Mapping Glioma Progression: Single-Cell RNA Sequencing Illuminates Cell-Cell Interactions and Immune Response Variability

doi:10.21203/rs.3.rs-4959179/v1

Download PDF

Research Article

Mapping Glioma Progression: Single-Cell RNA Sequencing Illuminates Cell-Cell Interactions and Immune Response Variability

https://doi.org/10.21203/rs.3.rs-4959179/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: Glioma, the most common primary tumor of the central nervous system, is marked by significant heterogeneity, presenting major challenges for therapeutic approaches and prognostic evaluations. This study explores the interactions between malignant glioma cells and macrophages/monocytes and their influence on tumor progression and treatment responses, using comprehensive single-cell RNA sequencing analysis.

Methods: We integrated RNA-seq data from the TCGA and CGGA databases and performed an in-depth analysis of glioma samples using single-cell RNA sequencing, functional enrichment analysis, developmental trajectory analysis, cell-cell communication analysis, and gene regulatory network analysis. Furthermore, we developed a prognostic model based on risk scores and assessed its predictive performance through immune cell infiltration analysis and evaluation of immune treatment responses.

Results: We identified 14 distinct glioma cellular subpopulations and 7 primary cell types, alongside 4 macrophage/monocyte subtypes. Developmental trajectory analysis provided insights into the origins and heterogeneity of both malignant cells and macrophages/monocytes. Cell communication analysis revealed that macrophages and monocytes interact with malignant cells through several pathways, including the MIF (Macrophage Migration Inhibitory Factor) and SPP1 (Secreted Phosphoprotein 1) pathways, engaging in key ligand-receptor interactions that influence tumor behavior. Stratification based on these communication characteristics showed a significant correlation with overall survival (OS). Additionally, immune cell infiltration analysis highlighted variations in immune cell abundance across different subgroups, which may be linked to differing responses to immunotherapy. Our predictive model, consisting of 29 prognostic genes, demonstrated high accuracy and robustness across multiple independent cohorts.

Conclusion: This study unveils the intricate heterogeneity of the glioma microenvironment, enhancing our understanding of the diverse characteristics of glioma cell subpopulations. It also lays the groundwork for the development of therapeutic strategies and prognostic models that specifically target the glioma microenvironment.

Glioma

Single-cell RNA Sequencing

Cell-cell Communication

Prognostic Model

Immune Microenvironment

Glioma is the most common primary brain tumor in the central nervous system, and it is typically characterized by significant heterogeneity, leading to a generally poor prognosis¹. Current treatment options for glioma primarily include surgical resection, radiotherapy, and chemotherapy. However, complete surgical removal is often challenging due to the tumor's invasive nature and its critical location within the brain. Radiotherapy uses high-energy radiation to destroy or inhibit the growth of tumor cells, while chemotherapy employs drugs to target and kill these cells². In addition, emerging therapies such as targeted therapy, immunotherapy, and gene therapy are making rapid strides, offering new hope for improving outcomes^3,4.

In recent years, with the growing understanding of the tumor microenvironment (TME) in cancer, the complex and crucial roles of immune cells within the TME have come into sharper focus. Macrophages, in particular, exhibit dual functionality within the TME. On one hand, they can exert anti-tumor effects by phagocytosing tumor cells and secreting pro-inflammatory cytokines⁵; on the other hand, they can be co-opted by tumor cells to adopt an immunosuppressive M2 phenotype, which secretes anti-inflammatory cytokines and other factors that promote tumor growth. Additionally, macrophages contribute to tumor invasion and metastasis through the secretion of enzymes such as matrix metalloproteinases (MMPs)⁶. Monocytes, another pivotal immune cell type, can migrate into the TME and differentiate into macrophages or other immune cells⁷. They play a significant role in inflammatory responses and immune regulation by secreting cytokines and chemokines, which influence the behavior of tumor cells and the recruitment of other immune cells. Immunomodulatory molecules further add to the complexity of the TME. These molecules include a variety of cytokines, chemokines, immune checkpoint molecules, and enzymes, which can be produced by tumor cells, immune cells, or other cells within the microenvironment. These molecules modulate the activation, proliferation, and function of immune cells, thereby regulating the intensity and duration of the immune response. For example, immune checkpoint molecules like PD-L1 can inhibit T-cell-mediated immune responses⁸, while pro-inflammatory cytokines like TNF-α and IL-1β may promote inflammation and tumor progression⁹,¹⁰. Collectively, macrophages, monocytes, and immunomodulatory molecules play pivotal roles in shaping the tumor and its immune microenvironment, creating a complex regulatory network that deeply impacts tumor development, immune response, and the efficacy of treatments¹¹.

Therefore, it is crucial to delve deeper into the cell-to-cell communication between malignant glioma cells and macrophages/monocytes, as well as to understand its functional implications. In this study, we first elucidate the heterogeneity of glioma in relation to macrophages and monocytes, further clarifying the complex characteristics of their interactions and identifying key communication pathways. We also classify gliomas to explore the variations across different classifications. Finally, we develop a prognostic model aimed at predicting patient survival, offering a potential tool for improving clinical outcomes.

Acquisition of Glioma RNA-seq and scRNA-seq Datasets

We obtained glioma RNA-seq data (TCGA-LGG/GBM), which included 704 tumor samples and 5 normal adjacent samples, from the TCGA database using the TCGAbiolinks R package. Additionally, we retrieved two glioma RNA-seq datasets (CGGA-693, CGGA-325) from the CGGA database website (http://www.cgga.org.cn/), consisting of 693 and 325 tumor samples, respectively. We also acquired four public glioma scRNA-seq datasets (GSE103224, GSE131928, GSE138794, GSE139448) from the TISCH2 database (http://tisch.comp-genomics.org/home/). The TCGA-LGG/GBM dataset was used as the training set for model construction, while the CGGA-693 and CGGA-325 datasets served as validation sets for subsequent model validation.

Single-Cell RNA Sequencing Analysis

Single-cell RNA sequencing analysis was performed using the Seurat R package¹². Single-cell data quality control was conducted based on the criteria of nFeature_RNA < 9000 & percent.mt < 25, with the remaining parameters set to default. After normalization using the "NormalizeData" function, the top 2000 highly variable features for each dataset were identified using the "FindVariableFeatures" function. Batch effect correction was performed using the harmony R package. Dimensionality reduction and visualization were carried out using the heuristic method and the UMAP algorithm. Malignant cells and macrophages/monocytes were identified and annotated based on the annotation information provided by the TISCH database using the "FindClusters" function.

Functional Enrichment Analysis

Functional enrichment analysis was conducted using the RunEnrichment R package¹³ for Gene Ontology (GO) functional enrichment analysis and the clusterProfiler R package¹⁴ for gene set enrichment analysis (GSEA) to assess significant differences between various tumor subtypes. The GSVA R package¹⁵ was utilized for gene set variation analysis (GSVA) to determine the enrichment of pathways in different scoring groups of the scoring model. Cancer hallmarks were obtained from the msigdb database.

Developmental Trajectory Analysis, Cell-cell Communication, and Gene Regulatory Networks (GRNs)

After identifying malignant cells and macrophages/monocytes, developmental trajectory analysis was performed using the RunSlingshot R software package (https://zhanghao-njmu.github.io/SCP/reference/RunSlingshot). The interactions between cell subpopulations and the active ligands on malignant cells and their target cells were calculated using the CellChat & NicheNet R packages¹⁶,¹⁷, based on the expression of known ligands, receptors, and their cofactors. GRN analysis was conducted using the SCENIC R package¹⁸.

Unsupervised Consensus Clustering

Based on the genes characteristic of the communication features of glioma malignant cells and macrophages/monocytes, unsupervised consensus clustering was performed on the TCGA-LGG/GBM dataset using the ConsensusClusterPlus R package¹⁹. A consensus matrix was generated, and samples with high similarity were grouped into a cluster. The optimal number of clusters was determined using the cumulative distribution function (CDF) curve and the partitioning around medoids (PAM) silhouette score.

Immune Cell Infiltration and Immune Treatment Response Analysis

Immune cell infiltration analysis was conducted on bulk RNA-seq datasets using eight TME deconvolution algorithms built into the IOBR R package (https://github.com/IOBR/IOBR): CIBERSORT, TIMER, xCell, MCPcounter, ESTIMATE, EPIC, IPS, quanTIseq. Cytolytic activity (CYT) and IFNG scores were calculated using the ssGSEA algorithm. Additionally, the immune response and scores in the TCGA-LGG/GBM data were predicted using the TIDE online analysis (http://tide.dfci.harvard.edu/).

Acquisition of Immunomodulatory Molecules

A total of 150 immunomodulators and chemokines were downloaded from the TISIDB database (http://cis.hku.hk/TISIDB/), including 41 chemokines, 24 immunoinhibitors, 46 immunostimulators, 21 MHC molecules, and 18 receptors.

Prognostic Model Construction and Validation

In the training set, differentially expressed genes (DEGs) were obtained using the RunDEtest R package (https://zhanghao-njmu.github.io/SCP/reference/RunDEtest). The number of genes was reduced using LASSO regression analysis, followed by the construction of a predictive model using multivariate Cox regression analysis. Glioma samples were divided into high-risk and low-risk groups based on the risk score. The overall survival (OS) between the two groups was determined using the Kaplan-Meier curve. The performance of the model was estimated using the receiver operating characteristic (ROC) analysis and validated in the validation set using the R packages "survival", "survminer" and "timeROC"²⁰,²¹,²².

Statistical Analysis

All statistical analyses were performed using R software (version 4.3.3). For paired independent samples, the Wilcoxon test was used for comparison. The Spearman or Pearson correlation analysis was employed to evaluate the relationship between two continuous variables. Survival differences between the two groups were determined using the Kaplan-Meier curve and the log-rank test. The performance of variables in predicting survival was assessed using the ROC curve. Statistical significance was set at a two-tailed p-value < 0.05.

In-depth Single-Cell RNA Sequencing Analysis of Gliomas

We performed a comprehensive single-cell RNA sequencing analysis of glioma samples by integrating data from four independent datasets (GSE103224, GSE131928, GSE138794, GSE139448), resulting in a compilation of data from 62,670 single cells. After a rigorous standardization and quality control process, we observed that the number of features (nFeature_RNA) and RNA counts (nCount_RNA) ranged from thousands to tens of thousands of transcripts (Fig. 1A). A heuristic approach was used to determine the optimal dimensionality of the datasets (Fig. 1B). From the merged dataset, we selected 2,000 highly variable features, with the top eight being MALAT1, TPSB2, CD74, SPP1, TPSAB1, APOE, and CRYGS (Fig. 1C). UMAP (Uniform Manifold Approximation and Projection) was utilized to display the distribution of single cells from the four datasets in a two-dimensional space (Fig. 1D). Further dimensionality reduction and clustering of the glioma single-cell data allowed us to successfully identify 28 distinct subpopulations (Fig. 1E, F). Through detailed annotation of these subpopulations, we identified 14 major cell types, including Malignant cells, Neurons, Oligodendrocytes, Astrocytes, and others (Fig. 1G). To validate our cell type annotations, we employed violin plots to demonstrate the expression patterns of cell type-specific marker genes. These markers, including CHCHD2P6, CLDN5, C1QB, MDM2, TRBC2, KNG1, MATR3, DLX6-AS1, and CHI3L1, exhibited distinct and high expression levels in their respective cell types (Fig. 1H).

Heterogeneity and Developmental Trajectory Analysis of Malignant Cells and Macrophages/Monocytes in Gliomas

In our detailed examination of the heterogeneity within malignant cell populations in gliomas, we discovered that these cells are not uniform but instead comprise multiple subpopulations with distinct developmental trajectories. The developmental trajectory analysis revealed that NB-like Malignant cells are situated at the basal part of the trajectory, suggesting they may serve as the origin for other subpopulations (Fig. 2A, B). By comparing gene expression across these subpopulations, we identified differentially expressed genes unique to each group. Figure 2C highlights the top 10 differentially expressed genes for seven identified subpopulations. Additionally, we performed Gene Ontology Biological Process (GO_BP) enrichment analysis for each malignant cell subpopulation. Figure 4D illustrates the top five most enriched GO_BP terms for four malignant cell subpopulations, which include processes such as wound healing, detoxification of copper ion, and oxidative phosphorylation. We also conducted a comprehensive analysis of macrophages and monocytes within the glioma microenvironment. The UMAP analysis identified four distinct macrophage and monocyte subtypes, labeled as MacroMono_0, MacroMono_1, MacroMono_2, and MacroMono_3 (Fig. 2E). The developmental trajectory diagram indicated that MacroMono_1 might be the progenitor of the other subtypes (Fig. 2F, G). Furthermore, we identified characteristic genes specific to each macrophage and monocyte subpopulation (Fig. 2H).

Cell Communication and Key Pathways Between Macrophages/Monocytes and Malignant Cells in Gliomas

Our exploration of cell-to-cell communication within the glioma microenvironment revealed a complex network of intercellular signaling pathways between macrophages/monocytes and malignant cells. The number and intensity of these interactions are depicted in Fig. 3A, where MacroMono_1 shows relatively low activity and fewer interactions with other cells. To further investigate the molecular mechanisms underlying these interactions, we conducted an in-depth analysis of ligand-receptor pairs between macrophages/monocytes and malignant cells. Key ligand-receptor interactions identified include MIF binding with CD74 + CXCR4 or CD74 + CD44, and SPP1 binding with CD44 or ITGAV + ITGB1 (Fig. 3B). Further pathway analysis highlighted the MIF signaling pathway's differential expression levels among various cell subpopulations within the glioma tumor microenvironment (Fig. 3C). Figures 3D and 3E underscore the significant role and higher intensity of MacroMono_0, MacroMono_2, and MES-like malignant cells in the MIF signaling pathway, suggesting their dominant influence and potential impact on tumor progression and treatment response. Additionally, our study identified the SPP1 signaling network as a critical component of the interaction between macrophages/monocytes and malignant cells, with notable activity in MacroMono_0, MacroMono_3, and MES-like malignant cells (Figs. 3F, G, H).

Ligand-Receptor-Target Networks and Gene Regulatory Networks (GRNs) in the Tumor Microenvironment (TME) of Gliomas

Utilizing the NicheNet tool, we analyzed ligands expressed by macrophages and monocytes within the glioma tumor microenvironment. Several ligands, including TGFB1, TGAM, VEGFA, HBEGF, IL1B, and PLXNB2, were identified as highly associated, with their expression levels in the TME potentially having a significant impact on tumor cell behavior. The interactions between ligand-receptor pairs from macrophages/monocytes and malignant cells offer insights into how these immune cells communicate with tumor cells via ligand secretion, potentially influencing the biological functions of the malignant cells. For instance, TGFB1 secreted by macrophages/monocytes can bind to EGFR and ERBB3 receptors on malignant cells, thereby activating downstream signaling pathways that promote tumor progression. A heatmap visualizes the regulatory potential of the top-ranked ligands and their downstream target genes in malignant cells. Notably, TGFB1 not only affects the expression of its direct receptors but also modulates other genes involved in tumor growth and immune regulation, such as SERPING1 and HLA-A (Fig. 4A). Further investigation into the gene regulatory networks (GRNs) in gliomas revealed distinct regulatory modules, including IKZF1_extended_30g, HDAC2_extended_141g, HDAC2_52g, and ZNF454_extended_69g, which exhibit varying activities and functions across different cell types (Figs. 4B, C).

Subgroup Stratification of the Tumor Microenvironment (TME) in Gliomas and Its Prognostic and Biological Characteristics

Based on the identified subpopulation characteristics, we employed an unsupervised consensus clustering method to stratify gliomas into two subpopulations. The optimal number of clusters was determined using the CDF curve and PAC scores (Figs. 5A, B, C). Survival analysis indicated that the second cluster was associated with a poorer overall survival (OS) (Fig. 5D). Using the ssGSEA method, we further evaluated the relative infiltration abundance of 28 immune cell subpopulations within the two glioma subpopulations. The results showed that, compared to C1, C2 exhibited a significantly higher number of infiltrating immune cells, all with statistical significance (Fig. 5E). Additionally, analysis of immune therapy response between the two clusters revealed that the activity of three immune therapy predictive factors—IFN-γ, GEP, and CYT—was significantly elevated in C2 (Fig. 5F).

Subgroup Stratification of the Tumor Microenvironment (TME) in Gliomas and Its Distinct TME Characteristics

Utilizing various algorithms, including CIBERSORT, MCP-counter, quanTIseq, EPIC, and TIMER, we assessed the infiltration abundance of immune cell subpopulations within the two glioma subpopulations. The analysis revealed that several immune cell subpopulations, such as macrophages, exhibited high infiltration levels in C2, while NK cells were more prominently expressed in C1 (Fig. 6A). GSVA was employed to evaluate the activity of the anti-cancer immune cycle and immune therapy predictive pathways between the two subpopulations. The results demonstrated significant differences in the activity of the anti-cancer immune cycle, with C2 showing a markedly higher enrichment score compared to C1. Additionally, there were notable differences in the activity of several immune therapy-related pathways, where C1 had significantly lower enrichment scores than C2, possibly reflecting variations in their anti-tumor immune responses (Figs. 6B, C). Figure 6D presents the expression levels of 150 immune regulatory factors across the two subpopulations. Moreover, GSEA identified significant differences in the activity of various cancer-related pathways between the two subpopulations, such as the interferon response, muscle generation, estrogen response, inflammatory response, and KRAS signaling pathway (Fig. 6E).

Glioma Prognostic Model Exhibits High Accuracy and Robust Performance in Predicting Prognosis

We further developed a prognostic model for glioma using LASSO regression analysis (Fig. 7A). The model ultimately incorporated 29 relevant genes, with the coefficients for each gene illustrated in Fig. 7B. Based on the risk scores, patients from the TCGA dataset were stratified into two groups. Survival analysis indicated that the low-risk group had significantly better survival outcomes. The prognostic model was also validated using the CGGA-693 and CGGA-325 datasets, demonstrating excellent predictive capability and accuracy. The ROC values at 1, 3, and 5 years were all above 0.9, underscoring the high accuracy and robust performance of our prognostic model in predicting patient outcomes (Figs. 7C, D, E).

Association of Risk Score with Immune Checkpoints, Immune Cell Infiltration Levels, and Upregulated and Downregulated Pathways in Glioma Patients

A high-risk score was found to be negatively correlated with inhibitory immune checkpoints and the abundance of immune cell infiltration (Figs. 8A, B). Key immune checkpoints, such as VTCN1, TIGIT, CD200R1, and BTLA, are crucial molecules that modulate immune responses and play significant roles in tumor immune evasion and immunotherapy. The negative correlation between a high-risk score and these immune checkpoints suggests that these checkpoints may have a positive role in the anti-glioma immune response (Fig. 8A). Additionally, a high-risk score was negatively correlated with the abundance of immune cell infiltration, providing insights into the role of different cell types in glioma progression and patient prognosis (Fig. 8B). Pathway analysis revealed that in high-risk glioma patients, the upregulated pathways predominantly involved Staphylococcus aureus infection, Systemic Lupus Erythematosus, and ECM-receptor interaction. In contrast, the downregulated pathways were mainly related to Nicotine Addiction, Glutamatergic Synapse, and GABAergic Synapse (Figs. 8C, D).

In this study, we conducted an in-depth single-cell RNA sequencing analysis to explore the intricate heterogeneity of glioma, focusing particularly on malignant cells and macrophages/monocytes within the tumor microenvironment. Our results not only reaffirm the diversity of immune cells previously observed in the glioma microenvironment but also, through developmental trajectory analysis, suggest possible origins and differentiation pathways for malignant cells, providing new perspectives on the biological behavior of glioma.

The subpopulations of malignant cells and the distinct macrophage/monocyte subtypes we identified highlight the significant heterogeneity within the glioma microenvironment. Notably, the identification of the NB-like Malignant cell population suggests its potential role as an initiator in tumorigenesis. This aligns with the cancer stem cell theory, which posits that tumor-initiating cells are pivotal in tumor initiation, progression, and therapeutic resistance²³. Additionally, our developmental trajectory analysis offers insights into the differentiation pathways of malignant cells, consistent with findings by Verhaak et al., who identified molecular subtypes of glioma through large-scale genomic analysis²⁴.

Our analysis of cell-to-cell communication uncovered the complex interactions between macrophages/monocytes and malignant cells. The identification of the MIF and SPP1 signaling pathways, in particular, offers new targets for glioma therapy. The immunomodulatory role of MIF in various tumors is well-documented, and our study further underscores its critical function within the glioma microenvironment²⁵,²⁶. Similarly, SPP1, an important extracellular matrix protein, has been implicated in the adhesion, migration, and invasion of tumor cells²⁷,²⁸.

Our investigation into gene regulatory networks (GRNs) revealed several transcription factors, such as IKZF1 and HDAC2, that may play crucial roles in glioma development. The aberrant activity of these transcription factors is closely linked to the proliferation, survival, and invasion of tumor cells, making them potential therapeutic targets for glioma.

Using unsupervised consensus clustering, we stratified gliomas into two subgroups with distinct survival outcomes. The higher levels of immune cell infiltration and enhanced immune treatment response in the C2 subgroup suggest that it may be more responsive to immunotherapy. This observation is consistent with the findings of Economopoulou et al., who emphasized the role of the immune microenvironment in tumor immune evasion and response to immunotherapy²⁹.

The prognostic model we developed demonstrated high accuracy and robustness in predicting glioma patient survival. This model not only serves as a valuable tool for clinical management but also identifies genes that could be targets for future research and drug development. The abnormal expression of these genes is closely associated with glioma invasiveness, therapeutic resistance, and prognosis³⁰,³¹,³²,³³,³⁴,³⁵. Moreover, the negative correlation between high-risk scores and immune checkpoints and immune cell infiltration levels highlights the significant role of the immune microenvironment in glioma prognosis. This finding aligns with the concept of immune editing, which suggests that tumors can evade immune surveillance by modulating the immune microenvironment³⁶. Additionally, the expression of immune checkpoint molecules may serve as biomarkers for predicting patient responsiveness to immunotherapy, offering opportunities for personalized treatment.

Despite the valuable insights gained from our study on glioma heterogeneity and the interactions within the tumor microenvironment, there are several limitations. Firstly, the sample size in our study is relatively limited, and future research should validate our findings with a larger cohort and across more diverse populations in terms of ethnicity and geographic distribution. Secondly, our results need to be corroborated by prospective clinical trials to ensure their robustness. Future studies should also delve deeper into the interactions between malignant cell subpopulations and macrophage/monocyte subtypes and how these interactions influence glioma progression and treatment response.

In conclusion, our study, leveraging single-cell RNA sequencing and comprehensive analysis, has illuminated the complex interactions between malignant cells and immune cells within the glioma microenvironment. The prognostic model we developed, alongside our analysis of cell-cell communication and gene regulatory networks, offers new strategies for personalized glioma treatment. These findings not only enhance our understanding of glioma heterogeneity but also provide valuable information for future research and therapeutic approaches. However, our research does come with certain limitations: (1) The dataset size used in this study was limited, and while we have drawn meaningful conclusions from the available data, a larger dataset would likely yield more robust and generalizable results; (2) The study lacks external validation, as the findings have not been corroborated by additional independent studies, which could impact the strength and reliability of our conclusions; (3) The results may not be broadly generalizable due to the specific characteristics of the dataset, and further research with more diverse samples is needed to confirm the applicability of these findings across different contexts.

ACKNOWLEDGMENTS

This work was supported by the South China Sea Rising Star Science and Technology Innovation Talent Platform Project(NHXXRCXM202351).

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Nicholson JG, Fine HA. Diffuse Glioma Heterogeneity and Its Therapeutic Implications. Cancer Discov. 2021;11(3):575-590. doi:10.1158/2159-8290.CD-20-1474
van den Bent MJ, Geurts M, French PJ, et al. Primary brain tumours in adults. Lancet Lond Engl. 2023;402(10412):1564-1579. doi:10.1016/S0140-6736(23)01054-1
Yang K, Wu Z, Zhang H, et al. Glioma targeted therapy: insight into future of molecular approaches. Mol Cancer. 2022;21(1):39. doi:10.1186/s12943-022-01513-z
Yasinjan F, Xing Y, Geng H, et al. Immunotherapy: a promising approach for glioma treatment. Front Immunol. 2023;14:1255611. doi:10.3389/fimmu.2023.1255611
Viola A, Munari F, Sánchez-Rodríguez R, Scolaro T, Castegna A. The Metabolic Signature of Macrophage Responses. Front Immunol. 2019;10:1462. doi:10.3389/fimmu.2019.01462
Yu-Ju Wu C, Chen CH, Lin CY, et al. CCL5 of glioma-associated microglia/macrophages regulates glioma migration and invasion via calcium-dependent matrix metalloproteinase 2. Neuro-Oncol. 2020;22(2):253-266. doi:10.1093/neuonc/noz189
Jakubzick CV, Randolph GJ, Henson PM. Monocyte differentiation and antigen-presenting functions. Nat Rev Immunol. 2017;17(6):349-362. doi:10.1038/nri.2017.28
Li CW, Lim SO, Xia W, et al. Glycosylation and stabilization of programmed death ligand-1 suppresses T-cell activity. Nat Commun. 2016;7:12632. doi:10.1038/ncomms12632
Wang Y, Che M, Xin J, Zheng Z, Li J, Zhang S. The role of IL-1β and TNF-α in intervertebral disc degeneration. Biomed Pharmacother Biomedecine Pharmacother. 2020;131:110660. doi:10.1016/j.biopha.2020.110660
Shen Y, Malik SA, Amir M, et al. Decreased Hepatocyte Autophagy Leads to Synergistic IL-1β and TNF Mouse Liver Injury and Inflammation. Hepatol Baltim Md. 2020;72(2):595-608. doi:10.1002/hep.31209
Fendl B, Berghoff AS, Preusser M, Maier B. Macrophage and monocyte subsets as new therapeutic targets in cancer immunotherapy. ESMO Open. 2023;8(1):100776. doi:10.1016/j.esmoop.2022.100776
Hao Y, Hao S, Andersen-Nissen E, et al. Integrated analysis of multimodal single-cell data. Cell. 2021;184(13):3573-3587.e29. doi:10.1016/j.cell.2021.04.048
Cogswell JP, Ward J, Taylor IA, et al. Identification of miRNA changes in Alzheimer’s disease brain and CSF yields putative biomarkers and insights into disease pathways. J Alzheimers Dis JAD. 2008;14(1):27-41. doi:10.3233/jad-2008-14103
Wu T, Hu E, Xu S, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innov Camb Mass. 2021;2(3):100141. doi:10.1016/j.xinn.2021.100141
Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics. 2013;14:7. doi:10.1186/1471-2105-14-7
Jin S, Guerrero-Juarez CF, Zhang L, et al. Inference and analysis of cell-cell communication using CellChat. Nat Commun. 2021;12(1):1088. doi:10.1038/s41467-021-21246-9
Browaeys R, Saelens W, Saeys Y. NicheNet: modeling intercellular communication by linking ligands to target genes. Nat Methods. 2020;17(2):159-162. doi:10.1038/s41592-019-0667-5
Aibar S, González-Blas CB, Moerman T, et al. SCENIC: single-cell regulatory network inference and clustering. Nat Methods. 2017;14(11):1083-1086. doi:10.1038/nmeth.4463
Wilkerson MD, Hayes DN. ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinforma Oxf Engl. 2010;26(12):1572-1573. doi:10.1093/bioinformatics/btq170
Therneau TM, until 2009) TL (original S >R port and R maintainer, Elizabeth A, Cynthia C. survival: Survival Analysis. Published online February 14, 2024. Accessed April 19, 2024. https://cran.r-project.org/web/packages/survival/index.html
Kassambara A, Kosinski M, Biecek P, Fabian S. survminer: Drawing Survival Curves using “ggplot2.” Published online March 9, 2021. Accessed April 19, 2024. https://cran.r-project.org/web/packages/survminer/index.html
Blanche P. timeROC: Time-Dependent ROC Curve and AUC for Censored Survival Data. Published online December 18, 2019. Accessed April 19, 2024. https://cran.r-project.org/web/packages/timeROC/index.html
Biserova K, Jakovlevs A, Uljanovs R, Strumfa I. Cancer Stem Cells: Significance in Origin, Pathogenesis and Treatment of Glioblastoma. Cells. 2021;10(3):621. doi:10.3390/cells10030621
Verhaak RGW, Hoadley KA, Purdom E, et al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell. 2010;17(1):98-110. doi:10.1016/j.ccr.2009.12.020
Xu L, Li Y, Sun H, et al. Current developments of macrophage migration inhibitory factor (MIF) inhibitors. Drug Discov Today. 2013;18(11-12):592-600. doi:10.1016/j.drudis.2012.12.013
Kang I, Bucala R. The immunobiology of MIF: function, genetics and prospects for precision medicine. Nat Rev Rheumatol. 2019;15(7):427-437. doi:10.1038/s41584-019-0238-2
Sathe A, Mason K, Grimes SM, et al. Colorectal Cancer Metastases in the Liver Establish Immunosuppressive Spatial Networking between Tumor-Associated SPP1+ Macrophages and Fibroblasts. Clin Cancer Res Off J Am Assoc Cancer Res. 2023;29(1):244-260. doi:10.1158/1078-0432.CCR-22-2041
Gao W, Liu D, Sun H, et al. SPP1 is a prognostic related biomarker and correlated with tumor-infiltrating immune cells in ovarian cancer. BMC Cancer. 2022;22(1):1367. doi:10.1186/s12885-022-10485-8
Economopoulou P, Kotsantis I, Psyrri A. Tumor Microenvironment and Immunotherapy Response in Head and Neck Cancer. Cancers. 2020;12(11):3377. doi:10.3390/cancers12113377
Li X, Chen G, Liu B, et al. PLK1 inhibition promotes apoptosis and DNA damage in glioma stem cells by regulating the nuclear translocation of YBX1. Cell Death Discov. 2023;9(1):68. doi:10.1038/s41420-023-01302-7
Wang F, Zhao F, Zhang L, et al. CDC6 is a prognostic biomarker and correlated with immune infiltrates in glioma. Mol Cancer. 2022;21(1):153. doi:10.1186/s12943-022-01623-8
Zheng XJ, Chen WL, Yi J, et al. Apolipoprotein C1 promotes glioblastoma tumorigenesis by reducing KEAP1/NRF2 and CBS-regulated ferroptosis. Acta Pharmacol Sin. 2022;43(11):2977-2992. doi:10.1038/s41401-022-00917-3
Luo H, Huang K, Cheng M, Long X, Zhu X, Wu M. The HNF4A-CHPF pathway promotes proliferation and invasion through interactions with MAD1L1 in glioma. Aging. 2023;15(20):11052-11066. doi:10.18632/aging.205076
Zhao L, Song C, Li Y, et al. BZW1 as an oncogene is associated with patient prognosis and the immune microenvironment in glioma. Genomics. 2023;115(3):110602. doi:10.1016/j.ygeno.2023.110602
Li T, Yang W, Li M, et al. Engrailed 2 (EN2) acts as a glioma suppressor by inhibiting tumor proliferation/invasion and enhancing sensitivity to temozolomide. Cancer Cell Int. 2020;20:65. doi:10.1186/s12935-020-1145-y
Dunn GP, Bruce AT, Ikeda H, Old LJ, Schreiber RD. Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol. 2002;3(11):991-998. doi:10.1038/ni1102-991

No competing interests reported.

Download PDF

Editorial decision: Revision requested
20 Sep, 2024
Reviews received at journal
18 Sep, 2024
Reviews received at journal
17 Sep, 2024
Reviews received at journal
17 Sep, 2024
Reviewers agreed at journal
11 Sep, 2024
Reviews received at journal
11 Sep, 2024
Reviewers agreed at journal
10 Sep, 2024
Reviewers agreed at journal
09 Sep, 2024
Reviewers agreed at journal
06 Sep, 2024
Reviewers agreed at journal
06 Sep, 2024
Reviewers invited by journal
06 Sep, 2024
Editor assigned by journal
03 Sep, 2024
Submission checks completed at journal
30 Aug, 2024
First submitted to journal
22 Aug, 2024

You are reading this latest preprint version

Mapping Glioma Progression: Single-Cell RNA Sequencing Illuminates Cell-Cell Interactions and Immune Response Variability

Status:

Version 1

Abstract

Figures

Introduction

Methods

Acquisition of Glioma RNA-seq and scRNA-seq Datasets

Single-Cell RNA Sequencing Analysis

Functional Enrichment Analysis

Developmental Trajectory Analysis, Cell-cell Communication, and Gene Regulatory Networks (GRNs)

Unsupervised Consensus Clustering

Immune Cell Infiltration and Immune Treatment Response Analysis

Acquisition of Immunomodulatory Molecules

Prognostic Model Construction and Validation

Statistical Analysis

Results

In-depth Single-Cell RNA Sequencing Analysis of Gliomas

Heterogeneity and Developmental Trajectory Analysis of Malignant Cells and Macrophages/Monocytes in Gliomas

Cell Communication and Key Pathways Between Macrophages/Monocytes and Malignant Cells in Gliomas

Ligand-Receptor-Target Networks and Gene Regulatory Networks (GRNs) in the Tumor Microenvironment (TME) of Gliomas

Subgroup Stratification of the Tumor Microenvironment (TME) in Gliomas and Its Prognostic and Biological Characteristics

Subgroup Stratification of the Tumor Microenvironment (TME) in Gliomas and Its Distinct TME Characteristics

Glioma Prognostic Model Exhibits High Accuracy and Robust Performance in Predicting Prognosis

Discussion

Declarations

References

Additional Declarations

Status:

Version 1