Quantification of Neocortical Slice Diffusion Characteristics Using Pharmacokinetic and Pharmacodynamic Modelling

Pharmacological brain slice experiments are complicated by the need to ensure adequate drug delivery deep into the healthy layers of the tissue. Because tissue slices have no blood supply, this is achieved solely by passive drug diffusion. The aim of this study was to determine whether pharmacokinetic/pharmacodynamic (PKPD) modeling could be adapted to estimate drug diffusion times in neocortical brain slices. No-magnesium seizure-like event (SLE) activity was generated in 41 slices (400？μm). Two anesthetic agents, etomidate (24？μM, ) and thiopental (250？μM, ), and magnesium ions ( ) were delivered to effect reversible reductions in SLE frequency. Concentration-effect hysteresis loops were collapsed using a first order rate constant model and equilibrium half-lives ( ) derived. The values obtained were consistent with expectations. The median (range) of 83.1 (19.4–330.1)？min for etomidate is in keeping with its known slow diffusion into brain slice tissue. Values for etomidate and thiopental (111.8 (27.8–198.0)？min) were similar, while magnesium had a significantly faster equilibration rate ( of 26.1 (8.6–77.0)？min) compared to the anesthetics, as expected for a simple ion. In conclusion, PKPD modeling is a simple and practical method that can be applied to brain slice experiments for investigating drug diffusion characteristics. 1. Introduction Characterisation of drug effects in in vitro brain slice experiments is complicated by the absence of a blood supply to the tissue. This means that drugs must be delivered to the tissue by passive diffusion via the perfusion solution. Because substances diffuse at different rates according to their physicochemical properties, it can be difficult to know how long to expose the tissue in order to achieve drug penetration into the healthiest layers of the slice—approximately the middle 100？μm for a 400？μm thick slice [1]. For drugs that diffuse slowly, there is a risk that a negative result will be spuriously attributed to an inactive drug, when in fact it never reached its site of action in the tissue. Directly measuring drug concentrations in tissue slicesnecessitates methodologies that are time consuming and complex and cannot realistically be applied to each and every drug one may want to test. This motivated us to explore alternative, simpler approaches for estimating drug diffusion in brain slice tissue. To this end, we have borrowed a pharmacological technique known as pharmacokinetic/pharmacodynamic (PKPD) modelling, which is used clinically to describe the relationship between the time course of drug