COMMENTARY

Cholesterol detoxification by the nuclear pregnane X receptor

Steven A. Kliewer* Departments of Molecular Biology and Pharmacology, University of Texas Southwestern Medical Center, Dallas, TX 75390-9041

e have a love–hate rela- sors induce a shunt pathway of sterol tionship with cholesterol. side chain shortening, which results in On one hand, cholesterol the formation of bile acids. Although is an essential component this finding demonstrates the existence ofW cell membranes and serves as the of a feed-forward regulatory pathway by precursor to steroid hormones and bile which potentially toxic bile acid interme- acids. On the other hand, cholesterol diates activate PXR and induce their can clog blood vessels and give rise to own metabolism, it is not known cardiovascular disease. Moreover, oxi- whether this pathway operates under dized metabolites of cholesterol, termed normal physiologic conditions. Finally, oxysterols, are cytotoxic in a variety of activation of PXR protects against the different cell types and contribute to hepatotoxic effects of high concentra- atherosclerosis (1). Thus, it is crucial tions of bile acids administered in the that excess cholesterol be removed from diet (8, 9). Together, these studies raise the body. In a recent issue of PNAS, the possibility that PXR evolved to de- Sonada et al. (2) described an unex- tect and detoxify not only xenobiotics pected role for the pregnane X receptor but also cholesterol precursors and me- (PXR) in protecting against cholesterol tabolites (Fig. 1). toxicity. To address whether PXR has a physi- PXR is a member of the steroid͞thy- Fig. 1. Model for PXR-induced detoxification of ological role in cholesterol metabolism, roid hormone receptor family of ligand- xenobiotics and cholesterol. Activation of PXR by Sonoda et al. (2) fed wild-type mice and Ϫ Ϫ activated transcription factors that is xenobiotics or cholesterol metabolites stimulates mice lacking PXR (Pxr / ) a diet en- expressed in liver, intestine, and kidney the transcription of genes encoding oxygenases, riched in cholesterol and cholic acid, a (3–5). Unlike the classical steroid hor- conjugases, and transporters, which detoxify xeno- bile acid that blocks cholesterol catabo- mone receptors, which are selectively biotics and cholesterol. PXR binds to DNA as a lism in the liver and enhances choles- heterodimer with the retinoid X receptor (RXR). activated by their cognate hormones at terol absorption in the intestine. The Ϫ/Ϫ nanomolar or even picomolar concen- Pxr mice have no overt phenotype trations, PXR is activated by micromolar the body itself. First, PXR is activated in under normal laboratory conditions (8, 14). Remarkably, the cholesterol͞cholic concentrations of a structurally diverse cell-based assays by many different acid diet killed Ϸ40% of the PxrϪ/Ϫ collection of foreign chemicals, or xeno- chemicals with steroid backbones, in- mice by day 10 and 100% of the mice by biotics, including the antibiotic rifampi- cluding bile acid and oxysterol metabo- day 60 while causing no lethality in wild- cin, the cancer drug taxol, and the herb lites of cholesterol (7–9). In fact, PXR type animals. Death in the PxrϪ/Ϫ mice Saint John’s wort. This remarkable pro- was named based on its activation by a was preceded by several days of leth- miscuity is facilitated by the unusual number of different C21 steroids, in- argy, hypothermia, and weight loss and ligand-binding pocket of PXR, which is cluding the hormone progesterone (10). by dramatic increases in the serum con- both very large and smooth, allowing it However, activation of PXR requires centrations of bilirubin and bile acids, to bind to many different chemicals (6). higher concentrations of these chemicals two markers of hepatobiliary defects. Once activated, PXR binds to DNA as a than are typically measured in vivo, Histologic examination showed that the heterodimer with the retinoid X recep- which draws into question the physiolog- livers from the cholesterol͞cholic-acid- tor and stimulates the transcription of ical relevance of these observations. fed PxrϪ/Ϫ mice were normal with re- genes encoding cytochrome P450 en- Second, Shenoy et al. (7) showed that spect to their overall morphology and zymes, conjugation enzymes, and trans- treatment of hepatocytes in vitro with the number of bile ducts. However, porters, which collectively promote 2,3-oxidosqualene:lanosterol cyclase in- these studies revealed hepatitis in the xenobiotic metabolism. These findings hibitors, which block the conversion of Ϫ Ϫ Pxr / mice manifested by infiltration of have led to a ‘‘xenosensor’’ model (Fig. squalene 2,3-oxide into the cholesterol mononuclear cells in the periportal and 1), in which PXR detects potentially precursor lanosterol, activates PXR, pre- midzonal areas of the hepatic lobes. The harmful xenobiotics and induces genes sumably through the accumulation of Ϫ Ϫ Pxr / mice also had markedly elevated that flush the offending chemicals from squalene intermediates that serve as concentrations of plasma aspartate the body (3–5). An undesirable side ef- PXR ligands (11). Third, there is evi- transaminase and alanine transaminase, fect of this xenobiotic detoxification dence that PXR can be activated by two markers of hepatocellular injury, system is that drugs that activate PXR cholesterol metabolites in vivo: Two and increased expression of genes in- can stimulate the metabolism of other groups showed that mice lacking the duced by inflammation. In addition to drugs, sometimes with life-threatening enzyme sterol 27-hydroxylase, which is causing hepatotoxicity, the cholesterol͞ consequences. required for the conversion of choles- Although PXR is proposed to func- terol to bile acids, have increased PXR tion as a xenosensor, there have been activity due to elevated concentrations See companion article on page 2198 in issue 6 of volume tantalizing hints that it might also detect of bile acid precursors such as 5␤- 102. and coordinate the detoxification of cholestane-3␣,7␣,12␣-triol (12, 13). By *E-mail: [email protected]. metabolic intermediates produced by activating PXR, these bile acid precur- © 2005 by The National Academy of Sciences of the USA

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0500159102 PNAS ͉ February 22, 2005 ͉ vol. 102 ͉ no. 8 ͉ 2675–2676 Downloaded by guest on September 23, 2021 cholic acid diet caused kidney damage the sinusoidal membrane of hepatocytes, not been described, mutations in a num- in the PxrϪ/Ϫ mice, as evidenced by in- where it transports a variety of endoge- ber of other transporters cause liver dis- creased concentrations of blood urea nous and exogenous compounds, in- ease (18). nitrogen and creatine and induction of cluding conjugated and unconjugated Cholesterol toxicity is usually thought proinflammatory genes. Taken together, bilirubin and conjugated steroids and of as occurring over long periods of these data demonstrate an essential role bile acids into the liver, where they are time as in cardiovascular disease. The for PXR in protecting against acute tox- further metabolized and excreted from work of Sonoda et al. (2) reveals an un- icity caused by a high-cholesterol diet. the body. Notably, activation of the expected acute toxicity caused by excess What is the molecular mechanism un- nuclear receptor CAR, a xenobiotic cholesterol, which raises a number of derlying the protective actions of PXR? interesting questions. What chemicals Chemicals that activate PXR such as kill the PxrϪ/Ϫ mice, and how do they pregnenolone 16␣-carbonitrile and spi- Pregnane X receptors cause death? What are the ligands gen- ronolactone increase both the produc- erated by the cholesterol͞cholic acid tion of bile and the concentration of protect against acute diet that activate PXR? Finally, and per- cholesterol in the bile, suggesting a role haps most importantly, do these studies for this receptor in regulating choles- toxicity caused by a translate to humans? Do polymorphisms terol homeostasis (15, 16). Sonoda et al. and mutations in PXR influence choles- (2) show that two genes regulated by high-cholesterol diet. terol metabolism and the onset and PXR, cytochrome P450 3A11 (Cyp3a11) progression of liver disease in people? and organic anion transporting peptide Notably, PXR agonists such as rifampi- 2(Oatp2), are induced in liver by the receptor closely related to PXR that cin and phenobarbital have long been cholesterol͞cholic acid diet in wild-type induces Cyp3a11 but not Oatp2, did not used clinically to treat pruritis associ- mice but not in PxrϪ/Ϫ mice, which sug- prevent the lethality of the cholesterol͞ ated with liver disease (19). The study gests that this diet produces a PXR li- cholic acid diet in the PxrϪ/Ϫ mice. by Sonoda et al. (2) raises the intriguing gand. CYP3A isozymes are best known These data point to OATP2 as a candi- possibility that PXR agonists may be for oxidizing drugs but also hydroxylate date for protecting against cholesterol useful for treating other aspects of liver various endogenous steroids, including toxicity. How this might occur is not disease, including hepatitis and renal cholesterol (17). OATP2 is localized on clear. Although OATP2 deficiency has failure associated with liver pathology.

1. Leonarduzzi, G., Sottero, B. & Poli, G. (2002) J. 8. Staudinger, J. L., Goodwin, B., Jones, S. A., 13. Dussault, I., Yoo, H. D., Lin, M., Wang, E., Fan, Nutr. Biochem. 13, 700–710. Hawkins-Brown, D., MacKenzie, K. I., LaTour, M., Batta, A. K., Salen, G., Erickson, S. K. & 2. Sonoda, J., Chong, L. W., Downes, M., Barish, A., Liu, Y., Klaassen, C. D., Brown, K. K., Rein- Forman, B. M. (2003) Proc. Natl. Acad. Sci. USA G. D., Coulter, S., Liddle, C., Lee, C.-H. & Evans, hard, J., et al. (2001) Proc. Natl. Acad. Sci. USA 98, 100, 833–838. R. M. (2005) Proc. Natl. Acad. Sci. USA 102, 3369–3374. 14. Xie, W., Barwick, J. L., Downes, M., Blumberg, B., 2198–2203. 9. Xie, W., Radominska-Pandya, A., Shi, Y., Simon, Simon, C. M., Nelson, M. C., Neuschwander-Tetri, 3. Xie, W. & Evans, R. M. (2001) J. Biol. Chem. 276, C. M., Nelson, M. C., Ong, E. S., Waxman, D. J. B. A., Brunt, E. M., Guzelian, P. S. & Evans, R. M. 37739–37742. & Evans, R. M. (2001) Proc. Natl. Acad. Sci. USA (2000) Nature 406, 435–439. 98, 4. Kliewer, S. A., Goodwin, B. & Willson, T. M. 3375–3380. 15. Klaassen, C. D. (1969) J. Pharmacol. Exp. Ther. 10. Kliewer, S. A., Moore, J. T., Wade, L., Staudinger, (2002) Endocr. Rev. 23, 687–702. 168, 218–223. J. L., Watson, M. A., Jones, S. A., McKee, D. D., 5. Dussault, I. & Forman, B. M. (2002) Crit. Rev. 16. Turley, S. D., Spady, D. K. & Dietschy, J. M. Oliver, B. B., Willson, T. M., Zetterstrom, R. H., Eukaryot. Gene Expr. 12, 53–64. (1983) Gastroenterology 84, 253–264. et al. (1998) Cell 92, 73–82. 6. Watkins, R. E., Wisely, G. B., Moore, L. B., 11. Shenoy, S. D., Spencer, T. A., Mercer-Haines, 17. Bodin, K., Andersson, U., Rystedt, E., Ellis, E., Collins, J. L., Lambert, M. H., Williams, S. P., N. A., Abdolalipour, M., Wurster, W. L., Runge- Norlin, M., Pikuleva, I., Eggertsen, G., Bjorkhem, Willson, T. M., Kliewer, S. A. & Redinbo, M. R. Morris, M. & Kocarek, T. A. (2004) Mol. Phar- I. & Diczfalusy, U. (2002) J. Biol. Chem. 277, (2001) Science 292, 2329–2333. macol. 65, 1302–1312. 31534–31540. 7. Shenoy, S. D., Spencer, T. A., Mercer-Haines, 12. Goodwin, B., Gauthier, K. C., Umetani, M., Watson, 18. Kullak-Ublick, G. A. & Meier, P. J. (2000) Clin. N. A., Alipour, M., Gargano, M. D., Runge- M. A., Lochansky, M. I., Collins, J. L., Leitersdorf, Liver Dis. 4, 357–385. Morris, M. & Kocarek, T. A. (2004) Drug Metab. E., Mangelsdorf, D. J., Kliewer, S. A. & Repa, J. J. 19. Bachs, L., Pares, A., Elena, M., Piera, C. & Rodes, Dispos. 32, 66–71. (2003) Proc. Natl. Acad. Sci. USA 100, 223–228. J. (1989) Lancet 1, 574–576.

2676 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0500159102 Kliewer Downloaded by guest on September 23, 2021