Structural Aspects of Heparin Responsible for Interactions With von Willebrand Factor
Abstract Unfractionated heparin (UFH) binds von Willebrand factor (vWF) and inhibits the vWF–platelet GP Ib interaction. For vWF, a heparin-binding domain has been identified, but for heparin, the structures that confer such activity are unknown. To investigate this, UFH was depolymerized by methods that yield structurally distinct fragments. The glycosaminoglycans (GAGs) produced were separated into five groups of homogeneous molecular weight (MW). Anti-Xa activity, vWF binding affinity, and vWF-dependent platelet agglutination were measured. Periodate oxidation but not heparinase digestion destroyed anti-Xa activity. At all MWs, periodate conferred greater vWF binding affinity and greater ability to inhibit platelet agglutination than heparinase. As an example, at MW 6100, the binding IC50 was 100±19 μmol/L for a periodate-derived GAG and 527±70 μmol/L for a heparinase-derived GAG. At the same MW, the agglutination IC50 was 17±5 μmol/L for periodate and 135±18 μmol/L for heparinase. This suggests that the disaccharide GlcNS[6S]-IdoA2S, destroyed by heparinase but not periodate, is crucial to heparin-vWF interactions. An MW dependency was also noted, with a minimum dodecasaccharide required for activity inhibition. To further investigate the heparin/vWF interaction, affinity fractionation of heparins was performed with an immobilized peptide derived from a heparin-binding domain of vWF. Disaccharide analysis of high-affinity heparins revealed an increased ratio of IdoA2S-GlcN[S/Ac]6S to IdoA2S-GlcN[S/Ac]. Affinity fractionation of oligosaccharides (MW 3500) diminished the relative content of all disaccharides except IdoA2S-GlcNS6S, which was increased. These data suggest that the disaccharide structures IdoA2S-GlcNS6S and GlcNS6S-IdoA2S are crucial to heparin/vWF interactions. Understanding the structural aspects that confer such activity may be useful in designing heparin-based antithrombotic drugs.
Presented in part at the 69th Scientific Sessions of the American Heart Association, New Orleans, La, November 10-13, 1996, and published in abstract form (Circulation. 1996;94[suppl I]:I-743).
- Received September 5, 1996.
- Accepted January 9, 1997.
Platelet interaction with the blood vessel wall after arterial injury is one of the initial events in thrombosis.1 The deposition of platelets on exposed subendothelium is dependent on vWF, especially at areas of high shear stress.2 3 4 5 Two distinct platelet receptors bind vWF, GP Ib, and GP IIb/IIIa.6 Receptor GP Ib mediates the initial platelet adhesion to exposed subendothelium, leading to activation.7 8 GP IIb/IIIa binds to vWF after activation, leading to platelet spreading and aggregation at sites of injury.9 10 vWF and both GP Ib and GP IIb/IIIa are necessary for shear-induced platelet aggregation.10 11 Ruggeri and others10 12 13 have suggested that inhibition of the vWF/GP Ib interaction may be an effective antithrombotic strategy, especially because vWF-mediated platelet adhesion is one of the early events leading to platelet thrombus formation at thrombogenic sites.
We previously discovered that heparin binds directly to a region within the A1 domain of vWF at a heparin-binding site that subsumes the domain for binding GP Ib.14 Using vWF/platelet binding assays and the technique of ristocetin-induced platelet agglutination as a measure of vWF activity, we previously showed that heparin inhibits vWF/platelet binding and vWF-dependent platelet function in vitro.15 We also showed that intravenous heparin inhibits ristocetin cofactor activity in patients.15 In the same study, experiments performed in the absence of ristocetin confirmed the inhibitory effect of heparin against asialo-vWf and bovine vWF, both of which do not require the cofactor ristocetin for platelet binding in vitro. Thus, heparin-ristocetin interactions are not the mechanism. Likewise, it does not appear that heparin bound to the platelet surface is responsible for the interference with vWF-mediated platelet function. For example, fractions of vWF prebound to heparin (in the absence of platelets) showed reduced agglutination activity when subsequently added to platelets. In contrast, platelets presaturated with heparin showed paradoxically more agglutination in response to ristocetin, and blockade of GP Ib with a monoclonal antibody did not reduce the quantity of heparin bound to the platelet.15 Thus, several lines of evidence elaborated in earlier work support a direct heparin-vWF interaction as the mechanism for the inhibitory anti-vWF activity of heparin.
The binding of heparin to vWF has been shown to be dependent on MW but independent of ATIII affinity.15 16 However, the structural aspects of heparin responsible for vWF interactions are not clearly defined. Also, the relationship between vWF binding and inhibition of vWF-dependent activity has not been determined. To investigate the structural aspects of heparin/vWF binding, UFH was depolymerized by methods that attack known sites on the heparin molecule. Heparinase breaks the α-1,4 linkage between GlcNS[6S] residues and IdoA2S,17 whereas periodate/alkali treatment preserves these disaccharides and instead breaks the carbon ring of unsulfated uronic acids.18 The GAGs derived from these different depolymerization techniques were then tested for vWF binding affinity and the ability to inhibit vWF-dependent platelet agglutination. Also, disaccharide analysis of heparins and oligosaccharides with high affinity for a synthetic vWF-derived peptide was performed to further investigate the structural specificity of the heparin/vWF interaction. These experiments provide evidence for the structural aspects of heparin that confer the ability to bind and inhibit vWF. Evidence was also found that heparin/vWF binding alone may not be sufficient for maximum inhibition of vWF/platelet activity.
The following items were purchased: unfractionated porcine mucosal heparin (179 anti-Xa U/mg; MW 13 500 D) from Celsus Co; ristocetin from Biodata Corp; [125I]NaI from Amersham Corp; Iodobeads from Pierce Chemical Co; heparinase I (EC 188.8.131.52) from Sigma Chemical Co; heparin-Sepharose CL-6B from Pharmacia; dialysis membranes from Fisher Chemical Co; and the remaining reagents from Sigma Chemical Co. Cryoprecipitate was from the Richmond Metropolitan Blood Bank.
The heparin oligosaccharide standards used for HPLC were generously supplied by Dr Per Ostergaard of Novo Nordisk (Denmark).
The MWs of the heparin and heparin-derived GAGs were determined by gel permeation chromatography–HPLC on a Shimadzu-6A system equipped with UV detector and recorder (Shimadzu SPD-6A and C-R4A). The column consisted of serial SynChropak 30/100 analytic columns (SynChrom, Inc). It was run at 1 mL/min in PBS (0.01 mol/L sodium phosphate, 0.2 mol/L sodium chloride, pH 7). Heparin oligosaccharides of known MW (Novo Nordisk) were used as standards.
Heparin is a linear-chain GAG composed of alternating uronic acids (IdoA or glucuronic acid) and glucosamines (GlcNAc or GlcN). The uronic acids may be sulfated at the second carbon; the glucosamines have various degrees of sulfation. In standard pharmaceutical heparin, the chain lengths vary from MW 5000 to 20 000 D. Thus, standard heparin is a heterogeneous material. A variety of chemical and enzymatic methods for cleaving heparin have been developed for use in structural analysis and to obtain low-MW heparins for clinical use. In this study, we used heparinase digestion or periodate oxidation to cleave standard heparin. Heparinase attacks only the α-1,4 linkage between nonacetylated GlcNS[6S] and IdoA2S.17 Periodate oxidation followed by alkali treatment preserves this disaccharide unit and instead cleaves heparin by breaking the carbon ring of unsulfated uronic acids.18 Each depolymerization method yields heterogeneous products of various chain lengths, mostly small pieces of the original heparin. Although heterogeneous, the products of heparinase digestion will have fewer disaccharide units of the type GlcNS[6S]-IdoA2S, whereas the products of periodate oxidation will retain these disaccharides and instead have fewer unsulfated uronic acids. The products of each depolymerization method can then be separated by size and compared by various assay techniques to assess the relative importance of these specific portions of the heparin molecule.
In this study, standard porcine UFH of average MW 13 500 was depolymerized by these two methods. Briefly, heparinase digestion was performed as follows: 8 mg/mL UFH was mixed with 0.8 U/mL heparinase (1 U is defined here as the amount that will form 0.1 μmol of unsaturated uronic acid per hour) in buffer (0.25 mol/L sodium acetate, 0.0025 mol/L calcium acetate, pH 7) at 30°C for 52 hours. This time was chosen after the depolymerization was monitored by application of samples to the HPLC system as described above until no change was noted in the HPLC profile. Periodate oxidation was performed as follows: UFH (2 mg/mL) was dissolved in buffer (0.05 mol/L sodium phosphate, pH 7). Sodium perchlorate (0.2 mol/L) and sodium periodate (0.02 mol/L) were added, and the reaction was allowed to proceed for 6 hours at 37°C. A molar excess of D-mannitol was then added to terminate the reaction. Alkaline elimination was performed by addition of 2 mol/L NaOH to raise the pH to 12 for 30 minutes at room temperature. This step cleaves the heparin chains at the sites of periodate oxidation to yield heparin oligosaccharides.
The products of each depolymerization method were desalted and lyophilized and separated by size on a G-50 Sephadex (superfine) column. Fractions from each depolymerization were pooled into five groups with peak MWs as noted in Table 1⇓. Within each group, the MW ranged from ≈1000 below to ≈1000 above the peak value. The derived GAGs were then dialyzed against deionized distilled water and lyophilized. Since periodate destroys unsulfated uronic acids, purity was confirmed by a GlcNS assay19 instead of a uronic acid assay.
Anti–Factor Xa Activity
The anti–factor Xa activity of the heparin-derived GAGs was determined with a chromogenic substrate assay described by Teien and Lie20 and adapted to microtiter plates. Reagents were purchased from Pharmacia.
Purification and Labeling of vWF
Human vWF was purified from cryoprecipitate by a modification of the procedures of Newman et al21 and Switzer and McKee.22 In brief, vWF was isolated by a series of ethanol and polyethylene glycol precipitations and by gel filtration on a Sepharose CL-4B column. Fractions from the column were monitored for ristocetin cofactor activity (a measure of vWF activity) by use of fixed platelets and for protein by bicinchoninic acid assay.23 Purity was assessed by reduced 5% SDS-PAGE,24 which gave a single band at MW 220 000, which is the expected location of vWF monomers. Fractions containing the highest specific ristocetin cofactor activity (typically 100 to 150 plasma equivalent units/mg) and greatest purity (>95%) were pooled and stored frozen in aliquots at −80°C.
Purified vWF (25 μg) was radioiodinated with Iodobeads25 with 250 μCi [125I]NaI as previously described.15 Radioiodination was confirmed by phosphorimaging of 6% SDS-PAGE of the protein under reducing conditions, which gave a single band at MW 220 000. The resulting labeled vWF had a specific activity of 2 Ci/g. It was stored in aliquots at −80°C for use within 4 weeks.
Heparin/vWF Binding Assay
The vWF binding affinity of GAGs was determined by a competitive binding assay as described by Fujimura et al,26 with some modification. Briefly, heparin-Sepharose beads (6% vol/vol) and 125I-vWF (0.2 μg/mL) were incubated in 200 μL buffer (0.05 mol/L Tris-HCl, 0.1 mol/L NaCl, 0.1% ovalbumin, pH 7.4), with varying concentrations of unlabeled heparin or heparin-derived GAGs added to compete with the heparin-Sepharose. After 1 hour of incubation at room temperature (1 hour was chosen because no further vWF binding was observed after 30 minutes), duplicate 80-μL aliquots were layered onto 250 μL 30% sucrose in microfuge tubes and centrifuged at 10 000g for 1.5 minutes. The tube tips containing the beads and bound vWF were separated from supernatant with a tube tip cutter and counted in an LKB-Wallac 1282 Compugamma gamma counter. Nonspecific binding was assessed with heparin-free Sepharose CL-6B and found to be equivalent to using a large excess of free UFH (20 mg/mL) in the assays. This was typically <1% of the total counts. Specific binding was determined by subtracting the nonspecific binding from the total binding. The maximum specific binding (no free heparin or GAG) was 29±2.3% of total counts. The competitive, inhibitory effects of GAGs were expressed as a percentage of the maximum specific binding. The mean±SD of at least three separate experiments was determined at each GAG concentration tested. The IC50 (the concentration inhibiting the specific binding by 50% of the maximum) was interpolated from the experimental curves for each GAG.
The ability of the heparin-derived GAGs to inhibit vWF-dependent platelet aggregation was determined by use of ristocetin-induced agglutination of fixed platelets. In earlier work, the inhibitory effects of UFH were found to be the same with either fresh or fixed platelets, so formaldehyde-fixed platelets were prepared and used as previously described.15 27 Briefly, fixed platelets were suspended in 0.15 mol/L Tris-buffered saline at a final concentration of 200 000/μL. For each batch of platelets, a standard amount of normal pooled, citrated plasma (the vWF source) was chosen on the basis of titrations to the lowest concentration needed to achieve maximum agglutination with ristocetin. Platelets were preincubated with plasma and standard heparin or a derived GAG for 5 minutes at 37°C. Agglutination in response to ristocetin (1 mg/mL final concentration) was measured with a Scienco aggregometer in standard aggregation units. Data for each concentration of UFH or derived GAG were expressed as a percent of maximum agglutination (without UFH or GAG). Results shown for each concentration are from at least two separate experiments, each performed in duplicate. The IC50 (the concentration that inhibited agglutination by 50%) for each GAG was interpolated from the experimental curves.
Affinity Chromatography of Heparins and Oligosaccharides
In a complementary method to discern the structural features of heparins with high affinity for vWF, we subjected standard and chemically modified heparins to affinity chromatography. In previous work, we localized a heparin-binding domain of vWF near the GP Ib binding domain and synthesized a 23-residue peptide (Y565 to A587) that binds heparin with an affinity comparable to that of native vWF.14 In further studies, a truncated version (K569 to I580) was found to possess comparable heparin-binding properties.27 All synthetic peptides were prepared by standardized solid-phase methods, and their purity and composition were established as previously described.14 28 29 For affinity chromatography, the synthetic peptides were coupled to Sepharose beads through a spacer arm. Briefly, Sepharose CL-6B was first activated with 1,1′-carbonyldiimidazole in anhydrous dioxane solution as described30 and substituted with 1,6-diaminohexane. The resulting pendant amino group was derivatized with iodoacetic acid to affect condensation. After >90% of amino groups were derivatized, the gel was washed extensively, and underivatized amino groups were then covalently modified with an excess of acetic anhydride. The activated and substituted Sepharose gel now bearing an iodoacetyl functional group was coupled to the peptide (extended at the C-terminus by the sequence b-Ala-Cys). The incorporation of peptide was monitored by disappearance of reaction with 5, 5′-dithiobis-2-nitrobenzoic acid (Ellman's reagent), and the absolute concentration of peptide on each gel was determined by quantitative amino acid analysis of a portion of acetone-dried, weighed substituted Sepharose. It was previously shown27 that optimal binding of heparin took place at a substitution level of peptide of about 1.5 mmol/L. Thus, the desired level of peptide was controlled by use of an excess of Sepharose with a fixed amount of peptide. Nonderivatized iodoacetyl functional groups were blocked with 2-mercaptoethanol.
For chromatography, the affinity gels and different heparins used as starting materials were equilibrated in 50 mmol/L Tris buffer, pH 7.5. After elution of the nonbinding (breakthrough) fraction of heparin, the vWF peptide columns were washed exhaustively, and then the bound fraction was eluted sequentially with 1 and 2 mol/L NaCl in equilibrating buffer. The elution profile was monitored at 206 nm, and the appropriate fractions were pooled and desalted by exhaustive dialysis (1000 MW cutoff dialysis membrane; Fisher). The heparins were recovered by lyophilization, and the heparin content was determined by the carbazole method. Sepharose matrices that were identically derivatized and prepared but without conjugated peptides did not bind heparin at all.31
In previous work, heparin fractions that eluted at 1 and 2 mol/L NaCl were found to have increased ability to inhibit vWF/platelet binding and ristocetin-induced platelet agglutination.31 Also, using isothermal titration calorimetry, the Kd for binding between the high-affinity (high NaCl elution) heparin and the peptide was lower than that for unfractionated heparin.27
For this study, three heparins were subjected to this affinity chromatography: (1) standard heparin (MW 13 500); (2) a periodate-oxidized heparin subsequently treated by reduction with sodium borohydride, which maintains chain length (MW 11 000); and (3) a periodate-oxidized heparin depolymerized to yield an oligosaccharide of MW 3500.
Disaccharide analysis was performed on the three heparins subjected to affinity chromatography. For the larger heparins (standard heparin and periodate oxidized/borohydride reduced), the disaccharide content of the starting material and the highest-affinity fraction (that which eluted at 2 mol/L NaCl) were compared. For the low-MW oligosaccharide, nonbinding fractions as well as moderate-affinity (eluting between 0.1 and 1.0 mol/L NaCl) and high-affinity (eluting between 1 and 2 mol/L NaCl) fractions were analyzed.
Standard chemical disaccharide analyses performed were based on the procedures of Guo and Conrad,32 33 with the technical assistance of Drs H.E. Conrad and Lowell Hager. In brief, the samples were first treated with N2H4 to remove N-acetyl groups from GlcNAc residues, then treated with nitrous acid at pH 1.5 to cleave GlcNSO3 glycosidic bonds, forming anhydromannose reducing sugars. Subsequent depolymerization was then carried out at pH 4.0 to cleave the remaining GlcN bonds, completely depolymerizing the sample to disaccharides. [3H]NaBH4 was then used to reduce and label their end groups to anhydromannitol residues. These labeled disaccharides were then separated, identified, and quantified by reverse-phase HPLC and flow-through liquid scintillation counting. For each heparin tested, the total content of unsulfated, monosulfated, and disulfated disaccharides was determined.
At each concentration of heparin oligosaccharide tested for vWF binding and for inhibition of platelet agglutination, data are presented as a mean with error bars to denote the SD. For IC50s interpolated from the experimental curves, 95% CIs were constructed for the curves, and these are given at each IC50 value reported.
Anti–Factor Xa Activity
The anti–factor Xa activities of each MW group from the different depolymerization methods are shown in Table 1⇑. As can be seen, periodate/alkali treatment destroyed the anti-Xa activity. In contrast, heparinase digestion only mildly diminished the anti-Xa activity in proportion to the extent of depolymerization.
vWF Binding Affinity
Both periodate-derived GAGs and heparinase-derived GAGs were able to compete for the binding of labeled vWF (Table 2⇓ and Fig 1⇓). At all MWs, periodate-derived GAGs were better able to compete for labeled vWF than heparinase-derived GAGs (Fig 1⇓), and size for size, they showed significantly lower IC50s (Table 2⇓). Also, an MW dependency for binding affinity was noted for both types of GAGs, with larger species having higher affinity. However, the binding affinities of the largest periodate-derived GAGs were higher than that of UFH despite the lower MWs of the periodate oxidized heparins.
Inhibition of vWF-Dependent Platelet Agglutination
Both periodate- and heparinase-derived GAGs were able to inhibit ristocetin-induced platelet agglutination (Table 2⇑ and Fig 2⇓). At all MWs, periodate-derived GAGs were better able to inhibit agglutination than heparinase-derived GAGs, as shown by their lower IC50s (Table 2⇑). An MW dependency similar to that seen with binding affinity for both treatment methods was also noted. There was a large drop in inhibitory potency below an MW of 4000 (Fig 3⇓).
Disaccharide Analysis of Affinity Fractionated GAGs
When the larger species of heparins (UFH, MW 13 500; periodate oxidized/borohydride reduced, MW 11 000) were affinity refined, no clear-cut differences in individual disaccharide content could be discerned. However, the overall ratio of the sulfated disaccharides IdoA2S-GlcN[S/Ac]6S to IdoA2S-GlcN[S/Ac] changed with affinity fractionation. This ratio increased from 4.5 to 6.4 for UFH and 2.3 to 4.2 for periodate-oxidized/borohydride-reduced heparin as a result of affinity fractionation.
In contrast, when the smaller, more homogeneous oligosaccharide fraction of MW 3500 was subjected to vWF peptide-affinity separation, significant differences in individual sulfated disaccharides were seen (Fig 4⇓). Disaccharide analysis showed a steady increase in the proportion of the disaccharide IdoA2S-GlcN[S/Ac]6S as the fractions increased from nonbinding to low to high affinity. It is noted that the method of disaccharide analysis cannot distinguish between IdoA2S-GlcNAc6S and IdoA2S-GlcNS6S, but it is known that the N-acetylated glucosamine residues in native heparin represent <10% of the total. Thus, most of the disaccharide detected is expected to be IdoA2S-GlcNS6S.
In these experiments, we have used chemical and enzymatic methods to prepare heparin-derived GAGs with specific structural characteristics to deduce which aspects of heparin structure confer ability to interact with vWF. We have also looked at the relationship between binding and biological inhibitory activity. Structural specificity was also investigated by disaccharide analysis of heparins with high affinity for a synthetic vWF-derived peptide.
With regard to vWF binding, GAGs prepared by periodate/alkali treatment had greater affinity for binding vWF than the same size GAGs prepared by heparinase treatment. From this, we deduce that the GlcNS[6S]-IdoA2S disaccharides, which are destroyed by heparinase but not periodate, are probably important to vWF binding. An MW dependency was also noted. However, the largest periodate-derived GAGs actually had higher vWF binding affinity than UFH, even though the MW was less, emphasizing the importance of structural specificity.
With regard to inhibition of vWF-dependent platelet agglutination, periodate-derived GAGs also showed greater ability to inhibit vWF-dependent platelet agglutination than heparinase-derived GAGs of the same size. This again confirms the importance of the GlcNS[6S]-IdoA2S disaccharides in heparin/vWF interactions. As with the binding assay, an MW dependency was noted, with severely reduced ability to inhibit agglutination at MW <4000, suggesting that a minimum chain length of 12 monosaccharides is important for the biological inhibition of vWF activity.
The different treatments had divergent effects on anti– factor Xa activity. The anti–factor Xa activity is dictated primarily by the ATIII-binding pentasaccharide sequence of heparin.34 35 36 37 In this study, we found that the anti–factor Xa activity is destroyed by periodate treatment but preserved by heparinase digestion. This is expected, because the unsulfated uronic acid moiety attacked by periodate is part of the ATIII-binding pentasaccharide, whereas the GlcNS[6S]-IdoA2S disaccharide attacked by heparinase is not part of the pentasaccharide. Thus, periodate treatment can create a GAG with little conventional anticoagulant activity, but it preserves or enhances anti-vWF activity, whereas heparinase treatment creates a GAG with moderate anticoagulant activity and reduced anti-vWF activity. This shows that the anti–factor Xa and anti-vWF activity of heparins can be modified independently by chemical or enzymatic treatment. It also confirms the previous observations by us and others15 16 that the ATIII pentasaccharide domain is not essential for anti-vWF activity.
We also observed that binding alone, although necessary for inhibition of activity, may not be sufficient for biological inhibitory activity. For example, for the 6100-MW periodate-derived GAGs, the binding affinity was 1.3 times less than that of UFH (IC50 100±19 versus 78±7.0 μmol/L), whereas the ability to inhibit vWF activity was 4.5 times less than UFH (IC50 17±5.0 versus 3.8±0.8 μmol/L). We surmise that this may be due to the lower MW of all the derived GAGs; those molecules that do bind vWF may not be large enough to effectively inhibit platelet interactions with the nearby GP Ib platelet receptor binding domain, through either steric hindrance or induction of conformational changes. Another possibility is that the above discrepancy between binding and activity inhibition is due to heparin binding to sites on vWF that do not interfere with its ability to bind GP Ib. Our studies to date do not rule out the participation of other domains of vWF in binding heparin. In fact, there is evidence of a low-affinity heparin binding site within the first 272 amino acid residues of vWF,38 remote from the GP Ib binding site.
Disaccharide analysis found increased amounts of the disaccharide IdoA2S-GlcNS6S in heparins and oligosaccharides that were affinity fractionated for binding to a synthetic vWF-derived peptide. This pair of monosaccharides is the same as that found to be important in the binding and platelet agglutination experiments (although the order of the residues is reversed). Thus, distinctly independent methods have yielded concordant results. The data suggest that a critical sequence of heparin that confers vWF binding and inhibitory activity consists of alternating residues of GlcNS6S-IdoA2S and IdoA2S-GlcNS6S.
The vWF/platelet interaction may have important clinical relevance. The platelet GP Ib/vWF interaction mediates the initial adhesion of platelets to exposed subendothelium, leading to platelet activation,7 8 whereas the platelet GP IIb/IIIa interaction with vWF is important after activation to promote platelet spreading and aggregation.9 10 Pharmacological suppression of the initial GP Ib/vWF interaction may be a promising method to prevent thrombosis in stenosed, injured arteries. A heparin-derived GAG with enhanced ability to inhibit this interaction may prove useful. Knowledge of the structural aspects of heparin responsible for inhibiting vWF/platelet interactions will be helpful in designing GAGs with enhanced anti-vWF activity but reduced conventional anticoagulant activity. We are currently testing the in vivo antithrombotic potential of heparin-derived GAGs that have been affinity fractionated for high vWF binding ability.31
Selected Abbreviations and Acronyms
|GlcNS[6S]||=||N-sulfated glucosamine±C6 sulfation|
|HPLC||=||high-performance liquid chromatography|
|IdoA2S||=||iduronic acid sulfated at C2|
|vWF||=||von Willebrand factor|
This study was supported in part by grants from the National Institutes of Health (HL-39903, Dr Sobel), the Veterans Administration Research Service (Dr Sobel), and the National Institutes of Health Individual National Research Service Award (HL-09395, Dr Poletti).
Hawiger J. Adhesive interactions of blood cells and the vascular wall. In: Colman RW, Hirsh J, Marder VJ, Salzman EW, eds. Hemostasis and Thrombosis. Philadelphia, Pa: JB Lippincott Co; 1994:762-796.
Ruggeri ZM, De Marco L, Gatti L, Bader R, Montgomery RR. Platelets have more than one binding site for von Willebrand factor. J Clin Invest. 1983;72:1-12.
De Marco L, Girolami A, Russell S, Ruggeri ZM. Interaction of asialo von Willebrand factor with glycoprotein Ib induces fibrinogen binding to the glycoprotein IIb/IIIa complex and mediates platelet aggregation. J Clin Invest. 1985;75:1198-1203.
Sakariassen KS, Nievelstein PF, Coller BS, Sixma JJ. The role of platelet membrane glycoproteins GpIb and IIb-IIIa in platelet adherence to human artery subendothelium. Br J Haematol. 1986;63: 681-691.
Ikeda Y, Handa M, Kawano K, Kamata T, Murata M, Araki Y, Anbo H, Kawai Y, Watanabe K, Itagaki I, Sakai K, Ruggeri ZM. The role of von Willebrand factor and fibrinogen in platelet aggregation under varying shear stress. J Clin Invest. 1991;87:1234-1240.
Peterson DM, Stathopoulos NA, Giorgio TD, Hellums JD, Moake JL. Shear-induced platelet aggregation requires von Willebrand factor and platelet membrane glycoproteins Ib and IIb-IIIa. Blood. 1987;69:625-628.
Ruggeri ZM. von Willebrand factor as a target for antithrombotic intervention. Circulation. 1992;86(suppl III):III-26-III-29.
Sobel M, Soler D, Kermode J, Harris R. Localization and characterization of a heparin binding domain peptide of human von Willebrand Factor. J Biol Chem. 1992;267:8857-8862.
Sobel M, McNeill PM, Carlson PL, Kermode JC, Adelman B, Conroy R, Marques D. Heparin inhibition of von Willebrand factor-dependent platelet function in vitro and in vivo. J Clin Invest. 1991;87:1787-1793.
Linhardt RJ, Rice KG, Merchant ZM, Kim YS, Lohse DL. Structure and activity of a unique heparin-derived hexasaccharide. J Biol Chem. 1986;26:14448-14454.
Switzer ME, McKee PA. Studies on human antihemophilic factor: evidence for a covalently linked subunit structure. J Clin Invest. 1976;57:925-937.
Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:682-690.
Fujimura Y, Koiti T, Holland LZ, Roberts JR, Kostel P, Ruggeri ZM, Zimmerman TS. A heparin-binding domain of human von Willebrand factor. J Biol Chem. 1987;262:1734-1739.
Bethell GS, Ayers JS, Hancock WS, Hearn MT. A novel method of activation of cross-linked agaroses with 1,1′-carbonyldiimidazole which gives a matrix for affinity chromatography devoid of additional charged groups. J Biol Chem. 1979;254:2572-2574.
Sobel M, Bird KE, Tyler-Cross R, Marques D, Toma N, Conrad E, Harris RB. Heparins designed to specifically inhibit platelet interactions with von Willebrand factor. Circulation. 1996;93:992-999.
Jordan RE, Beeler D, Rosenberg RD. Fractionation of low molecular weight heparin species and their interaction with antithrombin. J Biol Chem. 1979;254:2902.
Rosenberg RD, Lam L. Correlation between structure and function of heparin. Proc Natl Acad Sci U S A. 1979;76:1218-1222.
Lindahl U, Backstrom G, Hook M, Thunberg L, Fransson L, Linker A. Structure of the antithrombin-binding site in heparin. Proc Natl Acad Sci U S A. 1979;76:3198-3202.
Casu B, Oreste P, Torri G, Zoppetti G, Choay J, Lormeau JC, Petitou M, Sinai P. The structure of heparin oligosaccharides with high anti-Xa activity containing the minimum antithrombin III-binding sequence. Biochem J. 1981;197:599-605.
Fretto LJ, Fowler WE, McCaslin DR, Erickson HP, McKee PA. Substructure of human von Willebrand factor: proteolysis by V8 and characterization of two functional domains. J Biol Chem. 1986;261:15679-15689.