Gene Polymorphisms Predicting High Plasma Levels of Coagulation and Fibrinolysis Proteins
A Study in Centenarians
Abstract Gene polymorphisms associated with the plasma levels of fibrinogen, factor VII, and plasminogen activator inhibitor 1 (PAI-1)—hemostasis proteins that help to predict the risk of atherothrombotic disease—were compared in 124 healthy individuals ≥100 years old and 130 young, healthy individuals to identify genetic influences on extreme longevity. We investigated the restriction fragment length polymorphism G/A−455 located in the promoter of the β-fibrinogen gene, the guanine insertion/deletion polymorphism 4G/5G in the promoter of the PAI-1 gene, and the R353Q substitution polymorphism in exon 8 of the factor VII gene. Alleles and genotypes associated with elevated plasma levels of fibrinogen and factor VII were found with similar frequencies in centenarians and in the comparison group. However, in centenarians there was a significantly higher frequency of the 4G allele and of the homozygous 4G4G genotype associated with high PAI-1 levels. Since high PAI-1 is considered a predictor of recurrent myocardial infarction in young men, it is intriguing that the corresponding genetic marker is more frequent in centenarians who have escaped major age-related atherothrombotic disease and reached the extreme limits of human life. Homozygosity for the 4G allele, despite its association with impaired fibrinolysis, is compatible with successful aging.
- Received April 16, 1995.
- Accepted July 31, 1996.
Prospective studies have demonstrated a positive association between the plasma levels of proteins involved in hemostatic mechanisms and the development of atherothrombotic disease in individuals of either sex. The best-established predictors of risk are elevations of the coagulation factors fibrinogen and factor VII and the fibrinolysis component plasminogen activator inhibitor 1 (PAI-1).1
Common biallelic polymorphisms of the genes encoding for fibrinogen, factor VII, and PAI-1 influence the plasma levels of these proteins.2 The polymorphic genotypes associated with elevated plasma levels of these hemostasis components might prove more reliable as predictors of the thrombotic risk than the plasma levels. In the general population, plasma levels have a continuous distribution, and so it is difficult to establish cutoff values that can identify the individuals at risk. In addition, variation of plasma levels determined by environmental factors may further complicate risk assignment. To assess the value of genetic markers as risk predictors, studies have been carried out in relatively small groups of individuals with or without atherothrombotic disease, but the results have been conflicting.3 4 5 6
The purpose of this study was to evaluate the allele and genotype frequencies of polymorphisms influencing plasma levels of fibrinogen, factor VII, and PAI-1 in a cohort of Italian healthy centenarians, who are considered a natural model of extreme longevity and positive selection.7 Results obtained in these exceptional individuals, who have escaped age-related atherothrombotic disease and reached the extreme limits of life, were compared with those obtained in healthy individuals who were young enough that none of them would be lost from the study sample because of symptomatic atherothrombotic disease. We surmised that the study of genetic markers of hypercoagulability and hypofibrinolysis should establish whether or not successful aging is associated with the predominance of genotypes predicting lower plasma levels of hemostasis proteins.
In Italy, there is currently a multicenter project on centenarians; it is coordinated by one of the authors (C.F.). The aim of this project is to study the mechanisms of successful aging. Centenarians identified from the General Registry Offices are studied in special units, generally established in university departments. Approximately 2000 centenarians have been identified.7 Several biological systems of these individuals are being investigated by biochemical and molecular methods.
The present study is part of the hemostasis investigations.8 9 DNA polymorphisms for two coagulation factors, factor VII and fibrinogen, and the fibrinolysis inhibitor PAI-1 were studied in 124 healthy centenarians from four metropolitan communities in northern Italy (Milan, Modena, Parma, and Genoa). Allele and genotype frequencies were compared with those of 130 young, healthy individuals recruited from the same communities and ethnically homogeneous with the centenarians but unrelated to them. Approval for these studies was obtained from the Institutional Review Board of the University of Milan, and informed consent was given after the experimental nature of the studies had been explained.
Centenarians (96 women and 28 men 100 to 109 years old, all white) were ambulatory, self-sufficient, and living in their homes. None of them were current smokers; 31 had stopped smoking between 5 and 10 years before blood sampling, whereas the remaining 93 had never smoked or had stopped smoking at least 10 years before. They were selected from the whole group of centenarians identified in the same communities because they were judged to be healthy on the basis of the Senieur protocol,10 which establishes strict enrollment criteria based on clinical history, physical examination, and laboratory data for studies on healthy elderly people. Exclusion criteria were chronic disorders such as dementia, malignancy, infection, diabetes, a history of arterial and venous thrombosis, and current or past cardiovascular, renal, and liver disease. Hemocytometry and erythrocyte sedimentation rate and plasma concentrations of glucose, sodium, potassium, albumin, alkaline phosphatase, aminotransferases, urea, and creatinine were measured. Centenarians with values outside the 2.5 and 97.5 percentiles of the reference values of these measurements were excluded. An electrocardiogram was also obtained. Values of serum triglycerides, a measurement not included in the Senieur protocol, were available in only 101 centenarians and in 60 control subjects.
Young individuals (82 women and 48 men) were all <31 years old (mean age, 25.3±3.7 years; range, 14 to 30 years) and were randomly selected from the same metropolitan communities (hospital staff, high-school students). Arterial or venous thrombotic disease was excluded on the basis of clinical history and the results of physical examination. None of them were current or recent smokers.
Blood samples for DNA analysis were drawn directly into vacuum containers containing sodium citrate as anticoagulant. Genomic DNA was prepared from whole blood by a standard salting-out procedure.11
The G/A−455 biallelic polymorphism is located in the promoter region of the β-fibrinogen gene. The allele containing the polymorphic cutting site for the restriction enzyme HaeIII is identified as G−455, and the allele without the cutting site is identified as A−455. Carriers of the less frequent allele A−455 (≈0.20 in the general population) have higher fibrinogen levels (≈20% to 30%) than carriers of the G−455 allele.12
The method used was that described by Thomas et al,12 with some modifications. Briefly, a portion of the β-fibrinogen gene, −551 to −109 bp from the start of transcription, was amplified by polymerase chain reaction (PCR) using the primers 5′-ATAGAATAGGGTATGAATTTG-3′ and 5′-GAACCATTTTATCATTTAAGC-3′ in a 50-μL mixture containing 200 to 500 ng DNA, 0.1 mg/mL BSA, 200 μmol/L of each nucleotide triphosphate, 100 pmol of each primer, 2 mmol/L MgCl2, 10 mmol/L Tris, 50 mmol/L KCl (pH 8.3), and 2.5 U Taq polymerase. Amplification involved 33 cycles of 93°C (20 seconds), 50°C (10 seconds), and 70°C (45 seconds). Subsequently, the PCR product was digested with the restriction enzyme HaeIII (Boehringer Mannheim). Finally, the digestion products were separated by electrophoresis on 3% agarose gel and stained with ethidium bromide. The fragments of the G−455 allele contained 346 and 96 bp and the fragment of the A−455 allele, 442 bp.
Factor VII Polymorphism
The biallelic polymorphism is located in exon 8 of the factor VII gene and results from a missense mutation, leading to the substitution of 353R by Q in the mature protein.3 Factor VII plasma levels are 20% to 25% higher in 353RR homozygotes than in carriers of the 353Q allele, which has a population frequency of ≈0.10.3
The method used was that described by Green et al.3 Genomic DNA was analyzed. Oligonucleotide primers were those described by Humphries et al.13 Amplification involved 36 cycles of 93°C (60 seconds), 55°C (60 seconds), and 72°C (120 seconds). Subsequently, the PCR product was digested with the restriction enzyme Msp I (Boehringer Mannheim), and the digestion products were separated by electrophoresis on ethidium bromide–stained 3% agarose gel. The fragment of the 353R allele was 205 bp and that of the 353Q allele, 272 bp.
The guanine insertion/deletion polymorphism is located at position −675 in the promoter region of PAI-1, where one allele has a sequence of four guanines (4G) and the other has a fifth guanine inserted (5G).14 The polymorphism is associated with plasma PAI-1 levels.6 14 Individuals homozygous for the 4G allele (frequency ≈0.50) have 25% to 30% higher PAI-1 levels than individuals with at least one 5G allele.
The PAI-1 polymorphism was analyzed by the allele-specific oligonucleotide melting technique described by Dawson et al,14 with some modifications. An 890-bp region of the PAI-1 promoter was amplified by PCR using the primers described by Dawson.14 Amplification involved 35 cycles of 95°C (60 seconds), 55°C (60 seconds), and 72°C (60 seconds). Samples (3 to 5 μL) of each PCR product were spotted onto two sheets of Hybond N+ nylon membranes (Amersham). Samples were denatured in 1.5 mol/L NaOH, 0.5 mol/L NaCl for 5 minutes, neutralized with 3× SSC, and fixed with ultraviolet cross-linking. Two 14-mer oligonucleotides were designed to cover the polymorphic site of each allele: 5′-CACGTGGGGAGTCA-3′ and 5′-ACGTGGGGGAGTCA-3′. T4 polynucleotide kinase (Promega) was used to phosphorylate the 5′ ends with [32P]ATP. Membranes were prehybridized individually by immersion for at least 2 hours in a solution containing 5× SSPE buffer, 5× Denhardt's solution (0.1% Ficoll 400, 0.1% polyvinylpyrrolidone, 0.1% BSA), 100 μg/mL salmon sperm DNA, and 0.5% SDS. Then the labeled probes were added separately to each solution at a final concentration of 105 cpm/mL for at least 1 hour. Membranes were washed in 5× SSC, 0.1% SDS for 20 minutes to remove background hybridization. Prehybridization, hybridization, and washing were carried out at 55°C for the oligonucleotide complementary to the 4G allele and at 58°C for the oligonucleotide complementary to the 5G allele. Autoradiograms were obtained by exposing films to membranes with intensifying screens for 16 to 20 hours at −70°C. Samples from individuals with known genotypes determined by allele-specific oligonucleotide melting14 and kindly supplied by S. Humphries (London, UK) were used as controls.
The allele and genotype frequencies of the two study groups were compared by χ2 analysis by a method designed for small or even zero populations.15 The odds ratio and 95% confidence intervals were calculated to measure the association between each genotype and being a centenarian. The data were adjusted for sex by multiple logistic regression analysis. The association between triglycerides and genotypes was evaluated by ANOVA. For circumstantial technical reasons, not all the results could be obtained for all the samples, so that in some instances the numbers of centenarians and control subjects studied are <124 and <130, respectively (see Tables⇓⇓⇓).
Table 1⇑ shows that for the fibrinogen and factor VII polymorphisms, there was no significant difference between centenarians and control subjects in the allele frequencies or in the frequencies of the six possible genotypes. However, there was a difference for the PAI-1 polymorphism: the 4G allele and the homozygous 4G4G genotype, associated with elevated plasma levels of the fibrinolysis inhibitor, were significantly more frequent in centenarians than in young individuals (P=.021 and P=.049, respectively) (Table 1⇑). The odds ratios, which give a measure of the relative chance of subjects with or without a given allele or genotype becoming centenarians, were 1.5 and 2.6 for the 4G allele and 4G4G genotype, respectively (Table 1⇑). Our population of centenarians was composed mainly of women, with a male-to-female ratio of 1:3. There was no significant sex-related difference in any of the allele or genotype frequencies (data not shown).
In centenarians and control subjects, we also analyzed the distribution of the 27 combinations of the nine possible genotypes for fibrinogen, factor VII, and PAI-1 polymorphisms. There was no significant difference between the two groups in the frequency of any combination (Table 2⇑). Some combinations were present only in a small number of cases; six combinations were absent in both centenarians and young individuals (Table 2⇑).The effect of smoking on the distribution of the three genotypes of the fibrinogen polymorphism was not analyzed because centenarians (and control subjects) were nonsmokers or had stopped smoking for ≥5 years. Mean levels (±SD) of serum triglycerides were significantly higher in centenarians (n=101) than in young individuals (n=60) (114±51 versus 74±28 mg/dL, P<.001). In neither group was there any significant difference (ANOVA) in triglyceride levels in relation to the different genotypes of the factor VII polymorphism (Table 3⇑). Mean triglyceride values were higher in centenarians homozygous for the 4G PAI-1 polymorphism than in those with the 5G5G genotype (P<.05), but this difference was not seen in young individuals.
Three gene polymorphisms associated with the plasma levels of coagulation and fibrinolysis proteins that help to predict atherothrombotic disease have been evaluated in exceptionally long-lived individuals. We investigated polymorphisms of the coagulation factors, fibrinogen and factor VII, shown by the prospective Northwick Park Heart Study to be independent predictors of ischemic heart disease in middle-aged individuals.1 Also, because a prospective cohort investigation has shown that high levels of plasma PAI-1 activity are associated with early recurrence of myocardial infarction in young men,16 we investigated a PAI-1 polymorphism. Only a limited amount of blood can be obtained from such elderly persons, and so the simultaneous measurement of the corresponding plasma proteins was not done.
The frequencies of the two alleles (A−455 and 353R) and homozygous genotypes (AA−455 and 353RR) associated with higher levels of fibrinogen and factor VII were similar in centenarians and young individuals. These alleles and genotypes are associated not only with high plasma levels of these proteins but also with larger increments when carriers are exposed to environmental triggers such as smoking (for fibrinogen) and hypertriglyceridemia (for factor VII and PAI-1).2 17 The effect of smoking could not be evaluated in this study because centenarians were nonsmokers or had stopped smoking for a period of time long enough that any effect of smoking would be offset. In centenarians, triglyceride levels were higher than in young individuals and were associated with 4G4G genotype of the PAI-1 polymorphism, whereas there was no significant association between triglycerides and the factor VII genotype 353RR. Hence, the interaction between triglycerides and PAI-1 genotypes found in patients with non–insulin dependent diabetes mellitus17 was also present in centenarians. The relatively small sample is perhaps an explanation for the lack of association between triglyceride levels and the 4G4G genotype in young, healthy individuals.
On the whole, our findings in centenarians indicate that during the process of selection that allows a few individuals to reach the extreme limits of human life, one is not disadvantaged by the polymorphic genotypes associated with high levels of coagulation factors and high responses to environmental triggers. The 4G allele and 4G4G genotype associated with elevated levels of PAI-1, which predict recurrence of myocardial infarction in young men, were even more frequent in centenarians than in young individuals. It is intriguing that genotypes associated with “high-risk” phenotypes are at least as frequent or even more frequent in centenarians than in young individuals who carry a lifelong risk of atherothrombotic disease. This paradoxical finding is not isolated, because other genotypes with potentially unfavorable cardiovascular effects have been found more frequently in centenarians. Schächter et al,18 for instance, found that the homozygous genotype for the deletion polymorphism of the angiotensin-converting enzyme, which predisposes to coronary heart disease,19 20 is more frequent in French centenarians than in adults 20 to 70 years old (39.6% versus 25.6%). On the other hand, the ε4 allele of the gene encoding apolipoprotein E, which has been associated with ischemic heart disease,21 22 was less frequent in centenarians, a finding consistent with its status as a risk marker.18
The results of this study in centenarians on the frequency of gene polymorphisms associated with cardiovascular risk factors are compatible with the belief that the basis of successful aging is complex and multifactorial. It is likely that unidentified factors compensate for potentially unfavorable genotypes in centenarians. For instance, it can be hypothesized that the association of 4G with longevity arises because linkage disequilibrium with a locus counteracting the putative unfavorable effect of elevated PAI-1 levels offsets the risk conferred by hypofibrinolysis and creates favorable long-term effects. We should also consider the possibility of a cohort effect when comparing centenarians with individuals <30 years old, because young individuals may not have been exposed to the same environmental factors to which centenarians had been exposed at the corresponding ages. It is also possible that among elderly people who have escaped atherothrombotic disease, it is actually advantageous to have reduced fibrinolytic activity. The general conclusion that can be drawn from the study is that there is no evidence that possession of genotypes associated with increased circulating levels of fibrinogen, factor VII, and PAI-1 reduces the chance of survival to the age of 100 years without evident atherothrombotic disease.
This work was supported in part by CNR grants, Progetto Finalizzato Invecchiamento and MURST grants (60%). We thank Dr Kathryn Boyd for reading the manuscript critically.
Green F, Kelleher C, Wilkes H, Temple A, Meade TW, 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, Henney A, Humphries S. Genetic variation at the plasminogen activation inhibitor locus is associated with altered levels of plasma plasminogen activator inhibitor-1 activity. Arterioscler Thromb. 1991;11:183-190.
Ye S, Green FR, Scarabin PY, Nicaud V, Bara L, Dawson SJ, Humphries SE, Evans A, Luc G, Cambou JP, Arveiler D, Henney AM, Cambien F. The 4G/5G genetic polymorphism in the promoter of the plasminogen activator inhibitor 1 (PAI-1) gene is associated with differences in plasma PAI-1 activity but not with risk of myocardial infarction in the ECTIM study. Thromb Haemost. 1995;74:837-841.
Mari D, Mannucci PM, Coppola R, Bottasso B, Bauer KA, Rosenberg RD. Hypercoagulability in centenarians: the paradox of successful ageing. Blood. 1995;85:3144-3149.
Mari D, Mannucci PM, Duca F, Bertolini S, Franceschi C. Mutant factor V (Arg506Gln) in healthy centenarians. Lancet. 1995;347:1044.
Miller SA, Dykes DD, Polesky HF. A simple salting-out procedure for extracting DNA from human mutated cells. Nucleic Acids Res. 1988;16:1215-1218.
Humphries SE, Lane A, Green FR, Cooper J, Miller GJ. Factor VII coagulant activity and antigen levels in healthy men are determined by interaction between factor VII genotype and plasma triglyceride concentration. Arterioscler Thromb. 1994,14:193-198.
Dawson S, Wiman B, Hamsten A, Green F, Humphries S, Henney AM. The two allele sequences of a common polymorphism in the promoter of the plasminogen activator inhibitor-1 (PAI-1) gene responds differently to IL-1 in Hep G2 cells. J Biol Chem. 1993;268:10739-10745.
Roff DA, Bentzen P. The statistical analysis of mitochondrial DNA polymorphisms: chi square and the problem of small samples. Mol Biol Evol. 1989;6:539-545.
Cambien F, Poirier O, Lecerf L, Evans A, Cambou JP, Arveiler D, Luc G, Bard SM, Bara L, Ricard S, Tiret L, Amougel P, Alhenc-Gelas F, Soubrier E. Deletion polymorphism in the gene coding for angiotensin-converting enzyme is a potent risk factor for myocardial infarction. Nature. 1992;359:641-644.
Davignon J, Gregg RE, Sing CF. Apolipoprotein E polymorphism and atherosclerosis. Arteriosclerosis. 1988;8:1-21.