Abstract
Proteins such as membrane transporters, enzymes, receptors and channels play key roles in drug absorption, distribution, metabolism, and elimination, and also influence efficacy and the likelihood of adverse reactions. Therefore, if we can quantify the activities of these molecules, it may be possible to predict the behavior of candidate drugs in humans in disease states; such methodology would be extremely helpful for efficient drug development. We have developed an in silico method to select appropriate peptides within amino acid sequences in order to quantify targeted proteins by LC-MS/MS in selected reaction monitoring (SRM) mode. We have applied this method for the quantification of functional proteins in order to validate various in vitro and in vivo models. We found fairly good correlation between protein amounts and the enzymatic activities of microsomal cytochrome P450 (CYP) isoforms and uridine 5′-diphospho-glucuronosyltransferase (UGT) in human liver, as well as between protein amounts and the transport activities of multiple transporters in human lung cells. These results suggest that protein quantification can be useful in predicting activity. We have applied this approach to evaluate the usefulness and limitations of an immortalized human brain capillary endothelial cell line (D3 cells) and a P-glycoprotein humanized (hMDR1) mouse model by comparing the amounts of functional proteins in the models with those in isolated capillaries from human brain. In order to obtain sufficient human tissue specimens for further studies leading to clinical applications, we believe that international collaboration will be crucial.
