Dental pulp stem cells alleviate Schwann cell pyroptosis via mitochondrial transfer to enhance facial nerve regeneration

2.1. Establishment of a facial nerve transection model in crab-eating macaques

The commercial Neurolac® nerve guides and hDPSCs were obtained from Polyganics B.V., Groningen, The Netherlands, and Youtooth, Shanghai, China, respectively. The main component of Neurolac® nerve guides is poly (DL-lactide-co-caprolactone), which was manufactured by ring-opening polymerization of ε-caprolactone and lactide. All experiments related to the crab-eating macaque model were approved by the Ethics Research Committee of the Shanghai Ninth People's Hospital (Approval Nos. 2021-A030-01 and PL21-0613). Twelve adult macaques were randomly divided into three groups: Neurolac®, autologous nerve transplantation (Autograft), and DPSCs@Neurolac®. Facial nerves were surgically excised, creating a 5 mm defect, followed by the implantation of either a nerve conduit or autologous graft. DPSCs were cultured and injected into the proximal part of the nerve conduit at a density of 7.5 × 10⁷ cells/mL. After 12 months, the animals were sacrificed, and facial nerve tissues were harvested for H&E staining, immunohistochemistry (IHC), and TEM to assess myelination and axonal regeneration. Myoelectrical activity was recorded using an electrophysiological system (RM6240, China). A 10 mV electrical stimulus was applied to the gastrocnemius muscle proximal to the injured nerve, and compound muscle action potentials (CMAPs) were recorded and analyzed.

2.2. Establishment of a rat FNI model

All rat experiments were conducted in accordance with the guidelines approved by the Institutional Animal Care and Use Committee (IACUC) of Wuhan Beisai Model Biotechnology (Approval Nos. 2024-04-12D and 2024-09-04A). Sixteen male Sprague-Dawley (SD) rats, weighing 200–250 g, were randomly allocated into two groups: the FNI group and the DPSC-treated group. Randomization was performed using a random number table, and standardized surgical and evaluation protocols were followed to ensure experimental consistency and minimize bias. In the FNI group, a sham operation was performed on the right side, while the left side underwent facial nerve injury. In the DPSC-treated group, bilateral facial nerve injury was induced, with the right side treated with human DPSCs and the left side receiving TNFα-preconditioned DPSCs (20 ng/mL for 12 h). Both sides of each rat were observed and evaluated independently.

The FNI model was established under general anesthesia using isoflurane to minimize pain and stress. The facial nerve was exposed via a postauricular incision under sterile conditions, and uniform pressure was applied to the facial nerve trunk using hemostatic forceps for 45 s. Immediately following the injury, a 10 μL suspension of DPSCs (5 × 10⁵ cells) was injected at the injury site in the DPSC-treated group.

2.3. Facial nerve functional assessment

Facial nerve function was evaluated on postoperative days 3, 7, 14, 21, and 42 using three key indicators: blink reflex, whisker movement, and facial symmetry. A total of 12 rats were used in the study, with assessments conducted on specific sides of their faces, evaluated in three groups. Scoring was performed by three independent evaluators. The blink reflex was assessed by blowing air toward the eye, with eyelid closure scored from 0 (normal) to 3 (no response). Whisker movement was observed in a resting state and scored from 0 (normal) to 3 (no movement). Facial symmetry was visually assessed and scored from 0 (symmetric) to 2 (severe asymmetry). A double-blind design was employed, ensuring that both evaluators and handlers were blinded to group assignments, and an independent third party conducted the functional assessments to minimize bias.

2.4. Isolation and culture of DPSCs

All third molars were obtained as discarded biological samples from Renmin Hospital of Wuhan University, with approval from the Institutional Ethics Medical Committee. DPSCs were isolated from dental pulp tissues and digested with 3 mg/mL collagenase type I and 4 mg/mL dispase at 37 °C for 1 h. The resulting cell suspension was centrifuged and resuspended in high-glucose DMEM supplemented with 10 % FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. Cells were cultured at 37 °C in a humidified atmosphere of 5 % CO₂ and passaged at 80–90 % confluency using 0.25 % trypsin-EDTA. Cells from passages 3 to 6 were used for experiments. For differentiation, DPSCs were cultured in high-glucose DMEM and induced with OriCell® osteogenic, adipogenic, and chondrogenic differentiation kits (HUXXC-90021, HUXXC-90041, HUXXC-90031), following the manufacturer's protocols.

2.5. Transmission electron microscopy (TEM)

Fourteen days after the facial nerve injury, the facial nerve tissues were harvested and fixed in 2.5 % glutaraldehyde at 4 °C. The samples were then post-fixed with osmium tetroxide, dehydrated in ethanol, and embedded in epoxy resin. Ultrathin sections (70–90 nm) were stained with uranyl acetate and lead citrate. Myelin sheath structures were visualized using a transmission electron microscope (TEM), and myelin thickness, along with the G-ratio (axon diameter/total fiber diameter), was analyzed using MyelTracer software to assess remyelination (https://doi.org/10.1523/ENEURO.0558-20.2021).

2.6. Histological staining experiment

Facial nerve tissues were fixed in 4 % paraformaldehyde, paraffin-embedded, and sectioned at 5 μm thickness. H&E staining was performed following standard procedures. Antigen retrieval was carried out using citrate buffer (pH 6.0) at high temperature, followed by overnight incubation at 4 °C with primary antibodies against Ki67 (bs-23103R, Bioss), GSDMD (15146-1-AP, Proteintech), S100 (15146-1-AP, Proteintech), and MPZ (10572-1-AP, Proteintech). Fluorescently labeled secondary antibodies were then applied to the sections, and fluorescence microscopy (IX83, Olympus) was used to capture and observe fluorescence. Immunohistochemistry (IHC) staining was performed using an immunohistochemistry kit (Maixin, China), and the analysis of Ki67 staining was based on nuclear localization using ImageJ software. To detect mitochondrial ROS levels, fresh rat facial nerve tissues were collected and incubated with 10 μM MitoSox for 30 min. The tissues were cryopreserved using an optimal cutting temperature (OCT) compound and cryosectioned into thin slices. The tissue sections were counterstained with DAPI to visualize nuclei, and fluorescence microscopy (IX83, Olympus) was employed to observe protein expression and mitochondrial ROS levels. Fluorescence intensity was quantified using ImageJ software.

2.7. Cell culture and co-culture systems

RSC96 cells were cultured in high-glucose DMEM supplemented with 10 % fetal bovine serum (FBS). Pyroptosis was induced by treating RSC96 cells with 1 μg/mL LPS (HY-D1056, MCE) for 24 h, followed by exposure to 2 mM ATP (HY-B2176, MCE) for an additional 24 h, resulting in the generation of LA-RSC96 cells (LPS- and ATP-treated RSC96 cells). Preliminary experiments involved varying the concentration of LPS and the duration of induction. To mitigate mitochondrial ROS accumulation during pyroptosis induction, cells were pre-treated with 10 μM Mito-TEMPO (HY-112879, MCE) for 4 h. To impede TNTs formation, cells were pre-treated with 2 μM Cytochalasin B (HY-16928, MCE) for 4 h. To inhibit NF-κB signaling pathway, cells were pre-treated with 1 μM Bay 11–7082 (HY-13453, MCE) for 4 h. To block TNFα in the transwell system, TNFα neutralizing antibodies (AF-510-SP, R&D Systems) were added at a concentration of 10 μg/mL in cell culture. All in vitro experiments were performed in triplicate, with at least three independent repeats.

Two co-culture systems were established: direct contact and transwell. In the transwell system, DPSCs were seeded in the upper chamber, while RSC96 cells were placed in the lower chamber to allow indirect interaction. For the direct contact co-culture, RSC96 cells and DPSCs were mixed at a 7:3 ratio in a six-well plate. In the LA-RSC96+DPSC co-culture, LPS-pretreated RSC96 cells were co-cultured with DPSCs in ATP-containing medium for 24 h, ensuring that DPSCs were only exposed to ATP without prior contact with LPS. To generate ρ⁰-DPSCs (mitochondrial DNA-deficient), DPSCs were cultured for one month in medium containing 1 μg/mL ethidium bromide (HY-D0021, MCE), 100 μg/mL pyruvic acid (HY-Y0781, MCE), and 50 μg/mL uridine (HY-B1449, MCE). For TNFα or Metformin preconditioning, DPSCs were treated with human TNFα (HY-P70426A, MCE) or Metformin (HY-B0627, MCE) for 24 h. Fluorescence-activated cell sorting (FACS) was performed using a FACSAria III cell sorter (BD Biosciences) to isolate RSC96 cells from the co-culture systems.

To visualize mitochondrial transfer, DPSCs were transfected with pLV2-CMV-Mito-mCherry-Puro lentivirus for mitochondrial tracking, while RSC96 cells were labeled with CFSE. To visualize the cytoskeleton, we used either iFluor™ 405 phalloidin (40783ES75, Yeasen) or phalloidin-488 (G1248-100T, Servicebio) to stain F-actin. Cell nuclei were stained with DAPI. Mitochondrial transfer was analyzed using a confocal microscope (FV1200, Olympus), and observations were conducted using a fluorescence microscope (IX83, Olympus).

2.8. Quantitative PCR (qPCR) and RNA sequencing (RNA-seq)

Total RNA was isolated from tissues or cells using Trizol (15596018, Invitrogen) following the manufacturer's protocol. Genomic DNA contamination was removed, and the RNA was reverse transcribed into cDNA using a reverse transcription kit (R323, Vazyme). The qPCR reaction was prepared with ChamQ SYBR qPCR Master Mix (Q311-02, Vazyme) and conducted on the CFX96 Real-Time System (Bio-Rad). Rat Rplp0 was utilized as the internal control, and relative expression levels were calculated using the 2^⁻ΔΔCt method. Primer sequences are provided in Table S1. For RNA sequencing, samples were sent to Wuhan Benagen Technology Service Co., Ltd. for Illumina sequencing. Subsequently, Gene Ontology (GO) and KEGG pathway analyses were conducted on the DAVID website (http://david.abcc.ncifcrf.gov/). RNA-seq data are presented in Table S2.

2.9. Western blot (WB) analysis

Total cellular and tissue proteins were extracted using lysis buffers appropriate for preserving protein integrity, and concentrations were determined using the bicinchoninic acid (BCA) assay. Proteins were separated by SDS-PAGE and transferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were probed with primary antibodies against β-actin (66009-1-Ig, Proteintech) as a loading control, Caspase-1 (ab179515, Abcam), GSDMD (ab219800, Abcam), NLRP3 (68102-1-Ig, Proteintech), IL1β (A16288, ABclonal), COX4 (11242-1-AP, Proteintech), TFAM (23996-1-AP, Proteintech), COX2 (55070-1-AP, Proteintech), ND1 (19703-1-AP, Proteintech), TOM20 (sc-17764, Santa Cruz), TNFα (AF-510-SP, R&D Systems), Phospho-IKKα/β (ab194528, Abcam), IKKβ (A19606, ABclonal), Phospho-IKBα (2859, CST), IKBα (4814, CST), Phospho-p65 (3033, CST), p65 (10745-1-AP, Proteintech), PGC-1α (66369-1-Ig, Proteintech), CAT (sc-271803, Santa Cruz), SOD1 (sc-101523, Santa Cruz), Parkin (sc-32282, Santa Cruz), and PINK1 (sc-518052, Santa Cruz). After secondary antibody incubation, protein bands were visualized using enhanced chemiluminescence (ECL) and quantified through densitometric analysis using ImageJ software.

2.10. Flow cytometry analysis

Flow cytometry was conducted to analyze stem cell markers using antibodies against CD146 (PE-65181, Proteintech), CD44 (11-0441-81, Invitrogen), CD90 (APC-65088, Proteintech), CD34 (550761, BD Biosciences), CD45 (11-0459-41, Invitrogen), and CD105 (561443, BD Biosciences). Cells were incubated with these antibodies at 4 °C for 30 min, followed by washing and analysis on a Beckman Coulter CytoFLEX flow cytometer. Mitochondrial ROS production and membrane potential were assessed using MitoSox (M36008, Invitrogen) and TMRM (T668, Invitrogen), respectively. After staining, cells were analyzed by flow cytometry to quantify changes in mitochondrial function. To evaluate mitochondrial transfer between DPSCs and RSC96 cells, DPSCs were labeled with MitoBright LT Deep Red (MT12, Dojindo), while RSC96 cells were labeled with CFSE. The percentage of mitochondrial transfer was calculated by measuring fluorescent events in CFSE-positive cells, using the gating strategy for CFSE-labeled RSC96 cells in cocultured cells and mitochondrial transfer rate analysis (Fig. S4M). Flow cytometry data analysis and quantification were performed using FlowJo software.

2.11. Enzyme-linked immunosorbent assay (ELISA), ATP and caspase-1 activity assays

The level of TNFα in cell culture supernatant was measured with a rat TNFα ELISA kit (KE20018, Proteintech) according to the manufacturer's instructions. ATP levels were determined with an ATP assay kit (S0026, Beyotime), and caspase-1 activity was evaluated with a caspase-1 activity assay kit (C1101, Beyotime), following the manufacturer's instructions. Both ATP and caspase-1 levels were quantified through luminescence or absorbance measurements and compared across experimental groups in a microplate spectrophotometer (Bio-Rad).

2.12. Statistical analysis

All data are presented as mean ± standard deviation (SD). Statistical significance between two groups was determined using a two-tailed Student's t-test, while comparisons among multiple groups were performed using one-way ANOVA. A p-value of <0.05 was considered statistically significant.

3.1. DPSCs promote facial nerve regeneration in non-human primates

We first established a crab-eating macaque model of facial nerve transection to evaluate the therapeutic potential of DPSCs in nerve regeneration. The surgical procedure for facial nerve injury in non-human primates is detailed in Fig. 1A. The facial nerve was excised and repaired using either a Neurolac® conduit (Neurolac®), an autologous nerve graft (autograft), or a Neurolac® conduit with DPSC injection at the proximal site (DPSCs@Neurolac®), as shown in Fig. 1B. Neurolac® and autograft were selected as controls to provide comparisons with a clinically used synthetic nerve conduit and the gold standard in nerve repair, respectively. H&E staining demonstrated an increased cell number in the DPSCs@Neurolac® group compared to both the Neurolac® and autograft groups (Fig. 1C and D). Ki67 immunohistochemical analysis further confirmed significantly enhanced proliferative activity in the DPSCs@Neurolac® group relative to the Neurolac® and autograft groups (Fig. 1E and F). TEM further showed increased myelin sheath and axon thickness, along with a reduced G-ratio, in the DPSCs@Neurolac® group compared to the Neurolac® and autograft groups, indicating improved nerve regeneration (Fig. 1G and H). At 12 months post-surgery, electrophysiological tests revealed that no CMAP was detected on the injured side in the Neurolac® group, while the DPSCs@Neurolac® group exhibited significantly higher CMAP amplitude and conduction velocity compared to the Neurolac® group, though lower than the autograft group (Fig. S1). These results suggest that DPSC transplantation significantly outperforms the Neurolac® conduit in enhancing nerve survival, myelin regeneration, and functional recovery, achieving outcomes comparable to those of autologous nerve grafts.

3.2. DPSCs alleviate Schwann cell pyroptosis in a rat FNI model

To elucidate the pathological factors underlying nerve regeneration and investigate the therapeutic effects of DPSCs on Schwann cell pyroptosis following peripheral nerve injury, we established a rat FNI crush model and a DPSC transplantation-based rescue model (Fig. 2A; Figs. S2A–C). DPSC treatment significantly accelerated facial nerve functional recovery compared to the injury group (Fig. 2B). TEM analysis at two weeks post-injury demonstrated increased myelin sheath thickness in the DPSC-treated group (Fig. 2C and D). To assess the potential role of Schwann cell pyroptosis in impaired nerve regeneration, we examined pyroptosis-related protein expression in Schwann cells four days post-injury. Immunofluorescence (IF) staining showed increased expression of the pyroptosis marker GSDMD and a corresponding reduction in Schwann cell markers S100 and MPZ after injury, with DPSC treatment effectively mitigating these changes (Fig. 2E–G). Furthermore, mitochondrial ROS levels were significantly reduced in the DPSC-treated group, as shown by MitoSox staining (Fig. 2E–H), suggesting that DPSCs alleviate oxidative stress in the injured nerve. WB analysis further confirmed the downregulation of pyroptosis markers GSDMD, cleaved-Caspase1, and IL1β in the DPSC-treated group (Fig. 2I and J). Following FNI, mRNA levels of markers for pyroptosis, ferroptosis, and necroptosis were all upregulated to varying degrees, with pyroptosis-related markers exhibiting the most pronounced fold changes. Notably, DPSC treatment significantly attenuated the upregulation of these markers, with the greatest reduction observed in pyroptosis-related genes (Figs. S2D–G). TEM revealed typical morphological changes of pyroptosis in Schwann cells, including significant cell swelling, loss of membrane integrity, cytoplasmic content leakage and bubble-like structures (Fig. S2H).

3.3. DPSCs alleviate mitochondrial damage in pyroptosis-induced Schwann cells

To further investigate the mechanisms by which DPSCs mitigate Schwann cell pyroptosis, we developed an in vitro model using LPS and ATP to induce pyroptosis in RSC96 Schwann cells (LA-RSC96 cells) (Figs. S3A, B, E). Pyroptosis markers were significantly upregulated in cells treated with 1 μg/mL LPS and 2 mM ATP. To determine whether mitochondrial oxidative stress contributes to Schwann cell pyroptosis, we treated RSC96 cells with MitoTEMPO, a mitochondrial-targeted antioxidant, to inhibit mitochondrial ROS production. The results showed that MitoTEMPO alleviated pyroptosis and restored mitochondrial homeostasis, as evidenced by the suppression of MitoSox upregulation and the prevention of mitochondrial membrane potential reduction in LA-RSC96 cells (Figs. S3C and D; Figs. S4A–H). Next, we co-cultured DPSCs with pyroptosis-induced Schwann cells using both direct contact and transwell systems (Fig. S3E). We hypothesized that DPSCs inhibit pyroptosis by reducing mitochondrial oxidative damage in LA-RSC96 cells. Similar to MitoTEMPO treatment, co-culture significantly reduced mitochondrial ROS production, membrane potential loss, and pyroptosis marker expression, underscoring the importance of direct cell-cell interaction (Fig. S3F-L; Fig. S4I-M). Additionally, changes in caspase-1 activity over time demonstrated that the presence of DPSCs began to exert effects at 8 h, with a significant suppression of caspase-1 upregulation observed at 16 h (Fig. S3N). WB and qPCR analyses confirmed that mitochondrial biogenesis and function-related markers were significantly restored in Schwann cells after co-culture with DPSCs (Fig. S3M-N; Fig. S4O-R).

3.4. Mitochondrial transfer from DPSCs to pyroptosis-induced Schwann cells

Mitochondrial transfer can restore damaged cell homeostasis by supplying functional mitochondria through various mechanisms such as TNTs or extracellular vesicles [29,32]. To explore whether DPSCs restore mitochondrial function in LA-RSC96 cells via this mechanism, we labeled DPSCs with Mito-mCherry to trace mitochondrial transfer and co-cultured them with CFSE-labeled RSC96 cells (Fig. 3A). Confocal microscopy and flow cytometry confirmed mitochondrial transfer between DPSCs and Schwann cells in the direct contact model, while no transfer was observed in the transwell system (Fig. 3B–D, Fig. S5A). F-actin staining indicated that mitochondrial transfer occurred through TNTs (Fig. 3C). Moreover, the transfer rate increased over time and was significantly higher in LA-RSC96 cells compared to non-pyroptosis-induced cells (Fig. 3D–G). Notably, the mitochondrial transfer rate significantly increased after 8 h, coinciding with caspase-1 activity suppression, indicating a potential link between mitochondrial transfer and pyroptosis attenuation. Notably, ATP alone did not induce mitochondrial transfer (Figs. S5B–C).

3.5. Mitochondrial transfer from DPSCs alleviates mitochondrial damage in pyroptosis-induced Schwann cells

Given that DPSCs exhibit mitochondrial transfer capacity more efficiently to pyroptosis-induced Schwann cells, we hypothesized that this transfer alleviates mitochondrial damage in LA-RSC96 cells. To test this hypothesis, we generated mitochondrial DNA-deficient DPSCs (ρ⁰-DPSCs) and confirmed mitochondrial DNA deletion (Figs. S5D–G). Despite intact mitochondrial transfer rates (Figs. S5H–I), ρ⁰-DPSCs failed to rescue mitochondrial dysfunction (Fig. 4A–D) and pyroptosis in Schwann cells, as indicated by increased caspase-1 activity and pyroptosis marker expression (Fig. 4E–G, Fig. S5J). This suggests that the protective effects of DPSCs on LA-RSC96 cells are dependent on the functional integrity of transferred mitochondria. Furthermore, we used Cytochalasin B (CB) to disrupt the structure of actin filaments and block the formation of TNTs (Fig. 4H). This inhibition of TNT formation impaired mitochondrial transfer (Fig. 4I and J) and reduced the protective effects of DPSCs, as shown by enhanced mitochondrial ROS production, membrane potential loss, and increased pyroptosis marker expression (Fig. 4I-Q, Fig. S5K).

3.6. TNFα signaling activates NF-κB in DPSCs to promote mitochondrial transfer

Our observation that pyroptosis induction in Schwann cells increases mitochondrial transfer inspired us to hypothesize that pyroptosis-induced RSC96 cells enhance mitochondrial transfer by secreting signaling molecules. RNA sequencing and Gene Ontology (GO) analysis comparing LA-RSC96 cells and RSC96 cells revealed enrichment in cytokine production processes, suggesting a paracrine mechanism (Fig. 5A). KEGG pathway analysis further highlighted TNFα signaling as a key pathway activated in pyroptosis-induced Schwann cells (Fig. 5B),which is known to activate NF-κB in donor cells and promote mitochondrial transfer [30]. To investigate how Schwann cell-derived TNFα affects DPSCs, ELISA, WB and qPCR were performed to confirm the elevated expression of TNFα in pyroptosis-induced RSC96 cells (Figs. S6A–C). Inhibition of TNFα signaling with neutralizing antibodies (a-TNFα) blocked NF-κB activation, as indicated by reduced phosphorylation of P65 and Ikkα/β, and inhibited TNT formation in DPSCs, demonstrating the paracrine role of TNFα (Fig. 5C and D; Fig. S6D). Furthermore, we used the NF-κB pathway inhibitor Bay 11–7082 (Bay) to suppress NF-κB signaling (Figs. S6F and G), which impaired the ability of DPSCs to alleviate the expression of pyroptosis-related markers in pyroptosis-induced RSC96 cells (Fig. 5E and F). Inhibition of NF-κB signaling in DPSCs also reduced their protective effects on LA-RSC96 cells and diminished the occurrence of mitochondrial transfer (Fig. 5G–J; Fig. S6E, H-J, M). Additionally, exogenous TNFα treatment of DPSCs enhanced mitochondrial transfer and NF-κB activation (Fig. 5K-M; Figs. S6K and L). These findings underscore the critical role of TNFα in enhancing mitochondrial transfer from DPSCs to Schwann cells, facilitating DPSC-mediated neuroprotection.

3.7. TNFα enhances mitochondrial adaptive stress response in DPSCs

We further hypothesized that TNFα not only promotes mitochondrial transfer but also primes transferred mitochondria to better restore mitochondrial homeostasis in recipient cells. TNFα treatment in DPSCs significantly increased mitochondrial ROS levels and ATP production in DPSCs while lowering mitochondrial membrane potential, indicating enhanced mitochondrial stress response (Fig. 6A–E). WB and qPCR analyses revealed increased expression of mitochondrial biogenesis markers (PGC1α, NRF1, TFAM) and mitochondrial function-related proteins (COX4, TOM20, mt-Co2) in TNFα-treated DPSCs, indicating enhanced mitochondrial capacity in DPSCs (Fig. 6F–H). Notably, TNFα-treated DPSCs exhibited enhanced antioxidant capacity, as shown by increased expression of CAT and SOD1, as well as upregulated autophagy markers Parkin and Pink1 (Fig. 6I-L). Metformin was included as a positive control to evaluate protein level changes due to its role in alleviating mitochondrial dysfunction under stress conditions [33]. TNFα preconditioning also improved the protective effects of DPSCs on LA-RSC96 cells, as evidenced by enhanced mitochondrial function (Fig. 6M-Q; S7A-D), reduced pyroptosis (Fig. 6R-U), and increased antioxidant activity in both DPSCs and recipient Schwann cells (Figs. S7C and D). These results further support the notion that TNFα preconditioning enhances DPSC mitochondrial function and protective capacity, setting the stage for evaluating its therapeutic efficacy in nerve injury models.

3.8. TNFα preconditioning improves the therapeutic efficacy of DPSCs in repairing facial nerve injury

To validate the therapeutic potential of TNFα preconditioning in vivo, we transplanted TNFα-preconditioned DPSCs into the FNI model. TNFα-preconditioned DPSCs significantly accelerated facial nerve functional recovery compared to untreated DPSCs (Fig. 7A). TEM analysis revealed thicker myelin sheaths in the TNFα-preconditioned group (Fig. 7B and C). WB and IF staining further confirmed reduced pyroptosis marker expression and improved Schwann cell survival in the preconditioned group (Fig. 7D–H). These findings suggest that TNFα preconditioning enhances DPSC-mediated peripheral nerve repair by promoting mitochondrial transfer and restoring Schwann cell mitochondrial homeostasis.