Molecular Basis of the Clotting Defect in a Bleeding Patient Missing the Asp-185 Codon in the Factor X Gene

Abstract

Factor X (FX) is a vitamin K-dependent plasma zymogen, which following activation to factor Xa (FXa), converts prothrombin to thrombin in the blood clotting cascade. It was recently demonstrated that a natural variant of FX carrying the Asp-185 deletion (FX-D185del, chymotrypsinogen numbering) was associated with mild bleeding in a patient with severe FX deficiency. In this study, we expressed FX-D185del in mammalian cells and characterized its properties in appropriate kinetic assays in purified systems. We discovered that while the FX variant can be normally activated by physiological activators; both amidolytic and proteolytic activities of the mutant are dramatically impaired. Interestingly, factor Va (FVa) significantly improved the proteolytic defect when the mutant protease was assembled into the prothrombinase complex. Thus, in contrast to >50-fold catalytic defect in the absence of FVa, the variant activated prothrombin with only ~2.5-fold decreased catalytic efficiency in the presence of the cofactor. The FXa variant dramatically lost its susceptibility to inhibition by antithrombin and tissue factor pathway inhibitor, thus exhibiting ~2–3 orders of magnitude lower reactivity with the plasma inhibitors. Further studies revealed that Na⁺ no longer activates the variant protease, suggesting that the functionally important allosteric linkage between the Na⁺-binding and the P1-binding sites of the protease has been eliminated. These results suggest that the lower catalytic efficiency of FXa-D185del in the bleeding patient may be partially compensated by the loss of its reactivity with plasma inhibitors, possibly explaining the basis for the paradoxical severe FX deficiency with only mild bleeding tendency for this mutation.

Introduction

Factor X (FX) is a vitamin K-dependent plasma zymogen, which upon activation to factor Xa (FXa), assembles into the prothrombinase complex (FVa, negatively charged phospholipid vesicles and calcium ion) to activate prothrombin to thrombin at vascular injury sites (1,2). Similar to other vitamin K-dependent coagulation proteases, the structure of FXa consists of a light chain and a heavy chain linked together by a disulfide bond (3). Following activation by either one of its physiological activators, tissue factor (TF)-factor FVIIa (FVIIa), or factor IXa (FIXa)-factor VIIIa (FVIIIa), FXa is capable of slowly activating prothrombin on negatively charged membrane phospholipids (exposed by the injury), thereby generating a small amount of thrombin (4). Thrombin then activates the procofactors FV and FVIII to their active forms in order to facilitate the assembly of the prothrombinase and intrinsic Tenase complexes, respectively, thereby effectively mediating thrombin generation during the propagation phase of the clotting cascade (4). The complex formation of FXa with FVa is essential for the generation of physiological levels of thrombin at vascular injury sites in a timely fashion as the cofactor promotes the catalytic efficiency of FXa in the prothrombinase complex by five orders of magnitude (2,4). The proteolytic activity of FXa is regulated by two natural plasma inhibitors, antithrombin (AT) and tissue factor pathway inhibitor (TFPI) (5,6). Both inhibitors play central roles in the regulation of FXa in the clotting cascade since their deficiency is associated with hypercoagulability and severe thrombotic disorders (7–9).

Hereditary FX deficiency is a very rare bleeding disorder which is inherited as an autosomal recessive trait with a frequency of approximately 1 per million in general populations (10). Molecular defects in the gene coding for FX are the main causes of FX deficiency and thus far more than 100 different mutations have been identified, with the homozygous carriers exhibiting either mild or severe bleeding phenotypes (10–13). Most of the identified mutations are located in either the N-terminal light chain (in particular in the γ-carboxyglutamic acid domain) or in the C-terminal heavy chain which contains the catalytic domain of the protein (12). FX deficiencies are classified as type I, characterized by a lower FX antigen level; type II, characterized by lower FXa activity level; or a combination of both type I and type II where both the expression and activity levels of FX are decreased (10–13). In a recent study, a type II FX-deficient bleeding patient was identified who had deletion of a GAC codon in the FX gene, resulting in the deletion of Asp-185 (D185del) (chymotrypsinogen numbering system) from the amino acid sequence of the patient’s FX protein (14,15). The plasma clotting activity level of the patient’s FX was less than 1% in both intrinsic and extrinsic pathways (14). Nevertheless, despite such a severe FX deficiency, the patient did not experience severe bleeding diathesis, but rather mild bleeding episodes such as recurrent epistasis, bleeding of gums after brushing and dental extraction (14). To understand the molecular basis of this mild clotting defect, in this study we expressed the D185del mutant of FX in HEK-293 cells, and after purification to homogeneity characterized its properties in appropriate kinetic assay systems. We discovered that the mutant zymogen is activated normally by both of the physiological activators of FX. However, the catalytic activity of the FXa variant toward the natural substrate prothrombin and small synthetic substrates was dramatically impaired. Moreover, the reactivity of the FXa variant with specific plasma inhibitors of FXa was impaired by 2–3 orders of magnitude. Further studies revealed that the loss of Na⁺ activatability may account for the catalytic defect of the FXa variant since unlike wild-type, Na⁺ ion did not influence the activity of the variant. Nevertheless, FVa restored most of the catalytic defect of the variant FXa toward prothrombin since the variant exhibited only 2–3 fold slower catalytic efficiency in the prothrombinase complex. This result, together with the lack of the susceptibility of the FXa variant to neutralization by the plasma inhibitors, accounts for the mild clinical phenotype of this severely FX deficient subject.

Materials and Methods

Mutagenesis and expression of recombinant FX

The expression and purification of wild-type FX in HEK-293 cells has been described (16). The FX variant containing a GAC codon deletion, coding for Asp-185 (FX-D185del) in the chymotrypsinogen numbering system (15), was constructed by standard PCR mutagenesis methods and expressed in the same expression/purification vector system as described (16). After confirmation of the accuracy of the mutagenesis by DNA sequencing, the construct was introduced into HEK-293 cells and the mutant protein was isolated from 20-L cell culture supernatants by a combination of immunoaffinity and ion exchange chromatography using the HPC4 monoclonal antibody and a Mono Q ion exchange column as described (16). The fully γ-carboxylated protein was eluted from the ion exchange column at ~0.4 M NaCl as described (17). Both wild-type FX and FX-D185del were activated by RVV-X and FXa derivatives were separated from the snake venom on a Mono Q column chromatography as described (17). Active-site concentrations of FXa derivatives were determined by an amidolytic activity assay and titrations with human AT assuming a 1:1 stoichiometry as described (17).

Human plasma proteins including FVa, FVIIa, FIXa, prothrombin, antithrombin (AT), the factor X-activating enzyme from Russell’s viper venom (RVV-X) were purchased from Haematologic Technologies Inc. (Essex Junction, VT). Tissue factor pathway inhibitor (TFPI) was from Monsanto Chemical Co. (St. Louis, MO). Phospholipid vesicles containing 80% phosphatidylcholine and 20% phosphatidylserine (PC/PS) were prepared as described (18). Recombinant tissue factor (TF) was incorporated into PC/PS vesicles as described (18). Recombinant tick anticoagulant peptide (rTAP) was a generous gift from Dr. G. Vlasuk (Corvas International Inc., San Diego, CA). Human recombinant FVIIIa was a generous gift from Dr. Philip Fay (University of Rochester, Rochester, NY). The active AT-binding pentasaccharide (H₅) fragment of heparin (fondaparinux sodium) and unfractionated heparin (heparin sodium injection, USP, 10,000 units/mL) were purchased from Quintiles Clinical Supplies (Mt. Laurel, NJ). The chromogenic substrates, Spectrozyme FXa (SpFXa) was purchased from American Diagnostica (Greenwich, CT); S2765, S2222 and S2238 were purchased from Kabi Pharmacia/Chromogenix (Franklin, OH).

Activation by the FVIIa-TF complex

The initial rate of the concentration-dependence of the activation of recombinant wild-type and mutant FX by the FVIIa-TF complex was evaluated as described (17,18). Briefly, FX derivatives (6–800 nM) were incubated with FVIIa (0.1 nM) in complex with relipidated TF (2 nM) in 0.1 M NaCl, 0.02 M Tris-HCl (pH 7.5) containing 0.1 mg/mL bovine serum albumin, 0.1% polyethylene glycol 8000 and 5 mM Ca²⁺ (TBS/Ca²⁺) for 2–15 min at room temperature in 30-μL reactions in 96-well assay plates. The activation reactions were terminated by an addition of 20 μL of EDTA to obtain a final concentration of 20 mM, and the rate of FXa generation was determined by an amidolytic activity assay using 50 μL of 0.2 mM SpFXa.

Activation by the FIXa-FVIIIa complex

The initial rate of activation of recombinant FX derivatives (6–800 nM) by FIXa (0.1 nM) in complex with FVIIIa (10 nM) was monitored on PC/PS vesicles (50 μM) in TBS/Ca²⁺ for 2–10 min at room temperature in 30 μL reactions in 96-well assay plates as described (17). Activation reactions were terminated by an addition of 20 μL EDTA and the concentration of FXa generated was determined from standard curves as described above.

Activation by RVV-X

The initial rate of the concentration-dependence of activation of recombinant wild-type FX and FX-D185del by RVV-X was studied as described (17). Briefly, the activation of increasing concentrations of FX derivatives (6–800 nM) by RVV-X (0.1 nM) was monitored in TBS/Ca²⁺ for 2–10 min at room temperature in 30-μL reactions in 96-well assay plates. Following termination of the reaction by EDTA, the rate of FXa generation was determined as described above.

Cleavage of chromogenic substrates

The steady-state kinetics of hydrolysis of chromogenic substrates by FXa derivatives (1–10 nM) was determined in TBS/Ca²⁺. The rate of hydrolysis was measured at 405 nm at room temperature by a V_max kinetic microplate reader (Molecular Devices, Menlo Park, CA) as described (16,17). The K_m and k_cat values for the substrate hydrolysis were calculated from the Michaelis-Menten equation and the specificity constant for each protease was expressed as the ratio of k_cat/K_m.

Apparent dissociation constant (K_d(app)) for Na⁺

K_d(app) for the interaction of Na⁺ with each protease was determined from the effect of increasing concentrations of Na⁺ on the activity of the protease using the chromogenic substrate S2765 (50 μM for FXa-WT and 1 mM for FXa-D185del) at room temperature in 0.05 M Tris-HCl, pH 7.5, 0.1 mg/mL BSA, 0.1% PEG 8000 as described (19). The amidolytic activity of FXa derivatives was not influenced by the ionic strength of the reaction buffer as demonstrated previously (20).

Activation of prothrombin

The initial rate of prothrombin activation by both the wild-type and mutant FXa was determined in both the absence and presence of FVa on PC/PS vesicles (prothrombinase) at room temperature as described (17). Briefly, in the absence of FVa, prothrombin (1 μM) was incubated with each FXa (10 nM) in TBS/Ca²⁺ on PC/PS vesicles (25 μM). The concentration dependence of prothrombin activation was monitored from the rate of thrombin generation by an amidolytic activity assay using the chromogenic substrate S2238 (100 μM). The concentration of thrombin generated was determined from a standard curve prepared from the cleavage rate of S2238 by known concentrations of recombinant thrombin under exactly the same conditions. The initial rate of prothrombin activation in the presence of human FVa (20 nM) was measured by incubating each FXa derivative (50 pM) with increasing concentration of prothrombin (30–2000 nM) in TBS/Ca²⁺ on PC/PS vesicles (25 μM). Following 0.5–2 min incubation at room temperature, EDTA was added to a final concentration of 20 mM and the concentrations of thrombin generated were determined from a standard curve as described above. The same prothrombinase assay was used to measure the FVa concentration dependence of prothrombin activation by both wild-type FXa and FXa-D185del.

Inactivation by antithrombin

The reactivity FXa-D185del with AT was compared with that of wild-type FXa in both the absence and presence of heparin under pseudo-first order rate conditions by a discontinuous assay as described (19,20). Briefly, FXa (1 nM wild-type and 10 nM D185del) was incubated with 0.25–1 μM AT in the absence, or 5–500 nM AT in the presence of 2 μM fondaparinux or 0.5 μM unfractionated heparin in TBS/Ca²⁺. All reactions were carried out at room temperature in 50 μL volumes in 96-well polystyrene plates. After a period of time (10–240 min in the absence and 0.25–30 min in the presence of heparin), 50 μL of S2765 (100 μM for FXa and 1 mM for FXa-D185del) in TBS was added to each well and the remaining enzyme activity was measured using a V_max kinetic plate reader as described above. The observed pseudo-first-order, and second-order rate constants (k₂) of FXa inhibition in the absence and presence of heparin cofactors were calculated as described (19,20).

Interaction with TFPI and rTAP

The interaction of FXa derivatives with either TFPI or rTAP was evaluated by a two-step assay mechanism. In the first stage, each FXa derivative (0.1 nM) was incubated with different concentrations of TFPI or rTAP (0.15–400 nM) in TBS/Ca²⁺ in 20 μL volumes in 96-well polystyrene plates. Following 30 min incubation at room temperature, which was sufficient to establish equilibrium, 10 μL of the protease inhibition reaction was used to activate prothrombin (1 μM) in a prothrombinase assay (20 nM FVa and 25 μM PC/PS) for 0.5–2 min. The K_i values were estimated from the decrease in the rate of thrombin generation (monitored by the hydrolysis of S2238) by nonlinear regression analysis of kinetic data using a quadratic equation describing tight binding interactions as described (19). Neither one of the inhibitors can inhibit thrombin.

Results

Expression and characterization of FX-D185del

To understand the molecular basis of the clotting defect in the patient carrying the Asp-185 deletion in the FX protein, an FX construct carrying the mutation was expressed in HEK-293 cells and the mutant protein was purified from the cell culture supernatants by an immunoaffinity and ion exchange chromatography as described (17). SDS-PAGE analysis under non-reducing conditions indicated that the FX variant has been purified to homogeneity and that it migrates with an apparent molecular mass of ~75 kDa as observed for the wild-type counterpart (Fig. 1A). Analysis of activation profiles of the mutant (FX-D185del) by the physiological activators of FX in both the intrinsic (FIXa-FVIIIa) and extrinsic (FVIIa-TF) pathways as well as by RVV-X suggested that the FX variant can be activated with rates similar to those observed for wild-type recombinant FX (Fig. 1B-D). In the extrinsic pathway, the activation of FX-D185del exhibited ~2-fold lower k_cat under experimental conditions described under the legend of Fig. 1C. However, the lower k_cat of activation was compensated by ~2-fold lower K_m (Fig. 1C). These results indicate that the FX variant has been, most likely, folded properly. This is expected since Asp-185 is located near the catalytic pocket in a hydrophilic loop far away from the activation peptide of the zymogen which is recognized and cleaved by the physiological activators.

Figure 1.

SDS-PAGE of the recombinant wild-type FX and FX-D185del proteins and analysis of their activation profiles by different activators. A, under non-reducing conditions: lane 1, recombinant wild-type FX; lane2, FX-D185del; and lane 3, molecular mass standards in kDa. B, the activation of FX derivatives by the FIXa-FVIIIa complex was monitored as described under “Materials and Methods”. C, the activation of FX derivatives by the FVIIa-TF complex was monitored as described under “Materials and Methods”. D, the activation of FX derivatives by RVV-X was monitored as described under “Materials and Methods”. All data are the average of at least 3 measurements ± SD.

Amidolytic activity

Unlike its normal activation properties, the enzymatic activity of the mutant was dramatically impaired. Analysis of the amidolytic activities suggested that the FXa variant hydrolyzes all three FXa-specific chromogenic substrates, S2765, S2222 and SpFXa with markedly slower catalytic efficiencies (Table 1). Thus, the FXa variant hydrolyzed all three chromogenic substrates with ~200–300-fold decreased specificity constants (k_cat/K_m). The catalytic defect in the substrate hydrolysis involved both kinetic parameters (k_cat and K_m), suggesting that deletion of Asp-185 adversely affects both the reactivity of the catalytic triad and the substrate binding pocket of the mutant protease. Noting that Asp-185 is located in a loop that influences the Na⁺ binding properties of FXa (19), we monitored the catalytic activity of FXa toward the chromogenic substrates in the Tris-HCl buffer containing increasing concentrations of NaCl as described in our previous studies (19,20). Interestingly, we discovered that the FXa variant cannot bind to Na⁺, thus its catalytic activity was insensitive to the presence of Na⁺ in the reaction buffer, explaining the dramatic catalytic defect observed in the activity of the mutant (Fig. 2).

Table 1.

Kinetic constants for the cleavage of chromogenic substrates and apparent dissociation constant (K_d(app)) for interaction with Na⁺.

	Km (μM)	kcat (s−1)	kcat/Km (μM−1 s−1)	Kd(app) Na+ (nM)
FXa-WT
S2765	53 ± 2.8	130 ± 3.8	2.4	80 ± 6.8
SpXa	93 ± 11	124 ± 4.2	1.3	-
S2222	237 ± 20	100 ± 4.6	0.42	-
FXa-D185del
S2765	3864 ± 220	26.8 ± 0.8	0.0069	ND
SpXa	506 ± 66	2.9 ± 0.1	0.0057	-
S2222	915 ± 56	1.3 ± 0.04	0.0014	-

The kinetic constants were determined from the steady-state kinetics of hydrolysis of chromogenic substrates (30–4000 μM for S2765, 15–2000 μM for both SpFXa and S2222) by each FXa derivative (1 nM for FXa-WT and 15 nM for FXa-D185del) in TBS/Ca²⁺. The apparent dissociation constant (K_d(app)) for the interaction of each FXa with Na⁺ was determined by the same chromogenic activity assay using S2765 (50 μM for FXa-WT and 1 mM for FX1-D189del in 5 mM Tris-HCl (pH 7.5) containing 0.1% PEG 8000, 5 mM CaCl₂ and increasing concentrations of NaCl. The K_d(app) for FXa-WT is derived from Fig. 2. Kinetic values are the average of at least 3 measurements ± SD. ND; not determinable since the amidolytic activity of FXa-D185del was not affected by the presence of Na⁺ in the buffer.

Figure 2.

The Na⁺ dependence of the amidolytic activity of FXa. The amidolytic activity of wild-type FXa (○) and FXa-D185del (●) was monitored in the presence of increasing concentrations of Na⁺ at room temperature using S2765 as described under “Materials and Methods”. The solid line for FXa is derived from nonlinear regression fits of the kinetic data to the Langmuir isotherm equation. The K_d(app) value for Na⁺ binding to FXa is presented in Table 1.

Prothrombin activation

The catalytic activity of the FXa variant toward prothrombin was evaluated in both the absence and presence of FVa on PC/PS vesicles. First, the affinity of FXa for interaction with FVa was evaluated by monitoring the activation of prothrombin by each FXa derivative as a function of increasing concentrations of FVa on PC/PS vesicles. The results presented in Figs. 3A and 3B suggest that the affinity of the FXa variant for interaction with FVa was decreased by nearly an order of magnitude. Thus, in contrast to a K_d(app) of 0.45 nM for FVa interaction with wild-type FXa, the corresponding value was increased to 6.1 nM for the FXa variant (Table 2). Next, the catalytic activity of the FXa variant toward prothrombin was evaluated in the absence and presence of a saturating concentration of FVa (20 nM) on PC/PS vesicles by the same assay. The results presented in Fig. 4A in the absence of FVa indicate that, similar to poor hydrolysis of small synthetic substrates (Table 1), both the K_m(app) and k_cat parameters for prothrombin activation by the FXa variant have been dramatically impaired. Thus, in contrast to a k_cat/K_m ratio of 0.8 × 10⁻³ nM⁻¹ min⁻¹ for activation of prothrombin by wild-type FXa, the corresponding value for the FXa variant was decreased ~53-fold (Table 2). Interestingly, the activation of prothrombin by the FXa-D185del variant was relatively efficient in the prothrombinase complex, suggesting that the cofactor function of FVa overcomes some of the catalytic defects of the FXa variant caused by the Asp-185 deletion (Fig. 4B). Thus, in contrast to a 53-fold impairment in k_cat/K_m of prothrombin activation by the FXa variant in the absence of FVa, the catalytic defect in prothrombin activation by the FXa variant in the presence of FVa was limited to ~3-fold reduction of the activation rate (k_cat), suggesting that FVa restores the catalytic defect of the mutation by improving both the rate and the binding of the substrate to the mutant protease. The mutant actually exhibited a slightly improved K_m(app) for interaction with prothrombin, thus the overall catalytic activity of the FXa variant in the prothrombinase complex was only impaired ~2.5-fold (Table 2). In contrast to FVa restoration of the catalytic defect of prothrombin activation, the cofactor did not influence the defective amidolytic activity of the mutant toward small synthetic substrates (data not shown).

Figure 3.

FVa concentration dependence of prothrombin activation by FXa derivatives. A, the activation of human prothrombin (1 μM) by FXa-WT (○) (50 pM) was monitored in the presence of different concentrations of human FVa (0–5 nM) on PC/PS vesicles (25 μM) in TBS/Ca²⁺. Following 0.5 min activation at room temperature, EDTA was added to a final concentration of 20 mM and the rate of thrombin generation was measured from the cleavage rate of S2238 as described under “Materials and Method”. B, the same as A except that the activation reaction of prothrombin by FXa-D185del (●) (50 pM) was monitored using the same experimental conditions for 2 min. Solid lines in both panels are nonlinear regression fits of kinetic data to the Michaelis-Menten equation. The kinetic values are presented in Table 2.

Table 2.

Kinetic constants for the activation of prothrombin by FXa derivatives in the absence and presence of FVa and the apparent dissociation constants (K_d(app)) for their interaction with the cofactor.

	Km(app) (nM)	kcat (min−1)	kcat/Km (nM−1 min−1)	FVa, Kd(app) (nM)
FXa-WT
Prothrombin, PC/PS, Ca2+	283 ± 33	0.24 ± 0.01	0.8× 10−3	-
Prothrombin, PC/PS, FVa, Ca2+	50.6 ± 4.7	1186 ± 25	23.4	0.45 ± 0.03
FXa-D185del
Prothrombin, PC/PS, Ca2+	2401 ± 166	0.036 ± 0.002	0.015× 10−3	-
Prothrombin, PC/PS, FVa, Ca2+	38 ± 3.0	348 ± 6.0	9.2	6.1 ± 0.7

The kinetic parameters K_m(app) and k_cat were determined from the concentration dependence of prothrombin activation by FXa derivatives on PC/PS vesicles in TBS/Ca²⁺ in the absence and presence of a saturating concentration of FVa as described under “Materials and Methods”. The K_d(app) values for the interaction of FXa with FVa were determined from the saturable cofactor concentration dependent thrombin generation by each FXa derivative at room temperature by the same assay. All values are the average of at least 3 measurements ± SD.

Figure 4.

Prothrombin activation by FXa derivatives in the absence and presence of FVa. A, in the absence of FVa, different concentrations of human prothrombin (0.02–2 μM) was incubated with 5 nM of either FXa-WT (○) or FXa-D185del (●) on PC/PS vesicles (25 μM) in TBS/Ca²⁺. Following 20–80 min activation at room temperature, EDTA was added to a final concentration of 20 mM and the rate of thrombin generation was measured from the cleavage rate of S2238 as described under “Materials and Methods”. B, the same as A except that the activation reactions by each FXa derivative (50 pM) were carried out in the presence of FVa (20 nM) for 0.5–2 min. Solid lines in both panels are nonlinear regression fits of kinetic data to the Michaelis-Menten equation. The kinetic values are presented in Table 2.

Inhibition by AT and TFPI

Similar to amidolytic activities and interaction with Na⁺, the FXa variant exhibited a dramatic impairment in reactions with both physiological plasma inhibitors of FXa, AT and TFPI. The second-order rate constant for the AT inhibition of the FXa variant was impaired by nearly three orders of magnitude in both the absence and presence of heparin cofactors (Table 3). Similarly, analysis of the TFPI concentration dependence of protease inhibition suggested that TFPI inhibits the FXa variant with an equilibrium inhibition constant (K_i) that is >400 fold weaker than that observed with wild-type FXa (Fig. 5A and Table 3).

Table 3.

Kinetic constants for the inhibition of FXa derivatives by antithrombin (AT), tissue factor pathway inhibitor (TFPI) and recombinant tick anticoagulant protein (rTAP).

	k2,uncat (AT) (M−1 s−1)	k2,H5 (AT) (M−1 s−1)	k2,Hep (AT) (M−1 s−1)	Ki (TFPI) (nM)	Ki (rTAP) (nM)
FXa-WT	(2.5±0.2)×103	(4.7±0.6)×105	(7.0±0.5)×107	0.26 ± 0.05	0.51 ± 0.03
FXa-D185del	2.3 ± 0.7	(7.2±0.4)×102	(6.1±0.3)×104	108 ± 16	160 ± 17

The second-order rate constants for the AT inhibition of FXa derivatives (1–5 nM) in the absence of cofactor (k_2,uncat) were determined from the residual activities of the proteases after their incubation at room temperature with AT (0.5–2.0 μM) for 10–240 min in TBS/Ca²⁺ as described under “Materials and Methods”. The k_2,H5 and k_2,Hep values were determined by the same procedures except that AT (5–500 nM) was in complex with 2 μM pentasaccharide fondaparinux (H₅) or 500 nM of unfractionated heparin (Hep). The K_i values for binding to TFPI and rTAP were determined by incubating each FXa derivative (0.1 nM) with either TFPI or rTAP (0.6–400 nM) or in the same buffer. Following 30 min of incubation at room temperature aliquots of the inhibition reaction was used in a prothrombinase assay to activate prothrombin. K_i values were estimated from the decrease in the rate of thrombin generation by nonlinear regression analysis of data using a quadratic equation describing the tight binding interactions as described in “Materials and methods”. All values are the average of at least 3 measurements ± SD.

Figure 5.

Inhibition of FXa derivatives by TFPI and rTAP. A, the inhibition of FXa-WT (0.1 nM) by either TFPI (○) or rTAP was monitored in TBS/Ca²⁺ at room temperature by a two-step assay. In the first step, the protease was incubated with increasing concentrations of the inhibitors for 30 min, a time sufficient to establish equilibrium. In the second stage, an aliquot of the protease-inhibitor reaction was used to activate prothrombin (1 μM) in a prothrombinase assay (20 nM FVa and 25 μM PC/PS) for 0.5–2 min. The K_i values were determined from the rate of thrombin generation using S2238 (100 μM) as described under “Materials and Methods”. B, the same as A except that the inhibition of FXa-D185del variant by either TFPI (○) or rTAP was monitored in TBS/Ca²⁺ under identical conditions. Solid lines in both panels are the best fit of kinetic data to a quadratic tight binding equation.

It is known that recombinant tick anticoagulant protein (rTAP) can effectively inhibit FXa (Fig. 5A). This inhibitor has been extensively used for probing determinants of the specificity of FXa interaction with macromolecular substrates and inhibitors (21,22). Similar to TFPI, a dramatic impairment in the K_i for the rTAP inhibition of the FXa variant was also observed (Fig. 5B and Table 3).

Discussion

In this study, we expressed the D185del variant of FX in mammalian cells and elucidated the molecular basis of the mild clotting defect in a bleeding patient with severe FX deficiency who carried this mutation. Our results suggest that the deletion of Asp-185 dramatically impairs the catalytic activity of FXa, thus the mutant protease cannot effectively interact with synthetic or natural macromolecular substrates and inhibitors. The effect of the mutation was particularly pronounced in the FXa recognition of its specific plasma inhibitors, AT and TFPI, as evidenced by the FXa mutant exhibiting more than 2–3 orders of magnitude lower reactivity with these plasma inhibitors. The mutant protease also interacted with its natural substrate prothrombin with more than a 50-fold lower catalytic efficiency in the absence of FVa. Interestingly, however, FVa overcame most of the catalytic defect of the mutation so that the FXa mutant activated prothrombin with only 2.5-fold lower catalytic efficiency when the protease was assembled into the prothrombinase complex. The FVa restoration of the catalytic defect, together with a lack of susceptibility of the FXa-D185del variant with both specific physiological inhibitors of FXa may explain the basis for the clinically mild bleeding phenotype of the patient who was diagnosed with severe FX deficiency based on in vitro coagulation assays (14).

It has been demonstrated that FXa requires Na⁺ for its catalytic activity (19,20). The monovalent cation is known to bind to an exposed surface loop (225-loop) in the vicinity of the catalytic pocket to modulate the activity of FXa by an allosteric mechanism (20,23). Asp-185 is located in a loop below the active-site pocket of FXa, known as 185–189-loop, which also harbors the S1 subsite residue, Asp-189 (24). This residue interacts with the P1-Arg residues of substrates and inhibitors reminiscent of the S1-P1 recognition mechanism that has been established for all other members of the family of trypsin-like serine proteases (15,24,25). Previous data have indicated that the 185–189-loop of FXa is energetically linked to the Na⁺ binding site of the protease (19). The results presented above suggest that FXa-D185del has lost its ability to interact with Na⁺. This was evidenced by the observation that the catalytic activity of the mutant was not sensitive to the presence of the monovalent cation in the reaction buffer. Thus, it appears that the functionally important allosteric linkage between the Na⁺-binding loop and the 185–189-loop has been lost in the FXa variant. These results are consistent with the hypothesis that the deletion of Asp-185 has decreased the flexibility of the 185–189-loop so that the loop is locked in a conformation that is not suitable for interaction with the metal ion. Thus, the conformational cofactor effect of Na⁺, which is required for the protease recognition of the substrate, is not relayed to the S1 site of the mutant protease in order to facilitate its interaction with the P1-Arg of the substrate. In this context, FVa may restore the catalytic defect of FXa-D185del through modulation of the 185–189-loop and/or the Na⁺-binding loop of the mutant protease. This hypothesis is consistent with our previous mutagenesis data (19).

It has been demonstrated that FVa binds to the 162-helix of FXa to modulate the catalytic function of the protease (26). Interestingly, thermodynamic linkage analysis has indicated that an allosteric coupling also exists between FVa, Na⁺ binding and S1 sites of FXa (27). The observation that FXa-D185del interacted with FVa with an order of magnitude weaker affinity supports this hypothesis and further suggests that a conformational change in the loop harboring Asp-185 is also part of the same allosteric network which links the protein and/or metal ion cofactor functions to the mechanism and specificity of substrate recognition by the protease in the prothrombinase complex. This hypothesis is consistent with previous molecular modeling data predicting that deletion of Asp-185 can alter the conformation of 185–189-loop of FXa (14). Interestingly, FVa effectively restored most of the catalytic defect of Asp-185 deletion toward the natural substrate, but not toward the small synthetic substrates, when the mutant FXa assembles into the prothrombinase complex. The mechanism by which FVa restores the catalytic defect of FXa-D185del toward prothrombin is not known. However, some insight may be gleaned from our previous observation that the catalytic function of FXa becomes independent of Na⁺ when the protease assembles into the prothrombinase complex (20). Nevertheless, complex formation with FVa did not influence the amidolytic activity of the FXa variant in the prothrombinase complex, suggesting that the assembly of the natural substrate, prothrombin, to the activation complex is required in order that FVa can render FXa to function independent of the Na⁺ ion in the prothrombinase complex. It is possible that the ternary FXa-FVa-prothrombin complex formation stabilizes the energetically linked 225-loop, 185–189-loop and the S1 site in the Na⁺-activated conformations, thereby effectively accommodating the substrate in the catalytic pocket of the mutant protease.

In summary, our results suggest that the catalytic function of FXa-D185del is dramatically impaired because the deletion of Asp-185 decreases the flexibility of 185–189-loop, thereby stabilizing the Na⁺-binding 225-loop in a conformation that does not support its interaction with the monovalent cation. The mutation causes a dramatic 2–3 orders of magnitude decreased reactivity for the FXa variant in interaction with the specific substrates and inhibitors of FXa present in plasma. However, complex formation with FVa restores the catalytic defect of the FXa mutant toward prothrombin in the prothrombinase complex. Thus, the impaired catalytic activity of the FXa mutant toward the natural substrate prothrombin is compensated by its dramatically impaired reactivity with the plasma inhibitors. The inability of the FXa variant to interact with TFPI may have important ramifications for the regulation of the proteolytic activity of the FVIIa-TF complex since prior complex formation of FXa with TFPI is thought to be required for the physiological inhibition of the extrinsic Tenase complex (8,9). Thus, the activity of FVIIa is expected to remain up-regulated during the initiation phase of the clotting cascade in the patient’s plasma. These unique features in FXa-D185del may explain the basis for the clinically mild bleeding phenotype of the patient carrying this mutation who has a severe FX deficiency as diagnosed by in vitro coagulation assays.

Highlights.

• Asp-185 deletion in FX predisposes FX deficient patient to mild bleeding phenotype.

• A recombinant form of FX Asp-185 deletion mutant is expressed in mammalian cells.

• The catalytic activity of the recombinant mutant protease is severely impaired.

• FVa restores the catalytic defect of the FXa mutant in the prothrombinase complex.

• The results provide the basis for mild bleeding tendency in FX deficient patient.