Dysregulated cholesterol regulatory genes in hepatocellular carcinoma

The carcinogenic effect of HMGCR has been reported in many tumors, such as gastric cancer, ovarian cancer and breast cancer [11]. HMGCR was higher in HCC and positively correlated with poor prognosis in patients with HCC [12, 13]. MYC is an underlying target for several malignancies. Interfering with HMGCR inhibited HCC growth by activation of MYC, which suggested that HMGCR might be an effective target for the treatment of HCC [14]. Heat shock protein 90 (HSP90) promoted proliferation and inhibited apoptosis of HCC cell, but its mechanism was not clear. Li et al. confirmed that HSP90 promoted the proliferation and migration of HCC cell by increasing the expression of HMGCR [15]. Forkhead Box M1 (FoxM1) was recognized as a factor that promoted the development and progression of HCC. One study showed that HMGCR-inhibition increased HCC cell death via decreasing the expression and the transcriptional activity of FoxM1 [16].

As the second rate-limiting enzyme of cholesterol synthesis, cholesterol metabolism disorders caused by the dysregulation of SQLE expression are closely related to tumor development, which are involved in various processes such as tumor proliferation, apoptosis, epithelial mesenchymal transition and metastasis [17, 18]. The upregulation of SQLE expression was associated with advanced HCC histological grade, advanced AJCC stage, elevated α-fetoprotein and poor clinical outcome [19‐21]. Liu et al. found that SQLE increased the expression of SOAT1/2 to promote intracellular cholesteryl ester synthesis, which in turn accelerated HCC cells growth [19]. SQLE also increased Akt activity by knockdown PTEN. Specifically, the conversion of squalene to 2,3-epoxy squalene catalyzed by SQLE required consumption of NADPH, and the NADPH deficiency induced oxidative stress in HCC cells, which upregulation of DNMT3A expression leaded to PTEN silence and contributed to Akt–mTOR pathway activation [19]. P53, the most common mutant gene in malignancies, regulates multiple functions such as apoptosis, cell cycle arrest and senescence. P53 repressed HCC growth by directly inhibiting SQLE transcription and reducing cholesterol synthesis. SQLE over-expression accelerated HCC growth, which further indicated that SQLE played an important role in p53-mediated tumor suppression [22]. In addition, SQLE promoted the development of HCC by activating TGFβ/SMAD pathway [23]. The above studies suggested that SQLE might be a potential therapeutic target for HCC.

SREBP2 is engaged in the regulation of genes involved in cholesterol homeostasis. Moon et al. found that p53 also regulates SREBP2. Specifically, P53 inhibited the mevalonate pathway by preventing SREBP2 maturation, ultimately slowing down the progression of HCC [24]. ASPP2 interacted with p53 and stimulated the p53-mediated anti-tumor effects. Bai et al. found that the anti-tumor activity of ASPP2 was closely related to SREBP2. They proved that ASPP2 interacted with SREBP2 in the nucleus and reduced the transcriptional activity of SREBP2, which resulted in the down-regulation of the expression of the key enzymes in the mevalonate pathway, ultimately inhibiting the progression of HCC [13]. Chen et al. indicated that FASN deletion promoted nuclear localization and activation of SREBP2, which triggered de novo lipogenesis and cholesterol biosynthesis, eventually leading to hepatocarcinogenesis [25]. Epithelial–mesenchymal transition (EMT) plays an important role in the heterogeneity and tumor metastasis. SREBP2 promoted HCC cells invasion and metastases by inducing EMT [26]. Moreover, recent studies have shown that XBP1-u promoted tumorigenesis. Mechanically, XBP1-u colocalized with SREBP2 and suppressed its ubiquitin/proteasome degradation, leading to cholesterol synthesis and lipid accumulation, finally inducing liver carcinogenesis [27].

NPC1L1, a membrane protein with 1332 amino acids, is abundantly expressed in the small intestine and liver. It plays an important role in intestinal absorption of cholesterol. Most of the studies about NPC1L1 in tumors focus on clinical research and bioinformatics analysis. Chen et al. found that HCC patients with a relatively low expression level of NPC1L1 had a poor clinical outcome and were more prone to occur HCC recurrence [28]. HCV infection is a major risk for the development of HCC. NPCL1 not only contributed to HCV enter into cells, but also promoted HCV cell-to-cell spread, might contributing to development of HCV-HCC [29]. Drug-resistant recurrence is a major challenge in treatment of tumor. Zhang et al. found that NPC1L1 was closely associated with drug-tolerant persister state of tumor cells. Inhibition of NPC1L1 induced oxidative stress‐mediated cell death and disrupted adaptive responses of drug-tolerant persister cells to chemotherapy [30]. The molecular mechanism of NPC1L1 in HCC has not yet been reported. Whether the abnormal expression of NPC1L1 will induce drug resistance in HCC cells, it seems to be worth exploring.

LDLR is a transmembrane glycoprotein located on the cell membrane. Its binding to LDL promotes cholesterol uptake by endocytosis. The expression level and prognostic value of LDLR varies in different types of tumors. The high expression of LDLR indicated poor prognosis in patients with ovarian cancer [31]. In renal cancer tissues, LDLR expression was higher than in normal renal tissues. Knockdown LDLR inhibited renal cancer cell proliferation and induced cell cycle arrest [32]. LDLR was considered as an essential co-receptor for the HCV entry into cells. Knockdown LDLR hindered HCV infection [33]. Moreover, another study demonstrated that LDLR regulated the activation of MAPK/ERK pathway triggered by HCV E2, thereby maintaining the growth and survival of Huh-7 cells [34]. These resulted showed that LDLR might played an important role in progression of HCV-HCC. Mamatha et al. found that the expression of LDLR expression was higher in HCC tissues, which contributed to the uptake of exogenous cholesterol in HCC [35]. In contrast, Chen et al. observed that the expression level of LDLR was lower in HCC samples. The lower LDLR expression might be an indicator of poor clinical outcome in HCC patients. Silence LDLR activated MEK/ERK pathway and facilitated cholesterol synthesis, ultimately promoting HCC cell proliferation and metastasis [36]. The difference of findings might be attributed to the strong heterogeneity of HCC in different studies. The molecular mechanisms of LDLR in HCC still need to be explored.

The researches of NPC1 in HCC focus on bioinformatics analysis. Based on analysis of ICGC and TCGA databases, the higher NPC1 suggested unfavorable prognosis of HCC patients. The results of proteomic analysis also showed that NPC1 expression was upregulated in HCC tissues [37, 38]. Du et al. found that NPC1 inhibited the proliferation and metastasis of Huh7 cells in vitro, but the mechanism has not been reported [39]. NPC2, a small secretory glycoprotein, is widely located on lysosomal compartment. The HCC patients with lower NPC2 had higher α-fetoprotein, later histological stage and poorer prognosis. NPC2 silence promoted HCC cell proliferation, migration and xenograft tumorigenesis by regulating ERK1/2 activation [40]. Meanwhile, a study reported that NPC2 knockdown attenuated the therapeutic effect of sorafenib by activating MAPK/AKT signal pathway in HCC cells [41]. Moreover, other study reported that NPC2 interaction with GNMT triggered cholesterol accumulation, which might may provide novel therapeutic strategies for HCC [42].

GRAMD family proteins consist of GRAMD1A, GRAMD1B, GRAMD1C, GRAMD2, and GRAMD3, of which GRAMD1A, GRAMD1B, and GRAMD1C mediate cholesterol transport from plasma membrane to endoplasmic reticulum. Fu et al. found that the expression of GRAMD1A was increased in HCC tissues, and its higher expression was positively associated with unfavorable clinical outcome of HCC patients [43]. Mechanically, GRAMD1A accelerated self-renewal of HCC stem cells and progression of HCC through upregulating STAT5 transcriptional activity [43].

LXRs, including LXRα and LXRβ isoforms, performs a critical function in maintenance intracellular cholesterol homeostasis. LXR is activated by LXR agonists, and then LXR combines with retinoid X receptor (RXR) to form a heterodimer. The LXR–RER heterodimer binds to LXR-responsive element (LXRE) and regulates the expression of genes involved in cholesterol metabolism [44]. Hepatitis virus infection and liver steatosis are the most main risk factors for HCC. Na et al. found that LXR interacted with HBV X protein (HBx) in the nucleus and enhanced the transactivation function of LXR, inducing upregulation of SREBP-1c and FAS to accelerate lipid droplets accumulation, finally resulting in HBV-associated hepatic carcinogenesis [45]. Kin et al. also observed similar results [46]. HCV core protein might contribute to hepatic steatosis and HCV replication through LXR-regulated lipogenesis [47]. One study demonstrated that HCV core protein promoted the binding LXR between LXRE to activate SREBP-1c promoter activity, finally facilitating occurrence of hepatic steatosis and HCC [48]. Bakiri et al. proved that c-Fos down-regulated expression and activity of LXRα, resulting in alteration of hepatocyte morphology, infiltration of immune cells and formation of necrotic foci, ultimately promoting proliferation, dedifferentiation and DNA damage [49]. FoxM1 is related to progression of HCC by increasing the expression of cell cycle genes such as cyclin D1 and cyclin B1. Hu et al. found that LXRα binds to an inverted repeat IR2 in the promoter region of FoxM1 gene and decreased the expression of FoxM1, cyclin D1 and cyclin B1, thereby inhibiting proliferation and growth of HCC cells [50]. He et al. demonstrated that LXRα upregulated miR-134-5p while down-regulate FoxM1 by decreasing HULC, eventually suppressing HCC cells growth [51]. Moreover, LXRα increased expression of SOCS3 by enhancing the mRNA stability, then suppressing cyclin D1 and elevating p21 and p27, finally leading to cell cycle arrest and inhibiting HCC cells growth [52]. The upregulation of LXRα reduced TGFβ-induced Snail expression, leading to suppressing mesenchymal differentiation, the generation of reactive oxygen and promoting apoptotic response [53]. Morén et al. demonstrated that LXRα activation suppressed TGFβ-induced differentiation of cancer-associated fibroblasts by repressing the promoter activity of ACAT2, resulting in limiting HCC growth [54]. Lin et al. observed that LXRα upregulated the transcription level of miRNA-378a-3p and enhanced anti-tumor efficacy of sorafenib [55]. Drug induced lipotoxicity might be a potential therapeutic strategy for the treatment of HCC. A research demonstrated that LXRα activation and Raf inhibition leaded to accumulation of toxic saturated fatty acid, inducing the apoptosis of HCC cells [56]. Furthermore, LXR might serve as a marker for HCC prognosis. Long et al. found that LXR was much lower in HCC. The 5-year survival rate of patients with low LXR expression was lower than that of patients with high LXR expression [57].

ATP-binding cassette (ABC) transporters include a large family of membrane-bound proteins, some of which are associated with cholesterol efflux. ABC transporters are divided into seven subfamilies, known as ABCA-ABCG. Li et al. observed that ABCA1, cholesterol efflux transporter, upregulated in tumor monocytes/macrophages, thereby resulting in the production of immature and immunosuppressive monocytes/macrophages. High numbers of ABCA1 + monocytes/macrophages in HCC decreased CD8 + T cell infiltration, consequently leading to an unfavorable prognosis for HCC [58]. Cancer stem cells (CSCs) play a vital role in mediating unrestrained cell proliferation and chemoresistance. Hu et al. demonstrated that the over-expression of ABCA1 facilitated drug resistance of Lgr5 + HCC-CSCs cells to doxorubicin [59]. Xi et al. proved the higher ABCG1 predicted poor prognosis in patients with HCC [60]. One study demonstrated that ABCG1 knockdown leaded to decreased oxaliplatin resistance, indicating that that it played an important role in acquiring drug resistance of HCC cells [61]. Liao et al. observed similar result that ABCG1 silence reversed the oxaliplatin resistance in HCC [62]. Zhao et.al found that the down-regulation of ABCC6 accelerated cell proliferation, suppressed cell cycle arrest and cell apoptosis by deactivating PPARα [63]. Additionally, Researchers also found that the low expression of ACAB8 contributed to epithelial–mesenchymal transition by mediating ERK/ZEB1 pathway, consequently promoting proliferation and metastasis of HCC [64].

SR-B1, widely expressed in liver and steroidogenic cells, promoted cholesterol efflux from peripheral tissues, such as macrophages back to liver. Higher SR-B1 expression has been recognized as a biomarker of cancer progression, such as melanoma and lung cancer [65, 66]. The SR-B1 interaction with E1/E2 heterodimer triggered HCV enter into liver cell by endocytosis [67]. However, the molecular mechanism of SR-B1 in HCC needs to be further explored.

SOAT, also known as acyl coenzyme A cholesterol acyltransferase (ACAT), is a membrane-bound enzyme that promotes esterification of cholesterol and fatty acids into cholesterol esters. Two isoforms of SOAT have been identified, SOAT1 and SOAT2. SOAT2 is predominantly found in fetal hepatocytes and intestinal epithelial cells, while SOAT1 is widely expressed in most cells [6]. The over-expression of SOAT1 has been recognized as an indicator of tumor progression in glioblastoma, pancreatic cancer, and prostate cancer [68‐70].The high expression of SOAT1 was positively correlated with poor prognosis of HCC patients [71]. Jiang et al. found that SOAT1 knockdown inhibited growth and migration of HCC via down-regulating the integrins and TGFβ signal pathway [72]. Reprogramming energy metabolism is a hallmark of cancer [73]. Wang et al. demonstrated that target SOAT1 remodeled cholesterol metabolism and enhanced immune cells to suppress HCC tumor growth [74]. The P53 plays an important role in mediating lipid metabolism and energy generation. Zhu et al. proved that P53 deficiency upregulated the expression of SOAT1 and facilitated cholesterol esterification and fatty acid synthesis, consequently leading to HCC growth [75]. These results suggested that SOAT1 might be a potential target for P53-deficient HCC. In addition, Chen et al. analyzed the relationship between SOAT1 genetic variants and HCC. They found that SOAT1 rs10753191 and a haplotype TGA were related to decreased HCC risk [76]. SOAT2 is predominantly distributed in hepatocytes and intestinal epithelial cells. Liu et al. suggested SOAT2 induced oxysterol accumulation and mediated oxysterol secretion, leading to promoting HCC growth [77].