Asporin regulates glycolysis and stemness of gastric cancer cells

Abstract

Asporin, a small leucine-rich proteoglycan encoded by the ASPN gene, is a vital player in cancer biology. Its oncogenic role in gastric cancer (GC) has been well-documented, particularly in promoting proliferation, apoptosis resistance, and migration/invasion. Nevertheless, ASPN’s function in regulating glycolysis and stemness in GC cells remains underexplored. Here, we initially analyzed published transcriptomic datasets, confirming overexpression of ASPN in GC tissues compared with noncancerous stomach tissues. Furthermore, results from bioinformatics demonstrated that differentially expressed genes (DEGs) linked to ASPN expression regulate epithelial-mesenchymal transition (EMT) and stemness-related pathways. Notably, ASPN expression predicted poor patient survival. Using ASPN-deficient HGC27 and GCIY GC cell lines, which express high ASPN levels, we conducted comprehensive cell biology assays. ASPN knockdown significantly impaired GC cell viability, reduced their migratory and invasive capacities, and increased apoptosis. We also observed that ASPN deficiency disrupted glycolysis and diminished the stemness of GC cells, reflected in reduced colony and sphere formation capacity and lower expression of stemness markers. Functionally, HIF1α overexpression partially rescued the deficits from ASPN loss, positioning HIF1α as a key downstream effector. To validate our in vitro findings, we employed a xenograft GC mouse model. Consistent with our in vitro data, ASPN-deficient GC cells displayed reduced tumor growth and stemness in vivo. In conclusion, our data suggest that ASPN is oncogenic in GC, enhancing proliferation, migration, invasion, glycolysis, and stemness while inhibiting apoptosis. These findings define a crucial oncogenic function of ASPN and underscore the therapeutic potential of targeting the ASPN-HIF1α axis in GC.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10735-026-10732-3.

Introduction

Gastric cancer (GC) is a major global health challenge, ranking as the second leading cause of cancer-related mortality, just behind lung carcinoma [1]. More than one million GC cases were diagnosed in 2020, causing 770,000 deaths [2, 3]. Risk factors, including Helicobacter pylori infection, exposure to industrial chemicals, smoking, alcohol consumption, and more, have been linked to GC development [4]. Unfortunately, often diagnosed at advanced stages, GC patients had a poor prognosis [5], despite advances in novel therapies such as targeted therapy and immune therapy in addition to conventional surgical resection and systemic chemotherapy [6]. Therefore, identifying novel druggable targets and predictive biomarkers is critical for the development of novel treatments for this fatal disease.

Asporin, a small leucine-rich proteoglycan encoded by the ASPN gene, is primarily found in cancer-associated fibroblasts (CAFs) and has been implicated in various cancers [7]. In GC, ASPN fulfills its oncogenic function by facilitating cell division and motility, as well as suppressing apoptotic cell death through multiple mechanisms [8, 9]. Moreover, ASPN regulates the reprogramming of GC cells, conferring resistance to oxidative stress [10]. However, the precise mechanisms underlying ASPN’s oncogenic role in GC, particularly in regulating glycolysis and stemness in GC cells, are not fully understood.

Glycolysis is a critical metabolic pathway used by cells to metabolize glucose into energy in the form of adenosine triphosphate (ATP) and to produce intermediates for other metabolic pathways [11]. Altered glycolysis, often referred to as the Warburg effect, is commonly observed in many cancers, including GC [12]. GC cells frequently promote their survival and metastasis by metabolic reprogramming [13]. Although ASPN has been shown to affect glycolysis in other cancers [14, 15], its role in regulating glycolysis in GC cells remains unclear.

The stemness of cancer cells is a crucial factor in tumorigenesis, contributing to recurrence, therapy resistance, and metastasis [16]. Cancer stemness is governed by multiple signaling cascades, such as the WNT and TGF-β pathways [17]. ASPN deficiency in mice has been shown to cause decreased cancer stem cells (CSCs) [18]. Yet, ASPN’s function in maintaining GC cell stemness remains elusive and requires thorough investigation.

Herein, we first assessed ASPN expression by analyzing two publicly available GC transcriptomic datasets and IHC. We then examined ASPN levels in several GC cell lines, followed by knocking down ASPN in two different lines of GC cells with high ASPN expression. We investigated the impact of ASPN deficiency on viability, cell cycle, and motility of these GC cells. Furthermore, we explored how ASPN influences the glycolysis and stemness of GC cells.

Materials and methods

Bioinformatics

The RNA-sequencing dataset TCGA-STAD was retrieved from The Cancer Genome Atlas (https://portal.gdc.cancer.gov/), while the GSE27342 and GSE63089 datasets were obtained from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). Immunohistochemistry (IHC) data for ASPN were sourced from the Human Protein Atlas database (HPA, https://www.proteinatlas.org/). For enrichment analyses of Gene Ontology (GO) and KEGG pathways, samples from the TCGA-STAD dataset were categorized into ASPN-high and ASPN-low groups based on median ASPN mRNA levels. Differentially expressed genes (DEGs) between the two populations were screened and visualized with the R package clusterProfiler (v4.14.4).

Cell culture

Human GC cell lines (MKN-45, MGC803, MKN-7, NCI-N87, GCIY, and HGC-27), along with gastric epithelial cells (GES-1), were obtained from Xiamen Immocell Biotechnology. GCIY cells were maintained in MEM containing 15% FBS (Gibco, USA). HGC-27 and GES-1 cells were maintained in DMEM with 10% FBS. MKN-7, NCI-N87, and MGC803 cells were cultured in RPMI-1640 medium containing 10% FBS, while MKN-45 cells were cultured in RPMI-1640 with 20% FBS. All cells were placed in a cell culture incubator at 37 °C filled with 5% CO₂.

Plasmid construction

The coding sequence of HIF1α was cloned into the pLV3-CMV-mcs-Puro lentiviral vector to generate the HIF1α overexpression plasmid (designated HIF1α OE). Similarly, specific short hairpin RNAs (shRNAs) targeting ASPN were constructed into the pLKO.1 vector, yielding three knockdown constructs named shASPN-1, shASPN-2, and shASPN-3, respectively. A lentiviral vector containing a non-specific shRNA (shNC) was also generated to serve as a control. The accuracy of all plasmid constructs was confirmed through sequencing. Lentiviral particles were packaged by Anti-hela Biotechnology (China). The viral titer was determined using a kit from Takara (631476, Japan). TaqMan real-time PCR was used to measure the virus multiplicity of infection (MOI). The primers to construct plasmids are present in Table 1.

Table 1.

Primer name and sequence plasmid construction

Plasmid	Sequence (5’ to 3’)
shASPN-1	CCGGGGAGTATGTGCTCCTATTATTCTCGAGAATAATAGGAGCACATACTCCTTTTT
	AATTAAAAAGGAGTATGTGCTCCTATTATTCTCGAGAATAATAGGAGCACATACTCC
shASPN-2	CCGGCAATCAACTAAGTGAAATACCCTCGAGGGTATTTCACTTAGTTGATTGTTTTT
	AATTAAAAACAATCAACTAAGTGAAATACCCTCGAGGGTATTTCACTTAGTTGATTG
shASPN-3	CCGGGGAATGTAATAATTAGTAATTCTCGAGAATTACTAATTATTACATTCCTTTTT
	AATTAAAAAGGAATGTAATAATTAGTAATTCTCGAGAATTACTAATTATTACATTCC
shNC	CCGGTTCTCCGAACGTGTCACGTTTCTCGAGAAACGTGACACGTTCGGAGAATTTTT
	AATTAAAAATTCTCCGAACGTGTCACGTTTCTCGAGAAACGTGACACGTTCGGAGAA
HIF1α	TAGAGCTAGCGAATTCGCCACCATGGAGGGCGCCGGCGGCGCGAACGAC
	TCTTTGTAGTCGGATCCGTTAACTTGATCCAAAGCTC

Stable cell construction

HGC27 and GCIY cells were infected with lentivirus at an MOI of 15. After 24 h of culture, the medium was refreshed by fresh prepared medium containing 8 µg/mL polybrene (40804ES76, YEASEN, China) to enhance infection efficiency. For stable cell line selection, puromycin (2 µg/mL, 60210ES25, YEASEN) was applied to the culture medium 72 h post-infection. The medium was refreshed every other day during the selection process. After one week of puromycin selection, the surviving cells were maintained in regular medium and cultured until subsequent analyses.

Western blot

Cells and tumor tissues were treated using RIPA solution (20–188, Sigma-Aldrich, USA) containing the PMSF (329-98-6, Sigma-Aldrich) and subjected to intermittent sonication on ice. The samples underwent centrifugation to collect the supernatant. Protein levels were quantified using the Pierce™ BCA kit (23225, Thermo Fisher Scientific, USA). The samples were then denatured, separated by SDS-PAGE, transferred onto PVDF membranes (Sigma-Aldrich), and blocked with 5% non-fat milk in TBST at room temperature (RT) for 0.5 h. The membranes were further placed in primary antibody solutions overnight at 4 °C. The next day, following TBST washing, the membranes were placed in secondary antibody solutions for 1 h at RT. Finally, the membranes were washed again, and the signals were visualized using ECL substrate (32106, Thermo Fisher Scientific) and X-ray film exposure. Signal grayscale values were analyzed by ImageJ software (NIH, USA). The Western blot antibody details are listed in Table 2.

Table 2.

Antibody information for Western blot

Protein	Species	Vendor	Catalog number	Diluting ratio
ASPN	Rabbit	Proteintech (China)	27602-1-AP	1:1000
GLUT1	Rabbit	Proteintech	21829-1-AP	1:1000
GLUT3	Rabbit	Proteintech	20403-1-AP	1:1000
GLUT4	Rabbit	Abcam (UK)	ab33780	1:1000
SCO2	Rabbit	Proteintech	21223-1-AP	1:1000
NANOG	Rabbit	Proteintech	14295-1-AP	1:1000
MYC	Rabbit	Proteintech	10828-1-AP	1:1000
CD44	Rabbit	Proteintech	15675-1-AP	1:1000
HIF1α	Rabbit	Proteintech	20960-1-AP	1:2000
Actin	Rabbit	Proteintech	20536-1-AP	1:1000
HRP-Anti-Rabbit IgG	Goat	Proteintech	SA00001-2	1:10000

Quantitative polymerase chain reaction (qPCR)

The TRIzol reagent (Invitrogen, USA) was used to extract total RNA. cDNA was synthesized with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, USA). qPCR was performed on the QuantStudio™ qPCR System (Thermo Fisher Scientific, USA), using SYBR^® Green (Roche, Switzerland). The relative expression of target genes was measured based on the 2^−ΔΔCt method, with 18 S ribosomal RNA serving as the housekeeping gene. All qPCR assays involved biological and technical triplicates. Primers are listed in Table 3.

Table 3.

qPCR primer sequences

Primer ID	Sequence 5’-3’
ASPN-F	CTCTGCCAAACCCTTCTTTAGC
ASPN-R	CGTGAATAGCACTGACATCCAA
GLUT1-F	TCTGGCATCAACGCTGTCTTC
GLUT1-R	CGATACCGGAGCCAATGGT
GLUT3-F	GCTGGGCATCGTTGTTGGA
GLUT3-R	GCACTTTGTAGGATAGCAGGAAG
GLUT4-F	GCCATGAGCTACGTCTCCATT
GLUT4-R	GGCCACGATGAACCAAGGAA
SCO2-F	CTCCACCAAACAGGTTGCC
SCO2-R	CTGCTCAGCCGATCTGCTC
18 S-F	CGACGACCCATTCGAACGTCT
18 S-R	CTCTCCGGAATCGAA CCCTGA

Dual-luciferase assay

Luciferase activity was measured using the Dual Luciferase Reporter Assay Kit (Vazyme, DL101-01, China). HGC-27 GC cells were co-transfected with a HIF1α-responsive firefly luciferase reporter plasmid, a Renilla luciferase control plasmid, and the indicated shASPN plasmids. Twenty-four hours post-transfection, cells were divided and cultured under either normoxia (21% O₂) or hypoxia (1% O₂) for an additional 24 h. Thus, the total transfection time was 48 h. Cells were then lysed following the kit’s protocol. Luciferase activity was measured sequentially: firefly luciferase activity was recorded immediately after adding Luciferase Assay Reagent II and mixing, followed by measurement of Renilla luciferase activity after adding Stop & Glo Reagent and mixing. Firefly luciferase signals were normalized to the corresponding Renilla luciferase values, and data are presented as relative luciferase activity.

Cell viability measurement

MTT assay was used to determine cell viability. 1 × 10⁴ cells were plated into each well of 96-well plates and cultured for 24 h. Next, MTT reagent (20 µL, 5 mg/mL, ST316, Beyotime, China) was applied to each well and cultured for 4 hours. The solutions were then discarded, and 0.15 mL DMSO was added to dissolve the formazan crystals. Finally, the optical density (OD) at 490 nm was recorded using a microplate reader (A51119500C, Thermo Fisher Scientific).

Flow cytometry assay

Cell cycle status was assayed using a 7-AAD assay kit (ORCCY014A, ORiCells Biotechnology, China). Briefly, approximately 5 × 10⁵ cells were fixed in ethanol at -20 °C for 1 h and treated with RNase A at 37 °C for 0.5 h. The fixed cells were then filtered, pelleted, and resuspended in PBS, followed by incubation with the 7-AAD working solution prepared from the kit. Single-cell suspensions were maintained at 37 °C for 0.5 h and processed using a 1300 NovoCyte FACS cytometer with NovoExpress Software (Agilent, USA).

Apoptotic cell death was assessed using a kit from Solarbio (CA1020, China) as per the manufacturer’s guidelines. In brief, 1.5 × 10⁵ cells were digested, pelleted, and resuspended in 100 µL of Annexin V-FITC binding buffer. Subsequently, 10 µL of Annexin V-FITC/PI solution (1:1 mixture) was pipetted to the cell suspension and mixed gently. After incubation in the dark for 10 min, the samples were processed by flow cytometry.

To assess CD44 expression, cells were fixed, permeabilized, washed, and incubated with a CD44 antibody (397517, BioLegend, USA, 1:1000) at RT for 0.5 h. Finally, the cells were rinsed with PBS, resuspended in 2% BSA, and processed by flow cytometry.

Migration and invasion assay

3 × 10⁴ cells in 0.2 mL of serum-free culture medium were inoculated into the upper compartment of each Transwell chamber with an 8 μm pore size (Corning, USA). 0.7 mL of complete culture medium was added to the lower chamber. After 2 days of culture, the upper chamber membranes were carefully removed, rinsed with PBS, and fixed with methanol for 0.5 h at RT. The membranes were then stained with crystal violet (0.1%, Merck, USA) for 20 min, washed with PBS, dried, and imaged. ImageJ was used to quantify the results. The invasion assay followed the identical procedure as the migration assay, with the modification of using pre-coated upper chambers, which were treated with a 1:5 diluted Matrigel (356235, BD Sciences, USA) at 37 °C for 12 h.

Colony formation assay

Cells (500 cells per well) were plated into 6-well plates and cultured for two weeks. After incubation, the cell culture was rinsed with PBS, treated with 4% paraformaldehyde for 15 min, and then incubated with crystal violet solution. The cells were washed again with PBS, and colony formation was observed and imaged using a Motic microscope. Colony numbers were quantified using ImageJ software.

Sphere formation assay

A 100 µL cell suspension (3 × 10⁶ cells) was mixed with 100 µL of Matrigel and seeded into 96-well plates, which were then placed in a cell culture incubator until spheres formed. Once spheres had developed, they were washed several times with PBS. The spheres were cultured in culture medium containing B27 supplement (Gibco, 17504044, USA), basic fibroblast growth factor (bFGF, 20 ng/mL, Novoprotein, China), epidermal growth factor (EGF, 20 ng/mL, Novoprotein), and insulin (4 µg/mL, Gibco). After two weeks of culture, the spheres were imaged using a Motic microscope.

Measurement of glucose consumption, lactic acid (LA), ATP, Glucose-6-phosphate dehydrogenase (G6PD), and pH levels

The levels of glucose consumption, LA, ATP, G6PD, and pH were tested using the glucose uptake colorimetric assay kit (MAK083-1KT, Merck), L-Lactic Acid (LA) kit (E-BC-K044-M, Elabscience, USA), G6PD kit (BB-4720, BestBio, China), ATP Assay Kit (S0026B, Beyotime), and a pH meter (PHSJ-6 L, INESA, China), respectively, following the manufacturer’s protocols.

Extracellular acidification rate (ECAR) and O₂ consumption rate (OCR) measurement

The ECAR and OCR were determined using the XFe96 Extracellular Flux Analyzer (Agilent). 5 × 10⁴ cells were inoculated into each well of Seahorse XF96 V3 PS cell culture microplates (101085-004, Agilent) and cultured overnight. The following day, cells were maintained in sugar-free medium, and baseline ECAR and OCR values were recorded. Next, the cells were sequentially treated with Oligomycin (495455, Merck), FCCP (370-86-5, Merck), Rotenone (PHL22642, Merck), and Antimycin A (77332, Merck). ECAR and OCR were measured after each treatment. For glycolysis stress assays, cells were kept in medium supplemented glucose, 2-DG, Oligomycin, and FCCP. The parameters were analyzed using the accessory Seahorse software.

Xenograft assay

All animal experiments gained approval from the Ethics Committee of Jinjiang Municipal Hospital (Shanghai Sixth People’s Hospital Fujian) IACUC (Approval No. 2025-0072). BALB/c nude mice of six weeks old were sourced from the Animal Center of Xiamen University (China). After acclimatization, the mice were randomly categorized into two groups, with 10 mice in each group. Each mouse was subcutaneously injected into the right flank with 2 × 10⁶ ASPN-deficient or control stable cells. The dimensions of the xenograft tumors were measured at specified time points, and tumor volume was calculated as follows: (length×width²)/2. 36 days post-injection, the mice were euthanized, and the tumors were then dissected, weighed, and imaged. Half of the tumors were fixed in 4% paraformaldehyde for histological analysis, while the remaining tumors were subjected to Western blotting.

IHC and TUNEL

Xenograft tumor tissues were processed for paraffin sectioning into 5 μm slices. After deparaffinization and dehydration, the slices were treated with 3% H₂O₂ for 15 min to minimize endogenous peroxidase activity. Following antigen retrieval, the slices were treated with 5% goat serum and incubated with Ki67 antibody solution (Ab15580, Abcam, 1:500) overnight at 4 °C. Afterward, the slices were washed with PBST and treated with secondary antibody solution (B900210, Proteintech, 1:500) for 1 h at RT. Following washing with PBST, the slices were stained with DAB solution and counterstained with Meyer’s hematoxylin, followed by dehydration and mounting. The results were observed and photographed using a Motic AE2000 microscope equipped with a digital camera. Apoptosis in tissue slices was evaluated using the TUNEL assay with a TUNEL kit (E-CK-A320, Elabscience) according to the manufacturer’s instructions. The % cells positive for Ki-67 or TUNEL was measured using Image J.

Statistical analysis

Statistical analysis and graph generation were done by GraphPad Software (v8.0, Prism, USA). Results are presented as mean ± SD unless otherwise specified. The bar-dot plots contain data points representing the values of a biological replicate and/or the mean of three technical replicates. Unpaired bilateral Student’s t-test and one-way ANOVA followed by Tukey’s HSD were employed to assess differences between two and multiple datasets, respectively. A P value < 0.05 was considered statistically significant.

Results

ASPN is overexpressed in GC tissues and links to genes regulating signaling pathways related to epithelial-mesenchymal transition (EMT) and cancer cell stemness

Previous studies have demonstrated elevated ASPN expression in GC tissues [9, 19]. To corroborate these findings, we analyzed ASPN expression in GC and non-cancerous stomach tissues using two additional cohorts, GSE27342 and GSE63089. Our analysis revealed a marked increase in ASPN expression in GC tissues compared to normal stomach tissues (Fig. S1A). These results were in line with IHC data downloaded from the HPA platform, which also showed increased ASPN expression in GC tissues (Fig. S1B). Furthermore, our correlation analysis identified strong positive associations between ASPN expression and both GLUT3 (cor = 0.37, p < 0.0001) and GLUT4 (cor = 0.53, p < 0.0001) (Fig. S4A–B). Finally, by correlating ASPN expression with patient survival using the public TCGA dataset, we observed that patients with high ASPN expression had significantly worse overall survival (OS), disease-specific survival (DSS), disease-free interval (DFI), and progression-free interval (PFI) than those with low ASPN expression (Fig. S4C).

To identify genes potentially regulated by ASPN, we conducted functional enrichment analyses using the TCGA-STAD cohort. GO analyses revealed that DEGs associated with ASPN expression regulate EMT-related biological processes (BPs), such as collagen fibril organization, cell adhesion, and vasculogenesis. These DEGs also perform molecular functions (MFs) such as extracellular matrix binding and growth factor binding and are categorized into cellular components (CCs) like collagen trimers and actin filament bundles (Fig. S2A–C). Further KEGG analysis indicated that these DEGs modulate critical signaling pathways implicated in cancer cell stemness, including the Hh, WNT, and TGF-β pathways (Fig. S2D). These findings validate ASPN’s overexpression in GC and support its role as an oncogene.

ASPN deficiency impairs the malignancy of GC cells

To study the function of ASPN in GC cells, we began by examining its expression across multiple lines of GC cells and GES-1 cells. qPCR showed that ASPN expression was significantly increased in GCIY and HGC27 cells compared with GES-1 cells, while its mRNA was decreased in MKN45, MKN7, and NCI-N87 cells. No significant differential expression of ASPN was detected in MGC803 cells (Fig. S3A). Western blotting confirmed ASPN overexpression in GCIY and HGC27 cells, while reduced ASPN levels were observed in MKN45 cells. However, the decreases in MKN7 and NCI-N87 cells were not statistically significant (Fig. S3B, C). Thus, we chose GCIY and HGC27 cells for subsequent knockdown experiments.

Three shRNAs were designed to deplete ASPN expression, and their efficacy was evaluated by qPCR. Among these, shASPN-2 demonstrated the best efficacy and was subsequently used for all functional assays (Fig. S3D). The effectiveness of ASPN depletion was confirmed by qPCR and Western blotting, which showed a significant reduction in ASPN expression in GCIY and HGC27 cells transfected with shASPN-2 (Fig. S3E–G). Functional experiments were performed to evaluate the influence of ASPN knockdown on GC cell behavior. MTT assays revealed that ASPN deficiency decreased cell viability in both GCIY and HGC27 cells. 7-AAD staining and Annexin V/PI staining demonstrated that ASPN knockdown caused cell cycle arrest and increased apoptosis. Additionally, Transwell assays indicated that ASPN deficiency mitigated the motility of GCIY and HGC27 cells (Figs. 1A–E and 2A–D). These data collectively demonstrate that ASPN exerts an oncogenic function in GCIY and HGC27 cells by promoting cell viability, cell cycle progression, survival, and motility.

Fig. 1.

ASPN depletion causes reduce viability, cell cycle arrest, and increased apoptosis in HGC27 and GCIY cells. A MTT assay data showing the viability of control and ASPN-deficient HGC27 and GCIY cells. B, C 7-AAD staining assay results indicating the cell cycle status of control and ASPN-deficient HGC27 and GCIY cells. D, E Annexin V/PI staining data illustrating the apoptosis of control and ASPN-deficient HGC27 and GCIY cells

Fig. 2.

ASPN deficiency attenuates migration and invasion of HGC27 and GCIY cells. A, B Transwell assay data showing the migration of control and ASPN-deficient HGC27 and GCIY cells. C, D Transwell assay result indicating the invasion of control and ASPN-deficient HGC27 and GCIY cells

ASPN depletion mitigates the glycolysis of GC cells

To investigate the impact of ASPN on GC cell glycolysis, we performed ECAR and OCR assays in control and ASPN-deficient GCIY and HGC27 cells. The ECAR assay revealed that ASPN knockdown markedly decreased glycolysis, glycolytic capacity, and glycolytic reserve in both cell lines (Fig. 3A–D). Similarly, OCR assay data demonstrated that ASPN-deficient cells showed decreased maximal respiration, basal respiration, proton leak, and ATP production, along with an increased spare respiration capacity compared to control cells (Fig. 3E–H). Biochemical assays further confirmed that ASPN is essential for maintaining normal levels of ATP production, glucose consumption, lactic acid metabolism, G6PD activity, and pH balance in these cells (Fig. 4A–E). At the molecular level, ASPN depletion led to reduced expression of glucose transport regulators, including GLUT1, GLUT3, GLUT4, and SCO2, as shown by qPCR and Western blotting (Fig. 4F–H). These data highlight ASPN as a crucial regulator of GC cell glycolysis.

Fig. 3.

ASPN knockdown mitigates glycolysis in HGC27 and GCIY cells. A– D ECAR assay data showing the glycolysis, glycolysis capacity, and glycolysis reserve in control and ASPN-deficient HGC27 and GCIY cells. E–H OCR assay results indicating the maximal respiration, basal respiration, proton leak, ATP production, and spare respiration capacity in control and ASPN-deficient HGC27 and GCIY cells

Fig. 4.

ASPN regulates glucose metabolism and glucose transport proteins in HGC27 and GCIY cells. A– E Plots showing the levels of ATP (A), glucose consumption (B), supernatant lactic acid (C), G6PD activity (D), and pH(E) in control and ASPN-deficient HGC27 and GCIY cells. F qPCR data indicating the expression of GLUT1, GLUT3, GLUT4, and SCO2 in control and ASPN-deficient HGC27 and GCIY cells. G, H Western blot results illustrating the expression of GLUT1, GLUT3, GLUT4, and SCO2 in control and ASPN-deficient HGC27 and GCIY cells

ASPN maintains GC cell stemness

In order to investigate the effect of ASPN deficiency on GC cell stemness, colony formation and sphere formation assays were carried out. The outcomes showed that ASPN-deficient GCIY and HGC27 cells formed fewer colonies and smaller spheres than those from control GC cells (Fig. 5A–D). Additionally, flow cytometry analysis uncovered that the level of CD44, a CSC marker, was greatly reduced in ASPN-deficient cells (Fig. 6A–B). Western blotting further confirmed lower expression of additional CSC markers (OCT4, NANOG, and MYC) in ASPN-deficient GCIY and HGC27 cells (Fig. 6C–D). These results illustrate that ASPN is a critical factor in maintaining the stemness of GC cells.

Fig. 5.

ASPN deficiency impairs colony formation and sphere formation capacity of HGC27 and GCIY cells. A, B Colony formation assay data showing the colony formation status in control and ASPN-deficient HGC27 and GCIY cells. C, D Sphere formation assay results indicating the sphere formation ability of control and ASPN-deficient HGC27 and GCIY cells

Fig. 6.

ASPN depletion mitigates the stemness of HGC27 and GCIY cells. A, B Flow cytometry analysis data indicating the expression of CD44 in control and ASPN-deficient HGC27 and GCIY cells. C, D Western blot results showing the expression of OCT4, NANOG, and MYC in control and ASPN-deficient HGC27 and GCIY cells

ASPN regulates GC cell growth, glycolysis, and stemness in vivo

To confirm these in vitro findings, we conducted an in vivo analysis using a xenograft model with control and ASPN-deficient stable HGC27 cells. The ASPN-deficient tumors were smaller than the control tumors (Fig. 7A–C). IHC and TUNEL assays indicated that control tumor cells exhibited greater proliferation and less apoptosis compared to ASPN-deficient tumor cells (Fig. 7D–G). Western blotting further revealed that ASPN-deficient tumors expressed lower levels of OCT4, NANOG, and MYC and showed decreased expression of SCO2, GLUT1, GLUT4, and GLUT3 (Fig. 7H–I). These results reinforce the key function of ASPN in regulating GC cell glycolysis and stemness.

Fig. 7.

ASPN depletion impairs the growth, glycolysis, and stemness of GC cells in the xenograft model. A Image showing the morphology of control and ASPN-deficient xenograft tumors. B, C Plots illustrating the volume at the indicated time points B and weight of control and ASPN-deficient xenograft tumors on day 36 (C). D, E IHC data indicating the expression of proliferation marker Ki-67 in control and ASPN-deficient xenograft tumors. F, G TUNEL assay data depicting the apoptosis in control and ASPN-deficient xenograft tumors. H, I Western blot results demonstrating the expression of OCT4, NANOG, MYC, GLUT1, GLUT3, GLUT4, and SCO2 in control and ASPN-deficient xenograft tumors

ASPN promotes gastric cancer stemness and Glycolysis with HIF1α as a key downstream effector

To investigate the role of ASPN in regulating glycolysis and stemness, we performed sphere-formation assays and Western blot analysis in gastric cancer cells under various conditions. First, to directly assess the impact of ASPN on HIF1α transcriptional activity, we conducted a HIF-responsive luciferase reporter assay. ASPN knockdown significantly attenuated HIF1α-driven luciferase activity under hypoxic conditions, confirming its role in modulating the transcriptional output of the HIF1α pathway (Fig. S5F). Consistently, the ASPN-deficient group exhibited a significant reduction in both sphere-forming ability and glycolytic capacity compared to the control. Notably, this inhibitory effect was not rescued by the empty vector but was partially reversed upon HIF1α overexpression (Fig. S5A–D). At the molecular level, Western blot analysis confirmed that ASPN depletion downregulated the expression of HIF1α itself, stemness markers (OCT4, NANOG, MYC), and key glycolytic regulators (GLUT1, GLUT3, GLUT4, SCO2). Importantly, this downregulation followed the same pattern: it was not restored by the empty vector but was partially rescued by HIF1α overexpression (Fig. S5E). Collectively, by demonstrating that ASPN controls HIF1α transcriptional activity, and that HIF1α overexpression rescues the functional and molecular defects caused by ASPN loss, these results establish HIF1α as a key functional effector mediating the promotion of glycolysis and stemness by ASPN in gastric cancer cells.

Discussion

ASPN’s role in cancer appears to be highly context-dependent, functioning as an oncogene or a tumor suppressor in different cancer types. For example, in breast cancer, ASPN suppresses tumorigenesis by inhibiting TGF-β1 signaling [20]. Conversely, ASPN is generally considered oncogenic in GC. Our study reinforces this notion, demonstrating that ASPN is overexpressed in GC and contributes to tumor progression by enhancing cell viability, promoting cell cycle, increasing motility, and suppressing apoptotic cell death. Despite these insights, the broader role of ASPN in GC tumorigenesis remains partially elucidated, especially concerning its regulation of glycolysis and stemness, two critical hallmarks of cancer aggressiveness.To address this gap, we extended our investigation to explore how ASPN influences these processes in GC cells, thereby providing a more comprehensive understanding of its oncogenic function.

Glycolysis in cancer cells is a key metabolic pathway to support their high energy consumption, primarily generating ATP through this process [21]. Targeting glycolysis has demonstrated significant anti-tumor effects in various cancers [22, 23], and for GC, inhibiting glycolytic pathways is emerging as a potential therapeutic strategy [24]. Our study confirms the crucial role of ASPN in facilitating GC cell glycolysis and extends previous findings by demonstrating this effect across multiple cell models [10].More importantly, we provide mechanistic insight by showing that HIF1α overexpression can partially rescue the glycolytic defects induced by ASPN depletion, thereby positioning HIF1α as the key downstream effector. This establishes a functional, HIF1α-dependent axis for ASPN in promoting aerobic glycolysis, a common metabolic adaptation in cancer cells for energy production[25].

Targeting CSCs remains a crucial strategy in advancing cancer therapies [26]. Numerous factors contribute to maintaining CSC properties [27], but the role of ASPN in this context has been relatively unexplored. Previous findings demonstrated that ASPN knockout in prostate tumor allografts leads to a reduction in CSC populations [18]. Expanding on this, our bioinformatic analysis suggests that ASPN-associated genes in GC are involved in signaling pathways critical for stemness maintenance. Consistently, our experiments revealed that ASPN-deficient GC cells exhibited reduced expression of stemness markers and a diminished capacity for colony and sphere formation. These results position ASPN as a pivotal regulator of stemness in GC, raising questions about its function in other cancers and the corresponding molecular mechanisms.

Despite ASPN’s frequent dysregulation in cancer, the factors driving its aberrant expression are largely unknown. Several studies have shown that ASPN mRNA is targeted by miR-129-5p in cardiac fibroblasts and miR-4303 in chondrocytes [28, 29]. Interestingly, these miRNAs are also implicated in suppressing GC progression and glycolysis [30, 31]. Investigating whether these miRNAs or other regulatory mechanisms are responsible for ASPN upregulation in GC could offer valuable insights into its pathological role.

In conclusion, this study defines an oncogenic role for ASPN in GC, demonstrating that it enhances glycolysis and stemness largely in a HIF1α-dependent manner.These findings expand our understanding of ASPN’s role as an oncogene and suggest its potential as a druggable target in GC treatment. Further research into the mechanisms underlying ASPN dysregulation and its involvement in other cancers could provide a broader context for its role in cancer biology.

This study has several limitations that warrant consideration. Most importantly, while we have established HIF1α as a critical functional downstream mediator of ASPN in promoting glycolysis and stemness, the precise molecular mechanism linking ASPN to HIF1α activation remains to be determined. It is unclear whether this involves direct protein interaction, regulation of HIF1α mRNA, or modulation of upstream signaling pathways. Furthermore, due to constraints in timeline and resources, we could not directly confirm whether the ASPN/HIF1α axis exerts direct transcriptional control over key target genes such as GLUT1 and LDHA using ChIP or promoter-specific luciferase assays. Additionally, while we validated key effectors, a more comprehensive exploration of the transcriptomic and metabolomic landscape downstream of the ASPN-HIF1α axis is needed to fully elucidate this regulatory network. Therefore, future work will prioritize the validation of direct transcriptional regulatory relationships and conduct more comprehensive transcriptomic and metabolomic analyses centered on the ASPN–HIF1α signaling axis, with the aim of fully elucidating the dynamic regulatory mechanisms of this network in gastric cancer progression.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Guoxi Xu: Conceptualization, Funding acquisition, Project administration, Supervision, Writing—original draft, Writing—review & editing; Zhicong Cai: Data curation, Formal Analysis, Investigation, Methodology.Writing—review & editing; Yixiang Zhuang: Formal Analysis, Software, Validation, Visualization, Writing—review & editing; Huaishuai Wang: Investigation, Methodology, Resources, Writing—review & editing; Qiyi Lin: Data curation, Formal Analysis, Validation, Writing—review & editing; Tao Guo: Resources, Visualization, Writing—review & editing.Houquan Tao: Conceptualization, Funding acquisition, Supervision, Writing—review & editing.

Funding

This work was supported by the Quanzhou City Science & Technology Program of China (Grant No.2023N022S) and JinJiang Municipal Hospital(Shanghai sixth People’s Hospital Fujian) technology project (Grant No.2022JC01).

Data availability

All data supporting the conclusions of this study has been included in the manuscript and are available upon reasonable request from the corresponding author.

Declarations

Conflict of interest

The authors declare that they have no conflicts of interest with the contents of this article.

Competing interest

The authors declare no competing interests.

Consent to participate

N/A.

Ethical approval

This study was approved by the Ethics Committee of Jinjiang Municipal Hospital (Shanghai Sixth People’s Hospital Fujian)IACUC (Approval No. 2025-0072) in February 2023.

Generative AI in scientific writing

Not applicable.