Finding a New Role for NEMO: A Key Player in Preventing Hepatocyte Apoptosis and Liver Tumorigenesis by Inhibiting RIPK1

NEMO/IKKγ is a regulatory subunit of IKK complex, upstream of NF-κB, which participates in regulating immune and inflammatory responses, cell death, survival, and carcinogenesis. Previous reports demonstrated that specific deletion of NEMO in liver parenchymal cells results in spontaneous liver injury and carcinogenesis. It has been suggested that abolished NF-κB activity by NEMO deficiency is associated with this liver phenotype. In the recent article entitled “NEMO Prevents Steatohepatitis and Hepatocellular Carcinoma by Inhibiting RIPK1 Kinase Activity-Mediated Hepatocyte Apoptosis” published in Cancer cell by Kondylis et al., the investigators provided new insight into the in vivo role of NEMO in NF-κB-independent RIPK1-mediated hepatocyte apoptosis and liver tumorigenesis (1).

The in vivo role of NEMO has been well-characterized in the liver. Several studies have shown that mice with ablation of NEMO, specifically in liver parenchymal cells (NEMO^LPC-KO mice), are very sensitive to LPS-induced hepatotoxicity and spontaneously develop liver inflammation, steatosis, fibrosis, and hepatocellular carcinoma (2,3). Since previous in vitro studies strongly suggest that NEMO could regulate NF-κB-independent pathway (4), Kondylis et al. first investigated the in vivo role of NF-κB in liver pathophysiology by generating mice with combined deletion of the three NF-κB subunits (RelA, RelB, c-Rel) capable to activate gene transcription, specifically in liver parenchymal cells (NF-κB ^LPC-KO mice). While severe liver damage is observed in NEMO^LPC-KO mice, NF-κB ^LPC-KO mice had much less liver damage than NEMO^LPC-KO mice. Furthermore, after 1 year, tumorigenesis was observed in the livers of NEMO ^LPC-KO mice, but not in NF-κB ^LPC-KO mice. Thus, these results suggest that NEMO prevents hepatocyte apoptosis and tumorigenesis via mechanisms independent of the canonical NF-κB pathway.

To further clarify the role of NF-κB in NEMO^LPC-KO mice, the investigators generated NEMO^LPC-KO mice constitutively overexpressing IKK2 (IKK2ca^LPC), which induced NF-κB activation, even in the absence of NEMO (Figure.1). These mice had less serum ALT levels and hepatocyte apoptosis than NEMO^LPC-KO mice. Moreover, no visible liver tumors were observed even in one-year-old NEMO^LPC-KO /IKK2ca^LPC mice. They also found that additional deletion of RelA in liver parenchymal cells of these mice resulted in loss of the rescue effect induced by the overexpression of IKK2. These results indicate that IKK2 overexpression protects NEMO deficient-hepatocytes from spontaneous liver injury and carcinogenesis via RelA.

Figure 1. NEMO prevents hepatocyte apoptosis and liver tumorigenesis by inhibiting the formation of the complex of RIPK1/FADD/Caspase-8.

In hepatocytes, NEMO physiologically regulates NF-κB activation and also prevents the formation of the complex IIb (RIPK1, FADD, Caspase-8 [Casp8]) to inhibit hepatocyte apoptosis. NF-κB pathway also controls TRADD-mediated apoptosis. When NEMO is absent, RIPK1 interacts with FADD and Caspase-8 to form Complex IIb, leading to hepatocyte apoptosis and compensatory hepatocyte proliferation, resulting in hepatocarcinogenesis. The formation of Complex IIb requires RIPK1 kinase activity. When both NEMO and whole RIPK1 protein are deleted, Complex IIa is formed (TRADD, FADD, Casp8), thereby inducing hepatocyte apoptosis and liver tumorigenesis, suggesting another kinase-independent scaffolding function of RIPK1. In addition, overexpression of IKK2 (Ikk2ca) can inhibit RIPK1 kinase-mediated hepatocyte apoptosis in NEMO^LPC-KO mice through induction of NF-κB-dependent survival factors. TRADD: TNF receptor-associated death domain, RIPK1: Receptor-interacting protein kinase 1. FADD: Fas-associated protein with death domain.

Next, to investigate how RIPK1 contributes to hepatocyte death and tumorigenesis in NEMO^LPC-KO mice, they crossed NEMO^LPC-KO mice with RIPK1D138N knock-in mice which express a kinase inactive mutant form of RIPK1. In the absence of RIPK1 kinase activity, NEMO^LPC-KO mice had less hepatocyte apoptosis and almost no tumor formation. In contrast, mice with a double knock-out of whole RIPK1 protein and NEMO in the liver exhibited severe liver damage and proliferation. These results suggest that both kinase-dependent and -independent functions of RIPK1 are involved in hepatocyte death in NEMO^LPC-KO mice as two distinct mechanisms. Since inhibition of RIPK1 kinase activity prevented hepatocyte apoptosis, RIPK1 kinase activity is required for hepatocyte apoptosis in NEMO^LPC-KO mice. Interestingly, when RIPK1 protein is ablated, loss of its scaffolding function activates an alternative RIPK1 kinase-independent apoptotic pathway in NEMO^LPC-KO mice (Figure.1).

The investigators also determined that the formation of RIPK1/FADD/Caspase-8 complex (Complex IIb) was increased in NEMO-null hepatocytes and inactivation of RIPK1 kinase activity inhibited the complex formation and reduced apoptosis. Although the role of FADD was not discussed in this article, an earlier study reported that a double knock-out of NEMO and FADD in the liver attenuated inflammation and apoptosis and suppressed HCC development compared to deletion of NEMO alone (2,3). This supports the idea that RIPK1/FADD/Caspase-8 complex assembly is a crucial step for hepatocyte apoptosis in NEMO-null livers (2). Taken together, this study concluded that in the presence of intact RIPK1, NEMO protects hepatocytes from apoptosis by interfering with the formation of RIPK1/FADD/Caspase-8 complex. However, when NEMO is abolished and RIPK1 retains its kinase activity, RIPK1/FADD/Caspase-8 complex forms and mediates apoptosis (Figure.1).

Since TRADD/FADD/Caspase-8 complex had previously been shown to induce RIPK1 kinase-independent apoptosis (5), the authors examined the role of TRADD in RIPK1^LPC-KO /NEMO^LPC-KO mice. Surprisingly, additional deletion of TRADD dramatically attenuated hepatocyte apoptosis and showed no liver tumor formation in RIPK1^LPC-KO /NEMO^LPC-KO mice. These results indicate that in the absence of both NEMO and RIPK1, TRADD-mediated apoptosis is alternatively induced. Notably, only when both TRADD and RIPK1 are abolished, NEMO deficiency-mediated hepatocyte apoptosis and tumorigenesis is prevented. A previous in vivo study has reported that a kinase-independent scaffolding function of RIPK1 is essential for inhibiting epithelial cell apoptosis and necroptosis (6). When NF-κB is inactivated, TRADD/FADD/caspase-8 complex (complex IIa) forms and induces apoptosis, which has been shown to be RIPK1-independent (5); but based on this article, we know RIPK1 can also interfere in this pathway through the RIPK1’s scaffolding function (Figure.1).

Additionally, the investigators generated TNFR1^−/−/NEMO^LPC-KO mice to address the role of TNFR1 in the absence of NEMO and found that TNFR1 ablation could not rescue hepatocyte death and tumorigenesis. These results support their previous finding (3) in which death receptor-independent FADD signaling drives hepatitis and hepatocellular carcinoma in NEMO^LPC-KO mice. Since it’s been previously shown that TRAIL, CD40, other death receptors, Toll-like receptors, and T cell receptors can activate the RIPK1-mediated cell death pathway, multiple receptor signaling may also contribute to RIPK1-mediated cell death in the absence of NEMO (5). Therefore, deletion of single death receptor pathway might not be enough to rescue the liver phenotype in NEMO^LPC-KO mice.

In summary, Kondylis et al. provided new mechanistic insight into the role of NEMO, NF-κB, and RIPK1 in hepatocyte apoptosis and liver tumorigenesis. NEMO demonstrates a protective function in hepatocytes through both NF-κB-dependent and -independent mechanisms. In addition, RIPK1 is essential for NEMO deficiency-mediated liver pathophysiology and both RIPK1 kinase activity and its scaffolding function participate in this mechanism. This article also discussed the downregulation of NEMO as a predictor for prognosis and survival of HCC patients (1,7). Thus, NEMO and RIPK1 are not only potential targets for future therapies for chronic liver injury, fibrosis, and HCC, but may also be biomarkers for these liver diseases.