MiR-22/galectin-9 Axis Suppresses Hepatocellular Carcinoma Progression And Is Epigenetically Regulated By EZH2

Part 1 IFN-? stimulation up-regulates the expression of galectin-9 and EZH2 in HCC cell linesObjectives: To study the effect of IFN-? stimulation on galectin-9 and EZH2 expression in HCC cell lines.Methods: To assess galectin-9 and EZH2 expression in HCC under interferon(IFN)-? stimulation,we exposed two HCC cell lines,namely,the Hep G2 and Hep3 B cell lines,to recombinant human IFN-? to mimic the microenvironment of hepatitis virus-associated HCC,an environment in which high concentrations of IFN-? have been detected.HCC cell lines were exposed to IFN-? of increasing concentration and duration.To further elucidate the influence of IFN-? on HCC cell lines,immunofluorescence was utilized to exhibit the galectin-9 and EZH2 expression directly.Results: We found that both galectin-9 and EZH2 expression was significantly up-regulated in concentration-dependent.We then stimulated the cells with 10 ng/ml IFN-? for increasing periods of time.As expected,we noted that galectin-9 m RNA and protein expression levels increased in a time-dependent manner.We noted that the fluorescence intensity of galectin-9 and EZH2 in IFN-?-treated cells was significantly higher than that in control cells.Conclusion: IFN-? stimulation up-regulates galectin-9 and EZH2 expression in HCC cells.Part 2 EZH2 modulates galectin-9 expression in HCC cells by repressing mi R-22Objectives: To investigate how EZH2 modulates the expression of galectin-9 in HCC cells.Methods: Firstly,three si RNAs targeting EZH2 were tested for their knockdown efficiency,the most efficient two were selected for further experiments.We detected the expression of galectin-9 at protein and RNA levels after knockdown and overexpression of EZH2,respectively.Then,HCC cells were transfected with EZH2 specific si RNAs after stimulating with IFN-?,followed by western blot and q PCR to determine the level of galectin-9.With bioinformatics tools,several mi RNAs were predicted to target galectin-9.And dual-luciferase vectors were constructed by cloning 3'-UTR of galectin-9 into the multiple cloning site of psi CHECK-2,followed by dual-luciferase assay to confirm the functional relationship between mi RNAs and galectin-9.Then specific mimics and inhibitors were transfected to detect the expression of galectin-9 at protein levels.After confirming the specific mi RNA targeting galectin-9 in HCC,gain and loss of function of EZH2 was conducted to determine the expression of this mi RNA.Results: Our results showed that after knockdown of EZH2,galectin-9 expression was significantly down-regulated in HCC cells,while over-expression of EZH2 led to up-regulation of galectin-9 at both RNA and protein levels.What's more,knockdown of EZH2 could partially block the up-regulation of galectin-9 induced by IFN-? stimulating.Online bioinformatics predicted that mi R-22-3p,mi R-296-3p,mi R-455-5p,mi R-491-5p could target galectin-9,and dual-luciferase assays confirmed that only mi R-22-3p lead to significant decrease in the luciferase activity.Further study found that the level of EZH2 was inverse correlated with mi R-22 both in HCC tissues and cell lines,EZH2 gain and loss of function study showed that EZH2 could repress mi R-22 and MIR22 HG expression in HCC cell lines,and EZH2 modulated galectin-9 expression indirectly through repressing the expression of mi R-22.Part 3 EZH2 epigenetically repressing mi R-22 through H3K27 tri-methylation but not DNA hyper-methylationObjectives: To study the molecular mechanisms by which EZH2 repress the expression of mi R-22.Methods: The mi R-22 expression in HCC was parallel with its host gene,MIR22 HG.We retrieved the sequence of MIR22 HG and determined an 858-bp Cp G island covers the promoter region of the gene.Specific primers were designed and MSP assays were conducted in both HCC cell lines and HCC tissues and paired adjacent normal tissues.To further confirm the finding of MSP,BSP assays were conducted.Then,a series of 5' dissection of mi R-22 promoter were cloned into the multiple cloning sites of p GL3 Basic vector.These vectors were co-transfected with p RL-TK vector,followed by dual-luciferase reporter gene assay to determine which region was the core region with the highest promoter activities.Then specific primers of this core promoter region were designed followed by Ch IP assays with anti-EZH2,anti-H3K27me3,anti-Ig G antibodies in HCC cell lines transfected by EZH2 si RNA,p CMV-EZH2 and control,respectively.To further confirm our Ch IP results,Ch IP-seq datas from ENCODE database were retrieved.Results: We found an 858 bp long Cp G island near the transcription start site of MIR22 HG.The MSP results of HCC cell lines and HCC tissues and paired adjacent normal tissues showed that the promoter region of mi R-22 promoter was hypo-methylated,which was further verified by BSP assays.Through dual luciferase reporter gene assays of 5'-dissections of the promoter region,we defined the-337~-65 segment upstream of the transcription start site was the core element of promoter.Ch IP assays in EZH2-knockdown cells showed that the occupation status of EZH2 and H3K27me3 on the above-mentioned promoter region was significantly decreased compared with control,while over-expression of EZH2 un-regulated the enrichment of EZH2 and H3K27me3.To further confirm our findings,the Ch IP-seq datas of anti-EZH2,anti-H3K27me3 antibody were retrieved fromENCODE database.The EZH2 and H3K27me3 peaks were also found upstream of the transcription start site.Conclusion: EZH2 epigenetically represses mi R-22 transcription by tri-methylation of H3K27me3 at the core promoter element,not DNA hyper-methylation.Part 4 Effects of mi R-22 and galectin-9 on the malignant behaviors of HCC cells in vitro and in vivoObjective: To determine the effects of mi R-22 and galectin-9 on biological behaviors of HCC cells both in vitro and in vivo.Methods: Hep G2 cells transfected with p Enter-control and p Enter-pre-mi R-22 were selected by puromycin(2ug/ml).Then,galectin-9 overexpression plamids were co-transfected to the above-mentioned two stable cell lines.Galectin-9 expression in these four cell line were determined by western blot.Proliferation ability of these cell lines were detected by CCK-8 assays;Flow cytometry apoptosis analysis was used to evaluate the percentage of apoptoic cells,and levels of Bcl-2,BAX,caspase3,activated caspase3 were determined by immunoblot assay;Colony formation assay was utilized to measure the clonogenecity ability;Transwell migration and invasion assays were used to evaluate the migration and invasion ability in these four cell lines,respectively.Xenograft tumor formation and lung metastasis assays were used to detect the tumorigenecity and metastasis ability of mi R-22 and galectin-9 overexpression cells.Results: Enforced expression of mi R-22 decreased galectin-9 expression,as demonstrated by westernblotting.CCK-8 and FACS assays showed that enforced mi R-22 expression decreased cell viability and induced apoptosis compared to stable transfection of empty vector.The elevated levels of cleaved-caspase3 and BAX while declined BCL-2 was noted.Moreover,colony formation assay showed that mi R-22 overexpression inhibited Hep G2 and Hep3 B cell clonogenicity.Transwell assay showed that mi R-22 overexpression suppressed HCC cell migration and invasion ability.We subsequently co-transfected cells with galectin-9-overexpression vectors and pre-mi R-22 or empty vector.We found that co-transfection with pre-mi R-22 and galectin-9-overexpression plasmids restored galectin-9 protein expression.In xenograft tumor assay,we found that stable transfection of pre-mi R-22 into Hep G2 cells led to decreases in the sizes and weights of subcutaneousxenograft tumors in athymic nude mice compared to stable transfection of empty vector(mock).We co-transfected galectin-9-stable cells with pre-mi R-22 or empty vectors and found that these cells displayed lower tumor weights and smaller tumors than cells transfected with pre-mi R-22 or mock alone mi RNA,galectin-9-overexpression and pre-mi R-22 transfection suppressed HCC cell proliferation and migration.Conclusion: Although galectin-9 is negatively regulated by mi R-22,a well identified tumor suppressive mi RNA,both galectin-9-overexpression and pre-mi R-22 transfection suppressed the malignant behaviors of HCC cells.