Abstract
ZPI is a recently characterised inhibitory serpin present in plasma. In vitro studies have shown that ZPI inhibits both factor Xa (in the presence of protein Z) and factor XIa. Deficiency of ZPI is predicted to enhance coagulation and may be a risk factor for venous thrombosis. To test this hypothesis we carried out mutation screening within the coding region of the ZPI gene in a cohort of 150 patients with a history of DVT or PE (first event before the age of 60 years) and 150 matched controls. Five PCR products were produced for each subject. Heteroduplex analysis was performed using dHPLC and the PCR products were sequenced directly if a heteroduplex mismatch was identified.
Sixteen mutations/polymorphisms were identified within the coding region of the ZPI gene. Two mutations were regarded as functionally significant as they produced stop codons at arginine 67 and tryptophan 303 and would be expected to result in the loss of ZPI activity. These stop codon mutations were found in 8 patients and 2 controls. Based on these results, a power calculation was performed that showed that an additional 100 patients and controls would be necessary to confirm that the stop codon mutations were statistically significant risk factors for thrombosis. Bidirectional allele specific PCR was developed to rapidly identify these stop codon mutations and a further 100 patients and 100 controls were examined using this technique. The stop codon mutations were found in 11 patients (W303X n=8: R67X n=3) and only 2 controls (R67X n=2) (two sided Fishers exact p=0.02) with an odds ratio = 5.7 (95%CI, 1.25 to 26.0). In addition two further non-conservative mutations were identified in two other thrombosis patients. These resulted in a serine to tyrosine (codon 122) and a phenylalanine to leucine (codon 124) change in the region homologous to the D helix of other heparin activated serpins. These changes, involving bulky aromatic amino acid residues, have the potential to disrupt the structure and function of the D helix.
We identified a further 12 mutations/polymorphisms (C454G, C574T, A603G, A647G, G752A, A947T, C1276T, G1277A, G1438A, A1617C, G1789T and C1811T (italics: not previously described)). The significance of these mutations is uncertain. We have identified mutations causing stop codons, which will result in loss of ZPI function, in 4.4% of patients who present with venous thrombosis before the age of 60 years, compared with an incidence of only 0.8% in controls. Further studies are ongoing to measure the plasma concentration of ZPI in the affected individuals. Our results support an association between mutations in the ZPI gene and venous thrombosis. We propose that ZPI deficiency as a result of these mutations is potentially a new form of thrombophilia.
Author notes
Corresponding author