Twenty adult-type diffuse glioma specimens were collected through surgical resection at the 3201 Hospital, Affiliated with Xi’an Jiaotong University, from January 2021 to July 2022. The tissue specimens were routinely fixed in 10% neutral formalin and subsequently embedded in paraffin. Screening criteria: ① Patients with an initial diagnosis of glioma; ② Patients with no prior history of other associated malignancies or surgeries; ③ Patients with relatively comprehensive pathology-related data.

The study received approval from the Ethics Committee of the 3201 Hospital (approval number 2022082). All participants provided written informed consent. The clinical characteristics of the patients are presented in Supplementary Table 1.

Four fresh adult-type diffuse glioma tissues and four non-tumor brain tissue specimens obtained through internal decompression of brain injury were collected as controls. All samples were stored according to the standard in a liquid nitrogen tank for 40 minutes in vitro. Protein extraction was conducted, followed by the concentration determination utilizing the bicinchoninic acid assay (BCA) method. The denatured protein was placed on ice and subjected to electrophoresis. Then, the membrane was probed with the NOP14 primary antibody (1:2000, Proteintech Group, Rosemont, IL, USA) and $\beta$-actin primary antibody (1:2000, Proteintech Group) overnight. After Smembrane washing, the corresponding rabbit and rat secondary antibody (1:2000, Proteintech Group) was added for further incubation. Following a 2-hour incubation at room temperature, the Enhanced ChemiLuminescence (ECL) chemiluminescent solution was added, and the exposure was performed using a Bio-Rad chemiluminometer (thermofisher.cn).

All adult-type diffuse glioma samples were prepared following standard procedures, completely fixed with 10% neutral formalin, dehydrated, embedded, and then continuously sectioned at a thickness of 3 µm. The sections were dried and subjected to pretreatment using a DAKO PT Link automated immunohistochemical preconditioning system. Then, the sections were treated utilizing an AutostainerLink 48 (agilent.com.cn) fully automated immunohistochemical staining instrument. The rabbit monoclonal antibody against NOP14 (1:2000, Proteintech Group) was acquired from Proteintech Group Company. For external control, HepG2 cells were utilized as the positive control, phosphate-buffered saline (PBS) was substituted for the primary antibody as the negative control (NC), and red blood cells within the blood vessels served as the internal control.

The mRNA expression data of NOP14 and clinical data of glioblastoma (GBM) patients were obtained from the TCGA database (https://cancergenome.nih.gov/), and the data was processed utilizing R software (v3.6.3) and Perl (R Foundation for Statistical Computing, Vienna, Austria). The dataset included 701 patients diagnosed with GBM and 5 normal brain tissue samples. The glioma patients were categorized into high-expression and low-expression groups based on the median NOP14 mRNA expression.

The Strawberry Perl software integrated the NOP14 expression data from the TCGA-GBM dataset with tumor stage, histological grade, tumor size, and survival time of patients with GBM. The relationship between NOP14 and tumor stage was tested utilizing the Kruskal (KS) test. The difference in survival between high and low NOP14 expression groups was evaluated utilizing the Kaplan-Meier method. The predictive value of NOP14 was assessed utilizing multivariate Cox regression.

The gene expression data was classified into high and low groups using the median value of NOP14 expression in the TCGA-GBM dataset. Gene Set Enrichment Analysis (GSEA) software (gsea-msigdb.org) was utilized for c2.cp. Kegg.v7 in Molecular Tabel database (Molecular Signature Database, Msig-DB). The 2.symbols.gmt data set was utilized as a set of functional genes for conducting a signaling pathway enrichment analysis of NOP14. The single gene enrichment results of NOP14 were filtered based on the absolute normalized enrichment score (NES) values of 1. O, NOM p-val 0.05 and False Discovery Rate (FDR) q-val 0.25.

The relationship between NOP14 expression and the extent of immune cell infiltration, including neutrophils, macrophages, B cells, CD4${}^{+}$T cells, CD8${}^{+}$T cells, and dendritic cells, was explored with the help of the gene modules available in the TIMER database. Furthermore, the correlation module was adopted to analyze the association between NOP14 expression and the expression of cell markers related to immune cell infiltration. At 0.1 or 1.0, p 0.05 indicates statistical significance.

The Tumor Immune Single-cell Hub (TISCH) database (http://tisch.comp-genomics.org/) integrated single-cell sequencing data from 27 cancers [11]. In this study, the Glioma_GSE148842 dataset was selected to test the NOP14 expression in individual cells within the GBM microenvironment, with the results visualized using a heatmap.

Human normal astroglial cells (HA1800) were grown in an RPMI 1640 medium containing 10% fetal calf serum (FCS). Human glioma cell lines U87, U251, and LN229 were cultivated in high glucose Dulbecco’s Modified Eagle Medium (DMEM) with 10% FCS. All cell lines were validated by short tandem repeat (STR) profiling and tested negative for mycoplasma. All cells were kept at 37 °C with 5% CO${}_2$ and 95% humidity. Human glioma cell lines (LN229) and (U251) were purchased from Catalpa chinensis (Wuhan, China), and no mycoplasma (Myco-Zero ${}^{\text{TM}}$ Mycoplasma removal reagent (Beyotime, Shanghai, China)). RNA extraction was implemented upon 80% confluency. RNA extraction was carried out using the Mabio RNT412-02 RNA extraction kit (vazyme.com), followed by reverse transcription using the Vazyme R222-01 kit (vazyme.com). ChamQ Universal SYBR qPCR Master Mix (Vazyme Q711-03) was chosen for qPCR analysis. The primers (synthesized by Shanghai Qingke) and cells were provided by the Biochemistry Laboratory of Chongqing University. Primer sequence:

NOP14 F: 5${}^{\prime }$-GAAGGCCAACTCCAATCCGTT-3${}^{\prime }$, R: 5${}^{\prime }$-AAGTCTGTGTACGCTTCCTGA-3${}^{\prime }$; GAPDHF: 5${}^{\prime }$-TCACCAGGGCTGCTTTTA-3${}^{\prime }$ and R: 5${}^{\prime }$-TTCACACCCATGACGAACA-3${}^{\prime }$.

The cells were seeded into six-well cell culture plates one day prior to transfection, with a density of 6 $\times$ 10${}^5$ cells per well. Upon 80% confluency, cell transfection was conducted following the instructions of the Lipofectamine${}^{\text{TM}}$3000 transfection kit (thermofisher.cn). NC cells were transfected with NC Small interfering RNA (siRNA) (NC). Cells in the siRNA-NOP14 group were transfected with NOP14 siRNA (NOP14-Si), and those in the blank control group were not transfected. After 8 hours of transfection, the regular medium containing 10% fetal bovine serum was replaced. Cell transfection efficiency was assessed by measuring the levels of NOP14 in cells with the help of quantitative reverse transcription polymerase chain reaction (qRT-PCR) and Western blot, respectively.The NOP14 interference sequence was designed by Ying et al. [12]. The NOP14-si1 sequence: 5${}^{\prime }$-CCAGUGACCCUGAGAGCAATT-3${}^{\prime }$, and the NOP14-si2 sequence: 5${}^{\prime }$-GGUUUAUACCUGAGCUUAUTT-3${}^{\prime }$, were synthesized by Shanghai Engine Biotech. The NC-siRNA sequence was 5${}^{\prime }$-CCGGGAACTCGAGT-3${}^{\prime }$.

Cells from the NC and NOP14-siRNA groups were seeded onto 6-well plates and cultured at 37 °C in a 5% CO${}_2$ incubator. Cell fragments were eliminated by washing with PBS, and the serum-free culture medium was collected from the same field at 0 hours, 24 hours, and 48 hours, followed by cell migration calculation.

The cells were washed twice with PBS and subjected to serum-free DMEM starvation for 12 hours. The treated cells were harvested utilizing trypsin digestion. A single-cell suspension was then prepared and seeded in the Transwell upper chamber (8 µm, 24-well insert; 5 $\times$ 10${}^3$ cells per well; Transwell; Corning, NY, USA). A total volume of 600 µL of complete culture medium was supplemented to the lower chamber and incubated for 24 hours. Subsequently, the chambers were removed, and non-migrated cells were discarded. The remaining cells were fixed in 4% paraformaldehyde for 10 minutes. After air-drying, the cells were stained with crystal violet and examined under a light microscope for quantification.

Single-cell suspensions of the NC, NOP14-si1, and NOP14-si2 groups were seeded in 6-well plates (1000 cells per well). The cells were cultured in the incubator for 10 days after 24 hours. After fixation with 4% paraformaldehyde, the cells were counterstained with crystal violet. Clones of 50 cells were observed and counted under a microscope to construct the growth curve.

The statistical significance of the included data was determined by employing either Student’s t-test (for pairwise comparisons) or two-way ANOVA (for comparisons between multiple groups). Data are presented as fold-changes, reported as mean $\pm$ Structural Equation Modeling (SEM) or $\pm$ Standard Deviation (SD). Significant differences were determined using a p-value 0.05 threshold.

As illustrated in Fig. 1A,B, the expression of NOP14 was markedly elevated in glioma cell lines (U87, U251, LN229) compared to the normal astrocyte HA1800 (Fig. 1A,B). Similarly, tissue samples collected from 4 cases exhibited a high NOP14 in tumor tissue (Fig. 1C,D). Moreover, we tested NOP14 expression in various grades of glioma (n = 20). As the glioma grade level increased, there was an observed increase in NOP14 expression in glioma immunohistochemical staining. Our results showed a significant upregulation of NOP14 expression in glioma tissues compared to non-glioma tissues (all p 0.05).

Fig. 1.

The nucleolar protein 14 (NOP14) expression is upregulated in the Glioma. (A) Western Blot for the expression of NOP14 protein in several different glioma cell lines and normal astrocytes. (B) Quantitative map of NOP14 mRNA expression and protein expression. (C) Western blotting detection of Glioma clinical adjacent four cases of benign and cancer tissue. (D) Quantitative map of NOP14 protein expression; Immunohistochemical detection of NOP14 expression of different grades. *** p 0.001, ** p 0.01, * p 0.05, ${}^{\text{ns}}$p $\ge$ 0.05. ANCT, Adjacent Non-Cancerous Tissue; WHO, The World Health Organization.

Table 1 summarizes the clinical data of the 696 GBM patients. In this study, a total of 398 men and 298 women were analyzed. The chi-square test results demonstrated a significant correlation between NOP14 and WHO grade, isocitrate dehydrogenase (IDH) status, 1p/19q codeletion, and histological type (all p 0.001). The Wilcoxon rank-sum test revealed a significant association between NOP14 and age (p 0.001). There was no significant association between NOP14 expression and other clinicopathological features.

The expression of NOP14 was analyzed in patients with various clinicopathological features in this study. Among them, Levene’s test of homogeneity of variance and one-way ANOVA test showed that the expression of NOP14 mRNA increased with the WHO level (Fig. 2A). Moreover, NOP14 was differentially expressed regarding the primary therapy outcome for glioma and the histological type (Fig. 2B,C). The Mann-Whitney U test analysis revealed a higher expression of IDH wild-type compared to mutant NOP14 mRNA (Fig. 2D), non-codel 1p/19 compared to codel (Fig. 2E). Older adults demonstrated higher expression of NOP14 (Fig. 2F), and NOP14 expression was higher in patients with a dead outcome regarding overall survival (OS). Furthermore, consistent findings were observed when analyzing the Chinese Glioma Genome Atlas (CGGA)-based data (Supplementary Fig. 1A–D).

Fig. 2.

Correlation between NOP14 expression and different clinicopathological features and the prognostic value of NOP14 in OS and DSS of GBM. Correlation between (A) NOP14 expression and WHO GBM grade, (B) primary treatment, (C) histological type, (D) IDH status, (E) 1p/19p cedar, (F) age, and (G) OS events. (H,I) Prognostic value of NOP14 in the OS and DSS of GBM. (J–M) High expression of NOP14 in different subgroups was associated with worse OS. OS, overall survival; DSS, disease-specific survival; GBM, glioblastoma. *** p 0.001, * p 0.05, ${}^{\text{ns}}$p $\mathrm{\ge }$ 0.05.

To assess the prognostic value of NOP14 in GBM patients, the study analyzed the correlation between NOP14 expression with OS and disease-specific survival (DSS). The findings revealed that high NOP14 mRNA expression implied a significant association with shorter OS and DSS in the TCGA cohort (OS: hazard ratio [HR] = 2.09, 95% CI = 1.63–2.67, p 0.001; DSS: HR = 2.12, 95% CI = 1.64–2.75, p 0.001) (Fig. 2H,I). Similar results were observed in the CGGA-based data and the PrognoScan database (Fig. 2E,F). Additionally, high NOP14 mRNA expression showed linkage with shorter OS (Fig. 2G) only in WHO grade 3 gliomas (Supplementary Fig. 1 E,F). Besides, patients with lp/19 non-codel, male gender, and age 60 showed a higher expression of NOP14 mRNA and were associated with shorter OS (Fig. 2J–L). Moreover, higher expression of NOP14 mRNA in WHO G2, G3, and G4 gliomas was also correlated with shorter OS (Fig. 2M). In conclusion, NOP14 is significantly upregulated in glioma and is strongly linked to prognosis.

The results of the univariate Cox regression analysis uncovered that high expression of NOP14 showed a linkage with increased risk among adverse clinical prognostic factors (Table 2). Significant variables discovered in the univariate analysis were subsequently included in the multivariate Cox regression analysis. It was evident that NOP14 expression was an independent risk factor for overall survival (HR: 0.530, p = 0.009), thus indicating a predictive value for overall survival in the disease. Moreover, the WHO grade, particularly in the G3 stage, IDH status, histological type, and TP53, demonstrated a predictive advantage for the clinical outcome (Fig. 3A). The results of the univariate logistic regression analysis revealed significant clinicopathological differences between the groups, including WHO grade (odds ratio [OR] = 4.040, 95% CI = 2.857–5.765), 1p/19q codeletion (OR = 3.016, 95% CI = 2.093–4.398), IDH status (OR = 0.334, 95% CI = 0.240–0.461), and age (OR = 2.347, 95% CI = 1.604–3.471) (all p 0.001) (Table 3).

Fig. 3.

The prognostic value of NOP14 in GBM. (A) Multivariate Cox regression visualized in the forest plot. (B) NOP14 expression distribution and survival status. 0: dead, 1: alive. (C) Diagnostic receiver operating characteristic curve (ROC) curve of NOP14. (D) Time-dependent ROC curve of NOP14.

Fig. 3B displays the distribution of NOP14 expression, the survival status of GBM patients, and the NOP14 expression profiles. The group with a low-risk score showed a low expression of the dashed line, while the group with a high-risk score exhibited a high expression of the dashed line. With an increase in the risk score of GBM patients, there was a gradual rise in the number of green points, indicating an escalation in the mortality rate among GBM patients. This indicates the survival rate and high risk.

There receiver operating characteristic curve (ROC) curve for the NOP14 score demonstrated its diagnostic accuracy with an Area Under Curve (AUC) of 0.814 (95% CI = 0.793–0.835) (Fig. 3C). The accuracy of this study was also assessed over time to evaluate the predictive capability of NOP14 for overall survival at 1, 2, and 3 years (Fig. 3D).

In the NOP14 group, a total of 5892 genes showed differential expression, with 3126 DEGs upregulated (53.0%) and 2766 DEGs downregulated (46.9%) (adjusted p-value 0.05, log${}_2$ fold change $|$Log${}_2$-FC$|$ $>$2) (Supplementary Fig. 2A). The Gene Ontology (GO) analysis revealed that DEG-related NOP14 significantly regulated various biological processes, including neurotransmitter transport, regulation of postsynaptic membrane potential, calcium ion-regulated exocytosis, ion channel complex formation, synaptic membrane function, substrate-specific channel activity, ion-gated channel activity, and neuroactive ligand-receptor interaction (Fig. 4A, Table 2). The heat map illustrates the correlation between NOP14 and the 15 genes, as demonstrated in Fig. 4B.

Fig. 4.

Enrichment analysis of NOP14 in GBM. (A) Biological process enrichment related to NOP14-related genes. (B) A network of NOP14 and its 20 potential co-interaction proteins. (C–F) The results of enrichment analysis from GSEA. GSEA, Gene Set Enrichment Analysis.

To further elucidate the biological function of NOP14, GSEA was implemented on the disparities between the high and low expression datasets of NOP14 (Table 4). This analysis enabled us to identify the specific REACTOME and KEGG pathways associated with NOP14. Besides, 244 pathways exhibited significant differences in the enrichment of the REACTOME and KEGG pathways in the samples with high NOP14 expression (FDR 0.05, adjusted p 0.05). Table 3 displays the REACTOME and KEGG pathways that are significantly enriched based on NES. The GSEA analysis of the REACTOME depicted that high levels of NOP14 were positively correlated with various aspects of the cell cycle, such as cell cycle progression, mitotic division, and cell cycle checkpoint regulation, as well as with M phase and RHO GTPase effectors (Fig. 4C). On the other hand, high levels of NOP14 were negatively correlated with neuronal system function, chemical synapse transmission, neurotransmitter receptor activity, postsynaptic signal transmission, and G Protein-Coupled Receptor signaling (Fig. 4D). A GSEA of the KEGG pathway revealed significant enrichment in several pathways, including the Cell Cycle, Cytokine-Cytokine Receptor Interaction, Focal Adhesion, ECM Receptor Interaction, and P53 Pathway, which exhibited a positive correlation with high levels of NOP14 (Fig. 4E). On the other hand, the following pathways were found to be negatively correlated with high levels of NOP14: Neuroactive ligand-receptor interaction, Calcium signaling pathway, Amyotrophic lateral sclerosis, Long-term potentiation, Olfactory transduction, and PPAR signaling pathways (Fig. 4F). Our findings indicate a strong association between NOP14 expression and neural conduction, cell cycle control, and cytokine receptor interactions.

A Spearman correlation analysis was implemented to investigate the relationship between NOP14 and the levels of immune cell infiltration, quantified by the ssGSEA score. The infiltration level of Th2 cells was positively correlated with NOP14 expression (Spearman’s R = 0.448) (Fig. 5A) and was elevated in the high NOP14 expression group (all p 0.001) (Fig. 5B). Conversely, follicular helper T cells (Tfh) exhibited a negative correlation with NOP14 expression (Spearman’s R = –0.342) (Fig. 5C), and NOP14 expression declined in the high NOP14 expression group (all p 0.001) (Fig. 5D). Th2 cells, T helper cells, aDC, macrophages, neutrophils, Tgd, NK cells, and eosinophils exhibited a positive relationship with NOP14. Besides, a negative correlation existed between TFH, NK, CD56bright cells, mast cells, B cells, Th1 cells, TReg, Tem, Tcm, pDC, DC, and NOP14 (Fig. 5E). Thus, NOP14 plays a significant role in the immune infiltration of GBM. The heatmap was used to assess and observe the varying degrees of correlation between the proportions of the 24 types of tumor-infiltrating immune cell subsets (Fig. 5F).

Fig. 5.

The results of analysis between NOP14 expression and immune infiltration. (A) The positive correlation between NOP14 expression and Th2 cells. (B) Th2 cells’ infiltration level in different NOP14 expression groups. (C) The negative correlation between NOP14 expression and Follicular helper T cell (TFH). (D) TFH’infiltration level in different NOP14 expression groups. (E) Correlation between NOP14 expression level and the relative abundances of 24 immune cells. (F) Heat map of 24 immune infiltration cells in GBM. ***, p 0.001.

Research has demonstrated that relying solely on clinicopathological characteristics does not yield accurate prognostic information for tumor patients, resulting in excessive or inadequate treatment. Emerging evidence indicates that incorporating tumor microenvironment (TME)-related immune scores and the proportion of tumor-infiltrating immune cells (TIC) in the TME can enhance the predictive accuracy of conventional staging systems [11, 12, 13, 14, ]. TME comprises tumor cells, stromal cells, immune cells, extracellular matrix (ECM), signaling molecules, and cytokines. Immune cells in the TME can influence tumor growth [15, 16, 17]. To explore the association between NOP14 levels and stromal cell infiltration, we initially analyzed the Tumor Immune Single Cell Center [13] database to determine the main cell subset expressing NOP14. The GSC_GSE148842_Smartseq2 dataset (a glioma single-cell dataset) was explored. We discovered the expression of NOP14 in both single-cell subsets of immune and stromal cells (Fig. 6A–C). Subsequently, we examined the specific cell subsets of NOP14 species exhibiting high glioma expression. The results demonstrated that CD8${}^{+}$Tex cells predominantly expressed NOP14. Additionally, NOP14 exhibited higher expression in CD8${}^{+}$Tex cells than other cell subsets (Fig. 4C). These findings strongly clarify the pivotal role of NOP14 in modulating the functions of CD8${}^{+}$Tex cells.

Fig. 6.

Correlation between NOP14 levels and stromal cell infiltration. (A–E) GFPT2 was expressed in both immune and stromal cell single-cell subpopulations in a Glioma single-cell GSE dataset (GSC_GSE148842_Smartseq2) from the Tumor Immune Single-cell Hub (TISCH) dataset. (D) Corrplot was used to perform the correlation between NOP14 levels and CD8T-associated factors. (E) A corrplot was used to perform the correlation between NOP14 levels and EMT-related factors.

We further pinpointed the correlation between NOP14 and biomarkers linked to CD8${}^{+}$T cells. CD8${}^{+}$ tumor-infiltrating lymphocytes (CD8${}^{+}$TILs) are the primary immune cell population responsible for tumor eradication. However, numerous studies indicate that continuous antigen stimulation in the TME, along with the presence of immunosuppressive factors, leads to the progressive exhaustion of CD8${}^{+}$T cells. Consequently, the functions of these exhausted CD8${}^{+}$T cells, including their ability to kill tumors and proliferate, gradually diminish, eventually resulting in their complete functional loss [18]. Functional identification of killing T cells can be based on various physiological states. For instance, effector T cell cytokines such as CCL3 (MIP-1$\alpha$), CCL4 (MIP-1$\beta$), CCL5 (RANTES), effector memory T cell surface markers including CD44, low expression of CD197 (CCR7), and KLRG1, and the tissue-resident memory T-cell transcription factor KLF2 can be used. Studies have shown that CTLA-4 can suppress CD28 co-stimulatory signals by binding to B7 molecules, thereby inhibiting the activation of CD8T cells [19, 20]. Our study noted a strong negative correlation between NOP14 and the CD8${}^{+}$T marker genes CCL3, CCL4, and CCL5. Additionally, we found a positive correlation between NOP14 expression and forkhead box protein 3 (Foxp3) (Fig. 6D).

CD8${}^{+}$Tex bears great responsibility for tumor development, and the synergistic effects of these inhibitory features induce Epithelial–mesenchymal transition (EMT) and impact tumor cell survival [21]. To validate the impact of NOP14 expression on EMT, we investigated the correlation between NOP14 levels and EMT-associated markers. The findings indicated a strong correlation between NOP14 expression and EMT-related factors (Fig. 6E). These findings unveil that NOP14 could serve as a potential biomarker for modifications in the tumor immune microenvironment, especially those related to CD8${}^{+}$T cells.

In vitro, interference of NOP14 was observed in LN229 and U251 cells (Supplementary Fig. 3A,B). The cell wound scratch assay demonstrated that the inhibition of NOP14 resulted in a reduction in cell migration (Fig. 7A,B,G). Similarly, the Transwell assay indicated a decrease in cell invasion upon reduction of NOP14 expression (Fig. 7C,D,H). Furthermore, the clone formation assay provided evidence that the inhibition of NOP14 impaired the colony formation ability of GBM cells (Fig. 7E,F,I).

Fig. 7.

NOP14 promotes glioma cell migration and invasion in vitro. (A,B) Images of wound healing experiments in the NC and NOP14 gene interference groups. (G) Quantification of the wound closure condition. (C,D) Images of migrated transwell experiments in NC and NOP14 gene interference groups. (H) Quantitative analysis of the number of migrating glioma cells. (E,F) Plate cloning experiments of the NC and NOP14 gene interference groups. (I) Quantitative analysis of the plate clones. *** p 0.001.