving in-depth NMR experiments, relatively little is still known regarding the interplay between structural features such as O-, N-sulfation and N-acetylation and their conformation and dynamic consequences. Some cations are known to interact with unmodified heparin in a structure-specific manner, and in a few cases, NMR has been used to probe the consequent structural changes in the presence of various cations (Rabenstein et al. 1995; Angulo et al. 2000). Generally, however, the effect of cations is ignored in relation to biological activity, not least because of the difficulty of monitoring it in even the most well-defined biological systems.
Heparin, the abundant pharmaceutical, is considered by some to be a highly sulfated form of HS; the two polysaccharides have the same underlying structure and are thought to share a common biosynthetic pathway. Heparin is widely used as a proxy for HS in biochemical investigations (Guimond et al. 1993; Patey et al. 2006), because of its relative abundance and the fact that it usually exhibits high activity in processes in which HS is active. Heparin is richer in iduronate residues and, although lacking a domain structure, resembles the highly sulfated S-domains of HS. The activity of heparan sulfate/heparin derivatives is usually considered to stem from their substitution pattern and the outstanding example of this is the pentasaccharide sequence that inhibits factor Xa (Grootenhuis et al. 1995). However, in most other cases, the structure–activity relationship is not so clear-cut. Indeed, in some cases, apparently ambiguous or contradictory results have been reported (Yates et al. 2007). The way in which a particular sequence influences conformation and flexibility and how cations interact to influence these properties and activity is largely unknown. In order to begin to address some of these issues, we have devised strategies for exploring structure–function relationships in these molecules involving libraries of model compounds (Yates et al. 1996, 2004). In the first of these, a series of modified heparin polysaccharide derivatives were generated. Particular substitutions were made throughout the molecule allowing preliminary structure–activity relationships to be explored for a number of activities (Irie et al. 2002; Davies et al. 2003; Scholefield et al. 2003; Patey et al. 2006). This library has also permitted investigations of a range of different spectroscopic techniques, which might help illuminate the solution properties of these biologically important molecules.
One such technique, which is known to be sensitive to conformational changes in glycosaminoglycans, particularly around the uronic acid residues, is circular dichroism (CD) (Stone et al. 1970; Morris et al. 1975; Chung and Ellerton 1976; Buffington et al. 1977; Villanueva et al. 1984; Stone et al. 1985; Grant et al. 1992), and its development which employs synchrotron radiation circular dichroism; SRCD (Wallace 2000). The conformation of heparin derivatives, both in terms of rotation around the glycosidic bonds (Yates et al. 2000) and changes in the conformational equilibria of iduronate-2-O-sulfate residues (Ferro et al. 1986) may reasonably be assumed to influence the overall chain flexibility, although this relationship has not been studied in detail for most derivatives. The nature of the counter-ion associated with these acidic derivatives was also found to alter the appearance of the CD spectra and is known to influence some biological activities (Landt et al. 1994). Earlier work on some heparin derivatives, using CD and NMR (Mulloy et al. 1993) has been reported, but no systematic study of SRCD has been undertaken that combines a range of substitution patterns and cation forms. Here, we have exploited the library of modified heparin derivatives, which serve as a series of simplified, well-defined model polysaccharides for the naturally occurring HS. In this initial study, the polysaccharides were converted to saturation into the appropriate cation form for four common physiologically relevant cations: Na+, K+, Mg2+, Ca2+ and three less well distributed, but nonetheless physiologically present cations; Fe3+, Mn2+ and Cu2+. SRCD offers greatly improved sensitivity and spectra were recorded for each of the eight modified heparin derivatives in each ion form between 175 and 240 nm (Scheme 1, Table I). Data have been used to illustrate particular points throughout the text, but a complete set of SRCD spectra and their analyses are available as Supplementary Data. In selected cases, additional structural information was obtained by NMR and biological activity (signalling through the fibroblast growth factor-receptor tyrosine kinase pathway (FGF/FGFR)) was measured.
|
|
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgements/FundingReferences |
|---|
Origin and nature of SRCD spectra of heparin derivatives
CD arises from the differential interaction between the magnetic transition moment of the chromophore with right and left circularly polarised light, each of which possess a phase difference of +
/2 or –/2 radians, respectively, between their electric and magnetic perturbations. In the UV region, SRCD spectra effectively report differences in the chiral environment of the carboxylic acid and N-acetyl chromophores arising from n to * and to * transitions in the carbonyl bonds (Morris et al. 1975). This established the principles governing CD spectra of uronic acid derivatives between 190 and 260 nm and showed that they were sensitive both to configuration at C-4 and C-5 (for example, D-GlcA and L-IdoA monosaccharides gave very similar spectra, but with opposite sign) and conformation of the uronic acid residues. In heparin, there are three forms of uronic acid residue: IdoA2S, IdoA and GlcA. GlcA and IdoA monosaccharides are known to give essentially identical, but inverted spectra, reflecting their respective configuration and conformation (Morris et al. 1975). It is also well known that IdoA2S residues exhibit conformational equilibria which are influenced by the identity of adjacent residues (Ferro et al. 1986; Mulloy et al. 1993; Yates et al. 2000). The appearance of SRCD spectra of D-GlcA and D-GlcNAc monosaccharides are shown in Figure 1, together with that of unmodified heparin. The spectrum of heparin is not, however, a simple composite of the contributions made by these chromophores. Different polysaccharides in a common cation form exhibited modified peak amplitudes more noticeably than peak positions, while varying the cation form, for given polysaccharides, resulted in both altered peak amplitudes and positions. In selected cases, these properties were investigated further.
|
Substitution pattern of heparin derivatives (with invariant cation) affects SRCD spectral appearance
Taking the first case, in which changes in spectra with varying substitution pattern, but invariant cation were observed, combined with the known conformational stability of glucosamine, particularly GlcNAc (in the relevant derivatives) (Mulloy et al. 1993
; Yates et al. 1996, 2000) and GlcA, demonstrates that conformational changes in the carboxylic acid chromophore (IdoA or IdoA2S) can be induced by other remote structural modifications. A striking example of this is the modification of the otherwise intact heparin, 1, by the substitution of N-sulfate with N-acetyl groups to form N-acetylated derivative, 2 (Figure 2). Somewhat surprisingly, the addition of this chromophore (which, in the absence of other changes, would be expected to result in peak increases around 180 nm and troughs around 210 nm (see Figure 1)), was not directly apparent in the spectrum. Rather, other changes were induced, presumably arising from altered conformation around the carboxyl chromophores of the uronate residues, because the carboxylic acid group contains the only other CD chromophores active at these wavelengths. A change in conformation in the IdoA2S residue was identified using NMR NOE experiments and coupling constant measurements (Hricovini and Tvaroska 1990; Tvaroska and Taravel 1992). The through-space interactions between hydrogen atoms across the glycosidic bonds linking glucosamine to iduronate were clearly distinct between the two derivatives (Figure 3). Altered conformation of the IdoA2S residues was also confirmed by changes in 1JCH coupling constants at I-2 changing from 148 to 154 Hz and at I-3, from 149 to 151 Hz, while 3JHH values changed from 3.0 to <2 Hz between protons at I-1 and I-2 and from 3.6 to 2.9 Hz between those at I-4 and I-5. Several other examples in which altered substitution pattern (e.g. 6-O-de-sulfation of intact heparin) also resulted in modified SRCD spectra were observed and, presumably, these were also the consequence of modified uronate conformation. One comparison, between 5 and 7 (Figure 4), demonstrated that changes in substitution pattern (6-O-sulfation to 6-unsubstituted), which did not introduce or remove a chromophore-containing group, nevertheless altered the spectral appearance, suggesting that a conformational change also occurred in the unsubstituted (IdoA) iduronate residue. Thus, IdoA2S residues appear not to be unique in possessing conformational flexibility in heparin derivatives. This is confirmed by 3JHH coupling constant measurements for 5 and 7; 3JH1-H2 of 2.5 and 1.9 Hz and 3JH4-H5 of 2.5 and 2.4 Hz respectively, corresponding to ratios of chair to skew-boat conformers (1C4:2S0) of 85:15 and 95:5 (Yates et al. 2000). The change in substitution pattern from 5 to 7 also resulted in altered 1JCH coupling constant values at positions adjacent to glycosidic bonds; 1JCH for A-1 changing from 172 to 176 Hz and for I-1 from 147 to 156 Hz, indicating altered glycosidic geometry (Hricovini and Tvaroska 1990; Tvaroska and Taravel 1992).
|
|
|
SRCD spectra of particular polysaccharides vary with different cations
Examples of SRCD spectra in which the identity of the cation was varied for two selected polysaccharide derivatives; 1 and 2 (intact heparin and N-acetylated heparin) are shown in Figure 5A (see Supplementary Data 1 for all derivatives). Changing the associated cation resulted in altered peak position and modified the form of the SRCD curves. All the spectra were analysed by principal component analysis (PCA) and examples of these analyses are shown in Figure 5B for heparin and N-acetylated heparin (see Supplementary Data 2 for all the derivatives).
|
|
The degree of similarity in the SRCD curves for each cation form with a particular derivative has also been analysed by cluster (Table II) and nearest-neighbour analyses (Table III). One striking observation is that the majority of combinations of substitution pattern and cation form can be considered structurally unique at the level detected by SRCD. However, the analysis did reveal some similarities between the spectra of a number of cations, particularly Mg2+ and Ca2+, likewise Fe3+ and Mn2+, while the effects of others, in particular Na+, K+ and Cu2+ were more distinct.
|
|
Different cation forms of heparin derivatives exhibit subtly different conformations as detected by NMR
The four common cation forms (Na+, K+, Mg2+ and Ca2+) of 2-de-O-sulfated heparin derivative (5), which exhibited interesting differential abilties to signal through combinations of growth factors and receptors in a recent study (Guimond et al. 2006
) also exhibited distinct SRCD spectra, and were studied further by NMR. Unlike SRCD, NMR detects structural effects throughout the molecule. 2D heteronuclear single quantum correlation (HSQC) experiments correlating 1H and 13C chemical shift values (Supplementary Data 3) for the Na+ and K+ forms were similar, exhibiting only minor changes (0.01–0.02 ppm in 1H and 0.1–0.2 ppm in 13C) in the chemical shift values of signals arising from the iduronate residues (positions I-2, I-3 and I-5). The Mg2+ form of 5 also exhibited NMR spectra that were similar, but not identical to either the Na+ or K+ forms, showing variations at more positions, including A-2, A-4, I-2 and I-4. A more dramatic change was observed, however, in the Ca2+ form of 5, evident from the 1D 1H spectrum (Supplementary Data 3). In the 1H and 13C spectra, larger chemical shift differences were observed for A-1, A-3, A-4 and I-1, I-2, I-4 and I-5. Some of these, for example in I-1, I-5, A-3 and A-4 signals, were up to 0.1 ppm in 1H spectra and 0.4 ppm in 13C (Supplementary Data 3). NOE experiments (Table IV) confirmed that between the Na+, K+ and Mg2+ forms, the inter-proton distances were similar (the largest difference being a change in NOE of only 2.2% between K+ and Mg2+ forms for the I-1 to A-4 inter-glycosidic distance) both within the iduronate ring and across the glycosidic bonds, but were more distinct for the Ca2+ form. NOEs related to inter-glycosidic geometries, A-1/I-3 (10.5%) and A-1/I-4 (6.5%) exhibited the most significant differences. The magnitude of the NOE enhancement between protons across the iduronate ring were also measured (the distance between protons at A-1 and A-2 was considered invariant), where the ratio between I-5/I-4 divided by I-5/I-2 reports the equilibrium between 1C4 and 2S0 chair and skew-boat conformers, indicated that for Na+, K+ and Mg2+ forms of Ido-2-de-O-sulfated heparin, there were only very slight differences (ratios of 3.3, 3.4 and 3.5). On the other hand, for Ca2+, the ratio was 10.0, indicating a distinct shift in favour of the 1C4 conformation. T2 relaxation time measurements on selected positions (Table V) in (5), for the Na+, K+, Mg2+ and Ca2+ ion forms, revealed that the monovalent cation forms (Na+/K+) possessed similar values, but on binding either of the divalent cations, Mg2+ or Ca2+, the polysaccharide became more rigid.
|
|
Changing the associated cation of particular heparin derivatives can alter conformation and drastically modify activity
The possibility that structural differences arising from cation binding might result in altered activity was also investigated. Several of the cation forms of 5 were tested for their ability to support FGF/FGFR stimulated cell proliferation using the BaF3 cell assay, one of the experimentally most well-defined signalling systems available. BaF3 cells lack endogenous HS and FGFR and, when transfected with cDNA encoding a specific FGFR, FGF-stimulated cell proliferation occurs only when both a FGF and saccharide capable of supporting formation of a functioning signalling complex are supplied via the culture medium. This system, while not strictly an in vivo experimental system, does offer the advantage of reporting signalling in living cells, as opposed to binding events, which do not correlate well with activity (Yates et al. 2007
). An example of the dramatic effect on the activity of changing cation form is shown in Figure 6. Intact heparin (1) (as a positive control) together with derivative 5, in the Na+, K+, Mg2+, Ca2+ or Cu2+ forms, were tested for their ability to support signalling with FGF-2/FGFR1c. Derivative 5 showed a dramatic change from inactive to active when converted from the Na+ to the Cu2+ form. Furthermore, the effect was not due to Cu2+ ions in free solution affecting signalling; free Cu2+ ions (in both the absence and presence of heparin derivatives) at comparable concentrations resulted in cell death (data not shown). Optical biosensor inhibition experiments of 5 in the Na+ and Cu2+ forms showed no significant difference in their ability to inhibit binding of FGF-2 to a heparin surface derivatised with 1 (data not shown). Binding of Cu2+ ions altered the conformation of the saccharide (as evinced by distinct SRCD spectra) and one hypothesis is that this then enables it to support the formation of a functional complex but, the effect is clearly not due to a difference in binding ability to FGF-2. The mechanism of signalling in the FGF/FGFR tyrosine kinase pathway has not been unequivocally established and there are other possible causes, arising for example from effects on the receptor. Addition of free Cu2+ ions to the cell-culture medium with or without saccharide (but not bound to it initially) did not provide the same change in activity. The heparin derivatives must, therefore, be directly involved, if only as chaperones. The effects of altering cation forms are illustrated for Na+, K+, Mg2+, Ca2+ and Cu2+, some of which exhibited distinct activities. These effects may be of wider biological relevance given that, for example, the K+ form of Ido-2-de-O-sulfated heparin exhibited contrasting activity to the other forms. Some of these forms had very similar conformations (e.g. by NMR) suggesting that, in this case, the effects on the sugar and/or the proteins involved may be subtle and could involve the combined effects of sequence and cation, possibly modifying protein binding and/or structure. The concentration of K+ ions in mammalian physio- logy range from 3.5–5 mM extracellularly to 150 mM intracellularly, while Cu2+ ions in human serum normally vary between 10 and 25 µM (Trumbo et al. 2002) and are thought to be bound to a number of carrier proteins. However, it has long been known that Cu2+ ions can influence angiogenic processes, particularly in the presence of heparin (Alessandri et al. 1983, 1984). Cu2+ levels also become elevated during tumourogenesis (Goodman et al. 2005; Lowdnes et al. 2005) and it has been implicated in prion diseases (Gabizon et al. 1993), which also involve HS. It is notable that heparin derivatives were capable of retaining a substantial proportion of Cu2+ ions following exposure to cell culture medium containing serum and that Mg2+ ions were the most avid in replacing Cu2+. The biological effects seen for Cu2+ were not, however, reproduced by derivatives 1 and 5 in the Mg2+ form (data not shown) and cannot therefore be attributed to the exchange of Cu2+ by Mg2+ ions.
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgements/FundingReferences |
|---|
SRCD spectra of heparin derivatives between 175 and 240 nm are very sensitive to the chiral environment of the carbonyl groups of uronic acids and SRCD therefore probes selected functional groups that occur in well defined, but restricted locations, revealing conformational factors that are not readily accessed by other techniques. These findings suggest that changes in the chiral environment of the carboxylic acid group dominate the SRCD spectra and that the binding of almost every cation to each derivative results in a subtly distinct structure. Previous NMR studies have shown that proportions of conformers of iduronic acid are sensitive to sequence and sulfation pattern (Ferro et al. 1990
) and that in the antithrombin binding pentasaccharide sequence, distinct flexibilities can be detected for residues in different positons (Ragazzi et al. 1990). NMR studies on one selected example, Ido-2-de-O-sulfated heparin (5), showed that the changes in conformation detected by SRCD varied from slight, e.g. between Na+ and Mg2+ forms, to more substantial (in the case of the Ca2+ form) and also influenced the chain flexibility. If we permit the reasonable assumption that the conformation of neither the GlcA nor GlcN residues deviate significantly from the 4C1 chair conformation (Yates et al. 1996, 2000), the results presented here suggest that the conformation of IdoA2OH residues, like their IdoA2S counterparts, also change with substitution pattern in adjacent residues. The overall conformational consequences of these changes for the polysaccharide are not necessarily simple or predictable, but altered 1JCH coupling constants at the anomeric positions are consistent with influences on overall polysaccharide characteristics. This point emphasises that the structure–function relationships of these compounds should not be treated as though the carbohydrate scaffold were rigid and only variations in N-O-sulfate and N-acetyl appendages occur. In fact, changing the substitution pattern can influence the conformation and flexibility of the whole or particular regions of the molecule.
Principle component analysis (PCA) of the SRCD spectra also revealed some similarities in a limited number of instances. The most similar spectra were Ca2+/Mg2+ and Fe3+/Mn2+ forms. In the former pair, the similarities in ionic radii, charge and coordination geometry appear not to be strongly differentiated by SRCD for the structures tested. In the latter, although Fe3+ and Mn2+ have different charges, they both possess similar electron configuration (d5) in their outer electron shell, comparable effective ionic radii (0.61–0.77 nm and 0.67–0.82 nm, respectively) and similar octahedral coordination geometries (Bell 1977). More surprising perhaps, is the finding that the SRCD spectra of Na+ and K+ forms of these derivatives appeared distinct while NMR showed only slight differences in uronate conformation, emphasising the fundamentally distinct origins and sensitivities of these two spectroscopic techniques. These subtle changes may be due to the different effective ionic radii (0.102 nm and 0.138 nm respectively) of these two monovalent ions interacting with the carboxylic acid chromophore, as well as reflecting the ability of K+ ions to adopt both cubic and octahedral coordination and this latter property may be involved directly in the interaction between polysaccharide and protein. The more distinct properties of Ca2+ probably result from its effective ionic radius (0.10 nm) differing from that of Mg2+ (0.072 nm) and may also reflect particular spacings between charged groups in heparin derivatives.
One consequence of the binding of these divalent ions is their effect on the flexibility of the polysaccharides. NMR T2 relaxation time measurements of the Na+, K+, Mg2+ and Ca2+ cation forms of derivative 5 indicated that there was significant stiffening of the polysaccharide on binding divalent cations, but little difference between ion forms of the same valency. The progressive nature of the variation of structural changes with cation type, most marked for the Ca2+ form (Rabenstein et al. 1995), are also noteworthy and it is possible that a degree of cross-linking may occur between divalent forms of the polysaccharides. These results also suggest that not only are cations alone capable of altering ring and linkage geometries to different extents, but they may also impose modified dynamics on the polysaccharide chains, although T2 relaxation times would need to be measured at two different field strengths for a more detailed analysis to be made. Either of these factors, or their combined effect, could have consequences for activity via altered binding energies that could consist of both enthalpic and entropic contributions. An ability to control the attachment of particular cations (either by delivery of selected cations to the polysaccharide, or through their innate binding characteristics) would, therefore, appear to offer the means of modifying the properties of the polysaccharides, both conformational and dynamic, which could lead to altered activity and may be of biological significance, because of the preponderance of these ions in biological systems.
Cu2+ derivatives showed markedly distinct SRCD spectra and Cu2+ presents additional opportunities from the experimental point of view, which are not accessible for Na+, K+, Mg2+ or Ca2+, although its paramagnetic nature excludes its heparin derivatives from accurate dynamic and geometric measurements. Magnetic resonance techniques are sensitive to the line broadening effect of Cu2+ ions on signals arising from nuclei in their proximity and, in favourable cases, this can be used to help deduce its mode of binding. Experiments relating viscosity to cation form for heparin, but not chemically modified derivatives, have been reported (Chung and Ellerton 1976), in which viscosity increased with the cation forms; Na+, K+, Mg2+ and Ca2+. This work also assumed that Cu2+ ions bound to the N-sulfate groups initially, but without causing any conformational changes. The present work challenges this assumption and NMR (Supplementary Data 4) revealed that Cu2+ ions bind heparin (1), preferentially at IdoA2S-GlcNS6S sequences, specifically incorporating four groups; the carboxylic acid oxygen, the ring oxygen of IdoA2S, the glycosidic oxygen linking IdoA2S to GlcNS6S and the neighbouring 6-O-sulfate. In contrast, in N-acetylated heparin (2), the ions bind between the carboxylic acid group and the adjacent N-acetyl group (Supplementary Data 5), while in Ido de-O-sulfated heparin (derivative 5), Cu2+ ions bind in a more complex fashion (Supplementary Data 6). The particular electron configuration of the outer electron shell of Cu2+ ions (d9) renders them susceptible to changes in coordination geometry when binding ligands compared to their octahedral arrangement when free in aqueous solution. In heparin (1), the NMR spectral changes, involving interaction with the four groups mentioned above, may indicate a tetragonal arrangement, consistent with the Jahn–Teller effect (Jahn and Teller 1937), in which an unfavourable strain in the coordination shell is relieved by adoption of distorted tetragonal coordination geometry.
The consequences of cation binding by heparin on its biological and medical properties are known to be significant (Landt et al. 1994), but were particularly dramatic for K+ and Cu2+ in cell signalling through the FGF/FGFR system on its 2-de-O-sulfated derivative 5. Possible explanations are that the bound ion may be involved in forming the signalling complex, but only when delivered by particular heparin derivatives and not as a free ion in solution), or may induce a newly active conformation, or both. The effect is clearly not due to the influence of free Cu2+ ions in the experimental culture medium, however, because free Cu2+ does not induce these effects for any of the derivatives tested nor does the Cu2+ form of 5 have improved FGF-2 binding properties (data not shown). Moreover, at comparable levels, free Cu2+ ions cause cell death.
It is interesting to consider the role of monovalent sulfate and carboxylic acid groups in these complex polysaccharides. Sulfate groups are capable of binding monovalent cations of different radii and coordination geometry (with other, non-charged parts of the molecule and/or surrounding water molecules, or other molecules) but, importantly, each sulfate group is able to neutralise only one of the charges of a divalent cation, providing a mechanism whereby divalent cation binding can occur across glycosidic linkages to affect the orientation and conformation of neighbouring residues and the intervening glycosidic linkages. The effects of this on geometry and flexibility are clearly subtle, but distinct for different cations. This can give rise to a scale of conformational changes and flexibility that is ultimately determined by a combination of substitution pattern and the identity of the associated cations. The consequences of these structural changes on activity can also be dramatic and manipulation of such properties could provide an additional layer of control in biological systems.
A major conclusion, therefore, is that the sequence alone does not define conformation or flexibility for this class of molecules; the associated cations also need to be considered. For the heparin derivatives studied here, the presence of particular cations almost always gave rise to unique structures and was capable of grossly altering activity in cell-based signalling assays, even to the extent of converting former inactive saccharides into active ones. It is conceivable that the combined use of substitution patterns and various cations results in an increased repertoire of structures, but the extent to which these are utilised in biological systems, or indeed, whether a high level of degeneracy exists, remains unknown. The variations in SRCD spectra for different derivatives and cation forms will also complicate attempts to study protein secondary structural changes in protein–heparin complexes by SRCD (Yates et al. 2006) because their dependence on substitution pattern and cation form renders subtraction of their spectra from that of the complex invalid. The findings presented here have important ramifications for the way that heparin, heparan sulfate, their derivatives and activities are viewed and may explain some of the contradictory structure–activity data reported in the literature. Exploitation of these phenomena could provide an additional means of controlling them for biotechnological or tissue engineering purposes.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgements/FundingReferences |
|---|
Materials
Recombinant human FGF-1 and murine recombinant IL-3 were purchased from R&D Systems Europe (Abingdon, Oxon, UK). BaF3 cells transfected with FGFR1c were kindly provided by David Ornitz (Washington University, St. Louis, MO) (Orniz et al. 1996
). Porcine intestinal mucosa heparin was purchased from Celsus Labs (Cincinnati, OH). 3-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT), Fluka HPLC grade water, CuCl2 and Dowex 50W X8–400 cation exchange resin were purchased from Sigma-Aldrich (Gillingham, UK). RPMI-1640, foetal calf serum and 100x penicillin-G/streptomycin sulfate/L-glutamine were purchased from Invitrogen (Paisley, UK).
Polysaccharides
The starting material for all chemical modifications was porcine intestinal mucosal heparin (Celsus Labs, Cincinnati, OH; lot PH-42800). Modified polysaccharides were prepared as described and characterised by 1H and 13C NMR (Yates et al. 1996) and disaccharide analysis (Patey et al. 2006). Each polysaccharide was subjected to size-exclusion chromatography on TSK gel G2000SWXL (7.8 mm x 30 cm with 0.5 µm particle size, Supelco) eluting with water at 1 mL/min and detecting at 190 nm. All samples exhibited a single major peak, with very similar retention times (mean, 6.07 min; 
n-1 = 0.05). The SRCD spectra have been converted to molar ellipticity using a conversion factor, based on their precise content as previously determined, to calculate relative molecular weight (Patey et al. 2006; Skidmore et al. 2006). Polysaccharides were transformed into the appropriate cation form to saturation (three additions over 4 h in HPLC grade water with stirring) using Dowex (50W X8–400) cation exchange resin, which had previously been converted from the acid form following exhaustive washing with the relevant chloride salt (1 M). Full characterisation details are given in supplementary (Supplementary Data 7).
BaF3 proliferation assay
BaF3 cells, which do not produce heparan sulfate, expressing FGF receptor 1c (FGFR1c) isoform were maintained as previously described (Ornitz et al. 1996). For proliferation assays, 10,000 cells per well were plated in a 96-well format with 100 µL of culture medium (RPMI-1640 supplemented with 10% foetal calf serum, 2 mM L-glutamine, 100 U/mL penicillin-G and 100 µg/mL streptomycin sulfate, without IL-3) and incubated with 2 ng/mL IL-3, 1 nM FGF-2 and heparin derivatives at the indicated concentrations. After 72 h at 37°C, MTT was added to measure cell proliferation (Orniz et al. 1996). The cells are presented with an exogenous FGF and a test saccharide (in this case, in a particular cation form). Cell proliferation occurs only when the test sugar is competent to form a functional signalling complex with the growth factor and receptor. In the case of the assays with Na+ and Cu2+ saccharides, the effects observed were not due to Cu2+ ions leaching from the saccharide. Comparable levels of Cu2+ ions added to the culture medium without the saccharide and added at the same time as the saccharide (but not bound) resulted in cell death. Furthermore, both heparin (1) and ido-2-de-O-sulfated heparin (5) were found to retain Cu2+ ions following exposure to the culture medium. The proportion of ions that were identified by elemental analysis as Cu2+ were 78% before exposure to media and 62% afterwards for 1, and 63 and 43% for 5. For 1, the ions that replaced Cu2+ from the media were Mg2+, Ca2+, K+ and Na+ in the relative proportions 15:11:5:1 and for 5, Mg2+, K+, Ca2+ and Na+ in the proportions 17:7:4:3. Cation treated polymers in the culture medium were also shown to have suffered no significant depolymerisation resulting from the Fenton reaction. This requires a source of transition metal ion, free radicals and a reducing agent to function efficiently. The redox conditions throughout the assay were monitored by reference to a colourimetric assay using the redox sensitive dye resazurin, calibrated by reference to a set of standard solutions poised to defined values of Eh7 using cysteine (Goldner et al. 1997). The redox conditions were found to remain clearly oxidising (greater than +150 mV) throughout.
SRCD
SRCD spectra were recorded on the CD-12 beam line at Daresbury Laboratory, a purpose built SRCD facility, using quartz sample cell 0.02 cm path length, 1 nm resolution), between 240 nm and 175 nm. CD spectra of sugar samples were recorded at 10 mg/mL and are relative to (+)-10-camphorsulfonic acid (1.0 mg/mL). CD values are presented as molar circular dichroism (per mole per cm) and have been corrected to allow for differences in molecular weight according to composition analysis (Patey et al. 2006; Skidmore et al. 2006).
NMR
13C and 1H NMR spectra together with 2D 1H-1H NOE, COSY, TOCSY and HSQC experiments were recorded as previously described (Yates et al. 1996, 2000). Chemical shift values were measured downfield from trimethylsilyl propionate sodium salt as standard at 308 K. For 13C measurements, 500 mg of polysaccharide was dissolved in 2.5 mL D2O. All proton-detected spectra were measured at 308 K on a Bruker Avance 600 instrument equipped with cryogenic TXI 5 mm probe. The 2D-NOESY spectra were measured using the Bruker library sequence with mixing times of 250 ms and pre-saturation of the residual HOD signal. The 2D experiments matrix size of 2000 x 320 was zero filled to 4000 x 2000 by application of a squared cosine function before Fourier transformation. Spin–spin carbon relaxation times were determined using a pseudo-3D version of a modified double INEPT sequence, with suppression of the effects of cross-relaxation between dipolar and chemical shift anisotropy relaxation mechanisms (Kay et al. 1992; Dayie et al. 1994). The CPMG pulse trains were 22.4, 44.8, 89.6, 112, 134.4, 179.2, 224 and 268.8 ms. A shifted sine-bell-squared function was applied before Fourier transformation of each T2 experiment. The cross-peak volumes were measured using Bruker Topspin 2.0 software.
Principal component analysis (PCA) and cluster analysis
PCA was conducted using SPSS 13.0 (SPSS UK Ltd., Woking, UK) and the analysis was limited to two components. Cluster analysis was performed using Minitab (Minitab Ltd., Coventry, UK) with single linkage and Euclidean distances were measured.
Supplementary data for this article is available online at www.glycob.oxfordjournals.org
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgements/FundingReferences |
|---|
None declared.
|
Acknowledgements/Funding |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgements/FundingReferences |
|---|
The Biotechnology and Biological Sciences Research Council (SCIBS grant and BB/D020794/1), Cancer and Polio Research Fund, Human Frontiers Science Programme and the North West Cancer Research Fund for financial support are thanked for the provision of funding.
|
Abbreviations |
|---|
BACE-1, beta-site APP-cleaving enzyme 1; CD, circular dichroism; CD-12, beamline-12 at Daresbury Laboratory; CPMG, Carr-Purcell-Meiboom-Gill; FGF, fibroblast growth factor; FGFR, fibroblast growth factor receptor; GDNF, glial cell derived neurotrophic factor; GlcA, D-glucuronic acid; GlcA2S, 2-sulfated D-glucuronic acid; GlcN, D-glucosamine; GlcNAc, N-acetyl-D-glucosamine; GlcNS, N-sulfated D-glucosamine; HPLC, high pressure liquid chromatography; HS, heperan sulfate; HSQC, heteronuclear single quantum correlation; IdoA, L-iduronic acid; IdoA2S, 2-sulfated L-iduronic acid; IL-3, Interleukin-3; INEPT, insensitive nuclei enhanced by polarization transfer; K-5, E. coli capsular polysaccharide; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; NMR, nuclear magnetic resonance; NOE, nuclear Overhauser effect; PCA, principle component analysis; RPMI, Roswell Park Memorial Institute medium; SRCD, synchrotron radiation circular dichroism; T2, transverse relaxation time
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementAcknowledgements/FundingReferences |
|---|
Alessandri G, Raju K, Gullino PM. Angiogenesis in vivo and selective mobilization of mapillary endothelium in vitro by heparin-copper complex. Microcirc Endoth Lym (1984) 1(3):329–346.
Alessandri G, Raju K, Gullino PM. Selective mobilization of capillary endothelium in vitro and induction of angiogenesis in vivo by heparin-copper complex. P Am Assoc Canc Res (1983) 24:22–22.
Angulo J, De Paz JL, Nieto PM, Martin-Lomas M. Interaction of heparin with Ca2+: A model study with a synthetic heparin-like hexasaccharide. Isr J Chem (2000) 40(3–4):289–299.[CrossRef]
Arndt ER, Stevens ES. Vacuum-ultraviolet circular-dichroism studies of simple saccharides. J Am Chem Soc (1993) 115(17):7849–7853.[CrossRef][Web of Science]
Arndt ER, Stevens ES. Vacuum ultraviolet circular dichroism of gellan-family polymer films from water and dimethyl sulfoxide. Carbohydr Res (1996) 280(1):15–26.[CrossRef][Web of Science][Medline]
Bell CF. Principles and applications of metal chelation. (1977) Oxford, UK: Oxford University Press.
Buffington LA, Pysh ES, Chakrabarti B, Balazs EA. Far UV circular-dichroism of N-acetylglucosamine, glucuronic acid, and hyaluronic-acid. J Am Chem Soc (1977) 99(6):1730–1734.[CrossRef][Web of Science][Medline]
Capila I, Linhardt RJ. Heparin – protein interactions. Angew Chem Int Ed (2002) 41(3):391–412.
Chung MCM, Ellerton NF. Viscosity at low shear and circular-dichroism studies of heparin. Biopolymers (1976) 15(7):1409–1423.[CrossRef][Web of Science][Medline]
Davies JA, Yates EA, Turnbull JE. Structural determinants of heparan sulphate modulation of GDNF signalling. Growth Factors (2003) 21(3–4):109–119.[CrossRef][Web of Science][Medline]
Dayie KT, Wagner G. Relaxation-rate measurements for N-15-H-1 groups with pulsed-field gradients and preservation of coherence pathways. J Magn Reson Ser A (1994) 111(1):121–126.[CrossRef][Web of Science]
Delehedde M, Lyon M, Sergeant N, Rahmoune H, Fernig DG. Proteoglycans: pericellular and cell surface multireceptors that integrate external stimuli in the mammary gland. J Mammary Gland Biol Neoplasia (2001) 6(3):253–273.[CrossRef][Web of Science][Medline]
Ferro DR, Provasoli A, Ragazzi M, Torri G, Casu B, Gatti G, Jacquinet JC, Sinay P, Petitou M, Choay J. Evidence for conformational equilibrium of the sulphated L-iduronate residue in heparin and in synthetic heparin monosaccharides and oligosaccharides – NMR and force-field studies. J Am Chem Soc (1986) 108(21):6773–6778.[CrossRef][Web of Science]
Ferro DR, Provasoli A, Ragazzi M, Casu B, Torri G, Bossennec V, Perly B, Sinay P, Petitou M, Choay J. Conformer populations of L-iduronuc acid residues in glycosaminoglycan sequences. Carbohydr Res (1990) 195:157–167.[CrossRef][Web of Science][Medline]
Gabizon R, Meiner Z, Halimi M, Bensasson SA. Heparin-like molecules bind differentially to prion-proteins and change their intracellular metabolic-fate. J Cell Physiol (1993) 157(2):319–325.[CrossRef][Web of Science][Medline]
Gallagher JT. Heparan sulphate: Growth control with a restricted sequence menu. J Clin Invest (2001) 108(3):357–361.[CrossRef][Web of Science][Medline]
Gallagher JT. Multiprotein signalling complexes: Regional assembly on heparan sulphate. Biochem Soc Trans (2006) 34:438–441.[CrossRef][Web of Science][Medline]
Goldner M, Mingot N, Emond JP, Dublanchet A. Influence of different levels of redox potential on fermentative products formed by bacteroides fragilis. Clin Infect Dis (1997) 25:S147–S150.[Web of Science][Medline]
Goodman VL, Brewer GJ, Merajver SD. Control of copper status for cancer therapy. Curr Cancer Drug Targets (2005) 5(7):543–549.[CrossRef][Web of Science][Medline]
Grant D, Long WF, Moffat CF, Williamson FB. Cu2+-heparin interaction studied by polarimetry. Biochem J (1992) 283:243–246.[Web of Science][Medline]
Grootenhuis PDJ, Westerduin P, Meuleman D, Petitou M, Vanboeckel CAA. Rational design of synthetic heparin analogs with tailor-made coagulation-factor inhibitory activity. Nat Struct Biol (1995) 2(9):736–739.[CrossRef][Web of Science][Medline]
Guimond S, Maccarana M, Olwin BB, Lindahl U, Rapraeger AC. Activating and inhibitory heparin sequences for Fgf-2 (Basic Fgf) – Distinct Requirements for Fgf-1, Fgf-2, and Fgf-4. J Biol Chem (1993) 268(32):23906–23914.
Guimond SE, Turnbull JE, Yates EA. Engineered bio-active polysaccharides from heparin. Macromol Biosci (2006) 6:681–686.[CrossRef][Web of Science][Medline]
Hricovini M, Guerrini M, Torri G, Casu B. Motional properties of E. coli polysaccharide K5 in aqueous solution analyzed by NMR relaxation measurements. Carbohydr Res (1997) 300(1):69–76.[CrossRef][Web of Science][Medline]
Hricovini M, Guerrini M, Torri G, Piani S, Ungarelli F. Conformational-analysis of heparin epoxide in aqueous-solution – an NMR relaxation study. Carbohydr Res (1995) 277(1):11–23.[CrossRef][Web of Science][Medline]
Hricovini M, Tvaroska I. Conformational dependence of the one-bond carbon proton coupling-constants in oligosaccharides. Magn Res Chem (1990) 28(10):862–866.[CrossRef]
Ilan N, Elkin M, Vlodavsky I. Regulation, function and clinical significance of heparanase in cancer metastasis and angiogenesis. Int J Biochem Cell Biol (2006) 38(12):2018–2039.[CrossRef][Web of Science][Medline]
Irie A, Yates EA, Turnbull JE, Holt CE. Specific heparan sulphate structures involved in retinal axon targeting. Development (2002) 129(1):61–70.
Jahn H, Teller E. Stability of polyatomic molecules in degenerate electronic states. I. Orbital Degeneracy. Proc R Soc A (1937) 161(905):220–235.
Jakobson AM, Hahnenberger R. Antiangiogenic effect of heparin and other sulphated glycosaminoglycans in the chick-embryo chorioallantoic membrane. Pharmacol Toxicol (1991) 69(2):122–126.[Web of Science][Medline]
Kay LE, Nicholson LK, Delaglio F, Bax A, Torchia DA. Pulse sequences for removal of the effects of cross-correlation between dipolar and chemical-shift anisotropy relaxation mechanism on the measurement of heteronuclear T1 and T2 values in proteins. J Magn Reson (1992) 97(2):359–375.[Web of Science]
Landt M, Hortin GL, Smith CH, Mcclellan A, Scott MG. Interference in ionized calcium measurements by heparin salts. Clin Chem (1994) 40(4):565–570.
Lowndes SA, Harris AL. The role of copper in tumour angiogenesis. J Mammary Gland Biol (2005) 10(4):299–310.[CrossRef][Web of Science][Medline]
Morris ER, Rees DA, Sanderson GR, Thom D. Conformation and circular-dichroism of uronic acid residues in glycosides and polysaccharides. J Chem Soc, Perkin Trans (1975) 2(13):1418–1425.
Mulloy B, Forster MJ, Jones C, Davies DB. NMR and molecular-modeling studies of the solution conformation of heparin. Biochem J (1993) 293:849–858.[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 modeling study. Carbohydr Res (1994) 255:1–26.[CrossRef][Web of Science][Medline]
Ornitz DM, Xu JS, Colvin JS, McEwen DG, MacArthur CA, Coulier F, Gao GX, Goldfarb M. Receptor specificity of the fibroblast growth factor family. J Biol Chem (1996) 271(25):15292–15297.
Parish CR. The role of heparan sulphate in inflammation. Nature Rev Immun (2006) 6(9):633–643.[CrossRef][Medline]
Patey SJ, Edwards EA, Yates EA, Turnbull JE. Heparin derivatives as inhibitors of BACE-1, the Alzheimer's beta-secretase, with reduced activity against factor Xa and other proteases. J Med Chem (2006) 49(20):6129–6132.[CrossRef][Web of Science][Medline]
Powell AK, Yates EA, Fernig DG, Turnbull JE. Interactions of heparin/heparan sulphate with proteins: Appraisal of structural factors and experimental approaches. Glycobiology (2004) 14(4):17r–30r.[CrossRef][Web of Science][Medline]
Rabenstein DL, Robert JM, Peng J. Multinuclear magnetic resonance studies of the interaction of inorganic cations with heparin. Carbohydr Res (1995) 278(2):239–256.[CrossRef][Web of Science][Medline]
Ragazzi M, Ferro DR, Perly B, Sinay P, Petitou M, Choay J. Conformation of the pentasaccharide corresponding to the binding site of heparin for antithrombin III. Carbohydr Res (1990) 195:169–185.[CrossRef][Web of Science][Medline]
Rej RN, Holme KR, Perlin AS. Marked stereoselectivity in the binding of copper ions by heparin – contrasts with the binding of gadolinium and calcium-ions. Carbohydr Res (1990) 207(2):143–152.[CrossRef][Web of Science][Medline]
Rickard SM, Mummery RS, Mulloy B, Rider CC. The binding of human glial cell line-derived neurotrophic factor to heparin and heparan sulphate: Importance of 2-O-sulphate groups and effect on its interaction with its receptor, GFR alpha 1. Glycobiology (2003) 13(6):419–426.
Scholefield Z, Yates EA, Wayne G, Amour A, McDowell W, Turnbull JE. Heparan sulphate regulates amyloid precursor protein processing by BACE1, the Alzheimer's beta-secretase. J Cell Biol (2003) 163(1):97–107.
Skidmore MA, Guimond SE, Dumax-Vorzet AF, Atrih A, Yates EA, Turnbull JE. High sensitivity separation and detection of heparan sulphate disaccharides. J Chromatogr A (2006) 1135(1):52–56.[CrossRef][Web of Science][Medline]
Stevens ES, Morris ER, Rees DA, Sutherland JC. Synchrotron light-source applied to measuring the vacuum ultraviolet circular-dichroism of heparin. J Am Chem Soc (1985) 107(10):2982–2983.[CrossRef][Web of Science]
Stone AL. Far-ultraviolet circular-dichroism and uronic-acid components of anticoagulant decasaccharide, dodecasaccharide, tetradecasaccharide, and octadecasaccharide heparin fractions. Arch Biochem Biophys (1985) 236(1):342–353.[CrossRef][Web of Science][Medline]
Stone AL, Constant G, Sotsky SM, Dekaban A. Optical rotatory dispersion of mucopolysaccharides. 4. Optical rotatory dispersion and circular dichroism of glycosaminoglycans and heparan sulphate fractions from urine of patients with mucopolysaccharidosis (Hurler-Syndrome). Biochim Biophys Acta (1970) 222(1):79–89.[Medline]
Stringer SE. The role of heparan sulphate proteoglycans in angiogenesis. Biochem Soc Trans (2006) 34:451–453.[CrossRef][Web of Science][Medline]
Tiwari V, Clement C, Ding X, Valyi-Nagy T, Yue BYJT, Jian L, Shukla D. Role for 3-O-sulphated heparan sulphate as the receptor for herpes simplex virus type 1 entry into primary human corneal fibroblasts. J Virol (2006) 80(18):8970–8980.
Turnbull JE, Powell A, Guimond SE. Heparan sulphate: Decoding a dynamic multifunction nal cell regulator. Trends Cell Biol (2001) 11(2):75–82.[CrossRef][Web of Science][Medline]
Trumbo P, Schlicker S, Yates AA, Poos M. Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein and amino acids. J Am Diet Assoc (2002) 102(11):1621–1630.[CrossRef][Web of Science][Medline]
Tvaroska I, Taravel FR. One-bond carbon proton coupling-constants–angular-dependence in beta-linked oligosaccharides. J Biomol NMR (1992) 2(5):421–430.[CrossRef][Web of Science]
Varela-Nallar L, Toledo EM, Chacon MA, Inestrosa NC. The functional links between prion protein and copper. Biol Res (2006) 39(1):39–44.[Web of Science][Medline]
Villanueva GB, Allen N. Effect of heparin modification on its circular-dichroism spectrum. Biochem Biophys Res Commun (1984) 123(3):973–980.[CrossRef][Web of Science][Medline]
Vogt AM, Winter G, Wahlgren M, Spillmann D. Heparan sulphate identified on human erythrocytes: A plasmodium falciparum receptor. Biochem J (2004) 381:593–597.[CrossRef][Web of Science][Medline]
Wadstrom T, Ljungh A. Glycosaminoglycan-binding microbial proteins in tissue adhesion and invasion: Key events in microbial pathogenicity. J Med Microbiol (1999) 48(3):223–233.
Wallace BA. Conformational changes by synchrotron radiation circular dichroism spectroscopy. Nat Struct Biol (2000) 7(9):708–709.[CrossRef][Web of Science][Medline]
Westling C, Lindahl U. Location of N-unsubstituted glucosamine residues in heparan sulphate. J Biol Chem (2002) 277(51):49247–49255.
Yates EA, Guimond SE, Turnbull JE. Highly diverse heparan sulphate analogue libraries: Providing access to expanded areas of sequence space for bioactivity screening. J Med Chem (2004) 47(1):277–280.[CrossRef][Web of Science][Medline]
Yates EA, Rudd TR, Ducesne L, Skidmore MA, Nichols RJ, Clarke DT, Fernig DG. Revealing the hidden dimensions of GAGs in solution and in protein complexes with saccharide libraries, FTIR and SRCD. In: Glycobiology—Samson C, Markham O, eds. (2007) Bloxham, UK: Scion.
Yates EA, Santini F, De Cristofano B, Payre N, Cosentino C, Guerrini M, Naggi A, Torri G, Hricovini M. Effect of substitution pattern on H-1, C-13 NMR chemical shifts and (1)J(CH) coupling constants in heparin derivatives. Carbohydr Res (2000) 329(1):239–247.[CrossRef][Web of Science][Medline]
Yates EA, Santini F, Guerrini M, Naggi A, Torri G, Casu B. H-1 and C-13 NMR spectral assignments of the major sequences of twelve systematically modified heparin derivatives. Carbohydr Res (1996) 294:15–27.[Web of Science][Medline]
Yates EA, Terry CJ, Rees C, Rudd TR, Duchesne L, Skidmore MA, Levy R, Thanh NTK, Nichols RJ, Clarke DT, Fernig DG. Protein-GAG interactions: New surface-based techniques, spectroscopies and nanotechnology probes. Biochem Soc Trans (2006) 34:427–430.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. M. Thompson, D. G. Fernig, E. C. Jesudason, P. D. Losty, E. M. A. van de Westerlo, T. H. van Kuppevelt, and J. E. Turnbull Heparan Sulfate Phage Display Antibodies Identify Distinct Epitopes with Complex Binding Characteristics: INSIGHTS INTO PROTEIN BINDING SPECIFICITIES J. Biol. Chem., December 18, 2009; 284(51): 35621 - 35631. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Scarpellini, R. Germack, H. Lortat-Jacob, T. Muramatsu, E. Billett, T. Johnson, and E. A. M. Verderio Heparan Sulfate Proteoglycans Are Receptors for the Cell-surface Trafficking and Biological Activity of Transglutaminase-2 J. Biol. Chem., July 3, 2009; 284(27): 18411 - 18423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. R Rudd, M. A Skidmore, S. E Guimond, C. Cosentino, G. Torri, D. G Fernig, R. M Lauder, M. Guerrini, and E. A Yates Glycosaminoglycan origin and structure revealed by multivariate analysis of NMR and CD spectra Glycobiology, January 1, 2009; 19(1): 52 - 67. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||