HBx is a key player in the etiology of HCC and manipulates various biochemical pathways within host cells, which are vital for physiological and biochemical functions. HBx redirects these pathways to promote cell proliferation and invasion, impair DNA repair, and evade immune responses, collectively contributing to tumorigenesis.

In particular, the Wnt/β-catenin, Notch signaling pathways, and the insulin-like growth factor (IN/IGF) pathway have been associated with the organization of HBx’s carcinogenic influence. The Wnt/β-catenin pathway is upregulated in over 90% of HCC cases (67), it is activated via HBx ectopic expression combined with Wnt-1, nevertheless, this activation is crucial for stabilizing the effect of HBx on β-catenin (27). In turn, the inhibition of Wnt/β-catenin pathway antagonists, SFRP1 and SFRP5, by HBx increases HCC cell proliferation and epithelial-mesenchymal transition (EMT) (28). Concurrently, HBx also leads to prolonged activation of the oncogenic pathway Notch by upregulating its receptors and ligands, particularly Notch-1 and Jagged-1 (29, 30). This Notch pathway activation is further amplified by HBx’s effect on the Erk (MEK1/2) and PI3K/AKT pathways (68). Furthermore, HBx alters the IN/IGF pathway, which acts in tandem with the Wnt/β-catenin pathway. IN a double-transgene mouse model of HBx and IRS1 that mimic human hepatocellular carcinoma, persistent activation and cross-talk of IN/IGF1, WNT/β-catenin, and Notch enhance HCC cell metastasis and invasiveness, and up-regulation of aspartate β-hydroxylase (ASPH) is central to these signaling cascades (31). The above mechanism exploration is primarily based on mouse models and cells; however, further clinical experiments are required to validate its potential as a therapeutic target for liver cancer patients.

HBx also compromises DNA repair, a critical tumorigenic facet of HBV-induced HCC. The HBx protein can impede the repair of damaged DNA mediated by the anti-oncogene p53 (32), inhibition of transcription factor IIH (TFIIH) (33), and degradation of Smc5/6 complex (34). This disruption of DNA repair processes leads to the accumulation of DNA damage, subsequently instigating the onset of hepatocarcinogenesis.

On the immune evasion front, HBx uses several strategies to avoid the innate immune response, particularly suppression of interferons. It suppresses Toll-like receptor 3 (TLR3) and its adaptor protein, TIR-domain containing adaptor protein inducing interferon-β (TRIF), thereby reducing the anti-HBV immune responsee (35). Furthermore, HBx inhibits interferon β (IFN-β) signaling by interacting with IPS-1, suppressing interferon induction during HBV infection (36). Moreover, HBx facilitates adenosine deaminases acting on RNA 1 (ADAR1)-mediated viral RNA editing, which prevents the recognition of HBV RNA by pattern recognition receptors, thereby inhibiting interferon production (37). Patients with HBV immune tolerance experience a higher incidence of HCC compared to those who receive immune activation therapy (69). Type I interferons are approved as a first-line treatment for chronic HBV. However, it is speculated that the effectiveness of IFN therapy may be limited due to inhibition by HBx, and its significant side effects limit its broader clinical application. Thus, there is a need to develop more effective immunotherapeutic drugs to curb the progression of HBV-associated HCC.

In the context of HBV or HCV infection, HCC development and progression are supported by epigenetic dynamics. This is associated with DNA methylation, histone modifications, microRNAs (miRNAs), and long noncoding RNAs (lncRNAs).

In liver tissue affected by chronic hepatitis or cirrhosis due to HBV or HCV infection, DNA methylation can be detected, which is considered a precancerous state of HCC (70). Significantly elevated HBx expression is associated with methylation abnormalities in HBV-infected HCC patients (38). DNMT1, responsible for maintaining DNA methylation during replication, and DNMT3A/3B, also known as de novo methyltransferases, are involved in methylation modifications (11). HBx upregulates DNMT1/3A1/3A2, resulting in the hypermethylation of tumor suppressor-related genes (38, 39), such as increasing CpG island methylation of the cytokine signaling-1 (SOCS-1) promoter (41) and hypermethylation of p14’s promoter (42). HBx also inhibits Tyrosine-protein phosphatase nonreceptor type 13 (PTPN13) transcription by upregulating DNMT3A and promoting DNA methylation, thereby facilitating the progression of HCC progression (40).

Epigenetic modification of histones contributes to multiple malignant tumor pathogeneses and metastases, including HCC (71–73). HBx inhibits H3 lysine 4-methyltransferase complex core subunit WDR5 ubiquitination, resulting in increased HBV transcription via H3K4me3 modification of cccDNA (43). A positive feedback loop is formed when the HBx-WDR5-H3K4me3 axis increases ALKBH5, a demethylase enzyme that catalyzes m6A demethylation of HBx mRNA (74). However, other reports indicate that H3K4me3 is generally diminished in HBV infections. HBx can alleviate the reduction in H3 acetylation and H3K4me3, as well as the methylation of histone 3 lysine 9 (H3K9me), following HBV infection. This action helps to mitigate the transcriptional silencing of cccDNA (45). HBx inhibits cell apoptosis by inducing DLL3 silencing via histone acetylation in HBV-associated HCC (46).

MicroRNAs (miRNAs) and long noncoding RNAs (lncRNAs) are two important classes of noncoding RNAs that substantially impact cellular processes, including differentiation, proliferation, and survival (75, 76).Notably, irregular expressions of these RNA types are characteristic of liver disorders, including HCC (77, 78). HBx is a key mediator in hepatocarcinogenesis, influencing the levels of both miRNAs and lncRNAs to promote or inhibit HCC development (79, 80). For example, HBx can induce the expression of miR-5188 through Wnt signaling, thereby establishing a feedback loop that promotes HCC stemness, metastasis, and chemoresistance (47). Additionally, it can modulate the expression of miR-21, miR-1269b, and miR-106b, thereby affecting tumor suppression and the progression of HCC (48–50, 81)^(p12). In contrast, downregulated miRNAs affected by HBx, such as miR-1270, miR-216b, miR-122, miR-18b, and miR-148a, are involved in HCC carcinogenesis via numerous pathways (51–55). HBx influences transcription, translation, and epigenetic regulation of lncRNAs, with implications for HBV replication, immune evasion, drug resistance, and the activation of numerous signaling pathways (80). Noteworthy examples include DLEU2, LINC01431, TRERNA1, LINC01010, lncRNA-Dreh, and lncIHS (56, 57) ^(p1), (58–61). Consequently, these interactions between HBx and noncoding RNAs emphasize their significance in the molecular mechanisms underlying the development and progression of HBV-associated HCC.

HBV DNA integration into the host genome induces genomic instability and direct insertion mutations in cancer-related genes in early-stage clonal HCC expansion (82). Variation in the HBx sequence, specifically at the 3’-terminus, substantially affects the HCC development. Chimeric transcripts of HBx and human genes, often including 3’-terminal deletions, are frequently expressed and can encode alternative HBx versions implicated in transcriptional regulation (83, 84).

The carboxyl terminus of HBx plays multiple roles in protein-protein interactions, transcriptional transactivation, DNA repair, cellular signaling, and HCC pathogenesis (85). COOH-terminal mutations are associated with HCC and may lead to reactive oxygen species (ROS) production, which damages mitochondrial DNA (86, 87). The clustering pattern of HBx 3’ end generates a truncated X protein (Ct-HBx) (62) that promotes HCC progression by mediating glucose metabolism reprogramming, increasing matrix metalloproteinase 10 (MMP10) transcription (62, 63), and enhancing Wnt/β-Catenin signaling (64). HBX-C30 (30 aa deletion from HBx C terminus) co-activates anticancer FXR signaling less effectively than full-length HBx (65). Additionally, mutations in the N-terminal domain, such as F30V, increase HBx’s antiapoptotic activity and diminish HBV’s replication efficacy. These HBx mutations facilitate the development of HCC and immune evasion (66).

Currently, there is a lack of drugs and therapeutic methods targeting HBx, which awaits further development by researchers. During HBV infection, the host SMC complex, SMC5/6, suppresses viral transcription. HBx promotes ubiquitination and degradation of SMC5/6, enhancing cccDNA transcription and increasing viral replication. This destabilization of SMC5/6 aids HBV in evading host immune surveillance and facilitates viral persistence and propagation (15, 88, 89). Nitazoxanide, an FDA-approved thiazolide antimicrobial for protozoal enteritis, and Pevonedistat, an NAE inhibitor approved for myelodysplastic syndromes, have demonstrated in vitro efficacy in inhibiting the HBx/SMC5/6 axis. This inhibition restores SMC5/6 protein levels, effectively suppressing HBV transcription and translation (90–92). Dicoumarol, the precursor to warfarin and a competitive NQO1 inhibitor, reduces HBx level and cccDNA transcription in both HBV-infected hepatocytes and humanized mouse models (93).

The TGF-β branch, classified as the TGF-β family, is activated by three ligands: TGF-β1-3, with TGF-β1 being the most abundant and typical isomer as it is secreted by almost all cells (94). The TGF-β ligand synthesis is in a longer precursor protein that is cleaved by furin protease; additionally, TGF-β disulfide-bonded dimers and the latency-associated peptide (LAP) are joined by non-covalent bonds to form a small latent complex, which is cross-linked with the latent TGF-β binding protein (LTBP) forming a large latent complex. This inactive complex is then secreted and associated with the extracellular matrix (ECM) (95). Activation occurs when particular signals initiate the release of active TGF-β from the complex, involving elements such as extreme pH, proteases, and integrins (96). Upon TGF-β ligand binding, TβRII dimerizes and recruits TβRI to produce a heterotetrameric TβRI-TβRII complex (97, 98).

TβRI phosphorylates R-Smads (Smad2/3) at their extreme C-terminal Ser-X-Ser motifs in the canonical Smad signaling, resulting in oligomerization with Co-Smad (Smad4) and nuclear translocation (99). The activated Smad4-R-SMAD complex regulates the transcription of target genes through interactions with DNA-binding transcription factors (20). The Smad linker can also be phosphorylated to activate numerous signaling pathways involved in pathological or pathophysiological gene expressions (100). Inhibitory Smad6/7 (I-Smads) modulate canonical Smad signaling in a feedback manner via various mechanisms (101, 102). Besides canonical Smad signaling, TGF-β functions by activating non-canonical Smad pathways, including the mitogen-activated protein kinase (MAPK), extracellular signal-regulated kinases1/2 (Erk1/2), Rho-like, phosphatidylinositol-3-kinase (PI3K)/AKT, c-Jun amino-terminal kinase (JNK), p38/MAPK (103), and the Src tyrosine kinase signaling pathways (104, 105) ( Figure 3 ).

TGF-β signaling is essential at every stage of liver disease progression, from initial inflammation and damage to fibrosis, cirrhosis, and ultimately liver cancer. Early on, it inhibits tumor growth by inducing senescence and apoptosis. It promotes tumor growth, EMT, and metastasis in advanced stages. Nevertheless, Smad-dependent and independent pathways are involved in the complex signaling (106). ( Figure 4 ).

The TGF-β pathway plays a crucial role in the progression of HBV-associated HCC. Many drugs targeting the TGF-β pathway have shown promising results in treating HCC (164). However, the clinical applicability and efficacy of these treatments require further validation through extensive clinical trials. Studies have shown that various chemotherapeutic agents, including Fluorofenidone (AKF-PD) (165), Sanguinarine (San) (166), Aspirin (167), Praziquantel (PZQ) (168), and Ursodeoxycholic acid (UDCA) (169), along with small molecule inhibitors such as Galunisertib (LY2157299) (170–172), LY2109761 (173), SKLB023 (174), and LY3200882 (175), are effective in treating HBV-HCC. The therapeutic vaccine Belagenpumatucel-L (Lucanix) (176) also exhibits efficacy in this setting. Studies show significant activation of the TGF-β pathway in immunotherapy-resistant tumors, with TGF-β often implicated in establishing suppressive tumor microenvironments (177, 178). The TGF-β and PD-1 pathways operate through independent, yet complementary, immunosuppressive mechanisms, enhancing cancer immune evasion (179, 180). Consequently, combining TGF-β inhibitors with PD-1 monoclonal antibodies presents a promising treatment approach for HBV- HCC.

As mentioned, the TGF-β pathway is involved in tumor suppression during early-stage tumorigenesis and tumor promotion in advanced cancers (94, 182–184). This dichotomous effect is determined by the phosphorylation status of the Smad3 protein, particularly its c-terminus or linker region (185). The linker region of Smad2/3 contains numerous conserved motifs subject to regulatory factors and post-translational modifications, such as phosphorylation (186, 187). The linker domain is phosphorylated at specific serine/threonine residues by cytoplasmic MAPKs and nuclear CDKs (20, 35–38). The phosphorylation of Smad2/3 in the linker region generates three types of phospho-isoforms: C-terminally phosphorylated Smad2/3 (pSmad2C/3C), linker-phosphorylated Smad2/3 (pSmad2L/3L), and dually phosphorylated Smad2/3 (pSmad2L/C and 3L/C) (188).

In normal epithelial homeostasis, TGF-β mediates pSmad3C signaling, inhibiting cell proliferation by interfering with cell cycle progression (189–191). This is accomplished via activating CDK inhibitors, such as p15INK4B and p21CIP1, and inhibiting c-Myc gene expression and cell cycle-related molecules (192–194). The pSmad3C pathway protects against cancer development, leading to transient Ras activation followed by growth inhibition and apoptosis. Furthermore, pSmad3C can regulate apoptosis-related protein expressions, including Bcl2 (195)^(p2).

Cytoplasmic Ras-related kinase activation during carcinogenesis, including MAPKs, transforms Smad3 signaling from an antitumor pSmad3C state to the oncogenic pSmad3L and pSmad2L/C pathways. JNK, a serine/threonine kinase activated by Ras, plays a crucial role in this conversion by defeating TβRI/pSmad3C-mediated growth arrest (196, 197). Smad3 is phosphorylated at Ser-213 upon activation of JNK (198), a site where RTK pro-inflammatory cytokines, growth factors, and, to a lesser extent, TGF-β can increase phosphorylation levels. It has been demonstrated that c-Myc overexpression can inhibit the Smad3-dependent transcription of p15INK4B and p21WAF1 proteins, thereby opposing cell cycle arrest (199). The JNK/c-Myc mitotic pathway inhibits the TRI/pSmad3C/P21WAF1-mediated growth arrest (200). Ser-213 phosphorylation of Smad3L via TβRI perturbs COOH-tail phosphorylation (26, 198, 201), promoting nuclear translocation and accelerating cell proliferation signals mediated by pSmadL (198). JNK-activated pSmad3L-mediated cell proliferation signal and TβRI-activated pSmad3C-mediated cell cycle arrest signal are mutually antagonistic. Mutations in essential pathway components can cause persistent Smad3 linker phosphorylation, so highly phosphorylated Smad3L likely reduces pSmad3C’s sensitivity to growth inhibition in tumor cells (26, 202–204). The overexpression of receptors by cancer cells modifies Smad3 phosphorylation (205). The Ras/JNK pathway controls both pSmad3C and pSmad3L. Strong Smad2L/C and Smad3L/C phosphorylation is observed in colorectal cancer EMT-related tumors (206). CDK4 converts TGF-β signal-mediated pSmad2/3C to malignant pSmad2L/C and 3L/C pathways (207). The interaction between pSmad2L/C and pSmad3L induces fibrogenic signals and liver fibrosis via PAI-1 (208). Increased PAI-1 transcription and ECM synthesis positively modulate liver fibrosis in hepatocytes (209).

It has been demonstrated that HBx overexpression induces the development of hepatic tumors by stimulating DNA synthesis (210). Studies have revealed that HBx shifts TGF-β signaling from the TβRI-dependent pSmad3C tumor-suppressive pathway to the JNK-dependent pSmad3L oncogenic pathway during carcinogenesis, as observed in biopsy samples from chronically HBV-infected patients and HBx transgenic mice with liver lesions (26). The proto-oncogene c-Myc, a target of TGF-β/SMAD signaling (211, 212), is involved in HCC malignant progression (213). pSmad3L/Smad4 triggers c-Myc transcription, whereas pSmad3C/Smad4 inhibits it. The two complexes antagonize each other and govern c-Myc expression. HBx’s presence may cause the signal to upregulate c-Myc and promote cancer cell growth (214) ( Figure 5 ).

In conclusion, the interaction between HBx and TGF-β plays an essential role in hepatocarcinogenesis at various stages of HBV infection. TGF- β signaling has a dual function, promoting tumor suppression in the early stages of tumorigenesis while promoting tumor growth in advanced malignancies. Smad3’s function is determined by its phosphorylation status, with C-terminally phosphorylated Smad3 (pSmad3C) functioning as a tumor suppressor and linker-phosphorylated Smad3 (pSmad3L) contributing to oncogenesis. HBx modifies the TGF-β signaling pathway, redirecting it to the oncogenic pSmad3L pathway. This perturbation promotes the progression of the cell cycle, inhibits apoptosis, and accelerates the development of hepatocellular carcinoma.

In prior studies, researchers revealed a correlation between HBx and TGF-β expression in HBV-infected cells. TGF-β1 level correlates positively with HBx protein expression in early-stage HBV infection, suggesting that HBx may directly or indirectly promote TGF-β1 expression. HBx forms a complex with Egr-1 protein and transactivates the TGF-β1 promoter via the Egr-1 binding site (25). HBx overexpression during HBV infection correlates with the release of secreted factors, particularly TGF-β, from HBx-transfected HCC cells or adjacent endothelial cells through increased CD133 expression, which induces invasion of these cells by EMT (210).

TGF-β1 inhibited cell proliferation and invasion in a study involving trophoblast cells (HTR-8/SVneo). Nevertheless, HBx activated the Smad pathway in HBx-transfected cells, resulting in the downregulation of E-cadherin and the upregulation of vimentin and N-cadherin, which reduced the apoptotic capacity and increased the invasive capacity of HTR-8/SVneo cells. The mechanism by which HBx shifts TGF-β signaling in HTR-8/SVneo cells needs additional investigation (211). Additionally, HBx inhibits the expression of the E-cadherin gene (CDH1) by activating TGF-β, which may be an additional mechanism for its downregulation of E-cadherin and promotion of tumor metastasis (212). The HTR-8/SVneo cell line is derived from cells that were grown from early human placental chorionic villi explants and transfected with a gene encoding the Simian Virus 40 large T antigen. Since it is not sourced from a liver cancer cell line, further foundational experiments are necessary to determine whether it can represent the general mechanisms within HBV-infected and liver cancer patients. This includes using human-derived liver cancer cell lines and replicating the studies in mouse models that simulate liver cancer to validate the results.

Epigenetic and genetic events affect the interaction between HBx and TGF-β. HBx can increase TGF-β expression via autophagy induction, and increased TGF-β upregulates lncRNA-ATB, thereby enhancing liver cancer cell migration and invasion (215). Both HBx and TGF-β1 stimulation induces significant overexpression of miR-199a-3p in hepatic progenitor cells (HPCs), thereby promoting HPCs oncogenic transformation via a JNK/c-Jun/miR-199a-3p-dependent pathway (216). C-terminal truncated mutants (ctHBx) commonly found in HCC tissue samples (217) decrease bone activin membrane-bound inhibitor (BAMBI) more than HBx alone. ctHBx significantly inhibits BAMBI promoter activity in the absence of the Wnt/β-catenin pathway, thereby reducing the inhibition of TGF-β1 and β-catenin and promoting malignancy (218).

In an HBx transgenic mouse model undergoing partial hepatectomy, TGF-β, Smad2, and phosphorylated Smad3/4 (ser423/425) were significantly overexpressed in the HBx transgenic mice liver compared to non-transgenic mice, indicating the impact of HBx on the TGF-β/Smad pathway, which promotes the progression of HCC (213). Furthermore, HBx disrupts the negative feedback loop between TGF-β and protein phosphatase magnesium-dependent 1A (PPM1A) by enhancing PPM1A ubiquitination and degradation, resulting in TGF-β pathway overactivation, HCC migration, and invasion (219). HBx also stabilizes the binding of the Smad complex to the transcriptional machinery and facilitates the nuclear transposition of Smads, thereby amplifying TGF-β signaling (220).

Contrary to previous findings, some studies suggest a negative correlation between HBx and TGF-β activation. For example, HBx inhibits TGF-β-induced apoptosis by linking Src to PI3K, thereby activating the PI3K/Akt signaling pathway (221, 222). And another study showed that cells transfected with HBx exhibited reduced expression of the TGF-β type II receptor, resulting in a weaker TGF-β1 response and reduced growth inhibition in response to TGF-β1 (214). Furthermore, research also shows that HBx plays different roles at hepatocyte cell line. In normal liver cells, HBx induces cell cycle arrest by elevating TGF-β and p27 levels, leading to G1 or G2 phase blocks that facilitate HBV replication. In HBV-infected HCC cells, however, HBx may accelerate carcinogenesis by promoting cell cycle progression through the downregulation of TGF-β and the p53/27/21 pathway (223). Despite these findings, the overall trend indicating tumor progression via the HBx/TGF-β axis aligns with earlier results. This may be attributed to the paradoxical roles of TGF-β at different stages of tumor progression or variations in experimental conditions.

It is evident that the TGF-β signaling pathway is primarily influenced and regulated by HBx, particularly within HBV-infected cells. It is worth noting that HBx has been shown to lead to malignant transformation of HCC by influencing dysregulation of TGFB and thereby activating multiple cancer-promoting mechanisms. including EMT, anti-apoptosis, proliferation, inflammatory responses, metastasis, invasion, and fibrosis. Therefore, researchers need to pay more attention to the interaction mechanism of these HBx-TGF-β axis, and develop anti-tumor drugs that can target the common pathway of this axis, bringing new hope to the drug treatment of hepatocellular carcinoma. ( Figure 6 ).