This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chiang, J. Y. L. Search for Related Content PubMed PubMed Citation Articles by Chiang, J. Y. L.

Bile Acid Regulation of Gene Expression: Roles of Nuclear Hormone Receptors

Department of Biochemistry and Molecular Pathology, Northeastern Ohio Universities College of Medicine, Rootstown, Ohio 44272

Correspondence: Address all correspondence and requests for reprints to: John Y. L. Chiang, Ph.D., Department of Biochemistry and Molecular Pathology, Northeastern Ohio Universities College of Medicine, 4209 SR 44, P.O. Box 95, Rootstown, Ohio 44272. E-mail: jchiang{at}neoucom.edu

	Abstract

Top Abstract I. Introduction II. Bile Acid Synthesis... III. Nuclear Hormone Receptor... IV. Molecular Mechanisms of... V. Drug Therapies Targeted... VI. Conclusion and Future... Note Added in Proof References

 
Bile acids derived from cholesterol and oxysterols derived from cholesterol and bile acid synthesis pathways are signaling molecules that regulate cholesterol homeostasis in mammals. Many nuclear receptors play pivotal roles in the regulation of bile acid and cholesterol metabolism. Bile acids activate the farnesoid X receptor (FXR) to inhibit transcription of the gene for cholesterol 7-hydroxylase, and stimulate excretion and transport of bile acids. Therefore, FXR is a bile acid sensor that protects liver from accumulation of toxic bile acids and xenobiotics. Oxysterols activate the liver orphan receptors (LXR) to induce cholesterol 7-hydroxylase and ATP-binding cassette family of transporters and thus promote reverse cholesterol transport from the peripheral tissues to the liver for degradation to bile acids. LXR also induces the sterol response element binding protein-1c that regulates lipogenesis. Therefore, FXR and LXR play critical roles in coordinate control of bile acid, cholesterol, and triglyceride metabolism to maintain lipid homeostasis. Nuclear receptors and bile acid/oxysterol-regulated genes are potential targets for developing drug therapies for lowering serum cholesterol and triglycerides and treating cardiovascular and liver diseases.

I. Introduction

II. Bile Acid Synthesis and Regulation

A. Bile acid biosynthetic pathways

B. Regulation of bile acid synthesis and cholesterol homeostasis

C. Bile acid synthesis deficiency

III. Nuclear Hormone Receptor Regulation of Bile Acid Synthesis

A. Structure and function of nuclear hormone receptors

B. Nuclear receptors involved in regulation of genes in bile acid synthesis

C. Nuclear receptor regulation of cholesterol homeostasis

IV. Molecular Mechanisms of Regulation of Bile Acid Metabolism

A. Nuclear receptor-mediated mechanism

B. Cell-signaling mechanism

V. Drug Therapies Targeted to Nuclear Receptors and Genes in Bile Acid Metabolism

A. Bile acid synthesis

B. Bile acid transport

C. Reverse cholesterol transport

D. Cholesterol absorption

VI. Conclusion and Future Perspectives

	I. Introduction

Top Abstract I. Introduction II. Bile Acid Synthesis... III. Nuclear Hormone Receptor... IV. Molecular Mechanisms of... V. Drug Therapies Targeted... VI. Conclusion and Future... Note Added in Proof References

 
CONVERSION OF CHOLESTEROL to bile acids in the liver and biliary excretion of cholesterol for eventual disposal in stool are two major routes for removing excess cholesterol from the body. Recent studies have shown that bile acids not only serve as the physiological detergents that facilitate the absorption, transport, and distribution of lipid-soluble vitamins and dietary fats, but also are the signaling molecules that activate nuclear receptors and regulate bile acid and cholesterol metabolism. In addition, bile acids induce the cytochrome P450 3A (CYP3A) family of cytochrome P450 enzymes that detoxify bile acids, drugs, and xenobiotics in the liver and intestine, and also induce hepatocyte apoptosis. Bile acids are synthesized in the liver, excreted into the bile, reabsorbed in the ileum, and transported back to the liver via portal circulation to inhibit bile acid synthesis by suppressing the gene encoding the rate-limiting enzyme, cholesterol 7-hydroxylase (CYP7A1) (1). The mechanisms of bile acid feedback regulation have been studied in animal and tissue culture models for more than three decades. Recent studies suggest that bile acids are able to activate a bile acid receptor, farnesoid X receptor (FXR), which regulates the target genes in bile acid synthesis, transport, and cholesterol metabolism (2, 3, 4, 5, 6). Oxysterols are derived from cholesterol and bile acid biosynthetic pathways and are potent ligands that activate oxysterol receptor, liver X receptor (LXR), which induces genes involved in reverse cholesterol transport (7, 8, 9). FXR and LXR may coordinately regulate bile acid synthesis and cholesterol homeostasis (10, 11, 12, 13). This review will focus on the molecular mechanisms of nuclear receptor regulation of bile acid and cholesterol homeostasis. Diseases caused by bile acid synthesis defects and the potential drug therapies targeted to nuclear receptors for lowering serum cholesterol levels will also be discussed.

	II. Bile Acid Synthesis and Regulation

Top Abstract I. Introduction II. Bile Acid Synthesis... III. Nuclear Hormone Receptor... IV. Molecular Mechanisms of... V. Drug Therapies Targeted... VI. Conclusion and Future... Note Added in Proof References

 
A. Bile acid biosynthetic pathways
The conversion of cholesterol to bile acids occurs exclusively in hepatocytes by a cascade of 12 reactions catalyzed by enzymes located in the endoplasmic reticulum, mitochondria, cytosol, and peroxisomes. Detailed descriptions of the reactions and enzymes involved in bile acid biosynthetic pathways can be found in recent reviews (1, 14, 15, 16, 17). Figure 1 shows several of the intermediates and important regulatory enzymes in two major bile acid biosynthetic pathways. The main bile acid biosynthetic (classic or neutral) pathway is initiated by CYP7A1, which is only expressed in the liver, whereas the alternative (or acidic) pathway is initiated by sterol 27-hydroxylase (CYP27A1), which is expressed in many tissues. In the classic pathway, modifications of the steroid nucleus, including hydroxylation at 7- and 12-positions, epimerization of the 3ß-hydroxyl group, and saturation of the steroid nucleus, precede the oxidative cleavage of a three-carbon side chain. In the alternative pathway, oxidative cleavage of the side chain precedes the modifications of the steroid nucleus. Cholic acid (CA) and chenodeoxycholic acid (CDCA) are two major primary bile acids found in human bile.

View larger version (28K): [in this window] [in a new window]   Figure 1. Bile acid biosynthetic pathways in the liver. Two major bile acid biosynthetic pathways are shown. Only major regulatory enzymes, CYP7A1, CYP8B1, CYP27A1, CYP7B1, and 3ß-HSD, and their substrates and products are shown.

2. The alternative or acidic pathway.
The alternative pathway was originally suggested by the identification of many acidic intermediates, which were not intermediates of the classic pathway (21, 22). The alternative pathway produces mainly CDCA (23, 24). In this pathway, CYP27A1 converts cholesterol to both 27-hydroxycholesterol and 3ß-hydroxy-5-cholestenoic acid (25). Oxysterol 7-hydroxylase (CYP7B1) then converts these two intermediates to 7,27-dihydroxycholesterol and 3ß,7-dihydroxy-5-cholestenoic acid, respectively. It is believed that the same enzymes of the classic pathway catalyze subsequent modifications of the sterol nucleus (26). Recent studies suggest that the acidic pathway also produces CA (27, 28, 29).

The relative contribution of the acidic pathway to overall bile acid synthesis is not certain. Metabolites of the acidic pathway are accumulated in patients with chronic liver diseases and are an indication of a larger contribution of this pathway to bile acid synthesis (21). The acidic pathway may contribute as much as 50% of total bile acid synthesis in primary cultures of rat and human hepatocytes (30). However, the alternative pathway contributes only less than 18% of total bile acid synthesis in humans (31). The neutral pathway is highly regulated and is stimulated by bile fistula or by feeding cholestyramine, a bile acid-binding resin, whereas the acidic pathway is not induced as much (32, 33, 34). In Cyp7a1-/- mice, bile acid synthesis is markedly reduced and the acidic pathway may be activated after weaning to provide 7-hydroxylated bile acids (35, 36). In contrast, bile acid synthesis, pool size, and composition are not altered in Cyp7b1-/- mice (37), and Cyp7a1 expression is increased to maintain bile acid homeostasis. These genetic knockout experiments support the suggestion that the neutral pathway involving CYP7A1 is the major regulated pathway, whereas the acidic pathway involving CYP7B1 is a constitutive pathway (37).

B. Regulation of bile acid synthesis and cholesterol homeostasis
The rate of bile acid synthesis parallels the activity of CYP7A1, which is the only rate-limiting enzyme of the bile acid biosynthetic pathway (19). Interruption of enterohepatic circulation of bile acids by biliary diversion or treatment with bile acid sequestrants increases the rate of bile acid synthesis and the activity of CYP7A1 by about 3- to 4-fold. Intraduodenal infusion of bile acids inhibits the rate of bile acid synthesis to the normal level (19). Enterohepatic circulation of bile acids is the most important physiological mechanism for controlling the overall rate of bile acid biosynthesis (38).

1. Bile acid feedback regulation of bile acid synthesis.
Bile acids excreted from the liver are reabsorbed in the intestine and transported back to the liver by a process called enterohepatic circulation of bile (39, 40). Conjugated bile acids synthesized in the liver are excreted into the bile canaliculi and stored in the gallbladder. After each meal, gallbladder contraction releases bile acids into the intestine for digestion of fats. Portions of CA and CDCA are converted to the secondary bile acids, deoxycholic acid (DCA) and lithocholic acid (LCA), respectively, by 7-dehydroxylase in the bacterial flora. These bile acids, with the exception of LCA, are efficiently reabsorbed in the ileum (41) and transported back to hepatocytes via portal venous circulation (42, 43, 44). Bile acids bind to hepatic bile acid-binding proteins and are transported to canalicular membrane for secretion into bile (45). This process is repeated several times after each meal and reabsorbs about 95% of bile acids in humans. The remaining 5% lost in feces is replenished by de novo bile acid synthesis. The details of bile acid transport systems in hepatocytes and intestine can be found in recent reviews (40, 46, 47, 48, 49).

2. Cholesterol homeostasis in the liver.
Figure 2 illustrates the central role that the liver plays in maintaining cholesterol homeostasis. Major pathways for input and output of cholesterol are shown. Four major cholesterol input mechanisms in the liver are 1) uptake serum cholesterol esters by a low-density lipoprotein (LDL) receptor-mediated endocytosis; 2) reverse cholesterol transport from peripheral tissues to the liver by the selective uptake of high-density lipoprotein (HDL) by the scavenger receptor subtype B1 (SR-B1) (50, 51); 3) absorption of dietary cholesterol in intestine and transport to the liver as chylomicrons (CM) by LDL receptors-mediated mechanism; and 4) de novo synthesis of cholesterol from acetyl-coenzyme A (CoA).

View larger version (84K): [in this window] [in a new window]   Figure 2. Bile acid synthesis and cholesterol homeostasis. Liver synthesizes cholesterol from acetyl-CoA, and HMG-CoA reductase is the rate-limiting enzyme of the pathway. The cholesterol pool in the liver is contributed by four input mechanisms. Serum cholesterol esters (CE) carried by LDL and IDL are taken up into the liver by LDL receptor (LDLR) or LDL receptor-related protein (LRP)-mediated endocytosis. Oxidized LDL is taken up into peripheral tissues by scavenger receptors SR-A1 and CD34. CEs are hydrolyzed to free cholesterol by cholesterol ester hydrolase (CEH) in peripheral tissues. Excess cholesterol is effluxed from peripheral tissues by ABCA1 transporter to form HDL. Lecithin-cholesterol acyltransferase (LCAT) converts cholesterol to CE, which is selectively taken up into the liver by HDL receptor, SR-B1. Dietary free cholesterol (FC) is absorbed into intestine and is reesterified to CEs by acyl-CoA-cholesterol acyltransferase 2 (ACAT2). CEs, ApoB48, and triglycerides are assembled to form chylomicron (CM). Triglycerides in CM are hydrolyzed by lipoprotein lipase (LPL) in the capillary of the adipose and muscle tissues to form free fatty acids, and CM is converted to chylomicron remnants (CMR). CEs in CMRs are taken up into liver by LDL receptor and LRP (apoE receptors). Two mechanisms are involved in output of cholesterol. Of the daily cholesterol catabolized, about 50% is converted to bile acids, which facilitate the excretion of 40% cholesterol into bile. The canalicular transport system is illustrated in detail in Fig. 4. Bile acids, cholesterol, and phospholipids form mixed micelles in the gallbladder (not shown), and are secreted into the intestine after each meal. About 95% of the bile acids are reabsorbed in the ileum, excreted into portal circulation, and up-taken into hepatocytes by sodium-dependent taurocholate cotransport peptide (NTCP). CEs, triglycerides, and ApoB100 are assembled to form VLDL in the liver. VLDL excreted in the serum is subsequently converted to IDL, LDL, and oxidized LDL. They are taken up by scavenger receptors, SR-A1 or SR-B1, into macrophages for disposal or into liver or adrenal for synthesis of bile acids or sex hormones.

Hydrophobic bile acids are toxic if accumulated in large quantities in hepatocytes. Therefore, bile acid synthesis and transport must be tightly regulated. Cholesterol is important for synthesis of bile acids, biological membranes, and steroid hormones, and its homeostasis needs to be maintained in tissues. The liver plays a central role in maintaining bile acid and cholesterol homeostasis. Interruption of the enterohepatic circulation of bile acids leads to an increase in bile acid synthesis and a reduction of plasma LDL cholesterol concentration (52, 53). Increased input of cholesterol and decreased output of bile acids may cause hypercholesterolemia, atherosclerosis, cholestasis, and cholelithiasis in humans (38, 46, 54).

3. Oxysterol regulation of cholesterol homeostasis.
Oxysterols are potent regulators of cholesterol synthesis and lipid metabolism (55). Oxysterols are derived from cholesterol, and the intermediates of the cholesterol and bile acid synthesis pathways by either enzymatic or nonenzymatic oxidations (7, 56). The most abundant oxysterols in human plasma are 27-hydroxycholesterol, 24(S)-hydroxycholesterol, and 7-hydroxycholesterol, which are generated predominately by CYP27A1 in the lung (57), sterol 24-hydroxylase in the brain (58, 59), and CYP7A1 in the liver, respectively. Another abundant oxysterol, 25-hydroxycholesterol, is synthesized by microsomal sterol 25-hydroxylase, a noncytochrome P450 enzyme (60). Oxysterols regulate cholesterol and fatty acid syntheses through a mechanism involving sterol response element-binding proteins (SREBPs) (61, 62, 63). When oxysterol levels are low in cells, SREBP is translocated from endoplasmic reticulum to the Golgi where the N-terminal 58-kDa fragment is cleaved by sterol-sensitive proteases, which are regulated by SREBP cleavage-activating protein. The matured SREBP enters the nucleus and binds to the sterol response elements of LDL receptor, 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase, and other genes in cholesterol and fatty acid synthesis (63, 64, 65, 66, 67). In liver and peripheral tissues, CYP7B1 hydroxylates 27- or 25-hydroxycholesterol, whereas another 7-hydroxylase, CYP39A1, hydroxylates 24-hydroxycholesterol (68). It has been reported that the recombinant human CYP7A1 can function as an CYP7B1 of 20S-, 24-, 25-, or 27-hydroxycholesterol (69, 70). Conversion of monohydroxycholesterols to dihydroxycholesterols reduces the toxicity of monohydroxyl oxysterols. In atherosclerotic plaques, 27-hydroxycholesterol, 7-ketocholesterol, and 7ß-, and 7-hydroxycholesterol are the most abundant oxysterols, which cause foam cell formation from macrophages and lead to atherosclerosis in humans (56). Some oxysterols in peripheral tissues are excreted to circulation, transported to the liver, and converted to bile acids. This is a process analogous to reverse cholesterol transport and has been suggested as a defense against atherosclerosis in humans (71, 72).

C. Bile acid synthesis deficiency
1. Inborn errors of bile acid biosynthesis.
Several inborn errors of bile acid synthesis have been described in infants and children with various clinical presentations including advanced liver diseases, neonatal hepatitis, progressive cholestasis, and biliary atresia (73, 74, 75). Primary defects in bile acid synthesis may result in decreased bile formation, malabsorption of fat-soluble vitamins and fats, and accumulation of toxic, abnormal steroid intermediates in the liver, which may interfere with bile acid transport processes and lead to cholestasis and cirrhosis (76). The primary defects in bile acid biosynthesis are the defects in modifications of the sterol nucleus, including 3ß-HSD (77, 78, 79) and ⁴-3-oxosteroid-5ß-reductase deficiencies (80, 81), and the defects in side-chain oxidation due to CYP27A1 gene mutations (82). Defects in peroxisome biogenesis and enzymes in peroxisomal ß-oxidation may manifest as the secondary defect in bile acid synthesis, including Zellweger syndrome and related infantile Refsum disease and neonatal adrenoleukodystrophy (83). A defect in de novo cholesterol synthesis also causes the secondary defect in bile acid synthesis, the Smith-Lemli-Opitz syndrome due to a defect in 7-dehydrocholesterol ⁷-reductase (84, 85). Defects in bile acid transporters or 3ß-HSD cause progressive familial intrahepatic cholestasis (40, 46, 74, 77).

2. Deficiency of 7-hydroxylases.
Mice deficient in Cyp7a1 activity have been obtained by knockout of the Cyp7a1 gene (35). These mice display a complex phenotype including oily coats, hyperkeratosis, vision defects, and behavioral irregularities, which are consistent with malabsorption of vitamins E and D₃. Most Cyp7a1-/- mice died within 18 d; 40% of them died between d 1 and 4, and 45% died between d 11 and 18. Vitamin supplement to nursing mothers prevented deaths in the early period, and bile acid supplement prevented deaths in the later period. However, several 7-hydroxylated bile acids were detected in the bile and stool of adult Cyp7a1-/- mice. This was explained by the expression of hepatic CYP7B1 after weaning and accounted for the synthesis of abnormal 7-hydroxylated bile acids in these mice. The newborn Cyp7a1-/- mice developed neonatal cholestasis, which may be due to accumulation of monohydroxylated bile acids, 3-hydroxy-5-cholenoate and 3-hydroxy-5ß-cholanoate, and 27-hydroxycholesterol (86). An inherited deficiency of CYP7A1 has not been described in the literature (see Note Added in Proof, no. 1).

The CYP7A1 is a candidate gene for familial hypertriglyceridemia (87), gallstone disease (88, 89, 90), and hypercholesterolemia (52, 91). Several single-stranded conformation polymorphisms of the CYP7A1 have been identified (92). Polymorphisms in the 5'-flanking region and coding region were reported (93, 94). Genetic linkage analysis indicates a significant linkage between CYP7A1 and high plasma LDL-cholesterol concentrations (95). Two polymorphisms in the 5'-flanking region (-278CA and -554CT) may contribute to heritable variation in plasma LDL-cholesterol concentrations. The -278C alleles are associated with increased plasma LDL cholesterol concentration.

Setchell et al. (96) reported an inborn error of bile acid metabolism due to a defect of CYP7B1 in a child with severe neonatal cholestasis and cirrhosis. The absence of primary bile acid conjugates and accumulation of 3ß-hydroxy-⁵-cholenoic acids, the products of the acidic pathway, in serum and urine indicated a defect in 7-hydroxylation. In addition, the 27-hydroxycholesterol levels were 4500-fold higher than normal; however, there were no 7-hydroxylated bile acids. Neither CYP7A1 nor CYP7B1 activities were detectable. Analysis of the CYP7B1 identified a C-to-T mutation in exon 5, which converts Arg388 to a premature termination codon. The mechanism of liver injury in this patient is likely due to the accumulation of high levels of hepatotoxic monohydroxylated bile acids. These monohydroxylated bile acids may inhibit bile acid transport across canalicular membranes and reduce bile flow. There is no mutation in the coding exons of the CYP7A1 gene in this patient. It is possible that CYP7A1 may not be expressed in the neonatal human liver, and bile acid synthesis in early human development may proceed mainly via the acidic pathway (96).

3. Deficiency of CYP27A1.
Mutations of the CYP27A1 have been found in patients with cerebrotendinous xanthomatosis (CTX), a rare autosomal recessive defect of cholesterol metabolism manifested by tendon xanthomatosis, progressive neurological dysfunction, accumulation of cholesterol in the tissues, premature atherosclerosis, osteoporosis, and cholesterol gallstones (18, 82, 97). The defect leads to excessive accumulation of 7-hydroxycholesterol, 7-hydroxy-4-cholesten 3-one, 5ß-cholestane-3, 7, 12-triol, cholesterol, and cholestanol. The synthesis of bile acids, particularly CDCA, is reduced and leads to up-regulation of CYP7A1 and the accumulation of both 7-hydroxycholesterol and 7-hydroxy-4-cholesten-3-one. The precursor 7-hydroxy-4-cholesten-3-one is converted to cholestanol. Despite the normal circulating cholesterol levels in CTX patients, they develop xanthoma and premature atherosclerosis. This may be due to the reduced elimination of cholesterol from macrophages by CYP27A1. Mutations in the CYP27A1 of CTX patients have been identified (82, 98). CDCA therapy has been used to prevent or reverse neurological symptoms associated with this disease. Despite the link of CYP27A1 mutations to CTX, the etiology of this disease is still not known. Disruption of the Cyp27a1 gene in mice markedly reduced bile acid synthesis and fecal bile acid excretion by 80% (99). However, Cyp27a1-/- mice do not accumulate cholestanol and do not exhibit the progressive neurological defects observed in human CTX patients. The Cyp27a1-/- mice have enlarged livers and kidneys and have increased triglyceride levels, fatty acid synthesis, cholesterol absorption, and cholesterol synthesis (100). SREBP expression in livers of Cyp27a1-/- mice is elevated. Feeding CA reverses hepatomegaly and hypertriglyceridemia. It is concluded that CYP27A1 plays an important role in triglyceride metabolism.

	III. Nuclear Hormone Receptor Regulation of Bile Acid Synthesis

Top Abstract I. Introduction II. Bile Acid Synthesis... III. Nuclear Hormone Receptor... IV. Molecular Mechanisms of... V. Drug Therapies Targeted... VI. Conclusion and Future... Note Added in Proof References

 
Bile acid synthesis is highly regulated by many factors, including diets, nutrients, bile acids, and hormones, mainly by regulating CYP7A1 gene transcription (1). Many liver-specific transcription factors, mostly nuclear receptors, have been found to bind to and play important roles in regulating CYP7A1 transcription (1, 101, 102, 103, 104, 105, 106, 107, 108). Analysis of nucleotide sequences of two bile acid response elements (BAREs) identified in the rat CYP7A1, BARE-I (109) and BARE-II (110), revealed many AGGTCA-like repeating sequences. Chiang and co-workers (108) first reported that these hormone response elements in the BAREs bound retinoic acid receptor (RAR), chicken ovalbumin upstream promoter-transcription factor II (111, 112), hepatocyte nuclear factor 4 (HNF4) (108, 112), and peroxisome proliferator-activated receptor (PPAR) (113). They suggested that nuclear receptors might be involved in regulation of basal transcription as well as bile acid feedback regulation of the CYP7A1 gene (1, 109, 110). Subsequently, the rat CYP7A1 was identified as the first target gene of oxysterol receptor, LXR (9), and bile acid receptor, FXR (4, 5, 6). Further studies also identified pregnane X receptor (PXR), -fetoprotein transcription factor (FTF), and small heterodimer partner (SHP) as the bile acid-regulated nuclear receptors. Nuclear receptors involved in bile acid and cholesterol metabolism are described below.

A. Structure and function of nuclear hormone receptors
Nuclear receptors have a typical modular structure (Fig. 3), which contains a highly conserved DNA-binding domain in the N-terminal region and a moderately conserved ligand-binding domain in the C-terminal region. Ligand-independent activation function-1 and ligand-dependent activation function-2 are located in the N-terminal and C-terminal regions, respectively. Two cysteine-coordinated Zn²⁺ finger motifs located in the DNA-binding domain are directly involved in DNA binding and dimerization. The E region is also involved in dimerization and coregulator interaction. Nuclear receptors bind to the consensus hormone response elements located in genes. Upon ligand binding, nuclear receptors undergo conformational changes to release corepressors and recruit coactivators to bind to the activation function-2 helix (114, 115).

View larger version (31K): [in this window] [in a new window]   Figure 3. The general structures of nuclear hormone receptors. Upper figure shows the domain structure of a nuclear receptor. It contains activation function domain 1 (AF1), DNA binding domain (DBD), hinge region (D), ligand binding domain (LBD), and activation function domain 2 (AF2). With the exception of estrogen receptors (ERs), all classical steroid hormones receptors, i.e., glucocorticoid receptor (GR), mineralocorticoid receptor (MR), progesterone receptor (PR), androgen receptor (AR), bind to the palindromic repeating sequence, AGAACAN3TGTTCT. ERs bind to a direct repeat (DR) of the AGGTCA motif. Nonsteroid receptors bind to the DR, inverted repeat (IR), or everted repeat (ER) spacing by one to five nucleotides. LXRs, FXR, PPARs, retinoic acid receptors (RARs), and PXR form heterodimers with RXRs and bind to DR, ER or IR sequences as indicated. HNF4 binds to DNA as homodimers. NR5A2 monomeric receptors, -FTF, CPF, and LRH bind to an extended half-site preceded by a A/T-rich sequence.

B. Nuclear receptors involved in regulation of genes in bile acid synthesis
Nuclear receptors that have been identified to regulate genes in bile acid synthesis pathways and cholesterol metabolism are listed in Table 1 and described in detail below. These nuclear receptors are selectively expressed in the enterohepatic and peripheral tissues involved in bile acid synthesis, absorption, and transport, as well as cholesterol and lipoprotein transport (13). The NR1 family of nuclear receptors, including PXR, PPAR, LXR, and FXR, are activated by micromolar concentrations of bile acids, lipids, or steroids, which are 1000-fold higher than that for activation of the classic steroid hormone receptors, but are within the physiological or pathological concentrations.

RAR/RXR has been shown to bind to a DR5 motif and stimulates rat Cyp7a1 transcription (108, 120). Liver-specific disruption of Rxr in mice alters the expression of Cyp7a1, ApoaI, and ApocIII genes in liver (121). Cyp7a1 mRNA levels increase more than 8-fold in Rxr-/- compared with wild-type mice. This suggests that Cyp7a1 expression in the liver is under negative control mediated predominately by RXR and its partners FXR and PPAR. These two nuclear receptors negatively regulate CYP7A1 transcription. When fed a diet high in cholesterol, Cyp7a1 mRNA expression levels increase less than 2-fold in Rxr-/- mice, much less than the 4- to 5-fold increase in wild-type mice. This implies that the inhibitory effect, presumably by FXR and PPAR, must dominate over the stimulatory effect by LXR (122). This study reveals that RXR is involved in diverse physiological pathways regulating cholesterol, bile acids, and fatty acids, as well as steroid metabolism and homeostasis.

2. LXR (NR1H3).
NR1H3 subfamily receptors are activated by oxysterols (8, 123, 124). LXR has two isoforms, LXR (or RLD-1) (123, 125) and LXRß (UR, NER, RIP15, and OR-1) (126, 127, 128, 129). LXR is expressed in liver, spleen, adipose tissue, lung, and pituitary, whereas LXRß is expressed ubiquitously. Many oxysterols have been identified as the ligands of LXR (130). Among naturally occurring oxysterols, 22 (R)-hydroxycholesterol, 24 (S)-hydroxycholesterol, and 24 (S), 25-epoxycholesterol are the most potent LXR ligands. However, the physiological relevance of these oxysterols as the LXR ligands is not certain. The most abundant oxysterol in circulation, 27-hydroxycholesterol, has been shown to be a LXR ligand and may be the more relevant natural LXR ligand (131). In addition, 6-hydroxy bile acid analogs and cholestenoic acid have been identified as the selective ligands of LXR (132, 133).

LXR can act as either a positive or a negative regulator by binding different metabolites of the mevalonate pathway (134). LXR binds to a DR4 and stimulates rat CYP7A1 transcription (9). In Lxr-/- mice, the Cyp7a1 mRNA level is expressed normally in the liver, but is not stimulated by a high-cholesterol diet as in the wild-type mice (122), which leads to massive accumulation of cholesterol in the liver. Lxrß apparently is unable to compensate for Lxr deficiency in Lxr-/- mice. It was concluded that LXR might function as a cholesterol sensor, which stimulates Cyp7a1 expression to convert excess cholesterol to bile acids in response to high cholesterol (8, 9). However, LXR has much less effect on hamster and human CYP7A1, which lacks a DR4 motif (109). Therefore, the rat and mouse are unique in that they have the ability to efficiently convert excess cholesterol to bile acids by LXR-mediated stimulation of Cyp7a1 transcription. The role of LXR in regulation of CYP7A1 in humans remains elusive. In Lxr-/- mice, Srebp-1 and stearyl-CoA desaturase mRNA are reduced, suggesting that LXR plays a role in regulating triglyceride synthesis (122). This is confirmed by the identification of an LXR response element in SREBP-1c gene (135). Endogenous oxysterols derived from mevalonate pathway, most likely 24(S), 25 epoxycholesterol, activates LXR and induces SREBP-1c, which stimulates lipogenesis and leads to hypertriglyceridemia (136). This may explain why LXR-selective ligands induce SREBP-1c and hypertriglyceridemia in mice and hamsters (137). Disruption of Lxrß does not result in accumulation of cholesterol esters on a high-cholesterol diet as was observed in Lxr-/- mice (138).

Wild-type mice treated with a rexinoid, LG268, exhibit marked changes in cholesterol homeostasis including inhibition of intestinal cholesterol absorption and repression of bile acid synthesis (139). The observation that LG268 reduces Cyp7a1 mRNA levels is consistent with the report that CYP7A1 gene transcription is repressed by LG268 in transfection assays in HepG2 cells (140), and that the negative effect of RXR/FXR must dominate over the positive effect of RXR/LXR (121). Interestingly, the levels of mRNA hybridized with Abca1 cDNA probe in the intestine are increased. These authors suggest that rexinoids prevent the accumulation of cholesterol in liver and serum by both depleting bile acids, thus reducing intestinal reabsorption of cholesterol, and by Lxr induction of Abca1 transporter that efflux cholesterol from enterocytes. ATP-binding cassette transporter type A1 (ABCA1) functions as a cholesterol and phospholipid efflux regulator involved in HDL synthesis. Mutations of the ABCA1 gene have been identified in Tangier disease patients (141). However, the identity of ABCA1 transporter as a cholesterol efflux regulator in the intestine has not been firmly established. It has been reported that knockout of Abca1 gene in mice reduces intestinal cholesterol absorption (142). In complete contradiction, another laboratory reported increase of cholesterol absorption by ablation of the Abca1 gene (143). Recently, studies of sitosterolemia, an autosomal recessive disorder characterized by an increased intestinal absorption and decreased biliary excretion of dietary sterols, hypercholesterolemia, and premature coronary atherosclerosis, have identified mutations in the genes coding for ABCG5 and ABCG8 half-transporters. These transporters function as biliary sterol efflux regulators that limit intestinal absorption and promote biliary excretion of plant sterols (sitosterols) (144, 145). It is not known whether the same ABC transporters also regulate intestinal cholesterol absorption.

Several genes involved in reverse cholesterol transport are regulated by LXRs. Both Lxr and Lxrß regulate mouse Abca1 gene involved in cholesterol efflux in peripheral tissues (146, 147), human macrophage White protein (ABCG1), and the murine homolog Abc8 (148). LXR also controls lipid-induced expression of the apolipoprotein E (ApoE) gene in macrophages and adipocytes (149), the human cholesterol ester transfer protein (CETP) that mediates the exchange of cholesterol esters for triglycerides between HDL and triglyceride-rich lipoproteins (150), and the lipoprotein lipase (LPL) gene involved in hydrolysis of triglycerides carried by VLDL and CM (151). Interestingly, LXR regulates its own synthesis in macrophages, but not in adipocytes, hepatocytes, and other cell types (152).

3. FXR (NR1H4).
FXR (also named RIP14 and HRP1) was isolated by low-stringency screening of a liver cDNA library using oligonucleotides directed to the conserved DNA-binding domain of nuclear receptors (153) and by its ability to heterodimerize with RXR using yeast two-hybrid screening (128). FXR is closely related to the Drosophila ecdysone receptor and preferentially binds to an IR1 motif (128, 153). FXR and LXR are closely related and belong to the same NR1H subfamily of nuclear receptors. FXR is highly expressed in the liver, intestine, adrenal, and kidney (128, 153). Farnesol, juvenile hormone III, all-trans-retinoic acid, and TTNPB activate FXR at high concentrations (154). Recently, bile acids have been identified as the endogenous ligands for FXR (4, 5, 6). The hydrophobic bile acid, CDCA, is the most effective activator of FXR, with an EC₅₀ of about 10–20 µM, tested in kidney CV1 cells (6). The secondary bile acids, LCA and DCA, are less effective, and hydrophilic bile acids, ursodeoxycholic acid, and muricholic acids, are inactive.

When assayed in liver-derived cell lines, bile acid/FXR repressed CYP7A1 (4, 6, 140, 155), CYP27A1 (156), CYP8B1 (157), and NTCP gene transcription (158). In human embryonic kidney 293 cells, however, bile acids and cotransfection of FXR had no effect on CYP7A1 transcription. The FXR binding sequence IR1 is not present in the CYP7A1 gene and FXR does not bind to the BARE-II of CYP7A1 gene. Chiang et al. (140) suggested that FXR suppressed CYP7A1 transcription by an indirect mechanism involving other liver-specific factors (Section IV.A). It has been reported that FXR activates target genes by binding to the IR1 motifs in genes encoding ileum bile acid binding protein (IBABP) (6, 159), canalicular bile salt export pump (BSEP) (160), phospholipid transport protein (PLTP) (161), and ApoCII (162). These findings are consistent with elevated serum bile acids, cholesterol and triglycerides, reduced bile acid pool and fecal bile acid secretion, and lack of bile acid inhibition of Cyp7a1 expression in fxr-/- mice (163). These observations suggest that FXR plays a key role in lipid metabolism.

4. PPAR (NR1C1).
Three forms of PPAR, , , and (or ß), have been identified (164). PPAR is expressed in the liver, heart, and adipose tissues (164, 165), all of which have an active fatty acid ß-oxidation pathway. Fatty acids, eicosanoids, and hypolipidemic agents are ligands of PPARs (166, 167). PPAR is highly expressed in adipose tissues. PPAR (or ß) is expressed in most tissues. Fibrates are hypolipidemic drugs that affect many genes in lipid metabolism by activation of PPAR (164, 168).

Bile acid synthesis and pool sizes are reduced in gallstone and hypercholesterolemia patients treated with certain fibrates (169, 170). The PPAR ligand, Wy14,643, suppresses CYP7A1 mRNA levels and CYP7A1 luciferase reporter activity in HepG2 cells (113). A functional PPAR-responsive element has been mapped to the DR1 in BARE-II, which is also a HNF4 binding site. However, PPAR/RXR does not bind to this DR1 motif. It appears that PPAR interferes with HNF4 activation of the CYP7A1 by reducing the amount of HNF4 expressed (113, 171). Fibrate treatment changes bile acid composition by increasing CA and decreasing CDCA synthesis. This may be because PPAR stimulates CYP8B1 activity and increases CA synthesis in the rat (172). However, PPAR binds to the rat CYP8B1 gene rather weakly. In Ppar-/- mice, inhibitory effects of fibrates on bile acid synthesis and Cyp7a1 and Cyp27a1 expression were abolished (173).

PPAR appears to mediate fatty acid stimulation of LXR expression by binding to several PPAR response elements located in the 5'-upstream sequence of the LXR gene (174). PPAR has been shown to induce LXR, which then induces ABCA1 and ABCG1 expression in macrophages (149, 175). These findings suggest that the PPAR-LXR-ABC1 cascade is involved in cholesterol efflux in macrophages. Bile acids have been shown to antagonize PPAR activity; however, the physiological role of bile acids on PPAR regulation is not clear (176).

5. HNF4 (NR2A1).
HNF4 is the most abundant orphan nuclear receptor expressed in the liver. HNF4 homodimer binds to the DR1 motif and regulates the liver-specific expression of many genes involved in lipoprotein metabolism, including ApoAI, ApoB, and ApoCIII (177, 178), and glucose metabolism (179, 180). HNF4 has constitutive activity and is able to transactivate genes without ligand binding (181). Fatty acyl-CoA thioesters have been shown to activate HNF4; however, the physiological relevance of these ligands has been questioned (182). HNF4 binds to a DR1 sequence in the BARE-II and stimulates rat CYP7A1 promoter/reporter activity (108, 112). Mutation of the HNF4 binding site markedly reduced CYP7A1 promoter activity, indicating that HNF4 is crucial for basal level transcription (108, 112, 183). HNF4 binding sites have also been identified in the CYP8B1 (157, 184) and CYP27A1 genes. HNF4 has been shown to mediate bile acid repression of CYP8B1 transcription (157, 184). HNF4 activity is also regulated by posttranscriptional mechanisms, i.e., phosphorylation of the DNA-binding domain of HNF4 by protein kinase A reduced HNF4 transactivation activity (185). Bile acids or TNF has been shown to inhibit the transactivation potential of HNF4 via MAPK cascade (186).

Mutations of the HNF4 gene have been linked to maturity onset diabetes of the young (MODY1) (187, 188). HNF4 is an upstream regulator of HNF1 gene, the mutation of which has been linked to MODY3 (189, 190). Disruption of Hnf4 in mice is embryonic lethal, because HNF4 is critical for early liver development and differentiation. Liver-specific conditional disruption of the Hnf4 gene results in marked accumulation of lipids in the liver, reduction of serum cholesterol and triglycerides, and accumulation of bile acids in serum. These phenotypes may be explained by reduction of mRNA levels for Cyp7a1, Hnf4, ApoaII, ApocIII, Apob100, Ntcp (SLC10A1), Oatp1 (SLC21A1), and microsomal triglyceride transport protein (191). This is consistent with the important role that HNF4 plays in basal transcription of CYP7A1 and underscores the importance of this nuclear receptor in regulation of lipoprotein metabolism.

6. FTF (NR5A2).
The Fushi-tarazu factor-1 (Ftz-F1) family of monomeric nuclear receptors plays important roles in steroidogenesis, liver growth, endocrine development, and differentiation (192, 193). Two Ftz-F1 genes have been identified. Ftz-F1 (NR5A1) was first identified in Drosophila as a factor that activates the homeobox gene, fushi tarazu (194). A mouse homolog, steroidogenic factor 1 (SF-1), was first cloned from an adrenal gland cDNA library (195). The Ftz-F1ß (NR5A2) gene encodes -FTF and its homologs, rat FTF (196), human CPF (197), hepatitis B virus enhancer 1 factor (hB1F) (198), human FTF (hFTF) (199), mouse LRH (200), Xenopus laevis xFF1 (200), and the zebra fish (zFF1) (201). These NR5A2 variants differ in their N-terminal amino acid sequences and C-terminal truncation due to differential promoter usage and alternative mRNA splicing (201). All hFTFs lack a sequence corresponding to exon 2 of mouse FTF. hFTF is similar to CPF variant 1 and hB1F2 (541 amino acid residues) (202). CPF (495 amino acids) is identical with hB1F (198); they lack a sequence corresponding to both exons 2 and 3 of mouse FTF. CPF variant 2 is a truncated form (323 amino acid residues) lacking C-terminal 172 amino acid residues of ligand-binding domain and AF2 domains and is similar to hFTFs (199). FTF is the name recommended by the Genome Database Nomenclature Committee, and it is used here unless specified otherwise.

FTF is expressed in liver, intestine, and pancreas, and is most related to SF-1 expressed in steroidogenic tissues (203). FTF has intrinsic transcriptional activity and its ligand has not been identified. FTF binding sites have been identified in SF-1 (204), SHP (205), HBV (206), HNF3ß, HNF4, and HNF1 genes (202). The binding site for FTF in human CYP7A1 has been mapped to –₁₃₄TCAAGGCCA-₁₂₆ (197), which overlaps with the HNF4-binding site (-₁₄₄TGGACTtAGGTCA-₁₃₂) by three nucleotides (underlined). In rat CYP8B1, two FTF binding sites are identified (207). Embedded in the FTF site is a HNF4 binding site (184). In human CYP8B1, there is an overlapping HNF4 and FTF binding site (157). FTF is a weak transcription factor that, when transfected at high concentration, stimulates CYP7A1 reporter activity by about 2-fold in nonliver cells (197). It has been suggested that FTF functions as a competence factor for sterol regulation of mouse Cyp7a1 (2) and human CETP gene (208) by LXR. It is interesting that bile acids could induce FTF mRNA expression in rat livers and HepG2 cells (155, 184) and functioned as a repressor that inhibited human CYP7A1 and rat CYP8B1 transcription when assayed in HepG2 cells (155, 157, 184). Thus, FTF may directly inhibit rat CYP7A1 and human CYP8B1 in response to bile acids. The inhibitory effect of FTF is likely due to competition for HNF4 binding to the overlapping binding sites in the BAREs. Bile acids also induce FTF mRNA expression in the intestine (209). It is interesting that FTF induces the multidrug-resistant protein-3 (MRP3) gene involved in excretion of bile acids across basal lateral membrane into portal blood, and FXR does not regulate MRP3 gene (209). Thus, FTF may play a direct role not only in feedback inhibition of bile acid synthesis but also in stimulation of bile acid transport and absorption in the intestine. FTF gene transcription is regulated by GATA and basic helix-loop-helix transcription factors (202). FTF, in turn, regulates HNF4 and HNF1 gene transcription. Hence, FTF is an upstream regulator of the genes involved in early liver development. FTF may protect liver and intestine from cytotoxicity of bile acids during the development of the gastrointestinal tract.

7. SHP (NR0B2).
Using the mouse nuclear receptor, constitutive androgen receptor (CAR), as bait, two-hybrid screening identified SHP as an interacting factor (210). SHP is a unique orphan nuclear receptor that lacks a conserved DNA-binding domain but contains a receptor-interacting domain and a repressor domain (211). SHP is known to inhibit transactivation activity of RAR, CAR, HNF4, estrogen receptor and ß, PPAR, and thyroid hormone receptor (211, 212, 213, 214, 215). Thus, SHP is a promiscuous inhibitory heterodimer partner of nuclear receptors. SHP is closely related to DAX-1, a nuclear receptor expressed in steroidogenic tissues. DAX-1 was originally identified in X-linked adrenal hypoplasia congenita patients who have deletions or mutations of DAX-1 (216, 217, 218). Two mechanisms have been suggested for repression of nuclear receptor activity by SHP. First, SHP competes with other nuclear receptors for coactivators such as steroid receptor coactivator families of steroid receptor coactivators. Second, SHP represses nuclear receptors directly by its repressor function located at the C-terminal region (215). SHP transcription is stimulated by monomeric nuclear receptors bound to DNA, i.e., SF-1 and FTF (205). Because SHP interacts with FTF, SHP inhibits its own transcription by inhibiting FTF activity (155). Hence, FTF, FXR, and SHP tightly regulate the expression of SHP in the liver, similar to the regulation of CYP7A1 gene by these receptors. It would be interesting to disrupt the Shp gene or overexpress Shp in the mouse liver to verify the genes and pathways regulated by SHP (see Note Added in Proof, no. 2). SHP mRNA levels are relatively high in mouse livers but lower in rat livers. This may explain a much higher CYP7A1-specific activity in the rat than in mouse livers.

Recently, Goodwin et al. (3) reported that the FXR agonist, GW4064, repressed CYP7A1 mRNA but stimulated SHP mRNA expression in rats. They found an inverse relationship between CYP7A1 and SHP mRNA expression levels. FXR has been shown to bind mouse and human SHP promoter and stimulates SHP reporter activity (2, 3). In Fxr-/- (163) and in Cyp7a1-/- mice (2), SHP expression is reduced. Furthermore, SHP represses CYP7A1 in a dose-dependent manner by inhibiting the transactivating activity of FTF (LRH) (2, 3, 155). This is analogous to DAX-1 inhibition of SF-1 in steroidogenic tissues. Recently, Chen et al. (155) reported that feeding CDCA to rats had no effect on SHP mRNA expression in livers. However, overexpression of FXR in the presence of CDCA stimulated SHP mRNA expression levels in HepG2 cells. Species differences in SHP expression in response to bile acids may explain these descriptions. SHP mutations have been identified in obese Japanese with early onset of diabetes (219). It was suggested that SHP was a MODY gene that might regulate HNF4 activity and energy metabolism in the pancreas. A cascade mechanism of FXR regulation of CYP7A1 involving SHP will be described in Section IV.A.

8. PXR (NR1I2).
Mouse PXR or the human ortholog, steroid xenobiotic receptor, is a promiscuous xenobiotic receptor that is activated by structurally unrelated steroids, xenobiotics, and drugs, such as phenobarbital and antibiotics, and expressed predominantly in the liver and intestine (220, 221). PXR is most closely related to CAR, and shares common ligands and function. PXR ligands also induce the CYP3A family of cytochrome P450 enzymes. CYP3A4 is the most abundant cytochrome P450 isozyme expressed in human liver and intestine and metabolizes about 60% of clinical drugs in the liver and intestine (222). Dexamethasone, pregnenolone 16-carbonitrile (PCN), rifampicin, phenobarbital, and other drugs activate PXR, which forms a heterodimer with RXR and binds to the promiscuous response elements consisting of DR3, DR4, DR5, or ER6 in the CYP3A genes. Recently, LCA has been identified as a ligand of PXR (221, 223). It was suggested that PXR might function as a bile acid sensor that induced CYP3A4 to convert LCA to a hydrophilic bile acid, hyodeoxycholic acid (HDCA). PXR also induces the OATP2 (SLC21A6) in sinusoidal membrane. This was consistent with the observation that PCN did not affect bile acid secretion in Pxr-/- mice (224) and that Pxr null mice developed inflammatory response and liver damage upon LCA treatment. However, Pxr null mice were responsive to LCA induction of CYP3A (225). These results suggest that LCA induces OATP2, which transports LCA into hepatocytes to induce PXR to inhibit CYP7A1 transcription (225). Thus, PXR may play a protective role against hepatotoxicity and cholestasis induced by LCA (225). This is consistent with a previous report by Chiang et al. (226) that PCN and dexamethasone strongly inhibit CYP7A1 activity and protein expression in rat livers. Phenobarbital is known to stimulate CYP7A1 activity (226, 227). However, the mechanism by which PXR inhibits and phenobarbital induces CYP7A1 expression is not known at present. PXR also induces multidrug-resistant protein 1 (MDR1) and MRP2, which transport sulfate-conjugated tauco-CDCA and tauro-LCA to canaliculi (228).

C. Nuclear receptor regulation of cholesterol homeostasis
Figure 4 illustrates the central roles LXR and FXR play in coordinate regulation of bile acid synthesis, transport, and absorption in the liver and intestine, and cholesterol metabolism in the liver and peripheral tissues. When cholesterol levels increase in hepatocytes, oxysterols activate LXR, which stimulates the conversion of cholesterol to bile acids by inducing CYP7A1 transcription. LXR induces SREBP-1c to stimulate triglyceride synthesis by inducing genes involved in fatty acid synthesis. LDL receptor, HMG-CoA reductase, and other genes in cholesterol synthesis pathway may be also induced. LXR also induces genes involved in lipoprotein metabolism, including LPL, CETP, and PLTP.

View larger version (80K): [in this window] [in a new window]   Figure 4. LXR and FXR regulation of bile acid synthesis, transport, and absorption, as well as cholesterol homeostasis in the liver. In the liver, cholesterol is converted to bile acids (BA) and also oxidized to oxysterols by sterol hydroxylases. Oxysterols activate LXR, which induces transcription of genes for CYP7A1 and SREBP-1c. SREBP induces genes involved in fatty acid synthesis, LDL receptor, HMG-CoA reductase, and other genes in cholesterol synthesis. LXR also induces CETP and LPL involved in lipoprotein metabolism. Bile acids activate FXR and inhibit CYP7A1 and NTCP transcription. On the other hand, FXR induces the expression of bile salt export pump (BSEP), which excretes bile acids into bile. Cholesterol is excreted into bile by an unknown mechanism. Phosphatidylcholine (PC) is excreted by MDR2 to bile canaliculi. Cholesterol, PC, and bile acids form mixed micelles and are stored in the gallbladder. Bile acids secreted from the gallbladder are reabsorbed in the intestine by ASBT located in the brush border membrane. FXR induces ileum bile acid binding protein (IBABP), which binds and facilitates the efflux of bile acids by truncated ASBT (tASBT) located in the basolateral membrane into portal circulation to hepatocytes where bile acids are taken up by NTCP. LCA, a secondary bile acid formed in the intestine, induces OATP2 and transports LCA into hepatocytes for conversion to hyodeoxycholic acid by CYP3A, which is induced by PXR (not shown). Dietary cholesterol is absorbed into the intestine by an unknown mechanism. In the intestine, LXR induces ABCG5/G8 transporters and perhaps also ABCA1, which effluxes sitosterol (plant sterol) and cholesterol, respectively, from enterocytes. In peripheral tissues, oxidized LDL is taken up by SR-A1, and LXR induces ABCA1/G1, which effluxes cholesterol and phospholipids to form pre-HDL with ApoAI/ApoE. FXR induces PLTP, which transfers phospholipids from VLDL and LDL to pre-ßHDL and HDL3 to form HDL3 and HDL2, respectively. LXR induces CETP, which exchanges triglycerides for cholesterol between HDLs and other lipoproteins.

Increases in bile acid synthesis and pool size stimulate FXR, which inhibits CYP7A1 to decrease bile acid synthesis but stimulates BSEP expression to excrete bile acids into bile. Cholesterol is excreted into bile by an unknown mechanism, possibly involving ABCG5/G8, and MDR2 excretes phosphatidylcholine into bile to form mixed micelles with bile acids. In the intestine, bile acids are reabsorbed into enterocytes by sodium-dependent bile acid transporter (ASBT, SLC10A2) located in the brush border membrane and bind to IBABP, which may facilitate bile acid efflux by the truncated ASBT (tASBT), located in the basolateral membrane, to portal circulation (233). FXR inhibits the expression of NTCP in sinusoidal membrane to reduce reabsorption of bile salts into hepatocytes. Therefore, FXR may play major roles in bile acid metabolism, reverse cholesterol transport, and protect hepatocytes against cholestasis by 1) feedback inhibition of bile acid synthesis by CYP7A1; 2) stimulation of bile acid efflux from hepatocytes by BSEP; 3) inhibition of bile acid uptake into hepatocytes by NTCP; and 4) regulation of reverse cholesterol transport by inducing ApoCII and PLTP.

It should be emphasized that cholesterol metabolism in rats and mice is very different from humans and other species. Rats and mice have very little LDL and do not express CETP. Stimulation of CYP7A1 by a high-cholesterol diet is observed only in rats and some inbred strains of mice. The bile acid pools of rats and mice are more hydrophilic, containing mostly muricholic acids, and thus less effective in activation of FXR. Therefore, the positive effect of LXR may dominate over the negative effect of FXR to explain the high efficiency in conversion of cholesterol to bile acids in the rat and mouse. In contrast, a high-cholesterol diet does not stimulate, but represses CYP7A1 in the monkey, guinea pig, rabbit, and human. In the latter species the inhibitory effect of FXR may dominate over the stimulatory effect of LXR to explain the inhibition of CYP7A1 gene transcription by cholesterol. Therefore, cholesterol may indirectly activate FXR by stimulating the synthesis of bile acids. Thus, rats are resistant to diet-induced hypercholesterolemia, whereas rabbits, hamsters, and some humans develop hypercholesterolemia on a diet high in cholesterol.

	IV. Molecular Mechanisms of Regulation of Bile Acid Metabolism

Top Abstract I. Introduction II. Bile Acid Synthesis... III. Nuclear Hormone Receptor... IV. Molecular Mechanisms of... V. Drug Therapies Targeted... VI. Conclusion and Future... Note Added in Proof References

 
Bile acid feedback regulation of bile acid synthesis has been studied for more than three decades. Despite that, the molecular mechanism of bile acid feedback is poorly understood. During the last decade, cloning of the CYP7A1 cDNAs and the genes has contributed significantly to our understanding of the molecular mechanism of bile acid synthesis and regulation (1). Several mechanisms have since been proposed to explain bile acid feedback regulation of CYP7A1 transcription. The receptor-mediated mechanism originally proposed by Chiang and Stroup (109) is based on the finding that hormone response element-like repeats are present in the BAREs identified in the CYP7A1. Bile acids have been shown to activate protein kinase C (PKC) signaling pathway (234) and inflammatory cytokines (235, 236). A receptor-mediated mechanism might regulate bile acid synthesis under physiological conditions, whereas a cell-signaling mechanism possibly provides a rapid response to stress that is induced by bile acid overload (such as in cholestasis). These two mechanisms may converge to down-regulate the genes through the same transcription factors.

A. Nuclear receptor-mediated mechanism
Chiang and associates (1, 109) proposed that bile acids might bind to and activate a nuclear bile acid receptor, which interacts with a bile acid-responsive protein that transactivates CYP7A1 gene transcription. Interaction between bile acid receptor and bile acid-responsive protein might prevent a transactivating factor from binding to the BARE, thus inhibiting CYP7A1 transcription. It was further proposed that bile acid receptors and bile acid responsive proteins might be liver-enriched transcription factors or orphan nuclear receptors (1, 110). Identification of FXR as a bile acid receptor supports this mechanism. The comparison of nucleotide sequences of the BAREs identified in rat and human CYP7A1 (110, 155), rat and human CYP8B1 (157, 184), and human CYP27A1 show similar characteristics: they all contain overlapping binding sites for HNF4 and FTF. These two nuclear receptors may differentially regulate these genes by competing for binding to the BAREs. The relative expression levels of these two nuclear receptors in liver may also regulate these genes under different physiological conditions.

Figure 5 illustrates the receptor-mediated mechanisms of bile acid regulation of gene transcription based on the original mechanism proposed by Chiang and co-workers (1, 109, 110) and modified according to the FXR/SHP cascade mechanism proposed recently (2, 3, 155, 157). CDCA-activated FXR binds to the IR1 sequences and stimulates IBABP, BSEP, PLTP, and SHP transcription in the liver. SHP then interacts with FTF to repress CYP7A1, or with HNF4 to repress CYP8B1 and CYP27A1 transcription. It should be noted that bile acids also induce FTF, which directly inhibits genes in bile acid synthesis in the liver (155, 157, 184) but stimulates MRP3 gene in the intestine (209). Figure 5 also shows an SHP-independent mechanism by which the PXR activated by LCA represses CYP7A1 transcription by an unknown mechanism. On the other hand, PXR induces OATP2 to facilitate the transport of LCA to hepatocytes to induce CYP3A family enzymes that convert LCA to hyodeoxycholic acid. Vitamin D receptor has recently been identified as a LCA-activated receptor, which may also regulate bile acid synthesis (see Note Added in Proof, no. 3). Therefore, bile acids regulate bile acid synthesis, transport, absorption, and detoxification in the liver and intestine. It is intriguing that bile acids activate a very specific receptor, FXR, which induces a nonspecific, negative receptor, SHP. SHP then interacts with other nonspecific receptors (FTF and HNF4) and specifically inhibits the genes regulated by bile acids. The unique structures of BAREs in bile acid-repressed genes must be critical to provide specificity for bile acid inhibition. Tissue-specific expression of SHP, HNF4, and FTF may also provide specificity for this cascade mechanism of gene transcription. Further study by knocking out the shp gene in mice to study bile acid feedback regulation of genes transcription would provide more convincing evidence for the FXR/SHP-dependent mechanism (see Note Added in Proof, no. 2). In addition, bile acids may down-regulate a gene by SHP-independent mechanism, i.e., reducing HNF4 expression level and stimulation of FTF (184). When bile acid pool in the liver is reduced, increasing HNF4 expression and decreasing FTF expression would allow HNF4 to bind to the BARE and stimulate gene transcription. When bile acid pool increases, HNF4 expression is reduced and FTF is increased to allow FTF to bind to the BARE and down-regulate gene transcription.

View larger version (37K): [in this window] [in a new window]   Figure 5. Nuclear receptor-mediated mechanism. FXR and PXR are bile acid receptors. SHP, HNF4, and FTF are bile acid-responsive proteins that are regulated by bile acids. FXR/RXR binds to the IR1 sequence in IBABP, BSEP, PLTP, and SHP gene and activates their gene transcription. FXR indirectly represses gene transcription by induction of a negative receptor, SHP, which interacts with HNF4 or FTF and represses CYP7A1, CYP8B1, and CYP27A1 transcription. Bile acids also induce FTF, which interact with SHP or functions as a negative regulator that inhibits human CYP7A1, rat CYP8B1, and human CYP27A1 transcription. Bile acids also inhibit HNF4 gene transcription and contribute to the inhibition of HNF4-activated rat and human CYP8B1 and human CYP27A1. In SHP-independent mechanism, LCA activates PXR, which binds to an ER6 sequence in CYP3A and OATP2 genes. PXR inhibits CYP7A1 transcription by an unknown mechanism.

View larger version (41K): [in this window] [in a new window]   Figure 6. Cell signaling mechanism. Bile acids activate PKC, which initiates a MAPK signal transduction pathway to phosphate JNK1, 2. Bile acids also induce inflammatory cytokines, TNF and IL-1, which also activate MAPK cascade involving MEEK1 and SEK/MKK4, and phosphorylate JNK1/2. JNK1/2 phosphorylates c-Jun, which may interact with FTF (or other unknown factors) and repress CYP7A1. The phosphorylated c-Jun may induce SHP, which interacts with FTF and represses CYP7A1 transcription as in the nuclear receptor-mediated mechanism. JNK1/2 may phosphorylate HNF4 and inhibit its transactivation activity, leading to repression of CYP7A1 transcription. MEKK1, MAPK kinase kinase 1; SEK, stress-activated protein kinase kinase; MKK4, MAPK kinase kinase; JNK, c-Jun N-terminal kinase.

	V. Drug Therapies Targeted to Nuclear Receptors and Genes in Bile Acid Metabolism

Top Abstract I. Introduction II. Bile Acid Synthesis... III. Nuclear Hormone Receptor... IV. Molecular Mechanisms of... V. Drug Therapies Targeted... VI. Conclusion and Future... Note Added in Proof References