A genomewide study of body mass index and its genetic correlation with thromboembolic risk

SummaryThrombosis and obesity are complex epidemiologically associated diseases. The mechanism of this association is not yet understood. It was the objective of this study to identify genetic components of body mass index (BMI) and their possible role in the risk of thromboembolic disease. With the self-reported BMI of 397 individuals from 21 extended families enrolled in the GAIT (Genetic Analysis of Idiopathic Thrombophilia) Project, we estimated the heritability of BMI and the genetic correlation with the risk of thrombosis. Subjects were genotyped for an autosomal genome-wide scan with 363 highly-informative DNA markers. Univariate and bivariate multipoint linkage analyses were performed. The heritability for BMI was 0.31 (p= 2.9×10–5). Thromboembolic disease (including venous and arterial) and BMI had a significant genetic correlation (ρG= 0.54, p= 0.005). Two linkage signals for BMI were obtained, one at 13q34 (LOD= 3.36, p= 0.0004) and other at 2q34, highly suggestive of linkage (LOD= 1.95). Bivariate linkage analysis with BMI and thrombosis risk also showed a significant signal at 13q34 (LOD= 3), indicating that this locus influences at the same time normal variation in the BMI phenotype as well as susceptibility to thrombosis. In conclusion, BMI and thrombosis are genetically correlated. The locus 13q34, which showed pleiotropy with both phenotypes, contains two candidate genes, which may explain our linkage pleiotropic signal and deserve further investigation as possible risk factors for obesity and thrombosis.

Evidence for QTLs Influencing the Endogenous Thrombin Potential (ETP) and Protein C Concentrations in a Dutch Family with Unexplained Thrombophilia.

Introduction: In Dutch families with unexplained thrombophilia (GENES study), plasma levels of the endogenous thrombin potential (ETP) discriminated affected from non-affected family members. A high variance, i.e. >60%, attributable to genetic factors was found for the ETP in one extended family. Increases in ETP levels could not be readily explained by known determinants of ETP from the coagulation cascade or anticoagulant pathways. In the present study, a genome-wide scan was performed in this family to identify quantitative trait loci (QTL) influencing ETP and several other coagulation factors. Methods: Phenotypic data, plasma, and DNA were available for 135 of the 185 individuals in the pedigree. Patients using anticoagulation medication were excluded from the analyses. If necessary, data were transformed prior to the analyses. Multipoint Identity by Descent (IBD)-estimation and variance component multipoint (bivariate) linkage analysis including covariates were performed using SOLAR. LOD scores were empirically adjusted. Results: For ETP levels, suggestive evidence for linkage was found at 36 cM (one-lod support interval: 1 – 61 cM) on chromosome 20 (LOD = 2.57). Interestingly, significant evidence for linkage with protein C levels (LOD=6.6) was found in the same region at 48 cM (41 – 59 cM). A bivariate LOD score of 5.8 (p<2x10−6) was obtained for this region of chromosome 20, but the correlation between the two traits due to QTL effects was low (0.10 ± 0.28, p=0.734). This finding, together with the fact that the test for complete pleiotropy had to be rejected (p<0.0001), indicates that ETP and protein C levels are most likely influenced by two different QTLs (i.e. coincident linkage) on chromosome 20. Conclusion: In this Dutch family, ETP and protein C levels were influenced by two putative major genes, both localized in the same region of chromosome 20. This region is known to harbor several liver-enriched transcription factors, e.g. hepatocyte nuclear factor (HNF) 3b (20p11) and HNF4a (20q12-q13.1). Since most of the plasma proteins that determine the level of ETP are synthesized in the liver, these candidate genes are now further explored within this family with unexplained thrombophilia and inherited high ETP levels.

Pleiotropic effects of novel trans-acting loci influencing human sympathochromaffin secretion

Family studies have suggested a genetic contribution to variation in blood pressure, but the genes responsible have thus far eluded identification. The use of intermediate phenotypes associated with hypertension, such as chromogranin plasma concentrations, may assist the discovery of hypertension-predisposing loci. We measured the concentrations of four chromogranin A (CHGA) and B (CHGB) peptides in 742 individuals from 235 nuclear families. The CHGA- and CHGB-derived peptides displayed significant heritability and revealed significant genetic correlations, most strikingly observed between CHGA361–372 (catestatin) and CHGB439–451. A 5-cM microsatellite genome scan revealed significant and suggestive evidence for linkage on several chromosomes for three of the peptides. Subsequent bivariate linkage analysis for peptides CHGA361–372 and CHGB439–451, which showed evidence for convergent linkage peaks on chromosomes 2, 7, and 13, resulted in increased evidence for linkage to these regions, suggesting pleiotropic effects of these three loci on multiple chromogranin traits. Because CHGA itself is on chromosome 14q32, and CHGB itself is on chromosome 20pter-p12, the pleiotropic regions on chromosomes 2, 7, and 13 must represent trans-acting quantitative trait loci coordinately affecting CHGA/CHGB biosynthesis and/or exocytotic secretion, likely by regulating efferent sympathetic outflow, a conclusion consistent with the in vitro studies presented here of the dual control of both exocytosis and transcription of these peptides by secretory stimuli in chromaffin cells. The results suggest a new approach to heritable autonomic control of circulation and the genetic basis of cardiovascular diseases such as systemic hypertension.