Identification of hub genes and potential molecular mechanisms in medulloblastoma by integrated bioinformatics analysis

doi:10.21203/rs.3.rs-2067914/v1

Download PDF

Research Article

Identification of hub genes and potential molecular mechanisms in medulloblastoma by integrated bioinformatics analysis

https://doi.org/10.21203/rs.3.rs-2067914/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Medulloblastoma (MB) is the most prevalent malignant brain cancer in children. Alterations in genes are critical in the molecular pathogenesis of medulloblastoma. Herein, we propose to explore the potential biomarkers for medulloblastoma and further elucidate their possible molecular mechanism. Methods: Medulloblastoma data sets were abstracted from the GEO (Gene Expression Omnibus) data resource and merged with DEGs (differentially expressed genes) from individual microarrays. Then, GO (Gene Ontology) along with KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analyses were employed to analyze the data set. Results: We selected 3 GEO data sets of 15 normal cerebellum tissues and 66 medulloblastoma tissues. Overall, 47 DEGs were identified. Of these, 26 were upregulated, whereas 21 were downregulated DEGs. Based on the GO analysis, the primary biological focus of DEGs consisted of “Positive regulation of neuron “differentiation,” “Intracellular,” as well as “Calcium ion binding.” The most remarkable cascade based on the KEGG analysis was the “P53 signaling cascade”. The PPI network demonstrated DTL, MELK, CDK1, KIF11, NDC80, PBK, NUSAP1, TOP2A, TTK, and RRM2 as the most remarkable hub genes. Conclusion: Identification of the most remarkable hub genes and significant cascades in medulloblastoma provides critical information on the pathogenesis of medulloblastoma. In addition, this information can be used to identify diagnostic biomarkers.

Medulloblastoma

GEO data

Integrated bioinformatics

bioinformatics analysis

hub genes mechanisms

potential molecular mechanisms

Medulloblastoma (MB) is the most common and deadliest primary pediatric central nervous system tumor and accounts for approximately 15–20% of all brain tumors in children^{(1, 2)}. Medulloblastoma patients have a low rate of survival and higher mortality and are commonly found in Male children between the ages of 4 and 7⁽³⁾. The regimens include surgery, radiation, and chemotherapy. Despite recent advances in multimodal treatment, the 5-year overall survival does not improve significantly. Hence, exploring the responsible molecular mechanisms is critical to help develop reliable and efficient treatment approaches to enhance the survival of patients. New therapies are urgently needed to improve outcomes and quality of life for patients with MB.

In the last decade, there have been significant advancements in the use of microarrays in gene expression profiling to identify genes along with biological cascades linked to diverse diseases^{(4, 5)}. This technique is suitable for the identification of diagnostic as well as prognostic biomarkers of diseases. Numerous studies have been conducted regarding medulloblastoma. A significant number of these studies have documented DEGs (differentially expressed genes) which participate in the onset along with the progress of this disease^{(6, 7)}. Nonetheless, different studies have reported conflicting findings on the expression patterns of mRNAs. This could be attributed to many factors, including the use of different approaches to process data, different technological detection platforms, diversity of samples or tissues among each independent experiment, and the samples originating from varying sources. Hence, integrated DEG analyses were performed on the microarray cohorts of human medulloblastoma available in public data resources to unearth an accurate picture of gene expression signatures linked to medulloblastoma. The results identified good hub genes and pathways involved in medulloblastoma, which could serve as potential treatment targets.

Microarray data information

DNA microarray is the most recent approach that can be applied in genome analysis and gene expression mapping. The GEO (Gene Expression Omnibus) data resource was employed in the present study to provide the microarray data sets. Selection of the microarray data sets was based on: samples must have been human medulloblastoma tissues; sample size must have been more than five samples, the patients must not have received any specialized treatment consisting of chemotherapy and radiotherapy before surgery; study type of data set must have been expression profile studies. The exclusion criteria included non-human studies, medulloblastoma studies involving cells, and combined analysis of expression profile investigations. Two autonomous investigators abstracted data from the initial studies. The abstracted information included GEO accession number, Gene expression data, Number of cases along with the controls, type of sample and references, and the platform. In case of inconsistencies between the investigators, consensus or a third researcher resolved the disagreement. Approval of this study was granted by The Review Boards of the Binzhou Medical University Hospital.

The search keyword in the GEO data resource was “human medulloblastoma,” and the gene expression patterns of GSE39182, GSE74195, and GSE109401 were abstracted. GSE39182 includes five normal cerebellum tissues and 20 medulloblastoma tissues. GSE74195 consists of 5 standard cerebellum samples and 27 medulloblastoma samples. GSE109401 includes five normal cerebellum tissues and 19 medulloblastoma tissues(Supplemental Table 1).

Differentially expressed genes (DEGs) analysis

The Robust Multiarray Analysis algorithm (RMA) in the R Affy package was employed to normalize the CEL file raw data. The cohorts were randomly grouped into two groups, namely regular and medulloblastoma groups. T-test was implemented with the R limma package to determine DEGs between two study groups⁽⁸⁾. The cut-off selection criteria of the DEGs were P < 0.05, along with |log2foldchange (FC)|>1. Afterward, RobustRankAggreg was employed to perform gene integration on three gene chips determined as differentially expressed⁽⁹⁾.

GO and KEGG pathway enrichment analyses of DEGs

Clusterprofiler was employed to determine the biological processes participating in the pathogenesis of medulloblastoma through the advancement of the functions and cascades. Clusterprofiler analyses and visualizes critical data on GO along with KEGG analyses⁽¹⁰⁾. P-value < 0.05 signified significance.

Protein interaction network analysis

The STRING web resource (http://string-db.org/) is employed to determine predicted proteins' relationships with known proteins. Predictive computational and experimental data, along with text mining, yield the results. DEGs are incorporated into the PPI network. The cross talks the Cytoscape web resource evaluated the candidate proteins. MCODE (Molecular Complex Detection) was employed to determine the network of clusters. The MCODE plug-in is critical in scoring and determining the parameters that yield the optimal results for the network.

Microarray data and determination of DEGs in medulloblastoma

GSE39182, GSE74195, and GSE109401 (medulloblastoma expression microarray datasets) were standardized, as indicated in Fig. 1. The GSE39182 cohort was screened with the limma package (adjusted P-value༜0.05 along with |logFC|༞1), and 3721 DEGs were identified. Among these datasets, 1957 were downregulated, whereas 1764 were upregulated. In the GSE74195 dataset, 921 DEGs were uncovered, of which 377 were upregulated, and 544 were downregulated. In the GSE109401 data set, there were 122 DEGs, with 46 being upregulated and 66 being downregulated. The differential expression of numerous genes from the two cohorts in the three microarrays is indicated in Fig. 2.

Identification of DEGs for the 3 microarray datasets

The R “limma” package was employed to analyze three medulloblastoma gene expression microarray data sets. P < 0.05, along with |log2fold-change (FC)|>1, was employed as the cut-off criteria of genes identified as DEGs. We pre-processed the data, including background correction, normalization, and summarization. After that, the RobustRankAggreg package (adjusted P-value < 0.05) was employed to analyze the data. The RobustRankAggreg approach assumes that every gene is ordered randomly in every experiment. The highest-ranking gene across the experiments has the smallest P-value and has the highest possibility of differential gene expression. We uncovered 47 DEGs by rank analysis, consisting of 26 upregulated and 21 downregulated genes. The R-heatmap software was employed to develop a heatmap of the top 20 up and downregulated genes, as illustrated in Fig. 3.

Functional annotation and KEGG pathway for DEGs

The cluster profile package was employed in comparing gene clusters based on abundant biological processes. P < 0.05 signified significance. The GO analysis data illustrated that overlapped DEGs were remarkably linked to "Positive regulation of neuron differentiation" in the BP category; "Intracellular" in the CC category; and "Calcium ion binding" in the MF category (Fig. 4A-C, Supplemental Table 2). Besides, the enriched KEGG pathway data primarily included "Pyrimidine metabolism," "P53 signaling pathway," and "Insulin secretion" (Fig. 4D, Supplemental Table 2). The remarkably enriched terms and cascades may help us further explore the function of the DEGs in medulloblastoma.

PPI network and cluster analysis

Through the STRING web resource, 47 DEGs were incorporated into the DEGs PPI network, which had 29 nodes and 106 edges (Fig. 5A). We applied the Cytoscape web resource to investigate the cross-talk and associations of candidate proteins. Next, the MCODE plug-in was employed to score and locate parameters designed to give the best outcomes for the network. Eight clusters were determined congruent with okay-center = 2. Of these, cluster1, with 14 nodes and 87 edges, had the best score in these clusters (Fig. 5B). In this cluster, DTL, MELK, CDK1, KIF11, NDC80, PBK, NUSAP1, TOP2A, TTK and RRM2 were the top 10 genes on the basis of the mcode score. Therefore, these genes have pivotal roles in medulloblastoma.

Medulloblastoma is the most prevalent pediatric brain cancer. An extensive range of omics profiling techniques has become available to study MB⁽¹¹⁾. However, the pathogenesis of the disease has remained unclear. Similarly, there has been no breakthrough in its treatment. Establishing the molecular mechanisms of MB, with unique gene expression signatures, is critical for targeted diagnosis and treatment.

Microarray, along with high-throughput sequencing approaches that reveal the expression patterns of millions of human genes, has been extensively employed to predict possible targets for medulloblastoma treatment. In the recent decade, numerous studies on medulloblastoma have been conducted, but the five-year survival rate of medulloblastoma remains low. Moreover, there is yet to be a practical treatment approach, given that most studies have focused on the outcomes obtained from a unit cohort or on a single event. Three gene expression datasets were integrated from diverse groups and analyzed using bioinformatics and R software.

Herein, by using a comprehensive analysis of microarray data analysis, we understand a total number of 47 DEGs, including 21 downregulated genes and 26 upregulated genes in medulloblastoma compared to standard brain samples. Through PPI network construction, essential hub genes were identified. DTL, MELK, CDK1, KIF11, NDC80, PBK, NUSAP1, TOP2A, TTK, and RRM2 were hub proteins in this network. A recent study by Liu et al. found that MELK is closely associated with an intracranial malignant tumor, and MELK can regulate medulloblastoma cancer stem-like cell proliferation⁽¹²⁾. Another study showed CDK inhibitors (VMY-1-103) could inhibit cell proliferation by regulating the medulloblastoma cell cycle⁽¹³⁾. Besides this, TOP2A may be closely related to the aggressiveness of medulloblastoma, and it may become a breakthrough point for its targeted therapy⁽¹⁴⁾. These findings coincide with our study. Assessment of the DEGs in medulloblastoma was then conducted via GO functional annotation and demonstrated that DEGs were primarily involved in the modulation of "Positive regulation of neuron differentiation," Intracellular, and "Calcium ion binding." Moreover, the enriched KEGG cascades of DEGs included "Pyrimidine metabolism," "P53 signaling cascade", and "Insulin secretion." Previous investigations have reported a close association between P53 signaling and medulloblastoma⁽¹⁵⁾. Zhang al⁽¹⁶⁾ determined that EZH2 inhibits medulloblastoma growth via the p53 signaling pathway. BAI1 suppresses medulloblastoma formation by protecting p53⁽¹⁷⁾. In addition, P53 also plays a vital role in medulloblastoma chemoradiotherapy and can protect against chemotherapeutic stress and apoptosis⁽¹⁸⁾.

Bioinformatics employs biological and computational approaches to analyze biological data. This reduces the utilization of material as well as financial resources. Although numerous strengths exist, there are also some study limitations. First, the sample size was small. Besides, there is a lack of correlation analysis with clinical data for these genes and basic experiments required to explore the roles of these genes in medulloblastoma. Therefore, exploring this issue further would be a fruitful direction for future research.

Collectively, this study significantly advances our understanding molecular mechanism of the development of medulloblastoma and provides novel molecular targets for the treatment of medulloblastoma.

Funding

This study was supported by the projects of medical and health technology development program in Shandong province（2019WS323）

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Data availability statement

Data are available upon reasonable request.

Data deposition information item

GSE39182: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE39182

GSE74195: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi

GSE109401: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi

Author Contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Chenglong Li, Zefu Li. The first draft of the manuscript was written by Wenjing Su and Zhengbo Yuan. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Roussel MF, Stripay JL. Modeling pediatric medulloblastoma. Brain pathology. 2019.
Smoll NR, Drummond KJ. The incidence of medulloblastomas and primitive neurectodermal tumours in adults and children. Journal of clinical neuroscience: official journal of the Neurosurgical Society of Australasia. 2012;19(11):1541–4.
Vinchon M, Leblond P. Medulloblastoma: Clinical presentation. Neuro-Chirurgie. 2019.
Wang Y, Fan H, Zheng L. Biological information analysis of differentially expressed genes in oral squamous cell carcinoma tissues in GEO database. Journal of BUON: official journal of the Balkan Union of Oncology. 2018;23(6):1662–70.
Sun Z, Zhu Y, Aminbuhe, Fan Q, Peng J, Zhang N. Differential expression of APE1 in hepatocellular carcinoma and the effects on proliferation and apoptosis of cancer cells. Bioscience trends. 2018;12(5):456–62.
Sahin A, Vercamer C, Kaminski A, Fuchs T, Florin A, Hahne JC, et al. Dominant-negative inhibition of Ets 1 suppresses tumor growth, invasion and migration in rat C6 glioma cells and reveals differentially expressed Ets 1 target genes. International journal of oncology. 2009;34(2):377–89.
Jiang CM, Wang XH, Shu J, Yang WX, Fu P, Zhuang LL, et al. Analysis of differentially expressed genes based on microarray data of glioma. International journal of clinical and experimental medicine. 2015;8(10):17321–32.
Diboun I, Wernisch L, Orengo CA, Koltzenburg M. Microarray analysis after RNA amplification can detect pronounced differences in gene expression using limma. BMC genomics. 2006;7:252.
Kolde R, Laur S, Adler P, Vilo J. Robust rank aggregation for gene list integration and meta-analysis. Bioinformatics. 2012;28(4):573–80.
Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. Omics: a journal of integrative biology. 2012;16(5):284–7.
Ward E, DeSantis C, Robbins A, Kohler B, Jemal A. Childhood and adolescent cancer statistics, 2014. CA: a cancer journal for clinicians. 2014;64(2):83–103.
Liu H, Sun Q, Sun Y, Zhang J, Yuan H, Pang S, et al. MELK and EZH2 Cooperate to Regulate Medulloblastoma Cancer Stem-like Cell Proliferation and Differentiation. Molecular cancer research: MCR. 2017;15(9):1275–86.
Ringer L, Sirajuddin P, Heckler M, Ghosh A, Suprynowicz F, Yenugonda VM, et al. VMY-1-103 is a novel CDK inhibitor that disrupts chromosome organization and delays metaphase progression in medulloblastoma cells. Cancer biology & therapy. 2011;12(9):818–26.
Hashimoto Y, Penas-Prado M, Zhou S, Wei J, Khatua S, Hodges TR, et al. Rethinking medulloblastoma from a targeted therapeutics perspective. J Neurooncol. 2018;139(3):713–20.
Baxter EW, Milner J. p53 Regulates LIF expression in human medulloblastoma cells. J Neurooncol. 2010;97(3):373–82.
Zhang H, Zhu D, Zhang Z, Kaluz S, Yu B, Devi NS, et al. EZH2 targeting reduces medulloblastoma growth through epigenetic reactivation of the BAI1/p53 tumor suppressor pathway. Oncogene. 2020;39(5):1041–8.
Zhu D, Osuka S, Zhang Z, Reichert ZR, Yang L, Kanemura Y, et al. BAI1 Suppresses Medulloblastoma Formation by Protecting p53 from Mdm2-Mediated Degradation. Cancer cell. 2018;33(6):1004–16 e5.
Waye S, Naeem A, Choudhry MU, Parasido E, Tricoli L, Sivakumar A, et al. The p53 tumor suppressor protein protects against chemotherapeutic stress and apoptosis in human medulloblastoma cells. Aging. 2015;7(10):854–68.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Identification of hub genes and potential molecular mechanisms in medulloblastoma by integrated bioinformatics analysis

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Microarray data information

Differentially expressed genes (DEGs) analysis

GO and KEGG pathway enrichment analyses of DEGs

Protein interaction network analysis

Results

Microarray data and determination of DEGs in medulloblastoma

Identification of DEGs for the 3 microarray datasets

Functional annotation and KEGG pathway for DEGs

PPI network and cluster analysis

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1