Nucleotide sequence encoding a slow allele of Adh1 in pearl millet

Abstract

As in maize, two genes encode for alcohol dehydrogenase in pearl millet, Adh-1 and Adh-2. Each gene has two different allelic forms (F and S) in cultivated species. Moreover, a rare C form exists in another type of Pennisetum americanurn [1]. cDNA clones of Adh 1 have been isolated from Massue stock, an imbred line with Adh-lS/ Adh-1 S and Adh-2F/Adh-2F, originating from Bambey (Srnrgal). The cDNA bank was constructed in 2gtl0 from anaerobically treated seedling root mRNAs. Identification was made with pZML793 [2], a maize Adh clone (kindly supplied by Drs. M.M. Sachs and W.I. Peacock, CSIRO, Australia). Seventy clones from 17780 pfu hybridized to the insert isolated from the pZML793 and both strands from the longest insert (MSH10) were M 13-dideoxy sequenced. The insert, 1484 bp, is a full-length coding clone including 5' (96 nucleotides) and 3' (248 nucleotides) non-coding regions (Fig. 1). The coding region shows 92.4~o homology with that of Adhl-S from maize and encodes a polypeptide of 379 amino acids. Alignement of the predicted amino acid sequence with that of the maize Adhl-S is obvious (Fig. 2). Of the triplets which show nucleotide substitution (in the coding region), 72 bear a substitution in the third base and 3 in the first one (T, C and T at positions 439, 490 and 703 respectively) but without any change in the encoded amino acid. The remaining substitutions lead to 11 changes in the nature of the amino acid giving rise to a 97.1 ~o homology at the amino acid level. Of those 11 changes, 9 have already been detected in several species [3] and at the same positions. A simple substitution at position 458 (G!C instead of GAC, Val versus Asp) and a double substitution at 1070-1071 (TTC instead of TAT, Phe versus Tyr) do not seem to have been detected in other Adhl coding regions in maize or other species [3]. These overall amino acid changes do not give rise to a polypeptide whose electrophoretic behaviour is significantly modified; the theoritical pI (from PCgene program, Genofit) is 6.27 and 6.43 for the pearl millet and maize polypeptides respectively. We are currently analysing the F cDNA and the modifications affecting the cDNA sequence of an allele from a pearl millet variant obtained by androgenesis and characterized by a new allele of Adh migrating toward the cathode, different from the S and F forms, and noted C.