WO1998039456A1

WO1998039456A1 - Factor x variant

Info

Publication number: WO1998039456A1
Application number: PCT/US1998/003939
Authority: WO
Inventors: Joseph P. Miletich
Original assignee: Washington University
Priority date: 1997-03-07
Filing date: 1998-03-05
Publication date: 1998-09-11
Also published as: AU6673898A

Abstract

A Factor X variant is disclosed which has a single substitution in which the arginine residue at amino acid position 347 of human Factor X is replaced with asparagine. The Factor X variant has substantially reduced affinity for Factor Va, although the catalytic impact of Factor Va binding remains essentially intact.

Description

FACTOR X VARIANT

ACKNOWLEDGMENT OF SUPPORT

The invention herein was made in part with government support under grant number HL 14147 by the National Heart , Lung and Blood Institute and by a grant from Monsanto Company .

BACKGROUND OF THE INVENTION

The present invention relates to a variant of blood-clotting Factor X and, more particularly , to a human Factor X variant with substantially reduced affinity for Factor Va , but without substantial reduction in the catalytic impact of Factor Va binding .

(Note: Literature references on the following background information and on conventional test methods and laboratory procedures well-known to the person skilled in the art, and other such state-of-the-art techniques as used herein, are indicated by reference numbers in parentheses , and appended at the end of the specification) .

Factor X (fX) is one of the four classical vitamin K-dependent blood clotting factors (Factors II , VII , IX and X) . It is the zymogen ( inactive enzyme precursor) plasma glycoprotein essential for the conversion of prothrombin (fll) to thro bin during blood coagulation. Thrombin is the final protease in the blood coagulation cascade. Thrombin induces platelet aggregation and converts circulating fibrinogen to insoluble fibrin, resulting in the formation of a blood clot. Activated Factor X (fXa) thus holds a pivotal position in blood coagulation since it is the only known physiological activator of prothrombin (fll) .

The single gene for fX spans approximately 25 kb. The 8 exons of the fX gene specifically code for a 488-amino acid polypeptide.

The zymogen precursor, fX, is activated as fXa by cleavage of a single peptide bond (Arg¹⁹⁴-Ile¹⁹⁵) on a phospholipid surface by activated factor VII and tissue factor (fVIIa/TF, extrinsic pathway) or activated factors IX and VIII (fIXa/fVIIIa, intrinsic pathway) . The same bond can also be hydrolyzed by the Factor X- activating protein from Russell's viper venom (RW-X) (1-6).

Activated fX is inhibited by antithrombin III (ATIII) and tissue factor pathway inhibitor (TFPI, 7-10). The light chain fX contains eleven gamma-carboxylated glutamic acid residues in the amino terminal "Gla domain" , which is necessary for the membrane binding capacity of the molecule, followed by two epidermal growth factor-like domains (EGF1 and EGF2) . The heavy chain is joined to the light chain by a single disulfide bond and contains the activation peptide and the serine protease domain.

As the enzymatic component of the prothrombinase complex, fXa interacts with the surface (membrane phospholipid) , the cofactor (fVa) and the substrate (prothrombin) for maximally efficient thrombin formation (11) . Determining the contribution of specific inter olecular interactions within the prothrombinase complex to the overall catalytic activity of fXa is, therefore, complex. Ideally, each interaction should be characterized independently and combinatorily.

The methodology for efficient expression of recombinant fX has been developed as described by Miletich and co-workers (12) , but it affords the production of only limited quantities. This precludes direct analysis of the binding between recombinant fXa and other components of the prothrombinase complex by conventional means.

BRIEF DESCRIPTION OF THE INVENTION

In accordance with the present invention, a novel human Factor X (fX) variant is provided with a single substitution in which the arginine residue at amino acid position 347 is replaced with asparagine (fX 347N) . The activated fXR347N (fXaR347N7) unexpectedly has a substantially reduced affinity for Factor Va (fVa) , although the catalytic impact of fVa binding remains essentially intact.

In view of this significant biological activity in the coagulation cascade, the novel variant of the invention is useful for several specific reasons. For example, it is useful for in vitro assay and diagnostic applications as follows:

First, the novel variant is useful for quantitating the importance of the interaction between co-factor, Factor Va, and the serine protease domain of Factor Xa in reactions involving Factor Xa. This can be accomplished by comparison to wild type Factor Xa. Second, the novel variant is useful for quantitating the impact of specific inhibitors of the Factor Va - Factor Xa interaction, again by comparison of wild type to fXR347N in the presence of inhibitors. Quantitating the effect of inhibitors also is a useful adjunct to therapeutic treatment with inhibitors such as, for example, heparin and tissue factor pathway inhibitor (TFPI) .

Third, since the novel variant has no significant interaction with Factor Va in the absence of an appropriately charged phospholipid surface, it is highly useful in reactions where absolute specificity of thrombin activation to membrane surfaces is required.

The properties of the novel Factor X variant of the invention which make it useful for the above and other such purposes is demonstrated herein by various test methods as follows:

The mutation at position 347 is selective as demonstrated by normal activation and inhibition, except in the presence of sub- saturating heparin where the rate of inhibition by antithrombin III (ATIII) is 15% of normal.

The reactivity of fXaR347M towards prothrombin is equivalent to wild type fXa (fXaWT) in the absence of cofactor and phospholipid. Addition of cofactor, but no phospholipid, dramatically increases the catalytic efficiency of fXaWT towards prothrombin but has negligible effect on fXaR347N. With cofactor and 3:1 phosphatidylcholine:phosphatidylserine (PC: PS) vesicles, fXaR347N has full catalytic potential but reduced apparent affinity for fVa relative to fVaWT (Kd_app of 2.3 versus 0.1 nM, respectively) .

On an activated platelet surface, fXaWT and fXaR347N have similar thrombogenic activity. In a competitive binding assay based on the ability to displace labeled fXa from fVa on a phospholipid surface, fXaR347N also has ~10-fold lower affinity than fXaWT.

Thus, in accordance with the present invention, substitution of fXa at position 347 selectively attenuates the interaction between fXa and fVa without affecting its catalytic activity.

As used herein, the numbering of the amino acid positions in fX is based on its well-known sequence as shown, for example, by Marion R. Fung, et al., Proc. Nat'l. Acad. Sci. USA 82, 3591-3595 (1985) .

DETAILED DESCRIPTION OF THE INVENTION

While the specification concludes with claims particularly pointing out and distinctly claiming the subject matter regarded as forming the present invention, it is believed that the invention will be better understood from the following preferred embodiments of the invention taken in conjunction with the accompanying drawings . BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1. SDS-PAGE of WT and variant fX species. Proteins were analyzed using a 12.5% separating gel. The gel was stained with Coomassie Blue. Each lane contains 1.5 μg of protein. Lane a is reduced fXWT. Lane b is reduced fXR347N. Lane c is non-reduced fXWT. Lane d is non-reduced DCR347N.

Fig. 2. Generation of thrombin by fXaWT and fXaR347N in the absence of phospholipid and cofactor. 10 nM of fXaWT (— •— ) or fXaR347N (-- O--) was incubated with varying concentrations of prothrombin (0-450μM). Aliquots were removed from the reactions over time (0-10 in), diluted 10-fold into EDTA-containing buffer, and incubated with 0.5 mM Spectrozyme TH™. The initial rate of hydrolysis was monitored at 405nm. The concentration of thrombin was determined from a standard curve prepared from dilutions of maximally activated prothrombin. The rate of thrombin formation at each prothrombin concentration was then calculated and expressed as mol thrombin (IIa)/sec/mol Xa.

Fig. 3. Thrombin generation by WT and variant (Xa in the absence of phospholipid and in the presence of cofactor. 0.5 nM of fXaWT (— •— ) or fXaR347N («o~) was incubated with 10 μM prothrombin and varying concentrations of fVa (0-1 μM). Aliquots were removed from each reaction over time (0-8 min) and the initial rate of thrombin formation was determined at the fVa concentrations indicated as described in the legend to figure 2. Fig. 4. Thrombin generation on PC:PS vesicles in the presence of cofactor. 20 pM of fXaWT (— •— ) or fXaR347N (--θ--) was incubated with 20 μM PC:PS vesicles, 1 μM prothrombin, and varying concentrations of fVa (0-1 nM). Aliquots of each reaction were removed over time (0-3 min) and quenched into EDTA-containing buffer. Initial rates of thrombin formation were determined for each fVa concentration as described in the legend to figure 2.

Fig. 5. Generation of thrombin by fXaWT and fXaR347N on an activated platelet surface. Washed platelets were incubated at lO^/ml with 0.5 U/ml thrombin and fXa at varying concentrations (0-1 nM). The reaction was initiated by addition of 1.0 μM prothrombin. Thrombin generation was quantitated for fXaWT (— •— ) or fXaR347N (-- O-) by quenching aliquots of each reaction over time (0-3 min) into EDTA- containing buffer. Each aliquot was subsequently incubated with 0.5 mM Spectrozyme TH™ and the initial rate of thrombin formation was quantitated from Spectrozyme TH™ hydrolysis as described in the legend to figure 2.

Fig. 6. Competition binding assay. PC.PS coated latex beads (0.2%, v/v) were incubated for 10 min in the presence fVa (1 nM). I- fXS379a was added with various concentrations of unlabeled fXaWT (0-7 M, O) or fXaR347N (0-50 nM, •) and incubated for an additional 10 min. The beads were collected by centrifugation and counted. The amount of specific binding (fVa, phosphatidylserine, and calcium-dependent) was determined and expressed as the percent of binding in the absence of unlabeled fXa. Non-specific binding (fVa-independent) was subtracted from all reactions.

Fig. 7. Model of fXa and Arg347. The image of fXa was generated using RasMol v2.5. In the illustration, the serine protease domain is orange, EGF- 1 is yellow, and the surface loops identified by Chattopadhyay et al. as fVa binding epitopes are depicted in dark and light blue (31). The catalytic triad is represented in white and the position of Arg347 is highlighted in green.

In order to further illustrate the invention, the following detailed examples were carried out although it will be understood that the invention is not limited to these specific examples or the details described therein. EXAMPLES

Materials — Crude snake venoms, L-α-phosphatidylcholine, and L-α- phosphatidylserine were purchased from Sigma. Spectrozyme TH™ (H-D- hexahydrotyrosyl-L-alanyl-L-arginine- -nitroanalide-diacetate)and Spectrozyme FXa™ (methoxycarbonyl-D-cyclohexylglycyl-glycyl-arginine-p-nitro- anilide-acetate) were purchased from American Diagnostica Inc. (Greenwich, CT). Full-length heparin was obtained from Elkins-Sinn Inc. (Cherry Hill, NJ). Heparin pentasaccharide was a generous gift of J.C. Lormeau of Sanofi Recherche (Gentilly Cedex, France). All other reagents and chemicals were of the highest quality available commercially.

Proteins — Prothrombin was purified, concentrated (to >1 mM), and immunodepleted of residual fX (<1 nM) as described (12). Factor VII was purified from human plasma (13). Antithrombin III was purchased from Kabi Pharmacia Diagnostics (Piscataway, NJ) and recombinant TFPI was a gift of Monsanto/Searle Company (St. Louis, MO). Human fVa and fiXa were purchased from Haematologic Technologies (Essex Jet., VT). Thrombin was prepared by cation exchange chromatography after activation of human prothrombin with Taipan snake (Oxyuranus scutellatus) venom. Porcine fVIII was obtained from Potion Products (Agoura Hill, CA) and was further purified (14) using an immobilized antibody (W-3, a generous gift from Dr. David Fass). Innovin™, a lipidated, recombinant tissue factor, was obtained from Baxter Diagnostics (Deerfield, IL). RVV-X was purified from Russell's viper venom (15). SDS-PAGE was performed by the method of Laemmli (16). Proteins were stained directly using Coomassie Blue or were transferred to nitrocellulose (0.45 nm), probed using antibodies described previously (12), and visualized with an immunochemiluminescent kit (Amersham, Buckinghamshire, UK). The secondary antibody utilized for immunoblotting was affinity-purified caprine anti-mouse IgG conjugated to horseradish peroxidase, Jackson ImmunoResearch (West Grove, PA).

FX was purified from culture media as previously described (12). Briefly, fX was removed from media by absorption to barium citrate. The resulting pellet was washed and resuspended in 32% saturated NtUSO* in the presence of 5 mM diisopropylfluorophosphate and incubated for one hour at 4°C. The precipitate was collected by centrifugation and the resulting supernatant was dialyzed into 1 mM benzamidine, 10 mM HEPES, 100 M NaCl, pH 7.0. The dialyzed supernatant was concentrated, clarified by centrifugation at 12,000 x g, and immunoaffinity purified using antibody 3698.1A8.10, a calcium dependent monoclonal antibody directed against the Gla domain of fX. Separate columns were used for fXWT and fXR347N. FX was eluted with 5 mM EDTA, lOmM HEPES, 100 mM NaCl, pH 7.0, and pooled fractions containing fX were loaded onto a Pharmacia Mono Q FPLC column. FX was eluted with a high-salt buffer (10 mM HEPES, 500 mM NaCl, pH 7.0). Fractions containing fX antigen were pooled and adjusted to pH 6.5 and final concentrations of 2.5 mM CaCl₂ and 20% glycerol for storage at -70°C.

Defined phospholipid vesicle — PC:PS (3:1) vesicles, of nominal 100 nm diameter, were synthesized from egg yolk phosphatidylcholine (PC) and bovine brain phosphatidylserine (PS) at a concentration of 1.0 mM total lipid in 5 mM CaCl₂, 10 mM HEPES, 100 mM NaCl, pH 7.5, by the method of

® membrane extrusion (17) using the Liposofast Basic device (Avestin, Ottawa, ON, Canada). Vesicles were adjusted to a final concentration of 10% (v/v) glycerol, snap frozen, and stored at -70°C.

Purity and concentratio — The overall quality of the purified fX species was determined by SDS-PAGE. The total protein concentration was determined using the reported (2) extinction coefficient for plasma fX (OD280nm,£^m8 ^l =1.16) and agreed with the fX concentration determined by a two-site immunoassay (12). Activation of recombinant fX was also monitored by following the loss of signal over time in an immunoassay using monoclonal antibodies directed against the activation peptide and the light chain (3514.2E12.1 and 3448.1D7.20, respectively, 12).

Construction and cell culture — The recombinant fX constructs were made as described (12). In short, mutations were engineered into the cDNA of fX, which had been cloned from a human hepatocyte library and inserted into pUC19. Mutations were introduced to change amino acids Arg347 and Ser379 to alanine and to create new Hindi and Xcml sites, respectively (GTGGAQCGC → GTCGACAAC and GACAGCGGG → GACGCTGGG). These mutations, as well as the entire wild type sequence, were verified by sequencing. Constructs were shuttled into a mammalian expression vector (ZMB3) which was kindly provided by Dr. Don Foster (Zy oGenetics). Human kidney cells (293 cells, ATCC-CRL-1573) were transfected using calcium phosphate precipitation (18) and cultured as previously described (12). The fX produced was screened for concentration, size and function by immunoassay, by immunoblotting after SDS-PAGE, and by a modified prothrombin time assay, respectively. Selected clones were expanded for mediaconditioning.

Activatio — FXR347N was activated by three methods and directly compared to wild type recombinant fX. Activation by RVV-X was achieved by incubating 100 nM fX with either 50 pM RVV-X for initial rate measurements or 2 nM RVV-X for endpoint activation. FX was also activated by fVIIa (40 pM) in the presence of 0.5 nM lipidated tissue factor and by flXa (2 nM) in the presence of fVIIIa (4 U/ml) on POPS vesicles (20 μM). Aliquots of the activation reactions were removed over time and quenched into an EDTA-containing buffer. Initial rates of activation were quantitated from the initial rate of Spectrozyme FXa™ hydrolysis using a Molecular Dynamics (Sunnyvale, CA) kinetic microplate reader.

Inhibition Studies — Both zymogens (100 nM in assay buffer) were activated to completion by RVV-X (2 nM). The enzymes were diluted to 0.5 nM in assay buffer and incubated with Spectrozyme FXa™ (100 μM) in the presence of inhibitor: ATIII (0-4 μM), ATIII (0-50 nM) in the presence of heparin pentasaccharide (0.5 μM), ATIII (0-16 nM) in the presence of subsaturating concentrations of heparin (2.5 mU/ml), or TFPI (0-50 nM). The appearance of the chromophore, -nitroanilide acetate (pNA), was monitored over time at 405 nm. Primary data were fitted, using the Marquardt-Levy algorithm (SigmaPlot, Jandel Scientific, Corte Madera, CA), to an exponential equation of the form:

A = A,(l-e- )+A_o,

where A is the observed absorbance, Ar is the amplitude of the curve, koj_* is the apparent first-order rate constant with units of s^"\ t is time in seconds, and A_o is the initial absorbance. Each calculated apparent rate constant (ko s) was then plotted as a linear function of the effective inhibitor concentration and the resulting slope was calculated as the apparent second order rate constant. Thrombin formatio — In all cases, the variant fX species was compared to recombinant wild type. All prothrombin was iinmunodepleted of plasma fX. Kinetic values were derived from the least squares fit of the data based on the equation:

V = Vmax[Z]/(Kapp+[Z])

where v is the observed initial rate of thrombin formation, Vmax is the maximal initialrate of thrombin formation, [Z] is the concentration of the component being varied in the experiment, and Kapp is the apparent concentration of the variable component required to reach half maximal thrombin formation under the conditions specified.

Thrombin generation by 10 nM fXa from prothrombin alone was characterized by varying the prothrombin concentration (0-450 μM). Thrombin formation from 10 μM prothrombin by 0.5 nM fXa in the presence of fVa was quantitated over a range of fVa concentrations (0-1.0 μM). Thrombin generation on 20 μM PC:PS vesicles was determined using 20 pM fXa, 1.0 μM prothrombin, and varying fVa from 0-1.0 nM. Thrombin formation from 1.0 μM prothrombin on 10⁸/ml thrombin activated (0.5 U/ml) platelets was tested over a range of fXa concentrations (0-1.0 nM). In all experiments, aliquots were removed from the reaction tube at specified times and diluted at least 10-fold into EDTA buffer (5 mM EDTA, 1 mg/ml PEG 8000, 1 mg/ml BSA, 10 mM HEPES, 100 mM NaCl, pH 7.0). A saturating concentration (500 μM) of the thrombin chromogenic substrate, Spectrozyme TH™, was added to quenched samples and the appearance of the chromophore, p-nitroanilide acetate (pNA), was monitored over time at 405 nm. The concentration of thrombin formed was calculated by comparison to a standard curve using purified thrombin. The assay bufTer used in all experiments, unless otherwise noted, was 5 M CaCl₂, 1 mg/ml PEG 8000, 1 mg/ml BSA, 10 mM HEPES, 100 M NaCl, pH 7.0.

Competition binding assa — FXS379 was labeled with I using Bolton- Hunter reagent, Amersham (Arlington Heights, IL) in a modified procedure. After the supplied solvent was evaporated under nitrogen, fXS379, pH 8.5, was added and incubated for one hour on ice. Next, 10 mM HEPES, 100 mM NaCl, 200 mM glycine, pH 7.4, was added for an additional 5 min on ice. Radioactivity not incorporated into protein was removed using a Bio- Spin 6 column, Bio-Rad Laboratories (Hercules, CA). The specific activity of the labeled protein was typically 2000 cpm/ng. Labeled fXS379 was activated using RVV-X as described above and stored at -70°C in 20% glycerol (v/v).

Carboxyl latex beads (1.0 μm) from Interfacial Dynamics Corp. (Portland, OR) were washed exhaustively in 1% ethanol, followed by water, and then lyophilized. The dried beads (20 mg) were mixed with 3.8 μmoles of PC:PS (3: 1) in hexane: ethanol (9:1) to provide a ratio of lipid surface area to bead surface area of -10. The suspension was sonicated and then dried under nitrogen. The dried bead pellet was washed in assay buffer and the beads were collected by centrifugation at 12,000 x g and resuspended in 1.0 ml of assay bufferBeads coated with PC only were also prepared and used in pilot experiments to demonstrate specificity; i.e.; no fVa-dependent binding without PS. For binding studies, PC:PS coated beads were added at a final concentration of 0.2% (v/v). This concentration of beads was determined to be equivalent to 20 μM PC:PS vesicles for supporting thrombin formation by fVa and fXaWT. Uncoated beads prepared in the same manner but with no lipid added were added at a final concentration of 0.8% (v/v) to aid centrifugation and pellet visualization; uncoated beads were otherwise without effect in the assay. Human fVa was used at a final concentration of 1 nM. The reaction was incubated with mixing for 10 min. I-fXaS379 (final concentration 1.0 nM) was mixed with increasing concentrations of either fXaWT (0-7.0 nM) or fXaR347N (0-50 nM). The fXa solution was added to the fVa/bead mixture and incubated with mixing for an additional 10 min. The beads were collected by centrifugation and the pellets and supematants were counted separately. Non-specific binding was determined from reactions not containing fVa and subtracted from all counts bound in the pellets. The percentage of specific bound fXa was quantitated based on reactions with no unlabeled fXa.

Structure — The structure of fXa was visualized from the published coordinates (19) using RasMol v2.5, developed by Roger Sayle at Glaxo Research and Development (Greenford, Middlesex, UK). In the structure, the Gla domain is missing, EGF1 is disordered, and residues 329-333 are disordered or missing, so these regions are not visualized.

RESULTS

Mutational methodology — The expression system affords production of variant fX species that have been selectively mutated. Residue 347 was changed to asparagine (PYVDRNSCK → PYVDNNSCK) because this eliminates a charge on the surface of fX. It also creates a new N-linked glycosylation site, and a new oligosaccharide side chain might effectively block inter-molecular interactions at this region of the protein. However, as indicated by the equivalent electrophoretic mobility of fXR347N and fXWT (Fig. 1), residue 347 is apparently inaccessible during intracellular processing. FXR347N involves only a single substitution of Arg347.

Activation — Since fXaR347N has normal activity toward Spectrozyme FXa™ (data not shown), hydrolysis of this substrate was used to monitor initial rates and extents of activation as described in Methods. Activation of fXR347N by RVV-X, the intrinsic Xase complex (fIXa/fVIIIa), and the extrinsic Xase complex (fVIIa TF) are all normal (data not shown). Substitution of Arg347 with asparagine does not influence the activation of fX.

Inhibitio — Inhibition of activated DCR347N by TFPI and ATIII was compared directly to inhibition of fXaWT (Table I.) TFPI and ATIII inhibit the fXaR347N and fXaWT at the same rate. ATIII in the presence of a heparin pentasaccharide also inhibits fXaR347N at a normal rate. In the presence of a sub-saturating concentration of full-length heparin, however, fXaR347N is only inhibited at 15% of the noπnal rate. Arginine 347 may therefore contribute to the heparin binding capacity of fXa.

Prothrombin activation in the absence of a lipid surface — In order to evaluate the function of fXaR347N relative to fXaWT, the initial rate of prothrombin activation by each enzyme was determined at different prothrombin concentrations (see Methods). The activity of fXaR347N towards prothrombin alone in the absence of fVa and phospholipid is normal (Fig. 2). The apparent Km of fXaWT for prothrombin is 237 μM, compared to 242 μM for fXaR347N; the turnover rate of fXaWT is 0.026 mol prothrombtn/sec/mol fXa, as compared to 0.022 mol prothrombin sec/mol fXa for fXaR347N.

The catalytic activity of fXaWT toward prothrombin is markedly enhanced by fVa; e.g., at 10 μM prothrombin, fXaWT is catalytically more than 1000 times faster in the presence of 500 nM fNa. In contrast, the activity of fXaR347N is not detectably affected at fVa concentrations up to 1 μM (Fig. 3). Since fXaR347N interacts with and cleaves prothrombin normally, it either has a much weaker affinity for fNa or fNa prothrombin compared ^"to fXaWT, or it cannot undergo structural rearrangements induced by tNa that increase the catalytic activity of fXaWT (20,21).

Prothrombin activation in the presence of phospholipid — It has been suggested that appropriate phospholipid surfaces increase the local concentrations of fXa and fVa, thus maximizing complex formation (22- 25). Defined phospholipid surfaces such as PC:PS (3: 1) vesicles can support thrombin generation by fXa and fVa in a manner equivalent to that of more biological but less defined surfaces such as activated platelets. On PC.PS vesicles fXaR347Ν has a markedly reduced apparent affinity for fVa as compared to fXaWT (2.3 nM vs 0.1 nM, respectively, Fig 4). However, the maximal catalytic potential of fXaR347N and fXaWT are equivalent in this system (29 mol thrombi /sec/mol fXa). Substitution at position 347 must therefore modify a key surface epitope that significantly contributes to the intermolecular interaction between fXa and fVa (or fVa/prothrombin) without influencing the catalytic potential of fXa.

Since activated platelets provide both an ideal surface for prothrombinase complex assembly and saturating amounts of cofactor, the generation of thrombin on platelets (10⁸/ml) was examined. Under these conditions, f aWT and fXaR347N are similar, with apparent affinities of 134 pM and 186 pM and maximal rates of thrombin formation of 2726 pM sec^'1 and 2907 pM sec^"1, respectively, in the experiment depicted in Fig. 5.

Competition binding assay — Based on these observations, the simplest explanation for the phenotype of fXR347N is a selective attenuation of the interaction with fVa. However, since the functional studies rely on the amplification inherent in generating thrombin from prothrombin, it is possible that the reduced affinity is for a fVa/prothrombin complex rather than for fVa alone. The affinity of fXa for fVa in the absence of all other prothrombinase complex components is roughly 1 μM (26,27), which precludes direct examination of the binary interaction by conventional methods given the limited quantities of the recombinant protein and fNa that are available. A competition binding assay was therefore developed to examine the interaction between fVa and the fXa species on a phospholipid surface, but in the absence of prothrombin.

In the assay, PC:PS coated latex beads provide the surface. First, the concentration of coated beads equivalent to 20 μM PC: PS vesicles in supporting thrombin generation was empirically determined. Next, the amount of fVa required to support approximately 80% of the maximal thrombin generation rate with 1.0 μM prothrombin and saturating fXaWT was established. Finally, the amount of ^,2SI-labeled fXaS379 needed to reach approximately 80% of maximal specific binding at this concentration of fVa (but no prothrombin or unlabeled fXa) was determined. Specific binding is defined as that component of total binding shown to be fVa, phosphatidylserine, and calcium dependent in control experiments. Recombinant fXa with an alanine substitution at the active-site serine was used as the labeled species for greater stability over time; i.e., to minimixe fXa-dependent proteolysis.

In the binding studies, the fNa was pre-incubated with the coated vesicles. Then, increasing concentrations of fXaWT or fXaR347Ν mixed with the fixed amount of ¹²⁵I-labeled fXaS379 were allowed to compete for the bound fVa. This assay format was chosen to eliminate the possibility that fXaWT and fXaR347N might not be altered identically if each were directly labeled. The concentration of fXaR347N required to compete with the ^,2SI-fXaS379 is considerably higher (~10-fold) than for fXaWT (Fig. 6). These data are a direct demonstration of the attenuated binding capacity of fXaR347N for fVa on a lipid surface in the absence of prothrombin.

DISCUSSION

We have characterized a recombinant fX with asparagine substituted for arginine at position 347. The mutation has no discernible impact on synthesis or secretion in our expression system. Moreover, normal activation by fNIIa TF, fiXa/fVUIa and RVV-X, and normal interaction of the activated species with Spectrozyme FXa™, prothrombin, ATIII and TFPI attest to normal folding and active site geometry.

Arg347 is part of a large basic epitope that Pad anabhan and coworkers have described as a heparin binding domain (19). The loss of the charge contributed by this residue most likely weakens the fXa-heparin interaction, thereby causing the reduced (--6-fold) impact of heparin on the inhibition of fXaR347N by ATIII. Acceleration of inhibition by the pentasaccharide, which need only interact with ATIII to have its effect, is normal. To insure sensitivity to any difference attributable to the mutation, the concentrations of pentasaccharide and heparin tested were empirically determined to support half maximal acceleration of inhibition of fXaWT. These data are not sufficient for rigorous analysis of the absolute contribution of Arg347 to heparin binding. Similarly, functional definition of the complete heparin binding domain will require additional mutagenesis and is beyond the scope and focus of this report.

In addition to the effect on heparin binding, we find that Arg347 is critically important for the interaction between fXa and fVa. Since fVa may enhance the catalytic activity of fXa by mediating conformational changes in the enzyme, we considered the possibility that Arg347 could somehow be required for those rearrangements but not for fVa binding. However, the competitive binding assay demonstrates the reduced ability of fXaR347N to compete for fVa on a phospholipid surface independent of its catalytic function. Moreover, if a crucial rearrangement were blocked by the substitution, the fNa accelerated rate of thrombin generation by fXaR347Ν should never be normal. But given a suitable phospholipoid surface and excess fVa, fXaWT and fXaR347A have equivalent catalytic potential.

Chattopadhyay et. al., based on the inhibition of functional assays by peptides mimicking linear sequences of fX, concluded that residues 211-222 and 254-274 in fXa are important for interactions with fNa (28). These sequences are shown in dark and light blue in Fig. 7, an illustration of the fXa structure reported by Padmanabhan and coworkers. They are some distance from Arg347, whose position is shown in green. It is possibile that substitution of Arg347 attenuates fVa binding by somehow perturbing the structure of a distant binding epitope; e.g., residues 211-222 and 254-274. But we think it is more likely that Arg347 actually is part of the fNa binding site of fXa. First, based on the fXa crystal structure, there is no obvious way changing Arg347 to asparagine would cause a rearrangement apparent across the molecule without having any impact on activation, catalysis or inhibition. Second, the sequence of fVIIa corresponding to residues 346- 349 in fX (164-167 in chymotrypsin numbering), together with parts of its EGF2 domain, forms a contact surface with TF, based on the recently solved fNIIa/TF crystal structure (29). The homologous residues in fX might well interact with fVa. Finally, fXa bound to fVa is protected from inhibition by antithrombin III and heparin (30). It is appealing to propose that part of this protection is a consequence of the overlap of heparin and fVa binding sites on fXa. Just as is the case for fVIIa and TF, it is likely that many other residues besides Arg347 are important for the fXa interaction with fVa. Definition of the entire surface contact will require additional mutagenesis or a convincing crystal structure of the complex.

This substitution also creates a potential, though unused, site for N- linked glycosylation. There was reason to believe that an oligosaccharide side chain might be added. Protein C is usually glycosylated (31-33) at Asn329 (PVWHNECS), even though it is unususal for cysteine to follow two residues after the asparagine instead of serine or threonine. In contrast, fX is not glycosylated at the homologous residue, Asn348 (PYVDRNSCK). All N-linked carbohydrate on fX is attached to the peptide released on activation, residues 143-194. Since there are few reported examples of glycosylated NXC sites, the impact of surrounding residues is less well understood. The difference in susceptibility to glycosylation of the two proteins could be a consequence of the local sequence variation rather than higher order structural constraints. However, fXR347N, with both a typical and the unususal site juxtaposed (PYVD NSCK), is still not glycosylated in this region. Based on the importance of Arg347 to fNa binding, it is probably no accident that Asn348 cannot be glycosylated, as the addition of an oligosaccharide side chain might well prevent binding to fNa. It may be that this sequence is less accessible in fX than in protein C because some strucural element unique to fX masks the region. The activation peptide is an attractive candidate. With several carbohydrate side chains and 40 extra amino acids compared to the activation peptide of protein C, it could easily "cover" Asn348 and the surrounding residues. When it is released by activation, long after the "threat" of intracellular glycosylation is over, the site would become accessible. And this in turn might explain why zymogen fX is ineffective in competing with fXa for fNa binding (34). The three peptide fragments designated herein for convenience by the conventional one-letter symbols are also shown by the required 37 CFR §1.821-1.825 three-letter symbols in the appended Sequence Listing as follows:

Pro Tyr Val Asp Arg Asn Ser Cys Lys [SEQ ID N0:1] 1 5

Pro Tyr Val Asp Asn Asn Ser Cys Lys [SEQ ID NO: 2] 1 5

Pro Val Val Pro His Asn Glu Cys Ser [SEQ ID NO: 3] 1 5

TABLE 1

Inhibition of fXaR3<7N and fXaWT

Inhibitor Second Order Rate Constants

M-'-sec 'xlO⁴

(mean±S.E.) fXaWT fXaR347N

TFPiα U.3±1.6 10.7±0.6

ATIIiα 0.16±0.02 0.14±0.03

ATIII+PS* 2 l.l±1.8 17.1±2.5

ATIII+HEPARINC 155.4±0.3 23.4±0.7 ^a 0.5 nM fXa was incubated with varying concentrations of recombinant TFPI (0-50 nM) or ATIII (0-4 μM).

* 0.5 nM fXa was incubated with varying concentrations of ATIII

(0-50 nM) in the presence of 0.5 μM heparin pentasaccharide. c 0.5 nM fXa was incubated with varying concentrations of ATIII

(0-16 nM) in the presence of full-length heparin (2.5 mU/ml).

Second order rate constants were detennined as described in Materials and

Methods. Various other examples will be apparent to the person skilled in the art after reading the present disclosure without departing from the spirit and scope of the invention. It is intended that all such other examples be included within the scope of the appended claims.

REFERENCES

1. Jesty, J., Spencer, A. K., and Ne erson, Y. (1974) J. Biol. Chcm. 249, 5614-5622

2. DiScipio, R. G., Hermodson, M. A., Davie, E. W. (1977) Biochemistry 16, 5253-5260

3. Pfeiffer, R. A., Ott, R., Gilgenkrantz, S., and Alexandre, P. (1982) Human Genet. 62, 358-360

4. Nemerson, Y., and Bach, R. (1982) Prog. Hemostasis Thromb. 6, 237- 261

5. Nemerson, Y. (1986) Blood 11, 1-8

6. Mann, K. G., Jenny, R. J., and Krishnaswamy, S. (1988) Annu. Rev. Biochemistry 57, 915-956

7. Rosenberg, R. D., and Da us, P. S. (1973) J. Biol. Chcm. 248 6490- 6505

8. Owen, B. A., and Owen, W. G. (1990) Biochemistry 29, 9412-9417

9. Broze, G. J., Jr., Warren, L. A., Novotny, W. F., and et al. (1988) Blood 71, 335-343 lO. Girard, j. j_^ y/arτen, L. A., Novotny, W. F., Higuchi, D. A., Girard, J. J, and Miletich, J. P. (1989) Nature 338, 518-520

11. Krishnaswamy, S. (1990) J Biol. Chcm. 265, 3708-3718

12. Rudolph, A. E., Mullane. M. P., Porche-Sorbet, R. P., Tsuda, T., and Miletich, j. P. (1996) J Biol. Chcm. Ill, 28601-28606

13. Broze, G. J., Jr. and Miletich, J. P. (1984) J. of Clin. Invest. 73, 933- 938

14. Fass, D. N., Knutson, G. J., and Katzmann, J. A. (1982) Blood 59, 594-600 15. Kisiel.W., Heπnodson, M. A., and Davie, E. W. (1976) Biochemistry 15, 4901-4906

16. Laemmli, U. K. (1970) Nature 227, 680-685

17. MacDonald, R. C, MacDonald, R. I., Menco, B. Ph.M., Takeshita, K, Subbarao, N. K., and Hu, L. (1991) Biochim. and Biophys. Ada 1061, 297- 303

18. Graham, F. L., and Eb, A. J. (1973) Virology 52, 456-467

19. Padmanabhan, K, Padmanabhan, K. P., Tulinsky, A., Park, C. H., Bode, W., Huber, R., Blankenship, D. T., Cardin, A. D., and Kisiel, W. (1993) J. of Mol. Biol. 232, 947-966

20. Husten, J. E., Esmon, C. T., and Johnson, A. E. (1987) J. Biol. Chem. 262, 12953-12961

21. Walker, R. K, and Krishnaswamy, S. (1993) J. Biol. Chem. 268, 13920-13929

22. Dahlback, B., and Stenflo, J. (1978) Biochemistry 17, 4938-4945

23. Tracy, P. B., Nesheim, M. E., and Mann, K. G. (1981) J. Biol. Chem. 256, 743-751

24. Rosing, J., Tans, G., Grovers-Riemslag, J. W., Zwaal, R. F., and Hemker, C. (1980) J Biol. Chem. 255, 274-283

25. Nesheim, M. E., Tracy, R. P., and Mann, K. G. (1984) J. Biol. Chem. 259, 1447-1453

26. Boskovic, D. S., Giles, A. R., Nesheim, M. E. (1990) J Biol. Chem. 265, 10497-10505

27. Pryzdial, E. C. G., and Mann, K. G. (1991) J. Biol. Chem. 266, 8969- 8977

28. Chattopadhyay, A., James, H. L., and Fair, D. S. (1992) J. Biol. Chem. 267, 12323-12329 29. Banner, D. W., D'Arcy, A., Chene, C, Winkler, F. K, Guha, A., Konigsberg, W. H., Nemerson, Y., and Kirchhofer, D. (1996) Nature 380, 41-46

30. Lindhout, T., Baruch, D., Schoen, P., Franssen, J., and Hemker, C. (1986) Biochemistry 25, 5962-5969.

31. Stenflo, J., and Fernlund, P. (1982) J Biol. Chem. 257, 12180-12190

32. Miletich, J.P. and Broze G.J. Jr. (1990) J. Biol. Chem. 265, 1 1397- 11404.

33. Grinnell, B. W., Walls, j. D., and Gerlitz, B. (1991) J Biol. Chem. 266, 9778-9785

34. Miletich, J.P., Jackson CM., and Majerus P.W. (1978) J. Biol. Chem. 253, 6908-6916.

SEQUENCE LISTING (1) GENERAL INFORMATION:

(i) APPLICANT:

(A) NAME: Washington University

(B) STREET: One Brookings Drive

(C) CITY: St. Louis

(D) STATE: Missouri

(E) COUNTRY: United States of America

(F) POSTAL CODE: (ZIP): 63130

(G) TELEPHONE: (847)470-6500 (H) TELEFAX: (847)470-6881

(A) NAME: Miletich, Joseph P.

(B) STREET: Division of Laboratory Medicine

Department of Pathology

Washington University School of Medicine

Box 8118

660 South Euclid Avenue

(C) CITY: St. Louis

(D) STATE: Missouri

(E) COUNTRY: United States of America

(F) POSTAL CODE (ZIP): 63110

(G) TELEPHONE: (847)470-6500 (H) TELEFAX: (847)470-6881

(ii) TITLE OF INVENTION: Factor X Variant

(iii) NUMBER OF SEQUENCES: 3

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Word Perfect 5.0

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION BUMBER:

(B) FILING DATE:

(C) CLASSIFICATION:

(vii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: US 60/040,047

(B) FILING DATE: 07-MAR-1997

(C) CLASSIFICATION: (2) INFORMATION FOR SEQ ID NO:l:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 9 amino acids

(B) TYPE: amino acid

(C) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:l:

Pro Tyr Val Asp Arg Asn Ser Cys Lys 1 5

(2) INFORMATION FOR SEQ ID NO: 2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 9 amino acids

(B) TYPE: amino acid

(C) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:

Pro Tyr Val Asp Asn Asn Ser Cys Lys 1 5

(2) INFORMATION FOR SEQ ID NO 3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 9 amino acids

(B) TYPE: amino acid

(C) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3:

Pro Val Val Pro His Asn Glu Cys Ser 1 5

Claims

WHAT IS CLAIMED IS:

1. A human Factor X variant having the arginine residue at amino acid position 347 replaced with asparagine.

2. The method of substantially reducing the affinity of human Factor X for Factor Va without substantially reducing the catalytic impact of Factor Va binding comprising replacing the arginine residue at amino acid position 347 with asparagine.