P-glycoprotein (P-gp) and CYP3A have considerable overlap in inducers in vitro. Characterizing P-gp induction in vivo and potential coregulation with CYP3A are important goals for predicting drug interactions. This study examined P-gp expression in mouse tissues and potential coinduction with CYP3A following oral treatment with 1 of 7 prototypical inducing agents for 5 days. P-gp expression in brain or liver was not induced by any treatment as determined by Western blot, whereas dexamethasone, pregnenolone-16alpha-carbonitrile (PCN), St. John's wort (SJW), and rifampin induced hepatic CYP3A expression. In intestine, rifampin and SJW induced P-gp expression 3.7- and 1.6-fold and CYP3A 3.5- and 2.4-fold, respectively, whereas dexamethasone and PCN induced CYP3A only. These observations suggest that P-gp in mouse small intestine is inducible by some, but not all, CYP3A inducers, whereas P-gp expression in liver or brain is not readily induced. Intriguingly, rifampin and SJW, both activators of the human pregnane X receptor (PXR), induced CYP3A in both liver and intestine but induced P-gp only in intestine, whereas PCN, an activator of murine PXR, did not induce P-gp in any tissue. Rifampin disposition was evaluated, and hepatic exposure to rifampin was comparable to intestine; in contrast, brain concentrations were low. Overall, these observations demonstrate that P-gp induction in vivo is tissue-specific; furthermore, there is a disconnect between P-gp induction and CYP3A induction that is tissue- and inducer-dependent, suggesting that PXR activation alone is insufficient for P-gp induction in vivo. Tissue-specific factors and inducer pharmacokinetic/pharmacodynamic properties may underlie these observations.