Hepatic Nuclear Receptors in Cholestasis-to-Cholangiocarcinoma Pathology

Abstract

Cholestasis, characterized by impaired bile flow, is associated with an increased risk of cholangiocarcinoma (CCA), a malignancy originating from the biliary epithelium and hepatocytes. Hepatic nuclear receptors (NRs) are pivotal in regulating bile acid and metabolic homeostasis, and their dysregulation is implicated in cholestatic liver diseases and the progression of liver cancer. This review elucidates the role of various hepatic NRs in the pathogenesis of cholestasis-to-CCA progression. It explores their impact on bile acid metabolism as well as their interactions with other signaling pathways implicated in CCA development. Additionally, it introduces available murine models of cholestasis/primary sclerosing cholangitis leading to CCA and discusses the clinical potential of targeting hepatic NRs as a promising approach for the prevention and treatment of cholestatic liver diseases and CCA. Understanding the complex interplay between hepatic NRs and cholestasis-to-CCA pathology holds promise for the development of novel preventive and therapeutic strategies for this devastating disease.

Cholangiocarcinoma (CCA) is a devastating primary liver cancer that accounts for approximately 15% of primary liver cancers. Unlike hepatocellular carcinoma (HCC), which has plateaued since 2015 because of a significant decrease in hepatitis B virus infections, the incidence and mortality rates of CAA are increasing in the United States (National Cancer Institute, https://seer.cancer.gov/statfacts/html/livibd.html, last accessed June 1, 2024).1, 2, 3 In 2023, approximately 8000 patients were newly diagnosed with CCA, representing approximately 0.5% of all new cancer cases in the United States (National Cancer Institute, https://seer.cancer.gov/statfacts/html/livibd.html, last accessed June 1, 2024). The 5-year survival rate of CCA is <15%, mainly attributable to the lack of effective therapies and diagnosis of patients with advanced CCA (National Cancer Institute, https://seer.cancer.gov/statfacts/html/livibd.html, last accessed June 1, 2024).1, 2, 3, 4, 5 Surgical interventions, such as partial hepatectomy targeting a localized tumor or liver transplantation, are most beneficial.^6,7 However, only limited patients are eligible for these options, because most patients present at an advanced stage with metastases at the time of diagnosis.^1,6, 7, 8 Until recently, a combination of gemcitabine and cisplatin has been recommended as the only first therapy for unresectable intrahepatic CCA (iCCA), even though very few patients respond to this treatment, and it extends survival by less than a year.^6,7 Although a fibroblast growth factor receptor (FGFR) inhibitor^5,9,10 and an isocitrate dehydrogenase 1/2 (IDH1/2) inhibitor^9,11, 12, 13 were approved as first-line targeted therapy, the impact of these regimens is limited because only approximately 20% of patients harbor these mutations, and long-term survival benefits are modest because of the adaptive resistance.^1,6,7 Extrahigh intertumoral and intratumoral heterogeneity of molecular signatures and resistance to targeted therapies and immune checkpoint inhibitors are considered major obstacles in identifying effective therapies for iCCA.^10,11,14, 15, 16, 17, 18, 19, 20, 21

Although the exact risk factors for CCA development remain unclear, diverse cholestatic diseases, including primary sclerosing cholangitis (PSC),^22,23 liver fluke infections, biliary inflammation, choledochal cysts, and cirrhosis, are strongly connected with CCA,^24,25 except the non-specific environmental factors, such as genetic factors, aging, or toxin exposures. Particularly, PSC is the most critical risk factor for CCA development, and strong hepatic inflammation is known to contribute to the malignant transformation of hepatic cells into CCA.26, 27, 28, 29 Despite this valid contribution of cholestasis/PSC to CCA formation, underlying molecular mechanisms, effective biomarkers, and substantial crosstalk with currently using therapeutic regimens, particularly in tumor development, remain largely elusive.

Several nuclear receptors (NRs), a family of ligand-activated transcription factors, are predominantly expressed in the liver, playing crucial roles in regulating various pathophysiological processes, including bile acid (BA) metabolism, lipid homeostasis, and inflammation.30, 31, 32 Importantly, diverse NR regulators are currently approved for clinical trials in several hepatic disorders related to CCA, such as liver cirrhosis33, 34, 35 and cholestasis.36, 37, 38 Emerging evidence suggests that dysregulations of liver-enriched NRs, such as farnesoid X receptor (FXR),³⁹ pregnane X receptor,40, 41, 42 hepatocyte nuclear factor 4 α (HNF4α),^33,43 and constitutive androstane receptor,^40,44, 45, 46, 47 play crucial roles in liver cancer development and maintenance by influencing tumor cell transformation, proliferation, apoptosis, and metastasis, as well as modulating the tumor microenvironment. Therefore, understanding the intricate interplay between hepatic NRs and cholangiocarcinoma holds promise for identifying novel biomarkers and therapeutic targets, which will ultimately advance personalized prevention/treatment strategies for this devastating disease. This review aims to provide a comprehensive overview of the current knowledge regarding the involvement of hepatic NRs in cholestasis/PSC-to-CCA pathobiology in preclinical murine models and their potential implications for clinical management (Figure 1).

Figure 1.

Pathobiology of cholestasis-to-cholangiocarcinoma (CCA) transition and the role of nuclear receptors (NRs). Cholestasis significantly increases the risk of CCA. Various hepatic NRs, such as farnesoid X receptor (FXR) and peroxisome proliferator-activated receptors (PPARs), generally mitigate cholestatic liver injury and suppress CCA progression. However, the precise mechanism linking cholestasis to CCA initiation remains largely unclear and still controversial. For instance, although it is well expected that PPARα suppresses CCA proliferation in vitro and tumor growth in the xenograft model, its activation unexpectedly promotes CCA development in the c-Jun N-terminal kinase 1/2 (JNK1/2) double-knockout (KO) liver. Therefore, comprehensive investigations should be warranted to elucidate the pharmacologic effects of NR ligands in cholestasis to CCA and to develop NR-based targeted therapeutics. The red blunt arrow and blue arrow represent inhibitory and stimulatory effects, respectively. Names of NRs described in the figure are listed in Table 4. The image was created with BioRender.com (Toronto, ON, Canada). DDC, 3,5-diethoxycarbonyl-1,4-dihydrocollidine; ER, estrogen receptor; HNF4α, hepatocyte nuclear factor 4 α; IDH-Mut, a mutation in isocitrate dehydrogenase gene; LN, lymph node; RAR, retinoic acid receptor; RXR, retinoic acid X receptor; VDR, vitamin D receptor.

Cholestasis/PSC-to-CCA Pathology

BA-related liver diseases, such as PSC, developmental bile duct malformation, like Alagille syndrome, and progressive familial intrahepatic cholestasis have been identified as risk factors for CCA. PSC is a rare form of chronic liver disease characterized by chronic inflammation and fibrosis of the bile ducts within the liver. The exact cause of PSC pathology is largely unknown, although combinations of genetic, environmental, and immunologic factors are believed to play significant roles. It is often associated with other autoimmune conditions, such as inflammatory bowel disease, particularly ulcerative colitis, both of which are closely correlated with CCA.²⁸ Approximately 20% of patients with PSC are reported to develop iCCA, with a calculated odds ratio of 22.9, as per findings from a meta-analysis.^1,29 Consequently, one of the most concerning risks associated with PSC–inflammatory bowel disease is the emergence of malignancies in the intestines and hepatobiliary system, specifically CCA, which has a survival rate of <10% over a span of 5 years.^1,28,29 PSC often provokes CCA in younger patients compared with non–PSC-CCA cases and shows a higher correlation with underlying chronic cirrhosis associated with ulcerative colitis compared with non–PSC-CCA. Moreover, most PSC-associated CCAs arise from iCCA in a multifocal form, whereas extrahepatic CCA is much less common. Unlike HCC, where >90% of HCCs arise from a singular nodule, multifocal PSC-CCA cases suggest the presence of pan-hepatic oncogenic signals along with baseline mutations, conceivably linked to specific inflammation signatures.

In murine models, it was revealed that the pathologic microenvironment driving regenerative hepatocyte (HC)–to–biliary epithelial cell (BEC) conversion during cholestasis is involved in a transformation of CCA, when combined with significant contribution of inflammatory/pro-oncogenic microenvironment.^48,49 Although cholestasis-associated murine liver cancer models, such as the Mdr2 knockout, tend to favor HCC development, recent studies have demonstrated that combined HCC-CCA originates from hepatic progenitor cells without biliary fate commitment signals, which potentially mimics human PSC pathology toward CCA.⁵⁰ Similarly, a cholestatic condition induced by partial bile duct ligation alongside the HCC-inducing carcinogen N-nitrosodiethylamine treatment preferentially develops CCA.⁵¹ Mechanistically, hepatic damage by excess BAs, directly and indirectly, induces the activation of extracellular signal-regulated kinase 1/2, AKT, and NF-κB pathways to promote BEC proliferation and CCA.^52,53 BAs also transactivate carcinogenic epidermal growth factor receptor signaling to induce cyclooxygenase 2 (COX2), which collectively contributes to the initiation and progression of CCA.^54,55 Therefore, understanding the cholestasis/PSC-to-CCA pathology is crucial to identifying key biomarkers and/or driving factors in cholangiocarcinogenesis in PSC livers, having a huge clinical impact on PSC society. In particular, signal cascades for HC-to-CCA transformation have been validated in murine model systems using the hydrodynamic tail vein injection technique, which will be discussed further in this review.

Cellular Origins of CCA and Murine PSC-CCA Models

Traditionally, CCA is classified into three groups based on anatomic locations: iCCA, perihilar CCA, and extrahepatic CCA. Although all classes displayed histologic similarity within the tumors, they show distinct pathologic progression/behavior, heterogeneous molecular signatures, and diverse responses to targeted/personalized therapeutic approaches.^1,14,16 Theoretically, CCA is believed to derive from cholangiocytes (ie, BECs), as tumor cells exhibit luminal structures and express BEC-specific markers, including cytokeratins, Epcam, and Sox9. However, given the frequent detection of human CCA in the pericentral area of the liver lobule, which is anatomically separated from the biliary trees, and the well-documented occurrence of HC-to-CCA fate transition in diverse murine cholestasis models, HCs have been proposed as the alternative source for CCA.^1,16,56, 57, 58, 59, 60

Murine HC-Derived CCA Models

Combined expression of proto-oncogenes and biliary fate genes in HC successfully generates CCA models, which include myristoylated AKT (AKT)–NOTCH intracellular domain (NICD),^57,60 AKT–constitutively active YAP1^S127A,⁶⁰ KRAS^G12D-Cas9-sgp19,⁶¹ AKT-Fbwx7^Δ,⁶² KRAS^G12D-shp53, and more. Overexpression of these combinations initially induces trans-differentiation of HCs into BEC-like cells, producing lesions with similar pathology to those observed in cholestasis/PSC.^57,60 This suggests that HC-to-BEC differentiation under the cholestasis/PSC conditions may be pathologically contributing to the development of clinical CCA, and NR regulators, such as FXR, involved in hepatic cell fate conversion^63,64 may play a role in HC-derived cholangiocarcinogenesis in PSC. Moreover, chromatin and epigenome are largely remodeled during fate reprogramming, potentiating a higher chance of transforming into malignant cells.^65,66 Notably, transcriptome of these murine models of CCA significantly represents a subset of clinical CCAs, validating the clinical relevance of studying human CCA using these models, although specific correlation with causes for HC-driven CCA models remains elusive.^49,61

Murine PCS-to-Spontaneous CCA Models

The Mdr2 gene encodes the canalicular phospholipid transporter,⁶⁷ whereas Abcb11 encodes the bile salt export pump,⁶⁸ responsible for exporting bile acids at the canalicular membrane of hepatocytes. Genetic elimination of the Mdr2 gene in mice abolishes phosphatidylcholine from bile, and ablation of Abcb11 impairs BA export, both leading to liver injury associated with fibrosis and exhibiting histologic lesions resembling clinical PSC pathology.^67,69 In the long-term, Mdr2 knockout animals mainly induce HCC, with a rare occurrence of CCA lesions,⁷⁰ whereas Abcb11 knockout mice provoke CCA resembling the sequelae of clinical CCA.⁷¹ Interestingly, a recent study using hepatic progenitor cell–specific elimination of Mdr2 in Foxl1-Cre;Mdr2^(f/f) mice demonstrated combined HCC-CCA formation is dominated by CCA, attributed to increased expression of Il6, highlighting its pivotal role in cholestasis/PSC-to-CCA pathology.⁵⁰ Considering FXR regulation of the IL-6 and STAT3 pathway in liver diseases,^50,72,73 investigating the effects of FXR activation on CCA formation in this advanced system would be intriguing. Several chemical models resembling the pathologic oncogenic microenvironments of cholestasis that eventually lead to CCA are available. These models are spontaneous and variable, with high penetrance, and involve long-term observations. Thioacetamide⁷⁴ and N-nitrosodiethylamine/dimethylnitrosamine^51,75 long-term exposure models are available in the mouse system, whereas the thioacetamide⁷⁶ and furan-CCA⁷⁷ model is established in the rat system. However, these models have yet to be used to investigate the effect of NR regulation thus far.

Farnesoid X Receptor

BA Therapy and the Role of FXR in Cholestasis

In clinical practice, the most reliable first-line therapy for patients with cholestasis, including those with primary biliary cholangitis (PBC) and PSC, is ursodeoxycholic acid (UDCA), a naturally occurring bile acid but not directly related to FXR, derived from hepatic cholesterol metabolism. In PBC, UDCA exhibits immunomodulatory effects, suppressing the autoimmune response that targets the bile ducts and reduces hepatic inflammation.78, 79, 80 Additionally, UDCA helps to improve liver function by reducing liver enzyme levels, such as alkaline phosphatase and γ-glutamyl transferase.⁷⁹ It also decreases bilirubin levels, which can alleviate jaundice symptoms.

FXR is a cognate BA receptor and modulates systemic BA homeostasis. FXR activation controls the expression of genes involved in BA synthesis, transport, and conjugation, helping to protect against BA toxicity in cholestatic conditions.⁸¹ In addition, FXR activation suppresses inflammation during cholestasis, generally ameliorating hepatic injury. Reduced FXR expression and activity have been found in human cholestatic liver diseases, and mutations in the FXR gene also cause severe early-onset cholestasis.⁸² FXR agonist obeticholic acid is US Food and Drug Administration approved for use in patients with PBC and recommended for those intolerants to the conventional UDCA treatment.⁸³ Encouraged by the reports demonstrating the beneficial effect of FXR activation in murine PSC models, there have been clinical trials for FXR agonists in PBC and PSC (Table 1); both displayed safety and improved cholestatic injury by reducing serum hepatic enzymes and clinical symptoms, suggesting further consideration for clinical use.^36,37 However, conflicting findings^84,85 from animal studies demonstrate the possibility of FXR exerting beneficial or detrimental effects in a context-dependent manner, highlighting the need for close investigation into its role in cholestasis.

Table 1.

Current and Completed Clinical Trials

NR	Ligands	Disease	Type/phase	Current status	ID∗
FXR	OCA	PBC	Phase 2	Completed in 2016	NCT01865812
			Phase 3	Completed in 2018	NCT01473524
			Observational	Completed in 2022	NCT05292872
		PSC	Phase 2	Completed in 2018	NCT02177136
	Cilofexor	PBC	Phase 1	Completed in 2018	NCT02808312
			Phase 2	Completed in 2020	NCT02943460
PPARα	Bezafibrate	PBC	Phase 3	Completed in 2016	NCT01654731
			Observational	Recruiting†	NCT04514965
		PSC	Phase 3	Recruiting†	NCT04309773
	Fenofibrate	PBC	Phase 2	Completed in 2010	NCT00575042
			Phase 2/3	Recruiting†	NCT06365424
VDR	Vitamin D with UDCA	PBC	N/A	Completed in 2023	NCT06309589

FXR, farnesoid X receptor; ID, identifier; N/A, not applicable; NR, nuclear receptor; OCA, obeticholic acid; PBC, primary biliary cholangitis; PPAR, peroxisome proliferator-activated receptor; PSC, primary sclerosing cholangitis; UDCA, ursodeoxycholic acid; VDR, vitamin D receptor.

^∗Available at https://clinicaltrials.gov, last acccessed July 15, 2024.

^†As of July 15, 2024.

Versatile Roles of FXR on Liver Cancer Development and Progression

Numerous preclinical studies underline the potential of FXR agonists as therapeutic agents for liver cancers, typically HCC. In HCC models, FXR agonists, such as obeticholic acid and GW4064, show anti-tumorigenic effects by inhibiting tumor cell proliferation, promoting apoptosis, and suppressing tumor growth. Various FXR downstream effectors, including Socs3-leptin,⁷² p53-Ccnd1,⁸⁶ Socs3-p21,⁸⁷ CTNNB1,⁸⁸ mTor,^89,90 transcription activity of NDRG2,⁹¹ and extracellular signal-regulated kinase pathway,⁹² have been implicated. Notably, global FXR knockout mice, when aged, induce hepatocyte apoptosis and chronic liver injury, eventually leading to spontaneous HCC through elevated expression of proto-oncogenes, such as Ctnnb1 and/or c-Myc, implying potential roles for BA-FXR signaling in hepatocarcinogenesis.^39,93 However, despite the encouraging results in preclinical studies, there are currently no clinical trials evaluating the efficacy and safety of FXR agonists in patients with HCC, which potentially requires meticulous trial design.

In CCA, several in vitro studies using established CCA cell lines have yielded mixed results, and the impact of FXR agonists on CCA formation, particularly in the context of cholestasis, remains largely unexplored. Table 294, 95, 96, 97, 98, 99, 100, 101 summarizes the available literature demonstrating FXR function on CCA. In summary, FXR and small heterodimer partner (SHP) expression are significantly reduced in murine and human CCA regions compared with nontumor regions but are significantly enhanced by pharmacologic FXR activation94, 95, 96^,101 (Figure 2). FXR activation typically inhibits the proliferation of CCA cell lines in vitro,^94,95,97,102 whereas its effect in xenograft models is still controversial.96, 97, 98 This discrepancy may be because of the reliance on in vitro cell lines or mouse xenograft models in most previous studies,94, 95, 96, 97, 98^,101 with a limited assessment of FXR function on BA metabolism and CCA initiation/progression using a pathologically relevant CCA model. To this end, Dr. Liu and colleagues recently reported that obeticholic acid treatment slightly diminished spontaneous HCC-CCA formation and hepatic fibrosis in Abcb11 knockout mice.¹⁰⁰ However, given the multifaceted effects of BAs on CCA and FXR function (eg, a proven PBC/PSC drug UDCA is FXR antagonist^78,99,103), the overall impact of FXR activation or inhibition on CCA development requires a careful examination, particularly regarding the cellular plasticity during cholangiocarcinogenesis under cholestatic conditions. Therefore, it will be crucial to comprehensively investigate the roles of FXR agonists and NR regulators in this process. Understanding these roles will be critical in assessing the clinical implications of approving and considering FXR agonists and other chemicals for patients with PSC in the clinic.

Table 2.

The Role of FXR in Various Human CCA Cells and in Vivo Models

Variable	Cholangiocarcinoma	Outcome
In vitro
OCA	Primary human iCCA cells	Proliferation ↓, apoptosis ↑94
	RBE, HuCCT1, CC-LP-1	Proliferation ↓95
	MzCHA-1, KMCH, HepG2, Hep3B	Proliferation ↑96
	EGI1 and TFK1	Proliferation ↓, migration ↓97
UDCA	Primary human iCCA cells	No effect94
CDCA	GBC-CD and RBE	Viability ↓98
GW4064	GBC-CD and RBE	Viability ↓98
CA, DCA, CDCA	QBC939	Proliferation ↓99
GCA, GDCA, GCDCA	QBC939	Proliferation ↑99
shFXR	RBE, HuCCT1, CC-LP-1	Proliferation ↑95
In vivo
OCA
Oral	Spontaneously HCC, iCCA, mixed HCC/iCCA mice model (Abcb11−/−)	Inflammation↓, tumor incidence↓100
	S.c. xenograft mice model (Mz-ChA-1, HepG2)	Tumor burden ↑96
Diet	S.c. xenograft mice model (primary human iCCA cells)	Tumor burden ↓94
	Orthotropic xenograft mice model (RBE-RFP)	Tumor burden ↓, lung metastasis ↓95
	Orthotopic xenograft mice model (EGI1, TFK1)	Proliferation ↓97
GW4064
I.p.	Orthotopic xenograft rat (QBC939)	No effect101
	S.c. xenograft mice model (GBC-SD)	Sensitivity to cisplatin↑98
	S.c. xenograft mice model (QBC939)	Tumor burden ↑99
GDCA
Diet	S.c. xenograft mice model (QBC939)	Tumor burden ↓99
CDCA
Diet	S.c. xenograft mice model (QBC939)	Tumor burden ↑99

↑, Increased; ↓, decreased; CA, cholic acid; CCA, cholangiocarcinoma; CC-LP-1, human intrahepatic cholangiocarcinoma; CDCA, chenodeoxycholic acid; DCA, deoxycholic acid; EGI-1, human cholangiocarcinoma; FXR, farnesoid X receptor; GBC-SD, human gallbladder carcinoma; GCA, glycocholic acid; GCDCA, glycochenodeoxycholic acid; GDCA, glycodeoxycholic acid; GW4064, FXR agonist; HCC, hepatocellular carcinoma; Hep3B, human hepatocellular carcinoma; HepG2, hepatoblastoma; HuCCT1, human intrahepatic cholangiocarcinoma; iCCA, intrahepatic CCA; KMCH, human combined hepatocellular carcinoma and cholangiocarcinoma; Mz-ChA-1, human gallbladder carcinoma; OCA, obeticholic acid; QBC939, human extrahepatic cholangiocarcinoma; RBE, human intrahepatic cholangiocarcinoma; shFXR, small hairpin RNA targeting FXR; TFK1, human extrahepatic cholangiocarcinoma; UDCA, ursodeoxycholic acid.

Figure 2.

Expressions of hepatic nuclear receptors (NRs) in cholangiocarcinoma (CCA) and hepatocellular carcinoma (HCC). The Cancer Genome Atlas Program (TCGA) data set [TCGA–cholangiocarcinoma study (CHOL) and TCGA–liver hepatocellular carcinoma study (LIHC)] was analyzed using the Gene Expression Profiling Interactive Analysis platform. Each NR expression in tumor (red) and normal (gray) regions was compared in CCA and HCC. Gene names of NRs described in the figure are listed in Table 4. ER, estrogen receptor; FXR, farnesoid X receptor; HNF4, hepatocyte nuclear factor 4; PPAR, peroxisome proliferator-activated receptor; RAR, retinoic acid receptor; RXR, retinoic acid X receptor; TPM, transcripts per million; VDR, vitamin D receptor.

Peroxisome Proliferator-Activated Receptors

Peroxisome proliferator-activated receptors (PPARs) play a crucial role in the regulation of genes involved in lipid utilization, lipoprotein metabolism, adipocyte differentiation, and insulin action. This family comprises three isoforms (PPARα, PPARγ, and PPARδ), each exerting distinct metabolic functions across diverse tissues and cell types.¹⁰⁴

PPARα

PPARα is predominantly expressed in the liver, heart, and muscle, playing a central role in regulating fatty acid transport, catabolism, and energy homeostasis. In general, it facilitates fatty acid oxidation and gluconeogenesis to mediate fasting responses.¹⁰⁵ Clinical applications using PPARα agonists (ie, fibrates) initially targeted hypertriglyceridemia¹⁰⁶ but have since expanded to encompass various disease conditions, such as cardiovascular disease.¹⁰⁷ In PBC and PSC, fibrates improve liver function and reduce toxicity¹⁰⁸ because PPARα activation suppresses de novo BA synthesis and enhance BA secretion,¹⁰⁹ highlighting the necessity for clinical trials (Table 1). However, despite the potential beneficial effects on PBC and PSC, PPARα activation in hepatic JNK1/2 knockout unexpectedly contributes to CCA development, whereas loss of Ppara normalizes BA levels and reduces CCA prevalence,¹¹⁰ necessitating further substantial investigation into the diverse cholestasis and/or CCA settings (Table 3).110, 111, 112, 113, 114, 115 Notably, the efficacy of fibrates in CCA development has not been explored in established CCA models, mandating the importance of careful evaluation of the model-specific function of PPARα in future studies.

Table 3.

The Role of PPARs in CCA

Target	Cholangiocarcinoma	Outcome
PPARα
Ppara knockout	Spontaneous CCA mice model (JNK1−/−;JNK2−/−)	Tumor burden ↓110
PPARδ
GW501516 (PPARδ agonist)	Mz-ChA-1	Proliferation ↑111
PPARγ
15d-PGJ(2) (PPARγ agonist)	SG231, CC-LP-1, HuCCT1	Proliferation ↓112
Troglitazone (PPARγ agonist)	SG231, CC-LP-1, HuCCT1	Proliferation ↓112
	HuCCT1, HuH-28	Proliferation ↓113
Pioglitazone (PPARγ agonist)	CC-LP-1, MzChA-1	Viability ↓114
	Gemcitabine-resistant cell (CC-LP-1-GR, Mz-ChA-1-GR)	Viability ↓114
miR-130a (PPARγ)	CC-LP-1, Mz-ChA-1	Gemcitabine resistance ↑114
siPPARγ (PPARγ)	SNU-1079 and SNU-1196	Viability↓, migration↓, invasion ↓115

↑, Increased; ↓, decreased; CCA, cholangiocarcinoma; HuH-28, human intrahepatic cholangiocarcinoma; PPAR, peroxisome proliferator-activated receptor; SG231, human intrahepatic cholangiocarcinoma; SNU-1079, human intrahepatic cholangiocarcinoma; SNU-1196, human hilar cholangiocarcinom.

PPARγ

PPARγ was initially recognized as a key regulator of adipocyte differentiation and lipid metabolism. Additionally, its function has been broadly implicated in many diseases, including obesity/diabetes, atherosclerosis, and cancer.¹¹⁶ Moreover, PPARγ has demonstrated anti-inflammatory and protective properties across a spectrum of disease models, such as atherosclerosis, insulin resistance, Parkinson, Alzheimer, and inflammatory bowel diseases, by directly interfering with NF-κB signaling.¹¹⁷ These anti-inflammatory and protective properties have also been observed in intrahepatic cholestasis models.118, 119, 120

In CCA models, multiple reports generally support the inhibitory effect of PPARγ on the CCA cell line proliferation in vitro, partly through the P53-dependent pathways.112, 113, 114 However, conclusive in vivo evidence in established CCA models examining the initiation and progression of CCA remains elusive. Recently, single-cell RNA sequencing of CCA and lymph node metastases unexpectedly revealed PPARγ as a critical metabolic regulator driving CCA colonization in lymph nodes.¹²¹ This underlines a context-dependent systemic impact of PPARγ, linking CCA development with the immune-suppressive microenvironment.

PPARδ

PPARδ, also referred to as PPARβ, is ubiquitously expressed across various tissues, including the intestine, liver, skin, heart, spleen, and skeletal muscle.¹²² In skeletal muscle, PPARδ activation stimulates fatty acid oxidation and mitochondrial biogenesis, leading to a shift toward more oxidative type I muscle fibers and enhancing endurance capacity on exercise training.¹²³ Similar to other PPARs, PPARδ is linked to the reduction of BA synthesis, suppression of inflammation, and inhibition of hepatic fibrosis,^124,125 paving the way for recent clinical trial initiatives.^109,126,127 Similar to other members of the PPAR family, the roles of PPARδ in CCA are generally limited, while there is a publication demonstrating the pro-CCA effect of PPARδ overexpression or its agonist (GW501516) treatment, which promotes the proliferation of human CCA cells.¹¹¹

Estrogen Receptors

Lithogenic Effect of Estrogen in Bile Diseases

Estrogen emerges as an important risk factor for cholesterol gallstones by promoting the secretion of biliary cholesterol and causing mitochondrial damage in BECs, primarily through hepatic estrogen receptor (ER) α signaling.^128,129 It also plays a role in the pathogenesis of intrahepatic cholestasis of pregnancy, suggesting its pathologic significance in understanding sexual dimorphism in cholestatic liver diseases.¹³⁰ Although there have been suggestions that ER antagonists (eg, tamoxifen) could improve liver function in PBC,¹³¹ their efficacy remains inconclusive because of conflicting outcomes and potential adverse effects.^129,132

ER Functions in CCA Development

In CCA models, estrogen stimulates CCA development partially by inducing IL6 expression, as similarly observed in patients with CCA.^129,133,134 In IDH1-mutated CCA cells, ER activation enhances CCA cell proliferation and tumor growth.¹³⁵ Molecular characterization has elucidated ER-positive CCA development.136, 137, 138 Although ER antagonists have shown promise in suppressing tumor growth and promoting apoptosis,^136,139,140 further studies are needed to certify the efficacy of ER antagonists and long-term adverse effects in established CCA models.

Other Nuclear Receptors

Hepatocyte Nuclear Factor 4 α

HNF4α is highly expressed in the hepatocytes, serving as a master regulator of liver development hepatic energy homeostasis.¹¹¹ The absence of HNF4α results in significant metabolic dysregulation and increased mortality,¹⁴¹ with implications in BA metabolism¹⁴² and HCC progression.¹⁴³ During cholestasis, HNF4α has been shown to modulate liver functions.^144,145 Restoring Hnf4a mRNA attenuates cholestasis-induced liver fibrosis.¹⁴⁶ As similarly demonstrated in the N-nitrosodiethylamine–induced HCC model,¹⁴⁷ inhibition of HNF4α contributes to impaired hepatocyte differentiation to promote CCA in IDH-mutated animals,¹⁴⁸ suggesting the essential roles in HC-driven cholangiocarcinogenesis. Considering the repressive roles of DNA methyltransferase 1 (DNMT1)–HNF4α cascade in HC transformation into CCA, it would be essential to evaluate its preventive and therapeutic potential on PSC-to-CCA pathology.

Vitamin D Receptor

In the liver, nonparenchymal cells, such as Kupffer cells, sinusoidal endothelial cells, and hepatic stellate cells, express Vdr, exerting protective effects against steatohepatitis, liver fibrosis, and cholestasis.^149,150 Loss of Vdr accelerates cholangiopathy by impacting BEC tight junction proteins,^151,152 nominating vitamin D receptor (VDR) as a potential therapeutic target for cholestasis. In patients with CCA, higher VDR expression correlates with better prognosis, and VDR ligands effectively suppress CCA cell proliferation,153, 154, 155 consistent with observation in thioacetamide-induced CAA animal models.¹⁵⁶

RARs and RXRs

Retinoic acid receptors (RARs) and retinoic acid X receptors (RXRs) play essential roles in embryonic development, cell fate determination, metabolic regulation, and cell death,^157,158 activated by 9-cis retinoic acids (RAR and RXR) and all-trans-retinoic acid (RAR only). Hepatic expression of Rara and Rxra is reduced in bile duct ligation–induced cholestasis,^159,160 whereas its expression does not reach statistical significance in patients with PBC.¹⁶¹ Retinoid treatments alleviate bile duct ligation–induced cholestasis,^162,163 suggesting a new therapeutic approach for patients with PBC.¹⁶⁴ In CCA, ligand-activated RARβ and RXRγ induce apoptosis and enhance chemotherapy sensitivity,165, 166, 167 whereas RARγ and RXRα exhibit the opposite effect.^168,169 Further studies are needed to determine the procarcinogenic and anti-carcinogenic effects of different RAR and RXR isoforms in established CCA models.

Concluding Remarks

Hepatic NRs (Table 4) provide one of the most rapid responses in regulating physiological aspects through diverse biological ligands, including hormones, thereby offering a platform to regulate precise and tight controls, such as BA and hepatic metabolic maintenance, which are directly linked to liver cancer development, including CCA. Basically, they share various ligands and their cognate receptors, such as BAs and fatty acids, thus inducing unexpected systemic effects and making it extremely difficult to anticipate the pharmacologic effects. Meanwhile, the development of numerous analogs through biochemical approaches is actively linked to beneficial effects in clinical trials for liver diseases, including cholestasis.

Table 4.

NRs Described in This Review

NRs	Abbreviation	Isotypes (gene names)
Farnesoid X receptor	FXR	FXRα (NR1H4), FXRβ (NR1H5)∗
Peroxisome proliferator-activated receptor	PPAR	PPARα (PPARA), PPARγ (PPARG), PPARδ (PPARD)
Estrogen receptor	ER	(ESR1)
Hepatocyte nuclear factor 4 α	HNF4α
Vitamin D receptor	VDR
Retinoic acid receptor	RAR	RARα (RARA), RARβ (RARB), RARγ (RARG)
Retinoic acid X receptor	RXR	RXRα (RXRA), RXRβ (RXRB), RXRγ (RXRG)

NR, nuclear receptor.

^∗FXRβ gene is not present in humans and is not expressed in the murine liver. In this review, FXR refers specifically to the FXRα isoform.

Hepatic metabolism, especially BA metabolism, primarily governed by NR families, is a critical pathologic factor for CCA. However, compared to the massive publications on HCC, publications on CCA are severely limited. Previous studies have often relied on limited in vitro (ie, cell line) and in vivo (ie, xenograft) models, which do not adequately reflect the pathogenesis preceding the cholestatic state. Addressing this gap in the knowledge necessitates focused research efforts on cholestasis-to-CCA pathology. Deciphering the molecular mechanisms by which NR function influences cholestasis and CCA pathobiology is essential for improving the basic understanding and clinical management of both cholestasis and CCA. By unraveling the molecular mechanisms underlying NR-mediated regulation of hepatic pathophysiology and CCA progression, we can pave the way for more effective treatments and improved outcomes for patients with these devastating liver diseases.

Disclosure Statement

None declared.