Defining the relationship between farsenoid X receptor, hepatitis B virus X protein and HCC: It’s complicated

Abstract

The relationship between hepatitis B virus X protein (HBx), farsenoid X receptor (FXR) and hepatocellular carcinoma (HCC) is a complicated one in that we have a viral protein interaction that can drive tumorigenesis or inhibit HCC depending upon transactivation of full-length or truncated HBx. In the current article the authors have elegantly described a system of HBx-FXR interaction that demonstrates inhibition of HCC tumor growth via activation of full-length HBx. The paper employs both in vivo and in vitro studies including using FXR knockout mice crossed with HBx induced mice. Overall, studies on the interaction between HBx and FXR have been riddled with complication and this paper sheds important light on the relationship that may be key in developing much needed therapies for HCC.

Hepatocellular carcinoma, Hepatitis B virus and HBx protein

Hepatocelluar carcinoma (HCC) is the second cause of cancer-related death worldwide and is associated (~50% of cases) with hepatitis B virus (HBV) infection (1, 2). The high rate of metastasis or recurrence, coupled with low chemotherapy response, contributes to the dismal prognosis in HCC patients (3). Hepatocyte transformation is induced by inflammation, DNA damage, senescence, chromosomal instability and epigenetic alterations (2). Patients with HBV infection have a greater chance of developing HCC, and reports estimate that the risk is increased 25-fold over patients that are not carrying HBV (2).

HBV-X, or HBx, protein has been implicated in the progression of HCC and regulates transcriptional gene expression modulating oncogenesis, proliferation, inflammation and immune responses (4). Patients with C-terminal truncated HBx, in contrast to full-length, have enhanced tumor metastasis and cell invasion (3). Full-length HBx has also been implicated in HCC but with contrasting findings. In one study, ATX transgenic mice, where human α1-antitrypsin gene promoter controls HBx expression, do not develop spontaneous HCC (5). However, HBx transgenic mice, where HBx is under control of its own promoter elements, have been shown to develop spontaneous HCC, indicating HBx as a promoter of HCC (6). The contrasting role of HBx during HCC progression has been controversial among researchers, and understanding the underlying mechanism whereby HBx either promotes or protects against HCC development is key for unlocking its therapeutic potential.

Farsenoid X receptor

FXR is a member of the nuclear receptor superfamily and is an essential component for liver and intestinal functions. Bile acids are the primary endogenous ligands of FXR and include chenodeoxycholic acid, lithocholic acid, deoxycholic acid and cholic acid (7). The role of FXR has been mostly defined as hepatoprotective since knockout of FXR in mice results in cholestasis, nonalcoholic steatohepatitis, liver tumors, and colon tumors (4).

Studies on the relationship between FXR and HBx are limited, and most discuss their separate roles during HCC progression. During HCC, 80–90% of patients have HBV integration and HBx genes are often partially lost during integration, resulting in C-terminal truncation of HBx (8). A recent study of in vitro HCC cell lines raised from patients with HBV found that the C-terminal truncated mutants of HBx have increased capacity to confer the expression of “stemness” genes through deregulation of FXR/RXR signaling (8). Further, overexpression of HBx truncated mutants increased hepatosphere formation and migration, which was rescued through use of Z-guggulsterone treatment (8). However, treatment with RXR inhibitor did not rescue these phenotypes indicating an FXR-specific event. These unique studies show that truncated HBx deregulates FXR signaling during cancer development and increases stemness, hepatosphere formation and migration.

Summary of Current Findings

The finding that HBx is a modulator of HBV-associated HCC has been demonstrated in numerous studies; however, the concept that HBx-induced FXR activation decreases tumor incidence is innovative. The major finding from this paper demonstrates that full-length HBx behaves as a transactivator of FXR that, when activated, decreases HCC development. The authors elegantly demonstrate that full-length HBx enhances FXR binding to its response elements, thereby increasing FXR transcriptional activity and target gene expression. Considering FXR has been indicated as a regulator of anti-inflammation, these findings support the argument that full-length HBx acts as a protective agent during HCC progression through FXR transactivation. To further define HBx-FXR interaction the authors thoroughly describe the shared binding sites between full-length HBx and FXR. While full-length HBx seems to be protective against hepatocarcinogenesis via FXR transactivation, the authors found that C-terminal truncated HBx did not provide any protective benefits and was unable to bind to or transactivate FXR. In support of these data, Niu et al. found that there was a significant loss of full-length HBx, but increased positive staining for C-terminal truncated HBx, in tumors from patients with HBV-positive HCC, suggesting that the truncated mutant of HBx may be tumorigenic, whereas the full length may play a protective role. This is further supported by previous work demonstrating that truncated HBx is critical to hepatocarcinogenesis (8).

When clarifying the protective benefits associated with full-length HBx-FXR interaction, the authors employed the use of the HBx-sensitized ATX mice, FXR^−/− mice and ATX-FXR^−/− mice. The authors found that ATX mice did not develop spontaneous HCC tumors; however, almost half of the ATX-FXR^−/− mice developed tumors by 15 months. Unlike previous studies that have demonstrated that FXR^−/− mice develop tumors, the current study found the opposite and explained that this discrepancy may be due to different backgrounds of mice when compared to previous reports (7). In the absence of FXR, ATX mice had increased inflammatory cytokine expression indicating that activation of HBx/FXR signaling may behave as a protective mechanism inhibiting inflammation and tumorigenesis. The authors further demonstrate that HBx may regulate other players involved in FXR signaling, including SHP. The expression of SHP was absent in ATX-FXR^−/− mice, confirming that loss of SHP may be another mechanism promoting HCC formation in ATX-FXR^−/− mice since SHP^−/− mice develop spontaneous HCC whereas SHP overexpression partially protects FXR^−/− mice from HCC formation (9, 10).

For every one female with HBV-associated HCC there are three afflicted males, suggesting underlying gender effects that protects females HCC susceptibility (2). Surprisingly, the authors herein found that 100% of female ATX-FXR^−/− mice had increased tumor burden compared to 44% of the male TX-FXR^−/− mice. Interestingly, it has also been noted previously that both male and female FXR^−/− mice develop an equal amount of tumors suggesting that total knockout of FXR interrupts estrogen-protected pathways (7). In addition, the authors found that females had significantly higher levels of inflammatory cytokines compared to male ATX-FXR^−/− mice. Since the authors found that 100% of their ATX-FXR^−/− females developed tumors and it has been demonstrated that total FXR knockout induces tumors in both male and female mice (7), the current study leaves one wondering what might happen if the authors used ATX mice combined with FXR liver specific knockout mice, thus eliminating the detrimental effects of FXR deficiency on estrogen production.

Conclusions

The study from Niu, et al. nicely demonstrates the relationship between either full-length or C-terminal truncated HBx and FXR during HBV-induced HCC. This study provides compelling evidence for the hepatoprotective nature of full-length HBx via FXR transactivation and further describes the direct binding sites between these two proteins, providing potential therapeutic targets for HCC. The complex relationship between full-length versus truncated HBx and FXR is a novel and interesting finding of the paper, which may help clear the confusion on the role of HBx during hepatocarcinogenesis. Therapies that combine chemotherapy and drugs that target molecules like HBx or FXR may offer promise for advanced treatment strategies, specifically during HBV-induced HCC. It is obvious that the intricacies of the co-activation of FXR by HBx (truncated or full-length) should be further examined to shed more light on the potential therapeutic benefits of this viral protein-nuclear receptor interaction. In addition, further questions should be answered regarding the role of HBx-activated FXR, including evaluation of other components that might alter tumor incidence. For example, what might happen with HBx-FXR co-activation where liver functions are altered, such as cholestatic liver injury, bile acid feeding or fatty liver disease? While this paper begins to define the differential properties of full-length versus truncated HBx during HCC, the exact relationship between these proteins and FXR-mediated events remains complicated.

Abbreviations

FXR

Farsenoid X receptor

HBx

Hepatitis B virus X protein

HCC

Hepatocelluar carcinoma