Screening And Mechanism Study Of Long Non-coding RNA Associated With Prognosis In Triple-negative Breast Cancer

Objective: Breast cancer(BC)is one of the most common malignant tumors in women worldwide,and ranks second in the causes of cancer death.Triple-negative breast cancer(TNBC)patients accounts for about 15-20% of all breast cancer patients.TNBC is a rapidly proliferating and highly invasive subtype of breast cancer.It is prone to recurrence and metastasis in the early stage and has a poor prognosis.Since TNBC is highly heterogeneous at the molecular level,the prognosis of patients with different molecular subtypes is significantly discrepant.At present,TNBC still lacks specific biomarkers and therapeutic targets,so the choice of individualized treatment and clinical curative effect are limited.To explore the molecular basis in the progress of TNBC and find effective prognostic markers will be of great significance for improving the treatment present situation,reducing the risk of recurrence and metastasis,and prolonging survival time.Long non-coding RNAs(lncRNAs)are a group of RNAs with a length of more than 200 nucleotides.They are similar to mRNA in structure,but lack the function of coding proteins.LncRNAs are involved in many important biological phenomena,such as genomic imprinting,conformation formation of chromosomes and allosteric regulation of enzyme activities.Specific patterns of LncRNAs expression can coordinate cell status,differentiation and development.Accumulating evidences have showed that over-expression,deletion and mutation of lncRNA gene are closely related to many human diseases.LncRNAs are a group of important gene expression regulators,which can regulate the expression level of target genes at the epigenetic,transcriptional and post-transcriptional levels by interacting with DNA,RNA and protein.For example,lncRNAs can be used as signal molecules to promote transcription,as inducers to inhibit transcription,as regulators of epigenetics,or as scaffolds to interact with multiple protein chaperones and form ribonucleoproteins.In recent years,a series of abnormally expressed lncRNAs have been widely reported in different types of cancer.They can participate in the proliferation,apoptosis,invasion and metastasis of cancer cells through a variety of mechanisms.In addition,increasing studies have found that lncRNAs could play a key role in TNBC.Therefore,finding new targets from lncRNAs may be a promising treatment option for TNBC patients.Our previous studies used bioinformatics methods to screen the lncRNAs related to the prognosis of breast cancer.Through further external validation by Kaplan Meier Plotter and meta-analysis based on GEO breast cancer datasets,we found that lncRNA FOXD2-AS1 might be closely associated with the prognosis of TNBC.LncRNA FOXD2-AS1 is located on chromosome 1p33 with a transcript length of 2527 bp.The present studies have shown that FOXD2-AS1 is up-regulated in gastric cancer,colorectal cancer,lung cancer,thyroid cancer and liver cancer,and it can promote proliferation,migration,invasion and drug resistance of cancer cells through competitive endogenous RNA(ceRNA)mechanism or transcriptional regulation mechanism.However,there is no research on the expression and biological function of FOXD2-AS1 in breast cancer.The Part I of this study focused on the expression level,biological function and molecular mechanism of FOXD2-AS1 in TNBC cells.Prognostic assessment is critical for selecting the appropriate treatment options.In routine clinical practice,age,histological grade and TNM stage are the key factors to predict the prognosis of TNBC patients.However,since TNBC is a highly heterogeneous disease,even in TNBC patients with the same clinicopathological characteristics,there are significant differences in drug sensitivity and risk of recurrence and metastasis,which indicates that the traditional prognostic factors are not enough to evaluate the prognosis of TNBC patients comprehensively and accurately.In recent years,with the development of high-throughput sequencing technology,more and more genes and non-coding RNA have been identified to be the biomarkers for prognosis evaluation and treatment choice of cancer patients.In addition,compared with a single molecule,combining multiple molecules to construct a prognostic model of multi-molecular characteristic will have better predictive value.At present,there are few studies on the prognostic models of the multiple-lncRNA molecular characteristic related to the risk of recurrence and metastasis of TNBC patients.In the Part II of this study,a novel predictive model based on the 5-lncRNA molecular characteristic was established and validated by bioinformatic methods to predict recurrence-free survival(RFS)/distant metastasis-free survival(DMFS)of TNBC patients.Materials and Methods: Part ?: 1.Screening the lncRNAs related to the prognosis of breast cancer patients by bioinformatic methods,and validating the prognosis values of them by Kaplan-Meier Plotter.2.A meta-analysis was conducted to evaluate the relationship between FOXD2-AS1 expression and the prognosis and clinicopathological characteristics of patients with malignant tumors.3.A meta-analysis was conducted on the GEO datasets of breast cancer to evaluate the relationship between the expression of FOXD2-AS1 and the prognosis and clinicopathological characteristics of breast cancer patients.Kaplan-Meier Plotter was used to analyze the effect of FOXD2-AS1 on RFS in breast cancer patients with different molecular subtypes.4.The relative expression of FOXD2-AS1 in breast cancer tissues and adjacent normal breast tissues was detected by qRT-PCR.Spearman correlation analysis was used to calculate the association between FOXD2-AS1 expression and clinicopathological parameters of surgical specimens.5.The expression of FOXD2-AS1 in normal breast cell lines and five breast cancer cell lines was detected by qRT-PCR.6.The subcellular localization of FOXD2-AS1 in TNBC cells was detected by RNA nucleoplasm separation and qRT-PCR.7.The effects of FOXD2-AS1 on proliferation,migration and invasion of TNBC cells were detected by MTT and Transwell methods,respectively.8.Western blot was used to detect the effect of FOXD2-AS1 on epithelial-mesenchymal transition(EMT)related proteins in TNBC;9.Screening the target microRNAs that may have binding sites with FOXD2-AS1 from Starbase database,and validating them by qRT-PCR;10.QRT-PCR was used to detect the expression of miR-150-5p in breast cancer tissues and adjacent normal breast tissues.And the correlation between FOXD2-AS1 and miR-150-5p was analyzed by Spearman;11.Double luciferase reporter gene assay and RNA immunoprecipitation(RIP)analysis were used to verify the binding of FOXD2-AS1 to miR-150-5p;12.Targetscan,miRDB and DIANA databases were used to screen the downstream target genes of miR-150-5p;13.QRT-PCR,Western blot and Transwell methods were used to analyze the effects of miR-150-5p/ZEB1 axis on the migration and invasion of TNBC cells.14.The effects of FOXD2-AS1 on ZEB1 expression in TNBC cells were detected by qRT-PCR and Western blot.15.After co-transfection of miR-150-5p inhibitor and si-FOXD2-AS1 into TNBC cells,the rescue experiments were carried out to analyze whether the inhibition of miR-150-5p could reverse the inhibition of knocking down FOXD2-AS1 on cell migration and invasion,and the down-regulation of ZEB1 expression by Transwell,qRT-PCR and Western blot.16.The correlation between FOXD2-AS1 and ZEB1 expression level was validated in GSE58812.17.Screening of the transcription factors upstream of FOXD2-AS1 by using Genecards and PROMO databases;18.Chemostatic immunoprecipitation(CHIP)analysis was used to verify the binding of YY1 to FOXD2-AS1 promoter region;19.QRT-PCR was used to analyze the transcriptional regulation of YY1 on FOXD2-AS1.Part ?: 1.The second part: 1.Screening eligible TNBC datasets GSE21653(training set)and GSE58812(validation set)from GEO database,re-annotating the gene expression profile data based on probes to obtain the lncRNA expression profile data;2.Screening the lncRNAs related to RFS/DMFS of TNBC patients through univariate and multivariate COX analysis,1000 rbsurv dimensionality reduction analysis and COX multivariate regression analysis.3.Scoring each sample in the training set and verification set based on risk score formula.ROC analysis of risk score was carried out to evaluate the predictive value and stability of prognostic model;4.ssGSEA analysis was carried out with R software package GSVA to calculate the correlation between each KEGG pathway score and risk score,and to speculate the potential biological function involved in prognostic model.Results: Part ?: 1.The expression of FOXD2-AS1 was significantly correlated with RFS and DMFS of breast cancer patients;2.The high expression of FOXD2-AS1 was correlated with the adverse prognosis and clinicopathological characteristics of cancer patients(including tumor size,invasion depth,distant metastasis and TNM stage);3.The high expression of FOXD2-AS1 was correlated with the poor prognosis and clinicopathological characteristics of breast cancer patients(T stage,histological grade and distant metastasis).The expression of FOXD2-AS1 in breast cancer tissues was significantly higher than that in adjacent normal breast tissues,and the overexpression of FOXD2-AS1 was significantly correlated with lymph node metastasis.6.The expression of FOXD2-AS1 in breast cancer cells was significantly higher than that in normal breast cells,and the expression of FOXD2-AS1 in TNBC cells was significantly higher than that in hormone receptor positive breast cancer cells.7.FOXD2-AS1 was mainly distributed in the cytoplasm of TNBC cells;8.FOXD2-AS1 could promote the migration and invasion of TNBC cells;9.FOXD2-AS1 could promote EMT of TNBC cells;10.After screening and validation,FOXD2-AS1 was confirmed to adsorb and bind to miR-150-5p;11.The expression of miR-150-5p in breast cancer tissues is significantly lower than that in adjacent normal breast tissues.And miR-150-5p was negatively correlated with FOXD2-AS1 expression;12.Double luciferase reporter gene assay and RNA immunoprecipitation(RIP)analysis confirmed that the interaction between FOXD2-AS1 and miR-150-5p existed;13.The target gene regulated by miR-150-5p was identified as ZEB1;14.MiR-150-5p inhibited the migration and invasion of TNBC cells by targeting ZEB1;15.FOXD2-AS1 could promote ZEB1 expression in TNBC cells.16.Rescue experiments showed that inhibition of miR-150-5p could partially reverse the suppression of knocking down FOXD2-AS1 on the migration and invasion,and the down-regulation of ZEB1 expression in TNBC cells;17.External validation of GSE58812 showed that FOXD2-AS1 was positively correlated with ZEB1 expression in TNBC patients;18.YY1 was identified as a transcription factor upstream of FOXD2-AS1;19.CHIP analysis showed that YY1 might be a transcription factor upstream of FOXD2-AS1 in TNBC patients.YY1 can bind to the promoter region of FOXD2-AS1 in TNBC cells;20.YY1 can promote the transcription of FOXD2-AS1 in TNBC cells and up-regulate its expression.Part ?: 1.For training set GSE21653,through multiple dimension reduction analysis,five lncRNAs related to RFS of TNBC patients were screened out,and a prognostic model of 5-lncRNA molecular characteristic was constructed;2.ROC analysis was conducted in training set for the risk score,and the results showed that the 5-lncRNA molecular characteristic could predict RFS of TNBC patients very well;3.Stratified analysis showed that the 5-lncRNA prognostic model had stable predictive value in most subgroups of TNBC patients;4.KEGG pathway analysis showed that the biological functions related to the 5-lncRNA molecular characteristic constructed by us may be associated with some metabolic and immune pathways in TNBC;5.ROC analysis was conducted in the validation set for the risk score,and the results showed that the 5-lncRNA molecular characteristics also had a stable predictive value for DMFS in TNBC patients.Conclusion: 1.FOXD2-AS1 is highly expressed in tissues and cells of triple-negative breast cancer.2.In triple-negative breast cancer cells,FOXD2-AS1 adsorbs miR-150-5p and inhibits its function,up-regulates ZEB1 expression and promotes the invasion and metastasis of tumor cells.3.YY1 participates in the invasion and metastasis of triple-negative breast cancer cells by inducing FOXD2-AS1 transcription.4.Based on the re-annotated lncRNAs expression profile data,the 5-lncRNA prognostic model can predict the recurrence/metastasis risk of triple-negative breast cancer patients steadily and it is an independent prognostic factor of triple-negative breast cancer.