The destruction of amino acids during the acid hydrolysis of peptides has been described on several occasions and summarized by Smyth and Elliott (1). Several amino acids are subject to change under these conditions but these amino acids usually possess susceptible functional groups in the side chain. The modification of phenylalaliine does not appear to have been reported hitherto.
While carrying out routine amino acid analysis of protected peptide derivatives used in the synthesis of a series of bradykinin analogs (2) it was noticed that the ratios of phenylalanine to glycine were not always close to the expected values. This effect was observed with compounds that were pure by other criteria, i.e., homogeneous by thin-layer chromatography, and t'hat gave satisfactory elemental analyses and had compatible IR spectra. This deviation was observed whenever phenylalanine and nitroarginine occurred together and examination of analyzer traces revealed that concomitantly three new peaks had appeared. The structures of the compounds giving rise to these peaks and the mechanism whereby t,hey arose form the subject of t,his report.
ESPERIMEKTAL
Peptides were hydrolyzed in evacuated sealed tubes at 115" for 16 hr in constant-boiling hydrochloric acid. Amino acid analysis was carried out on the Beckman 120C analyzer under the manufacturer's recommended conditions. Column numbers have the usual connotation, i.e., column 1 is used for acidic and neutral and column 2 for basic amino acids and column 3 is that usually employed for basic materials in the physiological fluid analysis. Eluting buffers were as recommended by Beckman but, in place of Beckman resin, columns 2 and 3 contained fractionated (3) Zeo-Karb 225 sulfonated polystyrene (8% cross-linked) spherical beads. This material was found to reproduce the performance