Introduction

Glioma is a primary central nervous system (CNS) tumor originating from brain glial cells1,2. With an annual incidence of approximately 6.4 per 100,000 individuals, it is the most common primary malignant tumor of the CNS in adults3. World health organization (WHO) classification of tumors of the CNS classified gliomas as grades 1–4. Grade 1 and 2 gliomas are classified as low-grade gliomas, characterized by lower malignancy and relatively favorable patient prognosis. In contrast, grade 3 and 4 gliomas are high-grade gliomas, with higher malignancy and poorer patient prognosis. Among them, grade 3 gliomas account for 10–15% of all gliomas, with a median survival time of approximately 37.6 months, and grade 4 gliomas, also known as glioblastomas multiforme (GBM), account for approximately 57% of all gliomas, with a median survival time of only 14.4 months4,5,6. At present, the conventional treatment of glioma is mainly via surgical resection, combined with radiotherapy and chemotherapy. However, unlike other solid tumors, the blood–brain barrier, immune microenvironment and unique tumor phenotype all made treatment of glioma more challenging than other cancers. Therefore, further revealing the molecular pathogenic mechanisms of glioma and searching for the potential therapeutic targets have important clinical significance for improving the prognosis of glioma patients and developing new therapies approaches.

Calnexin (CANX) is an endoplasmic reticulum (ER) specific type I transmembrane protein, which is frequently used as an ER marker7. It plays an important role in proteins folding and assembly, degradation of misfolded or unfolded proteins, and quality control of glycoproteins in the ER8,9,10. In addition, CANX is also involved in the regulation of intracellular Ca2+ homeostasis, activation of the unfolded protein response, and regulation of cell apoptosis11,12,13. Given its important role in various biological processes, increasing number of studies have shown that the dysregulation of CANX is closely related to the occurrence and development of human diseases, including cancer. Previous studies have shown that CANX is significantly up-regulated in a variety of tumors, and its overexpression is associated with poor prognosis of patients, such as hepatocellular carcinoma, head and neck squamous cell carcinoma, lung cancer, and so on14,15,16. Furthermore, in vitro function experiments have revealed the contribution of CANX in the proliferation and apoptosis of colorectal cancer and osteosarcoma cells17,18. These studies suggest that CANX may serve as a potential prognostic biomarker and therapeutic target for cancer intervention.

Recently, the role of CANX in glioma progression has also received attention. For example, Patel et al.19 reported CANX was highly expressed in high-grade gliomas and associated with shorter survival of patients. Zhang et al.20 used the sequencing data in TCGA, GTEx and CGGA databases and built a risk feature model of 6 genes, including CANX, which could stratify risk and predict overall survival of patients with primary low-grade glioma. Although the above studies indicated CANX may play a role in the progression of glioma, the biological function in glioma is largely unknown.

In this study, proteomic technology and bioinformatic analysis identified CANX as a differentially expressed protein associated with glioma progression. The role of CANX in glioma malignant progression and its molecular mechanism were further investigated through molecular and cellular function experiments. Additionally, the influence of CANX expression on the immune microenvironment of glioma was analyzed by bioinformatics. These findings provide new insights into the role of CANX in glioma progression and contribute to a better understanding of the molecular mechanisms underlying gliomagenesis.

Results

Proteomic analysis and screening of differentially expressed proteins

Based on LC–MS/MS proteomic analysis, a total of 99 (2.2%) up-regulated and 466 (10.5%) down-regulated differentially expressed proteins were identified in GBM samples comparing with paracancer samples (Fig. 1a). Protein–Protein interaction (PPI) network analysis of above 565 differentially expressed proteins was performed using String database (http://string-db.org), and the resulting PPI network was imported into Cytoscape software. Then, the hub genes were ranked using three algorithms from the Cytohubba plugin: Degree, Closeness and Betweenness21. The top 15 hub genes from each algorithm were selected (Fig. 1b). Venn diagram was used to identify the intersection of these hub genes (Fig. 1c), and 9 differentially expressed genes were obtained: HSPA8, SNAP25, YWHAB, MAPK3, PARK7, HRAS, SOD1, CANX and CALM3. Univariate Cox regression analysis and multivariate Cox regression analysis were used to further explore the correlation between the 9 differentially expressed genes (DEGs) and patient prognosis in TCGA database, showing that CANX, PARK7 and HSPA8 are independent prognostic factors for glioma (p < 0.01) (Fig. 1d). The above 3 DEGs were further verified based on the GEPIA2 database, and CANX was finally selected as the target protein (Fig. 1e).

Fig. 1
figure 1

Screening of DEGs in glioma. (a) Volcano plot of differentially expressed proteins in GBM tissues. Red dots indicate up-regulated proteins based on log2(Fold Change) > 1 and p-value < 0.05. Green dots indicate down-regulated proteins based on log2(Fold Change) < -1 and p-value < 0.05. Gray dots indicate proteins that are not significantly different. (b) The PPI showing the top 15 proteins according to the algorithms Degree, Closeness and Betweenness in CytoHubba. (c) Venn diagram showing the 9 differentially expressed genes according to the three algorithms. (d) The forest plot shows the results of the univariate Cox regression analyses of the correlation between 9 DEGs and prognosis and the multivariate Cox regression analyses of the correlation between 6 DEGs and prognosis in TCGA. (e) Comparison of DEGs expression levels in tumor and normal tissues in GEPIA2 database. LGG: low-grade glioma.

Full size image

CANX is significantly overexpressed in glioma and is associated with poor patient prognosis

Glioma patients in the TCGA database were divided into high-expression group and low-expression group based on the expression level of CANX, and correlation analysis of clinicopathological features was conducted after the incomplete survival time data was excluded. KM survival analysis showed that the overall survival of glioma patients in the CANX high-expression group was significantly shorter than in the low-expression group (p < 0.001) (Fig. 2a). Clinicopathological correlation analysis showed that except gender, transcriptome subtype, age, MGMT promoter methylation, IDH mutation, and WHO grading were all significantly correlated with CANX expression (Fig. 2b). The expression level of CANX in 4 pairs of GBM tissues and paracancer tissues were detected by western blotting (WB). The results showed that CANX expression levels was significantly up-regulated in most tumor tissues compared with paracancer tissues (Fig. 2c), which was consistent with the LC–MS/MS data. Next, WB was used to verify the expression level of CANX in glioma cell lines. The results showed that CANX expression levels was significantly up-regulated in T98G, A172 and U251 cell lines compared with HA (p < 0.001), but no significant difference was found in U87 and SHG44 cell lines (Fig. 2d).

Fig. 2
figure 2

CANX expression verification and its association with prognosis of glioma patients. (a) KM curve of the survival difference between CANX high-expression and low-expression group. (b) The correlations between CANX expression and clinical parameters. (c) CANX protein expression level in GBM tumor tissues and paracancer tissues was determined by WB. β-actin was used as an internal control. (d) CANX protein expression level in human glioma cell lines and HA was determined by WB. GAPDH was used as an internal control.

Full size image

CANX silencing inhibited the malignant phenotype of glioma

In order to further investigate the function of CANX in glioma, three siRNAs targeting different regions of CANX were transiently transfected into A172 and T98G cells. Real-time Quantitative PCR (Q-PCR) and WB results showed that siRNA-CANX-401 had higher silencing efficiency and was used in subsequent analyses (Supplementary Fig. S1). CCK-8 assay, matrigel invasion assay, wound healing assay, colony forming assay and flow cytometry were utilized to evaluate the effects of CANX on proliferation, invasion, migration and apoptotic ability of GBM cells. The results showed that CANX silencing could significantly inhibit the proliferation, invasion and migration of GBM cells, and significantly increased the apoptosis rate (Fig. 3a–e). Altogether, these in vitro experiments suggested that CANX down-regulation inhibits the malignant progression of glioma.

Fig. 3
figure 3

Effect of CANX silencing on malignant progression of glioma. (a) CCK-8 assay was used to detect the viability change of GBM cells after CANX silencing. (b) GBM cell invasion ability was assessed by transwell assay. (c) Wound healing assay was used to evaluate cell migration ability after CANX silencing at 24 and 48 h. (d) Colony formation assay was used to evaluate cell proliferation ability. (e) The change of apoptosis ability in GBM cells were detected by flow cytometry after CANX silencing.

Full size image

CANX overexpression enhanced the malignant phenotype of glioma

To further demonstrate the role of CANX in glioma progression, stable cell lines with stable overexpression of CANX were constructed in A172 and T98G cells. Afterward, Q-PCR and WB results confirmed that CANX overexpression stable cell lines were successfully constructed (Supplementary Fig. S2). In contrast to CANX silencing, CANX overexpression resulted in increased proliferation, invasion and migration ability of GBM cells (Fig. 4a–d). Furthermore, CANX overexpression significantly reduced the apoptosis rate of GBM cells (Fig. 4e). Together, these results demonstrated that CANX overexpression could promote glioma progression in vitro.

Fig. 4
figure 4

Effect of CANX overexpression on malignant progression of glioma. (a) CCK-8 assay was used to detect the effect of CANX overexpression on the viability of GBM cells. (b) Transwell assay was used to detect the invasion ability of GBM cells. (c) Wound healing assay was used to detect the migration ability of GBM cells at 24 and 48 h after CANX overexpression. (d) Colony formation assay was used to detect the proliferation ability of GBM cells. (e) The effect of CANX overexpression on the apoptotic ability of GBM cells was determined by flow cytometry. NC: negative control, OE: CANX overexpression.

Full size image

CANX induces glioma survival by activating the PI3K/AKT/mTOR signaling pathway

In GSEA enrichment analysis, we found that in the CANX high-expression group, DEGs were mainly enriched in multiple tumor-related and immune-related pathways, including apoptosis, ECM receptor interaction, focal adhesion, B cell receptor signaling pathway, JAK-STAT signaling pathway, antigen processing and presentation, natural killer cell mediated cytotoxicity and cytokine receptor interaction and other signaling pathways (Fig. 5a). Importantly, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses showed that B cell receptor signaling pathway, JAK-STAT signaling pathway and focal adhesion could all transmit signal to PI3K/AKT/mTOR signaling pathway (Supplementary Fig. S3). WB was performed to detect the expression levels of AKT and p-AKT in PI3K/AKT/mTOR signaling pathway. The results showed that the expression level of p-AKT protein was significantly decreased after CANX silencing compared to the control group, while the expression level of p-AKT protein was significantly increased after CANX overexpression (Fig. 5b). It suggests that CANX may promote the progression of glioma through activating the PI3K/AKT/mTOR signaling pathway.

Fig. 5
figure 5

Signaling pathway enrichment analysis. (a) The results of GSEA enrichment analysis in the CANX high-expression group. (b) The changes of AKT and p-AKT protein expression in PI3K/AKT/mTOR signal pathway.

Full size image

Immune infiltration analysis

Previous studies have suggested that the tumor microenvironment plays an important role in tumor development. The GSEA results also showed that many immune-related pathways were associated with the CANX high-expression group. Therefore, the relationship between the CANX and glioma immune microenvironment was further investigated. Various algorithms were used to analyze the differences of immune infiltrating cells between the CANX high-expression and low-expression groups. The cibersort algorithm showed that the CANX high-expression group contained fewer B cells, CD8+ T cells, natural killer cells and mast cells, etc. (Fig. 6a,b), and most of these cells played anti-tumor roles in the tumor microenvironment, suggesting that CANX may have immunosuppressive effects. The estimate algorithm showed that the CANX high-expression group had higher immune score, stromal score, estimate score and lower tumor purity (Fig. 6c), and lower tumor purity usually implied a shorter survival time of patient, it was consistent with the results of the prognostic analysis based on the glioma clinical samples. The ssGSEA algorithm showed that the CANX high-expression group had higher amount of tumor-promoting immune infiltration cells and more immune-related functions or pathways than the low-expression group (Fig. 6d,e). Because the two groups significantly differed in immune infiltration, the correlation between CANX and four common immune checkpoints was evaluated. The results showed that the expression of PD-L1, CTLA4, HAVCR2 and PD-1 in CANX high-expression group were significantly up-regulated (Fig. 6f), suggesting that the body’s immune function was suppressed, and again indicating that CANX may play an important role in the immune escape of glioma.

Fig. 6
figure 6

Relationship between CANX expression and immune cell infiltration in glioma. (a) The infiltration level of 22 immune cells in glioma patients calculated by Cibersort. (b) The fraction of 22 immune cells in CANX low and high expression groups calculated by Cibersort. (c) Estimate evaluated the immune score, stromal score, estimate score and tumor purity of the tumor tissues. (d, e) ssGSEA calculated the activity of infiltrating immune cells and immune-related pathway. (f) Differences of four immune checkpoint genes in different CANX expression groups.

Full size image

Single cell sequencing data revealed the cell clusters correlated with CANX

Based on single cell sequencing data, we further analyzed the expression of CANX in different glioma cell type. First, batch removal effect was performed on eight primary glioma samples, and PCA visualization results were shown in Fig. 7a. Next, all the tumor cells were classified by UMAP for dimensionless clustering, and four cell clusters of astrocytes, dendritic cells (DC), endothelial cells and T cells were finally identified (Fig. 7b, c). All the cell clusters were coming from 8 glioma patients. The exact expression of CANX in four cell clusters was visualized in Fig. 7d. The violin diagram showed that CANX expression was highly correlated with astrocyte and DC cells (Fig. 7e).

Fig. 7
figure 7

Identification of cell types in single cell sequencing data analysis. (a) PCA visualization of batch removal effect of 8 primary glioma samples. (b, c) UMAP dimensionality reduction clustering and cell annotation. The color code on the upper right indicates the z-scored expression values and cell types for each colored cluster. (d) The distribution map of CANX expression in 4 cell clusters. The purple dot represents cell with CANX expression. (e) The expression level of CANX in four cell clusters. DC: dendritic cell.

Full size image

Discussion

Gliomas, the most common malignant brain tumors, currently have poor clinical treatment outcomes. Ongoing research into the molecular pathogenesis and biomarker discovery of gliomas is critical for developing new therapeutic strategies.

CANX has long been known as an ER marker, but its role in tumorigenesis has not been fully recognized, especially in glioma. In this study, we analyzed the differentially expressed proteins in GBM tissues based on proteomics technology, and found that CANX was significantly overexpressed in GBM tissues and correlated with poor patient prognosis, suggesting that it may be an important molecule involved in the malignant progression of glioma. Similar to our study, Patel et al. also used LC–MS/MS to identify CANX exhibiting increased expression in GBM tissues and found it was significantly correlated with the shorter survival of GBM patients19. Next, we explored the role of CANX in glioma cells. A series of cell function experiments showed that CANX can promote the proliferation, migration and invasion of GBM cells, indicating CANX may act as an oncogene in glioma development and progression. The oncogenic effect of CANX was also verified in multiple types of tumors. For example, Akiko Okayama et al. demonstrated that high expression of CANX could promote the proliferation, migration and invasion process of lung adenocarcinoma cells22. In a study of metastasis biomarkers in breast cancer, Moradpoor et al. similarly found that CANX protein was frequently expressed and closely associated with tumor metastasis23. Although the role of CNAX in tumor progression has been reported, the underlying molecular mechanism in glioma has not been elucidated.

Then, we made a preliminary exploration of the potential molecular mechanism of CANX. Western blotting confirmed that CANX overexpression could activate AKT. AKT, also known as protein kinase B, is a key component of the PI3K/AKT/mTOR axis, which plays an essential role in regulating cell proliferation and apoptosis24. mTORC1 is overactivated in human cancers as a downstream effector of many commonly mutated oncogenic pathways. Akt has been proven to promote mTORC1 activation through a variety of mechanisms, such as increased the phosphorylation of TSC2 or PRAS4025,26. In another study, Yan et al.27 had also indicated that CANX is an important regulator for leucine-stimulated mTORC1 pathway, and its loss of function could render the mTORC1 signaling insensitive to leucine deprivation. This report further confirms our speculation that CANX may promote glioma progression by activating the PI3K/AKT/mTOR signaling pathway. Therefore, for glioma patients with CANX abnormally overexpressed, some mTOR kinase inhibitors may be available for treatment, such as paxalisib, dactolisib, onatasertib, bimiralisib and SF-1126, which may shed light on new therapeutic strategy of glioma.

After that, we analyzed the relationship between CANX expression and glioma immune microenvironment. The GSEA results suggested that CANX was associated with multiple immune signaling pathways, and immune infiltration analysis again suggested that CANX may play an important role in immune escape of glioma. Similarly, several other studies have also reported that CANX may be involved in immune escape of tumor cells. For example, in oral squamous cell carcinoma, Chen et al. found that CANX inhibited the infiltration and effector functions of CD4+ and CD8+ T cells in tumor microenvironment, enhanced the expression of PD-1 on T cells, and promoted tumor growth, and this finding was also confirmed in melanoma28. It is well known that the low expression of major histocompatibility complex class I (MHC-I) on the surface of tumor cells is one of the main mechanisms by which tumors escape from T-cell recognition and destruction29. A study of colorectal cancer reported that aberrant expressions of CANX could prevent successful assembly of MHC-I, antigen peptides processing and presentation on the tumor cell surface, thereby aiding tumor cell immune escape30. Similarly, another study also reported that CANX expression was inversely correlated with MHC-1, which found that increased expression of SMAR1 could positively regulate MHC-1 expression by down-regulating CANX31. In kidney renal clear cell carcinoma, CANX had also been confirmed to be an immune-autophagy-related gene with independent prognostic value, which was significantly associated with immune score, immune checkpoint, methylation, and could be used as a potential target and effective prognostic biomarker for immunotherapy combined with autophagy32. Collectively, these works revealed the role of CANX in regulating the anti-tumor response of T cells in tumor cells, suggesting that CANX may be a new potential target for anti-tumor immunotherapy.

Finally, based on single cell sequencing data, this study analyzed the expression differences of CANX in different glioma cell types and found that CANX was mainly expressed in astrocytes and DC cells. It has been shown that activation of the JAK/STAT signaling pathway can lead to precocious astrogliogenesis and promote astrocyte differentiation33. In the GSEA enrichment analysis, we noticed that in the CANX high-expression group, differentially expressed genes were also enriched in the JAK/STAT signaling pathway. Henrik et al. reported the immunomodulatory properties of tumor-associated astrocytes in the human glioblastoma environment. The astrocyte-microglia interaction in the tumor environment promoted release of anti-inflammatory cytokines, which could inhibit T cell activation, leading to the formation of glioblastoma immunosuppressor microenvironment34. Therefore, it is inferred that CANX may maintain the anti-inflammatory microenvironment of glioma and promote tumor immune escape by activating the JAK/STAT signaling pathway.

In summary, this study demonstrated that CANX is up-regulated in glioma and is associated with poor patient outcomes. It can promote the proliferation, migration, invasion and anti-apoptotic phenotypes of glioma cells, and this tumor-promoting effect may be exerted by activating the PI3K/AKT/mTOR signaling pathway. In addition, this study also found that CANX may be involved in immune escape of gliomas. The above results revealed that CANX plays an important role in promoting malignant progression and immune escape of gliomas, suggesting that CANX may be a valuable therapeutic target.

Methods

Clinical samples and cell lines

GBM tissue samples were collected from the Department of neurosurgery, Tangdu Hospital, Air Force Military Medical University (Xi’an China). All patients underwent primary surgery and did not receive radiotherapy or chemotherapy. Tumor tissues were confirmed as grade 4 glioblastoma by pathological diagnosis and were stored in liquid nitrogen. This study was performed in accordance with the principles of the Declaration of Helsinki and approved by the Ethics Committee of Tangdu Hospital. Informed consent was obtained from all participants.

Human normal astrocyte cell line HA, human glioma cell lines A172, T98G, U251, U87 and SHG44 were purchased from the cell repository of Wuhan Institute of Virology (Wuhan, China). The cell lines of HA, A172, U251 and SHG44 were cultured with DMEM high-glucose medium (Pricella Life Science & Technology Co., Ltd., China), containing 10% fetal bovine serum (Gibco, US). The cell lines of T98G and U87 were cultured with MEM (containing NEAA) medium (Pricella Life Science & Technology Co., Ltd., China), containing 10% fetal bovine serum (Gibco, US). All cell lines were cultured at 37℃ and 5% CO2.

Proteomic data analysis

Label-free quantitative proteomic analysis was performed on 3 primary glioblastoma tissues and 3 paracancer tissues by liquid chromatography-tandem mass spectrometry (LC–MS/MS). Detailed procedure was described in a published article by our lab35. The data screening conditions were |log2Fold Change|> 1 and p-value < 0.05.

Data collection

The transcriptome sequencing data and clinical information of 702 cases of gliomas samples, including 5 cases of paracancer samples, 529 cases of low-grade glioma (LGG) and 168 cases of GBM, were downloaded from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/). Gene sets of 22 immune cell types were obtained from CiberSort36. The single-cell expression dataset GSE138794 is derived from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). This dataset contains single-cell sequencing data of 9 gliomas, in which 8 cases were primary GBM and 1 case was secondary GBM37.

Survival analysis and verification

Univariate and multivariate Cox regression analysis were used to screen for genes significantly associated with prognosis in glioma patients, the hazard ratio (HR) was calculated. According to the median expression of CANX gene, glioma patients in the TCGA database were divided into high-expression group and low-expression group. Kaplan–Meier (KM) analysis was used to compare the prognosis between the two clusters, and the Log-Rank test was used to determine if they were significantly different. Based on the clinicopathology parameters, clinicopathological correlation analysis was performed to compared whether there are differences in gender, transcriptome subtype, IDH mutation, MGMT promoter methylation, age and WHO grading.

Total RNA extraction and real-time quantitative PCR

Total RNA was extracted using TRIzol reagent (Sigma, USA) and concentration of RNA was determined by Nanodrop 2000. The RNA (1 μg) was reverse transcribed for the synthesis of cDNA using the TransScript® All-in-One First-Strand cDNA Synthesis SuperMix reagent kit (AT341, TransGen Biotech, China). The qRT-PCR was performed using the TransStart® Green qPCR SuperMix (AQ101, TransGen Biotech, China) on ABI Prism 7500 Fast, and the expression of GAPDH was used as an internal control. Amplification protocols were as follows: 94 °C for 30 s; 40 cycles of 94 °C/5 s and 60 °C/34 s. The primer sequences were shown as follows: CANX-Forward: 5′-CATCATGCCATCTCTGCTAAA-3′, CANX-Reverse: 5′-GGTTTTTGTGTCGGAAGATG-3’; GAPDH-Forward: 5′-GTCTCCTCTGACTTCAACAGCG-3′, GAPDH-Reverse: 5′-ACCACCCTGTTGCTGTAGCCAA-3′. Relative mRNA expressions were calculated by 2−ΔΔCt method. All reactions were performed three times.

Western blotting

Total proteins from GBM cells were extracted in RIPA buffer (Biosharp, China) containing 1% phosphatase inhibitor and 1% protease inhibitor and quantified by bicinchoninic acid (BCA) kit (ZHHC, China). Proteins were isolated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and the gels were transferred onto polyvinylidene fluoride (PVDF) membranes. Then, PVDF membranes were incubated with specific primary antibodies and overnighted at 4 °C, followed by incubation of secondary antibodies for 1 h. Finally, bands were visualized using an enhanced chemiluminescence (ECL) kit (ZHHC, China). All assays were performed in triplicate with independent preparations. Antibodies used in this study were shown as below: CANX antibody was purchased from Affinity Biosciences (1:2000 dilution, Cat No. #AF5362). β-actin, GAPDH, AKT and p-AKT antibody were purchased from Proteintech (Cat No. #81,115–1-RR, #60,004–1-lg, #60,203–2-lg and #28,731–1-AP, respectively). HRP Goat Anti-Rabbit IgG (H + L) and HRP Goat Anti-Mouse IgG (H + L) were purchased from ABclonal (Cat No. #AS014 and #AS003).

Lentivirus and siRNA production and transduction

The lentivirus plasmid was designed and synthesized by Genechem. CANX overexpression plasmid was designed and inserted into GV248 vector to construct stable cell lines. Stable clones transfected with CANX overexpression lentivirus were selected for 3 weeks by puromycin.

The siRNA-CANX was designed and synthesized by Tsingke. Targeting sequences were: siRNA-CANX-F: GCCAAGAAAGACGAUACCGAU; siRNA-CANX-R: AUCGGUAUCGUCUUUCU UGGC.

Cell viability assay and colony formation assay

Cell viability was measured by Cell Counting Kit-8 (CCK-8) assay (TargetMol, USA). The cell number was calculated by cell counter (ICytal, China), then cells were seeded into 96-well plates at a density of 2 × 103 cells/100 μl per well and incubated for 24 h at 37 °C with 5% CO2. After 24 h, CCK-8 mixture was added to each well and the cells were incubated for 2 h. Finally, the optical density value was measured by a microplate reader (BioTek, USA) at 450 nm.

Cell proliferation ability was measured by colony formation assay. The cells were seeded into six-well plates at a density of 200 cells per well and incubated in incubators for 2–3 weeks, during which the fluid was changed every 3 days. When visible clones appeared in the six-well plates, the culture was stopped, the medium was discarded, after washing with PBS for 3 times, fixed with 4% paraformaldehyde for 15 min, stained with 0.1% (w/v) crystal violet for 30 min, the dye was slowly washed with running water, and after drying in the air, the number of clones containing more than 50 cells was counted under the microscope and photographed. Average number of clones in 3 repeated experiments were calculated.

$${\text{Colony}}\;{\text{forming}}\;{\text{efficiency = (Average}}\;{\text{number}}\;{\text{of}}\;{\text{clones/Seeded}}\;{\text{cell}}\;{\text{count)}}\;{\text{*}}\;{\text{100\% }}$$
(1)

Wound healing cell migration assay and transwell cell invasion assay

Cell migration and invasion assays were performed as previously described38,39. 48 h after transfection (~ 80% confluence), lines were marked in the six-well plate to create an artificial wound, and images of the wound area were captured using an inverted microscope for 0 h. Then, the six-well plate was cultured at 37 °C, 5% CO2, and the cells were photographed again after 24 and 48 h. The wound areas were calculated using ImageJ.

Matrigel was diluted and vertically dropped into the transwell for pre-treatment. Then, 200 μl of cell suspension with a density of 5 × 104 cells/ml was added to the upper chamber of transwell, and 500 μl of complete medium containing 20% fetal bovine serum was added to the lower chamber. After 48 h, the cells were fixed with 4% paraformaldehyde, stained with 0.1% (w/v) crystal violet, and photographed under inverted microscope. The number of invasion cells is calculated by counting five random regions on each membrane.

Flow cytometry

Flow cytometry assay was conducted as previously described40. Cells were harvested and resuspended in binding buffer, and stained with annexin V and PI according to the instructions of the FITC-Annexin V/PI Cell Apoptosis kit (Bioscience, China). Apoptotic cells were detected by flow cytometry (Beckman Coulter, USA), and results analyzed using Flowjo.

Signaling pathway analysis

The potential mechanism of CANX regulating glioma progression was explored through gene set enrichment analysis (GSEA). The screening conditions were |normalized enrichment score (NES)|> 1, nominal (NOM) p-value < 0.05 and FDR q-value < 0.25.

Immune infiltration analysis

Immune infiltration and Single cell sequencing data processing, analysis and plotting were performed by R studio 4.2.2. CiberSort algorithm was used to predict the composition of infiltrating immune cells in each tumor sample. The Estimate algorithm was used to calculate the immune score, stromal scores, estimate score and tumor purity according to the proportion of immune cells and stromal cells of each glioma sample. The Single sample gene set enrichment analysis (ssGSEA) algorithm was used to calculate the activity of immune cells and immune-related pathway. The marker genes were obtained from previous publications41,42. The relative abundance of immune cell type in each sample was represented by the enrichment score calculated by the ssGSEA algorithm. R package “ggplot2” was used to analyze the effects of gene expression levels on four common immune checkpoint-related genes. Differences between the two sets of data were analyzed using Wilcox rank sum test, p < 0.05 were considered to be statistically significant, and all tests were two-sided.

Single cell sequencing data analysis

R package “Seurat” was used to process the single-cell data expression matrix, the single-cell sequencing data of 8 patients with primary glioma were integrated. Firstly, the function “NormalizeData” was used to normalize the single-cell gene expression data, “FindVariableGenes” was used to identify 2000 highly variable genes. Then, “FindIntegrationAnchors” and “Integratedata” were used to merge 8 glioma sample data, “RunPCA” was used to perform principal component analysis (PCA). Finally, “UMAP” was used to visualize the single-cell data, “SingleR” was used to identify and annotate cell types and to analyze of the expression levels of CANX genes in different cell types.

Statistical analysis

SPSS 22.0 is used for statistical analysis. Statistical results were expressed by mean ± standard deviation (Mean ± SD), Student’s t-test was used for two-group independent comparisons. For comparisons among more than two groups, one-way ANOVA were used. A p-value of less than 0.05 was considered statistically significant.