Single-Cell Sequencing Reveals the Role of NUSAP1 in Glioma and Its Potential for Precision Diagnostic and Prognostic Applications

doi:10.21203/rs.3.rs-4948171/v1

Download PDF

Research Article

Single-Cell Sequencing Reveals the Role of NUSAP1 in Glioma and Its Potential for Precision Diagnostic and Prognostic Applications

https://doi.org/10.21203/rs.3.rs-4948171/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: Personalized precision medicine (PPPM) is a rapidly advancing field with significant potential. Gliomas, known for their poor prognosis, rank among the most lethal brain tumors. Despite advancements, there remains a critical need for precise, individualized treatment strategies.

Methods: We conducted a comprehensive analysis of RNA-seq and microarray data from the TCGA and GEO databases, supplemented by single-cell RNA sequencing (scRNA-seq) data from glioma patients. By integrating single-cell sequencing analysis with foundational experiments, we investigated the molecular variations and cellular interactions within neural glioma cell subpopulations during tumor progression.

Results: Our single-cell sequencing analysis revealed distinct gene expression patterns across glioma cell subpopulations. Notably, differentiation trajectory analysis identified NUSAP1 as a key marker for the terminal subpopulation. We found that elevated NUSAP1 expression correlated with poor prognosis, prompting further investigation of its functional role through both cellular and animal studies.

Conclusions: NUSAP1-based risk models hold potential as predictive and therapeutic tools for personalized glioma treatment. In-depth exploration of NUSAP1's mechanisms in glioblastoma could enhance our understanding of its response to immunotherapy, suggesting that targeting NUSAP1 may offer therapeutic benefits for glioma patients.

glioma

NUSAP1

diagnosis

prognosis

immune infiltration

tumor immune microenvironment

Gliomas represent the most common malignant tumors within the central nervous system, accounting for more than 30% of all primary brain tumors, and are characterized by high rates of incidence and mortality[1, 2]. Although the precise cause remains unknown, genetic predispositions, environmental factors, and gene mutations are thought to significantly contribute to glioma development[3, 4]. Pathologically, gliomas are highly heterogeneous and aggressive, especially in high-grade forms like glioblastoma (GBM), known for rapid growth and the ability to invade surrounding brain tissue[5, 6]. Despite standard treatments such as surgery, radiotherapy, and chemotherapy, the prognosis for most glioma patients is bleak, with a five-year survival rate below 10%[7]. Additionally, gliomas' resistance to conventional therapies highlights the urgent need for new treatment strategies to enhance patient outcomes[8].

Nucleosome- and spindle-associated protein 1 (NUSAP1) is a microtubule-associated protein essential for cell cycle regulation, particularly during mitosis, where it governs spindle assembly and stability to ensure proper chromosome segregation[9, 10]. Recently, abnormal NUSAP1 expression has gained significant attention in various cancers[11]. Studies have shown that elevated NUSAP1 expression is closely linked to heightened tumor cell proliferation, migration, and invasion, which is associated with a poor prognosis in various solid tumors, such as breast, prostate, and lung cancers. In gliomas, NUSAP1's role is becoming increasingly clear, with early studies suggesting that its elevated expression in glioma cells may drive malignant behaviors, such as invasiveness and resistance to apoptosis. Additionally, research into NUSAP1 as a potential therapeutic target is growing, indicating that modulating its expression and function could provide new treatment options for gliomas.

Single-cell sequencing technology, a state-of-the-art high-throughput analytical tool, has greatly advanced predictive, preventive, and personalized medicine (PPPM). Single-cell sequencing facilitates an in-depth examination of individual cells across genomic, transcriptomic, and epigenomic dimensions, uncovering cellular heterogeneity, identifying rare disease-related cell populations, and elucidating their molecular profiles. This approach provides essential insights for prediction and prevention in medical research. In PPPM, single-cell sequencing is widely used for early disease diagnosis and the development of personalized treatment plans. In oncology, for example, single-cell sequencing can uncover the complexity of the tumor microenvironment, identify resistant clones and potential therapeutic targets, and customize precise treatment regimens. This technology also plays a crucial role in monitoring treatment responses and evaluating recurrence risks, aiding clinicians in optimizing therapeutic strategies and improving long-term survival rates. As a result, single-cell sequencing has become an essential tool in shifting from traditional treatment models to precision medicine, especially in cancer care.

2.1 Data Acquisition

Single-cell RNA sequencing (scRNA-seq) data was accessed from the Gene Expression Omnibus (GEO) database under accession number GSE141383 (https://www.ncbi.nlm.nih.gov/geo/), Bulk RNA-seq data was retrieved from The Cancer Genome Atlas (TCGA) through the official GDC portal (https://portal.gdc.cancer.gov/).

2.2 Data Filtering and Processing

Each dataset was processed using the Seurat package (v4.3.0) in R (v4.2.2). Initially, potential doublets were excluded using the DoubletFinder package (v2.0.3), and low-quality cells were filtered out, maintaining cell quality within the following ranges: 300 < nFeature < 6000 and 500 < nCount < 100,000. Cells meeting these criteria were retained for further analysis. Additionally, cells were required to have less than 25% mitochondrial gene expression and less than 5% red blood cell gene expression to be included in subsequent analyses [12, 13].

Next, the expression matrix was normalized, and the top 2000 highly variable genes (HVGs) were identified and standardized. PCA analysis was then performed on these genes. To address batch effects across samples, the Harmony package (v0.1.1) was utilized, selecting the top 30 principal components (PCs) for dimensionality reduction and clustering [14–17]. Following this, UMAP was employed to project the results onto a two-dimensional plot, facilitating cell type identification. Relevant cell markers from the literature were used to annotate cell clusters, thereby identifying distinct cell types [18, 19], and examining their distribution and proportions [20].

2.3 Enrichment Analysis of Differentially Expressed Genes

Differentially expressed genes (DEGs) for each cell type were identified using the "FindAllMarkers" function with default settings, employing the Wilcoxon rank-sum test. Genes expressed in over 25% of cells within clusters and exhibiting a log fold change (logFC) greater than 0.25 were selected. Enrichment analysis of differentially expressed genes (DEGs) was performed using the clusterProfiler package, with an emphasis on pathways related to each cell type as defined by Gene Ontology (GO) Biological Processes (BP) [15, 21].

2.4 Subpopulation Analysis of Glia/Neuronal Cells

To investigate the heterogeneity among glial and neuronal cells, we performed additional stratification. Following cell isolation, we identified the top 2,000 highly variable genes and proceeded with data normalization. The Harmony method was applied to minimize batch effects across samples. We selected the top 30 principal components (PCs) for downsampling and clustering, utilizing UMAP to project the data onto a two-dimensional map, facilitating the investigation of intercellular heterogeneity.

2.5 Malignant Cell Identification via InferCNV

Copy number variation (CNV) analysis was employed to distinguish malignant cells from non-tumor cells. By applying the inferCNV algorithm, we evaluated copy number variability across cell subpopulations, using vascular endothelial cells as a reference. Subpopulations with significant CNV alterations were classified as glioma cells.

2.6 Differential and Enrichment Analyses of Cell Subpopulations

Subsequently, the "FindAllMarkers" function was utilized to identify differentially expressed genes within each subpopulation using the Wilcoxon rank sum test. Following this, Gene Ontology Biological Process (GO-BP) enrichment analysis was performed with the clusterProfiler package.

2.7 Trajectory analysis

Three distinct software packages were employed to evaluate the differentiation progression across oligodendrocyte subpopulations. Initially, the cytoTRACE algorithm was utilized to assess the stemness levels within each subpopulation. Subsequently, differentiation trajectories were mapped using the Monocle R package (version 2.24.0), with the DDRTree algorithm employed to trace developmental progress along predetermined pathways. Finally, we conducted additional trajectory analysis during glioma differentiation using the Slingshot package [22]. Minimum spanning trees (MSTs) were employed to infer cell lineages using the getLineages function, while the getCurves function was utilized to estimate temporal changes in gene expression within each lineage [23].

2.8 Cell Communication Analysis

To investigate the intricate intercellular communication among distinct cell subpopulations in GBM tumors and their interactions with the tumor microenvironment, we conducted a cross-talk analysis. The CellChat R package (version 1.6.1) and the CellChatDB.human reference database were utilized to examine ligand-receptor interactions. This analysis provided insights into cell-cell communication by evaluating signaling pathways and receptor-ligand interactions, revealing coordinated interactions among various cell types [24].

2.9 Prognostic Modeling of Glioma-Associated Cells

To assess the prognostic impact of glioma-associated cell subpopulations on patient survival, we identified key marker genes as predictive signatures for gliomas. We employed the survival R package to perform univariate Cox regression and Lasso regression for selecting significant prognostic genes. These genes were then incorporated into a prognostic model through multivariate Cox analysis. A risk score for each sample was computed using the following formula: risk score = (expression of gene 1 * coefficient 1) + (expression of gene 2 * coefficient 2) + ... + (expression of gene n * coefficient n). Samples were categorized into high-risk and low-risk groups according to the median risk score. We then performed survival analysis to assess patient prognosis across these groups. The model's accuracy was assessed using receiver operating characteristic (ROC) curves for 1, 3, and 5 years, employing the timeROC software package (version 0.4.0). Additionally, we examined the correlations between the prognostic genes, risk scores, and overall survival (OS).

2.10 Analysis of Tumor-Infiltrating Immune Cells

To thoroughly evaluate the immune microenvironment in patients, we employed CIBERSORT, ESTIMATE, and xCell to derive various immune-related scores. We assessed the levels of 22 immune cell types using the CIBERSORT algorithm, comparing high and low levels between groups. Additionally, we investigated the interactions among immune cells, risk scores, model genes, and overall survival (OS). We also analyzed TIDE scores and the expression of AODRA2A across different groups.

2.11 Differential and Enrichment Analyses of Bulk Genomic Data

We performed differential analysis separately for high-risk and low-risk groups using the DESeq2 R package, applying a threshold of |logFC| > 2 and a p-value < 0.05. Additionally, we conducted Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA) on the differentially expressed genes using the clusterProfiler package [25, 26].

2.12 Somatic Mutation Analysis

Somatic mutation data from the TCGA repository were utilized to assess the mutational landscape of highly mutated genes relative to reference genes. Glioma samples were stratified into high and low groups based on the median tumor mutational burden (TMB). Kaplan-Meier survival analysis was subsequently conducted to compare survival outcomes between these groups. Additionally, we investigated the copy number variation (CNV) patterns of the targeted genes.

2.13 Drug Sensitivity Analysis

IC50 values for various chemotherapeutic agents were estimated using the pRRophetic R software (version 0.5). Sensitivity evaluations for these agents were then conducted across different categories [27, 28].

2.14 Cell culture

U-251 and LN229 cell lines were cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. Cells were maintained at 37°C in a humidified atmosphere with 5% CO2. For subculturing, cells were detached using trypsin-EDTA, counted, and reseeded at a density of 1 x 10^5 cells per flask. Cultures were routinely checked for mycoplasma contamination and passaged when they reached 80–90% confluency.

2.15 Cell Transfection, Lentivirus Vector Construction, and Cell Infection

Cells were seeded in 6-well plates at 70–80% confluence and transfected with plasmid DNA using Lipofectamine 3000 (Invitrogen) according to the manufacturer’s protocol. After 24 hours, the medium was replaced (sequences in Supplementary Table 1). Lentiviral vectors were constructed by cloning target genes into a lentiviral expression plasmid. The plasmid was co-transfected with packaging plasmids (pMD2.G and psPAX2) into HEK293T cells using Lipofectamine 3000. Viral supernatants were collected 48–72 hours post-transfection, filtered, and concentrated using a lentivirus concentration kit (System Biosciences). Target cells were infected with lentivirus in the presence of 8 µg/mL polybrene (Sigma). After 24 hours, the virus-containing medium was replaced with fresh growth medium. Infection efficiency was assessed 48 hours later.

2.16 RT-qPCR Analysis

Total RNA was extracted from cells using the RNeasy Mini Kit (Qiagen) and converted to cDNA using the PrimeScript RT Reagent Kit (Takara) according to the manufacturer's instructions. Quantitative PCR was performed using SYBR Green Master Mix (Applied Biosystems) on a real-time PCR system (Applied Biosystems 7500). The cycling conditions included an initial denaturation at 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute. Gene expression levels were normalized to GAPDH, and relative expression was calculated using the ΔΔCt method. Statistical analysis was conducted, and p-values < 0.05 were considered significant, with primer sequences provided in Supplementary Table 1.

2.17 Cell Counting Kit-8 (CCK-8) Assay

Cell viability was assessed using the Cell Counting Kit-8 (CCK-8, Dojindo). Cells were seeded in 96-well plates at a density of 2 x 10³ cells per well and allowed to adhere overnight. After treatment, 10 µL of CCK-8 solution was added to each well, and cells were incubated for 2 hours at 37°C. The absorbance at 450 nm was measured using a microplate reader. Data were normalized to control wells, and statistical analysis was performed with p-values < 0.05 considered significant.

2.18 Colony Formation Assay

Cell proliferation was evaluated using the Cell Counting Kit-8 (CCK-8, Dojindo). Cells were seeded in 96-well plates at a density of 2 x 10³ cells per well. Following the appropriate treatment, 10 µL of CCK-8 solution was added to each well, and the cells were incubated for 2 hours at 37°C. Absorbance was measured at 450 nm using a microplate reader, and cell viability was calculated relative to the control group. Statistical significance was determined with p-values < 0.05 considered significant.

2.19 Wound Healing Assay

Wound healing assays were conducted by creating a uniform scratch in a confluent cell monolayer using a sterile pipette tip. Cells were then washed to remove debris and incubated in fresh medium. Wound closure was monitored at 0 and 24 hours using a phase-contrast microscope. Images were captured, and the wound area was measured using ImageJ software. The percentage of wound closure was calculated relative to the initial wound area, and statistical analysis was performed with p-values < 0.05 considered significant.

2.20 Transwell Migration Assay

Cell migration was assessed using a Transwell migration assay. Cells were suspended in serum-free medium and seeded into the upper chamber of Transwell inserts with an 8 µm pore size (Corning). The lower chamber was filled with medium containing 10% fetal bovine serum as a chemoattractant. After 24 hours of incubation at 37°C, non-migrated cells on the upper membrane were removed with a cotton swab, and migrated cells on the lower membrane were fixed with 4% paraformaldehyde and stained with crystal violet. The number of migrated cells was counted under a light microscope, and statistical analysis was performed with p-values < 0.05 considered significant.

2.21 Flow Cytometric Analysis of Apoptosis

Apoptosis was assessed by flow cytometry using the Annexin V-FITC/PI Apoptosis Detection Kit (BD Biosciences). Cells were harvested, washed with cold PBS, and resuspended in binding buffer. Annexin V-FITC and propidium iodide (PI) were added to the cell suspension, followed by incubation in the dark for 15 minutes at room temperature. Stained cells were analyzed on a flow cytometer (BD FACSCanto II) within one hour, and data were processed using FlowJo software. The percentages of early apoptotic (Annexin V+/PI-) and late apoptotic/necrotic (Annexin V+/PI+) cells were quantified. Statistical significance was determined with p-values < 0.05 considered significant.

2.22 Animal Experiment

In the metastasis model, 18 mice were randomly assigned to three groups: sh-NUSAP1-NC, sh-NUSAP1-1, and sh-NUSAP1-2 (n = 3 per group). Stable LN229 and U251 cells (1 × 10⁶ cells in 100 µl of PBS) were administered to the mice via tail vein injection. Five weeks after the injection, distant metastasis was evaluated using the IVIS imaging system (Caliper Life Sciences, Hopkinton, MA, USA). This research was conducted with the approval of the Anhui Animal Care and Use Committee at Anhui Medical University.

2.22 Statistical analysis

Biological analyses were performed using R software version 4.1.3, while experimental data analysis was conducted with GraphPad Prism version 8.0. Mean values and standard deviations were derived from three independent experiments. Comparisons between two groups were made using Student's t-test, whereas one-way ANOVA followed by Tukey's post hoc test was employed for comparisons among multiple groups. Statistical significance was indicated as follows: *p < 0.05, **p < 0.01, ***p < 0.001.

3.1 Key Cell Types Involved in Glioma Progression Identified via snRNA Sequencing

Single-nucleus RNA sequencing (snRNA-seq) was conducted on tumor samples from nine glioma patients to investigate the cellular diversity within these specimens. After quality control and filtering, we analyzed 22,392 cells through dimensionality reduction clustering, identifying 22 clusters corresponding to six different cell types: oligodendrocytes (771), glial neuronal cells (13,760), astrocytes (4,980), smooth muscle cells (106), vascular cells (519), and myeloid cells (2,256) (Fig. 1A, upper). Among the 22,392 cells analyzed, 14,611 exhibited the IDH1 mutation, while 20,931 were IDH1 wild-type (Fig. 1A). UMAP plots illustrated the distribution of these six cell types across various cell cycle phases: 5,051 cells in the S-phase, 13,728 in the G1 phase, and 3,613 in the G2M phase (Fig. 1A, bottom).

A bar chart was employed to depict the distribution of the six cell types within a cohort consisting of eight patients with IDH1 wild-type lesions and one patient with IDH1 mutant lesions (Fig. 1B). This chart highlights the heterogeneity of cell types among the glioma patients. Box plots further emphasized the distinct variations among these six cellular phenotypes across the different cohorts (Fig. 1C). Marker gene expression varied among the six cell types identified in the tumor samples, with the top 10 markers for each cell type displayed in bubble plots (Fig. 1D).

RNA characteristics, including nCount_RNA, nFeature_RNA, G2M score, and S score, were visualized through UMAP plots (Fig. 1E). Word clouds displayed Gene Ontology Biological Process (GO-BP) enriched pathway terms across the cell types while volcano plots highlighted distinct genes among the six cell types (Fig. 1F). Heatmaps presented GO-BP enrichment analysis results for uniquely expressed genes within the six cellular phenotypes (Fig. 1G).

3.2 Intracellular Heterogeneity in Glia-Neuronal Cells

To analyze glia-neuronal cells for malignancy, we applied an inferred CNV algorithm to identify malignant cells at the single-cell level. Based on inferred copy number variations (CNV), cells with elevated CNV levels were classified as glioma cells (Supplementary Fig. 1) revealing three subpopulations: C0 MALAT1 + glioma cells (2,508), C1 AKAP9 + glioma cells (1,788), and C2 NUSAP1 + glioma cells (1,566) (Fig. 2A). We examined the distribution of cell cycle stages, subsets, and lineages across these three subgroups.

UMAP representations were used to visualize the CNV score, nCount_RNA, S score, G2M score, and other relevant attributes of the three cellular subgroups (Fig. 2B). Additionally, the distribution of these three subtypes was visually examined across eight cases with IDH1 wild-type lesions and one case with an IDH1 mutant lesion (Fig. 2C, left). The C2 NUSAP + glioma subpopulation was predominantly found in patients with SF12264. Box plots showed varying proportions of the three subpopulations across the groups (Fig. 2C, right). Volcano plots highlighted differentially expressed genes in each of the three subpopulations (Fig. 2D). Word clouds displayed enriched Gene Ontology Biological Process (GO-BP) pathway terms within these subgroups (Fig. 2E). Marker genes with differential expression in the three subpopulations were visualized using bubble plots and heatmaps (Fig. 2F), A heatmap was generated to illustrate the results of GO-BP enrichment analysis for the distinct gene sets in the three subgroups (Fig. 2G).

3.3 Pseudotemporal Analysis of Glial and Neuronal Cell Subpopulations Using CytoTRACE and Monocle

To explore the differentiation and developmental trajectories of glioma cell subpopulations, we conducted CytoTRACE analysis (Fig. 3A), revealing a differentiation pattern from C1 to C0 and then to C2 (Fig. 3B). A bar graph illustrates the distribution of the three cell subpopulations across IDH1 mutant and IDH1 wildtype cohorts (Fig. 3C, left). Notably, the C2 NUSAP1 + glioma cell subpopulation was exclusive to the IDH1 wildtype group. This finding provides a starting point to explore the relationship between the IDH1 mutant and wildtype groups. Additional bar graphs illustrated the distribution of the three cell subpopulations across different stages of the cell cycle (Fig. 3C, right). nd to show the changes in cell percentages during different stages of trajectory differentiation (Fig. 3D). In state 1, the C0 MALAT1-expressing glioma cells were the most prevalent. In state 2, C1 AKAP9-expressing glioma cells dominated, while in state 3, C2 NUSAP1-expressing glioma cells were the most abundant..

To gain further insights into glioma development, we conducted additional trajectory analysis of the three cellular subtypes using monoclonal techniques (Fig. 3E). The three cellular subtypes displayed a continuous distribution along the pseudotemporal trajectory, eventually diverging at a branching point. The trajectory commenced in the upper right quadrant and advanced to the differentiation point designated as state 1. At this point, the lineage split: one branch continued caudally, corresponding with state 3, while the other moved leftward, aligning with state 2. Chronologically, the C1 AKAP9 + glioma cell subpopulation appears to represent the early stage of tumorigenesis. As tumorigenesis progresses, this subpopulation differentiates into other subtypes, eventually becoming C0 DOCK5 + or C2 NUSAP + glioma cells. We identified the genes associated with the three cellular subpopulations and mapped their dynamic trajectories using pseudo-time series scatter plots, UMAP plots, and pseudo-scatter plots. The analysis revealed that the C1 cell subset, marked by the AKAP9 gene, was predominantly present early in the temporal sequence, while the C2 subset, characterized by the NUSAP1 gene, was more prevalent towards the end of the pseudotemporal series (Fig. 3F).

3.4 Slingshot Analysis of Pseudotemporal Trajectories of Glioma Cell Subpopulations

We further analyzed the differentiation trajectories of the three glioma cell subpopulations (C0 to C2) using the slingshot method. These subpopulations were consistently distributed along the temporal axis, forming a lineage (Fig. 4A). Spectrum 1 illustrates the progression from C1 to C0, followed by a shift to C2 (Fig. 4B). To understand the biological mechanisms underlying these pseudotemporal trajectories, We conducted enrichment analysis for GO-BP. The C1 subpopulation was associated with processes like polymerization and microtubule dynamics. C2 was linked to signaling mediation and immune responses, while C3 was connected to pyrimidine metabolism, and C4 to mitotic processes (Fig. 4C). Scatter plots illustrated the distribution of different cell subpopulations along Spectrum 1, highlighting their unique differentiation patterns over pseudotime (Fig. 4D).

3.5 CellChat Analysis of Cell-Cell Interactions and PTN Signaling Pathway Visualization

To better understand complex cellular responses, we aimed to explore intercellular dynamics and ligand-receptor communication networks. Using CellChat analysis, we constructed intercellular communication networks involving various cell types, such as glioma subpopulations, oligodendrocytes, myeloid cells, astrocytes, smooth muscle cells, and vascular cells. We quantified interaction frequencies by measuring connection thickness and assessed interaction intensities based on line weights (Fig. 5A). To investigate the coordination of various cell populations and signaling pathways, we used CellChat's non-negative matrix decomposition technique for pattern recognition. Our analysis revealed communication patterns connecting cell populations as either signal transmitters (outbound) or recipients (inbound). Using CellChat's gene expression tool, we further explored these cellular interactions and signaling pathways. We first linked hypothesized communication patterns to secretory cell cohorts to clarify efferent signaling modalities. Then, we correlated these patterns with secretory cell populations. We identified three communication modes, each associated with a specific cell type: mode 1 (glioma subpopulations), mode 2 (vascular endothelial cells), and mode 3 (oligodendrocytes and myeloid cells) (Fig. 5B). For glioma efferent signaling, mode 1 was characterized by pathways such as PTN and NCAM. In contrast, glioma afferent signaling mainly involved mode 1 communication patterns, including pathways like VEGF, PTN, and JAM (Fig. 5D). Employing CellChat's pattern recognition methodology, we quantitatively analyzed the ligand-receptor interactions within gliomas to pinpoint key signaling pathways associated with the three cell subtypes. In gliomas, each cell variant functions as both a sender and receiver in the communication network, releasing various cytokines or ligands and responding to signals from other cells (Fig. 5C).

We created a heatmap to visualize the intensity of afferent and efferent signals across all cell interactions, specifically focusing on participation in the PTN signaling pathway (Fig. 5E). Utilizing a centrality measurement algorithm, we classified cell types according to their roles as mediators and influencers in intercellular communication. The Glioma C2 subset, characterized by NUSAP expression, was identified as a pivotal player in the PTN signaling cascade. Additionally, we found that the NCAM signaling pathway, involved in cell adhesion, and the VEGF pathway, associated with angiogenesis, were particularly prominent in the C1 AKAP9 + Glioma subpopulation (Fig. 5F). Violin plots highlight cellular interactions, showing the NUSAP + Glioma subpopulation in the C2 group with elevated activity in the PTN signaling cascade. In contrast, the AKAP9 + Glioma subset in the C1 group was notably involved in the NCAM and VEGF signaling pathways (Fig. 5G). We identified all eight cell types within glioma tissues as origins of the PTN signaling cascade. The three glioma subtypes, along with other cell types, were considered potential targets, highlighting their correlations within the PTN signaling pathway in a hierarchical plot. The findings suggest that, except for myeloid cells, oligodendrocytes, and vascular endothelial cells (VECs), various cell types act as signaling mediators within the PTN cascade. A heatmap displaying the complex network of cell-cell interactions is shown in Fig. 5I.

3.6 Development and Validation of a Prognostic Model

To assess the clinical significance of the identified cell types, we performed a univariate Cox analysis on the top 100 marker genes within the C2 NUSAP + Glioma subgroup. This analysis revealed three genes—RPA3, TUBA1C, and NUDT1—that are associated with patient outcomes (Fig. 6A). To address multicollinearity within the gene pool, we used lasso regression to refine the selection, identifying three key genes crucial to the NUSAP + Glioma scoring system. Lambda plots validated the robustness of this gene subset (Fig. 6B). Patients were categorized into two groups based on their expression levels of the selected genes: high and low NUSAP + Glioma score groups. Survival analysis indicated that individuals in the low NUSAP + Glioma score group experienced superior survival outcomes compared to those in the high NUSAP + Glioma score group (Fig. 6C). Survival analysis was conducted for the three genes that make up the NUSAP + Glioma score model (Fig. 6D). The results consistently demonstrated that elevated expression levels correlated with poorer survival, whereas reduced expression levels were associated with improved prognostic outcomes, thereby affirming their role as risk factors. The NUSAP + Glioma score for each patient in the TCGA-GBM cohort was determined by integrating gene expression levels with their corresponding regression coefficients. Subsequently, patients were categorized into high and low score groups based on the median value. A higher NUSAP + Glioma score was associated with reduced survival times. Expression levels of the three genes in the model are depicted in Fig. 6E. Correlation analysis revealed an inverse relationship between overall survival (OS) and both the NUSAP + Glioma score and the three genes. These relationships are visually represented in scatter plots (Fig. 6F). ROC curves were utilized to evaluate the predictive accuracy of the NUSAP + Glioma score for 1-year, 3-year, and 5-year survival, yielding AUC values of 0.579, 0.792, and 0.625, respectively (Fig. 6G). Scatter plots displayed the genetic factors correlated with NUSAP + Glioma scores (Fig. 6H), and Fig. 6I highlighted differences in gene expression levels between high and low NUSAP + Glioma score groups. Multifactorial Cox regression analysis was performed to determine if the NUSAP + Glioma score serves as an independent prognostic factor. This analysis incorporated variables such as age, race, T stage, N stage, M stage, and the NUSAP + Glioma score, revealing the latter as a significant independent predictor of prognosis in glioma patients (p < 0.05) (Fig. 6K). Additionally, a column chart was generated to integrate clinical and pathological risk factors with cell type characteristics, utilizing age, race, and T stage. This chart provides an effective prediction of patient survival probabilities at 1, 3, and 5 years (Fig. 6J).

3.7 Immune Infiltration Differences between High and Low NUSAP + Glioma Score Groups

To investigate immune infiltration in gliomas and assess differences in immune cell populations between high and low NUSAP + Glioma score groups, we utilized heatmaps to visualize the variations in immune cell infiltration within each group (Fig. 7A). We next evaluated immune cell infiltration in glioma patients by analyzing data from the TCGA repository using the CIBERSORT computational tool. Heatmaps were employed to display the distribution of 22 distinct immune cell types identified in the samples (Fig. 7B). Bar graphs were used to depict the relationships between immune cell types and glioma subpopulation markers. The NUSAP + Glioma scores showed a positive correlation with M0 macrophages, resting dendritic cells, and other immune cells, whereas they were negatively correlated with activated NK cells, monocytes, and additional cell types (Fig. 7C). Various methods of immune cell content assessment were used to compare and summarize the associations between the three genes under study and immune cells. These relationships were illustrated using heatmaps, where positive correlations were represented in red and negative correlations were depicted in blue (Fig. 7D). Violin plots demonstrated immune dysfunction and tumor rejection across both high and low scoring groups, with a significantly elevated TIDE score observed in the low scoring group compared to the high scoring group (Fig. 7E). The expression of ADORA2A was present in tumors from both groups, but it was notably lower in the high NUSAP + Glioma score group relative to the low scoring group.

3.8 Differential and Enrichment Analysis

To compare the high and low NUSAP + Glioma score groups, we employed volcano plots and heat maps to illustrate the expression patterns of distinct genes in each group (Fig. 8A, B). To investigate the role of the C2 NUSAP + Glioma subgroup in glioma pathogenesis, we conducted functional enrichment analysis on genes that distinguish these groups. Bubble plots presented the results of Gene Ontology (GO) enrichment analysis, highlighting that these genes are primarily involved in oligosaccharide binding, peptidoglycan binding, and pathways associated with auditory receptor cell development (Fig. 8D). KEGG enrichment analysis, depicted in bar graphs, revealed significant associations with pathways such as neuroactive ligand-receptor interaction and the cAMP signaling pathway (Fig. 8C). Additionally, Gene Set Enrichment Analysis (GSEA) scores indicated gene enrichment across various pathways, as shown in GO-BP-enriched entries for differentially expressed genes (Fig. 8E).

3.9 Mutation analysis

We analyzed and visualized gene mutations within the tumor microenvironment (TME) to determine their correlation with immune components across two cell cohorts. This analysis revealed differences in the top 30 genes with the highest mutation frequencies within these mesenchymal cell cohorts. The vertical bar shows the mutation burden per sample, while the horizontal bar indicates the overall mutation prevalence of each gene across the samples (Fig. 8F). No significant chromosomal copy number variations (CNVs) were detected in the genes analyzed, as shown by the lack of significant gains or losses in the CNV profile (Fig. 8G). Violin plots were employed to examine mutation burden variations between high and low NUSAP + Glioma score groups, revealing no statistically significant differences (Fig. 8H). However, scatter plots showed a statistically significant correlation (P < 0.05) between mutation load and NUSAP + Glioma scores (Fig. 8I). Tumors were classified into four categories based on mutation burden: high NUSAP + Glioma score with high TMB, high NUSAP + Glioma score with low TMB, low NUSAP + Glioma score with high TMB, and low NUSAP + Glioma score with low TMB. Survival analysis revealed that the group with a low NUSAP + Glioma score and high TMB exhibited the most favorable survival outcomes, whereas the group with a high NUSAP + Glioma score and low TMB showed the poorest prognosis (Fig. 8J).

3.10 Drug sensitivity analysis

Violin plots were utilized to depict the differential responses to various medications between high and low NUSAP + Glioma score groups, emphasizing variations in drug sensitivity (Fig. 8K). The IC50 value for Axitinib was elevated in the high NUSAP + Glioma score group relative to the low score group, indicating a decreased drug responsiveness in the former.

3.11 Knockdown of NUSAP1 Suppresses Proliferation, Migration, and Metastasis in Glioma Cells

To examine the role of NUSAP1 in glioma, we employed gene knockdown via transfection and validated its efficiency with RT-qPCR (Supplementary Fig. 2). Colony formation assays were performed on U251 and LN229 glioma cell lines, comparing a negative control (NC) group with an si-NUSAP1 experimental group (Fig. 9B). T Results demonstrated that NUSAP1 inhibition led to reduced colony size in both U251 and LN229 cells, indicating that downregulation of NUSAP1 impedes glioma cell proliferation (Fig. 9A). To evaluate NUSAP1's impact on glioma cell migration, we conducted scratch and transwell assays (Fig. 9C). These assays revealed a marked decrease in the migratory capacity of U251 and LN229 cells following NUSAP1 knockdown (Fig. 9A). Further, CCK-8 assays confirmed that NUSAP1 suppression impeded both proliferation and migration of glioma cells (Fig. 9D). To further validate NUSAP1’s regulatory role, we generated stable NUSAP1 knockdown cell lines and performed apoptosis flow cytometry assays. The data indicated a significant increase in apoptosis in glioma cells after NUSAP1 knockdown (Fig. 9E, F). Additionally, we assessed the impact of NUSAP1 knockdown on lung metastasis using tail vein injection in animal models. Results showed a notable reduction in the metastatic potential of glioma cells with NUSAP1 knockdown (Fig. 9G, H).

Glioma continues to be one of the most difficult cancers to manage, as current treatment approaches have limited efficacy due to its aggressive behavior and intricate biological characteristics. Despite advancements in oncology, the progress of personalized medicine in glioma has been relatively slow, largely due to the tumor's heterogeneity and the intricate interplay of genetic mutations. However, the advent of single-cell technologies has the potential to revolutionize personalized treatment for glioma. By enabling detailed analysis of individual tumor cells, these technologies can uncover unique cellular subpopulations and molecular pathways driving tumor growth and resistance to therapy. This detailed analysis not only deepens our comprehension of glioma biology but also aids in the creation of more targeted and effective personalized therapies, potentially enhancing patient outcomes.

In our research, single-nucleus RNA sequencing (snRNA-seq) was conducted from nine glioma patients, uncovering the primary cell types and their intricate heterogeneity throughout tumor progression. This investigation provided an in-depth understanding of distinct cellular subpopulations and highlighted the impact of IDH1 mutation status on the distribution of these subpopulations and disease progression. Notably, we identified three major glioma cell subpopulations (C0 MALAT1+, C1 AKAP9+, and C2 NUSAP1+), which were highly correlated with tumor malignancy. Through trajectory analysis and interaction network studies, we uncovered the pivotal roles of these subpopulations in glioma progression and therapeutic response.

These findings offer critical molecular insights for the development of personalized treatment strategies in glioma. For example, integrating specific cellular subpopulations identified in patient tumors with clinical prognosis could enhance the accuracy of predicting therapeutic responses, thereby supporting the formulation of individualized treatment plans. Moreover, the identification of intercellular signaling pathways, such as PTN and NCAM, and the specific gene expression patterns of NUSAP1 and AKAP9, accelerates the discovery of therapeutic targets and the development of novel drugs. Consequently, our study not only charts new directions for personalized glioma treatment but also presents an innovative approach to managing this highly heterogeneous tumor..

In parallel, our research has shown significant potential in advancing personalized treatment strategies for melanoma, particularly in the context of genetic interventions aimed at modulating tumor progression. While NUSAP1 is recognized as an oncogene in several cancers, its role in glioma remains inadequately understood. Through NUSAP1 knockdown experiments, we elucidated its pivotal role in regulating glioma cell proliferation, migration, and distant metastasis. Both in vitro and in vivo assays—such as colony formation, scratch wound healing, transwell migration, and flow cytometry—demonstrated that NUSAP1 silencing significantly suppressed malignant behaviors of glioma cells and induced a marked increase in apoptosis. These results further underscore NUSAP1's potential as a therapeutic target and lay the foundation for future precision medicine approaches.

Prior research has established a correlation between elevated NUSAP1 expression and adverse outcomes in various cancers, including breast, lung, and prostate. Our findings extend this knowledge to glioma, revealing similar mechanisms by which NUSAP1 contributes to tumorigenesis. Given the genetic and microenvironmental similarities between melanoma and glioma, these insights support the potential of targeting NUSAP1 in melanoma treatment. By suppressing NUSAP1 expression, more precise tumor control may be achieved in the context of personalized therapy, offering better outcomes for patients.

Furthermore, with the rapid development of single-cell technologies, researchers can now capture tumor heterogeneity more accurately and identify key genes associated with tumor progression. NUSAP1, as one of these genes, has demonstrated its broad role in malignancies. This single-cell-based approach not only improves our understanding of tumor evolution but also guides the design of personalized treatment regimens, providing more precise and effective therapeutic options for patients across various cancer types.

In conclusion, our findings provide new insights into the potential application of NUSAP1 in personalized melanoma treatment. Future research exploring the interactions between NUSAP1 and other tumor-related genes may help develop more effective therapeutic strategies, advancing the progress of personalized medicine.

Our study highlights the critical role of NUSAP1 in glioma progression, demonstrating that its silencing significantly impairs glioma cell proliferation, migration, and metastasis while promoting apoptosis. These findings underscore NUSAP1 as a potential therapeutic target in glioma, with implications for advancing personalized treatment strategies in melanoma and other cancers. Further exploration of NUSAP1 in the context of tumor heterogeneity and single-cell analysis could pave the way for more precise and effective therapeutic interventions.

7.1 Funding

This work was supported by the Postgraduate Innovation Research and Practice Program of Anhui Medical University [Nos.YJS20230074, Nos.YJS20230184] , Youth Foundation of National Natural Science Foundation of China [Nos. 82003795], Youth Project of Anhui Natural Science Foundation (Nos. 2108085QH330), University Natural Science Research Project of Anhui Province (2023AH053318), Scientific Research Fund, Scientific Research and Cultivation Project of Department of Nursing, and Doctor Research Project of the First Affiliated Hospital of Anhui Medical University [Nos. 2021xkj020, hlpy20210014], Quality Engineering Teaching Research Project of Anhui Medical University[No. 2021xjjyxm12 and 2022zyxwjxalk057], Early contact scientific research training plan for clinical medicine [No. 2021-ZQKY-123], Open Project of Key Laboratory of Anti-Inflammatory and Immune Medicine, Ministry of Education [No. KFJJ-2020-01], Clinical Science Foundation project [No. 2023xkj193] and Open Project of Key lab of Dermatology, Ministry of Education, Anhui Medical University[No. AYPYS2022-5 and KFJJ-2020-01]..

7.2 Conflict of Interest

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

7.3 Author Contributions

Fang Zhang, Wen-Ming Hong and Dong-Hui Chen designed and supervised the study. Meng-Yu Zhao, Tian-Hang Yu and Jia-Feng Xu carried out the bioinformatics analysis, Meng-Yu Zhao and Wan-Yan Xu performed the experiments. Wan-Yan Xu, Yu Gu, Tian-Hang Yu, Cun-Zhi Wang, Jie-Hui Zhao and Li-Na Wang analyzed the data. Meng-Yu Zhao and Zhao-Lei Shen wrote the paper. Wen-Ming Hong, Dong-Hui Chen and Fang Zhang edited the manuscript and provided critical comments. All authors read and approved the final manuscript. Meng-Yu Zhao, Zhao-Lei Shen and Wan-Yan Xu contributed equally to this work.

8 Data Availability

The datasets analyzed in the current study are available in the TCGA repository (http://cancergenome.nih.gov/), and GEO (https://www. ncbi.nlm.nih.gov/geo/).

9 Code availability

Analyses were conducted using R (version 4.2.1) and GraphPad (GraphPad Prism 8.0). The codes used to support the findings of this study is available from the corresponding author on reasonable request.

van den Bent MJ, Geurts M, French PJ, Smits M, Capper D, Bromberg JEC, Chang SM: Primary brain tumours in adults. Lancet 2023, 402:1564-1579.
Lapointe S, Perry A, Butowski NA: Primary brain tumours in adults. Lancet 2018, 392:432-446.
Miller AM, Shah RH, Pentsova EI, Pourmaleki M, Briggs S, Distefano N, Zheng Y, Skakodub A, Mehta SA, Campos C, et al: Tracking tumour evolution in glioma through liquid biopsies of cerebrospinal fluid. Nature 2019, 565:654-658.
Li G, Li L, Li Y, Qian Z, Wu F, He Y, Jiang H, Li R, Wang D, Zhai Y, et al: An MRI radiomics approach to predict survival and tumour-infiltrating macrophages in gliomas. Brain 2022, 145:1151-1161.
Sturm D, Capper D, Andreiuolo F, Gessi M, Kolsche C, Reinhardt A, Sievers P, Wefers AK, Ebrahimi A, Suwala AK, et al: Multiomic neuropathology improves diagnostic accuracy in pediatric neuro-oncology. Nat Med 2023, 29:917-926.
Haider AS, van den Bent M, Wen PY, Vogelbaum MA, Chang S, Canoll PD, Horbinski CM, Huse JT: Toward a standard pathological and molecular characterization of recurrent glioma in adults: a Response Assessment in Neuro-Oncology effort. Neuro Oncol 2020, 22:450-456.
Weller M, Wick W, Aldape K, Brada M, Berger M, Pfister SM, Nishikawa R, Rosenthal M, Wen PY, Stupp R, Reifenberger G: Glioma. Nat Rev Dis Primers 2015, 1:15017.
Lu C, Wei Y, Wang X, Zhang Z, Yin J, Li W, Chen L, Lyu X, Shi Z, Yan W, You Y: DNA-methylation-mediated activating of lncRNA SNHG12 promotes temozolomide resistance in glioblastoma. Mol Cancer 2020, 19:28.
Chen M, Cen K, Song Y, Zhang X, Liou YC, Liu P, Huang J, Ruan J, He J, Ye W, et al: NUSAP1-LDHA-Glycolysis-Lactate feedforward loop promotes Warburg effect and metastasis in pancreatic ductal adenocarcinoma. Cancer Lett 2023, 567:216285.
Zheng H, Wang M, Zhang S, Hu D, Yang Q, Chen M, Zhang X, Zhang Y, Dai J, Liou YC: Comprehensive pan-cancer analysis reveals NUSAP1 is a novel predictive biomarker for prognosis and immunotherapy response. Int J Biol Sci 2023, 19:4689-4708.
Chiu CL, Li CG, Verschueren E, Wen RM, Zhang D, Gordon CA, Zhao H, Giaccia AJ, Brooks JD: NUSAP1 Binds ILF2 to Modulate R-Loop Accumulation and DNA Damage in Prostate Cancer. Int J Mol Sci 2023, 24.
Xing J, Cai H, Lin Z, Zhao L, Xu H, Song Y, Wang Z, Liu C, Hu G, Zheng J, et al: Examining the function of macrophage oxidative stress response and immune system in glioblastoma multiforme through analysis of single-cell transcriptomics. Front Immunol 2023, 14:1288137.
Lin Z, Li X, Shi H, Cao R, Zhu L, Dang C, Sheng Y, Fan W, Yang Z, Wu S: Decoding the tumor microenvironment and molecular mechanism: unraveling cervical cancer subpopulations and prognostic signatures through scRNA-Seq and bulk RNA-seq analyses. Front Immunol 2024, 15:1351287.
Ding Y, Zhao Z, Cai H, Zhou Y, Chen H, Bai Y, Liu Z, Liu S, Zhou W: Single-cell sequencing analysis related to sphingolipid metabolism guides immunotherapy and prognosis of skin cutaneous melanoma. Front Immunol 2023, 14:1304466.
Ge Q, Zhao Z, Li X, Yang F, Zhang M, Hao Z, Liang C, Meng J: Deciphering the suppressive immune microenvironment of prostate cancer based on CD4+ regulatory T cells: Implications for prognosis and therapy prediction. Clin Transl Med 2024, 14:e1552.
Huang W, Kim BS, Zhang Y, Lin L, Chai G, Zhao Z: Regulatory T cells subgroups in the tumor microenvironment cannot be overlooked: Their involvement in prognosis and treatment strategy in melanoma. Environ Toxicol 2024.
Zhao Z, Ding Y, Tran LJ, Chai G, Lin L: Innovative breakthroughs facilitated by single-cell multi-omics: manipulating natural killer cell functionality correlates with a novel subcategory of melanoma cells. Front Immunol 2023, 14:1196892.
Korsunsky I, Millard N, Fan J, Slowikowski K, Zhang F, Wei K, Baglaenko Y, Brenner M, Loh PR, Raychaudhuri S: Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods 2019, 16:1289-1296.
Zhou Y, Yang D, Yang Q, Lv X, Huang W, Zhou Z, Wang Y, Zhang Z, Yuan T, Ding X, et al: Single-cell RNA landscape of intratumoral heterogeneity and immunosuppressive microenvironment in advanced osteosarcoma. Nat Commun 2020, 11:6322.
Lin Z, Sui X, Jiao W, Chen C, Zhang X, Zhao J: Mechanism investigation and experiment validation of capsaicin on uterine corpus endometrial carcinoma. Front Pharmacol 2022, 13:953874.
Liu P, Xing N, Xiahou Z, Yan J, Lin Z, Zhang J: Unraveling the intricacies of glioblastoma progression and recurrence: insights into the role of NFYB and oxidative phosphorylation at the single-cell level. Front Immunol 2024, 15:1368685.
Qiu X, Mao Q, Tang Y, Wang L, Chawla R, Pliner HA, Trapnell C: Reversed graph embedding resolves complex single-cell trajectories. Nat Methods 2017, 14:979-982.
Street K, Risso D, Fletcher RB, Das D, Ngai J, Yosef N, Purdom E, Dudoit S: Slingshot: cell lineage and pseudotime inference for single-cell transcriptomics. BMC Genomics 2018, 19:477.
Jin S, Guerrero-Juarez CF, Zhang L, Chang I, Ramos R, Kuan CH, Myung P, Plikus MV, Nie Q: Inference and analysis of cell-cell communication using CellChat. Nat Commun 2021, 12:1088.
Lin Z, Zou J, Sui X, Yao S, Lin L, Wang J, Zhao J: Necroptosis-related lncRNA signature predicts prognosis and immune response for cervical squamous cell carcinoma and endocervical adenocarcinomas. Sci Rep 2022, 12:16285.
Zou J, Lin Z, Jiao W, Chen J, Lin L, Zhang F, Zhang X, Zhao J: A multi-omics-based investigation of the prognostic and immunological impact of necroptosis-related mRNA in patients with cervical squamous carcinoma and adenocarcinoma. Sci Rep 2022, 12:16773.
Fu J, Li K, Zhang W, Wan C, Zhang J, Jiang P, Liu XS: Large-scale public data reuse to model immunotherapy response and resistance. Genome Med 2020, 12:21.
Zhao J, Zou J, Jiao W, Lin L, Wang J, Lin Z: Construction of N-7 methylguanine-related mRNA prognostic model in uterine corpus endometrial carcinoma based on multi-omics data and immune-related analysis. Sci Rep 2022, 12:18813.

No competing interests reported.

supplementarymaterials.zip
Supplementary Figure 1 NUSAP1 gene transfection knock-down low efficiency verification. Compared with untransfected cells, the mRNA level of NUSAP1 gene was significantly decreased in the transfected knockdown group. Supplementary Figure 2 The classification of GBM cells. According to the infer CNV results, we defined cells with high CNV levels as GBM cells. Note： C2 NUSAP+ gliomas show variations in drug sensitivities between high and low groups, as illustrated by violin plots. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001 indicate a significant difference, while "ns" indicates a non-significant difference. N：Differences in drug sensitivities between high and low C2 NUSAP+glioma score groups are illustrated through violin plots. *, p ≤ 0.05; **p ≤ 0.01; ***p≤ 0.001 indicate a significant difference, and "ns" indicates a non-significant difference.

Download PDF

Editor assigned by journal
27 Aug, 2024
Submission checks completed at journal
21 Aug, 2024
First submitted to journal
20 Aug, 2024

You are reading this latest preprint version

Single-Cell Sequencing Reveals the Role of NUSAP1 in Glioma and Its Potential for Precision Diagnostic and Prognostic Applications

Status:

Version 1

Abstract

Figures

1 Introduction

2 Methods

2.1 Data Acquisition

2.2 Data Filtering and Processing

2.3 Enrichment Analysis of Differentially Expressed Genes

2.4 Subpopulation Analysis of Glia/Neuronal Cells

2.5 Malignant Cell Identification via InferCNV

2.6 Differential and Enrichment Analyses of Cell Subpopulations

2.7 Trajectory analysis

2.8 Cell Communication Analysis

2.9 Prognostic Modeling of Glioma-Associated Cells

2.10 Analysis of Tumor-Infiltrating Immune Cells

2.11 Differential and Enrichment Analyses of Bulk Genomic Data

2.12 Somatic Mutation Analysis

2.13 Drug Sensitivity Analysis

2.14 Cell culture

2.15 Cell Transfection, Lentivirus Vector Construction, and Cell Infection

2.16 RT-qPCR Analysis

2.17 Cell Counting Kit-8 (CCK-8) Assay

2.18 Colony Formation Assay

2.19 Wound Healing Assay

2.20 Transwell Migration Assay

2.21 Flow Cytometric Analysis of Apoptosis

2.22 Animal Experiment

2.22 Statistical analysis

3 Results

3.1 Key Cell Types Involved in Glioma Progression Identified via snRNA Sequencing

3.2 Intracellular Heterogeneity in Glia-Neuronal Cells

3.3 Pseudotemporal Analysis of Glial and Neuronal Cell Subpopulations Using CytoTRACE and Monocle

3.4 Slingshot Analysis of Pseudotemporal Trajectories of Glioma Cell Subpopulations

3.5 CellChat Analysis of Cell-Cell Interactions and PTN Signaling Pathway Visualization

3.6 Development and Validation of a Prognostic Model

3.7 Immune Infiltration Differences between High and Low NUSAP + Glioma Score Groups

3.8 Differential and Enrichment Analysis

3.9 Mutation analysis

3.10 Drug sensitivity analysis

3.11 Knockdown of NUSAP1 Suppresses Proliferation, Migration, and Metastasis in Glioma Cells

4 Discussion

5 Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1