Three human alcohol dehydrogenase subunits: cDNA structure and molecular and evolutionary divergence

Class I human alcohol dehydrogenase (ADH; alcohol:NAD⁺ oxidoreductase, EC 1.1.1.1) consists of serveral homo- and heterodimers of α, β, and γ subunits that are governed by the ADH1, ADH2, and ADH3 loci. We previously cloned a full length of cDNA for the β subunit, and the complete sequence of 374 amino acid residues was established. cDNAs for the α and γ subunits were cloned and characterized. A human liver cDNA library, constructed in phage λgt11, was screened by using a synthetic oligonucleotide probe that was matched to the γ but not to the β sequence. Clone pUCADHγ21 and clone pUCADHα15L differed from β cDNA with respect to restriction sites and hybridization with the nucleotide probe. Clone pUCADHγ21 contained an insertion of 1.5 kilobase pairs (kbp) and encodes 374amino acid residues compatible with the reported amino acid sequence of the γ subunit. Clone pUCADHα15L contained an insertion of 2.4 kbp and included nucleotide sequences that encode 374 amino acid residues for another subunit, the α subunit. In addition, this clone contained the sequences that encode the COOH-terminal part of the β subunit at its extended 5' region. The amino acid sequences and coding regions of the cDNAs of the three subunits are very similar (~93-95% identity). A high degree of resemblance is observed also in their 3' noncoding regions. However, distinctive differences exist in the vicinity of the Zn-binding cysteine residue at position 46 - i.e., Cys-Gly-Thr in the α, Cys-Arg-Thr in the wild-type β₁, Cys-His-Thr in the Oriental-type β₂, and Cys-Arg-Ser in the γ, reflecting the differences in their kinetic properties. Based on the cDNA sequences and the deduced amino acid sequences of the three subunits, their structural and evolutionary relationships are discussed.