Abstract
Drug transporters are increasingly recognised as important drivers in the pharmaco- and toxico-kinetic characteristics of therapeutic agents. Preclinical assays are used to determine the effects of transport on the pharmacokinetics and toxicology of a new compound during drug development. However in vitro cell systems and in vivo animal models often give erroneous results due to the differential expression of important proteins within them. In silico models have been employed to help bridge the gap and make more accurate predictions for drug disposition and metabolism; however such models have not incorporated transport kinetics or induction by nuclear receptors. Many current models also use a reductionist approach, whereby a multiple component pathway is described by a single mathematical term, reducing the ability of the simulation to determine the effects of single parameters. The first aim of this research was therefore to assess the differences in drug transporter expression in a variety of human and rat hepatocellular systems compared to liver. RT-PCR and protein analysis of drug transporter levels in vitro and in vivo showed huge differences, with influx transporters generally being under-expressed and efflux transporters over-expressed in vitro compared to in vivo. These differences were more pronounced at the RNA level than protein level. Transporter expression was also shown to be dynamic, changing over time in culture and in response to nuclear receptor activation. The second aim was to generate an in silico model of cellular response to stimuli, using a whole-cell approach to investigate system dynamics. Kinetics for the lifecycle rhodamiae-123 were determined in vitro and incorporated into an in silico model of human hepatocyte disposition that could accurately simulate outcomes in a variety of cell types and make useful predictions of drug disposition from in vitro results. In conclusion, the in silico model revealed that drug transporter expression and activity may affect drug disposition, with uptake processes presenting as the fragile node in the network. Such a model would be useful in the early stages of drug development to improve extrapolation between biological systems and to identify the likely consequences of different expression levels and species inhibition and induction.