| Lung cancer is the leading cause of cancer-related death worldwide.Therefore,investagating the molecular mechanisms underlying progression,and precisely distinguishing squamous cell carcinoma(SCC)from adenocarcinoma(AC)of non-small cell lung cancer(NSCLC),are clinically important for the diagnosis and therapy for the malignancy.MicroRNAs(miRNAs)are endogenous non-coding RNAs that finely regulate gene expression.MiRNAs can serve as oncogenes or tumour suppressors involved in tumor progression and metastasis of NSCLC.Furthermore,mi RNAs can be used as useful biomarker for tumor detection,prognosis and therapeutic effectiveness,suggested that miRNA is an important molecular mechanisms related to cancer development and progression.Using next-generation deep sequencing to comprehensively characterize and compare miRNA profles of surgically resected SCC and AC tissues,we recently identifed mi R-944 exhibited distinctive expression levels in SCC versus AC.In addition,miR-944 is located in the introns of host genes p63 and is called intronic miR-944.However,the role of miR-944 in NSCLC cells,the relationships between mi R-944 and its host gene p63 expression and the significance of miR-944 for discriminating SCC from AC are still unknown.Here we focus on the above questions to develop our study,three parts are involved: 1 Investigating the biological effects of mi R-944 in NSCLC cells;2 Exploring the relationship between intronic mi R-944 and its host gene p63 expression and the regulatory mechanisms of intronic miR-944 on p63 in NSCLC;3 Developing miRNA biomarkers for accurately differentiating SCC from AC in surgical tissues and bronchoalveolar lavage(BAL).Our results may throw new lights on the pathogenesis of lung SCC,and identify specific miRNA biomarker of SCC to accurate subclassification of NSCLC.Part one: The biological effects of miR-944 in non-small cell lung cancer cells in vitroObjective: To investigate the effects of miR-944 on cell proliferation,migration,apoptosis and cell cycle distribution in NSCLC cells in vitro.Methods: Adenocarcinoma A549 cell,squamous cell carcinoma H1703,H520 and SK-MES-1cell,and cutaneous squamous cell carcinoma A431 cells(tipical SCC cell),were cultured in vitro.Cells were transfected with miR-944 mimics or inhibitors,respectively,and the efficiency of transfection was confirmed by Taqman microRNA assay.MTT assay and Real-Time Cellular Analysis(RTCA)was used to detected cell proliferation.Wound healing assay,Transwell and RTCA method was used to determine the ability of cell migration.Furthermore,cell apoptosis and cell cycle distribuation were measured by FCM assay.All experiments were repeated three times,and data were analyzed using SPSS Statistics17.0 statistical software,P<0.05 was considered statistically significant.Results:1.The efficiency of miR-944 mimics or inhibitors transfectionCompared with control group,mi R-944 level was significantly incresased in A549,H1703 and H520 cells after transfected with mi R-944 mimics,respectively(P<0.05).Conversely,miR-944 expression was inhibited obviously in SK-MES-1 cells transfected with miR-944 inhibitors than that of control group(P<0.05).2.Effects of miR-944 on cell proliferationMTT assay showed that overexpression of mi R-944 increased A431 cell proliferation compared with control cells at 24-96 h post-transfection(All P <0.05),respectively.In A549 cells,miR-944 overexpression could also increase cell proliferation at 48-96 h post-transfection(P <0.05).In addition,H1703 cells showed higher proliferation activity 72 or 96 h after miR-944 mimcs transfection(P <0.05).RTCA detection results are consistent with the MTT method,which ensures the repeated ability of the experimental results.In contrast,MTT and RTCA assays indicated that downregulation of miR-944 decreased SK-MES-1 cell proliferation at 48 and 72 h post-transfection,respectively,compared with control cells(P <0.05).3.Effects of miR-944 on cell apoptosisThe apoptosis rate of A549 cells transfected with miR-944 mimics was 7.93% ± 0.89%,which was significantly lower than that of control group(15.6% ± 0.93%,P <0.05).A431 cells apoptotic rate was significantly lower in miR-944 mimics group than that in control cells(14.2% ± 1.02% V.S.9.3% ± 0.83%,p<0.05).While in H1703 and H520 cells,no differences were found in cell apoptosis by miR-944 overexpression.In addition,the apoptosis rate of SK-MES-1 cells transfected with miR-944 inhibitor was 7.93% ± 0.89%,which was significantly lower than that of control group(14.3% ± 0.98%,P <0.05).4.Effects of miR-944 on cell migrationMonolayer wound healing assay showed that the speed which cells migrated towards the scratch was higher in A549,H1703 or A431-miR-944 mimics cells when compared with the control cells(P<0.05).Transwell assay further indicated that the number of A549,H1703 or A431-miR-944 mimics cells migrated into the lower chamber was significantly higher than that in the control cells(P<0.05),respectively.Additionally,RTCA method also confirmed that cell number were all increased in A549,H1703 or A431-miR-944 mimics cells compared to control group(P<0.05).Taken together,the observations suggested that overexpression of miR-944 promote cell migration ability.In contrast,inhibition of mi R-944 expression significantly inhibited SK-MES-1 cell migration ability by Monolayer wound healing,Transwell as well as RTCA assays(all P<0.05).5.Effect of miR-944 on cell cycle distributionFCM results showed that up or downregulation of miR-944 had no significant effect on cell cycle distribution.Summary:1.MiR-944 contributed to the process of cell proliferation.Overexpression of miR-944 promoted A549,H1703 and A431 cell proliferation,while knockdown of miR-944 inhibited SK-MES-1 cell proliferation.2.MiR-944 was involved in the regulation of cell migration.Overexpression of miR-944 could increase the migration ability of A549,H1703 and A431 cells,while knockdown of mi R-944 decreased the migration ability of SK-MES-1 cells.3.Overexpression or knockdown of miR-944 could inhibit or promote cell apoptosis,but has no significant effect on cell cycle distribuation.All toghter,miR-944 might function as oncogene and participates in the process of cell proliferation,apoptosis and cell migration of NSCLC cells.Part two: The relationships between miR-944 and its Host gene p63 expression in NSCLCObjective: To clarify the effects between intronic miR-944 and its Host gene p63 at transcription and expression level,and to explore the possible mechanism underlying intronic miR-944 on p63 expression.Methods: NSCLC cells and A43 cell were cultured in vitro;Taqman microRNA assay was used to detecte miR-944 expression after p63 was induced by TGF-? treatment,or inhibited by siRNA transfection;Real-time PCR and Western Blot assays were used to detected p63 expression after miR-944 overexpression.Furthermore,p63 expresson was detected in cells transfected with YAP1 or TAZ si RNA,and YAP1 or TAZ expression was determined after miR-944 overexpression.In addition,a total of 90 paraffin-embedded specimens with primary NSCLC who had not undergone radiotherapy and chemotherapy prior to surgery were collected.Immunohistochemistry method was used to detect YAP1 or p63 protein expression.The correlation between miR-944,YAP1 and p63 was analyzed further.SPSS Statistics 17.0 software was used for statistical analysis.P<0.05 was considered statistically significant.Results:1.The regulation effect of p63 on mi R-944 expressionAccording to the literature,p63 expression could be induced by TGF-? treatment in A431 cells.We did confirm that p63 expression at mRNA and protein levels were both increased in A431 cells treated with TGF-? for 20 h.Simultaneously,mi R-944 expression was increased significantly accompanied by p63 expression(P<0.05).In addition,miR-944 expression was significantly inhibited in A431 and SK-MES-1 cells after p63 siRNA transfection(P<0.05).The above results suggested that p63 can regulate the expression of miR-944.2.The effect of miR-944 on the expression of its Host gene p63Overexpression of miR-944 in A549,H520 and A431 cells can increase the expression of p63 both at mRNA and protein level,compared with each control group(all P<0.05).On the contrary,knockdown of miR-944 in SK-MES-1 cells significantly decreased p63 expression at mRNA and protein level(P<0.05),suggested that mi R-944 can regulate the expression of Host gene p63.3.Effect of YAP1/TAZ on p63 expressionThe expression of YAP1 or TAZ at mRNA and protein level was significantly decreased in A549,H520 and A431 cells after YAP1 or TAZ si RNA transfection.Meanwhile,p63 expression was increased significantly in all the three cells both at mRNA and protein levels(P<0.05).The above results suggested that YAP1 or TAZ can regulate the expression of p63 negatively.4.Effect of miR-944 on the expression of YAP1/TAZYAP1 and TAZ expression was further detected in A549,H520 and A431 cells after miR-944 mimcs transfection,respectively.In A549 cells,overexpression of mi R-944 inhibited both YAP1 and TAZ expression at mRNA and protein level compared to control group(P<0.05).In H520 cells,YAP1 expression at mRNA and protein level,as well as TAZ mRNA expression were decreased significantly(P<0.05).Additionally,overexpression of miR-944 inhibited TAZ mRNA expression in A431 cells.Taken together,mi R-944 can inhibit YAP1/TAZ expression in the above cells to certain extent.5.The expression of YAP1 and p63 in human NSCLC tissues and the correlation with clinical pathological parametersImmunohistochemistry test showed that the positive staining of YAP1 was expressed in 77% out of 90 NSCLCs cases(69/90),and half of the samples had high YAP staining(45/90).Further analyses revealed that YAP high expression is significantly higher in AC than that in SCC(79% V.S.25%,P<0.05).The positive staining of p63 was expressed in 53% out of 90 NSCLCs cases(48/90),and its expression in SCC was significantly higher than that in AC(100% V.S.0%,P<0.05).Further analysis indicated that YAP1 and p63 expression was both correlated with gender,YAP1 expression was higher in female than that in male(P<0.05),while p63 expression was higher in male than that in female(P<0.05).In addition,p63 protein expression was positively correlated with tumor size(P<0.05).6.Correlation between miR-944,YAP1 and p63 expression in NSCLC tissuesThe expression levels of miR-944 was signifcantly higher in SCC compared with AC specimens(P<0.05).Spearman correlation analysis indicated that mi R-944 is positively correlated with p63 expression in 90 cases of NSCLCs(r=0.955,P<0.05),while YAP1 is negatively correlated with both p63 and miR-944(mi R-944 and YAP1: r=?0.535,P<0.05;p63 and YAP1: r=?0.535,P<0.05).Summary:1.miR-944 expression was increased accompany by p63 induction or decreased after p63 inhibition,indicated that p63 can positively regulate the expression of miR-944.2.Over or knockdown of mi R-944 expression significantly increased or decreased p63 expression at mRNA and protein level,indicated that miR-944 could positively regulate p63 expression in multiple cells.The results suggested that mi R-944 and its Host gene p63 could regulate each other through positive feedback loop.3.YAP1/TAZ can inhibit the expression of p63 gene,while overexpression of miR-944 can inhibit the expression of YAP1 or TAZ,thus relieving the inhibitory effect of YAP1/TAZ on p63 gene,which might be one of the mechanisms of miR-944 promoting the expression of its Host gene p63.4.The expression of YAP1 protein in lung adenocarcinoma was significantly higher than that in squamous cell carcinoma of the lung,while p63 protein expression was almost seen in squamous cell carcinoma,and was positively correlated with tumor size or gender.5.There was a significant positive correlation between miR-944 and p63 expression in NSCLC tissues,and both of them were negatively correlated with YAP1 expression.Part three: The value of mi R-944 in distinguishing squamous cell carcinoma from adenocarcinom of non-small cell lung cancerObjective: To screen and validate the significance of miR-944 and other miRNAs in distinguishing SCC from AC.Methods: A total of 128 fresh specimens,112 paraffin-embedded tissues,and 127 bronchial lavage fluid(BAL)specimens with primary NSCLC who had not undergone radiotherapy and chemotherapy prior to surgery were collected.Total RNA were extracted from the above samples according to the manufacturer's instructions,respectively.MiRNA expression in NSCLC tissues was detected using Taqman microRNA assay,and the expression of miRNA in BAL cells was detected by ddPCR.The receiver operating characteristic(ROC)curve and the area under the ROC curve(AUC)was used as an accuracy index for evaluating the diagnostic performance of the miRNAs.Stepwise logistic regression models was used to construct diagnostic biomarker panels,and then used the stepwise backward model selection to identify the best discriminating combinations of mi RNAs for classification of SCC from AC.The McNemar chisquared test was employed to determine the signifcant differences between cytology and biomarker panel.In addition,the expression of Pri-miR-944 in 11 SCC fresh frozen tissues was further detected by Taqman probe method to evaluate the transcriptional mature process of miR-944.P < 0.05 was considered statistically significant.Results:1.Developing a miRNA-based prediction model for distinguishing SCC from AC in surgical tumor tissue specimens:Four mi RNAs(mi Rs-944,205-5p,135a-5p,and 577)identifed in our previous study and three mi RNAs(miRs-21,34 a,and 375)identifed by others were detected in 128 frozen tumor tissues(62 SCC and 66 AC).Of the seven miRNAs,six(miRs-944,205-5p,135a-5p,577,34 a,and 375)displayed a significantly different level in SCC versus AC tumors(all P < 0.05).Among them,miRs-205-5p and 944 were the top 2 mi RNAs which showed the most significant.Subsequently,miRs-205-5p and 944 were selected in the model,which had an AUC of 0.988 for distinguishing SCC from AC tumors.The use of miRs-205-5p and 944 in combination generated an accuracy of 96.48% with 96.55% sensitivity and 96.43% specificity for differentiating SCC from AC.2.Validating the miRNA-based prediction model in an external cohort of FFPE specimensThe two mi RNAs defned from the above developmental phase were validated in an independent cohort of 112 FFPE specimens consisting of 57 SCC and 55 AC tissues.The expression levels of miRs-205-5p and 944 were signifcantly higher in SCC compared with AC specimens(All P<0.05).Furthermore,the prediction model(miRs-205-5p and 944)had an AUC of 0.986 for distinguishing SCC from AC tumors,and created 96.43% accuracy with 96.43% sensitivity and 96.43% specifcity,therefore confrming the ability for discriminating SCC from AC.3.Applying the prediction model in BAL specimens for differentiating SCC from AC of NSCLCsThe prediction model with the two miRNAs was further detected in 127 BAL samples consisting of 82 SCC and 45 AC.It showed that the two miRNAs generated an AUC of 0.997 for the discrimination of SCC from AC,and had 96.12% accuracy with 95.00% sensitivity and 96.83% specifcity for classifcation of SCC from AC.By contrast,cytology was successfully performed on only 58 of the 127 BAL samples.Therefore,cytological analysis of the BALs had a lower diagnostic rate(45.6%)compared with the miRNAs-based prediction model(P<0.05).4.Relationship between Pri-miR-944 and mature miR-944 expression:Using miR-205 as a reference,the relative level of Pri-miR-944 and mature mi R-944 expression were analyzed in 11 fresh frozen tumor SCC tissues.It showed that the level of mature miR-205-5p was significantly higher than that of Pri-mi R-205.However,there was no difference between the expression of mature mi R-944 and Pri-miR-944.Compared with the ratio of miR-205 relatived to Pri-mi R-205,the ratio of miR-944/Pri-miR-944 was significantly decreased in 11 SCC tissues(P <0.05).Summary1.The expression of miRs-205-5p,944,34 a,135a-5p,375 and 577 displayed a significantly different level in SCC versus AC tumors.Among them,miRs-205-5p and 944 were the highest miRNAs in lung SCC.2.Two mi RNAs-based modle(miRs-205-5p and 944)was useful biomarker for distinguishing SCC from AC both in frozen lung tumor tissues and FFPE specimens.3.In bronchial lavage fluid specimens,miRs-205-5p and 944 panel could not only increased the accuracy at diagnosing NSCLC,but also has a higher accuracy for discriminating SCC from AC compared with cytology.4.The efficiency of mature miR-944 generating from its primary miRNA was lower than that of miR-205-5p,which might be one of the reason related with the relatively lower level of miR-944 in NSCLCs.Conclusions:1.MiR-944 might have oncogenic functions contributing to the process of cell proliferation,apoptosis and cell migration of NSCLC cells and cutaneous squamous cell carcinoma cells.2.P63 could positively regulate the expression of miR-944.Conversely,miR-944 could also positively regulate p63 expression in multiple cells.Furthermore,there was a significant positive correlation between miR-944 and p63 expression in NSCLC tissues,suggested that miR-944 and its Host gene p63 could regulate each other through positive feedback loop.3.The expression of YAP1 in lung adenocarcinoma was significantly higher than that in lung squamous cell carcinoma,while p63 expression was almost seen in squamous cell carcinoma.There was a negative correlation between YAP1 and p63 expression in NSCLCs.In addition,YAP1/TAZ knockdown significantly increased p63 mRNA and protein levels in multiple cells in vitro,which suggested that YAP1/TAZ could inhibit p63 expression.4.MiR-944 can inhibit the expression of YAP1 or TAZ in several cell lines,thus relieving the inhibitory effect of YAP1/TAZ on p63 gene expression.miR-944 YAP1/TAZ p63 pathway is one of the novel mechanisms of Intronic miR-944 promoting the expression of its Host gene p63.5.The expressions of miR-205-5p and miR-944 were significantly higher in lung squamous cell carcinoma than that in lung adenocarcinoma.miR-205-5p and mi R-944 in combination was a useful biomarker for distinguishing SCC from AC both in frozen lung tumor tissues and FFPE specimens.Especially in bronchial lavage fluid specimens,miRs-205-5p and 944 panel could not only increase the accuracy at diagnosing NSCLC,but also has a higher accuracy for discriminating SCC from AC compared with cytology. |
PDF Full Text Request