Characterization of the kafirin gene family from sorghum reveals extensive homology with zein from maize.

Electrophoretic analysis of translation products of polyadenylated RNA isolated from mid-maturation sorghum seed in the presence of [(35)S]met, [(3)H]leu, or [(3)H]val revealed two major proteins of kDa and 21 kDa. These products were not detected when [(3)H]lys was supplied as the radioactive substrate. Under similar electrophoretic conditions, kafirin (a major seed storage prolamin of sorghum), migrated as two bands of 22 kDa and 19 kDa. Sequence analysis of two cDNA clones (pSK8 and pSKR2) from sorghum seed mRNA revealed them to be highly homologous with each other and to the 22 kDa zeins from maize, suggesting that they represented kafirin cDNAs. Compared with pSKR2, pSK8 had an insertion of 24 nucleotides and a deletion of 24 nucleotides, so that the coding regions were nearly identical in length. The deduced amino acid sequence for these cDNA clones reveals that kafirin, like zein, is rich in glutamine and nonpolar amino acids, but contains no lysine. Both kafirin and zein have a 21 amino acid signal peptide exhibiting 80% homology and eight copies of a repetitive amino acid block in the C-terminal domain with the consensus: infI (supP) LL finP (supA) LN infQ (supP) LALANPAAYLQQQQ.The kafirin cDNAs were used as probes to screen a sorghum genomic library; one genomic clone (λGK.1) was sequenced and found to be very similar (97.8%) to the pSK8 cDNA clone. Clone λGK.1 contains features typical for a functional gene in that the intronless open reading frame encoding 268 amino acids is flanked at the 5' end by sequences corresponding to the CAAT and TATA promoter boxes (positioned at about -60 and -30 bp, respectively, from the transcriptional initiation site), and at the 3' end by a consensus polyadenylation signal. In common with zein genomic clones, kafirin clones contain a 15 basepair consensus sequence centered at postion -320 relative to the transcriptional initiation site. Under similar hybridization conditions, genomic reconstruction analysis using an oligonucleotide probe indicated the presence of less than 20 copies of kafirin per haploid sorghum genome compared with approximatley 140 copies of zein per haploid maize genome.

Expression of a bean storage protein 'phaseolin minigene' in foreign plant tissues.

Using the phaseolin gene and its cDNA counterpart we constructed a mutant phaseolin gene lacking the five introns but retaining its natural 5' and 3' plant-regulatory sequences. This mutant phaseolin gene (minigene) was inserted into the Ti-plasmid of Agrobacterium tumefaciens strain 15955 which allowed its transfer and integration into the tobacco genome. Full-length and correctly initiated phaseolin mRNA was found among the poly(A)+RNA isolated from plant callus transformed with the minigene construction by using RNA-DNA hybridization and S1 nuclease mapping techniques. The presence of phaseolin polypeptides in soluble protein extracts from transformed tobacco tissues was confirmed by immunological methods. These results demonstrate that phaseolin gene introns and intron splicing are not a necessary requirement for biogenesis of stable phaseolin mRNA and that no alternative splice site was introduced by the removal of five introns.

Nucleotide sequences from phaseolin cDNA clones: the major storage proteins from Phaseolus vulgaris are encoded by two unique gene families.

The nucleotide sequences of eight partial and five full-length phaseolin cDNA clones show that phaseolin polypeptides are encoded by two distinct gene families which differ in their coding regions by the presence or absence of two different size direct repeats. The alpha-type phaseolin polypeptides are encoded by genes containing direct repeats which encode 14 additional amino acids. Aside from these differences, the alpha-and beta-type phaseolin genes show a high degree of homology (98%) which is consistent with these genes being derived from a common ancestral gene. Much of the heterogeneity found in the phaseolin polypeptides appears to be due to post-translational processing. Nucleotide sequence analysis demonstrates that the alpha-type genes contain only a few amino acid replacement substitutions and that the beta-type genes appear to contain no amino acid replacement substitutions. S1 nuclease mapping shows a complex pattern for transcriptional initiation of phaseolin mRNA. Hydropathy analysis shows that phaseolin polypeptides are predominately hydrophilic, and that the two N-glycosyl recognition sites are located in different hydropathic environments.