MiRNAs control hepatocarcinogenesis by regulating hepatocyte nuclear factor 4α-inflammatory signal feedback loops

Hepatocellular carcinoma (HCC) is the most common type of liver cancer and the third leading cause of cancer mortality, with a prevalence of approximately 600,000 cases worldwide.¹ Over the last 20 years, HCC incidence has risen steadily in both western countries, such as the United States, and developing countries, including China. HCC is commonly associated with chronic hepatitis B and C virus infections, alcohol abuse, and obesity. In spite of the etiologies of this disease, most HCCs are associated with predisposing liver disease with unresolved inflammation.² This chronic inflammation status alters the hepatic microenvironment, favoring pro-oncogenic or (epi)genetic changes, driving hepatocarcinogenesis. Over the last twenty years, many studies have suggested that chronic liver inflammation is tightly controlled by the interaction of a variety of inflammatory mediators and their downstream signaling pathways. In this context, both NF-κB and STAT3 have been identified as two key signaling molecules that link inflammation and liver cancer.^{3, 4}

In contrast to the constitutive activation of chronic inflammatory signals that usually accelerate hepatocarcinogenesis, the expression of hepatocyte nuclear factor 4α (HNF4α), a master regulator of hepatocyte differentiation, is suppressed in liver cancer. HNF4α is primarily expressed in hepatocytes and indispensable for hepatic epithelium formation during embryonic development. HNF4α is widely associated with the transcriptional regulation of hepatocyte genes specifically implicated in lipid metabolism, glucose metabolism, differentiation, and morphogenesis.⁵ Apart from its function in hepatocyte differentiation, HNF4α can also inhibit hepatocyte proliferation and hepatocarcinogenesis by suppressing various pro-mitogenic genes.^6–8 Loss of HNF4α promotes chemically induced liver cancer in rodents and is associated with human HCC.^{7, 8}

Although it is well-documented that liver cancer is associated with constitutive inflammation and loss of HNF4α, the relationship between these two events has only been recently investigated by Hatziapostolou et al.⁹ Their findings suggest that inflammatory mediators such as IL-6 inhibit HNF4α, while HNF4α suppresses IL-6-STAT3 activation and liver inflammation by attenuating IL-6 receptor expression, thereby acting as a feedback circuit to regulate hepatocarinogenesis.⁹ This is the first study reporting that the HNF4α-STAT3 axis governs hepatic transformation and oncogenesis, mechanistically linking inflammation and liver cancer. Moreover, this study further demonstrated that several miRNAs (such as miR-124, miR-24, and miR-629) tightly regulate this HNF4α-STAT3 feedback circuit. Particularly, loss of HNF4α results in IL-6 receptor-STAT3 activation via the inhibition of miR-124, while activation of STAT3 upregulates the hepatic expression of miR-629 and miR-24, which subsequently maintain the suppression of HNF4α and sustain oncogenesis.⁹

In this issue of Hepatology, Ning et al. identified another important feedback loop in which the loss of HNF4α leads to NF-κB activation; this activation then further inhibits HNF4α.¹⁰ This feedback circuit maintains the sustained suppression of HNF4α and induces activation of NF-κB, thereby promoting oncogenesis. Interestingly, the authors also demonstrated that several miRNAs play important roles in maintaining this feedback loop. Specifically, HNF4α upregulates the expression of miR-124 and miR-7, which then attenuate NF-κB activation. Conversely, NF-κB activation induces the expression of miR-21, which subsequently downregulates HNF4α expression.

MiRNAs, a group of small non-coding RNA molecules, play a role in the transcriptional and post-transcriptional regulation of a variety of genes that regulate cell differentiation, proliferation, survival, and transformation. In the liver, various miRNAs have been implicated in the pathogenesis of alcoholic liver injury, nonalcoholic fatty liver disease, viral hepatitis, fibrosis, and liver cancer.¹¹ The role of miRNAs in hepatocarcinogenesis has been actively investigated over the last decade. Emerging evidence suggests that the function of miRNAs in hepatic carcinogenesis is very complex. Different miRNA profiles in HCC have been reported from various studies, which is in part due to the different types of experimental approaches used to investigate miRNA expression in biological samples^{11, 12}. A large number of miRNAs have been shown to regulate hepatocarcinogenesis by affecting the expression of a wide variety of target genes that contribute to liver cell differentiation, growth, survival and apoptosis.¹³ Hatziapostolou et al. and Ning et al. identified a novel mechanism by which miRNAs regulate liver oncogenesis, which is mediated by maintaining the HNF4α-inflammatory signal feedback loops (Fig. 1). This mechanism likely plays an important role in sustaining chronic liver inflammation during hepatocarcinogenesis, thereby promoting liver cancer progression.

Figure 1. A model depicting the regulation of HNF4α-inflammatory signal (STAT3 and NF-κB) feedback loops and hepatocarcinogenesis by miRNAs.

HNF4α upregulates expression of miR-124 and miR-7, which then downregulates expression of RelA and subsequently inhibits NF-κB activation. Conversely, NF-κB activation induces the expression of miR-21, which consequently downregulates HNF4α expression. Moreover, HNF4α-associated upregulation of miR-124 also inhibits IL-6 receptor expression and subsequently attenuates STAT3 activation. On the other hand, STAT3 activation upregulates miR-24 and miR-629, which then attenuates HNF4α expression. Factors written in red mean those promote hepatocarcinogenesis while factors written in blue mean those inhibit hepatocarcinogenesis.

Although the studies by Hatziapostolou et al. and Ning et al. nicely confirmed the regulation of the HNF4α-STAT3 and HNF4α-NF-κB feedback loops by various miRNAs in cultured liver cancer cells in vitro, the interactions among miRNAs, HNF4α, and inflammatory signals in the control of hepatocarcinogenesis in vivo are likely much more complex. For example, Hatziapostolou et al. demonstrated that HNF4α inhibits STAT3 via the induction of miR-124 and, subsequently, by the inhibition of the IL-6 receptor. However, in addition to IL-6, many other cytokines can also activate STAT3 in hepatocytes and liver cancer cells,¹⁴ including various IL-6 family cytokines (such as leukemia inhibitory factor, ciliary neurotrophic factor, oncostatin M, cardiotrophin-1, and IL-11) and IL-22.¹⁴ Additionally, hepatocyte growth factor, epidermal growth factor, and hepatitis viral proteins can also activate STAT3 in hepatocytes, but to a lesser extent.¹⁴ It is not clear whether inactivation of HNF4α also enhances the STAT3 activation induced by these cytokines and mediators. In the HNF4α-NF-κB feedback loop, HNF4α indirectly inhibits NF-κB activity via the upregulation of miR-7 and miR-124 and subsequent suppression of RelA (a subunit of NF-κB) expression.¹⁰ It is known that hepatic NF-κB can be activated by many inflammatory cytokines (e.g., IL-1 and TNF-α), TLR ligands (e.g., LPS), and hepatitis B X protein.¹⁵ Therefore, it is likely that these mediators and the loss of HNF-4α additively or synergistically promote NF-κB activation in hepatocytes and liver cancer cells. Although Ning et al. concluded that miRNAs regulate the HNF4α-NF-κB feedback loop in hepatocytes and that NF-κB promotes HCC metastasis, the actual role of NF-κB in hepatocarcinogenesis is much more complex.¹⁵ First, NF-κB may exert different effects in different cellular compartments, such as hepatocytes, nonparenchymal cells, and immune cells. Therefore, NF-κB likely affects liver cancer development and progression via cell-specific mechanisms.¹⁵ It is not clear whether the miRNAs that regulate NF-κB in hepatocytes also affect NF-κB expression in nonparenchymal cells and immune cells or whether they subsequently modulate liver inflammation and carcinogenesis. Second, in addition to the upregulation of miR-124 and miR-7 by HNF4α, many other miRNAs were also upregulated after overexpression of HNF4α in HCC cells, of which several suppress HCC cell proliferation and contribute to the anti-tumor effect of HNF4α.¹⁶ Finally, NF-κB plays a wide variety of roles in regulating liver injury, inflammation, fibrosis, and regeneration, thereby indirectly affecting hepatocarcinogenesis.¹⁵

In summary, the study by Ning et al. described a novel mechanism by which miRNAs affect liver carcinogenesis via the regulation of the HNF4α-NF-κB feedback loop and is of clinical relevance for the prognosis of HCC patients. The authors showed that HCC patients with HNF4α^highRelA^low levels exhibit better overall survival and disease-free survival rates than those with HNF4α^low or HNF4α^highRelA^high levels, thereby suggesting that assessment of both HNF4α and RelA levels yields better prognostic accuracy for HCC patients than HFN4α alone. Additionally, several miRNA signatures have been identified that are useful for the diagnosis and prognosis of HCC patients.¹³ It would be interesting to determine whether the combined analysis of HNF4α, RelA, and miRNAs exhibits greater prognostic accuracy than assessment of only HNF4α and RelA. Finally, modulation of the HNF4α-inflammatory signal feedback loop by targeting miRNAs may represent a novel strategy to treat and prevent HCC.