Carnitine palmitoyltransferase 1A facilitates the senescence of human dental pulp stem cells by Parkin succinylation-mediated mitophagy

Abstract

BACKGROUND

Senescence of human dental pulp stem cells (DPSCs) weakens the differentiation and regeneration ability, affecting their function and the tooth, and limiting their in vitro application. Carnitine palmitoyltransferase 1A (CPT1A) has been reported to regulate the aging of multiple cells.

AIM

To investigate the effects of CPT1A on DPSC senescence and the underlying mechanism, particularly PTEN-induced kinase 1-mediated mitophagy.

METHODS

Human DPSCs at seven (p7) and 15 (p15) passages were used for functional analysis. Cell senescence and mitophagy were evaluated using senescence associated β-galactosidase staining and immunoblotting. The effect of CPT1A on Parkin succinylation was assessed using co-immunoprecipitation (co-IP)/IP, immunoblotting, and cycloheximide chase experiment. The aging mouse model was established using D-galactose (D-gal), and both senescence and mitophagy were evaluated in dental pulp tissues.

RESULTS

CPT1A expression was increased in DPSCs at p15. Compared with p7 cells, p15 cells had higher senescence and lower mitophagy levels, while knockdown of CPT1A inhibited senescence and promoted mitophagy of p15 DPSCs. Additionally, silencing CPT1A reduced the succinylation of Parkin at K27 site and enhanced Parkin protein stability. CPT1A succinylates Parkin and suppresses mitophagy through its lysine succinyltransferase activity. Moreover, the mitophagy inhibitor Mdivi-1 or knockdown of Parkin abrogated the effects on senescence and mitophagy mediated by CPT1A knockdown. The in vivo experiments showed that silencing CPT1A inhibits D-gal-induced dental pulp aging by promoting Parkin-mediated mitophagy.

CONCLUSION

Silencing of CPT1A suppresses senescence of DPSCs by facilitating mitophagy, which is associated with the inhibition of succinylation of Parkin, suggesting a new mechanism and target for DPSC aging.

Key Words: Dental pulp stem cells; Senescence; Carnitine palmitoyltransferase 1A; Succinylation; Mitophagy; Parkin

Core Tip: This study reveals that carnitine palmitoyltransferase 1A (CPT1A) promotes senescence of human dental pulp stem cells (DPSCs) by inhibiting Parkin-mediated mitophagy. CPT1A is upregulated in senescent late-passage DPSCs, and its knockdown suppresses senescence while enhancing mitophagy. Mechanistically, CPT1A acts as a succinyltransferase that succinylates Parkin protein at lysine 27, reducing Parkin stability. Silencing CPT1A decreases Parkin succinylation, stabilizes Parkin, and restores mitophagy. In vivo, CPT1A knockdown alleviates D galactose-induced dental pulp aging via Parkin-dependent mitophagy. This research identifies a novel CPT1A-Parkin succinylation axis controlling DPSC senescence.

INTRODUCTION

Dental pulp stem cells (DPSCs) have garnered significant interest due to ease of harvesting from dental pulp and non-invasive collection from discarded teeth[1]. As a population of mesenchymal stem cells, DPSCs exhibit high self-renewal capacity, multi-lineage differentiation potential, and the ability to generate dentin/pulp-like complexes[2,3]. These properties make them promising candidates for regenerative dentistry, particularly tooth repair and regeneration[4,5]. However, the regenerative potential of DPSCs declines with senescence[6], a state characterized by irreversible cell cycle arrest and loss of cellular function[7]. Hence, elucidating the molecular mechanisms driving senescence is essential for improving the therapeutic utility of DPSCs.

Succinylation is a key post-translational modification which is implicated in cellular processes such as senescence[8]. It involves the attachment of succinyl groups to lysine residues by succinyltransferases and their removal by desuccinylases[9]. Among the enzymes regulating succinylation, carnitine palmitoyltransferase 1A (CPT1A) has been identified as a succinyltransferase. CPT1A stands out as a rate-limiting enzyme located on the outer mitochondrial membrane that catalyzes the conversion of long-chain fatty acids into acyl-carnitines, the critical first step in mitochondrial fatty acid β-oxidation[10]. Moreover, CPT1A plays a pivotal role in cellular energy metabolism, and acts as a potential therapeutic target for various diseases[11,12]. Interestingly, recent studies have suggested that CPT1A plays a critical role in cellular senescence. For example, CPT1A regulates tumor cell senescence by influencing lipid accumulation related to mitochondrial function[13]. In addition, silencing or inhibiting CPT1A promotes endothelial cell senescence by impairing fatty acid metabolism, which may thereby facilitate the development of cardiovascular diseases[14]. The reduction of mitochondrial fatty acid oxidation is closely related to stem cell aging[15]; thus, we speculate that CPT1A may play an important role in regulating stem cell aging. However, whether CPT1A can influence DPSC senescence is still unclear.

Mitochondrial dysfunction contributes to the cellular aging process[16]. Mitophagy, the selective autophagic clearance of mitochondria, serves as an essential quality control mechanism that maintains mitochondrial and intracellular homeostasis by eliminating damaged or dysfunctional organelles[17,18]. Several studies have reported that impaired mitophagy is linked to the senescence of cells[19]. The PTEN-induced kinase 1 (PINK1)/Parkin pathway plays a key role in regulating mitophagy. Parkin, an E3 ubiquitin ligase, is recruited to damaged mitochondria to target them for degradation[20]. However, the role of Parkin in DPSC senescence remains unclear.

The interplay between CPT1A, succinylation, and mitophagy in the context of DPSC senescence is poorly understood. Therefore, this study aimed to investigate how CPT1A-mediated succinylation of Parkin influences mitophagy and senescence in DPSCs. Our findings provide a theoretical foundation for identifying novel therapeutic targets to counteract DPSC senescence.

MATERIALS AND METHODS

Culture of human DPSCs

DPSCs were purchased from Lonza (Basel, Switzerland). The cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 mg/mL streptomycin (all purchased from Gibco, Grand Island, NY, United States) at 37 °C with 5% CO₂ and 95% air.

Alkaline phosphatase staining

For osteogenic differentiation, DPSCs were seeded into 24-well plates and cultured in osteogenic induction medium containing 100 nM dexamethasone, 10 mmol/L β-glycerol phosphate, and 50 μM vitamin C. After 2 weeks, the cells were incubated with the stain working solution from the alkaline phosphatase (ALP) stain kit (Yeasen, Shanghai, China) for 30 minutes in the dark. After washing with distilled water, the staining results were observed under a microscope.

Alizarin Red S staining

DPSCs were cultured in osteogenic induction medium for 3 weeks. The cells were fixed with 10% formaldehyde for 15 minutes and stained with 1% pH = 4.2 Alizarin Red S (ARS) solution (Solarbio, Beijing, China) for 15 minutes. After washing with distilled water, the results were observed under a microscope.

Oil Red O staining

For adipogenic differentiation, DPSCs were seeded into 24-well plates and cultured in complete medium supplemented with 1 μM dexamethasone, 10 μg/mL insulin, 200 μM indomethacin, and 0.5 mmol/L isobutyl-methylxanthine. After 3 weeks, lipid droplets were evaluated using an Oil Red O stain kit (Solarbio, Beijing, China). The cells were fixed with the specific fixative for 30 minutes, rinsed with 60% isopropanol for 30 seconds, and stained with Oil Red O solution for 20 minutes. Following a second rinse with 60% isopropanol, nuclei were counterstained with hematoxylin for 1 minute. Lipid droplets were visualized under a microscope.

Flow cytometry

The levels of stem cell surface markers were detected as previously described[21]. DPSCs were collected and washed with phosphate buffered saline (PBS). The cells were incubated at 4 °C for 30 minutes in the dark with the following fluorescently conjugated antibodies: FITC-conjugated mouse anti-human CD90 (Cat# 389803, Biolegend, San Diego, CA, United States), PE-conjugated mouse anti-human CD105 (Cat# 323205, Biolegend, San Diego, CA, United States), APC-conjugated mouse anti-human CD146 (Cat# 323207, Biolegend, San Diego, CA, United States), PE-conjugated mouse anti-human CD34 (Cat# 343505, Biolegend, San Diego, CA, United States), and FITC-conjugated mouse anti-human CD45 (Cat# 304005, Biolegend, San Diego, CA, United States). Antibodies including FITC, PE, or APC-conjugated mouse IgG1, κ (Cat# 981802, 981804, or 981806) served as the isotype controls. The cells were then washed and flow cytometry was performed using a CytoFLEX LX flow cytometer (Beckman Coulter, Miami, FL, United States). A minimum of 10000 observations were acquired per sample. Data were analyzed using FlowJo software.

Cell passage

DPSCs were digested with 0.05% trypsin-ethylene diamine tetraacetic acid solution (Sigma-Aldrich, MO, United States) at 37 °C until they detached from the culture surface. After adding complete medium to stop digestion, the samples were centrifuged at 500 × g for 6 minutes. The sediments were resuspended in the medium, and the cells were seeded in a T25 culture flask. The cells at seven (p7) and fifteen (p15) passages were collected.

Cell transfection

DPSCs at p15 were seeded in six-well plates and cultured to > 80% confluence. Subsequently, the cells were transfected with short hairpin RNAs (shCPT1A, shParkin, and their negative control shNC; Invitrogen, Carlsbad, CA, United States), CPT1A overexpression plasmid [oe-CPT1A, CPT1A coding sequence was amplified and cloned into pcDNA3.1(+) vector], and its negative control (oe-NC) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, United States) for 48 hours.

HEK293T cells (ATCC, Manassas, VA, United States) were cultured in Roswell Park Memorial Institute-1640 medium (Gibco, Grand Island, NY, United States) supplemented with 10% FBS, 100 U/mL penicillin, and 100 mg/mL streptomycin at 37 °C with 5% CO₂. Wild type (WT) Parkin plasmids and mutated plasmids at the Lys (K)27 site of Parkin [K27R, K27R mutant was generated using the pcDNA3.1(+) vector via site-directed mutagenesis] were constructed. HEK293T cells or DPSCs were transfected with these plasmids using Lipofectamine 2000 for 48 hours.

To clarify that the regulation of Parkin’s succinyl modification by CPT1A depends on its succinyltransferase activity, CPT1A-G710E [carnitine palmitoyltransferase (CPTase)-inactive but not inactive lysine succinyltransferase (LSTase)] and CPT1A-H473A (LSTase-inactive) mutants were generated using pcDNA3.1-CPT1A-WT as a template. These plasmids were transfected into DPSCs using Lipofectamine 2000.

Cell treatment

To inhibit mitophagy, DPSCs were treated with 20 μM mitochondrial division inhibitor 1 (Mdivi-1; Sigma-Aldrich, MO, United States) for 2 hours prior to transfection.

Senescence associated β-galactosidase staining

The cell senescence β-galactosidase staining kit (Yeasen, Shanghai, China) was used to evaluate DPSC senescence. The cells were washed with PBS and fixed with 1 mL β-gal staining fixative at room temperature for 10 minutes. Subsequently, the cells were incubated with 1 mL pre-heated stain working solution overnight at 37 °C. The cells were observed and counted under a light microscope. Senescence associated β-galactosidase (SA-β-gal) stained cells were counted under the microscope.

Quantitative real-time polymerase chain reaction

Total RNA was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA, United States) according to standard protocols. After detecting RNA concentration, 1 μg RNA was reverse transcribed to cDNA using the HiScript II 1^st Strand cDNA synthesis kit (Vazyme, Nanjing, Jiangsu Province, China). Subsequently, RNA expression was measured using the Taq Pro Universal SYBR qPCR Master Mix (Vazyme, Nanjing, Jiangsu Province, China) and calculated using the 2^-∆∆Ct method. β-actin was included as the internal control.

Immunoblotting

Total proteins were extracted from DPSCs or dental pulp tissues from mice using radio-immunoprecipitation assay buffer, and the concentrations were determined using a bicinchoninic acid kit (Yeasen, Shanghai, China). The protein samples (30 μg) were used for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride membranes (Vazyme, Nanjing, Jiangsu Province, China). The membranes were probed with primary antibodies at 4 °C overnight. The following day, after washing the membranes with Tris Buffered Saline and Tween 20 (TBST), the membranes were probed with the secondary antibody at room temperature for 1 hour. The membranes were washed with TBST again and the bands were developed using SuperPico electrochemiluminescence Master Mix (Vazyme, Nanjing, Jiangsu Province, China). The primary antibodies used here included anti-CPT1A (#12252), anti-β-actin (#4967), anti-p53 (#9282), anti-p21 (#2947), anti-p16 (#18769), anti-PINK1 (#6946), anti-Parkin (#2132), anti-LC3B (#2775), anti-hemagglutinin (i.e., HA-tag, #3724), and anti-polyhistidine (i.e., His-tag, #2365). The secondary antibody was anti-rabbit IgG, HRP-linked antibody (#7074). All antibodies were purchased from Cell Signaling Technology (Danvers, MA, United States).

Transmission electron microscopy

Transmission electron microscopy (TEM) was performed as previously described[22]. Briefly, DPSCs were fixed with 2.5% glutaraldehyde at room temperature for 2 hours. The cells were stained with 1% osmium tetroxide and dehydrated in gradient ethanol. The samples were embedded in epoxy resin. After double-staining with uranyl acetate and lead citrate, the samples were observed under a JEM-1200EX transmission electron microscope (JEOL, Tokyo, Japan). The number of autophagic vacuoles per cell was quantified.

Immunofluorescence staining

DPSCs were stained with 100 nM MitoTracker Red FM (Yeasen, Shanghai, China) and 50 nM LysoTracker Green DND-26 (Yeasen, Shanghai, China) at 37 °C for 30 minutes. The staining solution was replaced by fresh cell culture medium. The fluorescence was immediately observed using a confocal laser scanning microscope.

Co-immunoprecipitation and immunoprecipitation

The protein A/G magnetic IP/co-IP kit (Vazyme, Shanghai, China) was used for these experiments. For co-immunoprecipitation (co-IP), HEK293T cells were washed with PBS and lysed using lysis buffer (20 μL lysis buffer/10⁵ cells) on ice for 10 minutes. The lysate was collected after centrifuging at 13800 × g for 5 minutes at 4 °C. Approximately 1 mg of lysate was incubated with 10 μg of antibodies against HA (#3724), His (#2365), or IgG (#2729) at room temperature for 30 minutes to form the antigen-antibody immune complex. The complex was added to protein A/G beads and incubated at room temperature for 30 minutes. After washing with wash buffer, the beads were resuspended in SDS-PAGE loading buffer, and the sample was heated at 95 °C for 5 minutes to denature the proteins. Then, SDS-PAGE was performed and protein levels were measured using immunoblotting. For IP, DPSCs or HEK293T cells were lysed, and the antibody against succinyllysine (PTM-419, PTMBio, Hangzhou, Zhejiang Province, China) was used.

Bioinformatic analysis

The succinylation sites in Parkin were predicted using the GPSuc online tool (http://kurata14.bio.kyutech.ac.jp/GPSuc/index.php).

Cycloheximide chase experiment

DPSCs were treated with 30 μg/mL cycloheximide (Sigma-Aldrich, MO, United States) for 0, 8, 16, and 24 hours. The protein levels of Parkin were examined using immunoblotting.

Proximity ligation assay

The interaction between endogenous CPT1A and Parkin was detected using the Duolink In Situ Detection Reagents (Sigma-Aldrich, MO, United States). DPSCs (p7, p15, or p15 treated with 10 μmol/L CCCP for 2 hours) were seeded on coverslips, fixed with 4% paraformaldehyde for 20 minutes, permeabilized with 0.2% Triton X-100 for 10 minutes, and blocked with blocking solution for 2 hours at 37 °C. The cells were then incubated overnight at 4 °C with primary antibodies against CPT1A and Parkin. After washing, proximity ligation assay (PLA) probes were applied and incubated for 1 hour at 37 °C. Images were acquired using a confocal laser scanning microscope.

Animal study

Male C57BL/6 mice (8 weeks old), weighing 23-25 g, were purchased from Hangzhou Qizhen Laboratory Animal Technology Co., Ltd (Hangzhou, Zhejiang Province, China). After one week of acclimatization in a specific pathogen-free environment at 22 ± 2 °C with a 12-hour light/dark cycle and 45%-55% humidity, the mice were randomly divided into four groups: Control, D-galactose (D-gal), D-gal + sh-NC lentivirus (Lv-sh-NC), and D-gal + sh-CPT1A lentivirus (Lv-sh-CPT1A), with six mice per group. To generate the aging model, mice received a subcutaneous injection of 150 mg/kg D-gal (Sigma-Aldrich, MO, United States) on the back, once a day for 6 consecutive weeks. Mice in the control group were subcutaneously injected with the same volume of normal saline at the same site, same time and frequency as the model group. Lentivirus (1 × 10⁸ PFU/mL) was injected into mice via the tail vein on the day before modeling, once a week for 6 consecutive weeks. After modeling was completed, the mice were euthanized by inhalation of isoflurane. The dental pulp tissues were isolated for further use. The animal study was approved by the Experimental Animal Welfare Ethics Committee of Beijing MDKN (Approval No. MDKN-2024-270).

Histological analysis

The dental pulp tissue was fixed in 4% paraformaldehyde for 24 hours, and then embedded in paraffin. Paraffin sections with a thickness of 5 μm were prepared. The sections were stained with hematoxylin and eosin (H&E) to evaluate pathology. Additionally, antigen retrieval was performed on the sections using 0.01 M sodium citrate buffer solution. After blocking using 5% bovine serum albumin (Sigma-Aldrich, MO, United States), the sections were stained with CPT1A (#97361, Cell Signaling Technology, MA, United States) or Parkin (ab15494, Abcam, Cambridge, MA, United States) at 4 °C overnight, and stained with the secondary antibody (#7074, Cell Signaling Technology, MA, United States) at 37 °C for 30 minutes. After development using the DAB solution (Yeasen, Shanghai, China), the sections were visualized under a light microscope.

Statistical analysis

Three biological replicates and three technical replicates were performed in the in vitro experiments. The animal study was repeated in six mice per group. Data were analyzed using GraphPad Prism 8 and presented as means ± SD. The Student’s t-test or ANOVA followed by Tukey’s post hoc test were used to evaluate differences. A P value < 0.05 was considered statistically significant.

RESULTS

Characteristics of DPSCs

To characterize the DPSCs used in this study, we first assessed their multipotent differentiation capacity and surface marker expression. We found that the cells could differentiate into osteoblasts, which was analyzed by ALP and ARS staining (Figure 1A and B). Additionally, we observed obvious lipids in the cells after adipogenic differentiation (Figure 1C). Flow cytometric analysis indicated that the cells were positive for CD90, CD105, and CD146, and negative for CD34 and CD45 (Figure 1D). These results confirm that these cells have multilineage differentiation potential and express stem cell surface markers, suggesting that DPSCs were successfully obtained.

Figure 1 Characteristics of dental pulp stem cells by alkaline phosphatase and Alizarin Red S staining. A: Image of alkaline phosphatase; B: Alizarin Red S staining of dental pulp stem cells (DPSCs) after osteogenic differentiation induction; C: Image of Oil Red O staining of DPSCs after adipogenic differentiation induction; D: Flow cytometry analysis of stem cell surface markers, including CD90, CD105, CD146, CD34, and CD45. n = 3/group.

CPT1A is highly expressed in DPSCs at p15

Given that replicative senescence occurs during prolonged in vitro culture, we compared CPT1A expression in DPSCs at p7 and p15. It was shown that the mRNA and protein levels of CPT1A were higher in the p15 group than in the p7 group (Figure 2). These findings suggest that CPT1A may be a regulator of DPSC senescence.

Figure 2 Carnitine palmitoyltransferase 1A is highly expressed in dental pulp stem cells at passage 15. A: The expression of carnitine palmitoyltransferase 1A (CPT1A) was detected in dental pulp stem cells at passage 7 (p7) and p15 using quantitative real-time polymerase chain reaction; B and C: The protein levels of CPT1A were measured using immunoblotting, and the results were quantified. n = 3/group; P < 0.001 indicates statistically significant differences. CPT1A: Carnitine palmitoyltransferase 1A.

Knockdown of CPT1A suppresses senescence and promotes mitophagy

To investigate the functional role of CPT1A in senescence, we knocked down CPT1A expression in DPSCs using shCPT1A. According to the quantitative real-time polymerase chain reaction (qPCR) results, CPT1A expression was decreased in the shCPT1A group, compared with the shNC group (Figure 3A). Next, senescence was evaluated using SA-β-gal staining, and the results indicated a higher proportion of SA-β-gal-positive cells in p15 DPSCs than in p7 DPSCs. Moreover, knockdown of CPT1A reduced the number of SA-β-gal-positive cells in p15 cells (Figure 3B and C). The levels of senescence-related proteins, including p53, p21, and p16, were enhanced in p15 DPSCs, compared with p7 DPSCs; however, CPT1A silencing decreased their levels in p15 cells (Figure 3D and E). These data suggest that silencing CPT1A inhibits replicative senescence. Accumulating evidence has indicated that mitophagy is closely associated with cellular senescence[23,24]. Therefore, we analyzed whether CPT1A regulates mitophagy. The TEM results showed that the number of mitochondria in the cytoplasm was increased, while the number of mitochondria encapsulated by autophagosomes was decreased in the p15 group, suggesting mitophagy was inhibited, compared with the p7 group. Knockdown of CPT1A promoted mitophagy according to the TEM results (Figure 3F and G). Several mitophagy markers were measured using immunoblotting. The results showed that PINK1 levels, Parkin levels, and LC3BII/I ratio were lower in p15 DPSCs than that in p7 DPSCs. Silencing CPT1A elevated these indicators in p15 DPSCs (Figure 3H and I), suggesting that knockdown of CPT1A promotes mitophagy. Moreover, p15 DPSCs showed a lower co-localization of lysosome with mitochondria than that in p7 DPSCs, while CPT1A knockdown facilitated co-localization (Figure 3J). Taken together, these findings indicate that CPT1A knockdown contributes to inhibiting DPSC senescence and promoting mitophagy.

Figure 3 Knockdown of carnitine palmitoyltransferase 1A suppresses senescence and promotes mitophagy. A: Dental pulp stem cells (DPSCs) at passage 15 (p15) were transfected with shNC and shCPT1A, and quantitative real-time polymerase chain reaction was conducted to measure CPT1A expression; B and C: After transfection, senescence in DPSCs was observed using senescence associated β-galactosidase staining; D and E: The levels of aging markers (p53, p21, and p16) were measured using immunoblotting; F: Transmission electron microscopy evaluation of mitophagy; G: Quantification of autophagic vacuoles per cell; H and I: The levels of mitophagy markers (PINK1, Parkin, and LC3B) were measured using immunoblotting; J: Immunofluorescent assay was conducted to visualize mitochondria (red dots) and lysosomes (green dots). n = 3/group; P < 0.001 indicates statistically significant differences. CPT1A: Carnitine palmitoyltransferase 1A; SA-β-gal: Senescence associated β-galactosidase.

Overexpression of CPT1A promotes DPSC senescence and inhibits mitophagy

To investigate the role of CPT1A, we constructed oe-CPT1A which was transfected into p7 DPSCs. As shown in Supplementary Figure 1A, CPT1A expression was elevated in the oe-CPT1A group. Cell functions were then analyzed. The results showed that overexpression of CPT1A increased the percentage of SA-β-gal stained cells (Supplementary Figure 1B and C), and concurrently induced the upregulation of p53, p21, and p16 expression (Supplementary Figure 1D and E). Additionally, TEM images showed that CPT1A reduced the number of mitochondria wrapped by autophagosomes (Supplementary Figure 1F and G). PINK1 levels, Parkin levels, and LC3BII/I ratio were decreased after CPT1A overexpression (Supplementary Figure 1H and I). The co-localization of lysosome with mitochondria was inhibited by overexpressing CPT1A in p7 DPSCs (Supplementary Figure 1J). Taken together, these findings indicate that CPT1A accelerates the senescence of p7 DPSCs and inhibits mitophagy.

Mitophagy is a necessary process for CPT1A in regulating DPSC senescence

To verify whether mitophagy participates in cellular senescence, we used Mdivi-1 to treat p15 DPSCs to inhibit mitophagy. The results showed that Mdivi-1 counteracted the reduction of SA-β-gal positive staining cells and the decreased expression of aging markers mediated by CPT1A knockdown (Figure 4A and B). The levels of p53, p21, and p16 were increased by Mdivi-1 in CPT1A knocked down cells (Figure 4C and D). Additionally, Mdivi-1 inhibited mitophagy, downregulated the expression of PINK1, Parkin, and the LC3BII/I ratio, and hindered the co-localization of lysosome with mitochondria induced by CPT1A silencing in p15 cells (Figure 4E-I). These results indicate that silencing CPT1A inhibits replicative senescence of DPSCs by facilitating mitophagy.

Figure 4 Mitophagy is a necessary process for carnitine palmitoyltransferase 1A to regulate dental pulp stem cell senescence. A and B: Dental pulp stem cells (DPSCs) at passage 15 (p15) were transfected with shNC and shCPT1A and treated with mitophagy inhibitor Mdivi-1, senescence in DPSCs was observed using senescence associated β-galactosidase staining; C and D: The levels of aging markers (p53, p21, and p16) were measured using immunoblotting; E: Transmission electron microscopy evaluation of mitophagy; F: Quantification of autophagic vacuoles per cell; G and H: The levels of mitophagy markers (PINK1, Parkin, and LC3B) were measured using immunoblotting; I: Immunofluorescent assay was conducted to visualize mitochondria (red dots) and lysosomes (green dots). n = 3/group; P < 0.001 and P < 0.05 indicate statistically significant differences. CPT1A: Carnitine palmitoyltransferase 1A; SA-β-gal: Senescence associated β-galactosidase.

CPT1A is the succinyltransferase for Parkin and promotes its degradation

Given that CPT1A functions as a succinyltransferase, we investigated whether it regulates the succinylation of key mitophagy-related proteins, PINK1 and Parkin. We found that knockdown of CPT1A reduced the succinylation levels of Parkin but had no effect on PINK1 succinylation (Figure 5A and B). We then explored the interaction between CPT1A and Parkin using co-IP. The results showed that Parkin was expressed when IP with HA was used, while CPT1A was expressed when IP with His was used (Figure 5C), suggesting that CPT1A protein interacts with Parkin protein. To identify the specific succinylation site, we used the GPSuc database and predicted lysine 27 (K27) as a potential succinylation site on Parkin (Figure 5D). Mutation at K27 site decreased Parkin succinylation and elevated its protein abundance (Figure 5E and F), indicating that K27 serves as a major site for CPT1A-mediated succinylation. Furthermore, knockdown of CPT1A enhanced Parkin protein stability (Figure 5G and H). In addition, mutation of Parkin at K27 site enhanced the protein stability of Parkin, elevated LC3BII/I ratio, and downregulated TOM20 levels (Supplementary Figure 2A-E). In summary, knockdown of CPT1A stabilizes Parkin protein by suppressing the succinylation of Parkin at K27 site.

Figure 5 Carnitine palmitoyltransferase 1A is the succinyltransferase for Parkin and promotes its degradation. A and B: Dental pulp stem cells (DPSCs) were transfected with shNC and shCPT1A, and the succinylation levels of PINK1 and Parkin were measured using immunoprecipitation (IP) with anti-succinyllysine and immunoblotting with anti-PINK1 and anti-Parkin; C: The interaction between CPT1A and Parkin proteins was evaluated in HEK293T cells using co-IP; D: Potential succinylation sites in Parkin were predicted using the GPSuc database; E and F: After wild type or K27R Parkin plasmids were transfected into HEK293T cells, immunoblotting was performed to measure Parkin protein levels, and Parkin succinylation levels were detected by IP using anti-succinyllysine and immunoblotting using anti-Parkin; G and H: DPSCs were transfected with shNC or shCPT1A and then treated with cycloheximide for 0, 8, 16, and 24 hours, and Parkin protein levels were detected using immunoblotting. n = 3/group; P < 0.001 indicates statistically significant differences. CPT1A: Carnitine palmitoyltransferase 1A; WT: Wild type.

CPT1A physically interacts with Parkin in a mitophagy-activating context

To investigate whether CPT1A and Parkin are in close proximity under conditions relevant to mitophagy, we performed PLA. As shown in Supplementary Figure 3, p7 DPSCs exhibited a low basal level of the PLA signal between CPT1A and Parkin, and the PLA signal was increased in p15 DPSCs. Furthermore, p15 DPSCs were treated with the mitophagy inducer CCCP, and the PLA signal intensity was further enhanced. Thus, the interaction or proximity between CPT1A and Parkin is upregulated during replicative senescence and is further potentiated upon mitophagy activation.

CPT1A succinylates Parkin and suppresses mitophagy through its LSTase activity

To further confirm that the succinylation of Parkin is directly mediated by the LSTase activity of CPT1A rather than its canonical CPTase activity, we employed two well-characterized CPT1A mutants. CPT1A has been reported to possess intrinsic LSTase activity independent of its classical CPTase function. Specifically, the G710E mutation selectively inactivates CPTase activity while preserving LSTase activity, whereas the H473A mutation disrupts the putative binding pocket for the thioester sulfur of acyl-CoA and impairs both CPTase and LSTase activities[25]. The results showed that compared with the WT group, CPT1A-G710E did not affect Parkin succinylation levels, while CPT1A-H473A reduced Parkin succinylation levels (Supplementary Figure 4A and B). We next examined whether the LSTase activity of CPT1A is required for its regulation of mitophagy. As shown in Supplementary Figure 4C-F, CPT1A knockdown increased PINK1, Parkin, and LC3BII/I protein levels, re-expression of CPT1A-WT reversed these changes, and this effect was similar with G710E. In contrast, compared with the WT group, H473A enhanced the levels of mitophagy markers. Collectively, the LSTase activity of CPT1A, which mediates Parkin succinylation, is essential for CPT1A-induced suppression of mitophagy.

Parkin reverses the impacts on senescence and mitophagy caused by CPT1A

To determine whether Parkin is responsible for CPT1A-regulated senescence and mitophagy, we performed rescue experiments. DPSCs were transfected with shParkin and shNC, and it was found that Parkin expression was downregulated in the shParkin group (Figure 6A). The results of SA-β-gal staining showed that knockdown of CPT1A reduced the number of positively stained cells, which was abrogated by knockdown of Parkin (Figure 6B and C). Also, the decreased expression of p53, p21, and p16 caused by CPT1A knockdown was abrogated by silencing Parkin (Figure 6D and E). Additionally, CPT1A knockdown promoted the formation of autophagosomes to facilitate mitophagy, while knockdown of Parkin reversed the mitophagy caused by CPT1A knockdown (Figure 6F and G). Silencing CPT1A enhanced PINK1 and Parkin protein levels, and elevated the LC3BII/I ratio, whereas Parkin knockdown reversed these effects modulated by CPT1A silencing (Figure 6H and I). The co-localization of lysosome with mitochondria promoted by CPT1A knockdown in p15 DPSCs was counteracted by Parkin knockdown (Figure 6J). Taken together, these findings indicate that knockdown of Parkin reverses the inhibition of replicative senescence and the promotion of mitophagy caused by CPT1A knockdown, suggesting that silencing CPT1A inhibits DPSC senescence by promoting Parkin-mediated mitophagy.

Figure 6 Parkin reverses the impacts on senescence and mitophagy caused by carnitine palmitoyltransferase 1A. A: Dental pulp stem cells (DPSCs) at passage 15 (p15) were transfected with shNC and shParkin, and quantitative real-time polymerase chain reaction was conducted to measure Parkin expression; B and C: After transfection with shCPT1A and shParkin, senescence in DPSCs was observed using senescence associated β-galactosidase staining; D and E: The levels of aging markers (p53, p21, and p16) were measured using immunoblotting; F: Transmission electron microscopy evaluation of mitophagy; G: Quantification of autophagic vacuoles per cell; H and I: The levels of mitophagy markers (PINK1, Parkin, and LC3B) were measured using immunoblotting; J: Immunofluorescent assay was conducted to visualize mitochondria (red dots) and lysosomes (green dots). n = 3/group; P < 0.001, P < 0.01, and P < 0.05 indicate statistically significant differences. CPT1A: Carnitine palmitoyltransferase 1A; SA-β-gal: Senescence associated β-galactosidase.

Knockdown of CPT1A attenuates D-gal-induced senescence and facilitates mitophagy

The role of CPT1A in vivo was evaluated. We established an aging mouse model using D-gal, and Lv-sh-CPT1A was injected to interfere with CPT1A expression. We analyzed the data of six mice in each group. CPT1A mRNA and protein levels were upregulated in the D-gal group compared with the control group, while Lv-sh-CPT1A reversed CPT1A expression induced by D-gal (Figure 7A-C). The succinylation levels of Parkin were also elevated after D-gal treatment, which was abrogated by CPT1A knockdown (Figure 7B and D). The protein levels of p53, p21, and p16 were increased in the dental pulp tissues of D-gal-induced mice, and knockdown of CPT1A downregulated their levels (Figure 7E-H). Conversely, the levels of PINK1, Parkin, and the LC3BII/I ratio were reduced in the dental pulp tissues of D-gal-induced mice, which was reversed after CPT1A silencing (Figure 7I-L). H&E staining results showed that D-gal caused changes in dental pulp tissue, mainly manifested as disordered and loose cell arrangement, unclear cell boundaries, and local infiltration by immune cells; however, CPT1A knockdown alleviated these adverse changes in D-gal-induced mice (Figure 7M). Furthermore, D-gal induced the upregulation of CPT1A and downregulation of Parkin in dental pulp tissues, while knockdown of CPT1A abrogated these effects (Figure 7N). In summary, silencing CPT1A inhibits D-gal-induced pulp aging by promoting Parkin-mediated mitophagy.

Figure 7 Knockdown of carnitine palmitoyltransferase 1A attenuates D-gal-induced senescence and facilitates mitophagy. A: Carnitine palmitoyltransferase 1A (CPT1A) expression in the dental pulp tissues of mice was measured by quantitative real-time polymerase chain reaction; B-D: CPT1A protein levels in the dental pulp tissues were measured using immunoblotting, and Parkin succinylation levels were detected by immunoblotting after immunoprecipitation; E-L: The levels of aging markers (p53, p21, and p16) (E-H) and mitophagy markers (PINK1, Parkin, and LC3B) (I-L) in the dental pulp tissues of mice were examined using immunoblotting; M: The pathology of dental pulp tissues was evaluated using hematoxylin and eosin staining assay; N: CPT1A and Parkin levels in the dental pulp tissues were detected using immunohistochemistry. n = 6/group; P < 0.001 indicates statistically significant differences. CPT1A: Carnitine palmitoyltransferase 1A; HE: Hematoxylin and eosin.

DISCUSSION

Our study provides novel insights into the molecular mechanisms underlying senescence in DPSCs, focusing on the interplay between CPT1A, succinylation, Parkin, and mitophagy. Our findings offer a new perspective on the mechanism of DPSC aging and highlight potential therapeutic targets.

DPSCs have attracted wide interest in regenerative dentistry due to their accessibility and multilineage differentiation capacity. However, senescence during in vitro expansion remains a major obstacle that limits their clinical application[26]. Previous studies have begun to elucidate the complex mechanisms involved in DPSC aging. For instance, adrenomedullin overexpression was shown to protect against senescence via the miR-152/CCNA2 axis[27]. In addition, overexpression of matrixmetalloproteinase 3 (MMP3) promotes DPSC aging, while loss of MMP3 function suppresses aging due to expansion in vitro[28]. Environmental and biochemical factors such as hypoxia and pleiotrophin have also been implicated in modulating senescence[29,30]. These studies have revealed that the mechanism of DPSC aging is very complex.

The role of CPT1A in cellular senescence appears context-dependent. Lin et al[14] found that silencing or inhibition of CPT1A facilitates vascular endothelial cell senescence. Conversely, Jiang et al[31] discovered that inhibition of CPT1A attenuates chondrocyte senescence induced by oxidative stress. These studies suggest that the effects of CPT1A on cellular aging are paradoxical and may highly depend on different cell types. In the present study, we found that CPT1A expression was elevated in p15 DPSCs, and knockdown of CPT1A suppressed senescence, suggesting the critical role of CPT1A in the senescence of DPSCs, consistent with findings in placenta-derived mesenchymal stem cells[32], suggesting a cell-type-specific function.

Recent studies have identified that mitophagy, a critical quality-control process, is closely related to cellular senescence. During the aging process, mitochondrial dynamics and mitophagy are impaired, which leads to an increase in the number of damaged or abnormally functioning mitochondria, further contributing to the progression of aging[33]. However, whether mitophagy is related to DPSC senescence is not well documented. Here, we found that late-passage DPSCs exhibit reduced mitophagy and increased senescence, and that CPT1A knockdown restored mitophagic activity and attenuated senescence. These findings are consistent with previous reports that have shown a decline in mitophagy and an increase in cellular senescence[34,35]. Moreover, the anti-senescent effect of CPT1A silencing was reversed by mitophagy inhibition, functionally linking CPT1A to senescence through mitophagic regulation. Our study demonstrated for the first time that silencing CPT1A inhibits DPSC senescence by promoting mitophagy. Nevertheless, the limitations of using Mdivi-1 to inhibit mitophagy should be noted, as it may have off-target effects other than inhibiting mitophagy. Future research could use more specific genetic approaches to regulate mitophagy, for example, knockout of core mitophagy genes, will help to further prove our conclusion.

The in vivo experiments further demonstrated the inhibitory effect of silencing CPT1A on dental pulp aging and its promoting effect on mitophagy. The aging model induced by D-gal is the one most similar to natural aging[36]. Natural aging leads to the decline of tooth function, tooth loss, and the occurrence of periodontal disease[37]. Therefore, slowing down the aging of dental pulp is of great significance for improving quality of life in the elderly. Our in vivo experimental results not only confirmed our in vitro data, but also highlighted the core role of CPT1A in driving dental pulp aging by regulating mitophagy. Consequently, targeting CPT1A may represent a promising therapeutic strategy to mitigate age-related decline in dental pulp function. However, whether CPT1A specifically targets the senescence of DPSCs in vivo requires further study.

CPT1A is a well-known succinyltransferase. A key novel finding of our study is that Parkin is succinylated at K27 site, and that this modification is regulated by CPT1A, although the presence of additional minor sites cannot be excluded. To our knowledge, this is the first study to report that Parkin can be modified by succinylation. Using the well-characterized CPT1A mutants G710E (CPTase-inactive but LSTase-intact) and H473A (LSTase-inactive), we provide direct evidence that the LSTase activity of CPT1A, is responsible for Parkin succinylation and the subsequent suppression of mitophagy. These findings suggest that CPT1A regulates the succinylation modification of Parkin depending on its role as a succinyltransferase, rather than its canonical role in fatty acid oxidation or other scaffolding functions. Parkin has been reported to be modulated by several post-translational modifications, such as phosphorylation and ubiquitination[38,39], the regulatory role of succinylation has remained entirely unexplored until now. We show that CPT1A knockdown reduces Parkin succinylation, enhancing its stability and promoting mitophagy. This discovery of succinylation-dependent regulation unveils a novel mechanistic layer by which CPT1A controls mitophagy and cellular senescence. Our PLA results showed that CPT1A and Parkin were in close proximity, particularly in senescent cells with inhibited mitophagy. However, when mitophagy was induced, this close proximity was further enhanced. We consider that in p15 DPSCs, although mitophagy is inhibited, the expression of CPT1A increases, which binds to Parkin more effectively and catalyzes its K27 succinylation, thus enhancing its physical proximity. After inducing mitophagy in p15 DPSCs, Parkin is activated, thus further enhancing its physical proximity relationship with CPT1A. Furthermore, our data showed that knockdown of Parkin reversed the effects of CPT1A knockdown on mitophagy and cellular senescence, which underscores the central role of Parkin in these processes. These results are consistent with the established role of Parkin in mitophagy and cellular aging[40-42].

Several limitations should be noted in this study. Although we identified Parkin K27 succinylation as a key modification regulated by CPT1A, we did not directly examine how the K27R mutation affects Parkin mitochondrial translocation or its E3 ubiquitin ligase activity toward downstream substrates under CCCP-induced mitophagy conditions. Thus, the precise molecular mechanism by which K27 succinylation impairs Parkin function remains to be fully elucidated. Additionally, the in vivo knockdown of CPT1A was achieved by tail vein injection of lentivirus, which may cause off-target effects on systemic metabolism and other tissues. Therefore, the observed anti-aging effect in dental pulp cannot be unequivocally attributed solely to CPT1A silencing in DPSCs, and future studies using conditional or tissue-specific knockout approaches are warranted. Third, the mitophagy inhibitor Mdivi-1 used in this study primarily targets Drp1-mediated mitochondrial fission and has known off-target effects. Although we cannot rule out potential non-mitophagy-related actions of Mdivi-1, our conclusions are supported by multiple complementary approaches, such as TEM, double immunofluorescence assay, and immunoblotting. Future studies using genetic approaches (e.g., CRISPR/Cas9-mediated knockout of PARK2 or PINK1) or more specific mitophagy reporters (e.g., mt-Keima) are warranted to definitively establish the causal role of mitophagy in CPT1A-mediated senescence. Moreover, we identified K27 as a functional succinylation site using site-directed mutagenesis, which needs to be verified using mass spectrometry.

CONCLUSION

Our study reveals that silencing CPT1A inhibits senescence of DPSCs by inducing mitophagy. Mechanistically, CPT1A knockdown stabilizes Parkin protein by inhibiting its succinylation at K27 site. These findings not only advance our understanding of DPSC aging but also suggest that targeting the CPT1A-Parkin-mitophagy axis may offer novel strategies to enhance the regenerative potential of DPSCs. Given the evolutionary conservation of CPT1A, Parkin, and succinylation machinery, our findings may extend to improve the limitations of the clinical application of DPSCs and the potential impact on human aging-related diseases.