Genetic epidemiological approaches to the study of risk factors for cardiovascular diseases

Abstract

Cardiovascular disease (CVD) is one of the most common causes of death in most developed countries. Genetic and environmental factors influence CVD and its risk factors in a complex fashion. Parallel to this, new statistical methods are being developed to make use of the new molecular genetic information. The aim of this thesis was focused on developing and implementing these new statistical techniques and applying them to twin data from the population- based Swedish Twin Registry in order to study risk factors of CVD.

In the first study we investigated genetic and environmental effects in the variation of lipids andapolipoproteins in 725 twin pairs. Structural equation modelling, which also tested for age and gendereffects were applied to the data. Heritabilities for lipids and apolipoproteins ranged from 35-74%.Consistent sex differences were found in triglycerides. Total phenotypic variation increased across the age groups for Cholesterol and Apolipoprotein B due to an increase in unique environmental variance components, which is probably due to the accumulation of environmental experiences throughout life. In contrast, in Apolipoprotein At variance was highest in the middle age group with changes due to differences in genetic variance. This could imply that different genetic mechanisms might act at different time points in life.

In the second study we investigated genetic and environmental effects in repeated measures of diastolic and systolic blood pressure in a sample of 298 twin pairs. With the use of a Cholesky decomposition model we found that genetic influences were stable over time and they explained up to 46% of the phenotypic variance in diastolic and 63% of the phenotypic variance in systolic blood pressure. We also investigated the association between blood pressure and two polymorphisms, the angiotensin-I converting enzyme gene (ACE) and the angiotensin II type I receptor (AT1R) gene with a novel approach that simultaneously tests for linkage and association. The method tests whether the polymorphisms are the true QTLs or in linkage disequilibrium with the true QTLs. No association was found.

Linkage to obesity has been previously reported on chromosome 2 and 10. In the third study we investigated the findings in candidate regions on Chromosome 2 and 10 in a selected (based on their phenotypic value) sample of Swedish twin pairs. It is shown that sampling siblings in that way is a more cost-effective approach of doing linkage. We implemented a "combined" Haseman-Elston approach that is especially powerful and developed for studying selected samples. No linkage was confirmed in the candidate regions of chromosome 2 and 10.

The Barker hypothesis states that metabolic disorders such as cardiovascular disease, blood pressure and diabetes begin as a sub optimal or abnormal development during foetal and early neonatal life. There have been criticisms to the Barker hypothesis, above all because the genetic and early environmental influences could be responsible for previously observed associations. Twins are ideal for testing these alternative hypotheses. We investigated the relationship between self-reported birth weight and Type II diabetes, first in a cohort of 11226 same-sexed Swedish twins, and secondly in 142 pairs discordant for Type II diabetes by utilizing the co-twin control method. We found a direct effect of low birth weight on Type II diabetes that is independent of genetic and early environmental effects. In conclusion, the incorporation of molecular data in order to disentangle the complexity underlying biological mechanisms and the statistical models developed to accompany them will in the future be a challenging task.