A new nonlinear equation for the tissue/blood partition coefficients of neutral compounds

A nonlinear model equation based on tissue composition (a content of lipids, proteins, and water) for the tissue/blood partition coefficients of compounds was developed. Based on this model, our nonlinear regression analysis for neutral compounds partitioning into the kidney, brain, muscle, lung, liver, heart, and fat resulted in equations with high fitting power (training set: n = 166, r(2) = 0.851, s = 0.260, Q(2) = 0.833) and strong predictive power (test set: n = 49, r(2) = 0.851, s = 0.246, Q(2) = 0.836). This model shows that the tissue/blood partition coefficients of a compound depend strongly on tissue compositions. The magnitudes of partition coefficients of a compound in different tissues are mainly modulated by volume fractions or weight fractions of tissue compositions. (C) 2004 Wiley-Liss, Inc. and the American Pharmacists Association.