| Abstract: | LC‐HRMS‐based identification of the products of peptide catabolism is the key to drive the design of more stable compounds. Because the catabolite of a given peptide can be very different from the parent compound and from other catabolites in terms of physicochemical properties, it can be challenging to develop an analytical method that allows recovery and detection of the parent and all parent‐related catabolites. The aim of this study was to investigate how the recovery and the matrix effect of peptidic drugs and their catabolites are affected by different protein precipitation (PP) and solid‐phase extraction (SPE) protocols. To this purpose, four model peptides representative of different classes (somatostatin, GLP‐2, human insulin and liraglutide) were digested with trypsin and chymotrypsin to simulate proteolytic catabolism. The resulting mixtures of the parent peptides and their proteolytic products covering a wide range of relative hydrophobicity (HR) and isoelectric points (pI) were spiked in human plasma and underwent different PP and SPE protocols. Recovery and matrix effect were measured for each peptide and its catabolites. PP with three volumes of ACN or EtOH yielded the highest overall recoveries (more than 50% for the four parent peptides and all their catabolites) among all the tested PP and SPE protocols. Mixed‐mode anion exchange (MAX) was the only SPE sorbent among the five tested that allowed to extract all the peptides with recoveries more than 20%. Matrix effect was generally lower with SPE. Overall, it was observed that peptides with either high hydrophilicity (e.g., somatostatin catabolites) or hydrophobicity (GLP‐2 and lipidated liraglutide catabolites) had a much narrower choice of PP solvent or SPE protocol. Simulation of catabolism using recombinant enzymes together with in silico calculation of the HR and the pI of potential proteolysis products is recommended to select the optimal extraction conditions for the study of peptide catabolism. |
