Pregnane X receptor prevents hepatorenal toxicity from cholesterol metabolites

Abstract

Efficient detoxification and clearance of cholesterol metabolites such as oxysterols, bile alcohols, and bile acids are critical for survival because they can promote liver and cardiovascular disease. We report here that loss of the nuclear xenobiotic receptor PXR (pregnane X receptor), a regulator of enterohepatic drug metabolism and clearance, results in an unexpected acute lethality associated with signs of severe hepatorenal failure when mice are fed with a diet that elicits accumulation of cholesterol and its metabolites. Induction of a distinct drug clearance program by a high-affinity ligand for the related nuclear receptor, the constitutive androstane receptor, does not overcome the lethality, indicating the unique requirement of PXR for detoxification. We propose that the PXR signaling pathway protects the body from toxic dietary cholesterol metabolites, and, by extension, PXR ligands may ameliorate human diseases such as cholestatic liver diseases and the associating acute renal failure.

Fig. 1.

Acute death of CH/CA-fed pxr^-/- mice is associated with severe conjugated bilirubinemia. (A) Survival of WT (+/+; n = 20) and pxr^-/- (-/-; n = 23) mice on the CH/CA diet. (B) Ventral view of the head (Upper) and hind limb (Lower) of representative WT (+/+) and lethargic pxr^-/- (-/-) mice fed the CH/CA diet for 6 days. Note that the skin color of the pxr^-/- mouse is significantly darker compared with the WT mouse, indicating the presence of severe jaundice. (C–H) Total plasma bilirubin (C), plasma direct (conjugated) bilirubin (D), plasma total bile acid (E), hepatic cholesterol (F), plasma total cholesterol (G), and plasma phospholipid (H) concentrations in WT (+/+) and pxr^-/- (-/-) mice fed the indicated diets for 7–10 days. All values are expressed as mean ± SD, n = 5. *, significant differences compared with WT on the same diet; #, significant differences compared with mice of the same genotype on the chow diet (P < 0.01).

Fig. 2.

The liver phenotype of CH/CA-fed WT and pxr^-/- mice. (A) Gross morphology of livers from chow- or CH/CA-fed WT (+/+, Left) and pxr^-/- (-/-, Right) mice. Liver sections were prepared (n = 5 per group) for histology and stained with hematoxylin/eosin. Note the absence of abnormality in chow-fed WT (Left Upper) and pxr^-/- (Right Upper) mice and the presence of inflammatory infiltration in both CH/CA-fed WT (Left Lower) and pxr^-/- (Right Lower) mice. (B) Plasma activities of aspartate transaminase (AST; Upper) and alanine transaminase (ALT; Lower). Animals used in this study were the same as in Fig. 1. All values are expressed as mean ± SD, n = 5. *, significant differences compared with WT on the same diet; #, significant differences compared with mice of the same genotype on the chow diet (P < 0.05). (C) Northern blot analysis of total RNA isolated from livers of chow- or CH/CA-fed WT (+/+) and pxr^-/- (-/-) mice.

Fig. 3.

The kidney phenotype of CH/CA-fed WT and pxr^-/- mice. (A) Plasma levels of blood urea nitrogen (BUN; Upper) and creatinine (Lower). Animals used in this study were the same as in Fig. 1. All values are expressed as mean ± SD, n = 5. *, significant differences compared with WT on the same diet; #, significant differences compared with mice of the same genotype on the chow diet (P < 0.05). (B) Northern blot analysis of total RNA isolated from kidney of chow- or CH/CA-fed WT (+/+) and pxr^-/- (-/-) mice.

Fig. 4.

Regulatory target genes for PXR in CH/CA-fed mice. (A) Real-time qPCR analysis for cyp3a11 and oatp2 gene expression in the liver. RNA samples were isolated from livers of animals used in Fig. 1, except that RNA samples for CH/CA-fed pxr^-/- mice were collected from animals that had not yet developed jaundice as described in the text. Values for qPCR analysis were normalized by the level of u36b4 gene encoding a ribosomal protein subunit, are expressed as fold change over chow-fed WT control, and represent the mean ± SD, n = 4. *, significant differences compared with WT on the same diet; #, significant differences compared with mice of the same genotype on the chow diet (P < 0.05). (B)WT(+/+) and pxr^-/- (-/-) mice were subjected to i.p. injection of pregnane-16α-carbonitrile (PCN) (40 mg/kg), TCPOBOP (3 mg/kg), or solvent. After 24 h, total RNA was isolated from the liver and subjected to qPCR analysis for cyp3a11 and oatp2 mRNA expression. Values for qPCR analysis were normalized by the level of u36b4 gene, are expressed as fold change over chow-fed WT control, and represent the mean ± SD, n = 3.

Fig. 5.

The effects of CAR activation in CH/CA-fed pxr^-/- mice. (A) Survival of pxr^-/- mice pretreated with TCPOBOP or solvent before the feeding of the CH/CA diet for 20 days. No significant difference in the rate of death was observed. (B) Total bilirubin (Left) and total bile acid (Right) concentrations in plasma from WT (+/+) and lethargic pxr^-/- (-/-) mice pretreated with TCPOBOP or solvent on the CH/CA diet for 7–10 days. All values are expressed as mean ± SD, n = 5. *, significant differences compared with WT on the same treatment; #, significant differences compared with mice of the same genotype without TCPOBOP pretreatment (P < 0.005).