The use of measured genotype information in the analysis of quantitative phenotypes in man. III. Simultaneous estimation of the frequencies and effects of the apolipoprotein E polymorphism and residual polygenetic effects on cholesterol, betalipoprotein and triglyceride levels.

We have begun a measured genotype approach to the genetic analysis of lipid and lipoprotein variability. This approach enables one to simultaneously estimate the frequencies and effects of alleles at specific loci along with the residual polygenetic variance component. In this study we consider the contribution of three common alleles at the locus coding for apolipoprotein E to interindividual variation of total cholesterol, betalipoprotein, and triglyceride levels. A sample of 102 nuclear families consisting of 434 individuals was studied. The frequencies of the epsilon 2, epsilon 3, and epsilon 4 alleles in this sample are 0.137, 0.740, and 0.123, respectively. In separate analyses of cholesterol and betalipoprotein levels, a complete model that includes the effects of the six apo E genotypes, unmeasured polygenes, and individual specific environmental effects fits these data significantly better than a reduced model that does not include the effects of the apo E polymorphism or a reduced model that does not include the effects of polygenes. On the average the epsilon 2 allele lowers total cholesterol and betalipoprotein levels by 0.425 mmol/l and 0.811 units, respectively. The epsilon 4 allele is associated with an average increase of these phenotypes by 0.255 mmol/l and 0.628 units, respectively. Simultaneous estimates of the interindividual variability of total cholesterol levels attributable to the apo E polymorphism and to residual polygenic effects are 8% and 56%, respectively. For betalipoprotein levels, we simultaneously estimate these values to be 7% and 42%, respectively. A reduced model including the effects of polygenes but not the effects of the apo E polymorphism fitted the triglyceride data as well as the complete model. The estimate of the fraction of interindividual variability associated with polygenetic effects was 26.5%. We review our present understanding of the genetic architecture underlying variability of cholesterol levels in the population at large and infer that the majority of the genetic variability may be accounted for by polymorphic gene loci with moderate effects on cholesterol levels.