Genotypic Variation in the Promoter Region of the Protein C Gene Is Associated With Plasma Protein C Levels and Thrombotic Risk
Abstract Protein C is a vitamin K–dependent zymogen of a serine protease that inhibits blood coagulation by proteolytic inactivation of factors Va and VIIIa. Individuals with protein C deficiency are at risk for thrombophlebitis, deep-vein thrombosis, and pulmonary embolism. Genetic analysis of a number of randomly chosen healthy individuals revealed three polymorphisms, C/T at −654, A/G at −641, and A/T at −476, in the protein C promoter region. To investigate whether these genetic variations associate with the plasma protein C level, we determined the genotype for the three polymorphisms and measured plasma protein C levels in 240 individuals not deficient in protein C. The mean protein C level of these individuals was 103%. Interestingly, individuals with the homozygous CGT genotype (n=40) had a mean protein C level of 94%, whereas individuals with a homozygous TAA genotype (n=28) had a mean protein C level of 116%. This difference in mean protein C levels between the CGT and TAA groups (P<.001) could not be explained by environmental factors known to influence protein C levels in the normal population. Plasma factor II and factor X levels did not differ between the two groups, which makes a difference in liver function an unlikely cause. Finally, we tested whether the genotype associated with lower protein C levels is associated with higher thrombotic risks. This analysis showed that compared with the genetic variant associated with higher protein C levels (TT/AA/AA), the genetic variant associated with lower protein C levels (CC/GG/TT genotype) is indeed a risk factor for thrombosis (OR, 1.6; 95% confidence interval, 1.0 to 2.5).
- Received September 8, 1994.
- Accepted December 5, 1994.
After activation by the thrombin-thrombomodulin complex, protein C inhibits blood coagulation in the presence of protein S,1 phospholipids, and calcium ions by proteolytic inactivation of factors Va and VIIIa.2 3 Furthermore, activated protein C stimulates fibrinolysis, in part through neutralization of plasminogen activator inhibitor–1 (PAI-1).4
The physiological significance of the protein C anticoagulant activity is clearly shown in individuals with homozygous or compound heterozygous protein C deficiency. These individuals suffer from massive disseminated intravascular coagulation or neonatal purpura fulminans.5 6 Individuals with heterozygous protein C deficiency, although more mildly affected, are at risk for thrombophlebitis, deep-vein thrombosis, or pulmonary embolism.7 8
The diagnosis of heterozygous protein C deficiency depends on the measurement of protein C activity and/or antigen levels. However, the diagnosis of hereditary protein C deficiency from protein C levels alone is sometimes difficult. This has been shown by Allaart et al,9 who documented a considerable overlap in protein C levels between DNA-confirmed heterozygotes and their normal relatives. Moreover, it has been shown that constitutional, lifestyle, or biochemical factors such as age, sex, body mass index, LDL cholesterol, triglycerides, use of oral contraceptives, and race influence protein C levels in healthy individuals.10 11 Smoking might also influence protein C levels, although the results are not consistent.10 11
Another factor that might influence the distribution of plasma protein C levels is genetic variation in the protein C gene itself. Associations between genetic variation and coagulation protein levels have been reported. Green et al12 reported a strong association between a common DNA polymorphism in exon 8 of the factor VII gene and plasma factor VIIc levels. Thomas et al13 showed an association between a common polymorphism in the 5′ flanking region of the β-fibrinogen gene and fibrinogen plasma levels, while Dawson et al14 reported on the association between genetic variation at the PAI-1 locus and plasma levels of PAI-1 activity.
We previously reported three polymorphisms in the promoter region of the human protein C gene that showed a high degree of linkage disequilibrium.15 In this study, we used these three polymorphisms, C/T at −1654, A/G at −1641, and A/T at −1476, to determine the effect of genetic variation on the plasma protein C levels in individuals who were not deficient in protein C. We will show that differences in protein C plasma levels are at least partly associated with genetic variation in the promoter region of the protein C gene. These results immediately raised the question of whether genetic variations are also associated with the risk of developing thrombosis. This proved to be the case.
Individuals participating in this study were selected from a case-control study on venous thrombosis, the Leiden Thrombophilia Study (LETS).16 Briefly, patients were selected from the computer files of the anticoagulation clinics in Leiden, Amsterdam, and Rotterdam. In the Netherlands, anticoagulation clinics monitor coumarin treatment in virtually all patients with venous thrombosis in a defined geographic area.17 Included are all 474 consecutive outpatients younger than 70 years of age who were referred for anticoagulant treatment because of a first, objectively confirmed episode of deep-vein thrombosis. The median time between the occurrence of the deep-vein thrombosis and blood collection was 19 (range, 6 to 68) months. Ninety percent of the eligible patients were willing to take part in the study. The thrombosis patients were asked to find their own healthy control subject. This resulted in 474 population control subjects matched for age and sex.16
To study the influence of genotypic variation on plasma protein C levels, we selected the first 250 individuals entering the LETS. From these 250 individuals, we excluded 10 people, 7 who were protein C deficient, 1 on oral anticoagulant treatment, and 2 who had missing samples. The remaining 240 individuals consisted of 130 patients and 110 control subjects.
For risk factor analysis, we used all 948 individuals of the LETS except 1, who was excluded because of a degraded DNA sample.
Sample Collection and Analysis
Blood was collected from the antecubital vein into Sarstedt Monovette tubes containing 0.106 mmol/L trisodium citrate. Plasma was prepared by centrifugation for 10 minutes at 2000g at room temperature and stored at −70°C in 1.5-mL aliquots. High-molecular-weight DNA was isolated from leukocytes by standard methods.
Protein C activity was measured with Coamate (Chromogenix) on an ACL-200 (Instrumentation Laboratory), factor II activity with Substrate S-2238 (Chromogenix) and Echis carinatus snake venom (Sigma Chemical Co) on an ACL-200 (the ECAR method18 ), and factor X antigen with a Laurell electroimmunoassay.19 Protein C antigen was measured with an enzyme-linked immunosorbent assay technique with the monoclonal antibody C12 as capture antibody and rabbit anti–protein C IgG coupled to horseradish peroxidase (Dakopatts) as tagging antibody.
Protein C activity levels were measured in all 948 subjects; the person performing the test was blinded to the status (ie, case or control) of the sample and the protein C genotype.
Student’s t test (unpaired) was used to compare means. Odds ratios (ORs), as a measure of relative risk, were calculated in the standard unmatched fashion. Undoing the matching allows selection of subgroups of patients and presentation of raw data. Because the present study focuses on genotypes, unmatched analysis will not lead to attenuated effects. A 95% confidence interval (CI) was constructed according to Woolf.20
The protein C promoter genotype was determined as reported previously.15 In brief, three oligonucleotide primers in the protein C promoter region were designed: 5′-ACATCTGTCAAGGGTTTTGCCCTCACCTCCCTCCCAGCTGGA-3′ (−696 to −655; mutagenic primer 1), 5′-TTTTGCCCTCACCTCCCTCCCTGCTGGAT/CGGCCACCTTGGT-3′ (−1683 to −1642; mutagenic primer 2), and 5′-GGCGGGTCGTGGAGATACTG-3′ (−1451 to −1470; normal primer 3). The underlined nucleotides in the primer sequences are not present in the normal sequence. Pair 2 consists of a mixture (1:1) of two oligonucleotides: one contains a T and the other a C at position −1654. Amplification with primers 1 and 3 introduces a BstXI site (-CCAN5/NTGG-) in the amplified DNA fragment when the C/T polymorphism consists of a T. In this instance, the 246–base pair (bp) polymerase chain reaction (PCR) fragment is cut by BstXI into 205- and 41-bp fragments. The amplification with primers 2 and 3 introduces a BstXI site in the amplified DNA fragment when a G is present at the second polymorphic site (A/G polymorphism). The resulting 233-bp PCR fragment is cut by BstXI into 193- and 40-bp fragments. The content of the third polymorphism (A/T) is determined by Rsa I digestion of the PCR fragments, which resulted in 25- and 221-bp fragments (primers 1 and 3) or a 208-bp fragment (primers 2 and 3) when a T is present in the promoter region.
From the first 250 individuals used to determine the influence of genotypic variation on plasma protein C levels, all three polymorphisms were determined. The other 698 individuals were screened only for the presence of the TT/AA/AA or CC/GG/TT genotype. First, individuals were screened for the A/T polymorphism; those individuals who were homozygous at this site (AA or TT) were subsequently screened for the C/T polymorphism. Finally, individuals homozygous for both polymorphisms (TT/../AA or CC/../TT) were also screened for the A/G polymorphism.
Influence of Genotypic Variation on Plasma Protein C Levels
Fig 1⇓ shows the distribution of the plasma protein C activity levels in the 240 individuals. The overall mean was 103% of pooled normal plasma. Patients had a mean level of 105% (n=130), whereas control subjects had a mean level of 102% (n=110).
Samples of all these individuals were genotyped for the three previously reported protein C promoter polymorphisms (C/T, A/G, and A/T). As Table 1⇓ shows, of the 27 possible genotypes, only 9 were observed, of which 5 were most frequent. These five genotypes–CC/GG/TT, CT/AG/AT, CC/AG/AT, CT/AA/AA, and TT/AA/AA–together occurred in 95% of the tested individuals, while the CC/AA/AA genotype occurred in about 3% of these subjects. The other three genotypes–CT/AG/AA, CC/AG/AA, and CC/GG/AT–together accounted for almost 2% of the individuals. All three polymorphisms were in Hardy-Weinberg equilibrium.
Table 1⇑ summarizes the mean protein C activity levels in the different genotype groups. Individuals with the homozygous CGT genotype, ie, CC/GG/TT, had mean protein C levels of 94% (n=40). Complete heterozygotes, ie, CT/AG/AT, had a mean of 104% (n=87), while homozygous TAA individuals, ie, TT/AA/AA, had a mean protein C level of 116% (n=28). Individuals with one of the other six observed genotypes had mean protein C levels close to the overall mean of 103%. As Table 1⇑ shows, there was no notable difference in mean protein C levels between patients and control subjects in any of the genotype groups. These results indicate that a clear difference (22%) in mean protein C levels exists between individuals with the TT/AA/AA and CC/GG/TT genotypes (P<.0001, Fig 2⇓). Because of this difference and because these genotypes were the only homozygous genotypes that occurred frequently in the tested individuals, we focused our further experiments on individuals with one of these two genotypes.
To confirm or refute the association between genotype and protein C levels (based on protein C activity measurements), we determined the protein C antigen levels of the 28 TT/AA/AA and 40 CC/GG/TT individuals. A 17% difference (P<.0005) was present between the mean protein C antigen levels in the TAA and CGT groups (108% and 91% for the TAA and CGT groups, respectively). Again, no difference was observed between protein C antigen levels in the patients and control subjects. Table 2⇓ shows the number of individuals by low, middle, and high tertile of the protein C antigen level distribution for both the homozygous TAA and CGT genotypes. It shows that half of the individuals with the TAA genotype are in the higher tertile (101% to 166%), while most individuals with the CGT genotype are in the lower tertile (65% to 86%).
Factors known to influence plasma protein C levels in the healthy population, eg, age, sex, obesity, and smoking, did not account for the differences mentioned above (Table 3⇓). The fact that factor II and factor X levels (two other vitamin K–dependent coagulation factors) did not differ between the two groups also excluded liver function abnormalities as a likely cause of the difference in mean protein C levels.
Risk Factor Analysis
To ascertain whether the genotype with low protein C levels (CC/GG/TT) is associated with a higher risk for thrombosis than the genotype with higher protein C levels (TT/AA/AA), we compared the number of patients and control subjects with the TT/AA/AA and CC/GG/TT genotypes. We found that 34% (n=13) had the TT/AA/AA genotype, and 66% (n=25) had the CC/GG/TT genotype. In the control group, 50% (n=15) had the TT/AA/AA genotype, and 50% (n=15) had the CC/GG/TT genotype. This indicates that having the homozygous CGT genotype yields an OR of 1.9 (95% CI, 0.7 to 5.0) compared with the homozygous TAA genotype. To narrow this CI, we genotyped the other 698 of the 948 individuals and recalculated the OR. After genotyping all these individuals, we had 51 patients (35%) and 66 control subjects (45%) with the TT/AA/AA genotype and 97 patients (65%) and 81 control subjects (55%) with the CC/GG/TT genotype, which leads to an OR of 1.6 (95% CI, 1.0 to 2.5). Therefore, individuals with the homozygous CGT genotype have a 50% to 100% greater chance of developing venous thrombosis than individuals with the homozygous TAA genotype.
The aim of this study was to evaluate the effect of genotypic variation at the protein C locus on plasma protein C levels and to assess the relation with thrombosis. This answers the question about whether genetically determined low protein C levels, in the absence of protein C deficiency per se, are associated with higher thrombotic risks than genetically determined high protein C levels.
The initially tested group (n=240) consisted of both patients (n=130) and control subjects (n=110). In principle, it might have been better to test only control subjects. However, when we analyzed the obtained data, we did not observe a difference between patients and control subjects concerning both the mean (Table 1⇑) and the distribution of the protein C activity levels (Fig 3⇓). This encouraged us to pool all data, regardless of whether they were from patients or control subjects.
A clear correlation between protein C promoter genotype and protein C plasma levels is shown in the 240 individuals. The mean protein C activity level of individuals with the homozygous CGT genotype is about 22% less than that of individuals with the homozygous TAA genotype.
The association between the genotype of the promoter region of the protein C gene and plasma protein C levels can be explained in several ways. First, the polymorphisms determining the genotype could affect the affinity of nuclear proteins involved in the regulation of transcription. Second, the genotype of these polymorphisms could be in linkage disequilibrium with functionally important sequences close to the protein C gene. The genotype probably is not linked to a functional sequence in the protein C gene itself because in the healthy population no amino acid substitutions are known to occur in the coding region of the protein C gene.21
The results of this study clearly indicate that the variation in the normal distribution of plasma protein C levels is due partly to variations in the protein C gene. This genetic variation, however, is only one of a number of constitutional, lifestyle, or biochemical factors influencing the normal distribution of protein C. Young age, male sex, black race, leanness, and low LDL cholesterol and triglyceride levels contribute to lower protein C levels.10 11 Use of oral contraceptives, on the other hand, raises the protein C level.11 Smoking might also lower protein C levels,10 although Tait et al11 found no influence of smoking.
An individual’s protein C level is a consequence of the combined action of all the above-mentioned factors. Therefore, it is of no direct clinical relevance to determine the protein C genotype routinely. In individual cases, however, the genotype might be taken into account. For instance, one should be careful in diagnosing young boys with protein C levels slightly below the 2-SD range around the mean as protein C deficient. If these subjects, who have lower protein C levels because of their young age, have the homozygous CGT genotype, their protein C levels might be below the 2-SD range, although the subjects are not protein C deficient.
It is well established that individuals with low protein C levels caused by a protein C deficiency are at risk for thrombosis.7 8 Our results indicate that individuals without protein C deficiency but with low protein C levels also have higher thrombotic risks than individuals with high protein C levels. This finding not only illustrates the important anticoagulant activity of protein C but also shows that quite subtle changes in factors of the protein C anticoagulant system, in this case genetically determined, can have substantial consequences.
This work was supported by grant 92.004 from the Trombosestichting Nederland. The original LETS was supported by the Netherlands Heart Foundation (grant 89.063).
Walker FJ. Regulation of activated protein C by a new protein. J Biol Chem. 1980;255:5521-5524.
Walker FJ, Fay PJ. Regulation of blood coagulation by the protein C system. FASEB J. 1992;6:2561-2567.
Stenflo J. A new vitamin K-dependent protein: purification from bovine plasma and preliminary characterization. J Biol Chem. 1976;251:355-363.
Van Hinsberg VWM, Bertina RM, van Wijngaarden A, van Tilburg NH, Emeis JJ, Haverkate F. Activated protein C decreases plasminogen activator inhibitor activity in endothelial cell conditioned media. Blood. 1985;65:1914-1920.
Seligsohn U, Berger A, Abend M, Rubin L, Attias D, Zivilin A, Rapaport SI. Homozygous protein C deficiency manifested by massive venous thrombosis in the newborn. N Engl J Med. 1984; 310:559-562.
Marlar RA, Montgomery RR, Broekmans AW. Diagnosis and treatment of homozygous protein C deficiency. J Pediatr. 1989; 114:528-534.
Griffin JH, Evatt B, Zimmerman TS, Kleiss AJ, Wideman C. Deficiency of protein C in congenital thrombotic disease. J Clin Invest. 1981;68:1370-1373.
Green F, Kelleher C, Wilkes H, Temple A, Meade T, Humphries S. A common genetic polymorphism associated with lower coagulation factor VII levels in healthy individuals. Arterioscler Thromb. 1991;11:540-546.
Dawson S, Hamsten A, Wiman B, Henny A, Humphries S. Genetic variation at the plasminogen activator inhibitor-1 locus is associated with altered levels of plasma plasminogen activator inhibitor-1 activity. Arterioscler Thromb. 1991;11:183-190.
Woolf B. On estimating the relation between blood group and disease. Am J Hum Genet. 1955;19:251-253.