A genome-wide search for genes affecting circulating fibrinogen levels in the Framingham Heart Study

Thrombosis Research 110 (2003) 57 – 64

Regular Article

A genome-wide search for genes affecting circulating fibrinogen levels in the Framingham Heart Study Qiong Yang a,b,*, Geoffrey H. Tofler c, L. Adrienne Cupples a, Martin G. Larson d,e, DaLi Feng d, Klaus Lindpaintner f,g, Daniel Levy d,e, Ralph B. D’Agostino h, Christopher J. O’Donnell e,i a

Department of Biostatistics, Boston University School of Public Health, Boston, MA, USA b Department of Neurology, Boston University School of Medicine, Boston, MA, USA c Royal North Shore Hospital, Sydney, Australia d Department of Medicine, Boston University School of Medicine, Boston, MA, USA e National Heart, Lung and Blood Institute’s Framingham Heart Study, Framingham, MA, USA f Max Delbrueck Center for Molecular Medicine, Berlin, Germany g F. Hoffmann-La Roche, Ltd., Basel, Switzerland h Department of Mathematics and Statistics, Boston University, Boston, MA, USA i Cardiology Division, Massachusetts General Hospital, Boston, MA, USA Received 30 January 2003; received in revised form 30 April 2003; accepted 1 May 2003

Abstract Introduction: Circulating levels of fibrinogen are associated with atherosclerosis and predict future coronary heart disease and stroke. Levels of fibrinogen are correlated among family members, suggesting a heritable component. Variants of the h-fibrinogen gene subunit on 4q28 are associated with fibrinogen levels but explain only a small proportion of the total genetic variability. It remains unknown what role, if any, is played by other genetic variants in the inter-individual variability in levels of fibrinogen in the general population. Materials and methods: We conducted a 10-cM spaced genome-wide scan using 402 original cohort subjects and 1193 offspring subjects from 330 extended families of the Framingham Heart Study. Heritability and linkage analyses were carried out using variance component methods. Regression analyses were performed to adjust for traditional risk factors and HindIII h-148 genotypes. Results and Discussions: The total heritability was estimated as 0.24. The highest and second highest LOD scores of linkage were found on chromosomes 2 (LOD = 1.5 at 243 cM) and 10 (LOD = 2.4 at 87 cM) using only offspring subjects in the analysis, and on chromosomes 2 (LOD = 2.1 at 242 cM) and 10(LOD = 1.4 at 86 cM), 17 (LOD = 1.4 at 96 cM) and 20 (LOD = 1.4 at 80 cM) using both original cohort and offspring. These results suggest that there may be influential genetic regions on these chromosomes. While no linkage with genome-wide significance was detected, further research to confirm our findings is warranted. D 2003 Elsevier Science Ltd. All rights reserved. Keywords: Genome-wide scan; Genetic linkage; Fibrinogen; HindIII h-148; Cardiovascular disease; Variance component analysis; Framingham Heart Study

1. Introduction Acute coronary thrombosis and thrombotic stroke are major causes of death and disability in the western world. Fibrinogen has been extensively studied in relation to atherothrombosis since fibrinogen is essential for the plate-

Abbreviations: QTL, quantitative trait locus; cM, centimorgan; BMI, body mass index; CVD, cardiovascular disease. * Corresponding author. Department of Biostatistics, Boston University School of Public Health, 715 Albany Street, T-4E, Boston, MA 02118, USA. Tel.: +1-617-414-1258; fax: +1-617-638-4458. E-mail address: [email protected] (Q. Yang).

let adhesion and aggregation and subsequent thrombus formation. Increased circulating levels of fibrinogen are associated with coronary risk factors [1], presence and progression of subclinical atherosclerosis [2,3] and are prospectively associated with the development of coronary heart disease, stroke and other cardiovascular diseases independent of the traditional risk factors [4 –8]. There has been much interest in identification of genetic determinants of fibrinogen levels. Inconsistent results, however, have been reported in the literature. For example, Hamsten et al. [9] reported that additive genetic factors explain 51% of the total variance. Livshits et al. [10] suggested that fibrinogen levels are in Mendelian transmis-

0049-3848/03/$ - see front matter D 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0049-3848(03)00288-3

58

Q. Yang et al. / Thrombosis Research 110 (2003) 57–64

sion and co-dominant alleles at a major gene locus account for 39% of the total variance. Less impressive heritabilities, 27% and 30%, respectively, were reported in Berg and Kierulf [11] and Reed et al. [12]. It has been widely reported that common genetic variants in the h-fibrinogen gene on 4q28 are associated with variability in circulating levels of fibrinogen [13 – 17]. For example, the H2 allele of HindIII h-148 polymorphism was associated with elevated fibrinogen levels [17]. However, the proportion of genetic variability explained by any of these subunit variants is small to modest [13 – 17]. Genome-wide linkage analysis is a powerful tool for identifying chromosomal regions that may be linked to circulating levels of fibrinogen. As all human genes and their common variants are catalogued through the Human Genome project and related efforts, genome-wide linkage results will focus research by assigning a high priority to specific candidate genes to pursue for mutation screening and functional assays. There are limited data on candidate quantitative trait loci (QTLs), with the exception of chromosome 4 that are associated with regulation of circulating levels of fibrinogen in human populations. Few published genome-wide linkage data are available for fibrinogen. The availability of a 10 centimorgan (cM) density genomewide scan, detailed information regarding traditional risk factors, as well as the genotypes of the HindIII h-148 polymorphism and circulating levels of fibrinogen in family members from a large number of pedigrees in the Framingham Heart Study provides a unique opportunity to conduct an analysis of genetic linkage for fibrinogen levels.

2. Materials and methods 2.1. Study subjects The study subjects are participants in the Framingham Heart Study. The selection criteria and study design of the Framingham Heart Study have been detailed previously [18,19]. The study began in 1948 with the enrollment of 5209 men and women, referred as original cohort, who have undergone examination biennially. Starting in 1971, 5124 offspring and their spouses, referred as offspring, of the original cohort were recruited and examined every 4 years (except an 8-year gap between the first and the second examinations). All subjects provided informed consent prior to each clinic visit, and the examination protocol was approved by the Institutional Review Board at Boston Medical Center (Boston, MA). The genome-wide linkage analyses were based on 330 largest extended families consisted of 1672 offspring and 1213 original cohort subjects, where 1308 offspring and 394 original cohort members in the 330 families were

genotyped in the genome-wide scan. There were 1193 offspring and 402 original cohort members in the 330 families who had circulating fibrinogen levels measured, where 1023 offspring and 204 original cohort subjects were also genotyped for HindIII h-148 polymorphism. The circulating fibrinogen levels of offspring subjects were measured during 5th examination cycle between 1991 and 1995, while those of original cohort subjects were measured during 10th examination cycle between 1968 and 1970. 2.2. Determination of circulating fibrinogen level For measurement of fibrinogen in the offspring subjects, venous blood samples were obtained in the morning to minimize the circadian effects on hemostatic factors levels. Blood was drawn in 3.8% sodium citrate solution with a blood/sodium citrate ratio of 9:1 in volume. Platelet-poor plasma was obtained by centrifugation at 2000  g and stored at 80 jC for later analysis. Fibrinogen levels were determined by the Clauss method [20]. The coefficients of variation of intra-assay and inter-assay were 2.6% and 4.7%, respectively, in our laboratory. Fibrinogen level in the original cohort subjects was measured by a modified method of Ratnoff and Menzie, as previously described [4]. The plasma was clotted with thrombin and recalcified and the clot was wrapped on a glass rod and washed with saline. The dried clot was put in sodium hydroxide, heated and hydrolyzed and the fibrinogen level read in a spectrophotometer. 2.3. Clinical definitions of traditional risk factors Body mass index (BMI, in kg/m2) was calculated using the measurements of weight and height. Prevalent diabetes was defined as a fasting blood glucose level >7.8 nmol/ l (140 mg/dl) for offspring subjects and as a non-fasting blood glucose level >11.1 nmol/l (200 mg/dl) for original Table 1 Characteristics of offspring subjects in 330 families who had circulating fibrinogen levels Variable

Mean F Standard deviation Men (n = 573) Women (n = 620)

Age (years) Body mass index (kg/m2) Smoking (%) Alcohol consumption (oz/week) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Anti-hypertensive therapy (%) Conjugated estrogen use (%) Presence of diabetes mellitus (%) Presence of CVD (%) h-148 genotype H1/H2 or H2/H2 (%) Circulating fibrinogen level (mg/dl)

50.9 F 10.0 28.1 F 4.1 20.5 3.8 F 4.5 126.6 F 16.8 77.4 F 9.8 12.7 N.A. 5.5 2.2 36.5 290.8 F 57.6

51.8 F 10.2 27.0 F 5.9 19.3 1.7 F 2.6 121.6 F 19.5 72.9 F 10.1 12.9 11.8 4.8 1.5 33.3 301.5 F 57.3

Abbreviations: CVD = cardiovascular disease, N.A. = not applicable.

Q. Yang et al. / Thrombosis Research 110 (2003) 57–64

59

Fig. 1. Genome-wide multipoint LOD scores for fibrinogen based on offspring subjects: with vs. without adjustment for fibrinogen HindIII h-148 genotype. Abbreviation: Chrom, chromosome number.

60

Q. Yang et al. / Thrombosis Research 110 (2003) 57–64

cohort subjects, or treatment with dietary modification, insulin or oral hypoglycemic agents at any examination up to the time of fibrinogen-level determination for both offspring and original cohort subjects. Prevalent cardiovascular disease (CVD) was defined as the presence of coronary heart disease (coronary insufficiency, angina, myocardial infarction), cerebrovascular disease (stroke or transient ischemic attack), congestive heart failure or peripheral vascular disease (intermittent claudication). Diastolic and systolic blood pressures were defined by the average of two physician-obtained measurements of seated blood pressure with a mercury column sphygmomanometer. Alcohol consumption was reported as the usual number of drinks (of comparable ethanol content) per day. Smoking (yes/no) was defined by regular cigarette smoking during the past year. Estrogen use (yes/no) was recorded if a woman was taking conjugated estrogen therapy after menopause. 2.4. Genotyping methods for genome-wide scan and HindIII b-148 polymorphism DNA specimens were obtained from blood samples routinely collected during examinations of original and offspring subjects between 1987 and 1991, and between 1995 and 2000. DNA was extracted from buffy coat of whole blood specimens using a Qiagen Blood and Cell Sulture DNA Maxi Kit. A genome-wide scan was conducted by the Mammalian Genotyping Service in Marshfield, WI. The genome-wide scan was based on a genetic map (set 8A) with average 10 cM density and 399 microsatellite markers over the 22 chromosomes. The screening set and genotyping protocols are available at the website of the Center for Medical Genetics, Marshfield Medical Research Foundation (http://research.marshfieldclinic.org/). A modified PCR-based RFLP analysis was used to detect the HindIIIh-148 polymorphism, which results in a C for T substitution at position-148 in the promoter region of the fibrinogen h gene. The sequences of the sense primer and anti-sense primer were 5Vattatgtcattgtcagaaaacataagcatttatg3Vand 5Vtctgctaggaatgacttcagaaatggttac3V, respectively. DNA was amplified using PCR. In the presence of the HindIII restriction endonuclease recognition site that represents the more common allelic variant (H1), the 400 base pair (bp) amplification product was cleaved into fragments of 114 and 286 bp. The H2 allele was not cleaved by HindIII. The HindIII-digested amplification product was size-fractionated on an agarose gel. PCR results were scored without knowledge of circulating fibrinogen level. Ninetyeight percent of the subjects were successfully genotyped [21]. 2.5. Heritability and linkage analyses Different assays were used to measure circulating fibrinogen levels in the original cohort and offspring subjects.

There were many more offspring subjects than original cohort subjects in the 330 families, so we focused our analyses mainly on offspring subjects. Secondary analyses were performed on a combined sample of both original cohort and offspring subjects. To reduce the variability of fibrinogen levels caused by known risk factors, we derived normal scores for linkage analyses based on the ranks of studentized residuals from sex-specific (and separately for original cohort and offspring in secondary analyses) multiple linear regression models [22]. Traditional risk factors including age (including squared and cubic terms), body mass index, cardiovascular disease, diabetes, hypertension treatment, diastolic and systolic blood pressure, alcohol consumption, cigarette smoking and estrogen usage were used as covariates in the multivariable regression. In addition, we conducted analyses with and without genotype of HindIII h-148 polymorphism and analyses with and without lipid variables (total cholesterol, HDL cholesterol and triglycerides) adjusted in the regression. Linkage analyses were conducted in SOLAR (Version 1.6.6) using the variance-components methodology that requires the trait to be normally distributed to avoid inflated false positive rates [23]. SOLAR calculates a LOD score by taking log10 of the likelihood ratio of the maximum likelihood with additive variancecomponents for both the trait locus and residual polygenic effect to that with only an additive variance-component for polygenic effect. A non-zero estimate of additive variance component for the trait locus (corresponding to a large LOD score in SOLAR) indicates linkage between the trait locus and the marker under study. In multipoint analyses, linkage Table 2 Multipoint LOD scores peaks z 1.0 from analyses with various samples and adjustments Sample

No. subjects

Additional a adjustment

LOD score peaks ( z 1.0) Chromosome

cM

LOD

Offspring

1193

NA

Offspring

1023

HindIII h-148

Offspring

1008

183 141 94 85 243 142 87 80 240 141 86

1.2 1.0 1.0 1.3 1.5 1.1 2.4 1.1 1.3 1.0 2.0

Offspring and Original Cohort

1227

HindIII h-148 Lipid Variables HindIII h-148

5 6 15 19 2 6 10 20 2 6 10

Offspring and Original Cohort

1177

2 10 17 19 20 2 10 17 20

242 86 96 33 80 239 67 96 80

2.1 1.4 1.4 1.0 1.4 1.7 1.3 1.5 1.5

a

HindIII h-148 Lipid Variables

Adjustments in addition to the traditional risk factors.

Q. Yang et al. / Thrombosis Research 110 (2003) 57–64

61

Fig. 2. Multipoint LOD scores of HindIII h-148 adjusted fibrinogen on chromosomes 2, 10, 17 and 20 from analyses with and without additional adjustment of lipid variables and with and without including original cohort.

62

Q. Yang et al. / Thrombosis Research 110 (2003) 57–64

to adjacent markers was also considered to evaluate the linkage to the current marker using a regression approach [24]. We repeated the multipoint analyses using Genehunter to address the possibility that the multipoint IBDs may be better estimated by Genehunter than SOLAR [25], but nine of the largest families were split into smaller units to overcome the prohibitively large amount of computer memory required otherwise.

3. Results The mean and standard deviation of circulating fibrinogen levels and of all risk factors are displayed in Table 1. The mean levels of fibrinogen were 290.8 (S.D. = 57.7) and 301.5 (S.D. = 57.3) mg/dl for men and women, respectively. The frequencies of fibrinogen HindIII h-148 genotypes were 65.0% for H1 homozygotes and 32.5% for H1/H2 heterozygotes subjects. The heterozygous and H2 homozygous state were associated with similar levels of fibrinogen, so we combined all genotypes containing H2 and compared all H2-containing subjects with the H1 homozygous subjects. Taken together, multiple covariates, including the HindIII h-148 genotype, accounted for 30.6% of the total phenotypic variance in female and 23.5% in male offspring subjects. The HindIII h-148 genotype accounted for 0.6% of the total variance in women and 1.4% in men. The total heritability of fibrinogen adjusted for HindIII h-148 genotype and other risk factors was 0.24 (95% confidence interval 0.09 –0.39) based on the sample consisting of only offspring subjects. With adjustment for the HindIII h -148 genotype and with only offspring in the analyses, the highest two multipoint LOD scores were found on chromosomes 10 (LOD = 2.4 at 87 cM) and 2 (LOD = 1.5 at 243 cM) (see Fig. 1). Without adjustment for HindIII h-148 genotype, the highest two multipoint LOD scores were found on chromosomes 19 (LOD = 1.3 at 85 cM) and 5 (LOD = 1.2 at 183 cM). Other LOD score peaks z 1 were found on chromosomes 6 and 20 and on chromosomes 6 and 15 with and without adjustment for HindIII h-148 genotype, respectively (Table 2). When we repeated above analyses of h-148 genotype adjusted fibrinogen in Genehunter, the maximum multipoint LOD score was found on chromosome 2 (LOD = 1.5 at 227 cM), close to the peak found in SOLAR analyses. The LOD score peak on chromosome 10 (LOD = 0.9 at 81 cM) in Genehunter analyses was also close to the location of the peak we found in SOLAR analyses, but the LOD score was lower. Secondary analyses were conducted using the sample consisting of both original cohort and offspring subjects who had both fibrinogen levels and HindIII h-148 genotype. In these analyses, the LOD score curves revealed a similar

pattern but some differences in magnitudes compared to analyses based upon offspring only. The highest LOD score was obtained on chromosome 2 (LOD = 2.1 at 242 cM). Both offspring-only and cohort-and-offspring analyses found z 1 LOD scores at same regions on chromosomes 10 and 20 (Table 2 and Fig. 2). Results were similar in models with and without adjustment for lipid variables in addition to HindIII h -148 genotype (Table 2 and Fig. 2) for cohort-and-offspring analyses. Using both original cohort and offspring, the magnitude of the LOD scores was generally slightly lower than that of using offspring only, with exceptions for linkage peak regions on chromosomes 17 and 20 (Fig. 2).

4. Discussion In the genome-wide search for circulating levels of fibrinogen in the 330 extended families of Framingham Heart Study, we found suggestive evidence of linkage (LOD>2.2) on chromosome 10, according to criteria proposed by Lander and Kruglyak [26]. We also found evidence of other potentially interesting linkages (LOD>1.5) on chromosomes 17 and 20. There was no evidence for significant linkage to the region of chromosome 4 harboring the genes encoding the a, h and g subunits of fibrinogen, including 4q28 that contained the HindIII h-148 polymorphism significantly associated with fibrinogen level in our study and other studies [13]. This could be due to a small amount of total phenotypic variance explained by those genes. Adjusting for the fibrinogen HindIII h-148 genotype resulted in a substantial increase in multipoint LOD scores in the two regions of chromosome 10 and 2. This indicates that the effect of the QTLs, if any, in the two regions with suggestive linkage was either confounded by the HindIII h-148 genotype and/or had an interaction with the genotype. Although the fibrinogen HindIII h-148 genotype itself only accounted for a small percent of the total phenotypic variance, the adjustment increased the difference in the maximum likelihood between the models with and without the QTL effects from the two linkage suggestive regions. It was also of interest to note that the only consistent LOD score peak z 1 before and after the adjustment for HindIII h-148 genotype was the one on chromosome 6 (Fig. 1). Since the fibrinogen HindIII h-148 gene is not highly polymorphic, we did not evaluate potential gene –gene interactions by linkage analyses, given the poor power for a cohort of our sample size [27]. Future investigation that constructs a finer map than the current one in the linkage interesting regions would be needed in order to further refine the location of the QTLs and to test the interaction between the QTLs and fibrinogen HindIII h-148 genotype by more powerful association analysis methods. Offspring-only analyses using Genehunter confirmed the linkage evidence we found on chromosome 2, but

Q. Yang et al. / Thrombosis Research 110 (2003) 57–64

not that on chromosome 10. Genehunter computes multipoint IBD more accurately [25] than SOLAR, but it has to split and trim some of the large families into smaller families, which may cause reduction in power to detect linkage in our case. We conducted additional analyses using subjects from two generations because few if any other community-based studies have this capability. When both original cohort and offspring subjects were included in the analysis, there were changes of magnitude of multipoint LOD scores on chromosomes 2, 10, 17 and 20. One potential limitation to this approach is that differences in the assays used to measure the circulating fibrinogen levels between the two generations may bias the results. There are a number of other common polymorphisms than the HindIII h-148 in h-fibrinogen gene associated with fibrinogen levels [15,16,28,29]. Adjusting for one of them, the HindIII h-148, may be adequate to some extent, since many of them are in tight linkage disequilibrium. However, future research should aim to evaluate the impact of all common variants in the a, h and g subunits that underlie variability in fibrinogen levels. There are potential study limitations. First, our study was conducted in a largely Caucasian population. It is uncertain whether findings in our largely Caucasian population can be generalized to other racial or ethnic groups in which there may be differential effects of specific genes due to gene – environment effects and differential prevalences of genes important in thrombosis. Second, our LOD scores are below the threshold value of 3.0 considered by some to be the minimal requirement for genome-wide significance. While LOD scores nearly 2.0 or above may represent true linkage, it is also possible that LOD scores of this magnitude may represent false positives. To address this, we conducted a simulation to estimate type I error for genome-wide linkage analyses using SOLAR with our families structure and genome-wide scan. The simulation result shows (based on 100 replicates) that multipoint LOD scores of 2.4 and 1.5 in our genome-wide linkage analyses correspond to type I errors of 0.13 and 0.81, respectively. Thus, the probability is 13% that a LOD score of 2.4 is a false positive. Substantial evidence from prospective studies has linked levels of circulating fibrinogen to the incidence of coronary heart disease and stroke [4 –8], but the genetic determinants of fibrinogen levels remain incompletely understood and our findings indicate chromosomal segments of interest on chromosomes 10 and 2. In this regard, an examination of the region of chromosome 10 reveals several potentially interesting candidates for genetic variability in fibrinogen levels, including urokinase plasminogen activator (PLAU) and proteoglycan 1, secretory granule (PRG1), and neurogenin 3 (NEUROG3). Furthermore, within the vicinity of the peak linkage of chromosome 2 resides a cluster of calpain 10 isoforms—calpain 10 has recently been linked to diabetes mellitus—as well as high-density lipoprotein binding protein (HDLBP). Thus, the findings in our study,

63

while not conclusive, merit comparison and confirmation in independent genome-wide linkage analyses.

References [1] Stec JJ, Silbershatz H, Tofler GH, et al. Association of fibrinogen with cardiovascular risk factors and cardiovascular disease in the Framingham Offspring Population. Circulation 2000;102(14):1634 – 8. [2] Chambless LE, Folsom AR, Davis V, et al. Risk factors for progression of common carotid atherosclerosis: the Atherosclerosis Risk in Communities Study, 1987 – 1998. Am J Epidemiol 2002;155(1): 38 – 47. [3] Folsom AR, Wu KK, Shahar E, et al. Association of hemostatic variables with prevalent cardiovascular disease and asymptomatic carotid artery atherosclerosis. Arterioscler Thromb 1993;13: 1829 – 36. [4] Kannel WB, Wolf PA, Castelli WP, et al. Fibrinogen and the high risk of cardiovascular disease: the Framingham study. JAMA 1987;258: 1183 – 6. [5] Lee AJ, Lowe GDO, Woodward M, et al. Fibrinogen in relation to personal history of prevalent hypertension, diabetes, stoke, intermittent claudication, coronary heart disease, and family history: the Scottish Heart Health Study. Br Heart J 1993;69:338 – 42. [6] Meade TW, Brozovic M, Chakrabarti RR, et al. Hemostatic function and ischemic heart disease: principal results of the Northwick Park Heart Study. Lancet 1986;6:533 – 7. [7] Danesh J, Collins R, Appleby P, et al. Association of fibrinogen, Creactive protein, albumin, or leukocyte count with coronary heart disease: meta-analyses of prospective studies. JAMA 1998;279(18): 1477 – 82. [8] Wilhelmsen L, Sva¨rdsudd K, Korsan-Bengtsen K, et al. Fibrinogen as a risk factor for stroke and myocardial infarction. N Engl J Med 1984;311:501 – 5. [9] Hamsten A, Iselius L, de Faire U, et al. Genetic and cultural inheritance of plasma fibrinogen concentration. Lancet 1987;2(8566): 988 – 91. [10] Livshits G, Schettler G, Graff E, et al. Tel-Aviv-Heidelberg threegeneration offspring study: genetic determinants of plasma fibrinogen level. Am J Med Genet 1996;63:509 – 17. [11] Berg K, Kierulf P. DNA polymorphisms at fibrinogen loci and plasma fibrinogen concentration. Clin Genet 1989;36(4):229 – 35. [12] Reed T, Tracy RP, Fabsitz RR. Minimal genetic influences on plasma fibrinogen level in adult males in the NHLBI twin study. Clin Genet 1994;45(2):71 – 7. [13] Cook DG, Cappuccio FP, Atkinson RW, et al. Ethnic differences in fibrinogen levels: the role of environmental factors and the beta-fibrinogen gene. Am J Epidemiol 2001;153(8):799 – 806. [14] Humphries SE, Cook M, Dubowitz M, et al. Role of genetic variation at the fibrinogen locus in determination of plasma fibrinogen concentrations. Lancet 1987;1(8548):1452 – 5. [15] Behague I, Poirier O, Nicaud V, et al. h-Fibrinogen gene polymorphisms are associated with plasma fibrinogen and coronary artery disease in patients with myocardial infarction: the ECTIM Study. Circulation 1996;93:440 – 9. [16] Baumann RE, Henschen AH. Linkage disequilibrium relationships among four polymorphisms within the human fibrinogen gene cluster. Hum Genet 1994;94:165 – 70. [17] van’t Hooft FM, von Bahr SJF, Silveira A, et al. Two common, functional polymorphisms in the promoter region of the h-fibrinogen gene contribute to regulation of plasma fibrinogen concentration. Arterioscler Thromb Vasc Biol 1999;19(12):3063 – 70. [18] Dawber TR, Meadors GF, Moore FEJ. Epidemiological approaches to heart disease: the Framingham Study. Am J Public Health 1951; 41:279 – 86. [19] Kannel WB, Feinleib M, McNamara PM, et al. An investigation of

64

[20] [21]

[22] [23]

[24] [25]

Q. Yang et al. / Thrombosis Research 110 (2003) 57–64 coronary heart disease in families: the Framingham Offspring Study. Am J Epidemiol 1979;110:281 – 90. Von Clauss A. Gerinnungsphysiologische Schnellmethode zur Bestimmung des Fibrinogens. Acta Haematol 1957;17:237 – 46. O’Donnell CJ, Larson MG, Feng DL, et al. Genetic and environmental contributions to platelet aggregation: the Framingham Heart Study. Circulation 2001;103(25):3051 – 6. SAS/STAT user’s guide, Version 6. 4th ed. Cory, NC: SAS Institute; 1989. Allison DB, Neale MC, Schork NJ, et al. Testing the robustness of the likelihood-ratio test in a variance-component quantitative-trait locimapping procedure. Am J Hum Genet 1999;65(2):531 – 44. Almasy L, Blangero J. Multipoint quantitative-trait linkage analysis in general pedigrees. Am J Hum Genet 1998;62(5):1198 – 211. Sobel E, Sengul H, Weeks DE. Multipoint estimation of identity-bydescent probabilities at arbitrary positions among marker loci on general pedigrees. Hum Hered 2001;52:121 – 31.

[26] Lander E, Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 1995; 11:241 – 7. [27] Almasy L, Dyer TD, Blangero J. Bivariate quantitative trait linkage analysis: pleiotropy versus co-incident linkages. Genet Epidemiol 1997;14:953 – 8. [28] Fellowes AP, Brennan SO, George PM. Identification and characterization of five new fibrinogen gene polymorphisms. Ann NY Acad Sci 2001;936:536 – 41. [29] Thomas A, Lamlum H, Humphries S, et al. Linkage disequilibrium across the fibrinogen locus as shown by five genetic polymorphisms, G/A2455(HaeIII), C/T2148(HindIII/AluI), T/G11689(AvaII), and BclI b-fibrinogen and TaqI a-fibrinogen, and their detection by PCR. Hum Mutat 1994;3:79 – 81.