Factor VII Polymorphisms in Populations With Different Risks of Cardiovascular Disease
Abstract Increased plasma factor VII coagulant activity (FVII:C) has been associated with the risk of ischemic heart disease (IHD). Differences in plasma FVII:C among individuals are associated with three common polymorphisms in the FVII gene. Therefore, we investigated FVII polymorphisms in four populations that differ in their risk of developing cardiovascular disease, namely, Europeans, Greenland Inuit, Gujarati Indians, and Afrocaribbeans. We studied (1) the promoter polymorphism, which is the result of a decanucleotide insertion in the FVII promoter at position −323 from the start of translation; (2) the hypervariable region 4 polymorphism (HVR4), which is the result of a variable number of tandem repeats in intron 7; and (3) the RQ353 polymorphism, a guanine-to-adenine substitution in the position of the codon for amino acid 353 resulting in an amino acid replacement of arginine (R) by glutamine (Q) in the FVII protein. The frequencies of these three polymorphisms and their linkage disequilibrium were different in the four populations studied. The frequencies of the alleles associated with higher plasma FVII:C were lower in the Europeans than in the Inuit, a population with a lower incidence of IHD. There was an association between both the promoter polymorphism and the RQ353 polymorphism and the plasma FVII:C in the Europeans, the Inuit, and the Gujarati Indians, and an association only between the RQ353 polymorphism and plasma FVII:C in the Afrocaribbeans. Only in the Inuit was the HVR4 polymorphism associated with plasma FVII:C. In multiple regression analysis, the additional information provided by the promoter polymorphism when the other polymorphisms were already included in the model was the most pronounced, suggesting that the promoter polymorphism may be the functional mutation having the greatest effect on determining plasma FVII:C.
- Received August 13, 1996.
- Accepted January 28, 1997.
A large variation is seen in the risk for developing IHD among different populations, with the risk being the highest in Finland1 and the Indian subcontinent,2 3 low among the Greenland Inuit, Japanese, Chinese, and Afrocaribbeans,4 and intermediate in Denmark, Great Britain, and the United States.1 5 Both a difference in lifestyle and of genetic factors may contribute to this difference.
Several studies have suggested a difference of the genetic background of several cardiovascular risk indicators between populations with different IHD risks. For example, the frequencies of the apolipoprotein E alleles were different in the Greenland Inuit and the Japanese compared with Europeans in the Netherlands or Denmark.6 Another example is provided by polymorphisms of the α- and β-fibrinogen genes, the allele frequencies of which are different in the European and Inuit populations.7 These polymorphisms are associated with differences in plasma fibrinogen levels; in the Greenland Inuit, lower frequencies were observed for the allele that was associated with the higher plasma levels of the cardiovascular risk indicator fibrinogen.
The coagulant activity of blood coagulation factor VII (FVII:C) has been shown to be an independent risk indicator for fatal IHD. In the Northwick Park Heart Study an increase of the plasma FVII:C was associated with an increased risk for fatal acute myocardial infarctions during follow-up (mean, 16.1 years).8 The predictive value of plasma FVII:C was strongest during the first 5 years of follow-up.9 In the Prospective Cardiovascular Münster Study (PROCAM), higher plasma FVII:C was associated with the risk of cardiac death during the first 6 years of follow-up.10
About 30% of the variation in plasma FVII:C in Europeans can be explained by polymorphisms of the FVII locus.11 12 13 Several studies have reported a strong association between a common polymorphism in exon 8 of the FVII gene and FVII levels.11 14 15 16 This polymorphism is caused by a guanine-to-adenine substitution in the codon for amino acid 353 of the FVII gene, which results in a substitution of arginine (R) by glutamine (Q) in the FVII protein. Heterozygotes have approximately 37% lower plasma FVII:C levels than individuals homozygous for the common R allele while individuals homozygous for the rare Q allele had 67% lower levels.11
Two other genetic FVII polymorphisms have been described, a decanucleotide insertion/deletion polymorphism in the promoter region of the FVII gene17 and a variable number of tandem repeats (37 bp) in intron 7 (HVR4 polymorphism).18 19 Bernardi et al12 described linkage disequilibrium among the three polymorphisms in an Italian population and showed that the presence of the rare alleles of the three polymorphisms was associated with lower plasma FVII:C.
In the present study, we assessed the frequencies and linkage disequilibrium of these three FVII polymorphisms and their relation to plasma FVII:C in four different ethnic groups (Europeans, Greenland Inuit, Gujarati Indians, and Afrocaribbeans), which are known to differ in their incidence of IHD.
One hundred eighty-two healthy Europeans were recruited in Copenhagen, Denmark, and Leiden, the Netherlands. One hundred ninety-two Inuit, living in Nanortalik, in the southwest of Greenland, were invited to participate in this study.6 One hundred twenty-three Afrocaribbeans and 142 Gujarati Indians were recruited from a general practice in the Brent and Harrow areas of northwest London as part of a cardiovascular and glucose-tolerance survey.20 Subjects were assigned to any one ethnic group if at least three of their grandparents belonged to that group. Individuals of mixed descent were not included.
DNA samples were available from 182 Europeans, 133 Inuit, 113 Afrocaribbeans, and 130 Gujarati Indians. Plasma FVII:C measurements were available from 144 Europeans, 133 Inuit, 78 Afrocaribbeans, and 93 Gujarati Indians. The populations characteristics are given in Table 1⇓.
Factor VII Clotting Activity (FVII:C)
FVII:C was measured using a one-stage clotting assay as follows: 100 μL plasma was diluted 1:10 in Tris hydrochloric acid buffer (50 mmol/L Tris, 100 mmol/L sodium chloride, pH 7.4) and mixed with 100 μL FVII-deficient plasma (Biopool, Umeå, Sweden) for the Inuit and Europeans; for the Gujarati Indians and Afrocarribeans, the method described by Thompson et al21 was used. The mixture was incubated for 3 minutes at 37°C in a coagulometer (Type 410A 4B, Amelung). Then 200 μL of a prewarmed (37°C) 1:1 mixture of human brain thromboplastin and calcium chloride (25 mmol/L) were added, and the clotting time automatically recorded on the coagulometer. Human brain thromboplastin was prepared as described by Poller and Thomson22 with slight modifications. Values are expressed as a percentage of a pooled normal plasma.23
Each 50 μL PCR contained 100 to 400 ng genomic DNA, 100 ng of each appropriate primer, 10 mmol/L Tris/HCl (pH 9.0), 1.5 mmol/L magnesium chloride, 50 mmol/L potassium chloride, 0.01 (w/v) gelatin, 0.1% Triton X-100, 0.02 mmol/L of each nucleotide, 0.1 U Taq polymerase (HT Biotechnology LTD, Cambridge, England). The reaction components were incubated at 95°C for 5 minutes, followed by 30 cycles at 95°C for 1 minutes, 55°C for 1 minutes, and 72°C for 2 minutes in a DNA thermal cycler (Perkin Elmer Cetus). PCR amplification primers for the FVII promoter polymorphism were 5′-GGC CTGGTCTGGAGGCTCTCTTC-3′ and 5′-GAGCGGACG GTTTTGTTGCCAGCG-3′. For the HVR418 and RQ35311 polymorphisms primers were used as described previously ). Ten microliters of the PCR product were digested with the appropriate restriction enzyme (StyI for the promoter and MspI for the RQ353 polymorphism) under the conditions described by the manufacturer. These digestion products and the PCR product of the HVR4 polymorphism were separated using electrophoresis through a 2% or 4% agarose gel, respectively, in 44 mmol/L tris-borate and 1 mmol/L EDTA containing 0.5 μg/mL ethidium bromide and visualized under UV light. The alleles with the restriction site and the noncleavable alleles were designated P0 and P10, respectively, for the promoter polymorphism and R and Q, respectively, for the RQ353 polymorphism. The alleles of the HVR4 polymorphism contained between four and eight repeats and were designated H4 to H8. The location of the 10 base pair insertion of the promoter polymorphism was determined by sequence analysis of the PCR product and was the same for all populations.
Allele frequencies were determined by gene counting; 95% confidence intervals (95% CI) of the allele frequencies were calculated from sample allele frequencies. Genotype distribution deviations from those expected on the Hardy-Weinberg equilibrium were analyzed using a χ2-test. Standardized disequilibrium statistics were used as described by Hill et al.24 Three-locus linkage analysis was performed using the 3locus.pas program.25
Multiple regression analysis and partial F test were performed to assess the extent to which the promoter, the HVR4, and the RQ353 polymorphisms influence the plasma FVII:C in these population samples. In the first step of the regression analysis, two polymorphisms (using dummy variables) were forced into the model together with sex, age, BMI, and triglycerides, variables that are known to be related to plasma FVII:C. Total triglyceride levels were skewed and therefore logarithmically transformed. In the second step, the third polymorphism was entered and the significance of adding this third polymorphism to the model was tested using a partial F test. The statistical analysis was performed using the “SPSS” and “Lotus1-2-3” computer programs. P values below .05 were considered statistically significant.
FVII Polymorphisms in Different Populations
There was a clear difference in the allele frequencies of the three FVII polymorphisms among different populations (Tables 2⇓ and 3⇓). The rare allele (P10) of the promoter polymorphism was more common in the Inuit, the Afrocaribbeans, and the Gujarati Indians than in the Europeans. The frequencies of the H7 allele of the HVR4 polymorphism and the rare Q allele of the RQ353 polymorphisms were similar in the Afrocaribbeans and the Europeans and higher in the Inuit and Gujarati Indians. In the four populations, the genotype distribution was in Hardy-Weinberg equilibrium.
There was a difference not only in allele frequency but also in linkage disequilibrium of the three polymorphisms among the different ethnic groups (Table⇑ 4). In the Europeans, the Inuit, and the Gujarati Indians there was strong linkage disequilibrium between the promoter polymorphism and the RQ353 polymorphism, but in the Afrocaribbeans there was no linkage disequilibrium. A similar observation was made for the linkage disequilibrium between the promoter and the HVR4 polymorphism. The linkage disequilibrium between the HVR4 and RQ353 polymorphisms was similar in these four populations.
The linkage disequilibrium among the three loci was different among the four populations (Table⇑ 5).
Association Between FVII Polymorphisms and Plasma FVII:C
For the promoter polymorphism, the association between genotype and plasma FVII levels differed among these populations (Figure⇓ ). In the Europeans, Inuit, and Gujarati Indians, lower plasma FVII:C was seen in heterozygous individuals than in individuals homozygous for the common allele, and the lowest plasma FVII:C was seen in individuals homozygous for the rare allele. No association between the promoter genotype and plasma FVII:C was observed in the Afrocaribbeans. For the HVR4 polymorphism the plasma FVII:C was highest in the Inuit homozygous for the rare allele and intermediate in the heterozygotes, but no association was observed between HVR4 genotype and plasma FVII:C in the other populations. The association between plasma FVII:C and the RQ353 polymorphism was similar in all populations, with the highest levels seen in individuals homozygous for the common allele and the lowest levels seen in individuals homozygous for the rare allele.
The proportion of the variation in plasma FVII:C that could be explained by the three polymorphisms was very different in the four populations. In the Europeans, 20.5% of the total variation was explained by the three polymorphisms, in the Inuit 8.3%, in the Afrocaribbeans 26.8%, and in the Gujarati Indians 41.4%. Also, the relative contribution of the three polymorphisms to the variance in plasma FVII:C differed among the four populations (Table⇑ 6 ). The only significant increments in the relative contribution were observed in the Afrocaribbeans and the Gujarati Indians when the promoter polymorphism was added to the HVR4 and the RQ353 polymorphism in the multiple regression model.
It has been demonstrated that high plasma FVII:C levels are associated with an increased risk of cardiovascular disease,8 9 10 and polymorphisms in the FVII gene can partly explain interindividual variability of plasma FVII:C.11 14 15 16 We therefore explored the associations between these FVII gene polymorphisms in four different populations known to have a different prevalence of cardiovascular disease.
No association was observed between the cardiovascular risk of the different populations and the allele frequencies of the FVII polymorphisms; however, these polymorphisms are associated with plasma FVII:C. Europeans, who have an intermediate risk of developing IHD, have the lowest frequencies of the rare allele of the FVII promoter polymorphism compared with the low-IHD-risk Inuit and Afrocaribbeans, and the high-IHD-risk Gujarati Indians. For the other polymorphisms, there is no direct relation between the frequency of the alleles and the incidence of IHD in the four populations.
For the promoter and the RQ353 polymorphisms there was a clear association between the genotype and the plasma FVII:C in each population. However, there were large differences in the percentage of variation that could be explained by the polymorphisms, with the variation being only 8% in the Inuit and as much as 41% in the Gujarati Indians. Thus, in the population with the highest IHD risk (Gujarati Indians), the highest genetic contribution to plasma FVII:C is observed, but the significance of this observation is unclear.
The populations studied had similar BMIs, sex distribution, and triglyceride levels but the Afrocaribbeans and the Gujarati Indians were older. The relation between these polymorphisms and FVII:C may depend on age. Furthermore, differences in the FVII:C assays may also affect the associations between the FVII polymorphisms and FVII:C. A group of elderly white subjects (mean age, 61 years) was compared with Gujarati Indians and Afrocaribbeans using the same FVII:C assay,21 results of which were similar to our findings in white subjects (unpublisheddata, 1997).
The difference in plasma FVII:C variation caused by adding a polymorphism to a multiple regression model with two otherpolymorphisms can be explained by the difference in linkage disequilibrium among the polymorphisms in the four populations studied. In the Inuit, the three polymorphisms showed strong linkage disequilibrium and adding a polymorphism to the multiple regression model provided no extra information. On the other hand, the Afrocaribbeans showed no linkage disequilibrium between the promoter and the other polymorphisms, and the addition of the promoter polymorphism increased the percentage of explained plasma FVII:C by 19.0%. These observations suggest that the FVII promoter polymorphism is the strongest determining polymorphism for the plasma FVII levels. Comparing different populations thus may provide a method to identify the functional polymorphism. This contribution of the promoter polymorphism to the FVII:C variance is supported by data from Pollak et al26 who found the promoter strength in transient expression assays of FVII promoter constructs containing the decanucleotide insertion to be lower than in those lacking the insertion. Humphries et al27 also reported that the plasma FVII:C was associated somewhat more strongly with the promoter polymorphism than with the RQ353 polymorphism in healthy middle-aged white men.
In the Afrocaribbeans there was no association between the promoter polymorphism and plasma FVII:C, while adding the promoter polymorphism to the multiple regression analysis explained an additional 19% of the FVII:C variation. This effect may be due to epistasis, that is, the interaction among polymorphisms which causes one polymorphism to interfere with the effects of the other. This will have to be studied in a larger population.
In conclusion, there are large differences in the genetic make-up of FVII in populations with different IHD risks. However, the genotype distribution or the relation between FVII polymorphisms and plasma FVII:C cannot be directly correlated with the IHD risk of the population. Studying the relation between FVII polymorphisms and plasma FVII:C in different populations suggests that the promoter polymorphism may be the functional mutation.
Selected Abbreviations and Acronyms
|BMI||=||body mass index|
|FVII:C||=||factor VII coagulant activity|
|IHD||=||ischemic heart disease|
|HVR4||=||hypervariable region 4|
|PCR||=||polymerase chain reaction|
We would like to thank Lars Johansen for collecting the Inuitdata, Peter Marckmann for collecting part of the Danish population data, J. Kennedy Cruickshank for collecting the data from the Gujarati Indians and the Afrocaribbeans. We also wish to thank Nico Lakenberg for technical assistance, Le Ahn Luong for localization of the 10 base pair insertion of the promoter polymorphism, and Prof Steve Humphries and Else M. Bladbjerg for valuable discussions.
Keys A, ed. Coronary heart disease in seven countries: American Heart Association Monograph 29. Circulation. 1970;41-42(suppl 1):I-1-I-211.
Balarajan R, Balusu L, Adelstien AM, Shukula V. Patterns of mortality among migrants to England and Wales from the Indian subcontinent. BMJ. 1984;289:1185-1187.
McKeigue PM, Marmot MG. Mortality from coronary heart disease in Asian communities in London. BMJ. 1988;297:903.
Marmot MG, Adelstein AM, Bulusu L. Immigrant mortality in England and Wales 1970-1978, OPCS Studies of Medical and Population Subjects No. 47, London, England:HMSO; 1984.
Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, Arveiler D, Rajakangas A, Pajk A. Myocardial infarction and coronary deaths in the World Health Organization MONICA project. Circulation. 1994;90:583-612.
de Knijff P, Johansen LG, Rosseneu M, Frants RR, Jespersen J, Havekes LM. Lipoprotein profile of a Greenland Inuit population. Arterioscler Thromb. 1992;12:1371-1379.
de Maat MPM, de Knijff P, Green FR, Thomas AE, Jespersen J, Kluft C. Gender-related association between β-fibrinogen genotype and plasma fibrinogen levels and linkage disequilibrium at the fibrinogen locus in Greenland Inuit. Arterioscler Thromb Vasc Biol. 1995;15:856-860.
Ruddock V, Meade TW. Factor VII activity and ischaemic heart disease: fatal and non-fatal events. Q J Med. 1994;87:403-406.
Heinrich J, Balleisen L, Schulte H, Assmann G, van de Loo J. Fibrinogen and factor VII in the prediction of coronary risk. Arterioscler Thromb. 1994;14:54-59.
Green F, Kelleher C, Wilkes H, Temple A, Mease T, Humphries S. A common genetic polymorphism associates with lower coagulation factor VII levels in healthy individuals. Arterioscler Thromb. 1991;11:540-546.
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.
Saha N, Liu Y, Heng CK, Hong S, Low PS, Tay JSH. Association of factor VII genotype with plasma factor VII activity and antigen levels in healthy Indian adults and interaction with triglycerides. Arterioscler Thromb. 1994;14:1923-1927.
Marchetti G, Gemmati D, Patracchini P, Pinotti M, Bernardi F. PCR detection of a repeat polymorphism within the F7 gene. Nucleic Acids Res. 1991;19:4570.
de Knijff P, Green F, Johansen LG, Grootendorst D, Temple A, Cruickshank JK, Humphries SE, Jespersen J, Kluft C. New alleles in F7 VNTR. Hum Mol Genet. 1994;3:384.
Poller L, Thomson J. Laboratory methods. In: Hemker HC, Loeliger EA, Veltkamp JJ, eds. Human Blood Coagulation. Leiden, the Netherlands: University Press; 1969;chap 3:290-295.
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.