Investigation of a proposed penicillin G acidic degradation scheme using high-pressure liquid chromatography and optimization techniques and mechanistic considerations

Optimization techniques were used to fit a recently proposed degradation scheme to recently published n.m.r. data for the time course of penicillin G and four degradation products at pH 2.5 and 37°C. Several conclusions arising from the n.m.r. analysis which were originally associated with the degradation scheme were not compatible with the optimized rate constants. It was necessary to change substantially the proportion of penicillin G degrading through benzylpenicillenic acid, benzylpenillic acid, and benzyl-penicilloic acid in order for the degradation scheme to fit the n.m.r. data. Benzylpenillic acid replaced benzylpenicillenic acid as the major product. The rate constants best describing the n.m.r. data showed benzylpenicillenic acid proceeding almost exclusively through benzylpenamaldic acid. Such optimization implied that the scheme could be simplified to three parallel reaction pathways, the dominant reaction occurring through benzylpenillic acid. However, mechanistic considerations indicate that the direct conversion of penicillin G into benzylpenillic acid is not possible and that a likely intermediate is benzyl-penicilloic acid. The degradation of benzylpenicilloic acid at pH 2.5 was consequently monitored by ionpair reversed-phase high-pressure liquid chromatography and rapid formation of benzylpenillic acid was detected. This observation is inconsistent with the recently proposed degradation scheme, even though the scheme can be made to fit the n.m.r. kinetic data.