Does the PI polymorphism alone control alpha-1-antitrypsin expression?

Whether genetic factors other than the protease-inhibitor (PI) polymorphism itself contribute to variation in alpha-1-antitrypsin is of both theoretical and practical interest. We have measured the quantity of alpha-1-antitrypsin (by an immunoturbidometric assay) and its activity (by assaying elastase inhibitory capacity [EIC]) in 583 individuals from 114 twin kinships who were also typed for PI by isoelectric focusing. Models of variation were fitted directly to the raw observations by a maximum-likelihood method. Specification of phenotypic means led to highly significant improvements in fit over models including only individual environment variance and additive genetic variance. The 29 phenotype means could also be described as the appropriate additive combinations of the 12 allelic effects. Only small improvements in fit could then be obtained by addition of polygenic components of variance. We conclude that nearly all genetic variation in alpha-1-antitrypsin quantity and activity can be explained by detectable variation at the PI locus and that this variance is largely additive. Bivariate analysis of alpha-1-antitrypsin and EIC revealed marginal evidence for differences in specific activities of molecules coded by different PI alleles. The correlation between environmental deviations for the two measures was only .63, which may reflect, in part, the rather low reliability of the assays and account for the modest heritabilities (less than .5) of the two measures. An intriguing finding was the presence of significant differences in E1 variance for different PI types, suggesting that different phenotypes have differing capacities to react to environmental challenges.