Introduction

Parkinson’s disease (PD) is a prevalent neurodegenerative disorder primarily affecting the elderly, characterized by progressive motor impairment that significantly diminishes quality of life and limits the capacity to perform daily activities. The underlying pathophysiology of PD is closely associated with mitochondrial dysfunction in neurons [1]. Mitochondria play a fundamental role in cellular energy production, ion regulation, and reactive oxygen species (ROS) control, which are essential for neuronal health and function. However, age-related decline and environmental factors, including exposure to neurotoxic chemicals, can damage mitochondrial structure and cause mutations in mitochondrial DNA [2]. Such mitochondrial damage disrupts cellular metabolism, increases ROS accumulation, and impairs ion balance, ultimately leading to the apoptosis of dopaminergic neurons responsible for movement control, resulting in the hallmark motor symptoms of PD. Current therapeutic approaches for PD are primarily center on managing dopamine levels in the brain through pharmacological treatment, surgical interventions, or physical therapy. While these treatments offer symptomatic relief, they do not halt disease progression or regenerate damaged neurons. This limitation has spurred interest in alternative therapeutic strategies, notably the restoration of mitochondrial function, given the pivotal role mitochondria play in neuronal survival and regeneration [3].

Stem cells, particularly their capacity for indefinite proliferation and differentiation into specialized cell types, are widely investigated for regenerative medicine applications. This field focuses on transplanting stem cells or functional tissues to restore or replace structures compromised by injury or disease. Beyond differentiating into target cell types, stem cells exhibit the unique ability to transfer healthy mitochondria to damaged cells, thereby restoring mitochondrial function and cellular viability [4]. Since stem cells generally operate in a low-energy state, their mitochondria are often young, functionally robust, and exhibit lower ROS levels, rendering them optimal for therapeutic transfer [5].

Mitochondrial transfer has demonstrated therapeutic potential across various disease models. For instance, mesenchymal stem cells (MSCs) from bone marrow were shown to transfer mitochondria to lung cells in a pneumonia mouse model, thereby improving respiratory cell function [6]. Similarly, MSC-derived mitochondria transplanted into the renal tissue of diabetic mice enhanced the morphology and function of renal tubular cells [7]. In neurological disease, studies indicate that astrocyte-derived mitochondria transfer to neurons can support neuronal survival and promote recovery in stroke-induced models [8]. Moreover, administering mitochondria from liver cells into the bloodstream of PD model mice improved mitochondrial respiratory function, reduced oxidative stress, and mitigated neuronal cell death, leading to improved motor behaviors [9].

Human dental pulp stem cells (hDPSCs) have garnered attention as a promising source of stem cells due to their accessibility and ease of isolation from routine dental procedures [10]. The process of harvesting and culturing hDPSCs is minimally invasive, cost-effective, and free from ethical concerns. Studies have shown that hDPSCs can differentiate into multiple cell types, including odontoblasts, osteoblasts, insulin-producing cells, etc [11, 12]. HDPSCs have been widely recognized for their strong immunomodulatory properties and neuroprotective effects, largely mediated through the release of bioactive factors such as cytokines, growth factors, and extracellular vesicles [13, 14]. Notably, recent studies have highlighted the pivotal role of mitochondria in the neuronal differentiation of hDPSCs [15], pointing to their broader potential in neuroregenerative therapies. Despite these promising findings, detailed investigations into the mitochondrial characteristics of hDPSCs, particularly their structural and functional properties, and their ability to transfer mitochondria to damaged cells remain scarce. This gap underscores the need for further research to evaluate their therapeutic potential more comprehensively.

Therefore, this study aims to investigate the mitochondrial properties of hDPSCs and evaluate their feasibility in restoring mitochondrial function within a PD cellular model. This research may serve as a foundation for developing clinical applications of hDPSC-derived mitochondria for PD treatment, with potential implications for other neurodegenerative diseases.

Methods

Tooth sample collection

This study received approval from the Ethical Committee on Human Rights Related to Human Experimentation of the Faculty of Dentistry/Faculty of Pharmacy, Mahidol University (Certificate of approval no. MUDT/PY-IRB 2020/037.2606). Healthy permanent teeth were obtained from Thai patients at the Faculty of Dentistry, Mahidol University. Individuals with a history of smoking, teeth with caries, teeth that had undergone root canal treatment or pulp removal, teeth without pulp, or signs of pulpal pathology were excluded from this study.

Isolation and maintenance of human dental pulp stem cells

The hDPSCs were isolated using the outgrowth method, cultured, and characterized as detailed in our previous work [16]. Briefly, pulp tissues were placed in a culture medium containing Dulbecco’s Modified Eagle Medium (DMEM, HyClone, Cytiva, USA), 10% fetal bovine serum (FBS, Sigma-Aldrich. USA), 100 U/mL penicillin, and 100 µg/mL streptomycin (Gibco, Thermo Fisher Scientific, USA). The cultures were maintained in a humidified incubator at 37 °C with 5% CO2. The medium was refreshed every other day until the cells reached 80% confluence, after which they were either sub-cultured or harvested using 0.25% Trypsin/EDTA (Gibco).

Characterization human dental pulp stem cells

Cell characterization was performed by detecting mesenchymal stem cell surface markers, conducting colony-forming unit (CFU) fibroblast assays, and assessing multipotent differentiation potential. The cells were incubated with antibodies targeting mesenchymal stem cell markers (CD44, CD73, and CD90) and a hematopoietic stem cell marker (CD34), all sourced from BioLegend (San Diego, CA, USA), and analyzed via flow cytometry. The CFU fibroblast assay involved culturing the cells at a low density (500 cells/mL) for two weeks, after which colonies were counted. To assess multipotential differentiation, cells were cultured in osteogenic differentiation medium (culture medium supplemented with 0.1 µM dexamethasone, 50 µg/mL ascorbate-2-phosphate, and 10 mM β-glycerophosphate, all from Sigma-Aldrich) or adipogenic differentiation medium (culture medium supplemented with 1 µM dexamethasone, 50 µM indomethacin, 1 µg/mL insulin, and 0.5 mM 3-isobutyl-1-methylxanthine, all from Sigma-Aldrich). Osteogenic and adipogenic differentiation were evaluated using Alizarin Red staining and Oil Red O staining, respectively.

Isolation of the mitochondria from human dental pulp stem cells

Isolation of mitochondria was conducted as described in previous studies [17, 18]. The hDPSCs at passage 3–6 were harvested using 0.25% Trypsin/EDTA, pelleted, and washed with ice-cold phosphate-buffered saline (PBS). The cell pellets were resuspended in ice-cold homogenizing buffer, consisting of 300 mM sucrose, 10 mM K-HEPES, and 1 mM K-EGTA, and homogenized by repeated aspiration with a syringe and needle. The homogenates were centrifuged at 600 g for 10 min at 2 °C. The supernatant was collected and centrifuged at 14,000 g for 10 min at 2 °C. The mitochondrial pellets were resuspended in ice-cold respiration buffer, consisting of 250 mM sucrose, 2 mM K-HEPES (pH 7.2), 0.5 mM K-EGTA (pH 8.0), 2 mM KH2PO4, and 10 mM MgCl2. The amount of isolated mitochondria was determined using the Bradford assay.

Mitochondrial staining

The hDPSCs were incubated with either 100 nM MitoTracker Red FM or 200 nM MitoTracker Green FM (both from Thermo Fisher) in the culture medium or PBS for 30 min. To visualize the shape and organization of mitochondria, the cells were observed under a confocal microscope (LSM 900 Airyscan 2, Zeiss, Germany). The isolated mitochondria were stained with MitoTracker Green FM and analyzed by flow cytometry to assess their viability.

Electron microscopy analysis

To examine the ultrastructure of the isolated mitochondria, the mitochondrial pellets were first fixed in 2.5% glutaraldehyde in PBS, followed by fixation with OsO4. They were then stabilized with 2% agarose in PBS, dehydrated with ethanol, infiltrated with Spurr’s resin, sectioned using an ultramicrotome, placed on copper grids, stained with uranyl acetate and lead citrate, and finally observed under a transmission electron microscope (JEM-1400, JEOL, Japan).

Cellular model of parkinson’s disease

Human neuroblastoma SH-SY5Y cells (ATCC®, CRL-2266™) were cultured in DMEM/F12 (Hyclone) supplemented with 10% FBS (Sigma-Aldrich), 100 U/mL penicillin, and 100 µg/mL streptomycin (Gibco). The cells were maintained in a humidified incubator at 37 °C with 5% CO2. The medium was changed every other day until the cells reached 80% confluence, at which point they were either sub-cultured or harvested using 0.25% Trypsin/EDTA (Gibco). To induce dopaminergic neuron characteristics, the cells were differentiated by culturing in DMEM/F12 (Hyclone) supplemented with 1% FBS (Sigma-Aldrich), 100 U/mL penicillin, 100 µg/mL streptomycin (Gibco), and 10 µM all-trans retinoic acid (RA, Sigma-Aldrich) for 7 days. Following RA treatment, 1000 µM 1-methyl-4-phenyl-pyridinium ion (MPP+) was added into the medium and maintained for 24 h to model PD [19].

Mitochondrial uptake in cellular model of parkinson’s disease

The mitochondria were stained with 200 nM MitoTracker Green FM (Thermo Fisher) in the respiration buffer for 30 min, followed by three washes with the same buffer. The stained mitochondria were then added to MPP+-treated, differentiated SH-SY5Y cells at concentrations of 0, 10, 20, 50, 100, and 200 µg per 105 cells for 24 and 48 h. After mitochondrial treatment, the cells were washed three times with PBS to remove any excess mitochondria, and the uptake of the mitochondria was observed under a fluorescent microscope (Zeiss). The optimal concentration, defined as the lowest amount of mitochondria that achieved the highest uptake within the shortest time, was selected for subsequent experiments.

To confirm mitochondrial internalization, co-localization with the endogenous mitochondria was assessed. The MPP+-treated, differentiated SH-SY5Y cells were stained with 100 nM MitoTracker Red FM and washed three times with PBS. The MitoTracker Green FM-stained mitochondria isolated from hDPSCs were then added. Co-localization of the mitochondria was visualized under a confocal microscope (Zeiss). The images were deconvoluted with DeconvolutionLab2 [20] and mitochondrial morphology were analyzed using Mitochondria Analyzer [21].

Assessment of mitochondrial complex I activity

Mitochondrial complex I activity was measured using a commercial Complex I Enzyme Activity Microplate Assay Kit (ab197243, Abcam, UK). Briefly, cells were harvested, and protein was extracted. Complex I activity was determined by monitoring the oxidation of NADH to NAD+, coupled with the reduction of a dye that leads to an increase in absorbance at OD 450 nm.

Intracellular ATP measurement

ATP levels were quantified with a commercial ATP assay kit (ab83355, Abcam). ATP-mediated phosphorylation of glycerol produced a colorimetric product, which was measured at OD 570 nm.

Reactive oxygen species (ROS) assay

ROS levels were assessed using the DCFDA/H2DCFDA - Cellular ROS Assay Kit (ab113851, Abcam). 2’,7’-Dichlorofluorescin diacetate (DCFDA) diffuses into cells, where it is oxidized by ROS to form highly fluorescent 2’,7’-dichlorofluorescein (DCF). Fluorescence intensity was visualized and quantified under a fluorescent microscope and ImageJ software.

Neurite length measurement

Cell morphology was visualized using phase-contrast microscopy, and neurite length was quantified with ImageJ software. Only primary neurites emerging directly from the soma were analyzed. Neurite tracing began at the soma origin and followed the neurite path to its termination point.

Immunocytochemistry

Samples were fixed in 4% paraformaldehyde (Sigma-Aldrich) in PBS for 30 min at room temperature, followed by three washes in PBS. Permeabilization and blocking were performed with 1% bovine serum albumin (BSA) and 0.1% Triton X-100 (both from Sigma-Aldrich) in PBS for 30 min at room temperature, followed by three PBS washes with 0.05% Tween-20 (PBST). Specimens were then incubated with primary antibodies diluted 1:300 in 1% BSA in PBST for 1 h at room temperature: anti-tyrosine hydroxylase (TH, Abcam), anti-growth associated protein-43 (GAP-43, Sigma-Aldrich), anti-synaptophysin (SYP, Cell Signaling Technology), and anti-dopamine transporter (DAT, Abcam). Following three washes in PBST, specimens were incubated with conjugated secondary antibodies at 1:500 dilution in 1% BSA in PBST for 1 h at room temperature, then washed with PBST three times. Samples were counterstained and mounted with Vectashield antifade mounting media containing DAPI (Vector Laboratories, USA), and images were acquired with a fluorescence microscope (Thunder Imager, Leica, Germany).

Calcium imaging

Intracellular calcium activity was assessed using the fluorescent calcium indicator dye Fluo-3 AM (Thermo Fisher Scientific). Cells were incubated with 3 µM Fluo-3 AM in the culture medium at 37 °C for 60 min. After incubation, the cells were gently washed with both the culture medium and PBS to remove excess dye, then maintained in Tyrode’s solution for imaging. Calcium influx was stimulated by adding 50 mM KCl, and time-lapse fluorescence recordings were captured using a confocal laser scanning microscope (Zeiss). Fluorescence intensity corresponding to calcium-bound Fluo-3 was quantified using ImageJ software to assess dynamic calcium responses. Quantitative measurements were expressed as normalized fluorescence changes ΔF/F₀ = (F – F₀)/F₀, where F₀ represents baseline fluorescence and F represents peak fluorescence after stimulation. For each culture, the mean peak ΔF/F₀ value was calculated from 30 cells (n). Independent biological replicates were obtained from three hDPSC donors (N = 3), each tested in triplicate. Statistical comparisons were performed on culture-level means using one-way ANOVA with Tukey’s post hoc test.

Statistical analysis

Data were presented as mean ± standard deviation (SD). To assess whether mitochondrial dose and treatment duration influenced mitochondrial uptake, a two-way analysis of variance (ANOVA) was performed, followed by Tukey’s multiple comparison test for determining significant differences. One-way ANOVA was used to compare differences among the experimental groups: Control, Control + MT, MPP+, and MPP++MT, with Tukey’s test applied for multiple comparisons. Statistical significance was set at p < 0.05. All analyses were conducted using GraphPad Prism 10.3.1.

Results

Culturing and characterization of human dental pulp stem cells

Isolate cells from human dental pulp adhered to the plastic surfaces of culture dishes and exhibited a fibroblast-like morphology (Fig. 1a). Flow cytometry analysis revealed that these cells expressed mesenchymal stem cell markers, with over 90% of cells expressing CD44, CD73, and CD90 (N = 3), and less than 0.5% expressing the hematopoietic stem cell marker CD34 (Fig. 1e). When seeded at low density to allow colony formation over two weeks, cell populations appeared as distinct clusters with defined boundaries, indicating self-renewal capability (Fig. 1b). Osteogenic induction followed by specific staining revealed that hDPSCs could differentiate into osteoblasts, as shown by the positive Alizarin Red staining, which binds to calcium deposits (Fig. 1c). When induced to form adipocytes, hDPSCs demonstrated adipogenic differentiation, as indicated by Oil Red O staining, which marked intracellular lipid droplets as red spots (Fig. 1d).

Fig. 1
figure 1

Culturing and characterization of hDPSCs. a Light microscopy reveals the cells exhibiting a fibroblast-like morphology. b Colony-forming unit-fibroblast assay shows distinct colonies within a 6-well plate under light microscopy. c Alizarin Red staining highlights osteogenic differentiation, indicated by red mineralized nodules. d Adipogenic differentiation is demonstrated by Oil Red O staining, revealing lipid droplet formation. e Flow cytometry analysis confirms the expression of CD44, CD73, and CD90, with minimal expression of CD34. Scale bar = 100 μm

Full size image

Mitochondrial characteristics of human dental pulp stem cells

Mitochondrial staining with MitoTracker displayed distribution of mitochondria throughout the cytoplasm and extending into cytoplasmic processes, as observed under fluorescent microscopy (Fig. 2a). Confocal microscopy revealed a bead-on-string-like mitochondrial arrangement within the cells (Fig. 2b). Quantifying isolated mitochondria using the Bradford assay showed an average mitochondrial concentration of 11.2 ± 0.2 mg/mL from 5 to 6 × 106 cells. Flow cytometry analysis of isolated mitochondria stained with MitoTracker Green FM, a dye specific to intact mitochondria, showed 70.1 ± 3.8% of mitochondria with preserved membrane integrity (n = 3) (Fig. 2c). TEM imaging revealed rounded mitochondria with underdeveloped cristae and no elongated forms (Fig. 2d), with diameters ranging from 37.6 to 1152.7 nm and an average of 237.5 nm.

Fig. 2
figure 2

Mitochondrial characteristics of hDPSCs. a Mitochondria of hDPSCs were stained with MitoTracker Green FM. Scale bar = 50 μm. b Confocal microscopy shows mitochondria within hDPSCs arranged in a bead-on-string pattern. Scale bar = 10 μm. c Flow cytometry analysis of MitoTracker Green FM demonstrates that most isolated mitochondria from hDPSCs are viable. d TEM image of isolated mitochondria from hDPSCs reveals a rounded shape with underdeveloped cristae. Scale bar = 1 μm

Full size image

Optimal mitochondrial dosage and time for uptake in cellular model of parkinson’s disease

Isolated mitochondria from hDPSCs were introduced into an MPP+-treated differentiated SH-SY5Y cells (a cellular model of PD) to verify mitochondrial uptake via co-localization with host mitochondria. Mitochondria of MPP+-treated cells were labeled with MitoTracker Red FM, while isolated hDPSC mitochondria were stained with MitoTracker Green FM. Confocal microscopy with orthogonal views demonstrated co-localization, with red and green fluorescence appearing in the same focal plane, confirming successful mitochondrial uptake (Fig. 3a). Fluorescence intensity of MitoTracker Green FM increased proportionally with mitochondrial concentration at 24 and 48 h (Fig. 3b-c). Two-way ANOVA revealed no significant difference in mitochondrial uptake between the two-time points. At mitochondrial doses of 10, 25, and 50 µg, fluorescence intensity did not differ significantly; however, doses of 100 and 200 µg produced significantly higher intensity than 10, 25, and 50 µg (p < 0.05). There was no significant difference between 100 and 200 µg doses, thus 100 µg and 24 h were chosen as optimal for further experiments.

Fig. 3
figure 3

Optimal mitochondrial dosage and time for uptake in cellular model of Parkinson’s disease. a Confocal microscopy image of MPP+-treated, differentiated SH-SY5Y cells with mitochondria stained using MitoTracker Red FM, while isolated hDPSC mitochondria were stained with MitoTracker Green FM. Orthogonal views from x/z (green box) and y/z (red box) planes show co-localization of the two mitochondrial sources, appearing as yellow/orange in merged images. Scale bar = 10 μm. b Scatter plots with bars show the uptake of isolated hDPSCs mitochondria at various doses and time points, represents as mean ± SD of MitoTracker Green FM intensity. Each dot represents independent replicate of three hDPSC donors (N = 3). Bars with different letters indicate significantly different (p < 0.05). c Fluorescence microscopic images show the uptake of isolated hDPSCs mitochondria at various doses and time points. Scale bar = 200 μm

Full size image

Mitochondria morphology and dynamic of cellular model of parkinson’s disease after receiving hDPSC-derived mitochondria

The experiment included four groups: Control (differentiated SH-SY5Y cells), Control + MT (differentiated SH-SY5Y cells treated with isolated hDPSC-derived mitochondria), MPP+ (MPP⁺-treated differentiated SH-SY5Y cells), and MPP++MT (MPP⁺-treated cells receiving isolated hDPSC-derived mitochondria). Endogenous mitochondria were labeled with MitoTracker Red FM, while exogenous hDPSC-derived mitochondria were stained with MitoTracker Green FM. Live-cell confocal microscopy enabled three-dimensional imaging for quantitative analysis of individual mitochondrial morphology. Sphericity was employed as a morphological parameter, with values approaching 1 indicating a more spherical structure (21). MPP⁺ treatment resulted in a significant increase in mitochondrial sphericity compared to the Control (p < 0.01) and Control + MT (p < 0.05) groups. Mitochondrial sphericity in the MPP++MT group remained comparable to the MPP⁺ group, suggesting that administration of exogenous hDPSC-derived mitochondria did not substantially alter the morphology of the existing dysfunctional mitochondria (Fig. 4a). Morphological examination further revealed that in the Control and Control + MT groups, host mitochondria exhibited diverse shapes and were distributed throughout the cytoplasm, including both the soma and neurites (Fig. 4c-i and c-ii). Conversely, MPP⁺ and MPP++MT groups predominantly displayed fragmented, spherical mitochondria confined to the soma (Fig. 4c-iii, 4c-iv), in line with pathological features observed in PD.

Fig. 4
figure 4

Mitochondrial morphology and distribution in differentiated SH-SY5Y cells following treatment with MPP⁺ and hDPSC-derived mitochondria. a Individual mitochondrial sphericity plot across Control, Control + MT, MPP⁺, and MPP⁺+MT. Higher sphericity values indicate more spherical mitochondria. b Percentage of co-localized mitochondria (yellow fluorescence from overlapping MitoTracker Red FM-labeled endogenous and MitoTracker Green FM-labeled hDPSC-derived mitochondria) in Control + MT and MPP⁺+MT groups. c Representative confocal images of mitochondrial morphology and distribution: (i) Top: Control group showing heterogeneous mitochondrial shapes (MitoTracker Red FM-label) dispersed throughout the soma and neurites. Bottom: Higher magnification area indicated by the rectangle of the top image. (ii) Control + MT group showing co-localization of endogenous and exogenous mitochondria (yellow fluorescence). Top left: MitoTracker Red FM-labeld endogenous SH-SY5Y mitochondria. Top right: MitoTracker Green FM-labeld isolated hDPSC-derived mitochondria. Bottom left: Merge image. Bottom right: Higher magnification area indicated by the rectangle of the bottom left image. (iii) Top: MPP⁺ group displaying fragmented, spherical mitochondria (MitoTracker Red FM-labeld) confined to the soma. Bottom - Higher magnification area indicated by the rectangle of the top image. (iv) MPP⁺+MT group showing dispersed hDPSC-derived mitochondria with reduced co-localization. Top left: MitoTracker Red FM-labeld endogenous SH-SY5Y mitochondria. Top right: MitoTracker Green FM-labeld isolated hDPSC-derived mitochondria. Bottom left: Merge image. Bottom right: Higher magnification area indicated by the rectangle of the bottom left image. Arrows indicate tubular mitochondria in the cytoplasmic process/neurite. N = 3, *p < 0.05, **p < 0.01. Scale bar = 25 μm

Full size image

Co-localization analysis revealed distinct integration patterns of hDPSC-derived mitochondria. In Control + MT cells, exogenous mitochondria co-localized with endogenous counterparts, producing a yellow signal from the overlap of red and green fluorescence (Fig. 4c-ii). However, MPP++MT cells exhibited a significantly reduced percentage of co-localized mitochondria compared to the untreated group (p < 0.01) (Fig. 4b). Notably, exogenous mitochondria in the MPP++MT group were observed in the cytoplasmic processes/neurites, while endogenous mitochondria remained restricted to the soma (Fig. 4c-iv). Although hDPSC-derived mitochondria initially appeared round under TEM, tubular forms were evident post-transfer in MPP⁺-treated cells, suggesting mitochondrial remodeling or fusion activity may occur in response to the intracellular environment (Fig. 4c-ii, 4c-iv).

Mitochondrial treatment restores mitochondrial function in cellular model of parkinson’s disease

MPP+ disrupts mitochondrial respiratory chain Complex I activity. The rate of the Complex I activity, measured change of NAD + production by time, was significantly decreased in the MPP+ group, while cells treated with isolated hDPSC mitochondria showed restored Complex I activity (p < 0.05) (Fig. 5a). ATP production, a mitochondrial respiratory chain product, was significantly lower in the MPP+ group compared to Control and Control + MT (p < 0.01). Mitochondrial treatment in the MPP++MT group restored ATP levels significantly above those in the MPP+ group (p < 0.05) (Fig. 5b). ROS levels, quantified by DCF fluorescence (an oxidized DCFDA product), were reduced in both Control and Control + MT groups compared to the MPP+ group (p < 0.001) (Fig. 5c-d).

Fig. 5
figure 5

Mitochondrial treatment restores mitochondrial function in cellular model of Parkinson’s disease. a Rate of Complex I activity across Control, Control + MT, MPP+, and MPP++MT groups. Each point represents the mean of OD 450 nm. b Scatter plots with bars showing mean ± SD of intracellular ATP levels and c ROS levels represented by the DCF intensity across groups. Each dot represents independent replicate of three hDPSC donors (N = 3), *p < 0.05, **p < 0.01, ***p < 0.001. d Fluorescence microscopic images show DCF intensity as pseudocolor scale in the range of 0–256 (arbitrary unit). Scale bar = 200 μm

Full size image

Mitochondrial treatment promotes neuronal recovery in cellular model of parkinson’s disease

Neurite length

Neurite length, an indicator of neuronal growth, was measured from phase-contrast microscopy images using ImageJ software. MPP+ treatment significantly reduced neurite length in differentiated SH-SY5Y cells compared to untreated controls (p < 0.0001). Mitochondrial administration (MPP++MT) significantly restored neurite length compared to both MPP+ and Control groups (p < 0.0001). Interestingly, neurite length was also significantly increased in Control + MT compared to Control and other groups (p < 0.0001). (Figure 6a and b)

Fig. 6
figure 6

Mitochondrial treatment promotes neuronal growth in cellular model of Parkinson’s disease. a Phase-contrast microscopy images illustrating cell morphology of Control, Control + MT, MPP+, and MPP++MT groups. Scale bar = 200 μm. b Violin plots of individual neurite lengths (gray dots) across groups; bold horizontal lines represent the mean, thin lines indicate SD. c Immunofluorescence microscopy images of Control, Control + MT, MPP+, and MPP++MT groups stained with DAPI (blue) and GAP43 (green). Scale bar = 25 μm. d Scatter plots with bars show mean ± SD intensity of GAP43. Each dot represents independent replicate of three hDPSC donors (N = 3), *p < 0.05, ****p < 0.0001

Full size image

Expression of growth associated protein 43 (GAP43)

GAP43, a protein marker of neuronal growth and plasticity, was assessed via immunofluorescent staining, showing expression within the cytoplasm, especially at branching points and growth cones (Fig. 6c). GAP43 expression, measured by fluorescence intensity, was significantly reduced in MPP+ treated cells compared to Control (p < 0.0001). Mitochondrial treatment in MPP++MT cells significantly increased GAP43 expression compared to both MPP+ (p < 0.0001) and Control groups (p < 0.05). Notably, GAP43 expression in Control + MT was significantly higher than in both Control (p < 0.05) and MPP+ (p < 0.0001), with no significant difference between Control + MT and MPP++MT (Fig. 6d).

Expression of tyrosine hydroxylase (TH)

TH, a rate-limiting enzyme for dopamine synthesis, was assessed via immunofluorescent staining, observed in both the cytoplasm and nucleus (Fig. 7a). TH expression was significantly reduced in MPP+ cells compared to both Control and Control + MT (p < 0.05). Mitochondrial treatment in MPP++MT cells significantly restored TH expression compared to the MPP+ group (p < 0.01) (Fig. 7c).

Fig. 7
figure 7

Mitochondrial treatment restores dopaminergic protein expression. a Immunofluorescence microscopy images of Control, Control + MT, MPP+, and MPP++MT groups stained for TH (green) and DAPI (blue). b Immunofluorescence microscopy images stained for SYP (green, left column), DAT (red, middle column, ), and DAPI (blue, merge, right column). Scale bar = 25 μm. c-e Scatter plots with bars showing mean ± SD intensity of TH, SYP, and DAT, respectively. Each dot represents independent replicate of three hDPSC donors (N = 3), *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Full size image

Expression of synaptophysin (SYP)

SYP, a presynaptic vesicle protein, was evaluated by immunofluorescence staining and detected in the cytoplasm, neurites, and synaptic terminals of differentiated SH-SY5Y cells (Fig. 7b). SYP expression, quantified by mean fluorescence intensity, was significantly decreased in MPP⁺-treated cells compared to both Control and Control + MT groups (p < 0.001). In contrast, treatment with hDPSC-derived mitochondria significantly restored SYP expression in the MPP⁺+MT group compared to MPP⁺ (p < 0.001) (Fig. 7d).

Expression of dopamine transporter (DAT)

DAT, a membrane protein involved in dopamine reuptake at synaptic terminals, was also assessed via immunofluorescence. Similar to SYP, DAT was observed in the cytoplasm, neurites, and synaptic terminals (Fig. 7b). MPP⁺ treatment significantly reduced DAT expression compared to Control and Control + MT groups (p < 0.0001). Notably, DAT expression was significantly elevated in the MPP⁺+MT group following mitochondrial treatment compared to the MPP⁺ group (p < 0.0001) (Fig. 7e).

Intracellular calcium activity

Upon KCl stimulation, cells in Control, Control + MT, and MPP++MT groups exhibited a rapid increase in cytoplasmic Fluo-3 fluorescence intensity (ΔF/F₀), indicating normal calcium influx. In contrast, MPP⁺-treated cells showed minimal changes, reflecting impaired calcium signaling (Fig. 8a and b). Quantification across independent cultures (N = 3 donors, each with 30 cells analyzed per culture) confirmed that Control group displayed a significant increase in fluorescence intensity (p < 0.05) compared to MPP⁺-treated cells. Mitochondrial treatment significantly enhanced ΔF/F₀ in both Control + MT (p < 0.001) and MPP++MT groups (p < 0.0001) compared to MPP⁺ (Fig. 8c). These results demonstrate that hDPSC-derived mitochondria restore calcium responsiveness in the PD cellular model.

Fig. 8
figure 8

Mitochondrial treatment restores intracellular calcium activity. a Fluorescence microscopy images of Control, Control + MT, MPP+, and MPP++MT groups show intensity of Fluo-3 before (left) and after KCl stimulation (right). Pseudocolor scale in the range of 0–256 (arbitrary unit). Scale bar = 25 μm. b Line graph with bars showing normalized fluorescence changes (ΔF/F₀, mean ± SD) over time (seconds). c Scatter plots with bars showing mean ± SD ΔF/F₀. Each dot represents independent replicate of three hDPSC donors (N = 3) with 30 cells analyzed per culture, *p < 0.05, ***p < 0.001, ****p < 0.0001

Full size image

Discussion

This study provides evidence supporting the therapeutic potential of hDPSC-derived mitochondria for restoring mitochondrial function and promoting neuronal recovery in a Parkinson’s disease cellular model. Previous studies have demonstrated that hDPSCs exert neuroprotective benefits through multiple mechanisms, including the release of trophic factors, immunomodulation, and potential differentiation into neuron-like cells. Our study specifically focused on mitochondria as the primary mediators of functional recovery, allowing us to dissect their unique contribution. While this approach provides mechanistic clarity, it does not capture the broader and potentially synergistic effects of whole-cell transplantation. Future work comparing cell-based versus mitochondria-based strategies in parallel — particularly in animal models — will be essential to determine which approach offers greater efficacy, safety, and feasibility for clinical translation.

Mitochondrial treatment into MPP+-treated dopaminergic neuron-like cells improves mitochondrial function, as evidenced by restored Complex I activity, ATP production, and reduced reactive oxygen species (ROS) levels. These findings highlight the role of mitochondrial treatment in repairing damaged mitochondria and restoring cellular energy metabolism in neurodegenerative models.

The isolation of mitochondria from hDPSCs reveals a significant capacity for cellular therapy applications, particularly in conditions involving mitochondrial dysfunction, like PD. HDPSCs are mesenchymal stem cells with a high mitochondrial content, allowing for the extraction of over 11 mg/mL of mitochondria from approximately 5–6 × 106 cells. This yield exceeds that of fibroblast-derived mitochondria from earlier studies, which produced only 2 mg/mL from 5 × 106 fibroblast cells [18]. The isolated mitochondria have an average size of 237.5 nm, exhibit a small, spherical shape, and lack prominent cristae. This observation is consistent with previous research suggesting that mesenchymal stem cell mitochondria exist in a low-energy inactive state that maintains redox balance, reduces ATP production, and minimizes oxidative stress, resulting in underdeveloped cristae. In contrast, highly active cells, such as liver and muscle cells, possess larger, elongated mitochondria with branching and well-defined cristae [5]. Moreover, over 70% of mitochondria isolated from hDPSCs retained their membrane integrity, as indicated by MitoTracker Green FM staining. The results suggest that hDPSCs contain a substantial number of mitochondria suitable for transferring into damaged cells and the extraction method is efficient, yielding large amounts of intact mitochondria. These mitochondrial properties make hDPSC mitochondria particularly well-suited for transfer into cells with mitochondrial damage, such as those affected by PD.

The internalization of mitochondria into recipient cells occurs via macropinocytosis, where cell membranes envelop and internalize mitochondria [22]. Each cell type exhibits different capacities and timing for mitochondrial uptake. The results show that 100 µg of hDPSC-derived mitochondria is the optimal amount for maximum uptake by PD model cells. Increasing the mitochondria to 200 µg did not significantly enhance uptake, and 24 h was sufficient for mitochondrial entry, as extending to 48 h showed no additional uptake.

Mitochondrial dysfunction is a hallmark of PD pathology, particularly the impairment of Complex I in the mitochondrial respiratory chain, which leads to ATP depletion, increased oxidative stress, and neuronal death [23]. MPP+ is a chemical used to induce Complex I disruption and ATP reduction, simulating the mitochondrial dysfunction observed in PD. In neurons, ATP is crucial for neurotransmitter synthesis, release, and electrical signal transmission. Mitochondrial dysfunction, indicated by reduced ATP levels, can impair neuronal signaling [24]. Additionally, mitochondrial dysfunction leads to ROS accumulation, causing oxidative stress and eventual cell death. In this study, MPP+ exposure significantly reduced Complex I activity and ATP levels, simulating the mitochondrial dysfunction observed in PD patients. However, 24-hour treatment of hDPSC-derived mitochondrial treatment significantly improved Complex I activity and ATP levels. This restoration may result from the incorporation of functional mitochondria, which support energy production and metabolic stability, in line with previous studies showing similar results with mitochondrial transplantation in other models [8, 25, 26]. ROS levels were notably higher in PD model cells but were substantially reduced after mitochondrial treatment, suggesting restored redox balance. Together, these findings suggest that hDPSC-derived mitochondria can alleviate mitochondrial damage and mitigate oxidative stress in PD-like conditions.

Enhanced neuronal recovery was further demonstrated by structural and functional improvements. In the SH-SY5Y PD, retinoic acid is used to differentiate the cells into dopamine-producing neurons, with neurite outgrowth from the cell body [27]. MPP+ treatment significantly reduced neurite length, with most cells exhibiting no neurites (Fig. 6a), consistent with previous studies showing that MPP+ induces neurite retraction [28]. MPP+ also significantly reduced GAP43 protein expression, a marker of neuronal growth, which is typically expressed in the cytoplasm, particularly at the growth cones (Fig. 6c). Previous research similarly observed reduced GAP43 expression in MPP+-treated SH-SY5Y cells [29]. However, hDPSC-derived mitochondrial treatment in the Parkinson’s model significantly increased neurite outgrowth and GAP43 expression comparable to the control group, indicating that hDPSC-derived mitochondria promote neuronal growth recovery. Mitochondria accumulate densely at the neurite growth cone to support neurite extension, synaptic growth, and neural connections. When mitochondrial function is inhibited, neurite growth halts [30]. The impact of hDPSC-derived mitochondria on neurogenesis provides insight into potential regenerative therapies for neurodegenerative diseases.

Dopamine is a critical neurotransmitter in brainstem function, particularly for skeletal muscle control. PD patients experience motor dysfunction due to dopaminergic neuron death, resulting in reduced dopamine levels. In the Parkinson’s disease model, SH-SY5Y cells treated with retinoic acid differentiate into dopamine-producing neurons, increasing TH production [19]. TH is an enzyme essential for dopamine synthesis, which significantly decreases following MPP+ treatment, as shown in the experiment (Fig. 7a and c), in line with previous studies [31]. To further validate dopaminergic function, we examined the expression of two key dopamine-related proteins: synaptophysin (SYP) and dopamine transporter (DAT). SYP is a presynaptic vesicle protein involved in neurotransmitter release, including dopamine, and is found in the substantia nigra [32, 33]. DAT is a transmembrane protein responsible for dopamine reuptake at the synaptic cleft and serves as a marker of dopaminergic terminal integrity [34, 35]. Our findings revealed a significant reduction in both SYP and DAT expression following MPP⁺ treatment (Fig. 7b and d-e), corroborating reports of decreased expression of these proteins in multiple PD models and postmortem human brain tissue [36,37,38,39,40,41]. In fact, reduced DAT levels have been correlated with striatal dopamine loss in PD patients, as shown in postmortem brain and PET imaging studies [42, 43]. Importantly, treatment with hDPSC-derived mitochondria (MPP⁺+MT group) significantly restored the expression of TH, SYP, and DAT to levels comparable to untreated controls. This suggests not only recovery of mitochondrial function but also the re-establishment of dopaminergic activity. The coordinated upregulation of TH (dopamine synthesis), SYP (dopamine release), and DAT (dopamine reuptake) strongly supports functional restoration of dopaminergic neurons. These protein markers serve as reliable surrogates for dopamine release and neuronal function, reinforcing the therapeutic potential of hDPSC-derived mitochondria in reversing dopaminergic deficits associated with PD.

Calcium imaging was employed to visualize the dynamic process of synaptic transmission. During an action potential, voltage-gated calcium channels open, allowing Ca²⁺ influx, which triggers synaptic vesicles to fuse with the presynaptic membrane and release neurotransmitters into the synaptic cleft [44]. In the MPP⁺-treated differentiated SH-SY5Y cells, this calcium influx was abolished, as evidenced by the lack of change in fluorescence intensity of the calcium indicator Fluo-3 following KCl stimulation. However, treatment with mitochondria isolated from hDPSCs restored calcium responsiveness in MPP⁺-treated cells similar to Control group, indicating that the cells were capable of responding to stimulation and likely releasing neurotransmitters [45]. This recovery of calcium signaling, together with the restored expression of dopamine-related proteins, highlights the critical role of healthy mitochondria in regulating neurotransmitter synthesis, storage, and release [46]. These findings suggest that hDPSC-derived mitochondria have the potential to restore dopaminergic function in PD. Although calcium imaging provided valuable insights into functional recovery, incorporating direct electrophysiological recordings in future studies would offer a more comprehensive understanding of neuronal activity.

Mitochondrial dynamics are critical for sustaining neuronal health and function. Mitochondrial fusion enables the exchange of mitochondrial contents, dilution of damaged components, and maintenance of ATP production. Conversely, fission facilitates mitochondrial distribution, quality control, and removal of dysfunctional mitochondria via mitophagy. A balanced interplay between these processes ensures optimal energy supply and neuroprotection. Disruption of this balance is associated with neuronal dysfunction and neurodegenerative diseases, including PD [47]. In this study, mitochondrial dynamics were evaluated by measuring mitochondrial sphericity, a metric indicative of fission events, where increased sphericity reflects fragmented, spheroid mitochondria [48]. MPP⁺-treated cells exhibited significantly higher sphericity values and swollen, rounded mitochondria—morphological features commonly observed in PD and linked to mitochondrial dysfunction and excessive fission [49, 50]. Notably, treatment with hDPSC-derived mitochondria did not significantly alter the sphericity in MPP⁺-treated cells, suggesting that the excessive fission state persisted. Furthermore, hDPSC-derived mitochondria remained largely separate from the host mitochondria, as indicated by a lower percentage of co-localization compared to untreated cells. Mitochondrial co-localization typically reflects fusion between exogenous and endogenous mitochondria, a process that requires an intact mitochondrial membrane potential [51]. In healthy cells, such fusion supports mitochondrial network integrity and function. The reduced fusion observed in MPP⁺-treated cells may indicate selective segregation of healthy hDPSC-derived mitochondria from dysfunctional host mitochondria, potentially allowing the exogenous mitochondria to restore neuronal function independently. Although isolated hDPSC-derived mitochondria appeared round in TEM images, tubular mitochondria were also present in both MPP+-treated and untreated cells, suggesting that once internalized, these mitochondria may become activated from dormant state and fuse together to enhances neuronal activity and synaptic plasticity [52] contributing to functional recovery of the PD cellular model. Additionally, most mitochondria found within the neurites of the MPP⁺+MT group were derived from hDPSCs. Previous study has shown a loss of synaptic mitochondria in neurons from postmortem PD brains [41]. In our model, MPP⁺-treated cells exhibited mitochondria confined primarily to the soma, while hDPSC-derived mitochondria were distributed throughout the cell, especially within neurites, indicating their role in supporting synaptic function [47]. Collectively, our findings suggest that hDPSC-derived mitochondria remain distinct from damaged endogenous mitochondria, properly distribute throughout neurite, and fuse together to restore neuronal growth and synaptic activity in a PD cellular model.

Conclusions

This study demonstrates the therapeutic potential of hDPSC-derived mitochondria in restoring mitochondrial function, reducing oxidative stress, enhancing dopaminergic protein expression, and promoting neuronal growth and synaptic activity. Upon introduction, the exogenous mitochondria remained distinct from damaged host mitochondria, fused, and distributed throughout the cells, thereby helping to re-establish cellular energy production, redox balance, and synaptic plasticity. These findings highlight their promise in treating neurodegenerative disorders, where mitochondrial dysfunction plays a central role in disease progression, and offer a foundational step toward the development of cell-free mitochondrial therapy for PD. Future studies should investigate long-term outcomes, elucidate the molecular mechanisms underlying mitochondrial dynamics, and assess in vivo efficacy to further advance this therapeutic strategy toward clinical application.