The Arg353Gln Polymorphism Reduces the Level of Coagulation Factor VII
In Vivo and in Vitro Studies
Abstract Factor VII levels are regulated by environmental and genetic factors. Two polymorphisms, a G-to-A transversion at nucleotide 10976 resulting in Arg353Gln and a decanucleotide insert at position -323 in the 5′-flanking region of the factor VII gene, have been associated with a 20% to 25% reduction in plasma factor VII levels. However Arg353Gln almost always segregates on alleles containing the insert in UK and Italian populations, thereby making it impossible to independently evaluate the impact of Arg353Gln on factor VII levels in these ethnic groups. We have evaluated the influence of genotype on factor VII levels in 99 healthy Polish blood donors and observed that Arg353Gln frequently occurs in the absence of the insert. In univariate analysis, the mean levels of factor VII coagulant activity (VII:C) and factor VII antigen (VII:Ag) were significantly lower in 16 people who were heterozygous for Arg353Gln and the insert compared with 72 normal subjects who had neither Arg353Gln nor the insert (88.8% of normal and 83.1% versus 102% and 100%, P=.019 and P=.0003, respectively). In nine subjects heterozygous for Arg353Gln alone, VII:C and VII:Ag were significantly decreased compared with the normal subjects (81.9% and 83%, respectively, P=.007 and P=.004). In multivariate analysis, Arg353Gln but not the insert significantly reduced VII:C and VII:Ag after adjustment for age and plasma triglycerides (P <.05 and P=.02, respectively). To evaluate the mechanism responsible for reduced factor VII levels in individuals with Arg353Gln, we performed transient transfection assays with factor VII cDNA containing the base substitution resulting in Gln353 and wild-type factor VII cDNA in COS-1 cells. The levels of VII:Ag in the cell lysates were similar, but the amino acid substitution significantly reduced factor VII secretion into the media to 74.9% of wild-type (P=.0001). Based on these in vivo and in vitro studies, we conclude that the Arg353Gln polymorphism alone can decrease plasma factor VII levels.
Address reprint requests to Kenneth A. Bauer, MD, Department of Veterans Affairs Medical Center, 1400 VFW Parkway, West Roxbury, MA 02132.
- Received April 11, 1997.
- Accepted August 26, 1997.
Factor VII is a vitamin K-dependent zymogen composed of 406 amino acids.1 It is synthesized by the liver and secreted into the blood where it circulates at a concentration of 0.5 μg/mL.2 On endothelial injury, factor VII plays a critical role in the initiation of blood coagulation by binding to exposed tissue factor. This interaction enables it to be rapidly activated to factor VIIa, and the factor VIIa-tissue factor complex then initiates blood coagulation by activating factor IX as well as factor X.3,4
The Northwick Park Heart Study found that elevated plasma VII:C is an independent risk factor for a first myocardial infarction in middle-aged men. This relationship was stronger for fatal than nonfatal coronary events.5,6 The Prospective Cardiovascular Münster Study found a trend toward higher VII:C only in subjects with fatal events, but this did not reach statistical significance.7 It has been suggested that increased sensitivity of the Northwick Park VII:C assay to the level of circulating factor VIIa underlies this difference.8
Plasma VII:C levels in normal individuals are influenced by a number of environmental and genetic factors. The former include dietary fat intake, age, obesity, diabetes, and in women, the onset of menopause and estrogen administration.9 Green et al10 first reported that an important genetic determinant of plasma VII:C is the Arg353Gln polymorphism in factor VII, which results from a single base change (G to A) at position 10976 in the factor VII gene. This transversion occurs with an allele frequency of 0.1 in normal individuals from the United Kingdom and the United States,10,11 and heterozygosity is associated with a reduction in VII:C and VII:Ag levels to ∼75% of normal. Homozygotes with Arg353Gln have levels that are ∼50% of normal. It was subsequently found that individuals from the United Kingdom with the Arg353Gln polymorphism almost always carry another polymorphism, the insertion of a decanucleotide at nucleotide −323 in the 5′-flanking region of the factor VII gene.12 However, subjects were identified who were heterozygous for the decanucleotide insert without Arg353Gln. This genotype was associated with a decrease in VII:C that was similar to Arg353Gln, thereby raising doubt about the role of the latter polymorphism in reducing factor VII:C levels.12
In this report, we investigated a cohort of healthy Polish blood donors in whom the Arg353Gln polymorphism is frequently not in allelic association with the decanucleotide insert. This has enabled us to demonstrate that Arg353Gln is independently associated with reduced factor VII levels. We also performed transient transfection assays with wild-type or Gln353 factor VII cDNA in vitro, which demonstrated that the presence of this polymorphism reduced factor VII secretion compared with the wild type.
Subjects and Laboratory Measurements
The individuals in this study were recruited from a random population of blood donors at the Institute of Hematology and Blood Transfusion in Warsaw, Poland. Fasting blood samples were drawn into plastic tubes containing 1/10 vol of 0.129 mol/L buffered trisodium citrate. Plasma was obtained by centrifugation at 2500g for 15 minutes at 4°C, transferred into plastic tubes, and stored along with the cellular elements at −80°C until use. Plasma VII:C was determined by a one-stage clotting assay using rabbit thromboplastin (ISI=1.78, Organon Teknika). VII:Ag was measured in plasma or lysates and media of transfected cells with an enzyme-linked immunoabsorbent assay (American Bioproducts Co). VII:C and VII:Ag were expressed as a percentage of normal pooled plasma that was constructed by mixing equal volumes of plasma from >30 healthy control subjects. Triglyceride levels were measured by enzymatic assay in a Beckman Synchron CX7/CX4CE chemistry analyzer (Beckman Instruments).
DNA Isolation, In Vitro Amplification Using PCR, and Detection of Polymorphic Alleles
DNA was purified from leukocyte nuclei obtained from whole blood by standard techniques.13 The conditions used to amplify the coding sequence and 5′-flanking region of the factor VII gene by PCR have been described previously.14,15 To identify the Arg353Gln polymorphism, a 239-bp fragment encoding the second portion of exon 8 was amplified using the primers 5′-GCAGCAAGGACTCCTGCAAG-3′ and 5′-CCACAGGCCAGGGCTGCTGG -3′ and digested with 10 U of Msp I (New England Biolabs). The presence of the normal allele results in two fragments of 53 and 186 bp, while the G-to-A substitution at position 10976 abolishes the restriction site. To detect the decanucleotide insert, a 431-bp fragment of the 5′-flanking region of the factor VII gene was amplified using the primers 5′-CCACACCCAAGCTTCGGTCTTGAG-3′ and 5′-GGCCTGGGGGATCCTGATGAA-3′ and digested with 10 U of EcoRI (New England Biolabs). The presence of the normal allele or the allele with the decanucleotide insert results in products of 328 and 103 bp or 328 and 113 bp, respectively. Restriction fragments were subjected to electrophoresis in nondenaturing 8% (w/v) polyacrylamide gels. After completion of the electrophoretic procedure, gels were stained in 0.5 μg/mL ethidium bromide for 5 minutes and photographed under ultraviolet transillumination.
Construction of Expression Vectors
The plasmid pEDFVIIwt containing a wild-type human factor VII cDNA with only 21 bp of the 3′-untranslated region1 cloned into the dicistronic mRNA mammalian expression vector pED-mtxr 16 has been described previously.17 To prepare a factor VII cDNA-containing Gln,353 exon 8 was amplified by PCR from genomic DNA of an individual with this polymorphism and cloned into PT7 to obtain PT7exon8Gln353. A 544-bp fragment containing the last 523 bp of exon 8 was obtained by BsiWI/EcoRI digestion of PT7exon8Gln353. This fragment was then substituted into pEDFVIIwt to obtain pEDFVIIGln353.
Cell Culture and Transient Transfection Assays
The pEDFVIIwt and pEDFVIIGln353 plasmids were transiently expressed in COS-1 cells (ATCC CRL 1650) or CHO cells (CHO-DUKX-B11 provided by Dr Barbara Furie, Boston, Mass). The COS-1 cells were grown in Dulbecco’s modified Eagle medium supplemented with 10% (v/v) fetal bovine serum, 4 μg/mL of vitamin K, 2 mmol/L l-glutamine, 10 mmol/L HEPES (pH 7.2), 100 U/mL of penicillin G, and 100 μg/mL streptomycin. The CHO cells were grown in α-modified essential medium supplemented with 10% (v/v) fetal bovine serum, 4 μg/mL vitamin K, 2 mmol/L l-glutamine, 10 mmol/L HEPES (pH 7.2), 10 μg/mL adenosine, 10 μg/mL deoxyadenosine, 10 μg/mL thymidine, 100 U/mL penicillin G, and 100 μg/mL streptomycin. The expression vectors were transfected into COS-1 or CHO cells with lipofectamine (GIBCO-BRL), according to the instructions of the manufacturer. Twenty hours before transfection, the cells were plated on 60-mm culture dishes at a density of 2×106 cells/plate. pEDFVIIwt or pEDFVIIGln353 (5 μg) was transfected into cells in media without serum. After 12 hours, the transfection medium was replaced by regular medium with 10% fetal bovine serum. After another 36 hours, the culture media were collected for assay, the cells were washed with phosphate-buffered saline, and cell extracts were prepared in reporter lysis buffer (Promega). The culture media and cell lysates were assayed for VII:Ag.
Levels of VII:C, VII:Ag, and triglycerides were compared in individuals with different genotypes by one-way ANOVA. For multivariate analysis, VII:C, VII:Ag, and triglyceride levels were log-transformed to reduce skewness and kurtosis. Multiple linear regression was used to assess the importance of Arg353Gln and the decanucleotide insert on log VII:C and log VII:Ag. By including an Arg353Gln × insert interaction term, we were able to perform standard two-way ANOVA. The multiple linear regression approach also allows us to adjust for other factors, eg, age, sex, log triglycerides, and log triglycerides × Arg353Gln interaction. We used P<.05 to indicate statistical significance, but retained variables with P<.10 in our model as they may confound the true effect of the factors of interest. The r2 statistic was used to estimate the proportion of variation in log VII:C and log VII:Ag explained by the model.
Informed consent was obtained from screened individuals. The study was approved by the Institute of Hematology and Blood Transfusion, Warsaw, Poland, and by the Human Studies Committee of the Brockton-West Roxbury VA Medical Center.
Studies of UK and Italian populations demonstrate that the decanucleotide insert at nucleotide −323 in the 5′-flanking region of the factor VII gene is nearly always in allelic association with the sequence alteration resulting in Arg353Gln. We sought to determine whether the Polish population differs with respect to this association by determining the prevalence of the two polymorphisms in a healthy blood donor population. The population included 83 men and 16 women with a mean age of 36.7 years (range, 19 to 69). The presence of the Arg353Gln polymorphism was demonstrated by subjecting PCR products encoding exon 8 of the factor VII gene to Msp I digestion. To evaluate for the presence of the decanucleotide insert, the 5′-flanking region of the factor VII gene was amplified by PCR and subjected to EcoRI digestion and gel electrophoresis (Fig. 1⇓). One woman was homozygous for the insert and heterozygous for Arg353Gln. Of the remaining 98 blood donors, 1 individual was heterozygous for the decanucleotide insert, 9 were heterozygous for Arg353Gln, and 16 were heterozygous for both polymorphisms (Table 1⇓). This calculates to an allele frequency of 9.6% for the decanucleotide insert and 13.1% for Arg353Gln.
We next determined the levels of VII:C, VII:Ag, and triglycerides in the 99 Polish individuals. The mean levels of VII:C, VII:Ag, and triglycerides were 97.7% of normal (±23.1), 96.4% (±16.9), and 1.17 mmol/L (±0.91), respectively. The correlation coefficient between the measurements of VII:C and VII:Ag was 0.815 (r2=0.665, P=.0001). The levels of VII:C and VII:Ag demonstrated a small increase with advancing age (r2=0.052, P=.023 and r2=0.068, P=.009, respectively). No significant differences in VII:C or VII:Ag levels were observed between men and women. Triglycerides had a significant effect on VII:C (r2=0.16, P=.0001) and to a lesser extent on VII:Ag (r2=0.066, P=.010).
The levels of VII:C, VII:Ag, and triglycerides as a function of genotype are shown in Table 1⇑, in which individuals are subdivided on the basis of Arg353Gln and the decanucleotide insert. Table 2⇓ shows the data for individuals with Arg353Gln or the insert without respect to the presence of the other polymorphism. The one woman who was homozygous for the insert and heterozygous for Arg353Gln was excluded from analysis. She had the lowest factor VII levels in the cohort (VII:C=54% of normal, VII:Ag=51%).
As 9 of 25 individuals had Arg353Gln without the decanucleotide insert, we were able to analyze the effect of the Arg353Gln polymorphism alone on factor VII levels. In these 9 subjects, the mean levels of VII:C and VII:Ag were significantly decreased compared with the normal subjects (81.9% and 83%, respectively, P=.007 and P=.004) by one-way ANOVA. The mean levels of VII:C and VII:Ag were also significantly lower in the 16 persons heterozygous for Arg353Gln and the insert compared with 72 normal subjects who had neither Arg353Gln nor the insert (88.8% and 83.1% versus 102% and 100%, P=.019 and P=.0003, respectively).
To assess the independent effect of the Arg353Gln polymorphism on factor VII levels, we performed a multivariate analysis on the data. For VII:C only, there was a significant interaction between the presence of Arg353Gln and triglycerides (P<.05). The presence of the insert and sex of the subject had no significant effect on VII:C or VII:Ag in this population. Heterozygosity for Arg353Gln, triglycerides, and subject age independently influenced VII:C and VII:Ag (P<.05 and P=.02, P=.05, and P=.14, and P=.04 and P=.01, respectively).
To determine whether the G-to-A transversion at position 10976 resulting in Gln353 decreases factor VII biosynthesis in vitro, we performed transient transfection assays in COS-1 cells with expression vectors containing factor VII cDNA with either the wild-type or Gln353 sequence (Fig. 2⇓). The amino acid substitution reduced secretion of VII:Ag into the media to 74.9±4.5% of wild-type (P=.0001, n=5 independent transfections), but VII:Ag levels in cell lysates were similar to those in cells transfected with the wild type plasmid (95.4±23% of wild-type, P=.3). Transient transfection assays in CHO cells also demonstrated a significant reduction in factor VII secretion for the Gln353 construct compared with that for the wild-type construct (54.6±7.0% of wild-type, P=.001, n=5 independent transfections, data not shown).
The association between the presence of the Arg353Gln polymorphism in factor VII and reduced plasma factor VII levels was initially reported in a UK cohort by Green et al.10 Similar observations were subsequently made in a US population,11 Japanese population,18 Gujarati Indians, and Afrocaribbeans residing in Britain.19
The presence of another polymorphism in the 5′-flanking region of the factor VII gene, the insertion of a decanucleotide at nucleotide −323, was first identified by Marchetti et al.20 In 88 residents from a rural Italian village, this polymorphism was found to be in strong allelic association with Arg353Gln.21 A recent investigation of 705 healthy men from the UK also showed that the decanucleotide insert frequently cosegregated on alleles with the codon for Arg353Gln.12 The mean levels of VII:C and VII:Ag in individuals heterozygous for both polymorphisms were reduced by ≈20% compared with subjects without either polymorphism. Twenty individuals (≈3% of the population) were heterozygous for the decanucleotide alone, and their mean levels of VII:C and VII:Ag were similar to those of individuals with both polymorphisms. Only two persons were heterozygous for Arg353Gln alone, a number insufficient to determine whether such individuals have lower plasma factor VII levels. Statistical analysis of the entire data set demonstrated that the factor VII-lowering effect of the decanucleotide insert was significant after adjustment for Gln,353 but the effect of Gln353 was no longer significant after adjustment for the insert.12
The above study therefore raised the possibility that the Gln353 allele was not actually responsible for lowering plasma factor VII levels, but was merely a marker for the presence of the decanucleotide insert. Moreover, Pollak et al22 had shown that the decanucleotide insert results in a 33% reduction in factor VII promoter activity compared with a wild-type construct in reporter gene assays in HepG2 cells, thereby providing mechanistic data for its effect on factor VII biosynthesis.
There are no previous data regarding factor VII levels in populations with the Arg353Gln polymorphism in the absence of the decanucleotide insert or studies of the effect of Arg353Gln on factor VII biosynthesis in vitro. We previously analyzed the genetic basis of factor VII deficiency in a cohort of Polish patients.14 In addition to a missense mutation resulting in Ala294Val and a single base deletion resulting in a frameshift mutation in the codon for amino acid 404, they were homozygous for Arg353Gln and did not have the decanucleotide insert (unpublished results). This suggested that the decanucleotide insert and Arg353Gln might not be in close allelic association in the normal Polish population.
Among 99 Polish blood donors, we found that Arg353Gln occurred in the absence of the insert in approximately one third of them. The allele frequency of Arg353Gln in this population was 13.1%, which was slightly higher than in the United Kingdon. In univariate analysis, the mean levels of VII:C and VII:Ag were significantly lower in people heterozygous for Arg353Gln in the absence of the insert compared with normal subjects without either polymorphism. In patients with similar factor VII genotypes, the mean level of VII:C was equivalent to VII:Ag, which indicates that Arg353Gln does not alter the specific activity of factor VII. This is consistent with data that this amino acid is located on the surface of the factor VII molecule, distant from its active enzymatic site.23
Several studies have found a significant positive correlation between plasma triglycerides and factor VII levels.24–27 However, this relationship appears to be confined to subjects carrying the Arg353 allele.28,29 We therefore used a multiple linear regression model in which factor VII levels were adjusted for these effects as well as for age and sex. Analysis of our data demonstrated that the effect of Arg353Gln in lowering VII:C and VII:Ag remained significant after adjusting for these variables. We did not find the decanucleotide insert to be associated with lower factor VII levels in the Polish population, but it should be noted that there was only one individual who had the insert without Arg353Gln.
To evaluate the mechanism responsible for reduced factor VII levels in association with Arg353Gln, we carried out transient transfection assays with a factor VII cDNA containing the base substitution resulting in Gln353 and compared the resulting levels of factor VII in culture media and cell lysates with those obtained from transient transfections with wild-type factor VII cDNA. In COS-1 cells, the Gln353 substitution significantly decreased VII:Ag in the media without significantly altering factor VII levels in cell lysates. The Gln353 substitution in factor VII changes an amino acid with a positively charged side group to one that is uncharged. This likely alters the conformation of the protein to diminish its secretion from the liver.
Based on our in vivo and in vitro studies, we conclude that the Arg353Gln polymorphism in the absence of the decanucleotide insert decreases plasma factor VII levels. The presence of the two polymorphisms on the same allele, however, does not appear to have a greater effect in reducing plasma factor VII levels than either one individually. The reason for this is unknown, but one possibility is that a reduction in factor VII promoter activity due to the decanucleotide insert masks the post-translational defect in factor VII biosynthesis that results from Gln.353
Selected Abbreviations and Acronyms
|CHO||=||Chinese hamster ovary|
|PCR||=||polymerase chain reaction|
|VII:C||=||factor VII coagulant activity|
|VII:Ag||=||factor VII antigen|
These studies were supported by the Medical Research Service of the Department of Veterans Affairs. Dr Hunault is supported by a grant from the Fondation de France-Fédération Nationale des Centers de Lutte contre le Cancer. We gratefully acknowledge the statistical assistance of Dr Robert Parker, Beth Israel Deaconess Medical Center, Boston, Mass.
Hagen FS, Gray CL, O’Hara P, Grant FJ, Saari GC, Woodbury RG, Hart CE, Insley M, Kisiel W, Kurachi K, Davie EW. Characterization of a cDNA coding for human factor VII. Proc Natl Acad Sci U S A. 1986;83:2412–2416.
Fair DS. Quantitation of factor VII in the plasma of normal and warfarin-treated individuals by radioimmunoassay. Blood. 1983;62:784–791.
Osterud B, Rapaport SI. Activation of factor IX by the reaction product of tissue factor and factor VII: additional pathway for initiating blood coagulation. Proc Natl Acad Sci U S A.. 1977;74:5260–5264.
Bauer KA, Kass BL, ten Cate H, Hawiger JJ, Rosenberg RD. Factor IX is activated in vivo by the tissue factor mechanism. Blood. 1990;76:731–736.
Heinrich J, Balleisen L, Schulte H, Assman G, van de Loo J. Fibrinogen and factor VII in the prediction of coronary risk: results from the PROCAM study in healthy men. Arterioscler Thromb. 1994;14:54–59.
Miller G, Stirling Y, Esnouf M, Heinrich J, van de Loo J, Kienast J, Wu K, Morrissey J, Meade T, Martin J, Imeson J, Cooper J, Finch A. Factor VII-deficient substrate plasmas depleted of protein C raise the sensitivity of the factor VII bio-assay to activated factor VII: an international study. Thromb Haemostas. 1994;71:38–48.
Meade TW. The Epidemiology of Haemostatic and Other Variables in Coronary Artery Disease. Leuven, Belgium: Leuven University Press; 1987:37–60.
Green F, Kelleher C, Wilkes H, Temple A, Meade T, Humphries S. A common genetic polymorphism associated with lower coagulation factor VII levels in healthy inidviduals. Arterioscler Thromb. 1991;11:540–546.
Kunkel LM, Smith KD, Boyer SH, Borgaonkar DS, Wachtel SS, Miller OJ, Breg WR, Jonew HW, Rary JM. Analysis of human Y chromosome-specific reiterated DNA in chromosome variants. Proc Natl Acad Sci U S A.. 1977;74:1245–1249.
Arbini AA, Bodkin D, Lopaciuk S, Bauer KA. Molecular analysis of Polish patients with factor VII deficiency. Blood. 1994;84:2214–2220.
Arbini AA, Pollak ES, Bayleran JK, High KA, Bauer KA. Severe factor VII deficiency due to a mutation disrupting a hepatocyte nuclear factor 4 binding site in the factor VII promoter. Blood. 1997;89:176–182.
Kaufman RJ, Davies MV, Wasley LC, Michnick D. Improved vectors for stable expression of foreign genes in mammalian cells by use of the untranslated leader sequence from EMC virus. Nucleic Acids Res. 1991;19:4485–4490.
Arbini AA, Mannucci PM, Bauer KA. A Thr359 Met mutation in factor VII of a patient with a hereditary deficiency causes defective secretion of the molecule. Blood. 1996;87:5085–5094.
Bernardi F, Marchetti G, Pinotti M, Arcieri P, Baroncini C, Papacchini M, Zepponi E, Ursicino N, Chiarotti F, Mariani G. Factor VII gene polymorphisms contribute about one third of the factor VII level variation in plasma. Arterioscler Thromb Vasc Biol. 1996;16:72–76.
Pollak ES, Hung HL, Godin W, Overton GC, High KA. Functional characterization of the human factor VII 5′-flanking region. J Biol Chem. 1996;271:1738–1747.
Hoffman CJ, Miller RH, Hultin MB. Correlation of factor VII activity and antigen with cholesterol and triglycerides in healthy young adults. Arterioscler Thromb. 1992;12:267–270.
Hoffman CJ, Lawson WE, Miller RH, Hultin MB. Correlation of vitamin K-dependent clotting factors with cholesterol and triglycerides in healthy young adults. Arterioscler Thromb. 1994;14:1737–1740.
Humphries SE, Lane A, Green FR, Cooper J, Miller GJ. Factor VII coagulant activity and antigen levels in healthy men are determined by interaction between factor VII genotype and plasma triglyceride concentration. Arterioscler Thromb. 1994;14:193–198.