Domain structure and sequence distribution in dentin phosphophoryn

Abstract

Phosphophoryn (PP) is a protein unique to the mineralized matrix of dentin. It also has a unique composition, with aspartic acid and phosphoserine comprising greater than 85% of all amino acid residues. Because of this unique composition and high content of phosphoserine, it has been difficult to apply direct peptide sequencing procedures effectively. However, to understand its function, and to prepare suitable probes for screening cDNA libraries, some sequence distribution information is required. To this end, using bovine (b) and rat incisor (ri) PPs, partial mild acid hydrolysis has been used to cleave at the aspartic acid residues and generate free amino acids and small peptides. The nature of the released amino acids and peptides has been determined. Peptides have also been generated by limited digestion with trypsin. Some of the peptides have been purified by h.p.l.c. techniques and sequenced. About 90% of the bPP and riPP were resistant to trypsin, and the large resistant fragment was sharply depleted of the non-aspartic acid and non-phosphoserine [(P)Ser] residues. All peptides isolated were acidic, but the remaining residues (other than aspartic acid and serine) appeared to be collected in regions flanking the trypsin-resistant core. These data show directly the presence of regions [Asp]n, [(P)Ser]m and [Asp-(P)Ser-Asp]k as prominent sequence features. A domain structure model is proposed.