Daniela Mari
Department of Internal Medicine, IRCCS Maggiore Hospital and University of Milan, Italy
Gene -Environment Interactions and Vascular Risk in the Elderly
to my son Pier Vittorio
Introduction
Perspective study have demonstrated a positive association between plasma levels of protein involved in hemostatic mechanisms and the development of atherothrombotic disease in individual of either sex 1 . The genetic factors that modulate the individual susceptibility to multifactorial diseases such as cardiovascular diseases, are common polymorphisms which generally have a modest effect at an individual level, but , because of their high frequency in population, can be associated with a high attributable risk. Potential synergistic gene-environment interaction can reveal or facilitate the phenotypic expression of such susceptibility genes. Common biallelic polymorphisms of many genes encoding for coagulation and fibrinolysis proteins influence the plasma levels of these proteins 2,3. Cardiovascular diseases (e.g., myocardial infarction, angina, and stroke) reach epidemic proportions in the elderly and are the primary limits to survival in man. In the last decades a special attention is blooming on cardiovascular risk factors in elderly and genetic epidemiological studies on disease of aging are increasing rapidly 4. Many aspects of aging involve inflammation 5. Age-related disease such as atherosclerosis are triggered or worsened by systemic chronic inflammation. Some traditional risk factors loose importance as predictors of atherothrombotic diseases and increases the role of novel cardiovascular risk factors as clotting and fibrinolytic proteins.
Polymorphisms in hemostatic system in elderly
Fibrinogen
Elevated plasma levels of fibrinogen are consistently associated with arterial thrombotic disorders. Prospective studies of both healthy subjects and patients who experienced atherothrombotic diseases have showed an association between fibrinogen levels and myocardial infarction , stroke and peripheral vascular disease. The Nortwhwick Park Heart Study was the first to report the association between high fibrinogen levels and ischemic heart disease (IHD). Subsequently many studies and meta-analyses confirmed the initial report 6-11. Fibrinogen levels are affected by age, physical inactivity, smoking , infection, body weight, alcohol consumption, hypertension, insulin resistance pattern, seasonal changes in elderly people 12. Fibrinogen, moreover, is a molecule that plays a role in acute-phase inflammation during which fibrinogen genes are coordinately up-regulated by IL-6 13. The G/A-455 biallelic polymorphism located in the promoter region of the ß-fibrinogen gene has been evaluated in many clinical studies. 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 ({approx.}0.20 in the general population) have higher fibrinogen levels ({approx.}20% to 30%) than carriers of the G-455 allele 14. In a population based sample who have enrolled 482 healthy middle-aged men, of whom 231 were smokers, the beta-fibrinogen G/A-455 polymorphism genotype along with plasma fibrinogen levels was determined in. Smokers had the highest plasma fibrinogen levels (2.92 g/l), ex-smokers the next (2.73 g/l), and never-smokers the lowest levels (2.66 g/l, P < 0.001). Those with one or two A-455 alleles had significantly higher plasma fibrinogen levels in never-smokers and ex-smokers (8.2% and 9.0%, respectively, P < 0.05), and the effect was larger in younger men (45 < 55 years, 11.6%, P = 0.002) than older men (> 65 years, 4.5%, NS), and was not significant in smokers (2.4%, P > 0.05). Allele frequencies were calculated and compared across age groups and between smokers and non-smokers. The difference in frequency of the -455 G/A allele between smokers and non-smokers varied significantly with age
(P < 0.01), with the frequency of the A-455 allele being significantly lower in smokers than in non-smokers in subjects aged > 65 years (P < 0.05), but not in younger men. 15. Recently,a population-based longitudinal study of 5888 older American adults, the Cardiovascular Health Study (CHS) (16,17) including all person 65 years of age or older , are followed up for 10 years to assess the roles of traditional and novel cardiovascular risk factors in the development of coronary heart disease. Primary end points of this study included "years of life" as well as "years of healthy life", defined as number of person in good health in the 10 years after the enrollment. In CHS plasma fibrinogen is a predictor of total mortality , confirming the results of other cohort study on older adults 18,19. On the contrary no association was observed with longevity for the -455 G/A beta-fibrinogen promoter polymorphism 20. The explanation of this negative result may depend on the small, insignificant part of the overall variance in fibrinogen levels explained by the -455G/A polymorphism in the promoter region of the fibrinogen gene. This fact was observed also in a twin study consisting of 129 monozygotic twin pairs and 153 dizygotic same-sex twins aged 73 to 94 years who participated in the Longitudinal Study of Aging of Danish Twins 21. As expected, the effects of genetic factors are smaller in the elderly twins because of the presence of many inflammatory triggers, such as cardiovascular disease and other age-associated diseases. It is not completely explained how much of the difference in heritability can be attributed to age versus other population characteristics. The association between G-455A polymorphism of the fibrinogen Bβ-gene and plasma fibrinogen levels and myocardial infarction, and potential synergistic gene-environment interaction involving this polymorphism have been explored in a case-referent study (Stockholm Heart Epidemiology Program study, SHEEP) 22. A total of 2246 cases (1485 men and 761 women) are enrolled of which 1643 (110 men and 538 women) survived at least 28 days after myocardial infarction . 1169 survivors (826 men aged 58.2 7.1 years and 324 women aged 61.6 6.8 years) are successfully genotyped for the G-455A polymorphism and their referents are 1517 (1028 men aged 58.8 7.1 years and 489 women aged 61.9 6.7 years). Smoking, job strain, overweight, diabetes mellitus, physical inactivity, hypertension, hypercholesterolaemia, high low-density lipoprotein cholesterol/high-density, lipoprotein cholesterol (LDL/HDL) quotient, and hypertriglyceridaemia are include as environment-gene interaction factors. Presence of the A allele at G-455A polymorphic site is associated with higher plasma fibrinogen levels than the presence of the G allele, but the difference was statistically significant for male subjects only. No association was found between the -455A allele and increased risk of myocardial infarction. In addition no synergistic interaction gene-environment interaction were detected with any of the environmental exposures studied.
Factor VII (FVII)
The coagulation cascade is triggered when circulating factor VII reacts with tissue factor who is usually not expressed in the intact vasculature. Factor VII plasma levels progressively increase with the age, from a mean of 95 units/dl in subjects 20 years to over 110 units/dl in subjects over 50 years 23.Other environmental factors than age influence plasma levels of factor VII, body max index and plasma tryglicerides. The correlation between plasma triglycerides levels and FVII coagulant activity might suggest the combined role of FVII and ipertriglyceridemia in arterial thrombotic disease. Thrombotic disorders have been shown to be more frequent in subjects with higher plasma levels of factor VII 24, but the data are been not confirmed in other studies 25; unfortunately there are differences in factor VII coagulant assay methodologies in the different studies. In Cardiovascular Heart Study 18, factor VII exhibited associations with incident angina (RR=1.44) in men and with death in women (RR, middle quintile compared with first =0.66). However, in general, factor VII was not consistently associated with cardiovascular events in this population 26.
A common polymorphism in FVII gene that is detected by the presence or absence of the cleavage site for Msp 1, is strongly associated with factor VII coagulant activity (factor VIIc)27. The variable site is located in exon 8 of the gene and is detected using in vitro gene amplification. The base change that creates the Msp I polymorphism is a G to A substitution, leading to the replacement of arginine (Arg) with glutamine (Gln) in the protein product of the M2 allele. There are conflicting data in literature regarding results of clinical studies carried out in young-adult population on the association between the FVII polymorphism and arterial thrombosis . Reports of a protective effect of the Q allele against MI in several Italian studies 28-30are counterbalanced by numerous reports showing no such findings in other populations 31-34 including the Framingham Heart study 35 Recently, the ongoing results of a case-control genetic study based on Jerusalem Longitudinal Cohort Study (JLCS) carried out in elderly are published 36 . DNA samples are available on approximately 400 of these subjects and R353Q factor VII polymorphism was carried out. The age of the population at the time of DNA sampling was 75 years. There were 150 males and 74 females in the older Ashkenazi group. The younger control group (21.89 years, range 13–33 years) included 121 Ashkenazi males and 320 females. The only significant result was decrease in the percentage (30.6% versus 19.2%) of the A allele in older Ashkenazi male subjects (chi-square = 4.027,
P = 0.045, 1d.f.) and a corresponding marked decrease (9.7–2.1%) in the percentage of AA homozygous (chi-square = 5.790, P = 0.055, 2d.f.). Thus, these data confirm that if there is any effect of FVII genotypes on atherothrombotic disease, it is very small and detectable only in highly selected populations. Jeffery et al, 2005 37 suggest that there is a protective contribution of the Q allele of the R353Q polymorphism of the Factor VII , a guanine-to-adenine substitution in the position of the codon for amino acid 353 resulting in an amino acid replacement of arginine (R) by glutamine (Q), in individuals with chronic stable angina . Patients attending the hospital for routine day case angiography over a 20 month period are follow up prospectively . Factor VIIa, VIIc and VIIAg levels, genotype for R353Q, lipid status, smoking history and the degree of vessel disease are determined. Of 519 cases, 400 had no previous myocardial infarction or revascularisation, including 153 with zero vessel disease; the results of genotype showed 9 (2%) QQ, 78 (20%) RQ and 313 (78%) RR. Compared with RR subjects, heterozygous were 2.7 years older (95% CI: 0.3, 5.0; P= 0.027), but no significant difference have been observed regarding gender, cholesterol, extent of vessel disease or smoking history. If those with vessel disease were considered, then the heterozygous were 3.5 years older than the RR homozygous (95% CI: 0.6-6.4, P= 0.016). There was a significant association between all measures of Factor VII and the R353Q polymorphism, with the Q allele associating with lower levels. There was no significant association between the degree of vessel disease and genotype. A moderate protective effect of the Q allele against stable angina was observed in that angiographic investigation occurs a few years later, for RQ heterozygous than RR homozygous. The effect may be mediated by reduced levels of Factor VII. When the entire cohort was investigated for age at presentation for angiogram, those with RQ genotype had a presentation 2.7 years later than those with an RR genotype (95% CI 0.3–3.5, P= 0.027). This was primarily observed in the group with vessel disease, where the effect was greater, there being a 3.5 year difference (95% CI 0.6–6.4, P= 0.016). QQ homozygous only numbered 5, and did not show any significant difference. The effect size was similar for each gender separately: females 3.9, P= 0.27; males 3.4, P= 0.03. Smoking status in those with vessel disease was not associated with age at presentation, smokers are 1.2 years older, 95% CI (1.3, 3.7) P = 0.3.
The relation with food intake with the RQ353 polymorphism, was explored in extreme quintile of coagulation factor VII clotting activity (FVII:C) in 1158 subjects. The subjects are enrolled within the Rotterdam Study , population based prospective cohort study of 7893 persons. In the total group (n=1158) FVII:C was inversely associated with fiber and protein intake and positively with saturated fat. The inverse association of factor VII with fiber was stronger in subjects with the RR genotype (= -0.76%PP/g, CI: -1.23, -0.29), than in those with RR/QQ genotypes ((= -0.19%PP/g, CI: -0.97, 0.59). The association of FVII:C with saturated fat was positive in those with the RR allele and inverse in subjects with the Q allele 38. In spite of the limits of the study, it is well known the RR genotype of the R/Q353 polymorphism of factor VII is a risk indicator for myocardial infarction through increased factor VII levels. Environmental dietary factors, like high intake of saturated fat and low intake of fiber and protein may interact with the thrombotic risk.
Factor XIII
Factor XIII catalyzes cross-linking of fibrin in the last step of the coagulation process. A common GT polymorphism leading to a valine (Val) to leucine (Leu) substitution 3 amino acids from the thrombin activation site, described in exon 2 of the FXIII A-subunit gene (FXIII Val34Leu), has been associated with a decrease risk of coronary artery disease and cerebrovascular disease 39. Because of complex gene-gene environment interaction between polymorphisms in the fibrinogen-chain gene and FXIII Val34Leu polymorhism, fibrinogen plasma levels are modulated , influencing the permeability of clot . As intermediaries for environmental effects, fibrinogen concentrations are an important candidate for interacting with fibrinogen and factor XIII coding polymorphisms to alter vascular risk 40. Clinical studies confirm this gene-gene and gene-environment interaction. Elbaz et al 41 study the relation between brain infarction (BI) and the FXIII Val34Leu polymorphism in 456 patients with a median age of 69 years, consecutively recruited with a BI confirmed by imaging, and 456 matched controls. The distribution of genotypes was different in cases (63. 2% Val/Val; 30.9% Val/Leu; 5.9% Leu/Leu) compared with controls (49. 8% Val/Val; 42.8% Val/Leu; 7.4% Leu/Leu; P < 0.001). Carrying the Leu allele was associated with an OR of 0.58 (95% CI = 0.44-0.75). In addition, the effect of smoking was modified by the polymorphism (P = 0.05); and was weaker among Leu carriers than among noncarriers, suggesting that among Leu carriers, the protective effect of the polymorphism outweighed the effect of smoking.
Factor V
A common missense mutation in coagulation Factor V(Arg 506 Gln), factor V Leiden, 42 creates phenotypic resistance to the anticoagulant effects of activated protein C and predispose carriers to venous thrombosis 43.
Women's Health Initiative Estrogen Plus Progestinic Group is a double-blind controlled trial of 16608 women between the ages of 50 and 79 years at the baseline, with the follow-up of 5.6 years. 44. Together with other gene variant related to thrombosis risk (prothrombin 20120A, methylenetetrahydrofolate reductase C677F), factor XIII Val34Leu, PAI-1 4G5G) FV Leiden was measured in the first 147 women who developed thrombosis and in 513 controls. Estrogen plus progestinic was associated with doubling the risk of venous thrombosis. Only factor V Leiden enhanced the hormone- associated risk of venous thrombosis in postmenopausal women, with a 6.69-fold increased risk compared with women in the placebo group without the mutation .
The role of factor V mutation in arterial thrombotic disease has been extensively studied clinically in patients who had an event at a young age, but the results are negative or affected by environmental risk factors 45 . The prevalence of the factor V mutation was determined in a group of elderly patients (649 years old) with stroke compared with its prevalence in several control patient group (older controls 66±8 years old, young controls 46±14 years old) 46. No increase in the prevalence of factor V Arg 506Gln mutation in elderly patient with stroke compared with its prevalence in several control patient groupwas found. In addiction, the frequency of factor V mutation in elderly individuals with either stroke or at risk of stroke was lower than in young control subjects. The result of this study on ischemic stroke in elderly suggests that there is no need to screen the typical elderly stroke patient for this prevalent allele.
Factor VIII/Von Willebrand factor (FVIII/VWF)
It is well known that increased plasma levels of FVIII are associated with an increased risk of venous thrombosis 47. The plasma FVIII levels are modulate by VWF antigen (VWFAg), the carrier molecule of FVIII in plasma 48, 49 and ABO blood group 50. In Cardiovascular Health Study, factor VIII showed a positive association with a risk of coronary heart disease among men and a risk of cerebrovascular disease, i.e. transient ischemic attacks and stroke among women 51.
VWF has been identified as a risk factor for recurrent myocardial infarction in the general population and is associated with cerebrovascular disease 52-54. The Rotterdam Study is a large community-based study of an elderly population. To investigate the presence of a prothrombotic state in atrial fibrillation (AF) among the elderly, von Willebrand factor, a marker of endothelial damage/dysfunction, was detected in a matched case-control study (162 patients, aged 78 years with AF, 324 controls, aged 77 years ). A positive relationship between AF and Plasma vWF was observed in elderly population, especially among women . A strong association between VWF and myocardial infarction , diabetes mellitus, current smoking was showed 55.
Recently, the results of a 20th year follow-up examination of all surviving men (aged 60–79 years), enrolled in the British Regional Heart Study were published 56. In this prospective study of cardiovascular disease, non-diabetic older men with no history of cardiovascular heart disease or stroke, showed a significant association between insulin resistance and inflammatory markers (CRP and white cell count), coagulation factors VII–IX, markers of endothelial dysfunction (VWF and tissue-plasminogen activator) and blood viscosity. These relationships were independent of age, smoking, physical activity, alcohol intake, social class, use of statin and aspirin and persisted after further adjustment for abdominal obesity. In particular fasting insulin was the factor associated most strongly with VWF and factor VIII. Only fasting insulin was independently associated with VWF, whereas factor VIII, which was highly correlated with VWF (r = 0.69), showed independent relationships with blood glucose and fasting insulin.
Plasminogen activator inhibitor -1 (PAI-1)
PAI-1 is a major determinant of fibrinolytic activity. There is substantial experimental and epidemiological evidence that PAI-1 might contribute to the development of ischaemic cardiovascular disease 57. Elevated PAI-1 play a role in the metabolic syndrome 58 and strong associations have been described between PAI-1 plasma levels and excess weight, abdominal obesity , diabetes mellitus and insulin concentration. A common deletion/insertion polymorphism (4G/5G) has been identified in the 5' promoter region . The guanine insertion/deletion polymorphism is located at position -675 , where one allele has a sequence of four guanines (4G) and the other has a fifth guanine inserted (5G) 59. The 4G allele has been correlated with high levels of gene transcription and elevated PAI-1 plasma levels in comparison with the 5G allele. This PAI-1 polymorphism modulate phenotypes associated with the metabolic syndrome 60.