Journal Publication

The Relationship of Three Genetic Traits to Venous Thrombosis in an African-American Population

by Anne Dilley, Harland Austin, W.Craig Hooper, Cathy Lally, Maria J. Ribeiro, Nanette Kass Wenger, Victor Silva, Peggy Rawlins, and Bruce Evatt

	Abstract
	Background
	Materials and Methods
	Results
	Discussion
	References

Abstract

A mutation in the Factor V gene (Factor V Leiden), a variant in the 5-10 methylene tetrahydrofolate reductase (MTHFR) enzyme, and an insertion/deletion polymorphism of the angiotensin I converting enzyme (ACE) gene may be related to abnormal blood clotting. We examined the associations between these genetic traits and venous thrombosis (VT) among African-Americans. This study comprised 93 patients with VT and 185 control subjects attending clinics at an urban, public hospital. Subjects' DNA was extracted from blood and assayed for these genetic traits. Odds ratios (ORs) were obtained from logistic regression and used as a measure of association between each genetic trait and VT. Factor V Leiden was unrelated to VT, but the mutation was too rare among our African-American subjects to evaluate adequately its relation to venous thrombosis. The homozygous and heterozygous genotypes for the V allele of the MTHFR gene was unrelated to VT (OR = 0.9, 95% CI: 0.5-1.8). Subjects with the deletion/deletion ACE polymorphism experienced a moderate increase in VT risk compared with persons with the other genotypes (OR = 1.5, 95% CI: 0.9-2.6). However, women with this ACE genotype experienced no increased risk (OR = 0.9, 95% CI: 0.5-1.9) whereas men with this genotype had nearly three times the risk (OR = 2.8, 95% CI: 1.2,6.2; P value for interaction = 0.06). These data indicate that the prevalence of Factor V Leiden and the V allele of the MTHFR gene is low among African-Americans. The D allele of the ACE gene is equally prevalent among African-Americans and Whites and may be related to venous thrombosis among African-American men.

Background

Complications of deep vein thrombosis (DVT), which include pulmonary embolism (PE) and chronic venous insufficiency, are associated with significant mortality and morbidity (1). Past studies have reported that while approximately 30 percent of DVTs are due to deficiencies in the anticoagulant proteins (protein S, protein C, and antithrombin III), the causes for the remaining cases are largely unknown (2).

Recent studies have identified three genetic traits which may directly or indirectly affect blood coagulability. Based on earlier work by Mariar et al. in which it was determined that activated protein C (APC) controlled clot formation by the proteolytic inactivation of the coagulation factors Va and VIIIa (3), Dahlback et al. described an impaired APC response, termed APC resistance (APC-R), in a patient with a history of unexplained thrombosis (4). In 1994, Bertina identified the molecular basis of APC-R as a defect in factor V (Factor V Leiden) involving the mutation of Arg506 to Gln506 (5). APC-R has been implicated as a cause of venous thrombosis (VT) in several epidemiologic studies (6-8).

Hyperhomocysteinemia is a consequence of either inherited or acquired alterations in the transsulfuration or remethylation pathway. A recently described polymorphism in the 5-10 methylene tetrahydrofolate reductase (MTHFR) gene which encodes for a key enzyme involved in remethylation is associated with reduced MTHFR activity and elevated levels of plasma homocysteine (9). The polymorphism is due to a C to T substitution at nucleotide 677 which converts alanine to a valine (9). Elevated levels of plasma homocysteine have been associated with an increased risk of VT (10-12).

In humans, an insertion/deletion polymorphism of the angiotensin I converting enzyme (ACE) gene has been identified in intron 16. Of the three genotypes I/I, I/D, and D/D, the D/D genotype has been associated with the highest plasma ACE activity, the I/D genotype with intermediate ACE activity and the I/I genotype with the lowest ACE activity (13). ACE converts angiotensin I to the potent vasoconstrictor, angiotensin II, and inactivates bradykinin, a vasodilator (14). Thus, one mechanism by which elevated ACE activity may cause chronic vascular disease is by inducing chronic vasoconstriction (15,16). High ACE levels are also thought to interfere with the fibrinolytic system by increasing levels of plasminogen-activator inhibitor -1 (17). Although the D/D genotype has been related to cardiovascular disease in several epidemiologic studies (18,19), the relation of the ACE polymorphism and VT has not been evaluated.

Almost all of the epidemiologic evidence evaluating the association between these genes and venous and arterial thrombosis pertains to White populations. The purpose of the present study was to determine if these genetic characteristics are associated with VT among African-Americans.

Materials and Methods

All patients with a VT attending an anticoagulant clinic at a large, urban public hospital in Atlanta, Georgia were eligible for inclusion as cases in the study. We approached 137 of these patients and 136 (99 percent) agreed to participate. Two patients were excluded as cases because their medical records did not support a VT diagnosis. Twenty seven patients were excluded because they had a history of heart disease, stroke, or other arterial thromboses. The present analysis is restricted to 93 African-American cases (14 White cases were excluded). Of these 93 cases, 51 had a DVT only, 34 had a DVT with a PE, and two had an inferior vena cava thrombus. Eighty-one of the cases were confirmed either by a venogram, ultrasound, an angiogram, or by a ventilation-perfusion scan. The two cases with an inferior vena cava thrombus were confirmed by computerized tomography. For 10 cases, we could not find radiologic confirmation of the diagnosis in the medical record and therefore relied on the clinical diagnosis.

We selected controls from among out-patients attending a clinical laboratory for routine blood tests and frequency matched them to cases on age (within 10 years), sex, and race. We excluded as controls persons with a history of heart attack, stroke, or blood clots. Of 225 eligible control subjects asked to participate, 200 (89 percent) agreed. The controls in the present analysis are restricted to 185 African-Americans (15 White controls were excluded).

Participation in the study entailed granting permission to review medical records, an in-person interview, and the collection of 20 milliliters of blood. The questionnaire elicited information on basic demographics, lifestyle habits, a personal history of thrombosis and other medical problems, and a family history of blood clots, stroke, or heart attack.

Laboratory Methods

The presence of Factor V Leiden, zygosity for the insertion/deletion polymorphism of the ACE gene, and zygosity for the alanine/valine polymorphism of the MTHFR gene were determined from DNA extracted from a blood sample. Blood samples were collected in 0.109 M sodium citrate. DNA was extracted from 3 ml of whole blood using the Gentra DNA Extraction kit (Minneapolis, MN) per the manufacturer's instructions and stored at -20 C. Polymerase chain reaction (PCR) was used to amplify DNA fragments in the factor V (20), MTHFR (9), and ACE (18) genes. Restriction enzyme analysis for Factor V Leiden and MTHFR was carried out using Mn1 and Hinf1, respectively. The digested products were then run on a ethidium bromide stained 3% metaphor gel and the results were determined from the restriction enzyme digestion pattern. A subset of samples was selected at random and confirmed by direct nucleotide sequencing. The insertion/deletion polymorphisms were determined by sizing the amplified product in a ethidium bromide stained 1.5 percent agarose gel. In order to ensure stringency for the DD genotype, DMSO was included in the PCR reaction mixture. Furthermore, all DD results were confirmed with the use of a third primer set that discriminated between ID and DD. PCR results for the genes were confirmed by both direct nucleotide sequencing and by GENESCAN software analysis following electrophoresis on the 377 automated ABI DNA sequencer (Applied Biosystems, Foster City, CA) on a subset of randomly selected samples. Quality control for the DNA analyses was maintained by the use of both positive and negative controls in each set of analyzed samples and results were confirmed independently by two different laboratory workers.

Statistical Methods

We created separate logistic regression models to assess the relationship between VT and Factor V Leiden, the ACE polymorphism, and the MTHFR polymorphism. Since all subjects were either heterozygous for the Factor V Leiden mutation or homozygous for its absence, the Factor V defect was classified simply as present versus absent. The MTHFR genotypes are denoted as alanine/alanine (A/A), alanine/valine (A/V), and valine/valine (V/V). The ACE genotypes are denoted as insertion/insertion (I/I), insertion/deletion (I/D), and deletion/deletion (D/D). We analyzed the MTHFR polymorphism by considering the expression of the V allele either as dominant (V/V and A/V versus A/A) or as recessive (V/V versus A/V and A/A). Additionally, we fit a logistic model with scores of 0 for the A/A genotype, 1 for the A/V genotype, and 2 for the V/V genotype. This model assumes that the presence of an additional V allele is associated with a multiplicative increase in VT risk. We performed an analogous analysis for the D allele for the ACE gene. The odds ratio (OR) was used as a measure of association between the genes and VT (21).

We included a continuous variable for age and an indicator for gender in our logistic models. Alcohol consumption, cigarette smoking, and family history of thrombosis were evaluated as potential confounders. Alcohol habit pertained to average lifetime use. Since these factors did not confound any of the gene-disease associations, we obtained the odds ratios relating the genetic traits and VT from a logistic model which included terms only for the matching factors (age and gender). All reported p values are two-tailed and 95 percent confidence intervals are used.

Results

The distribution of cases and controls according to basic demographics and history of other illnesses is displayed in Table 1. The mean age for cases and controls is 55 and 56 respectively (range 18-89 years for cases and 21-93 for controls). Fifty-four of the cases and 94 of the controls are women. Cases and controls were similar with respect to a history of high blood pressure or diabetes, a family history of any vascular disease, cigarette habit, and alcohol consumption. Eight cases had a history of cancer. Cancer was more common among controls, some of whom were obtained from the general medicine clinic which includes cancer patients.

In Table 2 the ORs relating VT and each of the three genetic traits is displayed. The Factor V mutation is not associated with VT among these study subjects. However, the OR pertaining to Factor V Leiden is imprecise due to the low prevalence of the mutation among study subjects. The prevalence of the V allele among our study subjects also was too low to evaluate adequately the recessive allele model (V/V versus A/V and A/A). However, the dominant allele model (V/V and A/V versus A/A) was more informative. There was little or no association between VT and the combined homozygous and heterozygous genotypes of the V allele compared with the A/A genotype (see Table 2). The overall prevalence of the V allele was 9.1% and 10.0% among cases and controls, respectively (p > 0.20). This observation further supports the belief that there is little or no association between the MTHFR polymorphism and VT.

The OR for the ACE D/D genotype compared with the other two ACE genotypes combined is moderately elevated, but not statistically significantly so (p value = 0.13). Among women, the OR for the D/D genotype compared with the other two combined is near its null value of unity (OR = 0.9; 95 percent CI: 0.5-1.9) whereas among men the D/D genotype is associated with about a tripling of VT risk (OR = 2.8; 95 percent CI: 1.2-6.2; p value = 0.01). This difference between the gender specific ORs for the D/D genotype approaches statistical significance (p = 0.06). The ORs for the D/D and I/D genotypes combined versus the I/I genotype are below unity both for men and women.

Discussion

This case-control study allowed us to examine the association of three genetic traits with VT in African-Americans, as well as to compare the prevalence of these traits with those observed in Whites. We did not find an association between Factor V Leaden and VT in this African-American study population. However, lack of statistical power precluded us from an adequate assessment of this gene (the minimal detectable odds ratio to obtain 80% power for a two-tailed test at a level of 5% for Factor V Leaden was about 6.3). We anticipated such low statistical power based upon our interim report that the overall prevalence of Factor V Leaden is significantly lower in African-Americans than in Whites (22).

We found no association between smoking and alcohol use and VT in this study. However, we believe that our use of hospital controls precludes a valid assessment of the relation of such lifestyle factors and VT. This is so because many controls were attending the hospital for treatment of tobacco and alcohol related diseases. Furthermore, we suspect that our study subjects' self-reported data may be inaccurate, biasing the Ors for such factors towards the null. For these reasons, we have not emphasized these findings in the paper, nor have we estimated the Ors relating these factors and VT. Nonetheless, despite this limitation of our study, we believe it is valid for the purpose of evaluating the role of these three genes in the etiology of VT in African-Americans.

Several epidemiological studies have found a direct association between elevated plasma homocysteine levels and VT (10-12). However, to our knowledge, the relation between the MOTHER polymorphism and VT has not been investigated. Frost et al. reported that persons homozygous for the V allele had higher plasma homocysteine levels than did persons with the A/V and A/A genotypes who had similar plasma homocysteine levels (9). We could not adequately evaluate the risk of VT associated with the V/V genotype because the prevalence of the V allele was too low among our study subjects (minimal detectable odds ratio of about 5.5). However, the dominant allele model (V/V and A/V versus A/A) was more informative and these genotypes were unrelated to VT (minimal detectable odds ratio of about 2.3). Thus, this analysis, as well as the observation that the prevalence of the V allele was nearly identical among cases and controls, provides moderate support against the belief that the heterozygous genotype for the V allele is associated with increased VT risk. In a recent study from Ohio, the prevalence of the V allele was significantly lower among African-Americans (10 percent) than it was among Whites (30 percent) (23). The prevalence of the V allele among our controls was also 10 percent, providing additional evidence that the V allele is much less common among African-Americans than it is among Whites in the United States.

The D/D ACE genotype has been linked to an increased risk of cardiovascular disease in numerous epidemiological studies (18,24,25), although one large prospective study did not find the positive association (26). We are unaware of any epidemiological studies of the ACE polymorphism and VT. The dominant allele model (D/D and I/D versus I/I) in our study was unrelated to VT. We found a moderate, though not statistically significant, elevated OR (1.5) for VT for persons with the D/D genotype compared to persons with the I/D or I/I genotypes for men and women combined. However, our gender specific analysis indicates that among African-American men, the risk of VT is about tripled for those with the D/D genotype. The D/D genotype is not associated with increased risk among African-American women. The interaction with gender was almost statistically significant. We have no biologic explanation for this finding and would be inclined to attribute it to chance were it not for the fact that two other studies of the ACE polymorphism also reported gender differences. In a cross-sectional study, Schunkert et al. found that the odds of left ventricular hypertrophy for men with the D/D genotype was increased by about 2.6-fold, while the corresponding OR for women was 1.2 (27). The difference in Ors between men and women in that study is nearly statistically significant (p = 0.07, our calculation). In a study of 182 White persons with coronary artery disease compared with 338 control subjects, Boer et al. reported an OR for the D/D genotype of 2.0 for men and about 0.9 for women (19). This gender difference also appears to be statistically significant (p = 0.02, our calculation). Thus, considered in the context of these two studies, our findings that the D/D ACE polymorphism may increase VT risk in African-American men, but not women, is more plausible.

The prevalence of the D allele among controls in our study is 55 percent and the distribution of the three genotypes is consistent with Hardy-Weinberg equilibrium. This prevalence is nearly identical to that (56 percent) reported for 2,340 U.S. physicians, most of whom are White (26). Among African-Americans living in Michigan and Florida, the prevalence of the D allele was reported as 58 percent and 60 percent, respectively (28,29), though 85 percent of these subjects (n=165) were hypertensive. The prevalence of the D allele was 49 percent among the 39 normotensive subjects in these two studies. Among 80 Nigerian bank workers, the prevalence of the D allele was 59 percent (30). Thus, in contrast to the Factor V Leaden mutation and the V allele for MOTHER gene, the prevalence of the ACE D allele is similar in Whites and African-Americans.

In summary, the present study was too small to investigate adequately the relation between VT and the Factor V Leaden defect or the V/V genotype among African-Americans. It does, however, support the growing body of evidence that indicates that the prevalence of the Factor V Leiden mutation and the V allele for the MTHFR gene are considerably lower among African-Americans compared with Whites, in contrast to the prevalence of the D allele for the ACE gene which is comparable among Whites and African-Americans. The study also suggests that the D/D genotype for the ACE gene is related to VT among African-American men. This latter finding requires confirmation in larger epidemiologic studies.

Table 1. Distribution of cases and controls according to basic demographic characteristics and a history of selected diseases

*A history of blood clots, stroke, or heart attack.

Table 2. Distribution of cases and controls and the odds ratios by gender according to Factor V Leiden, the ACE insertion/deletion polymorphism, and the MTHFR alanine/valine polymorphism

* The odds ratios and 95% confidence intervals for men and women combined adjusted for age and gender.
H Odds ratio from a logistic model containing terms for age and gender and a variable with scores of 0, 1, and 2 for the genotypes A/A, A/V, and V/V, respectively.
I Odds ratio from a logistic model containing terms for age and gender and a variable with scores of 0, 1, and 2 for the genotypes I/I, I/D, and D/D, respectively.

Acknowledgements

The authors wish to acknowledge Ms. Eileen Osinski for her diligent collection of the data for this project.

References

1. Goldhaber SZ, Morpurgo M. Diagnosis, treatment, and prevention of pulmonary embolism. Report of the WHO/International Society and Federation of Cardiology Task Force. Journal of the American Medical Association 1992;268(13):1727-33.
2. Bienvenu T, Ankri A, Chadefaux B, et al. Elevated total plasma homocysteine. A risk factor for thrombosis. Relation to coagulation and fibrinolytic parameters. Thrombosis Research 1993;70:123-29.
3. Mariar RA, Kjeiss AJ, Griffin JH. Mechanism of action of human activated protein C, a thrombin-dependent anticoagulant enzyme. Blood 1982;59:1067.
4. Dahlback B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: Prediction of a cofactor to activated protein C. Procedures of the National Academy of Sciences USA 1993;90:1004.
5. Bertina RM, Koelman BPC, Koster T, et al. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994;369:64-67.
6. Koster T, Rosendaal FR, de Ronder H, et al. Venous thrombosis due to poor anticoagulant response to activated protein C:Leiden Thrombophilia Study. Lancet 1993;342:1503-06.
7. Svensson PJ, Dahlback B. Resistance to activated protein C as a basis for venous thrombosis. New England Journal of Medicine 1994;330:517-22.
8. Ridker PM, Hennekens CH, Lindpainter K, et al. Mutation in the gene coding for coagulation for coagulation factor V and the risk of myocardial infarction, stroke, and venous thrombosis in apparently healthy men. New England Journal of Medicine 1995;332;192-17.
9. Frosst P, Blom HJ, Milos R, et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nature Genetics 1995;10:111-13.
10. Falcon CR, Cattaneo M, Panzeri D, et al. High prevalence of hyperhomocyst(e)inemia in patients with juvenile venous thrombosis. Arterioscler Thromb 1994;14:1080-1083.
11. Fermo I, C'Angelo SV, Paroni R, et al. Prevalence of moderate hyperhomocysteinemia in patients with early-onset venous and arterial occlusive disease. Annals of Internal Medicine 1995;123:747-753.
12. Den Heijer M, Koster T, Blom HJ, et al. Hyperhomocysteinemia as a risk factor for deep-vein thrombosis. New England Journal of Medicine 1996:334:759-62.
13. Rigat B, Hubert C, Alhenc-Gelas F, et al. An insertion/deletion polymorphism in the angiotensin I-converting enzyme gene accounting for half the variance of serum enzyme levels. Journal of Clinical Investigations 1990;86:1343-46.
14. Cambien F. The angiotensin-converting enzyme (ACE) genetic polymorphism: its relationship with plasma Ace level and myocardial infarction. Clinical Genetics 1994;46:94-101.
15. Iwai N, Ohmichi N, Uasuyuki N, et al. DD genotype of the angiotensin-converting enzyme gene is a risk factor for left ventricular hypertrophy. Circulation 1994;90(6):2622-28.
16. Sharma P, Carter ND, Barley J, et al. Polymorphisms in the gene encoding angiotensin 1-converting enzyme and relationship to its post-translational product in cerebral infarction. Journal of Human Hypertension 1994;8:633-34.
17. Ridker PM, Gadboury CL, Conlin PR, et al. Stimulation of plasminogen activator inhibitor in vivo by infusion of angiotensin II. Evidence of a potential interaction between the renin-angiotensin system and fibrinolytic function. Circulation 1993;87:1969-73.
18. Cambien F, Poirier O, Lecert L, et al. Deletion polymorphism in the gene for angiotensin-converting enzyme is a potent risk factor for myocardial infarction. Nature 1992;359:641-644.
19. Boehar N, Damaraju S, Prather A, et al. Angiotensin-I converting enzyme genotype DD is a risk factor for coronary artery disease. Journal of Investigative Med 1995;43:275-80.
20. Benson JM, Phillips DJ, Holloway BP, et al. Olignucleotide ligation assay for detection of the factor V mutation (Arg 506->Gln) causing protein C resistance. Thrombosis research 1996;83:87-96.
21. Hosmer DW, Lemeshow S. Applied Logistic Regression. New York: John Wiley and Sons, 1989.
22. Hooper WC, Dilley AB, Ribeiro MJ, et al. A racial difference in the prevalence of the Arg506>Gln mutation. Thrombosis Research 1996;81(5):577-81.
23. McAndrew PE, Brandt JT, Pearl DK, et al. The incidence of the gene for thermolabile methylene tetrahydrofolate reductase in African-Americans. Thrombosis Research 1996;83(2):195-98.
24. Ruiz J, Blanche H, Cohen N, et al. Insertion/deletion polymorphism of the angiotensin-converting enzyme gene is strongly associated with coronary heart disease in non-insulin dependent diabetes mellitus. Procedures of the National Academy of Sciences USA 1994;91:3662-5.
25. Nakai K, Itoh C, Miura Y, et al. Deletion polymorphism of the angiotensin I-converting enzyme gene is associated with serum ACE concentration and increased risk for CAD in the Japanese. Circulation 1994;90;2199-2202.
26. Lindpainter K, Pfeffer MA, Kreutz R, et al. A prospective evaluation of an angiotensin-converting-enzyme gene polymorphism and the risk of ischemic heart disease. New England Journal of Medicine 1995;332:706-11.
27. Schunkert H, Hense HW, Holmer SR, et al. Association between a deletion polymorphism of the angiotensin-converting-enzyme gene and left ventricular hypertrophy. New England Journal of Medicine 1994;330:1634-8.
28. Duru K, Farrow S, Wang JM, et al. Frequency of a deletion polymorphism in the gene for angiotensin converting enzyme is increased in African-Americans with hypertension. The American Journal of Hypertension 1994;7:759-62.
29. Rutledge DR, Kubilis P, Browe CS, et al. Polymorphism of the angiotensin I converting enzyme gene in essential hypertensive patients. Biochemistry and Molecular Biology International 1995;35(3):661-8.
30. Barley J, Blackwood A, Carter ND, et al. Angiotensin converting enzyme insertion/deletion polymorphism: association with ethnic origin. Journal of Hypertension 1994;12:955-57.