based candidate gene approach
Uitte de Willige, S.
Citation
Uitte de Willige, S. (2007, May 23). Novel genetic risk factors for venous
thrombosis; a haplotype-based candidate gene approach. Hemostasis and
Thrombosis Research Center, Department of Hematology, Faculty of Medicine,
Leiden University. Retrieved from https://hdl.handle.net/1887/11970
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis
in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/11970
Note: To cite this publication please use the final published version (if
applicable).
Chapter 5
General Discussion and Summary
Venous thrombosis is a common disease, which occurs mostly in the deep veins of the leg. The overall incidence of venous thrombosis increases with age.
Approximately 1-3 per 1000 individuals per year are affected in the Western world,1-
4 making it an important health issue. Venous thrombosis clusters within families and is a multicausal disease, in which both genetic and environmental factors interact in the onset of the disease.5-8 The investigation (of families) of thrombophilic patients has led to the identification of several genetic risk factors, such as deficiencies of anticoagulant proteins (antithrombin,9 protein C10,11 and protein S12), activated protein C (APC)-resistance associated with the factor V Leiden mutation13,14 and the prothrombin 20210A mutation.15 Still, underlying genetic risk factors could not be found in a substantial number of patients with a family history of venous thrombosis,16 despite the belief that thrombophilia families carry multiple defects.17 This suggested that there must be genetic determinants of venous thrombosis that have not yet been identified.
Aim and approach of this thesis
The aim of the study described in this thesis was to find new genetic risk factors for deep venous thrombosis. The key objectives were to identify genes that contribute to the genetic susceptibility to venous thrombosis and to identify the functional (i.e.
disease causing) mutations in these genes. With the completion of the human genome sequence18,19 and the technological revolution in genotyping,20 many projects aim to understand the genetic contribution to common diseases by identifying sequence variants associated with these disorders. The motivation to study the genetic contribution to venous thrombosis is twofold. First, the identification of new critical pathways in disease pathogenesis may lead to the identification of new drug targets. Second, the increased understanding of disease pathogenesis will lead to better and more focused treatment of patients. Prediction of disease outcome by genetic risk factors may lead to more individualized treatment protocols. Single genetic risk factors that confer small risks by themselves can become more important in the presence of additional risk factors, either genetic, phenotypic (quantitative) or environmental. For venous thrombosis this has been shown previously in subjects with multiple genetic defects,21,22 in a large Spanish family-based study23 and in women who carried the Factor V Leiden mutation and used oral contraceptives.24,25
To identify new genetic risk factors for venous thrombosis, we used a haplotype- based candidate gene approach. This approach has been shown to be successful in a previous study on SNPs and haplotypes of candidate gene coagulation factor V.26 The haplotype carrying the most common genetic risk factor for venous thrombosis, the Factor V Leiden mutation, was clearly more frequent in patients compared with control subjects. These findings suggested that a polymorphism in or near the factor V gene in this haplotype was associated with an increased risk of venous thrombosis. If the factor V gene had been sequenced in patients homozygous for this haplotype in order to locate the causal variation, the Factor V Leiden mutation would have been found.
The main hypothesis of the haplotype-based approach of this thesis was that relatively common functional variants exist that are the product of unique mutational events in a founder haplotype (the combination of marker alleles on a chromosome) and that the frequency of such haplotypes will be increased in a patient population. Our main source of haplotypes and haplotype-tagging SNPs was the public database of SeattleSNPs, a research group that focuses on identifying, genotyping, and modelling the associations between SNPs in candidate genes and pathways that underlie inflammatory responses in humans.27 By first examining all the common haplotypes of a candidate gene, and subsequently genotyping the smallest set of haplotype-tagging SNPs, we made use of the linkage disequilibrium (LD, the non-random association of nearby variants) that is present in the human genome. In this way, after finding an association, either the typed tagSNP or one of its proxies (SNP in LD with the typed tagSNP) may be the disease causing variant.
This approach has an advantage above testing single SNPs which are common in the population or that have a predictable function (e.g. non-synonymous or located in the promoter), because lack of association with a single candidate SNP does not rule out that nearby SNPs are functionally important. A limitation of the haplotype-based approach is that it is unlikely to find any rare variations or variations that lead to
‘loss of function’, which is most likely reflected by heterogeneous variations in the same gene.
In this thesis, candidate genes were selected on the basis of theoretical knowledge of the proteins encoded by the genes. This choice of a candidate gene has subjective elements, because the genes that are intuitively logical will be tested first. Since the biochemistry of blood coagulation has been well defined,28-30 there
are many potential candidate genes known that might confer a risk for deep venous thrombosis, although several of these potentials have already been thoroughly investigated (reviewed in refs 31 and 32). In addition, it is probable that pathways and proteins outside the coagulation system are involved in the pathogenesis of venous thrombosis.
Endothelial cell Protein C Receptor (EPCR)
The first candidate gene that we investigated was the Endothelial cell Protein C Receptor (EPCR) (Chapter 2). EPCR functions as an important regulator of the protein C anticoagulant pathway by binding protein C and enhancing activation by the thrombin-thrombomodulin complex.33 EPCR binds to both protein C and activated protein C (APC) with high affinity.34 In contrast to the membrane-bound form, a soluble form of EPCR (sEPCR) circulates in the plasma which inhibits APC anticoagulant activity.33,34 sEPCR does not enhance APC generation, since it blocks the binding of (A)PC to phospholipid membranes and thereby eliminates the ability of APC to inactivate activated factor V.35 For this thesis, EPCR was an attractive candidate gene, since it is a small gene of only four exons with a clear haplotype structure and without recombination within the gene. In addition, other abnormalities in the protein C pathway have been found that increase the risk of venous thrombosis, such as deficiencies of protein C10,11 and protein S,12 and low levels of circulating APC.36
It was previously reported that sEPCR levels were distributed bimodal in healthy populations,37,38 and a French study reported that this bimodal distribution was genetically controlled by haplotype A3 of the EPCR gene.39 In addition, in the French study population haplotype A3 was associated with an increased risk for venous thrombosis.39 To replicate these findings we assessed whether EPCR haplotypes or sEPCR levels were associated with the risk of deep venous thrombosis in a large case-control study, the Leiden Thrombophilia Study (LETS).40,41 In contrast to the three haplotypes (A1, A2 and A3) detected by the French study,39 SeattleSNPs reported a fourth common haplotype (H4).27 The other three haplotypes (H1, H2 and H3) were the same as A1, A2 and A3. All subjects of the LETS were genotyped for three haplotype-tagging SNPs, which enabled us to detect all four common haplotypes of the EPCR gene. We did not find a strong association between any of the EPCR haplotypes and thrombosis risk. We observed a 40% increase in risk for carriers of H4. This increase in risk will need confirmation in other studies. H1 was
previously found to be associated with a decreased risk of venous thrombosis and with increased levels of circulating activated protein C.42 We found only a slight reduction in risk.
Most of the studies on genetic EPCR variants focused on H3, the haplotype that previously was found to be associated with an increased risk of venous thrombosis and genetically controls sEPCR levels. In our study population, H3 increased the risk only slightly, but not significantly, while a Spanish study even found a slight decrease in risk.43 These discrepancies probably are related to the differences in allele frequency for H3 found in the different study populations, with an allele frequency gradient decreasing from the north to the south of Europe. In addition, both the French and the Spanish study did not include H4. Our study in a population of Dutch origin reported allele frequencies of 0.127 in the control population and 0.144 in the patient population. The French study found almost the same allele frequency in their patient population (0.138), while the frequency in their control population was lower (0.092), explaining the risk they found for haplotype A3. The Spanish study reported allele frequencies of 0.092 and 0.094 for their control and patient populations respectively, very similar to the allele frequencies of the French control population. A gradient across Europe has also been reported for the Factor V Leiden mutation, except that for this mutation the frequency is highest in the south of Europe.44
We replicated the finding that sEPCR levels are genetically controlled by H3 and in our control population this haplotype explained 86.5% of the variation in sEPCR levels. We observed a trimodal distribution of sEPCR levels, since in our control population we found ten control subjects who were homozygous for H3, having the highest sEPCR levels. Since the other studies had smaller numbers of subjects, only a few or even no H3 homozygotes were found, which made the sEPCR distribution bimodal.
It was hypothesised that the H3 tagging A>G variation at position 6936 in intron 4 of the EPCR gene is responsible for the elevated sEPCR levels seen in H3 carriers.39 The 6936A>G variation results in an amino acid change (Ser219Gly) in the transmembrane region of the receptor,45 and may introduce a conformational change in the molecule, which renders the molecule more susceptible to metalloproteolytic cleavage. Since residue 218 is also a glycine, these two adjacent
glycine residues may destabilize the helical transmembrane domain and thereby change the exposure of the cleavage site. This hypothesis has been explored in vitro and it was demonstrated that, at the tissue culture level, EPCR shedding is increased in cells bearing Gly219 compared to Ser219, explaining the elevated sEPCR levels in H3 carriers.46
We additionally assessed whether sEPCR levels were associated with the risk of venous thrombosis. We used quartiles, as measured in the control subjects, and found that compared to the first quartile, the risk associated with the other three quartiles was increased, but without evidence of a dose-response relationship.
Because of the absence of a gradient of risk over the levels of sEPCR it was unlikely that the observed increase in risk was a true effect. This may also explain why in our study population H3, which was strongly associated with elevated sEPCR levels, did not increase the risk of venous thrombosis. Expressing the results of these analyses in terms of risk reduction, low sEPCR levels (<81 ng/ml, P25) reduced the risk of venous thrombosis. These findings are in agreement with two studies on the effects of membrane EPCR and soluble EPCR on the hemostatic balance and endotoxemia in mice. It was found that transgenic mice with 11-fold elevated sEPCR levels did not exhibit any gross thrombotic or bleeding abnormalities and had normal growth and viability.47 Furthermore, elevated physiologically relevant levels of sEPCR did not measurably inhibit protein C activation or enhance coagulation.48
In conclusion, in our study population none of the haplotypes of the EPCR gene were associated with a strong thrombosis risk, but low sEPCR levels reduced the risk of deep venous thrombosis.
Fibrinogen
The three genes of the fibrinogen cluster (fibrinogen alpha (FGA), beta (FGB) and gamma (FGG)) were explored as candidate genes in Chapter 3. Fibrinogen is an abundant plasma protein that plays an important role in the hemostatic system, being the precursor of fibrin, the end product of blood coagulation.49-51 Of all the components of the coagulation system, elevated fibrinogen levels have been most consistently associated with vascular disorders,52-55 although they also have been considered a disease marker rather than a causal agent.56,57 However, recent findings suggest that there really is a cause-effect relationship between elevated fibrinogen levels and the risk of venous thrombosis. It has been reported that there
is an independent inverse association between circulating APC levels and fibrinogen levels.58 Protein C is activated by the thrombin-thrombomodulin complex, and thrombin binds to fibrinogen and thrombomodulin through a common region, anion- binding exosite-1.59 Therefore, fibrinogen could prevent the formation of thrombin- thrombomodulin complexes, thus impairing the generation of APC. These findings are consistent with previous observations in which fibrinogen was shown to hamper protein C activation by competitively inhibiting the binding of thrombin to thrombomodulin in solution.60
In total, fifteen haplotype-tagging SNPs were genotyped to catch all common haplotypes of the three fibrinogen genes (Chapter 3.1). In contrast to previous findings of effects of FGB promoter polymorphisms on fibrinogen concentration,61 we found no association of any of the fibrinogen haplotypes with total fibrinogen levels.
Although it has been reported that genetic factors contribute to the variation in plasma fibrinogen levels (heritability, h2 = 0.2-0.5),62-64 our findings are supported by the finding that no linkage was found between the fibrinogen locus and plasma fibrinogen levels.65,66
In each of the fibrinogen genes, homozygosity for haplotype 2 (H2) increased the risk of thrombosis approximately 2-fold. After adjustment for linkage disequilibrium between the three genes, only FGG-H2 homozygosity remained associated with thrombosis risk. The risk associated with FGB-H2 and FGA-H2 disappeared, which made us focus on FGG-H2. Previously, the 312Ala allele of a coding polymorphism in FGA (Thr312Ala), located in a region important for FXIIIa-dependent cross-linking at position328,67 was associated with more rigid and less porous fibrin gel structures.68 Ala312 fibrin clots had thicker fibers and more extensive α-chain cross-linking than Thr312 clots.68 These results suggested that Ala312 fibrin clots are more resistant to fibrinolysis, which might increase the thrombotic risk.53 However, because the Ala312 allele is located on FGA-H2, which is strongly linked to FGG-H2, it is very well possible that the risk reported to be associated with this polymorphism,69 is in fact due to linkage with FGG-H2.
Since FGG-H2 was associated with an increased thrombosis risk, but not with total fibrinogen levels, we reasoned that this haplotype should confer a qualitative defect.
Since none of the five FGG-H2 tagging SNPs changed the amino acid sequence, we sequenced the genes of ten thrombosis patients homozygous for FGG-H2 (20 FGG-
H2 alleles), including the promoter, 5' UTR, exons, intron/exon boundaries and 3' UTR, to look for the presence of additional variations in the coding region of the FGG gene, but no novel sequence variations were found. This indicated that most likely one of the FGG-H2 tagging SNPs itself was the risk-enhancing SNP.
Of all FGG pre-mRNA, 7-15% is alternatively spliced, resulting in an mRNA encoding the fibrinogen γ' chain, instead of the normal γA chain.70-73 We hypothesized that the efficiency of alternative splicing of the FGG pre-mRNA is influenced by SNP 9615C>T or SNP 10034C>T, since these two SNPs are located near the polyadenylation sites of the fibrinogen γ' and γA transcripts, respectively, and therefore could alter fibrinogen γ' expression. Therefore, we developed an ELISA for the measurement of fibrinogen γ' levels in plasma and measured fibrinogen γ' levels (i.e. γA/γ' plus γ'/γ') in all subjects of LETS. We found that risk haplotype FGG-H2 was strongly associated with reduced fibrinogen γ' levels and even more strongly with a reduced fibrinogen γ'/total fibrinogen ratio, with a clear allele-specific and dose-dependent effect of the FGG-H2 haplotype on both parameters. Additionally, we found that a reduced fibrinogen γ'/total fibrinogen ratio increased the risk of thrombosis more than 2-fold. These findings were in contrast to studies of the effect of fibrinogen γ' levels on the risk of arterial disease,74,75 and to predictions that an increase in fibrinogen γ' or in the fibrinogen γ'/total fibrinogen ratio would result in more stable and lysis-resistant clots and therefore would represent a prothrombotic state.
Several studies have reported that fibrinogen γA/γ' heterodimers behave differently from fibrinogen γA/γA homodimers during the early stages of fibrin polymerization, ultimately leading to an altered fibrin structure, which is more extensively cross- linked by activated factor XIII than γA/γA fibrin and also more resistant to fibrinolysis.76-78 This has been explained by the presence of a unique binding site for factor XIIIB in the carboxyterminus of the γ' chain.79,80 A possible explanation for this contradiction may be that the alternative carboxyterminus of the γ' chain defines not only a binding site for factor XIII but also a high-affinity non-substrate binding site for thrombin.77,78,81 Studies on γA/γ' fibrin(ogen) indicated that this binding site functioned as a thrombin inhibitor82,83 and importantly contributed to the antithrombin activity, which develops during fibrin formation (antithrombin I).84,85 FGG-H2 homozygosity therefore may be considered to confer a partial antithrombin I deficiency.
The association of FGG-H2 with reduced fibrinogen γ' levels suggested that a mutation present in the FGG-H2 pre-mRNA may be responsible for a reduced efficiency in the formation of the alternatively spliced γ' chain. After inspection of the SNPs present in this haplotype, we observed that 10034C>T is located in a GT-rich region (GGTA[C/T]CTTTATTGACCAT at nucleotides 10030-10047) just downstream from the second polyadenylation (pA2) site at nt 9997-10002 of the fibrinogen γA specific exon 10. Actually, this region shows a 78% match with the Cleavage stimulation Factor (CstF) binding site consensus 2a sequence.86 CstF is a multi- subunit complex required for efficient cleavage and polyadenylation of pre-mRNAs, which initially extend several hundred nucleotides beyond the ultimate polyadenylation site.86 It binds via its CstF-64 subunit to a G/U-rich downstream sequence element (DSE) and stabilizes the binding of the cleavage and polyadenylation specificity factor (CPSF) to the pA-signal.87,88 In FGG, the pA2- signal is used for the formation of γA mRNA in which intron 9 has been removed, whereas the pA1-signal, which is located in intron 9 at nucleotides 9558-9563, is used for the formation of γ' mRNA. In FGG-H2, which contains a T at nucleotide 10034, the DSE is strengthened (more GT-rich) compared to the other common FGG haplotypes (H1, H3, and H4), which all have a C at position 10034. The affinity of the CstF-64/DSE interaction is an important determinant of the relative strength of competing poly(A)-sites. Therefore we hypothesized that the 10034C>T change will increase the affinity of the DSE/CstF-64 interaction resulting in more frequent use of pA2 (more γA transcripts) at the expense of polyadenylation at pA1 (less γ' transcripts). FGG-H2 is therefore expected to produce relatively more γA transcripts and subsequently relatively less γ' transcripts. This would explain the reduced fibrinogen γ' levels and fibrinogen γ'/total fibrinogen ratios previously observed in homozygous carriers of FGG-H2 (Chapter 3.1). A second FGG-H2 specific polymorphism, 9615C>T [rs2066864], is located at a position downstream from pA1 in intron 9, two nucleotides downstream of a DSE homologous to CstF binding site consensus 2b,86 and might therefore also influence the relative use of the two pA- signals of the FGG transcript.
To investigate this hypothesis and the importance of the DSE downstream of pA2 (containing 10034C>T) for the regulation of pA1/pA2 usage, we transiently transfected FGG mini-gene constructs containing exon 9, intron 9 (containing pA1, the polyadenylation signal used for fibrinogen γ' formation), exon 10 and the 3' region (containing pA2, the polyadenylation signal used for fibrinogen γA formation)
in liver-derived HepG2 cells and used quantitative real-time PCR to measure the mean relative pA-signal usage (pA1/pA2-ratio) of the different constructs (Chapter 3.2). Compared to the pA1/pA2-ratio of the reference construct (9615C-10034C;
FGG-H1), the ratios of construct TT (9615T-10034T; FGG-H2) and construct CT (9615C-10034T) were decreased, while the ratio of construct TC (9615T-10034C) did not differ significantly from the reference construct. These results indicated that in our in vitro system SNP 10034C>T, and not SNP 9615C>T, was the major contributor to the reduction of the pA1/pA2-ratio. Compared to a reference construct, in which the DSE was still present, the pA1/pA2-ratio of a DSE deletion construct, missing the DSE was 2.2-fold increased, indicating that this site is involved in the regulation of pA2-usage. The functionality of the DSE was further confirmed using several mutant constructs in which the DSE was weakened or strengthened (less or more GT-rich) by introducing one or two point mutations.
Weakening the DSE resulted in a significant increase of the pA1/pA2-ratio, while strengthening the DSE resulted in a significant decrease of the ratio.
A potential weakness of our model was that we were unable to compare the absolute use of the separate pA-signals between constructs. By using cotransfection with a reporter construct this would have been possible. Instead we focused on the ratio by which the two polyadenylation sites were used. This should not depend on the transfection efficiency, because for each construct two transcripts are produced from the same template (“γA” from pA2-use and “γ'” from pA1-use). In our mini- gene model we found that 10034T reduces the pA1/pA2-ratio, and since this allele strengthens the putative CstF binding site, we concluded that pA2 is more frequently used. We cannot distinguish whether this is at the expense of polyadenylation at pA1 or not. Both scenarios would decrease the pA1/pA2-ratio.
However, since we previously only found a correlation of FGG-H2 with reduced fibrinogen γ'/total fibrinogen ratios, and not with plasma total fibrinogen levels (Chapter 3.1), we assumed that protein production and therefore total fibrinogen gamma mRNA synthesis does not change significantly, suggesting that the 10034T allele is responsible for a shift from pA1-usage to pA2-usage.
From the transfection results we concluded that 10034C>T is located in a GT-rich DSE involved in regulating the usage of the pA2-signal of FGG, which may represent a CstF binding site. Therefore, we propose that the 10034C>T change is the functional variation in the FGG-H2 haplotype and is responsible for the reduction in
the fibrinogen γ'/total fibrinogen ratio and the increased risk of deep venous thrombosis.
The genetic determinants of venous thrombosis in the African-American population are poorly characterized. The two most important genetic conditions predisposing to venous thrombosis among the Caucasian population, the Factor V Leiden mutation13 and the prothrombin G20210A mutation,15 are rare in the African-American population.89-91 Despite the rarity of known genetic risk factors among African- Americans, it has been found that the incidence of idiopathic venous thrombosis and the prevalence of family history of venous thromboembolism (VTE) is similar for African-American and Caucasian cases, suggesting that also among African- Americans, a strong genetic component contributes to the etiology of VTE.92,93 It has been reported that the minor allele frequency of SNP 10034C>T is higher in healthy African-American subjects compared to healthy Caucasian subjects.27,94 Compared to healthy Caucasian subjects, the minor allele frequency of SNP 9340T>C, which is tagging for FGG-H3 and was previously associated with a slight reduction in venous thrombosis risk, was reported to be increased in the African-American study population of SeattleSNPs,27 but decreased in the African population of HapMap.94 In Chapter 3.3 we investigated whether FGG 3'-end SNPs were associated with the risk of VTE in the African-American population and aimed to replicate the association of SNP 10034C>T with the risk of venous thrombosis in the Caucasian population. For this we used a large American case-control study on risk factors for venous thromboembolism, the Genetic Attributes and Thrombosis Epidemiology (GATE) study, which includes subjects of both the African-American and the Caucasian population.93
In the African-American population, SNPs 9340T>C and 10034C>T did not clearly influence the risk of VTE. In the Caucasian population SNP 9340T>C was associated with a slight decrease in risk, while homozygosity for SNP 10034C>T was associated with a non-significant 1.6-fold increase in risk. To compare the GATE results with the LETS results (Chapter 3.1), we additionally analyzed the risk of patients with a first event of DVT (+PE), without having a malignancy. In African-American subjects, the risk associated with SNP 9340T>C slightly increased, while SNP 10034C>T still showed no effect on risk. In Caucasian subjects, the influence of both SNPs did not change importantly, the risk for SNP 9340T>C still being slightly decreased and the risk for SNP 10034C>T still being non-significantly increased.
Finally, we analyzed the risk of VTE in idiopathic cases, since especially in the African-American population, there were many provoked events. These analyses showed that in both African-Americans and Caucasians SNP 9340T>C was associated with a slight reduction in VTE risk. Homozygosity for SNP 10034C>T was associated with a 1.5-fold increased risk of VTE in African-Americans, and with a 2.1-fold increase in Caucasians. This risk only reached statistical significance in the Caucasian population. For the Caucasian population these results are comparable with our findings in LETS, in which the risk for idiopathic homozygous carriers of FGG-H2, tagged by SNP 10034C>T, was 2.2-fold increased.
Since there are three more SNPs reported to be present in the 3'-end of the FGG gene in the African-American population (9595C>T, 9615C>T and 9765T>A),27 we additionally sequenced the 3'-end of the FGG gene in all subject of the GATE study.
This resulted in the identification of an additional rare variation at position 9937, conferring a C>T change, which was mainly present in the African-American population. For SNPs 9595C>T, 9765T>A and 9937C>T there were no differences in allele frequencies between cases and controls. For SNP 9615C>T, similar results as for SNP 10034C>T were obtained.
Additionally, we analyzed the association of the SNPs of the 3'-end of FGG with total fibrinogen levels in control subjects of both populations. Total fibrinogen levels were higher in African-American control subjects compared to Caucasian control subjects, while in neither races none of the FGG 3'-end SNPs were associated with total fibrinogen levels. This is in accordance with our findings in LETS (Chapter 3.1).
In the GATE study we found some recombination between SNPs 9615C>T and 10034C>T, while according to the data of SeattleSNPs, the rare T-alleles of SNPs 9615C>T and 10034C>T are completely linked, both in the subjects of European descent and in the subjects of African-American descent.27 This was also seen in LETS (Chapter 3.1). HapMap reported recombination between these two SNPs in only one of 228 African-American alleles.94 Unfortunately, the number of subjects with recombination between 9615C>T and 10034C>T in the GATE study was too low to find any significant differences between the two SNPs in their association with VTE.
We observed several important differences between the two ethnic populations.
Among patients, the sex ratio is different for African-Americans and Caucasians, since about 60% of African-American patients are female, whereas about 40% of Caucasian patients are female. Obesity is a risk factor for VTE among Caucasians, but it is not among African-Americans. Furthermore, African-American patients have slightly more first events and idiopathic events, while malignancies have been more frequently reported in Caucasian patients. In addition, the risks of the 10034T allele appear to be different in both populations. All this suggests that risks might be very different between populations. This is most likely not caused by different molecular mechanisms, but by different interactions among the risk factors, either genetic or environmental. What we perceive as the risk of the 10034T allele in the Caucasian population is in reality the crude risk of this allele in the genetic and environmental background of the Caucasian population. It is obvious that the effect of the 10034T allele might be affected by variations in the quality or quantity of splicing and polyadenylation factors. These variations might be different between ethnical groups. In addition, environmental risk factors will also be different between the ethnical groups, even though they usually will not differ between genotype-groups within a given population.
Since elevated levels of total fibrinogen have also been associated with an increased risk of arterial thrombosis, we additionally studied the association of the four common haplotypes of the FGG gene with the risk of myocardial infarction and with total fibrinogen levels in the 'Study of Myocardial Infarctions Leiden' (SMILE) (Chapter 3.4), and determined the role of fibrinogen γ' and FGG haplotypes in ischemic stroke in the COntrolled study of genetically determined COagulation disorders in patients with transient ischemic attack or ischemic Stroke (COCOS) (Chapter 3.5).
In both studies we did not find any association of FGG haplotypes with total fibrinogen levels, which is in accordance with our previous findings in LETS (Chapter 3.1) and the GATE study (Chapter 3.3). In SMILE none of the haplotypes was significantly associated with the risk of myocardial infarction, while in COCOS FGG- H3 carriers had a reduced risk of ischemic stroke. Mannila et al previously observed a similar association between FGG-H3 (called SNP2 in their study) and the risk of myocardial infarction,95 while in the SMILE we only found a slight decrease in risk for this haplotype in the subgroup of individuals less than 50 years of age.
In COCOS fibrinogen γ' levels were measured in 114 patients (samples collected during the acute phase) and in 120 healthy controls. In an unselected group of 48 patients, fibrinogen γ' levels were also measured in the samples collected at 3 months after the event (convalescent phase). We found that compared to the control subjects, the fibrinogen γ'/total fibrinogen ratio was increased in stroke patients in the acute phase, which may represent an antithrombotic defense mechanism of the human body. On the other hand, the fibrinogen γ'/total fibrinogen ratio was decreased in stroke patients in the convalescent phase compared to the control subjects. These findings suggest that there may be a different regulation of alternative pre-mRNA processing in the acute and the convalescent phase of stroke.
Such an effect has also been suggested for other genes.96-98 As previously suggested, it may also be that the clearance of fibrinogen γA and γ' from the circulation is different during various stages.74
Additionally, in COCOS we examined the association between FGG haplotypes and the fibrinogen γ'/total fibrinogen ratio. In all three subgroups, FGG-H2 was associated with a decreased fibrinogen γ'/total fibrinogen ratio as we found previously (Chapter 3.1), while there was no association of this haplotype with risk of stroke. None of the other haplotypes was significantly associated with the fibrinogen γ'/total fibrinogen ratio, although in all three subgroups a slight increase was observed for FGG-H3 carriers as previously observed in LETS (Chapter 3.1).
Selectins
The genes of the selectin family (E-selectin (SELE), L-selectin (SELL) and P-selectin (SELP)) and their most important counter-receptor P-selectin ligand (PSGL-1) were explored as candidate genes in Chapter 4. The selectins are a family of type I membrane proteins that facilitate the interaction of platelets, leukocytes and endothelial cells at inflammatory sites.99,100 All four genes contain many SNPs and there are recombination hotspots present within the SELP and PSGL-1 genes.27,94 The majority of studies on polymorphisms or haplotypes of the selectin genes focused on arterial disease and did not include the entire gene cluster.101-107
In the four genes, a total of 24 SNPs were genotyped to catch all common haplotype groups of the genes. Selection of the haplotype-tagging SNPs was a little harder compared to EPCR and fibrinogen, since all four genes contain many SNPs and there are recombination hotspots present within the SELP and PSGL-1 genes. Because of
the presence of a recombination hotspot in intron 8 of SELP, we analyzed the regions upstream (SELPup) and downstream (SELPdown) from the recombination hotspot separately. Because of the recombination hotspots, there were a lot of rare haplotypes present in the LETS population. We used the program TagSNPs (Version 2),108 which minimizes the uncertainty in predicting common haplotypes for individuals with unphased genotype data, to calculate the number of haplotypes for each gene, the frequency of the haplotypes and the statistic R2h, the squared correlation between true haplotype dosage (0, 1 or 2 copies of a haplotype) and the haplotype dosage predicted by TagSNPs.109 A high R2h indicates that a haplotype could be assigned with high certainty. An R2h>0.95 and overall haplotype frequency
≥1% were used as criteria for assigning haplotypes of SELE, SELL and SELP.
Because a recombination hotspot covers part of the PSGL-1 gene, haplotype construction for this gene was less accurate. Consequently, no haplotypes were constructed for PSGL-1, and the SNPs were analyzed separately. None of the PSGL- 1 SNPs was significantly associated with thrombosis risk.
In SELE and SELL none of the haplotypes was associated with the risk of venous thrombosis. In SELPup, H2-carriers had a slight increase in risk, whereas H4- carriers had a slight decrease in risk. In SELPdown, H2-carriers had a slight increase in risk, whereas H6-carriers had a slight decrease in risk, although this decrease was not significant.
Since the gene for coagulation Factor V is located 2.4 kb downstream of SELP, and the former gene harbors the most important genetic risk factor for venous thrombosis, the Factor V Leiden mutation (FVL),13 we additionally explored the degree of linkage disequilibrium between the selectin genes and FVL. We observed linkage between FVL and SNPs in all three selectin genes. Because of this high degree of linkage with FVL, a risk association of a selectin haplotype may reflect the effect of FVL on thrombosis risk (OR=8.0, 95%CI: 4.5-14.2 in LETS). Therefore we adjusted the risk estimates for this linkage by excluding all FVL carriers from the analyses and found that the risk associations for SELPup H4 and SELPdown H6 slightly increased as a result of linkage of the common allele of this haplotype to FVL. The risk associations for SELPup H2 and SELPdown H2 completely disappeared, showing that the increased risk associations were indeed a reflection of the effect of FVL on thrombosis risk. Mutual adjustment in one logistic regression model gave the
same results for the SELP haplotypes as after exclusion of all FVL carriers, whereas the risk of FVL remained.
Since PSGL-1 is the main counter receptor for each of the three selectin genes, we investigated the possibility of interaction between SNPs of PSGL-1 and SELE, SELL, and SELP, but no interactions between SNPs of the different genes were found.
Furthermore, soluble P-selectin levels were measured previously in a subgroup of 89 patients and 126 control subjects.110 Therefore we investigated the association between SELP haplotypes and P-selectin levels in control subjects, but none of the haplotypes was associated with soluble P-selectin levels, which may be due to the low number (n=126) of control subjects in whom P-selectin levels were measured.
Overall remarks
In this study we investigated several candidate genes to find new genetic risk factors for deep venous thrombosis. For this we used a haplotype-based candidate gene approach. The choice of candidate genes, based on theoretical knowledge of the proteins encoded by the genes, is becoming more difficult, since most of the potential candidate genes known from the biochemistry of blood coagulation have already been thoroughly investigated (reviewed in refs 31 and 32). So other methods to find candidate genes are warranted, such as whole genome scans. But we do show that the haplotype-based approach is a good way to find new genetic risk factors. This approach already had been shown successful previously,26 and also in this study we were able to find at least one new genetic risk factor, FGG-H2 SNP 10034C>T, although replication of our results is required.
Conclusion
The results described in this thesis hopefully contribute to a better understanding of the etiology of deep venous thrombosis. The ultimate goal of extending the knowledge on (genetic) risk factors for venous thrombosis is to come to individualized profiling which may lead to better treatment and maybe even prevention of the disease by informing patients on potential risk situations, and to individually treat these patients.
References
1. Anderson FA, Jr., Wheeler HB, Goldberg RJ, Hosmer DW, Patwardhan NA, Jovanovic B, Forcier A, Dalen JE. A population-based perspective of the hospital incidence and case- fatality rates of deep vein thrombosis and pulmonary embolism. The Worcester DVT Study. Arch Intern Med. 1991;151(5):933-938.
2. Nordstrom M, Lindblad B, Bergqvist D, Kjellstrom T. A prospective study of the incidence of deep-vein thrombosis within a defined urban population. J Intern Med.
1992;232(2):155-160.
3. Silverstein MD, Heit JA, Mohr DN, Petterson TM, O'Fallon WM, Melton LJ, III. Trends in the incidence of deep vein thrombosis and pulmonary embolism: a 25-year population- based study. Arch Intern Med. 1998;158(6):585-593.
4. Cushman M, Tsai AW, White RH, Heckbert SR, Rosamond WD, Enright P, Folsom AR.
Deep vein thrombosis and pulmonary embolism in two cohorts: the longitudinal investigation of thromboembolism etiology. Am J Med. 2004;117(1):19-25.
5. Seligsohn U, Zivelin A. Thrombophilia as a multigenic disorder. Thromb Haemost.
1997;78(1):297-301.
6. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet.
1999;353(9159):1167-1173.
7. Bovill EG, Hasstedt SJ, Leppert MF, Long GL. Hereditary thrombophilia as a model for multigenic disease. Thromb Haemost. 1999;82(2):662-666.
8. Zoller B, Garcia dF, Hillarp A, Dahlback B. Thrombophilia as a multigenic disease.
Haematologica. 1999;84(1):59-70.
9. van Boven HH, Lane DA. Antithrombin and its inherited deficiency states. Semin Hematol. 1997;34(3):188-204.
10. Griffin JH, Evatt B, Zimmerman TS, Kleiss AJ, Wideman C. Deficiency of protein C in congenital thrombotic disease. J Clin Invest. 1981;68(5):1370-1373.
11. Allaart CF, Poort SR, Rosendaal FR, Reitsma PH, Bertina RM, Briët E. Increased risk of venous thrombosis in carriers of hereditary protein C deficiency defect. Lancet.
1993;341(8838):134-138.
12. Schwarz HP, Fischer M, Hopmeier P, Batard MA, Griffin JH. Plasma protein S deficiency in familial thrombotic disease. Blood. 1984;64(6):1297-1300.
13. Bertina RM, Koeleman BP, Koster T, Rosendaal FR, Dirven RJ, de Ronde H, van der Velden PA, Reitsma PH. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature. 1994;369(6475):64-67.
14. Dahlbäck B. Inherited resistance to activated protein C, a major basis of venous thrombosis, is caused by deficient anticoagulant cofactor function of factor V.
Haematologica. 1995;80(2 Suppl):102-113.
15. Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3'- untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood. 1996;88(10):3698- 3703.
16. Bertina RM. Genetic approach to thrombophilia. Thromb Haemost. 2001;86(1):92-103.
17. Koeleman BP, Reitsma PH, Bertina RM. Familial thrombophilia: a complex genetic disorder. Semin Hematol. 1997;34(3):256-264.
18. International Human Genome Sequencing Consortium. Finishing the euchromatic sequence of the human genome. Nature. 2004;431(7011):931-945.
19. Gregory SG, Barlow KF, McLay KE, etal. The DNA sequence and biological annotation of human chromosome 1. Nature. 2006;441(7091):315-321.
20. Tsuchihashi Z, Dracopoli NC. Progress in high throughput SNP genotyping methods.
Pharmacogenomics J. 2002;2(2):103-110.
21. Emmerich J, Rosendaal FR, Cattaneo M, Margaglione M, De S, V, Cumming T, Arruda V, Hillarp A, Reny JL. Combined effect of factor V Leiden and prothrombin 20210A on the risk of venous thromboembolism--pooled analysis of 8 case-control studies including 2310 cases and 3204 controls. Study Group for Pooled-Analysis in Venous Thromboembolism. Thromb Haemost. 2001;86(3):809-816.
22. Tirado I, Mateo J, Soria JM, Oliver A, Borrell M, Coll I, Vallve C, Souto JC, Martinez- Sanchez E, Fontcuberta J. Contribution of prothrombin 20210A allele and factor V Leiden mutation to thrombosis risk in thrombophilic families with other hemostatic deficiencies.
Haematologica. 2001;86(11):1200-1208.
23. Souto JC, Almasy L, Borrell M, Blanco-Vaca F, Mateo J, Soria JM, Coll I, Felices R, Stone W, Fontcuberta J, Blangero J. Genetic susceptibility to thrombosis and its relationship to physiological risk factors: the GAIT study. Genetic Analysis of Idiopathic Thrombophilia.
Am J Hum Genet. 2000;67(6):1452-1459.
24. Vandenbroucke JP, Koster T, Briët E, Reitsma PH, Bertina RM, Rosendaal FR. Increased risk of venous thrombosis in oral-contraceptive users who are carriers of factor V Leiden mutation. Lancet. 1994;344(8935):1453-1457.
25. Aznar J, Vaya A, Estelles A, Mira Y, Segui R, Villa P, Ferrando F, Falco C, Corella D, Espana F. Risk of venous thrombosis in carriers of the prothrombin G20210A variant and factor V Leiden and their interaction with oral contraceptives. Haematologica.
2000;85(12):1271-1276.
26. van Hylckama Vlieg A, Sandkuijl LA, Rosendaal FR, Bertina RM, Vos HL. Candidate gene approach in association studies: would the factor V Leiden mutation have been found by this approach? Eur J Hum Genet. 2004;12(6):478-482.
27. Nickerson, D. SeattleSNPs. NHLBI Program for Genomic Applications, UW-FHCRC, Seattle, WA. http://pga.gs.washington.edu.
28. Mann KG. Biochemistry and physiology of blood coagulation. Thromb Haemost.
1999;82(2):165-174.
29. Esmon CT. Regulation of blood coagulation. Biochim Biophys Acta. 2000;1477(1-2):349- 360.
30. Oldenburg J, Schwaab R. Molecular biology of blood coagulation. Semin Thromb Hemost. 2001;27(4):313-324.
31. Franco RF, Reitsma PH. Genetic risk factors of venous thrombosis. Hum Genet.
2001;109(4):369-384.
32. Heit JA. Venous thromboembolism: disease burden, outcomes and risk factors. J Thromb Haemost. 2005;3(8):1611-1617.
33. Stearns-Kurosawa DJ, Kurosawa S, Mollica JS, Ferrell GL, Esmon CT. The endothelial cell protein C receptor augments protein C activation by the thrombin-thrombomodulin complex. Proc Natl Acad Sci U S A. 1996;93(19):10212-10216.
34. Esmon CT, Xu J, Gu JM, Qu D, Laszik Z, Ferrell G, Stearns-Kurosawa DJ, Kurosawa S, Taylor FB, Jr., Esmon NL. Endothelial protein C receptor. Thromb Haemost.
1999;82(2):251-258.
35. Liaw PC, Neuenschwander PF, Smirnov MD, Esmon CT. Mechanisms by which soluble endothelial cell protein C receptor modulates protein C and activated protein C function.
J Biol Chem. 2000;275(8):5447-5452.
36. España F, Vaya A, Mira Y, Medina P, Estelles A, Villa P, Falco C, Aznar J. Low level of circulating activated protein C is a risk factor for venous thromboembolism. Thromb Haemost. 2001;86(6):1368-1373.
37. Stearns-Kurosawa DJ, Swindle K, D'Angelo A, Della VP, Fattorini A, Caron N, Grimaux M, Woodhams B, Kurosawa S. Plasma levels of endothelial protein C receptor respond to anticoagulant treatment. Blood. 2002;99(2):526-530.
38. Stearns-Kurosawa DJ, Burgin C, Parker D, Comp P, Kurosawa S. Bimodal distribution of soluble endothelial protein C receptor levels in healthy populations. J Thromb Haemost.
2003;1(4):855-856.
39. Saposnik B, Reny JL, Gaussem P, Emmerich J, Aiach M, Gandrille S. A haplotype of the EPCR gene is associated with increased plasma levels of sEPCR and is a candidate risk factor for thrombosis. Blood. 2003;103(4):1311-1318.
40. Koster T, Rosendaal FR, de Ronde H, Briët E, Vandenbroucke JP, Bertina RM. Venous thrombosis due to poor anticoagulant response to activated protein C: Leiden Thrombophilia Study. Lancet. 1993;342(8886-8887):1503-1506.
41. van der Meer FJM, Koster T, Vandenbroucke JP, Briët E, Rosendaal FR. The Leiden Thrombophilia Study (LETS). Thromb Haemost. 1997;78(1):631-635.
42. España F, Medina P, Mira Y, Estelles A, Vaya A, Villa P, Royo M, Aznar J, Bertina RM. A new polymorphism in the 3'UTR region of the endothelial protein C receptor associated with increased levels of circulating activated protein c and decreased risk of venous thrombosis. Suppl Thromb Haemost , July 2001(ISSN 0340-6245). 2001;Abstract OC885.
43. Medina P, Navarro S, Estelles A, Vaya A, Woodhams B, Mira Y, Villa P, Migaud-Fressart M, Ferrando F, Aznar J, Bertina RM, Espana F. Contribution of polymorphisms in the endothelial protein C receptor gene to soluble endothelial protein C receptor and
circulating activated protein C levels, and thrombotic risk. Thromb Haemost.
2004;91(5):905-911.
44. Rees DC, Cox M, Clegg JB. World distribution of factor V Leiden. Lancet.
1995;346(8983):1133-1134.
45. Villoutreix BO, Blom AM, Dahlback B. Structural prediction and analysis of endothelial cell protein C/activated protein C receptor. Protein Eng. 1999;12(10):833-840.
46. Qu D, Wang Y, Song Y, Esmon NL, Esmon CT. The Ser219-->Gly dimorphism of the endothelial protein C receptor contributes to the higher soluble protein levels observed in individuals with the A3 haplotype. J Thromb Haemost. 2006;4(1):229-235.
47. Li W, Zheng X, Gu J, Hunter J, Ferrell GL, Lupu F, Esmon NL, Esmon CT. Overexpressing endothelial cell protein C receptor alters the hemostatic balance and protects mice from endotoxin. J Thromb Haemost. 2005;3(7):1351-1359.
48. Zheng X, Li W, Gu JM, Qu D, Ferrell GL, Esmon NL, Esmon CT. Effects of membrane and soluble EPCR on the hemostatic balance and endotoxemia in mice. Blood. 2006.
49. Mosesson MW. The roles of fibrinogen and fibrin in hemostasis and thrombosis. Semin Hematol. 1992;29(3):177-188.
50. Herrick S, Blanc-Brude O, Gray A, Laurent G. Fibrinogen. Int J Biochem Cell Biol.
1999;31(7):741-746.
51. Mosesson MW, Siebenlist KR, Meh DA. The structure and biological features of fibrinogen and fibrin. Ann N Y Acad Sci. 2001;936:11-30.
52. Koster T, Rosendaal FR, Reitsma PH, van der Velden PA, Briët E, Vandenbroucke JP.
Factor VII and fibrinogen levels as risk factors for venous thrombosis. A case-control study of plasma levels and DNA polymorphisms--the Leiden Thrombophilia Study (LETS). Thromb Haemost. 1994;71(6):719-722.
53. Danesh J, Collins R, Appleby P, Peto R. Association of fibrinogen, C-reactive protein, albumin, or leukocyte count with coronary heart disease: meta-analyses of prospective studies. JAMA. 1998;279(18):1477-1482.
54. Kamphuisen PW, Eikenboom JC, Vos HL, Pablo R, Sturk A, Bertina RM, Rosendaal FR.
Increased levels of factor VIII and fibrinogen in patients with venous thrombosis are not caused by acute phase reactions. Thromb Haemost. 1999;81(5):680-683.
55. van Hylckama Vlieg A, Rosendaal FR. High levels of fibrinogen are associated with the risk of deep venous thrombosis mainly in the elderly. J Thromb Haemost.
2003;1(12):2677-2678.
56. Reinhart WH. Fibrinogen--marker or mediator of vascular disease? Vasc Med.
2003;8(3):211-216.
57. Koenig W. Fibrin(ogen) in cardiovascular disease: an update. Thromb Haemost.
2003;89(4):601-609.
58. Diez N, Montes R, Alonso A, Medina P, Navarro S, Espana F, Hermida J. Association of increased fibrinogen concentration with impaired activation of anticoagulant protein C. J Thromb Haemost. 2006;4(2):398-402.
59. Fenton JW, Olson TA, Zabinski MP, Wilner GD. Anion-binding exosite of human alpha- thrombin and fibrin(ogen) recognition. Biochemistry. 1988;27(18):7106-7112.
60. Jakubowski HV, Owen WG. Macromolecular specificity determinants on thrombin for fibrinogen and thrombomodulin. J Biol Chem. 1989;264(19):11117-11121.
61. Humphries SE. Genetic regulation of fibrinogen. Eur Heart J. 1995;16 Suppl A:16-19.
62. de Lange M, Snieder H, Ariëns RAS, Spector TD, Grant PJ. The genetics of haemostasis:
a twin study. Lancet. 2001;357(9250):101-105.
63. Dunn EJ, Ariëns RA, de Lange M, Snieder H, Turney JH, Spector TD, Grant PJ. Genetics of fibrin clot structure: a twin study. Blood. 2004;103(5):1735-1740.
64. Souto JC, Almasy L, Borrell M, Gari M, Martinez E, Mateo J, Stone WH, Blangero J, Fontcuberta J. Genetic determinants of hemostasis phenotypes in Spanish families.
Circulation. 2000;101(13):1546-1551.
65. Yang Q, Tofler GH, Cupples LA, Larson MG, Feng D, Lindpaintner K, Levy D, D'Agostino RB, O'Donnell CJ. A genome-wide search for genes affecting circulating fibrinogen levels in the Framingham Heart Study. Thromb Res. 2003;110(1):57-64.
66. Soria JM, Almasy L, Souto JC, Buil A, Lathrop M, Blangero J, Fontcuberta J. A Genome Search for Genetic Determinants That Influence Plasma Fibrinogen Levels. Arterioscler Thromb Vasc Biol. 2005.
67. Baumann RE, Henschen AH. Human fibrinogen polymorphic site analysis by restriction endonuclease digestion and allele-specific polymerase chain reaction amplification:
identification of polymorphisms at positions A alpha 312 and B beta 448. Blood.
1993;82(7):2117-2124.
68. Standeven KF, Grant PJ, Carter AM, Scheiner T, Weisel JW, Ariëns RA. Functional analysis of the fibrinogen Aalpha Thr312Ala polymorphism: effects on fibrin structure and function. Circulation. 2003;107(18):2326-2330.
69. Carter AM, Catto AJ, Kohler HP, Ariëns RA, Stickland MH, Grant PJ. Alpha-fibrinogen Thr312Ala polymorphism and venous thromboembolism. Blood. 2000;96(3):1177-1179.
70. Francis CW, Marder VJ, Martin SE. Demonstration of a large molecular weight variant of the gamma chain of normal human plasma fibrinogen. J Biol Chem.
1980;255(12):5599-5604.
71. Mosesson MW, Finlayson JS, Umfleet RA. Human fibrinogen heterogeneities. 3.
Identification of chain variants. J Biol Chem. 1972;247(16):5223-5227.
72. Wolfenstein-Todel C, Mosesson MW. Human plasma fibrinogen heterogeneity: evidence for an extended carboxyl-terminal sequence in a normal gamma chain variant (gamma'). Proc Natl Acad Sci U S A. 1980;77(9):5069-5073.
73. Wolfenstein-Todel C, Mosesson MW. Carboxy-terminal amino acid sequence of a human fibrinogen gamma-chain variant (gamma'). Biochemistry. 1981;20(21):6146-6149.
74. Drouet L, Paolucci F, Pasqualini N, Laprade M, Ripoll L, Mazoyer E, Bal dit SC, Vanhove N. Plasma gamma'/gamma fibrinogen ratio, a marker of arterial thrombotic activity: a new potential cardiovascular risk factor? Blood Coagul Fibrinolysis. 1999;10 Suppl 1:S35-S39.
75. Lovely RS, Falls LA, Al Mondhiry HA, Chambers CE, Sexton GJ, Ni H, Farrell DH.
Association of gammaA/gamma' fibrinogen levels and coronary artery disease. Thromb Haemost. 2002;88(1):26-31.
76. Cooper AV, Standeven KF, Ariëns RA. Fibrinogen gamma-chain splice variant {gamma}' alters fibrin formation and structure. Blood. 2003;102(2):535-540.
77. Moaddel M, Farrell DH, Daugherty MA, Fried MG. Interactions of human fibrinogens with factor XIII: roles of calcium and the gamma' peptide. Biochemistry. 2000;39(22):6698- 6705.
78. Falls LA, Farrell DH. Resistance of gammaA/gamma' fibrin clots to fibrinolysis. J Biol Chem. 1997;272(22):14251-14256.
79. Mosesson MW, Siebenlist KR, Hainfeld JF, Wall JS. The covalent structure of factor XIIIa crosslinked fibrinogen fibrils. J Struct Biol. 1995;115(1):88-101.
80. Mosesson MW. Fibrinogen gamma chain functions. J Thromb Haemost. 2003;1(2):231- 238.
81. Liu CY, Nossel HL, Kaplan KL. The binding of thrombin by fibrin. J Biol Chem.
1979;254(20):10421-10425.
82. Mosesson MW. Antithrombin I. Inhibition of thrombin generation in plasma by fibrin formation. Thromb Haemost. 2003;89(1):9-12.
83. Meh DA, Siebenlist KR, Mosesson MW. Identification and characterization of the thrombin binding sites on fibrin. J Biol Chem. 1996;271(38):23121-23125.
84. Seegers WH, Nieft M, Loomis EC. Note on the adsorption of thrombin on fibrin. Science.
1945;101:520-521.
85. Klein PD, Seegers WH. The nature of plasma antithrombin activity. Blood.
1950;5(8):742-752.
86. Beyer K, Dandekar T, Keller W. RNA ligands selected by cleavage stimulation factor contain distinct sequence motifs that function as downstream elements in 3'-end processing of pre-mRNA. J Biol Chem. 1997;272(42):26769-26779.
87. MacDonald CC, Wilusz J, Shenk T. The 64-kilodalton subunit of the CstF polyadenylation factor binds to pre-mRNAs downstream of the cleavage site and influences cleavage site location. Mol Cell Biol. 1994;14(10):6647-6654.
88. Weiss EA, Gilmartin GM, Nevins JR. Poly(A) site efficiency reflects the stability of complex formation involving the downstream element. EMBO J. 1991;10(1):215-219.
89. Hooper WC, Dilley A, Ribeiro MJ, Benson J, Austin H, Silva V, Rawlins P, Wenger NK, Evatt BL. A racial difference in the prevalence of the Arg506-->Gln mutation. Thromb Res. 1996;81(5):577-581.
90. Dilley A, Austin H, Hooper WC, El Jamil M, Whitsett C, Wenger NK, Benson J, Evatt B.
Prevalence of the prothrombin 20210 G-to-A variant in blacks: infants, patients with venous thrombosis, patients with myocardial infarction, and control subjects. J Lab Clin Med. 1998;132(6):452-455.
91. Rees DC, Chapman NH, Webster MT, Guerreiro JF, Rochette J, Clegg JB. Born to clot:
the European burden. Br J Haematol. 1999;105(2):564-566.
92. White RH, Zhou H, Murin S, Harvey D. Effect of ethnicity and gender on the incidence of venous thromboembolism in a diverse population in California in 1996. Thromb Haemost. 2005;93(2):298-305.
93. Dowling NF, Austin H, Dilley A, Whitsett C, Evatt BL, Hooper WC. The epidemiology of venous thromboembolism in Caucasians and African-Americans: the GATE Study. J Thromb Haemost. 2003;1(1):80-87.
94. The International HapMap Project. Nature. 2003;426(6968):789-796.
95. Mannila MN, Eriksson P, Lundman P, Samnegard A, Boquist S, Ericsson CG, Tornvall P, Hamsten A, Silveira A. Contribution of haplotypes across the fibrinogen gene cluster to variation in risk of myocardial infarction. Thromb Haemost. 2005;93(3):570-577.
96. Bruggemann LW, Drijfhout JW, Reitsma PH, Spek CA. Alternatively spliced tissue factor in mice: induction by Streptococcus pneumoniae. J Thromb Haemost. 2006;4(4):918- 920.
97. Xiong Z, Shaibani A, Li YP, Yan Y, Zhang S, Yang Y, Yang F, Wang H, Yang XF.
Alternative splicing factor ASF/SF2 is down regulated in inflamed muscle. J Clin Pathol.
2006;59(8):855-861.
98. Mizgerd JP, Spieker MR, Lupa MM. Exon truncation by alternative splicing of murine ICAM-1. Physiol Genomics. 2002;12(1):47-51.
99. Zelensky AN, Gready JE. The C-type lectin-like domain superfamily. FEBS J.
2005;272(24):6179-6217.
100. Ley K. Functions of selectins. Results Probl Cell Differ. 2001;33:177-200.
101. Herrmann SM, Ricard S, Nicaud V, Mallet C, Evans A, Ruidavets JB, Arveiler D, Luc G, Cambien F. The P-selectin gene is highly polymorphic: reduced frequency of the Pro715 allele carriers in patients with myocardial infarction. Hum Mol Genet. 1998;7(8):1277- 1284.
102. Barbaux SC, Blankenberg S, Rupprecht HJ, Francomme C, Bickel C, Hafner G, Nicaud V, Meyer J, Cambien F, Tiret L. Association between P-selectin gene polymorphisms and soluble P-selectin levels and their relation to coronary artery disease. Arterioscler Thromb Vasc Biol. 2001;21(10):1668-1673.
103. Carter AM, Anagnostopoulou K, Mansfield MW, Grant PJ. Soluble P-selectin levels, P- selectin polymorphisms and cardiovascular disease. J Thromb Haemost.
2003;1(8):1718-1723.
104. Tregouet DA, Barbaux S, Escolano S, Tahri N, Golmard JL, Tiret L, Cambien F. Specific haplotypes of the P-selectin gene are associated with myocardial infarction. Hum Mol Genet. 2002;11(17):2015-2023.
105. Tregouet DA, Barbaux S, Poirier O, Blankenberg S, Bickel C, Escolano S, Rupprecht HJ, Meyer J, Cambien F, Tiret L. SELPLG gene polymorphisms in relation to plasma SELPLG levels and coronary artery disease. Ann Hum Genet. 2003;67(Pt 6):504-511.
106. Zee RY, Cook NR, Cheng S, Reynolds R, Erlich HA, Lindpaintner K, Ridker PM.
Polymorphism in the P-selectin and interleukin-4 genes as determinants of stroke: a population-based, prospective genetic analysis. Hum Mol Genet. 2004;13(4):389-396.
107. Bugert P, Vosberg M, Entelmann M, Jahn J, Katus HA, Kluter H. Polymorphisms in the P- selectin (CD62P) and P-selectin glycoprotein ligand-1 (PSGL-1) genes and coronary heart disease. Clin Chem Lab Med. 2004;42(9):997-1004.
108. Stram DO, Haiman CA, Hirschhorn JN, Altshuler D, Kolonel LN, Henderson BE, Pike MC.
Choosing haplotype-tagging SNPS based on unphased genotype data using a preliminary sample of unrelated subjects with an example from the Multiethnic Cohort Study. Hum Hered. 2003;55(1):27-36.
109. Epstein MP, Satten GA. Inference on haplotype effects in case-control studies using unphased genotype data. Am J Hum Genet. 2003;73(6):1316-1329.
110. Blann AD, Noteboom WM, Rosendaal FR. Increased soluble P-selectin levels following deep venous thrombosis: cause or effect? Br J Haematol. 2000;108(1):191-193.
