|
|||||||||
|
|
|
|
|
|
|
|
|
|
|
Prepublished online as a Blood First Edition Paper on June 14, 2002; DOI 10.1182/blood-2002-03-0909.
HEMOSTASIS, THROMBOSIS, AND VASCULAR BIOLOGY
From the Department of Haematology, Division of
Investigative Science, Hammersmith Campus, Imperial College of Science,
Technology, and Medicine, London, United Kingdom; and Service
d'Hématologie Biologique, Hôtel-Dieu, Paris,
France.
We have identified 2 PROS1 missense mutations in the
exon that encodes the vitamin K-dependent Gla domain of protein S
(Gly11Asp and Thr37Met) in kindred with phenotypic protein S deficiency and thrombosis. In studies using recombinant proteins,
substitution of Gly11Asp did not affect production of protein S but
resulted in 15.2-fold reduced protein S activity in a factor Va
inactivation assay. Substitution of Thr37Met reduced expression by
33.2% (P < .001) and activity by 3.6-fold. The Gly11Asp
variant had 5.4-fold reduced affinity for anionic phospholipid vesicles
(P < .0001) and decreased affinity for an antibody
specific for the Ca2+-dependent conformation of the protein
S Gla domain (HPS21). Examination of a molecular model suggested that
this could be due to repositioning of Gla29. In contrast, the Thr37Met
variant had only a modest 1.5-fold (P < .001), reduced
affinities for phospholipid and HPS21. This mutation seems to disrupt
the aromatic stack region. The proposita was a compound heterozygote
with free protein S antigen levels just below the lower limit
of the normal range, and this is now attributed to the partial
expression defect of the Thr37Met mutation. The activity levels were
strongly reduced to 15% of normal, probably reflecting the functional
deficit of both protein S variants. Her son (who was heterozygous only
for Thr37Met) had borderline levels of protein S antigen and activity,
reflecting the partial secretion and functional defect associated with
this mutation. This first characterization of natural protein S
Gla-domain variants highlights the importance of the high affinity
protein S-phospholipid interaction for its anticoagulant role.
(Blood. 2002;100:2812-2819) Protein S is a plasma glycoprotein that plays an
important role in the protein C anticoagulant pathway by acting as a
cofactor to activated protein C (APC) in the specific proteolytic
inactivation of factors Va and VIIIa, reviewed in Simmonds and
Lane.1 In healthy individuals, ~60% of
circulating protein S is found in complex with C4b binding protein, and
only free protein S has APC cofactor activity. The physiological
requirement for protein S is clearly demonstrated by the clinical
manifestations of purpura fulminans in infants who lack detectable
protein S at birth.2 Heterozygous deficiency of protein S
is found in 1% to 2% of consecutive patients with deep vein
thrombosis3 and has been classified into 3 subtypes. Type
I (total and free protein S below the lower limit of a normal range)
and type III (normal total protein S, but reduced free protein S
levels) deficiencies are now commonly considered to be
quantitative.4 Type II deficiency is qualitative and is
characterized by reduced activity in a specific functional assay.
Mature protein S has a modular structure and consists of a vitamin
K-dependent Gla domain that includes an aromatic stack region
(residues 1 to 46), a region sensitive to cleavage by thrombin (residues 47 to 75), 4 domains homologous to epidermal growth factor
(EGF-like domains, residues 76 to 242), and a region homologous to sex
hormone binding globulin (residues 243 to 635). Inherited quantitative
(type I and III) protein S deficiency is frequently caused by mutations
within the gene for protein S, PROS1.4,5 However, qualitative (type II) protein S deficiency is rarely reported
and only a handful of PROS1 defects have been
described.4 Only one of these, Lys9Glu, is located in the
Gla domain of protein S, but its function has not been
investigated.6 Here, we describe a kindred presenting with
type II protein S deficiency caused by 2 mutations. Both mutations were
present on separate alleles, in the exon that encodes the Gla domain of
protein S, and experiments with recombinant proteins showed that each
mutation resulted in a variant protein that was functionally defective
due to aberrant Ca2+-induced phospholipid binding.
Subjects, phenotypic, and clinical data
Identification and allelic distribution of
PROS1 mutations
Vector construction, mutagenesis, and transient expression
experiments
Stable expression of protein S Wild-type and variant protein S were stably expressed in Human Embryo Kidney (HEK) 293 cells (European Collection of Cell Cultures, Salisbury, United Kingdom). Complete growth medium for these cells was Eagle minimum essential medium (EMEM, Sigma, Poole, United Kingdom), supplemented with 1x MEM nonessential amino acid solution (Sigma), 2 mM L-glutamine (Sigma), 20 µg/mL vitamin K1 (Konakion, Roche, Lewes, United Kingdom), 10% (vol/vol) fetal calf serum (Imperial Laboratories, Andover, United Kingdom), 100 U/mL penicillin, and 100 µg/mL streptomycin sulfate (Sigma).Stable transfection was performed by standard techniques using 2 µg
DNA of each protein S construct and 10 µL Lipofectin (Invitrogen) reagent, according to the manufacturer's protocol. At 72 hours after transfection, the cells were transferred into a 100-mm
dish and maintained in complete growth medium containing 5 µg/mL
blasticidin (Invitrogen). Blasticidin resistant cells were grown to
confluence. Protein production for functional studies was performed in
175-cm2 flasks. When the cells were confluent, complete
medium was discarded, cells were washed once with phosphate-buffered
saline (PBS), and replaced with Opti-MEM I with Glutamax
(Invitrogen) supplemented with 20 µg/mL vitamin K1, 100 U/mL penicillin, 100 µg/mL streptomycin sulfate, and 5 µg/mL
blasticidin. After 5 days, cell culture supernatants were harvested,
centrifuged at 2000 rpm for 10 minutes, and filter sterilized. The
levels of protein S present in conditioned medium were assessed by
in-house ELISA and varied between 7.8 and 16.3 µg/mL. Aliquots of
conditioned medium were subsequently dialyzed against 40 mM Tris-HCl
(pH 7.4), 140 mM NaCl, and 3 mM CaCl2, and the ELISA
repeated to confirm protein S levels. Dialyzed medium was stored in
aliquots at Proteins and phospholipids Purified human plasma proteins (protein S, APC, factor Xa, and prothrombin) were obtained from Enzyme Research Laboratories (South Bend, IN). Factor Va was obtained from Haematologic Technologies (Essex Junction, VT). Phospholipids were obtained from Avanti Polar Lipids (Alabaster, AL) and were provided as premixed dioleoyl-phosphatidylethanolamine/dioleoyl-phosphatidylcholine/dioleoyl-phosphatidylserine (PE/PC/PS, 40:40:20) and PC/PS (80:20) in chloroform. The chloroform was evaporated under nitrogen vapor, and the lipids were resuspended in ice-cold sterile water then mixed vigorously for 1 hour with shaking at 4°C. The resuspended vesicles were sonicated at 20 kHz for 5 minutes and stored for short periods at 4°C.Western blotting Western blotting was performed as described previously.7 Briefly, 30 ng protein S was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) in a 4% to 20% polyacrylamide Tris-HCl gel (Bio-Rad, Hercules, CA). Proteins transferred to Hybond-P (Amersham Pharmacia Biotech, Little Chalfont, United Kingdom) were detected using an anti-human protein S peroxidase-conjugated antibody (Dako, Glostrup, Denmark) and enhanced chemiluminescence (ECL) detection reagents (Amersham Pharmacia Biotech).Factor Va inactivation assay Dialyzed recombinant protein S was diluted in buffer A (40 mM Tris-HCl (pH 7.4), 140 mM NaCl, 0.2% [wt/vol] NaN3, 3 mM CaCl2, 0.3% [wt/vol] bovine serum albumin [BSA]) to provide a range of final recombinant protein S concentrations (0 to 20 nM) and was incubated with 0.1 nM APC, 3 nM factor Va, and 25 µM phospholipid vesicles for 2 minutes at 37°C in buffer A. Following incubation, 2 µL of this reaction was transferred to a separate tube containing 72 pM factor Xa, 0.5 µM prothrombin, and 25 µM phospholipid vesicles in buffer A at 37°C. After 3 minutes' incubation, the reaction was stopped with 3 µL of 0.5 M EDTA. The entire reaction was then mixed with 100 µL 50 mM Tris-HCl (pH 8.3), 130 mM NaCl, 0.075 KIU/mL Trasylol in a microtiter plate and incubated at 37°C for 10 minutes. The chromogenic substrate S2238 (Chromogenix Instrumentation Laboratories, Milano, Italy) was then added to a final concentration of 0.444 mM and incubated at 37°C for 5 minutes. The reaction was stopped with 10% acetic acid and the absorbance was read at 405 nm. All experiments were performed in triplicate.Phospholipid binding assays The binding of protein S to anionic phospholipid vesicles was assessed using a microtiter-plate assay, adapted from van Wijnen et al.8 The washing steps were performed using buffer A. Phospholipid vesicles were coated onto the wells of microtiter plates at 25 µg/mL in 50 mM Na2CO3 (pH 9.6), overnight at 4°C. The plates were washed 3 times and blocked for 1 hour at room temperature with 315 µL buffer A supplemented with BSA to a final concentration of 3% (wt/vol). The plates were again washed 3 times, and protein S at a range of concentrations (0 to 26 nM) in buffer A was added and incubated for 2 hours at 37°C. After the plates were washed 3 times, 100 µL of rabbit anti-human protein S peroxidase-conjugated antibody diluted 1:2500 in buffer A was added to the wells and incubated for 1 hour at 37°C. Plates were washed 3 times and 100 µL peroxidase substrate, O-phenylenediamine (OPD; Dako) at 666 µg/mL in 33.3 mM citric acid, 66.7 mM Na2HPO4 (pH5.0), and 0.0002% H2O2 was added to each well and incubated at room temperature, in the dark, for 10 minutes. The reaction was stopped by addition of 75 µL 1.4 M H2SO4 and the absorbance at 492 nm was assessed. The binding of wild-type protein S to the phospholipid vesicles was also assessed in buffer A, where the CaCl2 was replaced with 5 mM EDTA. A further test of specificity was performed by adding increasing amounts of purified prothrombin, added as a competitor in the presence of 2.7 nM protein S. Minimal influence on protein S binding was observed up to equimolar amounts of prothrombin and protein S, after which prothrombin increasingly inhibited protein S binding to the phospholipid (50% at a 100-fold molar excess), as expected. All experiments were performed in triplicate. For each set of data, the Kd app and capacityapp were calculated using Enzfitter 2.0 software (Biosoft, Cambridge, United Kingdom). Saturation curves were fitted using a one-site ligand-binding model. The binding of 10 nM protein S to phospholipid vesicles at various concentrations of Ca2+ ions was also investigated. Here, several variations on buffer A were prepared containing either 0, 1, 2, 3, or 5 mM CaCl2, and the experiment was carried out as before, except that all reagents and washing steps for a particular well were performed using buffer A with the appropriate CaCl2 concentration.Antibody binding assays The binding of recombinant protein S to different monoclonal antibodies (kind gifts from Prof Björn Dahlbäck, University of Lund, Malmö, Sweden) was assessed. HPS21 is specific for the Ca2+-dependent conformation of protein S Gla domain, and HPS54 is specific for the Ca2+-dependent conformation of the first EGF-like domain of protein S. Each of these antibodies was coated onto the wells of microtiter plates at 10 µg/mL in 50 mM Na2CO3 (pH 9.6), overnight at 4°C and blocked with casein. Dialyzed protein S at a range of concentrations (0 to 20 nM) in 50 mM Tris-HCl (pH 7.5), 150 nM NaCl, 0.1% (wt/vol) NaN3, 3 mM CaCl2, 1% (wt/vol) casein, 0.05% (vol/vol) Tween 20 were added to the wells and incubated for 1 hour at 37°C. Plates were washed 4 times with 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 3 mM CaCl2, and 0.1% (vol/vol) Tween 20 and bound protein S was detected with a rabbit anti-human protein S peroxidase-conjugated antibody, as before.
The coding regions and intron/exon boundaries of PROS1 were amplified and sequenced in the symptomatic proposita (Figure 1). This revealed the presence of 2 heterozygous gene defects, both in exon 2 of PROS1 that encodes the Gla domain of the mature protein (data not shown). The mutations were 301 G to A and 380 C to T (numbering according to Schmidel et al5), and these were predicted to result in the amino acid substitutions Gly11Asp and Thr37Met. Using a PCR cloning strategy, the mutations were found to reside on different alleles. No other sequence abnormality was found. Analysis of exon 2 of PROS1 in a son of the proposita showed that he was heterozygous for the Thr37Met mutation but was normal at Gly11 (Figure 1 and data not shown). Transient transfection experiments in COS-1 cells were performed in
order to assess whether either of the identified mutations influenced
the ability of the cells to produce and secrete protein S (Figure
2). Protein S bearing the Gly11Asp
mutation was secreted at an equivalent level to the wild type
(98.4% ± 9.9% compared to 100% ± 7.1%; P = .62).
In contrast, the expression level of protein S with Thr37Met was
reduced by ~33% (66.9% ± 14.0% compared to 100% ± 7.1%;
P < .0001). This indicated that the latter mutation was
associated with a production defect, providing an explanation for the
borderline levels of free protein S antigen observed in the son of the
proposita. Recent findings from our laboratory indicate that the
production defect is likely to be due to impaired secretion,7 although it cannot be ruled out that the
reduced levels observed were due to defective transcription and/or
translation.
In order to investigate whether these mutations in PROS1
caused any functional deficit in protein S, wild-type and variant protein S was stably expressed in HEK293 cells. Each recombinant human
protein S and commercially available protein S purified from human
plasma appeared as doublets following reducing SDS-PAGE and Western
blot (Figure 3), as previously
described.9 There was an additional minor band for
Gly11Asp at ~46 kDa, the precise nature of which is not known. As it
was estimated to be less than 5% of the total protein S for this
sample, it was not considered to be problematic for functional
analysis.
A factor Va inactivation assay was employed to assess the ability of
the variants to act as a cofactor to APC (Figure
4). Under these conditions and short
incubation periods, 0.1 nM APC did not inactivate any factor Va in the
absence of protein S (Figure 4, compare 0 nM protein S for wild type
and variants with the open triangle [no APC]). The inclusion of
wild-type recombinant protein S to a final concentration of ~1.5 nM
was sufficient to promote the inactivation of 50% of the factor Va
present (Figure 4,
Because both the mutations identified caused substitutions within the
Gla domain of protein S, we postulated that the loss of function of the
variants was due to impaired Ca2+-induced binding to
phospholipid. In order to validate this, a microtiter plate-based
assay was used to assess saturation binding to PC/PS/PE (40:40:20)
vesicles (Figure 5A; Table
2). In the presence of EDTA,
wild-type protein S did not bind to the vesicles (Figure 5A,
The reduced phospholipid binding could have been caused by altered
affinity of the Gla domain for Ca2+. Therefore, the ability
of 10 nM protein S to bind phospholipid vesicles at various
concentrations of Ca2+ was assessed in the same
microtiter-plate assay (Figure 5B). No significant binding of proteins
to phospholipid was observed when they were diluted in a buffer that
did not contain Ca2+ ions (Figure 5B; 0 mM
CaCl2). For wild-type protein S, the presence of even low
concentrations of Ca2+ ions was sufficient to cause a large
increase in protein S binding (Figure 5B; A potential additional explanation for reduced phospholipid binding was
due to alteration of the Ca2+-dependent conformation of the
Gla domain. We therefore assessed the ability of wild-type and variant
protein S to bind to a Ca2+-dependent monoclonal antibody
that was directed toward the Gla domain of protein S (HPS21, Figure
6A). Both of the variants showed reduced
binding to this antibody. The Thr37Met variant had only a modestly
reduced affinity for HPS21 compared to the wild type, whereas that of
the Gly11Asp variant was more than 3-fold reduced. It is interesting to
note that both variants still bound detectably to this conformation-
and Ca2+-dependent antibody. As expected, there was no
alteration in the affinity of either variant for a
Ca2+-dependent monoclonal antibody directed toward the
first EGF-like domain (HPS54, Figure 6B), which indicated that this
adjacent domain was correctly folded.
Two missense defects (Gly11Asp and Thr37Met) have been identified within the exon of PROS1 that encodes the Gla domain in a proposita with phenotypic protein S deficiency and recurrent thrombosis. The effect of these mutations on expression and function of protein S have been investigated using recombinant techniques. These studies showed that substitution of Gly11Asp did not affect expression of protein S but did result in a molecule that was estimated to be ~15-fold less active than wild-type protein S in a factor Va inactivation assay. Substitution of Thr37Met reduced production by ~33%, and the variant protein that was exported successfully had ~3.5-fold reduced activity. This was not due to gross misfolding of the protein S variants, as the variants bound with similar affinity to a Ca2+-dependent monoclonal antibody directed toward the first EGF-like domain (known to be essential for protein S function; Dahlback et al10 and Figure 6B). The location of both of the amino acid substitutions suggested a
mechanism by which the activity was reduced, as the protein S
amino-terminal Gla domain is highly homologous to the corresponding regions of other vitamin K-dependent coagulation proteins, reviewed in
Zwaal et al,11 Nelsestuen et al,12 and
Stenflo and Dahlback.13 The primary essential function of
this domain is to provide the interaction site for anionic phospholipid
surfaces, such as those exposed when platelets or endothelial cells are
activated and/or damaged. Each Gla domain contains between 9 and 12 residues of Although these conformational changes are certain to be required for
phospholipid binding, the precise mechanism by which the binding occurs
is still controversial. Two different hypotheses are currently
proposed. The first hypothesis concerns a conserved patch of
solvent-exposed hydrophobic residues, at positions Phe4, Leu5, and Val8
(within the We hypothesized that the reduced activity of both identified protein S variants could be due to abnormal Ca2+-induced folding of the Gla domain with consequent defective phospholipid binding. This is supported by our results, which showed a variation in the severity of the latter between the 2 variants. Both variants demonstrated reduced ability to bind phospholipid vesicles containing PC/PS/PE (Figure 5A) in the presence of Ca2+, reflected in differences in the Kd app (Table 2). The Gly11Asp variant had a Kd app that was increased more than 5-fold compared to the wild type, but this was only increased 1.5-fold for Thr37Met (Table 2). This reflected the greater functional deficit of the Gly11Asp variant in the presence of PC/PS/PE vesicles (Figure 4) compared to Thr37Met. In addition to reduced affinity, both variants had an apparent decrease in the number of binding sites available to them on the surface of these vesicles, as the A492nm at saturation was more than 2-fold reduced for both variants, indicating that less protein S was bound (Table 2). This was not due to specific loss of PE-dependent sites on the vesicles, as both variants also showed reduced binding to vesicles that contained only PC and PS (data not shown). A reduction in the number of binding sites for protein S in the factor Va inactivation assay probably contributes to the functional deficit of these variants. In this study, using a microtiter plate-based assay, the apparent Kd app of wild-type recombinant protein S for PC/PS/PE-containing vesicles was ~4 nM. This agrees well with previous estimates of phospholipid affinity that varied between 4 and ~75 nM, using PC/PS-30-33 or PC/PS/PE-8 containing vesicles or endothelial cells34 and protein S from different sources. Our preliminary data suggest that the presence of PE in the vesicles increases the affinity of protein S, and to our knowledge this has not been addressed directly before. However, protein S has been shown previously to enhance the binding of protein C to phospholipids in a PE-dependent manner.35 The ability of the variants to bind HPS21, a monoclonal antibody specific for the Ca2+-dependent conformation of the protein S Gla domain,10 was tested to further investigate the mechanism for reduced phospholipid binding. Both variants were found to have reduced affinity (Figure 6A) and, again, the effect was greater for Gly11Asp compared to Thr37Met. However, as an interaction with this antibody (and indeed phospholipid) was detectable under our assay conditions, this suggested that the variants were still able to undergo some form of the Ca2+ transition. It seems that in both cases, the Ca2+-dependent conformation of the Gla domain is disturbed or less stable, resulting in reduced affinity with, and availability of, phospholipid binding sites. Although the defective binding to phospholipid of both variants could not be corrected readily by increasing the concentration of Ca2+ ions (Figure 5B), it remains possible that impaired affinity for Ca2+ ions also contributes to the folding defects. It is interesting that the Thr37Met variant bound to a consistently greater degree compared to Gly11Asp in the Ca2+ titration curves for phospholipid binding (Figure 5B). This corresponds well to its consistently greater activity in the factor Va inactivation assay (Figure 4) and affinity for phospholipid (Figure 5A) and HPS21 (Figure 6A). Direct Ca2+-binding studies would be required to confirm any reduction in affinities for these variants. The probable structural consequences of each amino acid substitution are considered below. In the absence of any 3D structures for protein S, a theoretical model
of the structure of the amino terminal domains of protein S (Gla
domain, thrombin sensitive region, and first EGF-like domain) has
previously been constructed.36 In this model, the
consequence of the Thr37Met substitution has already been examined.
Thr37 was noted to be strictly conserved between Gla domains and to be
buried in the protein S model. Therefore, when substituted for Met, the
much larger side chain could not be accommodated and steric clashes
were predicted with the backbone atoms of Arg28.36 This
would be expected to lead to the disruption of the Gly11Asp, at the carboxy-terminus of the
This study highlights the importance of the high affinity interaction between protein S and anionic phospholipids for its anticoagulant functions both in vitro and in vivo. Our findings with the recombinant proteins provide an explanation for the phenotypic and clinical data obtained from the family. The proposita was a compound heterozygote and had free protein S levels just below the lower limit of the normal range, and this is now attributed to heterozygosity for the Thr37Met mutation. The activity levels were strongly reduced, and this can now be attributed to the functional deficit characterized in both populations of protein S. The proposita has suffered early onset thrombosis associated with pregnancy, as well as recurrences. No other genetic defect could be detected, and it therefore seems that the compound heterozygous state for protein S presents sufficient risk to result in thrombosis when interacting with the acquired risk presented by pregnancy. In contrast, her clinically unaffected son (who was heterozygous only for the Thr37Met mutation) had borderline levels of protein S antigen and low/borderline levels of protein S activity, reflecting the partial secretion and functional defect associated with this mutation.
The authors gratefully acknowledge the assistance of Dr Geoff Kemball-Cook for the preparation of Figure 7. We also thank Prof Bjorn Dahlback for kindly providing monoclonal antibodies to protein S, and Dr Helen Philippou for advice regarding stable expression of protein S.
Submitted March 22, 2002; accepted May 31, 2002.
Prepublished online as Blood First Edition Paper, June 14, 2002; DOI 10.1182/blood-2002-03-0909.
Supported by a grant from the Brazilian government (Agencia CAPES) and grants from the British Heart Foundation.
The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked "advertisement" in accordance with 18 U.S.C. section 1734.
Reprints: Rachel E. Simmonds, Haematology Department, Division of Investigative Science, Faculty of Medicine, Imperial College of Science, Technology, and Medicine, Hammersmith Hospital, Du Cane Road, London W12 0NN, United Kingdom; e-mail: r.simmonds{at}ic.ac.uk.
1. Simmonds RE, Lane DA. Regulation of coagulation. In: Loscalzo J,Schafer AI, eds. Thrombosis and Hemorrhage. 2nd ed. Baltimore, MD: Williams and Wilkins; 1998:45-76. 2. Dahlback B. The protein C anticoagulant system: inherited defects as basis for venous thrombosis. Thromb Res. 1995;77:1-43[CrossRef][Medline] [Order article via Infotrieve].
3.
Koster T, Rosendaal FR, Briet E, et al.
Protein C deficiency in a controlled series of unselected outpatients: an infrequent but clear risk factor for venous thrombosis (Leiden Thrombophilia Study).
Blood.
1995;85:2756-2761 4. Gandrille S, Borgel D, Sala N, et al. Protein S deficiency: a database of mutations - summary of the first update: for the Plasma Coagulation Inhibitors Subcommittee of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Thromb Haemost. 2000;84:918[Medline] [Order article via Infotrieve]. 5. Schmidel DK, Tatro AV, Phelps LG, Tomczak JA, Long GL. Organization of the human protein S genes. Biochemistry. 1990;29:7845-7852[CrossRef][Medline] [Order article via Infotrieve].
6.
Simmonds RE, Ireland H, Kunz G, Lane DA.
Identification of 19 protein S gene mutations in patients with phenotypic protein S deficiency and thrombosis: Protein S Study Group.
Blood.
1996;88:4195-4204 7. Rezende SM, Simmonds RE, Zoller B, et al. Genetic and phenotypic variability between families with hereditary protein S deficiency. Thromb Haemost. 2002;87:258-265[Medline] [Order article via Infotrieve]. 8. van Wijnen M, Stam JG, Chang GTG, et al. Characterization of mini-protein S, a recombinant variant of protein S that lacks the sex hormone binding globulinlike domain. Biochem J. 1998;330:389-396. 9. Malm J, Cohen E, Dackowski W, Dahlback B, Wydro R. Expression of completely gamma-carboxylated and beta-hydroxylated recombinant human vitamin-K-dependent protein S with full biological activity. Eur J Biochem. 1990;187:737-743[Medline] [Order article via Infotrieve].
10.
Dahlback B, Hildebrand B, Malm J.
Characterization of functionally important domains in human vitamin K-dependent protein S using monoclonal antibodies.
J Biol Chem.
1990;265:8127-8135 11. Zwaal RFA, Comfurius P, Bevers EM. Lipid-protein interactions in blood coagulation. Biochem Biophys Acta. 1998;1376:433-453[Medline] [Order article via Infotrieve]. 12. Nelsestuen GL, Shah AM, Harvey SB. Vitamin K-dependent proteins. Vitam horm 2000;58:355-389[CrossRef][Medline] [Order article via Infotrieve]. 13. Stenflo J, Dahlback B. Vitamin K-dependent proteins. In: Stamatoyannopoulos G,Nienhuis AW,Mejerus PW,Varmus H, eds. The molecular basis of blood diseases. 2nd ed. Philadelphia, PA: WB Saunders; 1994:565-597.
14.
Furie B, Bouchard BA, Furie BC.
Vitamin K-dependent biosynthesis of gamma-carboxyglutamic acid.
Blood.
1999;93:1798-1808 15. Soriano-Garcia M, Padmanabhan K, deVos AM, Tulinsky A. The calcium ion and membrane binding structure of the gla domain of Ca-prothrombin fragment 1. Biochemistry. 1992;31:2554-2566 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
3 and G at
position +4, relative to the ATG initiation codon). Mutagenesis
was performed using the Quickchange Site Directed Mutagenesis Kit
(Stratagene, La Jolla, CA), using primers France 1A/1B and
France 2A/2B to construct Gly11Asp and Thr37Met, respectively (Table
). Extensive testing showed that purification of
protein S from the supernatants was not required for this assay to
remain specific for protein S function. Commercially available human plasma-purified protein S had comparable activity to the wild-type preparation, requiring 1.0 nM protein S for 50% inactivation (Figure 
). Furthermore, dialyzed conditioned medium from cells
transfected with the vector pcDNA6, diluted to the same extent as 20 nM
wild-type protein S, had no activity in this assay (Figure 
).
Under the conditions employed, factor Va was stable over the course of
the assay, and protein S did not have any APC-independent effects (Figure 
). When the variants were tested, 15.2-fold more of the
Gly11Asp variant was required to achieve a 50% reduction in factor Va
activity compared to wild-type protein S. This indicated that this
mutation resulted in a protein that acted as a poor cofactor to APC. In
the case of the Thr37Met variant, 3.6-fold more was required for a 50%
reduction in factor Va activity compared to the wild type, indicating a
milder, but still significant, impairment of function.
,
protein S with Gly11Asp; 
, protein S with Thr37Met; 
-carboxyglutamic acid (Gla), which are formed
by posttranslational 
-loop).
-helical structure of residues 34-45. Our results suggest that such a disruption impacts on secretion of the variant protein (Figure 