Both genetic and environmental factors contribute to the normal population variability of plasma von Willebrand Factor (vWF) levels, however, regulatory mechanisms at the vWF gene locus itself have not yet been identified. We have investigated the association between polymorphic variation in the 5′-regulatory region of the vWF gene and levels of plasma vWF:Ag in a study of 261 group O blood donors. Three novel single nucleotide polymorphisms (SNPs) were identified in the vWF promoter: C/T at -1234, A/G at -1185, and G/A at -1051. These SNPs had identical allele frequencies of 0.36 for the -1234C, -1185A, and -1051G alleles and 0.64 for the -1234T, -1185G, and -1051A alleles and were in strong linkage disequilibrium. In fact, these polymorphisms segregated as two distinct haplotypes: -1234C/-1185A/-1051G (haplotype 1) and -1234T/-1185G/-1051A (haplotype 2) with 12.6% of subjects homozygous for haplotype 1, 40.6% homozygous for haplotype 2, and 42.5% of subjects heterozygous for both haplotypes. Only 4.3% of individuals had other genotypes. A significant association between promoter genotype and level of plasma vWF:Ag was established (analysis of covariance [ANCOVA], P = .008; Kruskal-Wallis test,P = .006); individuals with the CC/AA/GG genotype had the highest mean vWF:Ag levels (0.962 U/mL), intermediate values of vWF:Ag (0.867 U/mL) were observed for heterozygotes (CT/AG/GA), and those with the TT/GG/AA genotype had the lowest mean plasma vWF:Ag levels (0.776 U/mL). Interestingly, when the sample was subgrouped according to age, the significant association between promoter genotype and plasma vWF:Ag level was accentuated in subjects > 40 years of age (analysis of variance [ANOVA], P = .003; Kruskal-Wallis test, P= .001), but was not maintained for subjects ≤ 40 years of age (ANOVA, P > .4; Kruskal-Wallis test, P > .4). In the former subgroup, mean levels of plasma vWF:Ag for subjects with the CC/AA/GG, CT/AG/GA, and TT/GG/AA genotypes were 1.075, 0.954, and 0.794 U/mL, respectively. By searching a transcription factor binding site profile database, these polymorphic sequences were predicted to interact with several transcription factors expressed in endothelial cells, including Sp1, GATA-2, c-Ets, and NFκB. Furthermore, the binding sites at the -1234 and -1051 SNPs appeared to indicate allelic preferences for some of these proteins. Electrophoretic mobility shift assays (EMSAs) performed with recombinant human NFκB p50 showed preferential binding of the -1234T allele (confirmed by supershift EMSAs), and EMSAs using bovine aortic endothelial cell (BAEC) nuclear extracts produced specific binding of a nuclear protein to the -1051A allele, but not the -1051G allele. These findings suggest that circulating levels of vWF:Ag may be determined, at least in part, by polymorphic variation in the promoter region of the vWF gene, and that this association may be mediated by differential binding of nuclear proteins involved in the regulation of vWF gene expression.

Both genetic and environmental factors contribute to the normal population variability of plasma von Willebrand Factor (vWF) levels, however, regulatory mechanisms at the vWF gene locus itself have not yet been identified. We have investigated the association between polymorphic variation in the 5′-regulatory region of the vWF gene and levels of plasma vWF:Ag in a study of 261 group O blood donors. Three novel single nucleotide polymorphisms (SNPs) were identified in the vWF promoter: C/T at -1234, A/G at -1185, and G/A at -1051. These SNPs had identical allele frequencies of 0.36 for the -1234C, -1185A, and -1051G alleles and 0.64 for the -1234T, -1185G, and -1051A alleles and were in strong linkage disequilibrium. In fact, these polymorphisms segregated as two distinct haplotypes: -1234C/-1185A/-1051G (haplotype 1) and -1234T/-1185G/-1051A (haplotype 2) with 12.6% of subjects homozygous for haplotype 1, 40.6% homozygous for haplotype 2, and 42.5% of subjects heterozygous for both haplotypes. Only 4.3% of individuals had other genotypes. A significant association between promoter genotype and level of plasma vWF:Ag was established (analysis of covariance [ANCOVA], P = .008; Kruskal-Wallis test,P = .006); individuals with the CC/AA/GG genotype had the highest mean vWF:Ag levels (0.962 U/mL), intermediate values of vWF:Ag (0.867 U/mL) were observed for heterozygotes (CT/AG/GA), and those with the TT/GG/AA genotype had the lowest mean plasma vWF:Ag levels (0.776 U/mL). Interestingly, when the sample was subgrouped according to age, the significant association between promoter genotype and plasma vWF:Ag level was accentuated in subjects > 40 years of age (analysis of variance [ANOVA], P = .003; Kruskal-Wallis test, P= .001), but was not maintained for subjects ≤ 40 years of age (ANOVA, P > .4; Kruskal-Wallis test, P > .4). In the former subgroup, mean levels of plasma vWF:Ag for subjects with the CC/AA/GG, CT/AG/GA, and TT/GG/AA genotypes were 1.075, 0.954, and 0.794 U/mL, respectively. By searching a transcription factor binding site profile database, these polymorphic sequences were predicted to interact with several transcription factors expressed in endothelial cells, including Sp1, GATA-2, c-Ets, and NFκB. Furthermore, the binding sites at the -1234 and -1051 SNPs appeared to indicate allelic preferences for some of these proteins. Electrophoretic mobility shift assays (EMSAs) performed with recombinant human NFκB p50 showed preferential binding of the -1234T allele (confirmed by supershift EMSAs), and EMSAs using bovine aortic endothelial cell (BAEC) nuclear extracts produced specific binding of a nuclear protein to the -1051A allele, but not the -1051G allele. These findings suggest that circulating levels of vWF:Ag may be determined, at least in part, by polymorphic variation in the promoter region of the vWF gene, and that this association may be mediated by differential binding of nuclear proteins involved in the regulation of vWF gene expression.

Summary.A role for steroid hormones has been proposed for the post-pubertal factor IX increment of ϳ25% seen in both normal males and females, as well as in the postpubertal phenotypic recovery seen in haemophilia B Leyden. We have evaluated androgen receptor binding to the factor IX promoter and have assessed transcriptional activation of the factor IX gene in hepatocytes through transient transfection studies and through expression of factor IX in a murine model of androgen insensitivity. Whereas transfection of the androgen receptor alone did not activate expression from the factor IX promoter, co-transfection with the CCAAT enhancer binding protein resulted in a synergistic 17-fold enhancement of transcriptional activity. Using liver nuclear extracts and recombinant androgen receptor protein we have confirmed binding of this protein to the factor IX proximal promoter and disruption of binding with a mutation at nucleotide ¹26. Finally, studies in normal and testicular feminized male mice showed different developmental patterns of factor IX expression. In normal mice, expression recapitulates that seen in humans, with early post-natal levels being ϳ50% of the adult values and with a post-pubertal increment of ϳ25%. In contrast, testicular feminized animals did not show a significant post-pubertal increment of factor IX. These studies provide further support for the role of androgen receptor binding to the factor IX promoter in regulating the developmental expression of factor IX.