Small heterodimer partner (SHP) links hepatitis C and liver fibrosis: a small protein on the big stage

The molecular pathogenesis of liver fibrosis is complex. Chronic injury to hepatocytes, e.g. by viral hepatitis or metabolic stress, activates intracellular signalling pathways including activation of several kinases, translocation of nuclear factor (NF)-kappa B and secretion of transforming growth factor (TGF)-β, subsequently sending profibrogenic signals to hepatic stellate cells and other non-parenchymal cells in the liver 1. In order to develop antifibrotic therapies, it is of utmost importance to unravel how these pathways are interconnected and how stress signals are integrated into profibrogenic signalling networks. One class of small molecules that might serve as master regulators within these pathways are micro-RNAs 2. In this issue of Liver International, experimental and translational data from Jung et al. indicate a crucial role for a yet overlooked small nuclear receptor termed Small heterodimer partner (SHP).

The small heterodimer partner (SHP, OMIM: 604630) also known as nuclear receptor subfamily 0, group B, member 2 (NR0B2) is an orphan receptor that was originally identified in a yeast 2-hybrid screen with a close mouse relative of the human nuclear hormone receptor MB67 as the bait 3. It belongs to the nuclear receptor family of intracellular transcription factors and modulates transcriptional activity of several genes without having a DNA binding domain suggesting that it functions as a negative regulator of receptor-dependent signalling pathways 3, 4. In humans, SHP is predominantly expressed in liver and, at lower levels, in heart and pancreas 3. Mice lacking SHP exhibit mild defects in bile acid homoeostasis and fail to repress cholesterol 7-α-hydroxylase expression in response to a specific agonist for the bile acid receptor FXR 5. Later studies in these mice showed that the loss of SHP accelerated renal fibrosis in a model of obstructive nephropathy while the overexpression of this orphan receptor inhibited the expression of unilateral ureteral obstruction (UUO)-induced plasminogen activator inhibitor-1 (PAI-1), type I collagen, fibronectin and α-SMA 6. Concerning liver it was already demonstrated that HNF4α and Egr-1 expression are dramatically increased in SHP null mice subjected to bile duct ligation-induced fibrosis and that in human cirrhotic livers SHP expression is significantly reduced 7.

In this issue of Liver International, Jung et al. investigated the biological effects of SHP during the pathogenesis of hepatitis C virus (HCV)-induced hepatic fibrogenesis 8. The authors analysed HCV positive liver samples from 23 patients in comparison to 7 HCV negative liver specimens from patients suffering from cholangiocarcinoma. This analysis revealed that the quantities of SHP mRNA and protein levels were significantly down-regulated in the HCV-infected livers. In addition, the expression of collagen, TGF-β and PAI-1 was elevated, the phosphorylation of Smad3 increased, while the quantities of phosphorylated AMPK (p-AMPK) was lowered in respective samples (Fig. 1). These same differences in profibrogenic gene expression arose in vitro when the highly permissive hepatoma cell line Huh-7.5 was infected with HCV genotype 2. However, the induction of collagen type I and PAI-1 expression and the diminished formation of activated AMPK that was observed upon HCV infection was repealed when SHP was overexpressed by an adenoviral vector. Interestingly, the incubation either with the AMP analogue 5-Aminoimidazole-4-carboxamide ribonucleotide (AICAR) or the biguanide metformin that both activate the AMPK 9, 10 completely restored the HCV-induced suppression of SHP and reduced HCV-induced profibrogenic genes. The authors further found that the transient overexpression of SHP or the incubation with AICAR or metformin diminished the HCV-associated activation of NF-κB as well as the production of TGF-β and connective tissue growth factor (CTGF/CCN2), again demonstrating the antifibrotic potential of SHP. Moreover, the overexpression of SHP and activation of AMPK ameliorated HSC invasion in a co-culture system with the genetically well-characterized immortalized human hepatic stellate cell line LX-2 11.

All these findings indicate that the loss of hepatic SHP in concert with reduced activity of AMPK is associated with profibrogenic marker gene expression, progression of hepatic fibrogenesis and cellular invasiveness. Further investigations should now expand the studies to primary cells, both hepatocytes and stellate cells, to confirm the relevance of these interactions. It should also be analysed whether the effects are even more relevant in primary hepatocytes infected with HCV genotype 3, which has a significantly higher intrinsic profibrogenic potential than genotype 2 12.

Small heterodimer partner is a crucial circadian clock protein that has a robust global impact on the expression of major liver metabolic genes 13, 14 suggesting that it is one of the key molecules that participates in maintaining hepatic homoeostasis. This study has shown that NF-κB is one of the signalling pathways by which SHP mediates its antifibrotic activities. The SHP physically interacts with the NF-κB component p65 and serves as a transcription co-activator of NF-κB 15. However, the finding that the SHP/NF-κB axis can modulate hepatic TGF-β expression and its target gene CTGF/CCN2 that both play profound roles in the formation and progression of hepatic fibrogenesis is new and offers potentially new avenues for therapeutic interventions. A proof-of-concept for the efficacy of SHP in targeting fibrotic lesions in kidney was already reported by the same group some years ago 6, possibly indicating that the observed mechanisms of SHP activity are organ-independent and universal for inflammatory and fibrotic disorders. Consistent with this assumption, a previous report has shown that the administration of the SHP-inducing lipid-lowering drug fenofibrate increased survival upon experimental sepsis in mice and that stimulation of cultured macrophages results in abundance of SHP through activation of the liver kinase B1 (LKB1) that is a primary upstream kinase of AMPK 16. The multifunctional diversity of SHP in NF-κB signalling is even more complex because SHP can not only act as a repressor of transactivation of the NF-κB subunit p65 but also can act as a potent inhibitor of polyubiquitination of the TNF receptor-associated factor 6 (TRAF6) that is involved in TNF receptor superfamily as well as formation of Toll/IL-1 family receptor signalling complexes and crucial for a diverse array of physiological processes including adaptive immunity and innate immunity 17.

Most interesting is the finding that SHP can interfere with Smad3 phosphorylation in HCV-infected Huh-7.5 cells. It might be possible that SHP sequesters unphosphorylated Smad3 preventing its binding to receptor kinase and subsequent phosphorylation. Alternatively, phosphorylated Smad3 is an interaction partner for SHP fostering its degradation. Although a direct physical interaction was not shown in the present article, other studies have shown that SHP can repress Smad-dependent transcription by directly interacting with Smad2 and Smad3 in transient transfection experiments and inhibit the activation of endogenous TGF-β-responsive gene promoters such as p21, Smad7 and PAI-1 18.

It will be interesting to follow up if AICAR, metformin or other AMPK activators such as high molecular weight adiponectin that all trigger AMPK activation in vitro 9, 10, 19 might be in vivo suitable to block profibrogenic gene expression in fibrogenesis or hepatic stellate cell invasiveness during development of HCC. Consistent with the presented data, a previous study has already unanimously reported that the activation of AMPK is effective to prevent profibrogenic signalling in cultured primary murine hepatic stellate cells 20. However, in the same study it was also shown that low AMPK activity alone that was induced by AMPKα1 deletion failed to prevent hepatic stellate cell activation in vitro and further was not associated with enhanced fibrogenesis in the carbon tetrachloride model 20. On the other side, mice that underwent UUO and received AICAR one day before the surgery and daily thereafter attenuated extracellular matrix protein deposition and the expression of α-smooth muscle actin, type I collagen and fibronectin leading to an overall reduction in tubulointerstitial fibrosis 21.

Very recent studies have also shown that SHP interacts with and negatively regulates NLRP3 activity through a mechanism involving physical interaction with NLRP3 and the maintenance of mitochondrial homoeostasis that is induced by loss of SHP 22. Based on these findings, it is most likely that the ‘crippled nuclear orphan receptor’ lacking a function DNA binding domain is a warrior that counteracts inflammatory and fibrotic processes at a multitude of early and late attack points.

Taken together, the findings of the interesting study by Jung et al. shed a beam of light onto a small protein that seems to be more than just a simple competitor of co-activators or transcriptional repressor. Augmentation of SHP expression or pharmacological activation of SHP might hold promising potential for the treatment of inflammatory and fibrotic liver diseases.