Highlighting Vitamin D Receptor-Targeted Activities of 1 α ,25-Dihydroxyvitamin D 3 in Mice via Physiologically Based Pharmacokinetic-Pharmacodynamic Modeling

• Published in: Drug metabolism and disposition 2018-01, Vol.46 (1), p.75-87
• Main Authors: Yang, Qi Joy, Bukuroshi, Paola, Quach, Holly P, Chow, Edwin C Y, Pang, K Sandy
• Format: Article
• Language: English
• ISSN: 0090-9556; 1521-009X
• DOI: 10.1124/dmd.117.077271

We expanded our published physiologically based pharmacokinetic model (PBPK) on 1 ,25-dihydroxyvitamin D [1,25(OH) D ], ligand of the vitamin D receptor (VDR), to appraise VDR-mediated pharmacodynamics in mice. Since 1,25(OH) D kinetics was best described by a segregated-flow intestinal model (SFM) that described a low/partial intestinal (blood/plasma) flow to enterocytes, with feedback regulation of its synthesis (Cyp27b1) and degradation (Cyp24a1) enzymes, this PBPK(SFM) model was expanded to describe the VDR-mediated changes (altered/basal mRNA expression) of target genes/responses with the indirect response model. We examined data on 1) renal Trpv5 (transient receptor potential cation channel, subfamily V member 5) and Trpv6 and intestinal Trpv6 (calcium channels) for calcium absorption; 2) liver 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase (Hmgcr) and cytochrome 7 -hydroxylase (Cyp7a1) for cholesterol synthesis and degradation, respectively; and 3) renal and brain Mdr1 (multidrug-resistance protein that encodes the P-glycoprotein) for digoxin disposition after repetitive intraperitoneal doses of 120 pmol 1,25(OH) D Fitting, performed with modeling software, yielded reasonable prediction of a dominant role of intestinal Trpv6 in calcium absorption, circadian rhythm that is characterized by simple cosine models for Hmgcr and Cyp7a1 on liver cholesterol, and brain and renal Mdr1 on tissue efflux of digoxin. Fitted parameters on the E , EC , and turnover rate constants of VDR-target genes [zero-order production (k ) and first-order degradation (k ) rate constants] showed low coefficients of variation and acceptable median prediction errors (4.5%-40.6%). Sensitivity analyses showed that the E and EC values are key parameters that could influence the pharmacodynamic responses. In conclusion, the PBPK(SFM)-pharmacodynamic model successfully characterized VDR gene activation and serves as a useful tool to predict the therapeutic effects of 1,25(OH) D .