Conversion of isoaspartyl peptides to normal peptides: implications for the cellular repair of damaged proteins.

The hypothesis that cellular protein carboxyl-methylation reactions recognize altered aspartyl residues as part of a protein repair pathway has been tested in an in vitro system using tetragastrin (Trp-Met-Asp-Phe-NH2) as a model sequence. The L-isoaspartyl form of tetragastrin, where the phenylalanine residue is linked to the side-chain carboxyl group of the aspartate residue ([iso-Asp3]tetragastrin), is a substrate for the erythrocyte protein carboxyl methyltransferases, while the normal form is not. The enzymatically produced alpha-methyl ester of [iso-Asp3]tetragastrin, [iso-Asp(OMe)3]tetragastrin, is unstable at pH 7.4 and 37 degrees C and spontaneously demethylates with a half-time of 41 min to an intermediate L-succinimide form ([Asu3]tetragastrin) that, in turn, spontaneously hydrolyzes with a half time of 116 min to give a mixture of normal tetragastrin (20%) and [iso-Asp3]tetragastrin (80%). This sequence of enzymatic and nonenzymatic reactions can be coupled in a single reaction mixture; the [iso-Asp3]tetragastrin that is produced upon succinimide hydrolysis can reenter the reaction sequence by enzymatic methylation, and the net result of the process is the conversion of the isomerized peptide to the normal peptide. The efficiency of this "repair" reaction is limited by a side reaction of racemization at the alpha-carbon of the succinimide (half-time = 580 min). In a 24-hr time period, normal L-aspartyl-containing tetragastrin is obtained in about 50% yield from the coupled reaction mixture; other products include [D-iso-Asp3]tetragastrin and [D-Asp3]tetragastrin. The versatile chemistry of succinimide peptides suggests that methylated L-isoaspartyl sites (and possibly methylated D-aspartyl sites) in cellular polypeptides can eventually yield "repaired" normal L-aspartyl sites through succinimide intermediates.