© 1999 American Heart Association, Inc.
Original Contributions |
Determinants of Plasma Levels of Plasminogen Activator Inhibitor-1
A Study of Normotensive Twins
From the Department of Clinical and Experimental Medicine, Clinica Medica 4 (M.C., G.P.R.) and Institute of Medical Semiotics, II Chair of Internal Medicine, University of Padua Medical School (M.T.S., G.M.P., S.V.), Italy.
Correspondence to G.P. Rossi, MD, FACC, Clinica Medica 4, Department of Clinical and Experimental Medicine, University Hospital, Via Giustiniani, 2, 35126 Padova, Italy. E-mail gprossi{at}ux1.unipd.it
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Abstract—We investigated whether plasma levels of the plasminogen activator inhibitor type 1 antigen (PAI-1:Ag) are genetically determined in monozygotic (MZ) and dizygotic (DZ) twins. Twenty-five pairs of healthy twins underwent measurements of PAI-1:Ag and other variables, including body mass index, mean blood pressure, plasma renin activity, insulin, and glucose. To ascertain the zygosity of twins, highly discriminating micro- and minisatellite systems with variable numbers of tandem repeats were analyzed by PCR amplification followed by polyacrylamide gel electrophoresis. Subjects were also genotyped for the 4G/5G polymorphism by PCR. Estimates of genetic variance and heritability were obtained for PAI-1:Ag, and for body mass index, mean blood pressure, plasma renin activity, glucose, and insulin by jointly examining data in a path analysis with TWINAN90. Results showed that 12 pairs of twins were MZ and 13 were DZ. All tests of genetic variance [within pair (WP): F=6.24, P=0.002; among component (AC): F=2.62, P=0.04; average absolute difference t test=3.00, P=0.004] showed significant genetic variance of PAI-1:Ag, but not of the other variables. Three tests of heritability (WP=0.837, P=0.002; AC=1.791, P<0.05; intraclass correlation: 1.180, P=0.001) consistently showed significant PAI-1:Ag heritability. Additive genetic influences (A), dominance genetic effect (D), and random environmental influences (E) accounted for 0.714, 0.154, and 0.132 of PAI-1:Ag variance, respectively. No effect of different 4G/5G genotypes was found. Thus, these results show significant genetic variance and heritability of PAI-1:Ag and suggest that A is more important than both D and E in determining PAI-1:Ag variance.
Key Words: plasminogen activator inhibitor type 1 • fibrinolysis • gene polymorphism • twins • humans
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Plasminogen activator inhibitor type 1 (PAI-1) is the key regulator1 of the activity of the fibrinolytic system, an important protective mechanism against thrombosis. Reduced fibrinolytic activity, mainly caused by increased plasma levels of PAI-1, is a common finding in many cross-sectional studies of patients with coronary artery disease (CAD)2 and recurrent myocardial infarction (MI),3 4 5 deep vein thrombosis,6 7 and stroke.8 Although extensive investigations have led to identification of the environmental factors regulating PAI-1:Ag and PAI-1 activity (PAI-1:act) levels in plasma (see for review7 9 ), only limited information exists on their genetic determinants. To our knowledge the following 8 polymorphisms of the PAI-1 gene have been identified10 : two (CA)n repeat polymorphisms, a HindIII restriction fragment length polymorphism, an insertion (5G)/deletion (4G) polymorphism on the promoter, ie, -675 bp upstream from the start of transcription of the PAI-1 gene, and 4 polymorphisms in the PAI-1 gene by use of the single-strand conformational polymorphism (SSCP) method.10 A major effect of the 4G/5G polymorphism on PAI-1:act was reported, but conflicting results are also available.9 10 11 12 13 In fact, a relationship between both PAI-1 antigen and activity levels and the 4G/5G polymorphism (the 4G homozygous having the highest mean plasma PAI-1 levels) has been shown in patients with CAD, non-insulin-dependent diabetes mellitus (NIDDM), and healthy control subjects.9 11 12 14 15 However, in contrast with these findings no significant association was found between both PAI-1 antigen and activity levels and the 4G/5G polymorphism or the HindIII RFLP in 189 patients with NIDDM, and between PAI-1:act levels and the 4G/5G polymorphism and four polymorphisms identified in the PAI-1 gene by use of the SSCP method, in healthy men aged 50 to 59 years.10 13 Thus, the issue of whether plasma PAI-1:act and/or plasma PAI-1:Ag are under the effect of a major gene is still under debate.
Studies of twins are increasingly popular in molecular genetics, because the comparison of monozygotic (MZ) and dizygotic (DZ) twins offers a unique and powerful method of partitioning genetic and environmental sources of covariance of quantitative traits which may be relevant for late-onset diseases, such as atherosclerosis, hypertension, and related cardiovascular disease.16 They can also provide estimates of the additive (A) genetic effects, which combine independently over different alleles at the same genetic locus and over different loci, dominant (D) effect, ie, allelic interactions at the same locus, the environmental effects common to cotwins (C, ie, shared family environment), and the environmental (E) variance unique to individuals on the variance of quantitative traits such as the components of the clotting cascade. Therefore, this study was undertaken to investigate the genetic and environmental components of the plasma levels of PAI-1 in identical and fraternal normotensive twins.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Twins
The twins investigated in this study were enrolled from an Italian association of twins based in Padua, Italy. To ascertain whether twins were MZ or DZ, some highly discriminating micro- and minisatellite systems with a variable number of tandem repeats were analyzed. Three amplified fragment-length polymorphisms (ApoB, D1S80, and YNZ22) and 4 short tandem repeats (HUMACTPB2, HUMTH01, HUMFES/FPS, and HUMMVWA31) were assessed by separate PCR amplifications of genomic DNA followed by polyacrylamide gel electrophoresis and silver staining. No subject was smoking or taking any drugs or medications known to affect the fibrinolytic tests for at least 15 days before the study. None participated in any competitive sports training or reported daily intake of alcohol. All subjects were healthy and showed normal fasting levels of serum glucose and insulin. Blood pressure (BP) was measured with a mercury sphygmomanometer using Korotkoff phase V for diastolic and according to the WHO guidelines. All gave an informed written consent, and the Institutional Human Subjects Review Committees of the University of Padua Medical School approved the study protocol.
Blood Sampling and Biochemical Measurements
Blood was collected after an overnight fast and 15-minute supine
rest without stasis from an antecubital vein. Five milliliters of whole
blood with 100 µL Na2EDTA 6% was immediately
put on ice. After centrifugation at 3000g
(at 4°C for 10 minutes), separated aliquots of plasma and buffy coat
were stored at -40°C and tested within 1 month. Plasma renin
activity was measured by a commercially available kit (Ares Serono,
Milan, Italy, supine normal values with a daily sodium intake of 100 to
200 mmol: 0.51 to 2.64 ng Ang I ·
mL-1 · h-1) as
reported.17 PAI-1:Ag levels were assayed by an ELISA
method (TintElize PAI-1, Biopool, Umeå, Sweden) following the
manufacturer's protocols18 (normal range 4 to 30
ng/mL). The within and between assay coefficients of variation of this
assay were 1.9% and 2.4%, respectively.
Extraction of DNA and 4G/5G Genotyping
DNA was isolated from 50 µL of blood with use of the DNA Blood
Extraction Fast Kit(TM) (AB Analitica Srl, Italy)
following the manufacturer's instructions. Genomic DNA was amplified
with allele-specific PCR following the procedure described by Falk
et al19 with minor modifications. In particular,
allele-specific oligonucleotides
5'-GTCTGGACACGTGGGGG-3' (PAI 5G) or 5'-GTCTGGACACGTGGGG-3' (PAI 4G)
coupled with a common upstream primer (PAI 2d:
5'-TGCAGCCAGCCACGTGATTGTCTAG-3') were used. A common upstream primer
(PAI 1u: 5'-AAGCTTTTACCATGGTAACCCCTGGT-3') was used to generate a
positive control in each PCR. The reaction was performed in a total
volume of 25 µL with 3 µL of extracted DNA. The mix contained
50 mmol/L KCl, 10 mmol/L Tris-HCl (pH 9), 2.5 mmol/L
MgCl2, 200 µmol/L dNTPs, 3 pmol primer PAI
1u, 25 pmol PAI 2d, and 50 pmol PAI 4G or PAI 5G. For each PCR, 1.5 U
Taq DNA polymerase (Pharmacia Biotech, Uppsala, Sweden) was
used after a hot start. The mixture was subjected to the following
cycling steps: 96°C, 5 minutes for 1 cycle; 94°C, 1 minute, 72°C,
2 minutes for 5 cycles; 94°C, 1 minute, 65°C, 45 seconds, 72°C, 1
minute 15 seconds for 25 cycles; 65°C 5 minutes for 1 cycle. The PCR
products were run in TBE 0.5x buffer in a 4% agarose gel stained
with ethidium bromide.
Statistical Analysis
Results are expressed as mean±SD, or SEM, as
appropriate.20 Comparison of subjects of the different
genotypes was carried out with Kruskall–Wallis test for plasma
renin activity, or with a 1-way ANOVA followed by Bonferroni's
multiple comparison test for the variables that showed a normal
distribution. Because subjects in each twin pair could not be regarded
as independent unrelated individuals, either twin 1 or twin 2 of each
pair was considered for statistical comparison. Analysis was
carried out with the SPSS for Windows(TM) statistical package (version
7.5, SPSS Inc, Chicago, Ill). A P value <0.05 was
considered statistically significant. Analysis of twin data was
carried out jointly with TWINAN90 (a generous gift of Prof. C.J.
Williams), a program specifically developed for conducting
analyses of twin data.21 To this end the
consistency of the variable of interest with the normal
distribution assumption and with the hypothesis of equal variance
between zygosity was verified beforehand. Estimates of genetic
variance, including the within-pair (WP) and among components (AC),
were attained thereafter, and a test for genetic variance based on the
average absolute difference between twins,22 which is
standardized to yield an approximate t test for the null
hypothesis of no genetic variance, was performed. Three estimates of
heritability were also calculated; the first two were derived from the
WP and AC estimates of genetic variance, respectively, according to the
following
equations: WP=2x(WDZ-WMZ)/(SMZ+SDZ)/4
and AC=(2AC)/(SMZ+SDZ)/4.
The third estimate of heritability (h2) is based on intrapair correlation coefficients calculated from the MZ and DZ twins as follows: h2=2(rMZ-rDZ).
This approach allows the determination of the proportion of variance of PAI-1:Ag accounted for by an additive genetic influence (A) which combines independently over different alleles at the same genetic locus and over different loci, nonadditive genetic influences (D) (ie, allelic interactions at the same locus), environmental influences shared by cotwins within a family (C) and unique to individuals (E).21 This ANOVA-based method was chosen because it was more accurate than path-maximum-likelihood methods to estimate and test the significance of A because it does not require the assumption that C=0.22 Finally, the results of the likelihood-based analyses were determined for models with A, D, and E (ADE); A, C, and E (ACE); A and E (AE), C and E (CE), and E.
This type of analysis was carried out also for body mass index (BMI), mean blood pressure (MBP), glycemia, and plasma fasting insulin.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
After ascertainment of zygosity, 12 twins were determined to be MZ and 13 DZ. The observed 4G/5G PAI-1 genotype distribution was in agreement with the Hardy–Weinberg proportion (4/4=0.14, 4/5=0.54, 5/5=0.32) in the whole population, as well as in the MZ and DZ subgroups (all P=NS by
2 test).
Their demographic and biochemical characteristics, including PAI-1:Ag
levels, are shown in Table 1
. No
significant difference between MZ and DZ twins and among different
4G/5G PAI-1 genotypes in all variables of interest was
found (Table 1).
|
Joint analysis of the MZ and DZ twins data was carried
out for the variable PAI-1:Ag with TWINAN90.21 A
Kolmogorov–Smirnov goodness-of-fit test applied separately to the
grouped pairs for each of the two zygosities confirmed that PAI-1:Ag
levels followed a normal distribution (P>0.15 for both
zygosities), and therefore no transformation of the data was needed.
The null hypothesis that the mean of the MZ and DZ twins absolute
difference did not differ from zero was thereafter accepted
(t=1.62, with 18.7 degree of freedom, NS) and the equal
variance hypothesis was not rejected (F=0.61, NS). Estimates of genetic
variance and intraclass correlation coefficients for both MZ and DZ
twins were obtained, and a test for genetic variance was performed
(Table 2). Both the preferred tests of
genetic variance WP and average absolute difference
tests21 and also the AC estimate resulted in being
statistically significant. The three estimates of heritability were
also calculated (Table 2), and the results were significant,
thereby leading to acceptance of the hypothesis that PAI-1:Ag
levels are heritable.21
|
The relationship between PAI-1:Ag in each pair of identical and
fraternal twins is shown as a scatter plot (Figure 1). A higher correlation coefficient was
observed in MZ (r=0.935, P<0.0001) than in DZ
(r=0.345, NS) twins. The twin pair difference in PAI-1:Ag in
MZ, and in DZ concordant and discordant for the 4G/5G genotype
was also examined (Figure 2). To further
investigate the components of PAI-1:Ag variance, the results of the
likelihood-based analyses for different models (ADE, ACE, and
the AE, CE, and E) were determined. Since the correlation coefficient
of DZ twins was less than half of that of MZ twins the ADE was the
preferred model (Table 2
). The latter indicated that the larger
proportion of PAI-1:Ag variance was caused by A, whereas smaller
proportions were accounted for by D and E.
|
|
The same statistical analysis applied to the same data set did not reveal any evidence of genetic variance and heritability for serum glucose, insulin, and mean blood pressure. Statistically significant correlation coefficients for mean blood pressure and serum glucose, both in MZ and DZ twins, and in either MZ or DZ twins (for body mass index and serum insulin, respectively) were found. However, all tests of genetic variance and heritability were not statistically significant for serum glucose, insulin, and mean blood pressure. Only BMI showed a statistically significant AC genetic variance and heritability (P=0.011) and a borderline significant intraclass correlation coefficient (P=0.042).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In this study, to gain insight into the relative effects of genes and environment on PAI-1:Ag, we compared the plasma levels of PAI-1:Ag of fraternal (DZ) and identical (MZ) twins selected from a Caucasian population, who were young, healthy, and normotensive and had normal plasma levels of PAI-1:Ag. Their zygosity status was carefully established both at the phenotypic and at the molecular level by assessing some highly discriminating variable number of tandem repeats micro- and minisatellite systems. In addition utmost care was also taken to ensure an accurate measurement of PAI-1:Ag both during sample collection and at assay. Data were analyzed with a strategy selected because it takes into consideration the contribution of shared environment and therefore allows investigation of the proportion of variance of PAI-1:Ag accounted for by an additive genetic influence (A), nonadditive genetic influences (D) (ie, allelic interactions at the same locus), environmental influences shared by cotwins within a family (common, C) and unique to individuals (E). At variance with a recent study of Swedish twins investigating the genetic influences of PAI-1:act with a multivariate genetic technique (LISREL),23 we used TWINAN90, an ANOVA-based method specifically developed for conducting analyses of twin data by Williams et al,21 because this approach was found to be more accurate than path-maximum-likelihood methods for estimating and testing the significance of A.22
By using this strategy, we found strong evidence of genetic variance
and heritability of PAI-1:Ag in these subjects (Table 2). When
phenotypic covariance was partitioned into additive genetic
effects (A), dominant genetic components (D), environmental effects
common to cotwins (C), and error variance (E) components, the estimates
indicated that more than 85% was accounted for by A and D (Table 2), thereby suggesting that most of PAI-1:Ag variance is
genetically determined. This contention is also supported by the
finding of high and statistically significant correlation coefficients
only in MZ and not in DZ twins. These results concerning PAI-1:Ag are
in sharp contrast with those concerning other demographic, biochemical,
and hemodynamic variables, which are of potential
interest for cardiovascular risk, such as fasting serum
glucose, insulin, and mean blood pressure.
In a recent study of white nuclear families from a healthy French population, variables related to the insulin resistance syndrome were found to explain a major part of PAI-1 variance (49% in fathers and 29% in mothers), whereas five PAI-1 gene polymorphisms accounted for a negligible proportion (3% in women and none in men) of PAI-1 variability.24 These findings differed from those of a Swedish study of twins in which the proportion of PAI-1:act levels accounted for by heritability and individual-specific environmental factors was 42% and 36%, respectively.23 Although these differences could be caused by several methodological factors, such as measurement of PAI-1:Ag versus PAI-1:act, study design, statistical technique used, in our view the most likely explanation relates to the different selection criteria of the study populations. It is in fact conceivable that our selection of young healthy normotensive twins with normal PAI-1 levels might have led to higher estimates of genetic variance, compared with those studies in which older subjects with higher BMI values and a much wider dispersion of PAI-1 levels were investigated.23 24
The 4G/5G PAI-1 polymorphism has been proposed as a determinant of
the plasma levels of PAI-1:Ag. It has also been put forward as a risk
factor for acute myocardial infarction in patients with angina pectoris
and NIDDM,11 12 25 although the prognostic role of the
4G/5G polymorphism in the development of MI is still under
debate.9 26 27 When classified by 4G/5G genotype,
our twins were demographically similar and had superimposable
blood pressure values (Table 1). The 4G or 5G allele
frequency was consistent with the Hardy–Weinberg proportion,
both in the whole twins population sample and when the twins were
classified according to their zygosity. Although this finding makes a
selection bias unlikely and suggests that the present results can
be representative of the general population, it must be
acknowledged that the observed 4G/5G genotype distribution in
our series corresponds with a 4G allele frequency of 0.41, which is
somewhat lower than the frequency of 0.54 to 0.56 reported in most
studies in Caucasian populations (see for review9 26 ).
Nevertheless, the present results do not support the contention
that 4G/5G polymorphism plays a major role in determining PAI-1:Ag,
because no significant difference between different 4G/5G
genotypes was found.
Were the 4G/5G polymorphism exerting a major effect on PAI-1:Ag,
the WP difference of 4G/5G concordant DZ twins would be expected to be
close to that of MZ twins and lower than that of 4G/5G discordant DZ
twins. In contrast, we found that the WP difference of 4G/5G concordant
DZ twins differs significantly from that of MZ and was similar to that
of DZ discordant twins (Figure 2). This finding suggests a role
of additional, possibly still unidentified polymorphisms on the
PAI-1 locus and/or of other quantitative trait loci contributing
significantly to PAI-1:Ag, an issue which obviously deserves further
specific research.
In conclusion, our results demonstrate that investigation of twins provides a useful model to assess the relative effects of environmental and genetic factors on quantitative traits such as PAI-1:Ag. Besides confirming that PAI-1:Ag levels are genetically determined, they indicate a major effect of A, ie, of an additive genetic influence which combines independently over different alleles at the same genetic locus and over different loci, and minor influences of both dominance genetic effect (ie, allelic interactions at the same locus) and environmental influence unique to individuals.
Received April 27, 1998; accepted July 1, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1. Chandler WL. A kinetic model of the circulatory regulation of tissue plasminogen activator. Thromb Haemost.. 1991;66:321–328.[Medline] [Order article via Infotrieve]
2. Wiman B. Plasminogen activator inhibitor 1 (PAI-1) in plasma: its role in thrombotic disease. Thromb Haemost.. 1995;74:71–76.[Medline] [Order article via Infotrieve]
3.
Hamsten A, Blomback M, Wiman B, Svensson J, Szamosi A,
de Faire U, Mettinger L. Haemostatic function in myocardial infarction.
Br Heart J.. 1986;55:58–66.
4. Hamsten A, Wiman B, de Faire U, Blomback M. Increased plasma levels of a rapid inhibitor of tissue plasminogen activator in young survivors of myocardial infarction. N Engl J Med. 1985;313:1557–1563.[Abstract]
5. Hamsten A, de Faire U, Walldius G, Dahlen G, Szamosi A, Landou C, Blomback M, Wiman B. Plasminogen activator inhibitor in plasma: risk factor for recurrent myocardial infarction. Lancet. 1987;2:3–9.[Medline] [Order article via Infotrieve]
6. Juhan-Vague I, Valadier J, Alessi MC, Aillaud MF, Ansaldi J, Philip-Joet C, Holvoet P, Serradimigni A, Collen D. Deficient t-PA release and elevated PA inhibitor levels in patients with spontaneous or recurrent deep venous thrombosis. Thromb Haemost. 1987;57:67–72.[Medline] [Order article via Infotrieve]
7. Wiman B, Hamsten A. Impaired fibrinolysis and risk of thromboembolism. Prog Cardiovasc Dis. 1991;34:179–192.[Medline] [Order article via Infotrieve]
8.
Fisher M, Francis R. Altered coagulation in cerebral
ischemia. Platelet, thrombin, and plasmin activity.
Arch Neurol. 1990;47:1075–1079.
9. Ye S, Green FR, Scarabin PY, Nicaud V, Bara L, Dawson SJ, Humphries, SE, Evans A, Luc G, Cambou JP, et al. 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. Etude CasTemoins de I'nfarctus du Mycocarde. Thromb Haemost. 1995;74:837–841.[Medline] [Order article via Infotrieve]
10.
Henry M, Chomiki N, Scarabin PY, Alessi MC, Peiretti F,
Arveiler D, Ferrieres J, Evans A, Amouyel P, Poirier O, Cambien F,
Juhan-Vague I. Five frequent polymorphisms of the PAI-1 gene: lack
of association between genotypes, PAI activity, and
triglyceride levels in a healthy population.
Arterioscler Thromb Vasc Biol. 1997;17:851–858.
11.
Ossei-Gerning N, Mansfield MW, Stickland MH,
Wilson IJ, Grant PJ. Plasminogen activator
inhibitor-1 promoter 4G/5G genotype and plasma
levels in relation to a history of myocardial infarction in patients
characterized by coronary angiography. Arterioscler
Thromb Vasc Biol. 1997;17:33–37.
12.
Eriksson P, Kallin B, van't Hooft FM, Bavenholm P,
Hamsten A. Allele-specific increase in basal transcription of the
plasminogen-activator inhibitor 1
gene is associated with myocardial infarction. Proc Natl Acad Sci
U S A. 1995;92:1851–1855.
13. Mansfield MW, Stickland MH, Grant PJ. Environmental and genetic factors in relation to elevated circulating leels of plasminogen activator inhibitor-1 in Caucasian patients with non-insulin-dependent diabetes mellitus. Thromb Haemost. 1995;74:842–847.[Medline] [Order article via Infotrieve]
14.
Dawson S, Hamsten A, Wiman B, Henney A, Humphries S.
Genetic variation at the plasminogen activator
inhibitor-1 locus is associated with altered levels of
plasma plasminogen activator
inhibitor-1 activity. Arterioscler Thromb. 1991;11:183–190.
15. Panahloo A, Mohamed-Ali V, Lane A, Green F, Humphries SE, Yudkin JS. Determinants of plasminogen activator inhibitor 1 activity in treated NIDDM and its relation to a polymorphism in the plasminogen activator inhibitor 1 gene. Diabetes. 1995;44:37–42.[Abstract]
16.
Ansley D. Aging twins offer clues to late-onset
disease. Science. 1993;259:1826–1828.
17. Rossi GP, Pavan E, Chiesura-Corona M, Bader M, Paganini G, Cesari M, De Toni R, Feltrin GP, Ganten D, Pessina AC. Renovascular hypertension with low-to-normal plasma renin: clinical and angiographic features. Clin Sci. 1997;93:435–443.[Medline] [Order article via Infotrieve]
18.
Declerck PJ, Alessi MC, Verstreken M, Kruithof EK,
Juhan-Vague I, Collen D. Measurement of plasminogen
activator inhibitor 1 in biologic fluids with a
murine monoclonal antibody-based enzyme-linked immunosorbent assay.
Blood. 1988;71:220–225.
19. Falk G, Almquist A, Nordenhem A, Svensson H, Wiman B. Allele specific PCR for detection of a sequence polymorphism in the promoter region of the plasminogen activator inhibitor-1 (PAI-1) gene. Fibrinolysis. 1995;9:170–174.
20. Rosner B. Fundamentals of Biostatics. 2nd ed. Boston, Mass: PWS Publisher; 1996.
21. Williams CJ, Christian JC, Norton JAJ. TWINAN90: a FORTRAN program for conducting ANOVA-based and likelihood-based analyses of twin data. Comput Methods Programs Biomed. 1992;38:167–176.[Medline] [Order article via Infotrieve]
22. Christian JC, Norton JAJ, Sorbel J, Williams CJ. Comparison of analysis of variance and maximum likelihood based path analysis of twin data: partitioning genetic and environmental sources of covariance. Genet Epidemiol. 1995;12:27–35.[Medline] [Order article via Infotrieve]
23.
Hong Y, Pedersen NL, Egberg N, de Faire U. Moderate
genetic influences on plasma levels of plasminogen
activator inhibitor-1 and evidence of genetic
and environmental influences shared by plasminogen
activator inhibitor-1,
triglycerides, and body mass index. Arterioscler
Thromb Vasc Biol.. 1997;17:2776–2782.
24.
Henry M, Tregouët DA, Alessi MC, Aillaud MF,
Visviskis S, Siest G, Tiret L, Juhan-Vague I. Metabolic
determinants are much more important than genetic polymorphisms in
determining the PAI-1 activity and antigen plasma concentrations.
Arterioscler Thromb Vasc Biol. 1998;18:84–91.
25. Mansfield MW, Stickland MH, Grant PJ. Plasminogen activator inhibitor-1 (PAI-1) promoter polymorphism and coronary artery disease in non-insulin-dependent diabetes. Thromb Haemost. 1995;74:1032–1034.[Medline] [Order article via Infotrieve]
26.
Ridker PM, Hennekens CH, Lindpaintner K, Stampfer MJ,
Miletich JP. Arterial and venous thrombosis is not
associated with the 4G/5G polymorphism in the promoter of the
plasminogen activator inhibitor
gene in a large cohort of US men. Circulation. 1997;95:59–62.
27.
Mannucci PM, Mari D, Merati G, Peyvandi F, Tagliabue L,
Sacchi E, Taioli E, Sansoni P, Bertolini S, Franceschi C. Gene
polymorphisms predicting high plasma levels of coagulation and
fibrinolysis proteins. A study in centenarians.
Arterioscler Thromb Vasc Biol. 1997;17:755–759.
This article has been cited by other articles:
 |
|
 |
|
  M. Cesari, K. Narkiewicz, R. De Toni, E. Aldighieri, C. J. Williams, and G. P. Rossi Heritability of Plasma Adiponectin Levels and Body Mass Index in Twins J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3082 - 3088. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Katoh, M. Lehtovirta, J. Kaprio, V. Harjutsalo, M. Koskenvuo, J. Eriksson, N. Tajima, and J. Tuomilehto Genetic and Environmental Effects on Fasting and Postchallenge Plasma Glucose and Serum Insulin Values in Finnish Twins J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2642 - 2647. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Dunn, R. A. Ariens, M. de Lange, H. Snieder, J. H. Turney, T. D. Spector, and P. J. Grant Genetics of fibrin clot structure: a twin study Blood, March 1, 2004; 103(5): 1735 - 1740. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Festa, R. D'Agostino Jr, S. S. Rich, N. S. Jenny, R. P. Tracy, and S. M. Haffner Promoter (4G/5G) Plasminogen Activator Inhibitor-1 Genotype and Plasminogen Activator Inhibitor-1 Levels in Blacks, Hispanics, and Non-Hispanic Whites: The Insulin Resistance Atherosclerosis Study Circulation, May 20, 2003; 107(19): 2422 - 2427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. L. Anderson and N. G. Anderson The Human Plasma Proteome: History, Character, and Diagnostic Prospects Mol. Cell. Proteomics, November 1, 2002; 1(11): 845 - 867. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Festa, R. D'Agostino Jr, R. P. Tracy, and S. M. Haffner Elevated Levels of Acute-Phase Proteins and Plasminogen Activator Inhibitor-1 Predict the Development of Type 2 Diabetes: The Insulin Resistance Atherosclerosis Study Diabetes, April 1, 2002; 51(4): 1131 - 1137. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Freeman, M. W. Mansfield, J. H. Barrett, and P. J. Grant Genetic Contribution to Circulating Levels of Hemostatic Factors in Healthy Families With Effects of Known Genetic Polymorphisms on Heritability Arterioscler. Thromb. Vasc. Biol., March 1, 2002; 22(3): 506 - 510. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Lane and P. J. Grant Role of hemostatic gene polymorphisms in venous and arterial thrombotic disease Blood, March 1, 2000; 95(5): 1517 - 1532. [Full Text] [PDF]
|
|
|
|
|
|
M. Cesari and G. P. Rossi Plasminogen Activator Inhibitor Type 1 in Ischemic Cardiomyopathy Arterioscler. Thromb. Vasc. Biol., June 1, 1999; 19(6): 1378 - 1386. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||