Alleles at the PRH1 locus coding for the human salivary-acidic proline-rich proteins Pa, Db, and PIF

We cloned and sequenced the entire exon and intron structures of Db and Pa genetic determinants at the PRH1 locus. Their derived amino acid sequences and that previously determined for the PIF protein completely explain the electrophoretic phenotypes of the acidic proline-rich proteins (PRPs) Pa, Db, and PIF. Thus, the Cys substitution near Arg 106 in the Pa protein sterically interferes with proteolytic cutting at Arg 106 and accounts for the single-banded phenotype. In contrast, the Db and PIF proteins are proteolytically cut at Arg 106 and show a double-banded phenotype. The Db protein has an extra 21-amino acid repeat that accounts for its larger size compared with the equal sized Pa monomer and PIF proteins. Several amino acid substitutions account for the charge and mobility differences of the Pa, Db, and PIF proteins in isoelectric-focusing gels. These DNA/protein correlations, as well as the extremely similar genomic-DNA sequences that differ by less than 1%, establish that Pa, Db, and PIF are alleles at the PRH1 locus. On the basis of the DNA sequences, we conclude that Db and Pa alleles diverged more recently from a common precursor than did the PIF allele from its precursor.