Joint bioinformatics analysis of underlying potential functions of hsa-let-7b-5p and core genes in human glioma

Background Glioma accounts for a large proportion of cancer, and an effective treatment for this disease is still lacking because of the absence of specific driver molecules. Current challenges in the treatment of glioma are the accurate and timely diagnosis of brain glioma and targeted treatment plans. To investigate the diagnostic biomarkers and prospective role of miRNAs in the tumorigenesis and progression of glioma, we analyzed the expression of miRNAs and key genes in glioma based on The Cancer Genome Atlas database. Methods Of the 701 cases that were downloaded, five were normal and 696 were glioma. Then, 1626 differentially expressed genes were identified, and 173 aberrantly expressed miRNAs were calculated by edgeR. GO and KEGG pathway enrichment analyses were performed using Cytoscape software. A coexpression network was built by weighted correlation network analysis (WGCNA). A cell scratch test and transwell, cell apoptosis and cell cycle assays were performed to validate the function of hsa-let-7b-5p. Results Based on crosstalk genes in the KEGG, PPI network, and WGCNA analyses, PLK1, CCNA2, cyclin B2 (CCNB2), and AURKA were screened as candidate diagnostic marker genes. The survival analysis revealed that high mRNA expression of PLK1, CCNA2, and AURKA was significantly associated with poor overall survival. Furthermore, hsa-let-7b-5p was identified as a core miRNA in the regulation of candidate genes involved in glioma development. We confirmed that hsa-let-7b-5p could inhibit the migration, invasion, and cell cycle of glioma cells. Conclusions This study provides four potential biomarkers for the diagnosis of glioma, offers a potential explanation of its pathogenesis, and proposes hsa-let-7b-5p as a therapeutic target. Electronic supplementary material The online version of this article (10.1186/s12967-019-1882-7) contains supplementary material, which is available to authorized users.

chemotherapy sensitivity of glioma patients [8][9][10][11]. However, these markers have limited sensitivity and accuracy [12]. Therefore, specific and novel early diagnostic markers are necessary to understand the pathogenesis of glioma.
At present, the study of gene expression and mutation in glioma has become an essential way to explore the development, progression, and prognosis of glioma. Thus, the identification of potential therapeutic targets for the treatment of glioma is desirable. A large number of abnormally expressed miRNAs can be found in tumor tissues or cells. MiRNAs are involved in the posttranscriptional regulation of genes, thus inhibiting the expression of target genes and indirectly affecting various molecules in the cell signaling pathway, leading to abnormal tumor cell survival pathways and tumor cell death.
High-throughput sequencing and bioinformatics analysis are important tools for cancer research, including early diagnosis of cancer, grading of cancer, and prognosis prediction.
In the present study, a microarray dataset of normal brain and glioma tissues was downloaded from the TCGA database, and differentially expressed genes were analyzed using edgeR [13,14]. Bioinformatics analysis of DEGs was performed to identify potential genes involved in glioma development. Then, GO analysis, including of the biological processes (BPs), molecular functions (MFs), cellular components (CCs), and KEGG pathways, of the DEGs was performed to reveal the differential functions and pathways of glioma compared with normal tissues [15]. Moreover, we established the PPI network of the DEGs and WGCNA coexpression network and selected hub genes with a high degree of connectivity [16][17][18]. The results showed that hsa-let-7b-5p could regulate target genes, including cell cycle protein A2 (CCNA2), CCNB2, PLK1, and AURKA, and further affect the progress of glioma. This study may provide novel insights into the molecular and biological processes and targets of glioma.

TCGA data information
The glioma data set was downloaded from the Broad GDAC Firehouse website (http://gdac.broad insti tute. org/). The database contained 5 normal samples and 696 glioma samples. The edgeR packages of Bioconductor analysis tools were applied to detect the DEGs, using FDR < 0.05 and |logFC| ≥ 2 as cut-off criteria. A total of 1626 DEGs were calculated, including 601 upregulated genes and 1025 downregulated genes.

Module selection from the PPI network
The CentiScape plug-in was used to calculate the node degree. The molecular complex detection (MCODE) plug-in was employed to find clusters in the whole PPI network with a node score cut-off = 0.2, k-core = 2, and maximum depth = 100 as cut-off criteria. Proteins in the central nodes might have important physiological regulatory functions and might be key candidate genes. Subsequently, the genes in the most significant module were extracted and subjected to GO function and KEGG pathway enrichment analysis at a significance of P < 0.05.

WGCNA
WGCNA was used for the scale-free network topology analysis of microarray expression data of glioma samples. Standard WGCNA parameters were used for analysis, with the exceptions of soft-thresholding power and the deep split. By using WGCNA, a coexpression module of genes related to sample characteristics was quickly extracted from the glioma data for subsequent analysis. In brief, genes with expression correlation were clustered into a module by calculating the expression correlation between genes. To broadly identify the coexpressed genes, all genes were used for analysis. The purpose of the module cluster analysis was to integrate the genes obtained by WGCNA with the key genes previously analyzed to further narrow the scope of key genes. The black cluster module is the most suitable for narrowing the range of key genes. A bridge between sample characteristics and changes in gene expression was constructed. A

miRNA-mRNA regulatory interaction networks
A miRNA-mRNA regulatory network was constructed using CyTargetLinker of Cytoscape. Briefly, a file was created containing the name of the miRNA and the name of the key genes and was imported into CytoScape. Then, an experimentally validated microRNA-target interaction relationship was downloaded from miRTarBase (http:// proje cts.bigca t.unima as.nl/cytar getli nker/regin s/) with 410,602 interactions, 2645 microRNAs, and 14,797 target genes. Then, the downloaded miRNA database was imported into the extend network of CyTargetLinker. According to the constructed miRNA-mRNA network, large amounts of prediction information can be acquired when predicting miRNA target genes. Here, miRNAs that could bind to most target genes were selected as key miRNAs. Other redundant information was further removed to construct the final miRNA-mRNA network.

Comparison of the expression level and correlation coefficient of candidate genes
The mRNA expression levels of PLK1, CCNA2, CCNB2, and AURKA were analyzed using the GEPIA (http:// gepia .cance r-pku.cn/index .html) web server. A boxplot was created to visualize the expression level.

Expression level of candidate genes in normal brain tissues and glioma specimens
The protein expression levels of PLK1, CCNA2, CCNB2, and AURKA in normal brain tissues and glioma specimens were investigated, and images were obtained from the Human Protein Atlas (http://www.prote inatl as.org) online database.

Survival analysis of PLK1, CCNA2 and AURKA
The clinical data were downloaded from the TCGA database. Cases were divided into two groups according to gene expression. Survival analyses were performed using GraphPad Prism 6.

Cell culture and transfection
U118MG cells were cultured in Dulbecco's Modified Eagle's Medium supplemented with 10% fetal bovine serum, penicillin (100 IU/mL), and streptomycin (0.1 mg/ mL) in a humidified incubator at 37 °C in a 5% CO 2 atmosphere. Cells were plated in 6-well plates (3 × 10 5 cells per plate) overnight prior to transfection. Transfection was performed using Lipofectamine 2000 reagent (Invitrogen, Carlsbad, USA) in accordance with the manufacturer's protocol.

RNA isolation and quantitative real-time polymerase chain reaction analysis
At 48 h after transfection with hsa-let-7b-5p mimic or negative control nucleotides, cells were collected and washed with PBS. Then, total RNA was isolated using the RNAprep pure cell/Bacteria kit (Tiangen Biotech Co., Ltd, Beijing, China) in accordance with the manufacturer's instructions.
For miRNA expression detection, miRNA was reverse transcribed with miScript II RT Kit (Qiagen). The relative expression levels of hsa-let-7b-5p were determined with the miScript SYBR Green PCR Kit (Qiagen) using a QuantStudio TM6 Flex real-time PCR system (Life Technologies), and hsa-let-7b-5p (MS00003122). RNU6 (MS00029204) was used as an internal control miRNA.
For mRNA expression detection, the GoScript ™ Reverse Transcription System (Promega Corporation, Madison, USA) was used to reverse transcribe RNA templates, and the relative expression levels of mRNAs were determined with GoTaq qPCR Master mix (Promega Corporation) using a QuantStudio TM6 Flex real-time PCR system. GAPDH was used as an endogenous control for mRNA expression. The primer sequences used in the current study are shown in Additional file 1: Table S1.
A comparative threshold cycle method (2 −ΔΔCT ) was used to calculate the relative expression of miRNAs and mRNAs.

Wound healing assay
Wound healing (scratch) assays were performed to detect cell migration. U118MG cells were seeded in 6-well plates and incubated to generate confluent cultures. At 24 h after transfection with hsa-let-7b-5p mimic or negative control nucleotides, scratches were made in the cell monolayer using a 200-μL sterile pipette tip. Then, the cells were washed with PBS buffer. Images of cell migration at the edge of the scratch were obtained at 0, 24, and 48 h.

Cell invasion assay
The invasion assay was carried out with a transwell chamber inserted with a PET filter membrane (8 μM pores, Corning, America) in 24-well plates (Corning, America). The top side of the filter was coated with Matrigel. At 24 h after transfection with hsa-let-7b-5p mimic or negative control nucleotides, the cells were collected and resuspended in serum-free medium (1 × 10 5 /mL). Then, 200 μL was added to the top well. The lower chamber was filled with 600 μL of medium containing 10% fetal bovine

Flow cytometry analysis
Flow cytometry analysis was performed using a BD LSR-Fortessa flow cytometer (BD Biosciences, San Jose, CA, USA) to determine the distribution of cells in the cell cycle and the proportion of apoptotic cells at 48 h after transfection with hsa-let-7b-5p mimic or negative control nucleotides. In brief, cells were collected and then fixed in 70% ethanol at 4 °C for 2 h. Then, the cells were treated with a 500 μL PI/RNase a working dyeing solution for 1 h at room temperature. Detailed information can be found in the cell cycle detection kit protocol (KeyGEN biotech, Nanjing, China). For examination of apoptosis, cells were stained using an Annexin V-FITC/PI apoptosis detection kit in accordance with the manufacturer's protocol. All samples were analyzed with the FACScalibur flow cytometer, and data were analyzed with Flowjo 7.6.1 software.

Identification of DEGs based on TCGA data in glioma
A total of 17,746 genes from 5 normal samples and 696 glioma samples were obtained. By using edgeR, the expression level of each gene was log 2 transformed. Following the calculation criteria, 1626 DEGs (601 upregulated and 1025 downregulated) were identified (Additional file 1: Table S2). As shown in the volcano plot, red dots indicate upregulated genes, and green dots indicate downregulated genes. Black dots show genes with expression of |log 2 FC| < 2 (Fig. 1a). We performed hierarchical cluster analysis to obtain an overview of the expression profile for the DEGs in the normal and glioma cases (Fig. 1b).

GO function and KEGG pathway enrichment analysis and crosstalk gene screening
For a more in-depth understanding of the selected DEGs, GO functional enrichment analysis was applied using DAVID [19,20], and KEGG pathway enrichment analysis was performed using ClueGo of Cytoscape [21,22]. The upregulated DEGs were particularly enriched in cell division, mitotic nuclear division, sister chromatid cohesion, and chromosome segregation (Fig. 2). Moreover, the downregulated genes were mainly involved in chemical synaptic transmission, neurotransmitter secretion, the G-protein coupled receptor signaling pathway, coupling to cyclic nucleotide secondary messenger, and the regulation of ion transmembrane transport. The top five GO pathways for the upregulated and downregulated DEGs are shown in Additional file 1: Table S3. Based on the KEGG pathway enrichment analysis, the upregulated DEGs were enriched in the p53 signaling pathway, cellular senescence, transcriptional misregulation in cancer, progesterone-mediated oocyte maturation, and the cell cycle (Fig. 3a). The downregulated DEGs were associated with the synaptic vesicle cycle, GABAergic synapse, calcium signaling pathway, cAMP signaling pathway, and neuroactive ligand-receptor interactions (Fig. 3b). The top KEGG pathways for the upregulated and downregulated DEGs are shown in Additional file 1: Table S4. Furthermore, the relationship between genes and pathways could be clearly observed in pathways, and all the crosstalk genes are shown in Additional file 1: Table S5. Crosstalk genes are associated with more than two pathways and serve as a bridge. Thus, they warrant further study.

Key candidate genes and pathway identification with DEG PPI and modular analyses
PPI relationships of the 1626 DEGs were established using STRING. A combined score of > 0.4 was selected in the PPI networks. Finally, 10,806 PPI relationships and 1310 nodes were obtained. The degree of connectivity of the DEGs was calculated using the CentiScape 2.2 plugin module of Cytoscape. DEGs with degree of connectivity > 100 were selected as the hub genes, which possibly play an important role in glioma progression and can be used as diagnostic markers. The hub genes, of which there were > 100 in the PPI network according to their degrees, are shown in Additional file 1: Table S6. The most significant submodules of the DEGs were selected using the MCODE plug-in module to understand further the biological significance of the PPI network. As shown in Fig. 4a, the most significant module (MCODE score = 67.370) was constructed with 74 nodes and 2459 edges (Fig. 4a). Further GO pathway enrichment analysis of the biological processes showed that the genes in the most significant module were mainly associated with mitotic nuclear division, cell division, sister chromatid cohesion, chromosome segregation, cell proliferation, and DNA replication (Fig. 4b).
Cell component analysis indicated that the genes were strongly enriched in the nucleoplasm, chromosome, centromeric region, condensed chromosome kinetochore, spindle, and kinetochore. Molecular function analysis showed that the genes were mainly involved in protein binding, ATP binding, microtubule binding, microtubule motor activity, and chromatin binding. KEGG analysis showed that the genes were mainly enriched in the p53 signaling pathway, cell cycle, oocyte meiosis, and progesterone-mediated oocyte maturation (Fig. 4c).

Gene-network modules identified by WGCNAs
Gene coexpression networks were collected by obtaining pathological information and relative gene expression.
The experimental results demonstrated that all genes were clustered into 34 modules (Fig. 5a, b). Different gene coexpression modules and the patient's age, sex, survival status, recurrence, and survival time were further selected to draw a clustering heat map (Fig. 5c). To further narrowing the range of key genes, the black module genes were extracted and performed Venn analysis to select hub key genes in glioma. Finally, the genes CDK1, PLK1, CCNA2, CCNB2, and AURKA were selected. Hub gene network analysis of the black module was performed, and those five candidate genes are presented in Fig. 5f (pink nodes). Then, the key controlling (candidate) genes in the modular network were identified. Target microRNAs were searched using Cytoscape software to analyze the interplay between CDK1, PLK1, CCNA2, CCNB2, and AURKA and microRNAs. miRNA expression data were downloaded from the TCGA database. The expression level of each miRNA was calculated and log2 transformed using DESeq R. A total of 173 aberrantly expressed miRNAs (84 upregulated and 89 downregulated) in glioma tissues compared with normal brain tissues were achieved (Additional file 1: Table S7).
In the volcano plot, red dots indicate upregulated genes, and green dots indicate downregulated genes. Black dots show genes with an expression of |log 2 FC| < 1 (Fig. 6a).
Hierarchical cluster analysis was further performed to obtain an overview of the expression profile for differentially expressed miRNAs in normal and glioma cases (Fig. 6b). Using the CyTargetLinker plug-in of Cytoscape, we constructed the miRNA-mRNA expression regulatory network. The miRNA data were downloaded from miRTarBase (http://mirta rbase .mbc.nctu.edu.tw/), a database that curates experimentally validated micro-RNA-target interactions. The mirtarbase-hsa-6.1 database was selected, including 2645 microRNAs, 14,797 target genes, and 410,602 interactions. The final regulatory relationship is shown in Fig. 6c, and the results showed that hsa-let-7b-5p could modulate the expression of CCNB2, CCNA2, PLK1, and AURKA simultaneously (Fig. 6c).

Validation of the DEGs in the human atlas dataset
The mRNA expression levels of PLK1, CCNA2, CCNB2, and AURKA in normal brain tissues and glioma cancer specimens were investigated by GEPIA (Fig. 7a). Immunohistochemical and staining index analyses showed that the protein expression levels of PLK1, CCNA2, CCNB2, and AURKA were higher in glioma cancer tissues than in nontumor tissues. Images were obtained from the Human Protein Atlas (http://www.prote inatl as.org) online database (Fig. 7b). Interestingly, glioma cases with high PLK1 expression demonstrated a poorer survival compared with those with low PLK1 expression (P = 0.0029, Fig. 7c). Similarly, high expression of CCNA2 and AURKA was associated with a worse overall survival (OS) of glioma patients (Fig. 7d, e). The expression level of CCNB2 was not associated with the OS of patients with glioma.
To assess the effect of hsa-let-7b-5p on the migration of U118MG cells in vitro, we transfected the U118MG cells with negative control nucleotides or hsa-let-7b-5p mimic. After 0, 24, and 48 h, images were obtained of cell migration at the edge of the scratch (Fig. 8c). After 24 and 48 h, the wound gap was markedly wider in the hsa-let-7b-5p mimic group than in the negative control group. This result showed that hsa-let-7b-5p inhibited the motility of the U118MG cells in vitro. Then, we investigated the effect of hsa-let-7b-5p on the invasion of U118MG cells by using Matrigel-coated transwell chambers. The number of cells that crossed to the Matrigel-coated filter membrane was significantly reduced in the hsa-let-7b-5p mimic group compared to that in the negative control group after 24 h (Fig. 8d). U118MG cells were transfected with negative control nucleotides or hsa-let-7b-5p mimic to investigate the role of hsa-let-7b-5p in apoptosis by flow cytometry analysis with an Annexin V-FITC/PI apoptosis detection kit. At 48 h posttransfection, cells were collected and analyzed in accordance with the manufacturer's protocol. As shown in Fig. 8e, the total apoptosis rate of the U118MG cells transfected with hsa-let-7b-5p mimic (23.83% ± 2.51%) was significantly higher than that of the cells transfected with negative control nucleotides (15.84% ± 0.67%) (P < 0.05). The result indicated that overexpression of hsalet-7b-5p increased the apoptosis of U118MG cells. Then, we determined the effect of hsa-let-7b-5p overexpression on the U118MG cell cycle. Compared with the control group, overexpression of hsa-let-7b-5p increased the proportion of cells in S phase and decreased the proportion of cells in G2 phase (Fig. 8f ). The results showed that hsalet-7b-5p could alter cell cycle dynamics in vitro.

Discussion
Glioma is the most common primary intracranial tumor [23][24][25]. Radical surgery combined with radiotherapy and chemotherapy is the basic means to treat glioma, but the treatment effect is not optimistic due to its easy recurrence and high mortality [25]. With continuous research and improved understanding of glioma in recent years, various treatment methods, such as gene therapy, immunotherapy, and molecular targeted therapy, have been proposed [26][27][28][29][30]. However, a method to cure glioma is currently lacking. Early diagnosis of glioma and timely surgical treatment or intervention with medicine is an important means to prolong the life of patients with this disease [31,32].
In this study, expression profile chip data of 696 glioma samples and five normal samples were downloaded from the TCGA database. A total of 1626 DEGs were calculated. GO function and KEGG pathway analyses were performed to acquire an in-depth understanding of these DEGs. The GO analysis showed that the upregulated DEGs were mainly involved in cell division, mitotic nuclear division, sister chromatid cohesion, embryonic skeletal system morphogenesis, anterior/posterior pattern specification, chromosome segregation, and cell proliferation. The downregulated genes were involved in chemical synaptic transmission, neurotransmitter secretion, G-protein coupled receptor signaling pathway, coupling to cyclic nucleotide second messenger, regulation of ion transmembrane transport, potassium ion transport, and calcium ion transmembrane transport. Furthermore, the KEGG pathways of the upregulated genes were included in the cell cycle, transcriptional misregulation in cancer, bladder cancer, small cell lung cancer, the p53 signaling pathway, and cellular senescence. Moreover, the downregulated DEGs were enriched in the calcium signaling pathway, serotonergic synapse, taste transduction, neuroactive ligand-receptor interaction, synaptic vesicle cycle, and retrograde endocannabinoid signaling.
Analysis of the most significant module of the PPI network showed that glioma was associated with the biological processes of mitotic nuclear division, cell division, sister chromatid cohesion, chromosome segregation, cell proliferation, DNA replication, and microtubule-based movement. KEGG pathway enrichment analysis for the most significant module was mainly involved in the cell cycle, oocyte meiosis, progesterone-mediated oocyte maturation, and the p53 signaling pathway.
Studies have shown that molecular target pathways, such as the growth factor pathway, Ras pathway, PI3K pathway, p53 pathway, and tumor metastasis invasion pathway, have great potential for the development of meaningful treatment strategies for glioma [33,34]. The p53 pathway is one of the important pathways for the molecular pathogenesis of glioma. P53 can be divided into wild-type p53 and mutant p53. Wild-type p53 is an anti-oncogene that rapidly induces apoptosis in damaged cells, prevents cells with potentially cancerous potential, and exerts anticancer effects. Once wild-type p53 is mutated, it no longer has a tumor suppressing effect but can accelerate the proliferation and growth of cancer cells. When wild-type p53 is transformed to mutant p53, the mutant p53 acts as a proto-oncogene, which is present for a long time in tumor cells and eventually leads to tumorigenesis. The KEGG pathway analysis enriches the p53 pathway, which may be due to increased p53 mutations in gliomas (Additional file 1: Figures S1, S2). Among these DEGs, five candidate genes with a high degree of connectivity and coexpression and crosstalk in pathways were selected, including CDK1, CCNA2, CCNB2, PLK1, and AURKA. miRNAs can change the expression of target proteins. MiRNAs have important clinical significance in the diagnosis of glioma, evaluation of the efficacy of chemotherapy drugs, anti-angiogenesis, treatment, and prognosis judgment. In the present study, the expression levels of CCAN2, CCNB2, PLK1, and AURKA in glioma were upregulated, and hsa-let-7b-5p binding sites were found in four gene sequences. Let-7b is a cancer suppressor gene that can inhibit the occurrence of cancer [35,36].
As the most significant candidate gene, polo-like kinase 1 (PLK1), is a highly conserved serine/threonine protein kinase. PLK1 is highly expressed in most malignant tumor cells and is closely related to the occurrence and development of tumors [37][38][39][40]. PLK1 is mainly expressed in the late G2 phase and M phase of cell mitosis and plays a crucial role in mitosis, such as mitotic initiation, centrosome maturation, spindle assembly, chromosome segregation, and cytokinesis. Clinical data show that inhibition of PLK1 can prevent the proliferation and promote the apoptosis of tumor cells. Therefore, PLK1 is an effective target for the treatment of gliomas. Cell cycle protein A2 (CCNA2) is encoded by the CCNA2 gene in humans. CCNA2 is expressed in cell division and interacts with CDK kinase to regulate cell cycle progression [41]. The CCNA2-CDK complex may promote tumorigenesis by phosphorylation of cancer proteins or tumor suppressor genes, such as p53. CCNA2 is highly expressed in a variety of tumor types, such as lung cancer, breast cancer, cervical cancer, and liver cancer [42][43][44][45]. In normal tissues, CCNA2 is expressed at low levels or is not expressed. Some studies have confirmed that CCNA2 has oncogene activity, and its high expression or overexpression is closely related to the malignant transformation of tumor cells. Therefore, CCNA2 may be a potential diagnostic marker and an important drug target for glioma. CCNB2 plays a role in the cell cycle. CCNB2 is highly expressed in a variety of human tumor tissues and peripheral blood, and it is associated with the clinical stage and metastasis of the tumor [46][47][48]. CCNB2 overexpression is negatively correlated with the prognosis of breast cancer patients and is an independent prognostic marker. In addition, CCNB2 mRNA is overexpressed in the tumor tissue of patients with lung adenocarcinoma, and its expression is closely related to the overall disease survival rate. High expression of CCNB2 in tumor tissues is closely related to the poor prognosis of gliomas. AURKA, a member of the Aurora kinase family, is an oncogene that plays an important role in cancer stem cell development, epithelial-mesenchymal transition, and distant metastasis [49][50][51]. AURKA protein kinase plays an important role in the development of malignant tumors. AURKA has been highly expressed in various malignant tumors, such as esophageal cancer, laryngeal cancer, liver cancer, and ovarian cancer [52][53][54]. AURKA may participate in the formation of tumors through two functions. One function is the suppression of cytokinesis during mitosis, which leads to an unstable tumor genome. The other function is the inhibition of cell cycle checkpoints, helping cells with unstable genomes continue to replicate and to enter mitosis, leading to the malignant proliferation of cells. AURKA protein kinase can also promote the formation of blood vessel mimicry in triple-negative breast cancer stem cells, and inhibiting its expression can weaken the formation of blood vessel mimicry capacity.

Conclusions
In summary, PLK1, CCNA2, CCNB2, and AURKA were screened as candidate diagnostic marker genes of glioma, and hsa-let-7b-5p could inhibit the invasion and metastasis of glioma cells. The underlying mechanism may be achieved by targeting the downregulation of PLK1, CCNA2, CCNB2, and AURKA protein expression. This discovery is expected to provide new therapeutic targets and biomarkers for the future treatment of gliomas.

Additional file
Additional file 1: Figure S1. P53 mutation in Glioma patients. Figure S2. Mutation site information of p53 in Glioma patients. Table S1. Primers used for quantitative real-time polymerase chain reaction. Table S2. Differentially expressed genes between normal brain and glioma samples. Table S3. Gene ontology enrichment analysis of upregulated and downregulated genes. Table S4. KEGG pathway enrichment analysis of upregulated and downregulated genes. Table S5. Cross-talk genes involved in different pathways. Table S6. Hub genes in the PPI network with degree more than 100. Table S7. Differentially expressed miRNAs between normal brain and glioma samples.