The Oncoprotein P28^GANK Establishes A Positive Feedback Loop In β-catenin Signalling

PartⅠThe oncoprotein p28^GANK establishes a positive feedback loop inβ-catenin signallingHepatocelluar carcinoma(HCC) is one of the most prevalent malignant cancers in our country.The development and progression of HCC is a complicated process involving multiple genes and multiple steps in human bodies.Up to now,though some suppressors in HCC were identified such as p53 and p16^INK4a,and confirmed their specific inactivation,few specific activated oncogenes in HCC was found and reported.In 2000,a novel gene named gankyrin with repeated sequences coding "ankyrin" motif was cloned in human HCC by Japanese scientists Fujita and his colleagues using subtractive hybridization method.We found this gankyrin gene was identical to the p28(Nas6p) gene reported by Hori in 1998 by searching the GenBank data,thus named it as p28^GANK in our work.p28^GANK(also known as PSMD10,p28,Nas6p,and Gankyrin),a non-ATPase subunit of the 19S/PA700 complex from the 26S proteasome,contains five ankyrin repeats mediating proteins interaction.20S proteasome is a protease degrading some incorrectly folding proteins or cell cycle regulators.Little is known about the mechanism how oncoprotein p28^GANK relates with proteins degradation by 26S proteasome,p28^GANK regulates the CDK4/CyclinD1/p16^INK4a/Rb1/E2F-1 pathway which connects with tumor development and progression,and p28^GANK-transfected cells are tumorigenic in nude mice.Recently,Nagao et al has identified MAGE-A4 binds to gankyrin and suppresses its oncogenic activity through yeast two-hybrid screen,p28^GANK also directly binds to MDM2,which increases p53-MDM2 association,inducing the ubiquitylation and degradation of p53.Theβ-catenin signalling pathway has a critical role in cell-fate determination,tissue homeostasis and tumorigenesis.In unstimulated cells,the cytoplasmic concentration ofβ-catenin is kept very low owing to phosphorylation by CK1 and GSK3βkinases. Activation of the Wnt pathway blocksβ-catenin degradation so thatβ-catenin accumulates and translocates to the nucleus,where it interacts with T-cell factor (TCF)/lymphoid enhancer factor(LEF) transcription factors.About 50%-70%of all HCC examined showed an abnormalβ-catenin protein accumulation in the cytoplasm and nucleus.However,in HCC theβ-catenin mutation rate is 13%-26%,the Axin mutation rate is 5%-10%and rare mutation of APC has been reported so far.This led us to speculate that other factors might be involved in the regulatingβ-catenin in HCC.To investigate the factors which might influence p28^GANK expression,we cloned a 1226 bp fragment of the 5'-flanking region of human p28^GANK gene into pGL3 luciferase reporter vector,which analyzed using Gene2Promoter(Genomatix) online software.To elucidate whether p28^GANK upregulation was correlated with cell cycle progression in normal or HCC cell lines,we examined the potential cell-cycle-dependence of p28^GANK expression in HepG2 and HEK293 cells after mitogenic stimulation.We found that EGF or HGF stimulation significantly increased p28^GANK report gene activity after the stimulation with different doses of either EGF or HGF.We also demonstrated that Ras activation could upregulate p28^GANK expression.Use of the PI3K inhibitor LY294002 abolishing the ability of growth factors and Ras activation induced PI3K-Akt cascade irritation and use of PD98059 or U0126 to inhibit ERK kinase cascade activation,we then approached p28^GANK expression by detection its promoter activation in HEK293 cells.The results show that PI3K inhibitor LY294002 significantly suppressed EGF or HGF induced p28^GANK reporter gene activities,while the ERK inhibitors PD98059 and U0126 show some enhancement instead of any inhibition.Reporter gene assays were performed on HEK293 ceils with introduced exogenousβ-catenin and c-Myc expression on p28^GANK expression.To our surprise,they both can significantly increased p28^GANK promoter activity.Furthermore,we found that theβ-cantenin/TCF could directly bind to the p28^GANK promoter in vivo.Then we asked whether p28^GANK expression could affectβ-catenin/TCF-dependent transcription.We found that introduced expression of p28^GANK induced activation ofβ-catenin/TCF-mediated transactivation in a concentration-dependent manner.Studies have revealed that elevating levels of wild-type p53 will downregulateβ-catenin in a variety of cell types.It was reported that p28^GANK could bind to Mdm2,facilitating p53-Mdm2 binding,and increased ubiquitylation and degradation of p53.To elucidate whether the effect dependent on p53,we detected theβ-catenin-OT activity in p53-negative Hep3B cells.The reporter activities were decreased with declined expression of p28^GANK and the RT-PCR assay also indicated that p28^GANK reduction resulted in a substantial decrease inβ-catenin-OT reporter activity which was accompanied by the downregulation of c-Myc,as well as cyclinD1.We assessed the correlation betweenβ-catenin signaling and the expression of p28^GANK in human primary hepatic turnouts.We found that there was a remarkable correlation between the activationβ-catenin signalling and the expression of p28^GANK,c-Myc and cyclinD1,which indicated thatβ-catenin signaling-induced p28^GANK expression might be an important mechanism for promoting cell cycles in human liver cancers.In order to confirm the positive feedback regulating effect,we examined the effect of p28^GANK expression on its own promoter activity in HEK293 and Hep3B cells.To our prediction,p28^GANK significantly increased its own expression in HEK293 cells.After endogenous p28^GANK knockdown in Hep3B cells,p28^GANK promoter activities decreased.In conclusion,we show that growth factors stimulation or Ras activation could upregulate p28^GANK expression through activationβ-catenin signalling.Inaddtion,β-cantenin was transcriptional activator of p28^GANK and the induced p28^GANK subsequently regulated theβ-canenin/TCF transcription activation.Thus the oncoprotein p28^GANK established a positive feedback loop inβ-catenin signaling and involved in tumorigenesis and progression in HCC. PartⅡSIRPαNegatively Regulates Both TLR3 and Cytoplasmic Pathways in TypeⅠIFN InductionSignal regulatory proteinα(SIRPα) was originally identified in rat cells by its association with cytoplasmic tyrosine phosphatase Src homology region 2 domain-containing phosphatase SHP-2 and was later shown to be highly conserved in other mammals,including human,mouse,and cattle.The cytoplasmic region of SIRPαcontains two immunoreceptor tyrosine-based inhibitory motifs(ITIMs) with four tyrosine residues that are phosphorylated in response to a variety of growth factors and ligand binding.This phosphorylation enables recruitment and activation of SHP-1 and SHP-2 that in turn dephosphorylates specific protein substrates involved in mediating various physiological effects.SIRPαis especially abundant in innate immune cells including macrophages and dendritic cells although it is also expressed in other cell types such as neurons and fibroblasts.Host antiviral responses are initiated through the detection of viral components by host pattern recognition receptors(PRRs).Upon recognition,PRRs activation results in expression of typeⅠinterferon(IFN-α/β),IFN-stimulated genes,and inflammatory cytokines that suppress viral replication and facilitate the adaptive immune responses. Double-stranded RNA(dsRNA),which is produced during replication of many viruses,is one of the viral components recognized by several PRRs,including Toll-like receptor(TLR)3 and the RNA helicases,namely retinoic acid-inducible geneⅠ(RIG-Ⅰ) and melanoma-differentiation-associated gene 5(MDA5).It is accepted that TLR3 detects extra-cellular viral dsRNA internalized into the endosomes,whereas RIG-Ⅰ/MDA5 detects intracellular viral dsRNA.Engagement of these receptors triggers the rapid production of typeⅠIFN and thus activates the innate immune response against infectious virus.SIRPαhas been shown to negatively or positively regulate MAPKs signaling initiated either by tyrosine kinase-coupled receptors for growth factors or by cell adhesion to extracellular matrix.Moreover,the expression of dominant negative form of SIRPαstimulates NF-κB activity and makes the cells resistant to TNF specific apoptosis.In our previousely study,we demonstrated the role of SIRPαin regulating TLR4-mediated signaling during innate immune responses.However,it is not known whether SIRPαis involved in regulating typeⅠIFN induction signaling.In the present study,we found that SIRPαexpression was lost after stimulation with polyriboinosinic:polyribocytidylic acid(poly(I:C)),a synthetic analogue of dsRNA. We also provide evidence that SIRPαfunctions as a specific inhibitor of dsRNA-activated signaling,transcription factor activation and gene induction in both the TLR3 and the cytoplasmic pathways in cells.Furthermore,the phosphorylation on the tyrosine residues of ITIM motifs in SIRPαand its binding to the p85 subunit of PI3K in a ligand-dependent fashion are required for its inhibitory function.Thus,we demonstrate an important function for SIRPαas an inhibitory protein in both TLR3 and cytoplasmic pathways in typeⅠIFN induction.To explore the involvement of SIRPαin regulating the induction of typeⅠIFN,we examined SIRPαprotein and mRNA levels after poly(I:C) treatment in mouse macrophage cell line(RAW 264.7) as well as in mouse peritoneal washout macrophages(PWCs).The results show that poly(I:C) treatment triggered a dosage-dependent decline in expression of SIRPαin protein extracts from RAW 264.7 and PWCs.Moreover,poly(I:C) treatment at 5μg/ml resulted in a decline in expression of SIRPαafter 6 hr(2-fold by densitometry),and SIRPαremained lower than control levels after 24 hr.quantitative real-time PCR analysis showed that there was nearly a 1.5-fold reduction of SIRPαmRNA expression after 30 min of exposure to poly(I:C) and treatment.Stimulation of TLR3 with poly(I:C) leads to IRF3 phosphorylation on multiple phosphorylation acceptor(phospho-acceptor) sites.We therefore investigated the possible involvement of SIRPαin regulating the activity of IRF3.We first synthesized small interfering RNAs(siRNA) targeting mouse SIRPαin order to suppress the endogenous SIRPαexpression in PWCs.We also constructed a shRNA vector that specifically downregulated SIRPαand then stably transfected it into RAW264.7 macrophage cells.The transfections were referred to here as SIRPI-KD(knockdown),OV(overexpression) and VT(vector) macrophages respectively.In response to poly(I:C),SIRPα-siRNA macrophages exhibited an enhanced phosphorylation of IRF3 whereas cells transfected with negative-control oligonucleotides had a much lower IRF3 activation in PWCs.There was an enhanced phosphorylation of IRF3 in SIRPα-KD compared to VT treated with poly(I:C). Therefore,we examined the effects of knockdown of SIRPαexpression on the activation of reporter genes of the ISRE,PRDⅢ-Ⅰand PRDⅡ.We found that all reporter genes were strongly activated in SIRPα-KD macrophages as compared with SIRPα-VT cells.To examine whether IFN-βand IFN-βinducible gene expression was indeed suppressed by SIRPα,we stimulated with poly(I:C) and analyzed the expression of IFN-βand IFN-βinducible genes,including ifit-1(encoding IFIT-1), ccl-5(encoding RANTES) and mip-1α(encoding MIP-1α).As expected, poly(I:C)-induced expression of these genes were higher in SIRPα-KD macrophages, whereas overexpression of SIRPαresulted in a reduction of the expression of these genes.We also analyzed the production of IFN-β,TNF-αand IL-6 after poly(I:C) treatment.All the three cytokines were significantly enhanced in SIRPα-siRNA PWCs compared to negative-control PWCs,as measured by ELISA.To further investigate the direct involvement of SIRPαin virus-induced typeⅠIFN production, we examined the effects of SIRPαknockdown on MAPK,NF-κB and STAT1 pathways which are all required for cytokine production during TLR3 signaling. SIRPα-siRNA macrophages displayed enhanced phosphorylation of p38,JNK and ERK,whereas negative-control macrophages displayed impaired activation.Because the phosphorylation of IκBαwas essential for the activation of NF-κB,we then examined the phosphorylation status of IκBαand NF-κB activity in macrophages with SIRPαknockdown.There was enhanced phosphorylation of IκBαand NF-κB activity in SIRPα-KD macrophages after poly(I:C) stimulation,which was consistent with the higher NF-κB activity,responsible PRDⅡactivity in SIRPα-KD macrophages.Furthermore,tyrosine phosphorylation of STAT1 was also displayed enhanced phosphorylation in PWCs transfected with SIRPα-siRNA.TypeⅠIFN and inflammatory cytokines are induced when poly(I:C) is administrated into the cytoplasm by lipofection.Notably,this induction is independent of TLR3 and dependent on RIG-Ⅰ/MDA5.To determine the effects of SIRPα,we used HEK293 cells lacking TLR3 expression.HEK293 cells were transfected with a SIRPαexpression plasmid or an empty plasmid and 24 hr later cells were re-transfected with poly(I:C) at different time points.Cells transfected with SIRPαexpression plasmids were found to exhibit decreased phosphorylation of IRF3 compared with cells transfected with empty plasmids.We next examined the effects of SIRPαon RIG-Ⅰ/MDA5-dependent gene induction at the level of promoter activation.We co-transfected cells with different doses of SIRPαexpression plasmids together with ISRE,PRDⅢ-Ⅰ,IFN-βor RANTES reporter genes for 24 hr and then re-transfected with poly(I:C) for another 6 hr.Consistent with the effect of SIRPαon RIG-Ⅰ/MDA5-dependent IRF3 activation,ISRE,PRDⅢ-Ⅰ,IFN-βand RANTES gene reporter activities were severely impaired in SIRPα-transfected cells in a dose-dependent manner.Impaired tyrosine-phosphorylated STAT1 was also observed in HEK293 cells transfected with SIRPαexpression plasmid.To further investigate whether the inhibition of typeⅠIFN induction by SIRPαwas dependent on the tyrosine phosphorylation in its ITIM motif,we adopted a SIRPα-4Y expression plasmid.HEK293 cells were co-transfected with different doses of SIRPα-WT or SIRPα-4Y expression plasmids together with ISRE,PRDⅢ-Ⅰ, RANTES or IFN-βreporter genes and then transfected with poly(I:C).Only SIRPα-WT inhibited poly(I:C) transfection-induced activation of ISRE,PRDⅢ-Ⅰ, RANTES and IFN-βreporter genes,while the SIRPα-4Y exhibited no inhibitory effects.Activation of PI3K plays a key role in the recruitment and activation of a wide variety of lipid and protein kinase-signaling cascades.The activity of PI3K has also been reported to be essential for IRF3-induced gene induction by dsRNA.Hence we investigated whether SIRPαmediated its negative regulatory effects via coupling PI3K.The requirement of phosphoinositide 3-kinase(PI3K) activity for the induction of IFN-βand IFN-β-inducible genes by dsRNA is supported by the observation that a PI3K inhibitor failed to activate IFN-βand IFN-β-inducible gene expression.Moreover,the activity of PI3K downstream kinase AKT was downregulated in the SIRPα-overexpression RAW264.7 and HEK293 cells.Given the observations that SIRPαwas tyrosine-phosphorylated in response to dsRNA stimulation and PI3K was required for gene induction in the production of typeⅠIFN-β,we investigate whether PI3K could interact with SIRPα.Our results demonstrated that PI3K was recruited to SIRPαafter poly(I:C) stimulation, suggesting that the recruitment was dependent on the phosphorylation of the tyrosine residue in the cytoplasmic domain of SIRPα.In conclusion,we have shown here that SIRPαplays a critically negative role in typeⅠIFN induction.SIRPαmay accomplish its inhibitory function in typeⅠIFN induction,in part,through its association and sequestration of the signal transducer PI3K.Therefore SIRPαmay be an extremely beneficial and potentially exploitable molecule to inhibit immune responses such as autoimmunity or to enhance the immune response during viral infections.