Modulation of factor V levels in plasma by intragenic polymorphisms

Abstract number: OC311

Lunghi^* B., Girelli† D., Scanavini^* D., Grandini^* A., Martinelli† N., Bernardi^* F.

^*University of Ferrara, Italy; †University of Verona, Italy

Background  

The study of the pseudohomozygous condition for FV Leiden both in human and in transgenic mice indicates that it could confer an increased risk for thrombosis. Functional polymorphisms lowering FV levels might produce a partially pseudohomozygous condition in FV Leiden carriers, potentially modulating thrombotic risk. We extensively investigated the association of intragenic FV polymorphisms with decreased FV levels in the general population.

Results  

1  Four hundred and two subjects were characterized for FV levels and genotyped for a VNTR marker in FV intron 11. anova conducted for genotype groups indicated the presence of significant (P = 0.006) genotype–phenotype associations. We screened for SNPs by restriction analysis and nucleotide sequencing in FV coding regions. Two frequent genetic components were identified, both located in the FV C2 domain (A) the M2120T polymorphism, found in the heterozygous condition in approximately 25% of subjects belonging to the lowest percentiles and (B) the D2194G polymorphism, the most 3' marker of the FVHR2 haplotype. A missense mutation (Y1702C), previously found to cause FV deficiency, was also detected in the heterozygous condition in one subject. Significant differences in FV levels associated with VNTR genotypes were not detectable after exclusion of 2120T- and 2194G-carriers from the analysis of variance.

2  The analysis of the FV level–SNPs association was extended to a larger sample (1013 subjects). Carriership of the M2120T (48 subjects, 4.7%) or of the D2194G (145 subjects, 14.3%) changes was associated, respectively, with 23% and 19% reduction in FV activity (P < 0.001) as compared with non-carriers. These markers were also found to predict significantly decreased FV antigen levels. Carriers (n = 10) of the FVHR2 haplotype but not of the D2194G polymorphism did not show decreased FV levels.

3  The functional role of these missense changes was confirmed by expression of recombinant 2120TFV (FV:c 57 ± 31% and FV:ag 65 ± 34% of wildtype FV) and of 2194GFV (FV:c 66 ± 24% and FV:ag 74 ± 26% of wildtype FV).

Conclusions  

Our data indicate that a significant proportion of FV plasma levels are determined by FV polymorphisms. Two markers, present in approximately 20% of normal subjects, explain 10.3% of FV activity variance (P < 0.001) in plasma. These functional polymorphisms, which might be frequent in FV Leiden carriers, are candidate markers for genotype–clincal phenotype association studies.