Deciphering hub genes as potential therapeutic target involved in progression of Glioblastoma multiforme disease: A multi-omics approach

doi:10.21203/rs.3.rs-4620933/v1

Download PDF

Research Article

Deciphering hub genes as potential therapeutic target involved in progression of Glioblastoma multiforme disease: A multi-omics approach

https://doi.org/10.21203/rs.3.rs-4620933/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Glioblastoma multiforme (GBM) is a highly aggressive primary brain tumor associated with high fatality rates, poor prognosis, and limited treatment options. To enhance our understanding of the disease and pave the way for targeted therapies, it is imperative to identify key genes influencing GBM progression. In this study, we harnessed RNA-Seq gene count data from GBM patients sourced from the GEO database, conducting an in-depth analysis of gene expression patterns. Our investigation involved the stratification of samples into two distinct sets, Group I and Group II, comparing low-grade and GBM tumor samples, respectively. Subsequently, we performed differential expression analysis and enrichment analysis to uncover significant gene signatures. To elucidate the protein-protein interactions that underlie GBM, we leveraged the STRING plugin within Cytoscape for comprehensive network visualization and analysis. By applying Maximal clique centrality (MCC) scores, we identified a set of 10 hub genes in each group. These hub genes were subjected to survival analysis, highlighting their prognostic relevance. In Group I, comprising BUB1, DLGAP5, BUB1B, CDK1, TOP2A, CDC20, KIF20A, ASPM, BIRC5, and CCNB2, these genes emerged as potential biomarkers associated with the transition to low-grade tumors. In Group II, encompassing LIF, LBP, CSF3, IL6, CCL2, SAA1, CCL20, MMP9, CXCL10, and MMP1, these genes were implicated in transforming adult glioblastoma. Kaplan–Meier's overall survival analysis of these hub genes revealed that modifications, particularly upregulation of these candidate genes, were associated with reduced survival in GBM patients. The findings underscore the significance of genomic alterations and differential gene expression in GBM, presenting opportunities for early diagnosis and targeted therapeutic interventions. This study offers valuable insights into the potential avenues for improving the clinical management of GBM.

Glioblastoma multiforme disease

differential expression analysis

survival analysis

network analysis

protein-protein interaction analysis

Glioblastoma multiforme (GBM) is a primary malignant brain tumor that is extremely prevalent and fatal. It constitutes almost 50% of all primary brain tumors and has a global occurrence rate of 3–4 cases per 100,000 people per year (Stupp et al., 2017). According to the International Agency for Research on Cancer's (IARC) most recent figures, glioblastoma contributes to around 3.2% of all new cancer cases and 2.5% of all cancer-related deaths globally (Ferlay et al., 2020). The incidence rate of glioblastoma varies by region, with the highest rates observed in Europe and North America and the lowest rates observed in Africa and Asia. An average of 64 years old is the age at diagnosis, and the incidence rate is a little greater in men than in women. For glioblastoma, the survival rate after five years is quite low, ranging from 0–4%, depending on the age and health status of the patient (Ostrom et al., 2020).GBM is an extremely aggressive and infiltrative tumor that can appear de novo or progress from lower-grade astrocytoma or oligodendroglioma (Ohgaki and Kleihues, 2013). Even with rigorous surgical resection, radiation therapy, and chemotherapy, the prognosis for GBM is not significant, having a median survival span of only around 15 months (Stupp et al., 2009). The pathogenesis of GBM is complex and multifactorial, involving genetic and epigenetic alterations and environmental and lifestyle factors (Wen and Kesari, 2008).

GBM is described by a high degree of cellular heterogeneity and genomic instability, contributing to the tumor's aggressiveness and treatment resistance (Gimple et al., 2019). The molecular basis of GBM has been extensively studied, and several key genetic alterations have been identified that contribute to its development and progression. One of the most common genetic changes in GBM is the loss of heterozygosity in chromosome 10, which results in the inactivation of the tumor suppressor gene PTEN (Parker et al., 2015). The PI3K/Akt/mTOR pathway is stimulated when PTEN function is lost, promoting cell proliferation, survival, and invasion (Braglia et al., 2020; Miricescu et al., 2020). Another important genetic alteration in GBM is the amplification and overexpression of the epidermal growth factor receptor (EGFR) gene, which occurs in around 40–50% of cases (Furnari et al., 2007). EGFR overexpression activates downstream signaling pathways, like the PI3K/Akt/mTOR and RAS/RAF/MAPK pathways, that stimulate cell proliferation, survival, and invasion. More recently, molecular profiling studies have identified additional genetic and epigenetic alterations in GBM that define distinct molecular subtypes of the disease (Network, 2015). These subtypes can be distinguished by distinctive DNA methylation patterns, somatic mutations, and gene expression profiles, which have important implications for patient prognosis and treatment.

In addition to genetic alterations, environmental and lifestyle factors have also been involved in the progression of GBM. One of the reliable risk factors for GBM, particularly at a young age, is exposure to ionising radiation (Little et al., 2009). Other potential risk factors include an introduction to certain chemicals, such as vinyl chloride, and certain viral infections, such as cytomegalovirus and human herpesvirus 6 (Wrensch et al., 2002). Despite advances in the interpretation of GBM pathogenesis and the development of new treatment modalities, the prognosis for GBM patients remains poor. Standard treatments, including surgery, radiation therapy, and chemotherapy, have limited effects on patient survival, and novel therapeutic approaches are urgently needed. The most frequent signs of GBM include headaches, ataxia, dizziness, and difficulty with vision, depending on the location and rising intracranial pressure resulting from the disease's clinical stage (Levine et al., 1987; Lakhan and Harle, 2009). Due to vague symptoms, Glioma could be misdiagnosed as infections, inflammatory processes, and circulatory and immunological conditions (Lakhan and Harle, 2009).

There has been a growth in interest in creating targeted therapeutics for GBM in recent years, including small molecule inhibitors of key signaling pathways and immunotherapy (Hambardzumyan and Bergers, 2015). The biomarkers for the prediction of cell senescence in GBM still require further study, nevertheless. We did the bioinformatics study based on the gene expression in cells for GBM transformation from normal to low-grade tumor and low-grade to adult glioma tissue, leading to the selection of mostly correlated genes as hub genes. We have taken RNA-Seq gene expression data from the GEO database. The DESeq2 package in R categorized the differentially expressed genes (DEGs) in the chosen dataset, followed by the execution of ‘annotationdbi’ package for annotation. ENRICHR, a web-based tool, was used for pathway and enrichment analysis. Further, for visualization of the STRING plugin in Cytoscape, we constructed the protein–protein interaction (PPI) networks and determined the hub genes using the Cytohubba plugin. Survival analysis was also implemented to validate the hub genes in both groups statistically. A graphical representation of the work pipeline used in this study is illustrated in Fig. 1.

Data Extraction

RNA-Seq data was retrieved from NCBI's Gene Expression Omnibus (Accession Id- GSE147352), which is a repository for functional genomics data (https://www.ncbi.nlm.nih.gov/geo). An extensive search was done to find datasets for GBM. The criteria were to find datasets containing Normal and tumor tissue samples belonging to Homo sapiens. Our dataset contains 85 adult glioblastomas, 18 lower-grade gliomas, and 15 normal brain tissue samples containing the expression of 58,303 coding and non-coding transcripts.

Differential Expression Analysis

Raw gene count data were obtained from GEO and must be transformed into a series matrix file. Using file handling in R, we transformed the htseq-counts file into a series matrix file. Further, NA values were removed from the series matrix file and lowly expressed genes were eliminated using rowSums() function to obtain significant results.

For Differential Expression Analysis, we have used DESeq2 package in R (Love et al., 2014). Here, we have provided a metadata file for samples and a series matrix file as input. We have analyzed our samples by comparing low-grade tumor vs normal tissues and Low-grade tumor vs adult glioblastoma. The standards for determining the differentially expressed genes (DEGs) were defined as an |log2(fold change) |>2 and an adjusted p-value < 0.05. Filtering of DEGs was performed using the criteria log2(fold change) > 2 for the up-regulated and log2(fold change) <-2 for the downregulated genes, respectively (Habib et al., 2022). The cut-off for the log2 fold change was set ± 2 to obtain a moderate number of DEGSs. The ggplot2 package in R was used to create a volcano plot of DEGs. Gene IDs available in the series matrix file were annotated from Ensembl using ‘annotationdbi’ package in R where EnsDb.Hsapiens.v75 database was used for mapping.

Pathways and Gene Enrichment Analysis

Enrichment Analysis of differentially expressed genes was done using ENRICHR, which is a freely available web-based program (https://maayanlab.cloud/Enrichr) that provides various types of summaries having combined functions of genes involved. Pathway analysis was executed using Reactome 2022. Gene ontology analysis was done using – GO Molecular Function 2021, GO Biological Process 2021 and GO Cellular Component 2021. The genes obtained in different analysis was studied thoroughly.

Protein-protein interaction (PPI) network construction and analysis

STRING, also known as Search Tool for the Retrieval of Interacting Genes/Protein (https://string-db.org), is a database of protein interactions that are already known and are listed in the literature or have direct and indirect associations derived from computational predictions. The gene IDs were imported to Stringdb, and the network was obtained with a confidence score (> 0.9), which was exported to Cytoscape (3.4.0) for visualization. Cytoscape is an open-source tool for visualizing networks and their interactions. The network imported from Stringdb was annotated with different colors based on their condition or involvement in the upregulation or downregulation of genes. Furthermore, styling and layout changes have been made to make the network more comprehensible.

Hub Gene Evaluation

Ten important hub genes were identified in each group executed using the Cytohubba plugin available in Cytoscape. Cytohubba utilizes different algorithms to find important genes based on their topological properties in the network (Chin, 2014). Here, the screening of hub genes was accomplished using the Maximal clique centrality (MCC) method where each gene is considered as a node. The MCC of ν is defined as MCC(ν) = \(\:{\sum\:}_{C\in\:S\left(\nu\:\right)}\left(\left|C\right|-1\right)!\), where ν is a node and S(ν) is the collection of maximal cliques which contain ν, and (|C| − 1)! is the product of all positive integers less than |C|. MCC(ν) is equal to the degree of v, if an edge does not exist between its neighbours.

Survival Analysis

Survival analysis for the hub genes obtained was implemented using GEPIA -Gene expression profiling interactive analysis (http://gepia.cancer-pku.cn), which contains gene expression profiles or datasets of cancer patients that employs log-rank test, also called Mantel-Cox test for the evaluation of the hypothesis. Survival analysis or time-to-event analysis in general, is the term used to signify a set of methodologies utilized for evaluating how long it will take for an event or point of interest to occur (Schober and Vetter, 2018). When the total survival time of a subject can not be precisely determined, censoring is applied (Rich et al., 2010). We have selected the GBM dataset for the evaluation. To find a better association between expressed hub genes and Glioblastoma prognosis, the ‘Survival’ component of GEPIA was executed to create survival curves.

Identification of differentially expressed genes

RNA-Seq data with GEO accession number GSE147352 (Glioblastoma multiforme disease) was retrieved from the database. The dataset contains 118 samples in which 15 normal brain tissue samples, 18 lower-grade gliomas and 85 adult glioblastomas were present. The series matrix file consisting of 58,303 transcripts was prepared using file handling in R. NA values from the samples were removed, after which 34,517 transcripts were obtained. To remove lowly expressed genes that may hinder our results, we used a rowSums(> 100) filter on the remaining genes, in which 22,549 were obtained. The cut-off value for rowSums() depends upon the size of the dataset. Our study used a fairly large dataset of 118 samples that would require a higher cut-off value. Furthermore, using Bioconductor’s DESeq2 package in R, we executed differential expression analysis on the remaining 22,549 transcripts.

A total of 2644 DEGs were retrieved in the first group, where the reference was normal tissues. Low-grade vs. normal tissue condition was opted, with 1276 and 1368 up-regulated and downregulated, respectively. In contrast, the low-grade tumor was taken as a reference in the second group. Glioblastoma vs low-grade tumor condition was selected, resulting in 889 total DEGs consisting of 700 up-regulated and 189 downregulated genes. The volcano plots portraying up, down, and non-regulating genes are depicted in Figs. 2a and 2b. The top 10 up-regulated and down-regulated genes in both conditions and their p values can be accessed through supplementary information (Supplementary Table S1-S4).

PPI network analysis

For the construction of the network, 2644 DEGs were submitted in STRING from group I, out of which only 1835 could be mapped from the database. Subsequently, the network was visualized in Cytoscape. The network comprises 509 nodes and 1493 edges (see Fig. 3a). Identification of up-regulated and downregulated DEGs was done in the network. In group II, 889 DEGs were submitted, of which only 645 could be mapped from the database. Upon visualization, the network consisted of 209 nodes and 354 edges, and the up-regulated and downregulated genes were identified in this network (see Fig. 4a).

Hub gene identification

Using the Cytohubba plugin, the high-value genes or hubs with dense connectivity were found by employing 11 analysis methods in Cytoscape. Node scores were calculated, and then the ranking of the top 10 Hubba nodes was done using the MCC method. Figure 3b depicts the hub genes for the Group I condition, and Fig. 4b depicts the hub genes for the Group II condition.

Pathway and gene enrichment analysis

In group I, 2644 obtained DEGs were submitted in the ENRICHR web-based tool to produce a combined functional summary of the genes. In pathway analysis, most genes were associated with Signal Transduction, Immune System, Disease, Metabolism, and signaling by GPCR. In the GO Biological Process, most genes were enriched in transcriptional regulation by RNA polymerase II, positive regulation of transcription, gene expression regulation, and chemical synaptic transmission (Supplementary Table S5). For GO Molecular Function, most genes were annotated in RNA pol II transcription regulatory region sequence-specific DNA binding, metal ion binding, and cis-regulatory region sequence-specific DNA binding (Supplementary Table S6). In the GO Cellular Component, most genes were involved in the intracellular membrane-bounded organelle, nucleus, an integral component of plasma membrane, neuron projection and the intracellular non-membrane-bounded organelle (Supplementary Table S7).

In group II, 889 DEGs were obtained and submitted to ENRICHR, and the summarised gene functions were obtained. In pathway analysis, most of the genes were associated with signal transduction, immune system, cytokine signaling in the immune system, metabolism of proteins and developmental biology. In the GO Biological Process, most genes were enriched in the regulation of transcription by RNA polymerase II, upregulation of cellular processes, cytokine-mediated signaling pathway and regulation of cell population proliferation (Supplementary Table S8). For GO Molecular Function, most genes were annotated in RNA pol II transcription regulatory region sequence-specific DNA binding, cytokine activity, receptor-ligand activity and cis-regulatory region sequence-specific DNA binding (Supplementary Table S9). In the GO Cellular Component, most genes were associated with an intracellular membrane-bounded organelle, a nucleus, an integral component of the plasma membrane, an extracellular matrix that contains collagen and an intracellular organelle lumen (Supplementary Table S7). Figure 5 depicts genes involved in various pathways for group I and II conditions. The gene ontology for both groups is illustrated in Fig. 6–8.

Survival Analysis

For the hub genes’ overall survival study and their relationship to the prognosis of GBM, we have used the GEPIA server as the GBM cancer patient survival dataset was present in GEPIA. In group I condition, CDK1 and TOP2A show no significant difference in survival of patients; the percent survival of both the genes is similar, but the survival duration of low expressing genes shows a reduced survival rate when compared with high expression of both the genes. Further, Fig. 9 displays a similar survival rate for DLGAP5, BUB1, and BUB1B, with nearly the same median survival duration. The survival period for these three cases when lowly expressed is between 50 to 70 months. The BUB family of genes containing (BUB1 and BUB1B) plays a major role in spindle checkpoint during mitosis. According to a study, BUB1B may exert a significant influence on the progression of GBM, where overall survival was found to increase when expressed low compared to the higher expression group (Ma et al., 2017). It has also been found that DLGAP5 was found to be up-regulated in gliomas, leading to poor survival chances (Zhou et al., 2021). Higher expression of KIF20A and ASPM suggests lower survival chances of patients in later stages of the observation period in comparison to the survival curves of other highly expressed genes. The survival plots for the group I condition are shown in Fig. 9.

In group II condition, CXCL10 shows poor survivability after 35 months of observation period and CCL20 shows low chances of survival after 30 months of diagnosis when highly expressed. However, SAA1 shows statistical significance with a p-value of 0.019 and HR of 1.5, where high expression of this gene reduces the survival chance of the patient. MMP1 also shows reduced survival in higher expression. Furthermore, MMP9 shows an opposite scenario where a higher expression may lead to greater patient survival chances where the p-value is 0.11, not statistically significant. Higher expression of MMP1, CCL2, and IL6 leads to reduced survivability in late-stage GBM patients but shows no statistical significance. The survival plots for the group II condition are shown in Fig. 10.

Glioblastoma is among the most incurable and fatal cancers that have been known for a very low survival duration in patients. The treatment available today includes surgery, after which radiotherapy and chemotherapy are carried out. Targeted therapy provides plenty of opportunities for the development of novel treatment approaches.

Our study revealed that upregulation of 20 genes, i.e., 10 genes in group I- BUB1, DLGAP5, BUB1B, CDK1, TOP2A, CDC20, KIF20A, ASPM, BIRC5, CCNB2 leads to lower grade glioma transformation. Gene ontology suggests BUB1 mitotic checkpoint serine/threonine kinase B (BUB1) is found to be involved in mitosis functional in microtubule skeleton organization, regulation of mitotic cell cycle phase transition, and cell cycle checkpoint signaling. BUB1 expressions enhance tumor development and stimulate radioresistance in GBM (Ma et al., 2017). The human 2q14 chromosome contains the BUB1 gene. The protein that is encoded by it serves as a base protein for spindle physical analysis. It is an organizing structure for precisely positioning other cell parts in the spindle. The BUB1 gene ensures proper chromosomal segregation and minimizes aneuploid formation in mitosis (Bouali et al., 2022). It is also found to be involved in gastric cancer prognosis, a key gene in colorectal cancer, a biomarker in breast cancer and found involved in the development of bladder cancer (Jiang et al., 2021; Gao et al., 2022; Hassan et al., 2022; Yun et al., 2022).

Gene ontology analysis shows DLGAP5 to be involved in signaling by notch regulation of metaphase transition. A mitotic spindle protein, also known as DLG7 or HURP, facilitates the synthesis of tubulin polymers (Santarella et al., 2007). Overexpression of this gene can act as resistance to radiotherapy and chemotherapy. The cells in the G0/G1 phase arose after DLGAP5 was downregulated; however, the cells in the S and G2/M phases substantially decreased, which suggests that DLGAP5 may accelerate the transition of the G0/G1 phase, therefore promoting glioma growth. It was also found that highly expressed DLGAP5 decreased patient survival rates (Zhou et al., 2021).

CDK1 is an essential regulator of cell cycle progression and regulation (Malumbres, 2014). GO-term analysis further shows it is involved in signal transduction, negative regulation of apoptosis, and mitotic cell cycle phase transition. It's essential to note that the overexpression of CDK1 reversed the inhibitory impact of the pathway inhibitor. The stimulation of GBM expansion by initiating the Akt signaling pathway was weakened by knocking out CDK1 to some extent in a study. These findings suggest that CDK1 contributed to the Akt signaling pathway, supporting the development of GBM tumors (Zhang et al., 2018). Many cancers have unusually elevated levels of the enzyme topoisomerase (DNA) II alpha (TOP2A), which modulates and modifies the topological states of DNA during transcription (Zhou et al., 2018). GO analysis suggests that TOP2A has a role in DNA binding, chromosome organization, condensation, and RNA binding. Several kinds of human cancers could be affected by TOP2A, a sensitive and specific marker observed in proactive cells that proliferate in the cell cycle's late S, G2, and M phases (Lazaris et al., 2002; Ravasz et al., 2002). CDC20, also known as cell division cycle 20, was found to be involved in the regulation of synapse maturation, cell cycle regulation, and mitotic spindle checkpoint as per GO analysis. CDC20 was stated as a target to overcome Temozolomide-resistant cells in GBM (Wang et al., 2017a). KIF20A belongs to the kinesin family.

In GO analysis, it is found to be involved in the Polo-like kinase 1 (PLK1) pathway, kinase binding, protein kinase binding, DNA replication and Aurora B signaling. A study found that overexpression of KIF20A promotes the proliferation of glial cells; hence, knocking out of these genes suppressed the PI3K/AKT pathway, leading to cell cycle arrest and apoptosis (Wang et al., 2017b). KIF20A is also a prognostic biomarker for malignant astrocytoma (Wang et al., 2017b). ASPM, according to ENRICHR, is a protein involved in glioblastoma. According to numerous studies, ASPM functions as an important Wnt signaling pathway regulator, thereby enhancing cancer tumor formation and stimulating neuron generation during brain development (Major et al., 2008; Buchman et al., 2011; Pai et al., 2019). It functions as an oncogene whose expression is elevated in certain types of human cancers, such as GBM, ovarian carcinoma, pancreatic cancer, gastric malignancy, and prostate cancer (Buchman et al., 2011; Wang et al., 2013; Vange et al., 2015; Pai et al., 2019). Additionally, it was discovered that sustained ASPM knockdown reduced both in vivo and in vitro proliferation of cells (Chen et al., 2020).

Furthermore, upregulation of 10 genes in group II, i.e., – LIF, LBP, CSF3, IL6, CCL2, SAA1, CCL20, MMP9, CXCL10, MMP1, leads to cancer development progressing to adult glioblastoma from lower grade glioma. According to GO analysis, LIF (Leukaemia inhibitory factor) is involved in cellular response to cytokine stimulus, cytokine-mediated signaling pathway, and interleukin signaling. LIF is a significant possible cancer treatment target that acts as a modulator for the immune system. In a study on glioblastoma, tumor-associated macrophages (TAMs) were found to be more prevalent when high amounts of LIF were present in the microenvironment of the tumor (TME) (Pascual-García et al., 2019; Christianson et al., 2021). Gene ontology suggests CSF3 acts as a growth factor for granulocytes and is involved in positive regulation of cellular processes, myeloid leukocyte differentiation, cytokine-mediated signaling pathways and immune systems. It has been revealed in a study that brain cells create CSF3 when there is a tumor present and that this synthesis causes hematopoiesis to shift towards granulocytic lineages, resulting in scarcity of WBC and favoring immunosuppression (Kast et al., 2017). The differences in Hub Genes between the two groups reflect the varying genetic characteristics and disease progression between low-grade and high-grade GBM. This divergence is a critical aspect of our findings as it sheds light on the heterogeneity of the disease and the potential variations in molecular mechanisms driving its progression. Group I, comprising low-grade Glioblastoma cases, is crucial as it sheds light on the early stages of the disease, providing insights into its initiation and potentially offering markers for early detection and intervention. On the other hand, Group II, representing high-grade Glioblastoma cases, is equally vital. High-grade Glioblastoma is associated with more aggressive progression and poorer prognosis, making it a major focus of research and clinical interest.

Moreover, in vitro, CSF3 enhanced glioma cell growth, migration, and invasion (Wang et al., 2012; Bacolod et al., 2016). IL6 is involved in inflammatory response, positive regulation of MAPK cascade, regulation of angiogenesis, Interleukin signaling, and Cytokine signaling revealed by GO analysis. Due to the invasive character of glioblastoma and a greater probability of recurrence, IL-6 signaling supports many pathways that support glioma formation, such as proliferation and migration (West et al., 2018). CCL2, as per GO analysis, is found in inflammatory response, granulocyte chemotaxis, regulation of T cell activation, GPCR ligand binding, and Signal transduction. Glioma cells produce several chemokine members, including CCL2, CXCL8, and CXCL12. Accordingly, glioma cells contribute to various biological characteristics of glial tumors, including invasiveness, survival, vascular development, and proliferation. They also produce chemokines and express chemokine receptors (Vakilian et al., 2017). An inflammatory-associated high-density lipoprotein is serum amyloid A1 (SAA1). Additionally, it is regarded as a prognostic indicator and indicator of cancer risk (Zhang et al., 2021). According to GO analysis, it is associated with cellular response to cytokine stimulus, neutrophil migration and chemotaxis. By controlling the production of proteins associated with apoptosis, like Bcl2 and Bax, SAA1 suppression may decrease serine/threonine protein kinase B (AKT) phosphorylation and can cause GBM cells to die. Moreover, Temozolomide (TMZ) sensitivity is elevated in glioma with decreased SAA1 activity (Knebel et al., 2017).

GO term shows MMP9 to be involved in extracellular structure organization, cellular response to cytokine stimulus, collagen formation and signaling by interleukins. The findings of one study suggested that the transformation to the more aggressive phenotype typical of WHO grade III gliomas may require the overexpression of MMP9, indicating the possible interference of the MMP9 gene in glioma formation and disease progression (Xue et al., 2017). The CXC chemokine family includes the CXCL10 protein; GO analysis shows the involvement of this protein in the inflammatory response, neutrophil chemotaxis, cytokine signaling in the immune system, signal transduction. When examined alongside regular astrocytes, CXCL10 is elevated in grade III and grade IV human glioma cells, respectively (Maru et al., 2008).

In vitro, CXCL10 promotes DNA synthesis and cell growth in human glioma cells. Although an in vivo connection between this chemokine system and glioma advancement has not been proven, these studies imply that CXCR3 is involved in glioma creation and progression(Liu et al., 2011). MMP-1 is found to be expressed in human GBM but not in normal brain tissue (McCready et al., 2005), and GBM cell movement is impaired when MMP-1 is knocked down (Pullen and Fillmore, 2010). In our GO analysis, it is found to be involved in extracellular structure organization, collagen degradation and cytokine signaling. In a non-permissive condition, the expression of MMP-1 greatly affects the incidence of tumors. Increased tumor development and tumor size were both found to correlate with elevated MMP-1 levels across all time durations studied (McCready et al., 2005). This study is based on the statistical analysis of the available GBM datasets in the GEO database. In-vitro and in-vivo validation would provide us with more insights about the differentially expressed genes and help us to identify potential targets for therapeutic intervention of GBM.

Limitations

Despite various useful findings, this study has a few critical limitations, including reliance on RNA-Seq data from a single dataset that potentially limits generalizability. Another limitation is that the nature of GBM is very heterogeneous, and this may lead to the lack of identification of some molecular subtypes. Some limitations can be related to methodological aspects, for instance, the choice of specific bioinformatics tools might produce different outcomes. The survival analysis based on the available datasets may be prone to selection bias and may not have all the information on the patients. Functional validation of these identified hub genes is further required to support the therapeutic perspective. Lack of coverage of some pathways and reduction of the protein-protein interactions may not mimic the real biological context appropriately. It was possible for the selection and processing of data to be biased, so the results needed to be verified with other datasets.

Our study collected RNA-Seq count data from the GEO database, after which filtration and analysis were done. Initially, two comparison groups comprised normal, low-grade, and adult glioblastoma tissue samples. Differential expression analysis revealed a total of 2644 DEGs in the first group, comparing normal and low-grade tissue samples, with 1276 and 1368 up-regulated and downregulated genes, respectively, Whereas, in the second group, glioblastoma and low-grade tumor samples compared, resulting in 889 DEGs consisting of 700 up-regulated and 189 downregulated genes. Functional analysis of the genes suggested that the highest number of genes clustered by the pathway analysis tools were of the signaling domain, followed by the involvement of genes in the immune response. Using the CytoHubba plugin in Cytoscape and STRING database, protein network visualization was executed. We selected the top 10 hub genes in both groups with the highest score according to the MCC method. Using the GEPIA2 server, we generated survival rate plots and analyzed them for potential biomarkers that can help in GBM prognosis. Genes such as MMP9, SAA1, CCL2, and MMP1 could have a prominent role in detecting and diagnosing GBM and may have a role in cancer progression. However, further investigation of their role in GBM is required. Overall, the current study suggests that the elucidated genes can be explored in therapeutic involvements of GBM after thorough validation.

Funding: This work is supported by the Indian Council of Medical Research (Grant No. ISRM/12(22)/2020). Researchers Supporting Project Number (RSP 2023R122), King Saud University, Riyadh, Saudi Arabia, for funding this study.

Acknowledgments

MFA acknowledges and extends his appreciation to the Researchers Supporting Project Number (RSP 2023R122), King Saud University, Riyadh, Saudi Arabia, for funding this study.

Authors' Contribution Statement

MUS: Conceptualization, Methodology, Software, Formal analysis, Data curation, Writing - Original Draft, Writing - Review & Editing. AC: Formal analysis, Writing - Original Draft, Writing - Review & Editing. MFA: Writing - Original Draft, Writing - Review & Editing. JA: Methodology, Visualization, Writing - Review & Editing. TM: Writing - Review & Editing. AH: Formal analysis, Data curation, Validation, Review & Editing; AS: Methodology, Visualization, Writing - Review & Editing. MIH: Conceptualization, Writing - Review & Editing, Supervision, Project Administration. All authors read and approved the final manuscript.

Data Availability Statement

The data underlying this article is available within the manuscript.

Competing interests The authors declare no competing interests.

Confict of interest The authors declare that the research was conducted in the absence of any commercial or fnancial relationships that could be construed as a potential confict of interest.

Ethical Approval Not Applicable

Bacolod, M.D., Talukdar, S., Emdad, L., Das, S.K., Sarkar, D., Wang, X.-Y., Barany, F., Fisher, P.B., 2016. Immune infiltration, glioma stratification, and therapeutic implications. Translational cancer research 5, S652.
Bouali, N., Hammouda, M.B., Ahmad, I., Ghannay, S., Thouri, A., Dbeibia, A., Patel, H., Hamadou, W.S., Hosni, K., Snoussi, M., Adnan, M., Hassan, M.I., Noumi, E., Aouadi, K., Kadri, A., 2022. Multifunctional Derivatives of Spiropyrrolidine Tethered Indeno-Quinoxaline Heterocyclic Hybrids as Potent Antimicrobial, Antioxidant and Antidiabetic Agents: Design, Synthesis, In Vitro and In Silico Approaches. Molecules 27.
Braglia, L., Zavatti, M., Vinceti, M., Martelli, A.M., Marmiroli, S., 2020. Deregulated PTEN/PI3K/AKT/mTOR signaling in prostate cancer: still a potential druggable target? Biochimica et Biophysica Acta (BBA)-Molecular Cell Research 1867, 118731.
Buchman, J.J., Durak, O., Tsai, L.-H., 2011. ASPM regulates Wnt signaling pathway activity in the developing brain. Genes & development 25, 1909-1914.
Chen, X., Huang, L., Yang, Y., Chen, S., Sun, J., Ma, C., Xie, J., Song, Y., Yang, J., 2020. ASPM promotes glioblastoma growth by regulating G1 restriction point progression and Wnt-β-catenin signaling. Aging (Albany NY) 12, 224-241.
Christianson, J., Oxford, J.T., Jorcyk, C.L., 2021. Emerging Perspectives on Leukemia Inhibitory Factor and its Receptor in Cancer. Frontiers in Oncology 11.
Ferlay, J., Ervik, M., Lam, F., Colombet, M., Mery, L., Piñeros, M., Znaor, A., Soerjomataram, I., Bray, F., 2020. Global Cancer Observatory: Cancer Today. Lyon: International Agency for Research on Cancer. IARC; 2018.
Furnari, F.B., Fenton, T., Bachoo, R.M., Mukasa, A., Stommel, J.M., Stegh, A., Hahn, W.C., Ligon, K.L., Louis, D.N., Brennan, C., 2007. Malignant astrocytic glioma: genetics, biology, and paths to treatment. Genes & development 21, 2683-2710.
Gao, F., Wang, J., Li, C., Xie, C., Su, M., Zou, C., Xie, X., Zhao, D., 2022. Risk-related genes and associated signaling pathways of gastrointestinal stromal tumors. International Journal of General Medicine, 3839-3849.
Gimple, R.C., Bhargava, S., Dixit, D., Rich, J.N., 2019. Glioblastoma stem cells: lessons from the tumor hierarchy in a lethal cancer. Genes & development 33, 591-609.
Habib, I., Anjum, F., Mohammad, T., Sulaimani, M.N., Shafie, A., Almehmadi, M., Yadav, D.K., Sohal, S.S., Hassan, M.I., 2022. Differential gene expression and network analysis in head and neck squamous cell carcinoma. Mol Cell Biochem 477, 1361-1370.
Hambardzumyan, D., Bergers, G., 2015. Glioblastoma: defining tumor niches. Trends in cancer 1, 252-265.
Hassan, M.I., Anjum, D., Mohammad, T., Alam, M., Khan, M.S., Shahwan, M., Shamsi, A., Yadav, D.K., 2022. Integrated virtual screening and MD simulation study to discover potential inhibitors of Lyn-kinase: targeting cancer therapy. J Biomol Struct Dyn, 1-11.
Jiang, N., Liao, Y., Wang, M., Wang, Y., Wang, K., Guo, J., Wu, P., Zhong, B., Guo, T., Wu, C., 2021. BUB1 drives the occurrence and development of bladder cancer by mediating the STAT3 signaling pathway. Journal of Experimental & Clinical Cancer Research 40, 378.
Kast, R.E., Hill, Q.A., Wion, D., Mellstedt, H., Focosi, D., Karpel-Massler, G., Heiland, T., Halatsch, M.-E., 2017. Glioblastoma-synthesized G-CSF and GM-CSF contribute to growth and immunosuppression: Potential therapeutic benefit from dapsone, fenofibrate, and ribavirin. Tumor Biology 39, 1010428317699797.
Knebel, F.H., Uno, M., Galatro, T.F., Bellé, L.P., Oba-Shinjo, S.M., Marie, S.K.N., Campa, A., 2017. Serum amyloid A1 is upregulated in human glioblastoma. Journal of neuro-oncology 132, 383-391.
Lakhan, S.E., Harle, L., 2009. Difficult diagnosis of brainstem glioblastoma multiforme in a woman: a case report and review of the literature. Journal of medical case reports 3, 1-3.
Lazaris, A., Kavantzas, N., Zorzos, H., Tsavaris, N., Davaris, P., 2002. Markers of drug resistance in relapsing colon cancer. Journal of cancer research and clinical oncology 128, 114-118.
Levine, S.A., McKeever, P.E., Greenberg, H.S., 1987. Primary cerebellar glioblastoma multiforme. Journal of neuro-oncology 5, 231-236.
Little, M.P., Wakeford, R., Tawn, E.J., Bouffler, S.D., Berrington de Gonzalez, A., 2009. Risks associated with low doses and low dose rates of ionizing radiation: why linearity may be (almost) the best we can do. Radiology 251, 6-12.
Liu, C., Luo, D., Reynolds, B.A., Meher, G., Katritzky, A.R., Lu, B., Gerard, C.J., Bhadha, C.P., Harrison, J.K., 2011. Chemokine receptor CXCR3 promotes growth of glioma. Carcinogenesis 32, 129-137.
Love, M.I., Huber, W., Anders, S., 2014. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 15, 550.
Ma, Q., Liu, Y., Shang, L., Yu, J., Qu, Q., 2017. The FOXM1/BUB1B signaling pathway is essential for the tumorigenicity and radioresistance of glioblastoma. Oncol Rep 38, 3367-3375.
Major, M.B., Roberts, B.S., Berndt, J.D., Marine, S., Anastas, J., Chung, N., Ferrer, M., Yi, X., Stoick-Cooper, C.L., von Haller, P.D., Kategaya, L., Chien, A., Angers, S., MacCoss, M., Cleary, M.A., Arthur, W.T., Moon, R.T., 2008. New regulators of Wnt/beta-catenin signaling revealed by integrative molecular screening. Sci Signal 1, ra12.
Malumbres, M., 2014. Cyclin-dependent kinases. Genome Biology 15, 122.
Maru, S.V., Holloway, K.A., Flynn, G., Lancashire, C.L., Loughlin, A.J., Male, D.K., Romero, I.A., 2008. Chemokine production and chemokine receptor expression by human glioma cells: role of CXCL10 in tumour cell proliferation. Journal of neuroimmunology 199, 35-45.
McCready, J., Broaddus, W.C., Sykes, V., Fillmore, H.L., 2005. Association of a single nucleotide polymorphism in the matrix metalloproteinase‐1 promoter with glioblastoma. International journal of cancer 117, 781-785.
Miricescu, D., Totan, A., Stanescu-Spinu, I.-I., Badoiu, S.C., Stefani, C., Greabu, M., 2020. PI3K/AKT/mTOR signaling pathway in breast cancer: from molecular landscape to clinical aspects. International journal of molecular sciences 22, 173.
Network, C.G.A.R., 2015. Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas. New England Journal of Medicine 372, 2481-2498.
Ohgaki, H., Kleihues, P., 2013. The Definition of Primary and Secondary GlioblastomaDefinition of Primary and Secondary Glioblastomas. Clinical cancer research 19, 764-772.
Ostrom, Q.T., Patil, N., Cioffi, G., Waite, K., Kruchko, C., Barnholtz-Sloan, J.S., 2020. CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2013–2017. Neuro-oncology 22, iv1-iv96.
Pai, V.C., Hsu, C.C., Chan, T.S., Liao, W.Y., Chuu, C.P., Chen, W.Y., Li, C.R., Lin, C.Y., Huang, S.P., Chen, L.T., Tsai, K.K., 2019. ASPM promotes prostate cancer stemness and progression by augmenting Wnt-Dvl-3-β-catenin signaling. Oncogene 38, 1340-1353.
Parker, N.R., Khong, P., Parkinson, J.F., Howell, V.M., Wheeler, H.R., 2015. Molecular heterogeneity in glioblastoma: potential clinical implications. Frontiers in oncology 5, 55.
Pascual-García, M., Bonfill-Teixidor, E., Planas-Rigol, E., Rubio-Perez, C., Iurlaro, R., Arias, A., Cuartas, I., Sala-Hojman, A., Escudero, L., Martínez-Ricarte, F., 2019. LIF regulates CXCL9 in tumor-associated macrophages and prevents CD8+ T cell tumor-infiltration impairing anti-PD1 therapy. Nature communications 10, 2416.
Pullen, N.A., Fillmore, H.L., 2010. Induction of matrix metalloproteinase-1 and glioma cell motility by nitric oxide. Journal of Neuro-oncology 96, 201-209.
Ravasz, E., Somera, A.L., Mongru, D.A., Oltvai, Z.N., Barabási, A.-L., 2002. Hierarchical organization of modularity in metabolic networks. science 297, 1551-1555.
Rich, J.T., Neely, J.G., Paniello, R.C., Voelker, C.C., Nussenbaum, B., Wang, E.W., 2010. A practical guide to understanding Kaplan-Meier curves. Otolaryngol Head Neck Surg 143, 331-336.
Santarella, R.A., Koffa, M.D., Tittmann, P., Gross, H., Hoenger, A., 2007. HURP wraps microtubule ends with an additional tubulin sheet that has a novel conformation of tubulin. Journal of molecular biology 365, 1587-1595.
Schober, P., Vetter, T.R., 2018. Survival Analysis and Interpretation of Time-to-Event Data: The Tortoise and the Hare. Anesth Analg 127, 792-798.
Stupp, R., Hegi, M.E., Mason, W.P., Van Den Bent, M.J., Taphoorn, M.J., Janzer, R.C., Ludwin, S.K., Allgeier, A., Fisher, B., Belanger, K., 2009. Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial. The lancet oncology 10, 459-466.
Stupp, R., Taillibert, S., Kanner, A., Read, W., Steinberg, D.M., Lhermitte, B., Toms, S., Idbaih, A., Ahluwalia, M.S., Fink, K., 2017. Effect of tumor-treating fields plus maintenance temozolomide vs maintenance temozolomide alone on survival in patients with glioblastoma: a randomized clinical trial. Jama 318, 2306-2316.
Vakilian, A., Khorramdelazad, H., Heidari, P., Rezaei, Z.S., Hassanshahi, G., 2017. CCL2/CCR2 signaling pathway in glioblastoma multiforme. Neurochemistry international 103, 1-7.
Vange, P., Bruland, T., Beisvag, V., Erlandsen, S.E., Flatberg, A., Doseth, B., Sandvik, A.K., Bakke, I., 2015. Genome‐wide analysis of the oxyntic proliferative isthmus zone reveals ASPM as a possible gastric stem/progenitor cell marker over‐expressed in cancer. The Journal of pathology 237, 447-459.
Wang, J., Yao, L., Zhao, S., Zhang, X., Yin, J., Zhang, Y., Chen, X., Gao, M., Ling, E.-A., Hao, A., 2012. Granulocyte-colony stimulating factor promotes proliferation, migration and invasion in glioma cells. Cancer biology & therapy 13, 389-400.
Wang, J., Zhou, F., Li, Y., Li, Q., Wu, Z., Yu, L., Yuan, F., Liu, J., Tian, Y., Cao, Y., Zhao, Y., Zheng, Y., 2017a. Cdc20 overexpression is involved in temozolomide-resistant glioma cells with epithelial-mesenchymal transition. Cell Cycle 16, 2355-2365.
Wang, M., Liu, K., Zhou, X.-L., Mei, S.-Y., Zhang, C.-J., Zhang, T.-G., 2017b. Downregulation of KIF20A induces cell cycle arrest and apoptosis by suppressing PI3K/AKT in human glioblastoma. Int J Clin Exp Med 10, 16133-16143.
Wang, W.Y., Hsu, C.C., Wang, T.Y., Li, C.R., Hou, Y.C., Chu, J.M., Lee, C.T., Liu, M.S., Su, J.J.M., Jian, K.Y., 2013. A gene expression signature of epithelial tubulogenesis and a role for ASPM in pancreatic tumor progression. Gastroenterology 145, 1110-1120.
Wen, P.Y., Kesari, S., 2008. Malignant gliomas in adults. New England Journal of Medicine 359, 492-507.
West, A.J., Tsui, V., Stylli, S.S., Nguyen, H.P.T., Morokoff, A.P., Kaye, A.H., Luwor, R.B., 2018. The role of interleukin-6-STAT3 signalling in glioblastoma. Oncol Lett 16, 4095-4104.
Wrensch, M., Minn, Y., Chew, T., Bondy, M., Berger, M.S., 2002. Epidemiology of primary brain tumors: current concepts and review of the literature. Neuro-oncology 4, 278-299.
Xue, Q., Cao, L., Chen, X.Y., Zhao, J., Gao, L., Li, S.Z., Fei, Z., 2017. High expression of MMP9 in glioma affects cell proliferation and is associated with patient survival rates. Oncol Lett 13, 1325-1330.
Yun, Z.-J., Wang, H.-J., Yu, Y.-X., Sun, Z.-Y., Yao, S.-K., 2022. Screening of differentially expressed genes for colorectal cancer and prediction of potential traditional Chinese medicine: based on bioinformatics. Zhongguo Zhong yao za zhi= Zhongguo Zhongyao Zazhi= China Journal of Chinese Materia Medica 47, 1666-1676.
Zhang, H., Xu, Y., Deng, G., Yuan, F., Tan, Y., Gao, L., Sun, Q., Qi, Y., Yang, K., Geng, R., Jiang, H., Liu, B., Chen, Q., 2021. SAA1 knockdown promotes the apoptosis of glioblastoma cells via downregulation of AKT signaling. J Cancer 12, 2756-2767.
Zhang, Y., Xia, Q., Lin, J., 2018. Identification of the potential oncogenes in glioblastoma based on bioinformatic analysis and elucidation of the underlying mechanisms. Oncol Rep 40, 715-725.
Zhou, D., Wang, M., Zhang, Y., Wang, K., Zhao, M., Wang, Y., Wang, X., Yu, R., Zhou, X., 2021. Screening and identification of LMNB1 and DLGAP5, two key biomarkers in gliomas. Bioscience Reports 41.
Zhou, T., Wang, Y., Qian, D., Liang, Q., Wang, B., 2018. Over-expression of TOP2A as a prognostic biomarker in patients with glioma. Int J Clin Exp Pathol 11, 1228-1237.

No competing interests reported.

SupplementaryTables.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Deciphering hub genes as potential therapeutic target involved in progression of Glioblastoma multiforme disease: A multi-omics approach

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Data Extraction

Differential Expression Analysis

Pathways and Gene Enrichment Analysis

Protein-protein interaction (PPI) network construction and analysis

Hub Gene Evaluation

Survival Analysis

Results

Identification of differentially expressed genes

PPI network analysis

Hub gene identification

Pathway and gene enrichment analysis

Survival Analysis

Discussion

Limitations

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1