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

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 β-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 β-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.

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 platelet 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], [5], [6], [7], [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 transmission 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 β-fibrinogen gene on 4q28 are associated with variability in circulating levels of fibrinogen [13], [14], [15], [16], [17]. For example, the H2 allele of HindIII β-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], [14], [15], [16], [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 genome-wide scan, detailed information regarding traditional risk factors, as well as the genotypes of the HindIII β-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.