Endophenotypes, Heritability, and Underlying Complexity in Hypertension

The article by Watkins et al.1 in this issue reports that heritability of plasma angiotensinogen (AGT) is extremely high in a population of European descent from Utah, and that much of this heritability may be accounted for by genetic variation in the AGT gene. Of interest is the fact that although the authors also report that blood pressure is heritable, it is much less so than plasma AGT, and there is no correlation between AGT and blood pressure—a somewhat unexpected result. Taken together, these data address several issues that have made the identification of genetic factors that predispose to hypertension problematic. In standard analyses of the genetics of hypertension, it is generally assumed that DNA variants alone can map to a clinical phenotype—i.e., increased blood pressure. This ignores the fact that multiple layers of genetic and protein regulation often intervene in the genotype–phenotype map. For example, interactions among genes (epistasis) can mean that what is encoded in the genome at a putatively critical gene cannot explain clinical outcome because its effect is modified by other gene products.2,3 However, the authors posit and demonstrate such DNA variants may be very good predictors of other phenotypes, such as protein concentrations. The relationship between endophenotypes and the clinical phenotype and what impacts that relationship then becomes key.