Structural differences in full-length cDNAs for two classes of sporamin, the major soluble protein of sweet potato tuberous roots

Abstract

Sporamin, which accounts for 80% of the total soluble proteins in sweet potato tuberous roots, consists of two polypeptide classes, A and B. The sporamin cDNA clones can also be classified into sporamin A and B subfamilies based on their sequence homologies, with intra-subfamily homologies being much higher than inter-subfamily homologies. The sequence of an essentially full-length cDNA for sporamin B was compared with that for sporamin A. The coding sequences of two cDNAs share 83% sequence homology. The sequences in the 5′- and 3′-noncoding regions show many deletions in addition to base substitutions. The endpoints of deletions longer than 4 bp match precisely to the endpoints of short direct repeats present in the other sequence, which suggests that these deletions are generated by slipped mispairing during DNA replication. In the 5′- and 3′-noncoding region of sporamin B cDNA, there are 5 bp direct repeats with sequences complementary to each other. Since most of these repeats are absent in sporamin A cDNA, these structural features may cause a difference in the secondary structure between A and B mRNAs and affect the translational efficiencies or stabilities of the mRNAs. Precursors for both classes of sporamin carry N-terminal extra-sequences which can be separated into a putative signal peptide segment and a segment enriched with basic amino acids. A two-step processing mechanism for the maturation of sporamin is suggested.