The xenobiotic receptors PXR and CAR in liver physiology, an update

Abstract

Pregnane X receptor (PXR) and constitutive androstane receptor (CAR) are two nuclear receptors that are well-known for their roles in xenobiotic detoxification by regulating the expression of drug-metabolizing enzymes and transporters. In addition to metabolizing drugs and other xenobiotics, the same enzymes and transporters are also responsible for the production and elimination of numerous endogenous chemicals, or endobiotics. Moreover, both PXR and CAR are highly expressed in the liver. As such, it is conceivable that PXR and CAR have major potentials to affect the pathophysiology of the liver by regulating the homeostasis of endobiotics. In recent years, the physiological functions of PXR and CAR in the liver were extensively studied. Emerging evidence has suggested the roles of PXR and CAR in energy metabolism, bile acid homeostasis, cell proliferation, to name a few. This review summarizes the recent progress in our understanding of the roles of PXR and CAR in liver physiology.

1. Introduction

The liver is an important organ that has numerous functions, ranging from detoxification of drugs and toxic chemicals to maintaining energy metabolism and bile acid homeostasis. Pregnane X receptor (PXR) and constitutive androstane receptor (CAR) are two chemical-sensing transcription factors that belong to the nuclear receptor superfamily. PXR and CAR are best known for their function as xenobiotic receptors through their transcriptional regulation of the expression of many Phase I and II drug-metabolizing enzymes and transporters [1, 2]. Both PXR and CAR are highly expressed in the liver, and they play important roles in supporting normal liver functions. This is because, in addition to metabolizing xenobiotics, the same PXR and CAR target enzymes and transporters are also responsible for the homeostasis of numerous endobiotics. In recent years, the roles of PXR and CAR beyond xenobiotic metabolism have been extensively investigated. Many studies have described the involvement of PXR and CAR in liver physiology, ranging from energy metabolism, bile acid and bilirubin homeostasis, cell proliferation, alcoholic liver disease, and hemorrhagic shock-induced liver injury. [3–5].

Similar to other members of the nuclear receptor family, PXR consists of N-terminal domain (NTD), DNA-binding domain (DBD), hinge region (H), and ligand-binding domain (LBD). PXR can bind to a wide range of ligands, including rifampicin, dexamethasone, PCN (pregnenolone-16a-carbonitrile), natural herb St. John’s Wort, and bile acids like lithocholic acid, possibly due to its large and flexible ligand-binding pocket [6]. Moreover, the ligand-binding pocket of PXR displays species difference. This is due to the low homology between human PXR LBD and mouse PXR LBD (73% at the amino acid level) [7]. Among typical species-specific PXR agonists, rifampicin has a high affinity towards the human PXR, whereas PCN is a potent activator of mouse PXR [8]. Upon binding to a particular ligand, PXR dimerizes with the retinoid X receptor (RXR) and then translocates into the nucleus and binds to its DNA response elements. Activation of PXR induces the expression of cytochrome P450 3A (CYP3As), a group of enzymes that catalyze the hydroxylation of a wide range of xenobiotic and endobiotic substrates and facilitate their clearance [9, 10]. Other genes regulated by PXR include organic anion transporting polypeptide 2 (Oatp2) [11] and multidrug resistance-associated protein 2 (Mrp1) [12, 13]. Additionally, the transcriptional activity of PXR can be affected by nuclear receptor co-activators such as SRCs and peroxisome proliferator-activated receptor (PPAR)-binding protein (PBP), or nuclear receptor co-repressors such as NcoR and SMRT [14].

Similarly, the structure of CAR also contains NTD, LBD, H region, and DBD. The nuclear translocation of CAR can be mediated by ligand binding and post-translational modification. The canonical ligands of CAR include TCPOBOP (1, 4-bis[2-(3, 5-dichloropyridyloxy)]benzene) and CITCO (6-(4-chlorophenyl)imidazo[2,1-b][1,3]thiazole-5-carbaldehyde-O-(3,4-dichlorobenzyl)oxime). TCPOBOP has a higher affinity towards the mouse CAR, while CITCO prefers binding to the human CAR [15, 16]. Phenobarbital, on the other hand, activates CAR without a direct receptor binding. Upon activation, CAR dimerizes with RXR, translocates into the nucleus, and binds to phenobarbital response element and transactivates target gene expression. The typical CAR target genes induce the cytochrome P450 2B (CYP2B) enzyme [17, 18]. Furthermore, the transcriptional regulation of CYP2B gene by CAR has been shown to involve many coactivator proteins such as peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC1α), growth arrest and DNA damage-inducible protein beta (GADD45β) and SRCs. In recent years, several studies focused on investigating the molecular mechanism of CAR nuclear translocation. It has been reported that PPAR-binding protein is required for the nuclear translocation of CAR [19]. In addition, it has been reported that dephosphorylation of serine 202 of CAR was critical for controlling the nuclear translocation of this receptor [20]. Specifically, YFP-tagged CAR S202D mutant failed to translocate into the nucleus in the livers of Car^−/− mice upon phenobarbital treatment [20]. A subsequent report from the same group showed that CAR activator phenobarbital triggered the dephosphorylation of threonine 48 of the mouse CAR, allowing for its translocation into the nucleus, whereas phosphorylation of CAR may destabilize the α-helix of CAR and retain it in the cytoplasm [21].

PXR and CAR share several similarities. First, both receptors can be activated by xenobiotics and induce the expression of drug-metabolizing enzymes and transporters to detoxify exogeneous chemicals. Second, regarding the endobiotic functions of PXR and CAR, both receptors have been demonstrated to be involved in energy metabolism, bile acid homeostasis, cell proliferation, and other pathophysiological processes. The roles of PXR and CAR in those processes under physiological and pathological conditions have been extensively studied. This review will mainly summarize the functions of PXR and CAR in the liver and provide an update on the most recent progress.

Table 1 summarizes shared and differential effects of PXR and CAR activation on liver physiology, whereas Table 2 summarizes shared and differential effects of PXR and CAR ablation.

Table 1.

Shared and differential effects of PXR and CAR activation on liver physiology

	PXR activation				CAR activation
	Function	Factors involved	Gene/Protein regulated	Notes	Function	Factors involved	Gene/Protein regulated	Notes
Energy metabolism	↓/↑ G luconeogenesis *	FOXO1[23]; HNF4α, PGC1α[82]; CREB[22];	↓C6Pase[22]1: ↓PEPCK1 [82]2; ↑G6Pase. fPEPCKl [83]3;	1. Rifampicin treatment in Huh7 cells 2. Rifampicin treatment in HepG2 cells 3. human liver cells	↓Gluconeogenesis	FOXO1[23, 84]; HNF4α, PGC1α[48, 49]; GADD45β[57]	↓G6pase. ↓Pepck [23, 54, 55,84]’;	1. HFD and ob/ob model; single dose of TCPOBOP in chow diet mice
	↑Lipogenesis	PPARγ, FOXA2[26]; SREBP1[85]	↑Scd-1[26, 85] ↑ SLC13A5[86]2;	1. PCN treatment in fasting mice 2. HepG2 cells	↓/↑Lipogenesis*	SREBP1[87]; LXR [88]; GADD45P[57]	↓Scd-I, ↓Srebp-1c, ↓Acc [54, 55]1; ↑Fasn, ↑Elov/6, ↑Gpat, ↑Pnpla3[50]2 ↑Insig1[50]3	1. HFD and ob/ob models; 2. TCPOBOP treatment for three days.; 3. Single dose of TCPOBOP;
	↓-β-oxidation	FOXA2[26],	↓Cptlα,Hmgcs2[26]1 ↓Pparα, Thlolase, ↑Pparγ[24]2;	1. PCN treatment in fasting mice; 2. FABP-VP-hPXR transgenic mice	↓β-oxidation		↓Pparα, ↓Acadl, ↓-Acadm, ↓Hmgcs2[54]1;	1. HFD model; TCPOBOP treatment decreased incomplete β-oxidation
	↑Fatty acid uptake		↑Cd36[24, 90]1;	1. hPXR mice + rifampicin; PCN; Alb-VP-hPXR; FABP-VP- hPXR	↑Glycolysis		↑Pkm2[91]1;	1. TCPOBOP treatment once a week for 8 weeks in chow diet mice
Bile acid homeostasis	↑Bile acid synthesis		↓Cyp7a1[37, 92]1; ↑Cyp7a1, ↓Shp ↓Ffg15[93]2;	1. PCN treatment for 4 days 2. Wild type mice were fed with lithogenic diet and pre-treated with PCN for 4 days;	↑Bile acid synthesis		↑Cyp71[52, 94]1; ↓Cyp8b1[94]2;	1. chow diet and lithogenic diet 2. TCPOBOP treatment for 4 days
	↑Bile acid detoxification		↑Cyp2b, ↑Cyp3a[32, 33, 37, 92]1; ↑Ugtlal[35]2;	1. PCN treatment; 2. VP-hPXR transgenic mice	↑Bile acid detoxification		↑Cyp7a1[52, 94]1; ↑Ugt1a1[35, 58, 95]1; ↑Sult2a, Papss2[61]2;	1. TCPOBOP treatment 2. VP-CAR mice;
	↑Bile acid export		↑Mrp2[96]1;↑Mrp3[96]1[33,96,97]; ↑Oatp2[37, 92, 93];		↑Bile acid export		↑Mrp2[58, 96]; ↑Mrp2[58, 96]; ↑Mrp2[58, 96]; ↓Mrp2[58, 96];
Cell proliferation	↑ Hepatomegaly/Liver regeneration	YAP[40]	↓CCND1, ↓CCNAl[40];		↑Hepatomegaly/Liver regeneration	YAP[69, 70, 98]; GADD45B[56, 64, 99];	↑Afp, ↑Ctgf[98]; ↓Birc5[70]; ↑Mdm2[102]; SRC-3[65]; c-Myc, FoxM1[67, 100]; HNF4α[101];
Other	↑Hemorrhagic shock-induced liver injury		↑Cyp3a4, ↑4-HNE [46]1;	1. VP-PXR mice; PCN; Rifampicin + hPXR;	↑Alcoholic liver disease		↓Adh1, ↓Aldh1a1, ↓Cat, ↓Aldh3a2[72];

↑: increase, up-regulation, sensitization;

↓: decrease, down-regulation, attenuation;

^*denote controversial results

Table 2.

Shared and differential effects of PXR and CAR ablation on liver physiology

	PXR ablation			CAR ablation
	Function	Gene/Protein regulated	Notes	Function	Gene/Protein regulated	Notes
Energy metabolism	↓Gluconeogenesis	↓G6pase, Pepck[29]1;	1. HFD and ob/ob model			Ablation of CAR in ob/ob background has no effects on body weight, fat-body weight ratio, serum triglyceride levels
	↓Lipogenesis	↓Srebp-1c, Scd-1[29]1;	l.HFD and ob/ob model	↑Lipogenesis	↑Srebp-1c, ↑Acc-1, ↑Fas, ↑Scd-1[88];
	↑/↓β-oxidation*	↑CPT1α[29]1; ↓Cpt1α, Hmgcs2[26]2;	l.HFD model; 2. Fasting chow diet mice
Bile acid homeostasis	↓Bile acid synthesis	↓Cyp7a1[93], ↑Shp, ↑Fgf151;	1. Lithogenic diet;	↓ Bilirubin clearance	↓UGT1A11[58]	1. phenylhydrazine-induced hyperbilirubinemia
	↑Bile acid export	↓Oatp2, ↑Mrp3[93]; ↓Oatp2, ↑Mrp4[93]1;	1.Lithogenic diet;
Cell proliferation	↓Hepatomegaly/Liver regeneration	↓Pparγ, ↓Cd36, ↓Acc-1[41];		↓Hepatomegaly/Liver regeneration	↓Foxm1b, ↑Cdkn1a, ↓Ccna2, ↓Ccnb2[67];
Other	↓ Alcoholic liver disease	↑Akr167, ↑Adh1[44]1; ↑ADH1, ↑ALDH1A1, ↑CATALASE [45]2;	1. Binge ethanol model; 2. Chronic ethanol model	↑Alcoholic liver disease [72]
	↓Hemorrhagic shockinduced liver injury [46]			↓NASH	↓iNos, ↓Tnfα, ↓Col1a, ↓Timp-1 [74]1;	1. MCD diet

↑: increase, up-regulation, sensitization;

↓: decrease, down-regulation, attenuation;

^*denote controversial results

2. Role of PXR in liver physiology

2.1. Role of PXR in energy metabolism

Growing evidence suggests that PXR plays essential roles in hepatic glucose and lipid metabolism. PXR is involved in the regulation of gluconeogenesis in the liver. Specifically, activation of PXR inhibits the expression of glucose-6-phosphatase (G6Pase) and phosphoenolpyruvate carboxykinase (PEPCK), two key enzymes involved in gluconeogenesis, in mouse primary hepatocytes and in mouse livers [22]. PXR was shown to interact with forkhead box protein O1 (FOXO1) and suppress gluconeogenic gene expression [23]. Furthermore, PXR was reported to interact with cAMP-response element binding protein (CREB) and hepatocyte nuclear factor 4 alpha (HNF4α) and interfere with their binding to the promoters of G6pase and Pepck, respectively [22]. However, overexpression of constitutively active human PXR in the liver was reported to repress G6pase and Pepck mRNA expression [24]. It was noticed that the conclusion that PXR suppresses the expression of G6pase and Pepck was made based on cell culture results. It remains to be determined whether the discrepancy was due to the species specificity of the PXR receptor, the constitutive/chronic activation nature of the system, or the differences of the in vitro and in vivo systems. A more recent study reported that besides the effects of PXR on gluconeogenic gene expression, PCN treatment decreases the expression of glucose transporter 2 (Glut2) and glucokinase (Gck), two genes involved in glucose uptake [25].

PXR also plays a role in regulating lipogenesis. Activation of PXR by rifampicin in humanized PXR transgenic mice induced lipid accumulation in the liver [24]. PCN treatment in fasting wild type mice downregulates the expression of two genes important for β-oxidation and ketogenesis, carnitine palmitoyltransferase 1A (Cpt1α) and 3-hydroxy-3-methylglutaryl-CoA synthase 2 (Hmgcs2), respectively [26]. At the same time, PCN treatment induces the expression of the lipogenic enzyme gene stearoyl-CoA desaturase-1(Scd1) [26]. Regulation of Cpt1a, Hmgcs2, and Scd1 by PCN treatment is PXR dependent. Interestingly, PXR knockout mice displayed decreased basal expression of Cpt1α and Hmgcs2 but increased expression of Scd1. It was further revealed that PXR could interact with insulin response forkhead box protein A2 (FOXA2) and then repress the FoxA2-mediated transcription of Cpt1α and Hmgcs2 [26]. Moreover, hepatic lipid accumulation was also observed in the transgenic mice expressing activated human PXR in the hepatocytes [24]. Interestingly, the expression of Srebp-1c and its downstream lipogenic enzymes Fas and Scd1 were unchanged, suggesting that the steatotic phenotype of PXR activation in vivo is Srebp-1c independent [24]. In the same study, it was revealed that activated PXR induces the expression of CD36, a fatty acid translocase responsible for fatty acids uptake and CD36 is a direct transcriptional target of PXR [24]. PXR deficiency improves diet-induced obesity in wild type mice and in the leptin-deficient ob/ob mice [27–29]. The alleviation of obesity by PXR ablation is accounted for by increased oxygen consumption, increased fatty acid oxidation, decreased hepatic lipogenesis, and improved insulin sensitivity [29].

2.2. Role of PXR in bile acid and bilirubin metabolism

Bile acids are hepatic catabolic byproducts of cholesterol. In addition to their roles in facilitating the absorption of lipophilic nutrients, bile acids are also important signaling molecular through their activation of bile acid receptors, such as farnesoid X receptor (FXR) and the plasma membrane-bound G protein-coupled bile acid receptor, Takeda G protein receptor 5 (TGR5), and sphingosine-1-phosphate receptor 2 (S1PR2) [30].

PXR plays a crucial role in bile acid metabolism. CYP3A, a direct transcriptional target of PXR, is involved in the detoxification of bile acid by facilitating bile acid hydroxylation [31]. Also, bile acid LCA has been shown to be an agonist of PXR, which may represent a self-detoxification mechanism to maintain homeostasis [32]. Moreover, PCN treatment upregulates the expression of basolateral bile acid efflux transporter MRP3, which may have protected mice from LCA-induced liver injury [33].

Bilirubin is a breakdown product of heme-containing proteins, such as hemoglobin. Normally, hydrophobic bilirubin is glucuronidated in the liver by UDP glucuronosyltransferase family 1 member A1 (UGT1A1) and then cleared from the body through bile ducts [34]. Accumulation of bilirubin in the liver will lead to jaundice. UGT1A1 is a direct transcriptional target of PXR [35]. Activation of PXR by rifampicin in humanized mice or liver-specific overexpression of constitutively activated hPXR induced the expression of UGT1A1, and protected mice from experimental hyperbilirubinemia [35].

2.3. Role of PXR in hepatocyte proliferation

As early as the 1970s, it was found that PCN increased liver weight and stimulated mitotic activity in rat livers. This was supported by a later study showing that dexamethasone induced mitogenesis in the liver [36]. Later on, it was shown that PCN treatment increased liver weight in wild type mice, but not in PXR knockout mice [37]. Furthermore, cell division and cell size were also induced by PCN treatment in WT mice, which was not seen in PXR knockout mice. In addition, rifampicin treatment in hepatocellular carcinoma cells induced mitogenesis. Shizu and colleagues reported that PCN treatment reduced mRNA expression of p27 and p130, driving the entry into the cell cycle [38]. Kodama and colleagues subsequently showed that two signaling cascades GADD45β-p38-MAPK and HNF4α-IGFBP1 were involved in the cell proliferation promoting effect of PXR activation [39]. In a more recent report, Jiang and colleagues demonstrated that both mouse and human PXR specific agonists increased hepatocyte size in mice in a PXR-dependent manner [40]. They further revealed that PXR activation induced yes-associated protein (YAP) mRNA expression and YAP nuclear translocation and PXR agonist induced liver enlargement in a YAP-dependent manner [40]. On the other hand, PXR ablation in mice after partial hepatectomy (PH) resulted in delayed liver regeneration, which was in part due to the decreased hepatocyte proliferation as indicated by reduced Ki67 expression, indicating the importance of PXR during liver regeneration [41]. Although the role of PXR activation in hepatocyte proliferation in rodents has been extensively studied, its role in human hepatocytes and human liver remains unclear. A recent study reported that treatment of cultured human hepatocytes with phenobarbital induced CYP enzyme activities, but did not induce replicative DNA synthesis [42].

2.4. Role of PXR in alcoholic liver disease (ALD)

It has been known that acute and chronic consumption of alcohol will lead to steatosis (fatty liver) which may progress into liver fibrosis, cirrhosis, or liver cancer [43]. Given the role of PXR in lipogenesis, recently a research group investigated the role of PXR in ethanol-induced hepatic steatosis and hepatotoxicity using wild type and Pxr-null mice. The authors found that in a binge ethanol-induced alcoholic liver disease (ALD) model, in which mice were fed with binge doses of ethanol (4.5g/kg) every 12 h and euthanized 4 h after the final dose, Pxr-null mice showed higher protein expression of alcohol dehydrogenase 1 (ADH1) and aldehyde dehydrogenase 1A1 (ALDH1A1) and higher serum triglyceride levels, indicating that PXR inhibits ethanol catabolism and protects against ethanol-induced hyperlipidemia [44]. Moreover, in a chronic ethanol-induced hepatosteatosis model, Pxr-null mice also showed higher protein levels of ADH1, ALDH1A1, and catalase [45]. However, in this chronic ALD model, Pxr-null mice showed lower hepatic triglyceride levels compared with wild type counterparts and were protected against ethanol-induced hepatosteatosis, which is opposite to what was seen in the binge ethanol model. The discrepancy of the PXR effect between binge drinking and chronic ethanol consumption remains to be understood. Moreover, the role of PXR in alcoholic liver disease needs to be further evaluated by pharmacological or genetic PXR activation models.

2.5. Role of PXR in hemorrhagic shock (HS) induced liver injury

Hemorrhagic shock (HS) is a clinical condition characterized by tissue hypoperfusion and often results in multiple organ injury, including injury to the liver. Many clinical drugs prescribed to trauma patients can activate PXR and induce CYP3A. Recently, we have revealed a novel function of PXR in HS-induced liver injury [46]. We showed that genetic or pharmacological activation of PXR exacerbated HS-induced liver injury in wild type or hPXR/CYP3A4 humanized mice, while the sensitization was attenuated by PXR ablation or pharmacological inhibition of PXR [46]. In understanding the mechanism by which PXR sensitizes mice to HS-induced liver injury, we found that the sensitizing effect of PXR activation was accounted for by PXR-responsive induction of CYP3A and increased oxidative stress in the liver. The sensitizing effect of PXR was attenuated by ablation or pharmacological inhibition of CYP3A, or treatment with the antioxidant N-acetylcysteine amide. These results suggest that cautions need to be applied for the unavoidable use of PXR-activating drugs in trauma patients because these drugs have the potential to exacerbate HS-induced hepatic injury, which can be mitigated by the coadministration of anti-oxidative agents, CYP3A inhibitors, or PXR antagonists.

Figure 1 summarizes the functions of PXR in liver physiology.

Figure 1.

Schematic summary of the functions of PXR in liver physiology

3. Role of CAR in liver physiology

3.1. Role of CAR in energy metabolism

Phenobarbital (PB), an activator of CAR, has been known to correlate with metabolic benefit for a long time. Activation of CAR by PB or TCPOBOP decreased gluconeogenic genes Pepck and G6pase expression. Regulation of gluconeogenic gene expression by CAR involved several coactivators. It was reported that CAR competed with FOXO1 for binding to insulin response elements in the promoter regions of G6pase or Pepck [47]. In addition, it has been proposed that CAR could compete with HNF4α for binding to the promoter region of G6pase and Pepck [48].

Our group recently reported that activation of CAR inhibited gluconeogenesis by facilitating the ubiquitination and degradation of pro-gluconeogenic nuclear receptor co-activator PGC1α [49]. Mechanistically, we showed that activated CAR translocated into the nucleus and served as an adaptor protein to recruit PGC1α to the Cullin1 E3 ligase complex for ubiquitination. The interaction between CAR and PGC1α also led to their sequestration within the promyelocytic leukemia protein-nuclear bodies, where PGC1α and CAR subsequently underwent proteasomal degradation. At the conceptual level, we reason both drug metabolism and gluconeogenesis are highly energy-demanding processes. The negative regulation of PGC1α by CAR may represent a cellular adaptive mechanism to accommodate energy-restricted conditions.

The role of CAR in lipogenesis may be context dependent. Under physiological conditions, the effects of CAR activation are controversial, which may be due to the difference of TCPOBOP dose regimens. In one study, wild type or Car-null mice were treated with TCPOBOP (3 mg/kg/day) for three days and it was found that activation of CAR by TCPOBOP caused lipid accumulation in the liver [50]. In the same study, it was further revealed that TCPOBOP treatment induced lipogenesis genes fatty acid synthase (Fasn) and ELOVL fatty acid elongase 6 (Elovl6) mRNA expression, which was abolished in CAR-null mice but not in LXR-null mice [50]. Our group also observed that in an acute kidney ischemia disease model, TCPOBOP pre-treatment (1 mg/kg/day) for two days in sham mice showed decreased β-oxidation genes peroxisome proliferator activated receptor alpha (Pparα) and Cpt1α expression in the liver compared to the vehicle group [51]. However, TCPOBOP treatment (1 mg/kg/week) in wild type mice fed on chow diet decreased hepatic triglyceride levels and suppressed the mRNA expression of lipogenesis gene Srebp-1c [52]. Moreover, Lukowicz and colleagues reported that after monitoring WT and CAR knockout male and female mice for 68 weeks, they found that deletion of CAR in male mice increased body weight from week 6 and worsened glucose tolerance from 12–13 weeks, suggesting the crucial role of CAR in maintaining energy homeostasis [53]. In a high fat diet-induced obesity model and ob/ob genetic obesity model, activation of CAR by TCPOBOP alleviated obesity and improved insulin sensitivity [54, 55].

In a very recent study, our group reported that the anti-obesity effects of CAR activation were partially dependent on its coactivator growth arrest and DNA damage-inducible gene 45b (Gadd45b). Gadd45b, a well-known anti-apoptotic factor, has been shown to be an inducible coactivator of CAR in promoting rapid liver growth and liver regeneration [56]. We showed that reduced body weight gain and improved insulin sensitivity by the CAR agonist 1,4-bis[2-(3,5 dichloropyridyloxy)] benzene (TCPOBOP) in the (HFD)-induced obesity model were markedly blunted in Gadd45b knockout mice. Mechanistically, the TCPOBOP-responsive inhibitions of hepatic lipogenesis, gluconeogenesis, and adipose inflammation observed in WT mice were largely abolished in Gadd45b knockout mice [57]. These results suggested that Gadd45b is required for the metabolic benefits of CAR activation.

3.2. Role of CAR in bilirubin and bile acid metabolism

Similar to PXR, activation of CAR also induces UGT1A1 expression in the liver, which contributes to accelerated bilirubin clearance, and this CAR agonist effect is abolished in CAR^−/− mice [58]. In addition, phenobarbital, a clinical drug and a ligand of CAR, has been shown to enhance bilirubin-conjugation and decreased bilirubin levels in humans [59, 60]. Besides induction of UGT1A1, CAR activation by TCPOBOP also induced the expression of other important components in the bilirubin clearance pathway, including glutathione S-transferase A1 (GSTA1), MRP2, and OATP2 [58].

The anti-cholestatic effect of CAR agonists such as phenobarbital and TCPOBOP was explained by their induction of sulfotransferases (SULTs) and PAPS synthetase 2 (PAPSS2) in the liver, which is considered to provide protective effects against lithocholic acid-induced toxicity [61].

Recently, Kim and colleagues reported that CAR is endogenously activated in Fxr and Shp double knockout (DKO) mice, a model of intrahepatic cholestasis that resembles human progressive familial intrahepatic cholestasis (PFIC5) [62]. The authors found that CAR ablation in Fxr/Shp DKO mice led to worsened hepatic and biliary damage accompanied by impaired bile salt export pump (BSEP) function and fecal bile acid secretion [62]. In another study, it was reported that both chemical activation of CAR and ablation of CAR decreased total bile acid concentrations in the livers of mice [63]. More specifically, CAR activation reduced 12-OH bile acid concentrations, while CAR ablation reduced 6-OH bile acid concentration in the liver [53].

3.3. Role of CAR in hepatocyte proliferation

Partial hepatectomy will trigger hepatocyte proliferation in order to restore the original liver mass. TCPOBOP can promote cell proliferation by increasing the expression of cyclins A and D1 [64]. It has been reported that SRCs, including SRC1, SRC2, and SRC3, are coactivators of CAR, and it was found that hepatocyte hyperplasia induced by TCPOBOP was attenuated in SRC3^−/− mice but not in SRC1^−/− and SRC2^−/− mice [65]. In addition, Tian and colleagues showed that GADD45β coactivated CAR’s transcriptional activity and the liver proliferative response by TCPOBOP was delayed in Gadd45β^−/− mice [56]. In a recent study, it was shown that chronic jet lag led to NAFLD and caused spontaneous HCC by activating oncogenic, steatotic, and inflammatory pathways in WT mice and this effect was CAR-dependent [66]. Interestingly, in the same study, CAR mRNA and protein levels showed circadian rhythm but were irregular in jet-lagged WT mice, indicating CAR is a clock-controlled gene. These results suggest an important role of CAR in circadian homeostasis [66].

When excessive liver is removed, liver failure will develop which is known as small-for-size-syndrome (SFSS). It was demonstrated that standard hepatectomy in Car^−/− mice resulted in lower survival compared with that in WT mice with excessive hepatectomy [67]. Moreover, pre-treatment of mice with CAR ligand TCPOBOP was sufficient to improve the survival rate of mice after excessive hepatectomy [67]. Mechanistically, it was shown that the CAR activation effect on experimental SFSS was mediated by the forkhead box protein M1 (FOXM1), a cell cycle promoter [67]. In an independent study, Bhushan and colleagues demonstrated that TCPOBOP-induced hepatomegaly, and hepatocyte proliferation required the receptor tyrosine kinases (RTKs) MET and epidermal growth factor receptor (EGFR) by using MET knockout mice in combination with EGFR inhibitor Canertinib [68]. In addition, studies from the same group showed that hepatocyte-specific deletion of YAP, a key transcription regulator controlling liver size, decreased TCPOBOP-induced hepatocyte proliferation and disrupted cell cycle activation driven by TCPOBOP [69]. This finding is consistent with the result of an earlier in vitro study in which the authors showed that YAP knockdown in mouse primary hepatocytes attenuated the pro-proliferative effect of CAR [70]. The effect of CAR activation on the proliferation of human hepatocytes is less clear. Soldatow and colleagues reported that in human primary hepatocytes, treatment with the CAR agonist CITCO induced the expression of CYP2B isoforms, but had little effect on the proliferation phenotype as assessed by direct DNA labeling with the nucleoside analog 5-ethynyl-2′-deoxyuridine (EdU) [71].

3.4. Role of CAR in alcoholic liver disease (ALD)

It has been reported that pre-activation of CAR by TCPOBOP increased hepatic toxicity induced by acute and chronic alcohol infusion in wild type mice, and this effect was abolished in CAR-null mice [72]. The sensitizing effect of TCPOBOP was associated with a decreased mRNA expression of enzymes that metabolize the alcohol in the liver, such as Adh1, Aldh1a1, and Cat. Surprisingly, Car ablation also sensitized mice to chronic alcohol-induced liver injury, with increased ethanol-induced steatosis and hepatocyte apoptosis [72]. The molecular mechanism behind the sensitizing effect of Car ablation is still unknown. Nevertheless, these results support the role of CAR in alcoholic liver injury and imply a risk of synergistic liver toxicity induced by alcohol and CAR activation.

3.5. Role of CAR in non-alcoholic steatohepatitis (NASH)

Non-alcoholic steatohepatitis (NASH) is also a common liver injury, which is characterized by steatosis, liver cell injury, inflammation and fibrosis [73]. Yamazaki and colleagues reported that compared to the CAR^−/− mice, wild type mice showed elevated ALT levels and infiltration of inflammatory cells when fed with methionine and choline-deficient (MCD) diet for 8 weeks [74]. Moreover, the MCD-induced liver fibrosis was also attenuated in CAR^−/− mice indicating CAR may play a critical role in the pathogenesis of NASH [74]. In understanding the sensitizing effect of CAR, the authors found that there was no significant difference in the lipid concentration of the liver - namely, “the first hit” between WT and CAR^−/− mice. In contrast, there was evidence suggesting that CAR activation caused the lipid peroxidation - namely, “the second hit” in the pathogenesis of NASH. The increased lipid peroxidation may have been explained by the induction of CYPs 2B10, 2C29, 3A11 in the WT mice.

3.6. Role of CAR in ischemic acute kidney injury (AKI) through the liver-kidney organ crosstalk

Acute kidney injury (AKI) is a disease condition with a high mortality rate which may cause injury in distant organs including the liver [75]. However, how AKI is associated with liver injury is not clear yet. We recently reported that AKI induced by renal ischemia-reperfusion (IR) down-regulated liver CAR expression, led to fatty liver and decreased hepatic very-low-density lipoprotein triglyceride (VLDL-TG) secretion [51]. Pre-treatment of mice with TCPOBOP protected mice from developing AKI-induced fatty liver and liver injury accompanied by lower VLDL-TG secretion, while this preventive effect was abolished in CAR^−/− mice [51]. Attenuation of fatty liver in turn improved kidney function. In addition, an increased serum interleukin-6 (IL-6) level after renal IR was reduced by TCPOBOP treatment in AKI mice, suggesting IL-6 may be a mediator of AKI-induced liver injury [51]. Moreover, CAR^−/− mice were more susceptible to AKI-induced kidney injury indicated by a lower survival rate [51]. Given that fatty liver increases the risks for developing chronic kidney disease [76], management of hepatosteatosis by CAR activation may be a new strategy to improve kidney function in AKI patients or kidney transplant patients.

Figure 2 summarizes the functions of CAR in liver physiology.

Figure 2.

Schematic summary of the functions of CAR in liver physiology

4. Crosstalk between PXR and CAR

As discussed above, PXR and CAR have overlapping yet distinct roles in liver physiology and pathophysiology. When considering PXR and CAR as xenobiotic and endobiotic receptors as well as therapeutic targets, the crosstalk between PXR and CAR should be paid attention to. Many xenobiotics, drugs and steroid-like ligands are dual activators of PXR and CAR. For example, phenobarbital, phenytoin and TCPOBOP can activate both PXR and CAR [77]. The overlapping functions of PXR and CAR are also in part accounted for by their share of DNA binding sites and binding protein partners. The share of DNA binding sites in the promoter regions of xenobiotic enzyme and transporter genes has been proposed to be an important mechanism to ensure the fail-safe of xenobiotic detoxification [78]. Among examples of PXR and CAR sharing binding protein partners, it has been reported that both PXR and CAR can inhibit gluconeogenesis by binding to the transcriptional factor FoxO1 and preventing its binding to the insulin-response element [23]. In addition, activation of both PXR and CAR can enhance hepatocyte proliferation in mice in a YAP-dependent manner. PXR has been shown to interact with YAP directly in primary mouse hepatocytes [40], but it remains unknown whether CAR also directly binds to YAP and whether this interaction will affect YAP activity. Furthermore, activation of PXR and CAR regulates the expression of similar sets of genes involved in energy metabolism and bile acid homeostasis (Table 1), but not always in the same directions, suggesting that there might be different co-activators or co-repressors involved in the regulation, which needs to be further investigated.

5. Conclusive remarks

PXR and CAR are two sister nuclear receptors that function as master regulators of xenobiotic detoxification by governing the expression of drug metabolizing enzymes and transporters in the liver and some other tissues. However, in recent years, more and more studies indicate that PXR and CAR also play significant roles in liver physiology and liver diseases. Specifically, it has been reported that activation of PXR and CAR by pharmacological or genetic means, or ablation of PXR and CAR in mice resulted in changes in the expression genes involved in energy metabolism, bile acid metabolism, cell proliferation, and alcohol metabolism. Moreover, these changes have led to numerous studies to investigate the roles of PXR and CAR in obesity, type 2 diabetes, cholestatic liver diseases, liver regeneration, hepatocarcinoma, and ALD [3–5]. PXR ablation is protective against obesity [27–29] and alcoholic liver disease in a chronic ethanol feed model [49]. Furthermore, activation of PXR has been shown to sensitize mice to hemorrhagic shock induced liver injury [46]. The hepatoxicity associated with rifampicin and isoniazid co-therapy was reported mediated through PXR activation [79], which further implicates PXR antagonists as potential therapeutics in several diseases. The tissue specific effect PXR and CAR is also intriguing. Although there are several cases in which activation of PXR in the liver enhances drug-induced hepatotoxicity, a recent report showed that PXR activation in the kidney was protective against cisplatin-induced acute kidney injury [80]. PXR activation by rifampicin has also been shown to enhance the glucose-lowering effect of metformin through the regulation of a metformin transporter [81]. These results suggest that the biological and/or therapeutic effects of PXR agonists or antagonists can be disease-, organ-, and drug-specific.

Among the unanswered questions, many of the effects of PXR and CAR on liver pathophysiology remain to be validated in humans or human patients. Another outstanding challenge is that although the physiological functions of PXR and CAR have been appreciated, the endogenous ligands, agonists or antagonists, remain largely elusive. This is particularly true for CAR whose reported ligands are mostly xenobiotics. Identification and characterization of endogenous PXR and CAR ligands will facilitate a better understanding of the mechanisms by which these two receptors affect liver physiology, as well as harness the therapeutic potentials of PXR and CAR in liver diseases.

Highlights.

• PXR and CAR are highly expressed in the liver.

• PXR and CAR are xenobiotic receptors that function as master transcriptional regulators of drug-metabolizing enzymes and transporters.

• PXR and CAR have equally important roles in governing the homeostasis of numerous endobiotics.

• PXR and CAR have overlapping yet distinct functions in liver physiology and pathophysiology.