Poirier Agnès, Lavé Thierry, Portmann Renée, Brun Marie-Elise, Senner Frank, Kansy Manfred, Grimm Hans-Peter, Funk Christoph
F. Hoffmann-La Roche Ltd., Non-Clinical Development, Drug Safety, Basel, Switzerland.
Drug Metab Dispos. 2008 Dec;36(12):2434-44. doi: 10.1124/dmd.108.020750. Epub 2008 Sep 22.
The use of in vitro data for quantitative predictions of transporter-mediated elimination in vivo requires an accurate estimation of the transporter Michaelis-Menten parameters, V(max) and K(m), as a first step. Therefore, the experimental conditions of in vitro studies used to assess hepatic uptake transport were optimized regarding active transport processes, nonspecific binding, and passive diffusion (P(dif)). A mechanistic model was developed to analyze and accurately describe these active and passive processes. This two-compartmental model was parameterized to account for nonspecific binding, bidirectional passive diffusion, and active uptake processes based on the physiology of the cells. The model was used to estimate kinetic parameters of in vitro transport data from organic anion-transporting peptide model substrates (e.g., cholecystokinin octapeptide deltorphin II, fexofenadine, and pitavastatin). Data analysis by this mechanistic model significantly improved the accuracy and precision in all derived parameters [mean coefficient of variations (CVs) for V(max) and K(m) were 19 and 23%, respectively] compared with the conventional kinetic method of transport data analysis (mean CVs were 58 and 115%, respectively, using this method). Furthermore, permeability was found to be highly temperature-dependent in Chinese hamster ovary (CHO) control cells and artificial membranes (parallel artificial membrane permeability assay). Whereas for some compounds (taurocholate, estrone-3-sulfate, and propranolol) the effect was moderate (1.5-6-fold higher permeability at 37 degrees C compared with that at 4 degrees C), for fexofenadine a 16-fold higher passive permeability was seen at 37 degrees C. Therefore, P(dif) was better predicted if it was evaluated under the same experimental conditions as V(max) and K(m), i.e., in a single incubation of CHO overexpressed cells or rat hepatocytes at 37 degrees C, instead of a parallel control evaluation at 4 degrees C.
作为第一步，使用体外数据对体内转运体介导的消除进行定量预测需要准确估计转运体的米氏参数V(max)和K(m)。因此，针对主动转运过程、非特异性结合和被动扩散（P(dif)），对用于评估肝脏摄取转运的体外研究的实验条件进行了优化。开发了一个机制模型来分析和准确描述这些主动和被动过程。这个双室模型基于细胞生理学进行参数化，以考虑非特异性结合、双向被动扩散和主动摄取过程。该模型用于估计来自有机阴离子转运肽模型底物（如胆囊收缩素八肽、德尔托啡肽II、非索非那定和匹伐他汀）的体外转运数据的动力学参数。与传统的转运数据分析动力学方法相比，通过该机制模型进行的数据分析显著提高了所有导出参数的准确性和精密度[V(max)和K(m)的平均变异系数（CVs）分别为19%和23%]（使用该方法时，平均CVs分别为58%和115%）。此外，发现通透性在中国仓鼠卵巢（CHO）对照细胞和人工膜（平行人工膜通透性测定）中高度依赖温度。对于某些化合物（牛磺胆酸盐、雌酮-3-硫酸盐和普萘洛尔），这种影响适中（37℃时的通透性比4℃时高1.5 - 6倍），而对于非索非那定，37℃时的被动通透性高出16倍。因此，如果在与V(max)和K(m)相同的实验条件下进行评估，即在37℃下对CHO过表达细胞或大鼠肝细胞进行单次孵育，而不是在4℃下进行平行对照评估，则能更好地预测P(dif)。
