DUPLICATION OF STRUCTURAL GENES FOR HEMOGLOBIN α AND β CHAINS IN MAN *

The prevalence of multiple loci governing a-chain synthesis in many other animal species has been adequately documented by earlier papers in this volume. Until rather recently the world of the human geneticist has been relatively simple, there being only one structural locus for each of the six types of polpeptide chains identified in man-the a, p, y, 6, E, and 2 chains. The work of Huisman and Schroeder and their colleagues1 assures us that there are at least two and perhaps four loci for the y chains of fetal hemoglobin; one or more of these have glycine at residue 136 and one or more have alanine. They further propose the existence of major and minor loci for each of these types.z The possibility of two Hb, loci was first raised by Lehmann and Carrella who drew attention to the fact that heterozygotes for many a-chain variants possess approximately 20% of the abnormal component, whereas heterozygotes for mutants of the &chain structural locus possess 30 to 40% of the variant. The proportions of abnormal components present in published aand b-chain mutants are shown in TABLE 1. ' With some notable exceptions the proportion of abnormal a-chain variants vary between 15 and 35%. A few variants fall below this range; for instance, Hb Ann Arbor, comprising 2 to 4%: Hb tor in^,^ comprising 6 to lo%, and Hb Bibba,6 comprising 11 % of the total hemoglobin, are unstable and are probably denatured within the red cell. At the other extreme Hb G Philadelphia and Hb G Chinese comprise as much as 40 to 50% of the total hemoglobin. The majority of the @-chain variants vary between 25 and 40% of the total hemoglobin. Again, variants such as Hb Geneva? comprising 25%, Hb Zurich,Io comprising 20-36% Hb Sabine," comprising 8% and Hb Ko1n,I2 comprising 10-15% of the total hemoglobin, are unstable. However, in no less than sixteen a-chain variants, the proportion of abnormal hemoglobin is 50% or more of the total hemoglobin. Published data must be interpreted cautiously, however, because, aside from the variation imposed by instability, a wide variety of techniques were employed in the various studies. Moreover, since there is probably only a single Hbs locus, in all likelihood the wide range of values among jl-chain variants also reflects additional intrinsic differences in gene expression from one variant to the next. Taking all of these factors into account, however, the differences in the expression between a-chain and @-chain variants seems valid. One possible explanation for at least part of the variability is that a-chain mutants are more sensitive to amino acid substitution. That possibility has not been excluded. Nevertheless, in a number of mutants the qualitative abnormali-