Bioinformatics studyGEPIA database analysis for gene expression profiling

To gain preliminary insights into CENPM expression patterns in GBM, we first queried the Gene Expression Profiling Interactive Analysis (GEPIA) web server. The TCGA and GTEx expression data were used to compare CENPM levels in GBM tissues versus non-tumorous brain samples. Expression data were normalized using transcripts per million (TPM). To increase the analysis's reliability, we included a matched normal dataset from GTEx for comparison. Statistical analysis used default parameters, with a significance cutoff of p-value < 0.05.

In vitro studyCell lines and culture conditions

We established an in vitro GBM model using two human GBM cell lines, LN229 and U251, alongside normal human astrocytes (NHAs) for comparison (Pollard et al. 2009). These cell lines were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). All cell lines were maintained in Dulbecco's Modified Eagle Medium (DMEM, Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS, Gibco Laboratories, Grand Island, NY, USA). Cultures were kept in a humidified incubator at 37 °C and 5% CO2.

We used NHAs as controls because GBM originates from astrocytes, making NHAs the most relevant non-tumorous comparison. Using NHAs as controls enables us to specifically identify the pathological changes in CENPM expression and cellular behavior unique to GBM cells, compared to their non-cancerous astrocytic counterparts.

Transfection strategies for gene modulation in GBM

To investigate the functional roles of CENPM and E2F1 in GBM, we used shRNA-mediated knockdown. Short hairpin RNAs (shRNAs) targeting CENPM (sh-CENPM#1, #2, #3) and E2F1 (sh-E2F1), along with a non-targeting control shRNA (sh-NC), were obtained from GenePharma (Shanghai, China). Additionally, we procured pcDNA3.1 expression vectors containing CENPM or E2F1 coding sequences, as well as an empty vector control, for overexpression studies.

We used Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) to deliver these constructs into GBM cells. Lipofectamine 2000 is a cationic lipid-based reagent that forms complexes with nucleic acids, facilitating their uptake into cells (Felgner et al. 1987). We carefully optimized transfection conditions to maximize gene silencing/overexpression efficiency while minimizing cytotoxicity. After transfection, cells were cultured to allow for altered gene expression before further analysis.

Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

We used reverse transcription quantitative polymerase chain reaction (RT-qPCR) to delve into CENPM mRNA expression levels within GBM cell lines. Using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), we first isolated total RNA. The Superscript™ II Reverse Transcriptase kit (Invitrogen, Carlsbad, CA, USA) then transformed this RNA into more stable complementary DNA (cDNA).

Quantitative PCR reactions used SYBR Premix Ex Taq™ (Takara Bio, Shanghai, China). This mix includes SYBR Green, a dye that fluoresces brightly when it binds double-stranded DNA, letting us track amplification in real-time. Melting curve analysis ensured signals detected came from our target amplicon and not non-specific amplification.

GAPDH, a commonly used housekeeping gene, served as our reference for normalizing CENPM mRNA expression. The 2 − ∆∆Ct method was used for relative expression calculations. Primer sequences specific to each gene were as follows: CENPM forward: 5′-CAGTCTCCAGAACACAGAGGAGTC-3′, CENPM reverse: 5′-CACCTGTGGCGAGGAAAC-3′, GAPDH forward: 5′-ACAGCCTCAAGATCATCAGC-3′, and GAPDH reverse: 5′-GGTCATGAGTCCTTCCACGAT-3′.

Western blot analysis

We used Western blot analysis to explore the protein profiles linked to CENPM's role in GBM cellular activity—particularly proliferation, invasion, and metabolic changes. First, protein lysates were carefully prepared with the preservation of protein structure in mind. SDS-PAGE (10% resolving gel) allowed us to separate these proteins by size. Next, proteins were moved onto PVDF membranes (Beyotime, Shanghai, China), followed by blocking to reduce non-specific antibody binding.

Primary antibodies targeted CENPM (1:1000, ab243820, Abcam, Shanghai, China), E-cadherin (1 µg/ml, ab231303), N-cadherin (1:5000, ab76011), hexokinase 2 (HK2, 1:1000, ab209847), Snail (1:1000, ab216347), Twist (1 µg/ml, ab50887), lactate dehydrogenase A (LDHA, 1:5000, ab52488), and β-actin (1 µg/ml, ab8226 – loading control). These incubated overnight at 4 °C. A horseradish peroxidase (HRP)-linked secondary antibody (1:2000; catalog ab7090) was then applied for 2 h at room temperature. Finally, a chemiluminescence kit (Thermo Fisher Scientific, Inc.) generated our signal. As always, β-actin allowed us to control for protein loading differences.

Cell viability assessment

The Cell Counting Kit-8 (CCK-8) assay gave us quantitative insight into how GBM cells' proliferative ability changed when CENPM and E2F1 expression was altered. This non-radioactive colorimetric assay cleverly utilizes cellular dehydrogenase activity, as only living cells convert a tetrazolium salt within the kit into a colored formazan dye. More dye means more live cells.

GBM cells were seeded (1,000 cells/well in 96-well plates). This density aimed to prevent overcrowding or nutrient limitations from affecting results. After allowing cells to adhere and recover, predetermined time points were selected (0, 24, 48, 72 and 96 h post-seeding). At each time point, CCK-8 solution (Dojindo Laboratories, Kumamoto, Japan) was added. 4 h at 37 °C gave the reaction sufficient time. Finally, we used a spectrophotometer (Thermo Fisher Scientific, MA, USA) to measure absorbance at 450 nm, correlating to the quantity of viable cells.

Assessment of cell proliferation using the EdU assay

To precisely quantify the proliferation rate of GBM cells, we utilized the 5-ethynyl-2'-deoxyuridine (EdU) incorporation assay. We chose the Cell-Light EdU DNA Cell Proliferation Kit (RiboBio, Guangzhou, China) for its proven efficiency and reliability in identifying actively dividing cells. As a thymidine analog, EdU integrates into the DNA of cells during the synthesis (S) phase, providing a clear marker for tracking DNA replication.

GBM cells were maintained under standard culture conditions prior to the addition of EdU solution (final concentration of 50 μM). Previous studies informed this concentration, aiming to maximize labeling of proliferative cells while minimizing potential cytotoxicity. Cells were incubated with EdU for 2 h, a timeframe selected to capture a substantial portion of cells in the S phase without unduly altering normal cell cycle dynamics.

After the EdU incorporation period, cells were fixed and permeabilized to facilitate the Apollo dye solution's entry. The Apollo dye specifically reacts with EdU's alkyne group, generating a fluorescent signal. We meticulously adhered to the manufacturer's protocol, ensuring reaction conditions were strictly controlled for consistent and dependable staining. DAPI (4',6-diamidino-2-phenylindole) was used as a nuclear counterstain, allowing us to easily distinguish EdU-positive (proliferating) cells from the total cell population.

Stained cells were visualized using a fluorescence microscope (Leica, Hilden, Germany) equipped with the filters necessary to differentiate the bright Apollo dye fluorescence (EdU-positive cells) from DAPI-stained nuclei.

Evaluation of cell invasion using transwell assay

We performed the Transwell invasion assay with 8 μm pore chambers (Corning, NY, United States) to analyze the invasive potential of GBM cells. This established research method evaluates cancer cells' ability to cross a physical barrier that simulates tissue structure. We coated the chambers with Matrigel (BD Biosciences, Franklin Lakes, NJ, United States), mimicking the extracellular matrix that cancer cells must degrade during the invasion process.

After suspending GBM cells in serum-free DMEM, we placed them in the Transwell's upper chamber. Lower chambers held DMEM with 20% FBS, establishing a chemoattractant gradient to draw cells downward. Cells were incubated for 48 h. Next, non-invasive cells remaining on the upper membrane surface were removed. We carefully fixed cells on the bottom of the membrane with 4% paraformaldehyde and stained them with crystal violet (0.1%) to highlight invaded cells.

To analyze invaded cells, we used microscopy (Olympus Optical Co., Ltd., Tokyo, Japan). This allowed direct count of cells and observation of morphological characteristics.

Wound healing assay

Using the LN229 and U251 GBM cell lines, we analyzed migration by creating a 'wound' in a confluent cell monolayer and monitoring the closure of this gap over 24 h. Cells were treated with either a control vector, CENPM overexpression vector, or CENPM overexpression vector plus the glycolytic inhibitor 2-deoxy-D-glucose (2-DG).

Metabolic profiling: glucose consumption, lactate production, and ATP levels

Characterizing cancer-related metabolic shifts is vital for GBM research. To this end, we used assay kits specific to measuring glucose consumption, lactate production, and ATP levels. This helps quantify aspects of cancer cells' metabolic reprogramming (the Warburg effect).

Glucose consumption assay

Using a Glucose Assay Kit (BioVision, Milpitas, CA, USA), we tracked changes in glucose concentration within the culture medium as an indicator of cellular uptake. This kit exploits the specific enzymatic oxidation of glucose to generate a colorimetric product (detected at 450 nm).

Lactate production assay

We determined lactate generated, a marker of anaerobic glycolysis, with a Lactate Assay Kit (BioVision, Milpitas, CA, USA). Based on a specific enzymatic reaction, it quantifies lactate levels within the culture medium.

ATP level determination

An ATP Assay Kit (BioVision, Milpitas, CA, USA) was used. ATP, the cell's main energy source, was detected through a luminescence reaction fueled by luciferase in proportion to available ATP. This highly sensitive assay directly reflects intracellular energy status.

Seahorse extracellular flux analysis to evaluate metabolic phenotype

The Seahorse Extracellular Flux Analysis was conducted to assess the metabolic phenotype of LN229 and U251 GBM cell lines. Cells were cultured, transfected with either shRNA for CENPM knockdown or an overexpression vector, and then prepared in Seahorse XF Cell Culture Microplates. Post-transfection, cells were incubated in unbuffered DMEM supplemented with glucose, pyruvate, and glutamine to equilibrate before measurement. Using the Seahorse XF Analyzer, the extracellular acidification rate (ECAR) was measured as an indicator of glycolytic activity. The protocol included injections of glucose to stimulate glycolysis, followed by oligomycin to inhibit ATP synthase and maximize glycolytic flux, and 2-deoxyglucose to terminate glycolysis. ECAR was recorded at multiple time points to establish a kinetic profile of glycolytic response due to genetic modifications. Data normalization was based on total protein content per well, quantified using a BCA assay. The glycolytic profiles under various genetic conditions were analyzed with ECAR values plotted over time.

Analysis of promoter activity via luciferase reporter assay

The luciferase reporter assay system offers a powerful tool to investigate the potential transcriptional regulation of CENPM by E2F1 in GBM cells.

Constructing reporter vectors

First, wild-type (WT) and mutant (MUT1/2) CENPM promoter sequences were integrated upstream of the luciferase gene within the pGL3 reporter vector (Promega, Madison, USA). This strategy links luminescence directly to the activity of the inserted promoter.

Co-transfection and measuring luciferase

For direct testing, we co-transfected GBM cells using Lipofectamine 2000. Two groups were compared: the reporter vectors paired alongside an E2F1 expression vector (pcDNA3.1/E2F1) and a control group using only the reporter and a blank vector. After a 48-h expression period, the luciferase assay system (Promega Corporation) quantified luminescence. The detected signal provides a highly sensitive measurement of E2F1's potential to boost or repress CENPM promoter activity.

Chromatin immunoprecipitation (ChIP) assay

To confirm the binding interaction between E2F1 and CENPM promoter, we conducted a ChIP assay using a commercial kit (Beyotime, Beijing, China). Briefly, LN229 or U251 cells were treated with 1% formaldehyde to cross-link DNA–protein complexes. Chromatin was subsequently sheared into 200–500 bp fragments via sonication. The cross-linked protein-DNA complexes were immunoprecipitated using either an anti-E2F1 antibody or IgG as a control. Finally, the purified chromatin DNA was analyzed by qPCR to assess the enrichment of CENPM promoter regions.

In vivo studyAnimal model and tumor cell implantation

To further investigate the role of CENPM in GBM tumor growth, we used a BALB/c nude mouse xenograft model. These immunocompromised mice are ideal for this type of study because they allow the growth of human tumor cells. We obtained twelve 4-week-old male BALB/c nude mice from the Vital River Company (Beijing, China). All animal experiments were approved by the Animal Ethics Committee of Beijing Viewsolid Biotechnology Co. LTD. All methods are reported in accordance with ARRIVE guidelines.

Experimental design

Mice were divided into two groups (n = 6 per group). One group received subcutaneous injections of U251 GBM cells with stable CENPM knockdown (sh-CENPM), while the control group received U251 cells transduced with a non-targeting shRNA (sh-NC). We injected cells into the right flank, a common site for xenografting that allows for easy tumor monitoring.

Tumor growth monitoring and assessment

After injection, we carefully monitored mice for tumor development. Tumor size was measured weekly using calipers. We calculated tumor volume using the standard formula (Volume = ½ x length x width2), providing a way to track tumor progression over time. After four weeks (a timeframe based on preliminary data), mice were euthanized and tumors were removed and weighed.

Sacrifice and tissue preparation

Mice were deeply anesthetized (75 mg/kg ketamine / 8 mg/kg xylazine, ip). We carefully performed transcardial perfusion to flush the circulatory blood with 0.9% saline, followed by fixation with 4% PFA (pH 7.4). Brains were then excised and postfixed overnight in PFA before storing them in a sucrose/PFA solution (pH 7.4) for protection. Using a freezing microtome (Leica Microsystems), we obtained precise 25 μm tissue sections. These brain sections were collected in an antifreeze solution and stored at − 20 °C to preserve them for later analysis.

Immunostaining protocol

Our immunostaining process began with a thorough KPBS wash. We then applied a blocking solution (KPBS-based, containing 1% BSA, 4% relevant serum, and 0.4% Triton X-100) for 2 h to reduce non-specific binding. Primary antibodies (CENPM, Ki-67, HK2, and LDHA all at 1/1000, Abcam, Shanghai, China) incubated overnight at 4 °C. After a washing step, we applied appropriate secondary antibodies for 2 h at room temperature. To aid visualization, DAPI (Sigma‒Aldrich) counterstaining and mounting with Fluoromount-G were performed. This prepared the sample for detailed analysis via Zeiss LSM800 confocal microscopy with Image J software (Ghareghani et al. 2023).

Statistical analysis

We used SPSS software (Version 25.0, IBM Corp., Armonk, NY, USA) and GraphPad Prism (Version 8.0, GraphPad Software, San Diego, CA, USA) for our statistical analyses. Data are presented as mean ± standard deviation (SD), with at least three independent experiments. Student's t-test was used for comparisons between the two groups, and one-way ANOVA with Tukey's post-hoc test was used for multiple comparisons. Pearson's correlation coefficient analyzed the relationship between E2F1 and CENPM expression.

P-values less than 0.05 were considered statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001). We also used Kaplan–Meier survival analysis with the log-rank test to assess differences in survival times between experimental groups. To ensure the robustness of our data, we verified normality (Shapiro–Wilk test) and homogeneity of variances (Levene's test) before applying parametric tests, and used non-parametric tests when necessary.