Genetic Influence on Inflammation Variables in the Elderly

From the Department for Thrombosis Research (M.P.M.dM., E.M.B., J.J.), University of Southern Denmark, and the Department of Clinical Biochemistry (M.P.M.d.M., E.M.B., J.J., Ribe County Hospital, Esbjerg, Denmark; the Department of Hematology (M.P.M.dM.), Erasmus Medical Center, Rotterdam, the Netherlands; and Institute of Public Health (J.v.B.H., K.C.), Research Unit of Epidemiology, University of Southern Denmark.

ABSTRACT

Background— Inflammation variables (C-reactive protein , fibrinogen, and soluble intercellular adhesion molecule-1 ) have been identified as risk factors for cardiovascular disease. It is still not known how much the regulation of inflammatory risk factors is determined by genetic factors, and the aim of this study was to determine the heritability of these inflammation variables and of the acute phase regulating cytokines interleukin-6 (IL-6) and tumor necrosis factor- (TNF-) at older ages.

Methods and Results— The heritability of CRP, fibrinogen, sICAM-1, IL-6, and TNF- was determined 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. Furthermore, we determined the influence of selected genetic polymorphisms on the plasma level variations. Genetic factors accounted for 20% to 55% of the variation in plasma levels of the inflammation variables. The highest heritability was found for sICAM-1. The genetic polymorphisms we studied explained only a small, insignificant part of the heritability.

Conclusions— This study in elderly twins provides evidence for a substantial genetic component of inflammatory cardiovascular risk factors among the elderly.

Inflammation variables have been identified as risk factors for cardiovascular disease. The heritability of inflammation variables at older ages was considered in a twin study on 129 monozygotic and 153 dizygotic same-sex twin pairs aged 73 to 94 years. Genetic factors accounted for 20% to 55% of the variation in levels of the inflammation variables studied.

Key Words: inflammation ? twins ? heritability ? cardiovascular disease

Introduction

A positive family history of cardiovascular disease is a strong determinant of the risk of cardiovascular events in epidemiological studies,1,2 and this observation was confirmed by a Swedish twin study in which death from coronary heart disease showed a substantial genetic component.3 Within families, the genetic background and environmental factors are shared, and twin studies are among the best studies for disentangling the effects of genetic and environmental components.

Inflammation is an important mechanism in cardiovascular disease. The acute phase proteins C-reactive protein (CRP),4,5 fibrinogen,6 and soluble intercellular adhesion molecule-1 (sICAM-1)7 are consistently associated with the risk of cardiovascular events in prospective population studies. The levels of these inflammatory markers reflect the severity of the underlying atherosclerosis in the vessel wall, but they may also causally contribute to the development of cardiovascular disease.8 As an example, CRP activates the classical pathway of the complement system, functions as a scavenger factor, and determines the uptake of low-density lipoprotein by macrophages.9 Fibrinogen levels determine the rigidity and fibrinolysis rate of clots; fibrinogen functions as a bridging molecule for many types of cell–cell adhesion events and provides a critical provisional matrix.10 ICAM-1 participates in the adhesion of neutrophils and monocytes to the endothelium, an essential step in extravasation of neutrophils and monocytes at the site of inflammation.

The plasma levels of these variables show a large interindividual and intraindividual variation.11 Inflammatory triggers are the main determinants of the levels of these factors, but also, genetic variations have been reported to explain part of the variation in inflammatory variables. Studies of CRP12–15 and fibrinogen13,16–18 in healthy young and middle-aged twins and in family studies suggest that a considerable part of the variation in these factors can be explained by genetic factors.

Because cardiovascular disease is primarily a disease of the elderly, it is important to assess how important genetic variation in risk factors is in the elderly because they are subject to many more triggers or risks than young individuals. It may be expected that genetic factors make a different contribution to the variation in the elderly compared with the younger age groups. On one hand, it is possible that functional genetic variations in the promoter region of the genes, which determine response to triggers, contribute more in the elderly because there are more triggers. On the other hand, the surplus and variety of triggers present in later age may also decrease the importance of genetic variation. Thus, it is not clear whether a larger or smaller contribution of genetic variation is expected, and evolution predicts the first, and traditional gerontology predicts the latter.

In addition to functional variations in genes of inflammation variables, functional genetic variations in factors that regulate these variables may contribute to the variation in plasma levels of inflammatory variables. Interleukin-6 (IL-6)19 and tumor necrosis factor- (TNF-)20 are 2 important regulatory cytokines for acute phase proteins21,22 that both have a genetic contribution to their levels23 and that both have functional genetic polymorphisms in their promoter region that have been associated with their plasma levels.24,25 These functional polymorphisms, for which an association with levels has been shown repeatedly and is biologically plausible, are the most likely genetic variants to explain part of the hereditary component that determines the plasma levels of the inflammatory variables.

The aim of this study was to assess the overall genetic role in variations of inflammation variables among the elderly. Therefore, we determined the genetic contribution to the variations in CRP, fibrinogen, sICAM-1 and the acute phase-regulating cytokines IL-6 and TNF- in elderly twins (73 to 94 years old).

Subjects and Methods

Participants

A total of 584 individuals (190 males and 394 females) aged 73 to 94 years participated in this study. These individuals made up 129 intact monozygotic (MZ) twin pairs and 153 intact dizygotic (DZ) same-sex twin pairs. The twin pairs were identified using the Danish Twin Registry and participated in the 1997 Longitudinal Study of Aging of Danish Twins.26 The aim was to include a random sample of twins so all twins who were able to finalize the interview and give informed consent were eligible for participation in the study.

The Danish Twin Registry, a nationwide and population-based registry established in 1954, comprises twins born between 1870 and 1930, surviving to the age of 6 years.26 Only same-sex twin pairs were included in the present study. Zygosity was established through a questionnaire on the degree of similarity between twins in a pair. Zygosity classification was evaluated by comparison with blood group determinants, and the misclassification rate was <5%.27

Methods

Blood samples (20 mL of EDTA blood) were collected within a 6-month period. The samples were centrifuged within 12 hours after sampling at 1000g for 10 minutes, and plasma was frozen in aliquots at –80°C within 2 days (usually within the same day). Plasma samples were rapidly thawed in a water bath at 37°C. The protein concentration of CRP was measured using a high-sensitivity in-house enzyme immunoassay using rabbit anti-human CRP IgG as capture and tagging antibody (DAKO). Human CRP Standard (Dade Behring) was used as a calibrator. Protein concentrations of IL-6, TNF-, and sICAM-1 were determined using high-sensitivity human ELISAs (R&D Systems). The protein concentration of fibrinogen was determined using a nephelometric method (Dade Behring). Interassay coefficients of variation for the assays we used were: for fibrinogen, <5.0%; IL-6, <10.1%; ICAM, <5.6%; TNF, <14.5%; CRP, <6.5% based on 17 determinations from a single normal plasma sample on different days.

Genomic DNA samples were analyzed using polymerase chain reaction (PCR)-based methods. The fibrinogen ?-455G/A promoter polymorphism was analyzed as described by Thomas et al.28 The CRP-1059G/C polymorphism was analyzed as described by Zee et al.29 The ICAM-1 K/E469 polymorphism was analyzed as described by Nishimura et al.30 The IL-6–174C/T promoter polymorphism was analyzed as described by Fishman et al.24 The TNF-–238G/A promoter polymorphism was analyzed as described by Wennberg et al.31 The genotype was determined in all individuals and monozygosity was confirmed. In each PCR run, samples with known genotypes were included. We also reanalyzed 10% of the samples (at random), and the results were confirmed.

Analyses of Twin Similarity

The similarity in MZ and DZ twins was assessed using intraclass correlations for inflammatory variables. The classic twin study methodology is based on the fact that MZ twins have identical genotypes, whereas DZ twins share, on average, half of their genes and are no more genetically related than ordinary siblings. A greater phenotypic similarity in MZ twins is to be expected if there is a significant genetic component in the etiology of the component. Analyses were adjusted (by linear regression) for age, sex, and body mass index (BMI). To approximate a bivariate normal distribution, data were transformed to the logarithmic scale, and probit plots showed that the marginal distributions could be assumed to be normal. Plasma levels of CRP >10 mg/L and levels of IL-6 >10 pg/mL (6% of the individuals) are considered to reflect "clinical inflammation"; these data and the data from the cotwin were removed from the data set.

Estimation of Heritability

Heritability of the 5 traits, effects of covariates on traits and the degree of dependence between the traits, were determined using multivariate variance components models.32 Phenotypic (co)variances are decomposed into genetic and environmental components by the Cholesky decomposition method. According to standard biometric practice, assuming no epistasis (genetic inter locus interaction), no gene–environment interaction or correlation, and no assortative mating, the phenotypic variance can be separated into 4 variance components: (A) variance attributable to additive genetic effects; (D) genetic dominance; (C) shared (family) environment; and (E) nonshared (individual-specific) environment.32 Only nonshared environments contribute to dissimilarity within MZ twin pairs because of their genetic identity, whereas the effects of additive genetic factors and genetic dominance may also contribute to dissimilarity within DZ pairs, who share, on average, half of the additive and one-quarter of the dominant genetic factors. For each trait, the heritability is the ratio of genetic variance to the total variance. Effects of the covariates of each trait are given by the regression coefficients coming from the covariate that is linearly added to the model (which corresponds to linear regression of the covariates on traits and subsequent heritability analysis of residuals thereof.

The multivariate model incorporates the 5 traits and associated covariates and allows estimating the degree of dependence between the traits (ie, correlations between the traits). Furthermore, the extent to which the same genetic factors contribute to the observed phenotypic correlation (ie, the degree of pleiotropy of possible genes influencing the traits) is obtained from the model by estimating the genetic correlations. The method for selecting the best model followed standard procedures (structural-equation analyses) using the Mx statistical program.32,33

Association Between Genotype and Phenotype

Association between the 5 polymorphisms and 5 trait levels in the whole population was initially assessed by 1-way ANOVA (with robust estimation of variance of phenotypes for individual twins clustered into twin pairs because of possible within-pair dependence). Sib-pair–based association analysis was performed using the variance components models of association implemented in the Quantitative Transmission Disequilibrium Test (QTDT) program,34 and in all analyses, gender, age, and BMI were included as covariates. We tested for association in QTDT using the model by Fulker et al,35 allowing for the effects of polygenes, shared family environment, nonshared environment, and effect of linkage at the locus under study. This model allows for the partitioning of observed association into between-pair and within-pair components, thus providing a check on whether results could have arisen because of unsuspected population stratification among the pairs studied. Subsequent permutation tests36 of association, which do not rely on the bivariate normal distribution assumption, were performed using the QTDT program. Secondly, the amount of variation in the plasma levels of the inflammatory variables attributable to the inflammatory polymorphisms was studied using parametric linkage analysis implemented in the Merlin program.36 There is no adjustment of the mean levels for zygosity. It is an assumption that MZ and DZ pairs come from the same population. There is no sign of violation of this assumption (Table 1).

TABLE 1. General Characteristics and Concentrations of Inflammation Variables in MZ and DZ Elderly Twins

Allele frequencies were determined by gene counting, and 95% CIs of the allele frequencies were calculated from sample allele frequencies. Deviations of the genotype distributions from that expected for a population in Hardy–Weinberg equilibrium were analyzed using the  2 test.

Results

Table 1 shows the general characteristics of our elderly twin population and the concentrations of inflammation variables in the MZ and DZ twins. All characteristics are comparable in the MZ and DZ twins. Concentrations of the inflammatory variables are comparable in MZ and DZ twins.

The model chosen to explain the factors contributing to the variance in the inflammatory cardiovascular risk factors is the model including additive genetic effects, common environmental effects, and nonshared environmental effects (ACE). In all the traits, the common environmental effects were small and nonsignificant (data not shown). The ACE model was then used to estimate the heritability coefficients, effects of covariates, and dependencies among the traits (ie, mutual correlation of the traits and genetic correlation of the traits).

As covariates in our multivariate variance components model, we included gender, age, and BMI. Plasma concentrations of IL-6 and TNF- increase with age, and the plasma concentration of IL-6 decreases with increasing BMI, whereas levels of CRP and fibrinogen increase with increasing BMI (Table 2). We observed no significant gender effect for any of the traits (Table 2).

TABLE 2. Heritability Estimates for Inflammatory Cardiovascular Risk Factors

The intraclass correlations in MZ twins were consistently higher than those in DZ twins for all variables (Table 2). The difference was statistically significant for ICAM-1 (P<0.001). Heritability estimates, when adjusting for effects of gender, age, and BMI, and when allowing for possible dependencies between the traits, varied between 0.17 for IL-6 and 0.55 for sICAM-1 (Table 2). Mutual correlations between the traits are given in Table 3. Between the plasma levels we observed, as expected, significantly positive correlations ranged from 0.17 to 0.55, indicating substantial relationship among the traits. Significant genetic correlations, ranging between 0.27 and 0.41, were seen between the cytokines and CRP, fibrinogen, and sICAM-1, indicating a substantial degree of pleiotropy (ie, a gene affecting 2 or more traits ). Small (and not significant) genetic correlations were observed among CRP, fibrinogen, and sICAM-1, which may indicate that the common genetic contribution to their regulation is insignificant.

TABLE 3. Correlations Between Cytokines and Acute Phase Proteins

Polymorphisms in the inflammatory genes (-455G/A in fibrinogen-?, -1059G/C in CRP, K/E469 in ICAM-1, -174C/T in IL-6, and -238G/A in TNF-) were all in Hardy–Weinberg equilibrium, and the frequencies of the alleles (Table 4) were comparable to those reported in other white populations.25,37–40

TABLE 4. Frequencies of Polymorphisms in Inflammation Genes

When we studied the effect of polymorphisms in factors upstream in the regulatory pathway of the cytokines and acute phase proteins, evidence for an association between genotype at the TNF–238G/A polymorphism and IL-6 was found by ANOVA (P=0.02; Table 5). This association was confirmed using the sib-pair–based analysis (QTDT, 2=5.7 ; P=0.02) performed using the genetic model by Fulker et al,35 which takes possible population stratification into account. Associations were also seen for TNF–238G/A polymorphism with CRP levels (P=0.03) and for the fibrinogen–455G/A polymorphism with fibrinogen levels (P=0.03) by ANOVA, but these associations were not confirmed in the subsequent sib-pair–based analysis.

TABLE 5. Relationship Between Polymorphisms and Plasma Protein Concentrations

No evidence for linkage between the polymorphisms and the plasma levels of the inflammation variables was observed, not even between the TNF- polymorphism and IL-6 levels. However, power is limited for linkage studies in our cohort.

Discussion

This study describes that there is a clear genetic component in levels of inflammatory cardiovascular risk factors in the elderly. Selected functional polymorphisms in the genes of fibrinogen-?, CRP, ICAM-1, IL-6, and TNF- only explained a very small part of this variation.

The contribution of genetic variation to plasma levels of fibrinogen in the elderly twins in our study was comparable or somewhat smaller than the genetic contributions published for younger, mostly middle-aged twins. For fibrinogen, the additive genetic variance in our elderly twins (0.21) was lower than the estimates determined in other twin studies, which varied from 0.28 to 0.44.13,16,18 Family studies reported heritabilities for fibrinogen between 0.20 and 0.51,16,41 and in the Kibbutzim Family Study, it was observed that the heritability decreased from 0.48 in 20-year-old individuals to 0.20 in 80-year-old individuals.42 It was expected that the effect of genetic factors could be smaller in the elderly twins because of the presence of many inflammatory triggers, such as cardiovascular disease and other age-associated diseases. The results of our study support this hypothesis, but it remains unclear how much of the difference in heritability can be attributed to age versus other population characteristics.

Only a small, insignificant part of this variance could be explained by the -455G/A polymorphism in the promoter region of the fibrinogen-? gene, in accordance with other reports.43 Polymorphisms in 2000 bp of the promoter region of the fibrinogen-? gene were identified, and in whites, these are in very strong linkage disequilibrium.44 Furthermore, we recently found using luciferase expression studies that the -148C/T polymorphism (which is in perfect linkage disequilibrium with the -455G/A polymorphism in white populations) is the functional polymorphism in the promoter region, explaining all promoter activity.45 Therefore, it is not expected that any new information will become available when other polymorphisms are studied in this region. Also polymorphisms in the promoter region of IL-6 and TNF-, 2 major determinants of fibrinogen synthesis, explained only an insignificant part of the variance in fibrinogen levels.

The genetic contribution to variation in the CRP concentration appears somewhat lower in our study than in middle-aged twins, for whom a clear genetic contribution has been reported in a UK study (intraclass correlation coefficient of 0.57 in MZ twins and 0.31 in DZ twins)12 and in healthy family studies.14,15 Polymorphisms in the promoter region of CRP, IL-6, or TNF- did not explain a major part of this genetic variance in our elderly twins, in contrast to the observation of Vickers et al that the IL-6–174C/T polymorphism explains 14% of the CRP concentration.14 For sICAM-1, no information was published on the heritability of the levels, but the high estimate in our study makes this an interesting candidate gene for further studies. For IL-6, the heritability estimate in our study (0.17) is similar to an estimate of heritability that is available from a family study (0.24).23 For TNF-, the heritability estimate in our study (0.26) was much lower than the estimates derived from family studies (0.68 and >0.80).23,46 The heritability of the in vitro production of TNF- in a whole-blood stimulation experiment was 0.60,47 which is also higher than the heritability we saw in vivo in elderly twins. This suggests that the responsiveness may have a stronger genetic component than the baseline levels.

The hypothesis that responsiveness may be genetically determined suggests that functional polymorphisms in the regulatory elements of inflammatory genes may be determinants of the plasma levels of inflammatory variables. However, the regulatory polymorphisms we studied in the genes for fibrinogen, IL-6, CRP, and TNF- did not explain much of the variance in their gene products, nor did polymorphisms in the IL-6 and TNF- genes explain the variance in the levels of the acute phase proteins. We studied polymorphisms for which an association with levels has been shown repeatedly and is biologically plausible. In addition to the fibrinogen promoter, also for the other factors, the gene and a reasonable part of the promoter region (on average up to 2000 bp) were studied for functional polymorphisms, and we selected polymorphisms for which it was most likely that they would explain a considerable part of the heritability.24,45,48 However, we cannot exclude that other genetic variants have a major impact on the plasma levels of these factors, and studies using full haplotype analysis will be able to give a better estimation of the contribution of genetic variation in a gene to its plasma levels.

We only saw 1 possible association, but no linkage, for any combination of polymorphism and plasma level, indicating that twin pairs that share their alleles are not necessarily close in trait values. This suggests that we did not study the functional polymorphism or that a combination of polymorphisms affects the plasma level. The intriguing question is what the genetic components are if they are not genetic variants in the genes, variations in the promoter regions, or in the regulating cytokine genes. The explanation for the high heritability may therefore be found in enhancer elements further away from the gene or in other genes that determine the regulation of inflammatory factors, such as nuclear proteins. The high genetic correlation between inflammatory variables in our study may indicate a common genetic regulatory mechanism.

BMI has a large genetic component and is also a major determinant of inflammation variables, for example, by increasing the cytokine production by adipose tissue.49 Furthermore, it is possible that the sensitivity of inflammatory variables to BMI is genetically determined. Adjustment for BMI in the statistical analysis may remove such a relationship. Therefore, we analyzed the data unadjusted and adjusted for BMI and noticed that the adjustment had no significant effect on the heritability estimated in our population. This suggests that different gene pools determine the inflammatory variables and the BMI.

When studying the elderly, it cannot be excluded that there has been a selection of our study population because only individuals that survived to be 73 years old were included in the study. However, that is a problem inherent in all aging research, and the present study benefited from a high response rate (79%).

The calculated heritability may be an underestimate of the underlying heritability because the procedure for processing blood was not optimal, which may have given a spuriously greater variation in the levels of the inflammatory markers. Furthermore, the phenotypes show a great day-to-day variation because the inflammatory variables are very sensitive to environmental factors.

In conclusion, levels of inflammatory cardiovascular risk predictors are, to a large extent, genetically determined, even in elderly individuals, and therefore, it is important to elucidate the genetic factors involved in the regulation of these proteins, especially in cardiovascular disease.

Acknowledgments

Funding from the US National Institute on Aging Research, grant NIA-P01-AG08761, and the European Commission, the fifth framework program (QLG2-CT-2002-01254) were received. We would like to thank Anette Larsen, Gunhild Andreasen, Kathrine Overgaard, Jakob Mortensen, and Berit Bents for their technical assistance.

References

Barrett-Connor E, Khaw K. Family history of heart attack as an independent predictor of death due to cardiovascular disease. Circulation. 1984; 69: 1065–1069.

Friedlander Y, Kark JD, Stein Y. Family history of myocardial infarction as an independent risk factor for coronary heart disease. Br Heart J. 1985; 53: 382–387.

Marenberg ME, Risch N, Berkman LF, Floderus B, de Faire U. Genetic susceptibility to death from coronary heart disease in a study of twins. N Engl J Med. 1994; 330: 1041–1046.

Danesh J, Wheeler JG, Hirschfield GM, Eda S, Eiriksdottir G, Rumley A, Lowe GD, Pepys MB, Gudnason V. C-Reactive protein and other circulating markers of inflammation in the prediction of coronary heart disease. N Engl J Med. 2004; 350: 1387–1397.

Danesh J, Muir J, Wong YK, Ward M, Gallimore JR, Pepys MB. Risk factors for coronary heart disease and acute-phase proteins. A population-based study. Eur Heart J. 1999; 20: 954–959.

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. J Am Med Assoc. 1998; 279: 1477–1482.

Ridker PM. Intercellular adhesion molecule (ICAM-1) and the risks of developing atherosclerotic disease. Eur Heart J. 1998; 19: 1119–1121.

Lagrand WK, Visser CA, Hermens WT, Niessen HWM, Verheugt FWA, Wolbink GJ, Hack CE. C-Reactive protein as a cardiovascular risk factor—more than an epiphenomenon? Circulation. 1999; 100: 96–102.

Zwaka TP, Hombach V, Torzewski J. C-Reactive protein-mediated low-density lipoprotein uptake by macrophages: implications for atherosclerosis. Circulation. 2001; 103: 1194–1197.

Mosesson MW, Siebenlist KR, Meh DA. The structure and biological features of fibrinogen and fibrin. Ann N Y Acad Sci. 2001; 936: 11–30.

De Maat MPM, De Bart ACW, Hennis BC, Meijer P, Havelaar AC, Mulder PGH, Kluft C. Interindividual and intraindividual variability in plasma fibrinogen, TPA antigen, plasminogenactivatorinhibitor (PAI) activity, and CRP in healthy, young volunteers and patients with angina pectoris. Arterioscler Thromb Vasc Biol. 1996; 16: 1156–1162.

MacGregor AJ, Gallimore JR, Spector TD, Pepys MB. Genetic effects on baseline values of C-reactive protein and serum amyloid a protein: a comparison of monozygotic and dizygotic twins. Clin Chem. 2004; 50: 130–134.

Retterstol L, Eikvar L, Berg K. A twin study of C-reactive protein compared to other risk factors for coronary heart disease. Atherosclerosis. 2003; 169: 279–282.

Vickers MA, Green FR, Terry C, Mayosi BM, Julier C, Lathrop M, Ratcliffe PJ, Watkins HC, Keavney B. Genotype at a promoter polymorphism of the IL-6 gene is associated with baseline levels of plasma C-reactive protein. Cardiovasc Res. 2002; 53: 1029–1034.

Pankow JS, Folsom AR, Cushman M, Borecki IB, Hopkins PN, Eckfeldt JH, Tracy RP. Familial and genetic determinants of systemic markers of inflammation: the NHLBI family heart study. Atherosclerosis. 2001; 154: 681–689.

de Lange M, Snieder H, Ariens RAS, Spector TD, Grant PJ. The genetics of haemostasis: a twin study. Lancet. 2001; 357: 101–105.

Hamsten A, Iselius L, de Faire U, Blomback M, Humphries SE, Cook M, Dubowitz M, Stirling Y, Meade TW. Genetic and cultural inheritance of plasma fibrinogen concentration. Lancet. 1987; 2: 988–991.

Reed T, Tracy RP, Fabsitz RR. Minimal genetic influences on plasma fibrinogen level in adult males in the NHLBI twin study. Clin Genet. 1994; 45: 71–77.

Castell JV, Andus T, Kunz D, Heinrich PC. IL-6. The major regulator of acute-phase protein synthesis in man and rat. Ann N Y Acad Sci. 1989; 557: 87–99.

Baumann H, Morella KK, Wong GH. TNF-, IL-1 ?, and hepatocyte growth factor cooperate in stimulating specific acute phase plasma protein genes in rat hepatoma cells. J Immunol. 1993; 151: 4248–4257.

Baumann H, Gauldie J. The acute phase response. Immunol Today. 1994; 15: 74–80.

Moshage H. Cytokines and the hepatic acute phase response. J Pathol. 1997; 181: 257–266.

Pantsulaia I, Trofimov S, Kobyliansky E, Livshits G. Genetic and environmental influences on IL-6 and TNF- plasma levels in apparently healthy general population. Cytokine. 2002; 19: 138–146.

Fishman D, Faulds G, Jeffery R, Mohamed-Ali V, Yudkin JS, Humphries S, Woo P. The effect of novel polymorphisms in the IL-6 gene on IL-6 transcription and plasma IL-6 levels, and an association with systemic-onset juvenile chronic arthritis. J Clin Invest. 1998; 102: 1369–1376.

Grove J, Daly AK, Bassendine MF, Day CP. Association of a tumor necrosis factor promoter polymorphism with susceptibility to alcoholic steatohepatitis. Hepatology. 1997; 26: 143–146.

Skytthe A, Kyvik K, Holm NV, Vaupel JW, Christensen K. The Danish Twin Registry: 127 birth cohorts of twins. Twin Res. 2002; 5: 352–357.

Christiansen L, Frederiksen H, Schousboe K, Skytthe A, von Wurmb-Schwark N, Christensen K, Kyvik K. Age and sex differences in the validity of questionnaire-based zygosity in twins. Twin Res. 2003; 6: 275–278.

Thomas A, Lamlum H, Humphries S, Green F. Linkage disequilibrium across the fibrinogen locus as shown by five genetic polymorphisms, G/A-455 (HaeIII), C/T-148 (HindIII/AluI), T/G+1689 (AvaII), and BclI (beta-fibrinogen) and TaqI (alpha-fibrinogen), and their detection by PCR. Hum Mutat. 1994; 3: 79–81.

Zee RY, Ridker PM. Polymorphism in the human C-reactive protein (CRP) gene, plasma concentrations of CRP, and the risk of future arterial thrombosis. Atherosclerosis. 2002; 162: 217–219.

Nishimura M, Obayashi H, Maruya E, Ohta M, Tegoshi H, Fukui M, Hasegawa G, Shigeta H, Kitagawa Y, Nakano K, Saji H, Nakamura N. Association between type 1 diabetes age-at-onset and intercellular adhesion molecule-1 (ICAM-1) gene polymorphism. Hum Immunol. 2000; 61: 507–510.

Wennberg P, Nordstrom P, Lorentzon R, Lerner UH, Lorentzon M. TNF- gene polymorphism and plasma TNF- levels are related to lumbar spine bone area in healthy female Caucasian adolescents. Eur J Endocrinol. 2002; 146: 629–634.

Neale MC, Cardon LR. In: Methodology for Genetic Studies of Twins and Families. Dordrecht: Kluwer Academic Publishers; 1992.

Neale MC. Mx: Statistical Modeling. 3rd ed. Richmond, Va: MCV; 1997.

Abecasis GR, Cardon LR, Cookson WO. A general test of association for quantitative traits in nuclear families. Am J Hum Genet. 2000; 66: 279–292.

Fulker DW, Cherny SS, Sham PC, Hewitt JK. Combined linkage and association sib-pair analysis for quantitative traits. Am J Hum Genet. 1999; 64: 259–267.

Abecasis GR, Cherny SS, Cookson WO, Cardon LR. Merlin–rapid analysis of dense genetic maps using sparse gene flow trees. Nat Genet. 2002; 30: 97–101.

de Maat MP, Bladbjerg EM, Drivsholm T, Borch-Johnsen K, Moller L, Jespersen J. Inflammation, thrombosis and atherosclerosis: results of the Glostrup Study. J Thromb Haemost. 2003; 1: 950–957.

Cao H, Hegele RA. Human C-reactive protein (CRP) 1059G/C polymorphism. J Hum Genet. 2000; 45: 100–101.

Margaglione M, Bossone A, Cappucci G, Colaizzo D, Grandone E, Di Minno G. The effect of IL-6 C/G-174 polymorphism and circulating IL-6 on fibrinogen plasma levels. Haematologica. 2001; 86: 199–204.

Jiang H, Klein RM, Niederacher D, Du M, Marx R, Horlitz M, Boerrigter G, Lapp H, Scheffold T, Krakau I, Gulker H. C/T polymorphism of the intercellular adhesion molecule-1 gene (exon 6, codon 469). A risk factor for coronary heart disease and myocardial infarction. Int J Cardiol. 2002; 84: 171–177.

Freeman MS, Mansfield MW, Barrett JH, Grant PJ. Genetic contribution to circulating levels of hemostatic factors in healthy families with effects of known genetic polymorphisms on heritability. Arterioscler Thromb Vasc Biol. 2002; 22: 506–510.

Friedlander Y, Elkana Y, Sinnreich R, Kark JD. Genetic and environmental sources of fibrinogen variability in Israeli families: the Kibbutzim Family Study. Am J Hum Genet. 1995; 56: 1194–1206.

Lane DA, Grant PJ. Role of hemostatic gene polymorphisms in venous and arterial thrombotic disease. Blood. 2000; 95: 1517–1532.

Green FR. Fibrinogen polymorphisms and atherothrombotic disease. Ann N Y Acad Sci. 2001; 936: 549–559.

Verschuur M, de Jong M, Vos HL, de Maat MP. The 148 C/T variation in the human B fibrinogen promoter influences basal and IL-6 induced transcription. Circulation. 2002; 106: II-80.

Stein CM, Guwatudde D, Nakakeeto M, Peters P, Elston RC, Tiwari HK, Mugerwa R, Whalen CC. Heritability analysis of cytokines as intermediate phenotypes of tuberculosis. J Infect Dis. 2003; 187: 1679–1685.

Westendorp RG, Langermans JA, Huizinga TW, Elouali AH, Verweij CL, Boomsma DI, Vandenbroucke JP, Vandenbrouke JP. Genetic influence on cytokine production and fatal meningococcal disease. Lancet. 1997; 349: 170–173.

Terry CF, Loukaci V, Green FR. Cooperative influence of genetic polymorphisms on IL 6 transcriptional regulation. J Biol Chem. 2000; 275: 18138–18144.

Yudkin JS, Stehouwer CDA, Emeis JJ, Coppack SW. C-Reactive protein in healthy subjects: associations with obesity, insulin resistance, and endothelial dysfunction—a potential role for cytokines originating from adipose tissue? Arterioscler Thromb Vasc Biol. 1999; 19: 972–978.