Title BIOLOGICAL ROLES of the VITAMIN D RECEPTOR in the REGULATION of TRANSPORTERS and ENZYMES on DRUG DISPOSITION, INCLUDING CY

Home , CYP3A43, Doxercalciferol

Title

BIOLOGICAL ROLES OF THE VITAMIN D RECEPTOR IN THE REGULATION OF

TRANSPORTERS AND ENZYMES ON DRUG DISPOSITION, INCLUDING

CYTOCHROME P450 (CYP7A1) ON CHOLESTEROL METABOLISM

Edwin Chiu Yuen Chow

A thesis submitted in conformity with the requirements

For the degree of Doctor of Philosophy

Department of Pharmaceutical Sciences

University of Toronto

Biological roles of the vitamin D receptor in the regulation of transporters and enzymes on drug disposition, including cytochrome P450 (CYP7A1) on cholesterol metabolism

Doctor of Philosophy (2012) Edwin Chiu Yuen Chow Department of Pharmaceutical Sciences, University of Toronto

ABSTRACT

Nuclear receptors play significant roles in the regulation of transporters and enzymes to balance the level of endogenous molecules and to protect the body from foreign molecules. The vitamin D receptor (VDR) and its natural ligand, 1,25-dihydroxyvitamin

D3 [1,25(OH)2D3], was shown to upregulate rat ileal apical sodium dependent bile acid transporter (Asbt) to increase the reclamation of bile acids, ligands of the farnesoid X receptor (FXR). FXR is considered to be an important, negative regulator of the cholesterol metabolizing enzyme, Cyp7a1, which metabolizes cholesterol to bile acids in the liver. In rats, decreased Cyp7a1 and increased P-glycoprotein/multidrug resistance protein 1 (P- gp/Mdr1) expressions pursuant to 1,25(OH)2D3 treatment was viewed as FXR effects in which hepatic VDR protein is poorly expressed. In contrast, changes in rat intestinal and renal transporters such as multidrug resistance associated proteins (Mrp2, Mrp3, and Mrp4),

Asbt, and P-gp after administration of 1,25(OH)2D3 were attributed directly as VDR effects due to higher VDR levels expressed in these tissues. Higher VDR expressions were found among mouse hepatocytes compared to those in rats. Hence, fxr(-/-) and fxr(+/+) mouse models were used to discriminate between VDR vs. FXR effects in murine livers. Hepatic

Cyp7a1 in mice was found to be upregulated with 1,25(OH)2D3 treatment, via the derepression of the short heterodimer partner (SHP). Putative VDREs, identified in mouse and human SHP promoters, were responsible for the inhibitory effect on SHP. The increase

in hepatic Cyp7a1 expression and decreased plasma and liver cholesterol were observed in mice prefed with a Western diet. A strong correlation was found between tissue Cyp7a1 and P-gp changes and 1,25(OH)2D3 plasma and tissue concentrations, confirming that VDR plays an important role in the disposition of xenobiotics and cholesterol metabolism.

Moreover, renal and brain Mdr1a/P-gp were found to be directly upregulated by the VDR in mice, and concomitantly, increased renal and brain secretion of digoxin, a P-gp substrate, in vivo. The important observations: the cholesterol lowering and increased brain P-gp efflux activity properties suggest that VDR is a therapeutic target for treatment of hypercholesterolemia and Alzheimer’s diseases, since beta amyloid, precursors of plague, are P-gp substrates.

iii

ACKNOWLEGMENTS

I would like to thank my supervisor, Dr. K. Sandy Pang who has given me great guidance and mentorship during my Ph.D. study. She has inspired me to enter the world of science and research.

I sincerely thank all the members in my advisory committees (Dr. David Hampson,

Dr. Reina Bendayan, Dr. Carolyn L. Cummins, and Dr. Cindy Woodland) for their kind help and suggestions.

I wish to thank my family members for their support, especially my late father, Hai

Woon Chow, who was a role model to me and who constantly reminded me that education is an asset that will lead to a better life. I thank my mother, Suk Lun Chow Wong, for her love and support throughout my studies.

I am very grateful to my colleagues, especially Drs. Hudaong Sun, Jianghong Fan,

Han-joo Maeng, and Matthew R. Durk, Cheng Jin, and Holly P. Quach, for their support and cooperation.

I wish to thank Lilia Magomedova, Rucha Patel, and Monika Patel in Dr. Cummins’ labortary for providing me with assistance in my Ph.D. project.

I would like to thank Dr. Geny M.M. Groothuis, Dr. Ansar K. Khan, and Myrte

Sondervan and the rest of Dr. Groothuis laboratory for their help and support when I was working at the University of Groningen, the Netherlands.

I wish to thank Dr. David D. Moore and Dr. Sayeepriyadarshini Anakk at the

Baylor College of Medicine at the Texas Medical Center and Dr. Reinhold Vieth and

Dennis Wagner at Mount Sinai Hospital, University of Toronto, for providing me assistance and resources for my Ph.D. project.

I also would like to thank the financial support from U of T open fellowship, NSERC

Canada Graduate Scholarship, and NSERC Michael Smith Foreign Study Supplements to travel aboard.

TABLE OF CONTENTS PAGE

Title ...... i ABSTRACT ...... ii ACKNOWLEGMENTS ...... iv TABLE OF CONTENTS...... vi LIST OF PUBLICATIONS...... xi ABBREVIATIONS AND TERMS ...... xiv LIST OF FIGURES ...... xix 1.1 INTRODUCTION...... 2 1.2 TRANSPORTERS AND ENZYMES AND THEIR REGULATION BY NUCLEAR RECEPTORS...... 3 1.2.1 Enzymes...... 4 1.2.2 Transporters ...... 5 1.2.3 The Bile Acid And Xenosensor Nuclear Receptors...... 6 1.2.3.1 The bile acid sensor: farnesoid X receptor (FXR)...... 6 1.2.3.2 The short heterodimer partner (SHP) ...... 7 1.2.3.3 The xenosenors: pregnane X receptor (PXR) and constitutive androstane receptor (CAR) ...... 8 1.3 THE VITAMIN D RECEPTOR (VDR)...... 9 1.3.1 VDR Protein Sequence Alignment in Human, Rat, and Mouse ...... 10 1.3.2 VDR Ligands...... 11

1.3.2.1 1,25-Dihydroxyvitamin D3 or 1,25(OH)2D3 ...... 11 1.3.2.2 1,25(OH)2D3 analogues...... 12 1.3.2.3 Alternative VDR ligands...... 13 1.3.3 The VDR Ligand Binding Pocket...... 13 1.3.4 Physiological Role of the 1,25(OH)2D3-Liganded VDR ...... 14 1.3.5 Disease Association with Vitamin D Deficiency...... 17 1.3.5.1 Hyperparathyroidism...... 17 1.3.5.2 Cardiovascular disease and hypertension...... 17 1.3.5.3 Inflammation and autoimmune disease...... 18 1.3.5.4 Diabetes...... 18 1.3.5.5 Cancer...... 19 1.4 VDR ON TRANSPORTERS, ENZYMES, AND NUCLEAR RECEPTORS...... 20 1.4.1 VDR Regulates Phase I and Phase II Enzymes ...... 20 1.4.2 VDR Regulates Transporters ...... 21 1.4.3 VDR and Cross-Talk with Other Nuclear Receptors...... 22 1.4.4 Significance of VDR in Transporters, Enzymes, and Nuclear Receptor Interactions...... 22 1.5 BILE ACIDS AND CHOLESTEROL HOMEOSTASIS...... 25 1.5.1 Metabolic Pathways of Cholesterol Metabolism ...... 26 1.5.2 Regulation of Cholesterol Metabolism ...... 27 1.5.3 Species Differences in Transporter and Enzyme Regulation...... 29 1.5.4 The Link between VDR, Bile Acids and Cholesterol...... 30 1.6 SIGNIFICANCE OF VDR IN REGULATION OF TRANSPORTERS AND ENZYMES ...... 32 2.1 STATEMENT OF PURPOSE OF INVESTIGATION ...... 34 2.2 HYPOTHESES ...... 36 vi

2.3 THESIS OUTLINE ...... 36 3. DIRECT AND INDIRECT EFFECTS OF THE VITAMIN D RECEPTOR (VDR) ON TRANSPORTERS AND ENZYMES IN THE RAT INTESTINE, LIVER AND KIDNEY IN VIVO...... 37 3.1 ABSTRACT ...... 38 3.2 INTRODUCTION...... 39 3.3 METHODS...... 44 3.3.1 Materials ...... 44 3.3.2 1,25(OH)2D3 and Vehicle (Corn Oil) Treatment in Rats In Vivo ...... 45 3.3.3 Blood Analysis and Preparation of Tissues ...... 45 3.3.4 Preparation of Subcellular Fractions from Enterocytes ...... 46 3.3.5 Preparation of Subcellular Fractions of Liver Tissue ...... 47 3.3.6 Liver Microsomal Cyp7a1 Activity ...... 48 3.3.7 Western Blotting...... 49 3.3.8 Quantitative Real-Time Polymerase Chain Reaction (qPCR) ...... 50 3.3.9 Statistical Analysis...... 51 3.4 RESULTS...... 54

3.4.1 Effect of 1,25(OH)2D3 Treatment on Portal Bile Acid and ALT Levels ...... 54 3.4.2 Effect of 1,25(OH)2D3 Treatment on Nuclear Receptors (NRs), Enzymes and Transporters in Intestinal Segments and Colon...... 54 3.4.2.1 Intestinal nuclear receptors, NRs...... 54 3.4.2.2 Intestinal enzymes...... 58 3.4.2.3 Intestinal apical absorptive transporter, Asbt ...... 60 3.4.2.4 Intestinal apical efflux transporter, Mdr1a (P-gp)...... 61 3.4.2.5 Intestinal apical efflux transporter, Mrp2...... 61 3.4.2.6 Intestinal basolateral efflux transporter, Mrp3 ...... 62 3.4.2.7 Intestinal basolateral efflux transporter, Mrp4 ...... 62 3.4.2.8 Intestinal basolateral efflux transporter, Ost-Ost...... 66 3.4.3 Effect of 1,25(OH)2D3 on Hepatic Nuclear Receptors, Enzymes, and Transporters ...... 66 3.4.3.1 Hepatic nuclear receptors, NRs...... 66 3.4.3.2 Hepatic cytrochrome P450, Cyps...... 67 3.4.3.3 Hepatic transporters...... 67 3.4.4 Effect of 1,25(OH)2D3 on Nuclear Receptors, Enzymes, Drug Transporters in the Kidney...... 68 3.4.4.1 Renal nuclear receptors, NRs ...... 68 3.4.4.2 Renal cytochrome P450, Cyps ...... 68 3.4.4.3 Renal transporters...... 69 3.5 DISCUSSION ...... 74 3.6 ACKNOWLEDGMENTS...... 81 3.7 STATEMENT OF SIGNIFICANCE OF CHAPTER 3...... 81

4. 1,25-DIHYDROXYVITAMIN D3 UPREGULATES P-GLYCOPROTEIN ACTIVITIES, EVIDENCED BY INCREASED RENAL AND BRAIN EFFLUX OF DIGOXIN IN MICE IN VIVO ..... 83 4.1 ABSTRACT ...... 84 4.2 INTRODUCTION...... 85 4.3 METHODS...... 87 4.3.1 Materials ...... 87 4.3.2 Induction Studies with 1,25(OH)2D3 in fxr(+/+) and fxr(-/-) Mice In Vivo...... 88 4.3.3 Preparation of subcellular fractions ...... 89 4.3.4 Western Blotting...... 90 vii

4.3.5 Quantitative Real-Time Polymerase Chain Reaction (qPCR) ...... 91 3 4.3.6 Pharmacokinetic Study of [ H]Digoxin in Vehicle or 1,25(OH)2D3 Treated Mice ...... 92 4.3.7 [3H]Digoxin Analyses...... 92 4.3.8 Modeling and Fitting ...... 94 4.3.8.1 Whole body physiologically-based pharmacokinetic modeling (PBPK) ...... 94 4.3.8.2 Fitting...... 96 4.3.9 Statistical Analysis...... 97 4.4 RESULTS...... 97 4.4.1 VDR and Mdr1a/P-gp mRNA and protein expression in the ileum, liver, kidney and brain of fxr(+/+) and fxr(-/-) mice ...... 97 4.4.1.1 Distribution of VDR protein expression among tissues ...... 97 4.4.1.2 Effects of 1,25(OH)2D3 on Mdr1 mRNA and P-gp protein expression in both fxr(+/+) and fxr(-/-) mice ...... 98 3 4.4.2 Effects of 1,25(OH)2D3 Treatment on the Pharmacokinetics of [ H]Digoxin in fxr(+/+) Mice...... 102 4.4.2.1 Blood decay profiles and excretion of [3H]digoxin after intravenous administration in fxr(+/+) mice ...... 102 4.4.3 Tissue Distribution...... 106 4.4.3.1 Estimation of area under the curves (AUCs)...... 106 4.4.3.2 Tissue to blood AUC vs. time profile...... 106 4.4.4 Whole Body PBPK Modeling...... 107 4.5 DISCUSSION ...... 116 4.6 APPENDIX ...... 120 4.7 ACKNOWLEDGMENTS...... 122 4.8 STATEMENT OF SIGNIFICANCE OF CHAPTER 4...... 122 5. INHIBITION OF THE SMALL HETERODIMER PARTNER (SHP) BY 1,25- DIHYDROXYVITAMIN D3-LIGANDED VITAMIN D RECEPTOR (VDR) REMOVED THE REPRESSION ON CYTOCHROME 7-HYDROXYLASE (CYP7A1) AND INDUCED CHOLESTEROL LOWERING ...... 124 5.1 ABSTRACT ...... 125 5.2 INTRODUCTION...... 125 5.3 METHODS...... 129 5.3.1 Materials ...... 129 5.3.2 Plasmids...... 129 5.3.3 1,25(OH)2D3 Treatment of Mice...... 130 5.3.4 Preparation of Subcellular Tissue Fractions ...... 131 5.3.5 Immunostaining ...... 131 5.3.6 Real-Time PCR (qPCR)...... 132 5.3.7 Western Blotting...... 132 5.3.8 Cyp7a1 Activity in Microsomes ...... 132 5.3.9 Blood Analyses...... 133 5.3.10 Liver Cholesterol ...... 133 5.3.11 Mouse Primary Hepatocyte Isolation...... 134 5.3.12 Cell Culture and Transfection Assays...... 134 5.3.13 Preparation of Nuclear Protein Extracts...... 135 5.3.14 Electrophoretic Mobility Shift Assay (EMSA)...... 135 5.3.15 Statistics...... 136 5.4 RESULTS...... 136 5.4.1 VDR Protein in Mouse Liver...... 136 viii

5.4.2 1,25(OH)2D3 Increases Hepatic Cyp7a1, Decreases Hepatic SHP and Lowers Cholesterol In Vivo...... 138 5.4.3 Correlation Between Cyp7a1 and SHP But Not FGF15...... 138 .4.4 1,25(OH)2D3 Lowers Cholesterol in C57BL/6 or fxr(+/+), fxr(-/-), and shp(-/-) Mice Fed a High Fat/High Cholesterol Diet...... 142 5.4.5 1,25(OH)2D3 Increases Cyp7a1 mRNA and Inhibits SHP Levels in Mouse Primary Hepatocytes .... 143 5.4.6 VDR Activation Strongly Represses Mouse and Human SHP Promoter Activity ...... 143 5.4.6.1 Binding to the VDREs of SHP...... 144 5.4.6.2 EMSA...... 145 5.5 DISCUSSION ...... 151 5.6 ACKNOWLEDGMENTS...... 155 5.7 STATEMENT OF SIGNIFICANCE OF CHAPTER 5...... 155

6. CORRELATION BETWEEN TISSUE 1,25-DIHYDROXYVITAMIN D3 LEVELS AND GENE CHANGES: A TEMPORAL STUDY...... 158 6.1 ABSTRACT ...... 159 6.2 INTRODUCTION...... 159 6.3 METHODS...... 162 6.3.1 Materials ...... 162 6.3.2 Pharmacokinetic Study of 1,25(OH)2D3 in Mice ...... 162 6.3.3 Plasma Calcium and Phosphorus Analysis ...... 163 6.3.4 Tissue 1,25(OH)2D3 Extraction and 1,25(OH)2D3 Enzyme-immunoassay (EIA) for Plasma and Tissue Samples ...... 163 6.3.5 Pharmacokinetic Analysis: Plasma Concentration-Time Profile ...... 164 6.3.6 Preparation of Subcellular Protein Fractions of Kidneys...... 164 6.3.7 Western Blotting...... 165 6.3.8 Quantitative Real-Time Polymerase Chain Reaction (real-time PCR or qPCR) ...... 166 6.4 RESULTS...... 168

6.4.1 Plasma Concentration of 1,25(OH)2D3 and Calcium in Single and Multiple Doses of 1,25(OH)2D3 in Mice ...... 168 6.4.2 Tissue Concentrations of 1,25(OH)2D3 in Single and Multiple Doses of 1,25(OH)2D3 in Mice ..... 170 6.4.3 Comparison of Renal Cyp24 mRNA and Protein in a Single or Multiple Doses of 1,25(OH)2D3 in Mice ...... 171 6.4.4 Temporal Changes in Ileal Cyp24, TRPV6 and FGF15, Liver Cyp24, Cyp7a1 and SHP, and Renal Cyp24, Mdr1 and TRPV6 mRNA Expressions after Multiple Doses of 1,25(OH)2D3 in Mice...... 171 6.5 DISCUSSION ...... 176 6.6 ACKNOWLEDGMENTS...... 178 6.7 STATEMENT OF SIGNIFICANCE OF CHAPTER 6...... 178 7. GENERAL DISCUSSION AND CONCLUSIONS...... 180 REFERENCES ...... 190 APPENDIX A1...... 205 APPENDIX A2...... 217 APPENDIX A3...... 236 APPENDIX A4...... 248

APPENDIX A5...... 267 APPENDIX A6...... 282 APPENDIX T1 ...... 294 APPENDIX T2 ...... 296 APPENDIX T3 ...... 298 APPENDIX T4 ...... 298 APPENDIX T5 ...... 299 APPENDIX T6 ...... 299 APPENDIX T7 ...... 300 APPENDIX T8 ...... 300 APPENDIX T9 ...... 301 APPENDIX F1 ...... 302 APPENDIX F2 ...... 303 APPENDIX F3 ...... 304 APPENDIX F4 ...... 305

LIST OF PUBLICATIONS Peer Reviewed 1. Fan J, Wong B, Hochman J, Chow ECY, and Pang K S (2012) Intracellular stability and hepatic uptake of MK-8544, a cationic lipid nanoparticle/siRNA complex, in rat liver. (submitted). 2. Durk MR, Chow ECY, Henderson JT, and Pang KS (2012) Induction of P- glycoprotein by the Vitamin D receptor ligand, 1α,25-dihydroxyvitamin D3, reduces brain accumulation of amyloid-beta peptides. (submitted). 3. Chow ECY, Durk MR, Cummins CL and Pang KS (2011) 1α,25-Dihydroxyvitamin D3 upregulates P-glycoprotein via the vitamin D receptor and not farnesoid X receptor in both fxr(-/-) and fxr(+/+) mice and increased renal and brain efflux of digoxin in mice in vivo. J Pharmacol Exp Ther 337:846-859. 4. Maeng HJ, Durk MR, Chow ECY, Ghoneim R and Pang KS (2011) 1α,25- Dihydroxyvitamin D3 on intestinal transporter function: studies with the rat everted intestinal sac. Biopharm Drug Dispos 32:112-125. 5. Chow ECY, Sondervan M, Jin C, Groothuis GM and Pang KS (2011) Comparative effects of doxercalciferol (1α-hydroxyvitamin D2) versus calcitriol (1α,25- dihydroxyvitamin D3) on the expression of transporters and enzymes in the rat in vivo. J Pharm Sci 100:1594-1604. 6. Khan AA, Chow ECY, Porte RJ, Pang KS and Groothuis GM (2011) The role of lithocholic acid in the regulation of bile acid detoxification, synthesis, and transport proteins in rat and human intestine and liver slices. Toxicol In Vitro 25:80-90. 7. Chow ECY, Sun H, Khan AA, Groothuis GM and Pang KS (2010) Effects of 1α,25- dihydroxyvitamin D3 on transporters and enzymes of the rat intestine and kidney in vivo. Biopharm Drug Dispos 31:91-108. 8. Chow ECY, Maeng HJ, Liu S, Khan AA, Groothuis GM and Pang KS (2009) 1α,25- Dihydroxyvitamin D3 triggered vitamin D receptor and farnesoid X receptor-like effects in rat intestine and liver in vivo. Biopharm Drug Dispos 30:457-475. 9. Khan AA, Chow ECY, van Loenen-Weemaes AM, Porte RJ, Pang KS and Groothuis GM (2009) Comparison of effects of VDR versus PXR, FXR and GR ligands on the regulation of CYP3A isozymes in rat and human intestine and liver. Eur J Pharm Sci 37:115-125. 10. Khan AA, Chow ECY, Porte RJ, Pang KS and Groothuis GM (2009) Expression and regulation of the bile acid transporter, OST-OST in rat and human intestine and liver. Biopharm Drug Dispos 30:241-258. 11. Sun H, Zhang L, Chow ECY, Lin G, Zuo Z and Pang KS (2008) A catenary model to study transport and conjugation of baicalein, a bioactive flavonoid, in the Caco-2 cell monolayer: demonstration of substrate inhibition. J Pharmacol Exp Ther 326:117-126.

12. Chow ECY, Liu L, Ship N, Kluger RH and Pang KS (2008) Role of haptoglobin on the uptake of native and beta-chain [trimesoyl-(Lys82)-(Lys82)] cross-linked human hemoglobins in isolated perfused rat livers. Drug Metab Dispos 36:937-945. 13. Chen Y, Whetstone HC, Youn A, Nadesan P, Chow ECY, Lin AC and Alman BA (2007) Beta-catenin signaling pathway is crucial for bone morphogenetic protein 2 to induce new bone formation. J Biol Chem 282:526-533.

Review article/Chapters 1. Chow ECY and Pang KS (2012) The segregated-flow model is important for consideration of drug absorption. Curr Drug Metab submitted 2. Maeng HJ, Chow ECY, Chen S, Fan J, and Pang KS (2011) Physiologically-based pharmacokinetic models In, Encyclopedia of Drug Metabolism and Interactions (Alexander V. Lyubimov, ed) Chapter 16, Wiley and Sons, New Jersey. 3. Pang KS, Sun H, and Chow ECY (2010) Impact of physiological determinants: flow, binding, transporters and enzymes on organ and total body clearances, in “Enzymatic- and Transporter-Based Drug-Drug Interactions: Progress and Future Challenges” (KS Pang, AD Rodrigues, and RM Peter, eds) Chapter 5, Springer, NY pp. 107-147. 4. Fan J, Chen S, Chow ECY and Pang KS (2010) PBPK modeling of intestinal and liver enzymes and transporters in drug absorption and sequential metabolism. Curr Drug Metab 11:743-761. 5. Sun H, Chow ECY, Liu S, Du Y and Pang KS (2008) The Caco-2 cell monolayer: usefulness and limitations. Expert Opin Drug Metab Toxicol 4:395-411.

Conference Abstracts 1. Chow ECY, Magomedova L, Patel R, Maeng HJ, Fan J, Durk MR, Irondi K, Cummins CL and Pang KS (2011) 1,25-Dihydroxyvitamin D3-liganded vitamin D receptor (VDR) derepressed cytochrome P450 7A1 (CYP7A1) and lowered cholesterol via inhibition of the small heterodimer partner (SHP). AAPS Annual Meeting, Washington, DC 2011. (recipient of 2011 AAPS Lilly Graduate Student Symposium Award) 2. Quach HP, Durk MR, Chow ECY, and Pang KS (2011) Vitamin D receptor (VDR) effects of lithocholic acid acetate (LCAa), an alternate VDR ligand, on liver, kidney and brain of mice mimic those of 1,25-dihydroxyvitamin D3. AAPS Annual Meeting, Washington DC, October 2011. 3. Durk MR, Chow ECY, and Pang KS (2011) Regulation of brain P-glycoprotein by the vitamin D receptor leads to reduced accumulation of beta amyloids. AAPS National Biotechnology Conference, San Francisco, CA, 2011. (received 2011 AAPS Innovation in Biotechnology Award) 4. Chow ECY, Maeng HJ, Khan AA, Groothuis GM.M, Pang KS (2009) 1α,25- Dihydroxyvitamin D3 triggered vitamin D receptor and farnesoid X receptor-like effects in rat intestine, liver, and kidney in vivo. AAPS Annual Meeting, Los Angeles, CA 2009. xii

5. Fan J, Wong B, Chow ECY, Hochman J, Pang KS (2009) Evaluation of intracellular stability and hepatic uptake of MK-8544, a cationic lipid nanoparticle/siRNA complex in the single pass perfused rat liver. AAPS Annual Meeting, Los Angeles, CA 2009. 6. Khan AA, Chow ECY, Pang KS, Groothuis GMM (2008) Expression and regulation of organic solute transporter (OST/) in rat intestine. ISSX Europe, Vienna, 2008. 7. Chow ECY, Liu S, Khan AA, Groothuis GMM and Pang KS (2008) Effects of 1α,25-dihydroxyvitamin D3, the natural ligand of the vitamin D receptor (VDR), on transporters, enzymes, and nuclear receptors in the rat intestine and liver. AAPS Annual Meeting, Atlanta, GA, 2008.

8. Chow ECY, Pang KS (2007) Effects of 1α,25-dihydroxyvitamin D3 on rat intestinal transporters and enzymes involved in bile acid homeostasis by the vitamin D receptor (VDR). Pittsburgh, PA July 27-29, 2007. (Second prize poster award) 9. Chow ECY, Khan AA., Pang KS, Elferink MGL, Groothuis GMM (2007) Precision cut rat intestine tissue slices to study regulation of transporters and enzymes involved in bile acid homeostasis by the vitamin D receptor (VDR): validation with in vivo data. AAPS Workshop, Transporters in ADME, from Bench to Bedside III. Bethesda, MD March 5-7, 2007. 10. Zhao C, Fischer H, Liu J, Liu L, Bergin C, Chow ECY, Chan WCW, Pang KS (2006) Nanogold particles of varying sizes to define changes in permeability of the normal, sham-operated, and metastatic perfused rat liver. Boston, June 2006. (poster award winner) 11. Sun H., Zhang L, Chow ECY, Lin G, Zuo Z, Pang KS. Modeling of transport and metabolism of baicalein (B), a bioactive flavone Chinese herbal medicine, and baicalein 7-glucuronide (BG) in Caco-2 cells. AAPS, October, San Antonio 2006. 12. Chow ECY, Ship N, Liu L, Kluger R, Pang KS. The role of haptoglobin on the uptake of native and cross-linked human hemoglobins in rat liver. First Great Lakes Symposium on Pharmaceutical Sciences at University of Toronto. Toronto July, 2005.

xiii

ABBREVIATIONS AND TERMS

1,25(OH)2D3 1α,25-dihydroxyvitamin D3 or calcitriol 7α-HCO 7α-hydroxy-4-cholesten-3-one 7β-HCO 7β-hydroxycholesterol ABC ATP-binding cassette ALT alanine aminotransferase APC antigen presenting cells Asbt/ASBT rodent/human apical sodium dependent bile acid transporter AUC area under the curve BARE bile acids response element B/P blood/plasma concentration ratio Bsep/BSEP rodent/human bile salt export pump CA cholic acid Caco-2 human epithelial colorectal adenocarcinoma cell line cAMP cyclic adenosine monophosphate CAR constitutive androstane receptor CaSR calcium-sensing receptor CDCA chenodeoxycholic acids CETP cholesteryl ester transfer protein Cyp24/CYP24 rodent/human cytochrome P450 24-hydroxylase cGMP cyclic guanosine monophosphate CHAPS 3-[(3-cholamidopropyl)dimethylammonio]-1- propanesulfonate CL clearance CNS central nervous system CT calcitonin Cyp/CYP cytochrome P450 enzyme DCA deoxycholic acid DDI drug-drug interaction DTT dithiothreitol

xiv

EHBR Eisai hyperbilirubinemic rat EHC enterohepatic circulation EIA enzyme-immunoassay fP and fB unbound fraction in plasma and blood, respectively FGF fibroblast growth factor FGFR fibroblast growth factor receptor FR fraction reabsorbed FXR farnesoid X receptor Gapdh/GAPDH rodent/human glyceraldehyde-3-phosphate dehydrogenase HDL-C high-density lipoprotein cholesterol HEK293 human embryonic kidney 293 cell line HepG2 human hepatocellular liver carcinoma cell line HIV human immunodeficiency virus HNF hepatocyte nuclear factor HPLC high pressure liquid chromatography ICP-AES inductively coupled plasma atomic emission spectroscopy INF-γ interferon gamma IL interleukin k elimination rate constant ka absorption rate constant

KTB tissue/blood partition coefficient LCA lithocholic acid LRH-1 liver receptor homolog-1 LXRα liver X receptor alpha Mdr1/MDR1/P-gp rodent/human multidrug resistance protein 1 or P- glycoprotein (P-gp) Mrp/MRP rodent/human multidrug resistance associated protein MSC model selection criterion NADPH nicotinamide adenine dinucleotide phosphate Ntcp/NTCP rodent/human sodium taurocholate co-transporting polypeptide xv

Oatp/OATP rodent/human organic anion transporting polypeptides Ost/OST organic solute transporter PBPK physiologically-based pharmacokinetic PMSF phenylmethylsulfonyl fluoride PPAR peroxisome proliferator-activated receptor PTH parathyroid hormone PTG parathyroid gland PXR pregnane X receptor Q organ flow rate RXRα retinoid X receptor alpha SHP short heterodimer partner SLC solute carriers Sult2a1/SULT2A1 rodent/human hydroxysteroid sulfotransferase 2A1 TBS-T Tris-buffered saline with 0.1% Tween 20 TGF-1 transforming growth factor beta 1 TNF-α tumor necrosis factor alpha TRPV transient receptor potential cation channel, subfamily V UGT UDP-glucuronosyltransferase V blood and tissue volumes VDR vitamin D receptor VDRE vitamin D response element

xvi

LIST OF TABLES

Chapter 1 Table 1-1 A summary table of common nuclear receptors/transcription factors and their ligands that when activated, changes the level of transporters and enzymes and other nuclear receptors/transcription factors Table 1-2 Regulation of enzymes by VDR activation Table 1-3 Regulation of transporters by VDR activation Table 1-4 Regulation of nuclear receptor cross-talked by VDR activation

Chapter 3 Table 3-1 Rat primer sets for quantitative real-time PCR Table 3-2 Changes in body weight and blood analysis with various intraperitoneal

injections of 1,25-dihydroxyvitamin D3 treatment for 4 days to the rat in vivo Table 3-3 Changes in mRNA expression of rat hepatic nuclear receptors, enzymes, and transporters, expressed as fold expression compared to vehicle treatment Table 3-4 Changes in mRNA expression of rat renal nuclear receptors, enzymes, and transporters, expressed as fold expression compared to vehicle treatment

Chapter 4 Table 4-1 Mouse primer sets for quantitative real-time PCR Table 4-2 Noncompartmental estimates of digoxin parameters in tissue and blood of

vehicle- and 1,25(OH)2D3-treated wild-type mice Table 4-3 Assigned parameters for PBPK modeling of [3H]digoxin for fxr(+/+) mice

which were treated i.p. with vehicle (0) or 2.5 μg/kg of 1,25(OH)2D3 every other day for 8 days Table 4-4 Fitted parameters (±SD) for [3H]digoxin for fxr(+/+) mice treated i.p. with

vehicle (0) or 2.5 μg/kg of 1,25(OH)2D3 every other day for 8 days based on the PBPK model shown in Fig. 4-1

xvii

Table 4-5 Correlation between fold-changes in protein expression of P-gp (from Western blotting in Fig. 4-3) and ratio of the estimated apparent efflux intrinsic clearances of P-gp (from PBPK modeling) between the

1,25(OH)2D3- and vehicle-treated mice

Chapter 5 Table 5-1 Mouse Primer Sequences

Chapter 6 Table 6-1 Mouse primer sets for quantitative real-time PCR

Table 6-2 Pharmacokinetic parameters of 1,25(OH)2D3 in mice

xviii

LIST OF FIGURES

Chapter 1

Figure 1-1 Common nuclear receptor structure

Figure 1-2 Bioactivation of vitamin D

Figure 1-3 Alternative Vitamin D analog and LCA derivative structures from Brown

and Sltopolsky (2008) and Ishizawa et al. (2008)

Figure 1-4 1,25(OH)2D3 and plasma calcium homeostasis

Figure 1-5 Regulation of Cyp7a1 by bile acids in the liver

Figure 1-6 Species differences in negative feedback regulation of rat and mouse Asbt

Chapter 3

Figure 3-1 Distribution and dose-dependent effects of 1,25(OH)2D3 on rat intestinal

VDR mRNA and protein

Figure 3-2 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal (A)

FXR (B) SHP and (C) LRH-1 mRNA

Figure 3-3 Dose-dependent effects of 1,25(OH)2D3 on intestinal FGF15 mRNA in the

ileum

Figure 3-4 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal Cyp3a

enzymes

Figure 3-5 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal Cyp24

mRNA

Figure 3-6 Dose-dependent effects of 1,25(OH)2D3 on intestinal mRNA and protein of

Asbt in the ileum

xix

Figure 3-7 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal Mdr1a

mRNA and P-gp protein

Figure 3-8 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal

mRNA and protein of Mrp2

Figure 3-9 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal

mRNA and protein of Mrp3

Figure 3-10 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal

mRNA and protein of Mrp4

Figure 3-11 Dose-dependent effects of 1,25(OH)2D3 on intestinal Ostα and Ostβ mRNA

in the ileum

Figure 3-12 Dose-dependent effects of 1,25(OH)2D3 on changes in protein of hepatic

cytochrome P450 isozymes (A), and sinusoidal (B) and canalicular (C)

transporters

Figure 3-13 Dose-dependent effects of 1,25(OH)2D3 on changes in (B) protein of renal

nuclear receptor, cytochrome P450 isozymes and transporters

Figure 3-14 A schematic diagram highlighting direct and indirect effects of 1,25(OH)2D3

on intestinal and hepatic nuclear receptors, drug transporters and enzymes

Chapter 4

Figure 4-1 Whole body PBPK modeling with enterohepatic circulation and renal

reabsoprtion of [3H]digoxin

Figure 4-2 Effects of 1,25(OH)2D3 on VDR (A) mRNA and (B) protein expression in

the ileum, liver, kidney, and brain

Figure 4-3 Effects of 1,25(OH)2D3 on Mdr1 (A) mRNA and (B) P-gp protein in the

brain, kidney, liver, and ileum

Figure 4-4 Plots of [3H]digoxin (A) blood concentration, and cumulative amounts in (B)

urine and (C) feces vs. time

Figure 4-5 Plot of the amount [3H]digoxin excreted to (A) urine and (B) feces vs. the

blood AUC(0→t)

Figure 4-6 Plots of amount [3H]digoxin in (A) small intestine, (B) liver, (C) kidney, (D)

brain, and (E) heart

Figure 4-7 Tissue to blood AUC ratio of [3H]digoxin over time profile for the (A) small

intestine, (B) liver, (C) kidney, (D) brain and (E) heart

Figure 4-8 Simulation of [3H]digoxin concentrations vs. time for (A) blood, (B) kidney

and (C) brain, and amounts vs. time in urine (D)

Chapter 5

Figure 5-1 Tissue distribution and localization of the VDR

Figure 5-2 Effect of 1,25(OH)2D3 treatment on (A) serum bile acids and plasma and

liver cholesterol and (B) Cyp7a1 mRNA and protein expressions and

activity in fxr(+/+) and fxr(-/-) mice

Figure 5-3 Changes in (A) hepatic FXR, SHP, LRH-1, LXR, and HNF-4 and (B)

ileal FXR, SHP, LRH-1, FGF15, and Asbt mRNA expressions in both

fxr(+/+) and fxr(-/-) mice

xxi

Figure 5-4 Correlation between murine Cyp7a1 mRNA, protein, and catalytic activity

vs. SHP mRNA (A, left column) and FGF-15 mRNA (B, right column) in

fxr(+/+) mice which were treated with 1,25(OH)2D3

Figure 5-5 Effects of 1,25(OH)2D3 treatment on (A) serum bile acids and plasma and

liver cholesterol and (B) Cyp7a1 mRNA and protein expressions in wild-

type, fxr(-/-) and shp(-/-) mice prefed with a high fat/high cholesterol diet (n

= 4-6) for 3 weeks

Figure 5-6 Changes in (A) hepatic FXR, SHP, LRH-1, LXR, and HNF-4 and (B)

ileal FXR, SHP, LRH-1, FGF15, and Asbt mRNA expressions in wild-type,

fxr(-/-) and shp(-/-) mice prefed with a high fat/high cholesterol diet (n = 4-6)

for 3 weeks

Figure 5-7 Gene expression changes in mouse primary hepatocytes treated with 100 nM

1,25(OH)2D3

Figure 5-8 1,25(OH)2D3 suppresses SHP expression via direct binding of VDR to a

DR3 response element located within the proximal SHP promoter

Chapter 6

Figure 6-1 Plasma 1,25(OH)2D3 and calcium concentration-time profiles after (A) a

single dose or (B) multiple doses of 2.5 µg/kg i.p. 1,25(OH)2D3 or vehicle to

mice.

Figure 6-2 Tissue (ileum, liver, kidney, and brain) 1,25(OH)2D3 concentration-time

profile from (A) a single dose or (B) multiple doses of 2.5 µg/kg i.p.

1,25(OH)2D3 to mice

xxii

Figure 6-3 Renal Cyp24 mRNA and protein expressions resulted in (A) a single dose or

(B) multiple doses of 2.5 µg/kg i.p. 1,25(OH)2D3 to mice

Figure 6-4 mRNA expressions of (A) ileal Cyp24, TRPV6 and FGF15, (B) hepatic

Cyp24, Cyp7a1 and SHP, and (C) renal Cyp24, Mdr1 and TRPV6 after

multiple doses of 2.5 µg/kg i.p. 1,25(OH)2D3 to mice

Chapter 7

Figure 7-1 Summary of nuclear receptor, transporter and enzyme changes in the

intestine, liver, and kidney of the rat treated with 1,25(OH)2D3

Figure 7-2 VDR increases cholesterol metabolism and lowers cholesterol via repression

of hepatic SHP (major mechanism) and possibly intestinal FGF15 (minor

mechanism)

xxiii

CHAPTER 1

THE VITAMIN D RECEPTOR (VDR)

1.1 INTRODUCTION

Transporters and enzymes play a critical role in the biological fates of endo- and exogenous molecules. In the past few decades, much was learnt on how the transporters and enzymes are regulated in the body, especially in the small intestine, liver, and kidney. It is now known that nuclear receptors and transcription factors play a crucial role in regulating the existence of these proteins (Mangelsdorf et al., 1995; Makishima, 2005). All nuclear receptors have a common structure (Fig. 1-1), which includes an amino terminal ligand-independent activation domain (AF-1) for the interaction with cofactors, a central

DNA binding domain (DBD) consisting of two zinc finger motifs and targets the nuclear receptor to highly specific DNA sequences or DNA response elements, a hinge region, and a carboxy-terminal ligand binding domain (LBD) that differs for every nuclear receptor and allows for specific hormonal and nonhormonal ligand binding, receptor dimerization and coregulator interactions (AF-2) for biological response (Mangelsdorf et al., 1995; Wagner et al., 2011). These nuclear receptors contain variable length and sequences in the N- terminal and C-terminal domains and the length of the hinge region between the DBD and

LBD. In absence of a ligand, nuclear receptors are either located unliganded in the cytoplasm or in the nucleus binding to their DNA response elements repressed by a corepressor complex. In the presence of a ligand, they can form as a homo- or heterodimers with each partner, acting as a transcription factor and binding to a specific response element sequence that is present as half-sites of direct or inverted repeats separated by different length of nucleotides (Olefsky, 2001). Currently, these nuclear receptors are categorized into four different classes (Mangelsdorf et al., 1995): Class 1 nuclear receptors are known as steroid hormone nuclear receptors, functioning as homodimers binding to response

element of inverted repeats; Class 2 nuclear receptors are adopted orphan nuclear receptors that form heterodimer with retinoid X receptors (RXR) and function in a ligand dependent manner; Class 3 and Class 4 receptors are orphan receptors that function as homodimers binding to direct repeats in the response element or as momers binding to a single site response element, respectively. The NR1 superfamily is commonly known for their role in the regulation of transporters and enzymes affecting the balance of not only endogenous molecules such as cholesterol, bile acids, and ions, but as well as the disposition of xenobiotics.

N AF-1 DBD Hinge LBD AF-2 C

Figure 1-1 Common nuclear receptor structure

1.2 TRANSPORTERS AND ENZYMES AND THEIR REGULATION BY NUCLEAR RECEPTORS

It has long been known that adaptive biological responses are present in the body to combat potential toxic effects of foreign substances such as xenobiotics. Early studies have illustrated that phenobarbtial treatment decreased the plasma concentration of phenytoin and coumarin anticoagulants, most likely due to upregulation of drug metabolism or elimination to increase clearance (Cucinell et al., 1963; Schoene et al., 1972; Remmer et al.,

1973). Studies later revealed that a transcriptional mechanism was involved in the increase in hepatic drug metabolism (Adesnik et al., 1981), likely due to presence of a xenobiotic sensing and response system. Today, we know that proteins in the cell such as CYP and membrane proteins on cell membrane bilayers, known as transporters, are responsible for the fate of both endogenous molecules such as bile acids and cholesterol and xenobiotics, 3

and that they are under the regulation of nuclear receptors that effect changes in levels of proteins and in turn, alter the disposition of these molecules. A summary table of the ligands for various nuclear transporters is listed in Table 1-1.

Table 1-1 A summary table of common nuclear receptors/transcription factors and their ligands, which when activated, could change levels of transporters and enzymes and other nuclear receptors/transcription factors

Nuclear Receptors/ Ligands Target Genes Transcription Factors Rifampin, ↑CYP3A4, ↑ CYP2C9, ↑ CYP2B6, PXR (Pregnane X Receptor) phenobarbital, ↑SULT2A1, ↑UGT1A1, ↑rOatp1a4, dexamethasone, PCN ↑MRP2, ↑P-gp, ↑CYP3A4, ↑CYP2C9, ↑ CYP2B6, CAR (Constitutive Androstane Phenobarbital ↑SULT2A1, ↑MRP2, ↑MRP3, ↑MRP4, Receptor) ↑P-gp 1,25(OH) D , VDR (Vitamin D Receptor) 2 3 ↑CYP3A4, ↑SULT2A1, ↑ASBT, ↑MRP3 lithocholic acid FXR (Farnesoid X Receptor ) Bile Acids ↑ SHP, ↓CYP7A1, ↑MRP2, ↑BSEP LXR (Liver X Receptor) Oxysterol ↑CYP7A1, ↑ABCG5, ↑ABCG8, ↑ ABCA1 ↓LRH-1, ↓HNF-4α, ↓CYP7A1, ↓NTCP, SHP (Short Heterdimer Partner) ↓OATP1B1 HNF-4α (Hepatocyte Nuclear ↑HNF-1α, ↑CYP27A1, ↑CYP8B1, Factor 4α) ↑CYP7A1, ↑CYP3A LRH-1 (liver receptor homolog-1) ↑CYP7A1, ↑ASBT

1.2.1 Enzymes

Drug metabolism is one of the major routes of drug clearance or elimination. These enzymes include the Phase I enzymes such as the cytochrome P450s (CYP) and reductases, sulfatases, and glucuronidases, and Phase II enzymes for sulfation (SULT), glucuronidation

(UGT), glutathione conjugation (GST), N-acetylation (NAT), and methylation

(methyltransferases) (Grant et al., 1991; Omura, 1999; Ritter, 2000; Venkatakrishnan et al.,

2001; Pang et al., 2010). CYP3A4 is the most abundant CYP isoform in the liver and

intestine and is involved in the biotransformation of many xenobiotics (Guengerich, 1999;

Dresser et al., 2000). In addition, endogenous substances, such as cholesterol and bile acids, are found to be metabolized by this enzyme (Araya and Wikvall, 1999; Furster and Wikvall,

1999). Many of these enzymes are under the regulation of many nuclear receptors, which will be discussed later. Thus, changes in the level of enzymes can ultimately change the disposition of endogenous and exogenous molecules.

1.2.2 Transporters

Drug transport proteins transport endogenous and exogenous molecules in and out of cells and are categorized into two major classes, the solute carriers (SLC) and ATP- binding cassette (ABC) transporters (Dean and Allikmets, 2001; Kim, 2002b; Tirona and

Kim, 2005; Fredriksson et al., 2008; Szakacs et al., 2008; Klaassen and Aleksunes, 2011).

Solute carriers include the apical sodium dependent bile acids transporter (ASBT;

SLC10A2) in the ileum and kidney, sodium taurocholate cotransporting polypeptide

(NTCP; SLC10A1) in the liver, organic anion transporting polypeptides (OATPs) in the intestine, liver and kidney, and organic solute transporters (OSTα-OSTβ) in the ileum and liver where they are major proteins for the transport of bile acids in enterohepatic recirculation. ASBT is present on the apical membrane for intestinal absorption and NTCP and major OATPs on the sinusoidal membrane transport bile acids into hepatocytes, whereas OSTα-OSTβ located on the basolateral membrane efflux bile acids out of the intestine and liver. In addition to the transport of bile acids, some of these transporters such as NTCP and OATPs are able to uptake exogenous molecules such as HMG-CoA reductase inhibitors (statins) (Ho et al., 2006; Ho et al., 2007), angiotensin-converting enzyme inhibitors (Liu et al., 2006a), angiotensin receptor II antagonists (Ishiguro et al., 2006) and

cardiac glycosides (König et al., 2006; Klaassen and Aleksunes, 2011). ATP-binding cassette (ABC) transporters, present normally at the apical membrane, are able to efflux molecules against a concentration gradient and utilize ATP. These transporters include the apical membrane transporters, the bile acids export pump (BSEP; ABCB11) in the liver, the multidrug resistance associated proteins 2 (MRP2; ABCC2) and the multidrug resistance protein 1 or P-glycoprotein (MDR1/P-gp; ABCB1), and the basolateral transporter, MRP3

(ABCC3) in the intestine, liver, and kidney (Dean and Allikmets, 2001; Szakacs et al., 2008;

Klaassen and Aleksunes, 2011). MRP4 (ABCC4) is located on the basolateral membrane in the intestine and liver, and on the apical membrane in the intestine and kidney (van Aubel et al., 2002; Maeng et al., 2011). These ATP efflux transporters - BSEP, MRP2, MRP3, and

MRP4 - reduce the cellular concentration of bile acids (Zöllner et al., 2006), whereas P-gp is found to play a role in the transport of cholesterol (Leon et al., 2006; Tamashevskii et al.,

2011). These transporters are cellular protective because they protect the cell from toxicity by effluxing harmful drug molecules and their metabolites out of the cell. Over the past decades, many studies have alluded to the fact that these SLC and ABC transporters are under the regulation of nuclear receptors, and these changes can affect the disposition of drug chemical entities.

1.2.3 The Bile Acid And Xenosensor Nuclear Receptors

1.2.3.1 The bile acid sensor: farnesoid X receptor (FXR)

The FXR is in the NR1 superfamily (NR1H4) and is known as a bile acids sensor that affects the balance of bile acids and cholesterol in the body by regulating many bile acid related transporters and enzymes, mainly in the liver and intestine. FXR is expressed in liver, kidney, intestine and adrenal gland (Zhu et al., 2011). Bile acids, such as the

chenodeoxycholic acids (CDCA), lithocholic acid (LCA), deoxycholic acid (DCA), and cholic acid (CA), and conjugates of the bile acids are the endogenous ligands of FXR

(Makishima et al., 2002). Activated FXR heterodimerizes with the retinoid X receptor

(RXR; NR2B1) to initiate in the transcription of genes (Goodwin et al., 2000). FXR activation directly induces BSEP and MRP2 in the liver and OSTα-OSTβ in the ileum to increase bile acid secretion (Zöllner et al., 2006). FXR indirectly reduces bile acid synthesis by downregulating the cholesterol metabolizing enzyme, CYP7A1, through the induction of another transcription factor, the short heterodimer partner (SHP; NR0B2), which inhibits

CYP7A1 transcription (Goodwin et al., 2000; Lu et al., 2000; Lee and Moore, 2002;

Zöllner et al., 2006). In mice and humans, intestinal ASBT is subject to negative feedback regulation by FXR via SHP-dependent repression of the liver receptor homolog-1 (LRH-1;

NR5A2), a transcription factor, that regulates ASBT transcription (Chen et al., 2003). In addition, bile acids can activate FXR to induce a hormonal signaling molecule, FGF15/19

(fibroblast growth factor 15 in rodents; FGF19 in humans) in the ileum to repress hepatic

CYP7A1 after entry into the liver via the membrane receptor FGFR4 (fibroblast growth factor receptor 4), which, when activated, decreases CYP7A1 through the c- Jun kinase signaling pathway in the liver (Wang et al., 2002; Inagaki et al., 2005).

1.2.3.2 The short heterodimer partner (SHP)

SHP is a unique transcription factor because it lacks an identified endogenous ligand and belongs to the orphan member of the nuclear receptor superfamily (Seol et al.,

1996; Zhang et al., 2011). Moore and colleagues were one of the first to discover that SHP functions as an atypical nuclear receptor because it lacks a DNA binding domain and consists of a putative ligand binding domain (Seol et al., 1996). SHP acts as a

transcriptional repressor or coregulator of various nuclear receptors by interacting with other transcription factors bound to DNA. SHP is expressed in the adrenal gland, stomach, intestine, gall bladder, liver, kidney, ovary, and heart (Bookout et al., 2006), and has been found to regulate mainly transporters and enzymes through protein-protein interactions with other nuclear receptors and transcription factors to inhibit transcription (Zhang et al.,

2011). SHP is repressive on liver receptor homolog 1 (LRH-1; NR5A2) and hepatocyte nuclear factor 4 (HNF-4; NR2A1) (Lee et al., 1998; Lee and Moore, 2002). SHP also interacts with the liver X receptor  (LXR; NR1H3), a stimulatory regulator of CYP7A1 that is activated by oxysterols (Brendel et al., 2002; Gupta et al., 2002; Schoonjans and

Auwerx, 2002) in the modulation of cholesterol absorption, transport, and elimination in rodents but not man (Peet et al., 1998; Chiang et al., 2001; Goodwin et al., 2003). Most studies focus on the role of SHP in the liver where it plays an important role in bile acids, fatty acid and triglyceride biosynthesis, cholesterol transport, and drug and hormone metabolism. SHP, as mentioned earlier, is a negative regulator of CYP7A1 in cholesterol metabolism (Goodwin et al., 2000; Lu et al., 2000).

1.2.3.3 The xenosenors: pregnane X receptor (PXR) and constitutive androstane receptor (CAR)

The pregnane X receptor (PXR; NR1I2) and the constitutive androstane receptor

(CAR; NR1I3) are members of the NR1 superfamily. These nuclear receptors heterodimerize with RXR, which, when activated by a ligand, initiates gene transcription.

These nuclear receptors are commonly known as xenobiotic sensors due to their roles in the regulation of transporters such as MRP and MDR1 and enzymes such as CYP, SULT, UGT in affecting drug disposition (Reschly and Krasowski, 2006). PXR and CAR serve not only

to protect the body from harmful xenobiotics but also provide alternative pathways to lower bile acid concentrations, mainly via enhanced metabolism (hydroxylation and conjugation), and protect the liver from bile acid induced toxicity (Staudinger et al., 2001; Xie et al.,

2001; Guo et al., 2003).

1.3 THE VITAMIN D RECEPTOR (VDR)

The vitamin D receptor (VDR) is also part of the NR1 superfamily (NR1I1). In rodents, the VDR is mainly localized in the intestine, kidney, skin, bone (Sandgren et al.,

1991), and low but detectable level in mouse and human livers (Gascon-Barré et al., 2003;

Song et al., 2009). 1,25-Dihydroxyvitamin D3 [1,25(OH)2D3] or calcitriol is the natural ligand of the VDR. The VDR becomes a ligand-activated transcription factor when it is bound to 1,25(OH)2D3, resulting in a conformational change of the receptor, which translocates into the nucleus and heterodimerizes with the RXR (Dusso et al., 2005). This complex then binds to the vitamin D response elements (VDREs) in the promoter region of

1,25(OH)2D3-responsive genes, recruiting nuclear proteins/coregulators into the transcriptional pre-invitation complex to initiate gene transcription. The VDR shares similar homology with PXR and CAR and can also be activated by lithocholic acid (LCA), a toxic bile acid, to induce CYP3A as detoxicification pathways of bile acids (Reschly and

Krasowski, 2006). Interests in the VDR as a regulator in drug disposition initially arose when 1,25(OH)2D3 was found to increase CYP3A4 and MDR1 in the intestinal Caco-2 cell line (Schmiedlin-Ren et al., 1997). Because enzymes and transporters are important in the absorption, distribution, metabolism and elimination of exogenous as well as endogenous compounds for biological, pharmacological, and toxicological events, there is the need for

more studies to determine whether VDR, similar to PXR and CAR, plays an important role in drug disposition.

UV 270-300nm

7-dehydrocholesterol CYP27A1 Liver Vitamin D (mitochondria) (Cholecalciferol)

Kidney

CYP1α or

1,25-dihydroxyvitamin D3 CYP27B1 Calcitriol 25-hydroxyvitamin D3 (mitochondria) [1,25(OH)2D3]

Figure 1-2 Bioactivation of vitamin D

1.3.1 VDR Protein Sequence Alignment in Human, Rat, and Mouse

Mouse, human and rat VDR protein alignments have been examined previously

(Kamei et al., 1995), showing that the DNA-binding domain is extremely highly conserved across species (100%). The mouse VDR ligand-binding domain is 89% identical to that in human and 96% identical to that in the rat (See APPENDIX F1). However, the mouse hinge region is different from that of man and rat, and is only 55% identical to that in human, and 78% identical to that in rat (Kamei et al., 1995).

1.3.2 VDR Ligands

1.3.2.1 1,25-Dihydroxyvitamin D3 or 1,25(OH)2D3

1,25(OH)2D3 is formed via the sequential metabolism by the liver and kidney prior to its binding and activation of the VDR (Fig. 1-2). There are two main forms of vitamin D; vitamin D2, which comes from plants and vitamin D3, which comes from animals (Jones et al., 1998). Vitamin D3 can be obtained from exogenous source such as milk and oily liver, or produced endogenously by the exposure of UV rays from sunlight, converting 7- dehydrocholesterol to vitamin D3 (Jones et al., 1998). Vitamin D3 is lipophilic and is stored mainly in adipose tissue rather than circulating in blood (Heaney et al., 2009). However, the first step in the bioactivation of vitamin D3 is by the hydroxylation of carbon 25, which occurs primarily in the liver by 25-hydroxylases, CYP27A1 and CYP2R1 (Cheng et al.,

2003) to form 25-hydroxyvitamin D3 [25(OH)D3] in the nM range. Over 99% of vitamin D metabolites are bind to plasma protein, mostly to the vitamin D binding protein (DBP) and also to albumin and lipoproteins to a lesser extent. In plasma, 25(OH)D3, which exhibits the highest binding affinity towards DBP and whose concentration is 20 times higher than other vitamin D metabolites, is highly bound to plasma DBP (Cooke and Haddad, 1989).

The low concentration of free vitamin D metabolite such as 25(OH)D3 in plasma contributes to reduced metabolism in tissues for elimination and a long circulating half life

(28 h) (Safadi et al., 1999). The 25(OH)D3-bound DBP complex is filtered through the glomerulus in the kidney and is recognized and taken up by endocytic receptor, megalin, present on the brush border of the renal proximal tubule cells (Safadi et al., 1999).

Once in tubular cells in the kidney, DBP is degraded by legumain (Yamane et al.,

2002), and the free 25(OH)D3 is metabolized by the mitochondrial 1-hydroxylase or

CYP27B1, to form the active 1,25(OH)2D3 (Fraser and Kodicek, 1970; Fu et al., 1997), though there is recent evidence that the enzyme is also present in the brain and skin (Eyles et al., 2005; Anderson et al., 2008). The plasma levels of vitamin D metabolites are controlled by further hydroxylation, mostly in the kidney, by the 24-hydroxylase (CYP24), at the C-24 position of both 25(OH)D3 and 1,25(OH)2D3 to produce 24,25-(OH)2D3 and

1,24,25(OH)2D3, respectively (DeLuca, 1988; Reddy and Tserng, 1989) or at other positions (Wang et al., 2011b). The expressions of CYP1 and CYP24 are tightly regulated in the kidney, which will be discussed later in the chapter.

1.3.2.2 1,25(OH)2D3 analogues

The four major vitamin D analogs (Fig. 1-3) have been used to treat hyperparathyroidism due to their lower hypercalcemic side effects are 22-oxa-1,25(OH)2D3 or oxacalcitriol, 19-nor-1,25(OH)2D2, 1(OH)D2 or doxercalciferol, and 1,25(OH)2-26,27-

F6-D3 or falecalcitriol (Brown and Slatopolsky, 2008). 1-Hydroxyvitamin D2 or 1(OH)D2, a prodrug of 1,25(OH)2D2, needs to be activated in the liver, and is less toxic than 1(OH)D3

(Sjoden et al., 1985). Similarly, 19-nor-1,25(OH)2D2 and 1(OH)D2 have been used to treat chronic kidney disease (Brown and Slatopolsky, 2008). Calcipotriol is used to treat psoriasis, having similar binding affinity for VDR compared to 1,25(OH)2D3 and 200 times less potent. In addition, 1,25(OH)2-16-ene-23-yne-D3 (EB1089) has been used to treat cancers such as leukemia, colon, breast and prostate cancer. MC1288 is used to suppress immune cells in transplantation.

Figure 1-3 Vitamin D analogs and LCA derivative structures from Brown and Sltopolsky (2008) and Ishizawa et al. (2008)

1.3.2.3 Alternative VDR ligands

Not all VDR analogs have a chemical structure similar to that of 1,25(OH)2D3.

Lithocholic acid (LCA), a secondary toxic bile acid, structurally dissimilar to 1,25(OH)2D3, is also a VDR ligand (Makishima et al., 2002). This bile acid is found to activate VDR at

µM concentrations rather than pM concentrations for 1,25(OH)2D3, and regulates VDR target genes such as Cyp24 and Cyp3a without triggering hypercalcemic effects

(Makishima et al., 2002; Nehring et al., 2007). There are increasing numbers of lithocholic acid derivatives synthesized as an alternative VDR ligand (Ishizawa et al., 2008). LCA acetate and LCA propionate (Fig. 1-3) are some of the recent developed VDR candidates to be used as VDR modulators (Ishizawa et al., 2008).

1.3.3 The VDR Ligand Binding Pocket

Activation of the VDR requires binding of the ligand to the binding pockets of the receptor. There is currently no crystal structure of the unoccupied ligand binding domain of the VDR. However, upon ligand binding with 1,25(OH)2D3, many crystal structures exist,

and subsequent analyses reveal that there are residues in the ligand binding pocket that interact with the ligand and trigger transcriptional activity (Vanhooke et al., 2004;

Yamagishi et al., 2006). In human VDR, 6 amino residues in the ligand binding pocket are found to interact strongly, due to hydrogen bonding, to 3 hydroxyl groups on the

1,25(OH)2D3 molecule (Yamagishi et al., 2006): Ser237 and Arg274 interact to the 1 hydroxyl group; Tyr143 and Ser278 interact with the 3 hydroxyl group; His305 and His397 interact with the 25-hydroxyl group. Arg274 is found to form the strongest hydrogen bond with 1,25(OH)2D3, and mutation of this residue is found in type II rickets (Yamagishi et al.,

2006). Similarly, the same type of amino acid residues are found in the ligand binding pocket of the rat VDR, but at different positions, are found to interact with hydroxyl groups of 1,25(OH)2D3 (Vanhooke et al., 2004): Ser233 and Arg270 interact to the 1 hydroxyl group; Tyr143 and Ser274 interact with the 3 hydroxyl group; and His301 and His393 interact with the 25-hydroxyl group. However, there is no information about mouse VDR ligand binding pocket; though it is speculated that the amino resides in the ligand binding pocket would be similar to the human and rat, due to the similarity in the amino acids binding sequence of the VDR protein (Kamei et al., 1995).

1.3.4 Physiological Role of the 1,25(OH)2D3-Liganded VDR

To date, vitamin D is not only known for its role for calcium and phosphate homeostasis, but also known to control cell proliferation and differentiation, as well as synthesis and secretion of cytokines and other hormones (Valdivielso et al., 2009). The presence of the VDR in many tissues suggests a definitive role of 1,25(OH)2D3 in the body

(Andress, 2006; Wang et al., 2008b). Although 1,25(OH)2D3 could be synthesized in many of these tissues, it is produced specifically for direct cell use, and not for mineral

requirements. The therapeutic use of 1,25(OH)2D3 for cancer, immune and endocrine modulation has been abandoned due to its hypercalcemic side effects. Thus, many vitamin

D analogs have been produced to separate those effects (Fig. 1-3) (Brown and Slatopolsky,

2008; Ishizawa et al., 2008).

The function of the vitamin D endocrine system is to ensure that calcium (Fig. 1-4) and phosphate are kept in balance for proper body functions. Thus, this system requires adequate communication between organs, such as the kidney, bone, parathyroid gland and intestine to maintain appropriate plasma levels of 1,25(OH)2D3. The physiological role of

1,25(OH)2D3 in plasma is to increase the absorption of calcium from the intestine, reabsorption in the kidney and bone. In enterocytes of the small intestine, 1,25(OH)2D3 and the VDR are required to induce epithelial calcium channels [transient receptor potential cation channel, subfamily V, member 6 or TRPV6 (den Dekker et al., 2003)] for calcium absorption from lumen, increase the transport of calcium across the cell by inducing calbindin D9K, a cytosolic calcium binding protein, and elevate basolateral plasma membrane ATPase (PMCA1) that transports calcium into the bloodstream (Dusso et al.,

2005). VDR activation also increases active phosphate transport through the induction of the apical Na-Pi cotransporter in the intestine (Yagci et al., 1992). In bone, 1,25(OH)2D3 activates osteoblasts and stimulates the maturation of osteoclasts to resorb calcium from bone and reverse transport calcium from bone compartment to plasma (Jones et al., 1998).

In kidney, 1,25(OH)2D3 upregulates TRPV5 and calbindin D28K to increase calcium reabsorption (Enomoto et al., 1992; Dusso et al., 2005).

PTH PTG Suppression Stimulation

Kidney

1 ,25(OH) D ↑Ca2+  2 3

Parafollicular cells

Figure 1-4 1,25(OH)2D3 and plasma calcium homeostasis

For control of plasma calcium, a negative feedback mechanism exists to reduce

1,25(OH)2D3 production and calcium reabsorption. The calcium-sensing receptor (CaSR) is present in the parathyroid glands for the detection of plasma calcium (Jones et al., 1998). A high concentration of 1,25(OH)2D3 and calcium in the plasma leads to inhibition of parathyroid hormone (PTH) synthesis and secretion (Silver et al., 1996; Dusso et al., 2005), and decreased plasma PTH leads to inhibition of 1-hydroxylase expression in the kidney to reduce 1,25(OH)2D3 synthesis (Henry and Norman, 1984). 1,25(OH)2D3 can also increase its own catabolism by upregulating the expression of its catabolic enzyme, CYP24, in the kidney (Chen and DeLuca, 1995). In addition, calcitonin (CT), a 32-amino acid polypeptide produced in the parafollicular cells of the thyroid gland is increased when plasma calcium is high, and inhibits calcium absorption from the intestine and osteoclast activity in bones (Jones et al., 1998). These events act as feedback controls and lead to decreases in the plasma levels of 1,25(OH)2D3 and calcium.

1.3.5 Disease Association with Vitamin D Deficiency

Approximately 30 to 50% of the population in the United States is vitamin D deficient (Lee et al., 2008; Wang et al., 2008b). Because 25(OH)D3 is the major metabolic form of vitamin D in the plasma and represents the summation of both vitamin D intake and vitamin D synthesized from sun exposure, it is used as an indicator to determine a patient’s vitamin D status (Sarkinen, 2011). In humans, a plasma level of less than 50 nM or 20 ng/ml of 25(OH)D3 is considered vitamin D deficient, whereas a level of more than

374 nM is considered toxic (Sarkinen, 2011). Bone diseases, such as rickets in children, and osteomalacia and osteoporosis in adults are diseases found to be associated to vitamin

D deficiency (Sarkinen, 2011). However, over the last decade, studies have found that numerous other diseases such as hyperparathyroidism, vascular stiffness, cardiovascular disease, hypertension, inflammation, diabetes, and cancer have been linked to vitamin D deficiency (Valdivielso et al., 2009).

1.3.5.1 Hyperparathyroidism

Increases in the parathyroid gland size and parathyroid hormone synthesis have been found to be associated with a decrease in circulating plasma calcium (Jones et al.,

1998). This is usually associated with patients with chronic kidney diseases, as low or inactive forms of 1,25(OH)2D3 are present in their plasma. Low concentrations of calcium in the plasma triggers the parathyroid gland to stimulate the production of parathyroid hormone, leading to hyperparathyroidism (Jones et al., 1998).

1.3.5.2 Cardiovascular disease and hypertension

Mice lacking the VDR show signs of hypertension and ventricular hypertrophy, which are markers of cardiovascular disease (Xiang et al., 2005). In vitro studies of isolated

cardiomyocytes from VDR knockout mice are found to be associated with accelerated contraction and relaxation rates (Tishkoff et al., 2008). However, 1,25(OH)2D3 treatment has been found to regulate contractility, growth, hypertrophy, collagen deposition, and differentiation of cardiomyocytes (O'Connell et al., 1995; Wu et al., 1996; O'Connell et al.,

1997; Rahman et al., 2007), suggesting an important role for the VDR in cardiac physiology. There is also an association between low 25(OH)D3 levels in plasma and higher risk of hypertension (Forman et al., 2007; Wang et al., 2008b), likely due to alterations of the renin-angiotensin system. Studies have shown that inhibition of

1,25(OH)2D3 synthesis elevates renin expression and plasma angiotensin II synthesis, leading to hypertension (Li et al., 2002; Yuan et al., 2007).

1.3.5.3 Inflammation and autoimmune disease

Vitamin D deficiency is associated with a number of autoimmune diseases, such as multiple sclerosis, Crohn disease, diabetes mellitus, systemic lupus erythematosus (SLE), asthma, Sjögren’s syndrome, systemic vasculitis and antiphospholipid syndrome (Zhang and Wu, 2011). Studies have found that vitamin D plays a role in the immune system by regulating functions of T cells, natural killer (NK) cells, B cells, and antigen presenting cells (APCs) by reducing the release of inflammatory factors, such as of interleukin (IL)-1,

IL-2, tumor necrosis factor alpha (TNF-α), and interferon (INF)-γ, and elevating IL-10 synthesis (Cutolo et al., 2007; Almerighi et al., 2009; Zhang and Wu, 2011).

1.3.5.4 Diabetes

There are two forms of diabetes - type 1 diabetes that is the result of autoimmune destruction of pancreatic β cells, which produced insulin, and type 2 diabetes in which there is resistance to insulin and insulin secretion from pancreatic β cells (Seshadri et al., 2011) -

both types have been associated with vitamin D deficiency. Studies found that children with serum 25(OH)D3 levels less than 15 ng/ml were more likely to have elevated blood glucose levels than those having levels greater than 26 ng/ml (Reis et al., 2009). One theory for this association might be that type 1 diabetes is associated with an imbalance of pro-/anti-inflammatory cytokines such as transforming growth factor beta 1 (TGF-β1), INF-

γ, IL-1α, IL-1β, IL-4, IL-6, IL-12, and tumor necrosis factor (TNF)-α. These factors have been found to be decreased with 1,25(OH)2D3 treatment (Zhang and Wu, 2011). In addition, the VDR is present in pancreatic β cells. Thus, 1,25(OH)2D3 may play a role in insulin secretion and insulin sensitivity in type 2 diabetes by either increasing the intracellular calcium concentration via non-selective voltage-dependent calcium channels in the β-cell to induce insulin secretion or by increasing the conversion of proinsulin to insulin (Seshadri et al., 2011).

1.3.5.5 Cancer

A comprehensive review of anti-tumor effects of 1,25(OH)2D3 is found in Deeb et al. (2007). Treatment with 1,25(OH)2D3 and vitamin D analogs has been found to disrupt the G0/G1 cell cycle, differentiation, induction of apoptosis, and inhibition of angiogenesis in tumors of prostate, ovary, pancreatic, breast and lung cancer (Colston et al., 1992;

Nakagawa et al., 2005; Zhang et al., 2005; Banach-Petrosky et al., 2006; Chiang and Chen,

2009). In addition, 1,25(OH)2D3 act synergistically with chemotherapeutic agents, including platinum analogues, taxanes, and DNA-intercalating agents (Deeb et al., 2007).

One theory is that 1,25(OH)2D3 increases apoptosis in tumor cells, which is increased with the genotoxic stimulus by chemotherapeutic drugs (Hershberger et al., 2002).

1.4 VDR ON TRANSPORTERS, ENZYMES, AND NUCLEAR RECEPTORS

1.4.1 VDR Regulates Phase I and Phase II Enzymes

Schmiedlin-Ren et al. (1997) studied the role of intestinal CYP3A4 in the oral bioavailability of midazolam, and observed that 1,25(OH)2D3 induced intestinal CYP3A4 expression and increased midazolam metabolism. It was not until several years later that this VDR mediated event was shown to be due to the presence of a VDRE in the human

CYP3A4 gene (Schmiedlin-Ren et al., 2001; Thummel et al., 2001). More studies (Table 1-

2) reported that 1,25(OH)2D3 treatment of human cell lines increased CYP3A4 in vitro

(Thompson et al., 2002; Pfrunder et al., 2003; Aiba et al., 2005; Wang et al., 2008a; Fan et al., 2009). Later studies revealed that the inductive role of the VDR is not limited to

CYP3A4. The expression of human CYP3A4 as well as CYP2B6 and CYP2C9 was shown to be induced with 1,25(OH)2D3 treatment in primary human hepatocytes (Reschly and

Krasowski, 2006). However, human CYP7A1, a cholesterol metabolizing enzyme, was reported to be downregulated with 1,25(OH)2D3 treatment in primary human hepatocytes and HepG2 cells, a human liver cancer cell line (Han and Chiang, 2009; Han et al., 2010).

In vivo studies found that rat Cyp3a9 (Zierold et al., 2006) and Cyp3a1/3a23 (Xu et al.,

2006), isoforms of the human CYP3A4, are induced in the intestine. Kutuzova and DeLuca

(2007) demonstrated that 1,25(OH)2D3 regulates genes for its detoxification in the rat intestine, such as inducing the expression of the 1,25(OH)2D3 catabolic enzyme, Cyp24, as well as Cyp3a1 and Cyp1a1. In rat and human intestinal slices, exposure of 1,25(OH)2D3 led to induced expression of Cyp3a1, Cyp3a2 and human CYP3A4 (Khan et al., 2009b).

Vitamin D analogs, such as 19-nor-1,25(OH)2D2, LCA, and 1(OH)D3, increased the expression of CYP3A and Cyp24 (Schmiedlin-Ren et al., 2001; Thummel et al., 2001;

Makishima et al., 2002; Zierold et al., 2006; Nehring et al., 2007; Nishida et al., 2009;

Khan et al., 2011). In addition, 1,25(OH)2D3 treatment brought about increase phase II enzymes such as sulfotransferases, human SULT2A1 (Echchgadda et al., 2004) and mouse

Sult2a2 (McCarthy et al., 2005), and the UDP glucuronosyltransferase, rat Ugt1a

(Kutuzova and DeLuca, 2007).

1.4.2 VDR Regulates Transporters

There are fewer studies on the regulation of transporters by the VDR. The classic targets of 1,25(OH)2D3 have been known to regulate channels and transporters involved in calcium, phosphate and sulfur homeostasis (Taketani et al., 1998; Dawson and Markovich,

2002; den Dekker et al., 2003). However, the role of VDR in xenobiotic transporters has been studied less. Although induction of P-gp by 1,25(OH)2D3 treatment was observed in various studies for a over a decade (Schmiedlin-Ren et al., 1997; Pfrunder et al., 2003; Aiba et al., 2005; Fan et al., 2009), it is not until recently that Saeki et al. (2008) reported that a

VDRE exists in the human MDR1 genome. Many more studies (Table 1-3) have since examined the role of VDR on transporters. In vivo, 1,25(OH)2D3 was shown to induce the expression of murine multidrug resistance-associated protein 3 (Mrp3) (McCarthy et al.,

2005). Previously, our laboratory noted that a VDRE existed in the rat apical sodium dependent bile acid transporter (Asbt) genome, and that induction of rat Asbt occurred with

1,25(OH)2D3 treatment to increase bile acid absorption in the ileum (Chen et al., 2006). Fan et al. (2009) found that 1,25(OH)2D3 treatment of Caco-2 cells induces human MRP2 and

MRP4, through a post transcriptional event. Induction of mouse Mrp2, Mrp3, and Mrp4 in vivo in rat and human intestinal slices was demonstrated to occur with vitamin D analogs,

LCA and 1(OH)D3 (Nishida et al., 2009; Ogura et al., 2009; Khan et al., 2011). One other

study further noted that the VDR played a role in intestinal folate transport in humans at rats (Eloranta et al., 2009).

1.4.3 VDR and Cross-Talk with Other Nuclear Receptors

The cross-talk between VDR and other nuclear receptors has recently been examined (Table 1-4). The VDR was found to inhibit the farnesoid X receptor (FXR)

(Honjo et al., 2006) and activation of the liver X receptor  (LXR) (Jiang et al., 2006), important regulators of bile acid and cholesterol homeostasis. Other studies further found that the VDR positively impacted the expression of peroxisome proliferator-activated receptor  and  (PPAR and PPAR) (Sertznig et al., 2009), and that PXR activation could lead to vitamin D deficiency (Holick, 2005). Recently, a VDRE was found in the fibroblast growth factor 15 (FGF15 or FGF19 in human), which is a negative regulator of

Cyp7a1 (Schmidt et al., 2010).

1.4.4 Significance of VDR in Transporters, Enzymes, and Nuclear Receptor

Interactions

The notion that the VDR may play a role in the regulation of transporters and enzymes is now viewed as an understatement. Alterations in enzymes, transporters and nuclear receptors by VDR ligands are listed in Table 1-2, 1-3, and 1-4, respectively. The

VDR regulates many enzymes such as CYPs, SULTs, and UGTs, and transporters such as

P-gp and MRPs, which are major players in the disposition of drugs. The role of the VDR is not only limited to transporter and enzyme regulations, but may also plays a role in bile acid and cholesterol homeostasis. Transporters and enzymes such as MRPs, CYP3A, and

CYP7A1 and nuclear receptors such as FXR and LXR, which are VDR targets, are important components in bile acid formation and transport and cholesterol metabolism.

Table 1-2 Regulation of enzymes by VDR activation VDR Ligands Target Genes Species References 1,25(OH)2D3 ↑CYP3A4 LS180 (in vitro) (Thummel et al., Caco-2 (in vitro) 2001) 1,25(OH)2D3 ↑CYP3A4 LS180 (in vitro) (Schmiedlin-Ren et HPAC (in vitro) al., 2001) Human Hepatocytes (in vitro) 1,25(OH)2D3 ↑CYP3A4 HT-29 colorectal cell (in vitro) (Thompson et al., 2002) 1,25(OH)2D3 ↑CYP7A1 HepG2 (in vitro) (Han and Chiang, Human Hepatocytes (in vitro) 2009) 1,25(OH)2D3 ↑CYP3A4 HepG2 (in vitro) (Wang et al., 2008a) Caco-2 (in vitro) Mouse Hepatocytes (in vitro) 1,25(OH)2D3 ↑SULT2A1 HepG2 (in vitro) (Echchgadda et al., (human, rat, mouse Caco-2 (in vitro) 2004) gene) 1,25(OH)2D3 ↑Cyp3a9 Rat (in vivo) (Zierold et al., 2006) 1,25(OH)2D3 ↑Cyp24a1 Rat (in vivo) (Kutuzova and ↑Cyp3a1 DeLuca, 2007) ↑Cyp1a1 ↑Ugt1a 1,25(OH)2D3 ↑Cyp3a1/23 Rat (in vivo) (Xu et al., 2006) 1,25(OH)2D3 ↑Cyp24a1 Mouse (in vivo) (Intestine and (Meyer et al., 2007) kidney) 1,25(OH)2D3 ↑CYP3A4 Human Hepatocytes (in vitro) (Drocourt et al., ↑CYP2B6 2002) ↑CYP2C9 1,25(OH)2D3 ↑CYP3A4 Caco-2 (in vitro) (Pfrunder et al., LS-180 (in vitro) 2003) 1,25(OH)2D3 ↑CYP3A4 Caco-2 (in vitro) (Aiba et al., 2005) LS-180 (in vitro) 1,25(OH)2D3 ↑Sult2a2 (colon) Mouse (in vitro) (McCarthy et al., 2005) 1,25(OH)2D3 ↑Cyp3a1 (rat Rat (in vitro) (Khan et al., 2009b) intestine) Human (in vitro) ↑Cyp3a2 (rat intestine) ↑CYP3A4 (human intestine) ↑CYP3A4 (human liver) 1,25(OH)2D3 ↓Cyp7a1 Human hepatocytes (in vitro) (Han et al., 2010) 1(OH)D3 ↑CYP3A4 Caco-2 (in vitro) (Schmiedlin-Ren et al., 2001) 1(OH)D3 ↑CYP3A4 Caco-2 (in vitro) (Thummel et al., 2001) 1(OH)D3 ↑Cyp3a11 Mouse (in vivo) (Makishima et al., 2002)

19-nor-1,25(OH)2D2 ↑Cyp3a9 Rat (in vivo) (Zierold et al., 2006) 1(OH)D3 ↑Cyp24 (intestine Mouse (in vivo) (Nishida et al., and kidney) 2009) ↑Cyp7a1 (liver) 1(OH)D3 ↑Cyp7a1 (liver) Mouse (in vivo) (Ogura et al., 2009) LCA ↑Cyp24 Rat (in vivo) (intestine and (Nehring et al., kidney) 2007) LCA acetate ↓CYP7A1 HepG2 (in vitro) (Han and Chiang, Human Hepatocytes (in vitro) 2009) LCA ↑Cyp3a11 Mouse (in vivo) (Makishima et al., (intestine and liver) 2002) LCA ↑Cyp3a1 (rat Rat (in vitro) (Khan et al., 2011) intestine) Human (in vitro) ↑Cyp3a2 (rat intestine) ↑Cyp3a9 (rat intestine and liver) ↓Cyp7a1 (rat liver) ↑CYP3A4 (human intestine)

Table 1-3 Regulation of transporters by VDR activation VDR Ligands Target Genes Species References 1,25(OH)2D3 ↑MDR1 LS180 (in vitro) (Pfrunder et al., 2003) 1,25(OH)2D3 ↑MDR1 LS180 (in vitro) (Aiba et al., 2005) 1,25(OH)2D3 ↑MDR1 Caco-2 (in vitro) (Saeki et al., 2008) 1,25(OH)2D3 ↑Mrp3 (colon) Mouse (in vivo) (McCarthy et al., MCA-38 (in vitro) 2005) 1,25(OH)2D3 ↑MDR1 LS180 (in vitro) (Thummel et al., 2001) 1,25(OH)2D3 ↑Asbt (intestine) Rat (in vivo) (Chen et al., 2006) 1,25(OH)2D3 ↓Ostα (rat Rat (in vitro) (Khan et al., 2009a) intestine) Human (in vitro) ↓Ostβ (rat intestine) ↓Ostα (human liver) 1,25(OH)2D3 ↑PepT1 (intestine) Rat (in vivo) (Maeng et al., 2011) ↑Mrp2 (Intestine) ↑Mrp4 (Intestine) 1,25(OH)2D3 ↑PCFT Caco-2 (in vitro) (Eloranta et al., 2009) 1(OH)D3 ↑Mrp2 (kidney) Mouse (in vivo) (Ogura et al., 2009) ↑Mrp4 (kidney) 1(OH)D3 ↑Asbt (intestine) Mouse (in vivo) (Nishida et al., ↑Mrp2 (kidney) 2009) ↑Mrp3 (kidney) ↑Mrp4 (intestine and kidney) LCA ↑Mrp3 (colon) Mouse (in vivo) (McCarthy et al., MCA-38 (in vitro) 2005) LCA ↑Mrp2 (rat Rat (in vitro) (Khan et al., 2011)

intestine) Human (in vitro) ↑Mrp2 (human intestine) ↑Mrp3 (human intestine)

Table 1-4 Regulation of nuclear receptor cross-talk by VDR activation VDR Ligands Target Genes Species References 1,25(OH)2D3 VDR→↓FXR HepG2 (in vitro) (Honjo et al., 2006) →↓SHP Caco-2 (in vitro) 1,25(OH)2D3 VDR→↓LXRα HepG2 (in vitro) (Jiang et al., 2006) →↓rCyp7a 1,25(OH)2D3 and LCA VDR→↓HNF- HepG2 (in vitro) (Han and Chiang, acetate 4α→↓CYP7A1 Human Hepatocytes (in vitro) 2009)

1.5 BILE ACIDS AND CHOLESTEROL HOMEOSTASIS

Cholesterol is an important component in the structure of cell membranes and is a precursor of steroids including corticosteroids, vitamin D, and bile acids. However, excess cholesterol in blood can lead to atherosclerosis and coronary heart disease as well as cerebrovascular disease. Methods to reduce blood cholesterol have become a popular research goal. Most medications currently used to treat hypercholesterolemia include the statins or HMG-CoA reductase inhibitors, niacin, and bile-acid sequestrants, which reduce the amount of cholesterol synthesis in the liver or reduce the amount of dietary cholesterol that is absorbed from the intestine (Gupta et al., 2010). Cholesterol homeostasis in mammals is maintained through the coordinate regulation of several major pathways in liver (Chiang, 2002): endocytosis of low-density lipoprotein (LDL) receptor uptake of serum cholesterol esters; reverse cholesterol transport from peripheral tissues to the liver and the uptake of high density lipoprotein (HDL) by the scavenger receptor subtype B1; absorption of dietary cholesterol from the intestine to the liver by LDL receptor-mediated mechanism of an endogenous biosynthetic pathway in which acetate is converted into cholesterol by HMG-CoA reductase.

1.5.1 Metabolic Pathways of Cholesterol Metabolism

In mammals, excess cholesterol is metabolically biotransformed to form bile acids, whereas a small portion of cholesterol is used to synthesize steroid hormones (Boggaram et al., 1984; Rezen et al., 2010). There are two major pathways in the formation of bile acids from cholesterol in the liver, the classic and alternative pathways (Chiang, 2004).

Cytochrome P450 7A1 or cholesterol 7-cholesterase (CYP7A1) catalyzes the first and rate-limiting enzymatic reaction in the classical pathway, whereas the alternate pathway is catalyzed by CYP27A1 and CYP7B1 (Russell and Setchell, 1992; Javitt, 1994). Cholic acid and chenodeoxycholic acid are the primary bile acids synthesized from cholesterol (Chiang,

2002). The bile acids are then conjugated with taurine or glycine before being excreted into bile by canalicular transporters such as BSEP and MRP2 (Kullak-Ublick et al., 2000). The function of bile acids in the small intestine is to act as detergent-like molecules to facilitate lipid uptake as mixed micelles (Hofmann, 1999). In humans, about 95% of bile acids are reclaimed through the ASBT in enterocytes of the ileum (Shneider, 2001; Chiang, 2002).

Once taken up into enterocytes, bile acids are bound to intracellular trafficking proteins, the ileal lipid binding protein or ileal bile acid-binding protein (ILBP or I-BABP) (Sacchettini et al., 1990), and are shuttled to the basolateral surface for efflux into the portal blood via

OST-OST (Dawson et al., 2005) and MRP3 (McCarthy et al., 2005). Hepatic uptake transporters on the sinusoidal membranes such as Na+-taurocholate cotransporting polypeptide (NTCP), which belongs to the same family of transporters as ASBT

(Hagenbuch and Meier, 1994), and organic anion-transporting polypeptides (OATPs)

(Zöllner et al., 2006) can mediate uptake of bile acids into hepatocytes and transport amphipathic endogenous and exogenous organic compounds (Hagenbuch and Meier, 2004),

completing the enterohepatic circulation of bile acids. MRP3 and MRP4, which are located at the basolateral membrane of hepatocytes, are capable of transporting bile acids back to blood, but are present at low expression (Zöllner et al., 2006) unless MRP2 is compromised or upon bile duct ligation (Akita et al., 2001; Akita et al., 2002; Ogura et al., 2009). Bile acids that escape hepatic uptake can be filtered by the glomerulus, excreted in urine by renal Mrp4 (Nishida et al., 2009), or be reabsorbed back to blood by ASBT and OST-

OST (Dawson et al., 2009), depending on the blood bile acid concentrations.

1.5.2 Regulation of Cholesterol Metabolism

Bile acids are detergent-like molecules and are toxic when they accumulate at high concentration (Hofmann, 1999). Thus, the body controls tightly bile acid synthesis and transport (Fig. 1-5). Numerous nuclear receptors, upon activation by their ligand, have been found to affect cholesterol metabolism by regulating the amount of CYP7A1. The promoter of the CYP7A1 gene contains a hexameric repeat of nucleotide sequence (AGGTCA), the bile acid response element (BARE) that is highly conserved among species (Chiang, 2003).

LRH-1 and HNF-4α are transcription factors that increase CYP7A1 transcription (Chiang and Stroup, 1994; Crestani et al., 1998; Goodwin et al., 2000). The murine liver X receptor

α (LXRα; NR1H3), upon transactivation with excess oxysterols, increases Cyp7a1 to increase cholesterol metabolism (Chiang et al., 2001). Excess bile acids evoke a negative regulatory pathway to reduce CYP7A1 transcription because bile acids are ligands of the

FXR that negatively regulates CYP7A1 (Goodwin et al., 2000; Lu et al., 2000). Upon FXR activation of SHP, inhibition of binding of LRH-1, a competency factor, and HNF-4α to the

CYP7A1 promoter region (Chiang, 2002) occurs. FXR can also reduce bile acid accumulation in liver directly by decreasing NTCP and inducing the bile acid efflux

transporters, BSEP and MRP2, which are present on the canalicular membrane of hepatocytes (Zöllner et al., 2006). In addition, HNF-4α stimulates rat CYP7A1 (Crestani et al., 1998) and CYP3A4 promoter activity (Tirona et al., 2003) whereas HNF-1α, which is highly dependent on HNF-4α, upregulates rat Ntcp and the organic anion transporting polypeptides, Oatp1a1 (Slc21a1) and Oatp1a4 (Slc21a10) (Trauner and Boyer, 2003), but these are downregulated by SHP (Fig. 1-5).

In the intestine, excess bile acids activate FXR that in turn induces the fibroblast growth factor (FGF19 in human and FGF15 in mouse), a hormonal signaling molecule, which travels in the portal blood and activates the liver surface fibroblast growth factor receptor 4 (FGFR4), and, acting through the c-Jun signaling pathway, decreases Cyp7a1

(Inagaki et al., 2005). PXR also negatively regulates Cyp7a1 (Staudinger et al., 2001). It is apparent that there are various nuclear receptors and transcription factors regulate CYP7A1 expression and cholesterol homeostasis.

Bile ↓Cyp7a1 ↑Bsep ↑Mrp2 Hepatocyte bile acid LRH-1 FGFR4 FGF15/19 HNF-4α

↓ SHP Oatp1a1 Nucleus HNF-1α

E ↓ bile acid FXR + RXR R e Oatp1a4 A n B e G

↓Ntcp

Figure 1-5 Regulation of Cyp7a1 by bile acids in the liver

1.5.3 Species Differences in Transporter and Enzyme Regulation

Species differences are found in the regulation of transporters and enzymes due to differences in transcriptional regulation of the conserved regions of the target genes. In the intestine, bile acids are reabsorbed into enterocytes through ASBT (Zöllner et al., 2006).

The increase in bile acids in enterocytes triggers the activation of FXR. FXR increases SHP,

Ostα and Ostβ, and FGF15 to increase bile acid efflux and decreases Cyp7a1 in the liver

(Goodwin et al., 2000; Inagaki et al., 2005; Rao et al., 2008). Due to the absence of a LRH-

1 cis-acting element binding in the rat Asbt promoter, the negative feedback for FXR and

SHP to downregulate rat Asbt to decrease bile acid absorption would be absent (Chen et al.,

2003), namely, the feedback regulation of Asbt by bile-acid-FXR-SHP cascade does not exist in the rat (Fig. 1-6). In addition, the stimulatory effect of LXR on Cyp7a1 only occurs in the rodent, and not the human (Chiang et al., 2001); HNF-4 and HNF-1 regulate rat Ntcp, but this does not occur in the human nor the mouse (Jung et al., 2004), whose transcriptional regulatory regions are different among species. FGF19 is found present in human hepatocytes and downregulates CYP7A1 (Song et al., 2009), but FGF15 is absent in the mouse liver (Inagaki et al., 2005). In addition, the expression of VDR is higher in the mouse and human compared to the rat (Gascon-Barré et al., 2003; Han and

Chiang, 2009), suggesting differences in the role of VDR in the liver among species. These species differences in the regulation of transporters and enzymes are important because a specific ligand to a nuclear receptor can eventually lead to different outcomes among species.

INTESTINE Rat APICAL BASOLATERAL

BA BA Asbt FXR BA LRH-1 SHP + BA BA BA - VDR no Lrh-1 in Asbt promoter; no feedback inhibition; ↑ bile acid

Mouse & Human VDR BA BA ASBT FXR BA + LRH-1 SHP BA BA -

feedback inhibition on Asbt Figure 1-6 Species differences in negative feedback regulation of rat, mouse, and human Asbt

1.5.4 The Link between VDR, Bile Acids and Cholesterol

There is certainly a link between the effects of nuclear receptors on bile acid and cholesterol homeostasis, as well as drug disposition. Nuclear receptors regulate drug transporters and drug metabolic enzymes, which are also involved in the handling of bile acid and cholesterol (Zöllner et al., 2006; Rezen, 2011; Tirona, 2011). Bile acids are substrates of ASBT, NTCP, BSEP, and OATPs (Zöllner et al., 2006), and some of these transporters can also transport xenobiotics such as pitavastatin, rifampin, enalapril, docetaxel (You and Morris, 2007). Thus, activation of FXR by bile acids could theoretically alter the disposition of these drugs. Changes in the level of cholesterol are known to affect P-gp activity (Tamashevskii et al., 2011), and P-gp is reported to play a role in the transport of cholesterol in the liver (Luker et al., 2001; Leon et al., 2006). The

VDR is a major regulator of the expression of CYP3A4 (Thummel et al., 2001), which, in

turn detoxifies xenobiotics and metabolizes bile acids to less toxic and more easily excreted derivatives, affecting the concentrations of bile acids in the body. Thus, the VDR is expected to play a role in the regulation of transporters and enzymes, as well as bile acid and cholesterol homeostasis.

There has been some controversy with regard to the role of the VDR in regulating human CYP7A1 and rodent Cyp7a1. The VDR has recently been found to potentially play a role in the negative regulation of Cyp7a1 in the rat. Our study showed that activated VDR induced Asbt, which increased bile acid transport in the ileum (Chen et al., 2006) and could potentially lead to indirect FXR effects in the liver where VDR is virtually absent, whereas higher VDR protein expression is shown to exist in both human and mouse livers (Gascon-

Barré et al., 2003), though the levels are still low compared to those in the intestine and kidney (Sandgren et al., 1991). Studies in human primary hepatocytes and HepG2 cells suggest that VDR inhibits CYP7A1 upon activation by 1,25(OH)2D3 (Han and Chiang,

2009; Han et al., 2010). However, in some of these studies, the proper time-matched control was absent and time-dependent changes on the stability of the gene or mRNA could affect the interpretation. Though, other in vitro studies showed human VDR inhibits FXR

(Honjo et al., 2006) and LXRα (Jiang et al., 2006), which would increase CYP7A1.

Recently, treatment in mice with 1α-hydroxyvitamin D3 [1α(OH)D3], a vitamin D prodrug, resulted in upregulation of Cyp7a1 mRNA (Nishida et al., 2009; Ogura et al., 2009), whereas inhibition of Cyp7a1 was observed after a high dose of 1,25(OH)2D3 to mice

(Schmidt et al., 2010). However, these two studies did not conclusively show whether the mRNA changes observed were due to VDR or FXR, or whether the changes could affect

Cyp7a1 protein and cholesterol levels. The time-course of determination of Cyp7a1

changes is another important factor when examining these data since different treatment periods or doses may result in opposite changes in Cyp7a1.

1.6 SIGNIFICANCE OF VDR IN REGULATION OF TRANSPORTERS AND ENZYMES

Nuclear receptors are viewed as important regulators of transporters and enzymes.

The VDR is one of the major players in the regulation of xenobiotic transporters and enzymes as a detoxification pathway. Studies have examined the involvement of VDR as a bile acid sensor and exert a role in cholesterol metabolism. There are various levels of VDR present in tissues. Changes in transporters and enzymes can be attributed to direct and indirect VDR effects or to FXR. Bile acids and cholesterol and their derivatives are ligands of many nuclear receptors. As a result, changes in the concentrations of bile acids and cholesterol result from VDR activation can also influence changes in other transporters and enzymes. Thus, VDR is an important regulator in transporters and enzymes as well as cholesterol metabolism.

CHAPTER 2

STATEMENT OF PURPOSE OF INVESTIGATION

2.1 STATEMENT OF PURPOSE OF INVESTIGATION

Nuclear receptors play a significant role in protection of the body by regulating the balance of endogenous molecules, such as cholesterol and bile acids and minimizing the potential toxic effects of foreign molecules, such as xenobiotics, by affecting the level of transporters and enzymes. Current findings have demonstrated that VDR is a potential candidate in the regulation of transporters and enzymes in vitro (Schmiedlin-Ren et al.,

1997; Echchgadda et al., 2004; Aiba et al., 2005; Fan et al., 2009) and in vivo (Xu et al.,

2006; Nishida et al., 2009; Ogura et al., 2009). VDR is known as a regulator of calcium and phosphate homeostasis (Jones et al., 1998), but has recently been shown to be a bile acid sensor, due to its ability to induce CYP3A and detoxify the toxic bile acids, LCA

(Makishima et al., 2002). In addition, VDR inhibits FXR and LXR activation, suggesting the involvement of VDR in cholesterol metabolism (Honjo et al., 2006; Jiang et al., 2006).

Thus, the role of VDR may be expanded from calcium and phosphate homeostasis, to that of a contributor of bile acid and cholesterol homeostasis.

To date, there exist only a few but inconclusive studies on the role of the VDR in the regulation of transporters and enzymes of the intestine and kidney and in cholesterol metabolism in the liver in vivo. My Ph.D. dissertation examines the role of the VDR, namely, the effects on transporters and enzymes that affect drug disposition and cholesterol metabolism. Previously, our laboratory had shown that transactivation of rat Asbt may potentially lead to an increase in intestinal bile acid absorption (Chen et al., 2006). Due to the low expression of VDR in rat liver and the unresponsive LRH-1 cis-acting element in the rat ileal Asbt promoter for negative feedback control of Asbt (Chen et al., 2003;

Gascon-Barré et al., 2003), an increase in bile acid absorption would induce secondary

FXR effects in liver because bile acids are ligands of FXR. The direct role of the VDR on rat intestine and tissue slices were examined, in collaboration with Dr. Geny Groothuis in

Groningen, The Netherlands, and the resulting published works are included in APPENDIX

A1, A2, and A3.

However, a direct effect of VDR cannot be ruled out, especially in species such as mouse and human that have a higher expression of liver VDR than the rat (Gascon-Barré et al., 2003). Because of the possibility of indirect FXR effects on hepatic transporters and enzymes after 1,25(OH)2D3 administration and the possible antagonism exerted by VDR on

FXR to increase the expression of CYP7A1 (Honjo et al., 2006), fxr(+/+) and fxr(-/-) mice with C57BL/6 background were used to isolate the VDR effects from those of FXR in vivo.

The correlation between Cyp7a1 vs. SHP or FGF15 was investigated. To examine the molecular mechanism of VDR in cholesterol metabolism and the potential cholesterol lowering properties, we chose to investigate the effects of VDR in wild-type, fxr(-/-) and shp(-/-) mice pretreated with a high (42%) fat and (0.2%) cholesterol diet/western diet.

Because P-gp or MDR1, a critical transporter in drug disposition, was found to be upregulated both by VDR and FXR (Martin et al., 2008; Fan et al., 2009), we also used the fxr(+/+) and fxr(-/-) mice to differentiate these effects, and interpreted the impact of pharmacokinetic changes with the use of a physiologically-based pharmacokinetic (PBPK) model. Lastly, to relate the changes observed in these studies to the activation of VDR, we examined and correlated the plasma and tissue levels of 1,25(OH)2D3 in mice to changes in gene expression.

2.2 HYPOTHESES

1. Due to the differences in the expression of VDR in the rat intestine, liver and

kidney, VDR regulates transporters and enzymes directly via the VDR in the

intestine and kidney, and indirectly via FXR in liver in vivo.

2. The change and regulation of P-gp/Mdr1 in mice are attributed only to the

direct actions of VDR, and this has direct implications on the in vivo

pharmacokinetics of P-gp substrates

3. VDR plays a direct role in the upregulation of Cyp7a1 via repression of SHP

on cholesterol metabolism in mice, leading to a cholesterol lowering effect.

2.3 THESIS OUTLINE

The major studies of my project were:

1) To study the effects of 1,25(OH)2D3 on rat intestine, liver, and kidney enzymes

and transporters: the low liver VDR expression in rat liver gives way to

secondary FXR effects in liver (Chapter 3, test hypothesis 1)

2) To determine changes in Mdr1/ P-gp in vivo greatly change the disposition of

digoxin after 1,25(OH)2D3 treatment in mice (Chapter 4, test hypothesis 2)

3) To examine changes in mouse liver Cyp7a1 and cholesterol lowering with

1,25(OH)2D3 treatment in mice (Chapter 5, test hypothesis 3)

4) To compare the concentrations of 1,25(OH)2D3 in plasma and tissue vs. changes

of VDR target genes in mice (Chapter 6, test hypotheses 1-3)

5) General Discussion and Conclusion (Chapter 7) 36

CHAPTER 3

3. DIRECT AND INDIRECT EFFECTS OF THE VITAMIN D RECEPTOR (VDR)

ON TRANSPORTERS AND ENZYMES IN THE RAT INTESTINE, LIVER AND

KIDNEY IN VIVO

Edwin C.Y. Chow1, Han-joo Maeng1, Huadong Sun1, Ansar A. Khan2, Geny M.M.

Groothuis2, and K. Sandy Pang1

1Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of

Toronto, Canada

2Pharmacokinetics, Toxicology and Targeting, Department of Pharmacy, University of

Groningen, The Netherlands

3.1 ABSTRACT

Bile acids are substrates of transporters and enzymes involved in bile acid homeostasis and are ligands of the farnesoid X receptor (FXR). Activation of the FXR triggers a negative feedback mechanism that regulates the expression of transporters and enzymes to decrease bile acid synthesis, increase bile acid secretion, and reduce bile acid uptake into cells. 1α,25-Dihydroxyvitamin D3 [1,25(OH)2D3], the natural ligand of the vitamin D receptor (VDR), was found to regulate many of the bile acid related transporters and enzymes in vitro. Previously, our laboratory found that VDR transactivated the rat ileal apical sodium-dependent bile acid transporter (Asbt) to increase bile acid absorption in vivo.

However, VDR effects on transporters and enzymes in the rat intestine and kidney, which are VDR-rich target organs, and in liver, whose expression is low, is ill-defined. Thus, protein and mRNA levels of various target genes in the rat small intestine, colon, liver, and kidney were measured by qPCR and Western blotting, respectively, after intraperitoneal dosing of 1,25(OH)2D3 (0 to 2.56 nmol/kg/day for 4 days) to rats. 1,25(OH)2D3 increased the protein expression of total cytochrome P450 3a (Cyp3a1), the multidrug resistance associated proteins (Mrp2, Mrp3, Mrp4) in proximal small intestine, and Asbt in ileum, as well as mRNA expression of the short heterodimer partner (SHP), fibroblast growth factor

15 (FGF15), and the organic solute transporters (Ost and Ost. 1,25(OH)2D3 treatment resulted in approximately 50% higher bile acid concentration (65.1 ± 14.9 vs. 41.9 ± 7.8

µM, P < .05) in portal blood and elevated hepatic FXR and SHP mRNA. Increased bile salt export pump (Bsep) and Ostα mRNA levels in liver tissue, and a >50% reduction in

Cyp7a1 protein and cholesterol metabolism in rat liver microsomes is likely a consequence of the bile acid-FXR-SHP cascade and activation of the c-Jun N-terminal kinase signaling 38

pathway by FGF15. Increased multidrug resistance protein 1 (Mdr1a/P-gp) expression was also observed in rat liver. In kidney, VDR, Cyp24, Asbt and Mdr1a mRNA and protein expression increased (2- to 20-fold) in 1,25(OH)2D3- treated rats, and a 28-fold increase of

Cyp3a9 mRNA, but not of total Cyp3a protein nor Cyp3a1 or Cyp3a2 mRNA was observed, suggesting that VDR plays a significant, renal-specific role in Cyp3a9 induction. We conclude that changes in hepatic transporters and enzymes are indirect, secondary effects of the liver FXR-SHP cascade because of increased intestinal absorption of bile acids, an event that leads to activation of FXR. In contrast, 1,25(OH)2D3 treatment resulted in direct

VDR effects in the intestine and kidney, VDR-rich organs, and changes in transporter and enzymes levels were tissue-specific.

3.2 INTRODUCTION

Nuclear receptors (NRs) play an important role in the regulation of enzymes and hepatobiliary transporters in bile acid homeostasis (Zöllner et al., 2006). The farnesoid X receptor (FXR; NR1H4) is the most important bile acid sensor and responds to high bile acid concentrations by inducing the short heterodimer partner (SHP; NR0B2), which in turn inhibits the liver receptor homolog-1 (LRH-1; NR5A2) (Goodwin et al., 2000). One of the major FXR effect is the downregulation of cholesterol 7α-hydroxylase, CYP7A1,

(Goodwin et al., 2000), the rate-limiting enzyme among a series of metabolic reactions in the formation of bile acids from cholesterol in liver (Chiang, 1998). In contrast, the LRH-1, liver X receptor- (LXR; NR1H3), and hepatocyte nuclear factor 4α (HNF-4α; NR2A1) in rat are known to increase CYP7A1 (Crestani et al., 1998b; Goodwin et al., 2000; Chiang et al., 2001). The activation of FXR in intestine increases fibroblast growth factor 15

(FGF15) or FGF19 in humans (Song et al., 2009), a signaling hormone that binds to fibroblast growth factor receptor 4 (FGFR4) on the liver membrane to decrease CYP7A1

(Inagaki et al., 2005; Song et al., 2009). FXR further counteracts the hepatic cytotoxicity of bile acids by decreasing the expression of sodium taurocholate cotransporting polypeptide (NTCP; SLC10A1) at the sinusoidal membrane to reduce bile acid uptake

(Denson et al., 2001), and by inducing the bile salt export pump (BSEP; ABCB11)

(Goodwin et al., 2002) and the multidrug resistance associated protein 2 (MRP2; ABCC2)

(Kast et al., 2002), the canalicular transporters, for the excretion of bile acids. Transcription factors are also affected by FXR: the hepatocyte nuclear factor 1-alpha (HNF-1α), which is highly regulated by HNF-4α, activates rat Ntcp and the organic anion transporting polypeptides, Oatp1a1 (Slco1a1) and Oatp1a4 (Slco1a4) (Trauner and Boyer, 2003), transporters that promote bile acid uptake into liver; inhibition of HNF-1α or HNF-4α would lead to decreased Ntcp and bile acid uptake into the rat liver.

In the intestine, bile acids are reabsorbed into enterocytes through the apical sodium dependent bile acid transporter (Asbt; SLC10A2) (Zöllner et al., 2006). The increase in bile acids in enterocytes could trigger the activation of FXR. FXR then increases SHP, Ostα and

Ostβ, and FGF15 to increase bile acid efflux, and downregulate Cyp7a1 in liver (Goodwin et al., 2000; Inagaki et al., 2005; Rao et al., 2008). Due to the absence of LRH-1 cis-acting element in the rat Asbt promoter, there is an absence of the negative feedback mechanism of FXR to downregulate Asbt (Chen et al., 2003). Therefore, FXR activation due to increased bile acid absorption led to an unabated, higher level of rat Asbt. FXR remains as an important nuclear receptor involved in the regulation of bile acid homeostasis, not only

on the reabsorption of bile acids in the intestine, but also on FXR effects in the liver, potentially on Ntcp, Bsep, as well as bile acid synthesis via Cyp7a1 in rat liver.

Vitamin D, the inert precursor of the active ligand, 1,25-dihydroxyvitamin D3

[1,25(OH)2D3], has been widely used as a nutraceutical in the prevention of cancer and prolongation of longevity (Holick, 2004; Thomas, 2006; Mullin and Dobs, 2007; Schwartz and Skinner, 2007). Much is known about the molecular actions of vitamin D to regulate calcium and phosphorus homeostasis, and its indirect feedback on parathyroid hormone

(Jones et al., 1998). The activation of vitamin D requires the consecutive metabolism by the liver and the kidney to form 25-hydroxyvitamin D3 and then 1-25-dihydroxyvitamin

D3, the ligand of the vitamin D receptor (VDR) (Prosser and Jones, 2004; Feldman et al.,

2005). The toxic bile acid, lithocholic acid, is also a VDR ligand that activates the VDR, albeit at µM rather than the nM concentrations required for 1,25(OH)2D3 (Makishima et al.,

2002; Nehring et al., 2007). VDR is present abundantly in the rat intestine and kidney

(Sandgren et al., 1991), but is expressed much less in liver, where VDR is found mostly in stellate cells, Kupffer cells, endothelial cells, and cholangiocytes and not hepatocytes

(Gascon-Barré et al., 2003). In contrast, in human or mouse livers, VDR is expressed in hepatocytes at low but measureable levels, and is also present in non-parenchymal cells

(Han and Chiang, 2009).

During the last few years, more in vitro and in vivo studies were performed to investigate the effects of 1,25(OH)2D3 on enzymes and transporters within first pass and elimination organs, namely the intestine, liver, and kidney. 1,25(OH)2D3 was shown to regulate calcium homeostasis (Abrams and O'Brien, 2004; Walters et al., 2007) and

involved in the regulation of transporters and enzymes (Schmiedlin-Ren et al., 2001;

McCarthy et al., 2005; Chen et al., 2006; Xu et al., 2006; Fan et al., 2009; Khan et al.,

2009a; Khan et al., 2009b). Activation of VDR by 1,25(OH)2D3 upregulates human cytochrome P450 (CYP3A4), hydroxysteroid sulfotransferase (SULT2A1), and drug transporters such as the multidrug resistance protein (MDR1 or P-gp) and the multidrug resistance associated proteins (MRP2 and MRP4) in Caco-2 cells (Schmiedlin-Ren et al.,

1997; Echchgadda et al., 2004; Aiba et al., 2005; Fan et al., 2009). A vitamin D response element has been identified in MDR1 (Saeki et al., 2008). In vivo, the VDR transactivates the murine Mrp3 (McCarthy et al., 2005) and the rat apical sodium dependent bile acid transporter (Asbt) (Chen et al., 2006). In addition, rat intestinal Cyp3a1 has been observed to be upregulated by 1,25(OH)2D3 treatment, both in vivo and in vitro (Xu et al., 2006;

Khan et al., 2009b). Interestingly, VDR activation is able to blunt LXRsignaling in

HepG2 cells (Jiang et al., 2006) and antagonize the activities of FXR (Honjo et al., 2006), suggesting that the cross-talk between these bile acid related nuclear receptors could lead to changes in transporter and enzyme levels in liver. LCA activation of VDR induces Cyp3a in murine liver and intestine, serving as a detoxification mechanism pathway of bile acids in colon (Makishima et al., 2002), and in intestinal and liver slices of rats and humans

(Khan et al., 2009b). However, these effects are difficult to be identified in vivo due to confounding effects and inter-organ interactions.

To date, there exists little or no systematic study that describes the effects of

1,25(OH)2D3-liganded VDR in vivo nor on the regulation of transporters and enzymes for bile acid homeostasis and drug disposition in vivo. Ogura et al. (2009) have recently reported on changes in hepatic and renal mRNA expression of transporters and enzymes in 42

mice that underwent sham or bile duct ligation and were treated with an extremely high dose (31 nmol/kg) of 1-hydroxyvitamin D3, an inert precursor of 1,25(OH)2D3. However, the physiology of the animal was compromised by bile duct ligation and protein expression and the functions of the transporters and enzymes were not evaluated. Moreover, 1- hydroxyvitamin D3 may not be an effective modulator of intestinal-related effects because the prodrug requires activation by 25-hydroxylase in liver to form the active compound,

1,25(OH)2D3 (Theodoropoulos et al., 2003).

The objective of this study was to examine the direct and indirect roles of VDR on transporters and enzymes in intestine, liver, and kidney of rats in vivo. The small intestine and kidney are two major target organs of VDR (Jones et al., 1998), and kidney plays a central role in the formation of 1,25(OH)2D3. In comparison, very low levels of VDR exist in rat hepatocytes (Gascon-Barré et al., 2003), and thus, little or no VDR-related change in genes of transporters and enzymes is expected to occur in rat livers. Previous in vitro studies with precision cut rat liver slices have failed to show observable induction of

Cyp3a1, Cyp3a2, or Cyp3a9 by 1,25(OH)2D3 (Khan et al., 2009b). However, the increased portal bile acid absorption due to elevated Asbt in the rat in vivo (Chen et al., 2003) could trigger indirect or secondary changes in hepatic transporters and enzymes due to the activation or inhibition of other nuclear receptors, specifically, FXR. In addition, bile acids regulate and utilize many of the transporters and enzymes involved in drug disposition such as Asbt, Ntcp, Oatps, Mrps and Cyp3a (Zöllner et al., 2006). Hence, my goal was to identify direct VDR inductive effects in vivo in rats with 1,25(OH)2D3 treatment on intestinal and renal transporters and enzymes and indirect VDR effects in liver associated with FXR by studying the changes in protein and mRNA levels of VDR and FXR target 43

genes in intestine, liver, and kidney. We tested the hypothesis that changes in the liver in vivo were secondary FXR-related effects and not VDR effects. We further examined the possible VDR effects on other NRs in the rat liver, because an antagonism of the VDR on chenodeoxycholate-activated FXR effect had been documented in FXR- and VDR- transfected HepG2 cells (Honjo et al., 2006).

3.3 METHODS

3.3.1 Materials

1,25(OH)2D3 powder was purchased from Sigma-Aldrich Canada (Mississauga, ON,

Canada). Antibodies to villin (C-19) and rat Cyp7a1 (N-17) were purchased from Santa

Cruz Biotechnology (Santa Cruz, CA); anti-Mrp2 (ALX-801-016-C250) was from Alexis

Biochemicals, San Diego, CA; anti-P-gp (C219) was from Abcam, Cambridge, MA; anti-

Gapdh (14C10) was from Cell Signaling Technology, Danvers, MA; anti-VDR (MA1-710) was from Affinity BioReagents, Golden, CO; and anti-Cyp3a2 antibodies (458223) that failed to distinguish from Cyp3a1 or Cyp3a9, were from BD Biosciences, Mississauga, ON.

Other antibodies were kind gifts from various investigators: anti-Oatp1a1, anti-Oatp1a4, and anti-Ntcp (Dr. Allan W. Wolkoff, Albert Einstein College of Medicine, the Bronx, NY), anti-Oatp1b2 (Dr. Richard B. Kim, University of Western Ontario, ON); anti-Asbt (Dr.

Paul A. Dawson, Wake Forest University, Salem, NC); anti-Mrp3 (Dr. Yuichi Sugiyama,

University of Tokyo, Japan); anti-Mrp4 (Dr. John D. Schuetz, St. Jude Children’s Research

Hospital, TN); and anti-Bsep (Dr. Bruno Stieger, University Hospital, Zurich, Switzerland).

All other reagents were purchased from Sigma-Aldrich Canada (Mississauga, ON, Canada) and Fisher Scientific (Mississauga, ON, Canada).

3.3.2 1,25(OH)2D3 and Vehicle (Corn Oil) Treatment in Rats In Vivo

1,25(OH)2D3 was dissolved in anhydrous ethanol and the concentration was quantified spectrophotometrically at 265 nm (UV-1700, Shimadzu Scientific Instruments,

MD); then the solution was diluted in filtered corn oil (Sigma-Aldrich, ON) for injection.

Male Sprague-Dawley rats (260-280 g), purchased from Charles River (St. Constant, QC), were given water and food ad libitum and maintained under a 12:12-h light and dark cycle in accordance to animal protocols approved by the University of Toronto (ON, Canada).

Rats (n = 4 in each group) were injected with 0, 0.64, 1.28, and 2.56 nmol/kg/day

1,25(OH)2D3 in 1 ml/kg corn oil intraperitoneally for 4 days. At 24 h following the last day of 1,25(OH)2D3 treatment, rats were anesthetized with ketamine and xylazine (90 mg/kg and 10 mg/kg, respectively) by intraperitoneal injection. An aliquot (0.5 ml) of portal and systemic blood was collected and centrifuged at 3000 rpm for 10 min to obtain serum.

3.3.3 Blood Analysis and Preparation of Tissues

Portal bile acid concentrations were determined using a Total Bile Acids Assay Kit

(BQ042A-EALD from BioQuant, San Diego, CA) following the manufacturer’s protocol.

Serum alanine aminotransferase was assayed with a Reagent Kit (BQ004A-CR from

BioQuant, San Diego, CA) following the manufacturer’s protocol.

After blood collection, the portal vein was cannulated and flushed with 50 ml of ice- cold physiological saline solution. The small intestine was removed and placed on ice, and cut into eight segments (Chen et al., 2006). Segment 1 (S1) is the duodenum, spanning from the pyloric ring to the ligament of Treitz; segment 2 (S2) is the proximal jejunum segment of equal length that is immediately distal to the ligament of Treitz. The remaining

small intestine was then divided into six segments of equal length (S3 to S8, with S8 representing the ileum proximal to the ileocecal junction) (Chen et al., 2006). Enterocytes were isolated by mucosal scraping of everted tissue with a tissue-scraper (Johnson et al.,

2006) (n = 4). The colon (taken 10-cm section after the ileocecal junction) was freed of fecal matters via flushing with 1 mM phenylmethylsulfonyl fluoride (PMSF) in 0.9% NaCl, and stored in the same solution. Enterocytes from the colon were obtained by mucosal scraping. Enterocytes from all segments were immediately snapped frozen in liquid nitrogen and stored at -80°C until analyses. The liver and kidney were removed, weighed, cut into small pieces, and snapped frozen in liquid nitrogen, and then stored in the -80°C freezer for future analyses.

3.3.4 Preparation of Subcellular Fractions from Enterocytes

Frozen mucosal scrapings (50-100 mg of tissue) from intestinal segments and the colon were immediately mixed with 1 ml of Trizma HCl (0.1 M, pH 7.4) buffer containing

1% protease inhibitor cocktail (Sigma-Aldrich, ON) and homogenized on ice for three 30 sec periods at 8,000 rpm using an Ultra-turrax T25 homogenizer (Janke & Kunkel, Staufen,

Germany), then sonicated for 10 sec. Samples were centrifuged at 1,000 g at 4°C for 10 min, and the pellet, or the crude nuclear protein fraction, was resuspended in nuclear buffer

[60 mM KCl, 15 mM NaCl, 5 mM MgCl2•6H2O, 0.1 mM EGTA, 300 mM sucrose, 0.5 mM DTT, 0.1 mM PMSF, 300 mM sucrose, 15 mM Trizma HCl pH 7.4] containing 1% protease inhibitor cocktail. The nuclear fraction was used for Western Blot analysis of

VDR. The supernatant was transferred to a new tube and spun again at 21,000 g at 4°C for

1 h to yield another supernatant (crude cytosolic faction) and pellet (crude membrane

fraction); the pellet was resuspended in the same homogenizing buffer and used for

Western blot analyses of the intestinal transporters.

3.3.5 Preparation of Subcellular Fractions of Liver Tissue

For preparation of the crude membrane fraction, ~0.5 g of liver or whole kidney tissue was homogenized in the crude membrane homogenizing buffer (250 mM sucrose, 10 mM HEPES, and 10 mM Trizma base, pH 7.4) containing 1% protease inhibitor cocktail as described above. The resultant homogenate was centrifuged at 3,000 g for 10 min at 4°C.

The supernatant was transferred to an ultracentrifuge tube and spun at 33,000 g for 60 min at 4°C. The resultant pellet yielded the crude membrane protein fraction, and was placed in resuspension buffer (50 mM mannitol, 20 mM HEPES, 20 mM Trizma base, pH 7.4) containing 1% protease inhibitor cocktail, and was used for Western blot analyses of hepatic transporters.

For preparation of the microsomal fraction, ~0.5 g of liver tissue was homogenized with the microsome homogenizing buffer [250 mM sucrose, 10 mM Trizma HCl, 1 mM

EDTA, pH 7.4] containing 1% protease inhibitor cocktail as described above. The homogenate was centrifuged at 9000 g for 10 min at 4°C. The supernatant was transferred to an ultracentrifuge tube and spun at 100,000 g for 60 min at 4°C. The resulting pellet containing microsomes was resuspended in the same homogenizing buffer and used for

Western blot analyses of cytochrome P450 enzymes.

Protein concentrations of the samples were assayed by the Lowry method (Lowry et al., 1951) using bovine serum albumin as the standard. Samples were then stored at -80°C until Western blot analyses.

3.3.6 Liver Microsomal Cyp7a1 Activity

Livers were homogenized with a Potter-Elvehjem homogenizer in the microsome homogenizing buffer described above in absence of protease inhibitors because we wished to test for the functional activity of Cyp7a1. The microsomal fraction, obtained by differential centrifugation as described above, was used to examine Cyp7a1 activity outlined by Hylemon et al. (1989). Microsomal protein (50 µl of 40 mg/ml solution) was suspended in microsomal reaction buffer [415 µl containing 100 mM potassium phosphate,

50 mM NaF, 1 mM EDTA, 0.015 % 3-[(3-cholamidopropyl)dimethylammonio]-1- propanesulfonate (CHAPS), 20 % glycerol, and 5 mM DTT] and preincubated for 5 min at

37°C. The reaction was initiated by the addition of a NADPH generating system (25 µl of solution A and 5 l of solution B, BD Biosciences) and 5 µl of 10 mM cholesterol in acetone, resulting in 2 mg of microsomal protein in 500 µl of incubation mixture. After 30 min of incubation, the reaction was terminated by the addition of 15 µl of ice-cold 20% sodium cholate and then 15 µl of the internal standard, 92 µg/ml 7-hydroxycholesterol

(Steraloids Inc, Newport, RI), was added. The 7-hydroxycholesterol formed in the reaction was converted to 7α-hydroxy-4-cholesten-3-one upon incubation with 44 µl 12.5

U/ml cholesterol oxidase (Sigma) in cholesterol oxidase buffer (10 mM potassium phosphate, 20% glycerol, and 1 mM DTT) for 10 min at 37°C. This reaction was terminated by the addition of 1 ml of ice-cold methanol. The product, 7α-hydroxy-4- cholesten-3-one, was extracted into 9 ml of hexane after mixing the contents vigorously for

15 min, followed by centrifugation at 4°C for 5 min at 5000 g. After evaporation of the hexane extract under N2 (Boc Canada, Ltd., Mississauga, ON), the residue was reconstituted in 100 µl mobile phase (acetonitrile: methanol 70:30 v/v), and 20 µl was 48

injected into a Shimadzu HPLC (LC 10AT pump, SPD-10A UV-Vis detector, SIL-10A autoinjector, and SCL-10A system controller). Separation was achieved with an Altex 10

m C-18 reverse phase column (4.6 mm × 250 mm) at flow rates varying from 0.7 to 2 ml/min, with detection carried out at 240 nm; data integration was performed by the software, Star Chrom Lite®. Standards (0.05 to 2.5 nmole) containing varying amounts of

7-hydroxycholesterol (Steraloids Inc, Newport, RI) were processed under identical conditions and were converted to 7α-hydroxy-4-cholesten-3-one for construction of the calibration curve. The plot of the peak area ratios of 7α-hydroxy-4-cholesten-3-one/internal standard against 7α-hydroxycholesterol concentration was linear. The reaction rate was normalized to the amount of microsomal protein and the reaction time.

3.3.7 Western Blotting

Protein samples (20 or 50 µg total protein) were mixed with the loading buffer containing 0.1 M of DTT and incubated at an optimized, denaturing condition (room temperature for 30 min, 37°C for 15 min or 95°C for 2 min). Loaded proteins (n = 3 or 4 for each treatment group) were separated by 7.5% or 10% SDS-polyacrylamide gels that were overlaid with a 4% polyacrylamide stacking gel at 100 V. After separation, proteins were transferred onto a nitrocellulose membrane (Amersham Biosciences, Piscataway, NJ).

The membrane was blocked with 5% (w/v) skim milk in Tris-buffered saline (pH 7.4) and

0.1% Tween 20 (TBS-T) (Sigma-Aldrich, ON) for 1 h at room temperature, and then washed with 0.1% TBS-T for 10 min. The membrane was incubated with primary antibody solution (1:1000 to 1:5000 in 2% skim milk in 0.1% TBS-T) overnight at 4°C. The next morning, the membrane was washed with 0.1% TBS-T three times for 10 min each before

incubation with the secondary antibody (1:2000; anti-rat for VDR; 1:2000; anti-rabbit for

Mrp3, Mrp4, Bsep, Oatp1a1, Oatp1a4, Oatp1b2, Ntcp, Asbt, GAPDH, and Cyp3a2; anti- goat for villin and Cyp7a1; anti-mouse for P-gp and Mrp2) with 2% skim milk in 0.1%

TBS-T for 2 h at room temperature, and again washed three times with 0.1% TBS-T for 10 min each. Bands were visualized using chemiluminescence reagents (Amersham

Biosciences, Piscataway, NJ) and quantified by scanning densitometry (NIH Image software; http://rsb.info.nih.gov/nih-image/). Band intensity of target protein was normalized against that of villin for intestinal samples or Gapdh for liver and kidney samples, to correct for loading errors.

3.3.8 Quantitative Real-Time Polymerase Chain Reaction (qPCR)

Total RNA was obtained from 50-100 mg of scraped enterocytes, liver and kidney tissue using the TRIzol extraction method (Sigma-Aldrich) according to the manufacturer’s protocol, with modifications. The extracted RNA pellet was air dried and then dissolved in

TE buffer (Applied Biosystems Canada, ON). Total RNA was quantified by UV spectrometry quantified at 260 nm. The purity was checked by ratios of the readings at 260

/280 nm and 260/230 nm (≥1.8). cDNA was immediately synthesized from RNA samples, using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems Canada,

ON) by following the instruction provided by the manufacturer. In brief, 1.5 µg of total

RNA was transcribed in a 20 µl reaction volume in the Applied Biosystems 2720 Thermal

Cycler. The program consisted of 10 min for annealing at 25°C, 120 min for reverse transcription at 37°C, and 5 min for inactivation at 85°C. Real-time quantitative polymerase chain reaction (PCR) was performed with two detection systems (SYBR Green or Taqman assay), depending on the availability of primer sets. Information on primer sequences is 50

summarized in Table 3-1. These primer sets were analyzed using BLASTn to ensure primer specificity for the gene of interest (http://www.ncbi.nlm.nih.gov/BLAST/). The PCR mixture (20 µl final volume) consisted of 75 ng cDNA, 1 µM of forward and reverse primers, and 1 Power SYBR Green PCR Master Mix (Applied Biosystems) was used to perform PCR analysis. Amplification and detection were performed using the ABI 7500.

The real-time PCR system was designated the following PCR reaction profile: 95°C for 10 min, and 40 cycles of 95°C for 15 sec and 60°C for 1 min, followed by the dissociation curve. All target mRNA data (n = 4 for each treatment group) were normalized to villin mRNA for intestinal samples and Gapdh mRNA for liver and kidney samples. Levels of villin or Gapdh in the intestine for each segment and liver, respectively, were not altered by

1,25(OH)2D3 treatment. Data were analyzed using the ABI Sequence Detection software version 1.4 to obtain critical threshold cycle (CT) value, the cycle number at which the fluorescence increased linearly. For intestinal samples, the CT value of villin was subtracted from the CT value of the target gene (∆CT = CT.Target – CT.villin). The ∆CT was then compared

- to the corresponding ∆CT of the vehicle control (∆∆CT) and expressed as fold expression 2

(∆∆CT). The same analytical method is applied for the analysis of liver and kidney samples.

3.3.9 Statistical Analysis

Results were expressed as mean ± standard deviation. Data comparing the difference between two groups were analyzed using the two-tailed Student's t test, ANOVA, and the Mann-Whitney U test for errors that were normally- or non-normally-distributed, respectively. For intestinal and colon mRNA and protein analyses, the vehicle-treated S1 sample (value set as unity for normalization) was used for normalization of other vehicle- and 1,25(OH)2D3-treated samples from other segments and colon. For liver and kidney 51

mRNA and protein analyses, the vehicle-treated sample was usually set as the control

(value set as unity), and was used for comparison with those of the treated samples. A P value of less than 0.05 was set as the level of significance.

Table 3-1 Rat primer sets for quantitative Real-Time PCR

Gene Bank Number Forward (5’→ 3’ Sequence) Reverse (5’→ 3’ Sequence) Gapdh XR_007996 CGCTGGTGCTGAGTATGTCG CTGTGGTCATGAGCCCTTCC Villin NM_001108224 GCTCTTTGAGTGCTCCAACC GGGGTGGGTCTTGAGGTATT Mrp2 NM_012833 CTGGTGTGGATTCCCTTGG CAAAACCAGGAGCCATGTGC Mrp3 NM_080581 ACACCGAGCCAGCCATATAC TCAGCTTCACATTGCCTGTC Mrp4 NM_133411 GCCCTTACCCAGCTGCTGA CAGAATCCAGAGAGCCTCTTTTACA Oatp1a1 NM_017111 CTACTGCCCTGTTCAAGGCC ATTGTATCTCTCAGGATTCCGAGG Oatp1a4 NM_131906 TGCGGAGATGAAGCTCACC TCCTCCGTCACTTTCGACCTT Oatp1b2 NM_031650 AGACGTTCCCATCACAACCAC GCCTCTGCAGCTTTCCTTGA Asbt NM_017222 TCAGTTTGGAATCATGCCTCTCA ACAGGAATAACAAGCGCAACCA Ost# NM_001107087 TGTCATCCTGACCGCCCT AAGCGATCTGCCCGCTG Ost XM_001076555 TATTCCATCCTGGTTCTGGCAGT CGTTGTCTTGTGGCTGCTTCTT Mdr1a AY582535 GGAGGCTTGCAACCAGCATTC CTGTTCTGCCGCTGGATTTC Bsep XM_579531 TGGAAAGGAATGGTGATGGG CAGAAGGCCAGTGCATAACAGA Ntcp NM_017047 CTCCTCTACATGATTTTCCAGCTTG CGTCGACGTTCGTTCCTTTTCTTG Cyp3a1 NM_013105 GGAAATTCGATGTGGAGTGC AGGTTTGCCTTTCTCTTGCC Cyp3a2 XM_573414 AGTAGTGACGATTCCAACATAT TCAGAGGTATCTGTGTTTCCT Cyp3a9 U60085 GGACGATTCTTGCTTACAGG ATGCTGGTGGGCTTGCCTTC Cyp7a1 X17595 CTGTCATACCACAAAGTCTTATGTCA ATGCTTCTGTGTCCAAATGCC Cyp24 NM_201635 GCATGGATGAGCTGTGCGA AATGGTGTCCCAAGCCAGC VDR NM_017058 ACAGTCTGAGGCCCAAGCTA TCCCTGAAGTCAGCGTAGGT LXR NM_031627 TCAGCATCTTCTCTGCAGACCGG TCATTAGCATCCGTGGGAACA FXR NM_021745 AGGCCATGTTCCTTCGTTCA TTCAGCTCCCCGACACTTTT SHP BC088117 CCTTGGCTAGCTGGGTACCA GTCCCAAGGAGTACGCATACCT LRH-1 NM_021742 GCTGCCCTGCTGGACTACAC TGTAGGGCACATCCCCATTC FGF15 AB078900 ACGGGCTGATTCGCTACTC TGTAGCCCAAACAGTCCATTTCCT HNF-1α X54423 CTCCTCGGTACTGCAAGAAACC TTGTCACCCCAGCTTAAGACTCT HNF-4α EF193392 CCAGCCTACACCACCCTGGAGTT TTCCTCACGCTCCTCCTGAA # Ost primer set includes probe 5’ FAM-CAGCCCTCCATTTTCTCCATCTTGGC-TAMRA 3’ for Taqman® Gene Expression Assay

3.4 RESULTS

3.4.1 Effect of 1,25(OH)2D3 Treatment on Portal Bile Acid and ALT Levels

Total portal bile acid concentrations for the highest 1,25(OH)2D3 dose (2.56 nmol/kg) were significantly higher (55%) compared to that of control (Table 3-2). However, there was no change in alanine aminotransferase (ALT) in serum, suggesting a lack of damage to the liver by the treatment of 1,25(OH)2D3 at the doses chosen.

Table 3-2 Changes in blood analysis with various intraperitoneal injections of 1,25- dihydroxyvitamin D3 treatment for 4 days to the rat in vivo

Daily Dose of 1,25(OH)2D3 (nmol/kg/day)

0 0.64 1.28 2.56 Portal Bile Acid (µg/ml) 41.9 ± 7.8 44.6 ± 10.1 49.9 ± 16.8 65.1 ± 14.9*

Alanine Aminotransferase, 11.7 ± 8.2 19.9 ± 2.3 12.0 ± 4.5 17.1 ± 7.1 ALT (IU/l) * P < .05, compared to vehicle treatment (0 nmol 1,25(OH)2D3/kg/day) using two-tailed Student’s t test

3.4.2 Effect of 1,25(OH)2D3 Treatment on Nuclear Receptors (NRs), Enzymes and Transporters in Intestinal Segments and Colon

The abundance of mRNA and protein of NRs, transporters, and enzymes in the small intestine was examined under control conditions and compared to those in the colon.

The effects of 1,25(OH)2D3 on transporter and enzyme expression were then determined in the segment(s) of greatest abundance.

3.4.2.1 Intestinal nuclear receptors, NRs

Intestinal NRs were found to be distributed differentially in the small intestine and colon. The mRNA distribution of VDR revealed a slight decreasing gradient, from the

duodenum (S1) to ileum (S8), and was higher in the colon (1.23-fold of duodenum) (Fig. 3-

1A). Less VDR mRNA was detected in kidney and liver than in the small intestine (S1)

(Fig. 3-1A). There was no difference in VDR protein levels in small intestine, colon, and kidney, however the VDR protein in liver was only 14% that of S1 (Fig. 3-1A). The

1,25(OH)2D3-treated rats exhibited little change in intestinal VDR mRNA and protein (Fig.

3-1B), although VDR protein increased 50% in the S2 portion of the small intestine for the

1.28 nmol/kg treated group (Fig. 3-1B).

Unlike VDR, FXR mRNA increased from duodenum to ileum, and was highest in colon (Fig. 3-2A). There was no induction of FXR mRNA except for a small increase in the colon for the 1.28 nmol/kg treatment group (Fig. 3-2A). Surprisingly, the distribution of

SHP in the small intestine is in sharp contrast to that of FXR (Fig. 3-2B). The expression of SHP mRNA was evenly distributed in the small intestine and was highest in the colon, as observed by others for the rat small intestine (Los et al., 2007). A 2- to 6-fold induction in

SHP mRNA was observed for the proximal jejunum and ileum relative to the control, as found previously (Chen et al., 2006). Similar to FXR, the expression of LRH-1 mRNA increased from the duodenum to ileum, and was highest in colon (Fig. 3-2C); LRH-1 mRNA was relatively unaltered by 1,25(OH)2D3 treatment in the small intestine, though a small (36%) increase occurred at the highest 1,25(OH)2D3 dose in S1 (Fig. 3-2C). Because the mRNA of FGF15 was highest in S8 in mice (Song et al., 2009), our appraisal of induction in tissue was based on that for the S8 segment, which showed at least a 3-fold increase in the 1,25(OH)2D3 treated rats over the control group (Fig. 3-3).

(A) 2.2 2.0 mRNA Protein 1.8 1.6 † 1.4 MW S1 S2 S7 S8 Colon Liver Kidney 1.2 1.0 VDR 52 kDa 0.8 † 0.6 Gapdh 37 kDa † 0.4 † Relative VDR Expression Relative 0.2 † 0.00 S1 S2 S7 S8 Colon Liver Kidney Tissue Distribution 6 Treatment (nmol / kg / day) (B) 2.2 Treatment (nmol / kg / day) † 2.0 0 0 5 0.64 1.8 0.64 † 1.28 1.28 2.56 2.56 1.6 4 1.4

1.2 3 1.0 0.8 † 2 0.6 † † * 0.4 1 0.2

Relative VDR mRNA Expression 0 0.0 Expression Relative VDR Protein 0 S1 S2 S7 S8 Colon S1 S2 S7 S8 Colon Intestinal Segments Intestinal Segments

Figure 3-1 Distribution and dose-dependent effects of 1,25(OH)2D3 on rat intestinal VDR mRNA and protein (n = 3 or 4 in each group). (A) mRNA and protein distributions of VDR in the small intestine [duodenum (S1), proximal jejunum (S2), distal jejunum (S7), and ileum (S8)] and colon, liver and kidney, normalized to Gapdh expression. (B) mRNA and protein changes of VDR, normalized to villin. The molecular weights of Gapdh, VDR, and villin bands were detected at 37, 52, and 95 kDa, respectively. “*” indicates P < 0.05 compared to vehicle control in the same segment whereas “†” indicates P < 0.05 compared to the level of vehicle-control56 S1 segment using the two-tailed Student’s t test.

(A)

35 Treatment (nmol / kg / day) 30 0 0.64 * 1.28 25 2.56 † 20

15 10 † 5 † †

Relative FXR mRNA Expression Relative 0 S1 S2 S7 S8 Colon Intestinal Segments (B)

5 7.0 Treatment (nmol / kg / day) Treatment (nmol / kg / day) 6.5 * 0 0 0.64 6.0 0.64 4 1.28 5.5 1.28 * 2.56 2.56 5.0 * * 4.5 3 † * * * 4.0 * 3.5 2 3.0 2.5 * 2.0 1 1.5 1.0 0.5

Relative SHP mRNA Expression 0 0.00 S1 S2 S7 S8 Colon mRNA Expression FGF15 Relative FGF15 Intestinal Segments (C) S8 Intestinal Segment Figure 3-3 Dose-dependent effects of 20 Treatment (nmol / kg / day) 1,25(OH)2D3 on intestinal FGF15 0 mRNA in the ileum (n = 3 or 4 in each 0.64 * 1.28 group). “ ” indicates P < 0.05 15 † 2.56 compared to vehicle control in the same segment using the two-tailed Student’s t test. 10

5 † † * 00 Relative LRH-1mRNA Expression S1 S2 S7 S8 Colon Intestinal Segments

Figure 3-2 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal (A) FXR (B) SHP and (C) LRH-1 mRNA (n = 3 or 4 in each group). “*” indicates P < 0.05 compared to vehicle control in the same segment whereas “†” indicates P < 0.05 compared to the level of vehicle-control S1 segment using the two-tailed Student’s57 t test.

3.4.2.2 Intestinal enzymes

Intestinal Cyp3a1 and Cyp3a9 mRNAs were preferentially localized in proximal segments of the small intestine, and was lowest in the colon (Figs. 3-4A and 3-4B), whereas Cyp3a2 mRNA was undetectable (data not shown). All doses of 1,25(OH)2D3 resulted in an induction (> 10-fold of control) of Cyp3a1 mRNA in the S1 and S2 segments

(Fig. 3-4A), as well as an increase of total Cyp3a protein (all Cyp3a1, Cyp3a2, Cyp3a9 isoforms) (Liu et al., 2006b). In contrast, intestinal Cyp3a9 mRNA was not altered in rats treated for 4 days after 1,25(OH)2D3 (Fig. 3-4B). Dose-dependent increases of total Cyp3a protein (> 2-fold) were clearly observed with 1,25(OH)2D3 treatment (Fig. 3-4C). The mRNA expression of Cyp24, a catabolic enzyme that inactivates 1,25(OH)2D3 (Healy et al.,

2003; Meyer et al., 2007), exhibited an increasing trend, from duodenum to ileum (3.6-fold of S1), and was highest in colon (41-fold of S1; Fig. 3-5). However, the mRNA expression of Cyp24 in intestine was low (CT value about 27 to 29 in the small intestine and ~24 in the colon). Cyp24 mRNA were induced with 1,25(OH)2D3 treatment along the length of the small intestine, but not in the colon at the highest dose. For the other doses, a trend of upregulation was seen, however, the results failed to reach statistical significance due to the high variation observed (1.6 to 90-fold compared to control; Fig. 3-5).

(A)

3000.0 # 1000 Treatment (nmol / kg / day) Treatment (nmol / kg / day) 500 0 0 0.64 0.64 1.28 1.28 2.56 1000.0 2.56 400 # # * * 50.0 * * 300 2.0

1.5 200 * 1.0 # † 100 # † 0.5 # † * 50 † * #* † † # 0.00 0 * * Relative Cyp24 mRNA Expression Relative Cyp3a1 mRNA Expression S1 S2 S7 S8 Colon S1 S2 S7 S8 Colon Intestinal Segments Intestinal Segments (B) Figure 3-5 Distribution and dose-dependent effects of 1,25(OH) D on intestinal Cyp24 mRNA (n = 3 1.6 2 3 Treatment (nmol / kg / day) or 4 in each group). “*” indicates P < 0.05

1.4 0 compared to vehicle control in the same 0.64 segment whereas “†” indicates P < 0.05 1.2 1.28 2.56 compared to the level of vehicle-control S1 1.0 segment using two-tailed Student’s t test. “#” indicates P < 0.05 compared to vehicle control 0.8 in the same segment using Mann-Whitney U 0.6 test. † † 0.4 †

0.2

0.00 Relative Cyp3a9 mRNA Expression S1 S2 S7 S8 Colon Intestinal Segments (C)

5.0 * Treatment (nmol / kg / day) 4.5 0 4.0 0.64 1.28 3.5 2.56 * 3.0 * Cyp3a S1 2.5 * Villin 2.0 Cyp3a 1.5 S2 Villin 1.0 0 0.64 1.28 2.56 0.5 0.00 Relative Cyp3a Protein Expression S1 S2 1,25(OH)2D3 (nmol/kg) Intestinal Segments

Figure 3-4 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal Cyp3a enzymes (n = 3 or 4 in each group). The distribution and changes in Cyp3a1 (A) and Cyp3a9 (B) mRNA and changes in Cyp3a protein in S1 and S2 segments (C) with 1,25(OH)2D3 treatments are shown. The Cyp3a protein band was detected at 56 kDa. Cyp3a protein for S1 and S2 contro59ls were viewed as unity. “*” indicates P < 0.05 compared to vehicle control in the same segment whereas “†” indicates P < 0.05 compared to the level of vehicle-control # S1 segment using the two-tailed Student’s t test. “ ” indicates P < 0.05 compared to vehicle control in the same segment using Mann-Whitney U test.

3.4.2.3 Intestinal apical absorptive transporter, Asbt

Asbt mRNA was most abundant in the ileum (S8) (Chen et al., 2006), and with only

1% of that mRNA in the colon (data not shown). These levels were unchanged after

1,25(OH)2D3 treatment (Fig. 3-6A). Asbt protein, present most abundantly in S8 (Chen et al., 2006), was increased significantly for the 2.56 nmol/kg treatment group (Fig. 3-6B).

2.0 (A) Treatment (nmol / kg / day) 1.8 0 0.64 1.6 1.28 2.56 1.4 1.2 1.0

0.8

0.6

0.4 0.2

Relative Asbt mRNA Expression mRNA Asbt Relative 0.00 S8 Intestinal Segment

(B) 2.0 Treatment (nmol / kg / day) * 1.8 0 0.64 MW 1.6 1.28 2.56 1.4 Asbt 48 1.2 1.0 Villin 95 0.8 0.6 0 0.64 1.28 2.56 0.4

0.2 1,25(OH)2D3 (nmol/kg)

Relative Asbt Protein Expression Protein Relative Asbt 0.00 S8 Intestinal Segment

Figure 3-6 Dose-dependent effects of 1,25(OH)2D3 on intestinal mRNA and protein of Asbt in the ileum (n = 3 or 4 in each group). Asbt mRNA (A) and protein (S8) with 1,25(OH)2D3 treatment are shown; the Asbt protein band was detected at 48 kDa in the ileum (S8). “*” indicates P < 0.05 compared to vehicle control in the same segment using the two-tailed Student’s t test.

3.4.2.4 Intestinal apical efflux transporter, Mdr1a (P-gp)

Levels of Mdr1a mRNA displayed an increasing trend, from the duodenum to the ileum, as observed by others (Liu et al., 2006b), and levels were highest in the colon: S1 =

S2 < S7 = S8 << colon (Fig. 3-7A). 1,25(OH)2D3 failed to alter the mRNA expression of intestinal Mdr1a, excepting a small decrease (34%) in the S7 segment at the highest

1,25(OH)2D3 dose (Fig. 3-7A). P-gp protein expression in the S8 segment (ileum) was the highest, but did not show any demonstrable trend of induction with 1,25(OH)2D3 treatment

(Fig. 3-7B).

(A) (B)

20 4.0 Treatment (nmol / kg / day) Treatment (nmol / kg / day) 0 3.5 † 0 0.64 0.64 1.28 15 3.0 1.28 2.56 2.56 2.5

10 2.0

1.5 † 5 † 1.0

* 0.5

0 Expression Protein P-gp Relative 0.00 Relative Mdr1a mRNA Expression mRNA Mdr1a Relative S1 S2 S7 S8 Colon S8 Intestinal Segments Intestinal Segment

Figure 3-7 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal Mdr1a mRNA and P-gp protein (n = 3 or 4 in each group). Mdr1a mRNA (A) and P-gp protein, detected at 170 kDa, for S8 are shown. “*” denotes P < 0.05 compared to vehicle control in the same segment, whereas “†” indicates P < 0.05 compared to the level of vehicle-control S1 segment using two-tailed Student’s t test.

3.4.2.5 Intestinal apical efflux transporter, Mrp2

Unlike Mdr1a, the distribution of Mrp2 mRNA and protein was found to decrease from the duodenum to ileum, as found earlier (Mottino et al., 2000), then to very low levels

in the colon; the mRNA of the transporter were highest in S1 and S2 (Figs. 3-8A, 3-8B, and

3-8C). 1,25(OH)2D3 did not alter the mRNA levels of Mrp2 among all of the intestinal segments (Fig. 3-8A) and colon, but the protein expressions of Mrp2 in the S1 and S2 segments were significantly induced by 1,25(OH)2D3 (Figs. 3-8B, and 3-8D).

3.4.2.6 Intestinal basolateral efflux transporter, Mrp3

The mRNA expression of Mrp3 exhibited an ascending distribution pattern, increasing from duodenum to ileum, then the colon (Fig. 3-9A). Mrp3 protein expression was highest at the duodenum, but this drastically dropped within the proximal jejunum, with levels gradually increasing towards the ileum (Figs. 3-9B and 3-9C). However, in the colon, protein levels of Mrp3 were even higher than in the jejunum. 1,25(OH)2D3 failed to perturb Mrp3 mRNA levels, but increased Mrp3 protein levels in both S1 and S2, especially with higher 1,25(OH)2D3 doses (Figs. 3-9B and 3-9D).

3.4.2.7 Intestinal basolateral efflux transporter, Mrp4

The mRNA and protein distributions of Mrp4 were similar to those of Mrp3: an ascending distribution pattern, increasing from the duodenum to ileum, as was observed for

Mrp4 mRNA (Fig. 3-10A). Significantly higher mRNA was observed in colon.

Interestingly, Mrp4 protein expression was highest in S1, but decreased precipitously in S2, followed by a gradual increasing trend towards S8 in the small intestine (Figs. 3-10B and 3-

10C). Mrp4 protein in the colon was quite high and was only 17% lower compared to that of the duodenum. Dose-dependent 1,25(OH)2D3 induction of Mrp4 was observed in the S2 segment (Figs. 3-10B and 3-10D).

(A) (B)

2.0 Treatment (nmol / kg / day) 8.0 0 * Treatment (nmol / kg / day) 0.64 7.0 1.28 0 2.56 0.64 1.5 6.0 1.28 2.56 5.0 * 1.0 4.0 * * 3.0 *

0.5 2.0 † † † 1.0 † † † †

Relative Mrp2 mRNA Expression mRNA Mrp2 Relative 0.00 0.00 Relative Mrp2 Protein Expression S1 S2 S7 S8 Colon S1 S2 S3 S4 S5 S6 S7 S8 Colon Intestinal Segments Intestinal Segments

S1 S2 MW MW Mrp2 180 Mrp2 180 Villin 95 Villin 95 S1 S2 S3 S4 S5 S6 S7 S8 Colon 0 0.64 1.28 2.56 0 0.64 1.28 2.56

1,25(OH) 2D3 (nmol/kg)

Figure 3-8 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal mRNA and protein of Mrp2 (n = 3 or 4 in each group). Mrp2 mRNA (A) and protein distribution (B,C) in the small intestine [duodenum (S1), jejunum (S2-S7), and ileum (S8)] and colon and (D) the inductive changes of Mrp2 protein in S1 and S2, detected at 180 kDa, with 1,25(OH)2D3 treatments are shown. “*” indicates P < 0.05 compared to vehicle control in the same segment, whereas “†” indicates P < 0.05 compared to the level of vehicle-control, S1 segment using two-tailed Student’s t test. 63

(A) (B)

20.0 5.0 Treatment (nmol / kg / day) Treatment (nmol / kg / day) 0 18.0 4.5 0 0.64 16.0 0.64 1.28 4.0 1.28 * 2.56 14.0 2.56 † 3.5 12.0 3.0 † 10.0 † 2.5 * 2.5 2.0 * 2.0 1.5 * 1.5 * † 1.0 † 1.0 † † 0.5 0.5 † † †

Relative Mrp3 mRNA Expression mRNA Mrp3 Relative 0.0 0.00

0 Relative Mrp3 Protein Expression S1 S2 S7 S8 Colon S1 S2 S3 S4 S5 S6 S7 S8 Colon Intestinal Segments Intestinal Segments (C) (D)

S1 S2 MW MW

Mrp3 170 Mrp3 170 Villin 95 Villin 95 S1 S2 S3 S4 S5 S6 S7 S8 Colon 0 0.64 1.28 2.56 0 0.64 1.28 2.56

1,25(OH) 2D3 (nmol/kg)

Figure 3-9 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal mRNA and protein of Mrp3 (n = 3 or 4 in each group). Mrp3 mRNA (A) and protein (B,C) distribution in the small intestine [duodenum (S1), jejunum (S2-S7), and ileum (S8)] and colon are shown. Inductive changes in Mrp3 protein in S1 and S2, detected at 170 kDa, with the 1,25(OH)2D3 treatments were observed (D). “*” indicates P < 0.05 compared to vehicle control in the same segment, whereas “†” indicates P < 0.05 compared to the level of vehicle-control S1 segment using two-tailed Student’s t test. 64

(A) (B)

Treatment (nmol / kg / day) 40 2.4 Treatment (nmol / kg / day) 0 2.2 35 0.64 0 1.28 2.0 0.64 2.56 1.28 30 1.8 2.56 † 1.6 * 25 1.4 20 1.2 1.0 † 15 0.8 * † 10 † 0.6 0.4 5 † † 0.2 † † † † †

Relative Mrp4 mRNA Expression mRNA Mrp4 Relative 0 0.0 Relative Mrp4 Protein Expression 0 S1 S2 S7 S8 Colon S1 S2 S3 S4 S5 S6 S7 S8 Colon Intestinal Segments Intestinal Segments

MW MW S1 S2 Mrp4 160 Mrp4 160 Villin 95 Villin 95 S1 S2 S3 S4 S5 S6 S7 S8 Colon 0 0.64 1.28 2.56 0 0.64 1.28 2.56

1,25(OH) 2D3 (nmol/kg)

Figure 3-10 Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal mRNA and protein of Mrp4 (n = 3 or 4 in each group). Mrp4 mRNA (A) and protein (B,C) distribution in the small intestine [duodenum (S1), jejunum (S2-S7), and ileum (S8)] and colon are shown. Changes in Mrp4 protein by 1,25(OH)2D3 treatments in S1 and S2 segments were observed (D). Mrp4 protein band was detected at 160 kDa. “*” indicates P < 0.05 compared to vehicle control in the same segment, whereas “†” indicates P < 0.05 compared to the level of vehicle-control S1 segment using two- tailed Student’s t test. 65

3.4.2.8 Intestinal basolateral efflux transporter, Ost-Ost

Ost and Ost mRNA, noted to be localized mostly in the ileum (Rao et al., 2008), was modestly induced by the lowest dose of 1,25(OH)2D3 (Fig. 3-11), but did not change at higher doses.

1.6 Treatment (nmol / kg / day) 0 # 0.64 1.4 1.28 # 2.56 1.2

1.0

0.8 0.6 0.4 0.2

Relative Ost mRNA Expression 0.00 Ost Ost. S8 Intestinal Segment

Figure 3-11 Dose-dependent effects of 1,25(OH)2D3 on intestinal Ostα and Ostβ mRNA in the ileum (n = 3 or 4 in each group). “#” indicates P < 0.05 compared to vehicle control in the same segment using the Mann-Whitney U test.

3.4.3 Effect of 1,25(OH)2D3 on Hepatic Nuclear Receptors, Enzymes, and

Transporters

3.4.3.1 Hepatic nuclear receptors, NRs

At the highest dose of 1,25(OH)2D3 (2.56 nmol/kg), FXR and VDR mRNA were increased 43% to 75%, respectively, over those of control rats (Table 3-3). SHP mRNA was also induced (2.5- to 5-fold) (Table 3-3). In addition, significant increases in LRH-1 and LXRα mRNA were observed at doses exceeding 1.28 nmol/kg; HNF-4α and HNF-1 mRNA were increased at the highest dose, though the increase (about 37%) was modest

(Table 3-3).

3.4.3.2 Hepatic cytrochrome P450, Cyps

In liver, only Cyp3a2 mRNA was significantly induced (66%) over that of the control at the highest dose, whereas Cyp3a1 mRNA remained unchanged; the small increase in mRNA of Cyp3a9 was insignificant (Table 3-3). A probable explanation is that

Cyp3a2 is present only in bile duct epithelial cells (Khan et al., 2009b) where VDR is present (Gascon-Barré et al., 2003), whereas Cyp3a1, Cyp3a2 and Cyp3a9 are found in hepatocytes (Khan et al., 2009b), where VDR is absent. In liver, the total Cyp3a protein was unchanged. Cyp24 mRNA was very low, and was not altered with 1,25(OH)2D3 treatment (data not shown).

The mRNA level of Cyp7a1 was unchanged (Table 3-3), but there was a significant reduction in Cyp7a1 protein (>50%) at all doses of 1,25(OH)2D3 treatment (Fig. 3-12A).

Moreover, Cyp7a1 activity in rat liver microsomes (4 mg/ml) was reduced from 0.81 ± 0.14 nmol/h/mg in controls to 0.32 ± 0.14 nmol/h/mg protein in 1,25(OH)2D3 (2.56 nmol/kg/day) treated rats (P = 0.002; n = 4 in each group). The 60% reduction in Cyp7a1 activity observed for the 1,25(OH)2D3-treated rats correlated well with the reduction of protein (Fig.

3-12A) and increased mRNA levels of FXR and SHP.

3.4.3.3 Hepatic transporters

Relatively little change was observed for the sinusoidal and cholangiocyte uptake and efflux transporters in the 1,25(OH)2D3 treated livers. The mRNA and protein expressions for the sinusoidal transporters, Ntcp, Oatp1a1, Oatp1a4, and Oatp1b2, and Ostβ, and the mRNA expression of the cholangiocyte uptake transporter, Asbt (Lazaridis et al.,

1997), were unaltered (Table 3-3 and Fig. 3-12B). The increase in Mrp3 and Mrp4 protein

was not significant in liver (Fig. 3-12B). Observable changes in the liver were confined to mRNA of Ost and Mrp3, and the canalicular transporters, Bsep and Mdr1, especially at the highest dose (Table 3-3). Moreover, induction of protein was observed for P-gp (Fig. 3-

12C, P < .05).

3.4.4 Effect of 1,25(OH)2D3 on Nuclear Receptors, Enzymes, Drug Transporters in the Kidney

3.4.4.1 Renal nuclear receptors, NRs

Renal VDR mRNA was significantly induced approximately 2-fold at all doses of

1,25(OH)2D3 (Table 3-4), whereas the nuclear VDR protein was significantly induced by

60% at the highest dose (Fig. 3-13), whereas LRH-1 mRNA remained relatively unchanged

(Table 3-4). In contrast, the mRNA expressions of other nuclear receptors, FXR, SHP, and the transcription factors, HNF-4α and HNF-1α, were reduced by 50 to 60% after treatment with doses of 1.28 and 2.56 nmol/kg of 1,25(OH)2D3.

3.4.4.2 Renal cytochrome P450, Cyps

The mRNA of Cyp24, an enzyme known to respond to 1,25(OH)2D3 induction, was

20-fold higher at all doses of 1,25(OH)2D3 (Table 3-4), and Cyp24 protein was increased significantly (4-fold) at the highest 1,25(OH)2D3 dose (Fig. 3-13). Interestingly, Cyp3a9 mRNA expression was significantly induced > 28-fold by 1,25(OH)2D3 at all doses (Table

3-4). Cyp3a2 mRNA was reduced (>95%; P < .05) at the highest dose whereas the decrease in Cyp3a1 mRNA was not significant (Table 3-4). Total Cyp3a protein tended to be increased, though the change was not significant (Fig. 3-13).

3.4.4.3 Renal transporters

Ostα mRNA was decreased by 60% at the highest dose whereas the decrease in Ostβ mRNA was insignificant (Table 3-4). The mRNA of the renal transporters, Asbt (4-fold) and P-gp (2-2.5 fold) were significantly increased (Table 3-4). Increased Asbt protein was observed (Fig. 3-13), and this correlated with the increase in mRNA levels with

1,25(OH)2D3 treatment, though being significant only at the 1,25(OH)2D3 dose of 0.64 nmol/kg. In contrast, P-gp mRNA induction was associated with a 2- to 4-fold increase in protein expression at the higher doses of 1,25(OH)2D3 (Fig. 3-13). However, downregulation of Mrp4 mRNA expression was observed (Table 3-4), whereas that of

Mrp2 and Mrp3 was unchanged with treatment. Changes in Mrp2, Mrp3 and Mrp4 protein in the kidney after 1,25(OH)2D3 treatment were minimal (Fig. 3-13).

Table 3-3 Changes in mRNA expression of rat hepatic nuclear receptors, enzymes, and transporters, expressed as fold expression compared to vehicle treatment

1,25(OH) D Dose (nmol/kg/day) 2 3 Gene 0 0.64 1.28 2.56 FXR 1.00 ± 0.13 1.36 ± 0.38 1.17 ± 0.23 1.43 ± 0.28* # Hepatic SHP 1.00 ± 0.55 2.66 ± 0.55* 2.65 ± 0.64* 4.93 ± 3.18 LXRα 1.00 ± 0.10 1.23 ± 0.24 1.23 ± 0.11* 1.71 ± 0.09* Nuclear HNF-1α 1.00 ± 0.06 0.96 ± 0.13 0.92 ± 0.14 1.37 ± 0.34# Receptors HNF-4α 1.00 ± 0.12 1.15 ± 0.17 1.15 ± 0.20 1.37 ± 0.22* LRH-1 1.00 ± 0.18 1.18 ± 0.23 1.41 ± 0.11* 1.63 ± 0.20* Hepatic Cyp7a1 1.00 ± 0.05 0.96 ± 0.41 1.36 ± 0.98 1.54 ± 0.66 Cytochrome Cyp3a1 1.00 ± 0.44 0.74 ± 0.16 0.90 ± 0.54 0.71 ± 0.11 P450s Cyp3a9 1.00 ± 0.85 1.00 ± 0.61 2.02 ± 1.62 2.15 ± 1.12 Ntcp 1.00 ± 0.10 1.09 ± 0.15 1.19 ± 0.26 1.29 ± 0.28 Hepatic Oatp1a1 1.00 ± 0.14 0.96 ± 0.07 0.96 ± 0.14 1.14 ± 0.07 Oatp1a4 1.00 ± 0.52 0.68 ± 0.15 0.72 ± 0.28 0.75 ± 0.37 Sinusoidal Oatp1b2 1.00 ± 0.20 0.90 ± 0.15 0.95 ± 0.13 1.10 ± 0.17 Transporters Mrp3 1.00 ± 0.22 1.11 ± 0.32 1.18 ± 0.32 2.14 ± 0.81* Mrp4 1.00 ± 0.09 0.98 ± 0.21 0.93 ± 0.47 1.27 ± 0.58 VDR 1.00 ± 0.11 1.34 ± 0.72 1.17 ± 0.29 1.75 ± 0.23* Cyp3a2 1.00 ± 0.21 0.95 ± 0.19 1.61 ± 0.65 1.66 ± 0.36* Cholangiocyte Asbt 1.00 ± 0.26 0.78 ± 0.24 0.89 ± 0.38 0.68 ± 0.29 Ostα 1.00 ± 0.11 1.53 ± 0.75 1.34 ± 0.33 2.24 ± 0.97* Ostβ 1.00 ± 0.34 0.65 ± 0.13 0.56 ± 0.10 0.79 ± 0.33 Hepatic Bsep 1.00 ± 0.07 1.04 ± 0.09 1.17 ± 0.23 1.36 ± 0.21* Canalicular Mrp2 1.00 ± 0.19 1.14 ± 0.21 1.05 ± 0.18 0.94 ± 0.26 Transporters Mdr1a 1.00 ± 0.27 1.98 ± 0.50* 1.78 ± 0.83 2.26 ± 0.84* * P < .05, compared to vehicle treatment (0 nmol 1,25(OH)2D3 /kg/day) using two-tailed Student’s t test # P < .05, compared to vehicle treatment (0 nmol 1,25(OH)2D3/kg/day) using Mann-Whitney U test

(A) Treatment (nmol / kg / day) 1.4 0 0.64 1.2 1.28 MW 2.56 1.0 Cyp7a1 50

0.8 Cyp3a 56 * * 0.6 Gapdh 37 * 0.4 0 0.64 1.28 2.56 0.2

Relative Protein Expression 0.00 1,25(OH)2D3 (nmol/kg) Cyp7a1 Cyp3a Hepatic Cytochrome P450 Enzymes MW (B) Uptake Efflux Ntcp 50 3.5 Treatment (nmol / kg / day) Oatp1a1 80 3.0 0 0.64 1.28 2.5 Oatp1a4 94 2.56

2.0 Oatp1b2 80

1.5 Mrp3 170

1.0 Mrp4 160 0.5 Gapdh 37 Relative Protein Expression Relative Protein 0.00 0 0.64 1.28 2.56 Ntcp Oatp Oatp. Oatp.. Mrp3 Mrp4 1a1 1a4 1b2 Hepatic Sinusoidal Transporters 1,25(OH)2D3 (nmol/kg)

(C) Efflux Uptake (Cholangiocyte) Treatment (nmol / kg / day) 3.5 0 * MW 0.64 3.0 1.28 2.56 * Bsep 160 2.5 P-gp 170 2.0 Mrp2 1.5 180 Asbt 1.0 48 37 0.5 Gapdh Relative Protein Expression 0 0.64 1.28 2.56 0.00 Bsep P-gp Mrp2 Asbt

Hepatic Canalicular Transporters 1,25(OH)2D3 (nmol/kg)

Figure 3-12 Dose-dependent effects of 1,25(OH)2D3 on protein changes in cytochrome P450 isozymes (A), and sinusoidal (B) and canalicular (C) transporters (n = 3 or 4 in each group) in rat liver. Gapdh, Cyp7a1, Ntcp, Oatp1a1, Oatp1a4, Oatp1b2, and Bsep bands were detected at 37, 50, 50, 80, 94, 80, and 160 kDa, their molecular weights, respectively. “*” indicates P < 0.05 compared to vehicle control using the two-tailed Student’s t test. 71

Table 3-4 Changes in mRNA expression of rat renal nuclear receptors, enzymes, and transporters, expressed as fold expression compared to vehicle treatment

1,25(OH) D Dose (nmol/kg/day) 2 3 Gene 0 0.64 1.28 2.56 VDR 1.00 ± 0.25 2.30 ± 0.26* 1.82 ± 0.12* 1.72 ± 0.66* FXR 1.00 ± 0.20 0.69 ± 0.15 0.58 ± 0.08* 0.50 ± 0.16* Renal Nuclear SHP 1.00 ± 0.35 0.49 ± 0.14 0.52 ± 0.14* 0.36 ± 0.14* Receptors HNF-1α 1.00 ± 0.25 0.56 ± 0.07* 0.47 ± 0.09* 0.37 ± 0.15* HNF-4α 1.00 ± 0.24 0.61 ± 0.10* 0.63 ± 0.08* 0.47 ± 0.13* LRH-1 1.00 ± 0.26 1.10 ± 0.21 1.09 ± 0.14 1.32 ± 0.22 Renal Cyp24 1.00 ± 0.35 21.6 ± 2.43* 20.5 ± 1.97* 19.4 ± 7.91* Cyp3a1 1.00 ± 1.13 0.96 ± 1.27 0.18 ± 0.06 0.29 ± 0.34 Cytochrome Cyp3a2 1.00 ± 1.23 0.65 ± 0.99 0.06 ± 0.09 0.02 ± 0.03* P450s Cyp3a9 1.00 ± 0.58 17.1 ± 24.12 21.0 ± 18.8 67.89 ± 53.5* Asbt 1.00 ± 0.24 3.71 ± 1.43* 3.87 ± 0.63* 5.24 ± 2.09* Ost 1.00 ± 0.46 0.75 ± 0.17 0.61 ± 0.14 0.40 ± 0.15* Renal Apical Ost 1.00 ± 0.46 0.54 ± 0.12 0.61 ± 0.05 0.81 ± 0.27 Transporters Mdr1a 1.00 ± 0.22 1.85 ± 0.38* 2.18 ± 0.35* 1.61 ± 0.24* Mrp2 1.00 ± 0.16 1.02 ± 0.12 1.26 ± 0.13 1.09 ± 0.19 Mrp3 1.00 ± 0.16 0.90 ± 0.29 1.03 ± 0.10 1.10 ± 0.45 Mrp4 1.00 ± 0.27 0.52 ± 0.09* 0.53 ± 0.05* 0.54 ± 0.16* * P < .05, compared to vehicle treatment (0 nmol 1,25(OH)2D3 /kg/day) using two-tailed Student’s t test

8 Treatment (nmol / kg / day) 7 0 * MW 0.64 6 1.28 2.56 Asbt 48 5 * Mrp2 180 4 * Mrp3 170 3 Mrp4 160 2 * P-gp 170 1

Gapdh 37 0

Relative in Kidney Protein Expression VDR Cyp24 Cyp3a P-gp Mrp2 Mrp3 Mrp4 0 0.64 1.28 2.56

MW 1,25(OH)2D3 (nmol/kg) MW Cyp24 50 VDR 52 Cyp3a 56 Gapdh 37 37 0 0.64 1.28 2.56 Gapdh

1,25(OH)2D3 (nmol/kg) 0 0.64 1.28 2.56

1,25(OH)2D3 (nmol/kg)

Figure 3-13 Dose-dependent effects of 1,25(OH)2D3 on changes in (B) protein of renal nuclear receptor, cytochrome P450 isozymes and transporters (n = 3 or 4 in each group). Gapdh, VDR, Cyp24, Cyp3a, Asbt, P-gp, Mrp2, Mrp3, and Mrp4 bands were detected at 37, 52, 50, 56, 48, 170, 180, 170, and 160 kDa, their molecular weights, respectively. “*” indicates P < 0.05 compared to vehicle control using two-tailed Student’s t test. 73

3.5 DISCUSSION

This study unequivocally demonstrated that, in addition to FXR, there is a direct and indirect role for 1,25(OH)2D3-liganded VDR on the regulation of transporters and enzymes associated with bile acid homeostasis in rats. The direct effects of 1,25(OH)2D3 activated VDR in rat intestine and kidney are tissue-specific and related to the presence of

VDR. VDR is much higher in the intestine than in the rat liver (Fig. 3-1A), wherein VDR is localized in cholangiocytes and stellate cells and virtually absent in hepatocytes (Gascon-

Barré et al., 2003). Humans, however, display a slightly higher liver VDR level (Gascon-

Barré et al., 2003). Thus, no change in VDR-target genes was expected for liver and larger changes are expected in small intestine and kidney (Table 3-2). Following the subcutaneous injection of radiolabeled 1,25(OH)2D3, the liver receives only a minor (one third or one quarter) exposure compared to the duodenum (Brown et al., 2004). In addition, VDR expression in the intestine is a >1000-fold that of the liver (Sandgren et al., 1991).

Although Gascon-Barré et al. (2003) reported that VDR in biliary cells is about one-third of that in the intestine, biliary cells represent only ~5% of the cells in a normal liver (Racanelli and Rehermann, 2006). Hence, the role of VDR in the liver may be negligible compared to the intestine due to the both low expression of VDR and exposure to 1,25(OH)2D3.

A correlation between VDR- and 1,25(OH)2D3-mediated induction of Cyp3a was clearly observed in the different organs (Figs. 3-1 and 3-4; Table 3-3 and 3-4).

Upregulation of intestinal Cyp3a1 but not Cyp3a9 mRNA was observed, and the same pattern was repeated in intestinal slices (Khan et al., 2009b). These results are in contrast with reports of increased Cyp3a9 mRNA in rats treated with 1,25(OH)2D3 in vivo (Zierold

et al., 2006). Total Cyp3a protein induction, speculated to be Cyp3a1 due to the induction of Cyp3a1 mRNA, and high levels of VDR were observed (Fig. 3-4), similar to the result reported in rat intestinal slices incubated with 100 nM of 1,25(OH)2D3 (Khan et al., 2009b).

In liver, induction of each of the Cyp3a isozymes was VDR- and site-specific. We observed induction of only Cyp3a2 mRNA expression (Table 3-3), reportedly present only in biliary epithelial cells, where VDR is localized (Gascon-Barré et al., 2003). However, Cyp3a1 and

Cyp3a9 mRNA remained unchanged in the liver after 1,25(OH)2D3 treatment, and this can be explained by the absence of VDR in hepatocytes. However, there may be two explanations why Cyp3a proteins are not induced in liver, but are increased in intestine.

First, Brown et al. (2004) showed that liver receives lower 1,25(OH)2D3 exposure than the intestine when radiolabeled 1,25(OH)2D3 was administrated subcutaneously to rats. Second,

VDR is present at very low levels in the liver (Gascon-Barré et al., 2003) and is absent in rat hepatocytes. Both of these reasons explain why Cyp3a protein was not increased in liver by 1,25(OH)2D3. Interestingly, Cyp3a9 mRNA in kidney was induced (Table 3-4) even though renal Cyp3a1 and Cyp3a2 mRNAs were not altered by 1,25(OH)2D3, results that are in contrast to observations on the induction of Cyp3a1 in intestine (Fig. 3-4) and Cyp3a2 in liver (Table 3-3) in rat in vivo and in rat intestinal slices (Khan et al., 2009b). These observations suggest that induction of Cyp3a9 isoform is renal-specific. At this juncture, we were unable to quantify the intracellular 1,25(OH)2D3 concentration in tissues because of the lack of a sensitive assay (Vieth et al., 1990). In sum, Cyp3a1 in small intestine,

Cyp3a2 in liver, and Cyp3a9 in kidney were induced by VDR in a tissue specific manner.

In rat intestine, the induction of Mrp proteins (Figs. 3-8, 3-9, and 3-10) in the proximal segments could be due to non-genomic effects of 1,25(OH)2D3, as it was shown 75

to be linked to an active calcium absorption mechanism triggered by the VDR in the

2+ duodenum and proximal jejunum (Marks et al., 2007). 1,25(OH)2D3 increases Ca uptake by the rat duodenum, an effect that may be linked to the cAMP-mediated activation of plasma membrane Ca2+ channels (Massheimer et al., 1994). The increased cAMP could result in elevated levels of Mrp2 (Roelofsen et al., 1998) and Mrp3 protein (Chandra et al.,

2005) in S1 and S2 segments, either by short term regulation or membrane vesicle trafficking. Furthermore, there may be a role for cAMP on Mrp4 protein expression after

1,25(OH)2D3 treatment, because Mrp4 has been shown to transport cAMP and cGMP (van

Aubel et al., 2002). Similar observations were made in human Caco-2 cells; incubation with

100 nM of 1,25(OH)2D3 for > 3 days increased MRP4 protein without a change in mRNA, likely due to the stabilization of MRP4 protein after 1,25(OH)2D3 treatment (Fan et al.,

2009).

In rat kidneys, there was induction of P-gp, which is localized in renal proximal tubules (Huls et al., 2007), and Asbt localized in the renal cortex (Anakk et al., 2003),

(Table 3-4; Fig. 3-13). Recently, Saeki et al. (2008) and Chen et al. (2006) showed the presence of a VDRE in the human MDR1 and Asbt genes, respectively. Thus, induction of rat renal P-gp and Asbt (Fig. 3-13) is most likely via VDR transactivation. Previously, we attributed that increased ileal Asbt with 1,25(OH)2D3 treatment (Chen et al., 2006).

Presently, we found that renal Asbt was also induced by VDR, and this could further increase bile acid reabsorption. This would further increase plasma bile acid concentrations leading to downstream FXR effects in the liver. Overall, renal VDR plays an important role in the regulation of renal transporters that are VDR target genes.

Changes observed in transporters and enzymes in the intestine and liver after

1,25(OH)2D3 treatment that are related to the indirect effects of VDR in the rat in vivo are summarized in Fig. 3-14. In the rat intestine, induction of Asbt (Fig. 3-6) by 1,25(OH)2D3- liganded VDR (Chen et al., 2006) led to an increase in portal bile acid concentrations

(Table 3-2). Due to the lack of a negative feedback mechanism of FXR-SHP-LRH-1 on

Asbt, because of the absence of a LRH-1 cis acting element in the rat Asbt promoter (Chen et al., 2003), increased Asbt protein is observed in response to 1,25(OH)2D3, together with increased Ost-Ost, triggered increase in bile acid absorption from enterocytes, leading to increased bile acid efflux into the portal blood (Table 3-2). In addition, increased bile acids in enterocytes activate FXR targeted genes. Inagaki et al. (2005) reported FXR activation in mice with GW4064 treatment (a FXR ligand) resulted in increased in the mRNA of its targets, SHP and FGF15, and decreased Cyp7a1 in the mouse liver in vivo without changes in FXR. Similarly, the present data showed that mRNA of the FXR-target genes, SHP,

Ost, Ost, and FGF15, were increased in the ileum.

Cyp3a HNF-4 Mrp3 Blood Hepatocyte LXR Mrp2 FXR Mrp4

Bsep Ntcp Induction Bile SHP

P-gp Oatp* Inhibition HNF-1 Cyp7a1 Unchanged LRH-1 FGFR4 FGF15 mRNA Cyp3a2 Ost Protein Asbt Ostβ mRNA + VDR Portal Vein protein

Cholangiocyte Inductive pathway

Inhibitory pathway Ostα Ostβ Enterocyte Cyp3a SHP FGF15 Enterocyte VDR FXR LRH-1

Asbt Lumen

Figure 3-14 A schematic diagram highlighting direct and indirect effects of 1,25(OH)2D3 on intestinal and hepatic nuclear receptors, drug transporters and enzymes. Cyp3a and Asbt protein were upregulated by VDR. As a result, absorption of bile acids into enterocytes was increased, which in turn, activated intestinal FXR and increased SHP, FGF15, Ostα and Ostβ. Intestinal FGF15 binds to FGFR4 in liver to inhibit liver Cyp7a1. In liver, FXR and SHP are activated by bile acids and resulted in upregulation of Bsep, Mrp3, and Ost mRNA, and downregulation of hepatic Cyp7a1 protein. “*” represents the Oatp isoforms, Oatp1a1, Oatp1a4, and Oatp1b2. 78

The increased bile acid present in the portal vein could elicit FXR-related changes.

Changes in liver after 1,25(OH)2D3 administration, other than Cyp3a2, were mostly FXR- related events. The elevated hepatic FXR-SHP (Table 3-3) and intestinal FGF15 (Fig. 3-3) levels would in turn downregulate Cyp7a1 in liver (Goodwin et al., 2000; Inagaki et al.,

2005; Song et al., 2009). Indeed, there was decreased Cyp7a1 protein (Fig. 3-12A) and lessened microsomal Cyp7a1 activity (60%) in the liver after 1,25(OH)2D3 administration; the lack of correlation with Cyp7a1 mRNA was, however, unexplained. Other factors might be involved in stabilizing the mRNA expression, and more studies are needed to clarify this mechanism. In other reports, a reduction (73%) in Cyp7a1 activity in rats has been attributed to the bile acid, deoxycholic acid (Hylemon et al., 1989). In addition, Inagaki et al. (2005) reported that FGF15 could downregulate Cyp7a1 in the liver through a mechanism involving FGFR4 and the c-Jun N-terminal kinase (JNK)-dependent pathway

(Holt et al., 2003). In this study, we suggest that the synergy of increased FXR and SHP in the liver (Table 3-3), due to elevated bile acids (Table 3-2) and the increase in FGF15 (Fig.

3-3) in the intestine, led to a decrease in Cyp7a1 protein (Fig. 3-12A) and reduction in

Cyp7A1 activity.

Furthermore, induced mRNA of Bsep and Ost, other FXR-target genes, after

1,25(OH)2D3 treatment were also observed (Table 3-3). Other effects of 1,25(OH)2D3 on liver transporters were confined to Mrp3 and Mdr1a mRNA and P-gp protein (Table 3-3 and Fig. 3-13C). Elevated Mrp3 expression was also observed in rat liver slices incubated with 1,25(OH)2D3 (unpublished data), and increased Mrp3 protein has been observed in the mutant EHBR rats, which increases the basolateral efflux transport of taurocholic acid in response to loss of Mrp2 function (Akita et al., 2001; Akita et al., 2002), suggesting 79

possible roles of 1,25(OH)2D3 and liver bile acids on the induction of Mrp3. Zöllner et al.

(2003) have also reported that hepatic Mrp3 levels in fxr(+/+) and fxr(-/-) mice are elevated with cholic acid and ursodeoxycholic acid treatment, suggesting a FXR-independent mechanism for Mrp3 induction. Martin et al. (2008) showed that chenodeoxycholic acid treatment induced MDR1 mRNA and protein expression in HepG2 cells. More studies are needed to examine whether the changes of these transporters are related to VDR or FXR activation.

In summary, we showed that 1,25(OH)2D3 is capable of directly and indirectly altering intestinal, hepatic, and renal transporters and enzymes in the rat in vivo. Tissue- and enzyme-specific induction by VDR occurred due to the differential abundances of

VDR and the dose of 1,25(OH)2D3 administered. Direct VDR changes in the transporters and enzymes were found in the rat intestine and kidney where higher levels of VDR exist.

FXR, SHP, Ost Bsep and Mrp3 mRNA levels were increased in the rat liver, suggesting that these secondary FXR effects are elicited as a result of increased bile acid absorption by

Asbt. Repression of hepatic Cyp7a1 protein and activity was observed, likely the result of secondary FXR effects in the liver as well as increased FGF15 in the intestine. Changes in hepatic Mrp3 and Mdr1a mRNA and P-gp protein by 1,25(OH)2D3 were, however, unexplained, but may be FXR-related. The changes found in the intestine, liver, and kidney associated with 1,25(OH)2D3 treatment could affect drug absorption, oral bioavailability, metabolism, and elimination. The consequences of the activation of VDR by 1,25(OH)2D3 or vitamin D analogues on first-pass elimination are paramount, especially with the administration of the vitamin D analogues, an emergent class of therapeutics for the treatment of cancer and other diseases (Trump et al., 2004; Brown and Slatopolsky, 2008). 80

This new information further opens up more queries. More experiments are needed to examine possible regulatory mechanisms as well as examine changes in drug disposition.

3.6 ACKNOWLEDGMENTS

We thank Dr. Jianghong Fan of our laboratory for assistance in tissue collection and

Dr. Carolyn Cummins of the University of Toronto for invaluable discussions. Dr. Han-Joo

Maeng was supported by the Government of Canada Post-doctoral Research Fellowship

(PDRF). This work was supported by the Canadian Institutes for Health Research,

MOP89850.

3.7 STATEMENT OF SIGNIFICANCE OF CHAPTER 3

This chapter examined the role of the VDR on the regulation of transporters and enzymes in the in vivo rat. The data showed that the VDR most likely played a direct role in changes in transporters and enzymes in rat intestine and kidney. Activation of VDR in the ileum triggered the induction of ileal Asbt, which led to an increase in bile acid absorption.

High concentration of bile acids in the enterocytes and liver activated intestinal and hepatic

FXR, which downregulated Cyp7a1, the cholesterol metabolizing enzyme. Due to changes in transporters and enzymes in intestine, liver, and kidney, more studies are needed to be performed to differentiate VDR and FXR effects. Some of these were addressed in experiments conducted in conjunction with Dr. G.M.M. Groothuis in The Netherlands (see

APPENDICES A1, A2, and A3). Other transporters such as PepT1, Oat1, and Oat3 have also been examined in the rat intestine and kidney (APPENDIX A4). Moreover the functional studies of increased intestinal transporters: PepT1, Mrp2, Mrp4, but not P-gp were confirmed in published, intestinal everted sac studies (Maeng et al., 2011) that appear 81

in APPENDIX A5. Moreover, doxercalciferol (HECTOROL®) and higher 1,25(OH)2D3 doses were examined in the rat, at a more protracted regimen (4 doses given every other day for 8 days) to rats to compare these effects. The results are summarized in a published paper (Chow et al., 2011b) (See APPENDIX A6).

In the study, I was involved in the treatment of 1,25(OH)2D3 to rats and tissue harvesting, performed mRNA and protein extractions, analyzed mRNA and protein expressions in the intestine, liver, and kidney as well as bile acid and ALT measurement.

Hanjoo Maeng contributed to microsomal preparation to examine rat Cyp7a1 microsomal activities.

CHAPTER 4

4. 1,25-DIHYDROXYVITAMIN D3 UPREGULATES P-GLYCOPROTEIN

ACTIVITIES, EVIDENCED BY INCREASED RENAL AND BRAIN EFFLUX OF

DIGOXIN IN MICE IN VIVO

Edwin C.Y. Chow, Matthew R. Durk, Carolyn L. Cummins, and K. Sandy Pang

Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of

Toronto, Canada

4.1 ABSTRACT

Farnesoid X receptor (FXR) effects, in addition to vitamin D receptor (VDR) effects, were observed in rat liver after treatment with 1α,25-dihydroxyvitamin D3 [1,25(OH)2D3], the natural ligand of VDR, due to increased bile acid absorption as a consequence of Asbt induction. To investigate whether the increased Mdr1/P-gp expression in the rat liver and kidney is due to VDR and not FXR, we examined changes in Mdr1/P-gp expression in fxr(+/+) and fxr(-/-) mice after intraperitoneal dosing of vehicle vs. 1,25(OH)2D3 (0 or 2.5

μg/kg every other day for 8 days). Renal and brain levels of Mdr1 mRNA and P-gp protein were significantly increased in both fxr(+/+) and fxr(-/-) mice treated with 1,25(OH)2D3, confirming that Mdr1/P-gp induction occurred independently of the FXR. Functional increases in P-gp were evident in 1,25(OH)2D3-treated fxr(+/+) mice given intravenous bolus doses of the P-gp probe, [3H]digoxin (0.1 mg/kg). Decreased blood (24%) and brain

(29%) exposure, estimated as AUCs, due to increased renal (74%) and total body (34%) clearances of digoxin were observed in treated mice. These events were predicted by physiologically-based pharmacokinetic (PBPK) modeling that showed increased renal secretory intrinsic clearances (3.45-fold) and brain efflux intrinsic clearances (1.47-fold) in the 1,25(OH)2D3-treated mouse, trends that correlated well with increases in P-gp protein expression in tissues. The clearance changes were less apparent due to high degree of renal reabsorption of digoxin. Nonetheless, the observations suggest an important role of the

VDR in the regulation of P-gp in the renal and brain disposition of P-gp substrates.

4.2 INTRODUCTION

P-Glycoprotein (P-gp), the gene product of the multidrug resistance protein 1

(MDR1 or ABCB1), is a 170-kDa membrane transporter and member of the adenosine triphosphate (ATP)-binding cassette (ABC) superfamily (Juliano and Ling, 1976). P-gp functions as an ATP-powered drug efflux pump that interacts with numerous large, nonpolar, and weakly amphipathic compounds and cations of no apparent structural similarity. P-gp is highly expressed in many major organs and tissues in the body, including intestine, liver, brain, kidney, colon, testes and placenta (Cordon-Cardo et al., 1990). For this reason, P-gp plays a critical role in drug absorption, distribution, and elimination and is recognized as an important target for drug-drug interactions (Yu, 1999).

Over the past few decades, multiple nuclear receptors and transcription factors have been found to regulate MDR1/P-gp expression (Reschly and Krasowski, 2006). These include the pregnane X receptor (PXR) and the constitutive androstane receptor (CAR), commonly referred to as xenobiotic-sensing nuclear receptors. Transactivation of the mouse Mdr1 gene by the humanized PXR was observed to result in altered drug disposition in transgenic mice (Bauer et al., 2006). The rodent Mdr1 or human MDR1 gene is also regulated by the farnesoid X receptor (FXR) (Landrier et al., 2006; Martin et al., 2008) and the FXR ligand, chenodeoxycholic acid (CDCA), increased MDR1 mRNA in HepG2 cells

(Martin et al., 2008). The vitamin D receptor (VDR) displays similar homology with PXR and CAR (Reschly and Krasowski, 2006), and is another nuclear receptor that transactivates MDR1. 1,25-Dihydroxyvitamin D3 [1,25(OH)2D3], the natural, active ligand of VDR (Tanaka et al., 1973), upregulates MDR1/P-gp expression in human colon

carcinoma cell lines, such as the Caco-2, LS180 and LS174T cells (Aiba et al., 2005; Fan et al., 2009; Tachibana et al., 2009), and elevated P-gp in human airway epithelium-derived

Caclu-3 cells (Patel et al., 2002). These results are consistent with the existence of a vitamin D response element (VDRE) in the MDR1 gene (Saeki et al., 2008).

In rats treated with 1,25(OH)2D3 in vivo, VDR led to increases in both Mdr1a mRNA and P-gp protein in liver and kidney (Chow et al., 2009; Chow et al., 2010). Due to low levels of VDR in rat liver (Gascon-Barré et al., 2003), induction of Mdr1a was suspected to be a result of indirect FXR effects elicited by induction of the apical sodium- dependent bile acid transporter (Asbt) in the rat intestine, culminating in increased absorption of bile acids into the portal blood (Chen et al., 2006) rather than from direct

VDR effects. Upon entering the liver, bile acids, ligands of FXR, elicit hepatic FXR effects

(Chow et al., 2009). Inasmuch as MDR1/P-gp may also be induced by bile acids (Martin et al., 2008), the increase in hepatic P-gp could be the consequence of both direct VDR effects and/or indirect FXR effects upon 1,25(OH)2D3 treatment to the in vivo rats.

To determine whether regulation of Mdr1/P-gp was via the direct action of the VDR or indirectly via the FXR, we carried out the present study to examine Mdr1/P-gp changes in fxr(-/-) mice and compared the changes in gene expressions with those for the wild-type fxr(+/+) mice after vehicle and 1,25(OH)2D3 treatment. FXR effects become obviated in fxr(-/-) knockout mice, despite that combined VDR and FXR effects would persist in fxr(+/+) wild-type mice in vivo. A protracted dosing regimen of 50 ng (2.5 µg/kg) of

1,25(OH)2D3 by intraperitoneal injection every other day for 8 days was chosen to lessen hypercalcemia (Chow et al., 2011b). Changes in protein and mRNA expression were first determined to rule out the contribution of FXR in the upregulation of Mdr1/P-gp in both 86

fxr(+/+) and fxr(-/-) mice. We further investigated the fate of an intravenous administration of tritiated [3H]digoxin, a P-gp probe, that is eliminated only via excretion in mice, to demonstrate changes in digoxin disposition due to elevated P-gp after 1,25(OH)2D3 treatment. In vivo tissue levels of [3H]digoxin, which undergoes enterohepatic recirculation and renal tubular reabsorption, in treated and untreated fxr(+/+) mice were fit to a physiologically-based pharmacokinetic (PBPK) model to appraise the role of the VDR in altering P-gp function. Special attention was given to P-gp levels in brain, the site of action of analgesics, antiepileptics, anticancer, and antiretroviral drugs, and in the heart, the site of action of digoxin.

4.3 METHODS

4.3.1 Materials

1,25(OH)2D3 was purchased in powder form from Sigma-Aldrich Canada

(Mississauga, ON). The primary antibodies for Western blotting were obtained from various sources: anti-P-gp (C219) and anti-Gapdh (6C5) and anti-Lamin B (Cat# ab45848) from Abcam, Cambridge, MA; villin (C-19) from Santa Cruz Biotechnology (Santa Cruz,

CA); anti-VDR (MA1-710) from Thermo Fisher Scientific Inc., Rockford, IL. cDNA synthesis and real-time PCR reagents were obtained from Applied Biosystems (Foster City,

CA). [3H]Digoxin (specific activity, 40 mCi/μmol) was purchased from PerkinElmer Life and Analytical Sciences (Boston, MA) and purified by HPLC to > 99% radiochemical purity (Liu et al., 2006b). All other consumable reagents, including unlabeled digoxin, were obtained from Sigma-Aldrich Canada (Mississauga, ON) and Fisher Scientific

(Mississauga, ON).

4.3.2 Induction Studies with 1,25(OH)2D3 in fxr(+/+) and fxr(-/-) Mice In Vivo

Both male and female fxr(-/-) mice were kind gifts from Dr. Frank J. Gonzalez

(National Institutes of Health, Bethesda, MD). The fxr(-/-) mice contained only the last exon of the FXR ligand binding domain and all the 3'-untranslated region of the FXR gene

(Sinal et al., 2000). The C57BL/6 pure strain male fxr(-/-) mice were genotyped using the following fxr primers: Forward 1 (wild-type allele) 5'-

TCTCTTTAAGTGATGACGGGAATCT-3'; Forward 2 (Null allele) 5'-

GCTCTAAGGAGAGTCACTTGTGCA-3'; Reverse 5'-

GCATGCTCTGTTCATAAACGCCAT-3', as described by Sinal et al. (2000). Male wild- type [fxr(+/+)] and knockout [fxr(-/-)] mice (8 – 12 weeks), bred in the animal facility of the University of Toronto (ON, Canada), were given water and food ad libitum and maintained on a 12:12-h light and dark cycle in accordance to approved protocols. Mice were injected 0 or 2.5 µg/kg of 1,25(OH)2D3 in sterile corn oil intraperitoneally (i.p.) every other day for 8 days. The concentration of 1,25(OH)2D3 in anhydrous ethanol was determined spectrophotometrically at 265 nm (UV-1700, Shimadzu Scientific Instruments,

MD), and the 1,25(OH)2D3 solution was diluted in sterile corn oil (Sigma-Aldrich, ON) for injection (Chow et al., 2009). The alternate- day regimen was chosen due to the lessened hypercalcemia observed in comparison to those given doses on consecutive days (Chow et al., 2011b).

On the 9th day, mice were anesthetized with an i.p. injection of ketamine and xylazine (150 and 10 mg/kg, respectively). After flushing the blood from the lower vena cava with 10 ml of ice-cold saline, the intestine, liver, brain, and kidneys were removed and placed on ice. The ileal segment, taken as 6 cm proximal to the ileocecal junction and 88

known to consist of the highest abundance of P-gp (Stephens et al., 2001; Liu et al., 2006b), was used for protein and mRNA analyses. The ileum was flushed with physiologic saline containing 1 mM phenylmethylsulfonyl fluoride (PMSF) to rid of feces, everted and placed into the same saline solution containing PMSF before being scraped with a tissue-scraper for the collection of enterocytes (Chow et al., 2009). The liver, brain, kidneys and heart were weighed and cut into small pieces. The scraped enterocytes and tissue pieces were snap-frozen with liquid nitrogen, and stored at -80°C until further analyses.

4.3.3 Preparation of subcellular fractions

Frozen mucosal scrapings (50-100 mg of tissue) were homogenized with 1 ml of

Tris-HCl (0.1 M, pH 7.4) buffer containing 1% protease inhibitor cocktail (Sigma-Aldrich,

ON) and sonicated, as described by Chow et el. (2009). After centrifugation at 1,000 g for

10 min at 4°C, the resulting supernatant was spun again at 21,000 g for 1 h at 4°C to yield a pellet or crude membrane fraction. The pellet was placed in a resuspension buffer (50 mM mannitol, 20 mM HEPES, 20 mM Trizma base, pH 7.4), which was premixed with 1% protease inhibitor cocktail (Sigma-Aldrich, ON). The liver, brain, and kidney tissue samples were homogenized (1:5 w/v) in a buffer (250 mM sucrose, 10 mM HEPES, and 10 mM Tris-HCl, pH 7.4), which was premixed with 1% protease inhibitor cocktail (Sigma-

Aldrich, ON). The homogenate was centrifuged at 3,000 g for 10 min at 4°C, and the resulting supernatant was spun again at 33,000 g for 60 min at 4°C. The crude membrane pellet was placed in the resuspension buffer. A unified procedure was used for preparation of the crude nuclear fraction. All tissues were homogenized in an identical fashion with the same homogenizing buffer (250 mM sucrose, 10 mM HEPES, and 10 mM Tris-HCl, pH

7.4) containing 1% protease inhibitor cocktail, and the homogenate was spun at 3,000 g 89

wherein the resultant pellet was resuspended in a nuclear buffer [60 mM KCl, 15 mM NaCl,

5 mM MgCl2•6H2O, 0.1 mM EGTA, 300 mM sucrose, 0.5 mM DTT, 0.1 mM PMSF, 300 mM sucrose, and 15 mM Trizma HCl pH 7.4] containing 1% protease inhibitor cocktail.

Protein concentration was measured by the Lowry method (Lowry et al., 1951).

4.3.4 Western Blotting

For determination of the changes in VDR and P-gp protein levels among tissues, 20 to 80 µg was used; linearity for the relative intensity was shown to exist for the different amounts of tissue used. The sample was loaded and separated by 7.5% and 10% SDS- polyacrylamide gels, respectively, for VDR and P-gp analyses and transferred onto nitrocellulose membranes (GE Healthcare, Chalfont St. Giles, Bukinghamshire, UK), as described by Chow et al. (2009). The membrane was blocked with 5% (w/v) skim milk in

Tris-buffered saline (pH 7.4) with 0.1% Tween 20 (TBS-T) (Sigma-Aldrich, ON), and washed three times with TBS-T before incubating with the primary antibody solution (2% skim milk) overnight at 4°C. Thereafter, the membrane was washed three times with TBS-

T and incubated with a secondary antibody (2% skim milk) at room temperature for 2 h.

The membrane was washed three times, and then incubated with chemiluminescence reagents from GE Healthcare for visualization of the band intensity, quantified by scanning densitometry (NIH Image software; http://rsb.info.nih.gov/nih-image/). Protein loading error was corrected by normalizing the target protein band against the protein band of the house-keeping gene: villin for intestinal samples, Lamin B for the comparison of VDR among tissues, and Gapdh for the same tissue type.

4.3.5 Quantitative Real-Time Polymerase Chain Reaction (qPCR)

The detailed procedure of RNA extraction has been described previously (Chow et al., 2009). For total RNA isolation, scraped enterocytes and other organ tissues were homogenized with TRIzol (50-100 mg/ml) solution and extracted with the TRIzol extraction method (Sigma-Aldrich, ON) according to the manufacturer’s protocol, with modifications. RNA purity of each sample was checked by 260 nm/280 nm and 260 nm/230 nm absorbance ratios (≥1.7). Exactly 1.5 µg of total RNA was converted to cDNA.

Real-time quantitative polymerase chain reaction (qPCR) was performed with the SYBR

Green detection system (Applied Biosystems 7500 Real-Time PCR System, Streetsville,

ON). Information on the primer sequence is summarized in Table 4-1, and the primer specificity was checked by BLAST analysis (http://www.ncbi.nlm.nih.gov/BLAST/).

Critical threshold cycle (CT) values of the target genes were collected using ABI Sequence

Detection software version 1.4. The target gene mRNA data was normalized to the housing-keeping gene: villin for intestinal samples and cyclophillin for other tissue samples.

The difference in CT values (∆CT) between target and house-keeping genes was compared

- to the corresponding ∆CT of the vehicle control (∆∆CT) and expressed as fold expression, 2

(∆∆CT), for relative mRNA quantification.

Table 4-1 Mouse primer sets for quantitative Real-Time PCR Gene Bank Number Forward (5’→ 3’ Sequence) Reverse (5’→ 3’ Sequence)

Mdr1a NM_011076 TACGACCCCATGGCTGGATC GGTAGCGAGTCGATGAACTG

VDR NM_009504 GAGGTGTCTGAAGCCTGGAG ACCTGCTTTCCTGGGTAGGT

Villin NM_009509 TCCTGGCTATCCACAAGACC CTCTCGTTGCCTTGAACCTC

Cyclophillin X58990 GGAGATGGCACAGGAGGAA GCCCGTAGTGCTTCAGCTT

3 4.3.6 Pharmacokinetic Study of [ H]Digoxin in Vehicle or 1,25(OH)2D3 Treated Mice

The digoxin study was performed with fxr(+/+) mice only (same as C57BL/6 pure) treated with vehicle (corn oil) or 2.5 μg/kg (or 50 ng/mouse) of 1,25(OH)2D3 given intraperitoneally (i.p.) every other day for 8 days. On the 9th day, each mouse received a bolus injection of 0.1 mg/kg digoxin and ~1.6 million dpm of [3H]digoxin in ~ 100 μl filtered saline solution containing 1.5% propylene glycerol via the tail vein and was placed inside a glass beaker atop a piece of suspended aluminum wire mesh for separate collection of feces and urine. Three to five mice were rendered unconscious in a carbon dioxide chamber and used for blood collection by cardiac puncture via a 1 ml syringe, which was pre-rinsed with heparin (1000 IU/ml). The lower vena cava was perfused with ice-cold saline to remove blood from tissues, and thereafter, the liver, kidney, heart, brain, and small intestine were removed rapidly, weighed, snap-frozen in liquid nitrogen and stored at -80°C for future analyses. The feces above the wire mesh were collected and pooled together with flushed luminal contents of the small intestine and colon with ice-cold saline into pre-tared

15 ml polyethylene tubes. The urine from the collection beaker was pooled, together with water rinses (twice with 1 ml of water).

4.3.7 [3H]Digoxin Analyses

Blood samples (0.25 ml) were deproteinized upon addition of methanol 1:4 (v/v).

The sample was mixed for 1 min and centrifuged at 14,000 g for 10 min at 4°C, and 0.9 ml of the supernatant was removed for liquid scintillation counting. Varying known counts of

[3H]digoxin were added to blank blood (same volume as samples) and used as standards; they were processed under identical conditions for construction of a calibration curve for

the determination of total radioactivity of [3H]digoxin in blood samples. Similarly, the liver, kidney, heart, brain, and small intestine tissue were homogenized 1:3 (w/v) in saline solution, and 0.25 ml (tissue) or 1.2 ml (brain) of the homogenate was deproteinized with

MeOH, 1:4 (v/v), while removing 1.0 to 4.8 ml for liquid scintillation counting. Blank tissue homogenate samples, spiked with the appropriate aliquots of dpm's for construction of calibration curves for each tissue, were processed according to the same deproteinization procedure. Counts in the fecal mixtures were estimated to denote the extents of biliary and luminal excretion; the contents were homogenized, and the radioactivity was extracted with ethyl acetate, 1:4 (v/v) after mixing (vortex) vigorously for 5 min and spun at 14,000 g for

10 min at 4°C. An aliquot (0.95 ml) of the upper layer was removed for liquid scintillation counting. Again, standards of known dpm's in fecal material were processed in an identical manner and used for the construction of the calibration curve. The total radioactivity in urine samples was determined directly by liquid scintillation counting.

The HPLC assay of Liu et al. (2006b) was used to separate digoxin from its di- and mono-digitosoxides and the aglycone. The dpm's in the deproteinized samples from blood, urine, feces, and tissue were separated by HPLC to resolve [3H]digoxin from its metabolite

(Liu et al., 2006b). The supernatant was evaporated under nitrogen gas and then reconstituted with 100 μl methanol. The reconstituted residue was centrifuged and 75 μl of the supernatant was injected into the HPLC Shimadzu system as described by Liu et al.

(2006b). Separation was achieved by a C18 reverse-phase column (Altech Associates,

Deerfield, IL; 4.6 × 250 mm, 10 μm particle size) and a binary gradient consisting of water and acetonitrile, with an initial condition of 18% acetonitrile maintained at a flow rate of 1 ml/min, then increased to acetonitrile to 28% (Liu et al., 2006b). The eluted fractions were 93

collected at 1 min intervals and counted, and fractions of digoxin and its metabolites recovered in the sample were multiplied by the total count of the sample to arrive at individual dpm of digoxin and the metabolites. Results from HPLC revealed that, on average, unchanged [3H]digoxin represented about 98% of the total radioactivity for all blood and tissue samples (data not shown). A previous report also showed that the metabolic clearance of digoxin in mice was only about 3% of total clearance (Kawahara et al., 1999). Thus, the total radioactivity of the sample was taken to represent unchanged

[3H]digoxin.

4.3.8 Modeling and Fitting

4.3.8.1 Whole body physiologically-based pharmacokinetic modeling (PBPK)

The concept of intestinal segregated-flow was incorporated in the PBPK or physiologically-based pharmacokinetic model (Li et al., 2002a; Liu et al., 2006b). In this model, a minor portion of the intestinal flow (5-30%) perfuses the enterocyte region whereas the majority of flow perfuses the serosal region (remaining flow, >70%) (Cong et al., 2000). The model consists of tissue compartments that describe digoxin concentrations in the blood (CB), heart (Cheart), kidney (CK), liver (CL), enterocyte (Cen) and serosal (Cs) tissues of the small intestine, other tissues (Cother), brain tissue (CBr) and blood of the brain

(CBr,B) (Fig. 4-1). The volume and tissue to blood partition coefficient (tissue concentration/blood concentration) of each tissue compartment, denoted as V and KTB respectively, are further qualified with the appropriate subscript for that tissue. The tissue/blood partition coefficient (KTB) of digoxin for the small intestine, liver, kidney, and heart are obtained experimentally from the tissue/blood ratios towards the end of the study

(close to 600 min). We recognize that the tissue/blood concentration ratio would underestimate the true partitioning ratio within eliminating organs (Khor et al., 1991; Chiba et al., 1998). For the intestine, KTB,I or CI/CB,I (intestine tissue/blood leaving intestine) is assumed to be identical for both the enterocyte and serosal tissue. This is an oversimplication since for KTB,I would not equal Cen/Cenb (ratio of the enterocyte tissue concentration relative to that in blood leaving the enterocyte) in view of the known, luminal secretion by the P-gp. The estimate of KTB,I should be a better estimated by Cs/Csb (ratio of the serosal tissue concentration relative to that in blood leaving the serosal tissue due to lack of elimination), because the serosa represents the non-eliminating tissue of the intestine. The distortion of KTB,I by luminal secretion should be low because the flow to the enterocyte region was low, rendering a flow-weighted average concentration in intestine that would be quite close to the true estimate, especially when luminal secretion is absent.

The intrinsic secretory clearances for kidney, liver, small intestine and for brain efflux that are representative of P-gp activities are denoted as CLint,sec,K, CLint,sec,H, CLint,sec,I,

Br Br and CLef , respectively; CLin represents the digoxin uptake intrinsic clearance into the brain tissue. Biliary secretion (CLint,sec,H) followed by reabsorption (rate constant, ka) allow for the enterohepatic circulation (EHC) of digoxin. For kidney, the filtrated and secreted digoxin is prone to reabsorption, and the net excretion is modified the fraction reabsorbed

(FR) within the renal tubule, as described by Levy (1980). The roles of urinary pH, the pKa of digoxin, and the urinary flow rate on digoxin reabsorption are not considered, and the extent of reabsorption is assumed to be the same for both the control and 1,25(OH)2D3- treated mice. The differential equations that relate to mass transfer across these organs/tissues are summarized in 4.6 APPENDIX. 95

4.3.8.2 Fitting

Fitting of the PBPK model to data from the vehicle control and 1,25(OH)2D3-treated fxr(+/+) mice was performed with the program, Scientist® (Micromath Version 2.0, St.

Louis, MO). Appropriate weighting schemes (unity, 1/observation, and 1/observation2) were used. Some parameters were fixed in the fitting procedure; these included the blood and tissue volumes (V) and organ flow rates (Q), unbound fraction of digoxin (fP and fB), and blood/plasma concentration ratio (B/P). The tissue/blood partition coefficient (KTB) of digoxin for the small intestine, liver, kidney, and heart were obtained experimentally from the tissue/blood ratios towards the end of the study (close to 600 min).

The first strategy was to estimate the parameters first with the control data. The blood, urine, feces and tissue data for the small intestine, liver, kidney, brain and heart in control mice were used in the fitting to obtain the tissue/blood partition coefficient of the

Br lumped tissues KTB,other, ka, CLin , FR, the fractional intestinal flow entering the enterocyte region (fQ), and the apparent intrinsic clearances. Because the unbound fraction of digoxin in the intestine, liver, kidney, and brain (fI, fL, fK, and fBr) are unknown, the fitted intrinsic

Br secretory clearances (CLint,sec) and efflux clearance (CLef ) were expressed as the product of tissue unbound fraction in tissue and intrinsic secretory clearance (for example, the apparent renal intrinsic secretory clearance, CL'int,sec,K is expressed as fKCLint,sec,K). With

Br the assumption that fQ, KTB,other, ka, CLin , and FR were identical between control and treated mice, these estimates obtained the first fit were used to estimate the apparent intrinsic clearances of the treated mice in a subsequent fit. The second strategy was to use both sets of control and treated data in the same (or forced) fit to arrive at parameter

Br estimates. In the fit, fQ, KTB,other, ka, CLin , and FR were again considered as common and unchanged parameters for both control and treated mice, and the apparent intrinsic secretory clearances of the small intestine, liver, and kidney, and efflux clearances of the

Br brain (fBr CLef ) were allowed to alter due to 1,25(OH)2D3 treatment. The weighting of two or 1/observation2 yielded the highest model selection criterion and lowest coefficient of variation (standard deviation/ parameter value).

4.3.9 Statistical Analysis

Protein and mRNA data are expressed as mean ± standard deviation. The two-tailed

Student's t test was used to compare differences between the vehicle control and treatment groups. For mRNA and protein analyses, data for the vehicle-treated sample from the fxr(+/+) mouse was set as the control (value set as unity) and used for comparison with those of other control and treatment samples. A P value of less than 0.05 was viewed as significant.

4.4 RESULTS

4.4.1 VDR and Mdr1a/P-gp mRNA and protein expression in the ileum, liver, kidney and brain of fxr(+/+) and fxr(-/-) mice

4.4.1.1 Distribution of VDR protein expression among tissues

With anticipation that the abundance of VDR would vary among tissues, samples containing 50 µg of total crude nuclear protein were analyzed to examine the distribution of

VDR protein in the nuclear fractions in ileum, liver, kidney and brain of fxr(+/+) and fxr(-/-) mice (data not shown). Linearity was shown to exist within the protein concentration range.

However, larger variations in Gapdh were found in comparison to those for Lamin B 97

among different tissues after loading of the same amount of tissue protein (data not shown).

Thus, VDR protein intensities among different tissues were normalized to Lamin B. VDR protein expression for the ileum and kidney in both fxr(+/+) and fxr(-/-) mice was high and similar; whereas VDR protein expression was significantly lower in the liver (29-35% of ileum control) and even lower for the brain (13-27% of ileum control) of both fxr(+/+) and fxr(-/-) mice. These results generally showed a lack of difference in VDR protein expression between the fxr(+/+) and fxr(-/-) mice.

Basal levels of VDR mRNA in the ileum, kidney and brain were similar in both fxr(+/+) and fxr(-/-) mice, whereas hepatic VDR mRNA in fxr(-/-) mice was three times higher that of fxr(+/+) mice (Fig. 4-2A). Upon 1,25(OH)2D3 treatment, a significant increase in VDR mRNA and protein (~3-fold) was observed for the kidney with

1,25(OH)2D3 treatment for both the fxr(+/+) and fxr(-/-) mice (Figs. 4-2), though there was a slightly lower VDR protein in the brain of the 1,25(OH)2D3-treated fxr(-/-) mouse compared to that of control fxr(-/-) mouse (Fig 4-2B). The reason for the latter was unknown.

4.4.1.2 Effects of 1,25(OH)2D3 on Mdr1 mRNA and P-gp protein expression in both fxr(+/+) and fxr(-/-) mice

Levels of Mdr1 mRNA and P-gp protein expression in ileum and liver of both the fxr(+/+) and fxr(-/-) mice remained unaltered with 1,25(OH)2D3 treatment (Fig. 4-3). In contrast, treatment of 1,25(OH)2D3 led to increased Mdr1 mRNA and P-gp protein expression in kidney and brain of both fxr(+/+) and fxr(-/-) mice (Fig. 4-3), though the natural abundance of Mdr1 mRNA in brain of fxr(-/-) mouse was considerably lower than that of the wild-type counterpart, the fxr(+/+) mouse (P < .05) (Fig. 4-3A). 98

Dose

Blood CB, VB

CBr, VBr Brain Tissue CLBr CLBr ef in Q Brain Br CBr,B, VBr,B Blood

Heart Q C , V , K heart heart heart TB,heart

Other Tissues Q other Cother, Vother, KTB,other

Kidney QK CK, VK, KTB,K

GFR CLint,sec,K A urine Liver QHV QHA CL, VL, KTB,L

QPV

Small Intestine Serosal C , (1-f )V , K CL s Q I TB,I int,sec,H (1-fQ)QPV Serosal blood

Mucosal blood Enterocyte fQQPV Cen, fQVI, KTB,I

CLint,sec,I ka Intestinal Alumen lumen

Figure 4-1 Whole body PBPK modeling with enterohepatic circulation and renal reabsoprtion of [3H]digoxin. Whole body PBPK model with blood and tissue (brain tissue, brain blood, heart, kidney, liver, small intestine, and other tissue) compartments. KTB, Q, and V represent the tissue to blood partition coefficient, blood flow to organ, and tissue volume, respectively; Br Br fQ is the fractional intestinal blood flow to the enterocyte region. CLin and CLef are the intrinsic influx and efflux clearances into and out of brain, respectively. Other tissues: other tissue, brain, kidney, liver, and small intestine, are denoted by other, Br, K, L, and I; CLint,sec,K, CLint,sec,L, and CLint,sec,I are the intrinsic secretory clearances for the kidney, liver, and small intestine, respectively; ka is absorption rate constant in the intestine. See the 4.6 Appendix and Tables 4-3 and 4-4 for details. 99

(A) (B)

2.5 Ileum 2.5 Ileum Vehicle control Vehicle control 2.0 2.0 1,25(OH)2D3 1,25(OH)2D3 1.5 1.5

1.0 1.0

0.5 0.5 Relative VDRVDRRelative Expression Expression Protein Protein Relative VDR mRNA Expression mRNA VDR Relative 0.0 0.00 0 ab ab X Data X Data 6.0 Liver 6.0 † Liver fxr(+/+) fxr(-/-) 5.0 Vehicle control 5.0 Vehicle control 1,25(OH)2D3 1,25(OH)2D3 4.0 4.0

3.0 3.0 Ileum 2.0 2.0 1.0 1.0 Liver Relative VDR VDR mRNA mRNAExpressionExpression 0.00 Relative VDR Protein Expression 0.00 + /ab + - / - ab X Data X Data Kidney 5.0 Kidney 5.0 Kidney Vehicle control 4.0 Vehicle control Brain 4.0 1,25(OH)2D3 1,25(OH)2D3 * * 3.0 3.0 0 2.5 0 2.5 * * 2.0 2.0 1,25(OH)2D3 (μg/kg) 1.0 1.0 Relative VDR mRNA Expression Relative Relative VDR mRNA Expression Expression mRNA mRNA VDR VDR Relative Relative VDR Protein Expression VDR Protein Relative

0.00 Expression Expression Protein Protein VDR VDR Relative 0 ab 0.0 fxr (+/+) fxr (-/-) X Data 3.0 Brain 3.0 Brain Vehicle control Vehicle control 2.5 1,25(OH) D 2.5 2 3 1,25(OH)2D3 2.0 2.0

1.5 1.5 * 1.0 1.0

0.5 0.5 Relative VDR Expression Expression mRNA mRNA VDR VDR Relative 0.00 Relative VDR VDR Protein Protein ExpressionExpression 0.00 fxr (+/+) fxr (-/-) fxr (+/+) fxr (-/-)

Figure 4-2 Effects of 1,25(OH)2D3 on VDR (A) mRNA and (B) protein expression in ileum, liver, kidney, and brain. The VDR protein band was shown at 50 kDa. The “*” denotes P < 0.05 compared to vehicle control using the two-tailed Student’s t test. Fifty µg of crude nuclear protein was loaded in each lane. Western blot bands of ileum, liver, kidney, and brain (from left to right): lane 1, vehicle control of fxr(+/+) mouse; lane 2, 2.5 µg/kg 1,25(OH)2D3 treated fxr(+/+) mouse; lane 3, vehicle control of fxr(-/-) mouse; lane 4, 2.5 µg/kg 1,25(OH)2D3 treated fxr(-/-) mouse.

100

(A) (B)

3.0 Ileum 3.0 Ileum

2.5 Vehicle control 2.5 Vehicle control 1,25(OH)2D3 1,25(OH)2D3 2.0 2.0 1.5 1.5

1.0 1.0

0.5 0.5

0.00 RelativeP-gpP-gp Protein Protein Expression Expression 0.00 RelativeMdr1a mRNA Expression ab ab X Data X Data 3.0 Liver 3.0 Liver fxr(+/+) fxr(-/-) Vehicle control 2.5 2.5 Vehicle control 1,25(OH)2D3 1,25(OH)2D3 2.0 2.0

1.5 1.5 Ileum

1.0 1.0 Liver 0.5 0.5 0.00 RelativeMdr1aMdr1a mRNA mRNA Expression Expression Relative P-gp Protein Expression Expression Protein Protein P-gp P-gp Relative 0.00 ab Kidney X Data fxr (+/+) fxr (-/-) 8.0 8.0 Kidney Kidney * 7.0 7.0 Brain Vehicle control * Vehicle control 6.0 1,25(OH) D 2 3 6.0 1,25(OH)2D3 5.0 * 5.0 0 2.5 0 2.5 4.0 4.0 * 3.0 3.0 1,25(OH)2D3 (μg/kg) 2.0 2.0 1.0 1.0

0.00 Expression Expression Protein Protein P-gp P-gp Relative 0 Relative Mdr1a mRNA Expression mRNA Mdr1a Relative ab 0.0 Relative P-gpRelative Expression Protein Mdr1 mRNAExpression Relative fxr (+/+) fxr (-/-) X Data 3.0 Brain 3.0 Brain Vehicle control 2.5 Vehicle control 2.5 1,25(OH)2D3 1,25(OH) D * 2 3 * 2.0 2.0 * 1.5 1.5 * 1.0 1.0 † 0.5 0.5

0.0 Expression Protein P-gp Relative 0.00 Relative Mdr1a mRNA Expression mRNA mRNA Expression Mdr1a Mdr1a Relative 0 fxr (+/+) fxr (-/-) fxr (+/+) fxr (-/-)

Figure 4-3 Effects of 1,25(OH)2D3 on Mdr1 (A) mRNA and (B) P-gp protein in brain, kidney, liver, and ileum. P-gp protein band was shown at 170 kDa. “*” indicates P < 0.05 compared to vehicle control, whereas “†” indicates P < 0.05 compared to the level of vehicle-control of fxr(+/+) mice using the two-tailed Student’s t test. Twenty µg of crude membrane protein for ileum, liver, and kidney was loaded in each lane; 80 µg of crude membrane protein was used for brain P-gp. Western blot bands of ileum, liver, kidney, and brain (from left to right): lane 1, vehicle control of fxr(+/+) mouse; lane 2, 2.5 µg/kg 1,25(OH)2D3 treated fxr(+/+) mouse; lane 3, vehicle control of fxr(-/-) mouse; lane 4, 2.5 µg/kg 1,25(OH)2D3 treated fxr(-/-) mouse.

101

3 4.4.2 Effects of 1,25(OH)2D3 Treatment on the Pharmacokinetics of [ H]Digoxin in fxr(+/+) Mice

4.4.2.1 Blood decay profiles and excretion of [3H]digoxin after intravenous administration in fxr(+/+) mice

Fig. 4-4 shows the blood concentration-time profiles and cumulative amounts recovered in urine and fecal matter vs. time after a single intravenous dose (0.1 mg/kg) of

[3H]digoxin. A biexponential decay of [3H]digoxin was noted in the blood concentration

(normalized to dose) vs. time profile (Fig. 4-4A). The concentration of [3H]digoxin was significantly lower in the 1,25(OH)2D3-treated group only at 360 min (6 h). The individual data points were averaged, and the mean blood values were used to calculate the area under the curve AUC(0∞) by the trapezoidal rule and extrapolation of the last concentration over the terminal half life () (Table 4-2). This AUC(0→∞) value for digoxin in 1,25(OH)2D3- treatment mice was only 76% that of control mice. The apparent total body and renal clearances were increased by 34% and 74%, respectively, and the terminal half life, decreased by 30% in the mice treated with 1,25(OH)2D3 (Table 4-2). The Vdarea (CLtotal/) for both the control and treatment group remained relatively unchanged.

The cumulative amounts of [3H]digoxin excreted in urine at 600 min and feces were comparable (Fig. 4-4B and 4-4C). Amounts in urine in 1,25(OH)2D3-treated mice were higher than those in control mice at almost every sampling time subsequent to 30 min (P

<.05), except at 360 min (Fig. 4-4B), and were significantly higher (42.6 ± 9.5 vs. 27.0 ±

3 5.6 %dose) at 600 min. The cumulative fecal amounts of [ H]digoxin in the 1,25(OH)2D3- treated group were increased only slightly compared to those of control mice (Fig. 4-4C).

The apparent renal (Fig. 4-5A) clearance, estimated as the slope upon plotting the

102

3 cumulative amount of [ H]digoxin excreted into urine vs. blood AUC of 1,25(OH)2D3- treated mice (0.074 ml/min) was 74% higher than that of the control mice (0.0426 ml/min).

Our control values were slightly lower than those estimated by others for renal clearance

(0.069 ml/min) (Kawahara et al., 1999) and total clearance (0.083 ml/min) of digoxin

(Griffiths et al., 1984). The filtration clearance, calculated as fPGFR [where fp is 0.78

(Davies and Morris, 1993; Kawahara et al., 1999)] was 0.22 ml/min, a value much higher in relation to the observed renal clearance (0.0426 ml/min), suggesting that reabsorption played a significant role in the net renal clearance of digoxin in mice.

The fecal (sum of net intestinal and biliary) clearance (0.0969 ml/min), estimated as the slope upon plotting the cumulative amount of [3H]digoxin recovered from feces vs. the blood AUC, was higher than the renal clearance of digoxin (Table 4-2). 1,25(OH)2D3- treatment increased the fecal clearance by 30% over the control group (0.0742 ml/min) (Fig.

4-5B). Reabsorption of the biliary and intestinal excreted digoxin by the intestine

(enterohepatic recirculation) must have occurred. This possibility was commented on by

Kawahara et al. (1999), and this could have affected the estimate of the fecal secretary clearances, which deviated from the slopes in Fig. 4-5B.

103

(A) (B) (C) 10.0 60 60 Vehicle control Vehicle control Vehicle control 5.0 50 1,25(OH)2D3 1,25(OH)2D3 50 1,25(OH)2D3 * H]DigoxinH]Digoxin H]DigoxinH]Digoxin H]Digoxin H]Digoxin H]Digoxin 33 33

* 33 2.0 40 40 1.0 30 * 30 0.5 * * (% dose(% dose / / ml) ml) 20 * 20 in Urine (% dose) in Urine (% dose) in Fecesin Feces (% (% dose) dose) 0.2 10 * 10

0.1 0 0 CumulativeCumulative Amount Amount of of [ [ BloodBlood Concentration Concentration of of [ [ CcumulativeCcumulative Amount Amount of of [ [ 0 100 200 300 400 500 600 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Time (min) Time (min) Time (min)

Figure 4-4 Plots of [3H]digoxin (A) blood concentration, and cumulative amounts in (B) urine and (C) feces vs. time. The “*” indicates P < 0.05 compared to vehicle control using the two-tailed Student’s t test. The solid and dashed lines represent the fitted lines for control and the 1,25(OH)2D3 treated groups, respectively, upon force-fitting of the data to the PBPK model.

(A) (B)

60 60 y = 0.0969x Vehicle control 2 Vehicle control R = 0.4724 50 1,25(OH)2D3 y = 0.0740x 50 1,25(OH)2D3 R2 = 0.9338 H]Digoxin H]Digoxin H]Digoxin H]Digoxin 33 40 33 40 y = 0.0742x 30 30 R2 = 0.6512

20 20 y = 0.0426x in Urine (% dose) in in Urine (% dose) in R2 = 0.921 (% dose) (% dose) in Feces in Feces 10 10

0 0 Cumulative Amount of of Amount Amount [ [ Cumulative Cumulative Cumulative Amount of [[ of of Amount Amount Cumulative Cumulative 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700 Blood AUC (% dose / ml * min) Blood AUC (% dose / ml * min)

3 Figure 4-5 Plot of the amount [ H]digoxin excreted to (A) urine and (B) feces vs. the blood AUC(0→t). The slope represents the apparent renal (A) and biliary (B) clearances. The solid line represents control group whereas the dashed line represents 1,25(OH) D treated group. 104 2 3

Table 4-2 Noncompartmental estimates of digoxin parameters in tissue and blood of vehicle- and 1,25(OH)2D3-treated wild-type mice

Vehicle control 1,25(OH)2D3 Treated/Control a Blood AUC(0→∞) (% dose·min/ml) 871 661 0.76 b Apparent total body clearance, CLtotal (ml/min) 0.105 0.140 1.34 c Filtration clearance, fPGFR (ml/min) 0.218 0.218 1.00 d Apparent renal clearance, CLR (ml/min) 0.0426 0.0740 1.74 e Apparent fecal (biliary) clearance, CLfeces (ml/min) 0.0742 0.0969 1.30 f t1/2β (h) 7.63 5.32 0.70 g Volume of distribution Vdarea (ml) 69.0 64.3 0.93 h Small intestine AUC(0→600 min) (% dose·min/g) 3733 3091 0.83 h Liver AUC(0→600 min) (% dose·min/g) 1508 1507 1.00 h Kidney AUC(0→600 min) (% dose·min/g) 708 677 0.96 h Brain AUC(0→600 min) (% dose·min/g) 68.5 48.9 0.71 h Heart AUC(0→600 min) (% dose·min/g) 785 712 0.91 a all data points were averaged to provide the mean, and the value was used for calculation; the blood AUC(0→∞) was estimated by addition of AUC(0→600 min) to AUCextrapolated to infinity or (C(600 min)/ ), where  is the terminal decay constant b calculated as dose/AUC(0→∞) c average of literatures values for fp (0.78) and GFR (0.28 ml/min) from (Davies and Morris, 1993; Kawahara et al., 1999) d renal clearance, slope in Fig. 4-5A e fecal clearance, slope in Fig. 4-5B f 0.693/β whereby was obtained upon regression of the three last data points of the averaged blood concentration g calculated as CLtotal/ h all data points were averaged to provide the mean, and the value was used for calculation; tissue AUC(0→600 min) was calculated by the trapezoidal rule

105

4.4.3 Tissue Distribution

4.4.3.1 Estimation of area under the curves (AUCs)

The amounts of [3H]digoxin, normalized to per gram (g) tissue in the small intestine, liver, kidney, brain, and heart tissues were plotted against time (Fig. 4-6). Levels of digoxin in the small intestine, liver, kidney and heart were generally similar, except for the brain. In both control and treated groups, [3H]digoxin levels in the small intestine, liver, kidney, and heart displayed instantaneous accumulation within the first few minutes, followed by a rapid decay by 10 min, then gradually decayed thereafter (Figs. 4-6A, 4-6B, 4-6C, and 4-

6E). However, the peaks occurred slightly later (~ 100 min) in the brain. In brain, levels of

[3H]digoxin in the treatment group were markedly lower than those of controls (Fig. 4-6D).

When the areas under the curve (AUCs) in the small intestine, liver, kidney, brain, heart, and blood, derived from observations on the amounts per g of tissue, were estimated by the trapezoidal rule (Table 4-2), an apparently lower AUC(0-600 min) (23%) was found for the brain in the treatment group. The AUC(0-600 min)s for the small intestine, liver, kidney, and heart remained relatively unchanged for both groups (Table 4-2).

4.4.3.2 Tissue to blood AUC vs. time profile

The tissue partitioning coefficient of digoxin was estimated as the ratio of AUC(0→t) of tissue normalized to that of blood AUC(0→t) (Fig. 4-7). In control mice, digoxin tissue/blood AUC ratios of 6.0, 2.5, 1.12, and 1.3 were reached at 600 min for the small intestine, liver, kidney, and heart, respectively (Figs. 4-7A, 4-7B, 4-7C, and 4-7E), and treatment with 1,25(OH)2D3 exerted only minimal effects on the tissue/blood AUC ratio. In the brain, a gradual rise in AUC ratio over time was observed for both treatment and

106

control groups (Fig. 4-7D). The ratio was consistently lower than unity in the brain, and was reduced with 1,25(OH)2D3-treatment. A plateau level was not reached for the brain to blood AUC ratio at 600 min (Fig. 4-7D).

4.4.4 Whole Body PBPK Modeling

For both types of fits, the blood, urine, feces, small intestine, liver, kidney, brain, and heart data were fit to the whole PBPK model (Fig. 4-1) with literature values of tissue volumes and blood flows (Table 4-3) and the observed tissue partition coefficients (KTBs)

Br (from Fig. 4-7). For the first of the sequential fits, fQ, KTB,other, CLin , ka, and FR were 0.185

± 0.082, 2.50 ± 0.17, 0.00303 ± 0.000491 ml/min, 0.00293 ± 0.00076 min-1, and 0.830 ±

1.39, respectively, for the vehicle control group, and these values were assigned to estimate the apparent intrinsic clearances for the 1,25(OH)2D3 treatment group. The composite fits revealed a higher [3H]digoxin elimination from blood due to an increased apparent renal intrinsic clearance fKCLint,sec,K (from 0.0323 ± 2.12 to 0.188 ± 0.027 ml/min) and a greater

Br efflux from brain tissue, fBr CL ef (from 0.00228 ± 0.00482 to 0.00329 ± 0.00409 ml/min) for the 1,25(OH)2D3-treatment group (Table 4-4). The apparent renal secretory intrinsic clearance (fKCLint,sec,K) after 1,25(OH)2D3-treatment was 3.65-fold that of control, a value that correlated well to the level of P-gp induction [2.65-fold for fxr(+/+) mice] and less so to the change in apparent renal clearance (1.74-fold) (Fig. 4-5A) due to the presence of filtration and the high degree of reabsorption of digoxin. The parameter estimated for intestinal (fICLint,sec,I) and hepatic (fHCLint,sec,H) secretions via P-gp were similar between the control and treatment groups (0.0227 ± 0.0085 vs. 0.0225 ± 0.0050 ml/min for the intestine and 0.0157 ± 0.00125 vs. 0.0152 ± 0.0086 ml/min for the liver; Table 4-4). These

107

estimates, even when summed (0.0384 to 0.0377 ml/min), were low in comparison to the fecal clearances (0.074 and 0.097 ml/min) estimated from the slopes of Fig. 4-5B, suggesting that the unbound fractions fI and fH must be very low. The estimates suggest that

P-gp activities for both intestinal and biliary excretion were relatively constant (Table 4-4).

Moreover, the fitted results showed that the weighting of 1/observation2 yielded the highest model selection criterion (MSC) (Table 4-4), and lowest coefficient of variation (standard deviation/parameter value).

Estimates from the simultaneous (force) fit were generally similar to those obtained

Br from the sequential fits (Table 4-4). The common parameters, fQ, KTB,other, CLin , ka, and

FR, were estimated to be 0.158 ± 0.347, 2.42 ± 0.82, 0.000301 ± 0.000389 ml/min, 0.00216

± 0.00162 min-1, and 0.858 ± 2.53, respectively (Table 4-4). Increased apparent renal intrinsic clearance fKCLint,sec,K (from 0.074 ± 5.34 to 0.255 ± 0.029 ml/min) and greater

Br efflux from the brain tissue, fBr CL ef (from 0.00230 ± 0.00422 to 0.00337 ± 0.00363 ml/min) were observed for the 1,25(OH)2D3-treatment group (Table 4-4). The apparent renal secretory intrinsic clearance (fKCLint,sec,K) after 1,25(OH)2D3-treatment was 3.45-fold that of control, a value that correlated well to the level of P-gp induction (2.65-fold; Table

4-5) but less so to the change in renal clearance (1.74-fold) (Fig. 4-5A). These changes were not exact matches due to the presence of filtration and a high degree of reabsorption of digoxin. Simulations performed using the FR as 0.857 and fQ as 0.158 and parameters obtained with the simultaneous fit showed that the blood and kidney and brain concentrations were generally insensitive to a doubling in fKCLint,sec,K (varied from 0.074 ml/min to ½, 2, 3, and 5x) (Fig. 4-8). In contrast, for drugs whose FR approaches 0,

108

changes in fKCLint,sec,K would significantly affect tissue levels and renal excretion

(simulation not shown). The ratio for brain efflux was 1.47x higher for the 1,25(OH)2D3- treatment group, a value similar to that for Western blotting (1.8x, Table 4-5). Values of the apparent fICLint,sec,I and fHCLint,set,H were of similar magnitude (0.0168 to 0.0189 ml/min;

Table 4-4) and were unchanged with 1,25(OH)2D3 treatment, as also found from Western blotting (Table 4-5). Again their sum was much less than the apparent fecal clearances

(0.074 and 0.097 ml/min), supporting the view that fI and fH must be very small. Overall, the fitted results showed that the weighting of 1/observation2 yielded the highest model selection criterion (MSC) (Table 4-4), and lowest coefficient of variation (standard deviation/parameter value).

In both types of fits, values of fQ (Table 4-4) were similar (0.185 and 0.158) to those found for the rat intestinal preparation (Cong et al., 2000; Liu et al., 2006b) that is less than

20% of the total flow, favoring the concept of segregated flow to the enterocyte (Cong et al.,

2000). Though the fitted fKCLint,sec,K for the force fit was almost twice that for the sequential fits, the change due to 1,25(OH)2D3, denoted as the ratio of the apparent intrinsic clearances, was similar (Table 4-4). In comparison, the force fit yielded higher coefficient of variations but a higher MSC (3.58), showing that this strategy was superior over that of the sequential fits. Moreover, when the traditional PBPK intestine model (Cong et al., 2000) was used, a slightly inferior fit with higher coefficients of variation, lower

MSC, and higher sum of squared residuals were observed (data not shown).

109

(A) (B) (C)

20 15 10 10 Vehicle control Vehicle control 5 Vehicle control 10 1,25(OH)2D3 5 1,25(OH)2D3 1,25(OH)2D3 2 2 5 1 1 H]H]in Liverin Liver Digoxin Digoxin 33 H]H]DigoxinDigoxin in Kidney in Kidney 0.5 33 0.5 H] Digoxin in Small Intestine Intestine Intestine Small Small in in Digoxin Digoxin H] H] 33 2 * (%dose / /(%dose (%dose g of g of tissue)tissue) (%dose(%dose // g g of of tissue)tissue) (%dose / g oftissue) (%dose / g oftissue) 0.2

AmountAmount of of [[ 1 0.1 AmountAmount of of[ [ 0.1 0 100 200 300 400 500 600 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Amount of [[ of of Amount Amount Time (min) Time (min) Time (min)

(D) (E)

0.16 15 0.12 10 0.09 * 5 Vechile control * 1,25(OH)2D3 0.06 2 0.04 1 H]H] Digoxin Digoxin in in BrainBrain H]H]DigoxinDigoxin in Heart in Heart 33 33 Vehicle control 0.5 0.02 1,25(OH)2D3 (%dose(%dose / g of / g oftissue)tissue) (%dose / g of tissue) (%dose / g of tissue) AmountAmount of [ of [ 0.01 AmountAmount of of [ [ 0.1 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Time (min) Time (min)

Figure 4-6 Plots of amount [3H]digoxin in (A) small intestine, (B) liver, (C) kidney, (D) brain, and (E) heart. The solid and dashed lines represent the fitted lines for the control and the 1,25(OH)2D3 treated groups, respectively, upon force-fitting of the data to the PBPK model. 110

(A) (B) (C)

8.0 3.0 2.0 Vehicle control 2.5 1,25(OH)2D3 6.0 1.5 2.0

4.0 1.5 1.0

Vehicle control 1.0 Vehicle control 2.0 1,25(OH)2D3 1,25(OH)2D3 0.5 0.5 Liver / Liver Blood AUC ratio Kidney / Blood AUC ratio AUC / Blood Kidney

Small Intestine / Blood AUC ratio 0.00 0.00 0.00 0 100 200 300 400 500 600 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Time (min) Time (min) Time (min)

(D) (E)

0.20 1.6 Vehicle control 1,25(OH)2D3 0.15 1.2

0.10 0.8 Vehicle control 0.05 0.4 1,25(OH)2D3

Brain / Blood AUC ratio Heart / Blood AUC ratio

0.000 0.00 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Time (min) Time (min)

Figure 4-7 Tissue to blood AUC ratio of [3H]digoxin over time profile for the (A) small intestine, (B) liver, (C) kidney, (D) brain and (E) heart. For each data point, the average value of tissue AUC of vehicle or treatment group is normalized to the average value of blood AUC of the same group. The ratios are plotted against time. The tissue to blood partition coefficients111 of the small intestine, liver, kidney, and heart was estimated at the plateau phases of the plot.

(A) (B) f CL (0.074 ml / min) 10 FR = 0.858 K int,sec,K 10 FR = 0.858 f CL (0.074 ml / min) 5X K int,sec,K 5 3X 5 5X 2X 3X

H]Digoxin H]Digoxin H]Digoxin 1X 2X

33 2 1/2X 2 1X 0 1/2X 1 0 1 0.5 H] Digoxin in Kidney H] DigoxinH] Digoxin in Kidney in Kidney 33 0.5 (% dosedose / ml) / ml)(% (% 0.2 (%dose / g of tissue) / g of (%dose tissue) / g of (%dose 0.1 Amount of [ [ of of Amount Amount BloodBlood Concentration Concentrationofof [ [ 0.05 0.1 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Time (min) Time (min) (C) (D) FR = 0.858 0.16 50 f CL (0.074 ml / min) 0.14 K int,sec,K FR = 0.858 0.12 5X 0.1 40 3X 0.08 2X H]Digoxin H]DigoxinH]Digoxin 33 1X 0.06 1/2X 30 0 0.04 fKCLint,sec,K (0.074 ml / min) H] DigoxinH] Digoxin in Brain in Brain

33 5X 20 3X in Urine (% (% dose) in Urine 0.02 2X (% in Urine dose) (%dose / g of tissue) / g of tissue) (%dose (%dose 1X 10 1/2X

Amount of Amount of [[ 0 0.01 CcumulativeCcumulative Amount Amount of of [ [ 0 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Time (min) Time (min) Figure 4-8 Simulation of [3H]digoxin concentrations vs. time for (A) blood, (B) kidney and (C) brain, and amounts vs. time in urine (D). The solid circles represent the observed, average [3H]digoxin blood, kidney, brain and urine data from the fxr(+/+) vehicle control mouse. The solid line represents the predictions according to force-fitting of data from both the control and 1,25(OH)2D3-treated mice. The dotted lines represented simulations using fxr(+/+) vehicle control parameters from Table 3 and 4 with varying values of CL′int,sec,K, from 0 to multiples (shown as numbers) of 0.074 ml/min (control mice). Note the relative insensitivity of the (A) blood, (B) kidney and (C) brain, and amounts vs. time in urine (D) even when

CL′int,sec,K was doubled. 112

Table 4-3 Assigned parameters for PBPK modeling of [3H]digoxin in fxr(+/+) mice which were treated i.p. with vehicle (0) or 2.5 μg/kg of 1,25(OH)2D3 every other day for 8 days

Assigned Parameters Values a Qbrian (ml/min) 0.089 a QK (ml/min) 1.3 a QHA (ml/min) 0.35 a Qheart (ml/min) 0.28 a QPV (ml/min) 1.5 a Qother (ml/min) 4.48 GFR (ml/min) a 0.28 a VB (ml) 1.7 a Vheart (ml) 0.095 a VBr, B (ml) 0.025 a VBr(ml) 0.226 a VK (ml) 0.34 a VL (ml) 1.3 a VI (ml) 1.5 a Vother (ml) 13.0 a B/P 0.898 a fP 0.78

fB = fP /(B/P) 0.87 b KTB,K 1.12 b KTB,L 2.5 b KTB,I 6.0 b KTB,heart 1.3 a average of literature values from Davies and Morris (1993) and Kawahara et al. (1999) b observed value from Fig. 4-7

113

3 Table 4-4 Fitted parameters (±SD) for [ H]digoxin in fxr(+/+) mice treated i.p. with vehicle (0) or 2.5 μg/kg of 1,25(OH)2D3 every other day for 8 days based on the PBPK model shown in Fig. 4-1

Sequential Fits Force Fit of Control and 1,25(OH)2D3 Data Fitted Vehicle Control 1,25(OH)2D3 Ratio Vehicle Control 1,25(OH)2D3 Ratio Parameters

fQ 0.185 ± 0.082 Same as control 0.158 ± 0.347

KTB,other 2.50 ± 0.17 Same as control 1 2.42 ± 0.82 1 -1 ka (min ) 0.00293 ± 0.00076 Same as control 1 0.00216 ± 0.00162 1 FR 0.830 ± 1.39 Same as control 1 0.858 ± 2.53 1

Br CLin (ml/min) 0.000303 ± 0.000491 Same as control 1 0.000301 ± 0.000389 1

Br fCLBr ef (ml/min) 0.00228 ± 0.00482 0.00329 ± 0.00409 1.44 0.00230 ± 0.00422 0.00337 ± 0.00363 1.47

CL′int,sec,K or fKCLint,sec,K (ml/min) 0.0323± 2.12 0.188 ± 0.027 3.65 0.074 ± 5.34 0.255 ± 0.029 3.45

CL′int,sec,I or fICLint,sec,I (ml/min) 0.0227 ± 0.0085 0.0225 ± 0.0050 0.991 0.0175 ± 0.0377 0.0168 ± 0.0380 0.96

CL′int,sec,H or fHCLint,sec,H (ml/min) 0.0157 ± 0.0125 0.0152 ± 0.0086 0.968 0.0189 ± 0.0688 0.0189 ± 0.0621 1.00 Model selection criterion, MSC 2.99 3.23 3.58 with weighting of 1/observation2 Residual sum of square 1.68 1.51 226

114

Fold Change in P-gp Protein Expression P-gp Apparent Efflux CLint's Tissue [1,25(OH)2D3/Vehicle Control] in fxr(+/+) Mice P-gp [1,25(OH)2D3/Vehicle Control] fxr(+/+) fxr(-/-) Ratio

Ileum Efflux 0.84 0.95 0.96

Liver Efflux 1.12 1.17 1

Kidney Efflux 2.65 5.90 3.45

Brain Efflux 1.80 2.45 1.47

115

4.5 DISCUSSION

In this study, we have employed both wild-type fxr(+/+) and knock-out fxr(-/-) mice to discern whether regulation of Mdr1a/P-gp observed in the rat after 1,25(OH)2D3 dosing was due to the direct actions of the VDR or through indirect actions by the FXR

(Chow et al., 2009; Chow et al., 2010). We appraised whether the VDR played a direct role in the upregulation of the Mdr1 gene in mice. We found that there are different responses between rats (Chow et al., 2010) and mice, due to species differences that could be the result of different regulatory responses from nuclear receptors. First, Asbt induction with

1,25(OH)2D3, though observable in the rat, is deemed nonexistent in the mouse due to the presence of LRH-1 cis-acting element on the mouse Asbt promoter in the intestine that exerts a negative feedback on Asbt upon FXR activation (Chen et al., 2003; Chen et al.,

2006). For this reason, FXR effects are less in the mouse than the rat, and both Asbt mRNA and protein expression and portal bile acid concentrations were unchanged in mice after

1,25(OH)2D3 administration (data not shown). By contrast, low levels of VDR present in the rat liver (Gascon-Barré et al., 2003; Chow et al., 2009) are unlikely to forge direct links between transactivation of the Mdr1 gene by the VDR. Rather, FXR effects are suspected to be operative, especially in rat liver due to the low VDR levels and increased portal bile acid concentrations (Chow et al., 2009; Chow et al., 2011b). Secondly, protein levels of

VDR in mouse liver are relatively higher than that in rat liver (Chow et al., 2010) in comparison to the ileum, and, upon activation, VDR could potentially be exerting a greater direct effect on Mdr1/P-gp expression in mouse liver than the rat. These perceived differences aptly explain the difference in responses to 1,25(OH)2D3 induced changes in

Mdr1/P-gp expression in mice and rats. There was no induction of ileal Mdr1/P-gp in the 116

present study, and a slight but insignificant increase in Mdr1 mRNA was observed in livers treated with 1,25(OH)2D3 in fxr(+/+) but not fxr(-/-) mice (Fig. 4-3); therefore, the involvement of FXR in the regulation of hepatic Mdr1 in rats cannot be ruled out. Our previous investigation had shown that intraperitoneal administration of 1,25(OH)2D3 to rats did not elevate P-gp in intestine (Chow et al., 2010), in spite that P-gp was increased in the liver (Chow et al., 2009). The lack of induction of intestinal Mdr1 in both the mouse and rat despite the abundance of VDR could be due to low levels of 1,25(OH)2D3 reaching the intestine, or the amount of 1,25(OH)2D3 needed for activation (Chow et al., 2009).

Of significance is that activation of VDR increased the Mdr1 and P-gp expression in the kidneys and brains of both the fxr(+/+) and fxr(-/-) mice (Fig. 4-3), and the mechanism of induction is independent of FXR. The induction of renal Mdr1/P-gp in 1,25(OH)2D3 treated mice led to increases in renal and total body clearances and lowered blood AUC(0→∞) of digoxin, and these shortened the elimination half life (Table 4-2). The upregulation of P- gp expression in kidneys is thus expected to exert a significant impact on the disposition of digoxin, a P-gp substrate that is primarily renally excreted. However, the lower sensitivity of the renal clearance to changes in efflux P-gp activity appears to be due to the high fraction of digoxin reabsorbed (Fig. 4-8; Table 4-5). For other renally excreted compound that are less reabsorbed, P-gp induction is expect to play a larger role in increasing the renal clearance. Increased in brain P-gp levels by 1,25(OH)2D3 also led to lower accumulation in the brain (Fig. 4-6D; Table 4-2) and lower brain/blood partitioning (Fig. 4-7D) of digoxin, despite that VDR mRNA and protein expression in the murine brain was low. There was a clear decrease in the brain/blood AUC ratio of the 1,25(OH)2D3-treated group vs. the

117

control group between 60 to 600 min, and the decrease in ratio was the greatest at 600 min

(29% decrease).

The induction by 1,25(OH)2D3 on the upregulation of the Mdr1 gene via the VDR is explained using PBPK modeling, which predicts the changes in digoxin disposition in various tissues as a result of increased P-gp activity in kidney and brain following

1,25(OH)2D3 treatment of mice. The PBPK model accurately predicted the data pertaining to digoxin disposition in blood, urine, feces, brain, liver, kidney, heart, and small intestine

(Figs. 4-4 and 4-6) and provided an accurate description of the insignificant P-gp activity in the intestine and the less than expected effect on renal secretion due to high extent of digoxin reabsorption by the kidney (Fig. 4-8; Tables 4-4 and 4-5). By contrast, the renal clearances of digoxin in humans (125 ml/min) and rats (1.25 ml/min) (Harrison and Gibaldi,

1977b; Harrison and Gibaldi, 1977a) were slightly larger than, or comparable to, their corresponding filtration clearances [90 ml/min in humans and 0.80 ml/min in rats estimated from the literature (Steiness, 1974; Evans et al., 1990; Davies and Morris, 1993)], suggesting that secretion in these species plays a major role in renal excretion. These observations point to species differences in renal excretion, and that higher reabsorption of digoxin occurs in murine than human kidneys.

This study demonstrates that 1,25(OH)2D3 treatment increased P-gp levels in both the brain and kidney of mice. This observation leads one to consider whether treatment with 1,25(OH)2D3 or the vitamin D analogs and a P-gp substrate drug would lead to increased renal clearance of the drug. The upregulation of brain MDR1/P-gp by

1,25(OH)2D3 would have a significant impact on drug disposition, cell homeostasis, and altered pharmacological and toxicological outcomes. Indeed, high doses of 1,25(OH)2D3 118

and vitamin D analogs have been used as a therapeutic class of drugs for treatment of hyperparathyroidism, kidney diseases, and cancer (Masuda and Jones, 2006). These treatments may change the expressions of transporters and enzyme and thus alter drug disposition (Chow et al., 2009; Chow et al., 2010; Chow et al., 2011b). Drug-Drug interactions (DDIs) involving P-gp in brain may be beneficial or detrimental. For example, vitamin D analogs given concomitantly with P-gp substrates targeting the brain will increase drug efflux from the tissue, rendering the drug ineffective. When 1,25(OH)2D3 is used in combination with anticancer agents such as paclitaxel, a known P-gp substrate, to treat cancer (Masuda and Jones, 2006), lower therapeutic effects may result in the brain. P- gp substrates targeting the brain include the antiretrovirals, such as atazanavir, ritonavir, and saquinavir for HIV treatment, antipsychotics such as risperidone (Kim, 2002a; Zastre et al., 2009), antiepileptic drugs such as topiramate (Luna-Tortos et al., 2009), or CNS drugs like morphine (Groenendaal et al., 2007), are all likely to be affected. In contrast, increased

P-gp activity may decrease brain concentrations of drugs such as oseltamivir (Tamiflu), a drug used for influenza treatment, which is a P-gp substrate (Morimoto et al., 2008).

Upregulation of P-gp may decrease brain concentrations of oseltamivir in this case.

In summary, the VDR can upregulate P-gp expression in kidney and brain of mice independently of FXR. Unequivocally, the induction of Mdr1/P-gp in rats and mice by the

VDR has been translated to observations in human cell lines, and the intestinal P-gp may be further involved in DDIs. VDR activation of MDR1 mRNA and P-gp had been observed in human colonic cell lines (Aiba et al., 2005; Fan et al., 2009; Tachibana et al., 2009), a notion consistent with the existence of VDREs in MDR1 (Saeki et al., 2008). As a result, it

119

is highly likely that 1,25(OH)2D3 treatment in combination with some other drugs would cause significant DDIs.

4.6 APPENDIX

In the equations below, B/P is the blood to plasma partition coefficient; V is the volume of the tissue; C is the concentration; Alumen and Aurine are the amounts of

[3H]digoxin excreted in feces and urine compartment, respectively; Q is the blood flow rate;

QHA and QPV are the blood flow rate of hepatic artery and portal blood; QHV is the sum of

QHA and QPV; KTB is the tissue to blood ratio, assessed as the AUC tissue/blood ratio; fB, fP, fBr, fK, fL, and fI, are the unbound fractions in blood, plasma, brain, kidney, liver, and intestine, respectively; fQ is the proportion of intestinal blood flow perfusing the enterocyte

Br Br region in the small intestine (Liu et al., 2006b); CLin and CLef are the intrinsic influx and efflux in the brain; ka is the absorption rate constant of the intestine; GFR is the glomerular filtration rate; FR is the fraction reabsorbed in the kidney. The mass balance equations are listed below.

In the blood (B) compartment: dC C C C C VB = Q C +Q K +Q L +Q other +Q heart - (Q +Q +Q +Q +Q )C Bdt Br Br,B K K HV K other K heart K Br K HV other heart B TB,K TB,L TB,other TB,heart (A1)

In the brain tissue (Br) compartment: dC VBr = fCCL-fCCLBr Br (A2) Brdt B Br,B in Br Br ef

In the brain blood compartment: dC VBr,B = Q(C- C) - fCLC+Br fCLC Br (A3) Br,Bdt Br B Br,B B in Br,B Br ef Br

120

In the heart compartment:

dCheart C heart Vheart = Q(C heart B - ) (A4) dt KTB,heart

In other tissue compartments: dC C other other (A5) Vother = Q(C other B - ) dt KTB,other

In the kidney compartment:

dCKKPB C f GFRC VK = Q K (C B - ) -  + f K C K CL int,sec,K  1 FR (A6) dt KTB,K  (B/P)

In the above equation, the arterial unbound plasma concentration equals fPCB/(B/P). The filtered and excreted components are reabsorbed according to the fraction reabsorbed, FR.

In the liver compartment, with segregated flows returning from the intestine:

dCLensL C C C VL = Q HA C B + Q PV [f Q +(1-f Q ) ] - Q HV - f L C L CL int,sec,L (A7) dt KTB,I K TB,I K TB,L where QHV is the sum of QHA and QPV.

In the intestinal compartments of the small intestine: Enterocyte region:

dCen C en fQ V I = f Q Q PV (C B - ) - f I C en CL int,sec,I + k a A lumen (A8) dt KTB,I

Serosal region:

dCss C (1-fQI )V = (1- f Q )Q PVB (C - ) (A9) dt KTB,I In the fecal compartment: dA lumen = f C CL + f C CL - k A (A10) dt Ienint,sec,I LL int,sec,H a lumen

121

4.7 ACKNOWLEDGMENTS

This work was supported by the Canadian Institutes for Health Research CIHR,

Grant [Grant MOP89850]. Edwin C.Y. Chow was supported by the University of Toronto

Open Fellowship and the National Sciences and Engineering Research Council of Canada

Alexander Graham Bell Canada Graduate Scholarship (NSERC-CGS), and Matthew R.

Durk was supported by the Canadian Institutes of Health Research (CIHR) Strategic

Training Grant in Biological Therapeutics.

4.8 STATEMENT OF SIGNIFICANCE OF CHAPTER 4

In this chapter, regulation of P-gp by either VDR or FXR was discriminated by the use of the fxr(-/-) mouse model. The study shows that the induction of P-gp in kidney and brain is independent of the presence of FXR, and the event was triggered by VDR activation in those tissues. The increase in P-gp levels in the mouse kidney and brain resulted in pharmacokinetic changes of digoxin, a P-gp substrate, in vivo. The change in P- gp in the rat liver (Chapter 3) was, however, likely due to FXR. The change in plasma, urine, feces and tissue concentrations in 1,25(OH)2D3-treated mice was interpreted using a physiologically-based pharmacokinetic (PBPK) model showing that the intrinsic efflux clearances in the kidney and brain were increased corresponding to VDR protein changes.

This study suggests that VDR activation can cause significant DDIs when vitamin D analogs are taken in combination with other drugs.

In this study, I was involved in the treatment of 1,25(OH)2D3 to mice, the determination of [3H]digoxin, and tissue harvesting, performed mRNA and protein extractions, analyzed mRNA and protein expressions in the intestine, liver, and kidney,

122

tissue extraction of digoxin and its analyses. I was responsible for the PBPK modeling and the design of the differential equations and data fitting. Matthew Durk examined mRNA and protein expressions in the brain; he assisted in tissue harvesting and the tissue extraction of digoxin. Dr. Cummins commented on the experimental design and gave suggestions to the paper.

123

CHAPTER FIVE

5. INHIBITION OF THE SMALL HETERODIMER PARTNER (SHP) BY 1,25-

DIHYDROXYVITAMIN D3-LIGANDED VITAMIN D RECEPTOR (VDR) REMOVED

THE REPRESSION ON CYTOCHROME 7-HYDROXYLASE (CYP7A1) AND

INDUCED CHOLESTEROL LOWERING

Edwin C.Y. Chow1, Lilia Magomedova1, Rucha Patel1, Matthew R. Durk1, Han-Joo

Maeng1,4, Holly Quach1, Kamdi Irondi1, Sayeepriyadarshini P. Anakk 3, Reinhold Vieth2,

David D. Moore3, Carolyn L. Cummins1, and K. Sandy Pang1

1Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, and

2Department of Nutritional Sciences, University of Toronto, Canada

3Department of Molecular and Cellular Biology, Baylor College of Medicine, Houston, Texas, USA

4College of Pharmacy, Inje University, 607 Obang-dong, Gimhae, Gyeongnam 621-749, South Korea

124

5.1 ABSTRACT

CYP7A1, the rate-limiting enzyme in cholesterol metabolism, is repressed by the

FXR-SHP-LRH1 cascade in liver and by intestinal FGF15/FGF19. Treatment with the

VDR ligand, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3; 2.5 µg/kg i.p. q2d for 8 days] to fxr(-/-) and fxr(+/+) mice on a normal or 3 week Western diet resulted in higher Cyp7a1 mRNA/protein expression and microsomal activity over those of vehicle controls, and these were accompanied by reduced SHP in liver without changes in intestinal mFGF15 mRNA expression. The same was observed in mouse primary hepatocytes treated 9 h with 100 nM

1,25(OH)2D3. Significant reduction in plasma and liver cholesterol concentrations occurred in all mice except for the fxr(+/+) mice on the normal diet. In 1,25(OH)2D3-treated shp(-/-) mice prefed with the Western diet, changes in Cyp7a1 were absent, and only plasma cholesterol was slightly reduced. Upon 1,25(OH)2D3 treatment, intestinal FGF15 was elevated and reduced in fxr(-/-)and shp(-/-) mice, respectively, prefed with Western diet.

The luciferase assay and truncation analysis revealed inhibitory effects of VDR on mSHP and hSHP promoters, and EMSA showed direct binding of VDR to the hSHP promoter.

The composite results implicate a role of the VDR in cholesterol lowering via inhibition of

SHP, by removing the repressive effect on Cyp7a1; VDR further increases FGF15 directly but antagonizes intestinal FXR to reduce FGF15, opposing effects that yield a net FGF15 decrease for increasing Cyp7a1 activity with 1,25(OH)2D3 treatment.

5.2 INTRODUCTION

Cholesterol is an important component of cell membrane structure and is a precursor of various steroids, including corticosteroids, vitamin D, steroid hormones, and 125

bile acids. However, excess cholesterol in blood can lead to atherosclerosis and coronary and cerebrovascular diseases. The therapeutic agents that are available to treat high cholesterol levels include the HMG-CoA reductase inhibitors, niacin, hypolipidemic drugs that act by blocking the breakdown of fat in adipose tissue (Brooks et al., 2010), and ezetimibe, which inhibits the Niemann-Pick C1-like 1 transporter in the intestinal absorption of cholesterol (Jia et al., 2010). Cholesteryl ester transfer protein (CETP) inhibitors that raise the high-density lipoprotein cholesterol (HDL-C) also pose as a new form of therapy (Masson, 2009). These inhibitors and bile-acid sequestrants reduce cholesterol synthesis in the liver or the amount of dietary cholesterol that is absorbed in the intestine. However, many of these drugs have serious side effects (Alsheikh-Ali et al., 2007;

Radcliffe and Campbell, 2008).

The regulation of cholesterol metabolism in the liver is paramount in the understanding of cholesterol homeostasis. In mammals, excess cholesterol follows a series of metabolic pathways to form bile acids, whereas a small portion of cholesterol forms steroid hormones (Boggaram et al., 1984; Rezen et al., 2010). Cytochrome P450 7A1

(CYP7A1) catalyzes the first and rate-limiting enzyme reaction in the classical, neutral pathway, whereas the alternate pathway is catalyzed by CYP27A1 (Russell and Setchell,

1992; Javitt, 1994). The promoter of the CYP7A1 gene contains a hexameric repeat of nucleotide sequence (AGGTCA) or the bile acids response element (BARE) that is highly conserved among species (Chiang, 2003). The pathways are tightly regulated at the transcriptional level by bile acids (Stravitz et al., 1993; Agellon and Cheema, 1997; Gupta et al., 2001) and other signaling molecules (Staudinger et al., 2001; Drover and Agellon,

2004; Ma et al., 2011). CYP7A1 is under negative feedback regulation by the farnesoid X 126

receptor (FXR; NR1H4), which can be activated by bile acids such as chenodeoxycholic acid (CDCA) to increase transcription of the short heterodimer partner (SHP, NR0B2)

(Chiang, 2003). SHP then prevents the binding of other transcriptional factors, liver-related homologue-1 (LRH-1 or fetoprotein transcription factor, FTF, NR5A2), a competence factor that binds as a monomer to the response element in the CYP7A1 promoter for its expression (Chiang and Stroup, 1994; Crestani et al., 1998; Chiang et al., 2000; Goodwin et al., 2000), and hepatocyte nuclear factor 4α (HNF-4αNR2A1), which is also important for the upregulation of CYP7A1 (Chiang, 2002; Abrahamsson et al., 2005). In addition

CYP7A1 expression is stimulated by the murine liver X receptor α LXRα,R, an oxysterol-activated transcriptional factor; SHP also can interact with LXRα by repressing its transcriptional activity (Brendel et al., 2002; Schoonjans and Auwerx, 2002) in the modulation of cholesterol absorption, transport and elimination in rodents (Goodwin et al.,

2003). In the intestine, the bile-acid-FXR activation leads to induction of the fibroblast growth factor (FGF15 in rodents; FGF19 in humans), a hormonal signaling molecule that represses hepatic CYP7A1 via the hepatic fibroblast growth factor receptor 4 (FGFR4) which in turn decreases CYP7A1 through activation of the liver c-Jun signaling pathway

(Inagaki et al., 2005).

The VDR (vitamin D receptor, NR1I1) has been examined with respect to its role in bile acid and cholesterol homeostasis (Thummel et al., 2001; Makishima et al., 2002; Chen et al., 2006; Chow et al., 2009; Han and Chiang, 2009; Nishida et al., 2009; Han et al., 2010;

Schmidt et al., 2010). Upon metabolism of vitamin D by the liver and kidney sequentially to form the active hormone ligand, 1α,25-dihydroxyvitamin D3 [1,25(OH)2D3] (Jones et al.,

1998), 1,25(OH)2D3 and various vitamin D analogues are found to bind to the VDR to 127

activate the transcription of VDR target genes (Echchgadda et al., 2004; McCarthy et al.,

2005; Zierold et al., 2006; Fan et al., 2009; Khan et al., 2009b; Chow et al., 2010; Chow et al., 2011b). However, there appear to be multiple pathways in which the VDR exerts its influence on the transcription of genes encoding the CYP7A1 pathway (Wagner et al.,

2010). 1,25(OH)2D3-Liganded VDR has been reported to inhibit/antagonize the CDCA- dependent transactivation of FXR (Honjo et al., 2006), blunt the LXRα-mediated induction of Cyp7a1 mRNA in rat hepatoma cells (Jiang et al., 2006), and repress the transcriptional activity of PPAR (Sakuma et al., 2003). Others have proposed inhibitory mechanisms on

CYP7A1 transcription in human hepatocytes and HepG2 cells, attributing these to ligand- activated VDR blocking HNF-4 activation of CYP7A1 gene (Han and Chiang, 2009; Han et al., 2010). Schmidt et al. (2010) reported on the inhibition of Cyp7a1 mRNA in mice 4 h following a single, high dose of 1,25(OH)2D3, attributing the observed induction of FGF15 and SHP to Cyp7a1 mRNA levels without correlating to Cyp7a1 protein or cholesterol lowering. By contrast, upregulation of Cyp7a1 mRNA was observed with treatment of 1α- hydroxyvitamin D3, a prodrug of 1,25(OH)2D3 in mice (Nishida et al., 2009; Ogura et al.,

2009), and doxercalciferol was reported to decrease the accumulation of triglycerides and cholesterol in murine kidney (Wang et al., 2011a). Thus, the role of the VDR in the regulation of CYP7A1 appears to be unclear.

In this study, we examined the role of the VDR on changes in liver SHP and intestinal FGF15 in wild-type and fxr(-/-) mice after treatment with 1,25(OH)2D3. After noting that VDR elevated Cyp7a1 mRNA and protein expression and microsomal function after 1,25(OH)2D3 treatment that was accompanied by repression of liver SHP, we fed wild- type, and fxr(-/-) and shp(-/-) mice a high fat/high cholesterol diet followed by 1,25(OH)2D3 128

treatment and examined the molecular mechanism of the role of the VDR on the alteration of SHP and by examining the binding of VDR to VDREs in SHP by EMSA.

5.3 METHODS

5.3.1 Materials

1,25(OH)2D3 powder was purchased from Sigma-Aldrich Canada (Mississauga, ON,

Canada). Anti-CYP7A1 (N-17) was purchased from Santa Cruz Biotechnology (Santa Cruz,

CA); and anti-GAPDH (6C5), from Abcam, Cambridge, MA. The fxr(-/-) and shp(-/-) mice were obtained from Dr. F. J. Gonzalez (National Institutes of Health, Bethesda) (Sinal et al.,

2000) and Dr. David D. Moore (Baylor College of Medicine in Texas Medical Center), respectively, and C57BL/6 mice was the control counterpart. Animals were given water and food ad libitum and maintained under a 12:12-h light and dark cycle in accordance to animal protocols approved by the University of Toronto (ON, Canada).

5.3.2 Plasmids

Expression vectors for pCMX, pCMX-hRXRa, pCMX-mLRH, pGEM, pCMX-β- galactosidase, hSHP(569)-luc, hSHP(371)-luc were kind gifts from Dr. David J.

Mangelsdorf (University of Texas Southwestern Medical Center). The mSHP(2kb)-luc was a kind gift of Dr. Li Wang (University of Utah). pEF-mVDR was a gift from Dr. Rommel

G. Tirona (University of Western Ontario, Canada). Human SHP promoter deletion constructs were generated by PCR amplification. The PCR fragments were ligated into the

HindIII and BglII sites of the luciferase reporter pGL3 (Promega) to generate hSHP(238)- luc and hSHP(138)-luc.

129

5.3.3 1,25(OH)2D3 Treatment of Mice

1,25(OH)2D3 powder were dissolved in anhydrous ethanol, and the concentration was spectrophotometrically measured at 265 nm (UV-1700, Shimadzu Scientific

Instruments, MD) before dissolving in sterile corn oil. For examination of the effects of

1,25(OH)2D3 in the presence and absence of FXR, male fxr(-/-) and fxr(+/+) mice (8 to 12 weeks; n = 5 - 10) were given intraperitoneal (i.p.) doses of 0 or 2.5 µg/kg 1,25(OH)2D3 dissolved in sterile corn oil (5 μl/g) every other day for 8 days. The alternate day regimen was chosen due to the lessened hypercalcemia observed in comparison to those given lower doses in consecutive days (Chow et al., 2011b). On the 9th day, mice were anesthetized with ketamine and xylazine by i.p. injection. Systemic and portal blood were taken and spun down at 605 g for 10 min to obtain plasma. Mice were flushed with ice-cold saline from the vena cava, and the intestine, liver, brain, and kidneys were harvested. The ileum was taken at a length of 6 cm anterior to the cecum. The ileal segment was flushed with saline solution containing 1 mM phenylmethylsulfonyl fluoride (PMSF), everted and scraped with a tissue-scraper for the collection of enterocytes (Chow et al., 2009). All enterocytes and tissue samples were snapped frozen in liquid nitrogen.

The effects of 1,25(OH)2D3 on plasma and liver cholesterol concentrations, VDR target genes and other nuclear receptors were examined in C57BL/6 or fxr(+/+), fxr(-/-) and shp(-/-) mice (about 6-8 weeks old; n = 6) that were fed a high fat (42%)/high cholesterol (0.2%) or Western diet (Harlan Teklad , #88137) for a total of 3 weeks. Mice in each vehicle- and treatment group, respectively, were given 0 or 2.5 µg/kg 1,25(OH)2D3 dissolved in sterile corn oil (5 μl/g) every other day for 8 days at the beginning of the 3rd week of a high fat/cholesterol diet pretreatment period. On the 9th day after the first 130

1,25(OH)2D3 treatment, blood and tissue samples were harvested using the procedures described earlier.

5.3.4 Preparation of Subcellular Tissue Fractions

To obtain liver microsomes for Western blot analyses, liver tissues were homogenized in the microsome homogenizing buffer followed by sequential centrifugation

(Chow et al., 2009). The resulting microsomes were suspended in the same microsome homogenizing buffer containing 1% protease inhibitor cocktail. For metabolism studies, the liver tissues were homogenized with similar procedure but without protease inhibitor in the homogenizing buffer. Protein concentration was determined by the Lowry method (Lowry et al., 1951).

5.3.5 Immunostaining

For identification of VDR in murine livers, 25 ml of ice-cold PBS was used to perfuse the mouse via the portal vein, followed by 50 ml of 4% paraformaldehyde prior to postfixation of the liver in 4% paraformaldehyde at 4°C overnight. Livers were embedded in paraffin and 7 µm sections were prepared. Following dewaxing, sections were incubated in 2N hydrochloric acid at 37°C for 30 min and pre-blocked with 5% goat serum in PBS containing 0.1% Tween-20, and incubated with a primary anti-VDR antibody (1:50 v/v) overnight. The pre-block sections were rinsed thrice with 5% goat serum and incubated with the secondary goat anti-rat HRP antibody for 2 h at room temperature, then stained using a metal-enhanced DAB kit (Thermo Scientific, Rockford, IL). Following washing, sections were imaged using a Nikon E1000R fluorescence microscope.

131

5.3.6 Real-Time PCR (qPCR)

The TRIzol extraction method was used to isolate total RNA from enterocytes and liver tissues (Chow et al., 2009). All RNA purities were checked, and 1.5 µg of total RNA was used for the synthesis of cDNA prior to qPCR with SYBR Green detection. Primer specificity (Table 5-1) was checked by BLAST analyses

(http://www.ncbi.nlm.nih.gov/BLAST/). The critical threshold cycle (CT) values of target genes were collected using the ABI Sequence Detection software 1.4 and normalized to cyclophillin for liver samples and to villin for ileum.

5.3.7 Western Blotting

Protein levels of mCyp7a1 were examined by Western blotting. About 50 µg of total protein was separated by 10% SDS-polyacrylamide gels and transferred onto nitrocellulose membranes (Chow et al., 2009). After blocking with 5% (w/v) skim milk in

Tris-buffered saline (pH 7.4) with 0.1% Tween 20 (TBS-T) and washing with TBS-T, the membrane was incubated with the primary antibody solution (2% skim milk) overnight at

4°C, then washed with TBS-T and incubated with a secondary antibody (2% skim milk) for

2 h, followed by washes and probed with chemiluminescence reagents for visualization and quantification of the band intensity by scanning densitometry (NIH Image software; http://rsb.info.nih.gov/nih-image/). Protein loading error was corrected by normalizing the

Cyp7a1 protein band against the protein band of Gapdh.

5.3.8 Cyp7a1 Activity in Microsomes

The method of Hylemon et al. (1989) was used to assay for Cyp7a1 activity in 2 mg of liver microsomal protein upon incubation with exogenously added cholesterol and a

132

NADPH generating system (Chow et al., 2009). The reaction was stopped with addition of ice-cold 20% sodium cholate. The reaction mixture was added the internal standard 7- hydroxycholesterol (7-HCO), incubated with cholesterol oxidase to convert the formed metabolite, 7α-hydroxycholesterol, to 7α-hydroxy-4-cholesten-3-one (7α-HCO), then stopped with ice-cold methanol. The sample was extracted with hexane followed by several centrifugation steps (Chow et al., 2009). The pooled extract was evaporated and reconstituted in the mobile phase (acetonitrile:methanol 70:30 v/v) and injected into the

HPLC. Separation was achieved with an Altex 10 m C-18 reverse phase column (4.6 mm x 250mm) at flow rates varying from 0.7 to 2 ml/min, with detection at 240 nm. The areas of 7α-HCO, obtained from the converted 7-hydroxycholesterol standards (from 0.05 to 2.5 nmole), were normalized to the 7-HCO areas for construction of the calibration curve, which was used to estimate the metabolic activities of Cyp7a1.

5.3.9 Blood Analyses

Total bile acid (portal) and cholesterol (systemic) concentrations were determined using the Total Bile Acids Assay (Diazyme, Cat#DZ042A-K) and Cholesterol (Wako

Diagnostics, Cat#439-17501) Kits, and enzyme (ALT) leakage in systemic plasma with the

ALT kit (BioQuant, Cat#BQ004A-CR).

5.3.10 Liver Cholesterol

Lipids were extracted from liver homogenized tissue (about 0.2 g) in chloroform: methanol (2:1, v/v) in a modified assay based on published methods (Folch et al., 1957;

Cho, 1983), and the bottom organic phase was harvested after centrifugation. The bottom phase was then collected consecutively after each centrifugation from extracts washed 133

once in 50 mM NaCl and twice in 0.36 M CaCl2/methanol. The collected organic phase was brought up to 5 ml with chloroform. In separate tubes in duplicate, 10 μl of chloroform:

Triton X-100 (1:1, v/v) was added, followed by addition of 100 μl aliquot of the extract or standards, and then were subsequently air dried overnight. Colorimetric enzymatic assays were performed using commercial reagents following manufacture protocols (Infinity,

Thermo Scientific, Cat#TR13421).

5.3.11 Mouse Primary Hepatocyte Isolation

Mouse primary hepatocytes were isolated by collagenase perfusion and purified by centrifugation (Horton et al., 1999). Freshly prepared hepatocytes were seeded (density of

0.5x106 cells per well) onto type I collagen coated 6-well plates in attachment media

(William’s E media, 10% charcoal stripped FBS, 1× penicillin/streptomycin and 10 nM insulin). Media was changed 3 h after plating, and experiments were performed on the following day. Ligands were added to cells in M199 media without FBS, and cells were harvested 9 h after treatment for RNA extraction.

5.3.12 Cell Culture and Transfection Assays

HEK293 cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum. Transfection assays were performed in media containing 10%-charcoal-stripped FBS using calcium phosphate in a 96-well plate. The following amount of plasmid DNA was used: 50 ng of reporter, 20 ng of pCMX-β- galactosidase, 15 ng of receptor, 15 ng pCMX-mLRH, and an appropriate amount of the pGEM filler plasmid for a total of 150 ng/well. Ligands were added at 6–8 h post- transfection in media containing 10% dextran-charcoal-stripped FBS. Cells were harvested

134

14–16 h later and assayed for luciferase and β-galactosidase activity. Luciferase values were normalized for transfection efficiency using β-galactosidase and expressed as RLU of triplicate assays (mean ± SD).

5.3.13 Preparation of Nuclear Protein Extracts

HEK293 cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum. Cells were transfected with 150 ng of mVDR, mRXR or CMX plasmid DNA using calcium phosphate in a 10 mm plate. At 30 h post- transfection, nuclear protein extracts were prepared using the NE-PER reagent (Pierce,

Rockford, IL).

5.3.14 Electrophoretic Mobility Shift Assay (EMSA)

The following oligonucleotides (along with their complementary sequence) were purchased from Sigma Aldrich: rOC-VDRE(+), 5′-

GCACTGGGTGAATGAGGACATTAC-3′, hSHP(-549)-VDRE (+), 5′-

GGCAAAGTCCTCCCAGCCCCCAGGG-3′, hSHP(-283)-VDRE (+), 5′-

GTTAATGACCTTGTTTATCCACTTG-3′, hSHP(-250)-VDRE (+), 5′-

GATAAGGGGCAGCTGAGTGAGCGGC-3′, hSHP(-169)-VDRE (+), 5′-

CGTGGGGTTCCCAATGCCCCCTCCC-3’. The oligos were biotinylated using the biotin

3′ end DNA labeling kit from Pierce, following the manufacturer’s instructions. The oligonucleotides were mixed at equimolar concentrations, denatured by heating to 95 °C for

5 min, and allowed to anneal by slow cooling at room temperature. DNA binding reactions were established using 2.5 µl of NE-PER nuclear extracts in 1X binding buffer (10 mM

Tris, 50 mM KCl, 1 mM DTT, 1 mM EDTA, 5 % glycerol, 1 µg BSA and 1 µg of poly(dI-

135

dC); pH 7.5). After a 20 min pre-incubation, 50 fmol of biotinylated DNA probe was added to each reaction and incubated for an additional 30 min on ice. Reactions were loaded onto a 6% polyacrylamide gel, transferred to Biodyne B membrane (Pierce) and crosslinked.

Detection was carried out using the LightShift Chemiluminescent EMSA Kit (Pierce) according to the manufacturer’s instructions.

5.3.15 Statistics

For comparison of results between two groups, the unpaired Student’s t-test was performed. A P value of less than 0.05 was set as the level of significance. For mRNA and protein analyses of fxr(+/+) and fxr(-/-) mice, the fxr(+/+) vehicle-treated sample was set as the control (value set as unity), and was used for comparison against both fxr(+/+) and fxr(-/-) treatment groups.

5.4 RESULTS

5.4.1 VDR Protein in Mouse Liver

VDR protein in livers of fxr(+/+) and fxr(+/+) mice were 71% and 65%, respectively, values lower than that of ileal VDR. There was no significant difference of

VDR in the kidney protein compared to that for the ileum in both mouse types (Fig. 5-1A).

Immunostaining analysis revealed that the VDR was present in the nucleus of mouse liver hepatocytes (Fig. 5-1B).

136

Table 5-1 Mouse Primer Sequences

Gene Bank Number Forward (5’→ 3’ Sequence) Reverse (5’→ 3’ Sequence)

FXR NM_009108 CGGAACAGAAACCTTGTTTCG TTGCCACATAAATATTCATTGAGATT

SHP NM_011850 CAGCGCTGCCTGGAGTCT AGGATCGTGCCCTTCAGGTA

LRH-1 NM_001159769 CCCTGCTGGACTACACGGTTT CGGGTAGCCGAAGAAGTAGCT

LXRα NM_013839 GGATAGGGTTGGAGTCAGCA GGAGCGCCTGTTACACTGTT

HNF-4α NM_008261 CCAAGAGGTCCATGGTGTTTAAG GTGCCGAGGGACGATGTAGT

FGF15 NM_008003 ACGGGCTGATTCGCTACTC TGTAGCCTAAACAGTCCATTTCCT

Cyp7a1 NM_007824 AGCAACTAAACAACCTGCCAGTACTA GTCCGGATATTCAAGGATGCA

Asbt NM_011388 GATAGATGGCGACATGGACCTC CAATCGTTCCCGAGTCAACC

Villin NM_009509 TCCTGGCTATCCACAAGACC CTCTCGTTGCCTTGAACCTC

Cyclophillin X58990 GGAGATGGCACAGGAGGAA GCCCGTAGTGCTTCAGCTT

137

5.4.2 1,25(OH)2D3 Increases Hepatic Cyp7a1, Decreases Hepatic SHP and Lowers

Cholesterol In Vivo

The fxr(-/-) mice chosen for this study were used to investigate the direct effects of

1,25(OH)2D3 in vivo in the absence of FXR. Basal levels of plasma bile acid and cholesterol were much higher in control fxr(-/-) mice compared to those for fxr(+/+) mice (P < .05), whereas that of tissue cholesterol was only marginally higher (Fig. 5-2A). Upon treatment with 1,25(OH)2D3, both plasma and liver cholesterol levels were significantly reduced in fxr(-/-) but not in fxr(+/+) mice. The fxr(-/-) mice also showed higher hepatic basal Cyp7a1 mRNA and protein expression and higher microsomal activity than fxr(+/+) mice, and correspondingly, higher plasma concentrations of bile acids (Fig. 5-2B). Remarkably,

1,25(OH)2D3 treatment resulted in upregulation of hepatic Cyp7a1 mRNA and protein expression and microsomal activity in both fxr(-/-) and fxr(+/+) mice. The mRNA expression of hepatic SHP, normally under control of FXR, was significantly decreased in both fxr(+/+) and fxr(-/-) treated mice (Fig. 5-3A). No significant change was observed for other genes or nuclear receptors regulating Cyp7a1 in liver and intestine (Figs. 5-3A and 5-

3B).

5.4.3 Correlation Between Cyp7a1 and SHP But Not FGF15

When a correlation was determined for the data for Cyp7a1 vs. SHP and FGF15 (Fig.

5-4), an inverse and significant correlation was found between Cyp7a1 mRNA, protein, or activity vs. the corresponding mRNA levels of SHP and not FGF15 for each control/treated fxr(+/+) and fxr(-/-) mouse. These results suggest that the VDR is able to repress SHP directly and independently of FXR in removing the inhibition on Cyp7A1.

138

(A) (B)

fxr(+/+) fxr(-/-)

1 2 3 4 5 6 (-) (+)

2.5 fxr (+/+) 2.0 fxr (-/-)

1.5

1.0 #

0.5 *

Relative VDR Protein Distribuion 0.00 Ileum Liver Kidney Figure 5-1 Tissue distribution and localization of VDR. (A) VDR protein expression in ileum (lane 1 and 4), liver (lane 2 and 5) and kidney (land 3 and 6) of male fxr(+/+) (lane 1,2, and 3) and fxr(-/-) (lane 4, 5 and 6) mice. (-) and (+) represent negative and positive control of VDR. TheVDR protein band was detected at 50 kDa. The symbols * and # denote significant differences (P < .05) between the ileum of its respective species, respectively. Data represent the mean ± SEM (n = 3-4). (B) Localization of VDR in hepatocytes of C57BL/6 mice. The VDR, as shown by the arrows, was localized within the nuclei of hepatocytes.

139

(A) (B)

M) Liver  120 Portal Serum 12 Vehicle control 100 10 Vehicle control 1,25(OH) D 2 3 1,25(OH)2D3 80 † 8 † 60 6 * 40 4 * 20 2

0 0 Relative Cyp7a1 mRNA Expression Serum Bile Acids Concentration ( fxr(+/+) fxr(-/-) fxr(+/+) fxr(-/-)

300 Plasma 12 Liver Vehicle control Vehicle control * 250 1,25(OH)2D3 10 ††† ** 1,25(OH)2D3 200 8 †

150 6

4 100 *** 50 2

Cholesterol Concentration (mg/dl) 0 0 fxr(+/+) fxr(-/-) Protein Expression Cyp7a1 Relative fxr(+/+) fxr(-/-)

10 Liver 2.0 Liver Vehicle control ** Vehicle control 1,25(OH) D 8 2 3 1,25(OH)2D3 1.5 ††† P = 0.053 * 6 1.0 4 ** 0.5 2 Hepatic Cyp7a1 Activity Tissue Cholesterol (mg/g)

(nmol / h / mg microsome protein) 0.00 Figure 5-20 Effect of 1,25(OH) D treatment on (A) serum bile acids and plasma and liver cholesterol 2 3 fxr(+/+) fxr(-/-) and (B)fxr(+/+) Cyp7a1 mRNA fxr(-/-) and protein expressions and activity in fxr(+/+) and fxr(-/-) mice. Plasma and tissue cholesterol were reduced after 1,25(OH)2D3 treatment (2.5 µg/kg or 6 nmol/kg 1,25(OH)2D3 every other day for 8 days ip), whereas these remained unchanged for fxr(+/+) mice. Higher basal expression of Cyp7a1 mRNA, protein, and catalytic activity existed in fxr(-/-) mice compared to fxr(+/+) mice due to absence of FXR; the expression of Cyp7a1 mRNA, protein, and catalytic activity were all increased after 1,25(OH)2D3 treatment. The symbols † and * denote significant differences between the two controls, and

(A) between the treated vs. vehicle control within the fxr(+/+) and fxr(-/-) mice, respectively. Data represent the mean ± SEM (n = 4-5). 140

2.0 Liver 2.0 Liver Vehicle control Vehicle control 1,25(OH)2D3 1.5 1,25(OH)2D3 1.5 ††† *** 1.0 1.0

0.5 0.5 ** Relative FXR mRNA Expression mRNA FXR Relative 0.00 RelativeSHP mRNA Expression 0.00 fxr(+/+) fxr(-/-) fxr(+/+) fxr(-/-)

2.0 Liver 2.0 Liver 2.0 Liver Vehicle control Vehicle control Vehicle control 1,25(OH) D 1,25(OH) D 1,25(OH) D 1.5 2 3 1.5 2 3 1.5 2 3

1.0 1.0 Expression mRNA 1.0 mRNA Expression mRNA   0.5 0.5 0.5

0 LXR Relative 0 0

Relative LRH-1 mRNA Expression mRNA LRH-1 Relative 0.0 0.0 0.0

HNF-4 Relative fxr(+/+) fxr(-/-) fxr(+/+) fxr(-/-) fxr(+/+) fxr(-/-)

2.0 Ileum 2.0 Ileum Vehicle control Vehicle control 1,25(OH)2D3 1,25(OH) D 1.5 2 3 1.5 ††

1.0 1.0

0.5 0.5 Relative FXR mRNA Expression FXR Relative 0.00 SHP mRNA Expression Relative 0.00 fxr(+/+) fxr(-/-) fxr(+/+) fxr(-/-) 2.0 Ileum 2.0 Ileum 2.0 Ileum Vehicle control Vehicle control Vehicle control 1,25(OH) D 1,25(OH)2D3 1,25(OH) D 1.5 2 3 1.5 † 1.5 2 3 * * 1.0 1.0 1.0

0.5 0.5 0.5

0 0 Asbt Expression mRNA Relative 0

Relative LRH-1 mRNA LRH-1 Expression Relative 0.0 0.0 0.0 Relative FGF15 mRNA Expression Relative fxr(+/+) fxr(-/-) fxr(+/+) fxr(-/-) fxr(+/+) fxr(-/-)

Figure 5-3 Changes in (A) hepatic FXR, SHP, LRH-1, LXR, and HNF-4 and (B) ileal FXR, SHP, LRH- 1, FGF15, and Asbt mRNA expressions in both fxr(+/+) and fxr(-/-) mice. The symbols † and * denote significant differences between the two controls, and between the treated vs. vehicle control within the wild-type or null mice, respectively. Data represent the mean ± SEM (n = 4-5).

141

.4.4 1,25(OH)2D3 Lowers Cholesterol in C57BL/6 or fxr(+/+), fxr(-/-), and shp(-/-) Mice Fed a High Fat/High Cholesterol Diet

A high fat/high cholesterol diet (Western diet) was given to male C57BL/6, fxr(-/-), and shp(-/-) mice for 3 weeks to elevate plasma and liver cholesterol and mimic the hypercholesterolemia condition (Fig. 5-5). The Western diet failed to increase portal bile acid or hepatic Cyp7a1 mRNA and protein levels in all genotypes. The Western diet significantly increased SHP mRNA expressions in the intestine and liver and intestinal

FGF15 level in mice (Fig. 5-6). 1,25(OH)2D3 treatment to these high cholesterol fed mice decreased plasma (27%) and liver (31%) cholesterol concentrations and increased the expression of Cyp7a1 mRNA and protein (Fig. 5-5). A significant decrease in liver SHP mRNA expression and a decrease trend in intestinal SHP and FGF15 levels were observed with 1,25(OH)2D3 treatment to the Western diet fed mice.

For fxr(-/-) mice, a 33% decrease in plasma cholesterol was observed in 1,25(OH)2D3 treated fxr(-/-) mice prefed a Western diet, and an increase in Cyp7a1 mRNA and protein expression (Fig. 5-5). However, the Western diet did not increase the basal level of SHP mRNA expression in the intestine and liver, nor the intestinal FGF15 mRNA. There was also a significant decrease in liver SHP mRNA expression in the Western diet fed fxr(-/-)

1,25(OH)2D3- treated mice, but increased intestinal SHP and FGF15 levels were observed, unlike results for the wild-type mice (Fig. 5-6).

Similar to wild-type mice, 1,25(OH)2D3 treated shp(-/-) mice prefed a Western diet showed a 33% decrease in plasma cholesterol (Fig. 5-5A), though no change in Cyp7a1 mRNA and protein expressions (Fig. 5-5B). The Western diet increased the basal level of

142

intestinal FGF15 mRNA (Fig. 5-6). Interestingly, 1,25(OH)2D3 administrated to the Western diet fed shp(-/-) mice decreased intestinal FGF15.

Small changes in LRH-1, LXRα, FXR, and VDR in the ileum and liver were observed in all genotypes with 1,25(OH)2D3 treatment (Fig. 5-6). The composite results suggest that the VDR is capable of reducing cholesterol via repression of SHP and/or

FGF15 to induce the level of Cyp7a1.

5.4.5 1,25(OH)2D3 Increases Cyp7a1 mRNA and Inhibits SHP Levels in Mouse Primary Hepatocytes

For examination of the direct VDR effect in mouse liver, we isolated and incubated mouse primary hepatocytes with 1,25(OH)2D3 for 9 h. The mRNA expression of Cyp24, a

VDR target gene, was induced 19-fold with 1,25(OH)2D3. 1,25(OH)2D3 also significantly downregulated SHP mRNA expression (35%) and increased Cyp7a1 by 3-fold (Fig. 5-7).

5.4.6 VDR Activation Strongly Represses Mouse and Human SHP Promoter

Activity

To establish whether modulation of SHP occurred directly at the level of the SHP promoter, we performed promoter reporter assays with the mouse (-2 kb) and human (-0.5 kb) SHP proximal promoters. As previously reported (Honjo et al., 2006), the FXR ligand,

CDCA, significantly increased SHP promoter activity when the hFXR was added (Figs. 5-

8A and 5-8B). Furthermore, the basal transcriptional activity of the SHP promoter was dramatically increased with the co-transfection of the known competence factor LRH-1 (Lu et al., 2000). In contrast, addition of 1,25(OH)2D3 strongly repressed the SHP promoter.

The repression of the mouse SHP promoter required the addition of VDR and was more

143

prominent in the presence of LRH-1 (Fig. 5-8A). In contrast, repression of the hSHP promoter by VDR was significant even in the absence of LRH-1, and was independent of

FXR activation by CDCA (Fig. 5-8B). Notably for mouse and human SHP promoters, the effects of 1,25(OH)2D3 dominated, and repression of the SHP-luc activity was observed when both FXR and VDR ligands were given together (Figs. 5-8A and 5-8B).

5.4.6.1 Binding to the VDREs of SHP

Sequence analysis of the human SHP promoter revealed the presence of four putative VDREs (direct-repeat 3 sequences) located within the first 550 bp of the proximal promoter. To identify the region on the human SHP promoter involved in repression by

VDR, promoter truncations were generated and assayed for the 1,25(OH)2D3-mediated repression (Fig. 5-8C). The ability of 1,25(OH)2D3 to repress the hSHP(-258)-luciferase reporter was significantly dampened compared to that of hSHP(-569)-luc, and repression was abolished in the shortest construct tested hSHP(-138)-luc.

We then examined whether any of these sequences could compete with the known interaction of VDR/RXR with the vitamin-D receptor response element from the rat osteocalcin gene (rOC-VDRE). As expected, the protein-DNA complex formed with labelled rOC-VDRE was dependent on the presence of both VDR and RXR proteins, and was not observed when a 500-fold excess of unlabeled rOC-VDRE oligonucleotide was included (Fig. 5-8D). The four potential VDREs that were identified in the proximal human

SHP promoter (-549,-283,-250,-169), only excess unlabeled hSHP-VDRE(-283) oligo abolished binding of the protein complex to the rOC-VDRE (Fig. 8D) whereas hSHP (-

169) oligos partially competed off the VDR/RXR binding to the rOC-VDRE. Consistent

144

with the truncation mutant analyses, these data suggest that the hSHP (-283) and hSHP (-

169) VDRE sites may be important for the 1,25(OH)2D3 mediated-repression of the human

SHP promoter.

5.4.6.2 EMSA

To test for the direct binding of the VDR/RXR complex to these putative VDREs,

EMSA experiments were carried out with biotinylated oligonucleotides of the hSHP sites at

-283 and -169. A visible protein-DNA complex was formed with hSHP(-283)-VDRE, consistent with direct binding of VDR/RXR to this site (Fig. 5-8E). Addition of 1000-fold excess of unlabeled hSHP(-283)-VDRE competitor abolished this interaction. No binding was observed for the putative site at -169 (data not shown). Taken together, we propose that the 1,25(OH)2D3 mediated suppression of SHP gene expression occurs through binding of

VDR to a DR3 response element of the SHP gene promoter.

145

(A) (B)

5.0 5.0 y = -0.02x + 2.73 y = -4.32x + 6.16 2 2 R = 0 R = 0.87 4.0 4.0 P = 0.0024 P = 0.988 Vehicle control 1,25(OH)2D3 3.0 3.0

2.0 2.0

1.0 Vehicle control 1.0 1,25(OH)2D3 0.00 0.00 Relative Cyp7a1 mRNA Expression mRNA Cyp7a1 Relative Relative Cyp7a1 Cyp7a1 mRNA mRNA Expression Expression 0.00 0.5 1.0 1.5 0.00 0.5 1.0 1.5 Relative SHP mRNA Expression Relative FGF-15 mRNA Expression

1.0 1.0 y = -0.14x + 0.61 Vehicle control R2 = 0.05 y = -0.80x + 1.13 1,25(OH)2D3 R2 = 0.84 0.8 0.8 P = 0.0035 0.6 0.6 P = 0.599

0.4 0.4

Vehicle control 0.2 0.2 1,25(OH)2D3 0.00 0.00 Relative Cyp7a1 Protein Expression Expression Protein Protein Cyp7a1 Relative Cyp7a1 Relative Cyp7a1 Protein Expression 0.00 0.5 1.0 1.5 0.00 0.5 1.0 1.5 Relative SHP mRNA Expression Relative FGF-15 mRNA Expression

0.6 0.6 y = 0.023x + 0.35 y = -0.40x + 0.67 2 R2 = 0.76 R = 0.01 P = 0.0097 P = 0.861 0.4 0.4

0.2 0.2 Vehicle control Vehicle control Liver Cyp7a1 Activity Activity Cyp7a1 Liver Liver Cyp7a1 Activity Activity Activity Cyp7a1 Cyp7a1 Liver 1,25(OH)2D3 1,25(OH)2D3

0 protein) microsome mg h / / (nmol 0.00 (nmol / h / mg microsome protein) microsome mg h / / (nmol (nmol / h / mg microsome protein) microsome / h mg (nmol 0.0 0.00 0.5 1.0 1.5 0.00 0.5 1.0 1.5

Relative SHP mRNA Expression Relative FGF-15 mRNA Expression

Figure 5-4 Correlation between murine Cyp7a1 mRNA, protein, and catalytic activity vs. SHP mRNA (A, left column) and FGF-15 mRNA (B, right column) in fxr(+/+) mice that were treated with 1,25(OH)2D3. Note the significant correlation between Cyp7a1 parameters and SHP mRNA and not FGF-15 mRNA. Data represent the mean ± SEM (n = 4-5).

146

(A) (B)

Portal Serum

M) Normal Diet Control 8 Liver

 200 High Cholesterol Diet Control Normal Diet Control High Cholesterol Diet Control High Cholesterol Diet 1,25(OH)2D3 High Cholesterol Diet 1,25(OH) D 150 6 2 3

P = 0.056 * 100 4

50 2

0 0 Relative Cyp7a1 mRNA Expression Cyp7a1 Relative Serum Acids Bile Concentration ( Wild Typefxr(-/-) shp(-/-) Wild Typefxr(-/-) fxr(-/-)shp(-/-) shp (-/-)

Plasma 400 Normal Diet Control 8 Liver High Cholesterol Diet Control Normal Diet Control High Cholesterol Diet 1,25(OH) D 2 3 High Cholesterol Diet Control 300 6 High Cholesterol Diet 1,25(OH) D † * 2 3 ††† ** ††† 200 ** 4 * * 100 2

Cholesterol Concentration (mg/dl) Cholesterol Concentration 0 0 Relative Cyp7a1 Protein Expression Protein Cyp7a1 Relative Wild Type fxr(-/-)fxr(-/-) shp(-/-)shp(-/-) Wild Type fxr(-/-)fxr(-/-) shp(-/-) Liver 30 Normal Diet Control High Cholesterol Diet Control 25 High Cholesterol Diet 1,25(OH)2D3 P = 0.05 20 15 *** ††† 10

Tissue Concentration (mg/dl) 0 Wild Type shp(-/-)shp(-/-)

Figure 5-5 Effects of 1,25(OH)2D3 treatment on (A) serum bile acids and plasma and liver cholesterol as well as (B) Cyp7a1 mRNA and protein expressions in wild-type, fxr(-/-) and shp(-/-) mice prefed with a high fat/high cholesterol diet (n = 4-6) for 3 weeks. After commencing treatment of 1,25(OH)2D3 at 2.5 µg/kg or 6 nmol/kg ip every other day for 8 days at the beginning of the 3rd week and harvesting on the 9th day of treatment, decreased plasma and liver cholesterol levels and increased Cyp7a1 mRNA and protein expressions were observed. The symbols † and * denote significant differences between the two controls, and between the treated vs. vehicle control, respectively. Data represent the mean ± SEM (n = 4-6).147

(A) (B)

40 Wild Type Ileum 6 Wild Type Liver Normal Diet Control Normal Diet Control High Cholesterol Diet Control 5 High Cholesterol Diet Control High Cholesterol Diet 1,25(OH)2D3 30 High Cholesterol Diet 1,25(OH)2D3 4 †† ††** 20 3

2 † †† †† †† 10 †† † 1 * Relative mRNA Expression * Relative mRNA Expression 0 0 FXR SHP VDR LRH-1 FGF15 Asbt FXRSHPVDRLXRLRH-1HNF-4

40 fxr(-/-) Ileum 6 fxr(-/-) Liver Normal Diet Control * High Cholesterol Diet Control Normal Diet Control 5 High Cholesterol Diet Control High Cholesterol Diet 1,25(OH)2D3 30 High Cholesterol Diet 1,25(OH)2D3 4 20 *** 3 ** 2 ** ** 10 * 1 Relative mRNA Expression Relative mRNA Expression 0 0 FXR SHP VDR LRH-1 FGF15 Asbt FXRSHPVDRLXRLRH-1HNF-4

40 shp(-/-) Ileum 6 shp(-/-) Liver Normal Diet Control Normal Diet Control 5 High Cholesterol Diet Control High Cholesterol Diet Control 30 High Cholesterol Diet 1,25(OH) D High Cholesterol Diet 1,25(OH) D 2 3 2 3 4

20 3

††** 2 10 †† † * 1 Relative mRNA Expression Relative mRNA Expression 0 0 FXR SHP VDR LRH-1 FGF15 Asbt FXRSHPVDRLXRLRH-1HNF-4

Figure 5-6 Changes in (A) hepatic FXR, SHP, LRH-1, LXR, and HNF-4 and (B) ileal FXR, SHP, LRH-1, FGF15, and Asbt mRNA expressions in wild-type, fxr(-/-) and shp(-/-) mice prefed with a high fat/high cholesterol diet (n = 4-6) for 3 weeks. The symbols † and * denote significant differences between the two controls, and between the treated vs. vehicle control, respectively. Data represent the mean ± SEM (n = 4-6). 148

30 * 20 Vehicle control 1,25(OH) D 2 3 * 4

1 * Relative mRNA Expression mRNA Relative 0 FXR SHP VDR LXR LRH-1 HNF-4  Cyp7a1 Cyp24

Figure 5-7 Gene expression changes in mouse primary hepatocytes treated with 100 nM 1,25(OH)2D3. Hepatocytes were treated with vehicle (Veh or vehicle, 0.1% EtOH) or 100 nM 1,25(OH)2D3 for 9 h in culture. Cells were harvested for RNA and gene expression analyzed by qPCR. The “*” denotes significant difference (P < .05) between treated vs. vehicle control. Data represent the mean ± SD (n=3).

149

Figure 5-8 1,25(OH)2D3 suppresses SHP expression via direct binding to a DR3 VDRE located within the proximal SHP promoter. (A and B), Reporter analysis of the mSHP and hSHP promoters. HEK293 cells were transiently transfected with either mSHP (-2 kb)-luciferase reporter (A) or hSHP (-569 bp)-luciferase reporter (B), in the presence or absence of mLRH-1, hFXR, hRXR and mVDR. After 6-8 h, cells were treated with vehicle (EtOH), 50 μM CDCA, 0.5 nM 1,25(OH)2D3 or 50 μM CDCA + 0.5 nM 1,25(OH)2D3. Data represent the mean ± SD (n=3). (C) Deletion analysis of the hSHP promoter. HEK293 cells were transiently transfected with the indicated hSHP promoter luciferase constructs in the presence of mLRH-1, hRXR and mVDR. Data represent the mean ± SD (n=3) for cells were treated with vehicle (EtOH) or 0.5 nM 1,25(OH)2D3. The boxes represent potential VDREs. (D and E) show the specificity of VDR/RXR binding to the hSHP VDRE by EMSA in HEK293 nuclear extracts. (D) VDR/RXR heterodimers were incubated with 40 nM rOC- VDRE biotin labelled probe. Where indicated, an unlabeled oligonucleotide competitor was also added to the binding reactions at a concentration equal to 500-fold that of the probe, except for lane 3, where the competitor concentration was 100-fold that of the probe. (E) VDR and RXR nuclear extracts (alone or in combination) were incubated with 40 nM of biotin-labelled rOC-VDRE or hSHP(-283)-VDRE. Where indicated, the matching unlabeled oligonucleotide competitor was also added to the binding reactions at a concentration150 equal to 1000-fold that of the probe.

5.5 DISCUSSION

This study has unequivocally demonstrated that 1,25(OH)2D3 induced hepatic

Cyp7a1 in mice. Among the mechanisms examined, it was found that decreased hepatic

SHP and a second, though minor, mechanism involving intestinal FGF15 might have resulted in the lowering of both plasma and liver cholesterol. Although various studies, including the present data (Fig. 5-1), have shown that the VDR is relatively low in liver compared to other major VDR organs such as the kidney and intestine (Gascon-Barré et al.,

2003; Chow et al., 2010), VDR is clearly identified to exist in mouse hepatocytes to elicit the direct inhibitory action on SHP, which leads to the derepression of Cyp7a1. The presence of VDR in mouse hepatocytes responded to 1,25(OH)2D3 treatment resulted in reduction of SHP in the induction of Cyp7a1 and lowering of cholesterol, strongly suggests a role of the VDR in hypercholesterolemia.

Species differences has attributed to the different 1,25(OH)2D3 responses observed in the regulation of rodent Cyp7a1 and human CYP7A1. In mouse and human, the presence of LRH-1 in the ASBT promoter was found to regulate ASBT levels in the ileum (Chen et al., 2003). Increased bile acid concentrations in the enterocytes triggered a negative feedback mechanism where the induction of SHP by FXR inhibited the binding of LRH-1 to the Asbt gene and negatively regulated ASBT. However, absence of the cis-acting element in LRH-1 in the rat Asbt promoter nullified the negative feedback inhibition of

Asbt. Hence, earlier rat studies have shown that 1,25(OH)2D3 treatment led to induced Asbt in rat ileum and increased bile acid absorption that further triggered secondary FXR effects in the liver to downregulate Cyp7a1 (Chow et al., 2009). In contrast, 1,25(OH)2D3 administration to mice did not result in significant increases in bile acids nor elicit 151

secondary FXR effects in the liver, and repression of SHP and inductive Cyp7a1 effects were observed in both wild-type and fxr(-/-) mice (Figs. 5-2 and 5-3). Similar studies found that 1(OH)D3, a vitamin D analog, when given to mice, increased Cyp7a1 mRNA level and decreased SHP mRNA expression in liver (Nishida et al., 2009; Ogura et al., 2009).

Our observations differed from those of Schmidt et al. (2010) who administered high doses of 50 and 75 µg/kg of 1,25(OH)2D3 (20 to 30-fold higher than our study) to mice and found opposite changes, that is increased FGF15 mRNA at 4 h. The discrepancy could be explained by the fact that our study was performed under steady-state, with low doses of 1,25(OH)2D3 given on a protracted schedule, a regimen that may have eliminated large changes and overt toxicity. The regulation of Cyp7a1 and FGF15 in mice after

1,25(OH)2D3 administration could undoubtedly be affected by the dose and dosing intervals.

We substantiate our present findings on Cyp7a1 induction by the VDR by showing that the induction in Cyp7a1 paralleled those for 1,25(OH)2D3 concentrations in the mouse liver

(Chapter 6; Fig. 6-4).

A dual inhibitory mechanism can be inferred for cholesterol lowering due to the

VDR: (i) the inhibition of hepatic SHP and (ii) inhibition of intestinal FGF15 by the VDR

(Fig. 5-2B, 5-3A, 5-6A and 5-6B). Our studies show that activated VDR decreases liver

SHP in vivo (Figs. 5-2 and 5-5) and in mouse hepatocytes (Fig. 5-7), and was bound to mouse and human SHP promoters to decrease SHP activity directly (Fig. 5-8). However,

Chiang’s group reported that VDR activation inhibited human CYP7A1 by genomic effects due to interaction with HNF-4α and the CYP7A1 promoter, and by non-genomic effects through the activation of the ERK pathway (Han and Chiang, 2009; Han et al., 2010).

Studies have also shown that VDR activation could either inhibit or activate the ERK 152

pathway (Wu et al., 2007; Han et al., 2010). Certainly, data from in vitro systems need to be carefully examined and properly interpreted. It could be argued that in vitro data and assays regarding SHP should be interpreted with caution because SHP has a short half life

(<30 min) due to rapid proteasomal degradation and it is plausible that activation of the

ERK pathway could reduce this degradation process (Miao et al., 2009). In contrast, we have provided strong evidence in mice in vivo and in primary mouse hepatocytes (Figs. 5-2 to 5-7) that VDR induces Cyp7a1 via SHP repression. There exists further evidence that

CYP7A1 mRNA and protein levels are increased in human hepatocytes incubated with 100 nM 1,25(OH)2D3 between 12 to 24 h (unpublished data, Fan and Pang). Our in vitro data further reveal direct binding of VDR/RXR to the hSHP promoter (Figs. 5-8D and 5-8E), and confirmed inhibitory effects of VDR activation towards mSHP and hSHP activities

(Figs. 5-8A and 5-8B). The inhibition of VDR on SHP was reversed by the removal of the

VDREs in hSHP promoter (Fig. 5-8C).

In addition to the direct inhibition of hepatic SHP expression by VDR, VDR may also play a role in intestinal FGF15 expression. In the absence of FXR, 1,25(OH)2D3 treatment to fxr(-/-) mice prefed with the Western diet markedly increased intestinal FGF15 expression (Fig. 5-6A), a finding similar to observations of Schmidt et al. (2010). However, the trend was opposite to those observed for the wild-type and shp(-/-) mice fed the

Western diet that were treated with 1,25(OH)2D3 (Fig. 5-6A). We reconciled the difference in observation along the lines of Honjo et al. (2006), who had previously observed VDR activation in the inhibition FXR and downregulation of SHP. There could exist opposing effects from VDR on intestinal FGF15, since VDR is able to positively upregulate intestinal FGF15 as well as inhibit intestinal FXR, which normally upregulates FGF15. In 153

animal models where the intestinal FXR level is high, the VDR inhibitory effect on FXR may predominate over the VDR stimulatory effect on FGF15, causing a net decrease of

FGF15, since the present data showed that the inhibitory role of the VDR towards FXR and

SHP lowering in liver is the major mechanism of Cyp7a1 induction. Moreover, for mice prefed the Western diet, there may be an increase in the bile acid pool size (Hartman et al.,

2009; Katona et al., 2009), as evidenced by the increase in mRNA of intestinal SHP and

FGF15 and hepatic SHP, all FXR-target genes (Fig. 5-6).

Vitamin D and its analogs have been shown to exert beneficial effects in liver. For an example, a study has suggested that administration of 1,25(OH)2D3 to diabetic rats caused hyperglycaemia evoked oxidative stress to increase LDL-cholesterol and liver injury

(increase in ALT) (Hamden et al., 2009), and was able to reduce cholesterol, triglycerides, and ALT (Hamden et al., 2009). Makishima’s group showed that vitamin D analogues can increase Cyp7a1 and reduce inflammatory factors in mice (Nishida et al., 2009; Ogura et al.,

2009). Recently, doxercalciferol, a vitamin D analogue, has been found to decrease the accumulation of triglycerides and cholesterol in kidney (Wang et al., 2011a). Additionally, basal levels of serum cholesterol are higher in VDR knockout mice than in wild-type mice, suggesting a role of VDR in cholesterol homeostasis (Wang et al., 2009). Clinically, a combination of atorvastatin and vitamin D supplement exerts a synergetic effect in lowering cholesterol in patients (Schwartz, 2009), and there may be a role of the VDR in hypercholesterolemia.

We showed in the present study that VDR unequivocally upregulated Cyp7a1 levels in the murine liver via repression of hepatic SHP directly, and decreased intestinal FGF15 via FXR antagonism, rendering the lowering of plasma and liver cholesterol in vivo. We 154

further showed that 1,25(OH)2D3 lowered cholesterol in a mouse model only when high concentration of cholesterol in plasma and liver were present. Several important messages can be deduced from our observations: (a) species differences exist between the rat and mouse, as shown in our work on 1,25(OH)2D3 treatment with the rat in vivo, evoking FXR rather than VDR effects in rat liver (Chow et al., 2009); (b) it must be recognized that results established in vitro may not reflect events in vivo; and (c) multiple mechanisms can prevail concomitantly and exert synergistic or opposing actions, and these pathways tend to be concentration-, time-, and dose-dependent, organ and route specific, and ligand- or solvent-dependent. Different results may be observed due to dose, vehicle, frequency, route of administration, and time of sampling.

5.6 ACKNOWLEDGMENTS

This work was supported by the Canadian Institutes for Health Research CIHR,

Grant [Grant MOP89850]. Edwin C.Y. Chow was supported by the University of Toronto

Open Fellowship and the National Sciences and Engineering Research Council of Canada

Alexander Graham Bell Canada Graduate Scholarship (NSERC-CGS), and Matthew R.

Durk was supported by the Canadian Institutes of Health Research (CIHR) Strategic

Training Grant in Biological Therapeutics. We like to thank Monica Patel from Dr. Carolyn

L. Cummins laboratory for providing assistance in the cholesterol measurements in the liver.

5.7 STATEMENT OF SIGNIFICANCE OF CHAPTER 5

Cytochrome 7-hydroxylase or CYP7A1 is the rate limiting enzyme in cholesterol metabolism in liver. The enzyme is known to be under regulation of the nuclear receptor 155

farnesoid X receptor (FXR) that negatively regulates CYP7A1 via the induction of the small heterodimer partner (SHP), which acts by attenuating the stimulatory effects of other nuclear receptors and transcription factors, such as the hepatocyte nuclear factor (HNF-4) and liver receptor homolog (LRH-1) in liver, and by intestinal FXR through induction of fibroblast growth factor 15 (FGF15), a hormonal signaling molecule that travels through the portal blood to negatively regulate Cyp7a1 in liver. In this chapter, we demonstrate that activation of the VDR in liver inhibited the expression of hepatic SHP to remove the inhibition on Cyp7a1. The inhibition results in increased cholesterol metabolism and decrease in plasma and liver cholesterol in a high cholesterol mouse model. We showed the lack of involvement of FXR in SHP repression but the involvement of SHP with use of both FXR and SHP knockout mouse models. The luciferase assay, truncation analysis, binding assay and EMSA of mouse and human SHP promoters revealed that direct binding inhibited their activities, suggesting that VDR directly binds and inhibits SHP transcriptional activity. These results suggest a potential role of VDR in cholesterol lowering by repression of SHP, removing the inhibition of SHP on Cyp7a1, increasing

Cyp7a1 protein and activities in vivo. Furthermore, examination of cholesterol related gene changes in the liver and intestine showed a decrease in intestinal ABCA1 transporter in

1,25(OH)2D3 treated mice prefed with a Western diet, which maybe another factor for the decrease in plasma cholesterol. (See APPENDIX F2 to F4). Primer sequences of these genes for qPCR are listed in Appendix T1. Western blotting conditions for Cyp7a1 protein and other related proteins are listed in APPENDIX T2. Changes in body weights of each treatment experiment, food intake, plasma calcium, phosphate and alanine transaminase

(ALT) and liver triglyceride are listed in APPENDIX T3-T9.

156

In this study, I was responsible for the investigation and analyses of all data from in vivo studies using wild-type, fxr(-/-) and shp(-/-) mice with or without the Western diet, including 1,25(OH)2D3 treatment to mice, tissue harvesting, mRNA and protein analyses, bile acids, cholesterol and ALT measurement. Lilia Magomedova examined the molecular mechanism of VDR on SHP: luciferase assay, truncation analysis, binding assay and

EMSA. Rucha Patel investigated the direct effect of VDR on primary mouse hepatocytes.

Matthew R. Durk performed immunostaining of VDR protein in mouse liver. Han-Joo

Maeng examined the microsomal activity of mouse Cyp7a1. Holly Quach ran the in vivo treatment of 1,25(OH)2D3 on fxr(-/-) pretreated with the Western diet.

157

CHAPTER 6

6. CORRELATION BETWEEN TISSUE 1,25-DIHYDROXYVITAMIN D3

LEVELS AND GENE CHANGES: A TEMPORAL STUDY

Edwin C.Y. Chow1, Reinhold Vieth2, and K. Sandy Pang1

1Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, and the

2Department of Nutritional Sciences, University of Toronto, Canada

158

6.1 ABSTRACT

Activation of the vitamin D receptor (VDR) has been found to regulate transporters and enzymes in many organs, including the multidrug resistance protein 1 or P-glycoprotein

(Mdr1/P-gp) in kidney and Cyp7a1, the cholesterol metabolizing enzyme in mouse liver in vivo. However, plasma and tissue levels of 1,25(OH)2D3 as well as changes in expression of target genes and nuclear receptors are largely unknown. A temporal study of 1,25(OH)2D3 and calcium concentrations in plasma and tissues (ileum, liver, kidney, and brain) as well as mRNA expression of target genes and nuclear receptors was conducted in wild-type mice after treatment of 2.5 µg/kg of 1,25(OH)2D3 i.p. every other day for 8 days. Blood (to provide plasma) and tissues were harvested at various time points during the treatment period for the determination of gene changes by qPCR; levels of calcium were measured by

ICP-AES and 1,25(OH)2D3 by EIA (enzyme immunoassay). The pharmacokinetics of

1,25(OH)2D3 was altered with time after 1,25(OH)2D3 because of induction of Cyp24, a catabolic enzyme for 1,25(OH)2D3 metabolism in tissues, especially in kidney. However,

1,25(OH)2D3 concentrations in tissues were much higher than basal levels during the 8-day treatment period. Induction of hepatic Cyp7a1 and renal Mdr1 mRNA correlated with

1,25(OH)2D3 concentrations and Cyp24 mRNA changes in the tissues, showing that the induction of hepatic Cyp7a1 and Mdr1 was the result of VDR activation.

6.2 INTRODUCTION

The vitamin D receptor (VDR) has been found to be distributed in a wide range of tissues and is now known to be one of the major players in the regulation of transporters and enzymes. The active form of vitamin D3, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3], is the 159

natural ligand for the VDR (Jones et al., 1998). When activated, the VDR becomes a transcription factor and is translocated into the nucleus where it heterodimerizes with the retinoid X receptor (RXRα) to form a complex that binds to the DNA response element to initiate transcription. Activation of the VDR has been associated with changes in numerous enzymes and transporters that include cytochrome P450 (human CYP3A4 and rodent

Cyp3a1, Cyp3a9, and Cyp24), sulfotransferase (SULT2A1), the multidrug resistance protein

1 or P-glycoprotein (MDR1/P-gp) (Chapter 4), and the multidrug resistance associated proteins (MRP2/Mrp2, Mrp3, MRP4/Mrp4) both in vitro and in vivo (Thummel et al., 2001;

Echchgadda et al., 2004; Chow et al., 2009; Fan et al., 2009; Chow et al., 2010). These changes have serious implications on cholesterol and bile acid homeostasis (Chapter 5) as well as amyloid-beta (Aβ) peptides transport (Durk et al., 2011b) in the brain.

Studies have revealed that the VDR is important in bile acid and cholesterol homeostasis (Makishima et al., 2002) (see Chapter 5). 1,25(OH)2D3 administration to rats resulted in increased apical sodium dependent bile acids transporter (Asbt) in the ileum that increased portal bile acid concentrations and activated the farnesoid X receptor (FXR), resulting in secondary effects in intestine and liver (Chen et al., 2006; Chow et al., 2009).

This resulted in a downregulation of Cyp7a1, the rate limiting enzyme for cholesterol metabolism, and a decrease in bile acid synthesis. However, in mice and humans, higher amounts of VDR are present in the liver compared to the rat, suggesting species variations of VDR in liver (Gascon-Barré et al., 2003). Nishida et al. (2009) reported that 1- hydroxyvitamin D3 treatment to mice resulted in increased expression of Cyp7a1 as well as

Mrp2, Mrp3, and Mrp4 in kidney for increase bile acid elimination from the liver (Nishida et al., 2009). In a previous study (Chapter 5), we demonstrated that repeated dosing of 160

1,25(OH)2D3 to mice prefed a Western diet resulted in decreased plasma and liver cholesterol concentrations and an increase in Cyp7a1 expression due to repression, of the

VDR on the hepatic short heterodimer partner (SHP), and an indirect repression of intestinal fibroblast growth factor 15 (FGF15), both being negative regulators of Cyp7a1 (Chapter 5).

The study suggests that the VDR plays a role in cholesterol metabolism in the liver.

Our laboratory showed that the VDR directly regulates P-gp independently of the

FXR in vivo, resulting in 3.45- and 1.47-fold increases in intrinsic secretary clearances of digoxin in kidney and brain, respectively (Chow et al., 2011a). Increased brain P-gp protein expression was associated with a decrease in accumulation of brain digoxin, a P-gp substrate, in vivo (Chow et al., 2011a). Moreover, the increased brain P-gp activity may play a role in the prevention of Alzheimer’s disease due to enhanced secretion of -amyloid peptides (Aβ)

(Durk et al., 2011a; Durk et al., 2011b), substrates of the P-gp (Lam et al., 2001).

Although the present findings suggest important roles of the VDR on cholesterol metabolism in liver and increased P-gp efflux in brain, there is no data existing to show the tissue levels of 1,25(OH)2D3 and how these relate to changes in gene expression. The plasma concentrations of 1,25(OH)2D3 are reported to be detectable at the pM range (Dusso et al.,

2005). Clinically, the 25-hydroxyvitamin D3, and not the 1,25(OH)2D3 concentration in plasma is used as a biomarker to reflect vitamin D status, because the conversion of 25- hydroxyvitamin D3 to 1,25(OH)2D3 is the rate limiting step (Wang et al., 2008b). However, the concentration of 25-hydroxyvitamin D3 does not reflect the concentration of

1,25(OH)2D3 in plasma nor in tissue. In order to implicate pharmacological effects of

1,25(OH)2D3, there is a need to quantify 1,25(OH)2D3 levels in tissues. Therefore, a

161

temporal study of 1,25(OH)2D3 concentrations in mice were determined to relate the tissue concentrations of 1,25(OH)2D3 on the changes in VDR target gene/nuclear receptor mRNA expression. In addition, we examined calcium ions and levels of Cyp24, a target gene of the

VDR and the catabolic enzyme that degrades 1,25(OH)2D3, after 1,25(OH)2D3 treatment.

The pharmacokinetic changes of exogenously administered 1,25(OH)2D3 in mice given multiple doses was studied.

6.3 METHODS

6.3.1 Materials

The 1,25(OH)2D3 powder was purchased from Sigma-Aldrich Canada (Mississauga,

ON, Canada). Anti-Cyp24 antibody (H-87) was purchased from Santa Cruz Biotechnology

(Santa Cruz, CA), and anti-Gapdh (6C5), from Abcam (Cambridge, MA). All other reagents were purchased from Sigma-Aldrich Canada (Mississauga, ON, Canada) and Fisher

Scientific (Mississauga, ON, Canada).

6.3.2 Pharmacokinetic Study of 1,25(OH)2D3 in Mice

Anhydrous ethanol was used to dissolve the 1,25(OH)2D3 powder, and the concentration was quantified at 265 nm spectrophotometrically (UV-1700, Shimadzu

Scientific Instruments, MD) before the final preparation of the 1,25(OH)2D3 in sterile corn oil. An in vivo pharmacokinetic study of 1,25(OH)2D3 was conducted in 8-week ago mice.

1,25(OH)2D3, dissolved in sterile corn oil, was given intraperitoneally (i.p.) to male,

C57BL/6 mice at either a dose of 0 or 2.5 µg/kg (5 μl/g) on Days 1, 3, 5, and 7 at 9 a.m. in the morning. Thereafter, one mouse was sacrificed for blood (plasma) and tissue measurements for the 1,25(OH)2D3 treatment group at various sampling time points at 0, 0.5, 162

1, 3, 6, 9, 12, 24, 48, 96, 192, and 360 h after the injected dose on Day 1, and at 0, 0.5, 1, 3,

6, 9, 12, 24, and 48 h after the injected doses on Days 3, 5, and 7. For the control group, sampling was conducted at 0, 3, 6, and 12 h on Day 1, and 0 h on Days 3, 5, 7, 9, and 15.

Each mouse was rendered unconscious in a carbon dioxide chamber, and blood was collected by cardiac puncture into a heparin-coated (1000 IU/ml) 1-ml syringe-23G 3/4” needle set. Plasma was obtained by centrifugation of blood at 3000 rpm for 10 min. The mouse was perfused with ice-cold saline through the lower vena cava. The liver, kidney, brain, and mucosal scraping from ileal enterocytes were harvested (Chow et al., 2009), weighed, cut into small pieces, snap- frozen in liquid nitrogen, and stored at -80°C for future analyses.

6.3.3 Plasma Calcium and Phosphorus Analysis

Quantification of calcium and phosphorus in plasma were determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES) (Optima 3000 DV, Perkin Elmer).

These plasma samples were diluted 350-fold with 1% nitric acid before each measurement.

Calcium was quantified at 317.9 nm and 315.9 nm, whereas phosphorus at 213.6 nm and

214.9 nm.

6.3.4 Tissue 1,25(OH)2D3 Extraction and 1,25(OH)2D3 Enzyme-immunoassay (EIA) for Plasma and Tissue Samples

The tissue extraction procedure for lipids was similar to Bligh and Dyer (1959), with modifications. Weighed brain, liver, kidney and scraped enterocyte samples were diluted with double distilled water (w/v) to 1 ml. The sample was homogenized with 3.75 ml of a mixture of methylene chloride and methanol (1:2 v/v). The homogenates were added to 1.25 ml of methylene chloride, vortexed for 1 min, and then 1.25 ml of double distilled water was 163

added and mixed for another min before centrifugation at 3000 rpm for 20 min at room temperature. The extracted solution (bottom phase) was collected by a glass syringe-metal needle set. The extraction procedure was repeated upon addition of 1.25 ml of methylene chloride, vortexed for 1 min and recentrifuged at 3000 rpm for 20 min at room temperature.

The recovered bottom phase was harvested and added to that from the previous extraction.

The pooled extract was dried under nitrogen gas and reconstituted in 0.3 ml of charcoal- stripped human serum.

The concentration of 1,25(OH)2D3 in mouse plasma or tissue sample, reconstituted in stripped human serum, was determined using an enzyme-immunoassay (EIA) kit (Cat#AC-

62F1 from Immunodiagnostics Systems (IDS) Inc., Scottsdale, AZ, USA) following the manufacture’s protocol. Plasma and tissue samples were diluted with charcoal stripped human serum before assay.

6.3.5 Pharmacokinetic Analysis: Plasma Concentration-Time Profile

A rapid absorption phase followed by an apparent, first-order decay was observed

(Fig. 6.1). The elimination rate constant (k) in plasma after each dose of 1,25(OH)2D3 was estimated from the slope of the log-linear phase, from between 1 to 12 h or 3 to 12 h time- points, and the elimination half life (t1/2) was calculated as 0.693/k. The plasma area under the curve (AUC(0-48)) between 0 to 48 h was estimated by the trapezoidal rule, assuming that the initial time point being the same as the basal level of 1,25(OH)2D3 of the control mouse.

6.3.6 Preparation of Subcellular Protein Fractions of Kidneys

For preparation of the crude membrane fraction for the assay of Cyp24, about 0.1 g of kidney tissue was homogenized in the crude membrane homogenizing buffer (250 mM

164

sucrose, 10 mM HEPES, and 10 mM Trizma base, pH 7.4) containing 1% protease inhibitor cocktail (Chow et al., 2011a). The resultant homogenate was centrifuged at 3,000 g for 10 min at 4°C. The pellet was resuspended in a buffer [60 mM KCl, 15 mM NaCl, 5 mM

MgCl2•6H2O, 0.1 mM EGTA, 300 mM sucrose, 0.5 mM DTT, 0.1 mM PMSF, 300 mM sucrose, and 15 mM Trizma HCl pH 7.4] containing 1% protease inhibitor cocktail for

Western blotting of Cyp24 protein expression. The protein concentrations of the samples were assayed by the Lowry method (Lowry et al., 1951) using bovine serum albumin as the standard. Samples were then stored at -80°C until Western-blot analyses.

6.3.7 Western Blotting

Total protein samples (50 µg) were separated by 10% SDS-polyacrylamide gels at

100 V. After separation, proteins were transferred onto a nitrocellulose membrane

(Amersham Biosciences, Piscataway, NJ), which was subsequently blocked with 5% (w/v) skim milk in Tris-buffered saline (pH 7.4) and 0.1% Tween 20 (TBS-T) (Sigma-Aldrich,

ON) for 1 h at room temperature, and then washed once with 0.1% TBS-T, followed by incubation with primary antibody solution in 2% skim milk in 0.1% TBS-T overnight at 4°C.

The membrane was washed with 0.1% TBS-T on the next day and then incubated with secondary antibody in 2% skim milk in 0.1% TBS-T for 2 h at room temperature, and again washed with 0.1% TBS-T. Bands were visualized using chemiluminescence reagents

(Amersham Biosciences, Piscataway, NJ) and quantified by scanning densitometry (NIH

Image software; http://rsb.info.nih.gov/nih-image/). The band intensity of the target protein was normalized against that of Gapdh to correct for loading errors.

165

6.3.8 Quantitative Real-Time Polymerase Chain Reaction (real-time PCR or qPCR)

Similar to the procedure described previously (Chow et al., 2011a), total RNA from scraped enterocytes, liver and kidney tissues were extracted with the TRIzol extraction method (Sigma-Aldrich) according to the manufacturer’s protocol, with modifications. Total

RNA was quantified by UV spectrometry at 260 nm. The purity was checked by ratios of the readings at 260/280 nm and 260/230 nm (≥1.7). About 1.5 µg of cDNA was immediately synthesized from the RNA samples, using the High Capacity cDNA Reverse Transcription

Kit (Applied Biosystems Canada, ON). qPCR was performed with SYBR Green detection system. A PCR mixture (20 µl final volume) consisting of 75 ng cDNA, 1 µM of forward and reverse primers, and 1 Power SYBR Green PCR Master Mix (Applied Biosystems) was used to perform PCR analysis. Information on primer sequences are listed in Table 6-1.

Amplification and detection were performed using an ABI 7500 system. The qPCR system was assigned the following PCR cycling temporal profile: 95°C for 10 min, and 40 cycles of

95°C for 15 sec and 60°C for 1 min, followed by the dissociation curve. Data were analyzed using the ABI Sequence Detection software version 1.4 to obtain critical threshold cycle (CT) values. Fold changes between vehicle control and treatment was expressed as 2-(∆∆CT). All target mRNA data were normalized to villin mRNA for intestinal samples and Gapdh mRNA for liver and kidney samples.

166

Table 6-1 Mouse primer sets for quantitative Real-Time qPCR

Gene Bank Number Forward (5’→ 3’ Sequence) Reverse (5’→ 3’ Sequence)

Cyp7a1 NM_007824 AGCAACTAAACAACCTGCCAGTACTA GTCCGGATATTCAAGGATGCA

Cyp24 NM_009996 CTGCCCCATTGACAAAAGGC CTCACCGTCGGTCATCAGC

Mdr1 NM_011076 TACGACCCCATGGCTGGATC GGTAGCGAGTCGATGAACTG

TRPV6 NM_022413 ATCGATGGCCCTGCCAACT CAGAGTAGAGGCCATCTTGTTGCTG

Villin NM_009509 TCCTGGCTATCCACAAGACC CTCTCGTTGCCTTGAACCTC

Cyclophillin X58990 GGAGATGGCACAGGAGGAA GCCCGTAGTGCTTCAGCTT

167

6.4 RESULTS

6.4.1 Plasma Concentration of 1,25(OH)2D3 and Calcium in Single and Multiple

Doses of 1,25(OH)2D3 in Mice

Fig. 6-1 shows the plasma concentration-time profiles of 1,25(OH)2D3 and calcium after a single dose or multiple i.p. doses of 2.5 µg/kg of 1,25(OH)2D3 to mice. A monoexponential decay of 1,25(OH)2D3 was noted in the plasma 1,25(OH)2D3 concentration-time profile (Fig. 6.1A). For the single dose treatment, the plasma

1,25(OH)2D3 level was similar to basal level at one day after injection, but fell below the basal level after two days and returned to the basal level eight days after 1,25(OH)2D3 administration. The calcium concentration rose above basal levels on Day 1 after treatment, but fell below basal level 2 days after treatment (Fig. 6-1A), and returned to basal level on the eighth day.

For multiple dose administration, plasma 1,25(OH)2D3 concentrations peaked around

60 nM 0.5 h after each injection and fell below the basal level at the end of each dosing period (Fig. 6-1B), possibly due to induction of Cyp24 (see Fig. 6.3 later). The concentrations of 1,25(OH)2D3 at the nadir fell below the corresponding control (basal) level, and this pattern persisted throughout the dosing regimen. The plasma calcium concentration peaked one day after dosing, but decreased to basal levels thereafter (Fig. 6-1B). The pattern was similar after subsequent injections, with the peak calcium concentrations about 11%-

50% above the basal level after subsequent administrations. The increase in calcium was about 10% above the basal level at two days after the last injection. The calcium exposure

[AUC(0→8days)] in the 1,25(OH)2D3-treated mice was 100.8 mg/dl•day compared to control,

84.2 mg/dl•day. 168

(A) (B)

1000000 Plasma 1000000 Plasma 100000 100000 10000 10000 1000 1000 100 100 10 10

(pM) Concentration 1 Concentration (pM) 1 3 3 Vehicle Control D D Vehicle Control 2 2 0.1 1,25(OH) D 0.1 2 3 1,25(OH) D 0.01 0.01 2 3 0.001 0.001 1,25(OH) 1,25(OH) 0.0001 0.0001 0246810121416 02468 Time (Day) Time (Day)

20 Plasma 20 Plasma

Vehicle Control Vehicle Control 1,25(OH) D 15 1,25(OH)2D3 15 2 3

10 10

5 5

Calcium Concentration (mg/dl) Calcium Concentration (mg/dl) 0 0 0246810121416 02468 Time (Day) Time (Day)

Figure 6-1 Plasma 1,25(OH)2D3 and calcium concentration-time profiles after (A) a single dose or (B) multiple doses of 2.5 µg/kg i.p. 1,25(OH) D or vehicle to mice. Data from vehicle-treated mice 2 3 were denoted as open circles, whereas those from 1,25(OH)2D3-treated mice were denoted as solid circles.

Because it is recognized that the 1,25(OH)2D3 concentrations subsequent to i.p.

administration is the sum of the basal and exogenously derived 1,25(OH)2D3, a very

simplistic approach was taken to address the question of increased 1,25(OH)2D3 metabolism

due to increased Cyp24. The pharmacokinetics of 1,25(OH)2D3, based on the total

1,25(OH)2D3 (exogenous + basal) concentration after the administered 1,25(OH)2D3 differed

slightly after each injection. The elimination rate constant (k) was 0.282 h-1 with an

elimination half life of 2.46 h after the first dose, and this was increased to 0.339 h-1 with a

169

decreased half life of 2.05 h for the last dose (Table 6-2). The plasma AUC0-48 (243 vs. 183 nM•h) was decreased after the last dose compared to the first dose (Table 6-2).

Table 6-2 Pharmacokinetic parameters of 1,25(OH)2D3 in mice Doses of 2.5 µg/kg 1,25(OH)2D3 1st Dose 2nd Dose 3rd Dose 4th Dose Plasma elimination rate constant, k (h-1) 0.282 0.291 0.253 0.339 Plasma elimination half-life, t1/2 (h) 2.46 2.38 2.74 2.05 Plasma AUC0-48 (pM*h) 243638 145949 173266 183626

6.4.2 Tissue Concentrations of 1,25(OH)2D3 in Single and Multiple Doses of

1,25(OH)2D3 in Mice

Fig. 6-2 illustrates the tissue concentrations (ileum, liver, kidney, and brain) of

1,25(OH)2D3- and calcium-concentration time profiles after a single intraperitoneal (Fig. 6-

2A) and multiple (Fig. 6-2B) doses of 2.5 µg/kg of 1,25(OH)2D3 to mice. Similar to the plasma concentration-time profile, a parallel, monoexponential decay of 1,25(OH)2D3 was observed in each tissue (ileum, liver, kidney, and brain). The single, i.p. 1,25(OH)2D3 injection to mice did not result in a drastic lowering of 1,25(OH)2D3 concentrations at the nadir in tissues below the basal level for the multiple doses, except in the liver, where the nadir dipped slightly below basal levels on the 2nd day but returned to basal level on the 4th day after administration (Fig. 6-2A). Multiple doses of 1,25(OH)2D3 resulted in

1,25(OH)2D3 tissue levels similar to those in plasma after each injection for all tissues (Fig.

6-2B). However, the half lives of the exogenously administered 1,25(OH)2D3 seemed to be generally faster (14-26%) in all tissues after the fourth dose compared to after the first dose.

170

6.4.3 Comparison of Renal Cyp24 mRNA and Protein in a Single or Multiple Doses

of 1,25(OH)2D3 in Mice

Cyp24 is the major catabolic enzyme that degrades 1,25(OH)2D3 (Jones et al., 1998).

Administration of 1,25(OH)2D3 to mice increased Cyp24 mRNA and protein levels by 80- fold at 1 hand 27-fold at 12 h after a single dose injection (Fig. 6-3A). The increase in mRNA peaked at around 3 h whereas the protein expression peaked at 12 h, and remained above basal levels for at least 8 and 16 days, respectively (Fig. 6-3A). Similar patterns were observed for multiple doses of 1,25(OH)2D3 to mice that showed high increases in renal

Cyp24 mRNA and protein levels, with expressions that were 40 and 5-fold those of the basal level, respectively, at the end of the treatment period (Fig. 6-3B).

6.4.4 Temporal Changes in Ileal Cyp24, TRPV6 and FGF15, Liver Cyp24, Cyp7a1 and SHP, and Renal Cyp24, Mdr1 and TRPV6 mRNA Expressions after

Multiple Doses of 1,25(OH)2D3 in Mice

Fig. 6-4 illustrates the temporal effect of VDR on target genes in different tissues after the multiple doses of 1,25(OH)2D3. There was no difference in the basal expression of ileal Cyp24 and TRPV6, liver Cyp24, and renal Mdr1 and TRPV6 mRNA from 0 to 12 h after injection of vehicle to control mice, likely due to the absence of circadian rhythms (Fig.

6-4). However, a circadian rhythm was observed in the liver Cyp7a1 mRNA, increasing to

20-fold during the first 12th hours after vehicle treatment, then later returning to basal level

(Fig. 6-4B).

In the ileum, the increase in Cyp24 mRNA expression peaked at approximately 6 h after each 1,25(OH)2D3 injection, and the mRNA returned to baseline after 1 day (Fig. 6-4A).

Similarly, TRPV6 expression increased at approximately 9 h after each dose, and the mRNA

171

returned to baseline after one day, with subsequent increases being dramatically higher

(about 500-fold) after the 3rd injection compared to the first and second injections (Fig. 6-

4A). Induction of Cyp24 mRNA in the liver was observed at 3 h after each 1,25(OH)2D3 injection and levels returned to basal level after a day (Fig. 6-4B).

An increase in renal Mdr1 mRNA was seen at the beginning of the 2nd injection, and the induction remained over 6-fold till the end of the experiment (Fig. 6-4C). Similar to the data from ileum, the increase of renal TRPV6 mRNA was observed 6 to 9 h after each treatment, and the induction was increased no more than 13-fold (Fig. 6-4C). The increase in renal TRPV6 remained above baseline until 24 to 48 h after each injection of 1,25(OH)2D3.

The patterns for intestinal FGF15 and hepatic SHP were erratic (Figs. 6-4A and 6-4B), and, due to the known instability of their expression shown in the control mouse during the vehicle treatment period, no conclusion could be made in regard to the data from the

1,25(OH)2D3- treated group.

172

(A) (B)

1000000 Ileum 1000000 Ileum 100000 100000 10000 10000 Vehicle Control Vehicle Control 1000 1,25(OH)2D3 1000 1,25(OH)2D3 100 100 10 10 1 1 Concentration (pmol/g) Concentration Concentration (pmol/g) 3 3 0.1 0.1 D D 2 2 0.01 0.01 0.001 0.001 0.0001 0.0001 1,25(OH) 1,25(OH) 0 2 4 6 8 10121416 02468 1000000 Liver Time (Day) 1000000 Liver Time (Day) 100000 100000 10000 Vehicle Control 10000 Vehicle Control 1000 1,25(OH)2D3 1000 1,25(OH)2D3 100 100 10 10 1 1 Concentration (pmol/g) Concentration (pmol/g) Concentration 3 3 0.1 0.1 D D 2 2 0.01 0.01 0.001 0.001 0.0001 0.0001 1,25(OH) 1,25(OH) 0246810121416 02468 1000000 Kidney Time (Day) 1000000 Kidney Time (Day) 100000 100000 10000 Vehicle Control 10000 Vehicle Control Concentration (pmol/g) Concentration 1000 1,25(OH)2D3 (pmol/g) Concentration 1000 1,25(OH)2D3 3 100 3 100 D 10 D 10 2 2 1 1 Concentration (pmol/g) Concentration Concentration (pmol/g) 3 0.1 3 0.1 D D 2 0.01 2 0.01 0.001 0.001 0.0001 0.0001 1,25(OH) 0246810121416 1,25(OH) 02468 1,25(OH) 1,25(OH) 1000000 Brain Time (Day) 1000000 Brain Time (Day) 100000 Vehicle Control 100000 Vehicle Control 10000 1,25(OH)2D3 10000 1,25(OH)2D3 1000 1000 100 100 10 10 1 1 Concentration (pmol/g) Concentration Concentration (pmol/g) 3 0.1 3 0.1 D D 2 0.01 2 0.01 0.001 0.001 0.0001 0.0001 1,25(OH) 0246810121416 1,25(OH) 02468 Time (Day) Time (Day)

Figure 6-2 Tissue 1,25(OH)2D3 (ileum, liver, kidney, and brain) concentration-time profile from (A) a single dose or (B) multiple doses of 2.5 µg/kg i.p. 1,25(OH)2D3 to mice. Data from vehicle treated mice were denoted as open circle while those from 1,25(OH)2D3 treated mice were denoted as solid circle

173

(A) (B)

100 Kidney 100 Kidney Vehicle Control Vehicle Control 1,25(OH) D 80 80 2 3 1,25(OH)2D3 60 60

40 40

20 20

0 0 0246810121416 02468 Relative Renal Cyp24 mRNA Expression Relative Renal Cyp24 mRNA Expression Time (Day) Time (Day)

Kidney 30 Kidney 30

25 Vehicle Control 25 Vehicle Control 1,25(OH) D 2 3 1,25(OH)2D3 20 20

15 15

10 10

5 5 0 0 0246810121416 02468 Relative Renal Cyp24 ProteinExpressionRelative Renal Cyp24 Relative Protein Cyp24 Renal Expression Time (Day) Time (Day)

Figure 6-3 Renal Cyp24 mRNA and protein expression after (A) a single dose or (B) multiple doses of 2.5 µg/kg i.p. of 1,25(OH)2D3 to mice. Data from vehicle treated mice were denoted as open circle, whereas those from 1,25(OH)2D3 treated mice were denoted as solid circle

The induction of liver Cyp7a1 mRNA levels after the first dose of 1,25(OH)2D3 were

higher than those control Cyp7a1 levels that showed a circadian rhythm (Fig. 6-4B); a more

drastic increase of liver Cyp7a1 mRNA was observed 12 h after the 2nd and 4th injections (60

and 80-fold). In fact, a 13-fold increase in liver Cyp7a1 mRNA expression was attained at

48 h after the last injection.

174

(A) (B) (C)

600 Ileum 180 Liver 100 Kidney Vehicle Control 160 Vehicle Control 500 1,25(OH)2D3 1,25(OH)2D3 140 Vehicle Control 80 1,25(OH)2D3 400 120 60 100 300 80 40 200 60 40 100 20 20 00 0 0

Relative Ileal Cyp24 mRNA Expression Ileal Cyp24 Relative 02468 Relative Liver Cyp24 mRNA Expression Relative Cyp24 Liver 02468 02468 Relative Renal mRNARelative Renal Cyp24 Expression Time (Day) Time (Day) Time (Day)

700 Ileum 100 Liver 14 Kidney Vehicle Control 600 12 80 1,25(OH)2D3 Vehicle Control Vehicle Control 500 10 1,25(OH) D 1,25(OH)2D3 2 3 400 60 8

300 40 6 200 4 20 100 2 00 0 0 02468 02468 02468 Relative Ileal TRPV6 mRNA Ileal Relative Expression Relative Liver Cyp7a1 mRNA Expression Cyp7a1 Liver Relative Relative Renal TRPV6 mRNA Expression Time (Day) Time (Day) Time (Day)

1.8 Ileum 1.2 Liver 18 Kidney Vehicle Control Vehicle Control 1.6 1,25(OH)2D3 1,25(OH) D 16 Vehicle Control 1.0 2 3 1,25(OH) D 1.4 14 2 3 1.2 0.8 12 1.0 10 0.6 0.8 8 0.6 0.4 6 0.4 4 0.2 0.2 2 0.00 0.00 0 Relative Liver SHP mRNA Expression Relative Ileal FGF15 mRNA Expression mRNA FGF15 Ileal Relative 02468 02468Expression mRNA Mdr1 Renal Relative 02468 Time (Day) Time (Day) Time (Day)

Figure 6-4 mRNA expression of (A) ileal Cyp24, TRPV6 and FGF15, (B) hepatic Cyp24, Cyp7a1 and SHP, and (C) renal Cyp24, Mdr1 and TRPV6 after multiple doses of 2.5 µg/kg i.p. of 1,25(OH)2D3 to mice. Data from vehicle treated mice were noted as open circles while those from 1,25(OH)2D3 treated mice were augmented as solid circles 175

6.5 DISCUSSION

In this study, we examined the plasma and tissue concentration-time profiles of

1,25(OH)2D3 and the pharmacokinetic changes after a single vs. multiple dosing schedule.

We appraised whether the 1,25(OH)2D3 tissue concentrations relate to changes in gene expressions in the ileum, liver, and kidney. Pharmacokinetic parameters of exogenously

st administered 1,25(OH)2D3 differed between the 1 and last injections (Table 6-1) due to induction of Cyp24 expression in ileum, liver, and kidney (Figs. 6-2 and 6-4). Cyp24 is an enzyme that metabolizes 1,25(OH)2D3 to inactive forms, 1,24,25-trihydroxyvitamin D3 and

25-hydroxyvitamin D3 to 24,25-dihydroxyvitamin D3 (Jones et al., 1998). The increase in

Cyp24 resulted in enhanced 1,25(OH)2D3 elimination and larger rate constants and shorter half-lives (Table 6-1). Although the fold-induction of Cyp24 mRNA in ileum was higher than that in the kidney, the increase in ileal Cyp24 was less sustainable compared to that in kidney, perhaps due to the higher turnover rate of Cyp24 in enterocytes (Ferraris et al.,

1992; Bonventre, 2003). As a result, the rise in renal Cyp24 mRNA and protein levels was maintained for at least one week. Because the kidney is considered the major organ for the elimination of 1,25(OH)2D3, the increase in Cyp24 from multiple dosing of exogenous

1,25(OH)2D3 increased the total body clearance significantly after the fourth dose.

1,25(OH)2D3 was found in intestine, liver, kidney, and brain. 1,25(OH)2D3 is lipophilic, but is tightly bound to the vitamin D binding protein (DBP) in plasma (Dusso et al., 2005). Despite this binding, the present study showed that exogenous 1,25(OH)2D3 distributes quite non-discriminately into tissues that contain either low or high expression of VDR. In addition, the concentration of 1,25(OH)2D3 in these tissues remained above

176

basal levels over a long period of time, suggesting that treatment with 1,25(OH)2D3 could result in sustainable pharmacological effects in these tissues.

The mRNA of VDR target genes in the intestine, liver, and kidney, altered by administration of 1,25(OH)2D3, were parallel to changes in the tissue concentrations of

1,25(OH)2D3. As seen in Figs. 6-3 and 6-4, induction of ileal and renal Cyp24 and TRPV6, known VDR targets (Meyer et al., 2007), occurred 3 to 6 h after each dose of 1,25(OH)2D3, with tissue concentrations of 1,25(OH)2D3 being much higher than that of the baseline. The peak 1,25(OH)2D3 tissue concentrations in ileum and kidney occurred at approximately 1 to

3 h after dosing, a few h prior to the observed induction of Cyp24 and Cyp7a1 mRNA. The delay could be due to the time needed for translocation of VDR into the nucleus and heterodimization with RXRα to initiate the increase in transcription. Fan et al. (2009) also reported that CYP3A4 mRNA in Caco-2 cells was highly induced with a longer time exposure of 1,25(OH)2D3. This was presently mirrored in our studies by the increase of renal Mdr1 mRNA, whose induction was sustained for 2 days after each injection, and the magnitude of induction depended on the number of doses given and the frequency.

Data on the induction of liver Cyp24 mRNA in Fig. 6-4B infers that there is VDR activation in the liver after 1,25(OH)2D3 administration. Similarly, Cyp7a1 mRNA was induced after each dosing interval. Although a circadian rhythm of Cyp7a1 was observed, as did others (Gielen et al., 1975), the higher increase in Cyp7a1 (60 to 80-fold) among the treated mice compared to controls at 12 h, suggests that the higher Cyp7a1 mRNA was not due to circadian rhythm. Activation of Cyp7a1 by VDR was previously found in vivo to involve direct inhibition of hepatic SHP (Chapter 5). Thus, the steps for VDR activation of

177

Cyp7a1 are reliant on the time needed for VDR to decrease hepatic SHP, which in turn, leads to increased Cyp7a1 expression.

In conclusion, this study conclusively shows that 1,25(OH)2D3 administration to mice increased the tissue levels of 1,25(OH)2D3 and increased the expression of VDR target genes. Chronic dosing of 1,25(OH)2D3, resulted in an increase in 1,25(OH)2D3 clearance, due to increased Cyp24 expression in all tissues, especially in the kidney, to degrade 1,25(OH)2D3. However, the increase in elimination did not significantly lower

1,25(OH)2D3 tissue concentrations compared to baseline levels, suggesting that

1,25(OH)2D3 treatment could maintain sustainable concentrations of 1,25(OH)2D3 in tissues to activate the VDR, and regulate VDR target genes.

6.6 ACKNOWLEDGMENTS

This work was supported by the Canadian Institutes for Health Research CIHR,

Grant [MOP89850]. Edwin C.Y. Chow was supported by the University of Toronto Open

Fellowship and the National Sciences and Engineering Research Council of Canada

Alexander Graham Bell Canada Graduate Scholarship (NSERC-CGS). We like to thank

Dennis Wagner for his support in the project.

6.7 STATEMENT OF SIGNIFICANCE OF CHAPTER 6

In this chapter, plasma and tissue (ileum, liver, kidney, and brain) concentrations of

1,25(OH)2D3, when given every two days, were quantified and correlated to gene changes in tissues. The resultant concentrations of 1,25(OH)2D3 in tissues were much higher than basal levels. However, the pharmacokinetic parameters of 1,25(OH)2D3 were changed in the multiple dosing scheme due to induction of the enzyme, Cyp24, that is present mostly 178

in the kidney. However, concentrations of 1,25(OH)2D3 in tissues remained above baseline throughout the treatment period. Changes in VDR target genes such as Cyp24, TRPV6,

Mdr1 and Cyp7a1 correlated well to 1,25(OH)2D3 concentrations in tissues. This study conclusively shows that the induction of VDR target genes depends on tissue concentrations of 1,25(OH)2D3 and the duration of treatment.

In the study, I performed the administration of 1,25(OH)2D3 to mice, tissue harvesting, mRNA and protein analyses, tissue lipid extraction, 1,25(OH)2D3 analyses in plasma and tissue and calcium analyses.

179

CHAPTER 7

7. GENERAL DISCUSSION AND CONCLUSIONS

180

Transporters and enzymes are important proteins that determine the biological fates of both endogenous molecules such as hormones, bile acids, and cholesterol as well as exogenous molecules, such as xenobiotics and their metabolites. In the past two decades, studies have shown that transcription factors, which are activated nuclear receptors, regulate the levels of transporters and enzymes and control the balance of endogenous and exogenous molecules in the body (Tirona and Kim, 2005; Tirona, 2011). However, the expression of various nuclear receptors varies among the organs and different species, and thus, changes in transporter and enzyme levels resulting from the administration of a nuclear receptor ligand differ in different organs and species.

The vitamin D receptor (VDR), a member of the nuclear receptor 1 superfamily, exhibits significant homology with the xenobiotic nuclear receptors, the pregnane X receptor (PXR) and the constitutive androstane receptor (CAR), and is found to be an important regulator of transporters and enzymes (Reschly and Krasowski, 2006; Zöllner and Trauner, 2009). VDR is present abundantly in rat intestine and kidney (Sandgren et al.,

1991), and much less exists in liver (Gascon-Barré et al., 2003) where the VDR exists mostly in stellate and Kupffer cells, endothelial cells, and cholangiocytes. However, the

VDR is in human hepatocytes and mouse liver (Gascon-Barré et al., 2003; Han and Chiang,

2009; Khan et al., 2009b). Recently, the VDR was suggested to be as an important bile acid sensor, due to its ability to control the balance of bile acids by regulating enzymes for detoxification and synthesis, as well as transporters for uptake and secretion. This dissertation examined the role of the VDR in drug disposition and cholesterol metabolism in different organs and species in vivo.

181

In Chapter 3, we examined the direct and indirect role of the VDR in the rat intestine, liver, and kidney in vivo (Fig. 7-1) (Chow et al., 2009; Chow et al., 2010). We first investigated the expression of VDR in tissues in which the VDR is highly expressed - intestine and kidney – as well as in liver, in which VDR is poorly expressed (Fig. 3-1). The tissue abundance of VDR observed were similar to those of previous studies (Sandgren et al., 1991). We also examined the mRNA and protein expression of enzymes and transporters in tissues and showed that induction of known VDR target genes was different in various tissues and organs. Rat Cyp3a1, Cyp3a2, and Cyp3a9, isoforms of human

CYP3A4, were increased by 1,25(OH)2D3 in a tissue-specific manner: Cyp3a1 in intestine,

Cyp3a2 in liver, and Cyp3a9 in kidney (Chapter 3). Efflux transporters such as Mrp2, Mrp3,

Mrp4 were induced with 1,25(OH)2D3 only in the proximal segments of the intestine (Figs.

3-8, 3-9, and 3-10), and these are speculated to be non-genomic, VDR effects (Chow et al.,

2010). In addition, activation of VDR in the intestine stimulates secondary, FXR effects in intestine and liver. Previously, our laboratory demonstrated that intestinal Asbt in the ileum was induced by VDR activation, which increased bile acid reabsorption from the rat intestine (Chen et al., 2006). Bile acids, which are FXR ligands, can trigger a negative feedback mechanism to decrease cholesterol metabolism in liver by downregulating the expression of liver Cyp7a1, the rate limiting enzyme for cholesterol metabolism (Goodwin et al., 2000; Lu et al., 2000). Chapter 3 reported that 1,25(OH)2D3 induction of rat ileal

Asbt stimulated the reabsorption of bile acids from the intestine (Chow et al., 2009). As a result, increased FXR target genes was observed in the intestine and liver, including intestinal FGF15, a hormonal signaling molecule that activates FGFR4 in liver, and hepatic

SHP, a transcription factor, which are both negative regulators of Cyp7a1. P-gp, which is

182

known to be induced by activated VDR in Caco-2 cells (Fan et al., 2009), was found to be increased in the rat kidney and liver (Chapter 3) (Chow et al., 2010). Due to the presence of

VDR and FXR in intestine, liver (very low), and kidney, the induction of Mdr1/P-gp is likely to be attributed to FXR and not VDR activation. This aspect was clarified with us of fxr knockout animals (Chapter 4).

Hepatocyte Renal Epithelium

↑FXR; SHP Blood ↔Mrp3

↔Mrp4 ↑VDR ↔Mrp2 ↑Cyp3a2 Bile ↑P-gp ↑Cyp3a9 ↑P-gp ↔Mrp3 ↑Bsep ↔Ntcp ↑Cyp24 ↓Cyp7a1 ↔Oatp* ↑Asbt ↓FXR; SHP Urine Portal Vein

Enterocyte ↑Mrp3 ↑Mrp4 protein ↑EnterocyteFGF15; SHP mRNA ↑Cyp3a1 Induction Inhibition

↑Mrp2 ↔P-gp ↑Asbt Lumen

Figure 7-1 Summary of nuclear receptor, transporter and enzyme changes in intestine, liver, and kidney of rats treated with 1,25(OH)2D3

In Chapter 4, we used FXR knockout [fxr(-/-)] mice to discriminate whether the regulation of P-gp is via the activation of VDR or FXR, and assessed the impact of P-gp changes on drug disposition in the body (Chow et al., 2011a). We showed that there was no change in intestinal and liver P-gp, and observed induction of P-gp in the kidney and brain

183

after 1,25(OH)2D3 administration in both fxr(+/+) and fxr(-/-) mice, suggesting that the induction is independent of FXR. We then evaluated the change in P-gp function with the use of a whole body, physiologically-based pharmacokinetics (PBPK) model after

3 administrating a bolus dose of [ H]digoxin, a P-gp substrate, to 1,25(OH)2D3-treated fxr(+/+) mice. We fitted the data to a model and showed that the intrinsic secretory clearances of the kidney and brain were indeed increased by 1,25(OH)2D3. Although the renal intrinsic secretory clearance was increased about 3.5-fold, the apparent total body and renal clearances were increased only 34% and 74%, respectively. We further demonstrated by simulations that, despite the sharp increase in the renal intrinsic secretory clearance, there was only minimal impact on the overall total body and renal clearance since digoxin reabsorption was high (Chow et al., 2011a). Chapter 4 reported mechanistically on the impact of P-gp changes by VDR to overall changes in drug disposition.

There remains the uncertainty on whether activation of the VDR increases or decreases cholesterol metabolism in the liver in vivo. The role of VDR in regulating

CYP7A1 remains controversial across different animal species. Previous results illustrate in rat liver, which consists of very low VDR expression, decreased Cyp7a1 resulted with

1,25(OH)2D3 treatment indirect via the Asbt-FXR mechanism (Chapter 3) (Chow et al.,

2009). When the direct VDR and FXR effects were investigated directly, in absence of the contribution of the intestine, in rat liver slices (Khan et al., 2009b; Khan et al., 2011), rCyp7a were found to be unaffected by 1,25(OH)2D3 treatment (unpublished data) due to the low protein level of VDR in rat liver. In studies utilizing human hepatocytes and

HepG2, where VDR was shown to be present, showed a decrease (Han and Chiang, 2009) or a potential increase in CYP7A1 (Honjo et al., 2006) with 1,25(OH)2D3 treatment. Also 184

Han and Chiang (2009) showed that deletion of the bile acids response elements (BAREs) abolished the inhibitory effect of VDR on CYP7A1. But the study overlooked the potential interaction of VDR on other nuclear receptors such as FXR, SHP, and LRH-1, which may further result in changes in CYP7A1. In mice, Makishima’s group found that treatment of high doses of 1-hydroxyvitamin D3, a vitamin D analog, resulted in an upregulation of mCyp7a1 mRNA (Nishida et al., 2009; Ogura et al., 2009), findings that are similar to ours in Chapter 5. Hence, we suggest that the Cyp7a1 downregulation observed in mice given a single, high 1,25(OH)2D3 dose (Schmidt et al., 2010) to be attributed to the extremely high dose given, and the time-course in the determination of Cyp7a1 changes could be a contributing factor to the difference of induction vs. downregulation among the data. In

Chapter 6, we unequivocally showed that induction of Cyp7a1 by the VDR had indeed occurred under our low and protracted dosing regimen used for 1,25(OH)2D3 treatment.

Moreover, the observational Cyp7a1 changes in papers such as Schmidt et al. (2010) differed in doses compared to our study and failed to conclusively show whether the mRNA changes observed were due to VDR or FXR, or whether these changes could affect

Cyp7a1 protein and cholesterol levels. Moreover, the level of toxicity under high doses of

1,25(OH)2D3 on nuclear receptor responses need to be further clarified.

Thus, in Chapter 5, we selected the use of fxr(+/+) and fxr(-/-) mice to differentiate between the FXR vs. VDR effects, in order to isolate the VDR effects from those of FXR in vivo. We examined the amount of VDR in different tissues in mice and found that VDR mRNA in mouse liver was similar to that in humans and was higher than that in rat.

Surprisingly, Cyp7a1 was increased by 1,25(OH)2D3 in both fxr(+/+) and fxr(-/-) groups, which indicates that the induction of Cyp7a1 is independent of FXR. To investigate the 185

molecular mechanism of Cyp7a1 induction, we found that hepatic SHP mRNA expression was decreased in both fxr(+/+) and fxr(-/-) mice treated with 1,25(OH)2D3. A strong, negative correlation was observed between liver Cyp7a1 and SHP mRNA from individual mouse samples. In addition, primary mouse hepatocyte data showed similar increases in

Cyp7a1 and decreases in SHP mRNA expression after 1,25(OH)2D3 incubation for 9 h.

These results suggest that VDR plays a direct role in liver, and the decrease in SHP is responsible for the increase in Cyp7a1 expression. Further in Chapter 5, we showed that

VDR activation, downregulated mouse and human SHP promoter activities using luciferase assays, and that truncation of the human SHP promoter abolished this inhibition. We also showed that a putative VDRE region (-283) was found in the human SHP promoter responsible for VDR binding and its inhibitory effect. In addition, Chapter 5 revealed that

VDR activation in vivo lowered plasma and liver cholesterol concentrations though decreases in hepatic SHP and intestinal FGF15 expression in mouse models, such as fxr(-/-) mice, and wild-type and shp(-/-) fed a high fat/high cholesterol diet. Finally, data in

Chapter 5 demonstrated that VDR activation in vivo can potentially lower cholesterol in mice and humans. This was confirmed in human hepatocyte studies that showed CYP7A1 activation after 12 h of incubation with 100 nM 1,25(OH)2D3 (unpublished data, Fan and

Pang).

186

Liver 1,25(OH) D Cholesterol 2 3 VDR BA FXR

VDR SHP CYP7A1

FGFR4

Portal Blood

Bile

Enterocytes BA FGF15 1,25(OH)2D3 VDR FXR

Figure 7-2 VDR increases cholesterol metabolism and lowers cholesterol via repression of hepatic SHP (major mechanism) and possibly intestinal FGF15 (minor mechanism) in mice

In Chapter 6, we correlated gene changes observed in mice, described in Chapters 4 and 5, to 1,25(OH)2D3 concentrations in plasma and tissues. First, we found that multiple doses of 1,25(OH)2D3 to mice can increase its own elimination by the induction of Cyp24 in tissues, especially in the kidney. However, 1,25(OH)2D3 concentrations in tissues remained higher than their basal concentration after 1,25(OH)2D3 administration and that the induction of Mdr1 in kidney and Cyp7a1 in liver strongly correlated to the timing of administrated doses and the duration of treatment. The suggested molecular events of these increases have been previously explained in Chapter 4 and 5. Thus, Chapter 6 illustrates that VDR related gene changes correlate with the tissue levels of 1,25(OH)2D3.

187

In conclusion, this thesis reveals that activation of VDR varies between different organs and tissues. In rats, we showed that specific transporters and enzymes are upregulated by the VDR in the intestine, liver, and kidney, and that induction of intestinal

Asbt triggered secondary FXR effects in intestine and liver. Transporter changes also occurred via VDR activation and greatly affected the disposition of digoxin, a P-gp substrate that displayed altered renal and brain efflux. In mice where higher VDR protein levels are present in the liver compared to rats, VDR affects liver and intestine VDR targets by decreasing the negative repressors of Cyp7a1, hepatic SHP and intestinal FGF15, to lower plasma and liver cholesterol concentrations. These changes correlate to the concentrations of 1,25(OH)2D3 in these tissues. My dissertation has demonstrated that VDR activation in vivo displays beneficial effects in the liver to treat hypercholesterolemia and in the brain to increase the excretion of brain -amyloids, a potential risk factor in Alzheimer's disease.

Future Directions

We have established that 1,25(OH)2D3 is an effective cholesterol lowering agent, and the concept may lead to a new therapeutic class of cholesterol lowering drugs with a new underlying mechanism. Parallel changes need to be established in humans with respect to cholesterol lowering. There are presently VDR ligands that are vitamin D analogues, and alternate VDR ligands such as the lithocholic acid derivatives that do not cause hypercalcemia (Ishizawa et al., 2008). Thus more studies are needed to examine these compounds or alternate ligands and their potential role in humans. More studies are also needed to dissect the lipoprotein fraction (LDL vs. HDL) that was lowered in plasma of mouse in previous studies. In addition, a combination therapy with VDR agonists and 188

statins, such as atorvastatin can be examined to investigate the potential synergistic effect, because a study has shown that a combination of vitamin D and atorvastatin greatly increases the lowering of cholesterol (Schwartz, 2009). The proposed combined use of

VDR ligands as adjunct drugs for therapy of hypercholesterolemia may be invaluable in the prevention and treatment of coronary heart diseases.

189

REFERENCES

Abrahamsson A, Gustafsson U, Ellis E, Nilsson LM, Sahlin S, Bjorkhem I and Einarsson C (2005) Feedback regulation of bile acid synthesis in human liver: importance of HNF-4 for regulation of CYP7A1. Biochem Biophys Res Commun 330:395-399. Adesnik M, Bar-Nun S, Maschio F, Zunich M, Lippman A and Bard E (1981) Mechanism of induction of cytochrome P-450 by phenobarbital. J Biol Chem 256:10340-10345. Agellon LB and Cheema SK (1997) The 3'-untranslated region of the mouse cholesterol 7-hydroxylase mRNA contains elements responsive to post-transcriptional regulation by bile acids. Biochem J 328 ( Pt 2):393-399. Aiba T, Susa M, Fukumori S and Hashimoto Y (2005) The effects of culture conditions on CYP3A4 and MDR1 mRNA induction by 1α,25-dihydroxyvitamin D3 in human intestinal cell lines, Caco-2 and LS180. Drug Metab Pharmacokinet 20:268-274. Akita H, Suzuki H and Sugiyama Y (2001) Sinusoidal efflux of taurocholate is enhanced in Mrp2-deficient rat liver. Pharm Res 18:1119-1125. Akita H, Suzuki H and Sugiyama Y (2002) Sinusoidal efflux of taurocholate correlates with the hepatic expression level of Mrp3. Biochem Biophys Res Commun 299:681- 687. Almerighi C, Sinistro A, Cavazza A, Ciaprini C, Rocchi G and Bergamini A (2009) 1,25- dihydroxyvitamin D3 inhibits CD40L-induced pro-inflammatory and immunomodulatory activity in human monocytes. Cytokine 45:190-197. Alsheikh-Ali AA, Maddukuri PV, Han H and Karas RH (2007) Effect of the magnitude of lipid lowering on risk of elevated liver enzymes, rhabdomyolysis, and cancer: insights from large randomized statin trials. J Am Coll Cardiol 50:409-418. Anderson PH, Hendrix I, Sawyer RK, Zarrinkalam R, Manavis J, Sarvestani GT, May BK and Morris HA (2008) Co-expression of CYP27B1 enzyme with the 1.5kb CYP27B1 promoter-luciferase transgene in the mouse. Mol Cell Endocrinol 285:1- 9. Andress DL (2006) Vitamin D in chronic kidney disease: a systemic role for selective vitamin D receptor activation. Kidney Int 69:33-43. Araya Z and Wikvall K (1999) 6-hydroxylation of taurochenodeoxycholic acid and lithocholic acid by CYP3A4 in human liver microsomes. Biochim Biophys Acta 1438:47-54. Banach-Petrosky W, Ouyang X, Gao H, Nader K, Ji Y, Suh N, DiPaola RS and Abate-Shen C (2006) Vitamin D inhibits the formation of prostatic intraepithelial neoplasia in Nkx3.1;Pten mutant mice. Clin Cancer Res 12:5895-5901. Bligh EG and Dyer WJ (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37:911-917. Boggaram V, Funkenstein B, Waterman MR and Simpson ER (1984) Lipoproteins and the regulation of adrenal steroidogenesis. Endocr Res 10:387-409. Bookout AL, Jeong Y, Downes M, Yu RT, Evans RM and Mangelsdorf DJ (2006) Anatomical profiling of nuclear receptor expression reveals a hierarchical transcriptional network. Cell 126:789-799.

190

Brendel C, Schoonjans K, Botrugno OA, Treuter E and Auwerx J (2002) The small heterodimer partner interacts with the liver X receptor alpha and represses its transcriptional activity. Mol Endocrinol 16:2065-2076. Brooks EL, Kuvin JT and Karas RH (2010) Niacin's role in the statin era. Expert Opin Pharmacother 11:2291-2300. Brown AJ, Ritter CS, Holliday LS, Knutson JC and Strugnell SA (2004) Tissue distribution and activity studies of 1,24-dihydroxyvitamin D2, a metabolite of vitamin D2 with low calcemic activity in vivo. Biochem Pharmacol 68:1289-1296. Brown AJ and Slatopolsky E (2008) Vitamin D analogs: therapeutic applications and mechanisms for selectivity. Mol Aspects Med 29:433-452. Chandra P, Zhang P and Brouwer KL (2005) Short-term regulation of multidrug resistance- associated protein 3 in rat and human hepatocytes. Am J Physiol Gastrointest Liver Physiol 288:G1252-1258. Chen F, Ma L, Dawson PA, Sinal CJ, Sehayek E, Gonzalez FJ, Breslow J, Ananthanarayanan M and Shneider BL (2003) Liver receptor homologue-1 mediates species- and cell line-specific bile acid-dependent negative feedback regulation of the apical sodium-dependent bile acid transporter. J Biol Chem 278:19909-19916. Chen KS and DeLuca HF (1995) Cloning of the human 1,25-dihydroxyvitamin D3 24- hydroxylase gene promoter and identification of two vitamin D-responsive elements. Biochim Biophys Acta 1263:1-9. Chen X, Chen F, Liu S, Glaeser H, Dawson PA, Hofmann AF, Kim RB, Shneider BL and Pang KS (2006) Transactivation of rat apical sodium-dependent bile acid transporter and increased bile acid transport by 1α,25-dihydroxyvitamin D3 via the vitamin D receptor. Mol Pharmacol 69:1913-1923. Cheng JB, Motola DL, Mangelsdorf DJ and Russell DW (2003) De-orphanization of cytochrome P450 2R1: a microsomal vitamin D 25-hydroxylase. J Biol Chem 278:38084-38093. Chiang JY (1998) Regulation of bile acid synthesis. Front Biosci 3:d176-193. Chiang JY (2002) Bile acid regulation of gene expression: roles of nuclear hormone receptors. Endocr Rev 23:443-463. Chiang JY (2003) Bile acid regulation of hepatic physiology: III. Bile acids and nuclear receptors. Am J Physiol Gastrointest Liver Physiol 284:G349-356. Chiang JY (2004) Regulation of bile acid synthesis: pathways, nuclear receptors, and mechanisms. J Hepatol 40:539-551. Chiang JY, Kimmel R and Stroup D (2001) Regulation of cholesterol 7α-hydroxylase gene (CYP7A1) transcription by the liver orphan receptor (LXRα). Gene 262:257-265. Chiang JY, Kimmel R, Weinberger C and Stroup D (2000) Farnesoid X receptor responds to bile acids and represses cholesterol 7-hydroxylase gene (CYP7A1) transcription. J Biol Chem 275:10918-10924. Chiang JY and Stroup D (1994) Identification and characterization of a putative bile acid- responsive element in cholesterol 7-hydroxylase gene promoter. J Biol Chem 269:17502-17507. Chiang KC and Chen TC (2009) Vitamin D for the prevention and treatment of pancreatic cancer. World J Gastroenterol 15:3349-3354.

191

Cho BH (1983) Improved enzymatic determination of total cholesterol in tissues. Clin Chem 29:166-168. Chow EC, Durk MR, Cummins CL and Pang KS (2011a) 1,25-dihydroxyvitamin D3 upregulates P-glycoprotein via the vitamin D receptor and not farnesoid X receptor in both fxr(-/-) and fxr(+/+) mice and increased renal and brain efflux of digoxin in mice in vivo. J Pharmacol Exp Ther 337:846-859. Chow EC, Maeng HJ, Liu S, Khan AA, Groothuis GM and Pang KS (2009) 1,25- Dihydroxyvitamin D3 triggered vitamin D receptor and farnesoid X receptor-like effects in rat intestine and liver in vivo. Biopharm Drug Dispos 30:457-475. Chow EC, Sondervan M, Jin C, Groothuis GM and Pang KS (2011b) Comparative effects of doxercalciferol (1-hydroxyvitamin D2) versus calcitriol (1,25- dihydroxyvitamin D3) on the expression of transporters and enzymes in the rat in vivo. J Pharm Sci 100:1594-1604. Chow EC, Sun H, Khan AA, Groothuis GM and Pang KS (2010) Effects of 1,25- dihydroxyvitamin D3 on transporters and enzymes of the rat intestine and kidney in vivo. Biopharm Drug Dispos 31:91-108. Colston KW, Chander SK, Mackay AG and Coombes RC (1992) Effects of synthetic vitamin D analogues on breast cancer cell proliferation in vivo and in vitro. Biochem Pharmacol 44:693-702. Cong D, Doherty M and Pang KS (2000) A new physiologically based, segregated-flow model to explain route-dependent intestinal metabolism. Drug Metab Dispos 28:224-235. Cooke NE and Haddad JG (1989) Vitamin D binding protein (Gc-globulin). Endocr Rev 10:294-307. Cordon-Cardo C, O'Brien JP, Boccia J, Casals D, Bertino JR and Melamed MR (1990) Expression of the multidrug resistance gene product (P-glycoprotein) in human normal and tumor tissues. J Histochem Cytochem 38:1277-1287. Crestani M, Sadeghpour A, Stroup D, Galli G and Chiang JY (1998) Transcriptional activation of the cholesterol 7α-hydroxylase gene (CYP7A) by nuclear hormone receptors. J Lipid Res 39:2192-2200. Cucinell SA, Koster R, Conney AH and Burns JJ (1963) Stimulatory Effect of Phenobarbital on the Metabolism of Diphenylhydantoin. J Pharmacol Exp Ther 141:157-160. Cutolo M, Otsa K, Uprus M, Paolino S and Seriolo B (2007) Vitamin D in rheumatoid arthritis. Autoimmun Rev 7:59-64. Dawson PA, Hubbert M, Haywood J, Craddock AL, Zerangue N, Christian WV and Ballatori N (2005) The heteromeric organic solute transporter -, Ost-Ost, is an ileal basolateral bile acid transporter. J Biol Chem 280:6960-6968. Dawson PA, Lan T and Rao A (2009) Bile acid transporters. J Lipid Res 50:2340-2357. Dawson PA and Markovich D (2002) Regulation of the mouse Nas1 promoter by vitamin D and thyroid hormone. Pflugers Arch 444:353-359. Dean M and Allikmets R (2001) Complete characterization of the human ABC gene family. J Bioenerg Biomembr 33:475-479. Deeb KK, Trump DL and Johnson CS (2007) Vitamin D signalling pathways in cancer: potential for anticancer therapeutics. Nat Rev Cancer 7:684-700. 192

DeLuca HF (1988) The vitamin D story: a collaborative effort of basic science and clinical medicine. FASEB J 2:224-236. den Dekker E, Hoenderop JG, Nilius B and Bindels RJ (2003) The epithelial calcium channels, TRPV5 & TRPV6: from identification towards regulation. Cell Calcium 33:497-507. Dresser GK, Spence JD and Bailey DG (2000) Pharmacokinetic-pharmacodynamic consequences and clinical relevance of cytochrome P450 3A4 inhibition. Clin Pharmacokinet 38:41-57. Drocourt L, Ourlin JC, Pascussi JM, Maurel P and Vilarem MJ (2002) Expression of CYP3A4, CYP2B6, and CYP2C9 is regulated by the vitamin D receptor pathway in primary human hepatocytes. J Biol Chem 277:25125-25132. Drover VA and Agellon LB (2004) Regulation of the human cholesterol 7-hydroxylase gene (CYP7A1) by thyroid hormone in transgenic mice. Endocrinology 145:574- 581. Durk MR, Chan GN, Fung AK, Bendayan R and Pang KS (2011a) The vitamin D receptor increases p-glycoprotein expression, leading to enhanced clearance of p- glycoprotein (p-gp) substrates in rodent and human brain capillary endothelial cell lines. 2011 AAPS Annual Meeting, Washington, DC. Poster#1272. Durk MR, Chow EC and Pang KS (2011b) Regulation of brain p-glycoprotein (p-gp) by the vitamin D receptor (VDR) leads to reduced accumulation of beta amyloids. 2011 AAPS National Biotechnology Conference, San Francisco, CA. Poster#T2103. Dusso AS, Brown AJ and Slatopolsky E (2005) Vitamin D. Am J Physiol Renal Physiol 289:F8-28. Echchgadda I, Song CS, Roy AK and Chatterjee B (2004) Dehydroepiandrosterone sulfotransferase is a target for transcriptional induction by the vitamin D receptor. Mol Pharmacol 65:720-729. Eloranta JJ, Zair ZM, Hiller C, Hausler S, Stieger B and Kullak-Ublick GA (2009) Vitamin D3 and its nuclear receptor increase the expression and activity of the human proton-coupled folate transporter. Mol Pharmacol 76:1062-1071. Enomoto H, Hendy GN, Andrews GK and Clemens TL (1992) Regulation of avian calbindin-D28K gene expression in primary chick kidney cells: importance of posttranscriptional mechanisms and calcium ion concentration. Endocrinology 130:3467-3474. Eyles DW, Smith S, Kinobe R, Hewison M and McGrath JJ (2005) Distribution of the vitamin D receptor and 1 alpha-hydroxylase in human brain. J Chem Neuroanat 29:21-30. Fan J, Liu S, Du Y, Morrison J, Shipman R and Pang KS (2009) Upregulation of transporters and enzymes by the vitamin D receptor ligands, 1,25- dihydroxyvitamin D3 and vitamin D analogs, in the Caco-2 cell monolayer. J Pharmacol Exp Ther 330:389-402. Folch J, Lees M and Sloane Stanley GH (1957) A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 226:497-509. Forman JP, Giovannucci E, Holmes MD, Bischoff-Ferrari HA, Tworoger SS, Willett WC and Curhan GC (2007) Plasma 25-hydroxyvitamin D levels and risk of incident hypertension. Hypertension 49:1063-1069.

193

Fraser DR and Kodicek E (1970) Unique biosynthesis by kidney of a biological active vitamin D metabolite. Nature 228:764-766. Fredriksson R, Nordstrom KJ, Stephansson O, Hagglund MG and Schioth HB (2008) The solute carrier (SLC) complement of the human genome: phylogenetic classification reveals four major families. FEBS Lett 582:3811-3816. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL and Portale AA (1997) Cloning of human 25-hydroxyvitamin D-1 -hydroxylase and mutations causing vitamin D-dependent rickets type 1. Mol Endocrinol 11:1961-1970. Furster C and Wikvall K (1999) Identification of CYP3A4 as the major enzyme responsible for 25-hydroxylation of 5-cholestane-3,7,12-triol in human liver microsomes. Biochim Biophys Acta 1437:46-52. Gascon-Barré M, Demers C, Mirshahi A, Neron S, Zalzal S and Nanci A (2003) The normal liver harbors the vitamin D nuclear receptor in nonparenchymal and biliary epithelial cells. Hepatology 37:1034-1042. Gielen J, Van Cantfort J, Robaye B and Renson J (1975) Rat-liver cholesterol 7- hydroxylase. 3. New results about its circadian rhythm. Eur J Biochem 55:41-48. Goodwin B, Hodgson E, D'Costa DJ, Robertson GR and Liddle C (2002) Transcriptional regulation of the human CYP3A4 gene by the constitutive androstane receptor. Mol Pharmacol 62:359-365. Goodwin B, Jones SA, Price RR, Watson MA, McKee DD, Moore LB, Galardi C, Wilson JG, Lewis MC, Roth ME, Maloney PR, Willson TM and Kliewer SA (2000) A regulatory cascade of the nuclear receptors FXR, SHP-1, and LRH-1 represses bile acid biosynthesis. Mol Cell 6:517-526. Goodwin B, Watson MA, Kim H, Miao J, Kemper JK and Kliewer SA (2003) Differential regulation of rat and human CYP7A1 by the nuclear oxysterol receptor liver X receptor-alpha. Mol Endocrinol 17:386-394. Grant DM, Blum M, Beer M and Meyer UA (1991) Monomorphic and polymorphic human arylamine N-acetyltransferases: a comparison of liver isozymes and expressed products of two cloned genes. Mol Pharmacol 39:184-191. Griffiths NM, Hewick DS and Stevenson IH (1984) The effect of immunization with digoxin-specific antibodies on digoxin disposition in the mouse. Biochem Pharmacol 33:3041-3046. Guengerich FP (1999) Cytochrome P-450 3A4: regulation and role in drug metabolism. Annu Rev Pharmacol Toxicol 39:1-17. Guo GL, Lambert G, Negishi M, Ward JM, Brewer HB, Jr., Kliewer SA, Gonzalez FJ and Sinal CJ (2003) Complementary roles of farnesoid X receptor, pregnane X receptor, and constitutive androstane receptor in protection against bile acid toxicity. J Biol Chem 278:45062-45071. Gupta A, Guyomard V, Zaman MJ, Rehman HU and Myint PK (2010) Systematic review on evidence of the effectiveness of cholesterol-lowering drugs. Adv Ther 27:348- 364. Gupta S, Pandak WM and Hylemon PB (2002) LXR alpha is the dominant regulator of CYP7A1 transcription. Biochem Biophys Res Commun 293:338-343.

194

Gupta S, Stravitz RT, Dent P and Hylemon PB (2001) Downregulation of cholesterol 7- hydroxylase (CYP7A1) gene expression by bile acids in primary rat hepatocytes is mediated by the c-Jun N-terminal kinase pathway. J Biol Chem 276:15816-15822. Hagenbuch B and Meier PJ (1994) Molecular cloning, chromosomal localization, and functional characterization of a human liver Na+/bile acid cotransporter. J Clin Invest 93:1326-1331. Hagenbuch B and Meier PJ (2004) Organic anion transporting polypeptides of the OATP/ SLC21 family: phylogenetic classification as OATP/ SLCO superfamily, new nomenclature and molecular/functional properties. Pflugers Arch 447:653-665. Hamden K, Carreau S, Jamoussi K, Miladi S, Lajmi S, Aloulou D, Ayadi F and Elfeki A (2009) 1,25 dihydroxyvitamin D3: therapeutic and preventive effects against oxidative stress, hepatic, pancreatic and renal injury in alloxan-induced diabetes in rats. J Nutr Sci Vitaminol (Tokyo) 55:215-222. Han S and Chiang JY (2009) Mechanism of vitamin D receptor inhibition of cholesterol 7-hydroxylase gene transcription in human hepatocytes. Drug Metab Dispos 37:469-478. Han S, Li T, Ellis E, Strom S and Chiang JY (2010) A novel bile acid-activated vitamin D receptor signaling in human hepatocytes. Mol Endocrinol 24:1151-1164. Hartman HB, Lai K and Evans MJ (2009) Loss of small heterodimer partner expression in the liver protects against dyslipidemia. J Lipid Res 50:193-203. Heaney RP, Horst RL, Cullen DM and Armas LA (2009) Vitamin D3 distribution and status in the body. J Am Coll Nutr 28:252-256. Henry HL and Norman AW (1984) Vitamin D: metabolism and biological actions. Annu Rev Nutr 4:493-520. Hershberger PA, McGuire TF, Yu WD, Zuhowski EG, Schellens JH, Egorin MJ, Trump DL and Johnson CS (2002) Cisplatin potentiates 1,25-dihydroxyvitamin D3-induced apoptosis in association with increased mitogen-activated protein kinase kinase kinase 1 (MEKK-1) expression. Mol Cancer Ther 1:821-829. Ho RH, Choi L, Lee W, Mayo G, Schwarz UI, Tirona RG, Bailey DG, Michael Stein C and Kim RB (2007) Effect of drug transporter genotypes on pravastatin disposition in European- and African-American participants. Pharmacogenet Genomics 17:647- 656. Ho RH, Tirona RG, Leake BF, Glaeser H, Lee W, Lemke CJ, Wang Y and Kim RB (2006) Drug and bile acid transporters in rosuvastatin hepatic uptake: function, expression, and pharmacogenetics. Gastroenterology 130:1793-1806. Hofmann AF (1999) Bile Acids: The Good, the Bad, and the Ugly. News Physiol Sci 14:24-29. Holick MF (2005) Stay tuned to PXR: an orphan actor that may not be D-structive only to bone. J Clin Invest 115:32-34. Honjo Y, Sasaki S, Kobayashi Y, Misawa H and Nakamura H (2006) 1,25- Dihydroxyvitamin D3 and its receptor inhibit the chenodeoxycholic acid-dependent transactivation by farnesoid X receptor. J Endocrinol 188:635-643. Inagaki T, Choi M, Moschetta A, Peng L, Cummins CL, McDonald JG, Luo G, Jones SA, Goodwin B, Richardson JA, Gerard RD, Repa JJ, Mangelsdorf DJ and Kliewer SA

195

(2005) Fibroblast growth factor 15 functions as an enterohepatic signal to regulate bile acid homeostasis. Cell Metab 2:217-225. Ishiguro N, Maeda K, Kishimoto W, Saito A, Harada A, Ebner T, Roth W, Igarashi T and Sugiyama Y (2006) Predominant contribution of OATP1B3 to the hepatic uptake of telmisartan, an angiotensin II receptor antagonist, in humans. Drug Metab Dispos 34:1109-1115. Ishizawa M, Matsunawa M, Adachi R, Uno S, Ikeda K, Masuno H, Shimizu M, Iwasaki K, Yamada S and Makishima M (2008) Lithocholic acid derivatives act as selective vitamin D receptor modulators without inducing hypercalcemia. J Lipid Res 49:763-772. Javitt NB (1994) Bile acid synthesis from cholesterol: regulatory and auxiliary pathways. FASEB J 8:1308-1311. Jia L, Ma Y, Rong S, Betters JL, Xie P, Chung S, Wang N, Tang W and Yu L (2010) Niemann-Pick C1-Like 1 deletion in mice prevents high-fat diet-induced fatty liver by reducing lipogenesis. J Lipid Res 51:3135-3144. Jiang W, Miyamoto T, Kakizawa T, Nishio SI, Oiwa A, Takeda T, Suzuki S and Hashizume K (2006) Inhibition of LXRα signaling by vitamin D receptor: possible role of VDR in bile acid synthesis. Biochem Biophys Res Commun 351:176-184. Johnson BM, Zhang P, Schuetz JD and Brouwer KL (2006) Characterization of transport protein expression in multidrug resistance-associated protein (Mrp) 2-deficient rats. Drug Metab Dispos 34:556-562. Jones G, Strugnell SA and DeLuca HF (1998) Current understanding of the molecular actions of vitamin D. Physiol Rev 78:1193-1231. Juliano RL and Ling V (1976) A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 455:152-162. Jung D, Hagenbuch B, Fried M, Meier PJ and Kullak-Ublick GA (2004) Role of liver- enriched transcription factors and nuclear receptors in regulating the human, mouse, and rat NTCP gene. Am J Physiol Gastrointest Liver Physiol 286:G752-761. Kamei Y, Kawada T, Fukuwatari T, Ono T, Kato S and Sugimoto E (1995) Cloning and sequencing of the gene encoding the mouse vitamin D receptor. Gene 152:281-282. Katona BW, Anant S, Covey DF and Stenson WF (2009) Characterization of enantiomeric bile acid-induced apoptosis in colon cancer cell lines. J Biol Chem 284:3354-3364. Kawahara M, Sakata A, Miyashita T, Tamai I and Tsuji A (1999) Physiologically based pharmacokinetics of digoxin in mdr1a knockout mice. J Pharm Sci 88:1281-1287. Khan AA, Chow EC, Porte RJ, Pang KS and Groothuis GM (2009a) Expression and regulation of the bile acid transporter, OST-OST in rat and human intestine and liver. Biopharm Drug Dispos 30:241-258. Khan AA, Chow EC, Porte RJ, Pang KS and Groothuis GM (2011) The role of lithocholic acid in the regulation of bile acid detoxication, synthesis, and transport proteins in rat and human intestine and liver slices. Toxicol In Vitro 25:80-90. Khan AA, Chow EC, van Loenen-Weemaes AM, Porte RJ, Pang KS and Groothuis GM (2009b) Comparison of effects of VDR versus PXR, FXR and GR ligands on the regulation of CYP3A isozymes in rat and human intestine and liver. Eur J Pharm Sci 37:115-125.

196

Kim RB (2002a) Drugs as P-glycoprotein substrates, inhibitors, and inducers. Drug Metab Rev 34:47-54. Kim RB (2002b) Transporters and xenobiotic disposition. Toxicology 181-182:291-297. Klaassen CD and Aleksunes LM (2011) Xenobiotic, bile acid, and cholesterol transporters: function and regulation. Pharmacol Rev 62:1-96. König J, Seithel A, Gradhand U and Fromm MF (2006) Pharmacogenomics of human OATP transporters. Naunyn Schmiedebergs Arch Pharmacol 372:432-443. Kullak-Ublick GA, Stieger B, Hagenbuch B and Meier PJ (2000) Hepatic transport of bile salts. Semin Liver Dis 20:273-292. Kutuzova GD and DeLuca HF (2007) 1,25-Dihydroxyvitamin D3 regulates genes responsible for detoxification in intestine. Toxicol Appl Pharmacol 218:37-44. Lazaridis KN, Pham L, Tietz P, Marinelli RA, deGroen PC, Levine S, Dawson PA and LaRusso NF (1997) Rat cholangiocytes absorb bile acids at their apical domain via the ileal sodium-dependent bile acid transporter. J Clin Invest 100:2714-2721. Lee HK, Lee YK, Park SH, Kim YS, Lee JW, Kwon HB, Soh J, Moore DD and Choi HS (1998) Structure and expression of the orphan nuclear receptor SHP gene. J Biol Chem 273:14398-14402. Lee JH, O'Keefe JH, Bell D, Hensrud DD and Holick MF (2008) Vitamin D deficiency an important, common, and easily treatable cardiovascular risk factor? J Am Coll Cardiol 52:1949-1956. Lee YK and Moore DD (2002) Dual mechanisms for repression of the monomeric orphan receptor liver receptor homologous protein-1 by the orphan small heterodimer partner. J Biol Chem 277:2463-2467. Leon C, Sachs-Barrable K and Wasan KM (2006) Does p-glycoprotein play a role in gastrointestinal absorption and cellular transport of dietary cholesterol? Drug Dev Ind Pharm 32:779-782. Levy G (1980) Effect of plasma protein binding on renal clearance of drugs. J Pharm Sci 69:482-483. Li YC, Kong J, Wei M, Chen ZF, Liu SQ and Cao LP (2002) 1,25-Dihydroxyvitamin D3 is a negative endocrine regulator of the renin-angiotensin system. J Clin Invest 110:229-238. Liu L, Cui Y, Chung AY, Shitara Y, Sugiyama Y, Keppler D and Pang KS (2006a) Vectorial transport of enalapril by Oatp1a1/Mrp2 and OATP1B1 and OATP1B3/MRP2 in rat and human livers. J Pharmacol Exp Ther 318:395-402. Liu S, Tam D, Chen X and Pang KS (2006b) P-glycoprotein and an unstirred water layer barring digoxin absorption in the vascularly perfused rat small intestine preparation: induction studies with pregnenolone-16α-carbonitrile. Drug Metab Dispos 34:1468- 1479. Los EL, Wolters H, Stellaard F, Kuipers F, Verkade HJ and Rings EH (2007) Intestinal capacity to digest and absorb carbohydrates is maintained in a rat model of cholestasis. Am J Physiol Gastrointest Liver Physiol 293:G615-622. Lowry OH, Rosebrough NJ, Farr AL and Randall RJ (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193:265-275. Lu TT, Makishima M, Repa JJ, Schoonjans K, Kerr TA, Auwerx J and Mangelsdorf DJ (2000) Molecular basis for feedback regulation of bile acid synthesis by nuclear receptors. Mol Cell 6:507-515. 197

Luker GD, Dahlheimer JL, Ostlund RE, Jr. and Piwnica-Worms D (2001) Decreased hepatic accumulation and enhanced esterification of cholesterol in mice deficient in mdr1a and mdr1b P-glycoproteins. J Lipid Res 42:1389-1394. Luna-Tortos C, Rambeck B, Jurgens UH and Loscher W (2009) The antiepileptic drug topiramate is a substrate for human P-glycoprotein but not multidrug resistance proteins. Pharm Res 26:2464-2470. Ma KY, Yang N, Jiao R, Peng C, Guan L, Huang Y and Chen ZY (2011) Dietary calcium decreases plasma cholesterol by downregulation of intestinal Niemann-Pick C1 like 1 and microsomal triacylglycerol transport protein and upregulation of CYP7A1 and ABCG 5/8 in hamsters. Mol Nutr Food Res 55:247-258. Maeng HJ, Durk MR, Chow EC, Ghoneim R and Pang KS (2011) 1,25-Dihydroxyvitamin D3 on intestinal transporter function: studies with the rat everted intestinal sac. Biopharm Drug Dispos 32:112-125. Makishima M (2005) Nuclear receptors as targets for drug development: regulation of cholesterol and bile acid metabolism by nuclear receptors. J Pharmacol Sci 97:177- 183. Makishima M, Lu TT, Xie W, Whitfield GK, Domoto H, Evans RM, Haussler MR and Mangelsdorf DJ (2002) Vitamin D receptor as an intestinal bile acid sensor. Science 296:1313-1316. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schutz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P and Evans RM (1995) The nuclear receptor superfamily: the second decade. Cell 83:835-839. Marks HD, Fleet JC and Peleg S (2007) Transgenic expression of the human Vitamin D receptor (hVDR) in the duodenum of VDR-null mice attenuates the age-dependent decline in calcium absorption. J Steroid Biochem Mol Biol 103:513-516. Massheimer V, Boland R and de Boland AR (1994) Rapid 1,25(OH)2-vitamin D3 stimulation of calcium uptake by rat intestinal cells involves a dihydropyridine- sensitive cAMP-dependent pathway. Cell Signal 6:299-304. Masson D (2009) Anacetrapib, a cholesterol ester transfer protein (CETP) inhibitor for the treatment of atherosclerosis. Curr Opin Investig Drugs 10:980-987. Masuda S and Jones G (2006) Promise of vitamin D analogues in the treatment of hyperproliferative conditions. Mol Cancer Ther 5:797-808. McCarthy TC, Li X and Sinal CJ (2005) Vitamin D receptor-dependent regulation of colon multidrug resistance-associated protein 3 gene expression by bile acids. J Biol Chem 280:23232-23242. Meyer MB, Zella LA, Nerenz RD and Pike JW (2007) Characterizing early events associated with the activation of target genes by 1,25-dihydroxyvitamin D3 in mouse kidney and intestine in vivo. J Biol Chem 282:22344-22352. Miao J, Xiao Z, Kanamaluru D, Min G, Yau PM, Veenstra TD, Ellis E, Strom S, Suino- Powell K, Xu HE and Kemper JK (2009) Bile acid signaling pathways increase stability of small heterodimer partner (SHP) by inhibiting ubiquitin-proteasomal degradation. Genes Dev 23:986-996. Morimoto K, Nakakariya M, Shirasaka Y, Kakinuma C, Fujita T, Tamai I and Ogihara T (2008) Oseltamivir (Tamiflu) efflux transport at the blood-brain barrier via P- glycoprotein. Drug Metab Dispos 36:6-9.

198

Mottino AD, Hoffman T, Jennes L and Vore M (2000) Expression and localization of multidrug resistant protein mrp2 in rat small intestine. J Pharmacol Exp Ther 293:717-723. Nakagawa K, Kawaura A, Kato S, Takeda E and Okano T (2005) 1,25-Dihydroxyvitamin D3 is a preventive factor in the metastasis of lung cancer. Carcinogenesis 26:429- 440. Nehring JA, Zierold C and DeLuca HF (2007) Lithocholic acid can carry out in vivo functions of vitamin D. Proc Natl Acad Sci U S A 104:10006-10009. Nishida S, Ozeki J and Makishima M (2009) Modulation of bile acid metabolism by 1- hydroxyvitamin D3 administration in mice. Drug Metab Dispos 37:2037-2044. O'Connell TD, Berry JE, Jarvis AK, Somerman MJ and Simpson RU (1997) 1,25- Dihydroxyvitamin D3 regulation of cardiac myocyte proliferation and hypertrophy. Am J Physiol 272:H1751-1758. O'Connell TD, Giacherio DA, Jarvis AK and Simpson RU (1995) Inhibition of cardiac myocyte maturation by 1,25-dihydroxyvitamin D3. Endocrinology 136:482-488. Ogura M, Nishida S, Ishizawa M, Sakurai K, Shimizu M, Matsuo S, Amano S, Uno S and Makishima M (2009) Vitamin D3 modulates the expression of bile acid regulatory genes and represses inflammation in bile duct-ligated mice. J Pharmacol Exp Ther 328:564-570. Olefsky JM (2001) Nuclear receptor minireview series. J Biol Chem 276:36863-36864. Omura T (1999) Forty years of cytochrome P450. Biochem Biophys Res Commun 266:690- 698. Pang KS, Rodrigues AD, Peter RM and ebrary Inc. (2010) Enzyme- and transporter-based drug-drug interactions progress and future challenges, pp xviii, 746 p., Springer : American Association of Pharmaceutical Scientists, New York. Patel J, Pal D, Vangal V, Gandhi M and Mitra AL (2002) Transport of HIV-protease inhibitors across 1,25di-hydroxy vitamin D3-treated Calu-3 cell monolayers: modulation of P-glycoprotein activity. Pharm Res 19:1696-1703. Peet DJ, Turley SD, Ma W, Janowski BA, Lobaccaro JM, Hammer RE and Mangelsdorf DJ (1998) Cholesterol and bile acid metabolism are impaired in mice lacking the nuclear oxysterol receptor LXR alpha. Cell 93:693-704. Pfrunder A, Gutmann H, Beglinger C and Drewe J (2003) Gene expression of CYP3A4, ABC-transporters (MDR1 and MRP1-MRP5) and hPXR in three different human colon carcinoma cell lines. J Pharm Pharmacol 55:59-66. Racanelli V and Rehermann B (2006) The liver as an immunological organ. Hepatology 43:S54-62. Radcliffe KA and Campbell WW (2008) Statin myopathy. Curr Neurol Neurosci Rep 8:66- 72. Rahman A, Hershey S, Ahmed S, Nibbelink K and Simpson RU (2007) Heart extracellular matrix gene expression profile in the vitamin D receptor knockout mice. J Steroid Biochem Mol Biol 103:416-419. Rao A, Haywood J, Craddock AL, Belinsky MG, Kruh GD and Dawson PA (2008) The organic solute transporter α-β, Ostα-Ostβ, is essential for intestinal bile acid transport and homeostasis. Proc Natl Acad Sci U S A 105:3891-3896.

199

Reddy GS and Tserng KY (1989) Calcitroic acid, end product of renal metabolism of 1,25- dihydroxyvitamin D3 through C-24 oxidation pathway. Biochemistry 28:1763-1769. Reis JP, von Muhlen D, Miller ER, 3rd, Michos ED and Appel LJ (2009) Vitamin D status and cardiometabolic risk factors in the United States adolescent population. Pediatrics 124:e371-379. Remmer H, Schoene B and Fleischmann RA (1973) Induction of the unspecific microsomal hydroxylase in the human liver. Drug Metab Dispos 1:224-230. Reschly EJ and Krasowski MD (2006) Evolution and function of the NR1I nuclear hormone receptor subfamily (VDR, PXR, and CAR) with respect to metabolism of xenobiotics and endogenous compounds. Curr Drug Metab 7:349-365. Rezen T (2011) The impact of cholesterol and its metabolites on drug metabolism. Expert Opin Drug Metab Toxicol 7:387-398. Rezen T, Rozman D, Pascussi JM and Monostory K (2010) Interplay between cholesterol and drug metabolism. Biochim Biophys Acta 1814:146-160. Ritter JK (2000) Roles of glucuronidation and UDP-glucuronosyltransferases in xenobiotic bioactivation reactions. Chem Biol Interact 129:171-193. Russell DW and Setchell KD (1992) Bile acid biosynthesis. Biochemistry 31:4737-4749. Sacchettini JC, Hauft SM, Van Camp SL, Cistola DP and Gordon JI (1990) Developmental and structural studies of an intracellular lipid binding protein expressed in the ileal epithelium. J Biol Chem 265:19199-19207. Saeki M, Kurose K, Tohkin M and Hasegawa R (2008) Identification of the functional vitamin D response elements in the human MDR1 gene. Biochem Pharmacol 76:531-542. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA and Cooke NE (1999) Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103:239-251. Sakuma T, Miyamoto T, Jiang W, Kakizawa T, Nishio SI, Suzuki S, Takeda T, Oiwa A and Hashizume K (2003) Inhibition of peroxisome proliferator-activated receptor  signaling by vitamin D receptor. Biochem Biophys Res Commun 312:513-519. Sandgren ME, Bronnegard M and DeLuca HF (1991) Tissue distribution of the 1,25- dihydroxyvitamin D3 receptor in the male rat. Biochem Biophys Res Commun 181:611-616. Sarkinen BE (2011) Vitamin D deficiency & cardiovascular disease. Nurse Pract 36:46-53. Schmidt DR, Holmstrom SR, Fon Tacer K, Bookout AL, Kliewer SA and Mangelsdorf DJ (2010) Regulation of bile acid synthesis by fat-soluble vitamins A and D. J Biol Chem 285:14486-14494. Schmiedlin-Ren P, Thummel KE, Fisher JM, Paine MF, Lown KS and Watkins PB (1997) Expression of enzymatically active CYP3A4 by Caco-2 cells grown on extracellular matrix-coated permeable supports in the presence of 1α,25-dihydroxyvitamin D3. Mol Pharmacol 51:741-754. Schmiedlin-Ren P, Thummel KE, Fisher JM, Paine MF and Watkins PB (2001) Induction of CYP3A4 by 1 α,25-dihydroxyvitamin D3 is human cell line-specific and is unlikely to involve pregnane X receptor. Drug Metab Dispos 29:1446-1453.

200

Schoene B, Fleischmann RA, Remmer H and von Oldershausen HF (1972) Determination of drug metabolizing enzymes in needle biopsies of human liver. Eur J Clin Pharmacol 4:65-73. Schoonjans K and Auwerx J (2002) A sharper image of SHP. Nat Med 8:789-791. Schwartz JB (2009) Effects of vitamin D supplementation in atorvastatin-treated patients: a new drug interaction with an unexpected consequence. Clin Pharmacol Ther 85:198-203. Seol W, Choi HS and Moore DD (1996) An orphan nuclear hormone receptor that lacks a DNA binding domain and heterodimerizes with other receptors. Science 272:1336- 1339. Sertznig P, Dunlop T, Seifert M, Tilgen W and Reichrath J (2009) Cross-talk between vitamin D receptor (VDR)- and peroxisome proliferator-activated receptor (PPAR)- signaling in melanoma cells. Anticancer Res 29:3647-3658. Seshadri KG, Tamilselvan B and Rajendran A (2011) Role of Vitamin D in Diabetes. J Endocrinol Metab 1:47-56. Shneider BL (2001) Intestinal bile acid transport: biology, physiology, and pathophysiology. J Pediatr Gastroenterol Nutr 32:407-417. Silver J, Moallem E, Kilav R, Epstein E, Sela A and Naveh-Many T (1996) New insights into the regulation of parathyroid hormone synthesis and secretion in chronic renal failure. Nephrol Dial Transplant 11 Suppl 3:2-5. Sinal CJ, Tohkin M, Miyata M, Ward JM, Lambert G and Gonzalez FJ (2000) Targeted disruption of the nuclear receptor FXR/BAR impairs bile acid and lipid homeostasis. Cell 102:731-744. Sjoden G, Smith C, Lindgren U and DeLuca HF (1985) 1-Hydroxyvitamin D2 is less toxic than 1-hydroxyvitamin D3 in the rat. Proc Soc Exp Biol Med 178:432-436. Song KH, Li T, Owsley E, Strom S and Chiang JY (2009) Bile acids activate fibroblast growth factor 19 signaling in human hepatocytes to inhibit cholesterol 7α- hydroxylase gene expression. Hepatology 49:297-305. Staudinger JL, Goodwin B, Jones SA, Hawkins-Brown D, MacKenzie KI, LaTour A, Liu Y, Klaassen CD, Brown KK, Reinhard J, Willson TM, Koller BH and Kliewer SA (2001) The nuclear receptor PXR is a lithocholic acid sensor that protects against liver toxicity. Proc Natl Acad Sci U S A 98:3369-3374. Stravitz RT, Hylemon PB, Heuman DM, Hagey LR, Schteingart CD, Ton-Nu HT, Hofmann AF and Vlahcevic ZR (1993) Transcriptional regulation of cholesterol 7-hydroxylase mRNA by conjugated bile acids in primary cultures of rat hepatocytes. J Biol Chem 268:13987-13993. Szakacs G, Varadi A, Ozvegy-Laczka C and Sarkadi B (2008) The role of ABC transporters in drug absorption, distribution, metabolism, excretion and toxicity (ADME-Tox). Drug Discov Today 13:379-393. Tachibana S, Yoshinari K, Chikada T, Toriyabe T, Nagata K and Yamazoe Y (2009) Involvement of Vitamin D receptor in the intestinal induction of human ABCB1. Drug Metab Dispos 37:1604-1610. Taketani Y, Segawa H, Chikamori M, Morita K, Tanaka K, Kido S, Yamamoto H, Iemori Y, Tatsumi S, Tsugawa N, Okano T, Kobayashi T, Miyamoto K and Takeda E (1998) Regulation of type II renal Na+-dependent inorganic phosphate transporters

201

by 1,25-dihydroxyvitamin D3. Identification of a vitamin D-responsive element in the human NAPi-3 gene. J Biol Chem 273:14575-14581. Tamashevskii AV, Kozlova NM, Goncharova NV, Zubritskaia GP and Slobozhanina EI (2011) Effect of cholesterol on the functional activity of proteins responsible for the resistance of human lymphocytes to xenobiotics. Biofizika 56:455-464. Tanaka Y, Frank H and DeLuca HF (1973) Biological activity of 1,25-dihydroxyvitamin D3 in the rat. Endocrinology 92:417-422. Thompson PD, Jurutka PW, Whitfield GK, Myskowski SM, Eichhorst KR, Dominguez CE, Haussler CA and Haussler MR (2002) Liganded VDR induces CYP3A4 in small intestinal and colon cancer cells via DR3 and ER6 vitamin D responsive elements. Biochem Biophys Res Commun 299:730-738. Thummel KE, Brimer C, Yasuda K, Thottassery J, Senn T, Lin Y, Ishizuka H, Kharasch E, Schuetz J and Schuetz E (2001) Transcriptional control of intestinal cytochrome P- 4503A by 1α,25-dihydroxy vitamin D3. Mol Pharmacol 60:1399-1406. Tirona RG (2011) Molecular mechanisms of drug transporter regulation. Handb Exp Pharmacol:373-402. Tirona RG and Kim RB (2005) Nuclear receptors and drug disposition gene regulation. J Pharm Sci 94:1169-1186. Tirona RG, Lee W, Leake BF, Lan LB, Cline CB, Lamba V, Parviz F, Duncan SA, Inoue Y, Gonzalez FJ, Schuetz EG and Kim RB (2003) The orphan nuclear receptor HNF4alpha determines PXR- and CAR-mediated xenobiotic induction of CYP3A4. Nat Med 9:220-224. Tishkoff DX, Nibbelink KA, Holmberg KH, Dandu L and Simpson RU (2008) Functional vitamin D receptor (VDR) in the t-tubules of cardiac myocytes: VDR knockout cardiomyocyte contractility. Endocrinology 149:558-564. Trauner M and Boyer JL (2003) Bile salt transporters: molecular characterization, function, and regulation. Physiol Rev 83:633-671. Valdivielso JM, Cannata-Andia J, Coll B and Fernandez E (2009) A new role for vitamin D receptor activation in chronic kidney disease. Am J Physiol Renal Physiol 297:F1502-1509. van Aubel RA, Smeets PH, Peters JG, Bindels RJ and Russel FG (2002) The MRP4/ABCC4 gene encodes a novel apical organic anion transporter in human kidney proximal tubules: putative efflux pump for urinary cAMP and cGMP. J Am Soc Nephrol 13:595-603. Vanhooke JL, Benning MM, Bauer CB, Pike JW and DeLuca HF (2004) Molecular structure of the rat vitamin D receptor ligand binding domain complexed with 2- carbon-substituted vitamin D3 hormone analogues and a LXXLL-containing coactivator peptide. Biochemistry 43:4101-4110. Venkatakrishnan K, Von Moltke LL and Greenblatt DJ (2001) Human drug metabolism and the cytochromes P450: application and relevance of in vitro models. J Clin Pharmacol 41:1149-1179. Vieth R, Kooh SW, Balfe JW, Rawlins M and Tinmouth WW (1990) Tracer kinetics and actions of oral and intraperitoneal 1,25-dihydroxyvitamin D3 administration in rats. Kidney Int 38:857-861. Wagner M, Zollner G and Trauner M (2010) Nuclear receptor regulation of the adaptive response of bile acid transporters in cholestasis. Semin Liver Dis 30:160-177. 202

Wang JH, Keisala T, Solakivi T, Minasyan A, Kalueff AV and Tuohimaa P (2009) Serum cholesterol and expression of ApoAI, LXR and SREBP2 in vitamin D receptor knock-out mice. J Steroid Biochem Mol Biol 113:222-226. Wang K, Chen S, Xie W and Wan YJ (2008a) Retinoids induce cytochrome P450 3A4 through RXR/VDR-mediated pathway. Biochem Pharmacol 75:2204-2213. Wang L, Lee YK, Bundman D, Han Y, Thevananther S, Kim CS, Chua SS, Wei P, Heyman RA, Karin M and Moore DD (2002) Redundant pathways for negative feedback regulation of bile acid production. Dev Cell 2:721-731. Wang TJ, Pencina MJ, Booth SL, Jacques PF, Ingelsson E, Lanier K, Benjamin EJ, D'Agostino RB, Wolf M and Vasan RS (2008b) Vitamin D deficiency and risk of cardiovascular disease. Circulation 117:503-511. Wang XX, Jiang T, Shen Y, Santamaria H, Solis N, Arbeeny C and Levi M (2011a) Vitamin D receptor agonist doxercalciferol modulates dietary fat-induced renal disease and renal lipid metabolism. Am J Physiol Renal Physiol 300:F801-810. Wang Z, Lin YS, Zheng XE, Senn T, Hashizume T, Scian M, Dickmann LJ, Nelson SD, Baillie TA, Hebert MF, Blough D, Davis CL and Thummel KE (2011b) An inducible cytochrome P450 3A4-dependent vitamin D catabolic pathway. Mol Pharmacol. Epub [doi:10.1124/mol.111.076356] Wu FS, Zheng SS, Wu LJ, Teng LS, Ma ZM, Zhao WH and Wu W (2007) Calcitriol inhibits the growth of MHCC97 heptocellular cell lines by downmodulating c-met and ERK expressions. Liver Int 27:700-707. Wu J, Garami M, Cheng T and Gardner DG (1996) 1,25(OH)2 vitamin D3, and retinoic acid antagonize endothelin-stimulated hypertrophy of neonatal rat cardiac myocytes. J Clin Invest 97:1577-1588. Xiang W, Kong J, Chen S, Cao LP, Qiao G, Zheng W, Liu W, Li X, Gardner DG and Li YC (2005) Cardiac hypertrophy in vitamin D receptor knockout mice: role of the systemic and cardiac renin-angiotensin systems. Am J Physiol Endocrinol Metab 288:E125-132. Xie W, Radominska-Pandya A, Shi Y, Simon CM, Nelson MC, Ong ES, Waxman DJ and Evans RM (2001) An essential role for nuclear receptors SXR/PXR in detoxification of cholestatic bile acids. Proc Natl Acad Sci U S A 98:3375-3380. Xu Y, Iwanaga K, Zhou C, Cheesman MJ, Farin F and Thummel KE (2006) Selective induction of intestinal CYP3A23 by 1α,25-dihydroxyvitamin D3 in rats. Biochem Pharmacol 72:385-392. Yagci A, Werner A, Murer H and Biber J (1992) Effect of rabbit duodenal mRNA on phosphate transport in Xenopus laevis oocytes: dependence on 1,25-dihydroxyvitamin-D3. Pflugers Arch 422:211-216. Yamagishi K, Yamamoto K, Yamada S and Tokiwa H (2006) Functions of key residues in the ligand-binding pocket of vitamin D receptor: Fragment molecular orbital- interfragment interaction energy analysis. Chemical Physics Letters 420:465-468. Yamane T, Takeuchi K, Yamamoto Y, Li YH, Fujiwara M, Nishi K, Takahashi S and Ohkubo I (2002) Legumain from bovine kidney: its purification, molecular cloning, immunohistochemical localization and degradation of annexin II and vitamin D- binding protein. Biochim Biophys Acta 1596:108-120.

203

You G and Morris ME (2007) Drug Transporters : Molecular Characterization And Role In Drug Disposition. Wiley-Interscience, Hoboken, N.J. Yu DK (1999) The contribution of P-glycoprotein to pharmacokinetic drug-drug interactions. J Clin Pharmacol 39:1203-1211. Yuan W, Pan W, Kong J, Zheng W, Szeto FL, Wong KE, Cohen R, Klopot A, Zhang Z and Li YC (2007) 1,25-dihydroxyvitamin D3 suppresses renin gene transcription by blocking the activity of the cyclic AMP response element in the renin gene promoter. J Biol Chem 282:29821-29830. Zastre JA, Chan GN, Ronaldson PT, Ramaswamy M, Couraud PO, Romero IA, Weksler B, Bendayan M and Bendayan R (2009) Upregulation of P-glycoprotein by HIV protease inhibitors in a human brain microvessel endothelial cell line. J Neurosci Res 87:1023-1036. Zhang HL and Wu J (2011) Role of vitamin D in immune responses and autoimmune diseases, with emphasis on its role in multiple sclerosis. Neurosci Bull 26:445-454. Zhang X, Jiang F, Li P, Li C, Ma Q, Nicosia SV and Bai W (2005) Growth suppression of ovarian cancer xenografts in nude mice by vitamin D analogue EB1089. Clin Cancer Res 11:323-328. Zhang Y, Hagedorn CH and Wang L (2011) Role of nuclear receptor SHP in metabolism and cancer. Biochim Biophys Acta 1812:893-908. Zhu Y, Li F and Guo GL (2011) Tissue-specific function of farnesoid X receptor in liver and intestine. Pharmacol Res 63:259-265. Zierold C, Mings JA and Deluca HF (2006) 19nor-1,25-dihydroxyvitamin D2 specifically induces CYP3A9 in rat intestine more strongly than 1,25-dihydroxyvitamin D3 in vivo and in vitro. Mol Pharmacol 69:1740-1747. Zollner G, Fickert P, Fuchsbichler A, Silbert D, Wagner M, Arbeiter S, Gonzalez FJ, Marschall HU, Zatloukal K, Denk H and Trauner M (2003) Role of nuclear bile acid receptor, FXR, in adaptive ABC transporter regulation by cholic and ursodeoxycholic acid in mouse liver, kidney and intestine. J Hepatol 39:480-488. Zöllner G, Marschall HU, Wagner M and Trauner M (2006) Role of nuclear receptors in the adaptive response to bile acids and cholestasis: pathogenetic and therapeutic considerations. Mol Pharm 3:231-251.

204

APPENDIX A1

Khan AA, Chow EC, van Loenen-Weemaes AM, Porte RJ, Pang KS and Groothuis GM (2009) Comparison of effects of VDR versus PXR, FXR and GR ligands on the regulation of CYP3A isozymes in rat and human intestine and liver. Eur J Pharm Sci 37:115-125

205

European Journal of Pharmaceutical Sciences 37 (2009) 115–125

Contents lists available at ScienceDirect

European Journal of Pharmaceutical Sciences

journal homepage: www.elsevier.com/locate/ejps

Comparison of effects of VDR versus PXR, FXR and GR ligands on the regulation of CYP3A isozymes in rat and human intestine and liver

Ansar A. Khan a,∗, Edwin C.Y. Chow b, Anne-miek M.A. van Loenen-Weemaes a,RobertJ.Portec, K. Sandy Pang b, Geny M.M. Groothuis a a Pharmacokinetics, Toxicology and Targeting, Department of Pharmacy, University of Groningen, Ant. Deusinglaan 1, 9713 AV, Groningen, The Netherlands b Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College Street, Toronto, ON M5S 3M2, Canada c Department of Hepatobiliary Surgery and Liver Transplantation, University Medical Center Groningen (UMCG), University of Groningen, Hanzeplein 1, 9700 RB Groningen, The Netherlands article info abstract

Article history: In this study, we compared the regulation of CYP3A isozymes by the vitamin D receptor (VDR) ligand 1␣,25- Received 2 December 2008 dihydroxyvitamin D3 (1,25(OH)2D3) against ligands of the pregnane X receptor (PXR), the glucocorticoid Received in revised form 19 January 2009 receptor (GR) and the farnesoid X receptor (FXR) in precision-cut tissue slices of the rat jejunum, ileum, Accepted 20 January 2009 colon and liver, and human ileum and liver. In the rat, 1,25(OH) D strongly induced CYP3A1 mRNA, Available online 30 January 2009 2 3 quantiﬁed by qRT-PCR, along the entire length of the intestine, induced CYP3A2 only in ileum but had no effect on CYP3A9. In contrast, the PXR/GR ligand, dexamethasone (DEX), the PXR ligand, pregnenolone-16␣ Keywords: carbonitrile (PCN), and the FXR ligand, chenodeoxycholic acid (CDCA), but not the GR ligand, budesonide Cytochrome P450 Induction (BUD), induced CYP3A1 only in the ileum, none of them inﬂuenced CYP3A2 expression, and PCN, DEX Intestinal slices and BUD but not CDCA induced CYP3A9 in jejunum, ileum and colon. In rat liver, CYP3A1, CYP3A2 and Liver slices CYP3A9 mRNA expression was unaffected by 1,25(OH)2D3, whereas CDCA decreased the mRNA of all

1␣,25-Dihydroxyvitamin D3 CYP3A isozymes; PCN induced CYP3A1 and CYP3A9, BUD induced CYP3A9, and DEX induced all three CYP3A isozymes. In human ileum and liver, 1,25(OH)2D3 and DEX induced CYP3A4 expression, whereas CDCA induced CYP3A4 expression in liver only. In conclusion, the regulation of rat CYP3A isozymes by VDR, PXR, FXR and GR ligands differed for different segments of the rat and human intestine and liver, and the changes did not parallel expression levels of the nuclear receptors. © 2009 Elsevier B.V. All rights reserved.

1. Introduction et al., 1995), while CYP3A9 and CYP3A62 expression is higher in female rats. CYP3A1 and CYP3A2 are predominantly expressed in The cytochrome P450 enzymes constitute a family of heme the rat liver, and CYP3A62, in female livers (Matsubara et al., 2004), protein oxygenases that display considerable similarities in their whereas CYP3A9 is highly expressed in the intestine relative to the molecular weights, immunohistochemical properties, and sub- liver (Mahnke et al., 1997; Wang and Strobel, 1997). The human strate specificities (Gonzalez, 1988). The CYP3A isoforms play CYP3A family which is expressed in the liver is composed of at least an important role in oxidation of endogenous steroids and toxic four isozymes: CYP3A4, CYP3A5, CYP3A7 and CYP3A43 of which hydrophobic bile acids. In the rat, the CYP3A family consists of CYP3A4 is the predominant isozyme expressed in adult human liver five isoforms: CYP3A1/CYP3A23 (Gonzalez et al., 1985), CYP3A2 (Guengerich et al., 1986). CYP3A4 and CYP3A5 isozymes are present (Gonzalez et al., 1986), CYP3A9 (Wang et al., 1996), CYP3A18 along the human digestive tract, with CYP3A5 mainly present in (Strotkamp et al., 1995) and CYP3A62 (Matsubara et al., 2004). the stomach and CYP3A4 along the intestine segments (Kolars et These enzymes are expressed predominantly in the liver and in the al., 1994). enterocytes of the intestine (Kolars et al., 1994). The distribution The expression of CYP3A isoforms in rats and humans was of CYP3A isozymes in the rat appears to be sex-, tissue- and age- reported to be modulated by exogenous and endogenous ligands dependent. CYP3A2 and CYP3A18 are predominantly expressed in through the pregnane X receptor (PXR) (Lu et al., 1972), the glu- male rats (Gonzalez et al., 1986; Nagata et al., 1996; Strotkamp cocorticoid receptor (GR) (Huss et al., 1999), and the vitamin D receptor (VDR) (Makishima et al., 2002; Thummel et al., 2001; Xu et al., 2006). Recently, a FXR response element (FXRE) was found in ∗ Corresponding author. Tel.: +31 50 363 7565; fax: +31 50 363 3247. the human CYP3A4 promoter, and induction by CDCA, a FXR ligand, E-mail address: [email protected] (A.A. Khan). was noted (Gnerre et al., 2004). The 5 flanking promoter regions of

206 116 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125 the rat and human CYP3A are characterized by direct repeats spaced Biomedicals, Inc. (Eschwege, Germany). Low gelling temperature by three base pairs (DR3) and everted repeats spaced by six base agarose, pregnenolone-16␣ carbonitrile and budesonide were pur- pairs (ER6) (Gnerre et al., 2004; Hashimoto et al., 1993; Thummel chased from Sigma–Aldrich (St. Louis, MO). RNAeasy mini columns et al., 2001). PXR, FXR and VDR directly bind to the respective were obtained from Qiagen, Hilden, Germany. Random primers response elements pursuant to the ligand binding and heterodimer- (500 ␮g/ml), MgCl2 (25 mM), RT buffer (10×), PCR nucleotide mix ization with retinoic acid X receptor ␣ (RXR␣)(Gnerre et al., 2004; (10 mM), AMV RT (22 U/␮l) and RNasin (40 U/␮l) were procured Lehmann et al., 1998; Thompson et al., 1999). In contrast, the GR from Promega Corporation, Madison WI, USA. SYBR green and Taq effects on CYP3A isozymes in rat and humans have been attributed Master Mixes were purchased from Applied Biosystems, Warring- indirectly to the induction of HNF4␣ and PXR (Huss and Kasper, ton, UK and Eurogentech, respectively. ATP Bioluminescence Assay 1998). kit CLS II is procured from Roche, Mannheim, Germany. All primers The effects of various ligands on rat and human CYP3A enzymes were purchased from Sigma Genosys. All reagents and materials in the intestine and liver have been studied in vitro in both pri- used were of the highest purity that was commercially available. mary cultured hepatocytes and enterocytes, and immortalized human cell lines such as HepG2 and Caco-2 cells. Immortalized 2.2. Animals intestinal cell lines derived from the different regions of the rat intestine were utilized to study the regulation of drug metaboliz- Male Wistar (HsdCpb:WU) rats weighing about 230–250 g were ing enzymes (Zhang et al., 1997). However, these cell lines lack the purchased from Harlan (Horst, The Netherlands). Rats were housed normal expression of nuclear receptors (NRs), metabolic enzymes in a temperature and humidity controlled room on a 12-h light/dark and transporters. For example, Caco-2 cells are PXR-deficient and cycle with food (Harlan chow no 2018, Horst, The Netherlands) and exhibit reduced levels of drug metabolizing enzymes (Li et al., tap water ad libitum. The animals were allowed to acclimatize for 2003). Furthermore, cell lines are unable to reflect the segmental 7 days before experimentation. The experimental protocols were expression of CYP3A isozymes and the gradients of activities along approved by the Animal Ethical Committee of the University of the length of the rat intestine (Liu et al., 2006; van de Kerkhof et al., Groningen. 2005). The induction/repression of CYP3A isoforms in the intact liver and intestinal tissue in response to ligands of the NRs has not been extensively investigated. Such a response is dependent 2.3. Rat liver and intestine not only on the presence of NR response elements, but also on the expression levels of the NRs and exposure of the particular cell to Under isoflurane/O2/N2O anaesthesia, the small intestine, colon the ligand. This exposure is the result of uptake, metabolism and and liver were excised from the rat. Small intestine and colon were excretion of the ligand and its metabolites and may differ between immediately placed into ice-cold carbogenated Krebs–Henseleit the various regions of the intestine and the liver as a result of dif- buffer, supplemented with 10 mM HEPES, 25 mM sodium bicarbon- ferences in expression of uptake and excretion transporters and ate and 25 mM D-Glucose, pH 7.4 (KHB) and stored on ice until the metabolizing enzymes. Different regions of the intestine and liver preparation of slices. Livers were stored in ice-cold University of are exposed to different concentrations of the ligands in vivo. For an Wisconsin solution (UW) until slicing. appreciation of the potential variation between the different organs and their sensitivity towards the NR ligands, these organs or tissues 2.4. Human liver and ileum tissue should be studied under identical conditions. Therefore, in this study, we compared the effects of various NR Pieces of human liver tissue were obtained from patients under- ligands on the intestine and liver of the rat and human in precision- going partial hepactectomy for the removal of carcinoma or from cut tissue slices. This model has been previously validated as a redundant parts of donor livers remaining after split liver trans- useful ex vivo model for induction studies (Olinga et al., 2008; van plantation as described previously by Olinga et al. (2008). Donor de Kerkhof et al., 2007b, 2008) that enables us to investigate the characteristics are given in Table 1. Human ileum was obtained as effects of inducing ligands under identical incubation conditions part of the surgical waste after resection of the ileo-colonic part for the liver and intestine. We tested the hypothesis that the reg- of the intestine in colon carcinoma patients, donor characteris- ulation of rat and human CYP3A isozymes by VDR ligands differed tics are given in Table 2. After surgical resection, the ileum tissue from those by PXR, GR and FXR ligands. We compared the induction was immediately placed in ice-cold KHB. The research protocols potential of PXR, FXR and GR ligands to that of VDR ligand, 1␣,25- were approved by the Medical Ethical Committee of the University dihydroxyvitamin D (1,25(OH) D ) on changes in mRNAs of the 3 2 3 Medical Center, Groningen with informed consent of the patients. various CYP3A isoforms in the small intestine (jejunum and ileum), colon and liver of the rat and the CYP3A4 in human ileum and liver slices, and investigated whether these responses correlated to the Table 1 expression levels of the NRs. Characteristics of human liver donor used: ATP contents after 3 and 24 h of incubation (each value is a mean ± S.D. of three slices per time point)a. 2. Materials and methods Human liver (HL) Gender Age ATP-content (pmol/␮g of protein)

2.1. Materials 3h 24h HL1b,c Female 54 9.2 ± 0.5 10.4 ± 1.5 HL2 Not available 4.2 ± 0.9 5.7 ± 1.9 1,25(OH)2D3 in ethanol was purchased from BIOMOL Research HL3c Female 72 3.4 ± 0.8 3.3 ± 1.2 Laboratories, Inc., Plymouth Meeting, PA. Chenodeoxycholic acid HL4 Female 64 7.2 ± 1.2 9.7 ± 1.8 was purchased from Calbiochem, San Diego, CA, dexamethasone HL5c Male 65 12.1 ± 1.2 12.1 ± 1.0 was from Genfarma bv, Maarssen. The solvents: ethanol, methanol Mean ± S.E.M. 7.2 ± 1.6 8.2 ± 1.6 and DMSO were purchased from Sigma–Aldrich Chemical Co. (St. P value 0.66 Louis, MO); Gentamicin and Williams medium E with glutamax- a Data were expressed as mean ± S.D. I and amphotericin B (Fungizone)-solution were obtained from b Human liver tissue for immunohistochemistry of VDR. c Gibco (Paisley, UK). D-Glucose and HEPES were procured from ICN Human livers responsive to CYP3A4 induction by 1,25(OH)2D3.

207 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125 117

Table 2 Rat intestinal slices were incubated for 12 h because the expres- Characteristics of human ileum donor used. The terminal ileum was obtained from sion of villin and GAPDH remained unchanged up to 12 h, whereas colon carcinoma patients as part of tumor resection. in pilot experiments, the expression of villin was significantly Human ileum (HIL) Gender Age Medical history decreased after 24 h of incubation, indicating loss of epithelial cells. HIL 1 F 85 Colon carcinoma; coronary disease Human ileum slices were incubated for 8 and 24 h, and showed HIL 2 M 60 Colon carcinoma that villin expression remained unchanged up to 24 h. Rat and HIL 3 F 61 Colon carcinoma human intestinal slices were incubated with 1,25(OH)2D3 (final HIL 4 Not available concentrations 5–100 nM), CDCA (final concentration 50 ␮M), DEX HIL 5 F 69 Colon carcinoma (final concentrations 1 and 50 ␮M) and BUD (final concentration 10 nM). Furthermore, rat intestinal slices were also incubated with 2.5. Preparation of slices PCN (final concentration 10 ␮M). All ligands were added as a 100- 2.5.1. Rat and human intestinal slices times concentrated, stock solution in ethanol (for 1,25(OH)2D3), Rat intestinal slices were prepared as published before (van de methanol (for CDCA) and DMSO (for DEX/BUD/PCN) and had no Kerkhof et al., 2005). In brief, the rat jejunum (at 25–40 cm from or only minor effects on villin expression. Higher concentrations ␮ the stomach), ileum (5 cm proximal to the ileocecal valve) and of CDCA (100 M) significantly reduced villin expression and con- colon (large intestine, distal to the ileocecal valve) tissues were sidered toxic. Control slices were incubated in William’s medium E separated. The jejunum, ileum and colon were divided into approx- with 1% ethanol, methanol and DMSO without inducers. From a sin- imately 3-cm segments. The lumen of the segments was flushed gle rat or human tissue sample, six (rat intestine) or three (human with ice-cold KHB that was aerated with carbogen. Thereafter, seg- intestine) replicate slices were subjected to each experimental con- ments were tied at one end and filled with 3% low gelling agarose dition. After the incubation these replicate slices were harvested, solution in saline that was kept at 37 ◦C, then cooled immedi- pooled and snap-frozen in liquid nitrogen to obtain sufficient total − ◦ ately in KHB allowing the agarose to solidify. Subsequently, the RNA for qRT-PCR analysis. Samples were stored in 80 C freezer agarose filled segments were embedded in agarose solution filled until RNA isolation. These experiments were replicated in 3–5 rats pre-cooled embedding unit (Alabama R&D, Munford, AL, USA). The and 3–5 human ileum donors. agarose filled solid embedded intestinal segments were then placed in the pre-cooled Krumdieck tissue slicer (Alabama R&D, Munford, 2.7. Incubation of rat and human liver slices AL, USA) containing carbogenated ice-cold KHB, and precision-cut slices were prepared with a thickness of approximately 200 ␮m Slices were incubated individually in 6-well, sterile tissue and wet weight of 2–3 mg (without agarose) (cycle speed 40: inter- culture plates (Grenier bio-one GmbH, Frickenhausen, Austria) con- rupted mode). Slices were stored in carbogenated ice-cold KHB on taining 3.2 ml William’s medium E supplemented with D-glucose ␮ ice until the start of the experiment which usually varies between to a final concentration of 25 mM, gentamicin sulfate (50 g/ml) 2 to 3 h after sacrificing the rat. and saturated with carbogen. The plates were placed in humidi- ◦ Human ileum slices were prepared according to the method fied plastic container kept at 37 C and continuously gassed with described for the jejunum (van de Kerkhof et al., 2006). In brief, carbogen and shaken at 80 rpm. Rat slices were incubated with ileum tissue was stripped of the muscular layer and the mucosal 1,25(OH)2D3 (final concentrations 10–200 nM), CDCA (final con- ␮ ␮ tissue was transferred to carbogenated ice-cold KHB. Mucosal tis- centrations 10–100 M) and DEX (final concentrations 1–50 M). sue was cut into rectangular pieces of ∼6–8 mm wide and these Apart from the above inducers, rat liver slices were also incubated ␮ were subsequently embedded in low gelling 3% agarose in saline with PCN (final concentration 10 M) and BUD (final concen- using pre-cooled tissue embedding unit (Alabama R&D, Munford, trations 10–100 nM). All inducers were added as a 100-times AL, USA) allowing the agarose solution to solidify. Precision-cut concentrated stock solution in ethanol (for 1,25(OH)2D3), methanol slices of approximately 200-␮m thick were prepared as described (for CDCA) and DMSO (for DEX/PCN/BUD). Control rat and human above for rat intestine. liver slices were incubated in William’s medium E with 1% ethanol, methanol, and DMSO, the vehicles. Rat and human liver slices were 2.5.2. Rat and human liver slices incubated for 8 and 24 h, respectively. From a single rat/single Cylindrical cores of 8 mm were prepared from rat livers and human liver donor three replicate slices were subjected to identical human liver tissue by advancing a sharp rotating metal tube in incubation conditions. At the end of the incubation these replicate the liver tissue and were subsequently placed in the pre-cooled slices were harvested, pooled and snap-frozen in liquid nitrogen Krumdieck tissue slicer. The slicing was performed in carbo- to obtain sufficient total RNA for qRT-PCR analysis. Samples were genated ice-cold KHB. The thickness of the liver slice was kept at stored in −80 ◦C freezer until RNA isolation. These experiments ∼200–300 ␮m and a wet weight of 10–12 mg were prepared with were replicated in 3–5 rats and 4–5 human liver donors. the standard settings (cycle speed 40: interrupted mode) of the Krumdieck tissue slicer. Subsequently, slices were stored in ice-cold 2.8. RNA isolation and quantitative real-time PCR (qRT-PCR) UW solution on ice prior to the start of the experiment, which usually varies from 1 to 3 h from sacrificing the rat and for human livers Total RNA from rat and human intestine and liver samples were 2 to 3 h post-surgery. isolated by using RNAeasy mini columns from Qiagen according to the manufacturer’s instruction. RNA quality and concentrations 2.6. Incubation of rat and human intestinal slices were determined by measuring the absorbance at 260, 230 and 280 nm using a Nanodrop ND100 spectrophotometer (Wilming- Slices were incubated individually in the 12-well, sterile tissue ton, DE, USA). The ratio of absorbance measured at 260 over 280 culture plates (Grenier bio-one GmbH, Frickenhausen, Austria) con- and 230 over 260 was always above 1.8. About 2 ␮g of total RNA taining 1.3 ml William’s medium E supplemented with D-glucose in 50 ␮l was reverse transcribed into template cDNA using ran- to a final concentration of 25 mM, gentamicin sulfate (50 ␮g/ml), dom primers (0.5 ␮g/ml), PCR nucleotide mix (10 mM), AMV RT ␮ amphotericin/fungizone (250 g/ml), and saturated with carbo- (22 U/␮l), RT buffer (10×), MgCl2 (25 mM) and RNAasin (40 U/␮l). gen. The plates were placed in humidified plastic container kept at Real time quantitative PCR (qRT-PCR) was performed for 37 ◦C and continuously gassed with carbogen and shaken at 80 rpm. genes of interest using primer sequences given in Table 3 by

208 118 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125

Table 3 Oligonucleotides for quantitative real-time PCR, rat and human genes (SYBR Green and Taqman® analysis).

Gene Forward primer(5–3) Reverse primer(5–3) Gene bank number r Villin GCTCTTTGAGTGCTCCAACC GGGGTGGGTCTTGAGGTATT XM 001057825 r GAPDH CTGTGGTCATGAGCCCCTCC CGCTGGTGCTGAGTATGTCG XR 008524 r CYP3A1 GGAAATTCGATGTGGAGTGC AGGTTTGCCTTTCTCTTGCC L24207 r CYP3A2 AGTAGTGACGATTCCAACATAT TCAGAGGTATCTGTGTTTCCT XM 573414 r CYP3A9 GGACGATTCTTGCTTACAGG ATGCTGGTGGGCTTGCCTTC U46118 r FXR CCAACCTGGGTTTCTACCC CACACAGCTCATCCCCTTT NM 021745 r PXR GATGATCATGTCTGATGCCGCTG GAGGTTGGTAGTTCCAGATGCTG NM 052980 h Villin CAGCTAGTGAACAAGCCTGTAGAGGAGC CCACAGAAGTTTGTGCTCATAGGC NM 007127 h CYP3A4 GCCTGGTGCTCCTCTATCTA GGCTGTTGACCATCATAAAAG DQ924960 h VDR GGAAGTGCAGAGGAAGCGGGAGATG AGAGCTGGGACAGCTCTAGGGTCAC NM 000376 r ␤ actina Assay-by-DesignTM FAM labelled, Part number 4331348 (Applied Biosystems) NM 031144 TGACCCCACCTACGCTGACT CCTTGGAGAATAGCTCCCTGTACT r VDRa 24873 Probe, 6FAM, ACTTCCGGCCTCCAGTTCGTATGGAC-TAMRA Assay-by-DesignTM ID, Hs99999905 m1 h GAPDHb NM 002046 Probe, 6FAM, GCGCCTGGTCACCAGGGCTGCTTTT, NFQ r, rat genes and h, human genes. a Primer sets for rat Taqman® Gene analysis. b Primer sets for human Taqman® Gene analysis. two detection systems based on the availability of primer sets; the method described earlier by de Kanter et al. (2002). In brief, villin and GAPDH were used as house-keeping genes for intesti- control human liver slices were incubated in 3.2 ml of William’s nal epithelial cells and liver cells, respectively, and CYP3A1, medium E, supplemented with D-glucose to a final concentration CYP3A2, CYP3A9, PXR and FXR were analyzed by the SYBR of 25 mM, gentamicin sulfate (50 ␮g/ml), and saturated with car- Green detection system. Primer sequences used for CYP3A1, bogen, as described in Section 2.7 for 3 and 24 h. At the end of CYP3A2 and CYP3A9 analysis were identical to those reported incubation time, three replicate slices were collected individually earlier by Mahnke et al. (1997). All primer sets were analyzed in 1 ml 70% ethanol (v/v) containing 2 mM EDTA (pH 10.9) and snap- using BLASTn to ensure primer specificity for the gene of inter- frozen in liquid nitrogen and stored at −80 ◦C freezer until analysis. est (http://www.ncbi.nlm.nih.gov/BLAST/). For qRT-PCR using the The samples were disrupted and homogenized by sonication, and SYBR Green detection system ∼50 ng of cDNA was used in a total ATP extracts were diluted 10 times with 0.1 M Tris–HCl contain- reaction mixture of 20 ␮l of the SYBR Green mixture (Applied ing 2 mM EDTA (pH 7.8) to reduce the ethanol concentration. The Biosystems, Warrington, UK). The PCR conditions are step 1: 95 ◦C ATP content was measured using the ATP Bioluminescence Assay for 10 min; and step 2: 40 cycles of 95 ◦C, 15 s; 56 ◦C, 60 s and 72 ◦C, kit CLS II from Roche (Mannheim, Germany) in a 96-well plate 40 s, followed by a dissociation stage (at 95 ◦C for 15 s, at 60 ◦Cfor Lucy1 luminometer (Anthos, Durham, NC, USA) using a standard 15 s and at 95 ◦C for 15 s) to determine the homogeneity of the ATP-calibration curve. PCR product. Further, the control consisting of water (with water Protein content of the slices was estimated in three identical, instead of total mRNA, which has been subjected to reverse tran- replicate slices which were not used for incubation. The slices scription protocol) and the mRNA control (isolated mRNA which were digested with 5 M NaOH and homogenized, and subsequently has not been subjected to reverse transcription protocol) were diluted with water to result in a concentration of 0.1 M NaOH. The used to determine primer dimer formation and contamination of protein content of the diluted homogenate was determined by the DNA in the isolated samples, respectively. Amplification plots and Bio-Rad protein assay dye reagent method (Bio-Rad, Munich, Ger- dissociation curves of the controls did not show any signal and many) using bovine serum albumin (BSA) for the calibration curve. dissociation product, suggesting the lack of primer dimer forma- The ATP content of the slice was expressed as pmol/␮g of protein. tion. In addition total RNA from the samples for the preparation of cDNA appeared to be free of DNA contamination. ␤-actin and VDR 2.10. Statistics genes were analyzed by Taqman® analysis using primer sequences All experiments were performed in 3–5 rats and in 4–5 human given in Table 3. For Taqman® analysis ∼250 ng of cDNA was used tissue samples. Values were expressed as mean ± S.E.M. All data in a total reaction mixture of 10 ␮l Taq Master Mix (2×). The qRT- ◦ were analyzed by the unpaired student’s t-test or Mann–Whitney PCR conditions for Taqman® analysis were: step 1, 95 C for 10 min; ◦ ◦ U-test to detect differences between the means of different treat- step 2, 40 cycles of 95 C for 15 s and 60 C for 60 s. All samples ments. The Student’s t-test was used to analyze the rat data where were analyzed in duplicates in 384 well plates using ABI7900HT the error distribution was found to be normal with equal vari- from Applied Biosystems. The comparative threshold cycle (CT) ance except for the CYP3A1 and CYP3A2 genes. Among experiments method was used for relative quantification since CT was inversely where non-equal error distribution and high variance (e.g. expres- related to the abundance of mRNA transcripts in the initial sam- sion of CYP3A1 and CYP3A2 genes in Wistar rats and CYP3A4 in ple. The mean CT of the duplicate measurements was used to human tissues due to age and habits) were observed, the non- calculate the difference in CT for gene of interest and the house parametric Mann–Whitney U-test was used. Statistical analysis was keeping gene, villin for intestine and GAPDH for liver (CT). This performed on fold induction as well as on CT with similar CT value of the treated sample was compared to the correspond- results. The P value <0.05 was considered as significant. ing CT of the solvent control (CT). Data are expressed as fold induction or repression of the gene of interest according to the − C 3. Results formula 2 ( T).

2.9. ATP and protein content of the human liver slices 3.1. Expression of nuclear receptors in rat intestine and liver

Viability of human liver slices during incubation was deter- VDR, PXR and FXR mRNA were detected in rat intestine as well mined by measuring the ATP contents of the slices according to as in liver. To analyze the expression of the NRs, enzymes and

209 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125 119

detectable (CT values ≥33 for CYP3A1) in all regions of the intestine. CYP3A2 was barely detectable in the ileum (≥35 for CYP3A2) but was undetectable in the jejunum and colon. Because CYP3A1 and CYP3A2 mRNA expression was decreased, whereas that of CYP3A9 expression was moderately elevated during incubation of the slices (data not shown), results on ligand-induced effects were expressed relative to “control” slices incubated with solvent for the same incubation period. Increasing concentrations of the VDR ligand, 1,25(OH)2D3 strongly induced CYP3A1 mRNA in all regions of the rat intestine (700-fold at 100 nM of 1,25(OH)2D3 in jejunum, 15,000-fold for the ileum, and 1000-fold for the colon; P < 0.05) (Fig. 2A), but the mRNA expression of CYP3A9 remained unchanged (Fig. 2C). In contrast, PCN, DEX and BUD strongly induced CYP3A9 mRNA in the jejunum and ileum, and to a much lesser extent, in the colon (Fig. 2C). PCN and DEX but not BUD induced CYP3A1 in the ileum (Fig. 2A), but had no effect on CYP3A1 in the colon. Although PCN, BUD and DEX induced CYP3A1 mRNA in the jejunum samples, the results failed to reach statistical signiﬁcance due to the high variation among the data (Fig. 2A). CDCA induced CYP3A1 mRNA only in the ileum and not in the jejunum and colon (Fig. 2A), and failed to affect the expression of CYP3A9 mRNA along the length of the intestine (Fig. 2C). CYP3A2 mRNA, though practi- cally undetectable after incubation with PCN, BUD, DEX and CDCA, was highly induced by 1,25(OH)2D3 in the ileum; however, CYP3A2 remained undetectable in the jejunum and colon for all situations (Fig. 2B).

3.2.2. Rat liver slices In the rat liver, the expression of CYP3A1 and CYP3A2 was very high compared to that in the intestine, and was detected at a CT value of 18–19, where as CYP3A9 was detected at a CT value Fig. 1. Expression of PXR, FXR and VDR mRNA in rat intestine was normalized to that of villin (A); the value of jejunum/villin was set to 1. The expression of PXR, of 22. The expression of CYP3A1, CYP3A2, and CYP3A9 mRNAs FXR and VDR mRNA in rat intestine and liver, after normalizing to GAPDH (B), was significantly decreased during incubation, but was not further with the liver value set to 1. Results were mean ± S.E.M. of three rats. “*” denotes affected by the presence of the solvent vehicle. Distinct from intesti- P < 0.05, compared to jejunum (A) or liver (B). “#” denotes P < 0.05, compared to ileum nal slices, incubation of rat liver slices with 1,25(OH)2D3 did not (A and B). change the expression of CYP3A1, CYP3A2 and CYP3A9 (Fig. 3A). DEX induced CYP3A1, CYP3A2 and CP3A9 mRNA expression in rat liver slices in a concentration-dependent manner (Fig. 3C). PCN transporters per enterocyte, their expression is expressed rela- induced CYP3A1 and CYP3A9 but not CYP3A2 mRNA expression tive to that of villin, which is present exclusively in the epithelial (Fig. 3C). However, BUD induced CYP3A9 expression without affect- cells of the intestine. In rat intestine, PXR, FXR and VDR expres- ing those of CYP3A1 and CYP3A2 (Fig. 3C). CDCA significantly sion varied along the length of the small intestine and colon. decreased the expression of CYP3A1, CYP3A2 and CYP3A9 with The expression of PXR and VDR mRNA relative to villin was 5- increasing concentration (Fig. 3B) to 0.7-fold, 0.5-fold and 0.7-fold, fold higher in the colon compared to the jejunum and the ileum respectively. (Fig. 1A). FXR expression relative to villin was 5-fold higher in the ileum compared to the jejunum and was similar to that in 3.3. Induction of PXR in rat intestine and liver slices the colon (Fig. 1A). For comparison of the expression in liver, GAPDH was used as a reference since villin is not expressed in The expression of PXR, a known GR-responsive gene, was hepatocytes. In the rat liver, the expression of FXR and PXR, rel- studied in the rat intestinal and liver samples treated with GR ative to GAPDH, was significantly higher (2–10-fold) compared (DEX/BUD) and PXR (PCN) ligands. DEX and BUD but not PCN to those in the small intestine and colon (Fig. 1B). However, the induced PXR expression in all the three regions of the intestine mRNA expression of VDR relative to GAPDH in the rat liver was and in the liver (Fig. 4A and B). Furthermore, PXR induction by very low, about 0.1% compared to those in the small intestine and DEX (1 ␮M) and BUD (10 nM) in the rat colon was lower com- colon, and was detected at approximate CT values of 32–34 cycles pared to that in the jejunum and ileum, but the fold-induction (Fig. 1B). at 50 ␮M DEX, in the jejunum, ileum and colon was comparable (Fig. 4A). 3.2. Expression and regulation of CYP3A isozymes in rat intestine and liver slices 3.4. Expression and regulation of CYP3A4 in human ileum and liver slices 3.2.1. Rat intestine slices Among the CYP3A isozymes in the rat intestine, CYP3A9 was CYP3A4 mRNA expression was constant up to 8 h of incubation clearly expressed (CT value ∼19–21) in all segments: the expres- in ileum slices, but decreased to 30–50% by 24 h, with only minor sion of CYP3A9 in rat intestine per enterocyte was in the rank differences between the control and the solvent-treated slices order of colon > jejunum ≥ ileum. CYP3A1 expression was low but (Fig. 5A). The FXR and PXR expression in human ileum and liver,

210 120 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125

Fig. 2. Slices from rat jejunum, ileum and colon were exposed to 1,25(OH)2D3 (5, 10 and 100 nM), CDCA (50 ␮M), DEX (1 and 50 ␮M), BUD (10 nM) and PCN (10 ␮M) for 12 h, after which total RNA was isolated and mRNA expression of CYP3A1 (A), CYP3A2 (B) and CYP3A9 (C) were evaluated by qRT-PCR. Results were expressed as fold induction after normalizing with villin expression and compared to the control slices of the same segment that was also incubated for 12 h; the control value was set as 1. Results were mean ± S.E.M. of 3–5 rats; in each experiment, 6 slices were incubated per condition. Signiﬁcant differences towards the control incubations are indicated with *P < 0.05 and **P = < 0.01. “†” denotes induction of CYP3A1 and CYP3A2 in all experiments, but the results failed to reach signiﬁcance due to the high variation between the experiments, ND—not detectable; “‡” denotes one or two out of three experiments showed induction.

when expressed relative to GAPDH, was higher in the liver com- 3.4.2. Human liver pared to that in the ileum (1.5–4 fold); the opposite was observed In human liver slices, CYP3A4 expression was significantly for the VDR expression, which was significantly higher in the ileum decreased to 10–20% upon incubation for 24 h in the solvent-treated than in the liver. controls (Fig. 6A). 1,25(OH)2D3 induced CYP3A4 in three out of four livers (Fig. 6B) (fold induction at 100 and 200 nM were: human liver 3.4.1. Human ileum (HL)1, 2.66/2.29; HL2, 2.89/0.83; HL3, 0.33/0.25; HL5, 1.43/1.41). Incubation of ileum slices with increasing concentrations of CDCA and DEX induced CYP3A4 significantly in all the five human 1,25(OH)2D3 induced CYP3A4 mRNA expression (Fig. 5B). DEX and livers studied (Fig. 6C and D). BUD but not CDCA also induced CYP3A4 mRNA expression in the The effects of the ligands for VDR, PXR, GR and FXR on the regula- ileum slices (Fig. 5C and D). tion of CYP3A isozymes in rat intestine (jejunum, ileum and colon),

211 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125 121

Fig. 4. Slices from rat intestine (jejunum, ileum and colon) were exposed to DEX (1 and 50 ␮M), BUD (10 nM) and PCN (10 ␮M) (A) for 12 h. Liver slices were exposed to DEX (1, 10 and 50 ␮M), BUD (10 and 100 nM) and PCN (10 ␮M) (B) for 8 h after, which total RNA was isolated and mRNA expression of PXR was evaluated by qRT- PCR. Results were expressed as fold induction after being normalized to the villin for the intestine and GAPDH for liver expression, and compared with the control slices (values set to 1) that were incubated for 12 and 8 h, respectively. Results were mean ± S.E.M. of 3–5 rats; in each experiment 6 intestinal and 3 liver slices were incubated per condition. Signiﬁcant differences towards the control incubations were denoted by *P < 0.05.

intestine and in the liver of the rat and humans with PXR-, GR- and FXR-speciﬁc ligands, and investigated whether the changes were related to expression levels of the NRs in the tissues. Most data con- cerning the regulation of CYP3A isoforms in the rat are restricted to the liver and the small intestine, mostly jejunum, and the data on the comparison of regulation of CYP3A isozymes across the intesti-

Fig. 3. Slices from rat liver were exposed to 1,25(OH)2D3 (10, 100 and 200 nM) (A), nal tract, jejunum, ileum and colon is scarce. In vivo, the extent of CDCA (50 ␮M) (B), and DEX (1, 10 and 50 ␮M), BUD (10 and 100 nM) and PCN (10 ␮M) exposure of the various organs to ligands of the NRs is rarely con- (C) for 8 h, after which total RNA was isolated and mRNA expression of CYP3A1 (A), trolled, and it is difficult to discriminate between the direct and CYP3A2 (B) and CYP3A9 (C) were evaluated by qRT-PCR. Results were expressed as fold induction, after being normalized to the GAPDH expression, and compared with indirect effects of the ligands. In this study, we compared the reg- the control slices that were incubated for 8 h, whose value was set to 1. Results were ulation of gene expression in different segments of the intestine mean ± S.E.M. of 3–5 rats; in each experiment, 3 slices were incubated per condition. and liver under identical experimental conditions using precision- Significant differences towards the control incubations are denoted by *, denoting cut tissue slices (Olinga et al., 2008; van de Kerkhof et al., 2007a, P < 0.05. 2008). Viability of the liver and intestinal slices during incubation was revealed by the stable expression of house-keeping genes, villin rat liver, human ileum and liver are summarized as an overview in as specific gene for enterocytes, and GAPDH for the intestinal and Table 4 to facilitate comparison of the effects of the various ligands liver tissue (data not shown). In addition, the ATP content of the in the different tissues. human liver slices was assessed as an additional viability marker during incubation; these levels were found to be constant during 4. Discussion incubation (Table 1). Furthermore, metabolism in tissue slices is comparable to in vivo (Graaf et al., 2007; van de Kerkhof et al., In this report, we compared the regulation of CYP3A isozymes 2007b) with adequate expression of transporters and enzymes. by the VDR-specific ligand, 1,25(OH)2D3, in different regions of the Therefore, the uptake and metabolism of ligands is expected to be

212 122 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125

Fig. 5. Slices from human ileum were exposed to control solvents (ethanol, methanol, and DMSO) (A), 1,25(OH)2D3 (10, 50 and 100 nM) (B), CDCA (50 ␮M) (C), DEX (1 and 50 ␮M) and BUD (10 nM) (D) for 12 and 24 h, after which total RNA was isolated and mRNA expression of CYP3A4 was evaluated by qRT-PCR. Results were expressed as fold induction after being normalized to the villin and compared to the control slices (values set as 1) that were incubated for 12 and 24 h. Results were mean ± S.E.M. of 4–5 human ileum donors; in each experiment, 3 slices were incubated per condition. Significant difference towards the control incubations is denoted by *P < 0.05 and **P < 0.01. “†” denotes induction of CYP3A4 in all experiments with high variation between the experiments, and failed to reach statistical significance. similar to in vivo and to reflect species differences between human induction responses, monitored by the induction of signature and rat. The concentrations of ligands for various nuclear recep- genes. tors used in this study are similar to those used in earlier studies In rat intestine, the mRNA expression of PXR, FXR and VDR was with metabolically active cells (Drocourt et al., 2002; Hoen et al., found to be present in varying abundances along the length of the 2000), which seem to be adequate to elicit nuclear receptor-specific small intestine (jejunum and ileum), with the highest expression of

Fig. 6. Slices from human liver were exposed to control solvents (EtOH, MeOH and DMSO) (A), 1,25(OH)2D3 (100 and 200 nM) (B), CDCA (100 ␮M) (C) and DEX (50 ␮M) (D) for 24 h, after which total RNA was isolated and CYP3A4 mRNA expression was evaluated by qRT-PCR. Results were expressed as fold induction after being normalized to GAPDH and compared to the control slices (values set as 1) that were incubated for 24 h. Results were mean ± S.E.M. of 5 human liver donors for all ligands except for1,25(OH)2D3 where n = 4; in each experiment 3 slices were incubated per condition. Signiﬁcant differences towards the control incubations are indicated with *P < 0.05 and **P < 0.01. “¥ ” denotes induction of CYP3A4 in three out of four human livers, which fails to reach statistical signiﬁcance.

213 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125 123

Table 4 Effect of VDR, PXR, FXR and GR ligands on the expression of CYP3A isozymes in rat and human intestine and liver.

Ligands Nuclear receptor Rat Human

CYP3A1 CYP3A2 CYP3A9 CYP3A4

Intestine Liver Intestine Liver Intestine Liver IL Liver

J IL Co J IL Co J IL Co

c 1,25(OH)2D3 VDR ↑ ↑ ↑ ↔ ↔↑ ↔ ↔ ↔↔↔ ↔ ↑ ↑ PCN PXR ↑a ↑↔ ↑ Not detectable ↔ ↑↑↑ ↑ Not done CDCA FXR/VDR ↔a ↑↔ ↓ Not detectable ↓ ↔↔↔ ↓ ↔ ↑ BUD GR ↔a ↔↔ ↔ Not detectable ↔ ↑↑↑ ↑ ↔ Not done DEX PXR/GR ↑b ↑↔ ↑ Not detectable ↑ ↑↑↑ ↑ ↑ ↑

J, jejunum; IL, ileum; Co, colon. ↑, induction; ↓, repression; ↔, no induction. a ↔, induction in one out of three experiment. b ↑, induction with high variation between the experiments. c ↑, induction in three out of four experiments.

all the NRs in the colon. In the rat, the expression of PXR and FXR was by 1,25(OH)2D3 along the length of the intestine, as reported by 2–10-fold lower in the intestine than in the liver, while in humans, Xu et al. (2006) and Chow et al. (2008) in rats in vivo. This finding the expression of PXR and FXR was 1.5–4-fold lower in the ileum contrasts that of Zierold et al. (2006). Our novel observation on the compared to the liver. Whether this lower liver to intestine ratio of induction of CYP3A2 by 1,25(OH)2D3 in the rat ileum emphasizes the NRs in humans is significantly different from that in rats could the segmental regulation of CYP3A isozymes in the small intestine. not be concluded from our data, because human samples showed In contrast to the effects of 1,25(OH)2D3 on CYP3A isozymes larger inter-individual differences and the liver and intestinal sam- in rat intestine, the prototypical PXR ligand, PCN, and DEX, which ples were obtained from different patients. The expression of VDR is a GR ligand at low concentrations (<1 ␮M) and a PXR ligand at mRNA showed an increasing gradient from rat jejunum to colon, in higher concentrations (>1 ␮M), induced CYP3A9 mRNA expression contrast to the reported gradual decrease of VDR receptor concen- in the jejunum, ileum and colon, CYP3A1 in the ileum only, but did tration from jejunum to ileum, as determined by the 1,25(OH)2D3 not affect the expression of CYP3A2 along the entire length of the binding assay (Feldman et al., 1979). The VDR expression relative intestine. BUD, a specific GR ligand, induced CYP3A9 expression to GAPDH was even higher (up to 2500-fold) in rat intestine and but not CYP3A1 and CYP3A2 in the jejunum, ileum and colon slices. human ileum compared to rat and human liver, respectively, as CDCA, the FXR ligand, induced CYP3A1 in the ileum slices but not reported earlier (Chan and Atkins, 1984; Sandgren et al., 1991). in jejunum and colon slices. Our results on the induction of CYP3A1 Immunohistochemical staining showed that the VDR protein was by PXR ligands, PCN and DEX, in rat intestine are consistent with exclusively localized in the bile duct epithelial cells (BECs) in rat earlier reports on rat jejunum explants (Schmiedlin-Ren et al., livers, whereas in the human livers, not only BEC cells but also 1993). Based on the BUD results, a synthetic GR ligand which did hepatocytes contained VDR protein, though to a lower extent (data not affect CYP3A1 mRNA expression, we conclude that CYP3A1 not shown), confirming the earlier findings of Gascon-Barre et al. is not regulated by GR. The induction by DEX at 1 ␮M, can be (2003). In human livers, the mRNA expression of VDR showed high explained as a PXR mediated effect, which was further confirmed inter-individual variations; only three out of four human livers by the observation that induction of CYP3A1 occurred with PCN, showed detectable VDR expression. the PXR ligand. The observation on induction of CYP3A9 by PCN, In the rat intestine, the VDR ligand, 1,25(OH)2D3 strongly DEX and BUD in rat intestine had not been reported earlier. Our induced CYP3A1 mRNA along the entire length of the intestine and results suggest that apart from PXR, CYP3A9 expression was also CYP3A2 only in ileum, but did not affect CYP3A9 expression. The regulated by GR. However our data failed to discriminate whether induction of CYP3A1 mRNA by 1,25(OH)2D3 which was found in BUD mediated regulation of CYP3A9 acted via the GRE in the rat jejunum slices is consistent with earlier in vivo report by Xu promoter or indirectly via the induction of HNF4␣. The induction et al. (2006) in Sprague–Dawley rats. We also report on CYP3A1 of PXR by DEX via GR, as reported previously by Huss and Kasper induction by 1,25(OH)2D3 in ileum and colon slices, with the high- (2000), was evident in our studies, since BUD and DEX both est induction occurring in ileum slices compared to the jejunum induced PXR (Fig. 4A). The induction potential of the PXR ligands, and colon slices, where induction was similar, despite the highest PCN and DEX (at 50 ␮M) on CYP3A1 in the ileum but not in the expression of VDR in colon. Recently, Chow et al. (2008) showed colon slices did not correlate with the higher expression of PXR in dose-dependent induction of CYP3A1 in the duodenum, jejunum, the colon compared to the jejunum. and ileum and not the colon in the Sprague–Dawley rats in vivo Although an FXRE has not been identified in the CYP3A1 pro- after intraperitoneal injections of 1,25(OH)2D3 for 4 days. This is moter, the FXR ligand, CDCA, was found to increase CYP3A1 mRNA likely explained by lower exposure of the colon than the small in the ileum. This observation could be the result of VDR-mediated intestine to 1,25(OH)2D3 than the small intestine in vivo. In vivo, regulation, since CDCA is also a VDR ligand, albeit of relatively low CYP3A2 mRNA levels were found to be very low and undetectable, affinity (Makishima et al., 2002). However, induction of CYP3A1 was rendering the study of the regulation of CYP3A2 along the length not observed in the colon with CDCA despite the high FXR and VDR of the rat intestine difficult (Chow et al., 2008). Recently Xu et expression. The possible explanation that CDCA is not efficiently al. (2006) and Chow et al. (2008) reported that CYP3A2 gene was taken up into the colonocytes is in contradiction with our finding ␣ ␤ not responsive to 1,25(OH)2D3 treatment. We also found very low that CDCA showed a strong upregulation of the Ost and Ost genes expressions of CYP3A2 mRNA along the length of the intestine of in rat colon slices (Khan et al., manuscript submitted). Among the the Wistar rats, and found, surprisingly, that 1,25(OH)2D3 signifi- VDR, FXR, PXR and GR ligands, the VDR ligand, 1,25(OH)2D3,was cantly induced CYP3A2 mRNA in the ileum, though not in jejunum by far the strongest inducer of CYP3A1 and also induced CYP3A2 in and colon slices. Furthermore, CYP3A9 expression was unaffected ileum. These results also showed that, although CYP3A1 was upreg-

214 124 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125 ulated by VDR, PXR and FXR ligands and CYP3A9 by PXR and GR In summary, studies in tissue slices showed that the overall ligands, dramatic differences in the extents of the induction were effects of ligands for NRs on regulation of CYP3A isozymes dif- found in the different segments of the intestine (Table 4). These differed in different regions of the rat intestine and liver, and human ferences were apparently not related to the differential expression ileum and liver slices, despite the incubation was conducted under of the respective NRs. The colon, although endowed with an abun- identical circumstances. This difference appears not to be directly dance of NRs, exhibited low induction potential of CYP3A isozymes related to the different expression levels of the nuclear receptors compared to those of the jejunum and ileum. involved. In the rat intestine, CYP3A1 expression is very sensitive The VDR-, PXR-, FXR- and GR-dependent regulation of CYP3A1, to the VDR ligand, and to a lesser extent, to PXR and GR ligands, CYP3A2 and CYP3A9 mRNA in rat liver slices differed dramat- whereas CYP3A2 expression is exclusively regulated by the VDR. ically from intestinal slices. The expression of CYP3A1, CYP3A2 CYP3A9 expression both in the liver and in all regions of the intes- and CYP3A9 mRNA was unchanged in liver slices incubated with tine appears to be mainly regulated by PXR and GR but not by VDR. 1,25(OH)2D3, as found in vivo by Xu et al. (2006) and Chow et al. In human tissue, however, CYP3A4 in ileum and liver was upregu- (2008). The lack of regulation of CYP3A1 in liver can be explained by lated by PXR, VDR and GR ligands. By contrast, CDCA elicited varying the absence of VDR in rat hepatocytes, the major site of the target effects, ranging from decreased expression in rat liver, lack of effect CYP genes, since VDR is found only in non-parenchymal cells and in human ileum, and increased expression in rat ileum and human biliary epithelial cells (Gascon-Barre et al., 2003). This was con- liver, effects that are not explained by the expression of FXR. Our firmed by immunohistochemisty of rat liver slices in our studies results suggest that prediction of the inducing potential of drugs where CYP3A1 was present exclusively in hepatocytes and CYP3A2, should not rely strictly on whether or not the drug under study is mainly in hepatocytes, and expressed at a much lower level in bil- a ligand for a certain NR and the expression levels of this NR in iary epithelial cells (unpublished observations). the target organ. Uptake, metabolism and excretion of the ligand as In contrast, PCN and DEX induced the expression of hepatic well as the availability of co-activators or repressors in the specific CYP3A1 and CYP3A9, and BUD induced CYP3A9, whereas CYP3A2 tissue and species may play a decisive role. expression was modestly induced only by DEX. These data on the induction of CYP3A1 and CYP3A9 by the PXR ligands, PCN and Acknowledgments DEX, agree with earlier reports on Sprague–Dawley rats (Huss and Kasper, 1998; Mahnke et al., 1997). However, the induction of The authors thank Dr. Vincent B. Nieuwenhuijs (University Med- CYP3A9 by BUD (GR), though suggested by Komori and Oda (1994), ical Center, Groningen) for providing the human ileum tissue. has not been reported earlier. Induction of CYP3A9 by PCN and DEX Grants: This work was supported in part by the Canadian Insti- in the rat liver and intestine implies that CYP3A9 is likely regulated tutes for Health Research, MOP89850. by PXR via a PXRE. However, it remains to be elucidated whether the inductive effect of BUD on CYP3A9 is directly mediated via GR References and a GRE in the CYP3A9 promoter, or indirectly via upregulation of PXR and HNF4␣ by BUD. The two stage induction by the GR on Chan, S.D., Atkins, D., 1984. The temporal distribution of the 1 alpha,25- the upregulation of CYP via induction of PXR has been suggested dihydroxycholecalciferol receptor in the rat jejunal villus. Clin. Sci. (Lond.) 67, 285–290. for CYP3A1/23 (Huss and Kasper, 2000), and is a likely possibility Chow, E.C.Y., Liu, S., Sun, H., Khan, A.A., Groothuis, G.M.M., Pang, K.S., 2008. Effects since PXR induction was also observed with GR ligands (Fig. 4B). of 1␣,25-dihydroxyvitamin D3 (calcitriol) and the vitamin D receptor (VDR) on Unlike the induction of CYP3A1 observed in the ileum, CDCA, an rat enzymes, transporters and nuclear receptors. In: Proceedings of the AAPS FXR ligand, showed repression of CYP3A1, CYP3A2 and CYP3A9 in Annual Meeting, Atlanta, GA. de Kanter, R., Monshouwer, M., Meijer, D.K., Groothuis, G.M., 2002. Precision-cut rat liver slices. This might be due to the CDCA mediated repression organ slices as a tool to study toxicity and metabolism of xenobiotics with special of PXR (data not shown), which in turn, affects the basal expression reference to non-hepatic tissues. Curr. Drug Metab. 3, 39–59. of CYP3A isozymes in rat liver. These results are in stark contrast Drocourt, L., Ourlin, J.C., Pascussi, J.M., Maurel, P., Vilarem, M.J., 2002. Expression of CYP3A4, CYP2B6, and CYP2C9 is regulated by the vitamin D receptor pathway to the mouse studies of Jung et al. (2006), where Cyp3a11 and pxr in primary human hepatocytes. J. Biol. Chem. 277, 25125–25132. were positively regulated by FXR in mice that were treated with FXR Feldman, D., McCain, T.A., Hirst, M.A., Chen, T.L., Colston, K.W., 1979. Characterization agonists, cholic acid and GW4064, suggesting species difference in of a cytoplasmic receptor-like binder for 1 alpha, 25-dihydroxycholecalciferol in rat intestinal mucosa. J. Biol. Chem. 254, 10378–10384. the regulation of these genes. Fukumori, S., Murata, T., Taguchi, M., Hashimoto, Y., 2007. Rapid and drastic induction In the human ileum, CYP3A4 was significantly induced by VDR, of CYP3A4 mRNA expression via vitamin D receptor in human intestinal LS180 PXR and GR ligands (Fig. 6B and D). Although the regulation of cells. Drug Metab. Pharmacokinet. 22, 377–381. Gascon-Barre, M., Demers, C., Mirshahi, A., Neron, S., Zalzal, S., Nanci, A., 2003. The CYP3A4 in vivo by PXR and GR ligands in the human intestine is normal liver harbors the vitamin D nuclear receptor in nonparenchymal and well known, we are the first to show the regulation of CYP3A4 biliary epithelial cells. Hepatology 37, 1034–1042. Gnerre, C., Blattler, S., Kaufmann, M.R., Looser, R., Meyer, U.A., 2004. Regulation of in human intestinal tissue by the VDR ligand, 1,25(OH)2D3. These CYP3A4 by the bile acid receptor FXR: evidence for functional binding sites in observations agree with findings in cultured monolayers such as the CYP3A4 gene. Pharmacogenetics 14, 635–645. Caco-2 and LS180 cells (Fukumori et al., 2007; Schmiedlin-Ren et Gonzalez, F.J., 1988. The molecular biology of cytochrome P450s. Pharmacol. Rev. 40, al., 2001). Also in human liver slices, CYP3A4 is upregulated by 243–288. Gonzalez, F.J., Kimura, S., Song, B.J., Pastewka, J., Gelboin, H.V., Hardwick, J.P., 1986. 1,25(OH)2D3, but only in those three liver samples that showed Sequence of two related P-450 mRNAs transcriptionally increased during rat expression of VDR, as reported for human hepatocytes (Drocourt development. An R. dre. 1 sequence occupies the complete 3 untranslated region et al., 2002). For the liver that did not express VDR, no upregula- of a liver mRNA. J. Biol. Chem. 261, 10667–10672. tion of CYP3A4 by 1,25(OH) D was found. This variation between Gonzalez, F.J., Nebert, D.W., Hardwick, J.P., Kasper, C.B., 1985. Complete cDNA and 2 3 protein sequence of a pregnenolone 16 alpha-carbonitrile-induced cytochrome samples was not due to viability differences because human liver P-450. A representative of a new gene family. J. Biol. Chem. 260, 7435–7441. slices treated with the PXR/GR ligand, DEX induced CYP3A4 in all Graaf, I.A., Groothuis, G.M., Olinga, P., 2007. Precision-cut tissue slices as a tool to the experiments, observations that were in agreement with earlier predict metabolism of novel drugs. Expert Opin. Drug Metab. Toxicol. 3, 879–898. Guengerich, F.P., Martin, M.V., Beaune, P.H., Kremers, P., Wolff, T., Waxman, D.J., report (Lehmann et al., 1998). CDCA induced expression of CYP3A4 1986. Characterization of rat and human liver microsomal cytochrome P-450 in human liver, confirming the FXR-dependent regulation in earlier forms involved in nifedipine oxidation, a prototype for genetic polymorphism reports by Gnerre et al. (2004). But this response was not observed in oxidative drug metabolism. J. Biol. Chem. 261, 5051–5060. Hashimoto, H., Toide, K., Kitamura, R., Fujita, M., Tagawa, S., Itoh, S., Kamataki, T., in the ileum. It appears that all NRs: VDR, PXR, GR and FXR, are able 1993. Gene structure of CYP3A4, an adult-specific form of cytochrome P450 in to induce CYP3A4 in the human liver. human livers, and its transcriptional control. Eur. J. Biochem. 218, 585–595.

215 A.A. Khan et al. / European Journal of Pharmaceutical Sciences 37 (2009) 115–125 125

Hoen, P.A., Commandeur, J.N., Vermeulen, N.P., Van Berkel, T.J., Bijsterbosch, Sandgren, M.E., Bronnegard, M., DeLuca, H.F., 1991. Tissue distribution of the 1,25- M.K., 2000. Selective induction of cytochrome P450 3A1 by dexamethasone dihydroxyvitamin D3 receptor in the male rat. Biochem. Biophys. Res. Commun. in cultured rat hepatocytes: analysis with a novel reverse transcriptase- 181, 611–616. polymerase chain reaction assay section sign. Biochem. Pharmacol. 60, Schmiedlin-Ren, P., Benedict, P.E., Dobbins 3rd, W.O., Ghosh, M., Kolars, J.C., Watkins, 1509–1518. P.B., 1993. Cultured adult rat jejunal explants as a model for studying regulation Huss, J.M., Kasper, C.B., 1998. Nuclear receptor involvement in the regulation of rat of CYP3A. Biochem. Pharmacol. 46, 905–918. cytochrome P450 3A23 expression. J. Biol. Chem. 273, 16155–16162. Schmiedlin-Ren, P., Thummel, K.E., Fisher, J.M., Paine, M.F., Watkins, P.B., 2001. Huss, J.M., Kasper, C.B., 2000. Two-stage glucocorticoid induction of CYP3A23 Induction of CYP3A4 by 1 alpha,25-dihydroxyvitamin D3 is human cell line- through both the glucocorticoid and pregnane X receptors. Mol. Pharmacol. 58, specific and is unlikely to involve pregnane X receptor. Drug Metab. Dispos. 29, 48–57. 1446–1453. Huss, J.M., Wang, S.I., Kasper, C.B., 1999. Differential glucocorticoid responses of Strotkamp, D., Roos, P.H., Hanstein, W.G., 1995. A novel CYP3 gene from female rats. CYP3A23 and CYP3A2 are mediated by selective binding of orphan nuclear Biochim. Biophys. Acta 1260, 341–344. receptors. Arch. Biochem. Biophys. 372, 321–332. Thompson, P.D., Hsieh, J.C., Whitfield, G.K., Haussler, C.A., Jurutka, P.W., Galligan, Jung, D., Mangelsdorf, D.J., Meyer, U.A., 2006. Pregnane X receptor is a target of M.A., Tillman, J.B., Spindler, S.R., Haussler, M.R., 1999. Vitamin D receptor displays farnesoid X receptor. J. Biol. Chem. 281, 19081–19091. DNA binding and transactivation as a heterodimer with the retinoid X receptor, Kolars, J.C., Lown, K.S., Schmiedlin-Ren, P., Ghosh, M., Fang, C., Wrighton, S.A., Merion, but not with the thyroid hormone receptor. J. Cell. Biochem. 75, 462–480. R.M., Watkins, P.B., 1994. CYP3A gene expression in human gut epithelium. Phar- Thummel, K.E., Brimer, C., Yasuda, K., Thottassery, J., Senn, T., Lin, Y., Ishizuka, H., macogenetics 4, 247–259. Kharasch, E., Schuetz, J., Schuetz, E., 2001. Transcriptional control of intestinal Komori, M., Oda, Y., 1994. A major glucocorticoid-inducible P450 in rat liver is not cytochrome P-4503A by 1alpha,25-dihydroxy vitamin D3. Mol. Pharmacol. 60, P450 3A1. J. Biochem. 116, 114–120. 1399–1406. Lehmann, J.M., McKee, D.D., Watson, M.A., Willson, T.M., Moore, J.T., Kliewer, S.A., van de Kerkhof, E.G., de Graaf, I.A., de Jager, M.H., Groothuis, G.M., 2007a. Induction 1998. The human orphan nuclear receptor PXR is activated by compounds that of phase I and II drug metabolism in rat small intestine and colon in vitro. Drug regulate CYP3A4 gene expression and cause drug interactions. J. Clin. Invest. 102, Metab. Dispos. 35, 898–907. 1016–1023. van de Kerkhof, E.G., de Graaf, I.A., de Jager, M.H., Meijer, D.K., Groothuis, G.M., 2005. Li, Q., Sai, Y., Kato, Y., Tamai, I., Tsuji, A., 2003. Influence of drugs and nutrients on Characterization of rat small intestinal and colon precision-cut slices as an in transporter gene expression levels in Caco-2 and LS180 intestinal epithelial cell vitro system for drug metabolism and induction studies. Drug Metab. Dispos. lines. Pharm. Res. 20, 1119–1124. 33, 1613–1620. Liu, S., Tam, D., Chen, X., Pang, K.S., 2006. P-glycoprotein and an unstirred water van de Kerkhof, E.G., de Graaf, I.A., Groothuis, G.M., 2007b. In vitro methods to study layer barring digoxin absorption in the vascularly perfused rat small intestine intestinal drug metabolism. Curr. Drug Metab. 8, 658–675. preparation: induction studies with pregnenolone-16alpha-carbonitrile. Drug van de Kerkhof, E.G., de Graaf, I.A., Ungell, A.L., Groothuis, G.M., 2008. Induction of Metab. Dispos. 34, 1468–1479. metabolism and transport in human intestine: validation of precision-cut slices Lu, A.Y., Somogyi, A., West, S., Kuntzman, R., Conney, A.H., 1972. Pregnenolone- as a tool to study induction of drug metabolism in human intestine in vitro. Drug 16-carbonitrile: a new type of inducer of drug-metabolizing enzymes. Arch. Metab. Dispos. 36, 604–613. Biochem. Biophys. 152, 457–462. van de Kerkhof, E.G., Ungell, A.L., Sjoberg, A.K., de Jager, M.H., Hilgendorf, C., de Mahnke, A., Strotkamp, D., Roos, P.H., Hanstein, W.G., Chabot, G.G., Nef, P., 1997. Graaf, I.A., Groothuis, G.M., 2006. Innovative methods to study human intestinal Expression and inducibility of cytochrome P450 3A9 (CYP3A9) and other drug metabolism in vitro: precision-cut slices compared with using chamber members of the CYP3A subfamily in rat liver. Arch. Biochem. Biophys. 337, preparations. Drug Metab. Dispos. 34, 1893–1902. 62–68. Wang, H., Kawashima, H., Strobel, H.W., 1996. cDNA cloning of a novel CYP3A from Makishima, M., Lu, T.T., Xie, W., Whitfield, G.K., Domoto, H., Evans, R.M., Haussler, rat brain. Biochem. Biophys. Res. Commun. 221, 157–162. M.R., Mangelsdorf, D.J., 2002. Vitamin D receptor as an intestinal bile acid sensor. Wang, H., Strobel, H.W., 1997. Regulation of CYP3A9 gene expression by estrogen Science 296, 1313–1316. and catalytic studies using cytochrome P450 3A9 expressed in Escherichia coli. Matsubara, T., Kim, H.J., Miyata, M., Shimada, M., Nagata, K., Yamazoe, Y., 2004. Iso- Arch. Biochem. Biophys. 344, 365–372. lation and characterization of a new major intestinal CYP3A form, CYP3A62, in Xu, Y., Iwanaga, K., Zhou, C., Cheesman, M.J., Farin, F., Thummel, K.E., 2006. Selec- the rat. J. Pharmacol. Exp. Ther. 309, 1282–1290. tive induction of intestinal CYP3A23 by 1alpha,25-dihydroxyvitamin D3 in rats. Nagata, K., Murayama, N., Miyata, M., Shimada, M., Urahashi, A., Yamazoe, Y., Kato, R., Biochem. Pharmacol. 72, 385–392. 1996. Isolation and characterization of a new rat P450 (CYP3A18) cDNA encod- Zhang, Q.Y., He, W., Dunbar, D., Kaminsky, L., 1997. Induction of CYP1A1 by beta- ing P450(6)beta-2 catalyzing testosterone 6 beta- and 16 alpha-hydroxylations. naphthoflavone in IEC-18 rat intestinal epithelial cells and potentiation of Pharmacogenetics 6, 103–111. induction by dibutyryl cAMP. Biochem. Biophys. Res. Commun. 233, 623–626. Olinga, P., Elferink, M.G., Draaisma, A.L., Merema, M.T., Castell, J.V., Perez, G., Zierold, C., Mings, J.A., Deluca, H.F., 2006. 19nor-1,25-dihydroxyvitamin D2 specifi- Groothuis, G.M., 2008. Coordinated induction of drug transporters and phase cally induces CYP3A9 in rat intestine more strongly than 1,25-dihydroxyvitamin I and II metabolism in human liver slices. Eur. J. Pharm. Sci. 33, 380–389. D3 in vivo and in vitro. Mol. Pharmacol. 69, 1740–1747.

216

APPENDIX A2

Khan AA, Chow EC, Porte RJ, Pang KS and Groothuis GM (2009) Expression and regulation of the bile acid transporter, OSTα-OST in rat and human intestine and liver. Biopharm Drug Dispos 30:241-258

217

BIOPHARMACEUTICS & DRUG DISPOSITION Biopharm. Drug Dispos. 30: 241–258 (2009) Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/bdd.663

Expression and Regulation of the Bile Acid Transporter, OSTa-OSTb in Rat and Human Intestine and Liver

Ã Ansar A. Khana, , Edwin C. Y. Chowb, Robert J. Portec, K. Sandy Pangb and Geny M. M. Groothuisa aPharmacokinetics, Toxicology and Targeting, Department of Pharmacy, University of Groningen, The Netherlands bDepartment of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, Canada cDepartments of Hepatobiliary Surgery and Liver Transplantation, University of Groningen, University Medical Center Groningen, The Netherlands

ABSTRACT: The regulation of the OSTa and OSTb expression was studied in the rat jejunum, ileum, colon and liver and in human ileum and liver by ligands for the farnesoid X receptor (FXR), pregnane X receptor (PXR), vitamin D receptor (VDR) and glucocorticoid receptor (GR) using precision cut tissue slices. The gradient of protein and mRNA expression in segments of the intestine for rOSTa and rOSTb paralleled that of rASBT. OSTa and OSTb mRNA expression, quantified by qRT-PCR, in rat jejunum, ileum, colon and liver, and in human ileum and liver was positively regulated by FXR and GR ligands. In contrast, the VDR ligand, 1,25(OH)2D3 decreased the expression of rOSTa-rOSTb in rat intestine, but had no effect on human ileum, and rat and human liver slices. Lithocholic acid (LCA) decreased the expression of rOSTa and rOSTb in rat ileum but induced OSTa-OSTb expression in rat liver slices, and human ileum and liver slices. The PXR ligand, pregnenolone-16a carbonitrile (PCN) had no effect. This study suggest that, apart from FXR ligands, the OSTa and OSTb genes are also regulated by VDR and GR ligands and not by PXR ligands. This study show that VDR ligands exerted different effects on OSTa-OSTb in the rat and human intestine and liver compared with other nuclear receptors, FXR, PXR, and GR, pointing to species- and organ-specific differences in the regulation of OSTa-OSTb genes. Copyright r 2009 John Wiley & Sons, Ltd. Key words: OSTa-OSTb; regulation; nuclear receptors; intestinal slices; liver slices

Introduction in the absorption of bile acids is the sodium dependent bile acid transporter, ASBT Bile acids (BA) undergo extensive enterohepatic (SLC10A2) [4], that is expressed along the apical cycling and are actively reabsorbed in the surface of ileocytes, BEC and renal proximal terminal part of the ileum, the bile duct epithelial tubular cells. In enterocytes, bile acids are cells (BEC) [1] and the renal proximal tubular effluxed out of the cells into the portal circula- cells [2,3]. They play an important role in the tion, and may be transported back to the regulation of bile acid synthesis and cholesterol intestinal lumen. Several basolateral bile acid homeostasis. The primary transporter involved transporters such as truncated ASBT (tASBT), MRP3 and MRP4, showing affinity towards bile acid transport have been proposed [5–8]. *Correspondence to: Department of Pharmacy, Pharmaco Although MRP3 was shown to transport bile kinetics, Toxicology and Targeting, A. Deusinglaan 1, 9713 salts and regulated by chenodeoxycholic acid AV Groningen, The Netherlands. E-mail: [email protected] (CDCA) in human ileum [7], its role in ileal bile

Received 19 January 2009 Revised 12 May 2009 Copyright r 2009 John Wiley & Sons, Ltd. Accepted 26 May 2009

218 242 A.A. KHAN ET AL. salt absorption may not be significant, since obtained indirectly from analysis of the livers of mrp3 / mice failed to show any apparent defect patients with cholestatic disease [19,20]. In the in bile acid absorption [9]. mouse in vivo, the regulation of Osta-Ostb by FXR Recently, Wang et al. [10] identified an organic in the intestine was shown [16]. In this study we solute transporter (OST) consisting of two half investigated whether OSTa and OSTb genes were transporters, a and b (OSTa and OSTb), in the regulated by ligands for the vitamin D receptor skate, Raja ernacea. Subsequently, rodent and (VDR) and glucocorticoid receptor (GR) in the rat human OSTa-OSTb orthologues that are able to and human liver and intestine, since these nuclear mediate sodium independent transport of organ- receptors were reported to regulate ASBT [21–23], ic anions, bile acids and sterols in transfected thebileacidtransporterthatwasundernegative Xenopus oocytes were identified [11]. The regulation by FXR in mouse, rabbit and human but expression of OSTa and OSTb are shown to not in rat intestine [24–26]. Precision-cut tissue parallel that of ASBT expression in enterocytes slices were used from the rat intestine (jejunum, along the length of the intestine and were ileum and colon) and liver and human ileum and co-incident with ASBT in BECs and renal liver, and the effects of VDR and GR ligands are proximal tubular cells of rat, mouse and human compared with those of FXR on the regulation of [11,12]. The OSTa and OSTb proteins are found to the mRNA expression of the OSTa and OSTb be localized at the basolateral membrane and genes. In addition, the involvement of PXR in the catalogued as the ileal bile acid basolateral regulation of OSTa and OSTb genes was also transporter in the mouse [12], since bile acid investigated. This ex vivo model enables to study homeostasis was perturbed in the Osta knockout the regulation of genes of interest under controlled mouse [13,14]. The mouse and human OSTa- and nearly physiological conditions directly, and OSTb genes are regulated by the farnesoid X allowed the comparison of direct effects of ligands receptor (FXR) and the liver X receptor a (LXRa) in different organs under identical conditions [15–17]. Both FXR and LXRa heterodimerize [27,28]. with the retinoic acid X receptor a (RXRa), and, upon ligand binding, the resulting complex binds to the inverted repeat-1 (IR1) in the promoters of OSTa and OSTb, thereby increasing their expres- Materials and Methods sion. Furthermore, human and mouse OSTa and OSTb promoters are endowed with binding sites Male Wistar (HsdCpb:WU) rats weighing about for the transcription factors, hepatocyte nuclear 230–250 g were purchased from Harlan (Horst, factor 4a (HNF4a) [17] and liver receptor homo- The Netherlands). Pieces of human liver and log protein-1 (LRH-1) [16,18]. ileum tissue were obtained as surgical waste Studies on rodent and human OSTa-OSTb from the University Medical Center, Groningen genes in the intestine and liver usually entail (UMCG) with the informed consent of the use of FXR and LXRa ligands on immortalized patients/donors. 1,25(OH)2D3 in ethanol was cell lines such as CT26, Caco-2, Huh-7 and purchased from BIOMOL Research Laboratories, HepG2 cells [15,17,19]. However, these cell lines Inc., Plymouth Meeting, PA. Chenodeoxy- lack the normal expression of various nuclear cholic acid (CDCA) and lithocholic acid (LCA) receptors, transporters and coactivators, and are were purchased from Calbiochem, San Diego, unable to reflect the regulation in distinct California, dexamethasone was obtained from segmental regions of OSTa and OSTb genes in Genfarma bv, Maarssen. Ethanol, methanol and intestine. In the rat, the regulation of rOSTa-rOSTb DMSO were purchased from Sigma-Aldrich genes has not been studied in great detail. Landrier Chemical Co. (St Louis, MO); gentamicin et al. [15] reported on the induction of hOSTa and Williams’s medium E with glutamax-I and hOSTb genes by CDCA, the FXR ligand, and amphotericin B (Fungizone)-solution were in human ileum biopsies after 4 h in culture. obtained from Gibco (Paisley, UK). D-Glucose However, evidence for the regulation of hOSTa- and HEPES were from ICN Biomedicals, Inc. hOSTb in human livers was predominantly (Eschwege, Germany). University of Wisconsin

219 REGULATION OF OSTa-OSTb IN RAT AND HUMAN INTESTINE AND LIVER 243 organ preservation solution (UW) was obtained et al. [29]. Human ileum tissue was obtained as a from Du Pont Critical Care, Waukegab, Illinois, part of the surgical waste after resection of the ileo- USA. Low gelling temperature agarose, budiso- colonic part of the intestine in colon carcinoma nide (BUD) and pregnenolone-16a carbonitrile patients. After surgical resection, the ileum tissue (PCN) were purchased from Sigma–Aldrich (St was placed immediately in Louis, MO). RNAeasy mini columns were ob- ice-cold carbogenated KHB. Human liver and tained from Qiagen, Hilden, Germany. Random ileum donor characteristics are as reported earlier m primers (500 g/ml), MgCl2 (25 mM), RT buffer [30]. Human liver and ileum slices were prepared (10X), the PCR nucleotide mix (10 mM), AMVRT within 30–60 min after resection. Rat and human (22 U/ml) and RNasin (40 U/ml) were purchased intestinal and liver slices were prepared according from Promega Corporation, Madison WI, USA. to the published methods [29,31,32]. Assay-on-DemandTM human GAPDH primers and probe for the Taqman analysis were pur- Induction studies chased from Applied Biosystems, Warrington, UK. All SYBR Green primers were purchased Precision-cut slices, prepared from rat intestine from Sigma Genosys. The Taq Master Mixes was (jejunum, ileum and colon) and human ileum procured from Eurogentech. The rabbit anti-rat were incubated individually in 12-well sterile Osta and Ostb antibodies were generous gifts tissue culture plates (Grenier bio-one GmbH, from Dr Ned Ballatori (Rochester, New York, Frickenhausen, Austria) containing 1.3 ml Wil- USA). The secondary antibody, Alexa Fluor-488 liams medium E supplemented with D-glucose anti-rabbit immunoglobulin (IgG) was purchased (final concentration of 25 mM), gentamicin sulfate from Invitrogen, Molecular Probes, Eugene, OR, (50 mg/ml), amphotericin/fungizone (250 mg/ml) USA. All reagents and materials used were of the and saturated with carbogen. The plates were highest purity that is commercially available. placed in humidified plastic container kept at 371C and continuously gassed with carbogen and Experimental protocols shaken at 80 rpm. Intestinal slices were incubated with 1,25(OH) D (final concentrations of 5 nM, All experimental protocols involving animals 2 3 10 nM and 100 nM), CDCA (final concentration of were approved by the Animal Ethical Committee 50 mM), LCA (final concentrations of 5 mM and of the University of Groningen. Experimental 10 mM), DEX (final concentrations of 1 mM and protocols involving human tissue (liver and 50 mM), BUD (final concentration of 10 nM) ileum) were approved by the Medical Ethical and PCN (final concentration of 10 mM) added Committee of the UMCG. as a 100-times concentrated stock solution in ethanol (1,25(OH) D ), methanol (CDCA and Preparation of rat and human intestinal and liver 2 3 LCA) or DMSO (DEX, BUD and PCN). Higher slices concentrations of CDCA (100 mM) and LCA The small intestine, colon and liver were excised (50 mM) were toxic to the intestinal slices. Rat from the rat under isoflurane/O2 anaesthesia. The intestinal slices were incubated for 12 h, since at small intestine and colon were immediately placed 24 h, the expression of villin was found to be into ice-cold Krebs-Henseleit buffer supplemented decreased. Human ileum slices were incubated with 10 mM HEPES, 25 mM sodium bicarbonate for 8 h and 24 h; villin expression was stable up to and 25 mMD-glucose, pH 7.4 (KHB), saturated 24 h. Data are presented for 24 h only, since the with carbogen (95% O2/5% CO2) and stored on ice results obtained at 24 h were not different from until preparation of the slices. The rat liver was those obtained at 8 h. Further, rat ileum slices stored in ice-cold UW until slicing. Pieces of were incubated in the presence of both human liver tissue were obtained from patients 1,25(OH)2D3 (final concentration of 100 nM) and undergoing partial hepatectomy for the removal of CDCA (final concentration of 50 mM). Control carcinoma (PH livers) or from redundant parts of slices were incubated in Williams medium E donor livers remaining after split-liver transplan- (supplemented with D-glucose and gentamicin tation (Tx livers) as described previously by Olinga sulfate) with 1% ethanol, methanol, DMSO and

220 244 A.A. KHAN ET AL. ethanol1DMSO without ligands. From a single rat presented for the 24 h time point, since the results or human tissue sample, six (rat intestine) or three were similar to those obtained at 8 h. Control liver (human ileum) replicate slices were subjected to slices were incubated in supplemented Williams each experimental condition. After the incubation, medium E with 1% ethanol, methanol and DMSO these replicate slices were harvested, pooled and without inducers. From a single rat/single human snap-frozen in liquid nitrogen to obtain sufficient liver donor, three replicate slices were subjected to total RNA for qRT-PCR analysis. Samples were identical incubation conditions. At the end of the stored at 801C until RNA isolation. These incubation these replicate slices were harvested, experiments were replicated in 3–5 rats and 3–5 pooled and snap-frozen in liquid nitrogen to obtain human ileum donors. sufficient total RNA for qRT-PCR analysis. Samples Liver slices (8 mm diameter and 250 mmthick) were stored at 801C until RNA isolation. These were incubated individually in sterile six-well experiments were replicated in 3–5 rats and 4–5 tissue culture plates (Grenier bio-one GmbH, human liver donors. Frickenhausen, Austria) containing 3.2 ml Williams medium E supplemented with D-glucose to a RNA isolation and quantitative real time PCR final concentration of 25 mM, gentamicin sulfate (qRT-PCR) (50 mg/ml) and saturated with carbogen. The plates were placed in humidified plastic container kept at Total RNA from rat and human intestine and liver 371C and continuously gassed with carbogen and samples was isolated using RNAeasy mini shaken at 80 rpm. Liver slices were induced with columns from Qiagen according to the manu- 1,25(OH)2D3 (final concentration, 100 nM), CDCA facturer’s instruction. The RNA concentration and (final concentration, 100 mM), LCA (final concentra- quality were determined by measuring the absor- tion, 50 mM), DEX (final concentration, 50 mM), BUD bance at 260 nm, 230 nm and 280 nm using a (final concentrations 10 nM and 100 nM)andPCN Nanodrop ND100 spectrophotometer (Wilming- (final concentration, 10 mM) added as a 100-fold ton, DE, USA). The ratios of absorbance measured concentrated stock solution in ethanol (for at 260 over 280 and 230 over 260 were found to be m m 1,25(OH)2D3), methanol (for CDCA and LCA) or above 1.8. About 2 g of total RNA in 50 lwas DMSO (DEX, BUD and PCN). Rat and human liver reverse-transcribed into template cDNA as re- slices were incubated for 8 h and 24 h. Data are ported earlier by van de Kerkhof et al. [33].

s Table 1. Sequence of oligonucleotides for quantitative Real-Time PCR, rat and human genes (SYBR and Taqman analysis)

Gene Forward primer (50-30) Reverse primer (50-30) Gene bank number rVillin GCTCTTTGAGTGCTCCAACC GGGGTGGGTCTTGAGGTATT XM_001057825 rGAPDH CTGTGGTCATGAGCCCCTCC CGCTGGTGCTGAGTATGTCG XR_008524 rOSTa CCCTCATACTTACCAGGAAGAAGCTAC CCATCAGGAATGAGAAACAGGC XM_221376 rOSTb TATTCCATCCTGGTTCTGGCAGT CGTTGTCTTGTGGCTGCTTCTT XM_238546 rSHP CTATTCTGTATGCACTTCTGAGCCC GGCAGTGGCTGTGAGATGC NM_057133 hVillin CAGCTAGTGAACAAGCCTGTAGAGGAGC CCACAGAAGTTTGTGCTCATAGGC NM_007127 ahGAPDH Assay-by-DesignTM ID - Hs99999905_m1 NM_002046 (Applied Biosystems) 6FAM–GCGCCTGGTCACCAGGGCTGCTTTT – NFQ arASBT ACCACTTGCTCCACACTGCTT CGTTCCTGAGTCAACCCACAT U07183 Probe - 6FAM - CTTGGAATGCCCCTTTGCCTCT- TAMRA ahOSTa AGATTGCTTGTTCGCCTCC TCACCACTTGGGGATCATTT NM_152672 Probe - 6FAM - CTCAAGTGATGAATTGCCACC TCCTCATACTGG-TAMRA ahOSTb CAGGAGCTGCTGGAAGAGAT GACCATGCTTATAATGACCACCA NM_178859 Probe - 6FAM - CGTGTGGAAGATGCATCTCC CTGGAATCATTC-TAMRA

s aPrimer sets for rat and human Taqman Gene analysis. r, rat genes; h, human genes.

221 REGULATION OF OSTa-OSTb IN RAT AND HUMAN INTESTINE AND LIVER 245

Quantitative real time PCR (qRT-PCR) for the (Lieca CM 3050) at 201C and placed on super- rat and human genes of interest was performed frostplusslides(Menzel,Braunchweig,Germany). using primer sequences listed in Table 1 by two Indirect immunofluorescence detection was per- s detection systems, SYBR Green or Taqman formed using Osta and Ostb antibodies according analysis according to the availability of primer to the protocol described previously [11]. In brief, sets. All primer sets were analysed using BLASTn tissue sections were fixed with acetone cooled to to ensure primer specificity for the gene of interest 201C for 10 min. Nonspecific binding sites were (http://www.ncbi.nlm.nih.gov/BLAST/). For the blocked with 1% bovine serum albumin (BSA) in SYBR Green, 50 ng of cDNA was used in a total phosphate buffered saline containing 0.05% Triton s reaction mixture of 20 ml. For the Taqman X 100. Primary antibodies were diluted in the analysis 250 ng of cDNA was used in a total blocking buffer, Osta (m315) (1:200) and Ostb reaction mixture of 10 ml. The PCR conditions (mB90) (1:150), and incubated with the sections for were similar to those described in an earlier report 2 h at room temperature. Subsequently, the sec- [30]. All samples were analysed in duplicates in tions were incubated with the secondary antibody 384-well plates using ABI7900HT from Applied (Alexa Flour-488) at a dilution of 1:50 in blocking Biosystems. Appropriate controls, consisting of buffer for 1 h at room temperature. water (with water instead of total mRNA, which has been subjected to the reverse transcription Data analysis protocol) and the mRNA control (isolated mRNA which has not been subjected to reverse transcrip- Allvalueswereexpressedasthemean7SD. All data DD tion protocol) were subjected to qRT-PCR to (fold-induction and CT)wereanalysedbypaired determine potential primer dimer formation and Student’s t-test using SPSS Version 16 for significant contamination of DNA in the isolated samples, differences between the means. The value po0.05 respectively. None of the primers showed dimer wasconsideredassignificant. formation. In addition, total RNA from the samples for the preparation of cDNA appeared to be free of DNA contamination. Dissociation Results curves showed a single homogenous product. The comparative threshold cycle (CT)method[34]was Expression of rASBT, rOSTa and rOSTb in rat used for relative quantification, where CT is intestine and liver inversely related to the abundance of mRNA a transcripts in the initial sample. The mean CT of The mRNA expression of rASBT, rOST and the duplicate measurements was used to calculate rOSTb genes was clearly detectable, not only in the difference between the CT for the gene of rat ileum but also in the jejunum and colon by interest and that of the reference gene (villin for qRT-PCR (Figure 1). Expressions of rASBT, D a b intestine and GAPDH for liver) ( CT), which was rOST and rOST mRNA were significantly D compared with the corresponding CT of the higher in rat ileum (average threshold cycles DD a solvent control ( CT). Data are expressed as fold (CT) 23 for rASBT, 17 for rOST and 16.5 for induction or repression of the gene of interest rOSTb) compared with those for the jejunum DD according to the formula 2 ( CT). (29 for rASBT, 20 for rOSTa and rOSTb) and colon (30 for rASBT, 24 for rOSTa and 22 for rOSTb). There was no difference between the C values in Immunolocalization of OSTa and OSTb in rat T the tissue and those in the slices at the start of the intestine and liver incubation. The gradient in expression of rOSTa, D The rat intestine was washed with 0.9% saline based on the CT values relative to villin and cut into small pieces. The intestinal tissue (jejunum:ileum:colon 5 1:3.8:0.2), was different was filled with Tissue Tek (Sakura Finetek from that of rOSTb (jejunum:ileum:colon 5 1:8:0.9) Europe, The Netherlands), then quickly frozen and rASBT (jejunum: ileum:colon 5 1:130:4) in cold isopentane (kept at 801C) and stored (Figure 1). In rat liver, the average threshold 1 m a b at 80 C. Sections of 5 m were cut in a cryostat cycles (CT)forrOST (30) and rOST (33) were

222 246 A.A. KHAN ET AL.

various solvents, did not alter the expression of rOSTa but rOSTb expression was significantly induced. In rat ileal slices, the expression of both rOSTa and rOSTb were significantly elevated (1.5- to 2-fold) (Figure 3A). In rat colon slices, the expression of both rOSTa and rOSTb was significantly decreased (3-fold) during incubation (Figure 3A). In rat liver slices, the mRNA expression of the rOSTa was not altered, whereas rOSTb was significantly downregulated during 24 h of incubation (4-fold) regard- less of the solvent used (Figure 3B). Figure 1. mRNA expression of rASBT, rOSTa and rOSTb The FXR ligand, CDCA, moderately induced transporters relative to villin expression in intestinal tissue (1.5- to 2-fold) the expression of rOSTa and (jejunum, ileum and colon) of the Wistar rat. The average rOSTb in the jejunum, ileum and liver, compared threshold cycles (CT) for rASBT in jejunum was 29, in ileum 23 a b with solvent treated control slices (Figures 4A, and in colon 30.The average CT for rOST and rOST in a b jejunum was 20 and 20, in ileum 17 and 16.5 respectively, and B,D). The induction of rOST and rOST was in colon 24 and 22, respectively. The mRNA expression of dramatically higher in the colon, amounting to rASBT, rOSTa and rOSTb transporters relative to villin in 25-fold for rOSTa and 45-fold for rOSTb (Figure 4C). ileum and colon was expressed relative to that in the jejunum, In contrast, incubation of rat intestine (jejunum, which was set to unity. Each bar represents the results of three Ã ÃÃ ileum and colon) and liver slices with LCA, an animals (n 5 3)7SD. po0.05; po0.001 FXR ligand, with affinity towards VDR [35], exhibited VDR dependent regulation of CYP3A much higher than in intestine. These distributions isozymes [36], displayed different effects on rOSTa were further confirmed at the protein level by and rOSTb genes. LCA significantly decreased the immunohistochemistry (Figure 2). In the rat expression of rOSTa and rOSTb in the ileum (Figure intestine, both rOSTa and rOSTb were detected at 4B), and rOSTa expression in the rat jejunum the basolateral membrane of the epithelial cells in without affecting rOSTb expression (Figure 4A). In all regions of the intestine (Figure 2). As expected, rat colon, LCA showed a strong, significant up- the highest expression was detected in ileum and a regulation (up to 10-fold) of the rOSTb gene low but clearly detectable expression was without significantly affecting the rOSTa gene observed in the colon and to a lesser extent in (Figure 4C). In liver slices, a small but significant the jejunum. In addition, a decreasing expression (1.5-fold) up-regulation of the rOSTa expression from the tip of villus to the crypts was found. In was found in the presence of LCA, whereas rOSTb the rat liver, rOSTa-rOSTb was visibly detectable at expression was decreased (Figure 4D). Further- the basolateral membrane of the BEC of the larger more, both CDCA and LCA significantly induced bile ducts and only a low expression was observed the SHP expression (2-fold induction) in slices of at the basolateral membrane of the hepatocyte. all regions of the rat intestine and liver (for the effect of CDCA on ileum, see Figure 6B; data not Regulation of rOSTa and rOSTb in rat intestine shown for other tissues). This observation on the and liver by bile acids up-regulation of SHP was expected for FXR a b The CT values of rOST and rOST genes were not ligands upon incubation with bile salts, and affected during the preparation of slices and similar confirms that the FXR pathway was intact in the CT values were found in the slices at the start of the slices, since these were able to respond to bile salts incubation and in freshly isolated tissue (data not as signalling agents of FXR. shown). However, the expression of rOSTa and rOSTb genes was significantly altered during Regulation of rOSTa and rOSTb in rat intestine incubation of slices at 371C, but the effects differed and liver by the VDR ligand in different segments of the rat intestine and liver slices (Figure 3A, B). Incubation of rat jejunum Incubation of rat intestinal slices in the presence slices for 12 h, either in the absence or presence of of the VDR ligand, 1,25(OH)2D3, resulted in a

223 REGULATION OF OSTa-OSTb IN RAT AND HUMAN INTESTINE AND LIVER 247

Figure 2. Indirect immunofluorescence showed that rOSTa and rOSTb were detected at the basolateral surface of the enterocytes in the jejunum, ileum and colon of the Wistar rat. In the liver, rOSTa and rOSTb proteins are predominantly localized in the bile duct epithelial cells and a low expression was observed in the hepatocytes. Control rat intestinal and liver sections incubated without primary antibodies for rOSTa and rOSTb did not show any fluorescence (results not shown) Copyright r 2009 John Wiley & Sons, Ltd. Biopharm. Drug Dispos. 30: 241–258 (2009) DOI: 10.1002/bdd

224 248 A.A. KHAN ET AL.

CDCA (50 mM), showed significant down-regulation of rOSTa and rOSTb (fold decrease– rOSTa 0.5 and of rOSTb 0.7; po0.05); the observations were identical to those for 1,25(OH)2D3 incubation alone (0.5-fold decrease of rOSTa and0.7-foldofrOSTb; po0.05) and contrasted those for CDCA, which induced rOSTa and rOSTb (1.5-fold induction of rOSTa and 1.55-fold of rOSTb; po0.05) (Figure 6A). SHP expression was induced by CDCA (fold o induction, 2.2; p 0.001) but not by 1,25(OH)2D3 (Figure 6B).

Regulation of rOSTa and rOSTb in rat intestine and liver by the GR and PXR ligands Incubation of rat intestinal (jejunum, ileum and colon) and liver slices with the GR/PXR ligand, DEX (1 mM and 50 mM for intestinal slices and 50 mM for liver slices), significantly induced the rOSTa and rOSTb expression in jejunum, colon and liver (Figures 7A, C, D) but not in the ileum (Figure 7B). These results in the intestinal slices were displayed again with BUD (10 nM), the specific GR ligand, which induced rOSTa and rOSTb expression in rat jejunum (Figure 7A) Figure 3. The effect of incubation at 371C on rat jejunum, and colon (Figure 7C) but not in the ileum ileum, colon (12 h) and liver slices (8 h and 24 h) on the a b (Figure 7B). However, the PXR ligand, PCN expression of rOST and rOST genes. The mRNA expression m a of rOSTa and rOSTb genes relative to villin (intestine) and (10 M), did not influence the rOST and GAPDH (liver) was quantified by real-time PCR and rOSTb expression in all regions of the intestine expressed with respect to the control slices without incubation (Figures 7A, B, C, D). In contrast to DEX, neither (0 h) for each of the intestinal segments, which were set to PCN nor BUD (10 nM and 100 nM) induced rOSTa unity. Results are expressed as mean7SD of 4–5 rats. Ã ÃÃ and rOSTb expression in liver slices during 8 h of po0.05; po0.001 incubation (data not shown), whereas BUD (100 nM) significantly induced rOSTa expression parallel decrease in expression of rOSTa and (fold induction 2.9; po0.05) (Figure 7D) during rOSTb in jejunum, ileum and colon in dose-a 24 h of incubation. Further, to confirm the intact- dependent manner (Figures 5A, B and C). Further- ness of the GR and PXR response in the rat more, 1,25(OH)2D3 significantly induced the intestinal (jejunum, ileum and colon) and liver expression of CYP3A1 (41000-fold induction), a slices, PXR, CYP3A1 and CYP3A9 mRNA VDR responsive gene in slices of all regions of the expression were analysed in these samples. The rat intestine [30]. The up-regulation of CYP3A1 by GR ligands, BUD and DEX, significantly induced 1,25(OH)2D3 confirmed that the slices were able to PXR and CYP3A9 mRNA in all the segments of respond to the VDR ligand. In contrast, incubation the intestine and in liver slices [30]. The PXR of liver slices in the presence of 100 nM ligands, PCN and DEX, induced CYP3A1 and a b 1,25(OH)2D3 did not affect rOST and rOST CYP3A9 in a region specific manner in rat expression (Figure 5D) but at the same time intestine and liver slices [30]. This observation induced VDR mRNA expression, as expected confirmed the intactness of the GR and PXR (unpublished observation). Rat ileal slices, when nuclear receptor pathways in the rat intestinal and co-incubated with 1,25(OH)2D3 (100 nM)and liver slices.

225 REGULATION OF OSTa-OSTb IN RAT AND HUMAN INTESTINE AND LIVER 249

Figure 4. rOSTa and rOSTb genes are induced by FXR ligands. Rat intestine slices (jejunum (A), ileum (B) and colon (C)) were treated with CDCA (50 mM) and LCA (5 mM and 10 mM) for 12 h, and liver slices (D) were treated with 100 mM of CDCA and 50 mM of LCA for 24 h. The mRNA expression of rOSTa and rOSTb genes relative to villin (intestine) and GAPDH (liver) was quantified by real-time PCR and expressed with respect to the solvent treated controls, which were set to unity. Results are expressed as Ã ÃÃ mean7SD of 4–5 rats. po0.05; po0.001

Expression and regulation of hOSTa, hOSTb in expression in each of the five human ileum slice the human ileum and liver experiments (2- to 3-fold induction), but the levels failed to reach statistical significance due The mRNA expression of hOSTa and hOSTb relative to GAPDH in the ileum was 2- to 3-fold to the larger variation existing among the human higher than that in human liver, which showed a ileum donor samples (Figures 8B and C). DEX a b low expression. The expression of hOSTa and and BUD induced hOST and hOST expression hOSTb mRNA in ileum slices was induced or in all but one of the human ileum donors (Figure remained unaltered upon incubation, with or 8D). CYP3A4 expression, a VDR, PXR and GR without the solvents for 24 h (average CT at 0 h, responsive gene was significantly induced in 26.0 for both hOSTa and hOSTb and at 24 h, 25.0 these samples by 1,25(OH)2D3, LCA, DEX and for hOSTa and 24.0 for hOSTb) (Figure 8A). BUD but not by CDCA [30]. Furthermore, CDCA CDCA induced hOSTa in ileum slices 6- to 7-fold, and LCA significantly induced SHP expression but the effect on hOSTb expression was smaller, (unpublished observations), confirming the intact- only a 2.5-fold induction was observed, and was ness of the PXR, VDR, GR and FXR pathways in consistently observed in all the human ileum the ileum slices. donors (Figure 8B). LCA, but not 1,25(OH)2D3 The incubation conditions significantly moderately induced the hOSTa and hOSTb decreased the expression (2-fold) of the hOSTa

226 250 A.A. KHAN ET AL.

a b Figure 5. The VDR ligand, 1,25(OH)2D3 decreases the expression of rat rOST and rOST genes in jejunum (A), ileum (B), colon (C) and liver (D) slices. Rat jejunum, ileum and colon slices were treated with 5 nM,10nM and 100 nM of 1,25(OH)2D3 for a b 12 h. Rat liver slices were treated with 100 nM of 1,25(OH)2D3 for 24 h. The mRNA expression of rOST and rOST genes relative to villin (intestine) and GAPDH (liver) were quantified by real-time PCR and expressed with respect to the solvent treated controls, Ã ÃÃ which were set to unity. Results are expressed as mean7SD of 4–5 rats. po0.05; po0.001

a b Figure 6. The VDR ligand, 1,25(OH)2D3, decreases the expression of rat rOST and rOST genes in ileum, also in the presence of FXR ligand, CDCA (A), 1,25(OH)2D3 did not affect the induction of SHP by CDCA in ileum slices (B). Rat ileum slices were m 1 m treated with 100 nM of 1,25(OH)2D3, 50 M of CDCA and 100 nM of 1,25(OH)2D3 50 M of CDCA for 12 h. The mRNA expression of rOSTa and rOSTb and short heterodimer protein (SHP) genes relative to villin were quantified by real-time PCR and expressed Ã with respect to solvent treated controls, which were set to unity. Results are expressed as mean7SD of 4–5 rats. po0.05; ÃÃ po0.001

227 REGULATION OF OSTa-OSTb IN RAT AND HUMAN INTESTINE AND LIVER 251

Figure 7. The GR ligands, dexamethasone (DEX) and budesonide (BUD), but not the PXR ligand, pregnane 16-a carbonitrile (PCN) induce the expression of rat rOSTa and rOSTb genes in jejunum (A), colon (C), but not in ileum (B) and liver (D) slices. Rat jejunum, ileum and colon slices were treated with 1 mM and 50 mM of DEX, 10 nM of BUD, and 10 mM of PCN for 12 h. Rat liver slices were treated with 50 mM of DEX, 10 nM and 100 nM of BUD, and 10 mM of PCN for 24 h. The mRNA expression of rOSTa and rOSTb genes relative to villin (intestine) and GAPDH (liver) were quantified by real-time PCR and expressed with respect to solvent treated controls, which were set to unity. Results are expressed as mean7SD of 4–5 rats. Ã ÃÃ po0.05; po0.001

b in human liver slices (CT at 0 h, 27.0 and at 24 h, and induced hOST expression in 3 out of 5 livers 28.0), whereas hOSTb expression was signifi- (Figure 9D). The intactness of the VDR, PXR, FXR cantly elevated (2-fold) (CT at 0 h, 32.0 and at 24 h, and GR pathways in the slices was confirmed by 31.0) (Figure 9A); these changes were not affected increased CYP3A4 [30], SHP and PXR expression by the type of solvent used. Incubation of human (unpublished observations). liver slices with CDCA strongly induced hOSTa (15-fold induction) and hOSTb (110-fold induction) expression. LCA moderately induced hOS- Discussion Ta and hOSTb expression (2.5-fold and 3.5-fold o respectively; p 0.05) (Figure 9B). 1,25(OH)2D3 In this study, precision-cut intact tissue slices of exerted only a minor decrease in hOSTa and rat intestine (jejunum, ileum and colon), and rat hOSTb expression in 3 out of 4 livers (Figure 9C). liver, and human ileum and liver were used to DEX significantly decreased hOSTa expression investigate the species, organ and region depen-

228 252 A.A. KHAN ET AL.

Figure 8. The effect of incubation time (24 h) and the solvent controls on the expression of human organic solute transporter, hOSTa and hOSTb genes in human ileum slices (A), The mRNA expression of hOST a and hOST b genes relative to villin was quantified by real-time PCR and expressed with respect to the control slices before incubation, which were set to unity. The FXR ligands, CDCA and LCA, induced (B), the VDR ligand, 1,25(OH)2D3 (C) did not induce, the GR/PXR ligand, DEX and the GR ligand, BUD and (D) induced hOSTa and hOSTb gene expression in all the experiments but failed to reach statistical significance. m m m m Human ileum slices were treated with 10 to 100 nM of 1,25(OH)2D3,50 M of CDCA, 1 M and 50 M of DEX and 1 M of BUD for 24 h. The mRNA expression of hOSTa and hOSTstb genes relative to villin were quantified by real-time PCR and expressed with respect to solvent treated controls, which were set to unity. Results are expressed as mean7SD of 4–5 human ileum donors. ÃÃpo0.001 dent regulation of the basolateral bile acid half earlier in the mouse [12], the Sprague-Dawley rat transporters, OSTa and OSTb by FXR, VDR, PXR and man [11], and their concomitant presence and GR ligands at the level of mRNA and the supports the hypothesis that they are both involved data are summarized in Table 2. As shown by in the facilitation of bile acid absorption. The ratio both qRT-PCR and immunohistochemistry (Figures of their expression in ileum relative to that in 1and2),rOSTa and rOSTb were expressed in all jejunum was higher in Wistar rats (3.8-fold for OSTa regions of the rat small intestine and colon of Wistar and 8-fold for OSTb, when normalized for villin rats, with the highest expression in ileum, where expression) (Figure 1) than the 2-fold difference most of the bile acids are actively reabsorbed [4]. reported for Sprague-Dawley rats [11]. The rOSTa Although the absolute expression of these genes and rOSTb are expressed at the basolateral surface cannot be determined by the applied qRT-PCR of the ileal enterocyte with a decreasing gradient of technique, the relative expression of these genes expression from the villus tip to the crypts in the along the length of the intestine can be assessed. Wistar rats, which is similar to the earlier reports in The expression patterns of rOSTa and rOSTb in the the mouse and the Sprague-Dawley rats [11,37]. In rat intestine paralleled that of rASBT as reported the livers of Wistar rats, rOSTa and rOSTb proteins

229 REGULATION OF OSTa-OSTb IN RAT AND HUMAN INTESTINE AND LIVER 253

Figure 9. The effect of incubation time (24 h) and the solvent controls on the expression of human organic solute transporter (hOST)a and hOSTb genes in human liver slices (A), The mRNA expression of hOSTa and hOSTb genes relative to GAPDH were quantified by real-time PCR and expressed with respect to solvent treated controls, which were set to unity. The FXR ligands, a b CDCA and LCA induced (B) and the VDR ligand, 1,25(OH)2D3 (C) did not affect the expression hOST and hOST genes. The GR/ PXR ligand, DEX (D) significantly decreased the expression of hOSTa, but induced hOSTb gene in three out of five livers. Human m m m liver slices were treated with 100 nM of 1,25(OH)2D3, 100 M of CDCA, 50 M of LCA and 50 M of DEX for 24 h. mRNA expression of hOSTa and hOSTb genes relative to GAPDH were quantified by real-time PCR and expressed with respect to solvent treated Ã ÃÃ controls, which was set to unity. Results are expressed as mean7SD of 4–5 human liver donors. po0.05; po0.001

Table 2. Summary of the effects of VDR, FXR, PXR and GR ligands on the OSTa and OSTb expression in rat and human intestine and liver; n 5 4–5 rats or 3–5 human ileum and liver donors Ligand(s) Nuclear receptor Rat Human

Intestine Liver Ileum Liver

Jejunum Ileum Colon

rOSTa rOSTb rOSTa rOSTb rOSTa rOSTb rOSTa rOSTb hOSTa hOSTb hOSTa hOSTb 1,25(OH)2D3 VDR kkkkkk2222k2 CDCA FXR/VDR mmmmmmmamm m m m LCA FXR/VDR k2kk2mmamm m m m DEX PXR/GR am am am2am amm am2amk bm PCN PXR 22222222ND ND ND ND BUD GR m am22am amm am2am ND ND induction; krepression; 2 no induction. amInduction in all experiments but with high variation between the experiments. bmInduction in three out of five experiments.ND, not done. were found to be expressed in detectable intensities Furthermore, rOSTa and rOSTb proteins were also at the basolateral membranes of the BEC (Figure 2). detected, albeit at lower intensities, at the basolateral

230 254 A.A. KHAN ET AL. membranes of the hepatocytes, as documented in and rFXR mRNA, not only in the ileum, but also earlier reports [11]. However, their functional sig- in the jejunum and colon of the rat intestine [30], nificance in hepatocytes has not been proven to date. and the reported induction of Osta and Ostb genes Previously, it was shown that both rat and by CDCA in cecum and colon of Slc10a2 / mouse human intestinal and liver slices adequately by Frankenberg et al.[16]promptedustoinvesti- reflect the regulation of drug metabolizing gate the regulation of rat rOSTa and rOSTb genes in enzymes and transporters as observed in vivo the rat jejunum and colon by the FXR ligand, [27,38], and VDR, PXR, GR and FXR pathways CDCA. Similar to what was observed for the rat remained intact [30]. The quality of total RNA ileum, the rOSTa and rOSTb genes were also isolated from fresh tissue (rat and human) and the significantly induced by CDCA in rat jejunum 0 h slices prior to incubation was similar and no and colon (Figure 4A, C). This shows that, although change was observed during the period of cold the basal expression pattern of rOSTa and rOSTb a storage and slicing in the average CT for rOST , genes varied widely along the length of the rat rOSTb, rVDR, rFXR and rPXR genes, as well as the intestine, the half transporters were responsive to signature genes of the various nuclear receptors. the FXR stimulus, albeit to a different extent in all Viability (ATP levels) and housekeeping genes regions of the intestine. These results also show that (villin and GAPDH) remained constant during the bile salts, despite being present at high concentra- incubation (data not shown). The expression of tions in both jejunum and ileum lumen in vivo,do rOSTa and rOSTb mRNA were moderately altered not play a decisive role in the basal expression of during incubation of rat intestinal and liver slices. rOSTa and rOSTb in the small intestine. In the rat In rat jejunum, the expression of rOSTb was colon, the response of the rOSTa and rOSTb significantly elevated after 12 h of incubation. In promoters to CDCA appeared to be remarkably contrast, rat liver slices showed a significant higher than in ileum, while the expression of FXR in decrease in the expression of rOSTb without colon was similar to ileum [30]. Based on these affecting rOSTa expression. However, in the rat results, it is speculated that bile acids play a role in ileum, rOSTa and rOSTb expression was signifi- the regulation of their own absorption by increasing cantly increased and in rat colon, rOSTa and their basolateral excretion in the intestine. The rOSTb expression was significantly decreased. difference in the CDCA-induced response of rOSTa These changes in rOSTa and rOSTb expression in and rOSTb between ileum and colon might be due different segments of the rat intestine and liver to a higher intracellular concentration of CDCA in during incubation of the slices suggest that the the colon which might be the result of a different expression of the rOSTa and rOSTb genes is balance between uptake, excretion and/or normally regulated in vivo by endogenous factors metabolism. which seem to be absent in the culture medium or Further, the role of the nuclear receptors, GR by endogenously generated factors whose avail- and VDR in the regulation of rOSTa and rOSTb ability is altered during incubation of the slices. genes is investigated by incubating rat jejunum, As reported earlier [15,16,19], CDCA, a high ileum, colon and liver, human ileum and liver affinity FXR ligand, was found to induce OSTa and slices with GR and VDR ligands. DEX was found OSTb genes in rat and human ileum and liver slices to significantly induce rOSTa and rOSTb expres- (Figures4B,4D,8B,9B).TheinductionofhOSTa sion in rat jejunum, colon and liver, but the and hOSTb genes in human ileum (6- to 7-fold) and moderate induction of rOSTa and rOSTb expres- liver (15- and 110-fold) slices by CDCA was much sion in rat ileum slices was found to be not stronger than that reported earlier in human ileum significant (Figures 7A–D). In addition, in human biopsies incubated for 4 h only, and in human ileum slices, both hOSTa and hOSTb gene hepatoma cell lines, Huh 7 and HepG2 [15]. These expression was induced by DEX (Figure 8D). In results show that the rat rOSTa and rOSTb genes, human liver slices, DEX significantly decreased similar to human and mouse OSTa and OSTb genes the expression of hOSTa but induced hOSTb [15,16] are responsive to CDCA in intact cells. expression (Figure 9D). These results are the first The presence of detectable amounts of rOSTa to show that rat and human OSTa and OSTb and rOSTb mRNA and protein (Figures 1 and 2) genes are regulated by DEX in intestine and liver.

231 REGULATION OF OSTa-OSTb IN RAT AND HUMAN INTESTINE AND LIVER 255

The induction of OSTa and OSTb genes by DEX in rat liver, LCA induced rOSTa expression in rat intestine and liver, and human ileum is without affecting the rOSTb expression likely to be attributed to GR and not to PXR, since (Figure 4D). This inductive effect of LCA on rOSTa BUD, a specific GR ligand, also induced the in rat liver was in contrast to that of 1,25(OH)2D3, expression of OSTa and OSTb genes. The PXR but paralleled that of CDCA suggesting that LCA ligand, PCN, failed to alter rOSTa and rOSTb in acted as a FXR ligand [39]. However, LCA showed rat intestinal slices. Furthermore GR ligands opposite results in rat jejunum and colon. In the rat induce the trans-acting factor, LRH-1 (A.A. jejunum, LCA significantly down-regulated rOSTa Khan et al, unpublished results), which is without affecting rOSTb, whereas in rat colon, LCA reported to be essential for the basal expression significantly induced rOSTb without affecting the of OSTa and OSTb [18]. However, whether the rOSTa expression. These mixed results suggest that effects of the GR ligands are indirectly mediated LCA affects the expression of rOSTa and rOSTb through induction of HNF4a and LRH-1, or genes via both VDR and FXR, giving rise to directly mediated through a potential GRE in inhibition and induction, respectively. The different the OSTa and OSTb genes needs to be ascer- effects of LCA on the rOSTa and rOSTb genes in rat tained. These results on the induction of OSTa intestine and liver are difficult to interpret but and OSTb by GR ligands further explain the suggest that the FXR-mediated effects predominate decreased loss of bile acids in the feaces and in rat colon and liver, whereas the VDR-mediated increased bile acid absorption in the patients effects predominate in jejunum and ileum. For the with Crohn’s disease treated with BUD and DEX. rat liver, this may be explained by the higher It was reported that these patients have induced expression of FXR compared with VDR in compar- ASBT expression [22]. Furthermore, induction of ison with the intestine [30]. In human liver slices, ASBT was found in human ileum slices treated LCA significantly induced the hOSTa and hOSTb with GR ligands (A.A. Khan et al., unpublished expression, similar to that of CDCA (Figure 9B), results) with subsequent induction of OSTa and however, LCA induced hOSTa and hOSTb expres- OSTb (Figure 8D). Thus, GR ligands simulta- sion in all the human ileum donors but failed to neously increase ASBT, OSTa and OSTb expres- reach statistical significance (Figure 8B), suggesting sion in human ileum slices. that the FXR regulation predominates in humans, In addition, our data also provide evidence on which is in line with the lack of VDR-mediated the involvement of the VDR in the regulation of effect by 1,25(OH)2D3 (Figure 8C). The results on a b a b rat rOST and rOST genes. 1,25(OH)2D3 exerted induction of hOST and hOST by LCA and CDCA an inhibitory effect on the expression of rOSTa in human livers are in line with those reported by and rOSTb genes in rat jejunum, ileum and colon Zollner et al. [40] in human cholestatic livers. Based slices in a dose-dependent manner (Figure 5A–C) on these results, it might be speculated that during but had no effect in liver slices (Figure 5D). In the cholestasis, bile acids might play a role in the rescue human ileum, hOSTa and hOSTb genes were not phenomenon by inducing the OSTa/OSTb trans- significantly altered by 1,25(OH)2D3 treatment, porter present in the basolateral membranes of whereas in the human liver slices, hOSTa expression human hepatocytes as was also suggested for was significantly decreased in all four livers, and MRP3 [41]. Together OSTa/OSTb and MRP3 play three out of four livers exhibited a 50% decrease a protective role by increasing the bile acid efflux in the expression of hOSTb (Figure 9C). Hence, into the blood from the hepatocytes. Furthermore, the role of VDR on the regulation of the OST thedifferenteffectsofLCAontheOSTa versus genesisdifferentforhOSTa and hOSTb genes in the OSTb expression in rat jejunum, ileum humans and appeared to differ among tissues and colon, and in human liver are noteworthy and in different species. The involvement of the because it is reported that the functional bile acid VDR was further investigated with another basolateral transporter is a heterodimer of OSTa natural VDR ligand, LCA, with affinity towards and OSTb proteins [10,42,43]. These results ne- FXR [35,39]. LCA was found to decrease cessitate further studies to investigate the effect the expression of rOSTa and rOSTb genes of LCA on the formation of the functional OSTa- significantly in rat ileum (Figure 4B). However, OSTb transporter.

232 256 A.A. KHAN ET AL.

To mimic the in vivo situation in which Acknowledgements 1,25(OH)2D3 was shown to increase the flux of bile acids into the rat ileocytes by inducing The authors thank Dr Ned Ballatori (Rochester, rASBT [21], rat ileum slices were co-incubated New York, USA) for the generous gift of the rat a b with both 1,25(OH)2D3 and CDCA. 1,25(OH)2D3 rOST and rOST antibodies, Dr Vincent B. completely abolished the CDCA-mediated Nieuwenhuijs (University Medical Center, induction of OSTa and OSTb (Figure 6A) despite Groningen) for providing the human ileum the presence of an intact FXR response, shown by tissue, and Mrs A. M. A. van Loenen-Weemaes the induced rSHP expression (Figure 6B). Alto- for her excellent technical assistance with the gether, the results led to the postulate of a immunohistochemical staining. This work was negative VDRE in the promoters of the rat rOSTa supported in part by the Canadian Institutes for and rOSTb genes, as reported earlier for the Health Research, MOP89850. parathyroid and CYP7A1 genes [44,45], that overrides the FXR-dependent positive regulation of rOSTa and rOSTb genes by CDCA. However, References indirect effects of 1,25(OH)2D3 on the expression of rOSTa and rOSTb genes cannot be ruled out. 1. Kullak-Ublick GA, Becker MB. Regulation of drug Further studies are needed to ascertain this and bile salt transporters in liver and intestine. hypothesis. Drug Metab Rev 2003; 35: 305–317. In conclusion, this study showed the induction 2. Dawson PA, Haywood J, Craddock AL, et al. of hOSTa and hOSTb genes by the FXR ligand, Targeted deletion of the ileal bile acid transporter CDCA, in intact human ileum and liver tissue, eliminates enterohepatic cycling of bile acids in mice. J Biol Chem 2003; 278: 33920–33927. and confirmed the earlier reports of human 3. Weinberg SL, Burckhardt G, Wilson FA. Taurocholate ileum biopsies and HepG2 cells [15]. Induction transport by rat intestinal basolateral membrane of rat rOSTa and rOSTb gene expression by vesicles.Evidenceforthepresenceofananionex- CDCA in rat jejunum, ileum, colon and liver change transport system. J Clin Invest 1986; 78: suggests that the rOSTa and rOSTb promoters are 44–50. 4. Shneider BL. Expression cloning of the ileal responsive to FXR ligand, observations that are sodium-dependent bile acid transporter. J Pediatr similar to the mouse and human. Furthermore, Gastroenterol Nutr 1995; 20: 233–235. the rat but not human OSTa and OSTb genes 5. Sun AQ, Ananthanarayanan M, Soroka CJ, are negatively regulated by the VDR ligand, Thevananther S, Shneider BL, Suchy FJ. Sorting 1,25(OH) D . This data suggest that the toxic of rat liver and ileal sodium-dependent bile acid 2 3 transporters in polarized epithelial cells. Am J bile salt, LCA acts as a VDR ligand on rat Physiol 275 a b 1998; : G1045–G1055. rOST and rOST genes rather than as an 6. Lazaridis KN, Tietz P, Wu T, Kip S, Dawson PA, FXR-ligand in jejunum and ileum, but acts as LaRusso NF. Alternative splicing of the rat an FXR-ligand in the rat colon and liver, and sodium/bile acid transporter changes its cellular in human ileum and liver. This study reports localization and transport properties. Proc Natl Acad Sci USA 2000; 97: 11092–11097. here, for the first time, that the rat and human a b 7. Inokuchi A, Hinoshita E, Iwamoto Y, Kohno K, OST and OST genes are not only positively Kuwano M, Uchiumi T. Enhanced expression of regulated by FXR, but also positively regulated the human multidrug resistance protein 3 by bile by GR ligands. In conclusion, apart from salt in human enterocytes. A transcriptional FXR, also VDR and GR ligands, which were control of a plausible bile acid transporter. J Biol Chem 2001; 276: 46822–46829. implicated in the regulation of ASBT expres- 8. Rius M, Hummel-Eisenbeiss J, Hofmann AF, sion in rat and human intestine and liver, Keppler D. Substrate specificity of human ABCC4 regulate the mRNA expression of the OSTa (MRP4)-mediated cotransport of bile acids and and OSTb genes, as summarized in Table 2. reduced glutathione. Am J Physiol Gastrointest However, the changes in expression of these Liver Physiol 2006; 290: G640–G649. 9. Zelcer N, van de Wetering K, de Waart R, et al. two half transporters is often not identical and Mice lacking Mrp3 (Abcc3) have normal bile salt the physiological consequences remains to be transport, but altered hepatic transport of endo- elucidated. genous glucuronides. JHepatol2006; 44: 768–775.

233 REGULATION OF OSTa-OSTb IN RAT AND HUMAN INTESTINE AND LIVER 257

10. Wang W, Seward DJ, Li L, Boyer JL, Ballatori N. 22. Jung D, Fantin AC, Scheurer U, Fried M, Kullak- Expression cloning of two genes that together Ublick GA. Human ileal bile acid transporter mediate organic solute and steroid transport in gene ASBT (SLC10A2) is transactivated by the the liver of a marine vertebrate. Proc Natl Acad Sci glucocorticoid receptor. Gut 2004; 53: 78–84. USA 2001; 98: 9431–9436. 23. Nowicki MJ, Shneider BL, Paul JM, Heubi JE. 11. Ballatori N, Christian WV, Lee JY, et al. OSTalpha- Glucocorticoids upregulate taurocholate transport OSTbeta: a major basolateral bile acid and steroid by ileal brush-border membrane. Am J Physiol transporter in human intestinal, renal, and biliary 1997; 273: G197–G203. epithelia. Hepatology 2005; 42: 1270–1279. 24. Chen F, Ma L, Dawson PA, et al. Liver receptor 12. Dawson PA, Hubbert M, Haywood J, et al. The homologue-1 mediates species- and cell line- heteromeric organic solute transporter alpha-beta, specific bile acid-dependent negative feedback Ostalpha-Ostbeta, is an ileal basolateral bile acid regulation of the apical sodium-dependent bile transporter. J Biol Chem 2005; 280: 6960–6968. 13. Rao A, Haywood J, Craddock AL, Belinsky MG, acid transporter. J Biol Chem 2003; 278: Kruh GD, Dawson PA. The organic solute 19909–19916. transporter alpha-beta, Ostalpha-Ostbeta, is 25. Neimark E, Chen F, Li X, Shneider BL. Bile acid- essential for intestinal bile acid transport and induced negative feedback regulation of the homeostasis. Proc Natl Acad Sci USA 2008; 105: human ileal bile acid transporter. Hepatology 3891–3896. 2004; 40: 149–156. 14. Ballatori N, Fang F, Christian WV, Li N, 26. Li H, Chen F, Shang Q, et al. FXR-activating Hammond CL. Ostalpha-Ostbeta is required for ligands inhibit rabbit ASBT expression via FXR- bile acid and conjugated steroid disposition in the SHP-FTF cascade. Am J Physiol Gastrointest Liver intestine, kidney, and liver. Am J Physiol Gastro- Physiol 2005; 288: G60–G66. intest Liver Physiol 2008; 295: G179–G186. 27. van de Kerkhof EG, de Graaf IA, de Jager MH, 15. Landrier JF, Eloranta JJ, Vavricka SR, Kullak- Meijer DK, Groothuis GM. Characterization of rat Ublick GA. The nuclear receptor for bile acids, small intestinal and colon precision-cut slices as FXR, transactivates human organic solute trans- an in vitro system for drug metabolism and porter-alpha and -beta genes. Am J Physiol induction studies. Drug Metab Dispos 2005; 33: Gastrointest Liver Physiol 2006; 290: G476–G485. 1613–1620. 16. Frankenberg T, Rao A, Chen F, Haywood J, 28. Rots MG, Elferink MG, Gommans WM, et al. An Shneider BL, Dawson PA. Regulation of the ex vivo human model system to evaluate specifi- mouse organic solute transporter alpha-beta, city of replicating and non-replicating gene Ostalpha-Ostbeta, by bile acids. Am J Physiol therapy agents. J Gene Med 2006; 8: 35–41. Gastrointest Liver Physiol 2006; 290: G912–G922. 29. Olinga P, Merema MT, Hof IH, et al. Effect of cold 17. Okuwaki M, Takada T, Iwayanagi Y, et al. LXR and warm ischaemia on drug metabolism in alpha transactivates mouse organic solute trans- isolated hepatocytes and slices from human and porter alpha and beta via IR-1 elements shared monkey liver. Xenobiotica 1998; 28: 349–360. with FXR. Pharm Res 2007; 24: 390–398. 30. Khan AA, Chow ECY, Van Loenen-Weemaes A, 18. Lee YK, Schmidt DR, Cummins CL, Choi M, Porte RJ, Pang KS, Groothuis GMM. Comparison Peng L, Zhang Y, et al. Liver receptor homolog-1 of effects of VDR versus PXR, FXR and GR regulates bile acid homeostasis but is not essential ligands on the regulation of CYP3A isozymes in for feedback regulation of bile acid synthesis. rat and human intestine and liver. Eur J Pharm Sci Mol Endocrinol 2008; 22: 1345–1356. 2009; 37: 115–125. 19. Boyer JL, Trauner M, Mennone A, et al. Upregula- 31. de Kanter R, Monshouwer M, Meijer DK, tion of a basolateral FXR-dependent bile acid Groothuis GM. Precision-cut organ slices as a efflux transporter OSTalpha-OSTbeta in cholesta- tool to study toxicity and metabolism of xenobio- sis in humans and rodents. Am J Physiol Gastro- tics with special reference to non-hepatic tissues. intest Liver Physiol 2006; 290: G1124–G1130. Curr Drug Metab 2002; 3: 39–59. 20. Zollner G, Wagner M, Moustafa T, et al. Coordi- 32. van de Kerkhof EG, Ungell AL, Sjoberg AK, et al. nated induction of bile acid detoxification Innovative methods to study human intestinal and alternative elimination in mice: role of FXR- drug metabolism in vitro: precision-cut slices regulated organic solute transporter-alpha/beta compared with Ussing chamber preparations. in the adaptive response to bile acids. Am J Physiol Drug Metab Dispos 2006; 34: 1893–1902. Gastrointest Liver Physiol 2006; 290: G923–G932. 33. van de Kerkhof EG, de Graaf IA, Ungell AL, 21. Chen X, Chen F, Liu S, et al. Transactivation of rat Groothuis GM. Induction of metabolism apical sodium-dependent bile acid transporter and transport in human intestine: validation of and increased bile acid transport by 1alpha, precision-cut slices as a tool to study induction of 25-dihydroxyvitamin D3 via the vitamin D drug metabolism in human intestine in vitro. Drug receptor. Mol Pharmacol 2006; 69: 1913–1923. Metab Dispos 2008; 36: 604–613.

234 258 A.A. KHAN ET AL.

34. Schmiedlin-Ren P, Benedict PE, Dobbins 3rd WO, bile acids and cholestasis: pathogenetic and thera- Ghosh M, Kolars JC, Watkins PB. Cultured adult rat peutic considerations. Mol Pharm 2006; 3: 231–251. jejunal explants as a model for studying regulation 41. Teng S, Piquette-Miller M. Hepatoprotective role of CYP3A. Biochem Pharmacol 1993; 46: 905–918. of PXR activation and MRP3 in cholic acid- 35. Makishima M, Lu TT, Xie W, et al. Vitamin D induced cholestasis. Br J Pharmacol 2007; 151: receptor as an intestinal bile acid sensor. Science 367–376. 2002; 296: 1313–1316. 42. Sun AQ, Balasubramaniyan N, Xu K, et al. 36. Matsubara T, Yoshinari K, Aoyama K, et al. Role Protein-protein interactions and membrane loca- of vitamin D receptor in the lithocholic acid- lization of the human organic solute transporter. mediated CYP3A induction in vitro and in vivo. Am J Physiol Gastrointest Liver Physiol 2007; 292: Drug Metab Dispos 2008; 36: 2058–2063. 37. Shneider BL, Dawson PA, Christie DM, Hardikar W, G1586–G1593. Wong MH, Suchy FJ. Cloning and molecular 43. Li N, Cui Z, Fang F, Lee JY, Ballatori N. characterization of the ontogeny of a rat ileal Heterodimerization, trafficking and membrane sodium-dependent bile acid transporter. J Clin Invest topology of the two proteins, Ost alpha and Ost 1995; 95: 745–754. beta, that constitute the organic solute and steroid 38. Olinga P, Elferink MG, Draaisma AL, et al. transporter. Biochem J 2007; 407: 363–372. Coordinated induction of drug transporters and 44. Sepulveda VA, Weigel NL, Falzon M. Prostate phase I and II metabolism in human liver slices. cancer cell type-specific involvement of the VDR Eur J Pharm Sci 2008; 33: 380–389. and RXR in regulation of the human PTHrP gene 39. Parks DJ, Blanchard SG, Bledsoe RK, et al. Bile via a negative VDRE. Steroids 2006; 71: 102–115. acids: natural ligands for an orphan nuclear 45. Han S, Chiang JY. Mechanism of Vitamin D receptor. Science 1999; 284: 1365–1368. receptor inhibition of cholesterol 7alpha-hydroxy- 40. Zollner G, Marschall HU, Wagner M, Trauner M. lase gene transcription in human hepatocytes. Role of nuclear receptors in the adaptive response to Drug Metab Dispos 2009; 37: 469–478.

235

APPENDIX A3

Khan AA, Chow EC, Porte RJ, Pang KS and Groothuis GM (2011) The role of lithocholic acid in the regulation of bile acid detoxification, synthesis, and transport proteins in rat and human intestine and liver slices. Toxicol In Vitro 25:80-90

236

Toxicology in Vitro 25 (2011) 80–90

Contents lists available at ScienceDirect

Toxicology in Vitro

journal homepage: www.elsevier.com/locate/toxinvit

The role of lithocholic acid in the regulation of bile acid detoxication, synthesis, and transport proteins in rat and human intestine and liver slices ⇑ Ansar A. Khan a, Edwin C.Y. Chow b, Robert J. Porte c, K. Sandy Pang b, Geny M.M. Groothuis a, a Pharmacokinetics, Toxicology and Targeting, Department of Pharmacy, University of Groningen, Ant. Deusinglaan 1, 9713 AV Groningen, The Netherlands b Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College Street, Toronto, ON M5S 3M2, Canada c Department of Hepatobiliary Surgery and Liver Transplantation, University Medical Center Groningen (UMCG), University of Groningen, Hanzeplein 1, 9700 RB Groningen, The Netherlands article info abstract

Article history: The effects of the secondary bile acid, lithocholic acid (LCA), a VDR, FXR and PXR ligand, on the regulation Received 22 March 2010 of bile acid metabolism (CYP3A isozymes), synthesis (CYP7A1), and transporter proteins (MRP3, MRP2, Accepted 26 September 2010 BSEP, NTCP) as well as nuclear receptors (FXR, PXR, LXRa, HNF1a, HNF4a and SHP) were studied in rat Available online 1 October 2010 and human precision-cut intestine and liver slices at the mRNA level. Changes due to 5 to 10 lMof LCA were compared to those of other prototype ligands for VDR, FXR, PXR and GR. LCA induced rCYP3A1 Keywords: and rCYP3A9 in the rat jejunum, ileum and colon, rCYP3A2 only in the ileum, rCYP3A9 expression in the Lithocholic acid liver, and CYP3A4 in the human ileum but not in liver. LCA induced the expression of rMRP2 in the colon VDR but not in the jejunum and ileum but did not affect rMRP3 expression along the length of the rat intes- FXR PXR tine. In human ileum slices, LCA induced hMRP3 and hMRP2 expression. In rat liver slices, LCA decreased GR rCYP7A1, rLXRa and rHNF4a expression, induced rSHP expression, but did not affect rBSEP or rNTCP Intestinal slices expression; whereas in the human liver, a small but signiﬁcant decrease was found for hHNF1a expres- Liver slices sion. These data suggests profound species differences in the effects of LCA on bile acid transport, synthe- Cytochrome P450 sis and detoxiﬁcation. An examination of the effects of prototype VDR, PXR, GR and FXR ligands showed Transporters that these pathways are all intact in precision cut slices and that LCA exerted VDR, PXR and FXR effects. Bile acid The LCA-induced altered enzymes and transporter expressions in the intestine and liver would affect the Induction disposition of drugs. Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction Because LCA is predominantly formed in the terminal part of the small intestine, ileal mucosal cells are exposed to very high Lithocholic acid (LCA) is a toxic secondary monohydroxy bile concentrations of LCA. However, most of the studies on LCA bio- acid formed by the bacterial biotransformation (7a-dehydroxyla- transformation were performed in liver microsomes of different tion) of the primary bile acid, chenodeoxycholic acid (CDCA) in species and comparable data is not available for intestine. Further- the terminal part of the small intestine (Danielsson and Gustafsson, more, the LCA-mediated regulation of the CYP isozymes involved 1981; Hirano et al., 1981). LCA is reported to be carcinogenic in in its metabolism is not completely understood in the intestine the intestine and cholestatic in the liver of animals and man (Javitt, and the liver. Some reports showed that the effects are concentra- 1966; Fisher et al., 1971; Narisawa et al., 1974), and is detoxified tion-dependent (Adachi et al., 2005) and route-dependent in vivo by cytochrome P450 (CYP) enzymes in the intestine and the liver (Owen et al., 2010). Makishima et al. (2002), Wolf (2002), and of humans (CYP3A4) and rats (CYP3A1, CYP3A2, CYP3A9, CYP2C6, Staudinger et al. (2001) independently showed that LCA and CYP2C11 and CYP2D1) to 6a- and 6b-hydroxy metabolites, respec- 3KCA can bind and transactivate the vitamin D receptor (VDR), tively (Araya and Wikvall, 1999; Deo and Bandiera, 2008), and con- the pregnane X receptor (PXR) and the farnesoid X receptor jugated by sulfotransferase (Hofmann, 2004). 3-Keto-5b-cholanic (FXR) (Adachi et al., 2005; Nehring et al., 2007). Hence, it is hypoth- acid (3KCA) was identified as the major LCA metabolite with hu- esized that LCA and its metabolites may coordinately regulate bile man recombinant CYP3A4 (Bodin et al., 2005), and found to bind acid detoxification, synthesis and transporter proteins in the intes- VDR (Adachi et al., 2005). tine and the liver possibly via the PXR,VDR, and FXR. In addition to the involvement of the PXR and glucocorticoid receptor (GR) in the regulation of CYP3A isoforms, the role of VDR in the regulation of ⇑ Corresponding author. Tel.: +31 50 363 7523; fax: +31 50 363 3247. CYP3A isozymes in the detoxification of LCA in human and rat E-mail address: [email protected] (G.M.M. Groothuis). intestinal cell lines and human fetal intestine explants is well

0887-2333/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.tiv.2010.09.011 237 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90 81 recognized (Catherine Theodoropoulos et al., 2003; Fukumori et al., plied Biosystems, Warrington, UK, Abgene Westbrug and Eurogen- 2007; Schmiedlin-Ren et al., 1997; Thummel et al., 2001). More re- tech, respectively. All primers were purchased from Sigma– cently, we also confirmed the involvement of VDR, as well as those Genosys by order on demand. All reagents and materials used were of PXR, FXR and GR in the regulation of CYP3A isoforms in preci- of the highest purity that was commercially available. sion-cut slices of rat and human intestine and liver (Chow et al., 2009b; Khan et al., 2009b; Xu et al., 2006), and in the Caco-2 cells 2.2. Animals (Fan et al., 2009). Matsubara et al. (2008) alluded that the induction of CYP3A isoforms was associated mostly with VDR and much Male Wistar (HsdCpb:WU) rats weighing about 230–250 g were less so with FXR and PXR. These nuclear receptors also play a role purchased from Harlan (Horst, The Netherlands) and were allowed in the regulation of transporters, namely, NTCP, ASBT, BSEP, MRP2, to acclimatize for 7 days before experimentation. Rats were housed 3, and 4, and Osta–Ostb in the enterohepatic cycle of bile acids and in a temperature and humidity controlled room on a 12 h light/ maintenance of the bile acid pool (Ananthanarayanan et al., 2001; dark cycle with food (Harlan Chow No. 2018, Horst, The Nether- Frankenberg et al., 2006; Inokuchi et al., 2001; Kast et al., 2002; lands) and tap water ad libitum. The experimental protocols were Landrier et al., 2006; McCarthy et al., 2005). approved by the Animal Ethical Committee of the University of The regulation of enzymes and transporters involved in bile acid Groningen. homeostasis had been studied in the liver of rodents after LCA administration (Beilke et al., 2008; Owen et al., 2010). In man, 2.3. Human liver and ileum tissue however, no direct data is available, though the role of LCA was indirectly studied in human cholestatic livers (Zollner et al., The research protocols were approved by the Medical Ethical 2007, 2006). In these in vivo studies, it is difficult to assess and Committee of the University Medical Center, Groningen, with in- control the exposure of the different tissues to LCA and it is virtu- formed consent of the patients. Pieces of human liver tissue were ally impossible to discriminate between direct effects of LCA from obtained from patients undergoing partial hepactectomy for the those of the LCA metabolites, cholestasis or other potential con- removal of carcinoma or from redundant parts of donor livers founding factors. Recently, distinctly different effects of the admin- remaining after split liver transplantation, as described previously istration route of LCA (intraperitoneal vs. oral) on the regulation of (Olinga et al., 2008). Human liver donor (n = 5) characteristics are enzymes and transporters involved in bile acid homeostasis in the as reported earlier (Khan et al., 2009b). Further, two additional hu- intestine and liver were demonstrated in mice (Owen et al., 2010), man livers were used for LCA experiments and their donor charac- but no data is available for the human and rat. Therefore we con- teristics (human livers HL6 and HL7) are given in Table 3. The ducted a systematic study to investigate the direct effects of LCA human ileum was obtained as part of the surgical waste after on the regulation of bile acid detoxification enzymes and trans- resection of the ileocolonic part of the intestine in colon carcinoma porters in the rat and human intestine and liver, especially on patients and the donor characteristics are similar to those reported CYP7A1, the bile acid synthetic enzyme, in rat and human livers, earlier (Khan et al., 2009b). and the nuclear receptors/transcription factors involved in the regulation of these proteins at the level of mRNA by exposing preci- 2.4. Preparation of slices sion-cut organ slices to different concentrations of LCA. Previously, this in vitro model was shown to be a valuable model 2.4.1. Rat and human intestinal slices to study regulation of genes of interest by ligands for several NR Rat intestinal and human ileum slices (200–300 lm thick) were in liver (Graaf et al., 2007; Jung et al., 2007; Olinga et al., 2008) prepared as published before (Khan et al., 2009a,b; van de Kerkhof and intestine (Khan et al., 2009a,b; Martignoni et al., 2006; van et al., 2005, 2006). The precision-cut slices were stored in carbo- de Kerkhof et al., 2005) under identical conditions. Further, we genated ice-cold KHB prior to the start of the experiment, which compared the LCA-mediated effects with those induced by other usually varies from 1 to 3 h from sacrificing the rat and for human nuclear receptor specific ligands: CDCA for FXR, PCN for PXR, livers 2–3 h post surgery. DEX for GR and PXR, and budesonide (BUD) for GR. 2.4.2. Rat and human liver slices Human and rat liver slices (200–300 lm thick) were prepared 2. Materials and methods as described earlier (Khan et al., 2009b; Olinga et al., 1998), from cylindrical cores of liver tissue (8 mm) using the Krumdieck tissue 2.1. Materials slicer. The slices were stored in ice-cold UW solution on ice prior to the start of the experiment, which usually varies from 1 to 3 h from Lithocholic acid and chenodeoxycholic acid were purchased sacrificing the rat and for human livers 2–3 h post surgery. from Calbiochem, San Diego, CA, dexamethasone was from Genfarma bv, Maarssen. Pregnenolone-16a carbonitrile, budesonide and the 2.5. Incubation of rat and human intestinal slices solvents: ethanol, methanol and dimethylsulfoxide were purchased from Sigma–Aldrich Chemical Co. (St. Louis, MO); gentami- Precision-cut slices from the rat intestine (jejunum, ileum and cin sulfate and Williams medium E with glutamax-I and colon) and human ileum were incubated individually in 12-well amphotericin B (fungizone)-solution were obtained from Gibco sterile tissue culture plates (Grenier bio-one GmbH, Frickenhausen, (Paisley, UK). D-Glucose, HEPES were procured from ICN Biomedi- Austria) containing 1.3 ml Williams medium E supplemented with cals, Inc. (Eschwege, Germany). University of Wisconsin organ D-Glucose to a final concentration of 25 mM, gentamicin sulfate, preservation solution (UW) was obtained from DuPont Critical 50 lg/ml amphotericin/fungizone, 250 lg/ml and saturated with Care, Waukegab, Illinois, USA. Low gelling temperature agarose carbogen at 37 °C, continuously gassed with carbogen and shaken was purchased from Sigma–Aldrich (St. Louis, MO). RNAeasy mini at 80 rpm. Rat intestinal slices were incubated with LCA (final con- columns were obtained from Qiagen, Hilden, Germany. Random centrations 5 and 10 lM), CDCA (final concentration, 50 lM), DEX primers (500 lg/ml), MgCl2 (25 mM), RT buffer (10), PCR nucleo- (final concentrations, 1 and 50 lM) and PCN (final concentration, tide mix (10 mM), AMV RT (22 U/ll) and RNasin (40 U/ll); were 10 lM) added as a 100 concentrated stock solution in methanol purchased from Promega Corporation, Madison WI, USA. SYBR (LCA and CDCA) and DMSO (DEX and PCN), and incubated for Green and Taqman Master Mixes (2) were purchased from Ap- 12 h. Human ileum slices were incubated with LCA (final 238 82 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90 concentration 10 lM) and CDCA (final concentration 50 lM) added ing to the earlier published method (Khan et al., 2009b). qRT-PCR as a 100x concentrated stock solution in methanol and incubated was performed for the rat and human genes using primer se- for 8 and 24 h. Control slices from rat intestine and human ileum quences listed in Tables 1 and 2, respectively, using SYBR Green were incubated in supplemented Williams medium E with 1% detection system as reported earlier by Khan et al. (2009b). Primer methanol or DMSO without inducers. From a single rat or human sequences used for CYP3A1, CYP3A2 and CYP3A9 analysis were as tissue sample, six (rat intestine) or three (human intestine) repli- reported earlier by Mahnke et al. (1997). All primer sets were cate slices were subjected to each experimental condition. After analyzed using BLASTn to ensure primer specificity for the gene the incubation these replicate slices were harvested, pooled and of interest (http://www.ncbi.nlm.nih.gov/BLAST/). Furthermore appro- snap-frozen in liquid nitrogen to obtain sufficient total RNA for priate controls were analyzed for all the primer sets to determine qRT-PCR analysis. Samples were stored in 80 °C freezer until dimer formation of the primer and homogeneity of the PCR prod-

RNA isolation. These experiments were replicated in 3–5 rats and ucts. The comparative threshold cycle (CT) method was used for with human ileum samples from 3 to 5 donors. relative quantification of the mRNA. CT is inversely related to the abundance of mRNA transcripts in the initial sample. The mean 2.6. Incubation of rat and human liver slices CT of the duplicate measurements was used to calculate the difference in CT for gene of interest and the reference gene (villin for D Rat and human liver slices were incubated individually in six- intestine and GAPDH for liver) ( CT), which was compared to the D DD well sterile tissue culture plates (Grenier bio-one GmbH, Fricken- CT of the corresponding solvent control ( CT). Data are ex- hausen, Austria) containing 3.2 ml Williams medium E supple- pressed as fold induction or repression of the gene of interest ðÞDDCT mented with D-Glucose to a final concentration of 25 mM, according to the formula 2 . No significant differences were gentamicin sulfate (50 lg/ml) and saturated with carbogen at observed in the expression of genes of interest between control 37 °C, continuously gassed with carbogen and shaken at 80 rpm. incubations with and without the solvents (methanol and DMSO), Rat liver slices were incubated with LCA (50 lM), DEX (50 lM), therefore all control incubation data was analyzed as one experi- BUD (10 and 100 nM) and PCN (10 lM). Control slices were incu- mental group. bated in supplemented Williams medium E with vehicle (methanol or DMSO) without inducers for 8 and 24 h. Human liver slices were 2.8. Statistics incubated with LCA and DEX (50 lM). Control slices were incubated in supplemented Williams medium E with 1% methanol or All experiments were performed in 3–5 rats and in 5–7 human DMSO without inducers for 24 h. From a single rat/single human tissue samples. Values were expressed as mean ± SEM. Data were liver donor, three replicate slices were subjected to identical incu- analyzed by the paired Student’s t-test or Mann–Whitney U-test bation conditions. At the end of the incubation, these replicate to detect the effect of the ligands. The Student’s t-test was used slices were harvested, pooled and snap-frozen in liquid nitrogen to analyze the rat data where the error distribution was found to to obtain sufficient total RNA for quantitative real time PCR be normal with equal variance, whereas the non-parametric (qRT-PCR) analysis. Samples were stored in 80 °C freezer until Mann–Whitney U-test was used for experiments where non-equal RNA isolation. These experiments were replicated in 3–5 rats and error distribution and high variance were observed (e.g. expression with liver samples from 4 to 5 human liver donors. of CYP3A1 and CYP3A2 genes in Wistar rats and all genes in human tissues). Statistical analysis was performed on fold induction as DD 2.7. RNA isolation and qRT-PCR well as on CT with similar results. A P value <0.05 was considered as significant. Total RNA from the rat and human intestine and liver samples was isolated with RNAeasy mini columns from Qiagen by following 3. Results manufacturer’s instruction. The ratio of absorbance measured at 260/280 and 260/230 using a Nanodrop ND100 spectrophotometer 3.1. Regulation of rCYP3A isozymes by LCA in rat intestine and liver (Wilmington, DE USA) was always above 1.8 for all the samples. slices The yield of mRNA was similar for the control slices, the solvent controls, and the slices exposed to the ligands. The total RNA Incubation of rat intestinal slices (jejunum, ileum and colon) (2 lg/50 ll) was reverse-transcribed into template cDNA accord- with LCA induced the expression of rCYP3A1 along the length of

Table 1 Oligonucleotides for quantitative real-time PCR, rat genes (SYBR Green analysis).

Gene Forward primer (50–30) Reverse primer (50–30) rVillin GCTCTTTGAGTGCTCCAACC GGGGTGGGTCTTGAGGTATT rGAPDH CTGTGGTCATGAGCCCCTCC CGCTGGTGCTGAGTATGTCG rCYP3A1 GGAAATTCGATGTGGAGTGC AGGTTTGCCTTTCTCTTGCC rCYP3A2 AGTAGTGACGATTCCAACATAT TCAGAGGTATCTGTGTTTCCT rCYP3A9 GGACGATTCTTGCTTACAGG ATGCTGGTGGGCTTGCCTTC rCYP7A1 CTGTCATACCACAAAGTCTTATGTCA ATGCTTCTGTGTCCAAATGCC rBSEP TGGAAAGGAATGGTGATGGG CAGAAGGCCAGTGCATAACAGA rNTCP CTCCTCTACATGATTTTCCAGCTTG CGTCGACGTTCGTTCCTTTTCTTG rMRP2 CTGGTGTGGATTCCCTTGG CAAAACCAGGAGCCATGTGC rMRP3 ACACCGAGCCAGCCATATAC TCAGCTTCACATTGCCTGTC rSHP CTATTCTGTATGCACTTCTGAGCCC GGCAGTGGCTGTGAGATGC rHNF1a CTCCTCGGTACTGCAAGAAACC TTGTCACCCCAGCTTAAGACTCT rHNF4a CCAGCCTACACCACCCTGGAGTT TTCCTCACGCTCCTCCTGAA rLXRa TGCAGGACCAGCTCCAAGTA GAATGGACGCTGCTCAAGTC rLRH-1 GCTGCCCTGCTGGACTACAC TGTAGGGCACATCCCCATTC rPXR GATGATCATGTCTGATGCCGCTG GAGGTTGGTAGTTCCAGATGCTG rFXR CCAACCTGGGTTTCTACCC CACACAGCTCATCCCCTTT

r, rat. 239 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90 83

Table 2 Oligonucleotides for quantitative real-time PCR, human genes (SYBR Green and TaqmanÒ analysis).

Gene Forward primer (50–30) Reverse primer (50–30) hVillin CAGCTAGTGAACAAGCCTGTAGAGGAGC CCACAGAAGTTTGTGCTCATAGGC hGAPDH ACCCAGAAGACTGTGGATGG TCTAGACGGCAGGTCAGGTC hCYP3A4 GCCTGGTGCTCCTCTATCTA GGCTGTTGACCATCATAAAAG hCYP7A1 GCTGTTGTCTATGGCTTATTCTT GCCCAGGTATGGAATTAATCCA hBSEP CAGTTCCCTCAACCAGAACAT TTTGATCATTTCGCTCTCGATG hNTCP CTCAAATCCAAACGGCCACAAT CACACTGCACAAGAGAATGATGATC hMRP2 CGGACAGCATCATGGCTTCT ACTCCTTCCTTGGCCAAGTTG hMRP3 GTCCGCAGAATGGACTTGAT TCACCACTTGGGGATCATTT hSHP TGAAAGGGACCATCCTCTTCA CAATGTGGGAGGCGGCT hHNF1a CAGAAAGCCGTGGTGGAGAC GACTTGACCATCTTCGCCACA hHNF4a CCTGGAATTTGAGAATGTGCAG AGGTTGGTGCCTTCTGATGG hLXRa CCCTTCAGAACCCACAGAGATC GCTCGTTCCCCAGCATTTT hPXR CCCAGCCTGCTCATAGGTTC GGGTGTGCTGAGCATTGATG hFXR AGAGATGGGAATGTTGGCTGA GCATGCTGCTTCACATTTTTTC hGAPDH Assay-on-Demand™ ID – Hs99999905_m1 Probe sequence (50FAM–30NFQ) – GCGCCTGGTCACCAGGGCTGCTTTT

h, human.

Table 3 Summary of the effects of LCA on the expression genes in human livers; n = 4–7 human liver donors; criteria for induction and repression are 1.5- and 0.7-fold, respectively.

Human livers HL1 HL2 HL3 HL4 HL5 HL6 HL7 Mean SEM P Gender Female N/A Male N/A Female Female Female Age 54 N/A 65 N/A 72 64 42 ATP pmol/lg of protein ± SD *10.4 ± 1.5 *5.7 ± 1.9 *12.1 ± 1.0 11.1 ± 0.9 *3.3 ± 1.2 *9.7 ± 1.8 ND

VDR (DCT) 11.7 16.2 14.8 16.2 NDE NDE ND Gene M CYP3A4 0.6 0.6 1.8 0.5 2.2 3.2 2.0 1.54 0.38 0.175 M CYP7A1 MM0.3 0.5 M 0.1 1.9 0.90 0.25 0.683 M SHP M 0.7 0.7 M 2.6 3.1 M 1.40 0.38 0.312 ; HNF1a MM0.3 MMND M 0.78 0.10 0.042 M HNF4a M 0.4 0.2 0.4 2.4 ND 0.4 0.80 0.34 0.523 M LXRa 0.6 0.7 0.4 ND MMND 0.77 0.19 0.166 M PXR 0.7 0.3 1.7 ND 1.5 0.4 ND 0.93 0.30 0.794 M FXR 0.6 M 0.5 ND MMND 0.86 0.16 0.336 M BSEP 0.4 M 1.7 M 0.7 0.2 M 0.84 0.19 0.413 M NTCP 0.3 0.30 M 0.1 2.3 0.1 2.2 0.86 0.36 0.712 M MRP2 0.5 M 0.5 MMM1.5 1.03 0.16 0.867

HL – human liver; ND – not done; NDE – not detectable; N/A – not available; M – no significant effect; – significant repression; ‘‘” data is taken from Khan et al. (2009b); all values are expressed as fold induction with respect to their solvent incubated controls. the intestine, showing a very high induction in the jejunum (400- 3.2. Expression and regulation of CYP3A4 in human ileum and liver fold at 10 lM LCA; P < 0.05) and ileum (550-fold at 10 lM LCA; slices P < 0.05), and a moderate induction in the colon (3.5-fold at 10 lM of LCA; P < 0.05) (Fig. 1A). LCA induced rCYP3A2 in the rat CYP3A4 expression was stable during control incubation of the ileum slices but not for the jejunum and colon (Fig. 1B). Induction human ileum slices for 8 h, but was decreased in human ileum and of rCYP3A2 in the rat ileum slices by LCA was found to be signifi- liver slices after 24 h of incubation (data not shown). Incubation of cant at 5 lM (5–40-fold; P < 0.05). At 10 lM of LCA, rCYP3A2 human ileum slices with LCA (10 lM) significantly induced induction was even higher but also highly variable, with values CYP3A4 expression (9- and 5-fold induction during 8 and 24 h, ranging from 6- to 124-fold and failing to reach statistical signifi- respectively; P < 0.05) (Fig. 3). Incubation of human liver slices cance. Furthermore, LCA induced rCYP3A9 modestly in jejunum, with LCA (50 lM) induced CYP3A4 mRNA expression in four of se- ileum and colon slices (2–3-fold; P < 0.001) with a slightly higher ven livers, and slightly reduced CYP3A4 mRNA in the other three effect in the colon slices (Fig. 1C). livers (Fig. 3 and Table 3). In contrast to the intestinal slices, the expression of rCYP3A1, rCYP3A2 and rCYP3A9 mRNA in the rat liver slices was not affected 3.3. Expression and regulation of rMRP2 and rMRP3 in rat intestine by LCA during 8 h of incubation. However, when liver slices were exposed to LCA (50 lM) for 24 h, rCYP3A9 expression was induced In rat intestine, rMRP2 and rMRP3 transporters are expressed in by 2-fold, whereas rCYP3A1 and rCYP3A2 expression remained reciprocal gradients along the length of the intestine, with rMRP2 unaltered (Fig. 2). As expected, the PXR ligands dexamethasone mRNA expression showing a decreasing gradient from the jejunum (DEX) and pregnenolone-16a carbonitrile (PCN) significantly towards the colon, and rMRP3 expression showed an increasing induced the expression of rCYP3A1 (100-fold) and rCYP3A9 (5–8- gradient from the jejunum to the colon (Fig. 4), as reported earlier fold; P < 0.001), but not rCYP3A2 (Fig. 2). Budesonide (BUD), a (Chow et al., 2009b). During control incubations of rat jejunum, synthetic GR ligand significantly induced rCYP3A9 expression (2– ileum and colon slices in Williams medium E without ligands, 3-fold; P < 0.001), decreased rCYP3A2 expression, and did not alter rMRP2 expression was significantly increased in jejunum and rCYP3A1 expression (Fig. 2). These results with PCN, DEX and BUD ileum, but decreased in colon slices with incubation time (data show that the PXR and GR mediated pathways are intact in rat liver not shown). The expression of rMRP3 was, however, significantly slices. increased in jejunum, ileum as well as in colon slices during control 240 84 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90

Fig. 1. Rat jejunum, ileum and colon tissue-slices were exposed to LCA (5 and 10 lM) for 12 h after which total RNA was isolated and mRNA expression of rCYP3A1 (A), rCYP3A2 (B) and rCYP3A9 (C) were evaluated by qRT-PCR. After normalizing for villin expression, the results were compared to that of 12 h incubated control slices of the same segment. Results showed mean ± SEM of 3–5 rats; in each experiment, and six slices were incubated per condition. Significant differences towards the control incubations are indicated with *P < 0.05, and **P < 0.001. ‘‘ ”denotes induction of rCYP3A1 and rCYP3A2 in all experiments, but failed to reach statistical significance due to high variation between the experiments, ND – not detectable. incubation (data not shown). LCA (10 lM) induced the expression hMRP3 in human ileum slices during 8 h of incubation but upon of rMRP2 compared to the control incubations in colon slices but prolonged (24 h) exposure of human ileum slices to CDCA, hMRP3 not in jejunum and ileum slices (Fig. 5A) nor rMRP3 expression but not hMRP2 expression was induced (Fig. 6A and B). DEX in- along the length of the intestine (Fig. 5B). In contrast, CDCA signif- duced hMRP2 expression in all the tested human ileum samples icantly decreased the expression of rMRP2 in jejunum and ileum without affecting hMRP3 expression after 24 h incubation slices but induced rMRP2 in colon slices (Fig. 5A and B), whereas (Fig. 6A and B). CDCA increased the rMRP3 expression only in ileum slices but did not affect the rMRP3 expression in jejunum or colon (Fig. 5A 3.5. Expression and regulation of the bile acid synthesis enzyme, and B). DEX significantly induced the rMRP2 expression in the jeju- transporters and nuclear receptors in rat liver slices num and colon slices and decreased the expression of rMRP3 in jejunum and ileum but not in colon slices (Fig. 5A and B). Similar The expression of rCYP7A1 mRNA in rat liver slices was highly to DEX, PCN significantly induced the rMRP2 expression in the sensitive to incubation. rCYP7A1 mRNA expression was decreased jejunum and colon slices (Fig. 5A) but did not affect the rMRP3 by 90% during 8 h of incubation, and upon 24 h incubation, rCY- expression along the length of the rat intestine (Fig. 5B). P7A1 mRNA was barely detectable (data not shown). Incubation of rat liver slices with LCA (50 lM) for 8 h significantly decreased 3.4. Expression and regulation of hMRP2 and hMRP3 in human ileum the rCYP7A1 expression when compared to control incubated slices slices (Fig. 7A). Furthermore, LCA induced rSHP and decreased rHNF1a, rLXRa and rLRH-1 expression without affecting the In the human ileum slices, the hMRP2 expression was signifi- rHNF4a expression after 8 h of incubation (Fig. 8A and B). Pro- cantly increased during control incubation for 8 h and returned longed exposure of rat liver slices to LCA for 24 h significantly de- to control values at 24 h, whereas hMRP3 was increased beyond creased rHNF4a expression (Fig. 8B). LCA decreased the rPXR and 24 h. LCA (10 lM) induced hMRP3 mRNA expression by 4-fold rFXR mRNA expression (Fig. 8A and B). DEX but not PCN signifi- after 8 h of incubation and hMRP2 expression by 4-fold after cantly decreased the rCYP7A1 expression with concomitant induc- 24 h of incubation as compared to the solvent treated controls tion of rSHP (Figs. 7A and 8A). Furthermore, DEX but not PCN (Fig. 6A and B). CDCA did not affect the expression of hMRP2 and induced the rPXR expression in liver slices upon 8 h of incubation 241 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90 85

Fig. 2. Rat liver slices were exposed to LCA (10–50 lM) for 8 and 24 h; and with DEX (50 lM), PCN (10 lM) and BUD (10 and 100 nM) for 24 h, after which total RNA was isolated and mRNA expression of rCYP3A1, rCYP3A2 and rCYP3A9 was evaluated by qRT-PCR. After normalizing for rGAPDH expression, the results were expressed as fold- induction and compared with the 8 and 24 h incubated control slices. Results showed mean ± SEM of 3–5 rats; three slices were incubated per condition in each experiment. Signiﬁcant differences towards the control incubations are indicated with *P < 0.05, **P < 0.001 and ***P < 0.0001. ‘‘ ”denotes induction of rCYP3A9 in all experiments, but failed to reach statistical signiﬁcance due to high variation between the experiments.

20 rMRP2 rMRP3 * 15

10 4 ** 3 2 1 mRNA expression of rMRP2 expression mRNA and rMRP3 / villin ** ** relative to jejunum in rat intestine relative 0 Jejunum Ileum Colon

Fig. 4. mRNA expression of rMRP2 and rMRP3 transporters relative to villin Fig. 3. Slices from human ileum and liver were exposed to 10 and 50 lM of LCA for expression in intestinal tissue (jejunum, ileum and colon) of the Wistar rat. The 8 and 24 h, respectively, after which total RNA was isolated and mRNA expression mRNA expression of rMRP2 and rMRP3 transporters relative to villin in ileum and of hCYP3A4 was evaluated by qRT-PCR. Results were expressed as fold-induction colon was expressed relative to that in the jejunum, which was set to unity. Each after normalizing with villin for ileum and hGAPDH for liver, and compared with bar represents the results of three animals ± SEM. Significant differences between the control incubated slices for the same length of time, which was set to unity. ileum and colon compared to jejunum are indicated with *P < 0.05 and **P < 0.001. Results showed mean ± SEM of four human ileum and seven liver donors; in each experiment three ileum and liver slices were incubated per condition. Significant differences towards the control incubations are indicated with *P < 0.05. ‘‘à”denotes induction of hCYP3A4 in four of seven experiments. rMRP3 expression was increased (data not shown). LCA decreased the expression of rMRP3 without affecting the expression of rMRP2 in liver slices upon 8 h incubation, but induced rMRP2 expression (Fig. 8A), whereas, both DEX and PCN significantly decreased the upon 24 h incubation (Fig. 7A and B). DEX and PCN induced the expression of rLXRa, rPXR, rFXR, rHNF1a, rHNF4a and rLRH-1 rMRP2 but not the rMRP3 expression (Fig. 7A and B). upon 24 h of incubation (Fig. 8B). In the rat liver slices, the expression of rBSEP and rNTCP mRNA was decreased during control incubation (data not shown). LCA 3.6. Expression and regulation of the bile acid synthesis enzyme, (50 lM) did not affect rBSEP and rNTCP expression (Fig. 7A and transporters and nuclear receptors in human liver slices B), whereas DEX but not PCN induced rNTCP and rBSEP expression (Fig. 7A and B). Furthermore, the induction of rBSEP by DEX was In the human liver slices, the expression of most genes includ- completely abolished upon 24 h incubation (Fig. 7B). During incu- ing hMRP2 and hMRP3 and hCYP7A1 was constant during control bation of rat liver slices, rMRP2 expression was decreased and incubations for 24 h (Table 3); FXR effects are usually associated 242 86 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90

Fig. 5. Rat jejunum, ileum and colon tissue-slices were exposed to LCA (5 and10 lM), CDCA (50 lM), DEX (1 and 50 lM) and PCN (10 lM) for 12 h after which total RNA was isolated and mRNA expression of rMRP2 (A) and rMRP3 (B) were evaluated by qRT-PCR. Results were expressed as fold-induction after Fig. 6. Human ileum slices were exposed to LCA (10 lM), CDCA (50 lM) and DEX (1 normalizing with villin expression and compared to the control incubated slices and 50 lM) for 8 and 24 h after which total RNA was isolated and mRNA expression of the same segment for 12 h, which was set to 1. Results showed mean ± SEM of 3– of hMRP2 and hMRP3 (A and B) were evaluated by qRT-PCR. Results were expressed 5 rats; in each experiment, six slices were incubated per condition. Significant as fold-induction after normalizing with villin expression and compared to the differences compared to the control incubations are indicated with *P < 0.05. ‘‘ control incubated slices for 8 and 24 h, which was set to 1. Results showed ”denotes induction of rMRP2 in all experiments, but failed to reach statistical mean ± SEM of 4–5 human ileum donors. In each experiment, three ileum slices significance due to large variation between the experiments. were incubated per condition. Significant differences compared to the control incubations are indicated with *P < 0.05. ‘‘ ”denotes induction of hMRP2 in all experiments, but failed to reach statistical significance due to large variation with decreased hCYP7A1 expression via elevated SHP expression. between the experiments. The effects of LCA (50 lM) on the expression of bile acid synthesis enzyme and transporters, and the transcription factors regulating consistent effect on hCYP7A1 (decrease in three of seven livers), their expression in human liver slices were quite variable among hSHP expression (induction in only two of seven livers), hHNF1a the individual livers and therefore the data is given for each indi- expression (reduced in three of five livers), hLXRa (decreased in vidual liver in Table 3 as fold induction with respect to the solvent three of five livers) and hHNF4a (decreased in four of six livers) incubated controls. This high variation does not seem to be caused (Table 3). DEX induced hHNF4a expression in three of four livers, by differences in viability of the livers (as judged by ATP concentra- but hLXRa expression was not affected by DEX (Table 4). The effect tion and morphology), nor does it seem to be related to the type of of LCA on the expression of hBSEP, hNTCP, hPXR, hFXR, hMRP2 and donor (transplantation or partial hepatectomy), or to the expres- hMRP3 was neither significant nor consistent (Table 3). DEX signif- sion level of hVDR, hFXR or hPXR. The expression levels of these icantly induced hBSEP and moderately induced hPXR expression in NRs varied up to 30-fold between the livers for hFXR and up to all the livers, and induced hNTCP in four of five livers, but had no 16-fold for hPXR (results not shown), whereas hVDR was low but effect on hMRP2 and hMRP3 expression (Table 4). detectable (CT 33–39) in four livers and undetectable (CT < 40) in two livers. In each individual liver, a fold induction of >1.5 is considered as up regulation and a fold induction of <0.7 is considered 4. Discussion as down regulation. Furthermore, the expression of a gene is considered as induced or down regulated, if it is induced or repressed In this report, rat and human precision-cut intestinal and liver in 50% of the human livers, and the others being non-responsive. slices were used to characterize the role of LCA in the regulation In contrast to the findings in rat liver slices, incubation of hu- of genes involved in bile acid detoxification, transport and synthe- man liver slices with LCA did not consistently increase CYP3A4 sis. Our data on the effects of 1,25(OH)2D3 (Khan et al., 2009a,b), expression (higher in four and lower in three livers) (Table 3), BUD, CDCA, PCN and DEX show that VDR, FXR, PXR and GR path- whereas DEX showed a strong upregulation of CYP3A4 expression ways are intact in the tissue slices. The observed changes in the in all the human livers (Table 4). In addition LCA did not have a expression of the CYP P450 isozymes and transporters during 243 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90 87

Fig. 7. Slices from rat liver were exposed to LCA (50 lM), DEX (50 lM) and PCN (10 lM) for 8 and 24 h, after which total RNA was isolated and mRNA expression of l l rCYP7A1, rNTCP, rBSEP, rMRP2 and rMRP3 were evaluated for (A) 8 h and (B) 24 h Fig. 8. Slices from rat liver were exposed to LCA (50 M), DEX (50 M) and PCN l incubations by qRT-PCR. Results were expressed as fold-induction after normalizing (10 M) for 8 and 24 h, after which total RNA was isolated and mRNA expression of a a a with rGAPDH and compared with the control slices that were incubated for 8 and rFXR, rLXR , rPXR, rSHP, rHNF1 , rHNF4 , and rLRH-1 was evaluated for 8 h (A) and 24 h, which was set to unity. Results showed mean ± SEM of 3–5 rats; three slices 24 h (B) incubations by qRT-PCR. Results are expressed as fold-induction after were incubated per condition in each experiment. Signiﬁcant differences compared normalizing with rGAPDH and compared with the control incubated slices for 8 and to the control incubations are indicated with *P < 0.05 and **P < 0.001. ‘‘#” indicates 24 h, which was set to unity. Results showed mean ± SEM of 3–5 rats; three slices rCYP7A1 is not detectable in samples incubated for 24 h. Note: the data for rPXR were incubated per condition in each experiment. Signiﬁcant differences compared induction in livers slices for 8 h is used from our recent publication (Khan et al., to the control incubations are indicated with *P < 0.05 and **P < 0.001. 2009b).

In human ileum slices, LCA was observed to induce CYP3A4 control incubations indicate that apparently the basal expression is expression (Fig. 3). However, the nuclear receptors (NRs) involved normally maintained by ligands that are absent in the culture med- in the induction of CYP3A4 by LCA from our studies is inconclusive ium. The pattern of LCA-mediated induction of rCYP3A1 and rCY- since CYP3A4 is induced also by PXR, GR and VDR ligands P3A2, and CYP3A4 in the rat and the human intestine, (Fukumori et al., 2007; Khan et al., 2009b). The involvement of respectively (Figs. 1A–C and 3) resembles that of the VDR ligand, VDR rather than PXR as the nuclear receptor responsible for LCA-

1,25(OH)2D3, and was clearly different from those of FXR, PXR or mediated CYP3A4 induction was suggested by Matsubara et al. GR ligands as reported earlier by us (Khan et al., 2009b) and others (2008), who showed that the regulation of human CYP3A4 by (Fukumori et al., 2007), conﬁrming the role of VDR in the regula- LCA in HepG2 cells is speciﬁcally mediated by VDR and not by tion of CYP3A isozymes by LCA in the rat and human intestine. PXR upon addition of siRNA to VDR and PXR. Moreover, Adachi However, the higher induction of rCYP3A1 mRNA by LCA in the et al. (2005) and Ishizawa et al. (2008) showed that LCA-mediated rat ileum compared to the colon cannot be explained by differ- effects were mostly associated with VDR and not PXR, and was ences in VDR expression, as VDR was shown to be higher in colon modest with FXR. (Khan et al., 2009b), thus other factors such as the presence of acti- In the rat liver slices, LCA did not affect rCYP3A1 and rCYP3A2 vators/repressors may also play a role. The induction of rCYP3A9 mRNA expression (Fig. 2), results congruent with those observed by LCA in the rat intestine is likely mediated via PXR, as the en- previously with the VDR ligand, 1,25(OH)2D3 (Khan et al., 2009b). zyme was induced by other PXR ligands, PCN and DEX, but not The absence of a VDR-mediated induction of rCYP3A isozymes in by 1,25(OH)2D3, the VDR ligand, or the CDCA, FXR ligand (Khan liver slices can be attributed to the low levels of VDR in rat liver et al., 2009b). The high induction of rCYP3A9 in the colon is consis- and explained by the localization of VDR in the bile duct epithelial tent with the higher expression of PXR for induction (Khan et al., cells and not hepatocytes, as reported earlier by Gascon-Barre et al. 2009b). (2003), and conformed to results in our studies (Khan et al., 2010), 244 88 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90

Table 4 Summary of the effects of dexamethasone on the expression genes in human livers; n = 4–5 human liver donors; criteria for induction and repression are 1.5- and 0.7-fold, respectively.

Human livers HL1 HL2 HL3 HL4 HL5 Mean SEM P Gender Female N/A Male Female Female Age 54 N/A 65 72 64 ATP pmol/lg of protein ± SD *10.4 ± 1.5 *5.7 ± 1.9 *12.1 ± 1.0 *3.3 ± 1.2 *9.7 ± 1.8

VDR (DCT) 11.7 16.19 14.8 NDE NDE Gene DEX " CYP3A4 17.6 4.0 15.0 1.7 9.1 9.52 3.05 0.023 " HNF1a MM1.6 1.7 ND 1.35 0.20 0.079 " HNF4a 1.9 M 6.5 3.9 ND 3.31 1.24 0.071 " PXR MM2.3 M 1.6 1.5 0.21 0.043 " BSEP 8.7 2.3 15.2 3.3 3.6 6.61 2.40 0.048 " NTCP 6.5 1.5 4.2 2.3 M 3.04 1.03 0.083 M MRP2 M 1.9 0.6 MM1.2 0.22 0.387 M MRP3 0.5 M 2.0 ND 0.5 0.95 0.35 0.861

HL – human liver; ND – not done; NDE – not detectable; N/A – not available; M – no effect; – signiﬁcant induction or induction in P50% of the livers; ‘‘” data is taken from Khan et al. (2009b); all values are expressed as fold induction with respect to their solvent incubated controls.

since the CYP3A expression is primarily in hepatocytes. However, lon slices (Fig. 5A and B) and decreased rOSTa–rOSTb in rat the LCA-induced rCYP3A9 expression upon 24 h of incubation ileum with LCA treatment (Khan et al., 2009a) tend to promote was similar to those from the PXR ligands, PCN and DEX (Fig. 2), the luminal excretion of bile acids. Although the VDR ligand, suggesting a PXR response prevailing at longer times of incubation. 1,25(OH)2D3 also induced rMRP2 without affecting rMRP3 expres- This delayed response in the induction of rCYP3A9 by LCA suggest sion (Khan et al., unpublished observation), we cannot conclude that this effect might not be directly mediated by LCA, but by its whether the effect of LCA on rMRP2 is mediated by VDR, PXR or metabolite, 3KCA, since it had been shown that LCA itself is a poor FXR, as DEX, PCN and CDCA also induced rMRP2 expression in PXR ligand and needs to be activated to 3KCA for PXR binding the rat colon (Fig. 5A). In the mouse intestine, mMrp2 was reduced (Makishima et al., 2002; Staudinger et al., 2001), a notion also sug- and mMRP3 was not significantly changed by LCA (Owen et al., gested to exist for the mouse by Owen et al. (2010). 2010), whereas in colon, mMrp3 was induced by 1,25(OH)2D3 In contrast to observations in the rat liver but consistent with (McCarthy et al., 2005), suggesting species differences in the regu- those in human ileum, LCA induced the CYP3A4 expression in hu- lation of MRP3. In the rat, LCA favors its own detoxication by man liver slices of four of seven liver donors (Fig. 3). This induction, inducing rMRP2 but not rMRP3 expression, thereby promoting api- in principle, could be explained by the observed abundance in cal efflux into the lumen of the colon (Fig. 5A and B). In contrast, expression of VDR and in human vs. rat hepatocytes (Gascon-Barre CDCA, the primary bile acid, stimulates absorption of bile salts by et al., 2003). But the induction of CYP3A4 did not correlate with the induction of rMRP3 and OSTa–OSTb expression and repression of VDR expression in these livers (Table 3), and it is unlikely that the rMRP2 expression in rat jejunum and ileum (Fig. 5A and B), favour- LCA mediated the induction of CYP3A4 in these four livers via the ing the reclamation of bile acids in the small intestine. Also in the VDR. Other nuclear receptors such as the PXR with its ligands, DEX human intestine, LCA increases the luminal excretion of BAs by (Table 4) and rifampicin (Olinga et al., 2008), would explain this inducing hMRP2 expression, whereas CDCA favors the basolateral induction data. Another aspect that contribute to the difficulty in transport of BAs by inducing hMRP3 expression in ileum slices data interpretation is the variability in the effects observed with (Fig. 6A and B), findings which are consistent with the earlier re- LCA, since the interindividual variation in enzyme activity involved ports (Inokuchi et al., 2001). Since CDCA did not affect hMRP2 in the metabolism of LCA by the sulfotransferases (Hofmann, 2004) expression in human ileum slices, the LCA effects are unlikely to and CYP3A4 (Lamba et al., 2002a,b) could result in variations in the be mediated by FXR. effective exposure to LCA or its metabolite, 3KCA. In addition to the evaluation of the effects of LCA on enzymes When we studied the effect of LCA on the regulation of the and transporters in rat and human intestine and colon, the effect basolateral half transporters, OSTa–OSTb (Ballatori et al., 2008; of LCA on rat and human CYP7A1, the rate limiting enzyme in bile Dawson et al., 2005) in intestinal slices concomitantly, we were acid synthesis and NTCP, BSEP, MRP2 and MRP3, the bile acid able to show decreased expression of rOSTa and rOSTb in the rat transporters, and the NR/transcription factors involved in the reg- ileum, and induced rOSTa and rOSTb in the rat colon and liver ulation of these proteins showed that in rat liver slices, LCA de- and human ileum and liver (Khan et al., 2009a). Decreased expres- creased the rCYP7A1 expression as observed with 1a,25- sion of rOSTa and rOSTb was also found in the intestine of LCA fed dihydroxyvitamin D3 treatment in the rat (Chow et al., 2009a) mice (Owen et al., 2010). Since rMRP2 and rMRP3, transporters and LCA treatment in the mouse (Owen et al., 2010) in vivo. Con- important for the apical excretion of monovalent and conjugated comitantly, rSHP was induced in the rat liver, an expected response bile acids across the apical and basolateral membranes, respec- for an FXR ligand, and further observed with CDCA (Khan et al., tively, of enterocytes (Brady et al., 2002; Cherrington et al., 2002; unpublished observations). Furthermore, LCA affected the rSHP- Hirohashi et al., 1998, 2000; Zelcer et al., 2006), the ascending independent pathways of rCYP7A1 regulation by decreasing the expression of rMRP3 and descending abundance of rMRP2 along expression of rHNF1a, rHNF4a, rLXRa and rLRH-1 (Fig. 8A and the length of the rat intestine (Fig. 4), together with the basolateral B), factors that are essential for the expression of rCYP7A1 (Abra- localization of rOSTa–OSTb in the ileum (Ballatori et al., 2005) sug- hamsson et al., 2005; del Castillo-Olivares and Gil, 2000a,b; Geier gest that bile acid transport in the small intestine favors ileal et al., 2008; Lehmann et al., 1997). However, LCA effects are unli- absorption, not withstanding the ileal abundance of rASBT (Chen kely to be related to PXR since PCN, the PXR ligand, failed to alter et al., 2006), whereas in the colon, net transport is more towards the expression of rCYP7A1 and rSHP. The effects of LCA on rCYP7A1 the lumen. The finding of unchanged rMRP3 expression in rat jeju- seem to decrease with increasing incubation time, suggesting that num, ileum and colon slices but induced rMRP2 expression in co- metabolism of LCA furnishes less efficient FXR agonists. The LCA- 245 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90 89 induced FXR effects on the repression of bile acid synthesis may importance of HNF-4alpha for regulation of CYP7A1. Biochem. Biophys. Res. take precedence over the PXR effects such as rCYP3A9 induction. Commun. 330, 395–399. Adachi, R., Honma, Y., Masuno, H., Kawana, K., Shimomura, I., Yamada, S., Furthermore, we found that LCA, unlike CDCA, decreased the Makishima, M., 2005. Selective activation of vitamin D receptor by lithocholic expression of rFXR and rPXR in rat liver slices (Fig. 8A) (Khan acid acetate, a bile acid derivative. J. Lipid Res. 46, 46–57. et al., unpublished observations), an observation reminiscent of Ananthanarayanan, M., Balasubramanian, N., Makishima, M., Mangelsdorf, D.J., Suchy, F.J., 2001. Human bile salt export pump promoter is transactivated by the reported VDR antagonism on the FXR (Honjo et al., 2006). Thus, the farnesoid X receptor/bile acid receptor. J. Biol. Chem. 276, 28857–28865. in addition to decreasing the expression of rCYP7A1, LCA also de- Araya, Z., Wikvall, K., 1999. 6Alpha-hydroxylation of taurochenodeoxycholic acid creased the rFXR expression, probably by a feedback loop mediated and lithocholic acid by CYP3A4 in human liver microsomes. Biochim. Biophys. a Acta 1438, 47–54. by the decreased expression of rHNF1 ,(Fig. 8A and B), which is Ballatori, N., Christian, W.V., Lee, J.Y., Dawson, P.A., Soroka, C.J., Boyer, J.L., required for its basal expression (Lou et al., 2007). The LCA-depen- Madejczyk, M.S., Li, N., 2005. OSTalpha–OSTbeta: a major basolateral bile acid dent rSHP induction and concomitant repression of rPXR is consis- and steroid transporter in human intestinal, renal, and biliary epithelia. Hepatology 42, 1270–1279. tent with results found in human hepatocytes and in mice on the Ballatori, N., Fang, F., Christian, W.V., Li, N., Hammond, C.L., 2008. Ost{alpha}– SHP-mediated repression of PXR (Ourlin et al., 2003); the observa- Ost{beta} is required for bile acid and conjugated steroid disposition in the tion is in contrast to the increased expression of both SHP and PXR intestine, kidney, and liver. Am. J. Physiol. Gastrointest. Liver Physiol. 295, found in mice fed with CDCA and GW4064 (Jung et al., 2006). G179–G186. Beilke, L.D., Besselsen, D.G., Cheng, Q., Kulkarni, S., Slitt, A.L., Cherrington, N.J., 2008. In human liver slices, effects of LCA on the NRs and CYP3A4 Minimal role of hepatic transporters in the hepatoprotection against LCA- were highly variable. The decrease in hHNF1a, hHNF4a and hLXRa induced intrahepatic cholestasis. Toxicol. Sci. 102, 196–204. was observed only in three of five of the tested human livers Bodin, K., Lindbom, U., Diczfalusy, U., 2005. Novel pathways of bile acid metabolism involving CYP3A4. Biochim. Biophys. Acta 1687, 84–93. (Table 3 and Fig. 8A and B). Unlike in rat liver slices, LCA did not Brady, J.M., Cherrington, N.J., Hartley, D.P., Buist, S.C., Li, N., Klaassen, C.D., 2002. affect the hMRP2 expression in human livers (Fig. 7A and B, Tissue distribution and chemical induction of multiple drug resistance genes in Table 3). In addition, the LCA-mediated effects on the expression rats. Drug Metab. Dispos. 30, 838–844. Catherine Theodoropoulos, C.D., Delvin, Edgard, Menard, Daniel, Gascon-Barre, of hSHP, hBSEP and hNTCP in human livers were not consistent, Marielle, 2003. Calcitriol regulates the expression of the genes encoding the whereas the CDCA effects were consistent with an intact FXR path- three key vitamin D3 hydroxylases and the drug – metabolizing enzymes way (Khan et al., unpublished observation). Hence, it is concluded CYP3A4 in the human fetal intestine. Clin. Endocrinol. 58, 1–3. Chen, X., Chen, F., Liu, S., Glaeser, H., Dawson, P.A., Hofmann, A.F., Kim, R.B., that LCA does not act as an FXR ligand in the human liver, as also Shneider, B.L., Pang, K.S., 2006. Transactivation of rat apical sodium-dependent reported by others (Ananthanarayanan et al., 2001; Jung et al., bile acid transporter and increased bile acid transport by 1alpha,25- 2007). These results suggest that the cholestatic effects of LCA dihydroxyvitamin D3 via the vitamin D receptor. Mol. Pharmacol. 69, 1913– 1923. are partly mediated by direct effect of LCA itself on the regulation Cherrington, N.J., Hartley, D.P., Li, N., Johnson, D.R., Klaassen, C.D., 2002. Organ of hBSEP and not hMRP2. Furthermore, DEX induced both hNTCP distribution of multidrug resistance proteins 1, 2, and 3 (Mrp1, 2, and 3) mRNA (Table 4), and rNTCP (Fig. 8A and B), likely via the GR as predicted and hepatic induction of Mrp3 by constitutive androstane receptor activators in by an earlier report (Eloranta et al., 2006). Hence, the GR pathway rats. J. Pharmacol. Exp. Ther. 300, 97–104. Chow, E.C., Maeng, H.J., Liu, S., Khan, A.A., Groothuis, G.M., Pang, K.S., 2009a. is intact in human liver slices. DEX also induced both hBSEP and 1Alpha,25-dihydroxyvitamin D(3) triggered vitamin D receptor and farnesoid X rBSEP (Table 4 and Fig. 8A), an observation that is reported for receptor-like effects in rat intestine and liver in vivo. Biopharm. Drug Dispos. the first time. 30, 457–475. Chow, E.C., Sun, H., Khan, A.A., Groothuis, G.M., Pang, K.S., 2009b. Effects of In conclusion, LCA plays an important role in the feed forward 1alpha,25-dihydroxyvitamin D3 on transporters and enzymes of the rat regulation of its detoxification pathways in the rat and human intestine and kidney in vivo. Biopharm. Drug Dispos. 31, 91–108. intestine by inducing CYP3A isozymes, thereby increasing its Danielsson, H., Gustafsson, J., 1981. Biochemistry of bile acids in health and disease. Pathobiol. Annu. 11, 259–298. metabolism, as was also found in the mouse in vivo (Owen et al., Dawson, P.A., Hubbert, M., Haywood, J., Craddock, A.L., Zerangue, N., Christian, W.V., 2010). In addition, LCA increases the luminal efflux of conjugated Ballatori, N., 2005. The heteromeric organic solute transporter alpha–beta, (toxic) bile acids via rMRP2 into the colon, while simultaneously Ostalpha–Ostbeta, is an ileal basolateral bile acid transporter. J. Biol. Chem. 280, 6960–6968. preserving the primary bile acid pool by inducing the expression del Castillo-Olivares, A., Gil, G., 2000a. Alpha 1-fetoprotein transcription factor is of rOSTb in the colon (Khan et al., 2009a) and rMRP3 in the ileum. required for the expression of sterol 12alpha-hydroxylase, the specific enzyme Distinct species differences were observed for the effects of LCA in for cholic acid synthesis. Potential role in the bile acid-mediated regulation of gene transcription. J. Biol. Chem. 275, 17793–17799. the rat and human liver. In the rat liver, LCA decreases bile acid del Castillo-Olivares, A., Gil, G., 2000b. Role of FXR and FTF in bile acid-mediated synthesis and excretion but its effects in the human liver were suppression of cholesterol 7alpha-hydroxylase transcription. Nucleic Acids Res. inconsistent and need further investigation. Thus, LCA was found 28, 3587–3593. to be a promiscuous ligand that can interact with FXR, VDR and Deo, A.K., Bandiera, S.M., 2008. Biotransformation of lithocholic acid by rat hepatic microsomes: metabolite analysis by liquid chromatography/mass PXR in the regulation of bile acid synthesis, metabolism and trans- spectrometry. Drug Metab. Dispos. 36, 442–451. port in the rat intestine and liver and the human intestine. This al- Eloranta, J.J., Jung, D., Kullak-Ublick, G.A., 2006. The human Na+-taurocholate tered expression of transporters and CYP3A enzymes may also cotransporting polypeptide gene is activated by glucocorticoid receptor and peroxisome proliferator-activated receptor-gamma coactivator-1alpha, and have consequences for the disposition of drugs, especially in situ- suppressed by bile acids via a small heterodimer partner-dependent ations when LCA is increased such as during cholestasis. mechanism. Mol. Endocrinol. 20, 65–79. Fisher, M.M., Magnusson, R., Miyai, K., 1971. Bile acid metabolism in mammals. I. Bile acid-induced intrahepatic cholestasis. Lab Invest. 25, 88–91. Acknowledgments Fan, J., Liu, S., Du, Y., Morrison, J., Shipman, R., Pang, K.S., 2009. Up-regulation of transporters and enzymes by the vitamin D receptor ligands, 1alpha, 25- dihydroxyvitamin D3 and vitamin D analogs, in the Caco-2 cell monolayer. J. The authors thank Dr. Vincent B. Nieuwenhuijs (University Pharmacol. Exp. Ther. 330, 389–402. Medical Center, Groningen) for providing the human ileum tissue. Frankenberg, T., Rao, A., Chen, F., Haywood, J., Shneider, B.L., Dawson, P.A., 2006. Regulation of the mouse organic solute transporter alpha–beta, Ostalpha– This work was supported by the Canadian Institutes for Health Ostbeta, by bile acids. Am. J. Physiol. Gastrointest. Liver Physiol. 290, G912– Research, MOP89850; Edwin C.Y. Chow is a recipient of the Alexan- G922. der Graham Bell NSERC fellowship, Canada. Fukumori, S., Murata, T., Taguchi, M., Hashimoto, Y., 2007. Rapid and drastic induction of CYP3A4 mRNA expression via vitamin D receptor in human intestinal LS180 cells. Drug Metab. Pharmacokinet. 22, 377–381. Gascon-Barre, M., Demers, C., Mirshahi, A., Neron, S., Zalzal, S., Nanci, A., 2003. The References normal liver harbors the vitamin D nuclear receptor in nonparenchymal and biliary epithelial cells. Hepatology 37, 1034–1042. Abrahamsson, A., Gustafsson, U., Ellis, E., Nilsson, L.M., Sahlin, S., Bjorkhem, I., Geier, A., Martin, I.V., Dietrich, C.G., Balasubramaniyan, N., Strauch, S., Suchy, F.J., Einarsson, C., 2005. Feedback regulation of bile acid synthesis in human liver: Gartung, C., Trautwein, C., Ananthanarayanan, M., 2008. Hepatocyte nuclear 246 90 A.A. Khan et al. / Toxicology in Vitro 25 (2011) 80–90

factor-4alpha is a central transactivator of the mouse Ntcp gene. Am. J Physiol. Makishima, M., Lu, T.T., Xie, W., Whitfield, G.K., Domoto, H., Evans, R.M., Haussler, Gastrointest. Liver Physiol. 295, G226–G233. M.R., Mangelsdorf, D.J., 2002. Vitamin D receptor as an intestinal bile acid Graaf, I.A.M.D., Groothuis, G.M., Olinga, P., 2007. Precision-cut tissue slices as a tool sensor. Science 296, 1313–1316. to predict metabolism of novel drugs. Expert Opin. Drug Metab. Toxicol. 3, 879– Martignoni, M., de Kanter, R., Grossi, P., Saturno, G., Barbaria, E., Monshouwer, M., 898. 2006. An in vivo and in vitro comparison of CYP gene induction in mice using Hirano, S., Masuda, N., Oda, H., 1981. In vitro transformation of chenodeoxycholic liver slices and quantitative RT-PCR. Toxicol. In Vitro 20, 125–131. acid and ursodeoxycholic acid by human intestinal flora, with particular Matsubara, T., Yoshinari, K., Aoyama, K., Sugawara, M., Sekiya, Y., Nagata, K., reference to the mutual conversion between the two bile acids. J. Lipid Res. 22, Yamazoe, Y., 2008. Role of vitamin D receptor in the lithocholic acid-mediated 735–743. CYP3A induction in vitro and in vivo. Drug Metab. Dispos. 36, 2058–2063. Hirohashi, T., Suzuki, H., Ito, K., Ogawa, K., Kume, K., Shimizu, T., Sugiyama, Y., McCarthy, T.C., Li, X., Sinal, C.J., 2005. Vitamin D receptor-dependent regulation of 1998. Hepatic expression of multidrug resistance-associated protein-like colon multidrug resistance-associated protein 3 gene expression by bile acids. J. proteins maintained in eisai hyperbilirubinemic rats. Mol. Pharmacol. 53, Biol. Chem. 280, 23232–23242. 1068–1075. Narisawa, T., Magadia, N.E., Weisburger, J.H., Wynder, E.L., 1974. Promoting effect of Hirohashi, T., Suzuki, H., Takikawa, H., Sugiyama, Y., 2000. ATP-dependent transport bile acids on colon carcinogenesis after intrarectal instillation of N-methyl-N0- of bile salts by rat multidrug resistance-associated protein 3 (Mrp3). J. Biol. nitro-N-nitrosoguanidine in rats. J. Natl. Cancer Inst. 53, 1093–1097. Chem. 275, 2905–2910. Nehring, J.A., Zierold, C., DeLuca, H.F., 2007. Lithocholic acid can carry out in vivo Hofmann, A.F., 2004. Detoxification of lithocholic acid, a toxic bile acid: relevance to functions of vitamin D. Proc. Natl. Acad. Sci. USA 104, 10006–10009. drug hepatotoxicity. Drug Metab. Rev. 36, 703–722. Olinga, P., Merema, M.T., Hof, I.H., De Jager, M.H., De Jong, K.P., Slooff, M.J., Meijer, Honjo, Y., Sasaki, S., Kobayashi, Y., Misawa, H., Nakamura, H., 2006. 1,25- D.K., Groothuis, G.M., 1998. Effect of cold and warm ischaemia on drug dihydroxyvitamin D3 and its receptor inhibit the chenodeoxycholic acid- metabolism in isolated hepatocytes and slices from human and monkey liver. dependent transactivation by farnesoid X receptor. J. Endocrinol. 188, 635–643. Xenobiotica 28, 349–360. Inokuchi, A., Hinoshita, E., Iwamoto, Y., Kohno, K., Kuwano, M., Uchiumi, T., 2001. Olinga, P., Elferink, M.G., Draaisma, A.L., Merema, M.T., Castell, J.V., Perez, G., Enhanced expression of the human multidrug resistance protein 3 by bile salt in Groothuis, G.M., 2008. Coordinated induction of drug transporters and phase I human enterocytes. A transcriptional control of a plausible bile acid and II metabolism in human liver slices. Eur. J. Pharm. Sci. 33, 380–389. transporter. J. Biol. Chem. 276, 46822–46829. Ourlin, J.C., Lasserre, F., Pineau, T., Fabre, J.M., Sa-Cunha, A., Maurel, P., Vilarem, M.J., Ishizawa, M., Matsunawa, M., Adachi, R., Uno, S., Ikeda, K., Masuno, H., Shimizu, M., Pascussi, J.M., 2003. The small heterodimer partner interacts with the pregnane Iwasaki, K., Yamada, S., Makishima, M., 2008. Lithocholic acid derivatives act as X receptor and represses its transcriptional activity. Mol. Endocrinol. 17, 1693– selective vitamin D receptor modulators without inducing hypercalcemia. J. 1703. Lipid Res. 49, 763–772. Owen, B.M., Milona, A., van Mil, S., Clements, P., Holder, J., Boudjelal, M., Cairns, W., Javitt, N.B., 1966. Cholestasis in rats induced by taurolithocholate. Nature 210, Parker, M., White, R., Williamson, C., 2010. Intestinal detoxification limits the 1262–1263. activation of hepatic pregnane X receptor by lithocholic acid. Drug Metab. Jung, D., Mangelsdorf, D.J., Meyer, U.A., 2006. Pregnane X receptor is a target of Dispos. 38, 143–149. farnesoid X receptor. J. Biol. Chem. 281, 19081–19091. Schmiedlin-Ren, P., Thummel, K.E., Fisher, J.M., Paine, M.F., Lown, K.S., Watkins, P.B., Jung, D., Elferink, M.G., Stellaard, F., Groothuis, G.M., 2007. Analysis of bile acid- 1997. Expression of enzymatically active CYP3A4 by Caco-2 cells grown on induced regulation of FXR target genes in human liver slices. Liver Int. 27, 137– extracellular matrix-coated permeable supports in the presence of 1alpha,25- 144. dihydroxyvitamin D3. Mol. Pharmacol. 51, 741–754. Kast, H.R., Goodwin, B., Tarr, P.T., Jones, S.A., Anisfeld, A.M., Stoltz, C.M., Tontonoz, P., Staudinger, J.L., Goodwin, B., Jones, S.A., Hawkins-Brown, D., MacKenzie, K.I., LaTour, Kliewer, S., Willson, T.M., Edwards, P.A., 2002. Regulation of multidrug A., Liu, Y., Klaassen, C.D., Brown, K.K., Reinhard, J., Willson, T.M., Koller, B.H., resistance-associated protein 2 (ABCC2) by the nuclear receptors pregnane X Kliewer, S.A., 2001. The nuclear receptor PXR is a lithocholic acid sensor that receptor, farnesoid X-activated receptor, and constitutive androstane receptor. protects against liver toxicity. Proc. Natl. Acad. Sci. USA 98, 3369–3374. J. Biol. Chem. 277, 2908–2915. Thummel, K.E., Brimer, C., Yasuda, K., Thottassery, J., Senn, T., Lin, Y., Ishizuka, H., Khan, A.A., Chow, E.C., Porte, R.J., Pang, K.S., Groothuis, G.M., 2009a. Expression and Kharasch, E., Schuetz, J., Schuetz, E., 2001. Transcriptional control of intestinal regulation of the bile acid transporter, OSTalpha–OSTbeta in rat and human cytochrome P-4503A by 1alpha,25-dihydroxy vitamin D3. Mol. Pharmacol. 60, intestine and liver. Biopharm. Drug Dispos. 30, 241–258. 1399–1406. Khan, A.A., Chow, E.C., van Loenen-Weemaes, A.M., Porte, R.J., Pang, K.S., Groothuis, van de Kerkhof, E.G., de Graaf, I.A., de Jager, M.H., Meijer, D.K., Groothuis, G.M., 2005. G.M., 2009b. Comparison of effects of VDR versus PXR, FXR and GR ligands on Characterization of rat small intestinal and colon precision-cut slices as an the regulation of CYP3A isozymes in rat and human intestine and liver. Eur. J. in vitro system for drug metabolism and induction studies. Drug Metab. Dispos. Pharm. Sci. 37, 115–125. 33, 1613–1620. Khan, A.A., Dragt, B.S., Porte, R.J., Groothuis, G.M., 2010. Regulation of VDR van de Kerkhof, E.G., Ungell, A.L., Sjoberg, A.K., de Jager, M.H., Hilgendorf, C., de expression in rat and human intestine and liver – consequences for CYP3A Graaf, I.A., Groothuis, G.M., 2006. Innovative methods to study human intestinal expression. Toxicol. In Vitro 24, 822–829. drug metabolism in vitro: precision-cut slices compared with using chamber Lamba, J.K., Lin, Y.S., Schuetz, E.G., Thummel, K.E., 2002a. Genetic contribution to preparations. Drug Metab. Dispos. 34, 1893–1902. variable human CYP3A-mediated metabolism. Adv. Drug Deliv. Rev. 54, 1271– Wolf, G., 2002. Intestinal bile acids can bind to and activate the vitamin D receptor. 1294. Nutr. Rev. 60, 281–283. Lamba, J.K., Lin, Y.S., Thummel, K., Daly, A., Watkins, P.B., Strom, S., Zhang, J., Xu, Y., Iwanaga, K., Zhou, C., Cheesman, M.J., Farin, F., Thummel, K.E., 2006. Selective Schuetz, E.G., 2002b. Common allelic variants of cytochrome P4503A4 and their induction of intestinal CYP3A23 by 1alpha, 25-dihydroxyvitamin D3 in rats. prevalence in different populations. Pharmacogenetics 12, 121–132. Biochem. Pharmacol. 72, 385–392. Landrier, J.F., Eloranta, J.J., Vavricka, S.R., Kullak-Ublick, G.A., 2006. The nuclear Zelcer, N., van de Wetering, K., de Waart, R., Scheffer, G.L., Marschall, H.U., Wielinga, receptor for bile acids, FXR, transactivates human organic solute transporter- P.R., Kuil, A., Kunne, C., Smith, A., van der Valk, M., Wijnholds, J., Elferink, R.O., alpha and -beta genes. Am. J. Physiol. Gastrointest. Liver Physiol. 290, G476– Borst, P., 2006. Mice lacking Mrp3 (Abcc3) have normal bile salt transport, but G485. altered hepatic transport of endogenous glucuronides. J. Hepatol. 44, 768–775. Lehmann, J.M., Kliewer, S.A., Moore, L.B., Smith-Oliver, T.A., Oliver, B.B., Su, J.L., Zollner, G., Wagner, M., Moustafa, T., Fickert, P., Silbert, D., Gumhold, J., Sundseth, S.S., Winegar, D.A., Blanchard, D.E., Spencer, T.A., Willson, T.M., 1997. Fuchsbichler, A., Halilbasic, E., Denk, H., Marschall, H.U., Trauner, M., 2006. Activation of the nuclear receptor LXR by oxysterols defines a new hormone Coordinated induction of bile acid detoxification and alternative elimination in response pathway. J. Biol. Chem. 272, 3137–3140. mice: role of FXR-regulated organic solute transporter-alpha/beta in the Lou, G., Li, Y., Chen, B., Chen, M., Chen, J., Liao, R., Zhang, Y., Wang, Y., Zhou, D., 2007. adaptive response to bile acids. Am. J. Physiol. Gastrointest. Liver Physiol. 290, Functional analysis on the 50-flanking region of human FXR gene in HepG2 cells. G923–G932. Gene 396, 358–368. Zollner, G., Wagner, M., Fickert, P., Silbert, D., Gumhold, J., Zatloukal, K., Denk, H., Mahnke, A., Strotkamp, D., Roos, P.H., Hanstein, W.G., Chabot, G.G., Nef, P., 1997. Trauner, M., 2007. Expression of bile acid synthesis and detoxification enzymes Expression and inducibility of cytochrome P450 3A9 (CYP3A9) and other and the alternative bile acid efflux pump MRP4 in patients with primary biliary members of the CYP3A subfamily in rat liver. Arch. Biochem. Biophys. 337, 62– cirrhosis. Liver Int. 27, 920–929. 68.

247

APPENDIX A4

Chow EC, Sun H, Khan AA, Groothuis GM and Pang KS (2010) Effects of 1,25- dihydroxyvitamin D3 on transporters and enzymes of the rat intestine and kidney in vivo. Biopharm Drug Dispos 31:91-108

248

BIOPHARMACEUTICS & DRUG DISPOSITION Biopharm. Drug Dispos. 31: 91–108 (2010) Published online 9 December 2009 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/bdd. 694

a Effects of 1 ,25-Dihydroxyvitamin D3 on Transporters and Enzymes of the Rat Intestine and Kidney In Vivo

Ã Edwin C. Y. Chowa, Huadong Suna, Ansar A. Khanb, Geny M. M. Groothuisb, and K. Sandy Panga, aDepartment of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, Canada bPharmacokinetics, Toxicology and Targeting, Department of Pharmacy, University of Groningen, The Netherlands

a ABSTRACT: 1 ,25-Dihydroxyvitamin D3 (1,25(OH)2D3), the natural ligand of the vitamin D receptor (VDR), was found to regulate bile acid related transporters and enzymes directly and indirectly in the rat intestine and liver in vivo. The kidney is another VDR-rich target organ in which VDR regulation on xenobiotic transporters and enzymes is ill-defined. Hence, changes in protein and mRNA expression of nuclear receptors, transporters and enzymes of the rat intestine and kidney in response to 1,25(OH)2D3 treatment (0 to 2.56 nmol/kg/day intraperitoneally in corn oil for 4 days) were studied. In the intestine, protein and not mRNA levels of Mrp2, Mrp3, Mrp4 and PepT1 in the duodenum and proximal jejunum were induced, whereas Oat1 and Oat3 mRNA were decreased in the ileum after 1,25(OH)2D3 treatment. In the kidney, VDR, Cyp24, Asbt and Mdr1a mRNA and protein expression increased significantly (2- to 20-fold) in 1,25(OH)2D3-treated rats, and a 28-fold increase of Cyp3a9 mRNA but not of total Cy3a protein nor Cyp3a1 and Cyp3a2 mRNA was observed, implicating that VDR played a significant, renal-specific role in Cyp3a9 induction. Additionally, renal mRNA levels of PepT1, Oat1, Oat3, Osta, and Mrp4, and protein levels of PepT1 and Oat1 were decreased in a dose-dependent manner, and the 50% concomitant reduction in FXR, SHP, HNF-1a and HNF-4a mRNA expression suggests the possibility of cross-talk among the nuclear receptors. It is concluded that the effects of 1,25(OH)2D3 changes are tissue- specific, differing between the intestine and kidney which are VDR-rich organs. Copyright r 2009 John Wiley & Sons, Ltd. a Key words: 1 ,25-dihydroxyvitamin D3; vitamin D receptor; intestine; colon; kidney; transporters; enzymes; nuclear receptors; Asbt; Mrp; PepT1; Oat; Mdr1a; P-gp; cytochrome P450 enzymes; Cyp24 and Cyp3a9

Introduction a Glossary: 1,25(OH)2D3,1,25-dihydroxyvitamin D3; Asbt/ ASBT, rat/human apical sodium dependent bile acid trans- Vitamin D, the inert precursor of the active ligand, porter; Cyp/CYP, rat/human cytochrome P450 enzyme; FXR, a farnesoid X receptor; Gapdh, rat glyceraldehyde-3-phosphate 1 ,25-dihydroxyvitamin D3 [1,25(OH)2D3], has dehydrogenase; HNF, hepatocyte nuclear factor; LRH-1, liver been used widely as a nutraceutical in the receptor homolog-1; LXRa, liver X receptor alpha; Mdr1a, rat multidrug resistance protein 1a or P-glycoprotein (P-gp); Mrp, prevention of cancer and prolongation of long- rat multidrug resistance-associated protein; NADPH, nicoti- evity [1–4]. Much is known about the molecular namide adenine dinucleotide phosphate; Oat, rat organic anion transporter; Oct, rat organic cation transporter; Ost, rat organic solute transporter; PepT, oligopeptide transporter; PMSF, phenylmethylsulfonyl fluoride; SHP, short heterodimer *Correspondence to: Leslie L. Dan Faculty of Pharmacy, partner; Sult2a1/SULT2A1, rat/human hydroxysteroid sulfo- University of Toronto, 144 College Street, Toronto, Ontario, transferase; TBS-T, Tris-buffered saline with 0.1% Tween 20; Canada M5S 3M2. VDR, vitamin D receptor. E-mail: [email protected]

Received 16 July 2009 Revised 2 October 2009 Copyright r 2009 John Wiley & Sons, Ltd. Accepted 5 November 2009

249 92 E. C. Y. CHOW ET AL. actions of vitamin D on the regulation of calcium VDR and FXR effects in the intestine and indirect and phosphorus homeostasis and its indirect FXR effects in the liver [18]. Induction of the ileal feedback on the parathyroid hormone [5]. The Asbt by 1,25(OH)2D3 via VDR [16] led to increased activation of vitamin D requires both the con- bile acid absorption that resulted in increased secutive metabolism by the liver and the kidney to intestinal fibroblast growth factor 15 (FGF15). a form 25-hydroxyvitamin D3 then 1 ,25-dihydroxy- These triggered various secondary, FXR-related vitamin D3, the ligand of the vitamin D receptor or non-related effects that led to down-regulation (VDR) [6,7]. The toxic bile acid, lithocholic acid, is of the cholesterol metabolizing enzyme, Cyp7a1, also a VDR ligand that activates the VDR, albeit at in the rat liver [18]. The composite findings mM rather than the nM concentration required for suggest the importance of VDR in the regulation 1,25(OH)2D3 [8,9]. VDR is present abundantly in of not only intestinal Asbt and Cyp3a enzymes the rat intestine and kidney [10], but is present at but also secondary FXR-effects, resulting in much lesser amounts in the liver, where VDR is reduction of Cyp7a1 in the liver in vivo [18]. found mostly in the stellate cells, Kupffer cells, The small intestine and the kidney are two endothelial cells and cholangiocytes and not major target organs of the VDR [5]. The kidney hepatocytes [11]. In contrast, in the human liver, plays a central role in the formation of VDRisexpressedinhepatocytesatlowlevel 1,25(OH)2D3. VDR mediates the regulation of in addition to expression in non-parenchymal calcium absorption in the intestine and reabsorp- cells [12,13]. tion in the kidney, with 1,25(OH)2D3 controlling There are increasingly more in vitro and in vivo levels of the calcium-binding proteins, calbindins studies investigating the effects of 1,25(OH)2D3 D9K and D28K, and the calcium ion channels, on enzymes and transporters within first-pass transient receptor potential vanilloid type 5 organs, the liver and the intestine. 1,25(OH)2D3 (TRPV5) and type 6 (TRPV6) [29–32]. In addition, was shown to regulate calcium homeostasis 1,25(OH)2D3 plays a major role in phosphate and [14,15] and to be involved in the regulation of sulfate homeostasis since the Type II renal transporters and enzymes [12,16–23]. The activa- sodium-dependent inorganic phosphate trans- tion of VDR by 1,25(OH)2D3 was reported to up- porter [33] and the sodium-sulfate cotransporter regulate the human cytochrome P450 enzyme, [34] are targets of the VDR. For the feedback CYP3A4, and the hydroxysteroid sulfotransfer- control of 1,25(OH)2D3 levels in the body, the ase (SULT2A1), and drug transporters such as the 1,25(OH)2D3-liganded VDR increases the cata- multidrug resistance protein (MDR1 or P-gp) and bolic enzyme, Cyp24, thereby increasing the the multidrug resistance associated proteins metabolism of 1,25(OH)2D3 to the inactive meta- a (MRP2 and MRP4) in Caco-2 cell studies bolite, 1 ,24,25-trihydroxyvitamin D3, to decrease [23–26]. In vivo, the VDR transactivated the intracellular levels of 1,25(OH)2D3 in the kidney murine Mrp3 [21] and the rat apical sodium [5]. Taken together, the composite information dependent bile acid transporter (Asbt) [16]. In suggests that 1,25(OH)2D3 and VDR play an addition, rat intestinal Cyp3a1 was observed to important role in the regulation of transporters be up-regulated by 1,25(OH)2D3 treatment both and enzymes in the intestine and kidney. in vivo and in vitro [17–19]. Interestingly, VDR However, the relative changes in target genes activation was able to blunt the liver X receptor for drug disposition by 1,25(OH)2D3 in the (LXRa) signaling in HepG2 cells [27] and could intestine and kidney, tissues which exhibit high also antagonize the activities of the farnesoid X VDR levels, have not been compared. This study receptor (FXR) [28], suggesting that cross-talk investigated the dose-dependent effects of interactions between these bile acids-related 1,25(OH)2D3 (from 0.64 to 2.56 nmol/kg/day) nuclear receptors could lead to changes in on intestinal and renal transporters and enzymes transporters and enzymes in the liver. However, and similar changes for VDR-target genes in the these effects are difficult to identify in vivo due to intestine and kidney were expected due to their confounding effects and inter-organ interactions. abundance in these tissues. These higher chosen In our previous study, it was found that rats doses in the rat are estimated, through pharma- treated with 1,25(OH)2D3 in vivo showed direct cokinetic analyses, to produce comparable blood

250 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT RENTAL AND INTESTINAL TRANSPORTERS AND ENZYMES 93

5 levels of 1,25(OH)2D3 to those in men who are n 4 in each group; from Charles River, St given slightly lower 1,25(OH)2D3 doses, as a Constant, QC) were injected intraperitoneally result of the higher clearance of 1,25(OH)2D3 with 0, 0.64, 1.28 and 2.56 nmol/kg/day of (4.6x) in the rat [35–39]. It was shown that Asbt, a 1,25(OH)2D3 in 1 ml/kg corn oil for 4 days. Rats VDR-targeted gene in the intestine, was also a were given water and food ad libitum and target in the kidney. However, opposite changes maintained under a 12:12 h light and dark cycle were also observed. The oligopeptide transporter, in accordance to animal protocols approved by PepT1, protein was increased in the intestine, the University of Toronto (ON, Canada). At 24 h a whereas PepT1 and PepT2 and Ost , as well as following the end of the last 1,25(OH)2D3 treat- the expression of organic anion transporters ment, the rats were anesthetized with an intra- (Oat1 and Oat3) were down-regulated by peritoneal injection of ketamine and xylazine (90 1,25(OH)2D3 in the kidney. The results infer and 10 mg/kg, respectively). The portal vein was possibilities of cross-talk or secondary effects of cannulated and flushed with 50 ml of ice-cold the nuclear receptors in the kidney. saline. The small intestine was removed and placed on ice, and divided into eight segments [segment 1 (S1) is the duodenum, spanning from the pyloric ring to the ligament of Treitz; segment Materials and Methods 2 (S2) is the proximal jejunum segment of equal length that is immediately distal to the ligament The active form of vitamin D, 1,25(OH) D ,in 2 3 of Treitz. The remaining small intestine was then powder form was procured from Sigma-Aldrich divided into six segments of equal length (S3 to Canada (Mississauga, ON, Canada). Antibodies S8, with S8 representing the ileum proximal were obtained from various sources: Cyp24 (H-87) to the ileocecal junction)] [16]. Each segment from Santa Cruz Biotechnology (Santa Cruz, CA); was everted and placed into a test tube contain- anti-Mrp2 (ALX-801-016-C250), from Alexis Bio- ing 1 mM phenylmethylsulfonyl fluoride (PMSF) chemicals, San Diego, CA; anti-Pgp (C219), from in physiological saline solution prior to scraping Abcam, Cambridge, MA; anti-GAPDH (14C10), of the enterocytes by a tissue-scraper [40]. The from Cell Signaling Technology, Danvers, MA; colon, taken as a 10 cm section adjacent to the anti-VDR (MA1-710) was purchased from Thermo ileocecal junction, was removed of fecal matters Fisher Scientific Inc., Rockford, IL; anti-Cyp3a2 via flushes of saline solution containing 1 mM (458223), from BD Biosciences, Mississauga, ON PMSF; the tissue was everted, placed into a test and OAT11-A was from Alpha Diagnostic Intl. tube containing 1 mM PMSF in saline and scraped Inc., San Antonio, TX. Other antibodies were kind for removal of the cells. The kidneys were also gifts from various investigators: anti-PepT1 harvested, cut into small pieces. All tissue samples (Dr Wolfgang Sadee, Ohio State University, were snap-frozen with liquid nitrogen, and stored Columbus, Ohio); anti-Asbt (Dr Paul A. Dawson, at 801C until further analysis. Wake Forest University School of Medicine, NC); anti-Mrp3 (Dr Yuichi Sugiyama, University of Tokyo, Japan); anti-Mrp4 (Dr John D. Schuetz, St Preparation of subcellular fractions Jude Children’s Research Hospital, TN). All other reagents were purchased from Sigma-Aldrich Enterocytes. The method of preparation of sub- Canada (Mississauga, ON, Canada) and Fisher cellular fractions had been described previously Scientific (Mississauga, ON, Canada). in detail [18]. Briefly, frozen mucosal scrapings ; ðOHÞ D (50–100 mg of tissue) were mixed with 1 ml of Induction study of 1 25 2 3 in rats in vivo Tris-HCl (0.1 M, pH 7.4) buffer containing 1% The 1,25(OH)2D3, in anhydrous ethanol solution, protease inhibitor cocktail (Sigma-Aldrich, ON) was analyzed spectrophotometrically at 265 nm and homogenized on ice and then sonicated for (UV-1700, Shimadzu Scientific Instruments, MD) 10 s. Samples were centrifuged at 1000 g at 41C and diluted in corn oil (Sigma-Aldrich, ON) for for 10 min and the resulting supernatant was injection. Male Sprague-Dawley rats (260–280 g; transferred to a new tube and spun again at

251 94 E. C. Y. CHOW ET AL.

21000 g at 41C for 1 h to yield the pellet (crude against the house keeping protein, villin, for membrane fraction). The crude membrane fraction intestinal samples and glyceraldehyde-3-phosphate was used for Western blot analyses of intestinal dehydrogenase (Gapdh) for kidney samples. transporters. Quantitative real-time polymerase chain reaction Kidney. One whole kidney was homogenized (1:5 (qPCR) w/v) in homogenizing buffer (250 mM sucrose, Scraped enterocytes from representative segments: 10 mM HEPES and 10 mM Tris-HCl, pH 7.4) mixed S1 for duodenum, S2 and S7 for proximal and distal with 1% protease inhibitor cocktail (Sigma- jejunum, respectively, and S8 for the ileum, as well Aldrich, ON) as described previously [18]. as the colon and the kidney were homo- Differential centrifugation was employed to genized with TRIzol (50–100 mg/ml). Total RNA obtain nuclear membrane, crude membrane and was isolated using the TRIzol extraction method crude cytosolic fractions for Western blot ana- (Sigma-Aldrich, ON) according to the manufac- lyses. Briefly, the kidney homogenate was cen- turer’s protocol, with modifications. The total trifuged at 9000 g for 10 min at 41C. The pellet RNA of each sample was then quantified by UV or the nuclear fraction was resuspended with spectrometry measured at 260 nm and the purity nuclear buffer, as described previously [18]. The was checked by 260 nm/280 nm and 260 nm/ supernatant was then spun at 33000 g for 230 nm ratio (X1.8). 1.5 mg of total RNA was used 60 min at 41C. The resultant supernatant or to synthesize cDNA using the High Capacity cytosol and the membrane fraction, resuspended cDNA Reverse Transcription Kit (Applied Biosys- with the resuspension buffer (50 mM mannitol, tems Canada, ON) on the Applied Biosystem 2720 20 mM HEPES, 20 mM Trizma base, pH 7.4), were Thermal Cycler. Real-time quantitative polymer- mixed with 1% protease inhibitor cocktail (Sig- ase chain reaction (qPCR) was performed with two ma-Aldrich, ON). detection systems (SYBR Green or Taqman assay), The method of Lowry et al. [41] was used to depending on the availability of primer sets. measure the protein concentration using bovine Information on the primers was described in Chow serum albumin as the standard. Samples were et al. [18] and shown in Table 1. BLAST analysis then stored at 801C until Western blot analysis. was run for the primer sets to ensure primer specificity for the gene of interest (http:// Immunoblotting www.ncbi.nlm.nih.gov/BLAST/). Aliquots of 75 ng of cDNA were mixed with 1 mM of forward Similar to the procedure of Chow et al. [18], and reverse primers, and 1 Power SYBR Green samples were loaded and separated by 7.5% or PCR Master Mix (Applied Biosystems) to perform 10% SDS-polyacrylamide gels and transferred PCR analysis. Amplification and detection were onto nitrocellulose membranes (Amersham Bio- performed using the ABI 7500 system at the sciences, Piscataway, NJ). Membranes were default thermal cycle setting: 951C for 10 min, blocked with 5% (w/v) skim milk in Tris- and 40 cycles of 951C for 15 s and 601C for 1 min, buffered saline (pH 7.4) with 0.1% Tween 20 followed by the dissociation curve. All target gene (TBS-T) (Sigma-Aldrich, ON), washed with TBS-T, mRNA data were normalized to villin in the and incubated with primary antibody solution intestine and Gapdh in the kidney by using ABI overnight at 41C. The membrane was washed Sequence Detection software version 1.4 to obtain again, followed by incubation with secondary critical threshold cycle (CT) values. Fold expres- antibody. After the second incubation, the mem- (DDCT) sion is represented as 2 for relative mRNA branes were washed and bands were visualized quantification. using chemiluminescence reagents from Amer- sham Biosciences (Piscataway, NJ) and quantified Statistical analysis by scanning densitometry (NIH Image software; http://rsb.info.nih.gov/nih-image/). To correct for Protein and mRNA data were expressed as protein loading error, the band intensity of the mean7standard deviation. The two-tailed target protein of each sample was normalized Student’s t-test and the Mann-Whitney U test

252 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT RENTAL AND INTESTINAL TRANSPORTERS AND ENZYMES 95

Table 1. Rat primer sets for quantitative real-time PCR Gene bank number Forward (50- 30 sequence) Reverse (50- 30 sequence) Gapdh XR_007996 CGCTGGTGCTGAGTATGTCG CTGTGGTCATGAGCCCTTCC Mrp2 NM_012833 CTGGTGTGGATTCCCTTGG CAAAACCAGGAGCCATGTGC Mrp3 NM_080581 ACACCGAGCCAGCCATATAC TCAGCTTCACATTGCCTGTC Mrp4 NM_133411 GCCCTTACCCAGCTGCTGA CAGAATCCAGAGAGCCTCTTTTACA Oat1 NM_017224 ATGCCTATCCACACCCGTGC GGCAAAGCTAGTGGCAAACC Oat3 BC081777 TGAGAAGTGTCTCCGCTTCG CTGTAGCCAGCGCCACTGAG Asbt NM_017222 TCAGTTTGGAATCATGCCTCTCA ACAGGAATAACAAGCGCAACCA PepT1a NM_057121 Applied Biosystems, Canada (Cat# Rn00589098_m1) PepT2 NM_031672 GGACCTTCCGAAGCGACAA GCGATGAGATGCTTTGGATATTT Ost-ab NM_001107087 TGTCATCCTGACCGCCCT AAGCGATCTGCCCGCTG Ost-b XM_001076555 TATTCCATCCTGGTTCTGGCAGT CGTTGTCTTGTGGCTGCTTCTT Mdr1a AY582535 GGAGGCTTGCAACCAGCATTC CTGTTCTGCCGCTGGATTTC Cyp3a1 NM_013105 GGAAATTCGATGTGGAGTGC AGGTTTGCCTTTCTCTTGCC Cyp3a2 XM_573414 AGTAGTGACGATTCCAACATAT TCAGAGGTATCTGTGTTTCCT Cyp3a9 U60085 GGACGATTCTTGCTTACAGG ATGCTGGTGGGCTTGCCTTC Cyp24 NM_201635 GCATGGATGAGCTGTGCGA AATGGTGTCCCAAGCCAGC VDR NM_017058 ACAGTCTGAGGCCCAAGCTA TCCCTGAAGTCAGCGTAGGT FXR NM_021745 AGGCCATGTTCCTTCGTTCA TTCAGCTCCCCGACACTTTT SHP BC088117 CCTTGGCTAGCTGGGTACCA GTCCCAAGGAGTACGCATACCT LRH-1 NM_021742 GCTGCCCTGCTGGACTACAC TGTAGGGCACATCCCCATTC HNF-1a X54423 CTCCTCGGTACTGCAAGAAACC TTGTCACCCCAGCTTAAGACTCT HNF-4a EF193392 CCAGCCTACACCACCCTGGAGTT TTCCTCACGCTCCTCCTGAA

s aPepT1 primer set is for Taqman gene expression assay. s bOsta primer set includes probe 50 FAM-CAGCCCTCCATTTTCTCCATCTTGGC-TAMRA 30 for Taqman gene expression assay. were used to compare differences between changes in transporter and enzyme expression by vehicle control and treatment groups. For in- 1,25(OH)2D3 since both mRNA and protein testinal and colon mRNA and protein analyses, expressions were present. the vehicle-treated S1 sample (value set as unity) was used for the normalization of other vehicle- Distribution of VDR in small intestine, colon, and 1,25(OH)2D3-treated samples in other seg- liver and kidney ments and colon. For renal mRNA and protein analyses, the vehicle-treated sample was set as Lower VDR mRNA levels in the kidney and liver the control (value set as unity), and was used were found when compared with that of the for comparison with those of other treatment small intestine (S1); the highest mRNA level samples. A value of p less than 0.05 was set as the existed in colon (Figure 1). However, there was significant level. no difference in VDR protein levels in the small intestine, colon and kidney, whereas the VDR protein level in the liver was only 14% that of S1 (Figure 1). Results

Distribution of enzymes and transporters in Distribution of HNF-1a in small intestine and ; ðOHÞ D the small intestine and colon and effects of colon and effects of 1 25 2 3 1; 25ðOHÞ D 2 3 The mRNA expression of HNF-1a was very low The distribution patterns of mRNA and protein and was evenly distributed in the small intestine; of the nuclear receptors, transporters and en- the levels were 2.2-fold higher in the colon zymes in the small intestine, if previously (Figure 2). Generally, 1,25(OH)2D3 treatment unknown, were first examined under control had no effect on HNF-1a along the length of the conditions and compared with the amount small intestine and colon although a minor present in the colon. Then the segments with significant decrease (15%) was observed in the the greatest abundance were chosen to examine ileum at 1.28 nmol/kg dose.

253 96 E. C. Y. CHOW ET AL.

Figure 3. Distribution and dose-dependent effects of 1,25(OH) D on intestinal Cyp24 mRNA (n 5 3 or 4 in each Ã2 3 group). indicates po0.05 compared with vehicle control in the same segment, whereas y indicates po0.05 compared with the level of vehicle-control S1 segment using the two-tailed Figure 1. Distribution of rat VDR mRNA and protein (n 5 3or Student’s t-test. #indicates po0.05 compared with vehicle 4 in each group). mRNA and protein distributions of VDR in control in the same segment using the Mann-Whitney U test. the small intestine [duodenum (S1), proximal jejunum (S2), distal jejunum (S7) and ileum (S8)], colon, liver and kidney are shown, normalized to Gapdh expression. y indicates po0.05 exhibited an increasing trend, from duodenum compared with the level of S1 segment using the two-tailed to ileum (3.6-fold of S1), and was highest in Student’s t-test. the colon (41-fold of S1; Figure 3). However, the mRNA expression of Cyp24 in the intestine was fairly low (CT value about 27 to 29 in the small intestine and 24 in the colon). Cyp24 mRNA levels were significantly induced with 1,25(OH)2D3 treatment along the length of the small intestine, but not in the colon at the highest dose. For the other doses, strong up-regulation was seen. But the results failed to reach statistical significance due to the high variation observed among the preparations (1.6 to 90-fold compared with control; Figure 3).

Distribution of apical, absorptive transporters in the small intestine and colon and effects of ; ðOHÞ D 1 25 2 3 Figure 2. Distribution and dose-dependent effects of 1,25(OH) D on intestinal HNF-1a mRNA (n 5 3 or 4 in each Ã2 3 group). indicates po0.05 compared with vehicle control in PepT1. The mRNA distribution of the oligopep- the same segment, whereas y indicates po0.05 compared with tide transporter, PepT1, an absorptive transpor- the level of vehicle-control S1 segment using the two-tailed ter, was evenly distributed in the rat small Student’s t-test. intestine, despite a previous report having suggested a proximal distribution [43]. The Distribution of Cyp24 in small intestine and mRNA of this absorptive transporter was vir- colon and effects of 1; 25ðOHÞ D 2 3 tually absent in the colon (Figure 4A). PepT1 The mRNA expression of Cyp24, a catabolic mRNA was mostly unchanged with 1,25(OH)2D3 enzyme that inactivates 1,25(OH)2D3 [32,42], (Figure 4A). By contrast, PepT1 protein levels in

254 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT RENTAL AND INTESTINAL TRANSPORTERS AND ENZYMES 97

5 Figure 4. Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal mRNA and protein of PepT1 (n 3 or 4 in each group). PepT1 mRNA (A) and protein (B), detected at 95 kDa in S1 and S2 segments, are shown with the 1,25(OH) D treatments Ã 2 3 indicates po0.05 compared with vehicle control in the same segment, whereas y indicates po0.05 compared with the level of vehicle-control S1 segment using the two-tailed Student’s t-test.

Figure 5. Distribution and dose-dependent effects of 1,25(OH) D on intestinal (A) Oat1 and (B) Oat3 mRNA (n 5 3 or 4 in each Ã 2 3 group). indicates po0.05 compared with vehicle control in the same segment, whereas y indicates po0.05 compared with the level of vehicle-control S1 segment using the two-tailed Student’s t-test.

S1 and S2 segments showed significant induction and Oat3, displayed an increasing trend, from at the highest 1,25(OH)2D3 dose (Figure 4B). duodenum to ileum (Figures 5A and 5B). The mRNA expression of Oat1 in the colon was 4-fold Oat1 and Oat3. Unlike PepT1, the mRNA dis- higher compared with that in the duodenum tribution of the organic anion transporters, Oat1 while there was no difference for Oat3. A small

255 98 E. C. Y. CHOW ET AL.

Figure 6. Distribution and dose-dependent effects of 1,25(OH) D on intestinal Mdr1a mRNA and P-gp protein (n 5 3 or 4 in each 2 3 Ã group). Mdr1a mRNA (A) and P-gp protein (B), detected at 170 kDa, for S8 are shown. denotes po0.05 compared with vehicle control in the same segment, whereas y indicates po0.05 compared with the level of vehicle-control S1 segment using the two-tailed Student’s t-test.

5 Figure 7. Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal mRNA and protein of Mrp2 (n 3 or 4 in each group). Mrp2 mRNA (A) and protein distribution (B, C) in the small intestine [duodenum (S1), jejunum (S2-S7), and ileum (S8)] and colon and (D) the inductive changes of Mrp2 protein in S1 and S2, detected at 180 kDa, with 1,25(OH) D treatments are Ã 2 3 shown. indicates po0.05 compared with vehicle control in the same segment, whereas y indicates po0.05 compared with the level of vehicle-control S1 segment using the two-tailed Student’s t-test.

256 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT RENTAL AND INTESTINAL TRANSPORTERS AND ENZYMES 99

5 o 5 oo increase in Oat1 mRNA (70%) with 1,25(OH)2D3 were highest in the colon: S1 S2 S7 S8 treatment in the duodenum and a 2.7-fold colon (Figure 6A). 1,25(OH)2D3 failed to alter the increase in Oat3 mRNA in the colon were mRNA expression of intestinal Mdr1a, except for observed at the highest dose (Figures 5A and a small decrease (34%) in the S7 segment at the 5B). However, in the ileum, both Oat1 and Oat3 highest 1,25(OH)2D3 dose (Figure 6A). P-gp mRNA levels were greatly decreased at all doses protein expression in the S8 segment (ileum) (456%); there was a 66% decrease in Oat3 was the highest, but did not show any demon- mRNA in S7 at the 1.28 nmol/kg dose. strable trend of induction with 1,25(OH)2D3 treatment (Figure 6B).

Distribution of apical efflux transporters, Mdr1a Mrp2. Unlike Mdr1a, the distribution of Mrp2 and Mrp2 in the small intestine and colon and ; ðOHÞ D mRNA and protein was found to decrease from effects of 1 25 2 3 the duodenum to ileum, as found earlier [45], then to very low levels in the colon; the mRNA Mdr1a (P-gp). Levels of Mdr1a mRNA displayed levels of these transporters were highest in S1 an increasing trend, from the duodenum to the and S2 (Figures 7A, 7B and 7C). 1,25(OH)2D3 did ileum, as observed by others [44], and the levels not alter the mRNA levels of Mrp2 among all the

5 Figure 8. Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal mRNA and protein of Mrp3 (n 3 or 4 in each group). Mrp3 mRNA (A) and protein (B, C) distribution in the small intestine [duodenum (S1), jejunum (S2-S7) and ileum (S8)] and colon are shown. Inductive changes in Mrp3 protein in S1 and S2, detected at 170 kDa, with the 1,25(OH) D treatments were Ã 2 3 observed (D). indicates po0.05 compared with vehicle control in the same segment, whereas y indicates po0.05 compared with the level of vehicle-control S1 segment using the two-tailed Student’s t-test.

257 100 E. C. Y. CHOW ET AL. intestinal segments (Figure 7A) and colon, but 8C). However, in the colon, protein levels of the protein expression of Mrp2 in the S1 and S2 Mrp3 were even higher than in the jejunum. segments was significantly induced by 1,25(OH)2D3 failed to perturb Mrp3 mRNA 1,25(OH)2D3 (Figures 7B and 7D). levels, but increased Mrp3 protein levels in both S1 and S2, especially with higher 1,25(OH)2D3 doses (Figures 8B and 8D). Distribution of basolateral efflux transporters in small intestine and colon and effects of ; ðOHÞ D Mrp4. The mRNA and protein distributions of 1 25 2 3 Mrp4 were similar to those of Mrp3: an ascending distribution pattern, increasing from the Mrp3. The mRNA expression of Mrp3 exhibited duodenum to ileum, was observed for Mrp4 an ascending distribution pattern, increasing mRNA (Figure 9A). A significantly higher from the duodenum to ileum, then the colon mRNA level was observed in the colon. Interest- (Figure 8A). Mrp3 protein expression was high- ingly, Mrp4 protein expression was highest in S1, est at the duodenum, but this drastically dropped but decreased precipitously in S2, followed by a within the proximal jejunum, with levels gradu- gradual increasing trend towards S8 in the small ally increasing towards the ileum (Figures 8B and intestine (Figures 9B and 9C). Mrp4 protein in the

5 Figure 9. Distribution and dose-dependent effects of 1,25(OH)2D3 on intestinal mRNA and protein of Mrp4 (n 3 or 4 in each group). Mrp4 mRNA (A) and protein (B, C) distribution in the small intestine [duodenum (S1), jejunum (S2-S7) and ileum (S8)] and colon are shown. Changes in Mrp4 protein by 1,25(OH) D treatments in S1 and S2 segments were observed (D). Mrp4 Ã 2 3 protein band was detected at 160 kDa. indicates po0.05 compared with vehicle control in the same segment, whereas y indicates po0.05 compared with the level of vehicle-control S1 segment using the two-tailed Student’s t-test.

258 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT RENTAL AND INTESTINAL TRANSPORTERS AND ENZYMES 101 colon was quite high and was only 17% Cyps. The mRNA level of Cyp24, the enzyme lower compared with that of the duodenum. known to respond to 1,25(OH)2D3 induction, Dose-dependent 1,25(OH)2D3 induction of was 20-fold higher at all 1,25(OH)2D3 doses Mrp4 was observed in the S2 segment (Figure 11A), and Cyp24 protein was increased (Figures 9B and 9D). significantly (4-fold) at the highest 1,25(OH)2D3 dose (Figure 11B). Interestingly, Cyp3a9 mRNA ; ðOHÞ D 4 Effect of 1 25 2 3 on nuclear receptors, expression was significantly induced 28-fold enzymes, drug transporters in the kidney by 1,25(OH)2D3 for all doses (Figure 11A). Cyp3a2 mRNA was reduced (495%; po0.05) at Nuclear receptors. Renal VDR mRNA was signifi- the highest dose, whereas the decrease in Cyp3a1 cantly induced about 2-fold at all doses of mRNA was not significant (Figure 11A). Total 1,25(OH)2D3 (Figure 10A) while the nuclear Cyp3a protein was slightly increased, though the VDR protein was significantly induced by 60% change was not significant (Figure 11B). at the highest dose (Figure 10B). LRH-1 mRNA was relatively unchanged (Figure 10A). On the Transporters. The mRNA expression of Oat1 and other hand, mRNA levels of other nuclear Oat3, basolateral transporters, was much higher receptors, FXR, SHP, and the transcription fac- than those in the small intestine (CT values of 17 tors, HNF-4a and HNF-1a, were reduced by 50% and 20 compared with 29 and 34 in the intestine), to 60% after 1,25(OH)2D3 treatment with the and was reduced with 1,25(OH)2D3 treatments doses of 1.28 and 2.56 nmol/kg. (Figure 12A). Osta mRNA was decreased by 60%

Figure 10. Dose-dependent effects of 1,25(OH) D on changes in (A) mRNA and (B) protein of renal nuclear receptors (n 5 3or4 2 3 Ã in each group). Gapdh and VDR bands were detected at 37 and 52 kDa, their molecular weights, respectively. indicates po0.05 compared with vehicle control using the two-tailed Student’s t-test.

259 102 E. C. Y. CHOW ET AL.

b at the highest dose while the decrease in Ost Mrp4 protein in the kidney with 1,25(OH)2D3 mRNA was insignificant. The protein expression treatment were minimal (Figure 13B). of Oat1 was reduced 50% (Figure 12B), whereas no change was observed for Mrp3 mRNA and protein (Figure 12A and 12B). Discussion The mRNA levels of the renal reabsorptive apical transporters, Asbt (4-fold) and P-gp (2–2.5 fold) In this study, it was demonstrated that transpor- were significantly increased (Figure 13A). In- ters and enzymes, known to be regulated by VDR creased Asbt protein was observed, and this or by non-genomic VDR effects, could be correlated with the increase in mRNA levels with regulated in both the intestine and kidney. The 1,25(OH)2D3 treatment (Figure 13A and 13B), effects of VDR appeared to affect kidney trans- though being significant only at the 1,25(OH)2D3 porters and enzymes more so than those in the dose of 0.64 nmol/kg. By contrast, P-gp mRNA intestine. Moreover, some of the changes were induction was associated with a 2- to 4-fold found to be opposite and tissue-specific. The increase in protein expression at the higher doses modulations found in this study might suggest (Figure 13A and 13B). However, down-regulation tissue-specific changes in drug disposition. of PepT1, PepT2 and Mrp4 mRNA expression In the rat kidney, there was induction of P-gp, was observed (Figure 13A), while that of Mrp2 localized in renal proximal tubules [46], Asbt, was unchanged. A significant reduction in PepT1 and Cyp3a9, localized in the renal cortex [47], mRNA level was associated with a parallel (Figures 11 and 13). Recently, Saeki et al. [48] and reduction in protein level. Changes in Mrp2 and Zierold et al. [49] showed the presence of a VDRE

Figure 11. Dose-dependent effects of 1,25(OH)2D3 on changes in (A) mRNA and (B) protein of renal cytochrome P450 isozymes (n 5 3 or 4 in each group). Gapdh, Cyp24 and Cyp3a bands were detected at 37, 50 and 56 kDa, their molecular weights, Ã respectively. indicates po0.05 compared with vehicle control using the two-tailed Student’s t-test.

260 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT RENTAL AND INTESTINAL TRANSPORTERS AND ENZYMES 103 in the human multidrug resistance protein induction of Cyp3a1 in the intestine and Cyp3a2 (MDR1) and in the rat Cyp3a9 genes, respectively. in the liver in the rat in vivo [18,19] and in rat Thus, induction of rat renal Cyp3a9 and P-gp intestinal slices [17]. The observation suggests that (Figures 11A and 13B) is most likely via VDR the induction of Cyp3a9 isoform is renal-specific. transactivation. Previous reports on the rat small Overall, renal VDR plays an important role in the intestine showed induction of Asbt in vivo [18], regulation of renal transporters and enzymes that and the increased Asbt in the kidney is likely a are VDR target genes. result of VDR activation as well (Figure 13). In the intestine, induction of PepT1 and Mrp Previously, we had attributed that the increased protein (Figures 4B, 7B, 8B and 9B) in the ileal Asbt elevated portal bile acid concentrations proximal segments may be related to non- with 1,25(OH)2D3 treatment [18]. Presently, it was genomic effects of 1,25(OH)2D3 as it was shown found that renal Asbt was also induced, and this to be linked to an active calcium absorption could further augment bile acid reabsorption, mechanism triggered by the VDR in the duode- rendering higher plasma bile acid concentrations num and proximal jejunum [50]. 1,25(OH)2D3 1 and leading to downstream FXR effects in the could increase Ca2 uptake in the rat duodenal liver [18]. Interestingly, Cyp3a9 mRNA in the intestine, an effect that may be linked to the kidney was induced (Figure 11A) even though cAMP-mediated activation of plasma membrane 1 renal Cyp3a1 and Cyp3a2 mRNAs (Figure 11A) Ca2 channels [51]. The increased cAMP levels were not affected by 1,25(OH)2D3, and this could explain elevated protein levels of PepT1 observation contrasted with the results on the [52], Mrp2 [53] and Mrp3 [54] in S1 and S2

Figure 12. Dose-dependent effects of 1,25(OH)2D3 on changes in (A) mRNA and (B) protein of renal basolateral transporters (n 5 3 or 4 in each group). Gapdh, Oat1 and Mrp3 bands were detected at 37, 72 and 170 kDa, their molecular weights, Ã respectively. indicates po0.05 compared with vehicle control using the two-tailed Student’s t-test.

261 104 E. C. Y. CHOW ET AL.

segments in our study, either by short term that treatment with 1,25(OH)2D3 increased the regulation or membrane vesicle trafficking. activity of protein kinase C (PKC) in Madin Furthermore, there may be a role of cAMP on Darby bovine kidney (MDBK) cells, a normal Mrp4 protein expression with 1,25(OH)2D3 treat- renal epithelial cell line derived from bovine ment since Mrp4 has been shown to transport kidney [57]. Decreased renal organic anion cAMP and cGMP [55]. Similar observations were transport by Oat1 [58] and Oat3 [59] due to made in the human Caco-2 cell monolayer; PKC activation was observed in the rat. This 4 incubation with 100 nM of 1,25(OH)2D3 for 3 reasoning may be used to explain the down- days increased MRP4 protein without a change regulation of Oat1 and Oat3 mRNA levels in the in mRNA, likely due to the stabilization of MRP4 ileum (Figure 5). Additionally, other studies protein with 1,25(OH)2D3 treatment [23]. found that PepT1 transport was attenuated with Similar to the observations for the small increased cAMP levels that led to PKC activation intestine, renal Oat1, Oat3, PepT1 and PepT2 [60,61]; the same may apply to PepT2 that also (Figures 12 and 13) may also be affected by the contains a PKC recognition region [62]. These non-genomic effects of 1,25(OH)2D3. In the could explain the decrease in renal PepT1 and kidney, 1,25(OH)2D3 and cAMP were needed to PepT2. Therefore, non-genomic effects of up-regulate VDR mRNA and protein expression 1,25(OH)2D3 could regulate transporters in the to modulate Cyp24 [56]. Others had also found kidney.

5 Figure 13. Dose-dependent effects of 1,25(OH)2D3 on changes in (A) mRNA and (B) protein of renal apical transporters (n 3or4 in each group). Gapdh, Asbt, PepT1, P-gp, Mrp2 and Mrp4 bands were detected at 37, 48, 95, 170, 180 and 160 kDa, their Ã molecular weights, respectively. indicates po0.05 compared with vehicle control using the two-tailed Student’s t-test.

262 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT RENTAL AND INTESTINAL TRANSPORTERS AND ENZYMES 105

Based on our previous study [18], cross-talk (Figure 12A) might be the result of decreased a between nuclear receptors was not observed HNF-1 levels in the 1,25(OH)2D3 treatment for the rat intestine even though VDR and FXR groups (Figure 10A), inasmuch as HNF-1a was activations by 1,25(OH)2D3 and bile acids, res- known to be involved in the basal expression pectively, were observed. In the liver, cross-talk of human OAT3 in Caco-2 and HepG2 cell between VDR and other nuclear receptors was lines [63]. Hence, renal VDR could potentially unlikely due to low levels of VDR in hepatocytes down-regulate renal transporters through the [11]. On the other hand, higher levels of VDR and down-regulation of other renal nuclear receptors other nuclear receptors are present in the kidney, or transcription factors, such as FXR and and are conducive to cross-talk interactions. HNF-1a. A reduction in FXR and SHP mRNA levels by The data from this and previous rat in vivo 1,25(OH)2D3 was observed (Figure 10A), and also studies [18] are summarized in Table 2. by Honjo et al. [28] in the VDR-ligand-mediated Certainly, the molecular mechanisms by which decrease of FXR and SHP in the kidney cell line, 1,25(OH)2D3 and VDR regulate intestinal and CV1. This reduction may have conduced to the renal drug transporters and enzymes remain down-regulation of Osta mRNA (Figure 12A), a unclear and need to be clarified. The VDR effects FXR targeted gene. The decrease in Ostb was not on increased Asbt and Cyp24 were similar for significant, showing that Osta, was more respon- both the intestine and kidney, whereas opposite sive to FXR. This was observed in rats in vivo trends for Mrp2, Mrp4, PepT1, Osta and Ostb where FXR effects in the liver increased the Osta were observed. VDR induction led to increased mRNA level but not Ostb [18]. However, Cyp3a1 and Cyp3a2 in the intestine and Cyp3a9 1,25(OH)2D3, in the absence of bile acids, could for the kidney; P-gp changes were insignificant in decrease Osta mRNA directly in rat intestinal the intestine but were increased markedly in the slices [20]. Similarly, the inhibition of rat Oat3 kidney. This study demonstrates that changes in

Table 2. Comparison of VDR-related changes in mRNA and protein in the rat intestine, kidney and liver after 1,25(OH)2D3 treatment (fold increases) of the rat in vivo for 4 days. Some of the data on the intestine and liver were obtained from Chow et al. [18] Nuclear receptors, Intestine Kidney Liver transporters or enzymes mRNA Protein mRNA Protein mRNA Protein VDR 22mmm2 FXR 2 N/A k N/A m N/A SHP m N/A k N/A m N/A LXRa N/A N/A N/A N/A m N/A HNF-1a 2 N/A k N/A m N/A HNF-4a N/A N/A k N/A m N/A LRH-1 2 N/A 2 N/A m N/A Cyp3a1 m m 2222 Cyp3a2 km Cyp3a9 2m 2 Cyp24 m N/A mm2N/A Asbt 2a ma mm22 Mdr1a (P-gp) 22a mmmm PepT1 2m k k N/A N/A PepT2 N/A N/A k N/A N/A N/A Mrp2 2m 2222 Mrp3 2m 22m 2 Mrp4 2m k 222 Osta ma N/A k N/A m N/A Ostb ma N/A k N/A 2 N/A Oat1 ka N/A kkN/A N/A Oat3 ka N/A k N/A N/A N/A

N/A dictates unavailable; 2 means unchanged. aChanges in the ileum only (S8 segment).

263 106 E. C. Y. CHOW ET AL. transporters and enzymes of the intestine and HNF4a, CYP7A1 and NTCP in human but kidney by 1,25(OH)2D3 treatment are tissue- not rat liver slices. Drug Metab Dispos 2009: specific and could alter the pharmacokinetics of submitted. 13. Han S, Chiang JY. Mechanism of vitamin D drugs differentially. This new information further receptor inhibition of cholesterol 7a-hydroxylase opens up more queries. More experiments are gene transcription in human hepatocytes. Drug needed to examine possible transcriptional and Metab Dispos 2009; 37: 469–478. posttranscriptional regulation mechanisms as 14. Walters JR, Balesaria S, Khair U, Sangha S, well as to examine changes in drug disposition. Banks L, Berry JL. The effects of vitamin D metabolites on expression of genes for calcium transporters in human duodenum. J Steroid Biochem Mol Biol 2007; 103: 509–512. Acknowledgments 15. Abrams SA, O’Brien KO. Calcium and bone mineral metabolism in children with chronic This work was supported by the Canadian illnesses. Annu Rev Nutr 2004; 24: 13–32. Institutes for Health Research, MOP89850. 16. Chen X, Chen F, Liu S et al. Transactivation of rat apical sodium-dependent bile acid transporter and increased bile acid transport by 1a,25- References dihydroxyvitamin D3 via the vitamin D receptor. Mol Pharmacol 2006; 69: 1913–1923. 17. Khan AA, Chow EC, van Loenen-Weemaes AM, 1. Mullin GE, Dobs A. Vitamin D and its role Porte RJ, Pang KS, Groothuis GM. Comparison in cancer and immunity: a prescription for of effects of VDR versus PXR, FXR and GR sunlight. Nutr Clin Pract 2007; 22: 305–322. ligands on the regulation of CYP3A isozymes in 2. Holick MF. Sunlight and vitamin D for bone rat and human intestine and liver. Eur J Pharm health and prevention of autoimmune diseases, Sci 37 cancers, and cardiovascular disease. Am J Clin 2009; : 115–125. 18. Chow ECY, Maeng HJ, Liu S, Khan AA, Nutr 2004; 80: 1678S–1688S. a 3. Schwartz GG, Skinner HG. Vitamin D status and Groothuis GMM, Pang KS. 1 ,25-Dihydroxyvi- cancer: new insights. Curr Opin Clin Nutr Metab tamin D3 triggered vitamin D receptor and Care 2007; 10: 6–11. farnesoid X receptor-like effects in rat intestine 4. Thomas DR. Vitamins in aging, health, and and liver in vivo. Biopharm Drug Dispos 2009; 30: longevity. Clin Interv Aging 2006; 1: 81–91. 457–475. 5. Jones G, Strugnell SA, DeLuca HF. Current 19. Xu Y, Iwanaga K, Zhou C, Cheesman MJ, Farin F, Thummel KE. Selective induction of intestinal understanding of the molecular actions of a vitamin D. Physiol Rev 1998; 78: 1193–1231. CYP3A23 by 1 ,25-dihydroxyvitamin D3 in rats. 6. Feldman D, Pike JW, Glorieux FH. Vitamin D. Biochem Pharmacol 2006; 72: 385–392. 2nd edn. Elsevier Academic Press: Amsterdam; 20. Khan AA, Chow EC, Porte RJ, Pang KS, Boston; 2005. Groothuis GM. Expression and regulation of a b 7. Prosser DE, Jones G. Enzymes involved in the the bile acid transporter, OST -OST in rat and activation and inactivation of vitamin D. Trends human intestine and liver. Biopharm Drug Dispos Biochem Sci 2004; 29: 664–673. 2009; 30: 241–258. 8. Makishima M, Lu TT, Xie W et al. Vitamin D 21. McCarthy TC, Li X, Sinal CJ. Vitamin D receptor- receptor as an intestinal bile acid sensor. Science dependent regulation of colon multidrug resis- 2002; 296: 1313–1316. tance-associated protein 3 gene expression by 9. Nehring JA, Zierold C, DeLuca HF. Lithocholic bile acids. J Biol Chem 2005; 280: 23232–23242. acid can carry out in vivo functions of vitamin D. 22. Schmiedlin-Ren P, Thummel KE, Fisher JM, Proc Natl Acad Sci USA 2007; 104: 10006–10009. Paine MF, Watkins PB. Induction of CYP3A4 a 10. Sandgren ME, Bronnegard M, DeLuca HF. by 1 ,25-dihydroxyvitamin D3 is human cell Tissue distribution of the 1,25-dihydroxyvitamin line-specific and is unlikely to involve pregnane D3 receptor in the male rat. Biochem Biophys Res X receptor. Drug Metab Dispos 2001; 29: Commun 1991; 181: 611–616. 1446–1453. 11. Gascon-Barre M, Demers C, Mirshahi A, 23. Fan J, Liu S, Du Y, Morrison J, Shipman R, Pang Neron S, Zalzal S, Nanci A. The normal liver KS. Up-regulation of transporters and enzymes harbors the vitamin D nuclear receptor in by the vitamin D receptor ligands, 1a,25-dihy- nonparenchymal and biliary epithelial cells. droxyvitamin D3 and vitamin D analogs, in the Hepatology 2003; 37: 1034–1042. Caco-2 cell monolayer. J Pharmacol Exp Ther 2009; 12. Khan AA, Chow EC, van Loenen-Weemaes AM, 330: 389–402. Porte RJ, Pang KS, Groothuis GM. 1a,25-Dihy- 24. Aiba T, Susa M, Fukumori S, Hashimoto Y. The droxy vitamin D3 mediates down-regulation of effects of culture conditions on CYP3A4 and

264 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT RENTAL AND INTESTINAL TRANSPORTERS AND ENZYMES 107

a 3 MDR1 mRNA induction by 1 ,25-dihydroxyvita- administration of H-1,25-dihydroxyvitamin D3 min D3 in human intestinal cell lines, Caco-2 and in man. Gut 1985; 26: 1240–1245. LS180. Drug Metab Pharmacokinet 2005; 20: 268–274. 37. Barger-Lux MJ, Heaney RP, Dowell S, Chen TC, 25. Schmiedlin-Ren P, Thummel KE, Fisher JM, Holick MF. Vitamin D and its major metabolites: Paine MF, Lown KS, Watkins PB. Expression of serum levels after graded oral dosing in healthy enzymatically active CYP3A4 by Caco-2 cells men. Osteoporos Int 1998; 8: 222–230. grown on extracellular matrix-coated permeable 38. Kissmeyer AM, Binderup L. Calcipotriol (MC supports in the presence of 1a,25-dihydroxy- 903): pharmacokinetics in rats and biological vitamin D3. Mol Pharmacol 1997; 51: 741–754. activities of metabolites. A comparative study 26. Echchgadda I, Song CS, Roy AK, Chatterjee B. with 1,25(OH)2D3. Biochem Pharmacol 1991; 41: Dehydroepiandrosterone sulfotransferase is a 1601–1606. target for transcriptional induction by the vita- 39. Muindi JR, Peng Y, Potter DM et al. Pharmaco- min D receptor. Mol Pharmacol 2004; 65: 720–729. kinetics of high-dose oral calcitriol: results from 27. Jiang W, Miyamoto T, Kakizawa T et al. Inhibi- a phase 1 trial of calcitriol and paclitaxel. Clin tion of LXRa signaling by vitamin D receptor: Pharmacol Ther 2002; 72: 648–659. possible role of VDR in bile acid synthesis. 40. Johnson BM, Zhang P, Schuetz JD, Brouwer KL. Biochem Biophys Res Commun 2006; 351: 176–184. Characterization of transport protein expression 28. Honjo Y, Sasaki S, Kobayashi Y, Misawa H, in multidrug resistance-associated protein (Mrp) 2-deficient rats. Drug Metab Dispos 2006; 34: Nakamura H. 1,25-dihydroxyvitamin D3 and its receptor inhibit the chenodeoxycholic acid-de- 556–562. pendent transactivation by farnesoid X receptor. 41. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. J Endocrinol 2006; 188: 635–643. Protein measurement with the Folin phenol 29. Dupret JM, Brun P, Perret C, Lomri N, reagent. J Biol Chem 1951; 193: 265–275. Thomasset M, Cuisinier-Gleizes P. Transcrip- 42. Healy KD, Zella JB, Prahl JM, DeLuca HF. Regulation of the murine renal vitamin D tional and post-transcriptional regulation of receptor by 1,25-dihydroxyvitamin D and cal- vitamin D-dependent calcium-binding protein 3 cium. Proc Natl Acad Sci USA 2003; 100: gene expression in the rat duodenum by 9733–9737. 1,25-dihydroxycholecalciferol. J Biol Chem 1987; 43. Fei YJ, Kanai Y, Nussberger S et al. Expression 262: 16553–16557. cloning of a mammalian proton-coupled 30. Hoenderop JG, van der Kemp AW, Hartog A oligopeptide transporter. Nature 1994; 368: et al. Molecular identification of the apical 21 563–566. Ca channel in 1,25-dihydroxyvitamin D3- 44. Liu S, Tam D, Chen X, Pang KS. P-glycoprotein responsive epithelia. J Biol Chem 1999; 274: and an unstirred water layer barring digoxin 8375–8378. absorption in the vascularly perfused rat small 31. den Dekker E, Hoenderop JG, Nilius B, Bindels intestine preparation: induction studies with RJ. The epithelial calcium channels, TRPV5 and pregnenolone-16a-carbonitrile. Drug Metab Dis- TRPV6: from identification towards regulation. pos 2006; 34: 1468–1479. Cell Calcium 2003; 33: 497–507. 45. Mottino AD, Hoffman T, Jennes L, Vore M. 32. Meyer MB, Zella LA, Nerenz RD, Pike JW. Expression and localization of multidrug Characterizing early events associated with the resistant protein mrp2 in rat small intestine. activation of target genes by 1,25-dihydroxyvi- J Pharmacol Exp Ther 2000; 293: 717–723. tamin D3 in mouse kidney and intestine in vivo. 46. Huls M, Kramers C, Levtchenko EN et al. J Biol Chem 2007; 282: 22344–22352. P-glycoprotein-deficient mice have proximal 33. Taketani Y, Segawa H, Chikamori M et al. 1 tubule dysfunction but are protected against Regulation of type II renal Na -dependent ischemic renal injury. Kidney Int 2007; 72: inorganic phosphate transporters by 1,25-dihy- 1233–1241. droxyvitamin D3. Identification of a vitamin 47. Anakk S, Ku CY, Vore M, Strobel HW. Insights D-responsive element in the human NAPi-3 into gender bias: rat cytochrome P450 3A9. gene. J Biol Chem 1998; 273: 14575–14581. J Pharmacol Exp Ther 2003; 305: 703–709. 34. Dawson PA, Markovich D. Regulation of the 48. Saeki M, Kurose K, Tohkin M, Hasegawa R. mouse Nas1 promoter by vitamin D and thyroid Identification of the functional vitamin D hormone. Pflugers Arch 2002; 444: 353–359. response elements in the human MDR1 gene. 35. Brandi L, Egfjord M, Olgaard K. Pharmacoki- Biochem Pharmacol 2008; 76: 531–542. a netics of 1,25(OH)2D3 and 1 (OH)D3 in normal 49. Zierold C, Mings JA, Deluca HF. 19nor-1,25- and uraemic men. Nephrol Dial Transplant 2002; dihydroxyvitamin D2 specifically induces 17: 829–842. CYP3A9 in rat intestine more strongly than 36. Ledger JE, Watson GJ, Ainley CC, Compston JE. 1,25-dihydroxyvitamin D3 in vivo and in vitro. Biliary excretion of radioactivity after intravenous Mol Pharmacol 2006; 69: 1740–1747.

265 108 E. C. Y. CHOW ET AL.

50. Marks HD, Fleet JC, Peleg S. Transgenic expres- and cAMP or PTH. Am J Physiol 1999; 276: sion of the human vitamin D receptor (hVDR) in E793–E805. the duodenum of VDR-null mice attenuates the 57. Simboli-Campbell M, Franks DJ, Welsh J. age-dependent decline in calcium absorption. 1,25(OH)2D3 increases membrane associated J Steroid Biochem Mol Biol 2007; 103: 513–516. protein kinase C in MDBK cells. Cell Signal 51. Massheimer V, Boland R, de Boland AR. Rapid 1992; 4: 99–109. 1,25(OH)2-vitamin D3 stimulation of calcium 58. Uwai Y, Okuda M, Takami K, Hashimoto Y, uptake by rat intestinal cells involves a dihy- Inui K. Functional characterization of the dropyridine-sensitive cAMP-dependent path- rat multispecific organic anion transporter way. Cell Signal 1994; 6: 299–304. OAT1 mediating basolateral uptake of anionic 52. Gaildrat P, Moller M, Mukda S et al. A novel drugs in the kidney. FEBS Lett 1998; 438: pineal-specific product of the oligopeptide 321–324. transporter PepT1 gene: circadian expression 59. Takeda M, Sekine T, Endou H. Regulation by mediated by cAMP activation of an intronic protein kinase C of organic anion transport promoter. J Biol Chem 2005; 280: 16851–16860. driven by rat organic anion transporter 3 53. Roelofsen H, Soroka CJ, Keppler D, Boyer JL. (rOAT3). Life Sci 2000; 67: 1087–1093. Cyclic AMP stimulates sorting of the canalicular 60. Muller U, Brandsch M, Prasad PD, Fei YJ, organic anion transporter (Mrp2/cMoat) to the Ganapathy V, Leibach FH. Inhibition of the H1/ apical domain in hepatocyte couplets. J Cell Sci peptide cotransporter in the human intestinal cell 1998; 111: 1137–1145. line Caco-2 by cyclic AMP. Biochem Biophys Res 54. Chandra P, Zhang P, Brouwer KL. Short-term Commun 1996; 218: 461–465. regulation of multidrug resistance-associated 61. Brandsch M, Miyamoto Y, Ganapathy V, protein 3 in rat and human hepatocytes. Am J Leibach FH. Expression and protein kinase Physiol Gastrointest Liver Physiol 2005; 288: C-dependent regulation of peptide/H1 co- G1252–G1258. transport system in the Caco-2 human colon 55. van Aubel RA, Smeets PH, Peters JG, Bindels RJ, carcinoma cell line. Biochem J 1994; 299: 253–260. Russel FG. The MRP4/ABCC4 gene encodes a 62. Daniel H, Herget M. Cellular and molecular novel apical organic anion transporter in human mechanisms of renal peptide transport. Am J kidney proximal tubules: putative efflux pump Physiol 1997; 273: F1–F8. for urinary cAMP and cGMP. J Am Soc Nephrol 63. Kikuchi R, Kusuhara H, Hattori N et al. Regula- 2002; 13: 595–603. tion of the expression of human organic anion 56. Yang W, Friedman PA, Kumar R et al. transporter 3 by hepatocyte nuclear factor 1al- Expression of 25(OH)D3 24-hydroxylase in distal pha/beta and DNA methylation. Mol Pharmacol nephron: coordinate regulation by 1,25(OH)2D3 2006; 70: 887–896.

266

APPENDIX A5

Maeng HJ, Durk MR, Chow EC, Ghoneim R and Pang KS (2011) 1α,25-Dihydroxyvitamin D3 on intestinal transporter function: studies with the rat everted intestinal sac. Biopharm Drug Dispos 32:112-125

267

BIOPHARMACEUTICS & DRUG DISPOSITION Biopharm. Drug Dispos. 32: 112–125 (2011) Published online 14 January 2011 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/bdd.742

a 1 ,25-Dihydroxyvitamin D3 on Intestinal Transporter Function: Studies with the Rat Everted Intestinal Sac

Ã Han-Joo Maeng, Matthew R. Durk, Edwin C. Y. Chow, Ragia Ghoneim, and K. Sandy Pang Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, Ontario, Canada M5S 3M2

a ABSTRACT: Previous studies have shown that 1 ,25-dihydroxyvitamin D3 (1,25(OH)2D3) treatment (2.56 nmol/kg i.p. daily 4) increased PepT1, Mrp2, Mrp4, Asbt, but not Mdr1/P-gp in the rat small intestine. In this study, the intestinal everted sac technique, together with various select probes: mannitol (paracellular transport), glycylsarcosine (PepT1), 5(and 6)-carboxy-20,70- dichlorofluorescein (CDF) diacetate (precursor of CDF for Mrp2), adefovir dipivoxil (precursor of adefovir for Mrp4) and digoxin (P-gp) was used to examine the functional changes of these transporters. After establishing identical permeabilities (Papp) of mannitol for the apical-to- basolateral (A-to-B) and basolateral-to-apical (B-to-A) directions at 20 min in 1,25(OH)2D3-treated vs. vehicle-treated duodenal, jejunal and ileal everted sacs, a significant enhancement of net A-to-B transport of glycylsarcosine in the duodenum, increased B-to-A transport of CDF and A-to-B and B- to-A transport of adefovir in the jejunum were observed with 1,25(OH)2D3 treatment. However, the A-to-B and B-to-A transport of digoxin in the ileum was unchanged. These changes in transporter function in the rat intestinal everted sac corresponded well to changes in proteins that were observed previously. This study confirms that the rat intestinal PepT1, Mrp2 and Mrp4, but not P-gp are functionally induced by 1,25(OH)2D3 treatment via the vitamin D receptor (VDR). Copyright r 2011 John Wiley & Sons, Ltd. a Key words: transporters; intestine; everted sac; 1 ,25-dihydroxyvitamin D3 (1,25(OH)2D3); vitamin D receptor (VDR)

Introduction have been well established during the past decades. But it is only recently that VDR-mediated Ligand-activated transcription factors, including regulation of drug transporters and enzymes is a the pregnane X receptor (PXR), constitutive being investigated. 1 ,25-Dihydroxyvitamin D3 androstane receptor (CAR), farnesoid X receptor [1,25(OH)2D3, calcitriol] [2] and other vitamin D (FXR), glucocorticoid receptor (GR) and vitamin analogs [3–5] as well as the secondary bile acid, D receptor (VDR) are among the nuclear receptor lithocholic acid and its derivatives [5], are known superfamilies that are critical in the regulation to bind to the VDR that is expressed abundantly of drug transporters and drug-metabolizing in the rat intestine [6,7] and kidney [7] to result enzymes in the intestine and liver [1]. The roles in changes in target genes. of the VDR in calcium and bone homeostasis In terms of metabolism, the ligand-activated VDR was found to induce CYP3A4 in the Caco-2 [8,9] and LS180 cell [8] monolayers and human *Correspondence to: K. Sandy Pang, Faculty of Pharmacy, University of Toronto, 144 College Street, Toronto, Ontario, intestinal slices [10]. In the rat intestine, Cyp3a9 [3] Canada M5S 3M2. E-mail: [email protected] and Cyp3a1 [6,10,11] mRNA and total Cyp3a

Received 24 August 2010 Revised 9 November 2010 Copyright r 2011 John Wiley & Sons, Ltd. Accepted 23 November 2010

268 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT INTESTINAL TRANSPORTERS 113

protein [6] were upregulated by 1,25(OH)2D3 from 6.8 to 8) and suggesting enhanced P-gp treatment. The hydroxysteroid sulfotransferase functional activity due to VDR activation [9]. The (SULT2A1) was activated in intestinal cells upon cellular retention of CDF in the 1,25(OH)2D3- co-transfection with the liganded-VDR [12]. treated Caco-2 cells was significantly decreased In terms of transporters, 1,25(OH)2D3 and litho- (to 60%) due to the higher efflux function of cholic acid, both VDR ligands, have led to MRP2 [9]. By contrast, the physiological rele- increased multidrug resistance protein 1 (MDR1) vance of VDR activation on the function of mRNA levels in the human colorectal adenocarci- transporters and enzymes in drug disposition noma cell lines, LS174T [13] and Caco-2 cells [9] in vivo is rarely investigated. Chen et al. [16] that abundantly express the VDR, observations showed, pursuant to 1,25(OH)2D3 treatment to consistent with the presence of a vitamin D the rat in vivo, that the ileal absorption of response element in the human MDR1 gene [14]. cholylsarcosine, a non-metabolized synthetic bile In Caco-2 cells, 1,25(OH)2D3 treatment increased acid analog, was enhanced in the perfused rat the protein expression of MRP2 and MRP4 intestinal preparation, an observation consistent (multidrug resistance-associated protein 2 and 4) with the elevated mRNA and protein expression [9]. In the mouse, mRNA levels of the multidrug of Asbt. resistance-associated protein 3 (Mrp3) in the In this study, the everted rat intestinal sac colon were enhanced upon pretreatment with technique was used to examine the function of 1,25(OH)2D3 [15]. In the rat, 1,25(OH)2D3 tran- intestinal transporters after treatment in vivo.The scriptionally activated the apical sodium-depen- everted sac technique, introduced by Wilson and dent bile acid transporter (Asbt) [16] and increased Wisemen [18] and further improved with use of protein expression of members of the multidrug specific probes for different transporters [19,20], is resistance-associated protein (Mrp2, Mrp3 and useful for the study of drug absorption and Mrp4) and the oligopeptide transporter 1 (PepT1) permeation via passive diffusion and/or transpor- without altering the mRNA and protein expres- ters in the small intestine [20–22]. After establish- sion of Mdr1a/P-gp (P-glycoprotein) [6,7] and the ing the integrity of the everted sac preparation breast cancer resistance protein (Bcrp) (unpub- with [14C]mannitol, transport studies were con- lished data) in the rat small intestine. Treatment ducted with glycylsarcosine (GlySar, a PepT1 a with 1 -hydroxyvitamin D3,aprecursorof substrate), digoxin (a P-gp substrate), CDF-DA, 1,25(OH)2D3, also induced mRNA levels of Asbt precursor of CDF (a Mrp2 substrate) and adefovir and Mrp4 in the small intestine of mice [17]. dipivoxil, precursor of adefovir (a Mrp4 substrate) Collectively, these studies suggest that trans- in the B-to-A or A-to-B direction in vehicle- and porter and enzyme genes are targets of the VDR 1,25(OH)2D3-treated everted sacs. The segment and that species differences exist between ro- chosen for each of the transport studies corre- dents and man in the induction of enzymes and sponded to the intestinal segment shown pre- transporters of the intestine. Unfortunately, viouslytoexhibitthegreatestchangeinproteinor changes in mRNA or protein are seldom corre- mRNA expression in rats treated with 1,25(OH)2D3 lated to function. There exist only a few reports [6,7]. By defining the impact of 1,25(OH)2D3 on correlating protein expression and function due intestinal transporters, the importance of VDR in to VDR in cultured cells in vitro. Fan et al. [9] the induction or inhibition of intestinal transpor- demonstrated that the inductive effect of ters could be appraised for drug absorption as well 1,25(OH)2D3 on P-gp and MRP2 protein expres- as drug–drug interactions (DDIs). sion correlated to the increased efflux of digoxin and 5 (and 6)-carboxy-2,7-dichlorofluorescein diacetate (CDF-DA), respectively, across the Methods Caco-2 cell monolayer. Treatment of the Caco-2 cell monolayer with 1,25(OH) D (100 nM) for 3 2 3 Materials days significantly increased the apparent perme- 3 ability coefficient (Papp) of digoxin in the B-to-A Radiolabeled [ H]glycylsarcosine (specific activ- direction, yielding an increased efflux ratio (EfR, ity, 10.9 mCi/mmol), [8-3H]adefovir dipivoxil

269 114 H.-J. MAENG ET AL.

(specific activity, 5.5 mCi/mmol), [3H(G)]digoxin junction. Only the S1, S2, and S8 segments (specific activity, 40 mCi/mmol) and [1-14C]man- (6 cm in length for the sac preparation) were nitol (specific activity, 51 mCi/mmol) were pur- used in this study. Once rinsed with saline chased from Perkin Elmer Life and Analytical (0.9% NaCl solution), the intestine was everted Sciences (Waltham, MA), respectively. [3H]Di- with a glass rod and one end was tied using silk goxin was purified by HPLC to 499% radio- braided sutures before filling the sac with 0.3 ml chemical purity, as described by Liu et al. [23]. of freshly prepared, oxygenated buffer I 1,25(OH)2D3 and the non-radiolabeled drugs (117.6 mM NaCl, 25 mM NaHCO3, 1.2 mM MgCl2, (glycylsarcosine, adefovir dipivoxil, mannitol, 1.25 mM CaCl2,11mM glucose and 4.7 mM KCl, CDF-DA and digoxin) were procured from pH 7.4) [24] or drug-containing buffer I for the Sigma-Aldrich Canada (Mississauga, ON, Canada). A-to-B or B-to-A transport study. The other end All other reagents were obtained from Sigma- of the sac was immediately tied thereafter using Aldrich Canada (Mississauga, ON, Canada) and silk braided sutures, providing a sac of 4 cm Fisher Scientific (Mississauga, ON, Canada). length. The tied ends beyond the sutures were further clamped to prevent leakage. Immediately 1,25(OH)2D3 and vehicle (corn oil) treatment in thereafter, each sac was placed in 5 ml of rats oxygenated buffer II (117.6 mM NaCl, 25 mM NaHCO 1.2 mM MgCl , 1.25 mM CaCl ,11mM Male Sprague-Dawley rats (260–280 g), pur- 3, 2 2 glucose and 4.7 mM KCl, pH 6.8) [24] or drug- chased from Charles River (St Constant, QC), containing buffer II for the transport study in the were given water and food ad libitum and B-to-A or A-to-B direction, respectively. The sac maintained under a 12:12-h light and dark cycle 1 in accordance with animal protocols approved by was incubated at 37 C in a water bath with gentle shaking, with a 95% O /5% CO stream (1 l/min) the University of Toronto (Canada). 1,25(OH) D 2 2 2 3 bubbling through the mucosal (outside or A) was dissolved in anhydrous ethanol, and the solution. concentration was measured spectrophotometri- Low concentrations of [3H]glycylsarcosine cally at 265 nm (UV-1700, Shimadzu Scientific 3 (10 mM for S1), [ H]digoxin (10 mM for S8), Instruments, MD); then the solution was diluted 3 [ H]adefovir dipivoxil (1 mM for S2) and CDF- to 2.56 nmol/ml with filtered corn oil for intra- DA (50 mM for S2) were selected for study to peritoneal injection. Rats (n 5 4 in each group) were injected intraperitoneally for 0 (vehicle) or avoid saturation of the transporters. For verifica- tion of the integrity of the everted rat intestinal 2.56 nmol/kg 1,25(OH)2D3 in 1 ml/kg corn oil, sac and for paracellular passive diffusion, 100 mM daily for 4 days. mannitol (with [14C]mannitol) was applied to sacs prepared from S1, S2 and S8. For the Everted rat intestinal sac method transport study in the B-to-A-direction, 200 ml The rat everted intestinal sac study was carried was removed from the A compartment and out as described previously [18–21]. Briefly, replaced by buffer II at times of 10, 15, 20, 25 at 24 h after the last 1,25(OH)2D3 dose, rats and 30 min. For the transport study in the A-to-B- were anesthetized with ketamine and xylazine direction, the experiment was terminated at the (90 mg/kg and 10 mg/kg, respectively) given time of sampling (20 min), a time found to exhibit by intraperitoneal injection. The small intestine proportionality of uptake rates with time for the was removed quickly and placed on ice and cut probe. The time-linearity in the A-to-B direction into eight segments, as described in published was first determined for each compound, and reports [6,7]. Segment 1 (S1) represents the was established with use of multiple sac pre- duodenum and spans from the pyloric ring to parations after incubation times at 5, 10, 20, or the ligament of Treitz; segment 2 (S2) refers to the 30 min. At each time point, the experiment was proximal jejunum segment which is immediately terminated to obtain the contents of the probe in distal to the ligament of Treitz and of length the sac, donor and receiver compartments, and at identical to that for the duodenum. S8 denotes least three sacs were used to provide the mean the ileum segment proximal to the ileocecal value at each time point.

270 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT INTESTINAL TRANSPORTERS 115

At the end of the experiment, the total contents based on CDF efflux, and Mrp4 function, based of the mucosal and serosal compartments and the on adefovir efflux. Enterocytes were obtained sac tissue were removed for sampling. The sac from scrapings of the intestinal segment. Lysate, was washed in ice-cold buffer II, blotted dry, then in Tris-HCl buffer (50 mM, pH 7.4), was prepared cut open, and the serosal fluid inside the sac was from enterocytes by sonication by a cell disrup- drained into a pretared glass vial to determine tor, followed by centrifugation at 9000 g for the weight (translated to volume, assuming a 15 min. CDF was incubated with 2 mg protein, density of 1) of buffer I remaining in the sac assayed by the Lowry method [25] at 371C for 10, (basolateral side). The volume of the mucosal 20 and 30 min, and the reaction was stopped bath (apical side) was measured by taking the upon the addition of two volumes of ice-cold difference in weight of the mucosal fluid in the MeOH, followed by centrifugation at 14000 g containing vessel before and after the incubation for 10 min at 41C. Fluorescence values of the study. Recovery (%) of fluid volume was then supernatant obtained from the control and calculated as fluid weight after/before incuba- 1,25(OH)2D3 treatment groups were measured tion. For evaluation of amount recovery (% dose), and compared. samples of the mucosal and serosal fluids and in A similar incubation study was conducted for sac lysate were analysed for radioactivity by adefovir divipoxil (1 mM) to examine whether liquid scintillation counting (Beckman Coulter 1,25(OH)2D3 had an effect on hydrolysis. Data Canada, Inc., Mississauga, ON, Canada; model obtained from the incubation of adefovir dipi- 6500) using the Ready SafeTM scintillation fluid voxil in enterocyte lysate were fitted to a set of (Beckman Coulter, Canada). The same volume of differential equations to reveal km and km{mi}, the initial concentration, the timed-sample and the rate constants for sequential hydrolysis of blank buffer was used for liquid scintillation bis(POM)-PMEA (or adefovir dipivoxil) to counting to ensure similar quenching. The sacs mono(POM)-PMEA, then PMEA, with the pro- were solubilized with 1 N NaOH (5 ml) for 2 h at gram Scientists (Micromath, St Louis, MO). 371C, as described by Lafforgue et al. [22]. Known counts of these compounds were added to blank Assays sac lysate to ascertain the degree of quenching; For radiolabeled mannitol, glycylsarcosine and none was found. The amount in each compart- digoxin, the contents (dpm) recovered in the ment was estimated by the product of the volume sample in the donor and receiver compartments, and the concentration. The total recovery (%) for before and after the experiment, were assayed by the each drug was calculated as below: liquid scintillation counting (Beckman Counter Amount recovery ð%Þ model 6500, Beckman Coulter, Canada, Missis- sum of amounts in donor; sac; and receiver sides sauga, ON). The amount of compound trans- ¼ dose ported into the receiver side was quantified in 100% ð1Þ terms of its specific activity and expressed as nmol of drug/g intestine. The sac protein concentration was assayed by the Lowry method [25].

Effect of 1,25(OH)2D3 on CDF-DA and adefovir HPLC for adefovir dipivoxil, mono(POM)- divipoxil hydrolysis PMEA and PMEA or adefovir Since metabolite efflux and not parent drug An aliquot (0.1 ml) of the samples obtained from efflux was used to appraise transporter function the transport study was counted directly, after dosing of the precursors, CDF-DA and ade- whereas another aliquot was centrifuged, and fovir divipoxil, the effect of 1,25(OH)2D3 treat- 0.1 ml was injected directly onto the HPLC. The ment on CDF-DA (10 mM) hydrolysis to CDF and mono(POM)-PMEA and PMEA or adefovir were adefovir divipoxil (1 mM) hydrolysis to adefovir in identified according to their radioelution times in rat intestinal tissue was first investigated before the HPLC system published earlier by Annaert an interpretation was made of Mrp2 function, et al. [26,27]. The retention times of adefovir

271 116 H.-J. MAENG ET AL. dipivoxil and its metabolites, mono(POM)- CDF-DA to CDF by base was complete. The PMEA and PMEA or adefovir were around 28, calibration curve for CDF was constructed over 19 and 9 min, respectively. Briefly, the com- the concentration range 0.1–10 mM from CDF-DA, pounds were separated using a modified reverse- and was found to be linear (R2 5 0.994). phase HPLC method and each species was quantified by radioelution. The Shimadzu Drug permeability analysis HPLC system (LC 10AT pump, SIL-10A auto- From the everted rat intestinal sac data, the injector, and SCL-10A system controller) with apparent permeability coefficient (Papp,cm/s) an Altex 10 mM C-18 reverse phase column was calculated according to the following equation: (4.6 mm 250 mm), was used to achieve separation DA =D of adefovir dipivoxil and its metabolites. Mobile ¼ R t ð Þ Papp Area C 2 phase A consisted of 10 mM potassium dihydrogen 60 0 D phosphate with 2 mM tetrabutylammonium hydro- where AR is the amount accumulated in the gen sulfate, adjusted to pH 6 with 1 N NaOH; receiver side during the time interval, Dt. Area is mobile phase B was 100% acetonitrile. A linear the surface area of the mucosal membrane (cm2), m gradient program was performed from 5% to 38% and C0 is the initial drug concentration ( M) acetonitrile over 10 min, maintained at the constant placed into the donor side. The surface area of proportion of 38% over 20.5 min, followed by the mucosa was calculated by correlating the return to the initial condition over the next amount of sac protein to the surface area (1 mg 0.5 min and re-equilibration of the column over protein 5 9.955 mm2), as described by Barthe et al. thenext9.5min.TheHPLCfractionalrecoveryof [19]. The efflux ratio, EfR is given by the ratio of each species was multiplied by the direct count to Papp in the (B-to-A) to that in the (A-to-B) yield the corrected dpm of the species. direction. Papp;BtoA EfR ¼ ð3Þ Fluorescence assay for CDF Papp;AtoB The stability of CDF-DA in buffer was first investigated in a preliminary study. A minor and Statistical analysis negligible degradation (3%, estimated as the pro- Data were expressed as mean7standard devia- duct, CDF divided by the initial material in CDF tion. The difference between and among means equivalent from base hydrolysis) was observed at was analysed using the two-tailed Student’s t-test 1 37 Cafter30minincubationwithbufferI.For or ANOVA, respectively. A value of po0.05 was prevention of sample degradation, mucosal (apical) considered to be significant. samples taken at each time point were immediately stored at –201C. Upon completion of sampling, the Results collected samples were thawed over ice, mixed and 100 ml of sample was transferred into the fluores- Effect of 1,25(OH) D on mannitol transport cence reader plate and processed. The fluorescence 2 3 of CDF in the receiver side (mucosal fluid) was Mannitol, a probe of paracellular transport, was first measured by a microplate fluorescence reader at a studied to establish the effects of 1,25(OH)2D3 on l l wavelength of ex nm and em nm (SpectraMax passive transport in the everted rat intestinal sac Gemini XS; Molecular Devices, Sunnyvale, CA) preparation (Figure 1). The amount (% dose) and according to the method of Tian et al.[28].A volume recovery (%) at the end of incubation were calibration curve was constructed to assay for close to 100% (data not shown). The transport of fluorescence of CDF formed from CDF-DA. An mannitol in the A-to-B direction for 20 min or B-to-A aliquot (100 ml) of 10 N NaOH was added to each direction over 30 min in everted sacs of S1, S2 and S8 CDF-DA standard solution in a final volume of segments did not change with 1,25(OH)2D3 treat- 5 ml. The maximum fluorescence value was ment, evidenced by the virtually identical Papp,A-to-B reached rapidly within 1 min of incubation at and Papp,B-to-A for mannitol (values between 37 1C with base, suggesting that hydrolysis of 1.23 10 6 and 2.89 10 6 cm/s) (Figure 1).

272 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT INTESTINAL TRANSPORTERS 117

m Figure 1. Effects of 1,25(OH)2D3 vs. vehicle treatment (2.56 nmol/kg daily for 4 days) on the permeability of mannitol (100 M)in rat intestinal sacs (prepared from S1, S2 and S8; n 5 3 or 4 sacs). (A) and (B) denote the A-to-B and B-to-A transport of mannitol, respectively, in everted sacs prepared from the duodenum, S1 (left), jejunum, S2 (middle) and ileum, S8 (right). In (B), each data point for each time point was the mean of 3 or 4 sacs

5 7 Table 1. The apparent permeabilities (Papps) of the compounds defined by the everted rat intestinal sac technique (n 3or4, SD) Drug Treatment Segment Apparent permeability (10 6 cm/s) Efflux ratio EfR

Papp,A-to-B Papp,B-to-A Mannitol Control S1 1.2370.11 1.3870.50 1.12 7 7 1,25(OH)2D3 1.52 0.29 1.41 0.43 0.93 Control S2 2.8971.86 2.5071.05 0.87 7 7 1,25(OH)2D3 2.34 1.10 2.53 2.03 1.08 Control S8 1.4970.45 1.7370.49 1.17 7 7 1,25(OH)2D3 1.72 0.68 1.59 0.31 0.93 Glycylsarcosine (GlySar) Control S1 3.4170.86 0.8870.50 0.26 7 Ã 7 1,25(OH)2D3 5.54 1.04 0.85 0.57 0.15 Adefovir from precursor, adefovir dipivoxil Control S2 0.8170.28 0.5970.33 0.74a 7 Ã 7 Ã a 1,25(OH)2D3 1.68 0.39 1.45 0.47 0.86 CDF from precursor, CDF-DA Control S2 0.3670.03 2.1670.49 5.94a 7 7 Ã a 1,25(OH)2D3 0.42 0.22 4.33 1.53 10.2 Digoxin Control S8 0.5570.16 1.7471.23 3.15 7 7 1,25(OH)2D3 0.60 0.07 1.41 1.19 2.36 Ã po0.05 compared to the control group aApparent EfR for metabolite

As a result, the calculated EfR for the everted sacs tight junction between the enterocytes in the from S1, S2 and S8 were near unity for the control everted sac. and 1,25(OH)2D3-treated groups (Table 2). The Papp in the S2 segment (proximal jejunum) was Effects of 1,25(OH)2D3 on active absorptive slightly but insignificantly higher compared with transporters: PepT1 those for the S1 and S8 segments (Table 1, one- way ANOVA). Taken together, the data suggest PepT1. The linearity of transport with time in m that 1,25(OH)2D3 was devoid of influence on the the A-to-B direction for the transport of 10 M

273 118 H.-J. MAENG ET AL.

GlySar, a model substrate of PepT1, was By contrast, the B-to-A transport of GlySar verified using everted sacs from the duodenum was unchanged (Figure 2B). The Papp value (S1), and recovery of GlySar in the system for the A-to-B direction increased in the 1,25 was complete. Figure 2A shows that the absorp- (OH)2D3-treated group compared with the con- tion of GlySar was linear up to 30 min trol group (5.5471.04 10 6 vs. 3.4170.856 2 5 6 o (R 0.995) and the 20 min incubation time was 10 vs. cm/s, p 0.05), whereas the value of Papp chosen for subsequent A-to-B transport studies. for the B-to-A direction, which was lower in A significant increase in transport (48%) was comparison to values for the Papps for the A-to-B 7 6 observed with the 1,25(OH)2D3 treatment direction, was unchanged (0.848 0.569 10 compared with controls (po0.05) (Figure 2B). vs. 0.87670.502 10 6 cm/s), resulting in a

Table 2. Comparison of changes induced by 1,25(OH)2D3 between functional activities and published protein expression of transporters

a Transporter Segment Change of expression levels induced by 1,25(OH)2D3 Changes in functional activity in everted intestinal sac induced by 1,25(OH)2D3 mRNA Protein PepT1 S1 2b mc m(A to B) Mrp2 S2 2m m(B to A) Mrp4 S2 2m m(A to B) and (B to A) P-gp S8 222(B to A) aData from [7] bUnchanged cInduced

Figure 2. Effects of 1,25(OH)2D3 vs. vehicle treatment (2.56 nmol/kg daily for 4 days) on the function of PepT1 in the rat duodenum (S1), appraised via the absorption of glycylsarcosine or GlySar (10 mM) by the everted sac. (A) Rates of glycylsarcosine absorbed vs. time in the A-to-B direction; each time point represents the mean of 3 or 4 preparations (left); Glycylsarcosine transport at 20 min for the A-to-B direction (right); and (B) transport in the B-to-A direction (right) with each time point Ã o representing the mean of 3 or 4 preparations for 1,25(OH)2D3- and vehicle-treated rats. Statistically higher absorption ( p 0.05) was observed for the treatment group compared with the vehicle control

274 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT INTESTINAL TRANSPORTERS 119 slightly lower EfR (Table 1). The data suggests the presence of an absorptive transporter that is induced by 1,25(OH)2D3.

Effects of 1,25(OH)2D3 on active efflux transporter: Mrp2, Mrp4 and P-gp

Mrp2. Similar to the base-catalysed reaction, hydrolysis of CDF-DA in rat jejunal enterocyte lysate (2 mg protein) occurred rapidly and instantaneously. The maximum rate of hydrolysis was reached by the first sampling time (10 min), and the metabolic patterns were identical between Figure 3. Effects of 1,25(OH) D vs. vehicle treatment the control and 1,25(OH)2D3-treated group 2 3 (p40.05). By contrast, CDF-DA was stable in (2.56 nmol/kg daily for 4 days) on the function of Mrp2 in the rat jejunum (S2), appraised via the apical efflux of CDF buffer I over the 30 min of study, and recovery of formed from CDF-DA (50 mM) which was administered into mass in the system for the control and treated the everted sac (each time point represents the mean of n 5 7 Ã groups exceeded 75% and 85%, respectively or 8 sacs). Statistically higher efflux ( po0.05), was observed (p40.05), pointing to the soundness of the trans- for the treatment group compared with the vehicle control port data to denote changes in transporter function. After the administration of 50 mM CDF-DA into the everted sac prepared from the proximal lysate protein), whereas hydrolysis of the mono jejunum (S2), the appearance of the fluorescent (POM)-PMEA to PMEA was considerably slower 7 7 1 CDF, formed intracellularly from esterases and (km{mi}: 0.049 0.008 vs. 0.045 0.007 min per 4 probe for Mrp2, in the receiver or serosal mg lysate protein, p 0.05); the rate constant for compartment for A-to-B transport was unchanged PMEA formation was unaltered by 1,25(OH)2D3 4 treatment. with 1,25(OH)2D3-treatment (p 0.05, apparent 6 Upon incubation of 1 mM adefovir dipivoxil Papp of CDF was 0.36 vs. 0.42 10 cm/s and lower than that for mannitol Table 1). The with the everted jejunal sac for A-to-B transport, mucosal appearance of the fluorescent CDF for adefovir was found to be the only species B-to-A transport was significantly higher after detected inside the sac (serosal compartment) at 20 min. At 30 min incubation for B-to-A trans- 1,25(OH)2D3 treatment compared with the controls beyond 25 min of incubation (po0.05, ANOVA) port, both mono(POM)-PMEA and adefovir were (Figure 3). As a result, the calculated apparent EfR detected, though with only a negligible amount as mono(POM)-PMEA (o1% total amount in the of CDF was higher in the 1,25(OH)2D3-treated group vs. the control (10.2 vs. 5.94, Table 1). receiver side, data not shown). Since efflux of adefovir into both apical and basolateral direc- Mrp4. Since adefovir dipivoxil [bis(POM)- tions has been reported for Caco-2 cells [26,27], PMEA] is a di-ester prodrug of adefovir, the everted jejunal sac experiments with 1 mM adefo- intended Mrp4 substrate [29], the presence of vir dipivoxil were conducted for both the A-to-B mono(POM)-PMEA and adefovir, metabolites and B-to-A directions after 1,25(OH)2D3 and found in Caco-2 cells and rat Ussing chamber vehicle treatment to the rat. The time-linearity [26,27,30], hydrolysis of adefovir divipoxil was was first investigated with 1 mM adefovir dipi- first investigated between the control and voxil for A-to-B transport in the jejunal sac. 1,25(OH)2D3-treated groups with the lysate pre- The A-to-B appearance rate of adefovir was linear pared from rat jejunal enterocytes. Adefovir up to 30 min (R2 5 0.973, Figure 4A), and the divipoxil was converted to mono(POM)-PMEA 20 min incubation time was chosen in subsequent rapidly in rat enterocyte lysate at rates that were transport studies, wherein good recovery (470%) similar for both the control and treated groups was observed in the transport studies. The 5 7 7 1 (km 0.617 0.091 vs. 0.622 0.04 min per mg results showed that 1,25(OH)2D3 treatment led to

275 120 H.-J. MAENG ET AL.

Figure 4. Effects of 1,25(OH)2D3 vs. vehicle treatment (2.56 nmol/kg daily for 4 days) on the function Mrp4 in the rat jejunum (S2), appraised via apical and basolateral efflux of adefovir by the everted sac upon administration of the precursor, adefovir dipivoxil (1 mM), into the apical or basolateral compartments (each point is the mean of 3 or 4 sacs). (A) Rates of adefovir efflux vs. time in the A-to-B direction; each time point represents the mean of 3 or 4 preparations (left), and adefovir efflux at 20 min for the A-to-B direction (right); (B) transport in the B-to-A direction with each time point representing the mean of 3 or 4 preparations, for 5 Ã o 1,25(OH)2D3- and vehicle-treated rats (n 3 or 4 sacs). Statistically higher basolateral and apical efflux ( p 0.05) was observed for the treatment group compared with vehicle control group significantly higher 80% (po0.05) efflux of adefo- (Figure 5A), and there was no difference in vir basolaterally into the serosal side, likely due to transport between the treated and non-treated enhanced Mrp4 function in the proximal jejunum groups (Figure 5B). Recovery exceeded 83% dose. (Figure 4B). When the B-to-A transport was In addition, the B-to-A efflux of 10 mM digoxin by examined with 1 mM adefovir dipivoxil, the rate P-glycoprotein into the mucosal side over 30 min of efflux of adefovir into the mucosal side apically remained unchanged with 1,25(OH)2D3 treat- was also increased by 1,25(OH)2D3 treatment ment (Figure 5B). The Papp,B-to-A values were o (p 0.05) (Figure 4B). The apparent Papp value similar for the control- and 1,25(OH)2D3 treated for adefovir at 20 min for A-to-B direction was sacs (1.4171.19 10 6 vs. 1.7471.23 10 6 cm/s; increased significantly in 1,25(OH)2D3-treated Table 1) and there was no change in the EfR sacs compared with vehicle-treated sacs (3.15 vs. 2.36, Table 1), suggesting that the (1.6870.391 10 6 vs. 0.80670.284 10 6 cm/s, functional activity of P-gp had remained un- o p 0.05), and the apparent Papp value for adefovir changed with VDR treatment. The Papp for for the B-to-A direction was also enhanced with digoxin in the A-to-B direction was lower than 7 6 1,25(OH)2D3 treatment (1.45 0.471 10 vs. that of mannitol in the rat everted ileal sac, as 0.59370.327 10 6 cm/s, po0.05) (Table 1). previously observed by Lacombe et al. [20]. However, Papp in the B-to-A direction was similar P-gp. When the time-linearity of A-to-B transport to that of mannitol. These data conform to the with 10 mM digoxin was investigated in the rat notion of an apical transporter, P-gp, for the everted ileal sac, the net rate of A-to-B transport secretion of digoxin, and the transporter is 2 5 of digoxin was linear up to 30 min (R 0.969) unperturbed by 1,25(OH)2D3 treatment.

276 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT INTESTINAL TRANSPORTERS 121

Figure 5. Lack of effect of 1,25(OH)2D3 vs. vehicle treatment (2.56 nmol/kg daily for 4 days) on the function of P-gp in the rat ileum (S8), appraised via the efflux of digoxin (10 mM) in the everted rat sac in the B-to-A direction (n 5 5 sacs in each point). (A) Net rates of digoxin absorbed vs. time in the A-to-B direction; each time point represents the mean of 3 or 4 preparations (left) and digoxin net absorption at 20 min for the A-to-B direction (right); (B) digoxin efflux in the B-to-A direction, with each time 5 point representing the mean of 3 or 4 preparations, for 1,25(OH)2D3- and vehicle-treated rats (n 3 or 4 sacs)

Discussion junctions in Caco-2 cell monolayers and murine intestine due to an increase of the tight junction Drug absorption into the systemic circulation is proteins, zonula occluding-1 (ZO-1), claudin-1, controlled by intestinal transporters which are claudin-2, claudin-12 as well as adherens junction localized at the apical and/or basolateral mem- proteinE-cadherin[36,37],wasfirstexamined. brane to mediate drug absorption or secretion There was no effect of 1,25(OH)2D3 treatment on [31,32]. The presence of apical absorptive and the passive diffusion of paracellular transport basolateral efflux transporters could increase oral marker, mannitol in both the A-to-B and B-to-A bioavailability, whereas apical secretory transpor- directions among the S1, S2 and S8 segments ters, P-gp, MRP2 and BCRP, would decrease oral (Figure 1), thus alleviating the concern. Thus, it is bioavailability. However, induction of intestinal unlikely that alteration of membrane integrity is transporters has been well recognized and docu- responsible for the observed functional changes of mented on multiple occasions in studies in vivo transporters with the probe drugs (Figures 2–4). [33–35]. With respect to induction of transporters Our observation, which reports a lack of effect of by the VDR, significantly higher protein levels of 1,25(OH)2D3 on mannitol transport by the para- PepT1, Mrp2, Mrp4 and Asbt have been observed cellular route in the rat small intestine, is consistent to occur in the small intestine upon 1,25(OH)2D3 with the lack of effect of 100 nM of 1,25(OH)2D3 on treatment in the rat in vivo [6,7]. mannitol transport observed in Caco-2 cells after a The present, follow-up intestinal sac study was 3 day-treatment regime [9]. performed to investigate the corresponding func- Changes in functional activity of transporters tional changes of rat intestinal transporters with using the probe drugs in 1,25(OH)2D3-treated 1,25(OH)2D3 treatment. First, the persistent con- rats were then correlated to changes in expres- cern of activation of VDR on enhancement of tight sion of mRNA and protein from our previous

277 122 H.-J. MAENG ET AL. published data [6,7] (Table 2). A good correspon- MRP4 is expressed on both the basolateral and dence between changes in transporter function apical membranes of epithelial cells in the rat and protein expression with 1,25(OH)2D3 treat- intestine [51]. In addition, bidirectional efflux ment was observed. Namely, functions of the (i.e. both apical and basolateral efflux from the absorptive transporter, PepT1, and the efflux Caco-2 cells) of adefovir, with predominant efflux transporters, Mrp2 and Mrp4, were increased in into the apical side, was observed [26,27]. The 1,25(OH)2D3-treated rats compared with control same was found in rat jejunal membrane mounted rats, as expected (Figures 2–4). These data, in the Ussing chamber [30]. With Mrp4/MRP4 documenting the effect of 1,25(OH)2D3 on being expressed in both apical and basolateral absorptive and efflux transporters in small membranes of the small intestine [49–51], in- intestine, is informative in predicting potential creased efflux of adefovir due to 1,25(OH)2D3 drug–drug interactions (DDIs). For example, induction of Mrp4/MRP4 would drastically re- enhanced PepT1 function by VDR activation duce the accumulation of adefovir in the enter- would effectively promote a higher absorption ocyte. Although Mrp3 levels were found induced of di- and tripeptides drugs such as cloxacillin in S1 and S2 segments of 1,25(OH)2D3 treated rats [38], ceftibuten [39] and valacyclovir [40], [7], adefovir is not a substrate for Mrp3/MRP3 whereas induced Mrp2 efflux function will likely [29]. By contrast, Mrp5 is capable of transporting lower the absorption for organic anion drugs adefovir [52], but its expression in rat small such as statins (e.g. pravastatin and cerivastatin) intestine is very low [53]. Although Bcrp is also [41] or ACE inhibitors (e.g. enalapril and lisino- recognized as an apical efflux transporter of pril) [42], HIV protease inhibitors (saquinavir, adefovir [54], protein levels of Bcrp are found to ritonavir and indinavir) [43] and anticancer be unchanged by the VDR (unpublished data), drugs (etoposide and vincristine) [44] when used rendering the observed increase in adefovir in conjunction with calcitriol or other VDR transport in B-to-A direction unlikely to be due ligands. However, there was no change for to Bcrp. Taken together, the induced efflux of intestinal P-gp functional activity (Figure 5), adefovir in the rat everted intestinal sac with studied with digoxin as the test probe [45]. 1,25(OH)2D3 treatment may be mediated primar- Digoxin, a substrate of Oatp1a4 [46,47], is likely ily by increased expression of Mrp4. to enter the enterocyte by passive diffusion due In summary, using the everted intestinal sac to the absence/low levels of Oatp1a4 in the rat technique, we demonstrated that 1,25(OH)2D3 ileum [48], and the observed secretion in the exerted an inductive effect on drug absorptive everted ileal sac is likely to be influenced mostly and efflux transporters in the rat small intestine by P-gp function. These observations on the lack and was devoid of effect on the diffusion of of change in P-gp levels in the rat small intestine mannitol (Figure 1). It was verified that changes demonstrate a species difference between rat in transporter functional activity paralleled the enterocytes and Caco-2 cells to 1,25(OH)2D3 induction of transporter protein levels in treatment [9]. 1,25(OH)2D3-treated rats (Table 2) [6,7]. Induction Interestingly, the transport of adefovir in the of PepT1, Mrp2 and Mrp4 in the small intestine everted rat intestinal sac in both A-to-B and by 1,25(OH)2D3 via the VDR (Figures 2–4) could B-to-A directions was significantly enhanced by affect drug absorption and potentially lead to 1,25(OH)2D3 treatment (Figure 4), results which DDIs. Since vitamin D analogs are used clinically are consistent with the overall increase in protein in combination with other drugs [55], the effect of expression observed by Chow et al. [7] (Table 2). vitamin D analogs on drug absorption and DDIs The localization of Mrp4/MRP4 has been found requires further investigation in man in vivo. on both the apical and/or basolateral membranes [49–51]. Immunofluorescence imaging had recently demonstrated that MRP4 is localized Acknowledgments more on the basolateral membrane than the apical membrane in Caco-2 cells [29], and immunohis- This work was supported by the Canadian tochemical staining further revealed that Mrp4/ Institutes for Health Research, MOP89850.

278 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT INTESTINAL TRANSPORTERS 123

Dr Han-Joo Maeng is a recipient of the Govern- effects of VDR versus PXR, FXR and GR ligands ment of Canada Postdoctoral Research Fellow- on the regulation of CYP3A isozymes in rat and ship (PDRF) and a fellowship from the National human intestine and liver. Eur J Pharm Sci 2009; 12: 115–125. Research Foundation of Korea, funded by the 11. Xu Y, Iwanaga K, Zhou C, Cheesman MJ, Farin F, South Korean Government (NRF-2009-352- Thummel KE. Selective induction of intestinal a E0068). Edwin C. Y. Chow is a recipient of the CYP3A23 by 1 ,25-dihydroxyvitamin D3 in rats. Alexander Graham Bell NSERC fellowship, Biochem Pharmacol 2006; 28: 385–392. Canada, and Matthew Durk is the recipient of a 12. Echchgadda I, Song CS, Roy AK, Chatterjee B. fellowship from the Strategic Training Grant on Dehydroepiandrosterone sulfotransferase is a target for transcriptional induction by the vitamin Biologic Therapeutics from the Canadian Insti- D receptor. Mol Pharmacol 2004; 65: 720–729. tute for Health Research, CIHR. 13. Tachibana S, Yoshinari K, Chikada T, Toriyabe T, Nagata K, Yamazoe Y. Involvement of vitamin D receptor in the intestinal induction of human ABCB1. Drug Metab Dispos 2009; 37: 1604–1610. References 14. Saeki M, Kurose K, Tohkin M, Hasegawa R. Identification of the functional vitamin D response elements in the human MDR1 gene. 1. Urquhart BL, Tirona RG, Kim RB. Nuclear Biochem Pharmacol 2008; 15: 531–542. receptors and the regulation of drug-metabolizing 15. McCarthy TC, Li X, Sinal CJ. Vitamin D receptor- enzymes and drug transporters: implications for dependent regulation of colon multidrug interindividual variability in response to drugs. resistance-associated protein 3 gene expression J Clin Pharmacol 2007; 47: 566–578. by bile acids. J Biol Chem 2005; 280: 23232–23242. 2. DeLuca HF, Zierold C. Mechanisms and functions 16. Chen X, Chen F, Liu S, et al. Transactivation of rat of vitamin D. Nutr Rev 1998; 56: S4–S10. apical sodium-dependent bile acid transporter 3. Zierold C, Mings JA, Deluca HF. 19Nor-1,25- and increased bile acid transport by 1a,25- dihydroxyvitamin D2 specifically induces CYP3A9 in rat intestine more strongly than 1,25- dihydroxyvitamin D3 via the vitamin D receptor. dihydroxyvitamin D in vivo and in vitro. Mol Mol Pharmacol 2006; 69: 1913–1923. 3 17. Nishida S, Ozeki J, Makishima M. Modulation of Pharmacol 2006; 69: 1740–1747. a 4. Salusky IB. Are new vitamin D analogues in renal bile acid metabolism by 1 -hydroxyvitamin D3 bone disease superior to calcitriol? Pediatr Nephrol administration in mice. Drug Metab Dispos 2009; 2005; 20: 393–398. 37: 2037–2044. 5. Makishima M, Lu TT, Xie W, et al. Vitamin D 18. Wilson TH, Wiseman G. The use of sacs of everted receptor as an intestinal bile acid sensor. Science small intestine for the study of the transference of 2002; 17: 1313–1316. substances from the mucosal to the serosal 6. Chow EC, Maeng HJ, Liu S, Khan AA, surface. J Physiol 1954; 123: 116–125. Groothuis GM, Pang KS. 1a,25-Dihydroxyvitamin 19. Barthe L, Woodley JF, Kenworthy S, Houin G. An improved everted gut sac as a simple and D3 triggered vitamin D receptor and farnesoid X receptor-like effects in rat intestine and accurate technique to measure paracellular trans- liver in vivo. Biopharm Drug Dispos 2009; 30: port across the small intestine. Eur J Drug Metab 457–475. Pharmacokinet 1998; 23: 313–323. 7. Chow EC, Sun H, Khan AA, Groothuis GM, 20. Lacombe O, Woodley J, Solleux C, Delbos JM, a Pang KS. Effects of 1 ,25-dihydroxyvitamin D3 on Boursier-Neyret C, Houin G. Localisation of drug transporters and enzymes of the rat intestine and permeability along the rat small intestine, using kidney in vivo. Biopharm Drug Dispos 2010; 31: markers of the paracellular, transcellular and 91–108. some transporter routes. Eur J Pharm Sci 2004; 8. Thummel KE, Brimer C, Yasuda K, et al. Tran- 23: 385–391. scriptional control of intestinal cytochrome P- 21. Barthe L, Woodley J, Houin G. Gastrointestinal a 4503A by 1 ,25-dihydroxy vitamin D3. Mol absorption of drugs: methods and studies. Pharmacol 2001; 60: 1399–1406. Fundam Clin Pharmacol 1999; 13: 154–168. 9. Fan J, Liu S, Du Y, Morrison J, Shipman R, 22. Lafforgue G, Arellano C, Vachoux C, et al. Oral Pang KS. Up-regulation of transporters and absorption of ampicillin: role of paracellular route enzymes by the vitamin D receptor ligands, vs. PepT1 transporter. Fundam Clin Pharmacol a 1 ,25-dihydroxyvitamin D3 and vitamin D ana- 2008; 22: 189–201. logs, in the Caco-2 cell monolayer. J Pharmacol Exp 23. Liu S, Tam D, Chen X, Pang KS. P-glycoprotein Ther 2009; 330: 389–402. and an unstirred water layer barring digoxin 10. Khan AA, Chow EC, van Loenen-Weemaes AM, absorption in the vascularly perfused rat small Porte RJ, Pang KS, Groothuis GM. Comparison of intestine preparation: induction studies with

279 124 H.-J. MAENG ET AL.

pregnenolone-16a-carbonitrile. Drug Metab Dispos 37. Fujita H, Sugimoto K, Inatomi S, et al. Tight 2006; 34: 1468–1479. junction proteins claudin-2 and -12 are critical for 1 24. Veau C, Leroy C, Banide H, et al. Effect of chronic vitamin D-dependent Ca2 absorption between renal failure on the expression and function of enterocytes. Mol Biol Cell 2008; 19: 1912–1921. rat intestinal P-glycoprotein in drug excretion. 38. Luckner P, Brandsch M. Interaction of 31 b-lactam Nephrol Dial Transplant 2001; 16: 1607–1614. antibiotics with the H1/peptide symporter PEPT2: 25. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. analysis of affinity constants and comparison with Protein measurement with the Folin phenol PEPT1. Eur J Pharm Biopharm 2005; 59: 17–24. reagent. J Biol Chem 1951; 193: 265–275. 39. Bretschneider B, Brandsch M, Neubert R. Intestinal 26. Annaert P, Kinget R, Naesens L, de Clercq E, transport of beta-lactam antibiotics: analysis of the Augustijns P. Transport, uptake, and metabolism affinity at the H1/peptide symporter (PEPT1), the of the bis(pivaloyloxymethyl)-ester prodrug of uptake into Caco-2 cell monolayers and the trans- 9-(2-phosphonyl methoxyethyl)adenine in an in epithelial flux. Pharm Res 1999; 16: 55–61. vitro cell culture system of the intestinal mucosa 40. Ganapathy ME, Huang W, Wang H, Ganapathy V, (Caco-2). Pharm Res 1997; 14: 492–496. Leibach FH. Valacyclovir: a substrate for the 27. Annaert P, Van Gelder J, Naesens L, et al. Carrier intestinal and renal peptide transporters PEPT1 mechanisms involved in the transepithelial trans- and PEPT2. Biochem Biophys Res Commun 1998; port of bis(POM)-PMEA and its metabolites 246: 470–475. across Caco-2 monolayers. Pharm Res 1998; 15: 41. Itagaki S, Chiba M, Kobayashi M, Hirano T, 1168–1173. Iseki K. Contribution of multidrug resistance- 28. Tian X, Zamek-Gliszczynski MJ, Zhang P, associated protein 2 to secretory intestinal trans- Brouwer KL. Modulation of multidrug port of organic anions. Biol Pharm Bull 2008; 31: resistance-associated protein 2 (Mrp2) and Mrp3 146–148. expression and function with small interfering 42. Liu L, Cui Y, Chung AY, et al. Vectorial transport RNA in sandwich-cultured rat hepatocytes. Mol of enalapril by Oatp1a1/Mrp2 and OATP1B1 and Pharmacol 2004; 66: 1004–1010. OATP1B3/MRP2 in rat and human livers. 29. Ming X, Thakker DR. Role of basolateral efflux J Pharmacol Exp Ther 2006; 318: 395–402. transporter MRP4 in the intestinal absorption of 43. Huisman MT, Smit JW, Crommentuyn KM, et al. the antiviral drug adefovir dipivoxil. Biochem Multidrug resistance protein 2 (MRP2) transports Pharmacol 2010; 79: 455–462. HIV protease inhibitors, and transport can be 30. Annaert P, Tukker JJ, van Gelder J, et al. In vitro, enhanced by other drugs. AIDS 2002; 16: 2295–2301. ex vivo, and in situ intestinal absorption character- 44. Chen ZS, Kawabe T, Ono M, et al. Effect of istics of the antiviral ester prodrug adefovir multidrug resistance-reversing agents on trans- dipivoxil. J Pharm Sci 2000; 89: 1054–1062. porting activity of human canalicular multi- 31. Suzuki H, Sugiyama Y. Role of metabolic enzymes specific organic anion transporter. Mol Pharmacol and efflux transporters in the absorption of drugs 1999; 56: 1219–1228. from the small intestine. Eur J Pharm Sci 2000; 12: 45. Stephens RH, O’Neill CA, Bennett J, et al. 3–12. Resolution of P-glycoprotein and non-P-glycopro- 32. Shitara Y, Horie T, Sugiyama Y. Transporters as a tein effects on drug permeability using intestinal determinant of drug clearance and tissue distri- tissues from mdr1a (-/-) mice. Br J Pharmacol 2002; bution. Eur J Pharm Sci 2006; 27: 425–446. 135: 2038–2046. 33. Adibi SA. Regulation of expression of the 46. Sugiyama D, Kusuhara H, Shitara Y, Abe T, intestinal oligopeptide transporter (Pept-1) in Sugiyama Y. Effect of 17 beta-estradiol-D-17 beta- health and disease. Am J Physiol Gastrointest Liver glucuronide on the rat organic anion transporting Physiol 2003; 285: G779–G788. polypeptide 2-mediated transport differs depending 34. Ghanem CI, Go´mez PC, Arana MC et al. Induc- on substrates. Drug Metab Dispos 2002; 30: 220–223. tion of rat intestinal P-glycoprotein by spirono- 47. Kodawara T, Masuda S, Wakasugi H, et al. lactone and its effect on absorption of orally Organic anion transporter oatp2-mediated inter- administered digoxin. J Pharmacol Exp Ther 2006; action between digoxin and amiodarone in the rat 318: 1146–1152. liver. Pharm Res 2002; 19: 738–743. 35. Iizasa H, Genda N, Kitano T, et al. Altered 48. Cattori V, Van Montfoort J, Stieger B, et al. expression and function of P-glycoprotein in Localization of Oatp4 in rat liver and comparison dextran sodium sulfate-induced colitis in mice. of its substrate specificity with Oatp1, Oatp2 and J Pharm Sci 2003; 92: 569–576. Oatp3. Eur J Physiol 2001; 443: 188–195. 36. Kong J, Zhang Z, Musch MW, et al. Novel role of 49. Leggas M, Adachi M, Scheffer GL, et al. Mrp4 the vitamin D receptor in maintaining the integrity confers resistance to topotecan and protects the of the intestinal mucosal barrier. Am J Physiol brain from chemotherapy. Mol Cell Biol 2004; 24: Gastrointest Liver Physiol 2008; 294: G208–G216. 7612–7621.

280 a 1 ,25-DIHYDROXYVITAMIN D3 ON RAT INTESTINAL TRANSPORTERS 125

50. Kruijtzer CM, Beijnen JH, Schellens JH. Improve- by microglia. J Pharmacol Exp Ther 2004; 309: ment of oral drug treatment by temporary 1221–1229. inhibition of drug transporters and/or cyto- 53. Maher JM, Cherrington NJ, Slitt AL, Klaassen CD. chrome P450 in the gastrointestinal tract and Tissue distribution and induction of the rat liver: an overview. Oncologist 2002; 7: 516–530. multidrug resistance-associated proteins 5 and 6. Life Sci 2006; 78: 2219–2225. 51. Johnson BM, Zhang P, Schuetz JD, Brouwer KL. 54. Takenaka K, Morgan JA, Scheffer GL, et al.Substrate Characterization of transport protein expression in overlap between Mrp4 and Abcg2/Bcrp affects multidrug resistance-associated protein (Mrp) 2- purine analogue drug cytotoxicity and tissue dis- deficient rats. Drug Metab Dispos 2006; 34: 556–562. tribution. Cancer Res 2007; 67: 6965–6972. 52. Dallas S, Schlichter L, Bendayan R. Multidrug 55. Masuda S, Jones G. Promise of vitamin D resistance protein (MRP) 4- and MRP 5-mediated analogues in the treatment of hyperproliferative efflux of 9-(2-phosphonylmethoxyethyl) adenine conditions. Mol Cancer Ther 2006; 5: 797–808.

281

APPENDIX A6

Chow EC, Sondervan M, Jin C, Groothuis GM and Pang KS (2011) Comparative effects of doxercalciferol (1α-hydroxyvitamin D2) versus calcitriol (1α,25-dihydroxyvitamin D3) on the expression of transporters and enzymes in the rat in vivo. J Pharm Sci 100:1594-1604

282

PHARMACOKINETICS, PHARMACODYNAMICS AND DRUG METABOLISM

Comparative Effects of Doxercalciferol (1α-Hydroxyvitamin D2) Versus Calcitriol (1α,25-Dihydroxyvitamin D3) on the Expression of Transporters and Enzymes in the Rat In Vivo

EDWIN C.Y. CHOW,1 MYRTE SONDERVAN,2 CHENG JIN,1 GENY M.M. GROOTHUIS,2 K. SANDY PANG1 1Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, University of Toronto, Toronto, Ontario, Canada

2Pharmacokinetics, Toxicology and Targeting, Department of Pharmacy, University of Groningen, Groningen, the Netherlands

Received 3 June 2010; revised 29 July 2010; accepted 31 August 2010 Published online 21 October 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.22366

ABSTRACT: Effects of 1.28 nmol/kg doxercalciferol [1"(OH)D2], a synthetic vitamin D2 analog that undergoes metabolic activation to 1",25-dihydroxyvitamin D2, the naturally occurring, biologically active form of vitamin D2, on rat transporters and enzymes were compared with those of 1",25-dihydroxyvitamin D3 [1,25(OH)2D3, active form of vitamin D3; 4.8 and 6.4 nmol/kg] given on alternate days intraperitoneally for 8 days. Changes were mostly conﬁned to the intestine and kidney where the vitamin D receptor (VDR) was highly expressed: increased intestinal Cyp24 and Cyp3a1 messenger RNA (mRNA) and a modest elevation of apical sodium-dependent bile salt transporter (Asbt) and P-glycoprotein (P-gp) protein; increased renal VDR, Cyp24, Cyp3a9, Mdr1a, and Asbt mRNA, as well as Asbt and P-gp protein expression; and decreased renal PepT1 and Oat1 mRNA expression. In comparison, 1"(OH)D2 treatment exerted a greater effect than 1,25(OH)2D3 on Cyp3a and Cyp24 mRNA. However, the farnesoid X receptor -related repressive effects on liver Cyp7a1 were absent because intestinal Asbt, FGF15 and portal bile acid concentrations were unchanged. Rats on the alternate day regimen showed milder changes and lessened signs of hypercalcemia and weight loss compared with rats receiving daily injections (similar or greater amounts of 0.64–2.56 nmol/kg daily ×4) described in previous reports, showing that the protracted pretreatment regimen was associated with milder inductive and lesser toxic effects in vivo. © 2010 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 100:1594–1604, 2011 R Keywords: doxercalciferol or Hectorol ;1",25-dihydroxyvitamin D3; vitamin D receptor; farnesoid X receptor; P-glycoprotein; membrane transporters; induction; drug transport; drug metabolizing enzymes; cytochrome P450

INTRODUCTION and androgen-independent prostate cancer.5 The prodrug is metabolized by the 25-hydroxylase Doxercalciferol (HectorolR , Genzyme, Cambridge, and 24-hydroxylase, respectively, to form the ac- Massachusetts), 1"-hydroxyvitamin D or 1"(OH)D , 2 2 tive metabolites, 1,25(OH) D 6 and 1,24(OH) D ,3 is metabolized to 1",25-dihydroxyvitamin D 2 2 2 2 2 ligands of the vitamin D receptor (VDR).7 Unlike [1,25(OH) D ], the naturally occurring, biolog- 2 2 1,25(OH) D ,1"(OH)D was reported to be associated ically active form of vitamin D .Like1",25- 2 3 2 2 with lower calcemic and phosphatemic activities.8 dihydroxyvitamin D [1,25(OH) D or calcitriol, the 3 2 3 There is increasing evidence that activation of the active form of vitamin D ], doxercalciferol is used for 3 VDR by 1,25(OH) D is involved in the regulation of the treatment of secondary hyperparathyroidism,1,2 2 3 transporters and enzymes. In the Caco-2 cell mono- metabolic bone disease,3 as well as tumor growth4 layer, 1,25(OH)2D3 was found to activate the VDR and upregulate cytochrome P450 3A4 (CYP3A4) and xeno- Correspondence to: K. Sandy Pang (Telephone: 416-978-6164; biotic transporters such as the multidrug resistance Fax: 416-978-8511; E-mail: [email protected]) protein (MDR1 or P-gp) and the multidrug resistance Journal of Pharmaceutical Sciences, Vol. 100, 1594–1604 (2011) 9–11 © 2010 Wiley-Liss, Inc. and the American Pharmacists Association associated proteins (MRP2 and MRP4) and, in

1594 JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 283 DOXERCALCIFEROL AND CALCITRIOL ON RAT TRANSPORTERS AND ENZYMES 1595

HepG2 cells, induce the hydroxysteroid sulfotrans- els in the intestine and liver for 24-hydroxylation,21 ferase enzyme (SULT2A1).12 In precision-cut intesti- but is present abundantly in the kidney.18 Reports nal slice studies, 1,25(OH)2D3 induced rat Cyp3a1 have suggested that the presence of microsomal CYP and Cyp3a2 and human CYP3A4.13 Recently, Fan (CYP2R1) and the mitochondrial CYP27A22 acts as 11 23–26 et al. compared the effects of 1"(OH)D2 with those of 25-hydroxylases. Thus, it is possible that local 1,25(OH)2D3 in the Caco-2 cell monolayer and found concentrations of the active metabolites of 1"(OH)D2 that 1"(OH)D2 was equipotent to 1,25(OH)2D3 in the are higher at various tissue sites that are rich in induction of CYP3A4, MDR1, MRP2 messenger RNA the activation enzymes. The effect of 1"(OH)D2 on (mRNA), and protein expression as well as MRP4 its target genes is expected to be higher in the protein. The prodrug, 1"(OH)D2, and the active lig- kidney because Cyp24 activates, rather than deac- and, 1,25(OH)2D3, appear to exhibit similar induc- tivates, 1"(OH)D2 in contrast to the catabolism of tive effects on transporters and enzymes in Caco-2 1,25(OH)2D3 to form the inactive 1,24,25(OH)3D3.Al- 11 cells in vitro. However, 1,25(OH)2D3 was reported though studies have shown that 1,25(OH)2D2 has to inhibit the activity of the liver X receptor-alpha a similar terminal elimination half-life compared 14 27 (LXR-") in HepG2 VDR-transfected cells and the to 1,25(OH)2D3, the conversion of 1"(OH)D2 to farnesoid X receptor (FXR) in CV1 VDR-transfected 1,25(OH)2D2 is still the rate-limiting step. These cells,15 suggesting that there is cross-talk between the events would generate interesting differences on VDR nuclear receptors that may further lead to other indi- effects between 1"(OH)D2 and 1,25(OH)2D3 adminis- rect changes on levels of the transporter and enzyme. tration pursuant to their administration to rats. Recent in vivo studies showed additional complex- The vitamin D analogs, including 1"(OH)D2, are a ities in separating the effects elicited by the VDR di- burgeoning and important class of compounds used rectly and those indirectly via the FXR due to elevated for the treatment of hyperparathyroidism, kidney bile acids. VDR effects differ among tissues, because and bone diseases, and anticancer therapy, and are the VDR is present abundantly in the rat intestine sometimes used clinically with other drugs for vari- and kidney16 but is scant in the rat liver,17,18 whereas ous indications.28 If the assoicated changes in trans- both human hepatocytes and nonparenchymal cells porter or enzyme are left unstudied or unnoticed, we show detectable expressions of the VDR.19 Intestinal would not be aware that the vitamin D analogs, when proteins in the proximal segments of the small in- used concomitantly with other drugs, could lead to testine, for example, Cyp3a1, Mrp2, Mrp3, Mrp4 and drug–drug interactions. We therefore proposed to in- the oligopeptide transporter, PepT1, were found to be vestigate the effects of 1"(OH)D2 in rats in vivo on 18,20 upregulated in 1,25(OH)2D3-treated rats. In the transporters and enzymes and compared these with kidney, P-gp and Asbt protein levels were induced in those from 1,25(OH)2D3 treatment. Although we only 1,25(OH)2D3-treated rats, whereas PepT1 and PepT2 reported mRNA and protein changes, these changes and organic anion transporters, Oat1 and Oat3, were have been further translated to functional changes downregulated.18 Although the VDR was virtually ab- and pharmacokinetic situations (yet to be published). sent in rat hepatocytes, the VDR effect on intestinal For this class of drugs, hypercalcemia is a major con- Asbt induction led to elevated portal bile acid concen- cern, and was observed in earlier studies wherein trations and FXR effects in the intestine pursuant to rats given daily 1,25(OH)2D3 (0.64–2.56 nmol/kg) in- 20 20 1,25(OH)2D3 dosing. Additionally, increased intesti- traperitoneal (i.p.) injections for 4 days. For ap- nal ﬁbroblast growth factor 15 (FGF15) produced in praisal of the strategy of lessened toxicity, we utilized the ileum activated the FGF receptor 4 (FGFR4) in a slightly more protracted regimen: the compounds the rat liver, coupled with the bile acid–FXR acti- were administered i.p. every other day for 8 days to vation of the small heterodimer partner (SHP), trig- investigate whether higher doses of 1"(OH)D2 and gered the downregulation of Cyp7a1, the rate-limiting 1,25(OH)2D3 would evoke similar or lesser changes enzyme for cholesterol metabolism, in the rat liver.20 on hypercalcemia and on VDR target genes in the in- These composite ﬁndings suggest the importance of testine, liver, and kidney in vivo. the VDR-mediated regulation of transporters and enzymes in the intestine, liver, and kidney, both directly and indirectly, in bile acid homeostasis as well as drug MATERIALS AND METHODS disposition in vivo. Materials The manner in which the prodrug doxercalciferol [1"(OH)D2] alters the expression of transporters The 1,25(OH)2D3 powder was purchased from Sigma and enzymes is mostly unknown. Doxercalciferol re- –Aldrich Canada (Mississauga, Ontario, Canada). quires activation to 1,25(OH)2D2 and 1,24(OH)2D2, 1"(OH)D2 was a kind gift from Dr Peter Bonate the active primary and minor metabolites, respec- (Genzyme, Cambridge, Massachusetts). Antibodies to tively, in order to bind to the VDR prior to man- villin (C-19) and Cyp7a1 (N-17) were purchased from ifestation of its activity. Cyp24 exists at lower lev- Santa Cruz Biotechnology (Santa Cruz, California),

DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 284 1596 CHOW ET AL. anti-Mrp2 (ALX-801–016-C250) from Alexis Bio- After blood collection, the portal vein was cannu- chemicals (San Diego, California), anti-P-gp (C219) lated and flushed with 50 mL of ice-cold physiologi- and anti-glyceraldehyde 3-phosphate dehydrogenase cal saline solution. The segments of the small intes- (GAPDH) (6C5) from Abcam (Cambridge, Mas- tine of vehicle control and treated rats were removed sachusetts), anti-Cyp3a2 antibodies (458223) that and placed on ice and cut into eight segments, as failed to distinguish from Cyp3a1 or Cyp3a9 were pur- outlined by the procedure of Chow et al.20 Entero- chased from BD Biosciences (Mississauga, Ontario, cytes were isolated from the mucosal scrapings of ev- Canada), and OAT11-A against Oat1 was from Alpha erted intestinal segments of the S1 (duodenum) and Diagnostic Intl. Inc. (San Antonio, Texas). Other an- S8 (ileum) segments. Enterocytes were then imme- tibodies were kind gifts from various investigators: diately snapped frozen in liquid nitrogen and stored anti-Oatp1a1 and anti-Ntcp (Dr. Allan W. Wolkoff, at −80◦C until analyses. The liver and kidney sam- Albert Einstein College of Medicine, the Bronx, New ples were removed, weighed, cut to small pieces, and York), anti-Asbt (Dr. Paul A. Dawson, Wake Forest snapped frozen in liquid nitrogen, then stored in the University, Salem, North Carolina), anti-Mrp3 (Dr. −80◦C freezer for future analyses. Yuichi Sugiyama, University of Tokyo, Japan), anti- Mrp4 (Dr. John D. Schuetz, St. Jude Children’s Re- Preparation of Subcellular Fractions from Enterocytes search Hospital, Memphis, Tennessee), and anti- Bsep (Dr. Bruno Stieger, University Hospital, Zurich, Frozen mucosal scrapings from intestinal segments Switzerland). All other reagents were purchased from were homogenized with 1 mL of Trizma HCl (0.1 M, Sigma–Aldrich and Fisher Scientific (Mississauga, pH 7.4) buffer containing 1% protease inhibitor cock- Ontario, Canada). tail (Sigma–Aldrich) and then sonicated for 10 s, as previously described.20 Samples were centrifuged ◦ 1,25(OH)2D3 and Vehicle (Corn Oil) Treatment in Rats first at 1000 g at 4 C for 10 min and the supernatant In Vivo was transferred to a new tube and spun again at 21,000 g at 4◦C for 1 h. The resulting pellet was resus- The concentrations of 1,25(OH) D and 1"(OH)D in 2 3 2 pended in the same homogenizing buffer for Western anhydrous ethanol were measured spectrophotomet- blot analyses. rically at 265 nm (UV-1700, Shimadzu Scientific In- struments, Columbia, Maryland), and the solutions were diluted in filtered corn oil (Sigma–Aldrich) for Preparation of Subcellular Fractions of Liver injection. Male Sprague–Dawley rats (260–280 g), and Kidney Tissue purchased from Charles River (St. Constant, Quebec, For preparation of the crude membrane fraction, liver Canada), were given water and food ad libitum and and kidney tissues were homogenized in the crude maintained under a 12:12-h light and dark cycle in ac- membrane homogenizing buffer (250 mM sucrose, cordance to animal protocols approved by the Univer- 10 mM HEPES, and 10 mM Trizma base, pH 7.4) sity of Toronto (Ontario, Canada). Rats (n = 4 in each containing 1% protease inhibitor cocktail as described group) were injected with 0, 1.28 nmol/kg of 1"(OH)D 2 above. The resultant homogenate was centrifuged at or 4.8 or 6.4 nmol/kg of 1,25(OH) D in 1 mL/kg corn 2 3 3000 g for10minat4◦C. The supernatant obtained oil i.p. every other day for 8 days. At 48 h following the was transferred to an ultracentrifuge tube and spun last day of injection, rats were anesthetized with ke- at 33,000 g for 60 min at 4◦C. The resultant pellet tamine and xylazine (90 and 10 mg/kg, respectively) was placed in a resuspension buffer (50 mM man- by i.p. injection. An aliquot (0.5 mL) of portal and sys- nitol, 20 mM HEPES, 20 mM Trizma base, pH 7.4) temic blood was collected and centrifuged at 605 g for containing 1% protease inhibitor cocktail for West- 10 min to obtain plasma. ern blot analyses. For preparation of the hepatic microsomal fraction, liver tissue was homogenized with Blood Analysis and Preparation of Tissues microsomal buffer (250 mM sucrose, 10 mM Trizma Portal bile acid concentration was determined using HCl, 1 mM EDTA, pH 7.4) containing 1% protease in- a Total Bile Acids Assay Kit (BQ042A-EALD from hibitor cocktail, as described above. The homogenate BioQuant, San Diego, California) following the man- was centrifuged at 9000 g for10minat4◦C to pro- ufacturer’s protocol. Calcium and phosphorus mea- vide a supernatant, which was subsequently spun at surements in systemic plasma were determined 100,000 g for another 60 min at 4◦C. The resulting by inductively coupled plasma atomic emission pellet was resuspended in the same microsomal ho- spectroscopy (Optima 3000 DV, PerkinElmer Inc., mogenizing buffer containing protease inhibitor for Waltham, Massachusetts). These plasma samples Western blot analyses. The protein concentrations of were diluted 350-fold with 1% nitric acid before each the samples were assayed by Lowry method29 using measurement. Calcium was measured at 317.9 and bovine serum albumin as the standard. Samples were 315.9 nm, whereas phosphorus at 213.6 and 214.9 nm. then stored at −80◦C until Western blot analyses.

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 DOI 10.1002/jps 285 DOXERCALCIFEROL AND CALCITRIOL ON RAT TRANSPORTERS AND ENZYMES 1597

Western Blotting ABI Sequence Detection software version 1.4 (Applied Biosystems Canada, ON) to obtain critical threshold Intestinal, hepatic, and renal protein samples (20 cycle (C ) value. Fold changes between vehicle con- or 50 :g) were separated by 7.5% or 10% SDS- T trol and treatment was expressed as 2−(CT).Alltar- polyacrylamide gels at 100 V. After separation, get mRNA data were normalized to villin mRNA for proteins were transferred onto a nitrocellulose intestinal samples and GAPDH mRNA for liver and membrane (Amersham Biosciences, Piscataway, New kidney samples. Jersey) that was subsequently blocked with 5% (w/v) skim milk in Tris-buffered saline (pH 7.4) and 0.1% Tween 20 (TBS-T) (Sigma–Aldrich Canada) for 1 h Statistical Analysis at room temperature, and then washed with 0.1% ± TBS-T followed by incubation with primary antibody Data were expressed as mean standard deviation. solution in 2% skim milk in 0.1% TBS-T overnight Data comparing the difference between two groups at 4◦C. On the next day, the membrane was washed were analyzed using both the two-tailed Student’s t- with 0.1% TBS-T and then incubated with secondary test and the Mann–Whitney U-test, respectively. A antibody in 2% skim milk in 0.1% TBS-T for 2 h at one-way analysis of variance was used for the mRNA room temperature, and again washed with 0.1% TBS- and protein data to compare the treatment groups T. Bands were visualized using chemiluminescence versus the vehicle control. A P value of less than 0.05 reagents (Amersham Biosciences) and quantified by was set as the level of significance. scanning densitometry (NIH Image software; http:// rsb.info.nih.gov/nih-image/). The band intensity of target protein was normalized against that of villin RESULTS for intestinal samples or GAPDH for liver and kidney samples, to correct for loading errors. All target Effects of 1α(OH)D2 and 1,25(OH)2D3 on Body proteins were checked by linearity (range 10–60 :g) Weight, Calcium and Phosphorus Levels, and Portal Bile against band intensity, and were found to be associ- Acid Concentrations ated with good correlation coefficients (0.9–0.99) for Treatment with the 1.28 nmol/kg dose of 1"(OH)D , the plot of the intensity versus the amount of protein 2 given i.p. every other day for 8 days, resulted in a in the tissue. slightly but significantly lower body weight compared with the vehicle control at the beginning of day 2. The Quantitative Real-Time Polymerase Chain Reaction 1,25(OH)2D3 dose of 4.8 nmol/kg, given every other Similar to the procedures previously described,20 today for 8 days, also resulted in a modest but signifi- tal RNA from scraped enterocytes, liver, and kidney cant loss in body weight compared with vehicle control tissues were extracted with the TRIzol extraction at the beginning of day 6, and a small and significant method (Sigma–Aldrich) according to the manufac- weight loss compared with vehicle control at the be- turer’s protocol, with modifications. Total RNA was ginning at day 5 for the 6.4 nmol/kg dose (Fig. 1a). In quantified by ultraviolet spectrometry measured at comparison, rats reported earlier, which were given 260 nm. The purity was checked by the ratios of the daily, lower 1,25(OH)2D3 doses (1.28 and 2.56 nmol/ readings at 260/280 and 260/230 nm (≥1.7). The com- kg) consecutively for 4 days, suffered a significantly plementary DNA (cDNA) was immediately synthe- greater loss in body weight that was apparent after sized from the RNA samples, using the High Capacity day 3 of treatment (Fig. 1b).18,20 The calcium levels cDNA Reverse Transcription Kit (Applied Biosystems in plasma were significantly increased (12% and 20%) Canada, Ontario, Canada). Quantitative polymerase in rats treated with 1"(OH)D2 and the 1,25(OH)2D3 chain reaction (qPCR) was performed with two de- (1.28, and 4.8 nmol/kg, respectively, every other day tection systems (SYBR Green or Taqman assay), de- for 8 days), whereas the phosphorus levels had re- pending on the availability of the primer sets. A PCR mained unchanged (Fig. 1c). These changes were mixture (20 :L final volume) consisting of 75 ng milder in comparison with the daily treatment group cDNA, 1 :M of forward and reverse primers, and reported earlier (Fig. 1d), wherein higher calcium in- 1× Power SYBR Green PCR Master Mix (Applied creases (20%, 33%, and 37%) for all 1,25(OH)2D3- Biosystems) was used to perform PCR analysis. In- treated groups (0.64, 1.28, and 2.56 nmol/kg daily ×4) formation on primer sequences was summarized in were noted. Again, the phosphorus levels were un- the published lists of Chow et al.18,20 Amplification changed. Portal bile acid concentrations were signifi- and detection were performed using the ABI 7500 cantly increased only in the group receiving the high- system. The qPCR system was designated the follow- est dose of 6.4 nmol/kg of 1,25(OH)2D3 every other ing PCR reaction profile: 95◦C for 10 min, and 40 cy- day (Fig. 1e), and the same was observed in the group ◦ ◦ cles of 95 C for 15 s and 60 C for 1 min, followed by treated with 2.56 nmol/kg of 1,25(OH)2D3 daily for the dissociation curve. Data were analyzed using the 4 days (Fig. 1f).

DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 286 1598 CHOW ET AL.

Figure 1. Comparative effects of 1"(OH)D2 and 1,25(OH)2D3 treatment every other day (for 8 days) versus every day (for 4 days) on rat body weight (a and b), plasma calcium and phosphorus levels (c and d), and bile acids (e and f). ∗P < 0.05 compared with vehicle control using the two- tailed Student’s t-test.

Effects of 1α(OH)D2 and 1,25(OH)2D3 Treatment on transporters, and enzymes of the intestine. With Intestinal Nuclear Receptors, Enzymes, and 1"(OH)D2 treatment, the intestinal mRNA expres- Transporters sions of VDR, FXR, SHP, FGF15, Cyp3a9, Asbt, and Mdr1a in rat duodenum (S1) and/or ileum (S8) re- Treatment of 1"(OH)D2 and 1,25(OH)2D3 to rats remained unchanged, whereas levels for Cyp24 and sulted in different effects on the nuclear receptors, Cyp3a1 were increased signiﬁcantly in both the

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 DOI 10.1002/jps 287 DOXERCALCIFEROL AND CALCITRIOL ON RAT TRANSPORTERS AND ENZYMES 1599

Figure 2. Effects of 1"(OH)D2 and 1,25(OH)2D3 treatment every other day (for 8 days) on rat intestinal Cyp3a protein in S1 (duodenum) and Asbt and P-gp protein in S8 (ileum). Cyp3a, Asbt, and P-gp protein bands were detected at 56, 48, and 170 kDa, respectively. ∗P < 0.05 compared with vehicle control in the same segment using the two- tailed Student’s t-test. duodenum and ileum (Table 1). On the contrary, a slight decrease in VDR and FXR mRNA and a slight increase in SHP, FGF15, Cyp3a9, and Asbt mRNA were observed in the ileum of the 1,25(OH)2D3- treated rats, with greater changes at the higher dose. Upon comparison, the fold changes in Cyp24 and Cyp3a1 mRNA induced by 1"(OH)D2 were >300× and >7× those of 1,25(OH)2D3 (Table 1). However, " the total Cyp3a protein in the duodenum remained Figure 3. Comparative effects of 1 (OH)D2 and unchanged among all treatment groups (Fig. 2). The 1,25(OH)2D3 treatment every other day (for 8 days) on Asbt protein expression in the ileum was increased by rat hepatic cytochrome P450 isozymes (a) and sinusoidal and canalicular transporter (b) proteins. Cyp7a1, Ntcp, 45% only with 1"(OH)D treatment (Fig. 2), despite 2 Oatp1a1, Bsep, Mrp2, Mrp3, and Mrp4 bands were detected that the increase in Asbt mRNA was insignificant (Ta- at 50, 50, 80, 160, 180, 170, and 160 kDa, their molecular ble 1). Although no change was observed for Mdr1a weights, respectively. ∗P < 0.05 compared with vehicle con- mRNA for all treatment groups (Table 1), the P-gp trol using the two-tailed Student’s t-test. protein level was increased by 66% after 1"(OH)D2, but not after 1,25(OH)2D3 treatment (Fig. 2). polypeptide 1a1, in the 1"(OH)D treatment group. α 2 Effects of 1 (OH)D2 and 1,25(OH)2D3 Treatment on Protein levels of total Cyp3a and Cyp7a1 (Fig. 3a), and Hepatic Nuclear Receptors, Enzymes, and Transporters for the transporters, Mrp2, Mrp3, Mrp4, Bsep, Ntcp, and Oatp1a1 (Fig. 3b), were all unchanged. The only Treatment with 1"(OH)D2 and 1,25(OH)2D3 only exerted a minimum impact on the expression of hepatic notable change was the 2.3-fold increase in P-gp pro- nuclear receptors/transcription factors, enzymes, and tein in the 1,25(OH)2D3 treatment group (4.8 nmol/ 18 transporters in the rat liver. The mRNA expression of kg) (Fig. 3b). The same was observed previously. the VDR, FXR, and hepatocyte nuclear factor 4 alpha Effects of 1α(OH)D and 1,25(OH) D Treatment on (HNF-4") remained unchanged among the treatment 2 2 3 Renal Nuclear Receptors, Enzymes, and Transporters groups (Table 2). SHP mRNA was increased significantly in the 1,25(OH)2D3 treatment groups of 4.8 and Greater changes were observed for the renal nu- 6.4 nmol/kg. A twofold to threefold and highly variable clear receptors/transcription factor, enzymes, and increase was observed for Mrp2, Mrp3, and Cyp3a9 transporters of rats treated with 1"(OH)D2 and mRNA in the 1"(OH)D2 and 1,25(OH)2D3 treatment 1,25(OH)2D3. Notably, the mRNA expression of the groups (Table 2). The changes in hepatic Cyp3a1, VDR was significantly increased twofold to fourfold, Cyp3a2, Cyp7a1, Ntcp, Bsep, and Mdr1a mRNA ex- whereas levels of SHP, HNF-1", HNF-4", and liver repression were minor, except for the small change in ceptor homolog-1 remained relatively unchanged for mRNA of Oatp1a1, the organic anion transporting all groups (Table 3). Renal Cyp24 (23- to 38-fold),

DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 288 1600 CHOW ET AL.

Table 1. Changes in mRNA Expression of Rat Intestinal Nuclear Receptors, Enzymes, and Transporters in S1 (Duodenum) and S8 (Ileum), Expressed as Fold Expression Compared with Vehicle Treatment

Nuclear Receptor, Intestinal Enzymes, and Vehicle Every Other 1"(OH)D2 nmol/kg day 1,25(OH)2D3 nmol/kg day Every Segment Transporters Day for 8 Days Every Other Day for 8 Days Other Day for 8 Days ANOVA 0 1.28 4.80 6.40 S1 Duodenum VDR 1.00 ± 0.12 0.96 ± 0.06 0.84 ± 0.16 0.96 ± 0.23 Cyp24 1.00 ± 0.31 174 ± 137a 0.47 ± 0.07 0.53 ± 0.16 <.05 Cyp3a1 1.00 ± 0.64 7.52 ± 7.17a 0.71 ± 0.24 0.87 ± 0.75 <.05 Cyp3a9 1.00 ± 0.38 2.05 ± 1.38 1.41 ± 0.28 1.16 ± 0.75 Mdr1a 1.00 ± 0.17 0.72 ± 0.09 1.30 ± 0.14 0.99 ± 0.37 S8 ileum VDR 1.00 ± 0.09 1.01 ± 0.17 0.68 ± 0.03b 0.66 ± 0.15b <.05 FXR 1.00 ± 0.23 0.90 ± 0.13 0.67 ± 0.37 0.72 ± 0.10b SHP 1.00 ± 0.76 1.36 ± 1.33 1.78 ± 0.44 3.71 ± 1.44b FGF-15 1.00 ± 0.52 4.11 ± 3.17 1.24 ± 0.79 2.64 ± 1.21b <.05 Cyp24 1.00 ± 0.29 879 ± 720a 1.30 ± 0.64 1.49 ± 0.36 <.05 Cyp3a1 1.00 ± 0.47 6.38 ± 2.46b 0.86 ± 0.53 1.26 ± 1.06 <.05 Cyp3a9 1.00 ± 0.63 1.75 ± 0.47 1.84 ± 1.05 2.21 ± 0.42b <.05 Mdr1a 1.00 ± 0.22 1.43 ± 0.42 1.04 ± 0.13 1.46 ± 0.41 <.05 Asbt 1.00 ± 0.13 1.27 ± 0.14 1.34 ± 0.79 1.43 ± 0.26b

a P < 0.05 compared with vehicle treatment (0 nmol 1,25(OH)2D3/kg) using Mann–Whitney U-test. b P < 0.05 compared with vehicle treatment (0 nmol 1,25(OH)2D3/kg) using the two-tailed Student’s t-test. ANOVA, analysis of variance; VDR, vitamin D receptor; FXR, farnesoid X receptor; SHP, small heterodimer partner; Asbt, apical sodium-dependent bile salt transporter.

Cyp3a9 (8- to 32-fold), Asbt (about 2.5- to 5-fold), group] were observed (Table 3). The decrease in Oat1 and Mdr1a (twofold) mRNA were increased among and increase in P-gp protein (Fig. 4) paralleled the all 1,25(OH)2D3-treated rats. The inductive effects changes in mRNA levels (Table 3). The Mrp3, Mrp4, of 1"(OH)D2 were usually greater than those of and Mrp2 protein expressions were unchanged for 1,25(OH)2D3 (Table 3). However, expression levels of all groups, whereas Asbt protein levels were found Cyp3a1, Oat3, Ost-$, PepT2, and Mrp4 mRNA re- to increase signiﬁcantly at highest 1,25(OH)2D3 dose mained unchanged (Table 3). The positive changes (Fig. 4). noted were conﬁned to increased Ost-", Mrp2, and " Mrp3 mRNA for the 1 (OH)D2 treatment group only, DISCUSSION and not for the 1,25(OH)2D3 treatment groups. De- creased mRNA expressions of Oat1 [for the 4.8 nmol/ The VDR-mediated transcriptional changes in trans- kg of 1,25(OH)2D3 treatment group] and PepT1 [for porter and enzyme expression are increasingly be- the 4.8 and 6.4 nmol/kg of 1,25(OH)2D3 treatment ing investigated due to the importance of vitamin D

Table 2. Changes in mRNA Expression of Rat Hepatic Nuclear Receptors, Enzymes, and Transporters, Expressed as Fold Expression Compared with Vehicle Treatment

Nuclear Receptors, Vehicle Every Other 1"(OH)D2 nmol/kg Every 1,25(OH)2D3 nmol/kg Every Other Enzymes, and Transporters Day for 8 Days Other Day for 8 Days Day for 8 Days ANOVA 1.28 4.80 6.40 VDR 1.00 ± 0.26 1.22 ± 0.61 1.10 ± 0.35 0.82 ± 0.32 FXR 1.00 ± 0.26 1.15 ± 0.13 1.24 ± 0.24 1.12 ± 0.29 SHP 1.00 ± 0.35 2.02 ± 1.15 2.22 ± 0.88a 1.66 ± 0.32b HNF-4" 1.00 ± 0.22 1.24 ± 0.41 1.19 ± 0.19 0.97 ± 0.07 Cyp3a1 1.00 ± 0.49 1.82 ± 0.80 1.19 ± 0.16 1.37 ± 0.08 Cyp3a2 1.00 ± 0.37 0.79 ± 0.65 1.51 ± 0.38 0.92 ± 0.33 Cyp3a9 1.00 ± 1.06 0.99 ± 0.32 2.76 ± 2.34 3.64 ± 2.02a Cyp7a1 1.00 ± 0.32 0.94 ± 0.33 1.43 ± 0.58 1.95 ± 1.81 Oatp1a1 1.00 ± 0.17 0.73 ± 0.10a 1.23 ± 0.43 0.98 ± 0.24 Ntcp 1.00 ± 0.24 1.26 ± 0.07 1.34 ± 0.39 0.96 ± 0.10 Bsep 1.00 ± 0.14 0.92 ± 0.19 1.49 ± 0.59 1.17 ± 0.18 Mdr1a 1.00 ± 0.29 1.07 ± 0.03 1.33 ± 0.56 1.00 ± 0.32 Mrp2 1.00 ± 0.56 1.07 ± 0.50 1.89 ± 0.47a 1.58 ± 0.31 <.05 Mrp3 1.00 ± 0.30 2.02 ± 0.68a 1.48 ± 0.64 1.20 ± 0.11 <.05 Mrp4 1.00 ± 0.31 0.74 ± 0.16 1.34 ± 0.36 1.01 ± 0.08

a P < 0.05 compared with vehicle treatment (0 nmol 1,25(OH)2D3/kg/day) using two-tailed Student’s t-test. ANOVA, analysis of variance; VDR, vitamin D receptor; FXR, farnesoid X receptor; SHP, small heterodimer partner; HNF-4", hepatocyte nuclear factor 4 alpha (HNF-4").

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 DOI 10.1002/jps 289 DOXERCALCIFEROL AND CALCITRIOL ON RAT TRANSPORTERS AND ENZYMES 1601

Figure 4. Comparative effects of 1"(OH)D2 and 1,25(OH)2D3 treatment every other day (for 8 days) on rat renal basolateral and apical transporters proteins. Oat1 band was detected at 72 kDa. ∗P < 0.05 compared with vehicle control using the two-tailed Student’s t-test. †P < 0.05 using one-way analysis of variance. analogs in the treatment of secondary hyperparathy- sponse elements in these genes.31,32 In support of roidism, psoriasis, advanced malignancy, and an- these data in vitro,9–12 our laboratory also observed 28,30 ticancer therapy. In man, inductive effects on that 1,25(OH)2D3 exerts a positive transcriptional in- 33 CYP3A4 and MDR1/P-gp by 1,25(OH)2D3 were ob- duction on the rat Asbt and Cyp3a1 of the small served in Caco-2 cells,11 observations that are con- intestine via the VDR in vivo,20 and Cyp3a9, Mdr1a/ gruent with reports on presence of vitamin D re- P-gp, and Asbt in the rat kidney.18 The reduction of

Table 3. Changes in mRNA Expression of Rat Renal Nuclear Receptors, Enzymes, and Transporters, Expressed as Fold Expression Compared with Vehicle Treatment∗

Nuclear Receptors, Vehicle Every Other 1"(OH)D2 nmol/kg Every 1,25(OH)2D3 nmol/kg Every Other Enzymes, and Transporters Day for 8 Days Other Day for 8 Days Day for 8 Days ANOVA 1.28 4.80 6.40 VDR 1.00 ± 0.12 4.43 ± 0.64a 2.40 ± 0.86a 2.41 ± 0.38a <.05 FXR 1.00 ± 0.15 0.93 ± 0.15 0.71 ± 0.05a 0.85 ± 0.10 <.05 SHP 1.00 ± 0.10 0.94 ± 0.29 0.94 ± 0.16 0.86 ± 0.14 HNF-1" 1.00 ± 0.11 0.83 ± 0.14 0.87 ± 0.09 0.77 ± 0.11a HNF-4" 1.00 ± 0.12 0.84 ± 0.08 0.79 ± 0.15 0.93 ± 0.06 LRH-1 1.00 ± 0.29 1.01 ± 0.20 1.16 ± 0.36 0.84 ± 0.06 Cyp24 1.00 ± 0.88 38.2 ± 7.66a 23.1 ± 8.19a 27.8 ± 3.19a Cyp3a1 1.00 ± 0.82 0.94 ± 0.72 1.28 ± 0.95 2.22 ± 0.89 Cyp3a9 1.00 ± 0.32 31.8 ± 7.88a 9.42 ± 5.14a 8.36 ± 5.49a <.05 Asbt 1.00 ± 0.29 4.99 ± 0.40a 2.42 ± 0.74a 2.62 ± 0.97a <.05 Oat1 1.00 ± 0.29 0.72 ± 0.06 0.49 ± 0.17a 0.86 ± 0.13 <.05 Oat3 1.00 ± 0.44 1.27 ± 0.13 0.64 ± 0.41 1.30 ± 0.06 <.05 Ost-" 1.00 ± 0.22 2.15 ± 0.52a 0.91 ± 0.36 1.12 ± 0.13 <.05 Ost-$ 1.00 ± 0.28 0.89 ± 0.16 0.96 ± 0.27 0.78 ± 0.17 Pept1 1.00 ± 0.40 0.38 ± 0.12a 0.34 ± 0.20a 0.59 ± 0.09 <.05 Pept2 1.00 ± 0.18 0.88 ± 0.06 0.96 ± 0.31 0.84 ± 0.18 Mdr1a 1.00 ± 0.12 2.13 ± 0.34a 2.11 ± 0.56a 2.21 ± 0.35a <.05 Mrp2 1.00 ± 0.35 1.50 ± 0.20a 0.82 ± 0.10 1.38 ± 0.27 <.05 Mrp3 1.00 ± 0.09 1.48 ± 0.10a 1.01 ± 0.07 1.16 ± 0.23 <.05 Mrp4 1.00 ± 0.08 0.96 ± 0.09 0.84 ± 0.18 0.91 ± 0.04

∗Cyp3a2 was present at very low levels and was not reported. a P < 0.05 compared with vehicle treatment (0 nmol 1,25(OH)2D3/ kg) using the two-tailed Student’s t-test. ANOVA, analysis of variance; VDR, vitamin D receptor; FXR, farnesoid X receptor; SHP, small heterodimer partner; HNF-1", hepatocyte nuclear factor-1 alpha; LRH-1, liver receptor homolog-1; Asbt, apical sodium-dependent bile salt transporter.

DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 290 1602 CHOW ET AL.

Cyp7a1 in the rat liver after daily 1,25(OH)2D3 treat- increase in P-gp protein expression for the all treatment in vivo was explained by the elevated intesti- ment groups (Fig. 4). Although reports have shown nal FGF15 and hepatic FXR activation as a result of that the upregulation of Mdr1a/P-gp may be due to higher portal bile acid concentrations that were asso- the activation of the transcription factor hypoxia- 20 39 ciated with 1,25(OH)2D3 treatment. inducible factor-1 by calcium, we believe that the Moreover, parallel increases in plasma calcium lev- upregulation of Mdr1a/P-gp is mainly due to VDR ac- 32 els were observed in 1"(OH)D2 and 1,25(OH)2D3- tivation. First of all, Saeki et al. showed that a vita- treated rats (Fig. 1c). Previous studies had reported min D response element is found in MDR1 gene, and 11 on hypercalcemia and weight loss being associated Fan et al. have reported that both 1,25(OH)2D3 and with the treatment of 1,25(OH)2D3 after its daily ad- 1"(OH)D2 increased P-gp in vitro. Calcium absorp- ministration to rats (Figs. 1b and 1d).18,20 The vita- tion occurred in both intestine and kidney by TRP6 40,41 min D analog, 1"(OH)D3, a prodrug of 1,25(OH)2D3,is and TRP5, respectively. If elevated plasma cal- known to alter transporters and enzymes in mice,34,35 cium would increase P-gp in the intestine, liver, and and this prodrug displayed higher hypercalcemic kidney, then levels of P-gp would all be increased in 36,37 properties than 1"(OH)D2. It is therefore pru- these tissues. However, only P-gp in the kidney was dent to first assess the methods of dosing in vivo upregulated in the 1,25(OH)2D3-treated rats in this fore correlation to the changes in transporter and (Fig. 4) and other studies.18 enzyme target genes associated with treatment of When changes were appraised in the rat liver, a 17 1,25(OH)2D3 or vitamin D analogs. In this study, we VDR-poor tissue, 1"(OH)D2 was found to play only examined whether a longer, protracted dosing sched- a minor role in the regulation of transporter and en- ule of 1"(OH)D2 and 1,25(OH)2D3 treatment can re- zyme expressions (Table 2; Fig. 3), as observed for sult in lessened hypercalcemia and yet retain the ef- 1,25(OH)2D3 in this study. The changes were also fects on transporters and enzymes in vivo. Indeed, similar to those observed in previous studies wherein the alternate-day, protracted dosing regimen allevi- 1,25(OH)2D3 at lower doses was administered daily ated the hypercalcemic and weight loss effects in- for 4 days.20 However, the secondary FXR effect on the duced by 1,25(OH)2D3 and 1"(OH)D2 compared with reduction of Cyp7a1 (Fig. 3a) as a result of the high those from the daily administration of 1,25(OH)2D3 portal bile acid concentrations after daily injections 20 (Figs. 1a–1d), showing that the newly adopted reg- of 1,25(OH)2D3 was absent in the present study imen is effective in reducing the toxicity associated in both the 1"(OH)D2 and 1,25(OH)2D3 treatment with treatment of vitamin D analogs. groups. The induction of SHP in liver by 1"(OH)D2 When changes with 1"(OH)D2 were appraised in and 1,25(OH)2D3 for the protracted, alternate-day the rat kidney, a major target site for VDR trans- regimen (Table 2) was lower than those achieved with 20 activation, the alternate-day treatment regimen per- 1,25(OH)2D3 daily dosing at lower doses. The com- sistently increased Cyp24, Cyp3a9, Asbt, and Mdr1a, parison suggests lower FXR effects for the protracted and downregulated Oat1, observations that compared regimen. Indeed, the alternative day administration well with results from 1,25(OH)2D3 treatment at of 1,25(OH)2D3 to rats resulted in a lack of change higher doses and the same dosing regimen (Table 3; Asbt protein and a smaller induction of intestinal Fig. 4) as well as those reported previously for the FGF15 mRNA (Table 1), factors previously found to be 18 20 daily regimen with 1,25(OH)2D3. However, the ex- important in the downregulation of Cyp7a1. How- tents of induction of VDR, Cyp24, Cyp3a9, and Asbt ever, the extents of induction of intestinal FGF15 by the 1"(OH)D2 treatment were significantly greater mRNA (4.1-fold) and Asbt protein by 1"(OH)D2 treat- than those of 1,25(OH)2D3 administered at higher ment in this study (Table 1) were similar and com- doses (Table 3). Induction of Cyp24 due to the high parable to those previously observed with the daily 21 20 levels of VDR in the kidney and the activation of treatment regime; the lack of change for 1"(OH)D2 Cyp24 is expected to result in increased catabolism of on Cyp7a1 in the liver was unknown. 38 11,13 1,25(OH)2D3 but increased activation of 1"(OH)D2 The upregulation of intestinal CYP3A and re- 3 18 to 1,24(OH)2D2. As a result, 1"(OH)D2, the prodrug nal P-gp by the vitamin D analogs, 1,25(OH)2D3 or that is activated to 1,24(OH)2D2, may be more effi- 1"(OH)D2 (Fig. 4), could lead to potential drug–drug cacious than the native active ligand, 1,25(OH)2D3, interactions when these drugs are taken in combi- that is rapidly inactivated by Cyp24 in vivo. These nation with other xenobiotics. Currently, high doses changes can explain the reciprocal relationship be- of 1,25(OH)2D3 and the vitamin D analogs have been 6 tween 1,24(OH)2D2 and 1,25(OH)2D3. Perhaps due used clinically in combination with anticancer agents to these reasons, 1"(OH)D2 treatment, even at a such as docetaxel and paclitaxel that are also CYP3A4 lower dose, evoked dramatically greater changes in and P-gp substrates.28 The upregulation of CYP3A4 renal transporter and enzyme expressions more than and P-gp by 1,25(OH)2D3 or 1"(OH)D2 could reduce 1,25(OH)2D3 at high doses. The one significant change the therapeutic effects of these drugs and their as- in the kidney resides in Mdr1a mRNA and a twofold soicated toxicity. Thus, more studies are needed to

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 DOI 10.1002/jps 291 DOXERCALCIFEROL AND CALCITRIOL ON RAT TRANSPORTERS AND ENZYMES 1603

examine the potential drug–drug interactions of vita- activated dihydroxyvitamin D2 metabolites decreases endoge- " min D analogs and drugs that are P-gp substrates. nous 1 ,25-dihydroxyvitamin D3 in rats and monkeys. En- In conclusion, this study demonstrates that VDR docrinology 136:4749–4753. 7. Peter WL. 2000. Hectorol: A new vitamin D prohormone. regulation of transporters and enzymes in the small Nephrol Nurs J 27:67–68. intestine and kidney persisted when 1,25(OH)2D3 and 8. Salusky IB. 2005. Are new vitamin D analogues in renal bone the provitamin D analog, 1"(OH)D2, are given inter- disease superior to calcitriol? Pediatr Nephrol 20:393–398. mittently to the rat in vivo. In this study, we veri- 9. Aiba T, Susa M, Fukumori S, Hashimoto Y. 2005. The effects of culture conditions on CYP3A4 and MDR1 mRNA induction fied that 1"(OH)D2 showed a greater induction poten- by 1",25-dihydroxyvitamin D3 in human intestinal cell lines, tial toward transporters and enzymes in the kidney Caco-2 and LS180. Drug Metab Pharmacokinet 20:268–274. than 1,25(OH)2D3, an observation attributed to the 10. Schmiedlin-Ren P, Thummel KE, Fisher JM, Paine MF, Lown high levels of renal Cyp24 that converts 1"(OH)D2 to KS, Watkins PB. 1997. Expression of enzymatically active the active form, 1,24(OH)2D2 (Table 3). These obser- CYP3A4 by Caco-2 cells grown on extracellular matrix-coated " vations are significant in terms of considerations of permeable supports in the presence of 1 ,25-dihydroxyvitamin D3. Mol Pharmacol 51:741–754. drug–drug interactions for the concomitant use of an- 11. Fan J, Liu S, Du Y, Morrison J, Shipman R, Pang KS. 2009. ticancer drugs with the vitamin D analogs. With a pro- Up-regulation of transporters and enzymes by the vitamin D tracted dosing schedule of these vitamin D analogs, receptor ligands, 1",25-dihydroxyvitamin D3 and vitamin D changes in transporters and enzymes in the intestine analogs, in the Caco-2 cell monolayer. J Pharmacol Exp Ther and kidney were found to be less dramatic compared 330:389–402. 12. Echchgadda I, Song CS, Roy AK, Chatterjee B. 2004. Dehy- with those after daily administration. The protracted droepiandrosterone sulfotransferase is a target for transcrip- schedule is associated with reduced hypercalcemia tional induction by the vitamin D receptor. Mol Pharmacol and weight loss, and would prove beneficial for long- 65:720–729. term treatments. The benefits of even more sparse 13. Khan AA, Chow EC, van Loenen-Weemaes AM, Porte RJ, Pang administration of vitamin D analogs, as used clin- KS, Groothuis GM. 2009. Comparison of effects of VDR ver- 28 sus PXR, FXR and GR ligands on the regulation of CYP3A ically in the treatment of cancer, on transporters isozymes in rat and human intestine and liver. Eur J Pharm and enzymes need to be further appraised for their Sci 37:115–125. long-term use. 14. Jiang W, Miyamoto T, Kakizawa T, Nishio SI, Oiwa A, Takeda T, Suzuki S, Hashizume K. 2006. Inhibition of LXR" signaling by vitamin D receptor: Possible role of VDR in bile acid ACKNOWLEDGMENTS synthesis. Biochem Biophys Res Commun 351:176–184. 15. Honjo Y, Sasaki S, Kobayashi Y, Misawa H, Nakamura This work was supported by the Canadian Institutes H. 2006. 1,25-dihydroxyvitamin D3 and its receptor inhibit for Health Research, MOP89850. Edwin C.Y. Chow the chenodeoxycholic acid-dependent transactivation by far- is a recipient of the NSERC Alexander Graham Bell nesoid X receptor. J Endocrinol 188:635–643. Canada Graduate Scholarship. 16. Sandgren ME, Bronnegard M, DeLuca HF. 1991. Tissue distribution of the 1,25-dihydroxyvitamin D3 receptor in the male rat. Biochem Biophys Res Commun 181:611–616. 17. Gascon-Barre M, Demers C, Mirshahi A, Neron S, Zalzal S, References Nanci A. 2003. The normal liver harbors the vitamin D nu- 1. Martin KJ, Gonzalez EA. 2001. Strategies to minimize bone clear receptor in nonparenchymal and biliary epithelial cells. disease in renal failure. Am J Kidney Dis 38:1430–1436. Hepatology 37:1034–1042. 2. Frazao JM, Elangovan L, Maung HM, Chesney RW, Acchiardo 18. Chow EC, Sun H, Khan AA, Groothuis GM, Pang KS. 2010. " SR, Bower JD, Kelley BJ, Rodriguez HJ, Norris KC, Robertson Effects of 1 ,25-dihydroxyvitamin D3 on transporters and en- JA, Levine BS, Goodman WG, Gentile D, Mazess RB, Kyllo zymes of the rat intestine and kidney in vivo. Biopharm Drug DM, Douglass LL, Bishop CW, Coburn JW. 2000. Intermittent Dispos 31:91–108. 19. Han S, Chiang JY. 2009. Mechanism of vitamin D receptor doxercalciferol (1"-hydroxyvitamin D2) therapy for secondary hyperparathyroidism. Am J Kidney Dis 36:550–561. inhibition of cholesterol 7"-hydroxylase gene transcription in 3. Strugnell S, Byford V, Makin HL, Moriarty RM, Gilardi R, human hepatocytes. Drug Metab Dispos 37:469–478. LeVan LW, Knutson JC, Bishop CW, Jones G. 1995. 1",24(S)- 20. Chow EC, Maeng HJ, Liu S, Khan AA, Groothuis GM, Pang KS. 2009. 1",25-Dihydroxyvitamin D triggered vitamin D dihydroxyvitamin D2: A biologically active product of 1"- 3 receptor and farnesoid X receptor-like effects in rat intestine hydroxyvitamin D2 made in the human hepatoma, Hep3B. Biochem J 310 (Pt 1):233–241. and liver in vivo. Biopharm Drug Dispos 30:457–475. 4. Grostern RJ, Bryar PJ, Zimbric ML, Darjatmoko SR, 21. Feldman D, Pike JW, Glorieux FH. 2005. Vitamin D. 2nd ed. Lissauer BJ, Lindstrom MJ, Lokken JM, Strugnell SA, Al- Boston: Elsevier Academic Press , pp. 86-87 . bert DM. 2002. Toxicity and dose-response studies of 1"- 22. Theodoropoulos C, Demers C, Mirshahi A, Gascon-Barre M. 2001. 1,25-Dihydroxyvitamin D downregulates the rat hydroxyvitamin D2 in a retinoblastoma xenograft model. Arch 3 Ophthalmol 120:607–612. intestinal vitamin D3-25-hydroxylase CYP27A. Am J Physiol 5. Liu G, Wilding G, Staab MJ, Horvath D, Miller K, Dresen Endocrinol Metab 281:E315–E325. A, Alberti D, Arzoomanian R, Chappell R, Bailey HH. 2003. 23. Cheng JB, Levine MA, Bell NH, Mangelsdorf DJ, Russell DW. 2004. Genetic evidence that the human CYP2R1 enzyme Phase II study of 1"-hydroxyvitamin D2 in the treatment of advanced androgen-independent prostate cancer. Clin Cancer is a key vitamin D 25-hydroxylase. Proc Natl Acad Sci U S A Res 9:4077–4083. 101:7711–7715. 6. Knutson JC, Hollis BW, LeVan LW, Valliere C, Gould KG, 24. Gupta RP, He YA, Patrick KS, Halpert JR, Bell NH. 2005. CYP3A4 is a vitamin D-24- and 25-hydroxylase: Analysis of Bishop CW. 1995. Metabolism of 1"-hydroxyvitamin D2 to

DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 292 1604 CHOW ET AL.

structure function by site-directed mutagenesis. J Clin En- bile acid transport by 1",25-dihydroxyvitamin D3 via the vita- docrinol Metab 90:1210–1219. min D receptor. Mol Pharmacol 69:1913–1923. 25. Rahmaniyan M, Patrick K, Bell NH. 2005. Characterization 34. Nishida S, Ozeki J, Makishima M. 2009. Modulation of bile of recombinant CYP2C11: A vitamin D 25-hydroxylase and acid metabolism by 1"-hydroxyvitamin D3 administration in 24-hydroxylase. Am J Physiol Endocrinol Metab 288:E753– mice. Drug Metab Dispos 37:2037–2044. 760. 35. Ogura M, Nishida S, Ishizawa M, Sakurai K, Shimizu M, Mat- 26. Aiba I, Yamasaki T, Shinki T, Izumi S, Yamamoto K, Yamada suo S, Amano S, Uno S, Makishima M. 2009. Vitamin D3 S, Terato H, Ide H, Ohyama Y. 2006. Characterization of rat modulates the expression of bile acid regulatory genes and re- and human CYP2J enzymes as Vitamin D 25-hydroxylases. presses inflammation in bile duct-ligated mice. J Pharmacol Steroids 71:849–856. Exp Ther 328:564–570. 27. Upton RA, Knutson JC, Bishop CW, LeVan LW. 2003. Phar- 36. Sjoden G, Smith C, Lindgren U, DeLuca HF. 1985. 1"- macokinetics of doxercalciferol, a new vitamin D analogue Hydroxyvitamin D2 is less toxic than 1"-hydroxyvitamin D3 that lowers parathyroid hormone. Nephrol Dial Transplant in the rat. Proc Soc Exp Biol Med 178:432–436. 18:750–758. 37. Coburn JW, Tan AU, Jr., Levine BS, Mazess RB, Kyllo DM, 28. Masuda S, Jones G. 2006. Promise of vitamin D analogues Knutson JC, Bishop CW. 1996. 1"-Hydroxy-vitamin D2:A in the treatment of hyperproliferative conditions. Mol Cancer new look at an ‘old’ compound. Nephrol Dial Transplant (11 Ther 5:797–808. Suppl 3):153–157. 29. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Pro- 38. Zierold C, Mings JA, DeLuca HF. 2003. Regulation tein measurement with the Folin phenol reagent. J Biol Chem of 25-hydroxyvitamin D3–24-hydroxylase mRNA by 1,25- 193:265–275. dihydroxyvitamin D3 and parathyroid hormone. J Cell 30. Nagpal S, Na S, Rathnachalam R. 2005. Noncalcemic actions Biochem 88:234–237. of vitamin D receptor ligands. Endocr Rev 26:662–687. 39. Riganti C, Doublier S, Viarisio D, Miraglia E, Pescarmona 31. Thummel KE, Brimer C, Yasuda K, Thottassery J, Senn T, G, Ghigo D, Bosia A. 2009. Artemisinin induces doxorubicin Lin Y, Ishizuka H, Kharasch E, Schuetz J, Schuetz E. 2001. resistance in human colon cancer cells via calcium-dependent Transcriptional control of intestinal cytochrome P-4503A activation of HIF-1" and P-glycoprotein overexpression. Br J by 1",25-dihydroxy vitamin D3. Mol Pharmacol 60:1399– Pharmacol 156:1054–1066. 1406. 40. den Dekker E, Hoenderop JG, Nilius B, Bindels RJ. 2003. The 32. Saeki M, Kurose K, Tohkin M, Hasegawa R. 2008. Identifi- epithelial calcium channels, TRPV5 & TRPV6: From identification of the functional vitamin D response elements in the cation towards regulation. Cell Calcium 33:497–507. human MDR1 gene. Biochem Pharmacol 76:531–542. 41. Meyer MB, Zella LA, Nerenz RD, Pike JW. 2007. Characteriz- 33. Chen X, Chen F, Liu S, Glaeser H, Dawson PA, Hofmann AF, ing early events associated with the activation of target genes Kim RB, Shneider BL, Pang KS. 2006. Transactivation of rat by 1,25-dihydroxyvitamin D3 in mouse kidney and intestine apical sodium-dependent bile acid transporter and increased in vivo. J Biol Chem 282:22344–22352.

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 4, APRIL 2011 DOI 10.1002/jps 293

APPENDIX T1 Mouse Primer Sequences for qPCR APPENDIX T1 Gene Bank Number Forward (5’→ 3’ Sequence) Reverse (5’→ 3’ Sequence)

PPAR NM_011146.3 CATGCTTGTGAAGGATGCAAG TTCTGAAACCGACAGTACTGACAT

ApoE NM_009696.3 AAGCAACCAACCCTGGGAG TGCACCCAGCGCAGGTA

CETP NM_011125.2 GATGGTGTACGTGGCCTTTT TGGCCTCTAGCTTCAGCTTC

VLDLr NM_001161420.1 GAGCCCCTGAAGGAATGCC CCTATAACTAGGTCTTTGCAGATATGG

LDLr NM_010700.2 AGGCTGTGGGCTCCATAGG TGCGGTCCAGGGTCATCT

SR-B1 NM_016741.1 GGGAGCGTGGACCCTATGT CGTTGTCATTGAAGGTGATGT

Asbt NM_011388.2 GATAGATGGCGACATGGACCTC CAATCGTTCCCGAGTCAACC

Oatp1a4 NM_030687.1 CACGTCTGTAGTTGGGCTTATC CCCATAACTGCACATCCTACAC

Ost NM_145932.3 TACAAGAACACCCTTTGCCC CGAGGAATCCAGAGACCAAA

Ost NM_178933.2 GTATTTTCGTGCAGAAGATGCG TTTCTGTTTGCCAGGATGCTC

NPC1L1 NM_207242.2 TGGACTGGAAGGACCATTTCC GCGCCCCGTAGTCAGCTAT

Ntcp NM_011387.2 ATCTGACCAGCATTGAGGCTC CCGTCGTAGATTCCTTGCTGT

ABCA1 NM_013454.3 CGTTTCCGGGAAGTGTCCTA CTAGAGATGACAAGGAGGATGGA

ABCG5 NM_031884.1 TCAATGAGTTTTACGGCCTGAA GCACATCGGGTGATTTAGCA

ABCG8 NM_026180.2 TGCCCACCTTCCACATGTC ATGAAGCCGGCAGTAAGGTAGA

294

Bsep NM_021022.3 ACAGCACTACAGCTCATTCAGAG TCCATGCTCAAAGCCAATGATCA

Mrp2 NM_013806.2 GCTTCCCATGGTGATCTCTT CTTGGATTGTGGCTTCCAAG

Mrp3 NM_029600.3 CGCTCTCAGCTCACCATCAT GGTCATCCGTCTCCAAGTCA

Mrp4 NM_001163675.1 GGTTGGAATTGTGGGCAGAA TCGTCCGTGTGCTCATTGAA

TRPV6 NM_022413.4 ATCGATGGCCCTGCCAACT CAGAGTAGAGGCCATCTTGTTGCTG

Cyp3a11 NM_007818.3 TTTGGTAAAGTACTTGAGGCAGA CTGGGTTGTTGAGGGAATC

Cyp8b1 NM_010012.3 GCCTTCAAGTATGATCGGTTCCT GATCTTCTTGCCCGACTTGTAGA

HMG CoA NM_008255.2 CAAGGAGCATGCAAAGACAA GCCATCACAGTGCCACATAC Reductase

Sult2a1 NM_001111296.2 CTGGCTGTCCATGAGAGAAT GGCTTGGAAAGAGCTGTACT

295

APPENDIX T2 Antibodies Chart APPENDIX T2 Antibody, Species/Tissue % SDS- Incubation Molecular Primary Ab Secondary Ab Company Cat# Tested PAGE gel condition Weight (kDa) and Dilution and Dilution mouse and rat GAPDH anti-mouse; Abcam Inc. intestine, liver, and any any 37 1: 10000 (ab9245) 1:10000 kidney mouse and rat Santa Cruz Villin (C-19) intestine, liver, and any any 95 1:5000 anti-goat; 1:5000 Biotechnology kidney Lamin-B Abcam Inc. mouse liver any any 72 1: 5000 anti-rabbit; 1:5000 (ab45848) Thermo mouse and rat VDR (9A7) Fisher intestine, liver, and 10 95°C for 5 min 60 1:1000 anti-rat; 1:2000 Scientific kidney Dr. Paul A. mouse and rat Asbt 12 37°C for 15 min 48 1:5000 anti-rabbit; 1:5000 Dawson intestine and kidney Alpha Oat1 (OAT11-A) Diagnostic mouse and rat kidney 10 37°C for 15 min 72 1:1000 anti-rabbit; 1:2000 Intl. Inc. Dr. Allan W. Oatp1a1 mouse and rat liver 10 37°C for 15 min 80 1:2000 anti-rabbit; 1:2000 Wolkoff Dr. Allan W. Oatp1a4 mouse and rat liver 10 37°C for 15 min 94 1:2000 anti-rabbit; 1:2000 Wolkoff Dr Richard B. Oatp1b2 mouse and rat liver 10 37°C for 15 min 80 1:2000 anti-rabbit; 1:2000 Kim Dr. Wolfgang mouse and rat kidney PepT1 10 37°C for 15 min 95 1:5000 anti-rabbit; 1:5000 Sadee and rat intestine Dr. Bruno Bsep mouse and rat liver 7.5 37°C for 15 min 160 1:2000 anti-rabbit; 1:2000 Stieger Mrp2 (ALX- Alexis rat intestine, liver, anti-mouse; 7.5 37°C for 15 min 180 1:1000 801-016-C250) Biochemicals and kidney 1:2000 mouse and rat Dr. Yuichi Mrp3 intestine, liver, and 7.5 37°C for 15min 170 1:2000 anti-rabbit; 1:5000 Sugiyama kidney 296

mouse and rat Dr. John D. Mrp4 intestine, liver, and 7.5 37°C for 15 min 160 1:2000 anti-rabbit; 1:2000 Schuetz kidney mouse and rat anti-mouse; P-gp (C-19) Abcam Inc. intestine, liver, and 7.5 37°C for 15 min 170 1:1000 1:2000 kidney Santa Cruz Cyp24 (H-87) mouse and rat kidney 10 37°C for 15 min 50 1:1000 anti-goat; 1:2000 Biotechnology mouse and rat Cyp3a2 BD intestine, liver, and 10 37°C for 15 min 55 1:5000 anti-rabbit; 1:5000 (458223) Biosciences kidney Santa Cruz Cyp7a1 (N-17) mouse and rat liver 10 37°C for 15 min 50 1: 1000 anti-goat; 1:2000 Biotechnology

297

APPENDIX T3 Body weight of fxr(+/+) (wild-type) mice treated with 1,25(OH)2D3 i.p. q2d (every other day) for 8 days. Data represents mean ± SD. * denotes P < 0.05 using two-tailed Student t test compared to vehicle control. Batch 3 (Jan 19, 2009) Batch 4A (Mar 30, 2009) Batch 5A (Nov 15, 2010) Treatment Vehicle 2.5 µg/kg Vehicle 2.5 µg/kg Vehicle 2.5 µg/kg Day Control 1,25(OH)2D3 Control 1,25(OH)2D3 Control 1,25(OH)2D3 1 0.965 ± 0.041 0.967 ± 0.026 0.992±0.018 0.961 ± 0.011* 1.001 ± 0.021 1.002 ± 0.013 2 0.958 ± 0.048 0.964 ± 0.057 0.989±0.030 0.961 ± 0.019 0.990 ± 0.019 0.974 ± 0.015 3 0.961 ± 0.048 0.953 ± 0.034 0.985±0.036 0.952 ± 0.021 0.990 ± 0.022 0.967 ± 0.018 4 0.983 ± 0.032 0.957 ± 0.053 0.977±0.031 0.945 ± 0.024 0.977 ± 0.029 0.955 ± 0.026 5 1.010 ± 0.032 0.927 ± 0.053* 0.974±0.031 0.911 ± 0.022* 0.992 ± 0.026 0.953 ± 0.026* 6 1.022 ± 0.023 0.945 ± 0.044* 0.969±0.027 0.922 ± 0.021* 0.992 ± 0.024 0.955 ± 0.019* 7 1.025 ± 0.024 0.906 ± 0.066* 0.974±0.022 0.905 ± 0.042* 0.980 ± 0.021 0.898 ± 0.032* 8 1.030 ± 0.022 0.934 ± 0.060* 0.972±0.023 0.921 ± 0.049 0.991 ± 0.028 0.943 ± 0.038* * denotes P < 0.05 compared to vehicle control using Mann-Whitney U test APPENDIX T3 APPENDIX T4 Body weight of fxr(-/-) mice treated with 1,25(OH)2D3 i.p. q2d (every other day) for 8 days. Data represents mean ± SD. * denotes P < 0.05 using two-tailed Student t test compared to vehicle control. Batch 1 (Nov 8, 2007) Batch 2 (Dec 4, 2007) Batch 4B (Mar 24, 2009) Batch 5B (Nov15, 2010) Treatment Vehicle 2.5 µg/kg Vehicle 2.5 µg/kg Vehicle 2.5 µg/kg Vehicle 2.5 µg/kg Day Control 1,25(OH)2D3 Control 1,25(OH)2D3 Control 1,25(OH)2D3 Control 1,25(OH)2D3 1 0.993 ± 0.0195 0.972 ± 0.021 0.988 ± 0.013 0.980 ± 0.011 0.994 ± 0.032 0.986 ± 0.008 0.986±0.018 0.981 ± 0.014 2 1.021 ± 0.0207 0.976 ± 0.032* 1.001 ± 0.007 0.984 ± 0.024 1.007 ± 0.016 0.982 ± 0.014 0.994±0.014 0.982 ± 0.032 3 1.031 ± 0.0100 0.970 ± 0.026* 0.993 ± 0.014 0.964 ± 0.012* 0.996 ± 0.018 0.946 ± 0.027 0.992±0.023 0.951 ± 0.020* 4 1.035 ± 0.0252 0.959 ± 0.016* 1.002 ± 0.016 0.975 ± 0.014* 1.001 ± 0.038 0.982 ± 0.013 1.007±0.034 0.978 ± 0.019 5 1.040 ± 0.0319 0.923 ± 0.026* 0.992 ± 0.014 0.933 ± 0.013* 1.007 ± 0.022 0.933 ± 0.017* 1.015±0.031 0.947 ± 0.016* 6 1.035 ± 0.0315 0.960 ± 0.024* 0.995 ± 0.019 0.952 ± 0.015* 1.006 ± 0.027 0.945 ± 0.027* 1.010±0.026 0.961 ± 0.031* 7 1.038 ± 0.0292 0.942 ± 0.019* 0.999 ± 0.011 0.918 ± 0.013* 1.022 ± 0.047 0.894 ± 0.017* 1.021±0.029 0.959 ± 0.028* 8 1.020 ± 0.0201 0.974 ± 0.029* 0.996 ± 0.016 0.923 ± 0.032* 1.017 ± 0.049 0.932 ± 0.023* 1.019±0.026 0.956 ± 0.048* * denotes P < 0.05 compared to vehicle control using Mann-Whitney U test APPENDIX T4

298

APPENDIX T5 Body weight of wild-type mice treated with 1,25(OH)2D3 i.p. q2d (every other day) for 8 days in the last week of a 3 weeks pretreatment of Western diet and its cumulative food intake (g) per mouse. Data represents mean ± SD. * and † denote P < 0.05 using two-tailed Student t test compared to normal diet and Western diet vehicle control, respectively. Batch 6 (Oct 1, 2009) Batch 7A (Mar 2, 2010) Normal Diet; Western diet; Normal Diet; Western diet; Treatment Western diet; Western diet; Vehicle 2.5 µg/kg Vehicle 2.5 µg/kg Day Vehicle Control Vehicle Control Control 1,25(OH)2D3 Control 1,25(OH)2D3 1 0.975 ± 0.009 0.975 ± 0.043 0.980 ± 0.016 0.995±0.036 0.970 ± 0.037 1.002 ± 0.020 2 0.974 ± 0.009 0.985 ± 0.021 0.959 ± 0.029* 0.991±0.059 0.982 ± 0.020 0.985 ± 0.015 3 0.979 ± 0.015 0.992 ± 0.014 0.897 ± 0.042* 1.001±0.050 0.995 ± 0.025 0.952 ± 0.029* 4 0.991 ± 0.006 1.008 ± 0.026 0.905 ± 0.072* 1.010±0.043 1.009 ± 0.032 0.949 ± 0.022* 5 0.982 ± 0.023 1.024 ± 0.030† 0.841 ± 0.084* 1.018±0.038 1.010 ± 0.026 0.924 ± 0.024* 6 0.993 ± 0.017 1.033 ± 0.036† 0.861 ± 0.106* 1.013±0.026 1.010 ± 0.047 0.944 ± 0.022* 7 0.988 ± 0.014 1.033 ± 0.039† 0.824 ± 0.122* 1.025±0.020 1.014 ± 0.039 0.925 ± 0.031* 8 1.001 ± 0.017 1.037 ± 0.038 0.842 ± 0.146* 1.028±0.014 1.025 ± 0.047 0.953 ± 0.041* † indicates P < 0.05 compared to normal diet vehicle control using Mann-Whitney U test * denotes P < 0.05 compared to Western diet fed vehicle control using Mann-Whitney U test APPENDIX T5 APPENDIX T6 Cumulative food intake (g) per mouse of wild-type mice treated with 1,25(OH)2D3 i.p. q2d (every other day) for 8 days in the last week of a 3 weeks pretreatment of Western diet. Data represents mean ± SD. * and † denote P < 0.05 using two-tailed Student t test compared to normal diet and Western diet vehicle control, respectively. Batch 7B (Mar 2, 2010) Western diet; Treatment Normal Diet; Western diet; 2.5 µg/kg Day Vehicle Control Vehicle Control 1,25(OH)2D3 2 6.18 ± 1.38 4.25 ± 0.75 5.62 ± 0.43 4 12.68 ± 1.48 9.81 ± 1.65 9.90 ± 0.97 6 19.08 ± 1.66 15.23 ± 1.79 16.70 ± 1.85 8 25.40 ± 1.70 20.43 ± 2.10 21.82 ± 1.01 APPENDIX T6

299

APPENDIX T7 Body weight of shp(-/-) mice treated with 1,25(OH)2D3 i.p. q2d (every other day) for 8 days in the last week of a 3 weeks pretreatment of Western diet. Data represents mean ± SD. * denotes P < 0.05 using two-tailed Student t test compared to vehicle control. Batch 8 (July 4, 2011) Western diet; Western diet; Treatment Day 2.5 µg/kg Vehicle Control 1,25(OH)2D3 2 1.009 ± 0.015 1.005 ± 0.011 4 1.028 ± 0.026 0.983 ± 0.028 6 1.035 ± 0.031 0.943 ± 0.044* 8 1.039 ± 0.041 0.894 ± 0.024* * denotes P < 0.05 compared to Western diet fed vehicle control using Mann-Whitney U test APPENDIX T7

APPENDIX T8 Plasma calcium, phosphorus, and alanine aminotransferase (ALT) and liver triglyceride levels in fxr(+/+) and fxr(-/-) mice. Data represented mean ± S.E.M. (n=4-6). Calcium and phosphorus in systemic plasma, diluted 350-fold with 1% HNO3, was determined by inductively coupled plasma atomic emission spectroscopy, and showed about a 23-30% increase in calcium in mice treated with 1,25(OH)2D3 treatment. Alanine transaminase (ALT) leakage in systemic plasma was measured with ALT kit (Bioquant, Nashville, TN). 1,25(OH)2D3 treated fxr(-/-) mice showed a 81% decrease in ALT. Concentrations of liver triglyceride from samples of liver cholesterol extraction procedure were measured with a Triglyceride commercial kit (Thermo Scientific, Rockford, IL). Liver triglyceride concentration in fxr(-/-) vehicle control was 2-fold higher than wild-type control. fxr(+/+) (Batch 4A) fxr(-/-) (Batch 1) Vehicle control 1,25(OH)2D3 (2.5 µg/kg) Vehicle control 1,25(OH)2D3 (2.5 µg/kg) Plasma Calcium (mg/dl) 9.6 ± 0.2 12.5 ± 0.2* 8.1 ± 0.4# 10.0 ± 0.6* Plasma Phosphorus (mg/dl) 19.2 ± 0.9 18.7 ± 0.7 19.8 ± 1.3 17.3 ± 0.8 Plasma ALT (IU/ml) 10.2 ± 0.9 11.4 ± 1.2 173.2 ± 30.0 32.4 ± 4.7* Liver Triglyceride (mg/g) 17.1 ± 4.0 15.7 ± 2.6 48.6 ± 14.1# 36.2 ± 6.0 * denotes P < 0.05 compared to vehicle control using Mann-Whitney U test # indicates P < 0.05 compared to fxr(+/+) control using Mann-Whitney U test APPENDIX T8

300

APPENDIX T9 Plasma ALT and liver triglyceride levels in C57BL6, fxr(-/-), and shp(-/-) mice fed with a Western diet. Data represented mean ± S.E.M. (n=4- 8). The level of ALT was increased with Western diet in wild-type, but not fxr(-/-) and shp(-/-) mice. Mice treated with 1,25(OH)2D3 had not effect on ALT changes compared to its vehicle control. Liver triglyceride concentrations were increased in mice fed with a Western diet, but showed a decrease in trend when treated with 1,25(OH)2D3. (From Batch 7)

Normal Diet; Vehicle Control Western Diet; Vehicle Control Western Diet; 1,25(OH)2D3 (2.5 µg/kg) Wild-type mice (Batch 7) ALT (IU/ml) 7.2 ± 1.6 23.5 ± 2.3† 17.2 ± 2.7 Liver Triglyceride (mg/g) 9.7 ± 5.1 28.5 ± 16.6† 17.1 ± 12.1 Plasma Calcium (mg/dl) 9.1 ± 0.1 9.0 ± 0.1 12.8 ± 0.5* Plasma Phosphorus (mg/dl) 17.3 ± 0.9 21.1 ± 1.0† 19.5 ± 0.6 fxr(-/-) mice (Holly samples) ALT (IU/ml) 96.3 ± 9.0 81.4 ± 59.9 89.2 ± 54.3 Liver Triglyceride (mg/g) 19.7 ± 1.0 53.5± 4.4† 31.3 ± 8.9 shp(-/-) mice (Batch 8) ALT (IU/ml) 12.0 ± 2.0 23.7 ± 4.5 46.2 ± 14.7 Liver Triglyceride (mg/g) 4.2 ± 1.0 35.0 ± 7.7† 13.9 ± 3.5* † indicates P < 0.05 compared to normal diet vehicle control using Mann-Whitney U test * denotes P < 0.05 compared to Western diet fed vehicle control using Mann-Whitney U test APPENDIX T9

301

Human

Mouse

APPENDIX F1 APPENDIX F1 Comparsion of human SHP (hSHP) and mouse SHP (mSHP) promoter sequences used in Chapter 5 using BLAST Aligment software (http://blast.ncbi.nlm.nih.gov/). The result shows 78% homology between human and mouse SHP promoter sequences.

302

(A)

6 Liver fxr(+/+) Control 5 fxr(+/+) 1,25(OH)2D3 fxr(-/-) Control 4 fxr(-/-) 1,25(OH) D 2 3 3 †

2 * 1

Relative mRNA Expression 0 ApoE CETP VLDLr LDLr SR-B1 ABCA1 ABCG5 ABCG8 Cyp8b1HMG CoA Reductase

(B)

6 Ileum fxr(+/+) Control

5 fxr(+/+) 1,25(OH)2D3 fxr(-/-) Control 4 fxr(-/-) 1,25(OH) D 2 3 † 3 2 †

1 Relative mRNA Expression 0 NPC1L1 ABCA1 ABCG5 ABCG8

APPENDIX F2 APPENDIX F2 Changes in (A) hepatic apolipoprotein E (ApoE), cholesteryl ester transfer protein (CETP), very low density lipoprotein receptor (VLDLr), low density lipoprotein receptor (LDLr), scavenger receptor class B type 1 (SR-B1), cholesterol efflux transporters (ABCA1, ABCG5, and ABCG8), Cyp8b1, and 3-hydroxy-3-methyl-glutaryl-CoA reductase (HMG-CoA reductase) and (B) ileal Niemann-Pick C1-Like 1 (NPC1L1), ABCA1, ABCG5, and ABCG8 mRNA expressions in both fxr(+/+) and fxr(-/-) mice. The symbols † and * denote significant differences, using Mann- Whitney U test, between the two controls, and between the treated vs. vehicle control within the wild-type or fxr(-/-) mice, respectively. Data represent the mean ± SEM (n = 4-8).

303

6 WildtypeWild Type Mouse Liver Liver Normal Diet Control 5 High Cholesterol Diet Control High Cholesterol Diet 1,25(OH)2D3 4

3 † * † † * * † 2 * † † * * 1 Expression mRNA Relative 0 ApoE CETP VLDLr LDLr SR-B1 ABCA1 ABCG5 ABCG8 Cyp8b1HMG CoA Reductase

14 fxr(-/-)fxr(-/-) MouseMouse Liver Liver Normal Diet Control 12 High Cholesterol Diet Control

High Cholesterol Diet 1,25(OH)2D3 10

8 † 6 † * † † * 4 † † * † 2 Expression mRNA Relative 0 ApoE CETP VLDLr LDLr SR-B1 ABCA1 ABCG5 ABCG8 Cyp8b1HMG CoA Reductase .

6 shp(-/-)shp(-/-) MouseMouse Liver Liver Normal Diet Control 5 High Cholesterol Diet Control High Cholesterol Diet 1,25(OH)2D3 4

3 † * † 2 † † * 1 Relative mRNA Expression Relative 0 ApoE CETP VLDLr LDLr SR-B1 ABCA1 ABCG5 ABCG8 Cyp8b1HMG CoA Reductase

APPENDIX F3 APPENDIX F3 Changes in hepatic ApoE, CETP, VLDLr, LDLr, SR-B1, ABCA1, ABCG5, ABCG8, Cyp8b1, and HMG-CoA reductase mRNA expressions in wild-type, fxr(-/-) and shp(-/-) mice prefed the Western diet for 3 weeks. The symbols † and * denote significant differences using Mann- Whitney U test between the two controls, and between the treated vs. vehicle control, respectively. Data represent the mean ± SEM (n = 4-8).

304

Wild Type Mouse Ileum 6 Wildtype Mouse Ileum Normal Diet Control 5 High Cholesterol Diet Control High Cholesterol Diet 1,25(OH)2D3 4 † *

3 † * † 2

1 Relative mRNA Expression 0 NPC1L1 ABCA1 ABCG5 ABCG8

14 fxr(-/-) MouseMouse Ileum Ileum Normal Diet Control 12 High Cholesterol Diet Control High Cholesterol Diet 1,25(OH) D 10 2 3

8 6 † P=0.05 P=0.05 4 2 Relative mRNA Expression Relative mRNA 0 NPC1L1 ABCA1 ABCG5 ABCG8

shp(-/-) MouseMouse Ileum Ileum 6 Normal Diet Control 5 High Cholesterol Diet Control † * High Cholesterol Diet 1,25(OH)2D3 4

2 †

1 Relative mRNA ExpressionRelative mRNA 0 APPENDIX F4 NPC1L1 ABCA1 ABCG5 ABCG8

APPENDIX F4 Changes in ileal NPC1L1, ABCA1, ABCG5, and ABCG8 mRNA expressions in wild-type, fxr(-/-) and shp(-/-) mice prefed the Western diet for 3 weeks. The symbols † and * denote significant differences using Mann-Whitney U test between the two controls, and between the treated vs. vehicle control, respectively. Data represent the mean ± SEM (n = 4-8). The results show that the decrease in intestinal ABCA1 may be responsible for the reduction of plasma cholesterol in wild- type and shp(-/-) mice.

305