Background: Plasma homocyst(e)ine levels (pHo) can be a risk marker for car
diovascular diseases. Different factors affect pHo, but it remains unclear
whether pHo are genetically determined and whether they are related to othe
r risk markers, such as the angiotensin I converting enzyme (ACE) and the p
lasminogen activator inhibitor type-1 (PAI-1),
Methods: We measured fasting pHo, plasma levels of ACE (pACE), and PAI-1 an
tigen (PAI-1:ag) in 60 pairs of healthy monozygotic (MZ) and dizygotic (DZ)
normotensive twins. Twin zygosity was determined with polymerase chain rea
ction analysis of informative minisatellite markers. pHo data were first an
alyzed with TWINAN90 to obtain estimates of genetic variance and heritabili
ty and then examined jointly in a path analysis.
Results: Thirty-one twins were MZ and 29 DZ, The mean pHo were 10.48+/-4.07
mu mol/L (95% confidence interval, 9.73-11.24 mu mol/L). Two pairs had to
be excluded from further analysis because of overt hyperhomocyst(e)inemia r
esulting from concomitant drug treatment, Highly statistically significant
intraclass correlation coefficients were observed both in MZ (r=0.421; P=0.
008) and in DZ (r=0.488; P=0.004). Because all tests of genetic variance an
d heritability were not significant, the hypothesis of genetic variance and
heritability of pHo was rejected. The preferred model of a likelihood-base
d analysis included an additive genetic influence (A), a common environment
al influence (C), and an individually unique environmental influence (E), a
ccounting for 8%, 39%, and 53%, of pHo variance, respectively, No relations
hip between pHo and pACE or PAI-1:ag was detected,
Conclusions: These data do not support the contention that normal-to-border
line elevated pHo of healthy subjects are heritable and under major genetic
influence. They suggest that E and C are far more important than A in dete
rmining pHo variance. Furthermore, they provide no evidence of a relationsh
ip of pHo with pACE and PAI-1:ag.