Knockdown of GCNT2 promoted osteoblast differentiation by activating PI3K/AKT/mTOR pathway in osteoblasts

Abstract

Osteoporosis (OP) constitutes a systemic bone metabolic disorder characterized by complex pathogenesis and clinical recalcitrance. The regulatory mechanisms underlying OP require further investigation. Microarray profiling was employed to identify abnormally expressed genes in OP patients. OP patients and MC3T3-E1 osteoblasts were utilized for in vivo and in vitro research. Alkaline phosphatase (ALP) staining intensity, alizarin red staining (ARS) intensity, and expression of Runt-related transcription factor 2 (Runx2), osteocalcin (OCN), and osteopontin (OPN) were assessed to evaluate osteoblast differentiation. Quantitative reverse transcription polymerase chain reaction (qRT-PCR) was applied to measure mRNA expression, while protein expression was determined by Western blot analysis. Glucosaminyl (N-acetyl) transferase 2 (GCNT2) was upregulated in OP and Dex-treated MC3T3-E1 cells. Dex-treated MC3T3-E1 cells, whose osteogenic differentiation was impaired, served as the OP model. ALP and ARS intensities, together with Runx2, OCN, and OPN expression elevated by GCNT2 knockdown, were attenuated by the PI3K inhibitor LY294002. Suppressed GCNT2 promoted osteoblast differentiation by activating PI3K/AKT/mTOR signaling pathway in OP. These in-vitro findings suggest that GCNT2 knockdown merits further pre-clinical evaluation as a potential therapeutic strategy for OP.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-26289-2.

Introduction

Osteoporosis (OP) is a systemic skeletal disorder defined by diminished bone mass and microarchitectural deterioration, which increases bone fragility and fracture susceptibility^1,2. It is estimated that over 10 million people in the United States suffer from OP, and another 48 million people suffer from low bone mass. The prevalence of osteoporosis and low bone mass in the population over 50 years old is 16% and 37%, respectively³. Population ageing will further increase OP susceptibility^4–6; thus, its prevention and treatment require urgent investigation.

Current pharmacological strategies for OP target bone-resorption inhibition, bone-mineralization promotion, and aberrant bone-turnover correction^7,8. Systemic therapy, however, elicits delayed responses and adverse effects: bisphosphonates are associated with osteonecrosis of the jaw; hormonal agents elevate cardiovascular risk; traditional Chinese medicines exhibit complex composition and uncertain efficacy⁹. Gene therapy, an emerging modality, precisely targets genetic defects and pathological tissues, localizes to lesions, and manifests minimal adverse events, thereby attracting growing interest^10–12. Osteoclast differentiation and function can be suppressed by disrupting RANKL–RANK interaction, mitigating OP progression^13,14. Likewise, SOST gene secreted sclerostin to promoted bone formation thereby to treat OP¹⁵. Therefore, searching for potential differentially expressed genes (DEGs) has prospective significance for enriching the OP mechanism network and alleviating OP.

In current work, glucosaminyl (N-acetyl) transferase 2 (GCNT2) was identified to be aberantly upregulated in OP. GCNT2 encodes a β-1,6-N-acetylglucosaminyltransferase that catalyzes the biosynthesis of blood-group I antigens¹⁶. Mutations in GCNT2 are associated with a spectrum of disorders, including autosomal-recessive congenital cataracts that may be accompanied by facial dysmorphism and psychomotor delay¹⁷. Aberrant methylation of GCNT2 correlates with lymph-node metastasis in colorectal cancer, and in multiple malignancies—such as prostate and esophageal carcinomas—GCNT2 modulates tumor invasion and metastasis by regulating cell adhesion and key signaling pathways^18,19. However, the regulatory effects of GCNT2 on bone mineralization of OP remains unclear. Given that GCNT2 is expressed in human-induced pluripotent stem cell -derived mesenchymal stem cells with osteogenic differentiation potential²⁰, we postulate that GCNT2 may be involved in osteoblast differentiation in the pathogenesis of OP. These findings may help to identify candidate genes that may contribute to osteoblast dysfunction in OP.

Materials and methods

Bioinformatics

DEGs were identified using microarray datasets (accession number GSE100609) obtained from the Gene Expression Omnibus (GEO) repository. This series was designed to identify novel diagnostic and prognostic markers of post-menopausal OP in Asian-Indian women. Osteoclast precursor cells were isolated from peripheral blood monocytes of four osteoporotic patients and four age- and lifestyle-matched non-osteoporotic controls. It employs a single, standardized microarray platform (Affymetrix HG-U133 Plus 2.0), and offers a clinically homogeneous cohort that minimizes technical and biological confounders. Raw CEL files were preprocessed using the Robust Multi-array Average (RMA) algorithm for background correction, quantile normalization, and log2 transformation. Differential expression analysis was performed using the limma package²¹ (version 3.48.0) in R, with thresholds set at |log2 fold change (FC)| ≥ 1.5 and Benjamini–Hochberg adjusted p-value < 0.05 to control false discovery rate. DEGs were listed in Table S1. Significantly correlated DEG pairs were identified through Pearson correlation coefficient analysis (two-tailed, α = 0.01) using the Hmisc package (version 4.7-1)²², with absolute correlation coefficients > 0.8 considered biologically meaningful. After importing the DEGs into FunRich_3.1.3 software for analysis, the enriched signaling pathways of these DEGs were obtained using the Biological Process database, which is derived from Gene Ontology.

Primary osteoblast isolation

Primary osteoblasts were isolated from two distinct patient cohorts following approved protocols by the Institutional Review Board of Ethics Committee of Honghui Hospital, Xi’an Jiaotong University with written informed consent. Control samples (n = 40 donors, aged 45–60 years) consisted of mandibular bone fragments obtained during elective orthognathic procedures from individuals without metabolic bone disorders, while osteoporotic specimens (n = 40 patients, aged 62–75 years, dual-energy X-ray absorptiometry-confirmed T-score < − 2.5) were collected from vertebral bodies during spinal decompression surgeries. Clinical characteristics of the subjects are listed in Table 1. Within 2 h post-resection, bone fragments were rinsed in PBS containing 2× penicillin/streptomycin (Gibco, Carlsbad, USA), mechanically cleaned of periosteum and marrow using surgical curettes, and minced into 1–2 mm² particles. Sequential enzymatic digestion was performed through three-stage collagenase treatment in α-MEM: initial 30-min incubation with 0.2% collagenase type II (Worthington, Lakewood, NJ, USA) at 37 °C (discarded), followed by two digestions (90-min and 60-min durations) using 0.1% collagenase II + 0.05% trypsin-EDTA at 120 rpm shaking. Cell suspensions from latter digestions were pooled, filtered through 70 μm strainers, and centrifuged at 300 × g for 5 min before resuspension in osteogenic medium containing α-MEM supplemented with 10% FBS, 50 µg/mL ascorbic acid, 10 mM β-glycerophosphate, and 1% antibiotics. Cells were seeded at 5 × 10⁴ cells/cm² in collagen-coated flasks (Corning, NY, USA) and maintained at 37 °C/5% CO₂ with medium replacement every 72 h.

Table 1.

Clinical characteristics of the subjects.

	Healthy (n = 40)	Patients (n = 40)	X²	P value
Age (years)			22.170	< 0.001
< 45	32	11
≥ 45	8	29
Gender			8.205	0.004
Male	7	19
Female	33	21
Degree of education			6.753	0.080
Primary	1	5
Middle	3	7
High	15	16
College	21	12
Smoking status			3.566	0.168
Never	17	10
Ever	8	7
Current	15	23
Drinking status			1.971	0.373
Never	12	10
Ever	11	7
Current	17	23
Hypertension			0.228	0.633
Yes	12	14
NO	28	26
Diabetes			0.082	0.775
Yes	7	8
NO	33	32
History of osteoporosis			18.460	< 0.001
Yes	0	15
NO	40	25

Cell culture and treatment

The murine calvarial pre-osteoblast cell line MC3T3-E1 was procured from the Shanghai Institute of Biochemistry and Cell Biology (Chinese Academy of Sciences, Shanghai, China) with certificate of analysis verifying species origin and STR profile authentication. Cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM; HyClone, Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco), 100 U/mL penicillin, and 100 µg/mL streptomycin (Solarbio, Beijing, China). Cultures were incubated at 37 °C in a humidified 5% CO₂ atmosphere (Heracell™ 150i, Thermo Fisher) with medium replacement every 48 h. For subculturing, cells at 80–90% confluence were dissociated using 0.25% trypsin-EDTA (Gibco) for 2 min at 37 °C, followed by neutralization with complete medium. Cell counts and viability (> 95%) were determined via trypan blue exclusion using a Countess™ II FL Automated Cell Counter (Thermo Fisher). All experiments utilized cells between passages 8–12. Mycoplasma contamination was routinely tested using the MycoAlert™ PLUS Detection Kit (Lonza, LT07-118) with bi-monthly verification.

To investigate PI3K/Akt signaling involvement in osteogenic differentiation, MC3T3-E1 cells (passages 10–12) were treated with the specific PI3K inhibitor LY294002 (Selleckchem, S1105, ≥ 98% purity by HPLC). LY294002 stock solution (20 mM in DMSO, Sigma D2650) was diluted in osteogenic induction medium to working concentrations (10–50 µM) immediately before use. Cells seeded at 5 × 10³ cells/cm² in 6-well plates (Corning, 3516) were pre-treated with LY294002 or vehicle control (0.1% DMSO) for 2 h prior to osteogenic induction.

OP model cell establishment

For OP model induction, MC3T3-E1 cells at passage 8–10 were seeded in 6-well culture plates at a density of 1 × 10⁴ cells/cm². After 24 h of attachment, the medium was replaced with fresh medium containing 10⁻⁷ M dexamethasone (Dex)²³. The cells were then treated with DEX for 7 days, with the medium and Dex being refreshed every 2 days to maintain the stability of the experimental conditions. Osteogenic differentiation was initiated at 60% confluence by replacing medium with induction cocktail containing: 10⁻⁷ M Dex (Sigma-Aldrich), 10 mM β-glycerophosphate (Sigma-Aldrich), and 50 µg/mL l-ascorbic acid 2-phosphate (Sigma-Aldrich). The differentiation medium was refreshed every 48 h for 21 days under standard culture conditions (37 °C, 5% CO₂, 95% humidity).

RNA interference and transfection

Gene-specific small interfering RNAs (siRNAs) targeting murine GCNT2 were designed and synthesized by Genepharma (Shanghai, China). Dex-treated MC3T3-E1 cells (passage 10–12) were seeded in 6-well plates (Corning) at 1 × 10⁵ cells/well (1.9 × 10⁴ cells/cm²) in antibiotic-free α-MEM containing 10% FBS 24 h prior to transfection. Lipofectamine 3000 (Invitrogen) transfection complexes were prepared. After 15-min incubation at room temperature, complexes were added dropwise to cells at 60–70% confluence. Medium was replaced with complete osteogenic induction medium (containing 100 nM Dex) 6 h post-transfection. Knockdown efficiency was validated at 48 h through qRT-PCR analysis.

Osteogenic differentiation assessment

The differentiation medium was refreshed every 48 h for 21 days under standard culture conditions (37 °C, 5% CO₂, 95% humidity). Control groups received equivalent medium changes without osteogenic additives. Differentiation efficiency was monitored through alkaline phosphatase (ALP) activity quantification at day 7 and calcium deposition assessment at day 14 via alizarin red S staining (ARS)²⁴.

Cells were fixed in 4% paraformaldehyde (PFA; Leagene, DF0135) for 15 min at 25 °C, washed 3× with calcium-free PBS, and stained with 2% ARS (ARS; Leagene, BC0425, pH 4.2) for 45 min at 37 °C. Excess dye was removed by 5× washes with deionized water under gentle agitation (50 rpm). Mineralized nodules were imaged using a Nikon Eclipse Ti2 inverted microscope (10× objective, NIS-Elements AR 5.21.03 software) with consistent brightfield settings (exposure 50 ms, gain 1×). For quantitative analysis, ARS-stained calcium deposits were dissolved in 10% cetylpyridinium chloride (Sigma, C9002) for 30 min and absorbance measured at 562 nm (BioTek Synergy H1, Gen5 3.04 software) against a standard curve of 0–40 mM CaCl₂ (R² > 0.99).

For ALP staining, an ALP staining kit was used according to the manufacturer’s protocol (Sidansai, Shanghai, China). Cells were fixed in 4% PFA for 10 min at day 7 post-induction and stained using the BCIP/NBT Alkaline Phosphatase Color Development Kit (Sidansai, SDA1002) according to manufacturer specifications. Briefly, 200 µL working solution (BCIP: NBT = 1:50 in Tris-HCl buffer, pH 9.5) was applied for 30 min under light-protected conditions. Reaction was terminated by 3× PBS washes. Stained cells were imaged with a Zeiss Axio Observer 7 (20×/0.8 NA Plan-Apochromat objective, ZEN 3.4 blue edition) using standardized DIC settings. For quantitative analysis, p-nitrophenyl phosphate (pNPP; Sidansai, SDA1002) substrate was reacted with cell lysates (RIPA buffer, 30 min on ice) for 30 min at 37 °C. Reaction absorbance was measured at 405 nm with background subtraction at 655 nm using a BioTek Synergy H1 microplate reader. ALP activity was normalized to total protein concentration quantified via BCA assay (Pierce™ 23225).

qRT-PCR

Total RNA was isolated from MC3T3-E1 cells using TRIzol™ Reagent (Invitrogen) following the manufacturer’s protocol. RNA integrity was verified via 1% agarose gel electrophoresis (120 V, 20 min) showing distinct 28 S/18S ribosomal RNA bands. RNA purity was quantified using a NanoDrop OneC spectrophotometer (Thermo Fisher). Genomic DNA elimination and cDNA synthesis were performed using the PrimeScript™ RT Master Mix (Takara) with 1 µg total RNA in a 20 µL reaction volume under the following thermal conditions: 37 °C for 15 min (reverse transcription), 85 °C for 5 s (enzyme inactivation), followed by 4 °C hold. Quantitative PCR was conducted using TB Green™ Premix Ex Taq™ II (Takara) on an Applied Biosystems 7500 Real-Time PCR System (Thermo Fisher) with the following reaction parameters: 95 °C for 30 s (initial denaturation), 40 cycles of 95 °C for 5 s → 60 °C for 34 s. Melt curve stage: 95 °C for 15 s → 60 °C for 1 min → 95 °C for 30 s (continuous). Reactions contained 10 µL TB Green mix, 0.4 µM forward/reverse primers, 2 µL cDNA template (1:5 dilution), and nuclease-free water to 20 µL final volume. Each sample was analyzed in triplicate with inter-run calibrators. The relative expressions of genes were calculated using the 2^−ΔΔCt method. Primer sequences were as followed:

Human GCNT2: forward, 5′-ACTGTGTGCACCTGGATCAG-3′;

reverse, 5′-GCTGTAGGTGTCCTTGGACC-3′,

Mouse GCNT2: forward, 5′-TGTTCCTGGCTCTATGCCAAA-3′;

reverse, 5′-TTAGCAAACAGGCTTGGTGAAT-3′,

Mouse Runx2: forward, 5′-TTCAACGATCTGAGATTTGTGGG-3′;

reverse, 5′-GGATGAGGAATGCGCCCTA-3′,

Mouse OCN: forward, 5′-CTGACCTCACAGATCCCAAGC-3′;

reverse, 5′-TGGTCTGATAGCTCGTCACAAG-3′,

Mouse OPN: forward, 5′-CACTCCAATCGTCCCTACAGT-3′;

reverse, 5′-CTGGAAACTCCTAGACTTTGACC-3′,

Human GAPDH: forward: 5′-ACAACTTTGGTATCGTGGAAGG-3′;

reverse: 5′-GCCATCACGCCACAGTTTC-3′.

Mouse GAPDH: forward: 5′-TGACCTCAACTACATGGTCTACA-3′;

reverse: 5′-CTTCCCATTCTCGGCCTTG-3′.

Western blot assay

Total protein was isolated from Dex-treated MC3T3-E1 cells using RIPA (Sigma, NJ, USA) lysis buffer. Protein samples were separated by SDS-PAGE and transferred to polyvinylidene fluoride membranes (Millipore, MA, USA). Membranes containing proteins were blocked with 5% non-fat milk for 1 h and then incubated with primary antibodies purchased from abcam (Cambridge, UK) for anti-Runx2 (1/1000, ab236639), anti-OCN (1/1000, ab309521), anti-OPN (1/1000, ab218237), anti-PI3K (1/1000, ab191606), anti-p-PI3K (1/1000, ab278545), anti-AKT (1/10000, ab179463), anti-p-AKT (1/1000, ab38449), anti-mTOR (1/10000, ab134903), and anti-p-mTOR (1/1000, ab109268), and GAPDH (1/1000, ab8245) overnight at 4 °C. Then membranes were washed with TBST and incubated with secondary antibody (1:5000, ab6721). Protein bands were visualized using a gel imaging system (Bio-Rad, Hercules, USA).

Statistical analysis

All data performed at least 3 times were expressed as mean ± SD performed by SPSS version 17.0 (SPSS Inc., Chicago, USA). Kolmogorov–Smirnov and Shapiro–Wilk tests were used for normality test. Analysis between two groups was determined by student’s t-test. One-way ANOVA was used for comparison among different groups. Differences were considered as statistically significant when P values less than 0.05.

Results

GCNT2 is over-expressed in OP

We first screened genes with differential expression in OP by analyzing GSE100609. Figure 1A displays only the top 20 DEGs ranked by adjusted p-value. Among these significantly upregulated genes, GCNT2 is a gene involved in mesenchymal stem cell differentiation²⁰ but its role in OP is not yet clear. Currently, there is no research directly linking GCNT2 with osteoblast differentiation, which is novel (Fig. 1A). Subsequently, we isolated osteoblasts from the collected samples of healthy and osteoporotic patients, and detected the expression level of GCNT2 in them. The results showed that the expression of GCNT2 in the osteoporotic patient group was significantly higher than that in the healthy control group (Fig. 1B). Furthermore, there is a significant correlation between the expression of GCNT2 and age and history of osteoporosis (Table 1).

Fig. 1.

GCNT2 is over-expressed in OP. (A) Heat map of DEGs identified by GSE100609 dataset, showing multiple DEGs between the control and osteoporosis groups, with GCNT2 being significantly upregulated. (B) GCNT2 expression levels in osteoblasts isolated from 40 OP patients and 40 healthy controls detected by qRT-PCR. **P < 0.01.

DEX treatment inhibits osteoblast differentiation of MC3T3-E1 cells

Next, we induced an osteoporosis cell model in vitro using DEX. Both ARS intensity and ALP intensity of DEX-treated MC3T3-E1 cells were dramatically decreased, demonstrating that mineralization levels of MC3T3-E1 cells were declined under the treatment of DEX (Fig. 2A,B). Concomitantly, osteogenic proteins Runx2, OCN, and OPN were significantly down-regulated (Fig. 2C,D). Interestingly, GCNT2 was also over-expressed in DEX-treated MC3T3-E1 cells (Fig. 2E). Hence, DEX-treated MC3T3-E1 cells could serve as OP model cells in the subsequent experiments for studies on osteoblast differentiation ability.

Fig. 2.

DEX treatment inhibits osteoblast differentiation of MC3T3-E1 cells. MC3T3-E1 cells treated with 100 nM DEX for 7 days followed by 14 days osteogenic induction. DEX reduced (A) mineral nodules (ARS), (B) ALP activity, and (C,D) Runx2, OCN, OPN expression; (E) GCNT2 was concurrently up-regulated. **P < 0.01.

Silenced GCNT2 promotes osteoblast differentiation of MC3T3-E1 cells under DEX treatment

To probe the role of GCNT2 in Dex-induced osteoblastic differentiation, MC3T3-E1 cells were transfected with two distinct siRNAs (si-GCNT2 1#, si-GCNT2 2#) against GCNT2 or with si-NC. Both si-GCNT2 1# and si-GCNT2 2# markedly suppressed GCNT2 expression relative to si-NC, with si-GCNT2 2# exhibiting superior efficiency and hence selected for downstream assays (Fig. 3A). ARS staining revealed that Dex alone reduced calcium-nodule formation versus control, whereas GCNT2 silencing substantially restored nodule number; quantitative densitometry confirmed rescue of Dex-suppressed mineralization (Fig. 3B). ALP staining and intensity analysis showed Dex markedly diminished ALP activity, an effect abolished by GCNT2 knockdown, restoring ALP to control levels (Fig. 3C). Nonetheless, GCNT2 knockdown could abrogate such suppression, elevating ALP activity to a level comparable to the control group, suggesting the promotive effect of GCNT2 silencing on early osteogenic differentiation.

Fig. 3.

Silenced GCNT2 promotes osteoblast differentiation. (A) siRNA efficacy; (B) ARS; (C) ALP. (D,E) Expression levels of Runx2, OCN, and OPN measured by qRT-PCR and Western blot revealing DEX-induced downregulation of these osteogenic markers and their restoration by GCNT2 deficiency. *P < 0.05, **P < 0.01, vs. si-nc group. ^##P < 0.01, vs. DEX + si-nc group.

To further elucidate the molecular mechanisms underlying the promoting effect of GCNT2 silencing on osteoblastic differentiation, the mRNA and protein expression levels of key osteogenic markers were assessed. At the mRNA level (Fig. 3D), DEX treatment led to a significant reduction in the expression of Runx2, OCN, and OPN. However, GCNT2 deficiency remarkably reversed these changes, restoring the expression levels of these markers to near control levels. Consistent with the mRNA results, Western blot analysis revealed that GCNT2 silencing could markedly upregulate the protein expression of Runx2, OCN, and OPN, which were downregulated by DEX treatment (Fig. 3E). Collectively, these findings indicate that silencing GCNT2 effectively counteracts the inhibitory effect of dexamethasone on the osteoblastic differentiation of MC3T3-E1 cells by upregulating key osteogenic markers.

GCNT2 related genes are enriched in the mTOR signaling pathway

To explore the potential functional associations of GCNT2 with other genes, Pearson correlation analysis was performed on the gene expression profiles. As depicted in the heatmap (Fig. 4A), GCNT2 exhibited both positive and negative correlations with DEGs (Pearson r from − 0.62 to + 0.93). These correlations are descriptive and do not infer direct regulatory or causal relationships Notably, genes such as HIST2H3A, DPY19L3, and RGMA showed strong positive correlations with GCNT2, with Pearson correlation coefficients exceeding 0.9. In contrast, a few genes, including ENHO and PAPD5, displayed negative correlations, suggesting diverse regulatory mechanisms. Subsequently, pathway enrichment analysis was conducted to identify the biological pathways in which these GCNT2-related genes are involved. The analysis revealed that these genes are significantly enriched in ten biological pathways (Fig. 4B). Among these, the mTOR signaling pathway emerged as the most gene-enriched pathway, accounting for 28.6% of the total genes analyzed. Other pathways, such as BMP receptor signaling and the E2F transcription factor network, also showed enrichment but with relatively lower gene percentages (14.3% each). The mTOR signaling pathway is well - known for its critical role in regulating cell growth, proliferation, and survival, which are closely linked to osteoblastic differentiation and bone metabolism. The enrichment of GCNT2 - related genes in this pathway suggests a potential regulatory role of GCNT2 in osteoblastic processes via the mTOR pathway.

Fig. 4.

GCNT2 related genes are enriched in mTOR pathway. (A) Pearson correlation analysis between GCNT2 gene and other genes. (B) Enrichment analysis was calculated using FunRich (GO-Biological Process), demonstrating that GCNT2-related genes were significantly enriched in ten biological pathways.

Suppressed GCNT2 promotes osteoblast differentiation by activating PI3K/AKT/mTOR signaling pathway

To further elucidate the mechanism by which GCNT2 knockdown promotes osteoblastic differentiation, we investigated the activation of the PI3K/AKT/mTOR signaling pathway. As shown in Fig. 5A, Western blot analysis revealed that the protein levels of p-PI3K, p-AKT, and p-mTOR were significantly upregulated in the si-GCNT2 group compared to the si-NC group. Specifically, the relative protein levels of p-PI3K/PI3K, p-AKT/AKT, and p-mTOR/mTOR increased by approximately 3.1-fold, 3.3-fold, and 1.8-fold, respectively. These results indicate that GCNT2 suppression activates the PI3K/AKT/mTOR signaling pathway (Fig. 5A). To validate the role of the PI3K/AKT/mTOR pathway in GCNT2-mediated osteoblastic differentiation, MC3T3-E1 cells were treated with the PI3K inhibitor LY294002 after GCNT2 knockdown. ARS staining showed that the increase in calcium nodule formation induced by GCNT2 knockdown was significantly attenuated by LY294002 treatment (Fig. 5B). Similarly, ALP staining demonstrated that LY294002 treatment reduced the ALP activity in the GCNT2-knockdown group (Fig. 5C). The mRNA expression levels of Runx2, OCN, and OPN, which were upregulated by GCNT2 knockdown, were markedly decreased by LY294002 treatment (Fig. 5D). Consistently, Western blot analysis showed that the protein expression levels of these osteogenic markers were also reduced by LY294002 treatment (Fig. 5E). Collectively, these findings suggest that GCNT2 suppression promotes osteoblastic differentiation of MC3T3-E1 cells by activating the PI3K/AKT/mTOR signaling pathway. The use of the PI3K inhibitor LY294002 effectively reversed the promotive effects of GCNT2 knockdown on osteogenic differentiation, highlighting the critical role of the PI3K/AKT/mTOR pathway in this process.

Fig. 5.

Suppressed GCNT2 promotes osteoblast differentiation by activating PI3K/AKT/mTOR signaling pathway. (A) Protein expression of PI3K, AKT, mTOR as well as their phosphorylated forms measured by Western blot assay. (B) ARS, (C) ALP, (D,E). Expression levels of Runx2, OCN, and OPN measured by qRT-PCR and Western blot revealing that LY294002 treatment partly weakened GCNT2 deficiency effects. **P < 0.01, vs. si-NC group. ^##P < 0.01, vs. DEX + si-NC group. ^&P < 0.05, ^&&P < 0.01, vs. si-GCNT2 group.

Discussion

OP affects patients’ health and quality of life due to its increased risk of fracture, and its mechanism research and prevention are facing severe challenges²⁵. In our study, GCNT2 was verified to be up-regulated in OP patients and DEX-treated MC3T3-E1 cells. Furthermore, knockdown of GCNT2 promoted osteoblast differentiation by activating PI3K/AKT/mTOR signaling pathway.

GCNT2 belongs to a group of N-glycosyltransferases which play key roles in determining protein structure and function²⁶. Previous studies have shown that GCNT2 regulates glycosylation related genes expression in different ways in esophageal squamous cell carcinoma, colon cancer, melanoma and breast cancer, thereby affecting tumor development^27–29. However, the roles and mechanisms of GCNT2 in OP remain unclear. Our data suggested that GCNT2 was identified to be over-expressed in OP patients from microarray profile analysis, which was verified in isolated osteoblasts from the collected samples of OP patients obtained in this study.

During bone metabolism, the differentiation regulation of osteoblasts and osteoclasts is a critical factor in maintaining skeletal homeostasis. Recent research has revealed that the dual regulatory role of differential genes in both osteogenic and osteoclastic differentiation provides novel insights into the pathogenesis of bone diseases. For instance, Drg2 influences the differentiation of osteoblasts and osteoclasts by modulating Rac1 activity, thereby regulating bone mass in vivo. This finding indicates that Drg2 holds potential as a therapeutic target for bone loss-related diseases³⁰. Furthermore, miRNA-324 maintains bone homeostasis by regulating the activity of both osteoblasts and osteoclasts. Its deficiency leads to increased bone mineral density and accelerated bone formation rate, suggesting the potential application of miRNA-324 in the treatment of bone disorders³¹. Similarly, miR-196b-5p impacts bone homeostasis by targeting SEMA3A to regulate the differentiation of osteoblasts and osteoclasts. Its inhibition may offer therapeutic benefits for alleviating OP³². Therefore, in this study, we innovatively investigated the regulatory role of GCNT2, which is differentially expressed in osteoclast differentiation, in the process of osteogenic differentiation. In this study, Dex treatment inhibited the osteoblast differentiation of MC3T3-E1 cells and increased the expression of GCNT2, indicating that Dex treated MC3T3-E1 cells were established as OP model cells.

Subsequently, knockdown of GCNT2 promoted osteoblast differentiation of OP model cells, indicating that GCNT2 could regulate osteogenesis of OP. Then, enrichment analysis of signal pathways suggested that mTOR signaling pathway may involve in osteoblast differentiation of OP. Activation of the PI3K/AKT pathway stimulates osteoblast proliferation and differentiation, and inhibiting osteoclast formation, thereby is closely related to bone metabolism^33,34. For instance, lncRNA AK023948 has been reported to regulate osteoblast proliferation in estrogen-deficient OP rats by activating PI3K/AKT signaling pathway³⁵. PI3K/AKT/mTOR signaling pathway was also activated by glucocorticoids to promote osteoclast autophagy³⁶. In this study, the activation of PI3K/AKT/mTOR signaling pathway induced by silenced GCNT2 was suppressed by PI3K inhibitor, so did the osteoblast differentiation. These results indicated that suppressed GCNT2 promoted osteoblast differentiation via activating PI3K/AKT/mTOR signaling pathway, which was in line with previous studies³⁷.

The present study has several limitations that warrant consideration. First, the transcriptomic discovery phase relied exclusively on the GSE100609 dataset, which was generated from peripheral blood mononuclear cells enriched for osteoclast precursors; although GCNT2 was concordantly up-regulated in our osteoblast cultures, PBMC-derived signatures may not fully reflect osteoblast-specific biology. Second, the in-vitro model employed dexamethasone-treated MC3T3-E1 cells, a glucocorticoid-induced osteoporosis paradigm that captures only one facet of the multifactorial human disease. Third, we acknowledge that the control osteoblasts were isolated from mandibular bone of younger individuals (45–60 years), whereas osteoporotic cells were obtained from vertebral bodies of older patients (62–75 years). Sex distribution was identical (70% female in both groups), yet age and anatomical origin remain potential confounders. Because the sample size and study design did not allow statistical adjustment for these variables, the observed up-regulation of GCNT2 could be partially influenced by age-related transcriptional drift or site-specific bone remodeling signatures rather than osteoporosis per se. Future work will employ age- and site-matched cohorts, or utilize co-culture models that isolate the intrinsic osteoporotic milieu, to more rigorously attribute GCNT2 dysregulation to the disease phenotype. Future studies incorporating multi-tissue transcriptomic analyses, genetically modified animals, and patient-derived primary cells are required to address these limitations.

Conclusion

In conclusion, silenced GCNT2 promoted osteoblast differentiation of DEX-treated MC3T3-E1 cells via activating PI3K/AKT/mTOR signaling pathway. This study may provide clues to discovering new target to treat OP.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

OP

Osteoporosis

GCNT2

Glucosaminyl (N-acetyl) transferase 2

ARS

Alizarin red staining

ALP

Alkaline phosphatase

Runx2

Runt-related transcription factor 2

OCN

Osteocalcin

OPN

Osteopontin

Author contributions

S.S. conceived the study; Y.H. conducted the experiments; S.W., D.H. and L.Z. analyzed the data; Y.H. was a major contributor in writing the manuscript. All authors read and approved the final manuscript.

Funding

This study was supported by Natural Science Foundation of Shaanxi Province, Grant/Award Number: 2021SF-241, 2025JC-YBQN-1269.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval

The study was approved by the Ethics Committee of Honghui Hospital, Xi’an Jiaotong University (No.202403018). All experiments were performed in accordance with relevant guidelines and regulations.

Informed consent

Written informed consent was obtained from all patients.