【作者】 石萌；

【导师】 何继满；

【作者基本信息】 南方医科大学 ， 内科学， 2014， 博士

【摘要】 1.研究背景和目的虽然全球癌症总的发病及死亡率显出令人鼓舞的下降势头,但是,胃肠道肿瘤仍是一种顽固的癌症,其中大约2/3发生在发展中国家。中国的胃肠道肿瘤发生数占到了全球的42%,是影响健康和社会的一个极其严重的问题。胃肠道肿瘤初次诊断发现时,往往近半的患者已经局部恶化或已经转移,降低手术切除的功效。因此,胃肠肿瘤的早期诊断和切除预后是医学研究两个重要课题,而早期标志物及预后机制的认识则至为关键。Fas也被称为CD95,它是DR家族(肿瘤坏死因子受体(TNFR)超家族的一个亚家族)的一员。Fas信号通路被FasL (Fas受体的一个配体)激活后促进细胞凋亡。化疗药物可引起FasL与Fas的上调,凭借Fas诱导的细胞凋亡,有助于消除肿瘤细胞。然而,很多肿瘤细胞都对化疗后Fas介导的细胞凋亡具有抗性。同时有报导称Fas信号可以调节诸多非凋亡效应,特别是在Fas抵抗的肿瘤细胞中所发生的肿瘤生存、侵袭和转移等过程。在癌症发生过程中,上皮表型肿瘤细胞失去其上皮样特征而获得有较强侵袭和迁移能力,该过程称为上皮-间充质转化(Epithelial-mesenchymal transitions,EMT)。EMT过程包括E-cadherin的表达抑制和N-cadherin等表达强化的过程。包括Snail、Slug、Twist、ZEB1和ZEB2在内的多个转录因子被鉴定为EMT和肿瘤转移的诱导者。作为转录后调控的一部分,miRNA因为其靶向调控EMT事件中关键蛋白表达的能力而成为它们的重要调节因子miRNA是一类非蛋白编码小RNA,在生长、发育、细胞增殖和凋亡等多个生物过程中起重要作用。它们对于肿瘤形成的重要方面,如引发、肿瘤生长、肿瘤发展、分化等,主要通过干扰涉及关键环节(包括细胞周期、细胞凋亡、细胞迁移)的基因的表达,发挥着关键作用。除了miR-200基因族,其他几种miRNA基因族的成员也被证实通过在多种类型癌症中瞄准EMT关键基因,从而涉及EMT过程的调节。我们可以推测Fas信号的非凋亡效应可能通过EMT现象对胃肠道肿瘤细胞的侵袭和转移起到促进作用,于此同时,miRNA可能在一定程度上调控了Fas信号在其中的作用。在这项研究中,我们首先对Fas信号在胃肠道肿瘤细胞中的非凋亡效应进行了检测,同时对其能否引起EMT过程进行了检测。然后基于miRNA芯片检测Fas信号引起的miRNA表达谱的变化,通过预测靶基因和进行功能及通路分析,选择合适的靶基因及相应miRNA进行验证。2.材料和方法2.1.细胞胃癌细胞株AGS,结直肠癌细胞株SW480、DLD1、SW1116、HT-29、LoVo和SW620,胰腺癌细胞株SW1990、BxPC-3,肝癌细胞株HepG2购自ATCC。胃癌细胞株BGC-823、SGC-7901、MGC80-3,肝癌细胞株SMC7721,食管癌细胞株Eca-109购自中国科学院细胞库。上述细胞株由南方医院消化病研究所细胞培养室保存提供。2.2流式细胞仪检测将对数生长期细胞悬浮于2%BSA(PBS配制)中,用计数板细胞计数,将细胞密度调整为5×105个/mL。对照组加入PE标记的抗小鼠IgG,样品加入PE标记的小鼠抗人Fas抗体或小鼠抗人FasL抗体孵育30分钟后,利用流式细胞仪(BD公司)检测上述各细胞株Fas受体和mFasL配体的表达情况。2.3.细胞迁移试验将AGS和SW480细胞分为对照组、NOK-1组、NOK-1+sFasL组和FasL组4组,采用transwell小室进行迁移试验。对NOK-1组予以10μg/mL的NOK-1(FasL抗体)处理,NOK-1+sFasL组予以NOK-1(10ug/mL)预处理1小时后予低剂量sFasL (12.5ng/mL)处理,FasL组予以低剂量sFasL处理,对照组不做特殊处理。培养6小时后,行DAPI染色对穿过transwell小室的细胞进行计数。2.4.蛋白印迹试验选择胃癌细胞株AGS细胞和BGC-823细胞、肠癌细胞株SW480细胞、胰腺癌细胞株BxPC-3细胞和SW1990细胞作为研究对象。实验组给予低剂量sFasL(12.5ng/uL)处理,对照组不予特殊处理,然后分别提取1日、2日、3日后各细胞株的总蛋白,行免疫印迹检测检测EMT标记蛋白E-cadherin、Occludin、 Villin、N-cadherin、Vimentin、Snail1、MMP9的表达水平变化。2.5.MiRNA芯片将AGS及SW480细胞分为三组,分别用FasL(12.5ng/mL)处理0小时(对照组),6小时和12小时(实验组)。采用miRNA提取试剂盒抽提总RNA。通过miRCURYTM LNA Array平台进行微阵列分析。微阵列与标记的样品在微阵列杂交系统中56℃杂交16-20小时。然后使用GenePix4000B微阵列扫描仪扫描载玻片,并用GenePix Pro6.0软件对图像进行分析。使用标准化中值对表达数据标准化,对至少1.5倍变化的差异表达的miRNA进行鉴别。使用MEV软件(V4.6, TIGR)进行分层聚类。2.6.靶基因预测选择表达差异倍数>1.5的miRNAs,在GOmir6.1中选择TargetScan, miRanda, RNAhybrid, PicTar4way和PicTar5way五个数据库同时进行靶基因的预测,将同一个miRNA在至少3个数据库中发生重叠的靶基因作为预测结果。2.7.靶基因功能和通路富集分析利用DAVID6.7系统进行miRNA靶基因的GO分析,得出靶基因有显著联系的、低误判率的、靶向性的基因功能分类,选取P<0.05, FDR<0.05作为限定值。同样利用DAVID6.7系统进行基于KEGG的信号传导通路富集分析,选取P<0.05,基因数目>3,FDR<0.05作为限定值。2.8.实时荧光定量PCR将AGS细胞分为三组,分别予FasL(12.5ng/mL)处理6小时和12小时(实验组),对照组不予特殊处理。提取总RNA然后进行miRNA逆转录,然后通过定量RT-PCR检测目的miRNA的Ct值,检测目的miRNA的表达变化。以2-△△Ct(△Ct=CtmiRNA-CtU6,△△Ct=△Ct试验组-ACt对照组)结果表示。2.9.统计学处理数据统计采用PASW Statistics18.0软件分析,结果表述包括统计量值和具体P值,所有结果均以P<0.05表示差异有统计学意义。细胞迁移试验、荧光定量PCR结果组间比较采用单因素方差分析(One-way ANOVA),利用Levene法进行方差齐性分析,如方差齐,多重比较采用LSD法。方差不齐者组间比较采用Welch法,多重比较采用Dunnett’s T3法。细胞生长曲线结果采用析因设计方差分析,方差齐者,多重比较采用LSD法。特别的,荧光定量PCR结果各处理组与对照组比较采用单样本t检验,细胞生长曲线结果各时间点与对照组比较采用单样本t检验。miRNA验证试验中,组间比较采用独立样本t检验。3.结果3.1.胃癌细胞株AGS、BGC-823、SGC-7901,结肠癌细胞株SW480、DLD1、 SW1116、HT-29、LoVo和SW620,胰腺癌细胞株SW1990、BxPC-3,肝癌细胞株HepG2、SMC7721和食管癌细胞株Eca-109强表达Fas受体(>50%),不表达mFasL配体(<10%)。3.2.相对于对照组的细胞生长,低剂量sFasL(12.5ng/ml)不会对AGS和SW480细胞产生显著影响(P=0.575,P=0.373),而中剂量(25ng/ml)或高剂量(50ng/ml)的sFasL则会对AGS和SW480细胞产生显著抑制作用,差异有统计学意义,且P值均小于0.001。3.3.相对于对照组,低剂量sFasL能够显著增强AGS细胞的迁移能力(P<0.001),且这种作用能够被FasL的抗体NOK-1所抑制(P=0.359)。同样地,相对于对照组,低剂量sFasL能够显著增强SW480细胞的迁移能力(P<0.001),且这种作用能够被FasL的抗体NOK-1所抑制(P=0.681)。3.4.低剂量sFasL (12.5ng/uL)可以促进胃癌细胞株AGS、BGC-823,肠癌细胞株SW480、SW1116,胰腺癌细胞株SW1990和食管癌细胞株Eca-109的细胞形态由上皮样表型向间质样表型转变。3.5.低剂量sFasL处理可使AGS细胞、SW480细胞、BGC-823细胞、BxPC-3细胞和SW1990细胞的上皮细胞标记蛋白E-cadherin、Occludin、Villin表达下调,同时使间质细胞标记蛋白N-cadherin、Vimentin、Snaill和MMP9等表达上调。3.6.低剂量sFasL可导致AGS细胞中ERK/GSK3-β蛋白的磷酸化,SW480细胞中ERK/Akt蛋白的磷酸化。在AGS和SW480细胞中,相对于对照组,使用U0126预处理2小时,可以抑制sFasL处理导致的p-ERK蛋白的表达。进一步,ERK通路抑制剂U0126可以阻断低剂量sFasL导致的EMT标记蛋白的表达变化。3.7.低剂量sFasL处理对AGS和SW480细胞中EMT标记蛋白表达的影响,在24小时内即可发生。低剂量sFasL处理6小时和12小时后,上皮细胞标记蛋白E-cadherin、Occludin、Villin表达下调,间质细胞标记蛋白N-cadherin、 Vimentiin、Snaill和MMP9等表达上调。3.8.低剂量sFasL处理后AGS及SW480细胞的miRNA表达出现显著变化。AGS中,约三分之一的miRNA (35%,175/500)显示出至少1.5倍的表达差异,其中86.9%(150/175)表达下调,只有13.1%(25/175)表达上调。同时在SW480细胞中,24%的miRNA (120/500)显示出至少1.5倍的表达差异,这其中33.3%(40/120)表达下调,66.7%(80/120)表达上调。3.9.在TargetScan, miRanda, RNAhybrid, PicTar4way和PicTar5way五个数据库中同时进行靶基因的预测。在AGS中得到下调表达miRNA靶基因1805个,上调miRNA靶基因247个。在SW480中得到下调miRNA靶基因1088个,上调miRNA靶基因1788个。3.10.靶基因的GO基因功能分类表明,在AGS中,差异表达miRNA靶基因功能集中在regulation of transcription、cell morphogenesis和regulation of biosynthetic process等过程中。而在SW480中,差异表达miRNA靶基因功能集中在regulation of transcription、neuron differentiation和regulation of biosynthetic process等过程中3.11. KEGG信号通路富集分析表明,在AGS和SW480细胞中经低剂量sFasL处理后,差异表达miRNA的靶基因与Adherens junction和Neurotrophin signaling pathway通路相关。3.12.将予低剂量sFasL处理后AGS细胞和SW480细胞中下调表达的miRNA合并分析,利用KEGG将靶基因产物在adherens junction通路上进行标示,结合文献我们发现靶基因Smad3、Snail1、Rac1、Cdc42和MAPK3可能在EMT中起重要作用,其对应miRNA为miR-23a, miR-92a, miR-138和miR-155。3.13.在AGS细胞中,利用实时荧光定量PCR验证miRNA发现,与对照组的相比,低浓度sFasL处理6小时后或12小时后,miR-23a, miR-92a, miR-138和miR-155四条miRNA的表达量均显著降低,且差异有统计学意义,P值均小于0.05。这种抑制作用在使用FasL抑制剂NOK-1预处理后被显著地削弱,且差异有统计学意义,P值均小于0.05。3.14.在AGS细胞中,利用实时荧光定量PCR验证靶基因mRNA发现,与对照组相比,低浓度sFasL处理6小时或12小时后,Smad3、Snail1、Rac1、Cdc42和MAPK3基因的相应mRNA的表达均显著升高,且差异具有统计学意义,P值均小于0.05。3.15.在AGS细胞株中,与对照组相比,低浓度FasL处理6小时和12小时后,Smad3、Snail1、Rac1、Cdc42和MAPK3基因的相应蛋白的表达均有明显升高。4.结论在Fas受体高表达而mFasL低表达的胃肠道肿瘤细胞株中,低剂量的sFasL处理可以诱导EMT的产生,同时促进肿瘤细胞的迁移能力。这个过程中,ERK通路的激活起了关键作用。同时,Fas信号通路显著地改变了AGS细胞和SW480细胞的niRNA表达谱,基因功能和通路分析表明,差异表达miRNA靶基因功能集中在regulation of transcription和regulation of biosynthetic process等过程中,并且与Adherens junction通路密切相关。根据生物信息学分析和实时荧光定量PCR验证中发现,miR-23a, miR-92a, miR-138和miR-155可能通过调控靶基因Smad3、Snail1、Rac1、Cdc42和MAPK3在低剂量sFasL诱导EMT的过程中起到了调节作用。更多还原

【Abstract】 IntroductionDespite the encouraging downward trend for overall cancer incidence and mortality globally, gastrointestinal (GI) cancer still remains a formidable cancer with approximately two thirds of the cases occurring in developing countries. China alone accounts for42%of cases of GI cancer, which remains the most deadly form of cancer among both sexes in the country GI cancer is often diagnosed at a very advanced stage and close to half of the patients are diagnosed with unresectable, locally advanced, or metastatic disease. Identification of novel molecular and cellular mechanisms responsible for tumorigenesis and progression of GI cancer are thus critical to the improvement of the prognosis of GI ancer patients.Fas, also called CD95, is a member of the DR. family, which is a subfamily of the tumor necrosis factor receptor (TNFR) superfamily. Fas signaling triggered by FasL, a ligand of Fas receptor, promotes cellular apoptosis and modulates cell cycle progression, autophagy, inflammation and innate immunity. Chemotherapeutic drugs can cause upregulation of FasL and Fas, which contributes to the elimination of tumor cells by Fas-induced apoptosis. However, many tumor cells are still resistant to Fas-mediated apoptosis after chemotherapy. Therefore,it is possible that chemotherapy-mediated upregulation of FasL and Fasinduces tumor proliferation and metastasis in a subset of patients that are resistant to treatment.Epithelial-mesenchymal transition (EMT)is a good model to explain how solid tumors metastasize from the site of origin to a new site. EMT occurs by a series of orchestrated events in which cell-cell and cell-extracellular matrix interactions are altered, and the transition from an epithelial to a mesenchymal phenotype allows for cell movement. During cancer progression, advanced stage cancer cells frequently show downregulation of epithelial markers, which leads to loss of epithelial polarity, intercellular junctions, and reduced intercellular adhesion, and these alterations are often accompanied by increased cell motility and expression of mesenchymal markers, indicating EMT process may probably be involved in cancer metastasis. In malignant tumor, the Fas signaling pathway has recently been shown to promote cancer cell metastasis and motility through the EMT in GI cancer, suggesting the need for further evaluation of the Fas signaling pathway in relation to GIcancer.MiRNA (miRNA), a class of non-protein-coding small RNA, play a vital role in multiple biological processes such as development, cellular proliferation and apoptosis. MiRNAs may act as tumor suppressors or oncogenes and deregulated miRNA expression has been document in human cancers, including GI cancer. Recently, specific EMT or metastasis-regulating miRNAs have been found to be associated with EMT and cancer metastasis. For instance, miR-9increases cell motility and a context-dependent EMT-like conversion by directly targeting E-cadherin.In the current study, we examined whether Fas signaling can promote motility in GI cancer cells and the possible mechanisms required. Meanwhile, we examined the Fas signaling-induced changes in the global miRNA expression profile of GI cancer cells using a miRNA-based microarray chip assay and further identified the target genes of four miRNAs whose expression were shown to be downregulated. We also annotated the functional pathways associated with the four miRNAs and the predicted target genes.Material and MethodsCells culture All human GI cancer cell lines were obtained from American Type Culture Collection (Manassas, VA) andthe cell bank of Chinese academy of sciences (Shanghai),and routinely maintained in our lab. These cell lines werecultured at37℃in RPMI1640medium with2mM glutamine,10%fetal bovine serum (FBS), and1%penicillin and streptomycin in a humidified atmosphere containing5%CO2.Cell lines were serum-starved for24h before the assayin most experiments unless otherwise indicated.Flow cytometryThe cells were trypsinized and pelletedy centrifugation. After washing the pellet with PBS, the cells were counted, and stained according to the instruction of Flow Cytometry-BestProtocols of eBioscience (San Diego, CA). The surface expression of Fas and FasLwere detected using the FACS Calibur Flow Cytometer (BD, USA). Cell migration assayThe invasive potential of cells was evaluated using transwell inserts with8μm pores (Coring, NY).AGS or SW480cells (50,000) were added to the top chamber in serum-free medium (100μl) and the bottom chamber was filled with medium containing10%FBS.FasL was used at12.5ng/ml and the FasL inhibitor (NOK-1), which were added into the culture medium2h before treatment, was used at10μg/ml.After72h incubation, the migration cellswere were fixed with paraformaldehyde and stained with DAPI. Tenrandom fields for each insert were counted. Inserts were conducted in triplicate in three separate experiments. ImmunoblotTotal protein was extracted in RIPA lysis buffer with protease and phosphotase inhibitors. Protein quantification was determined using the BCA method. Protein (30μg) was resolved by4-20%SDS polyacrylamide gel electrophoresisand transferred to PVDF membrane. Membranes were incubated with various antibodies in specific concentrations according to the manufacturer’s instructions. Antibodies against the following protiens were used for the study:E-cadherin,Occludin,Villin,N-cadherin,Vimentin,Snaill,MMP9, Smad3, ERK, Akt, GSK3-β,p-ERK, p-Akt, p-GSK3-P, Snaill, Rac1, CDC42, MAPK3, and GADPH. Protein expression was detected by ECL. MiRNA microarrayTotal cellular RNA was isolated using TRIzol reagents (Invitrogen, Carlsbad, CA) and miRNeasy mini kit (QIAGEN, Valencia, CA) as instructed by the manufacturers. RNA levels, quality and purity were assessed by nanodrop spectrophotometry (ND-1000, Nanodrop Technologies) and standard denaturing agarose gel electrophoresis. The samples were labeled using the miRCURYTM Hy3TM/Hy5TM Power labeling kit (Exiqon, Vedbaek, Denmark). Microarray analysis was performed by applying the miRCURYTM LNA Array platform (v.18.0)(Exiqon) containing Tm-normalized capture probes for3100perfectly matched miRNAs annotated in miRBase18.0and additional capture probes for25miRPlusTM human miRNAs. Microarrays with labeled samples were hybridized for16-20h at56℃in a12-Bay Hybridization Systems (Hybridization System-Nimblegen Systems, Madison, WI) and washed using a commercial wash buffer kit (Exiqon). Two biological replicates were done for each treatment. The slides were then scanned using the Axon GenePix4000B microarray scanner (Axon Instruments, Foster City, CA) and images were analyzed using GenePix Pro6.0software (Axon). The miRNA replicates were averaged and miRNAs with intensities>=30were chosen for calculating normalization factor. Differentially expressed miRNAs with at least1.5-fold changes (either greater of less) and statiscal difference (P<0.05) between FasL-treated cells and control cells were identified. Hierarchical clustering was performed using the MEV software (v4.6, TIGR). Microarray experiments were performed by KangChen Bio-tech, Shanghai, China. Functional annotation and molecular network analysisMiRNA targets were predictied using the computer-aid algorithms Tarbase, miRecord, TargetScan PicTar4way, and PicTar5way. Only target genes that were predicted by at least three algorithms were accepted as potential targets of a selected miRNA.Functional annotation of the idnetified miRNA targets was done using DAVID6.7. KEGG molecular pathway analysis was also performed to identify possible enrichment of genes with specific biological themes.Quantitative Real-time RT-PCR.Total cellular RNA was isolated using miRNeasy Mini Kit as intructed by the manufacturer (Qiagen). Quantitative real-time-RT-PCR (qRT-PCR) of miRNA was performed with the Hairpin-itTM miRNAs qPCR kit(Gemma pharmaceutical technology) using Mx3005P (Stratagene). qRT-PCR of MiRNA targets mRNA was performed with the S YBR Premix Ex TaqTM II kit(TaKaRa) using ABI PRISM7500Real-Time PCR System. Primers used in the study were pruchased from GeneCopoeia. The2-△△Ct method was used to calculate relative gene expression levels. U6-snRNA was used as a reference for normalization of miRNAs.GAPDH was used as a reference for normalization of mRNAs. Statistical analysisExperiments were done in triplicates. Statistical analysis was conducted using PASW Statistics18.0. P-values less than0.05were considered statistically significantInthe cell migration assay and qRT-PCR, the Factorial analysis was applied by SNK method. Test of homogeneity of variance was applied by Levene method. Analysis of Variance was applied by ANOVA method or Welch method. Multiple comparisons was applied by LSD method or Dunnett T3method.In the MiRNA microarray, differences were statistically evaluated by Student’s t test.ResultsFas and mFasL expression in GI cancer cell linesGastric cancer cell lines AGS, BGC-823and SGC-7901, colon cancer cell line SW480, DLD1, SW1116, HT-29, LoVo and SW620, pancreatic cancer cell line SW1990, BxPC-3, hepatocellular carcinoma cell line HepG2, SMC7721and esophageal cancer cell lines Eca-109strongly express Fas receptor (>50%), not express mFasL ligand (<10%).Low dose of sFasL promotes proliferation in AGS and SW480cellsCompared with the cell growthof control group, low dose sFasL (12.5ng/ml) does not affect AGS and SW480cells (P=0.575, P=0.575), while the middledose (25ng/ml) orhigh dose (50ng/ml) of sFasLhave a significant inhibitory effect on AGS and SW480cells, the difference was statistically significant (P<0.001).Fas signaling promotes migration in AGS and SW480cellsLow-dose FasL (12.5ng/ml) treatment promotes migration in AGS and SW480(P<0.001). In addition, pretreatment with NOK-1inhibits the FasL-induced motility of AGS and SW480(P=0.359and P=0.681, respectivly). Fas signaling promotes transition from epithelial to mesenchymal phenotype in GI cancer cell linesAfter low dose FasL (12.5ng/uL) treatment for3days,AGS, BGC-823, SW480, SW1116, SW1990, and Eca-109changed from the epithelial phenotype to the mesenchymal phenotype. Fas signaling induces EMT in GI cancer cell linesAfter low dose FasL (12.5ng/uL) treatment, the immunoblot showed that Fas signaling inhibits epithelial markers (E-cadherin, Occludin, Villin) and enhances mesenchymal markers (N-cadherin, Vimentin, and MMP9) in AGS, SW480, BGC-823, BxPC-3and SW1990cells.Fas signaling activates ERK1/2to induce EMTWe found that ERK1/2was activated in AGS and SW480cells after FasL stimulation. Cells were then pretreated with U0126(10μM) before FasL treatment to inhibit ERK1/2activation. Immunoblot results show that ERK1/2inhibition suppresses Fas-induced EMT.Fas signaling induces broad changes in the miRNA profile in AGS and SW480cellsCompared to the contro, approximately one third (35%,175/500) of the human mature miRNAs exhibited at least a1.5-fold difference in sFasL treated AGS cells. The majority (86.9%;150/175) of these miRNAs was downregulatedand, and only25(13.1%) miRNAs were upregulated. In SW480cells,24%(120/500) miRNAs exhibited at least a1.5-fold difference,33.3%(40/120) of these miRNAs was downregulatedand, and66.7%(80/120)miRNAs were upregulated.Functional annotation and molecular network analysis of miRNA targetsGO function classification indicates that the miRNA target genes function is focused on the regulation of transcription and regulation of biosynthetic process in AGS and SW480cells.KEGG analysis showed that the miRNA target genes were mainly involved in the Adherens junction pathway.Fas signaling specifically downregulates the expression of miR-23a, miR-92a, miR-138, and miR-155in AGS and SW480cellsIn the adherens junction pathway, using miRNA-based chips and preliminary bioinformatics evaluation, we identified4miRNAs, miR-23a, miR-92a, miR-138, and miR-155might involved in the Fas signaling induced EMT.These miRNAs were significant downregulated in AGS and SW480cells treated with FasL. In the qRT-PCR analysis, the FasL markedly suppressed the expression of miR-23a, miR-92a, miR-138, and miR-155in AGS cells at6and12hours (P<0.05). This inhibitory effect, however, was significantly attenuated by pretreatment with NOK-1(P<0.05).Target genes expression in AGS cellsThe Smad3,Snaill,Rac1,Cdc42, and MAPK3are potential target genes of miR-23a, miR-92a, miR-138, and miR-155. In the qRT-PCR analysis, FasL causes a2to4fold increase in the mRNA transcript levels of the above five target genes in AGS cells. Consistenly, the proteins of these target genes were also significantly increased compared with the control cells.ConclusionThere was moderate Fas expression but undetectable mFasL expression in a majority of GI cancer cell lines. In these cell lines, low doses (12.5ng/uL)sFasL treatment promotes transition from epithelial to mesenchymal phenotype, and in inhibits epithelial markers and enhances mesenchymal markers. Meanwhile, Fas signaling might activates ERK1/2to induce the EMT process.Fas signaling induces broad changes in the miRNA profile in AGS and SW480cells, and the target genes is focused on the regulation of transcription and biosynthetic process, and mainly involved in the Adherens junction pathway. Furthermore, we identified four miRNAs and five potential target genes, which might involved in the cell-cell adhesion pathway regulation and the EMT process.Thus, our findings could provide a new linkage between regulation of molecules and Fas signaling activation as well as provide new clues for understanding molecular mechanism of Fas signaling regulation.更多还原