The Roles And Prognostic Significance Of Matrix Metalloproteinases Family Gene And MMPs-related Long Non-coding RNA In Laryngeal Squamous Cell Carcinoma

Laryngeal cancer was the second most prevalent tumor in otorhinolaryngology-head and neck surgery,with the squamous cell carcinoma subtype accounting for 95% of larynx cancer.With the implementation of smoking bans,the total number of new cases of LSCC had slightly reduced over the past three years.However,the incidence remains high,accounting for approximately 1% of all cancers.The causes of laryngeal cancer were complex and multifaceted.Besides,the significance of early clinical symptoms remains unclear,and survival rates at later stages were still low.Therefore,it is urgent to elucidate the pathogenesis of laryngeal cancer and metastatic carcinoma in particular,and it is necessary to explore novel therapeutic targets and effective prognostic biomarkers of laryngeal cancer.Matrix metalloproteinases(MMPs)family were highly conserved zinc-dependent endopeptidases and primary proteases participating in the degradation of extracellular matrix(ECM)and basement membranes.Moreover,it predominantly participates in tumor invasion and metastasis.The majority of MMPs were secreted as inactive zymogens that are activated extracellularly and play essential roles in various physiological and pathological processes.A total of 25 MMP genes in public database have been described in humans and classified into six types based on the structural domain composition and substrate specificity.Among the top 10 most up-regulated genes in head and neck squamous cell carcinoma(HNSCC)cancer tissues when compared with normal tissues,six genes belonged to the MMPs.The results suggested that the expression levels of MMPs might play a vital role in the onset and course of progression of HNSC.Therefore,it was valuable to search for new risk markers and potential therapeutic targets of LSCC according to the transcriptional profiling analysis,prognostic association analysis and mechanism research of MMPs family in LSCC.Long non-coding RNAs(Lnc RNAs)were longer than 200 nt and constitute a large quantity of mammalian transcriptomes.In humans,more than half of the Lnc RNAs adjacent to or overlapped with the location of protein-coding genes.Because the oligonucleotide sequence conservation of Lnc RNAs was relatively low,Lnc RNA-related researches in vitro and in vivo were difficult to translate into human trials.Although many Lnc RNAs might be transcriptional noise,a growing number of Lnc RNAs had been reported to have biological functions.For example,regulatory gene expression,defending the genome against foreign DNA invasion,direct DNA synthesis or genomic rearrangements.Lnc RNAs had been shown to be crucial player in different diseases,especially in cancer.Some Lnc RNAs had already been validated as oncogenes or tumor suppressor genes in LSCC,the aberrant expression of these Lnc RNAs had been associated with LSCC development and progression and was closely related to poor prognosis of LSCC.Therefore,in-depth study of Lnc RNAs expression,clinical significance and the associated regulatory mechanism,identifying potential therapeutic targets and biomarkers for LSCC was of great importance.In this study,we analyzed the differentially expressed MMP family members and Lnc RNAs in LSCC of GEO and TCGA cohorts and searched for tissue-specific genes of LSCC.Given than the TCGA database included complete gene expression profiles and clinical information of LSCC.Univariate and multivariate Cox regression analyses of TCGA-LSCC data were performed to develop Lnc RNAs-related prognostic model and MMPs-related prognostic model.Furthermore,we screened the relevant key genes MMP-1 and LINC00278 on the Y Chromosome from the prognostic model.We then performed MTS assay,clone formation assay,transwell assay and scratch assay to evaluate the cell proliferation and metastatic ability changes in LSCC cells by MMP-1 knockdown and overexpressing LINC00278.The results of the four parts are reported as follows:Part one The matrix metalloproteinase gene family: a significant prognostic gene lineage correlated with immune infiltrates in laryngeal squamous cell carcinomaObjective: The aim of the present study was to identify the key genes associated with the development of LSCC and elucidate its underlying mechanism.We then selected key gene MMP-1 for functional validation and downstream analysis.Methods:1.RNA-seq data were identified using the ‘edge R’ package and Microarray datasets were analyzed using R package ‘Limma’ in order to obtain DEGs.The six datasets were integrated and analyzed by the Robust Rank Aggreg(RRA)method to obtain DEGs.The common differential genes were obtained by intersecting the differential genes between the public databases and our own microarray dataset.2.Univariate and multivariate Cox regression analyses were performed with the ‘survival’ and ‘survminer’ packages.Receiver operating characteristic(ROC)curves were plotted by the ‘survival ROC’ package.Additionally,we compared the area under curve(AUC)for each ROC curve to assess the MMPs-related prognostic model performance.Cox regression analysis was performed to determine independent prognostic factors.Finally,we utilized the ‘rms R’ package to consolidate the risk score and clinical characteristics for nomogram construction.3.Risk-score-based DEGs and MMPs family genes were evaluated via the WGCNA following the tutorial on the official website.The networks were visualized by Cytoscape 3.7.2.4.Stromal scores,immune scores and the tumor purity of LSCC patients were calculated with the ESTIMATE algorithm using the estimate package.The quantification of infiltrating immune cells in human LSCC has been performed by CIBERSORT method.5.The m RNA levels of MMP-1,MMP-3,MMP-8 and MMP-10 in different tumors were identified by the Oncomine database and GEPIA database.MMP-1,MMP-3 and MMP-10 were selected for the next quantitative real-time PCR(q RT-PCR)validation in 40 paired LSCC and adjacent nontumor tissues.The relationship between the expression of MMP-1and clinical parameters was also investigated in 40 paired LSCC patients.Based on the q RT-PCR results and follow-up data to verify the reliability of the model predictions.6.The in vitro functional assays,including cell proliferation and invasion,were performed following si RNA transfections for 24?h.7.The MMP-1 co-expression genes were used for KEGG pathway enrichment with the ‘Cluster Profiler’ package in R software.The patient samples were divided into two cohorts according to the median expression of MMP-1(high vs.low expression)and the tumour immune microenvironment between the two groups were compared by the Wilcoxon rank-sum test.Results:1.Seven common DEGs in the intersection of the four databases mentioned above were identified,including MMP-1,MMP-3,MMP-9,MMP-10,MMP-11,MMP-12,and MMP-13.2.Multivariate Cox regression was used to establish a four-gene(MMP-1/-3/-8/-10)prognostic model.The AUC of the four-gene(MMP-1/-3/-8/-10)based model(area under red=0.641)was higher than the model based on TNM stages(area under blue curve=0.536),the model based on grade(area under green =0.523),and the model based on gender(area under yellow=0.374).The four-gene(MMP-1/-3/-8/-10)based model(P=0.015)was significantly associated with prognosis,dominated independent prognostic factor for overall survival.3.WGCNA analysis showed that the four prognostic model genes(MMP-1,MMP-3,MMP-8 and MMP-10)did not exist in isolation but instead was a complex interconnected network and MMP-1 or MMP-10 might fulfill pivotal roles in LSCC patients.4.The high-risk group showed a higher stromal component and lower tumor purity than those in the low-risk group.Our results show that the more advanced the risk,the higher the immune score will be,despite being no statistical significance.Additionally,five types of immune cells with differences in infiltration were detected between the two groups,such as plasma cells,CD8+ T cells,follicular helper T cells,resting NK cells,and M0 macrophages.5.The MMP-1,MMP-3 and MMP-10 expression,but not MMP-8 was the highest in head and neck squamous cell carcinoma(HNSC)than others cancers in the Oncomine and GEPIA dataset.The results demonstrated that MMP-3 and MMP-10 were significantly upregulated in 34/40,31/40 paired LSCC tissues.MMP-1 indicated significant upregulation(40/40)in paired LSCC tissues,and had the highest basal expression of LSCC samples in TCGA.Moreover,the TCGA database analysis indicated that MMP-1overexpression was correlated with lower progress-free survival and overall survival(OS)of patients with LSCC.Based on the q RT-PCR results and follow-up data,patients with high MMP-1 expression had a poorer OS than patients with low MMP-1 expression in our own dataset.Moreover,the validation set prediction correlation coefficient reached 0.847,and survival time was significantly different between groups(P=0.0042),indicating that the model had a good predictive ability.6.The relationship between the expression of MMP-1 and clinical parameters revealed that expression level of MMP-1 was significantly associated with smoking(P<0.05),TNM stage(P<0.001),lymphatic metastasis(P<0.001)and pathological differentiation(P<0.01).However,no relationship was observed between the age,alcohol use and location of carcinoma.MMP-1 downregulation inhibited cell viability,colony formation and cell migration in TU686 and Fa Du cells.7.According to KEGG pathway enrichment analysis,the MMP-1co-expression genes were significantly associated with PI3K-Akt signaling pathways.Moreover,the prognostic model was related to na?ve B cells,memory B cells,CD8+ T cells,follicular helper T cells,Tregs,resting NK cells,monocytes,M0 macrophages,activated dendritic cells and activated mast cells infiltration.Part two The role and potential mechanisms of long,non-coding RNA-LINC00278 in laryngeal squamous cell carcinomaObjective: The present study was to construct a prognostic risk assessment model of LSCC based on the expression of these Lnc RNAs and took LINC00278 on the Y Chromosome(LINC00278 was also a MMPs-related Lnc RNA)as a starting point to explore its expression in LSCC and its role in the process of proliferation,migration and invasion,providing a strong theoretical basis for mining potential diagnostic and therapeutic targets of LSCC.Methods:1.Gene differential expression analysis was performed using the R package ‘Edge R’.Univariate and multivariate Cox regression analysis were established in construction of prognostic model based on expression profiles and clinical information of LSCC in TCGA.ROC curves were plotted by the ‘survival ROC’ package.Additionally,we compared the AUC for each ROC curve to assess the Lnc RNAs-related prognostic model performance.Then the expression level of LINC00278 was validated with multiple datasets from the Gene Expression Omnibus(GEO)database.2.LINC00278 was selected for the next q RT-PCR validation in 72 paired LSCC and adjacent nontumor tissues.The relationship between the expression of LINC00278 and clinical parameters was also investigated LSCC patients.The follow-up information was analyzed by the Kaplan-Meier survival analysis with the log-rank test.3.The effects of overexpression LINC00278 on the proliferation,migration and invasion of cancer cells were verified by MTS,colony formation,wound healing and Transwell invasion assays.AMC-HN-8 cells were injected subcutaneously into the lower right flank of BALB/c nude mice to establish the tumor xenograft model.The expressions of EMT-related proteins were assessed by western blotting assays.4.The h TFtarget database was used to predict the upstream transcription factors of LINC00278 and the binding sites.The chromatin immunoprecipitation(Ch IP)and dual luciferase reporter assays were applied to demonstrate the binding of ETS proto-oncogene 1,transcription factor(ETS1)and LINC00278 promoter region.Results:1.A total of 790 differentially expressed Lnc RNAs were identified via screening,including 665 upregulated and 125 downregulated Lnc RNAs.Multivariate Cox regression was used to establish a six-gene(LINC00278,MYHAS,MNX1-AS1,LINC02086,LSAMP-AS1 and CASC20)prognostic model.2.The AUC of the six-gene based model(area under red=0.832)was higher than the model based on TNM stages(area under yellow curve=0.614),the model based on grade(area under green =0.633),and the model based on gender(area under orange=0.485),the model based on alcohol(area under blue=0.421),and the model based on cigarettes(area under pink=0.511).3.Among the 6 lnc RNA,only LINC00278 was low expression and CASC20 was high expression both in TCGA and GEO database.After comprehensive consideration,we thus selected LINC00278 gene(chromosome Yp11.2)as the research target in this study(LINC00278 was also a MMPs-related Lnc RNA).4.The expression levels of LINC00278 were markedly decreased in LSCC tissues compared with adjacent nontumor tissues.The relationship between the expression of LINC00278 and clinical parameters revealed that expression level of LINC00278 was significantly associated with TNM stage(P<0.001),lymphatic metastasis(P<0.01)and pathological differentiation(P<0.01).However,no relationship was observed between the age,smoking,alcohol use and location of carcinoma.Moreover,patients with low LINC00278 expression had a poorer OS than patients with high LINC00278 expression.5.The overexpression of LINC00278 significantly reduced LSCC cell proliferation,migration and invasion ability.The mean tumor weight and volume were markedly reduced in the LINC00278 overexpression group compared with those in the empty vector group.Compared with the pc DNA3.1 group,the protein expression level of E-cadherin was increased,while the protein expression level of N-cadherin,Vimentin,Zeb1 and Snail were down-regulated in pc DNA3.1-LINC00278 group.6.The dual luciferase reporter assays revealed that the promoter region 1(the first exon)was the most highly active,which was further validated using a Ch IP assay.Conclusion:1.Seven common DEGs in the intersection of the four databases mentioned above were identified.We then discovered a signature of four genes(MMP-1,MMP-3,MMP-8 and MMP-10)from 25 MMP genes family and the MMP-related prognostic model had a good predictive ability.These signature genes were predictive of clinical outcomes of LSCC and significantly associated with tumor immune.2.MMP-1 indicated significant upregulation in paired LSCC tissues with consideration of tissue specific expression pattern.Patients with high MMP-1expression had a poorer OS than patients with low MMP-1 expression.Knockdown MMP-1 significantly impaired the proliferation,migration and invasion capacities of LSCC cells,which may be associated with PI3K-AKT signaling and tumor microenvironment.3.Lnc RNAs-related prognostic model(LINC00278,MYHAS,MNX1-AS1,LINC02086,LSAMP-AS1 and CASC20)showed excellent predictive ability.Among the 6 lnc RNA,only LINC00278 was low expression and CASC20 was high expression both in TCGA and GEO database.4.The expression levels of LINC00278 were markedly decreased in LSCC tissues compared with adjacent nontumor tissues.Patients with low LINC00278 expression had a poorer OS than patients with high LINC00278 expression.LINC00278 on the Y Chromosome was upregulated by ETS1,inhibited LSCC growth in vivo and in vitro.