Encoding of human basic and glycosylated proline-rich proteins by the PRB gene complex and proteolytic processing of their precursor proteins.

Proline-rich proteins (PRPs) constitute a family of about 20 members in human saliva that are encoded by six genes. Assignment of genomic DNA coding regions is complicated because of the occurrence of many alleles and the great similarity of amino acid sequences of PRPs. To overcome these problems, the nucleotide sequences of the genes encoding basic and glycosylated PRPs from one person were determined and then aligned with her previously determined protein sequences. This, together with additional protein data, has also resolved various discrepancies between corresponding protein and DNA sequences. For the first time in one person it is now possible to account for all the regions in the PRB genes encoding basic and glycosylated PRPs, and the primary structures of all secreted basic and glycosylated PRPs have been determined. Each gene encodes a precursor protein that subsequently undergoes proteolytic cleavage, thereby giving rise to the secreted proteins. The results have allowed identification of all the proteolytic cleavage sites in the precursor proteins, which all conform to a consensus cleavage site for furin. To evaluate if furin is responsible for the precursor protein cleavages, a recombinant precursor protein was synthesized by in vitro transcription translation of a PRB1 allele. The protein was shown to be correctly cleaved by furin, giving rise to the expected secreted proteins.