Establishment And Analysis Of A Prognostic Model Of Lung Adenocarcinoma Using Tumor Metastasis-related LncRNAs Markers And Its Relevant Molecular Mechanism

OBJECTIVE:Local relapses and metastases are the leading causes of death in patients with lung cancer.Metastasis-associated lncRNAs are closely related to the relapse and prognosis of patients with lung cancer.In the present study,by analyzing high-throughput data from The Cancer Genome Atlas(TCGA),we aimed to screen and identify lncRNA molecular markers related to metastasis and recurrence free survival(RFS)of lung adenocarcinoma(LUAD),and then develop a relapse prognosis assessment model of LUAD.Moreover,we also verified the roles of the identified molecules in the lncRNA panel in the metastasis of LUAD cells by using the in vitro migration and invasion experiments.METHODS:1.Preliminary screening of lncRNAs related to metastasis of LUAD.Data for selected samples of 347 LUAD patients with complete lncRNAs data,clinicopathological characteristics and follow-up information(records of recurrence),were downloaded from the TCGA database(https://cancergenome.nih.gov/).The differentially expressed lncRNAs between LUAD patients with(n=158)and without metastasis(n=189)were identified by EdgeR in R language of |logFC|>1 and FDR<0.05.2.Construction of a prognostic lncRNA panel in LUAD.The patients were randomly divided the patients into a training set(n=231)and a testing set(n=116).In the training set,all the above-mentioned lncRNA molecules related to metastasis were included in the univariate Cox proportional hazards regression model analysis,and the correlation between the lncRNA expression and the patient's RFS was analyzed.The lncRNAs with P values less than 0.05 were selected.The selected lncRNAs were included in the stepwise multivariate Cox proportional hazards regression model analysis,and each IncRNA molecule was weighted according to the regression coefficient to construct an IncRNA panel for predicting the prognosis of LUAD patients.Kaplan-Meier survival analysis and time-dependent ROC curve were used to evaluate the prognostic value of the IncRNA panel.3.Validation of the 'ncRNA panel.Either in the testing set or the entire set(including all subjects),we used the IncRNA panel obtained from the training set to calculate the risk index of LUAD patients,and then divided the patients into two groups according to the optimal cut-off value obtained from the training set,namely the high-risk group and low-risk group.Kaplan-Meier analysis and ROC curve were used to detect the prognostic value of lncRNA panel in both the testing set and entire set.4.Testing the independence of the lncRNA panel to predict the prognosis.The univariate and multivariate Cox proportional hazards regression model analysis and Kaplan-Meier stratified survival analysis were used to detect the independence of lncRNA panel to predict the prognosis of LUAD patients in the entire set5.Prognostic performance of the lncRNA panel compared to TNM staging and the combined model.The lncRNA panel was further combined with TNM staging with independent prognosis effect to build a combined model.The ability to predict the prognosis of LUAD was compared by using time-dependent ROC curve among lncRNA panel,combined model and TNM staging.6.The roles of the lncRNAs of the panel in LUAD metastasis.We used in vitro migration and invasion experiments to specifically study the effect of the key lncRNAs on the biological invasion and migration of LUAD cells.RESULTS:1.After differential lncRNAs analysis betweentwo groups of LUAD patients with and without metastasis,735 eligible lncRNAs were obtained.These significantly different lncRNAs were considered as candidate biomarkers for prognosis in LUAD patients.2.In the training set,all the 735 differential lncRNAs were included in the univariate cox proportional hazards regression model,and 43 lncRNAs with P values less than 0.05 were screened.The 43 lncRNAs were analyzed by multivariate Cox proportional hazards regression model,and the IncRNA panel for the prognosis evaluation of LUAD patients was obtained:Risk score=(0.191404×LINC02323)+(0.256783×ZNF649.AS1)+(0.502202×HNF4A.AS1)+(-0.319007×LINC01819)+(-0.306173×FA M222A.AS1)+(-0.304856×LINC00672).The mean value of risk score was taken as the optimal cut-off value for the lncRNA panel to predict the prognosis,and then the patients were divided into the high-risk group(n=119)and low-risk group(n=112)according to the cut-off value.The Kaplan-Meier survival analysis curve showed that the RFS of the high-risk group was significantly lower compared with the low-risk group(log-rank test,P<0.001).The time-dependent ROC curve analysis showed that the 3-year AUC of the lncRNA panel was 0.6970(95%CI:0.6102-0.7821),and the 5-year AUC was 0.6981(95%CI:0.5866-0.8089).3.Kaplan-Meier survival analysis in the testing set showed that the RFS of patients in the high-risk group was significantly lower than that of the low-risk group(log-rank test,P<0.001).The time-dependent ROC curve analysis in the testing set showed that the 3-year AUC of the lncRNA panel was 0.7620(95%CI:0.5884-0.9459),and the 5-year AUC was 0.8421(95%CI:0.6534-0.9957).The results obtained in the entire set were consistent with the testing set.Kaplan-Meier survival analysis showed that the RFS of patients in the high-risk group was significantly lower than that of the low-risk group(log-rank test,P<0.001).The time-dependent ROC curve analysis showed that the 3-year AUC of the lncRNA panel was 0.7059(95%CI:0.6322-0.7801),and the 5-year AUC was 0.7307(95%CI:0.6368-0.8217).4.The univariate and multivariate Cox proportional hazards regression models in the entire set showed that TNM staging and lncRNA panel were independent prognostic factors for LUAD patients.Kaplan-Meier stratified survival analysis showed that the RFS of the lncRNA panel in the early group(TNM staging ?+?),the primary tumor status(T1+T2)group and regional lymph node status(N)group of patients with high risk was significantly lower compared with the low-risk patients(log-rank test,P<0.05).5.According to the prognostic abilities at the cut-off point of 3 years,the AUC of lncRNA panel in the entire set was 0.7059(95%CI:0.6322-0.7801),the AUC of TNM staging was 0.6678(95%CI:0.5769-0.7613),and the prognostic ability of lncRNA panel was significantly better compared with the TNM staging(P<0.05).At the cut-off point of 5 years,the AUC of lncRNA panel in the entire set was 0.7307(95%CI:0.6368-0.8217),and the AUC of TNM staging was 0.6111(95%CI 0.5044-0.7229),The prediction ability of IncRNA panel on the prognosis of LUAD was significantly better compared with the TNM staging(P<0.05).We used the lncRNA panel and TNM staging to constitute a combined model When the cutoff point was 3 years,the combined model AUC of the entire set was 0.7474(95%Cl:0.6745-0.8198).When the cutoff point was 5 years,the combined model AUC of the entire set was 0.7670(95%CI:0.6723-0.8564).The predictive ability of the combined model was significantly better compared with the TNM staging(P<0.05),while there was no significant difference in the predictive ability between the combined model and the single lncRNA panel.6.According to the results of migration and invasion experiments,all the six lncRNAs that made up the lncRNA panel had an effect on the migration and invasion abilities of LUAD cell lines,among which LINC02323 had the most significant ability to enhance the migration and invasion of LUAD cells.CONCLUSIONS:1.In this study,for the first time,we systematically screened lncRNA biomarkers related to the metastasis of LUAD from TCGA database.Furthermore,we performed the prognostic analysis and constructed a 6-lncRNA panel that could be served as a model for predicting tumour relapse and was an independent prognostic factor for LUAD patients.This panel had better prognostic value than TNM stage and may yield more powerful information in the clinical setting.2.The lncRNAs of this panel were related to not only tumor metastasis but also prognosis prediction of relapse in LUAD patients,which might aid in development of targeted therapies as a target in metastatic cancer.Part Two.The molecular mechanism of LINC02323 involved in OBJECTIVE:metastasis of lung adenocarcinomaBased on the encouraging results of the first part of this study,we aimed to further conduct a detailed functional analysis of the lncRNA panel.In this part,we focused on how LINC02323 was involved in TGF-?-induced EMT of LUAD cells.METHODS:1.Analysis of the clinical value of LINC02323 expression in various tumours.The expression of LINC02323 in multiple tumor tissues was analyzed by the starBase v2.0 software.Then,the correlation between the LINC02323 expression and prognosis of LUAD patients was studied by Kaplan-Meier survival analysis using TCGA database.2.Observation of the effect of LINC02323 expression on the biological behavior of LUAD cells.RT-qPCR was performed to assess the expression of LINC02323 in LUAD cell lines,and over-expression or RNA interference(RNAi)was conducted according to the low or high expression levels of LINC02323 in various cells,respectively.The distribution of LINC02323 in cells was detected by the nuclear-plasma separation experiment.The MTT proliferation assay was used to detect the proliferation,and transwell assays were applied to examine the role of LINC02323 on migration and invasion of LUAD cells.3.Investigation of molecular mechanisms of LINC02323 in the EMT of LUAD cells induced by TGF-?.The EMT of LUAD cells were induced by TGF-? stimulation.The expression of EMT markers and LINC02323 were detected by RT-qPCR or Western blot,respectively.RNAi or overexpression of LINC02323 was used to detect the role of LINC02323 in EMT of LUAD cells regulated by TGF-? using RT-qPCR and Western blot methods.StarBase v2.0 and miRDB software were used to predict the possible microRNAs interacting with LINC02323 and the downstream target genes,respectively.Pearson's correlation analysis was performed to analyze the correlation between LINC02323 and relevant microRNA levels using transcriptome expression profile data from TCGA.Then,the results were verified by a dual-luciferase reporter system and qRT-PCR,respectively.Finally,through rescue experiments,the role of LINC02323 as ceRNA in the EMT and metastasis of LUAD cells was elucidated by interfering with the expression of key molecules,RT-qPCR,Western blot and Transwell assays.RESULTS:1.Compared with corresponding normal tissues,the expression of LINC02323 in multiple tumor tissues,including LUAD,lung squamous cell carcinoma,bladder transitional cell carcinoma,pancreatic cancer,hepatocellular carcinoma,renal papillary cell,colon cancer and bile duct carcinoma,was significantly higher(all P<0.05).Besides,the high expression of LINC02323 was correlated with the poor prognosis of LU AD patients.Kaplan-Meier survival analysis showed that the patient RFS was significantly different between high-risk and low-risk LUAD patients(log-rank test,P<0.001).2.Nuclear separation experiments showed that LINC02323 was distributed in the cytoplasm and nucleus,with a ratio of 44.49%and 55.51%,respectively.Compared with the control group,the over-expression or depletion of LINC02323 did not affect the proliferation activity of LUAD cells(all P>0.05).However,according to the results of the Transwell cell migration and invasion experiment,the over-expression or depletion of of LINC02323 significantly affected the migration and invasion of LUAD cells(all P<0.05).3.TGF-? induced EMT in LUAD cells and up-regulated the expression of LINC02323(P<0.001)at the same time.TGF-? induced EMT process was significantly affected by both RNAi and overexpression of LINC02323.The predicted results showed there are binding sites between LINC02323 and miR-1343-3p,and miR-1343-3p could directly bind to TGFBR1-3'-UTR.The expression of LINC02323 was found to be negatively correlated with miR-1343-3p in LUAD by analyzing TCGA data(P<0.05).The double luciferase reporting system and RT-qPCR experiments confirmed that LINC02323 could bind to miR-1343-3p,miR-1343-3p could bind to TGFBR1-3'-UTR.Inhibition of miR-1343-3p reversed LINC02323 silence mediated suppressing migration,invasion and EMT,indicating LINC02323 acts as a ceRNA sponging miR-1343-3p to promote the EMT and metastasis in LUAD via targeting TGFBR1.CONCLUSIONS:1.Our study provided,for the first time,evidence for the high expression of LINC02323 in various tumours,including LUAD,and its poor prognosis in LUAD paitents.These findings suggest that it may act as an oncogene to promote cancer progression.2.LINC02323 was involved in the EMT process of tumor cells via regulating the miR-1343-3p-TGFBR1 axis in LUAD cells,which indicated that it was a key molecule in the process of invasion and metastasis of LUAD and may be used as a potential target in metastatic cancer.