Research On The Network And Mechanism Of Multidrug Resistance Associated MicroRNA In Gastric Cancer

Background:Multidrug resistance is one of the main obstacles for the treatment of gastric cancer andbrings about poor prognosis. Therefore it is of vital importance to clarify the mechanismsof multidrug resistance for gastric cancer and identify effect molecular target to overcomemultidrug resistance. In recent years the refreshing genomic and proteomic technologyacts as a useful tool to study the multidrug resistance associated moleculars, followed bythe reporting of a great number of multidrug-resistance associated moleculars. However itis still fair to say that the multidrug resistance is so complicated that the mechanism ofmultidrug resistance is far from being fully understood. In the meantime，the previousreported molecular targets for overcoming multidrug resistance is limited and moreeffective molecular targets is needed urgently.MicroRNAs (miRNAs) are a newly class of22nucleotide non-coding RNA moleculesthat negatively regulate the expression of target genes post-transcriptionally by binding tothe3’ untranslated region (3’UTR) of mRNA through inhibiting their translation orpromoting the degradation of mRNAs. It is estimated that about one third of the genes inmammalian genome are regulated by miRNAs. More and more evidence showed that themiRNAs plays an important role in cancer progression including the multidrug resistance.However it is still not reported yet that genomic analysis of multidrug resistance in gastriccancer by combining mRNA expression profiling and miRNA expression profiling. Aim:To analyse the multidrug resistance associated genes and miRNAs and obtain a primarylist of multidrug resistance associated genes and miRNAs, with the aim to provideresources for clarifying the mechanism of multidrug resistance and identifying effectivemolecular targets for multidrug resistance. To construct a primary multidrug resistanceassociated network of gastric cancer and identify potentially functional miRNA-mRNApairs for future studies.Methods:1. The differentially expressed genes between the drug resistant sublines and theirparental cell line were identified by the mRNA expression profiling using lncRNAmicroarray.2. The differentially expressed miRNAs between the drug resistant sublinesand their parental cell line were identified by miRNA expression profiling using themicroRNA microarray.3. The commonly differentially expressed genes and miRNAswere arbitrarily selected and validated by quantitive RT-PCR.4. The GO and KEGGpathway analysis of the commonly differentially expressed genes were performed usingthe GSEA software (Gene Set Enrichment Analysis).5. The targets prediction of thecommonly differentially expressed miRNAs was performed using the Targetscan, Pictarand MiRanda algorithms.6. Integrative analysis of the mRNA and miRNA expressionprofiling data were performed by combining the commonly differentially expression datawith the miRNA target prediction.7. The MAPK pathway and PTEN/AKT pathwayactivation in drug resistant sublines was validated by the western blotting.8. By MTTassay whether MAPK pathway was a therapeutic target was validated.Results:1. Of the3,855dysregulated genes detected in SGC7901/ADR cells when compared toSGC7901cells,1,728genes were up-regulated (fold change≥2) and2,127genes weredown-regulated (fold change≤0.5). Of the3,907dysregulated genes detected inSGC7901/VCR cells when compared to SGC7901cells,2,078genes were up-regulated(fold change≥2) and1,829genes were down-regulated (fold change≤0.5). Notably,2,215genes were overlapping between the dysregulated genes in SGC7901/VCR cells and the dysregulated genes in SGC7901/ADR cells. Among the2,215overlapping genes,1024genes were commonly up-regulated and1191genes were commonly down-regulated inthe two drug resistant sublines.2. Eighteen human microRNAs were detecteddysregulated in SGC7901/ADR when compared to SGC7901; in detail,11miRNAs weredown-regulated (fold change≤0.5) and6miRNAs were up-regulated (fold change≥2) inSGC7901/ADR cells. Twenty-eight human miRNAs were found dysregulated inSGC7901/VCR when compared to SGC7901; in detail,17miRNAs were down-regulated(fold change≤0.5) and11miRNAs were up-regulated (fold change≥2) in theSGC7901/VCR cells. Between the dysregulated microRNAs in SGC7901/VCR cells andthe dysregulated microRNAs in SGC7901/ADR cells,14were overlapping. Of theseoverlapping microRNAs,9were commonly down-regulated and6were commonlyup-regulated in the two drug resistant sublines.3. The commonly differentially expressedgenes ERBB3, ERBB2, CXCR4, FGFR1, ABCB1and ABCB4were selected andvalidated by qRT-PCR. The qRT-PCR results showed a good correlation between the twotechniques. The commonly differentially expressed miRNAs were arbitrarily selected andvalidated by qRT-PCR.The results also showed a good reponse between the twotechniques.4. GO analysis (molecular function)of the commonly differentially expressedgenes revealed that receptor activity, transmembrane receptor activity, receptor bindingactivity, DNA binding activity and transcription factor activity were the5most prominentterms, while cytokine-cytokine receptor interactions, pathways in cancer, melanogenesis,focal adhesion and MAPK signalling pathway were the top5over-represented terms fromthe KEGG pathway analysis.5. A total of2339genes were predicted to by the target genesof the commonly differentially expressed miRNAs, incluing1651for the commonlydownregulated miRNAs and688for the commonly upregulated miRNAs.6. GO analysis(molecular function)of the predicted target genes revealed that DNA binding activity,transferase activity, kinase activity, receptor activity, transcription factor activity were the5most prominent terms, while Neurotrophin signaling pathway, pathways in cancer,prostate cancer, focal adhesion and MAPK signalling pathway were among the mostover-represented terms from the KEGG pathway analysis.7. Western blot analyses detected significant levels of p-Erk1/2in both SGC7901/ADR and SGC7901/VCR cells,while p-Erk1/2levels in SGC7901cells were almost undetectable. A significantly higherlevel of phospho-Akt-S473was also detected in SGC7901/ADR cells, while slightlyincreased levels of phospho-Akt–S473were detected in SGC7901/VCR cells.8. MEKinhibition by PD98059attenuated drug resistance in SGC7901/ADR and SGC7901/VCRcells in response to chemotherapeutic drugs adriamycin and vincristine by MTT assay.Conclusions:We obtained a significant number of commonly dysregulated genes and a subset ofcommonly dysregulated miRNAs by comparing the mRNA and miRNA expressionprofiles of the drug resistant sublines, SGC7901/VCR and SGC7901/ADR, with that oftheir parental gastric cancer cell line, SGC7901. We further integrated the mRNA andmiRNA profiling results, followed by the construction of a primary multidrug resistancenetwork and identification of functional miRNA-mRNA pairs that may be involved inmultidrug resistance.