From a mid-maturation seed cDNA library we have isolated cDNA clones encoding two Triticum aestivum puroindolines. Puroindoline-a and puroindoline-b, which are 55% similar, are basic, cystine-rich and tryptophan-rich proteins. Puroindolines are synthesized as preproproteins which include N- and C-terminal propeptides which could be involved in their vacuolar localization. The mature proteins have a molecular mass of 13 kDa and a calculated isoelectric point greater than 10. A notable feature of the primary structure of puroindolines is the presence of a tryptophan-rich domain which also contains basic residues. A similar tryptophan-rich domain was found within an oat seed protein and a mammalian antimicrobial peptide. The ten cysteine residues of puroindolines are organized in a cysteine skeleton which shows similarity to the cysteine skeleton of other wheat seed cystine-rich proteins. Northern blot analysis showed that puroindoline genes are specifically expressed in T. aestivum developing seeds. No puroindoline transcripts as well as no related genes were detected in Triticum durum. The identity of puroindolines to wheat starch-granule associated proteins is discussed as well as the potential role of puroindolines in the plant defence mechanism.

The complete nucleotidic sequence of the yam mosaic virus (YMV) RNA was determined following the cloning of partial segments of the genome by reverse transcription and polymerase chain reactions (RT-PCR) using degenerate and/or specific oligonucleotide primers. YMV genomic RNA is 9,608 nucleotides in length and contains one open reading frame (ORF) encoding a polyprotein of 3,103 amino acids (aa) with a calculated Mr of 350,915. The 5' leader sequence of YMV RNA preceding the ORF is 134 nucleotides (nt) long while the 3' untranslated region (UTR) is 165 nt excluding the poly(A) tail. A computer algorithm predicted that the 3'UTR forms four stem loop structures which form a cloverleaf-like secondary structure. These structures apparently share some homologies with those observed in the 3'UTR of the potato virus Y-NL1 strain. Seven potential recognition sites for the NIa protease were found: one putative cleavage site for the P1 proteinase and one for the HC proteinase. The organization of the YMV genome is therefore similar to the other members of the genus Potyvirus based upon conserved sequence motifs common amongst members of this group. Despite its similarity with the other potyviruses in these conserved regions, YMV appears to be a distinct potyvirus species based upon a comparison of its sequence with those of other potyviruses.