VEGF secreted by human dental pulp stem cell promotes spinal cord injury repair by inhibiting microglial pyroptosis through the PI3K/AKT pathway

Abstract

Background

Spinal cord injury (SCI) remains a devastating central nervous system disorder. The complex pathological microenvironment following SCI, particularly the imbalance in neuroinflammation, contributes to its therapeutic challenges. Microglial pyroptosis, a type of programmed cell death, is pivotal in exacerbating neuroinflammation and secondary tissue damage after SCI. Our previous study demonstrated the inhibitory efficacy of conditioned medium (CM) derived from human dental pulp stem cells (DPSCs) on the microglial pyroptosis and its positive effects on the functional recovery in SCI models. However, the major secretory product in CM responsible for inhibiting microglial pyroptosis remains unclear.

Objective

We aim to investigate whether vascular endothelial growth factor (VEGF) secreted by human DPSCs can alleviate microglial pyroptosis through the PI3K/AKT signaling pathway and promote motor and electrophysiological function recovery in SCI mice.

Methods

Human DPSCs were isolated and cultured, and CM was collected for VEGF detection and further treatment. The BV2 cell line was established as a microglial pyroptosis model through the administration of lipopolysaccharide (LPS). SCI was induced in mice. Molecular and histological techniques were employed to evaluate pyroptosis and explore the underlying mechanisms both in vivo and vitro.

Results

Human DPSC-derived VEGF significantly inhibited microglial pyroptosis both in vitro and vivo, as evidenced by the decreased expression of pyroptosis-related markers, such as caspase-1 and IL-1β. The anti-pyroptotic effects of VEGF were closely associated with the activation of the PI3K/AKT signaling pathway, which was identified as a key regulatory mechanism. Importantly, treatment with DPSC-CM improved the recovery of motor function and electrophysiological conduction in SCI mice.

Conclusion

Human DPSC-derived VEGF alleviates microglial pyroptosis via the PI3K/AKT signaling pathway, thereby contributing to the repair of SCI. Our study provides new insights into the potential for therapy of DPSCs and their secreted factors, particularly VEGF, offering new perspectives on the treatment of SCI.

Graphical Abstract

Introduction

Spinal cord injury (SCI) is a devastating condition that often leads to permanent motor and sensory impairments, diminishing patients’ quality of life [1]. The primary injury occurs immediately after trauma, causing mechanical damage, hemorrhage, and disruption of the blood-spinal cord barrier. Following SCI, a series of pro-inflammatory factors are released, such as IL-1β and TNF-α, leading to excessive inflammation during the acute phase, which further aggravates the injury [2]. The secondary injury is characterized by ischemia, inflammation, oxidative stress, and neuronal death, which exacerbate the initial injury and hinder recovery [3, 4]. Neuroinflammation plays a central role in worsening tissue damage and functional deficits in both phase [5, 6].

Pyroptosis, an extremely inflammatory form of programmed cell death, plays a key role in neuroinflammation following SCI. Pyroptosis is primarily mediated by inflammasomes, which trigger a cascade of events leading to cell death and the secretion of pro-inflammatory cytokines. In the context of SCI, the NLRP3 inflammasome is particularly critical, as its activation in microglia drives caspase-1 production, facilitating the processing and activation of IL-1β and IL-18. These cytokines promote inflammation, further exacerbating the injury [2, 7–10]. Conversely, inhibition of microglial pyroptosis has emerged as a promising strategy to mitigate neuroinflammation and promote functional recovery after SCI [11–14].

Human dental pulp stem cells (DPSCs), which originate from the neural crest and exhibit regenerative potential, have been explored as a promising therapeutic option for SCI [15]. These cells exhibit high clonogenicity and proliferative potential, as well as the ability to form dense calcified colonies and occasional nodules [16]. Nosrat et al. first reported the use of DPSCs for SCI treatment by transplanting dental pulp tissue into transected rat spinal cords [17]. DPSCs may secrete a variety of neuroprotective and regenerative factors, including vascular endothelial growth factor (VEGF), insulin-like growth factor 1 (IGF-1), and various cytokines, which collectively support tissue repair and regeneration. Previous studies showed that human DPSC-derived conditioned medium (DPSC-CM) possesses significant therapeutic potential, improving recovery after SCI by inhibiting microglial activation and reducing neuroinflammation [18–22]. However, the key components within DPSC-CM that contribute to these beneficial effects remain to be fully elucidated.

Our previous study was the first to demonstrate that human DPSC- CM can promote SCI repair by inhibiting microglial pyroptosis [22]. Further proteomic analysis revealed that VEGF and the PI3K/AKT signaling pathway may play central roles in this process. VEGF, a acknowledged factor in angiogenesis and neuroprotection, has been implicated in promoting functional recovery following SCI [23]. It enhances vascular integrity, reduces ischemic damage, and supports neuroprotection and tissue regeneration [24]. Additionally, VEGF activates several signaling pathways, one of which is the PI3K/AKT pathway, crucial for regulating cell survival, growth, and metabolism [25].

In this study, we explored the hypothesis that VEGF secreted by human DPSCs can alleviate microglial pyroptosis by activating the PI3K/AKT pathway, thereby promoting functional recovery in SCI mice.

Methods and materials

Animals

Female C57BL/6 J Nifdc mice, aged 8 weeks and weighing 20–25 g, were obtained from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). A total of seventy-two mice were used in the study, and the animals were divided into four groups, with eighteen mice in each group. To eliminate potential sex-related variations and reduce the risk of urinary tract infections, only female mice were included in the study [26]. The mice were kept in standard cages with a 12-h light–dark cycle, with three mice per cage. Adequate food and water were provided ad libitum, and the temperature was controlled within the range of 22 to 24 °C. Ethical approval for the research was obtained from the Animal Experiments and Experimental Animal Welfare Committee of Capital Medical University (Beijing, China) under approval No. AEEI-2024-142, issued on May 24, 2024. All experimental procedures were conducted in accordance with the ethical guidelines set by the Committee for the Purpose of Control and Supervision of Experimentation on Animals (CPCSEA).

Isolation and culture of human DPSC

Normal third molar specimens were collected from healthy adults aged 18–25 at the Oral Center of Beijing Luhe Hospital, Capital Medical University. The study received ethical approval from the Medical Ethics Committee of Beijing Luhe Hospital, Capital Medical University, on July 22, 2024 (Approval No. 2024-LHKY-083–01). All human participants provided written informed consent before the study commenced. Freshly extracted teeth were preserved in phosphate-buffered saline (PBS) (Gibco, USA) containing penicillin/streptomycin (P/S) (Gibco, USA) and disinfected with medical ethanol for 5 min, followed by thorough rinsing with PBS. The teeth were sectioned using a dental drill to access the pulp, which was subsequently extracted and minced into approximately 1mm³ fragments. The extracted dental pulp was transferred into a solution containing collagenase type I (3 mg/mL, Sigma-Aldrich, USA) and dispase II (4 mg/mL, Sigma-Aldrich, USA), and digested at 37 °C with shaking for 45 min. the cell suspension was then filtered using a cell sieves (Falcon, Beijing, China), and centrifuged at 252 × g for 5 min. After this, the pellet was resuspended in α-modified Eagle medium (α-MEM) (Gibco, USA) containing 20% fetal bovine serum (FBS) (Gibco, USA), 1%P/S and cultured at 37 °C in 5% CO₂, the cells labeled as DPSC, P0. After approximately one week, the culture medium was changed, and when the cells reached 70–80% confluence, they were passaged using StemPro Accutase (Gibco, USA) and α-MEM with 10% FBS, 1% P/S. The DPSC used in this experiment were from passages P3-P6.

Flow cytometry

Flow cytometry was used to identify the DPSCs. The human DPSCs were dissociated into a single-cell suspension, and the cell density was adjusted to 1 × 10⁶ cells/100μL, then the cells were incubated at room temperature in the dark for 30 min with the following antibodies: FITC anti-human CD73 Antibody (Biolegend, USA), PerCP anti-human CD45 (Biolegend, USA), PE anti-human CD90 (Biolegend, USA), APC anti-human CD34 (Biolegend, USA). After incubation, the suspension was centrifuged at 252 × g for 5 min, then resuspended. The analysis was conducted using the BeamCyte-1026 flow cytometer (Changzhou BDAC Biotech Co., Ltd, China).

Human DPSC-CM collection

Human DPSCs from passages P3-P6 were chosen for the experiment. Upon reaching 70–80% confluence, the cells were cultured in serum-free Dulbecco Modified Eagle Media: Nutrient Mixture F-12 (DMEM/F12) (Gibco, USA) for an additional 48 h. The cell-free DMEM/F12 was incubated under the same conditions for 48 h and used as the control conditioned medium (Con-CM). After the incubation, the medium was harvested and centrifuged at 252 × g for 5 min to eliminate cellular debris. It was subsequently filtered through a 0.22 μm cell strainer to ensure sterility. The resulting filtrate was transferred to an amicon Ultra-15 Centrifugal Filter Unit (Millipore, USA) and centrifuged at 7000 × g for 20 min at 4 °C to isolate DPSC-CM. Afterward, the DPSC-CM was stored at − 80 °C for short-term use.

Mass spectrometry

As shown in our previous research, we collected human DPSC-CM samples from three volunteers of the same generation. The Con-CM samples were incubated with the cellular culture medium under the same culture conditions. Mass spectrometry analysis was performed by RayBiotech Co., Ltd. (Guangzhou, China). We conducted Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses to explore the functions of differentially expressed proteins (DEPs). Functional enrichment analysis was assessed using Fisher’s exact test in the clusterProfiler package of R/Bioconductor, with a threshold of count ≥ 2 and p-value < 0.05.

Enzyme-linked immunosorbent assay (ELISA)

The VEGF concentration in DPSC-CM was measured using the ELISA. The DPSC-CM was retrieved from the −80℃ to determine the thaw to room temperature. Following the manufacturer’s instructions (Lianke bio, Beijing, China), a standard curve was prepared. Subsequently, the VEGF concentration in the DPSC-CM was determined using a microplate reader for dual-wavelength detection. The VEGF levels were assessed based on absorbance value.

BV2 cell culture and pyroptosis model construction

BV2 cells were purchased from Procell Co., Ltd. (Wuhan, China) and cultured in DMEM/F12 with 10% FBS and 1% P/S. The experiment is divided into four groups: (a) Control group: treated with Con-CM; (b) LPS group: 1 μg/mL lipopolysaccharide (LPS)( Sigma-Aldrich, USA) induced pyroptosis and stimulated for 24 h; (c) DPSC-CM + LPS group: cultured with DPSC-CM for 12 h, followed by 1 μg/mL LPS stimulation for 24 h; (d) DPSCC-CM + LPS + BEV group: DPSC-CM combined with 1ug/ml Bevacizumab (BEV) (a humanized anti-VEGF monoclonal antibody that blocks the effects of VEGF) (MedChemExpress, China) for 12 h, followed by 1 μg/mL LPS stimulation for 24 h.

Cell counting kit-8 (CCK-8) assay

Cell viability was evaluated using the CCK-8 assay according to the manufacturer's instruction (Beyotime, Suzhou, China). BV2 cells were plated in 96-well plates at a density of 5 × 10⁴ cells /100 μL. The cells were treated with LPS, DPSC-CM + LPS, or DPSC-CM + LPS + BEV, and incubated for 24 h. Afterward, 10μL of CCK-8 solution was added to each well, followed by an additional 2 h incubation at 37 °C in the dark with 5% CO₂. Absorbance at 450 nm was measured using a microplate reader (Thermo Fisher Scientific, USA).

Spinal cord injury model and treatment

The SCI surgical procedure was performed as in our previous study [22], in brief, mice were anesthetized with a combination of isoflurane (Lunan Pharmaceutical Group Corporation, Linyi, China) and air (4% for induction and 2% for maintenance). The T10 vertebral spinous process, which was the most prominent, was located, and the tissue was dissected layer by layer to remove the lamina, thereby exposing the T10 segment of the spinal cord. SCI was induced using a weight-dropping method with an IMPACTOR MODEL III (State University of New Jersey, New Brunswick, NJ, USA), utilizing a 25 mm high, 5 g weight, and 2 mm diameter rod. Following the impact, both the tail-flick reflex and hind limb paralysis were observed, indicating successful modeling of SCI. Postoperatively, bladder compression was performed twice daily to assist urination until the mice were able to urinate spontaneously. The mice were subsequently randomly allocated into four groups: (a) Sham group, with laminectomy but no SCI; (b) SCI group, mice were intraperitoneally administered 200 μL of Con-CM daily for 3 consecutive days; (c) human DPSC-CM group, mice received intraperitoneal injections of 200 μL of human DPSC-CM daily for 3 consecutive days; (d) human DPSC-CM + BEV group, mice were treated with 200 μL of DPSC-CM containing BEV (10 mg/kg, MedChemExpress, Shanghai, China) via intraperitoneal injection daily for 3 consecutive days.

Behavioral assessment

The Basso Mouse Scale (BMS) and inclined plane test were employed to assess the motor function recovery in mice on 1, 2, 3, 5, 7, 14, 21and 28 days post-surgery. The BMS score is scored on a scale of 0 to 9, where 9 represents normal function and 0 indicates complete paralysis. Each mouse was observed for at least 5 min. For the inclined plane test, a plane with adjustable angles and sufficient friction was selected. The maximum angle at which the SCI mice could maintain a stable position was measured. Starting from a horizontal position (0°), the angle was increased by 5° increments. The mouse was placed on the plane, and the maximum angle at which it could stay in position for 5 s without descending was noted. This procedure was repeated three times, and the mean value of the angles was calculated.

Electrophysiological recording

Motor and sensory function were evaluated using motor-evoked potentials (MEP) and somatosensory-evoked potentials (SEP) at 28 days after SCI. Mice were anesthetized with isoflurane, and an electrophysiological detector (Iridi Technology, Zhuhai, China) was used to measure both MEP and SEP. For MEP measurement, recording electrodes were positioned on the Achilles tendon, and stimulating electrodes were positioned above the anterior fontanelle. For SEP measurement, stimulating electrodes were implanted in the skeletal muscles of the hind limbs, while recording electrodes were positioned over the sensory-motor cortices. Neurological function recovery was evaluated based on the amplitudes and peak latencies of MEP and SEP.

Spinal cord tissue collection

Spinal cord tissue was harvested at 3 and 28 days after SCI. Mice from each group were administered deep anesthesia using isoflurane, followed by cardiac perfusion with physiological saline or 4% paraformaldehyde (PFA). Next, we collected 0.5 cm of spinal cord tissue centered around the injury epicenter. Tissues designated for Western blotting (WB) analysis were immediately stored at −80 °C. For frozen sectioning, the tissues were fixed at room temperature for 4 h, then dehydrated in PBS containing 30% sucrose. After dehydration, the tissues were embedded in OCT compound (Aqlabtech, Beijing, China), and 10 μm longitudinal sections were prepared. The selected specimens were stored at −20 °C for short-term preservation.

Hematoxylin–eosin (HE) and Nissl staining

At 28 days after SCI, spinal cord tissues were processed for HE and Nissl staining. For HE staining, frozen spinal cord sections were thawed at room temperature. Following the manufacturer's guidelines (Solarbio, Beijing, China), The sections were incubated in hematoxylin solution to stain the cell nuclei blue-purple. After differentiation, the sections were transferred to eosin solution to stain the cytoplasm red. Dehydration and clearing were performed using xylene and alcohol, respectively, followed by mounting with neutral resin. The specimens were then observed under a microscope. For Nissl staining, spinal cord sections were thawed as described above, stained with Nissl staining (Servicebio, Wuhan, China) solution, differentiated with glacial acetic acid, air-dried, and then mounted with neutral resin before being observed under a microscope.

Western blotting

WB was performed as our previous studies described [22]. The protein expression levels of NLRP3, GSDMD, N-GSDMD, Caspase-1, CL-caspase-1, IL-1β, PI3K, P-PI3K, AKT, and P-AKT were measured in BV2 cells. Total protein was extracted and then quantified using the bicinchoninic acid method. In vivo, the same proteins were extracted from the spinal cord tissue. GAPDH served as the internal control. The relevant antibodies are listed in Table 1, BeyoECL Plus (Beyotime, Beijing, China) was used for developing immunoblots. Protein bands were visualized using the BLT GelView 6000Plus Intelligent Imaging Workstation (Guangzhou Biolytiger Biotechnology Co., Ltd., China), and the intensities were analyzed using Image J (National Institutes of Health, USA).

Table 1 Antibody information

Immunofluorescence (IF) staining

Spinal cord tissues were collected at 3and 28 days post-injury for immunofluorescence staining. Initially, the frozen sections were thawed at room temperature and then washed in PBS. Briefly, after thawing and washing the frozen sections, they were incubated with 0.5% Triton X-100 and 2% bovine serum albumin (BSA) solution at room temperature for 1 h to permeabilize and block the tissue. Primary antibodies were then applied and incubated overnight at 4 °C. Afterward, fluorescently labeled secondary antibodies were added. The specific antibodies used are listed in Table 1. Following the incubation, the sections were washed and stained with diaminidine phenyl indole (DAPI)for nuclear staining. Images were captured using a laser confocal microscope (Nikon, Tokyo, Japan), and fluorescence intensity was analyzed using ImageJ.

For cellular experiments, BV2 cells were seeded into confocal culture dish. After treatment, Cells were fixed with 4% PFA for 15 min at room temperature. Following this, cells were permeabilized and blocked using a solution containing 0.1% Triton X-100 and 0.1% BSA in PBS. Subsequently, cells were incubated with primary and secondary antibodies before being observed under a confocal microscope.

Statistical analysis

Data are expressed as means ± standard deviation (S.D.). Statistical differences between multiple groups were assessed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc tests. The BMS score and inclined plane results were analyzed using repeated measures and two-way ANOVA followed by Tukey's post hoc test to compare group differences on the same day. All statistical analyses were conducted using GraphPad Prism (version 9.3) (GraphPad Software, Inc., USA). P < 0.05 was considered statistically significant.

Results

Isolation and identification of human DPSCs and DPSC-CM

We successfully isolated and passaged human DPSCs (Fig. 1A, B). Flow cytometry analysis was performed to detect the expression of cell surface antigens. The results revealed that DPSCs expressed stem cell-associated markers CD73 and CD90, while they did not express hematopoietic cell markers CD34 and CD45 (Fig. 1C, D), which conforms to the established criteria for DPSC identification. In our previous studies, we collected DPSC-CM for mass spectrometry analysis [22]. Compared to Control-CM, a total of 158 DEPs were identified in DPSC-CM. Further GO analysis revealed that the top three biological processes enriched were signaling receptor activator activity, receptor-ligand activity and cytokine receptor binding (Fig. 1G). KEGG pathway analysis showed that the MAPK and PI3K signaling pathways were the two most enriched pathways in DPSC-CM (Fig. 1F). The PI3K signaling pathway, which is crucial for cell survival, growth, and metabolism [27, 28], was particularly notable. The most enriched proteins within the PI3K signaling pathway were VEGF, ANG-1 and HGF. Additionally, our ELISA results confirmed that the VEGF content in DPSC-CM was significantly higher than in Con-CM (Fig. 1E).

Fig. 1

Characterization of human DPSCs, CM extraction, and mass spectrometry analysis of DPSC-CM. A Schematic Diagram illustrating the extraction process of human DPSC-CM. B Representative Image of human DPSCs under microscopic examination; Scale bars: 250 μm. C, D Flow cytometry analysis of DPSCs surface markers CD90, CD73, and negative markers CD45, CD34. E Detection of VEGF expression in Control-CM and DPSC-CM groups by ELISA. F Kyoto Encyclopedia of Genes and Genomes pathway analysis. G Gene Ontology molecular function analysis. (DPSC: dental pulp stem cell; CM: conditioned medium; VEGF: vascular endothelial growth factor; ELISA: Enzyme-linked immunosorbent assay; DPSC-CM: dental pulp stem cell -derived conditioned medium.)

VEGF derived from DPSC-CM inhibits LPS-induced neuroinflammation and pyroptosis in BV2 microglial in vitro.

LPS-treated BV2 microglial were used to model microglial pyroptosis in vitro. CCK-8 assays demonstrated that DPSC-CM significantly improved cell viability, which was reduced by LPS stimulation. However, this protective effect was reversed upon treatment with BEV (Fig. 2A), indicating that VEGF derived from DPSC-CM plays a key role in mitigating microglial pyroptosis.

Fig. 2

VEGF derived from human DPSCs alleviates LPS-induced pyroptosis in BV2 microglial in vitro. A CCK-8 assay to assess BV2 microglial cell viability. B Representative co-immunofluorescence images of NLRP3 (green, Alexa Fluor® 488) and IBA-1 (red, rhodamine) in LPS-stimulated pyroptosis of BV-2 microglial following treatment with DPSC-CM and BEV; n = 3; Scale bars: 50 μm. C Quantitative analysis of NLRP3 immunofluorescence intensity. D Representative co-immunofluorescence images of GSDMD (green, Alexa Fluor® 488) and IBA-1 (red, rhodamine) in LPS-stimulated pyroptosis of BV-2 microglial following treatment with DPSC-CM and BEV; n = 3; Scale bars: 50 μm. E Quantitative analysis of GSDMD immunofluorescence intensity. F Representative co-immunofluorescence images of IL-1β (green, Alexa Fluor® 488) and IBA-1 (red, rhodamine) in LPS-stimulated pyroptosis of BV-2 microglial following treatment with DPSC-CM and BEV; n = 3; Scale bars: 50 μm. G Quantitative analysis of IL-1β immunofluorescence intensity. H Representative western blot images of NLRP3, GSDMD, N-GSDMD, CL-Caspase-1, and IL-1β protein expression in LPS-stimulated pyroptosis of BV-2 microglial treated with DPSC-CM and BEV; n = 3; GAPDH was used as the control. I-M Quantitative analysis of NLRP3, GSDMD, N-GSDMD, CL-Caspase-1, and IL-1β protein expression levels. Data are shown as means ± SD, Statistical significance was assessed using one-way ANOVA followed by Tukey’s post hoc tests. n = 3; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (CCK-8: Cell counting kit-8; LPS: lipopolysaccharide; BEV: bevacizumab; DPSC-CM: dental pulp stem cell -derived conditioned medium.)

Immunofluorescence staining revealed that the fluorescence intensities of NLRP3, GSDMD, and IL-1β in IBA-1-positive cells were significantly reduced in the DPSC-CM group compared to the LPS group. After BEV administration, the fluorescence intensities of NLRP3, GSDMD, and IL-1β increased (Fig. 2B–G).

Western blotting revealed that LPS stimulation upregulated the expression of pyroptosis-related proteins (NLRP3, CL-caspase-1, N-GSDMD, IL-1β) and while downregulating GSDMD expression. Treatment with DPSC-CM significantly mitigated the LPS-induced increase in these proteins and reversed the reduction in GSDMD expression, the effects of DPSC-CM were reversed by BEV treatment (Fig. 2H–M). Collectively, these results suggest that VEGF derived from DPSC-CM effectively inhibits neuroinflammation and pyroptosis in BV-2 microglial in vitro.

VEGF derived from DPSC-CM alleviates pathological damage and enhances functional recovery in SCI mice

The experimental design is outlined in Fig. 3A. To evaluate the recovery of motor function after SCI, behavioral and electrophysiological assessments were performed. The results demonstrated that the BMS and sub-BMS scores significantly improved after DPSC-CM treatment, compared to the SCI group. However, these effects were suppressed upon BEV administration (Fig. 3B, C). The results from the inclined plane test further supported these findings, with the DPSC-CM-treated mice showing a significantly higher maximum incline angle, whereas BEV-treated mice displayed a reduced maximum angle (Fig. 3D).

Fig. 3

The effects of VEGF derived from human DPSCs and BEV on functional recovery and tissue repair in SCI mice. A Schematic diagram of the in vivo experimental design. B-C BMS scores and sub-scores were used to evaluate motor function recovery in each group of mice at 1, 3, 5, 7, 14, 21, and 28 days; n = 4. D Inclined plane test was used to assess motor function recovery in each group of mice at 1, 7, 14, 21, and 28 days; n = 4. E Nissl staining images of the spinal cord surrounding the injury core region obtained at 28 days; n = 3; Scale bars: 100 μm and 20 μm. F Quantitative analysis of Nissl body count. G Representative MEP in each group of mice at 28 days post-injury; n = 3. H Representative HE staining images of the spinal cord surrounding the injury core region at 28 days; n = 3; Scale bars: 1 mm and 250 μm. I Quantification of the cavity area. (J-K) Statistical analysis of the peak and amplitude of MEP. Data are shown as means ± SD, B-D Statistical significance was assessed using repeated measures and two-way ANOVA. F, I-K Statistical significance was assessed using one-way ANOVA followed by Tukey’s post hoc tests. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (VEGF: vascular endothelial growth factor; DPSC: dental pulp stem cell; BEV: bevacizumab; SCI: Spinal cord injury; BMS: Basso Mouse Scale; MEP: motor evoked potentials; HE: Hematoxylin–eosin.)

We also monitored the MEP and SEP in mice 28 days after SCI. Briefly, the amplitudes of both MEP and SEP were significantly reduced, and latencies were notably prolonged after SCI. Following DPSC-CM treatment, the amplitudes increased and the latencies shortened. Compared to the DPSC-CM group, BEV treatment led to a further reduction in the peak amplitudes of both MEP and SEP, with increased latencies (Fig. 3G, J–K, Fig. S1 A, B-C). These findings suggest that VEGF derived from DPSC-CM can improve both motor and sensory functional recovery in mice after SCI.

HE and Nissl staining were conducted to assess the extent of SCI 28 days post-injury. Notably, the SCI group showed significant tissue loss and neuronal damage. In contrast, the DPSC-CM treatment group exhibited improved tissue integrity and reduced neuronal damage. However, these beneficial effects were diminished in the BEV-treated group (Fig. 3E–F, H–I), suggesting that VEGF derived from DPSC-CM may mitigate tissue loss and neuronal injury following SCI.

VEGF derived from DPSC-CM inhibits microglial pyroptosis 3 days after SCI

We then investigated whether VEGF derived from DPSC-CM could alleviate microglia pyroptosis after SCI using IF. The microglial biomarker IBA-1 was co-stained with NLRP3 at 3 days post-injury. The results indicated that DPSC-CM treatment reduced the fluorescence intensity of NLRP3 in IBA-1 positive cells when compared to the SCI group. However, after adding BEV, the fluorescence intensity of NLRP3 increased compared to the DPSC-CM group (Fig. 4A, E). Similar changes were observed for caspase-1, GSDMD, and IL-1β (Fig. 4B-D, F–H). These findings indicate that VEGF derived from DPSC-CM can inhibit microglial pyroptosis after SCI by suppressing the NLRP3/caspase-1/IL-1β pathway.

Fig. 4

The effect of VEGF derived from human DPSCs and BEV on microglial pyroptosis at 3 Days after SCI. A-D Representative co-immunofluorescence images showing the expression levels of NLRP3, GSDMD, Caspase-1, IL-1β (green, Alexa Fluor® 488), and microglia (IBA-1, red, rhodamine) in each group of mice at 3 days after SCI; n = 3; Scale bars: 100 μm and 50 μm. E–H Quantitative statistical analysis of NLRP3, GSDMD, Caspase-1, IL-1β immunofluorescence intensity. Data are shown as means ± SD, Statistical significance was assessed using one-way ANOVA followed by Tukey’s post hoc tests. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (VEGF: vascular endothelial growth factor; DPSC: dental pulp stem cell; BEV: bevacizumab; SCI: Spinal cord injury.)

VEGF derived from DPSC-CM promotes neural repair and reduces glial fibrillary acidic protein expression following SCI

At 28 days after SCI, IF staining was performed using beta III tubulin (Tuj1) and glial fibrillary acidic protein (GFAP) as neurons and astrocytes markers, respectively, alongside IBA-1 to assess microglial activation. The results showed a significant downregulation of Tuj1 expression and an upregulation of GFAP in the SCI group. However, treatment with DPSC-CM led to an upregulation of Tuj1 expression and a downregulation of GFAP expression. In the DPSC-CM combined with BEV treatment group, a trend similar to the SCI group was observed (Fig. 5A–D). These findings indicate that VEGF derived from DPSC-CM alleviates neuronal damage, promotes neural repair, reduces astrocyte activation, and suppresses glial scar formation following SCI.

Fig. 5

The effect of VEGF derived from human DPSCs and BEV on neural repair and scar formation at 28 days after SCI. A Representative co-immunofluorescence images showing the expression of GFAP (green, Alexa Fluor® 488) and microglia marker IBA-1 (red, rhodamine) at 28 days post-injury; n = 3; Scale bars: 1000 μm and 50 μm. B Quantitative analysis of GFAP immunoreactivity. C Representative co-immunofluorescence images of Tuj1 (green, Alexa Fluor® 488) and microglia (IBA-1, red, rhodamine) at 28 days post-injury; n = 3; Scale bars: 1000 μm and 50 μm. D Quantitative analysis of Tuj1 immunoreactivity. Data are shown as means ± SD, Statistical significance was assessed using one-way ANOVA followed by Tukey’s post hoc tests. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (VEGF: vascular endothelial growth factor; DPSC: dental pulp stem cell; BEV: bevacizumab; SCI: Spinal cord injury; GFAP: glial fibrillary acidic protein; Tuj1: beta III tubulin.)

VEGF derived from DPSC-CM alleviates neuronal damage and promotes angiogenesis

At 28 days post-SCI, IF staining was also performed to assess the expression of neurofilament-200 (NF200), myelin basic protein (MBP), microtubule-associated protein 2 (MAP2), and platelet endothelial cell adhesion molecule-1 (PECAM-1/CD31) in the spinal cord. Compared to the SCI group, the DPSC-CM treatment group exhibited significantly higher fluorescence intensities of NF200, MAP2, and MBP. However, these intensities were reduced when DPSC-CM treatment was combined with BEV (Fig. 6 A-C, E–G). CD31, a well-established marker for endothelial cells, is widely used to assess angiogenesis and plays a critical role in SCI repair [29, 30]. Notably, the fluorescence intensity of CD31 was significantly increased in the DPSC-CM-treated group compared to both the SCI and DPSC-CM + BEV groups (Fig. 6 D, H). These results indicate that VEGF derived from DPSC-CM contributes to alleviating neuronal damage, reducing myelin loss, and the promotion of angiogenesis, thereby enhancing SCI repair.

Fig. 6

The effects of VEGF derived from human DPSCs and BEV on neuronal survival, myelin regeneration, and angiogenesis in SCI mice at 28 days. A–D Representative co-immunofluorescence images of NF200, MAP2, MBP and CD31 (green, Alexa Fluor® 488) and microglia (IBA-1, red, rhodamine) at 28 days post-injury; n = 3; Scale bars: 100 μm and 50 μm. E–H Quantitative analysis of NF200, MAP2, MBP and CD31 immunoreactivity. Data are shown as means ± SD, Statistical significance was assessed using one-way ANOVA followed by Tukey’s post hoc tests. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (VEGF: vascular endothelial growth factor; DPSC: dental pulp stem cell; BEV: bevacizumab; SCI: Spinal cord injury; NF200: neurofilament-200; MBP: myelin basic protein; MAP2: microtubule-associated protein 2; PECAM-1/CD31: platelet endothelial cell adhesion molecule-1.)

VEGF derived from DPSC-CM alleviates microglial pyroptosis after SCI by regulating the NLRP3/GSDMD/Caspase-1 system via activation of the PI3K-AKT signaling pathway

WB analysis was performed to examine the expression of NLRP3, GSDMD, caspase-1, CL-casepase-1 and N-GSDMD proteins at 3 days after SCI. Compared to the DPSC-CM group, the levels of NLRP3, CL-caspase-1, and N-GSDMD were significantly elevated in both the SCI and DPSC-CM + BEV groups, while the levels of pro-caspase-1 and GSDMD decreased. Importantly, treatment with DPSC-CM alone reversed these changes (Fig. 7A, D–H).

Fig. 7

The effect of VEGF derived from DPSCs and BEV on pyroptosis in SCI and activation of the PI3K-AKT signaling pathway. A Western blot analysis of NLRP3, GSDMD, N-GSDMD, Caspase-1 and CL-Caspase-1 expression in the different groups 3 days after SCI; n = 3. B Western blot analysis performed for P-PI3K, PI3K, P-AKT and AKT expression in BV2 microglia treated with DPSC-CM and BEV; n = 3. C Western blot analysis of P-PI3K, PI3K, P-AKT and AKT expression in spinal cord obtained on 3 days post-injury; n = 3. D–H Quantitative analysis of NLRP3, GSDMD, N-GSDMD, Caspase-1, CL-Caspase-1 protein expression levels. I-L Quantification of the ratio of P-PI3K/PI3K and P-AKT /AKT. Data are shown as means ± SD, Statistical significance was assessed using one-way ANOVA followed by Tukey’s post hoc tests. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (VEGF: vascular endothelial growth factor; DPSC: dental pulp stem cell; BEV: bevacizumab; SCI: Spinal cord injury.)

To further elucidate the role of the PI3K-AKT signaling pathway in microglial pyroptosis, we assessed the expression of key signaling proteins both in vitro and in vivo using WB. In vitro, compared to the control group, the P-PI3K/PI3K and p-AKT/AKT ratios were significantly decreased in LPS group, indicating that LPS suppressed the phosphorylation of PI3K and AKT. Notably, the P-PI3K/PI3K and p-AKT/AKT ratios were significantly elevated in the DPSC-CM group compared to the LPS group, suggesting that DPSC-CM treatment effectively alleviated the LPS-induced suppression of phosphorylation. However, after DPSC-CM + BEV treatment, the P-PI3K/PI3K and p-AKT/AKT ratios were significantly reduced compared to the DPSC-CM group, indicating that BEV reversed the protective effects of DPSC-CM (Fig. 7B, I, J). A similar trend was observed in vivo after SCI (Fig. 7C, K, L). These results suggest that VEGF derived from DPSC-CM can alleviate microglial pyroptosis by activating the PI3K-AKT signaling pathway, thereby reducing the severity of SCI.

Discussion

Our results provide compelling evidence that human DPSC-derived VEGF significantly reduces microglial pyroptosis through the PI3K/AKT signaling pathway both in vitro and vivo, contributing to functional recovery in SCI mice. This study underscores the therapeutic promise of DPSCs and their secreted factors, particularly VEGF, in enhancing SCI recovery through modulation of neuroinflammation and promotion of neuroprotection.

In our previous study, we conducted mass spectrometry analysis on the human DPSC-CM and the Control-CM [22]. The analysis revealed 158 proteins with statistically significant differences between DPSC-CM and Control-CM. KEGG pathway analysis identified the PI3K/AKT signaling pathway as one of the most enriched signaling pathway among the DEPs, with VEGF being the most abundant protein within this pathway. ELISA experiments further confirmed the high abundance of VEGF.

The PI3K/AKT signaling pathway is widely recognized as a key regulator of cellular survival, growth, and metabolic processes, and its activation is critical for neuroprotection following SCI [27, 28]. Recent studies have underscored the importance of the PI3K/AKT pathway in inhibiting microglia pyroptosis [12, 13]. Our study supports this notion by showing that VEGF activates the PI3K/AKT pathway to exert its anti-pyroptotic effects on microglia after SCI. This finding is in line with other reports that demonstrate the protective effects of activating PI3K/AKT signaling in SCI [13, 31, 32]. The therapeutic potential of activating this pathway is well-documented, and our results suggest that VEGF, through its activation of PI3K/AKT signaling, may significantly contribute to enhancing neuroprotection and promoting functional recovery following SCI.

VEGF has long been recognized for its critical role in promoting angiogenesis and tissue repair, particularly in ischemic and inflammatory conditions. In the context of SCI, VEGF exerts multifaceted effects, including enhancing vascular integrity, reducing secondary tissue damage, activating neural stem cells, and facilitating neurovascular regeneration [23, 33]. Our findings further extend these observations, showing that VEGF secreted by DPSCs is crucial in regulating in the inflammatory response specifically by inhibiting microglial pyroptosis, thereby promoting angiogenesis and reducing neuroinflammation.

Recent studies have further supported the notion that VEGF's neuroprotective effects are not limited to angiogenesis but also involve direct interactions with glial cells and neurons to mitigate neuroinflammation and promote tissue regeneration [34, 35]. In this study, we demonstrated that VEGF derived from DPSC-CM could inhibit microglial pyroptosis and subsequently promote SCI repair. This finding identifies VEGF as a key therapeutic component within DPSC-CM, providing further theoretical support for its clinical translational research.

Given the complexity of human DPSC-CM's composition and the lack of prior studies identifying its primary reparative factors, our research is innovative in utilizing proteomic analysis to pinpoint VEGF through its enrichment in the PI3K/AKT signaling pathway and subsequently confirming its functional role. Our results add to this body of evidence, highlighting VEGF as a key mediator of the anti-pyroptotic effects observed in DPSC-CM, which may contribute to the observed improvement in motor function and electrophysiological recovery.

Microglial pyroptosis, a type of programmed cell death that leads to the release of pro-inflammatory cytokines, plays a important role in exacerbating neuroinflammation and tissue damage after SCI [36, 37]. Our findings align with the growing body of literature that identified pyroptosis as a key driver of secondary injury following SCI. The NLRP3 inflammasome, activated in microglia, is central to pyroptotic cell death. Inhibition of this process has been shown to reduce neuroinflammation and promote functional recovery in SCI models [13, 37, 38]. In our study, VEGF derived from DPSC-CM significantly decreased pyroptosis-related markers (e.g., caspase-1, IL-1β), and inhibited microglial pyroptosis after SCI via the PI3K-AKT pathway.

The findings of this study could provide valuable insights into the development of stem cell-based therapeutic strategies for SCI. DPSCs, which have demonstrated therapeutic promise in preclinical investigations, represent a valuable source of secreted factors capable of modulating the inflammatory microenvironment and promoting tissue repair. Our results support the therapeutic potential of DPSC-derived CM in treating SCI, with particular emphasis on its influence on VEGF and the PI3K/AKT signaling pathway. The ability of DPSC-CM to alleviate microglial pyroptosis and promote functional recovery provides a strong rationale for further exploration of DPSCs and their secreted factors in clinical applications.

Additionally, our study shows the potential of VEGF as a key therapeutic factor in SCI repair. While VEGF’s role in angiogenesis is well-established, its neuroprotective effects, particularly in modulating neuroinflammation and glial cell activity, warrant further investigation. Future studies exploring the optimal delivery methods of VEGF or DPSC-CM, as well as their long-term efficacy, will be crucial in translating these findings into clinical therapies.

While this study offers valuable insights into the mechanisms underlying DPSC-mediated SCI repair, several limitations warrant consideration. First, the use of VEGF-neutralizing antibodies, such as bevacizumab, to block VEGF activity may have introduced potential off-target effects that could influence data interpretation. Future studies using more specific VEGF inhibitors or gene-editing techniques could help further elucidate the role of VEGF in SCI repair. Additionally, although we have demonstrated the effects of DPSC-derived VEGF in vitro and vivo, the long-term effects of DPSC-based therapies on SCI recovery remain to be explored. Investigating the functional outcomes over extended periods and assessing the potential for chronic side effects will be essential in determining the clinical feasibility of this approach.

Conclusion

The current study provides compelling evidence that VEGF secreted by human DPSCs plays a pivotal role in alleviating microglial pyroptosis through the PI3K/AKT signaling pathway, thereby contributing to the repair process following SCI. Our findings suggest that DPSC-derived CM, through its secretion of VEGF, holds significant therapeutic potential for enhancing functional recovery following SCI. These results not only advance our understanding of the molecular mechanisms underlying SCI repair but also support the development of stem cell-based therapies targeting neuroinflammation and neuroprotection.