Liver X receptors as potential targets for cancer therapeutics (Review)

Introduction

The nuclear receptor (NR) superfamily consists of important ligand-inducible transcription factors that are involved in various physiological processes, including cell differentiation, embryonic development and metabolism (1,2). NR signaling serves important roles in a number of diseases including cancer, diabetes and metabolic diseases (1–3). Liver X receptors (LXRs) are important members of the NR family that were originally identified as major sensors of dietary cholesterol (4). Two LXRs, LXRα (NR1H3) and LXRβ (NR1H2), were initially identified in a human liver cDNA library screen as orphan receptors (5–7). LXRs are involved in cholesterol synthesis and transport, glucose homeostasis, and the modulation of inflammatory and immune responses (8–10). LXRs exhibit the typical structure of the NR superfamily. LXRα and LXRβ harbor four distinct domains: i) An N-terminal activation domain (AF-1); ii) a DNA-binding domain with two zinc fingers; iii) a hinge domain that binds co-repressors in the absence of ligand; and iv) a C-terminal domain that contains a hydrophobic ligand-binding domain and a transactivation domain (11,12).

LXR ligands and mechanism of action

LXRs were initially identified as orphan receptors. However, later studies identified oxysterols, or oxidized metabolites of cholesterol, as the endogenous ligands of LXRα and LXRβ (13,14). Furthermore, 22(R)-hydroxycholesterol, 20(S)-hydroxycholesterol, 24-hydroxycholesterol, 7α-hydroxycholesterol and 27-hydroxycholesterol also activate LXRs (13,15). Additionally, microbial and plant-derived sitosterol, sitostanol and acanthoic acid from Rollinia function as LXR agonists (16,17). Since the initial identification of endogenous LXR ligands, synthetic ligands have been developed by numerous pharmaceutical companies. T0901317 was the first synthetic LXR agonist to be developed, and is the most widely used in basic research. However, this compound may also activate two other NRs: Farnesoid X receptor and pregnane X receptor (18,19). A more selective synthetic agonist, GW3965, was identified in a screen of GlaxoSmithKline (Brentford, UK) compounds (20). However, neither of these agonists is used therapeutically owing to their temporary hypertriglyceridemic effects (20,21). More recently, promising synthetic ligands including 22(E)-ergost-22-ene-1α, 2β-diol and N,N-dimethyl-3β-hydroxycholenamide have been developed with fewer side effects and greater potential for use in clinical testing (22,23).

In the absence of a ligand, LXRs form obligate heterodimers with the retinoid X receptor. These heterodimers bind to target gene promoters harboring the conserved LXR-response element, which consists of two direct repeats of AGGTCA separated by four nucleotides or one nucleotide. The heterodimer forms a complex with NR co-repressor (NCoR) to block transcription (24). Upon ligand binding to the heterodimer, NCoR is released from the complex and co-activators are recruited, leading to chromatin remodeling and target gene transcription (12).

Effects of LXRs in different types of cancer

LXRs and colon cancer

LXR agonists suppress colon carcinogenesis by several mechanisms (Fig. 1A). Uno et al (25) demonstrated that LXR agonists block the activation of Wnt signaling by suppressing the transactivation activity of β-catenin. In HCT116 cells, LXR agonists decrease the mRNA expression of β-catenin target genes, including MYC proto-oncogene, matrix metalloproteinase 7 and bone morphogenetic protein 4. Notably, in colon cancer cells, LXR agonists inhibit endogenous β-catenin activity and cell proliferation without inducing apoptosis. Consistent with this result, LXR ligands induce pyroptosis, rather than apoptosis, in human and murine colon cancer cells. Pyroptosis is dependent on caspase-1 activation. More importantly, LXRs also serve non-genomic roles due to the differential cytoplasmic localization of LXRβ. LXRβ is predominantly located in the cytoplasm of colon cancer cells and induces pyroptosis when bound by ligand (26,27). Previously, a study revealed that LXR activation leads to cell cycle arrest in colon cancer cell lines. Furthermore, proliferation markers in the colon are significantly increased in LXRαβ(−/−) mice compared with wild-type mice (28). Together, inhibition of the Wnt signaling pathway and cell cycle arrest serve important roles during LXR ligand-dependent inhibition of colon cancer cell proliferation (Table I). Ligand-mediated activation of LXR may suppress the growth of colon cancer (29).

Figure 1. (A and B) Model of LXR agonist activation in various types of cancer and tumor immunity. LXR, liver X receptor; APOE, apolipoprotein E; p27, cyclin-dependent kinase inhibitor 1B; Skp2, S-phase kinase-associated protein 2; SOC3, suppressor of cytokine signaling 3; FOXM1, forkhead box M1; EST, estrogen sulfotransferase; SULT1E1, sulfotransferase family 1E member 1; FASN, fatty acid synthase; PI3K, phosphoinositide 3-kinase; MAPK, mitogen-activated protein kinase; IFN-γ, interferon-γ; AKT, protein kinase B; CCR7, C-C chemokine receptor type 7; DC, dendritic cell.

Table I. Effects of LXR in different types of cancer.

Table I.

Effects of LXR in different types of cancer.

Ligand	Tumor type	Molecular mechanism	(Refs.)
LXRα and LXRβ	Colon	Wnt signaling pathway, cell cycle arrest and pyroptosis	(25–29)
LXRα and LXRβ	Prostate	AKT survival signaling, MAPK transduction pathways	(29–34)
LXRα and LXRβ	Breast	EST, E2F family members	(35–39)
LXRβ	Pancreatic	Inhibits cell proliferation, cell cycle arrest	(40,41)

[i] LXR, liver X receptor; AKT, protein kinase B; MAPK, mitogen-activated protein kinase; EST, estrogen sulfotransferase; E2F, E2 factor.

LXRs and prostate cancer

Liao and colleagues were the first to report that the LXR agonist T0901317 inhibits the proliferation of human prostate LNCaP cells (30). T0901317 decreases the number of cells in S-phase and increases the expression of the cyclindependent kinase inhibitor p27 (Fig. 1A). LXRα expression levels are associated with the sensitivity to T0901317-mediated growth inhibition. Additionally, T0901317-treated nude mice exhibited decreased growth of xenograft LNCaP tumors. Additional studies revealed that protein kinase B survival signaling is downregulated following T0901317 treatment, and that T0901317 induces the apoptosis of LNCaP cells in a lipid raft-dependent manner in vitro and in vivo (31). LXR agonists were demonstrated to inhibit cell proliferation and induce G1/S arrest through lipogenic activity (32). Furthermore, a prior study identified cross-talk between the androgen receptor and LXR, and determined that this interaction influences cellular cholesterol levels (33). The suppressor of cytokine signaling 3 pathway is also involved in LXR agonist-mediated prostate carcinogenesis (34) (Table I). Taken together, these results suggest that LXRs may be promising pharmacological targets for the treatment of prostate cancer.

LXRs and breast cancer

Estrogen metabolism serves an important role in breast cancer and LXR controls estrogen homeostasis (35) (Table I). Estrogen sulfotransferase (EST) is a transcriptional target of LXR. In in vitro and in vivo breast cancer models, LXR inhibits estrogen-dependent cancer cell proliferation by regulating the hepatic expression of EST. These results suggest that LXR inhibits breast cancer growth through a novel mechanism involving the regulation of estrogen homeostasis (35). However, LXR agonists were demonstrated to decrease the proliferation of several breast cancer cell lines independent of lipid biosynthesis (36). LXR agonists decrease the proliferation of cells in S-phase and induce G1 arrest and apoptosis in MCF-7 cells (37,38). Microarray analysis of gene expression revealed that LXR ligands target a set of common responsive genes, including those regulated by E2 factor family members (39).

LXRs and pancreatic cancer

Pancreatic cancer is one of the most fatal types of cancer; the 5-year survival rate of patients with pancreatic ductal adenocarcinoma (PDAC) is only 5%. LXR agonists inhibit cell proliferation, cell-cycle arrest and colony formation in pancreatic cancer cell lines and regulate multiple gene networks involved in cell-cycle arrest and growth factor signaling (40) (Fig. 1A). However, LXR agonists do not induce apoptosis in PDAC cells (40). Further study is required to clarify the types of cell death induced by LXR agonists. Notably, a prior study revealed that the LXR agonist 22(R)-dihydroxycholesterol (Oxy16) inhibits pancreatic cancer cell-induced paracrine Hedgehog signaling (Table I). However, this inhibition is independent of LXR activation (41).

LXRs and tumor immunity

It has been reported that LXRs are involved in innate and adaptive immune responses in various diseases (42,43). LXRαβ(−/−) mice are highly susceptible to intracellular bacterial infection, suggesting that LXR-dependent gene expression serves an important role in innate immunity (42). A-Gonzalez et al (44) revealed that LXR signaling promotes apoptotic cell clearance by macrophages and the maintenance of immune tolerance. LXR signaling is also involved in the regulation of other types of immune cell (Fig. 1B). LXR activation inhibits lymphocyte proliferation, and loss of LXR expression confers a proliferative advantage on lymphocytes (45). LXR agonists induce interferon-γ expression in macrophages and T-cells, and increase the survival rate and the tumor-free population of mice inoculated with tumor cells (46,47). However, further studies are required to clarify whether these effects are conserved in other model systems. In Th17 cells, ectopic expression of LXR negatively regulates differentiation, whereas loss of LXR expression promotes differentiation (48). The role of LXR in dendritic cells (DCs) and tumor immunity is controversial. Villablanca et al (49) reported that LXR ligands released from tumor cells inhibit C-C chemokine receptor type 7 (CCR7) expression on maturing DCs. In mice, abrogating LXR agonist release from tumor cells controls tumor growth by recovering DC migration to the tumor area (49). However, another study revealed that treatment with LXR agonists increased CCR7 mRNA expression in an immature DC line (50). Additional investigations are required to determine the underlying molecular mechanism of the potentially opposing roles of LXR.

The antitumor effects of LXR agonists are also evident in the tumor microenvironment. LXR agonists impair the compartmentation of vascular endothelial growth factor receptor-2 in lipid rafts and decrease tumor growth by inhibiting angiogenesis (51). Furthermore, LXRβ agonist treatment induces apolipoprotein E secretion by stromal and tumor cells and blocks tumor growth, angiogenesis and metastasis (52). However, further studies are required in order to investigate whether other types of cell in the tumor microenvironment are influenced by LXRβ agonists.

Conclusion

LXR ligands have been developed to treat various diseases, including diabetes, Alzheimer's disease, atherosclerosis and cancer. In spite of progress, a number of fundamental questions remain with respect to cancer therapeutics. A number of natural and synthetic LXR ligands have been identified. T0901317 and GW3965 are the most widely used agonists for mechanistic and functional studies of LXR activity. However, T0901317 has been reported to increase triacylglycerol levels in the plasma and liver and was therefore suggested to be a poor candidate for clinical application (29). New synthetic agonists, including LXR-623, were developed to minimize the side effects on plasma triacylglycerol levels for clinical testing. The development of novel agonists, particularly for cancer treatment, may accelerate basic and clinical applications. The expression profiles of the two LXRs in clinical samples are unknown, and whether these expression profiles are associated with disease subtype, pathological parameters or disease outcomes requires clarification. Endogenous ligands in tumor and stromal cells must also be identified. Taken together, these results suggest that LXRs may be potential targets for cancer therapeutics.

Acknowledgements

The present review was supported by the National Natural Science Foundation of China (grant nos. 81502088, 81502621 and 81300287), Special Postdoctoral Science Foundation of China (no. 2017T100335), the Natural Science Foundation of Jiangsu Province (grant no. BK20140539) and the Start-Up Research Funding of Jiangsu University for Distinguished Scholars (grant nos. 14JDG065 and 15JDG021).

References