METTL3-mediated m⁶A modification of CACNA1E promotes osteosarcoma progression and chemoresistance by enhancing WNT7B-mediated Ca²⁺ signaling

Abstract

Background

Osteosarcoma (OS) is the most prevalent primary cancer of the bone. Metastasis and chemoresistance are the major obstacles to the improvement of OS prognosis, in which N⁶-methyladenosine (m⁶A) modification plays an important role, but the exact molecular mechanisms are still unclear.

Methods

MeRIP-seq and RNA-seq were conducted on OS and paired adjacent normal tissue samples, which determined CACNA1E as a key m⁶A-modified molecule. In vitro and in vivo models were established to evaluate the function of CACNA1E on OS growth, metastasis, and methotrexate (MTX) resistance, and to explore the upstream regulators and downstream effectors of CACNA1E.

Results

CACNA1E exhibited notable m⁶A hypermethylation and upregulated expression in OS than adjacent normal tissues. CACNA1E knockdown effectively hindered OS growth, lung metastasis, and MTX resistance. METTL3, an m⁶A “writer” boosted the mRNA stability of CACNA1E through m⁶A modification, and this process was recognized and enhanced by IGF2BP2, an m⁶A “reader”. WNT7B was identified as a downstream molecule of CACNA1E. CACNA1E facilitated OS progression and MTX resistance by enhancing the non-canonical Wnt/Ca²⁺ signaling through transcriptionally activating WNT7B. Furthermore, a novel combination treatment of targeted inhibition of CACNA1E with MTX had a synergistic effect on suppressing OS progression.

Conclusions

Collectively, our findings uncover that METTL3-mediated m⁶A modification of CACNA1E contributes to OS progression and chemoresistance through enhancing WNT7B-mediated non-canonical Wnt/Ca²⁺ signaling. Targeted inhibition of CACNA1E in combination with MTX may be a promising alternative therapeutic strategy for patients with MTX-resistant OS.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12943-025-02553-x.

Background

Osteosarcoma (OS) is the most prevalent primary malignant bone tumor and often affects adolescents and children [1]. The age-adjusted incidence of OS in children and adolescents aged 0 ~ 24 years is 4.4 per million people each year [2]. About 60% of patients are aged 10 ~ 20 years, with OS regarded as the second major cause of death in this age group [3]. Although the incidence of OS is relatively rare, the disability and mortality rates are high. The treatment mainstay of OS is neoadjuvant chemotherapy-surgery-consolidation chemotherapy, of which methotrexate (MTX) is one of the first-line chemotherapy agents [4, 5]. With the introduction of systematic chemotherapy, the five-year survival rate for patients with localized OS has increased to 60 ~ 70% [6]. However, OS has a high propensity to metastasize, with the lungs being the most common site of metastasis [7, 8]. Chemoresistance is a main obstacle to the improvement of OS treatment [9]. The five-year survival rate for patients with metastatic or recurrent OS is as low as 20 ~ 30% [10]. Given the bottleneck of traditional treatment options, it is urgently needed to develop new treatments to improve patient prognosis.

Ca²⁺ functions as a universal secondary messenger that controls various cellular processes, e.g., proliferation, migration, and cell death, all of which are indispensable for cancer formation [11]. Dysregulated Ca²⁺ homeostasis is a crucial driver in cancer occurrence and progression as well as affects the treatment response [12]. At current, Ca²⁺ signaling has become an attractive direction for the development of novel anticancer treatment strategies. Calcium voltage-gated channel subunit alpha 1E (CACNA1E) is a key regulator of Ca²⁺ signaling, which modulates intracellular Ca²⁺ via forming Ca²⁺ entry points, and initiates rapid synaptic transmission via directly interacting with G and SNARE proteins [13]. Accumulated evidence demonstrates that CACNA1E is involved in modulating the progression of several human cancers [14–16], but the specific mechanism by which CACNA1E modulates Ca²⁺ signaling and its contribution to the progression of OS remain indistinct. The non-canonical Wnt/Ca²⁺ signaling pathway is activated when the Wnt ligand binds to the Frizzled family proteins, triggering events that lead to the release of intracellular calcium, which is also implicated in cancer progression [17, 18]. Given the shared Ca²⁺ signaling, we hypothesize that CACNA1E may impact the non-canonical Wnt/Ca²⁺ signaling pathway.

N⁶-methyladenosine (m⁶A) is the most prevalent RNA modification in eukaryotic cells, which affects the fate and expression of RNA through regulating RNA splicing, nuclear export, stability, and translation [19, 20]. Aberrant m⁶A modification and its related modifiers, especially “writers”, “erasers”, and “readers”, are commonly found in human cancer types, including OS. In recent years, dysregulated gene expression induced by m⁶A modification has been extensively explored in OS and determined as an important factor for malignant progression and poor clinical outcomes [21–23]. It has been reported that Ca²⁺ signaling is potentially modulated by m⁶A modification through affecting Ca²⁺ signaling-related gene expression [24]. However, there is currently a lack of extensive knowledge regarding the regulatory mechanisms of m⁶A modification in Ca²⁺ signaling in OS.

In this study, we determined CACNA1E as a novel driver of OS progression and MTX resistance. Mechanistically, METTL3-driven m⁶A modification boosted CACNA1E mRNA stability in an IGF2BP2-depenent manner. The upregulation of CACNA1E subsequently activated the non-canonical Wnt/Ca²⁺ signaling pathway by transcriptionally activating WNT7B, thus facilitating OS growth, metastasis, and MTX resistance. Importantly, CACNA1E showed potential as a therapeutic target to overcome MTX resistance, and targeted suppression of CACNA1E in combination with MTX may be a promising treatment strategy, especially for patients with MTX-resistant OS.

Materials and methods

Human specimens

OS and paired adjacent normal tissue specimens were gathered at Guangxi Medical University Cancer Hospital (Nanning, China). The pathological examinations of the removed specimens were validated following the guidelines of the established protocols outlined by the World Health Organization. Adjacent normal tissue was taken from 2~ 5 cm away from the tumor lesion. Each participant signed written informed consent, and ethical approval of the study was granted by the Ethics Committee of Guangxi Medical University Cancer Hospital (2024-D100-01).

Cell culture

Human 293 T (HTX1559), SJSA-1 (HTX2137), MNNG (HTX1631), and U2OS (HTX1634) cell lines were obtained from Otwo Biotech (ShenZhen, China), and hFOB1.19 (CL-0353) and SAOS2 (CL-0202) cell lines were acquired from Procell (Wuhan, China). OS and 293 T cells were cultivated in DMEM/F-12 (Meilun Biotech, Dalian, China) and DMEM (Meilun Biotech), respectively. All the media were supplemented with fetal bovine serum (FBS; Solarbio, Beijing, China) and 1% penicillin–streptomycin solution (Solarbio). All the cell cultures were maintained in an incubator (ThermoFisher, MA, USA) with 5% CO₂ at 37 °C. MTX-resistant OS cell lines (MNNG^MR and SJSA-1^MR) were generated by gradually increasing the concentration of MTX (HY-14519, MCE, NJ, USA), as previously reported [9].

Plasmids, transfection, and lentiviral infection

Lentivirus-mediated gene silencing was employed to generate stable knockdown cell lines. Short hairpin RNA (shRNA) sequences targeting METTL3 (#sh1: 5’-GCCTTAACATTGCCCACTGAT-3’, #sh2: 5’- GCAAGTATGTTCACTATGAAA-3’), CACNA1E (#sh1: 5’-GTCCTTTGAGTACACCATTAT-3’, #sh2: 5’-CGCAACAAAGTCCTGAGGTAT-3’), and IGF2BP2 (#sh1: 5’-CTTAACCAGTGCAGAAGTCAT-3’, #sh2: 5’-CAGTGCTGAGATAGAGATTAT-3’, #sh3: 5’-CGGATCTTTGGGAAACTGAAA-3’) were provided by Genechem (Shanghai, China). 293 T cells were used for lentivirus packaging. Stable knockdown cell lines were selected with 2 μg/mL puromycin. A non-specific shRNA (shNC) was adopted as a control.

Overexpression of CACNA1E (oe-CACNA1E) and WNT7B (oe-WNT7B) was achieved through cloning the full-length CACNA1E and WNT7B coding sequence into the pcDNA3.1 vector (Invitrogen, Carlsbad, CA, USA). Empty vector (EV) was adopted as a control. FLAG-tagged human CACNA1E plasmid was obtained from Genechem. The plasmid was transfected via Lipofectamine 3000 reagent (Invitrogen).

Antibodies

Antibodies used were as follows: antibodies of CACNA1E (24697–1-AP), GAPDH (10494–1-AP), IGF2BP1 (22803–1-AP), IGF2BP2 (24744–1-AP), IGF2BP3 (14642–1-AP), YTHDF1 (17479–1-AP), YTHDC1 (14392–1-AP), and FLAG (20543–1-AP) from Proteintech (Shanghai, China); antibodies of N-cadherin (ab76011), Vimentin (ab8069), E-cadherin (ab231303), METTL3 (ab195352), WNT7B (ab227607), and P-gp (ab129450) from Abcam (Cambridge, MA, USA); CAMK2A antibody (#11945) from CST (Danvers, MA, USA); antibodies of CAMK2N (SAB4300530) and PLCB2 (HPA041298) from Sigma-Aldrich (St. Louis, MO, USA); antibody of PLCB4 (PA5-100855) from ThermoFisher.

Methylated RNA immunoprecipitation sequencing (MeRIP-seq) and MeRIP-qPCR

Total RNA was extracted from tissues or cells by TRIzol reagent, which was then purified with Dynabeads™ mRNA Purification Kit (61006, Invitrogen). RNA fragmentation was achieved with RNA fragmentation reagent (AM8740, Invitrogen), with subsequent immunoprecipitation using anti-m⁶A antibody. The input and immunoprecipitation RNA samples were subject to NEBNext® Ultra™ RNA Library Prep Kit for sequencing library preparation and sequenced on Illumina HiSeq 2500 or analyzed via MeRIP-qPCR using Magna MeRIP m⁶A Kit (Millipore, Billerica, MA, USA).

RNA sequencing (RNA-seq)

Total RNA was extracted, followed by cDNA library preparation. The paired-end reads were produced on Illumina Novaseq 6000 and mapped to the human genome hg38 utilizing hisat2. Differentially expressed transcripts between groups were screened following |log2fold change|> 1 and P < 0.05. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was conducted using the differentially expressed transcripts.

Quantitative real-time PCR (qRT-PCR)

Total RNA was extracted with TRIzol reagent (Invitrogen), with subsequent cDNA synthesis utilizing PrimeScript RT kit (Takara, Dalian, China). After that, qRT-PCR analysis was conducted through SYBR Green SuperMix (Roche, Basel, Switzerland) along with ABI 7900HT Fast Real-Time PCR System (Applied Biosystems, CA, USA). GAPDH was employed as a control. The primer sequence information is presented in Supplementary Table 1.

Western blot

Cells were lysed in RIPA buffer (P0013B, Beyotime, Shanghai, China) containing protease inhibitor (P1005, Beyotime). The extracted proteins (20 μg) were separated via SDS-PAGE gels and transferred onto PVDF membranes (Millipore). Subsequently, the membranes were blocked with 5% nonfat milk and incubated with primary antibody at 4 °C overnight and secondary antibody at room temperature (RT) for 1 h. Protein bands were visualized by enhanced chemiluminescence kit (Amersham Biosciences, Uppsala, Sweden).

Cell viability assay

Cells were seeded into a 96-well plate (1,000 cells/well). 10 μL of Cell Counting Kit-8 (CCK-8) solution (C0005, TargetMol, Shanghai, China) was added to each well and incubated for 2 h at 37 °C following the manufacturer’s instructions. Absorbance at 450 nm was measured.

Colony formation assay

Cells were inoculated in a 6-well plate (200 cells/well). Following 14-day cultivation, the cell cultures were subject to 4% paraformaldehyde fixation for 30 min, followed by 0.1% crystal violet staining for 10 min. The stained clones were photographed and counted.

Wound healing assay

Cells were seeded into a 6-well plate and grown to 90% confluence. Utilizing a 200 µL pipette tip, a wound on the cell surface was created. Photographs of the wound were acquired at 0, 6, 12, and 24 h under a microscope (Olympus, Tokyo, Japan).

Transwell assay

Migration and invasion assays were carried out by Transwell chambers (Corning, NY, USA) coated without or with Matrigel. The upper compartment was filled with 2.5 × 10⁵ cells in 100 µL of serum-free medium, and the lower compartment was added with 600 µL of medium supplemented with 10% FBS. Following 24-h cultivation at 37 °C, invaded cells underside of the membrane were subjected to methanol fixation and crystal violet staining, followed by image acquisition using a microscope (Olympus).

Animal experiments

Male BALB/c nude mice aged 4–5 weeks from the Animal Experiment Center of Guangxi Medical University were assigned randomly into different groups (n = 6 per group). Tumor measurements and data analysis were performed in a blinded manner.

For experiment 1, shNC- or shCACNA1E-transfected MNNG and SJSA-1 cells (2 × 10⁶) were subcutaneously injected into nude mice. Starting one-week post-injection, tumor size was measured at 7-day intervals with the formula ½(length × width²). Four weeks after injection, the animals were euthanized. Xenografts were removed and tumor weight was monitored.

For experiment 2, in the model of lung metastasis, shNC- or shCACNA1E-transfected MNNG and SJSA-1 cells (1 × 10⁶) labeled with luciferase were intravenously infected into each mouse via the tail vein. Bioluminescence imaging was conducted to track metastatic progression. After six weeks, the mice were euthanized, followed by the extraction of their lungs for subsequent imaging and paraffin embedding. The quantity of metastatic nodules in the pulmonary region was measured.

For experiment 3, nude mice were subcutaneously injected by shNC- or shCACNA1E-transfected MNNG and SJSA-1 cells (2 × 10⁶) and randomly assigned into different groups once the tumor volume reached ~ 200 mm³ after ~ 7 days. The mice received intraperitoneal injection of 10% DMSO or 5 mg/kg MTX every three days. Tumor size was monitored at 7-day intervals over a period of 4 weeks. After 4 weeks of subcutaneous injection, the animals were euthanized. Xenografts were removed and tumor weight was monitored.

All the animal experiments were conducted following the ARRIVE guidelines and approved by the Animal Care and Use Committee of the First Affiliated Hospital of Guangxi Medical University (2023-S476-01).

H&E staining and immunohistochemistry (IHC)

Tumor tissues were fixed, embedded in paraffin, and cut into 5-µm-thick sections. Hematoxylin and eosin staining was performed. Then, the sections were deparaffinized and rehydrated, and the endogenous peroxidase activity was blocked with 3% H₂O₂, followed by antigen retrieval. After blocking with 5% normal goat serum at 20 °C for 10 min, the sections were incubated with primary antibody at 4 °C overnight and secondary antibody at RT for 1 h. The mean optical density was calculated utilizing the ImageJ software.

Fluorescencein situhybridization(FISH)

Fluorescein isothiocyanate (FITC)-labeled human CACNA1E mRNA probe was synthesized by RiboBio (Guangzhou, China). Following the FISH kit (RiboBio) manufacturer’s instructions, cells were fixed, permeabilized with 0.5% Triton X-100 for 5 min at 4 °C, and blocked with pre-hybridization solution for 30 min at 37 °C. Next, the cells were subject to fluorescently labeled probe at 37 °C overnight, anti-METTL3 antibody at 4 °C for 12 h, and secondary antibody (ab150079, Abcam) for 1 h at RT. After rinsing, the nucleus was stained with DAPI. Photographs were captured utilizing a fluorescence microscope (Olympus).

Dual luciferase assay

293 T cells were inoculated into a 24-well plate and grown to 70% confluence. The cells were transfected with luciferase reporter vectors fusing wild-type (WT) or mutant (Mut) CACNA1E 3’ UTR. Following 48-h incubation, the luciferase activity was monitored through dual luciferase reporter assay system (E1910, Promega, Madison, WI, USA).

RNA stability assay

Cells were cultured in a 6-well plate overnight and then exposed to 5 μg/mL actinomycin D (HY-17559, MCE) for 0, 4, 8, and 12 h. Total RNA was isolated and subject to qRT-PCR analysis.

Co-immunoprecipitation (Co-IP) assay

Protein A/G agarose beads pre-bound with antibody was added to the cell lysates and incubated with anti-FLAG antibody at 4 ℃ overnight, with human anti-IgG as a negative IP control. Next, the beads were rinsed and subjected to boiling in SDS loading buffer. The immunoprecipitated proteins were monitored via Western blot.

Intracellular Ca²⁺ detection

Fluo-8 AM, Ca²⁺ fluorescent probe was adopted to measure intracellular Ca²⁺. Cells were cultured in a 96-well plate overnight and subject to Fluo-8 AM in HHBS at 37 ºC for 1 h. After rinsing, photographs were acquired utilizing a fluorescence microscope (Olympus).

Cell apoptosis assay

For TUNEL staining, TUNEL Apoptosis Assay Kit (C1088, Beyotime) was adopted to monitor cell apoptosis as instructed by the manufacturer. Cells (1 × 10⁴) were seeded in a 96-well plate, fixed with 4% paraformaldehyde, and permeabilized with 0.1% Triton X-100. Next, they were subject to TUNEL reagent (50 μL/well) for 1 h at 37 ºC and then stained by DAPI (C1002, Beyotime). Image acquisition was performed under a fluorescence microscope (Olympus).

For flow cytometry, cells were seeded in a six-well plate and fixed with ice-cold ethanol. They were then stained with Annexin V-FITC and propidium iodide (PI). This analysis was conducted using FACScan flow cytometer (Becton Dickinson, Heidelberg, Germany).

RNA immunoprecipitation (RIP) assay

RIP assay was carried out using RNA Immunoprecipitation Kit (Beyotime Biotechnology, Shanghai, China) as instructed by the manufacturer. 1 × 10⁷ cells were first digested and then lysed to release RNA–protein complexes. The lysate was centrifuged to separate the supernatant, followed by incubation with magnetic beads pre-coated with METTL3, IGF2BP2, or rabbit IgG antibody (used as a negative control). After allowing sufficient time for the RNA–protein complexes to bind to the beads, several wash steps were performed to remove unbound materials. Finally, the RNA associated with the immunoprecipitated proteins was eluted and purified for subsequent qRT-PCR analysis.

Chromatin immunoprecipitation (ChIP) assay

ChIP assay was conducted with a ChIP Assay Kit (P2083S; Beyotime) following the manufacturer’s instructions. 1 × 10⁷ cells were crosslinked with 1% formaldehyde for 10 min at RT. Chromatin was prepared and then incubated with CACNA1E (24697–1-AP; Proteinthch) or rabbit IgG antibody at 4 °C overnight. After elution, the bound DNA fragments were analyzed by ChIP-PCR.

Statistical analysis

All the statistical analyses were conducted utilizing GraphPad Prism v9.0.0 (GraphPad Software, La Jolla, CA, USA) and R v3.4.1 (https://www.r-project.org/). The data are presented as means ± standard deviation (SD). The variations between two or more groups were evaluated utilizing Student’s t-test or one- or two-way analysis of variance. Kaplan–Meier approach was adopted to analyze the survival curves, with log-rank test for comparison. Pearson’s correlation coefficient was utilized to evaluate the correlation between variables. Statistical significance was defined as P < 0.05.

Results

Identification of CACNA1E as an m⁶A-hypermethylated and upregulated gene in OS

To identify potential m⁶A-modified genes contributing to OS pathogenesis, we conducted integrated MeRIP-seq and RNA-seq analyses on 3 pairs of OS and adjacent normal tissue specimens. Among differentially m⁶A methylated and expressed genes, CACNA1E was found to exhibit notable hypermethylation and upregulation in OS tissues than in adjacent normal tissues (Fig. 1A, B and Supplementary Table 2). CACNA1E exhibited the highest expression level in OS samples. This distinctive expression pattern, combined with its statistically significant association, supports its potential role in OS biology. We subsequently assessed CACNA1E expression in 6 pairs of OS clinical specimens and confirmed the upregulation of CACNA1E in OS (Fig. 1C). Furthermore, CACNA1E was consistently upregulated across multiple OS cells (SJSA-1, MNNG, U2OS, and SAOS2) compared to human osteoblast cells (hFOB1.19) (Fig. 1D). Using the TARGET dataset, we assessed the association between OS patient survival and the expression of the main top-scoring genes, revealing that increased levels of CACNA1E were significantly linked to worse clinical prognosis (Fig. 1E and Supplementary Fig. 1). Notably, multivariate Cox regression analysis established CACNA1E as an independent prognostic factor after adjusting for age, gender, and metastasis status (Hazard Ratio (HR) = 3.436 (1.143–10.334), P = 0.028) (Fig. 1F). To gain a deeper understanding of the role played by CACNA1E in OS, we established stable knockdown of CACNA1E in MNNG and SJSA-1 cells through lentiviral infection of shCACNA1E, with shNC as a negative control (Fig. 1G-I and Supplementary Fig. 2A). After that, we focused on the effects of CACNA1E on cell proliferation, motility, and invasiveness. Our results strongly indicated that the absence of CACNA1E led to a notable decrease in proliferation of OS cells (Fig. 1J-M). Additionally, cell motility (Fig. 1N, O), migration, and invasion (Fig. 1P-R) were remarkably attenuated by CACNA1E knockdown. It is widely accepted that the capacity of many tumor cells to migrate is facilitated by epithelial-mesenchymal transition (EMT) process that involves a reduction in the expression of epithelial cell markers (e.g., E-cadherin) and an increase in the expression of mesenchymal cell markers (e.g., Vimentin and N-cadherin) [19, 25]. Our analysis confirmed that CACNA1E knockdown reduced N-cadherin and Vimentin expression and elevated E-cadherin expression (Fig. 1S-U). These findings highlight the contribution of CACNA1E to OS development and progression.

Fig. 1.

Identification of CACNA1E as an m⁶A-hypermethylated and upregulated gene in OS. A Starplot illustrating the distribution of genes with differential expression (up/down; |log2fold change|≥ 1 and P < 0.05) and differential m⁶A (hyper/hypo; |log2fold change|≥ 1 and P < 0.05) in OS tissues with adjacent normal tissues. B Heatmap showing the top5 up-/down-regulated genes in OS tissues versus adjacent normal tissues. C CACNA1E mRNA level examined by qRT-PCR in OS and adjacent normal tissues. D CACNA1E mRNA level examined by qRT-PCR in osteoblast cells (hFOB1.19) and OS cells (SJSA-1, MNNG, U2OS, and SAOS2). E Kaplan–Meier survival curves of OS patients (n = 85) stratified by high and low CACNA1E expression, with log-rank test for comparing survival rate. F Multivariate Cox regression analysis for determining CACNA1E as an independent prognostic factor after adjusting for clinical variables (age, gender, and metastasis status) based on the TARGET database. G-I qRT-PCR and Western blot assays for verifying the stable knockdown of CACNA1E in MNNG and SJSA-1 cells. J, K CCK-8 and (L, M) colony formation assays for determining the effect of CACNA1E on OS cell proliferation. N, O Wound healing assay for examining the migration of shNC- and shCACNA1E-transfected OS cells. Photographs are displayed at 0, 6, 12, and 24 h. Scale bar, 100 μm. P-R Analysis of OS cell migration and invasion by Transwell assays. Scale bar, 100 μm. S-U Western blot for examining the effect of CACNA1E on EMT markers in OS cells. These data are presented as the means ± SD of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 derived from Student’s t-test or one- or two-way analysis of variance

Targeted inhibition of CACNA1E effectively hinders OS growth and lung metastasisin vivo

To further evaluate the influence of CACNA1E on OS progression, we constructed a xenograft subcutaneous tumor mouse model by subcutaneously injecting shNC- and shCACNA1E-transfected MNNG and SJSA-1 cells into nude mice. After four weeks, knockdown of CACNA1E effectively impeded tumor growth, as evidenced by the reduction in tumor volume and weight (Fig. 2A-G). In addition, knockdown of CACNA1E resulted in a reduction in the expression of CACNA1E, Ki-67, N-cadherin, and Vimentin and an elevation in the expression of E-cadherin in subcutaneous tumors (Fig. 2H-J). Furthermore, to determine the impact of CACNA1E on lung metastasis of OS, shNC- or shCACNA1E-transfected MNNG and SJSA-1 cells labeled with luciferase were intravenously injected into nude mice. After a period of six weeks, it was noted that CACNA1E knockdown caused a notable reduction in lung metastasis, with the reduced quantity and size of pulmonary metastatic nodules (Fig. 2K-N). Collectively, these findings demonstrate that targeted inhibition of CACNA1E effectively impedes tumor growth and distant dissemination of OS in vivo.

Fig. 2.

Targeted inhibition of CACNA1E effectively hinders OS growth and lung metastasis in vivo. A-D Photographs of subcutaneous tumor models and dissected tumors. The models were established by subcutaneous injection of shNC- and shCACNA1E-transfected (A, B) MNNG and (C, D) SJSA-1 cells (n = 6) into nude mice. E–G Measurement of tumor volume and weight. H Representative H&E-stained images and IHC staining of CACNA1E, Ki-67, and EMT markers in subcutaneous tumor sections. Scale bar, 100 μm. I, J Quantification of IHC staining intensity of CACNA1E, Ki-67, and EMT markers. K Representative bioluminescence images of nude mice that received intravenous injection of shNC- or shCACNA1E-transfected MNNG and SJSA-1 cells labeled with luciferase. L Quantification analysis of bioluminescent images of lung metastases. M Photographs of dissected lungs from mouse tumor metastasis models through injection of shNC- and shCACNA1E-transfected MNNG and SJSA-1 cells via the tail vein. N Representative H&E-stained images of lung sections for showing metastatic cancer nodules. Scale bar, 100 μm. **P < 0.01; ****P < 0.0001 derived from Student’s t-test or two-way analysis of variance

METTL3-mediated m⁶A modification stabilizes CACNA1E mRNA in OS

Given the high expression and significant m⁶A modification of CACNA1E in OS, we speculated that m⁶A modification might regulate the mRNA stability of CACNA1E. Analysis of m⁶A peak distribution revealed that most m⁶A peaks were located on exons 13 to 15 of CACNA1E mRNA (Fig. 3A). Notably, OS tissues showed the higher enrichment of m⁶A peaks than adjacent normal tissue specimens. Further investigation determined that the m⁶A motifs were predominantly enriched in the “RRACH” sequence (Fig. 3B). Utilizing the SRAMP database (https://www.cuilab.cn/sramp/), the identified m⁶A motifs within the CACNA1E sequence had significant confidence levels (Fig. 3C). MeRIP-qPCR analysis showed that m⁶A antibody remarkably enriched CACNA1E mRNA than IgG (Fig. 3D). Among the well-known m⁶A regulators, METTL3 was found to present a positive association with CACNA1E at the transcriptional level among TCGA OS samples (Fig. 3E), and FISH assay confirmed the co-localization of CACNA1E mRNA and METTL3 in the nucleus (Fig. 3F), indicating that METTL3 may mediate m⁶A modification of CACNA1E. After stable knockdown of METTL3 (Fig. 3G and Supplementary Fig. 2B), the m⁶A-mediated enrichment of CACNA1E mRNA was remarkably attenuated (Fig. 3H, I). To further validate the m⁶A modification of CACNA1E by METTL3, luciferase reporter vectors with wild-type and mutant m⁶A sites were generated (Fig. 3J). The dual luciferase assay revealed that overexpressing METTL3 enhanced the transcription activity of wild-type CACNA1E but not the mutant construct (Fig. 3K). Consistent with this, METTL3 knockdown substantially reduced the mRNA and protein levels of CACNA1E (Fig. 3L-P). We subsequently investigated the impact of METTL3 on CACNA1E mRNA stability in OS cells using actinomycin D chase assay. The results showed that METTL3 knockdown accelerated CACNA1E mRNA decay (Fig. 3Q, R).

Fig. 3.

METTL3-mediated m⁶A modification stabilizes CACNA1E mRNA in OS. A IGV plot visualizing m⁶A peaks within CACNA1E transcript in OS and corresponding normal tissues based on MeRIP-seq. B Prediction of m⁶A motifs within CACNA1E. C Analysis of significant confidence levels of putative m⁶A motifs within CACNA1E. D Relative enrichment of CACNA1E m⁶A level normalized to input in OS cells detected by MeRIP-qPCR. E Pearson correlation analysis between the mRNA levels of METTL3 and CACNA1E in TCGA OS samples. F FISH for detecting CACNA1E mRNA and METTL3 protein in OS cells. Scale bar, 50 μm. G qRT-PCR for measuring the stable knockdown of METTL3 in OS cells. H, I MeRIP-qPCR for detecting CACNA1E m⁶A level in control and METTL3-knockdown OS cells. J Schematic representation of the m⁶A motif (WT) in the 3’ UTR of CACNA1E from MeRIP-seq as well as the mutant m⁶A motif (Mut). K Relative luciferase activity in 293 T cells co-transfected with CACNA1E m⁶A WT or Mut luciferase reporter vector and oe-METTL3 or NC plasmid. L Measurement of CACNA1E mRNA level by qRT-PCR in control and METTL3-knockdown OS cells. M–O Detection of METTL3 and CACNA1E expression by Western blot in control and METTL3-knockdown OS cells. P FISH for detecting CACNA1E mRNA and METTL3 protein in shNC- and shMETTL3-transfected OS cells. Scale bar, 50 μm. Q, R Actinomycin D assay for analyzing the effect of METTL3 silence on the mRNA stability of CACNA1E. These data are presented as the means ± SD of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, P > 0.05 derived from Student’s t-test or one- or two-way analysis of variance

We next elucidated the role of METTL3 in OS progression. After METTL3 silence, the proliferative and colony-forming capacities of OS cells were notably suppressed (Fig. 4A-D). Likewise, the migration and invasion of OS cells were remarkably declined after silencing METTL3 (Fig. 4E-I). Thus, METTL3 may potentially impact the growth, infiltration, and movement of OS cells. In addition, to further validate CACNA1E as a downstream target of METTL3, we conducted functional rescue experiments in OS cells with METTL3 knockdown and CACNA1E overexpression. The results revealed that CACNA1E overexpression markedly enhanced OS cell proliferation, migration, and invasion, while METTL3 knockdown attenuated the effects induced by CACNA1E overexpression (Fig. 4J-R). The impairment of migration aligns with the mesenchymal-to-epithelial transition (MET)-like molecular shift characterized by upregulated E-cadherin and downregulated N-cadherin and Vimentin upon axis inhibition (Fig. 1S-U), functionally substantiating the role of the METTL3/CACNA1E axis in promoting a pro-metastatic state. These findings reveal that METTL3 enhances CACNA1E mRNA stability through m⁶A modification, thus contributing to OS progression.

Fig. 4.

METTL3 boosts OS progression by mediating m⁶A modification of CACNA1E. A, B Cell viability of shNC- and shMETTL3-transfected OS cells determined by CCK-8. C, D Colony-forming analysis of OS cells. E, F Migratory ability of OS cells evaluated by wound healing assay. Photographs are displayed at 0, 6, 12, and 24 h. Scale bar, 100 μm. G-I Migratory and invasive potential of OS cells detected by Transwell assays. Scale bar, 100 μm. J, K CCK-8 assay of METTL3-knockdown or/and CACNA1E-overexpressing OS cells. L, M Colony formation of METTL3-knockdown or/and CACNA1E-overexpressing OS cells. N, O Wound healing assay of METTL3-knockdown or/and CACNA1E-overexpressing OS cells. Photographs are displayed at 0, 6, 12, and 24 h. Scale bar, 100 μm. P-R Migration and invasion of OS cells measured by Transwell assays. Scale bar, 100 μm. These data are presented as the means ± SD of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 derived from one- or two-way analysis of variance

IGF2BP2 recognizes and enhances METTL3-mediated m⁶A modification of CACNA1E in OS

To identify the specific m⁶A “readers” that directly recognize METTL3-mediated m⁶A modification of CACNA1E, we conducted Co-IP assay using FLAG-tagged CACNA1E (FLAG-CACNA1E) (Supplementary Fig. 3). The results indicated that IGF2BP2, not IGF2BP1, IGF2BP3, YTHDF1, or YTHDC1, directly interacted with CACNA1E (Fig. 5A), indicating that IGF2BP2 may directly bind to m⁶A motifs of CACNA1E and boost mRNA stability. Consistent with its oncogenic role, high IGF2BP2 expression was correlated to poorer OS outcomes among patients with OS (Fig. 5B). Additionally, the RIP experiments revealed a significant enrichment of CACNA1E mRNA in the METTL3 or IGF2BP2 immunoprecipitated samples, supporting their direct interactions (Fig. 5C, D). To pinpoint the m⁶A site in CACNA1E recognized by IGF2BP2, we conducted dual luciferase assay with reporters carrying wild-type or mutant CACNA1E m⁶A motifs. The results revealed that IGF2BP2 overexpression markedly elevated the transcriptional level of wild-type CACNA1E, but not affecting the mutant construct (Fig. 5E). Functionally, IGF2BP2 knockdown markedly reduced the mRNA level of CACNA1E in OS cells (Fig. 5F, G and Supplementary Fig. 2C). Furthermore, after exposure to actinomycin D for indicated intervals, IGF2BP2 silence notably accelerated CACNA1E mRNA decay (Fig. 5H, I). Hence, the above findings demonstrated the role of IGF2BP2 in boosting METTL3-mediated CACNA1E m⁶A modification. In addition, silencing IGF2BP2 led to a prominent decline in proliferation, colony forming, migration, and invasion of OS cells (Fig. 5J-R), revealing that METTL3-mediated CACNA1E m⁶A modification is boosted by IGF2BP2, thus facilitating OS progression.

Fig. 5.

IGF2BP2 recognizes and enhances METTL3-mediated m⁶A modification of CACNA1E in OS. A Analysis of the interactions between FLAG-CACNA1E and m⁶A “readers” (especially IGF2BP1/2/3, YTHDF1, and YTHDC1) by Co-IP assay in 293 T cells. B Kaplan–Meier plots of overall survival in high and low IGF2BP2 expression patients with OS. C, D RIP experiment for validating the interactions between METTL3 and CACNA1E as well as between IGF2BP2 and CACNA1E. E Relative luciferase activity in 293 T cells co-transfected with CACNA1E m⁶A wild-type (WT) or mutant (Mut) luciferase reporter vector and oe-IGF2BP2 or NC plasmid. F, G IGF2BP2 and CACNA1E mRNA levels measured by qRT-PCR in shNC- and shIGF2BP2-transfected OS cells. H, I Detection of the mRNA stability of CACNA1E by qRT-PCR in shNC- and shIGF2BP2-transfected OS cells after exposure to actinomycin D for indicated intervals. J, K CCK-8, (L, M) colony formation experiment, (N, O) wound healing assay, and (P-R) Transwell migration and invasion assays of shNC- and shIGF2BP2-transfected OS cells. Scale bar, 100 μm. These data are presented as the means ± SD of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, P > 0.05 derived from Student’s t-test or one- or two-way analysis of variance

CACNA1E facilitates OS progression through modulating WNT7B expression

To uncover the downstream molecular mechanisms underlying CACNA1E-mediated OS progression, we performed RNA-seq analysis on control and CACNA1E knockdown OS cells, and identified 1129 differentially expressed transcripts in CACNA1E-knockdown than control OS cells (|log2fold change|> 1 and P < 0.05) (Fig. 6A). We randomly selected six transcripts (CACNA1A, NKD2, RYR1, VDAC1, MITF, and MMP1) for qRT-PCR analysis to verify their expression, and the results showed that they were all consistent with the RNA-seq data, suggesting the accuracy of RNA-seq (Supplementary Fig. 4A-F). The KEGG analysis demonstrated a notable enrichment of the differentially expressed transcripts in cancer-related pathways, e.g., pathways in cancer and calcium signaling pathway (Fig. 6B). Significantly, WNT7B was the most down-regulated gene associated with cancer-related pathways in CACNA1E-knockdown OS cells. Additionally, the analysis conducted using the GeneMANIA database (https://genemania.org/) indicated a close functional interaction between WNT7B and CACNA1E (Fig. 6C). Consistent with CACNA1E expression pattern, WNT7B also presented upregulation in OS versus normal tissues (Fig. 6D). Furthermore, the upregulation of WNT7B was also detected in OS cells (Fig. 6E). These findings implied that WNT7B may function as a downstream molecule of CACNA1E. Silencing CACNA1E was observed to markedly decline the mRNA and protein levels of WNT7B in OS cells (Fig. 6F-H). Similar to CACNA1E, silencing METTL3 also declined the expression of WNT7B (Fig. 6I). Next, to determine whether CACNA1E exerts a tumor-promoting effect by regulating WNT7B, we carried out functional rescue experiments in vitro through co-transfecting cells with shCACNA1E and WNT7B overexpression plasmid. WNT7B overexpression effectively reversed the suppressive effects of CACNA1E knockdown on OS cell proliferation, colony formation, migration, and invasion (Fig. 6J-R). Clinically, consistent with CACNA1E, high WNT7B expression was connected to worse overall survival and disease-free survival among patients with OS (Fig. 6S, T). Therefore, these data indicate that CACNA1E facilitates OS progression via modulating WNT7B expression.

Fig. 6.

CACNA1E facilitates OS progression through modulating WNT7B expression. A Volcano plots for differentially expressed transcripts between shCACNA1E- and shNC-transfected OS cells based on RNA-seq data. B KEGG enrichment plots for signaling pathways enriched by the differentially expressed transcripts. C Proteins interacting with WNT7B and CACNA1E using the GeneMANIA database. D CACNA1E and WNT7B mRNA levels in OS and normal tissues. E WNT7B mRNA level examined by qRT-PCR in osteoblast cells (hFOB1.19) and OS cells (SJSA-1, MNNG, U2OS, SAOS2). F–H WNT7B mRNA and protein levels detected by qRT-PCR and Western blot in shCACNA1E- and shNC-transfected OS cells. I WNT7B mRNA level examined by qRT-PCR in shMETTL3- and shNC-transfected OS cells. J, K Cell viability of METTL3-knockdown or/and WNT7B-overexpressing OS cells determined by CCK-8. L, M Colony-forming potential of OS cells. N, O Migratory ability of OS cells assessed by wound healing assay. Photographs are displayed at 0, 6, 12, and 24 h. Scale bar, 100 μm. P-R Migratory and invasive potential of OS cells measured by Transwell assays. Scale bar, 100 μm. S, T Kaplan–Meier plots of overall survival and disease-free survival in OS patients stratified by WNT7B expression. These data are presented as the means ± SD of three independent experiments. **P < 0.01; ***P < 0.001; ****P < 0.0001 derived from one- or two-way analysis of variance

CACNA1E affects the non-canonical Wnt/Ca²⁺ signaling pathway by transcriptionally regulating WNT7B

Given that CACNA1E encodes R-type voltage-gated calcium channels, we assessed the impact of CACNA1E on the Ca²⁺ signaling pathway. Using Fluo-8 AM calcium assay, we measured Ca²⁺ level in OS cells and found that silencing CACNA1E prominently declined Ca²⁺ level (Fig. 7A, B). Additionally, we detected the influence of CACNA1E on molecules in the calcium signaling pathway, including CAMK2A, CAMK2N, PLCB2, and PLCB4. It was observed that silencing CACNA1E resulted in a decline in the expression of CAMK2A, PLCB2, and PLCB4 as well as an increase in the expression of CAMK2N (Fig. 7C-G). Importantly, WNT7B overexpression rescued the inhibitory effect of CACNA1E knockdown on Ca²⁺ level in OS cells (Fig. 7H, I). Moreover, the decline in the expression of CAMK2A, PLCB2, and PLCB4 and the increase in the expression of CAMK2N induced by CACNA1E knockdown were rescued by WNT7B overexpression (Fig. 7J-N).

Fig. 7.

CACNA1E affects the WNT7B-mediated non-canonical Wnt/Ca²⁺ signaling pathway by transcriptionally regulating WNT7B. A, B Fluo-8 AM calcium assay for determining intracellular Ca²⁺ level in shNC- and shCACNA1E-transfected OS cells. Scale bar, 100 μm. C-G CAMK2A, CAMK2N, PLCB2, and PLCB4 mRNA and protein levels detected by qRT-PCR and Western blot in shNC- and shCACNA1E-transfected OS cells. H, I Analysis of intracellular Ca²⁺ level by Fluo-8 AM calcium assay in OS cells co-transfected with shCACNA1E and WNT7B overexpression plasmid. Scale bar, 100 μm. J-N CAMK2A, CAMK2N, PLCB2, and PLCB4 mRNA and protein levels assessed by qRT-PCR and Western blot in OS cells co-transfected with shCACNA1E and WNT7B overexpression plasmid. O Schematic diagram of the truncated WNT7B binding motifs. P Relative luciferase activity in 293 T cells transfected with the truncated WNT7B luciferase reporter vectors. Q ChIP-PCR analysis for verifying the binding of CACNA1E with WNT7B promoter. These data are presented as the means ± SD of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, P > 0.05 derived from one- or two-way analysis of variance

To investigate the mechanism by which CACNA1E regulates WNT7B expression, we employed a series of truncated promoter plasmids of WNT7B to identify the specific binding region. The dual luciferase assay revealed that CACNA1E overexpression enhanced the transcriptional activity of WNT7B promoter, but did not affect the transcriptional activity of other truncated regions (Fig. 7O, P). Furthermore, the ChIP-PCR assay demonstrated that CACNA1E specifically bound to the region between −2000 bp and −1500 bp upstream of the transcription start site of WNT7B (Fig. 7Q). These results provide the first direct evidence that CACNA1E affects the non-canonical Wnt/Ca²⁺ signaling pathway by modulating the transcriptional activity of WNT7B.

Targeted inhibition of CACNA1E improves MTX sensitivity and overcomes MTX resistance through WNT7B-mediated Ca²⁺ signaling pathway

Drug sensitivity analysis showed that CACNA1E was negatively connected to MTX sensitivity using the GDSC database (https://guolab.wchscu.cn/GSCA/#/document) (Supplementary Fig. 5A), indicating that CACNA1E may be in relation to MTX resistance. To verify this hypothesis, we established MTX-resistant OS cells (MNNG^MR and SJSA-1^MR). After exposure to distinct concentrations of MTX (0, 5, 10, 20, 40, and 60 μM), the IC50 value of MTX in both MTX-resistant and parental cells was measured by CCK-8 assay. The IC50 value was notably higher in MNNG^MR and SJSA-1^MR cells than that in their parental cells (Supplementary Fig. 5B-E). Also, the higher expression of P-glycoprotein (P-gp) was detected in MNNG^MR and SJSA-1^MR cells than that in their parental cells (Fig. 8A, B). These data confirmed that MTX-resistant OS cells were successfully generated. The expression of CACNA1E was notably upregulated in MTX-resistant OS cells than that in their parental cells (Fig. 8A, C). Silencing CACNA1E improved the sensitivity of OS cells to MTX (Fig. 8D, E) as well as enhanced MTX-induced OS cell apoptosis (Fig. 8F, G and Supplementary Fig. 6). In MTX-resistant OS cells, silencing CACNA1E enhanced the inhibitory effect of MTX on colony-forming potential (Fig. 8H, I) as well as declined the IC50 value of MTX (Fig. 8J-M). To quantify the drug combination synergy between CACNA1E knockdown and MTX, we used the Chou-Talalay method to calculate the combination index (CI). Our results showed that the CI value was < 1, indicating that CACNA1E knockdown combined with MTX treatment had a synergistic effect (Fig. 8N-Q). TUNEL staining analysis showed the effect of CACNA1E silence on enhancing the pro-apoptosis of MTX in MTX-resistant OS cells (Fig. 8R, S). Subsequently, we investigated whether CACNA1E knockdown-induced MTX resistance inhibition is in relation to WNT7B-mediated calcium signaling pathway. Intracellular Ca²⁺ level was declined in both CACNA1E-knockdown and MTX-treated MTX-resistant OS cells, and the combination of the two was more pronounced (Fig. 8T, U). In addition, the combination of CACNA1E silence and MTX had a more pronounced effect on declining the expression of WNT7B in MTX-resistant OS cells (Fig. 8V, W). Taken together, these results suggest that targeted inhibition of CACNA1E overcomes MTX resistance potentially through modulating WNT7B-mediated calcium signaling pathway.

Fig. 8.

Targeted inhibition of CACNA1E improves MTX sensitivity and overcomes MTX resistance through WNT7B-mediated calcium signaling pathway. A-C P-gp and CACNA1E protein levels measured by Western blot in parental and MTX-resistant OS cells. D, E Cell viability of shNC-/shCACNA1E-transfected and MTX-treated OS cells. F, G Apoptosis of shNC-/shCACNA1E-transfected and MTX-treated OS cells analyzed by flow cytometry. H, I Colony-forming potential of shNC-/shCACNA1E-transfected and MTX-treated MTX-resistant OS cells. J-M IC50 value of MTX in shNC-/shCACNA1E-transfected MTX-resistant OS cells. N-Q Representative Chou-Talalay plots and CI-Fa plots showing the synergistic effects between CACNA1E knockdown and MTX analyzed using the CompuSyn software. R, S Apoptosis of shNC-/shCACNA1E-transfected and MTX-treated MTX-resistant OS cells measured by TUNEL staining. Scale bar, 100 μm. T, U Intracellular Ca²⁺ level in shNC-/shCACNA1E-transfected and MTX-treated MTX-resistant OS cells detected by Fluo-8 AM calcium assay. Scale bar, 100 μm. V, W WNT7B protein level measured by Western blot in the above cells. These data are presented as the means ± SD of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 derived from Student’s t-test or one- or two-way analysis of variance

Targeted inhibition of CACNA1E and MTX synergistically prevent tumor growth in vivo

To further clarify the impact of CACNA1E on MTX responsiveness, we constructed a xenograft subcutaneous model through subcutaneous infection of shNC-/shCACNA1E-transfected MNNG and SJSA-1 cells into nude mice. One week after tumor cell transplantation, we intraperitoneally injected MTX into the mice every three days. After four weeks, tumor weight and size were remarkably declined by both CACNA1E silence and MTX intervention, and the combination of the two was more pronounced (Fig. 9A-J). IHC staining of Ki-67 also demonstrated that both CACNA1E silence and MTX intervention suppressed tumor cell proliferation, with a better effect when the combination of the two (Fig. 9K-M). Moreover, CACNA1E silence in combination with MTX intervention synergistically down-regulated the expression of CACNA1E and WNT7B in tumors (Fig. 9K-M). To further investigate the clinical relevance of CACNA1E, we performed IHC staining of CACNA1E and WNT7B on 4 MTX-sensitive and 4 MTX-resistant clinical OS tissue samples. The results showed the significant upregulation of CACNA1E and WNT7B expression in MTX-resistant OS tissues (Fig. 9N, O). In conclusion, these findings demonstrate that targeted inhibition of CACNA1E and MTX intervention synergistically prevent tumor growth in vivo.

Fig. 9.

Targeted inhibition of CACNA1E and MTX synergistically prevent tumor growth in vivo. A-D Photographs of subcutaneous tumors of nude mice after subcutaneous infection of shNC-/shCACNA1E-transfected (A, B) MNNG and (C, D) SJSA-1 cells and intraperitoneal administration of MTX (n = 6). E, F The size of subcutaneous tumors at different time points. G-J The size and weight of subcutaneous tumors. K Representative images of H&E-stained subcutaneous tumors and IHC images of Ki-67, CACAN1E, and WNT7B. Scale bar, 100 μm. L, M Quantification of Ki-67, CACAN1E, and WNT7B expression. (N, O) IHC assay for detecting the proteins levels of CACNA1E and WNT7B in MTX-sensitive and -resistant OS tissues. Scale bar, 100 μm. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, P > 0.05 derived from Student’s t-test or one- or two-way analysis of variance

Discussion

OS is the most prevalent primary bone malignancy [26]. Despite the improvement in the five‐year survival of patients with OS, prognostic outcomes of patients with metastatic or resistant disease remain unsatisfactory [10]. An in-depth understanding of mechanisms and possible treatment targets is clinically significant for preventing metastasis, overcoming chemoresistance, and improving prognostic outcomes. Our study bridges this gap by identifying a novel and specific regulatory mechanism: METTL3/IGF2BP2-induced m⁶A modification stabilizes CACNA1E mRNA, which activates WNT7B-mediated non-canonical Wnt/Ca²⁺ signaling pathway to promote OS progression and MTX resistance. This work moves beyond the established paradigm of generic METTL3-downstream target relationships by precisely defining a previously unreported link between m⁶A machinery and a specific calcium channel in OS.

In the present study, we discovered that CACNA1E, a specific R-type calcium channel, was highly expressed and m⁶A-hypermethylated in human OS tissues based on MeRIP-seq combined with RNA-seq. The role of CACNA1E in OS remains largely unexplored. Our functional experiments established CACNA1E as a critical driver of OS growth, metastasis, and MTX resistance. The clinical relevance of our findings was bolstered by multivariate Cox regression analysis of the OS cohort, which confirmed that high CACNA1E expression acted as an independent prognostic factor for poor survival. Mechanistically, we demonstrated that the m⁶A “writer” METTL3 installs m⁶A modifications on CACNA1E transcripts, and the m⁶A “reader” IGF2BP2 recognizes these marks to enhance CACNA1E mRNA stability (Fig. 10). This establishes METTL3 as the upstream regulator of Cav2.3 (encoded by CACNA1E), thereby providing the mechanistic basis for its control over the downstream CACNA1E/WNT7B/Ca²⁺ signaling axis functionally characterized in this study. Since both METTL3 and IGF2BP2 converge on the post-transcriptional stabilization of CACNA1E mRNA, their silencing may attenuate the Cav2.3-dependent calcium signaling, which we have directly demonstrated to be pivotal for the WNT7B-mediated oncogenic phenotypes. Consequently, a key consideration is how increased CACNA1E mRNA and protein translates into enhanced functional calcium channel activity. Recent research demonstrated that the loss of Cav2.3 (encoded by CACNA1E) phosphorylation (a state of functional down-regulation) leads to a gain-of-function alteration in the channel, characterized by significantly slower inactivation kinetics [13]. This establishes an unequivocal paradigm where molecular regulation of CACNA1E directly causes predictable changes in its ion channel function and downstream cellular excitability. Our study reveals a novel, upstream layer of regulation operating within this established framework: epitranscriptomic control of CACNA1E abundance via m⁶A. The logical extension, supported by this paradigm, is that increased CACNA1E caused by m⁶A-mediated mRNA stabilization leads to an increase in functional channels at the plasma membrane. This is consistent with the robust enhancement of intracellular calcium flux and calcium-dependent signaling we observed upon CACNA1E upregulation. M⁶A modification has emerged as a critical epigenetic regulator in various cancers [27–29]. Several recent studies have explored the intricate regulatory mechanisms of m⁶A modification in OS and other malignancies, shedding light on its potential as a therapeutic target [30, 31]. This METTL3/IGF2BP2-CACNA1E axis represents a sophisticated epitranscriptomic regulatory module that expands our understanding of post-transcriptional control in OS. Our findings resonate with emerging evidence highlighting the importance of RNA metabolism in cancer progression, as exemplified by a recent study on miRNA-regulated networks in lung adenocarcinoma that demonstrates how post-transcriptional control of cytoskeleton-associated oncogenes influences malignant progression and therapeutic resistance across solid tumors [32].

Fig. 10.

A graphic schematic of the mechanism by which CACNA1E m⁶A-modified by METTL3 in an IGF2BP2-dependent manner activates WNT7B-mediated non-canonical Wnt/Ca²⁺ signaling and thus promotes OS progression and MTX resistance

To elucidate the functional consequences of CACNA1E upregulation, we employed transcriptomic profiling and identified WNT7B as a critical downstream molecule of CACNA1E. Mechanistically, CACNA1E transcriptionally activates WNT7B, thus activating the non-canonical Wnt/Ca²⁺ signaling pathway mediated by WNT7B. Importantly, although the Wnt/Ca²⁺ signaling pathway is recognized as one of the noncanonical Wnt signaling pathways [33, 34], the specific role of WNT7B in this pathway remains unclear. Our study provides the first direct evidence that WNT7B participates in modulating the non-canonical Wnt/Ca²⁺ signaling pathway. We also demonstrated that CACNA1E affects intracellular Ca²⁺ levels and the expression of Ca²⁺ signaling pathway-related genes (CAMK2A, CAMK2N, PLCB2, and PLCB4) in OS cells through WNT7B. This newly established CACNA1E/WNT7B/Ca²⁺ cascade provides a detailed mechanistic link that was previously missing, moving our conclusions from correlation to causation. It is noteworthy that the oncogenic signaling driven by this axis specifically operates via the non-canonical Wnt/Ca²⁺ signaling pathway, which is distinct from the canonical β-catenin-dependent pathway [35]. Whether there is crosstalk between this Ca²⁺-dependent axis and the β-catenin pathway in OS represents an interesting avenue for future research. An important question arising from our data is the precise subcellular origin of the calcium signal modulated by Cav2.3. The observed global reduction in cytosolic calcium upon CACNA1E knockdown could, in principle, stem from either diminished influx through plasma membrane channels or altered release from intracellular stores such as the endoplasmic reticulum. Given the well-established role of Cav2.3 as a voltage-gated calcium channel primarily mediating extracellular calcium entry, the most parsimonious interpretation is that its upregulation directly enhances constitutive and stimulated calcium influx [13, 36]. This primary influx likely serves as a critical trigger and sustained source for calcium-dependent processes, potentially including the amplification of signals via calcium-induced calcium release (CICR) from intracellular stores [37]. Therefore, dampening this initial trigger would account for the comprehensive attenuation of downstream calcium signaling we observed. Future studies employing compartment-specific calcium indicators or pharmacological tools to dissect store-operated calcium entry will be valuable to precisely delineate the spatial dynamics and contributions of different calcium pools within this oncogenic signaling axis. WNT7B, a ligand for Wnt signaling, is known to contribute to cancer progression [38–40]. Our data confirmed the high expression of WNT7B in OS and its importance for mediating the pro-tumorigenic effects of CACNA1E. Our in vivo findings, demonstrating that CACNA1E knockdown effectively suppressed subcutaneous tumor growth and experimental lung metastasis, provide crucial functional validation of its oncogenic role. The selection of these widely adopted models follows an established research paradigm in oncology for initial target validation, as they offer a controlled and reproducible system to definitively assess the cell-autonomous tumor-promoting and metastatic colonization capacity of a novel target [41–44]. The robust effects observed under these conditions firmly establish the biological relevance of the METTL3/CACNA1E/WNT7B axis in driving core malignant phenotypes. While the current models clarify fundamental questions about the axis’s intrinsic functionality, the bone microenvironment plays an indispensable role in OS pathogenesis. Therefore, future studies employing orthotopic/intraosseous models will be critical to elucidate how this axis interacts with the complex bone stroma to influence disease progression and therapy response.

A paramount clinical challenge in OS is the development of resistance to first-line chemotherapeutics like MTX [45–47]. Our study revealed that targeted suppression of CACNA1E improved the sensitivity of OS cells to MTX and overcame MTX resistance, which was in relation to the WNT7B-mediated Ca²⁺ signaling pathway. Importantly, formal synergy quantification using the Chou-Talalay method confirmed a strong synergistic effect (CI < 1) between CACNA1E inhibition and MTX treatment. Therefore, we proposed a novel combination treatment strategy of targeted inhibition of CACNA1E with MTX to improve treatment efficacy and overcome MTX resistance for OS. Currently, molecular targeted therapy has become a key part of comprehensive cancer therapy, and has demonstrated excellent treatment efficacy in clinical practice. For patients with OS, precise targeted therapy combined with chemotherapy represents a potential direction of future research. Our data uncovered that targeted inhibition of CACNA1E may improve the responsiveness of OS cells to MTX and combat MTX resistance and emphasized the potential of targeted inhibition of CACNA1E in combination with MTX as an alternative therapeutic strategy for MTX-resistant OS patients. The precise mechanism by which MTX interacts with the CACNA1E/WNT7B/Ca²⁺ axis warrants further investigation. While our study establishes this axis as a critical mediator of MTX resistance, it is plausible that MTX may also exert its effects through additional pathways, such as influencing mitochondrial membrane conductivity and calcium homeostasis as suggested in the previous literature [48]. Exploring whether and how the CACNA1E-dependent calcium signal integrates with or modulates mitochondrial function to affect chemosensitivity represents a compelling and important direction for future mechanistic studies, which could provide an even deeper understanding of the therapeutic synergy observed in this study. To realize the clinical translation of our findings, a larger clinical sample size and prospective clinical trials are needed to verify the study results.

Conclusion

In summary, our study emphasized the biological importance of CACNA1E in OS progression and MTX resistance and unveiled a novel and clinically relevant signaling pathway in OS. CACNA1E mRNA stability and translation were enhanced by METTL3-mediated m⁶A methylation in an IGF2BP2-depenent manner, which subsequently activated the non-canonical Wnt/Ca²⁺ signaling pathway mediated by WNT7B, eventually contributing to OS growth, metastasis, and MTX resistance. Collectively, CACNA1E shows promise as a therapeutic target for overcoming MTX resistance. Targeted suppression of CACNA1E combined with MTX may be a promising strategy for the treatment of OS.

Abbreviations

OS

Osteosarcoma

MTX

Methotrexate

m⁶A

N⁶-methyladenosine

shRNA

Short hairpin RNA

NC

Negative control

MeRIP-seq

Methylated RNA immunoprecipitation sequencing

RNA-seq

RNA sequencing

KEGG

Kyoto Encyclopedia of Genes and Genomes

qRT-PCR

Quantitative Real-time PCR

RT

Room temperature

CCK-8

Cell Counting Kit-8

IHC

Immunohistochemistry

FISH

Fluorescence in situ hybridization

FITC

Fluorescein isothiocyanate

WT

Wild-type

Mut

Mutant

Co-IP

Co-immunoprecipitation

PI

Propidium iodide

RIP

RNA immunoprecipitation

SD

Standard deviation

3’ UTR

3’ Untranslated region

Authors’ contributions

Li Zheng, Qingjun Wei and Jinmin Zhao are responsible for the design of this study; Chaotao Chen, Kai Xiong, Feiyuan Liang and Yanping Zhong are responsible for the writing of this manuscript; Chaotao Chen, Kai Xiong, Feiyuan Liang, Yanping Zhong, Xiong Qin and Nanchang Huang are responsible for the data analysis; Xiong Qin, Nanchang Huang, Yuqi Fang, Bo Zhu, Jianwen Cheng and Qingjun Wei are responsible for the experimental testing; Bo Zhu, Jianwen Cheng and Qingjun Wei are responsible for sample collection; Li Zheng support the funding for this study. All authors reviewed and approved the manuscript.

Funding

This work was funded by the Guangxi Natural Science Foundation (2023GXNSFAA026002), the Guangxi Scientific Research and Technological Development Foundation (GuikeAB21220062), the National Natural Science Foundation of China (52573316, 82360426).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Each patient provided written informed consent for the collection of clinical samples, and the study protocols were approved by the Ethics Committee of Guangxi Medical University Cancer Hospital (2024-D100-01). The animal experiments were approved by the Animal Ethics Committee of the First Affiliated Hospital of Guangxi Medical University (2023-S476-01).

Consent for publication

All patients provided written informed consent.

Competing interests

The authors declare no competing interests.