Intranasal delivery of DPSC-derived small extracellular vesicles-encased phloroglucinol attenuates non-motor and motor deficits and promotes neurogenesis in an in vivo rat model of Parkinson’s disease

Abstract

Background

Parkinson’s disease (PD) is characterized by dopaminergic (DA) neuron degeneration in the substantia nigra pars compacta (SNpc) driven by oxidative stress, inflammation, and impaired neurogenesis. Phloroglucinol, a polyphenolic antioxidant, has demonstrated neuroprotective effects in PD models but suffers from limited clinical applicability due to poor blood-brain barrier (BBB) permeability. Small extracellular vesicles (sEV) derived from dental pulp stem cells (DPSCs) exhibit neuroprotective and immunomodulatory properties and serve as promising vehicles for targeted drug delivery across the BBB. This study aimed to evaluate the therapeutic efficacy of intranasally administered sEV-encased phloroglucinol (sEV-Phl) in a chronic MPTP rat model of PD.

Methods

DPSC-derived sEV were isolated via density gradient ultracentrifugation and characterized using Transmission Electron Microscopy (TEM), Dynamic-Light-Scattering (DLS), and CD marker expression. Phloroglucinol was encased in sEV (sEV-Phl) using sonication. Antioxidant properties were tested in vitro using an H₂DCF.DA assay in SH-SY5Y cells exposed to 6-OHDA. Chronic MPTP-treated male Wistar rats received intranasal sEV-Phl, with motor and non-motor behaviours evaluated up to 4-weeks post-MPTP treatment. TH-positive neurons, neurogenesis (Ki67, BrdU and FOXA2), lipid-peroxidation, and neurotransmitter-levels were analyzed. sEV biodistribution was tracked via near-infrared imaging and localization in neuronal and glial cells was confirmed with PKH-26 labelling, with confocal-imaging further verifying localization in neuronal and glial cells. TNF-α expression was assessed as a marker of neuroinflammation.

Results

sEV displayed high purity and homogeneity. sEV-Phl significantly reduced oxidative stress both in vitro and in vivo, as indicated by decreased ROS and lipid peroxidation levels. sEV-Phl treated MPTP rats demonstrated marked improvement in motor and non-motor behaviours compared to MPTP rats. Immunohistochemical analysis revealed increased TH-positive neurons and enhanced neurogenesis in the SNpc of sEV-Phl-treated animals. Biodistribution studies confirmed efficient midbrain targeting of sEV, which were localized to dopaminergic-neurons, astrocytes and microglia. sEV-Phl also significantly reduced TNF-α expression, indicating decreased neuroinflammation.

Conclusion

This study provides the first instance of using DPSC-derived sEV as a delivery vehicle for phloroglucinol in a PD model. sEV-Phl demonstrated significant neuroprotective-effects, enhanced DA-neuron survival and neurogenesis, and reduced neuroinflammation. Intranasal delivery of sEV-Phl represents a promising non-invasive therapeutic strategy for PD, offering a dual benefit of antioxidative and neurogenic support.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-025-04573-2.

Introduction

Parkinson’s disease (PD) is a complex neurodegenerative disorder marked by the selective loss of dopaminergic (DA) neurons in the substantia nigra pars compacta (SNpc) of the midbrain, resulting in striatal dopamine depletion and disruption of basal ganglia circuitry [1, 2]. By the time motor symptoms of PD manifest, an estimated 60–70% of midbrain DA neurons have already been lost [3, 4]. Despite significant advances in recent years, current pharmacological and other therapeutic interventions are yet not able to halt PD progression or support the survival of existing DA neurons. Dopamine and its analogues are the mainstay in PD therapy, but their long-term use is associated with progressively severe side-effects and the production of reactive oxygen species (ROS) due to autoxidation, DA neurons being particularly susceptible to oxidative damage due to the presence of both dopamine and high levels of iron. While ROS (which contribute to oxidative stress) cannot be directly estimated in living patients or post-mortem tissues due to their short half-life, several indirect indicators of ROS activity in the post-mortem Parkinsonian brain support the significant role of oxidative stress in the disease. These include increased membrane peroxidation, as indicated by elevated levels of TBA (thiobarbituric acid)-reactive substances; increased cysteinyl adducts of dopamine; and ROS-mediated DNA damage, evidenced by higher levels of 8-hydroxy-2-deoxyguanosine in the substantia nigra pars compacta (SNpc) [5–7]. In addition to oxidative stress, inflammation and the lack of compensation for damaged/lost DA neurons are two other key factors that significantly contribute to the progression and worsening of neurodegeneration in PD. Unfortunately, dopamine analogues do not address these factors, and thus effective treatment in PD should incorporate supportive therapies that can provide antioxidative, anti-inflammatory and regenerative relief, ideally delivered through a non-invasive route.

Recent efforts in PD pharmacotherapy have thus increasingly focused on small molecules that mimic the neuroprotective (antioxidant) and neurogenic properties of endogenous neurotrophins and neurotrophic factors. For instance, the polyphenolic compound phloroglucinol has been shown to alleviate motor deficits in a 6-OHDA-induced PD model by enhancing Nrf2-mediated antioxidant activity when administered intrastriatally [8]. Similarly, the microneurotrophin BNN has been shown to induce adult neurogenesis in the SNpc of PD weaver mouse models [9]. The polyphenol oleuropein has also been found to inhibit α-synuclein fibrillation, thus reducing the risk of α-synuclein aggregation [10, 11]. However, clinical applications of these compounds face significant hurdles, primarily due to their limited ability to cross the blood-brain barrier (BBB) and their rapid degradation through the first-pass metabolism in the gut when ingested through the oral route (direct intrastriatal delivery being ruled out due to its invasive nature). Polyphenols such as phloroglucinol, due to their phytoestrogenic nature, also demonstrate high affinity for estrogenic (ERβ) receptors [12, 13], which are abundantly expressed in nasal mucosa [14]. This interaction significantly reduces the bioavailability of these compounds for DA neurons in the midbrain when administered directly intranasally. Traditional drug delivery approaches fail to address the critical challenge of organ specificity, as these molecules are indiscriminately absorbed by all cells regardless of disease state. Ideally, these promising compounds require a delivery mechanism that avoids the gut, can cross the BBB, and homes in to the injured or affected regions by responding to inflammatory cues from cells in those areas. Additionally, if the carrier or delivery vehicle itself possesses immunomodulatory properties, it could significantly enhance therapeutic effects on the injured cells while the compound delivers its antioxidative benefits.

Recent research efforts have thus increasingly explored the use of zwitterionic nanoliposomes and EV derived from mesenchymal stromal cells (MSCs) to address many of these issues and enhance the clinical viability of such compounds. While many other drug delivery routes — such as dendrimers, nanoparticles, liposomes, hydrogels and cell-penetrating peptides (CPPs) — have shown promise in this regard, many are limited by dysregulation of biochemical pathways or increased risk of side effects [15–18]. An ideal drug delivery system should be capable of site-specific delivery, while avoiding premature degradation or recognition by the body’s immune defenses.

Small EVs (sEV), which are nano-sized vesicles (30–150 nm) derived from adult tissue MSCs, offer a very promising solution in this situation due to their intrinsic ability to penetrate the BBB. MSCs are known for their strong immunomodulatory properties and have been widely used in cellular therapies for various human diseases. These same therapeutic effects have also been observed in sEV derived from MSCs, which mimic the immunomodulatory functions of their parent cells [19]. Mediated primarily through their paracrine functions, MSCs exert therapeutic effects by secreting growth factors, antioxidants, proteins, nucleic acids, and lipids that deliver protective signals to target cells [20]. MSC derived sEV, unlike nanoparticles, can regulate macrophage polarization, especially under disease conditions [21–27]. They also show minimal long-term accumulation in organs or tissues and low systemic toxicity while facilitating cellular uptake [15]. In particular, Pivoraitė U et al. (2015); [28] reported that sEV derived from Dental Pulp Stem Cells (DPSCs) exhibit similar neuromodulatory and neuroprotective effects as the DPSCs themselves. An earlier study from our lab [29] also demonstrated that DPSC-derived conditioned media can protect DA neurons from 6-OHDA toxicity in non-contact scenarios, suggesting that the beneficial effects are mediated by the MSC secretome. However, there are yet no studies on the use of DPSC-derived sEV as delivery vehicles for therapeutic agents to the midbrain via the intranasal route.

In this study, we aimed to utilize the beneficial role of the DPSC secretome by isolating sEV and using them as carriers for the antioxidant phloroglucinol, which has shown neuroprotective effects in PD models. Recognizing the need for an effective delivery vehicle capable of transporting antioxidants to the midbrain via the intranasal route, our objective was to evaluate the therapeutic efficacy and neurogenesis in SNpc and bio-distribution of intranasally delivered sEV-encased phloroglucinol in an in vivo PD model, aiming to develop a promising antioxidant delivery strategy for midbrain-targeted PD treatment. In this study, we investigated the therapeutic effects of intranasally administered sEV-encased phloroglucinol (sEV-Phl) in a chronic MPTP rat model of PD. The antioxidant properties of phloroglucinol within the sEV-Phl formulation were assessed both in vitro, using a H₂DCF.DA ROS assay in SH-SY5Y cells treated with 6-OHDA; and in vivo, through lipid peroxidation analysis in the midbrain of MPTP-treated rats. We evaluated the impact of sEV-Phl on both motor and non-motor behaviors at intervals from week 1 to week 4 post-MPTP administration, in addition to quantifying TH-positive neurons and neurogenesis in the SNpc using immunohistochemistry. We also measured neurotransmitter levels in the midbrain at corresponding time points. Near-infrared organ imaging was employed to track the biodistribution of sEV throughout the study, while PKH-26 labeling and confocal imaging validated the association of sEV-Phl with dopaminergic neurons, astrocytes, and microglia in the SNpc. We also examined the effect of sEV-Phl on the expression of the pro-inflammatory marker TNF-α in this region. This is the first reported study to utilize DPSC-derived sEV as a delivery vehicle for phloroglucinol, and demonstrates significant improvements in behavioral outcomes, increase in TH-positive neurons due to enhanced neurogenesis, as well as reduction in neuroinflammation in the SNpc.

Methods

Animals and cell lines

The work has been reported in line with the ARRIVE guidelines 2.0. Male Albino Wistar rats of 2.5 months, weighing within 280–300 g, were utilized in all in vivo experiments. These animals were bred, raised, and acquired in accordance with the guidelines set by the Institutional Animal Ethics Committee (IAEC) of NIMHANS, Bengaluru, India, and aligned with the regulations of the Committee for Control and Supervision of Experiments on Animals (CCSEA), India. Approval for this study was also obtained from the Institutional Biosafety Committee (IBSC) and IAEC for the use of MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine) as the inducing agent to replicate the sporadic model of PD in rats. All rats were caged in a 12 h light/ dark cycle standard condition, the temperature 22 ± 2 °C and the 55%−60% humidity with free access to standard rodent diet and sterile water. The SH-SY5Y secondary cell line was procured from the National Cell Repository facility at the National Centre for Cell Sciences (NCCS), Pune, and its authentication report, confirming alignment with the ATCC reference cell line, is presented in the supplementary Fig. 1. These cells were cultured in Dulbecco’s modified Eagle’s medium/F12 (DMEM: F12) supplemented with 10% fetal bovine serum (FBS), 1.25% GlutaMAX, and 1% penicillin/streptomycin, and maintained at 37 °C with 5% CO₂.

Human dental pulp stem cell (DPSCs) culture and characterisation

The Institutional Committee for Stem Cell Research (IC-SCR) at NIMHANS granted approval for the study utilizing commercially procured human DPSCs from HiMedia Laboratories, Mumbai, India. As reported in our earlier work [30] four vials of DPSCs, each from a different donor and at Passage 2, were acquired and cultured in KnockOut™ DMEM supplemented with 10% Fetal Bovine Serum (FBS), 1.25% GlutaMAX, and 1% Penicillin-Streptomycin (Invitrogen). These cells were subsequently expanded until reaching passage 4, at which point they were pooled together in equal proportions (1:1:1:1) during trypsinization before being plated at passage 5 and propagated till passage 10 [30].

As per Shekari et al., (2023) [31] the cell identity was provided by characterization of the DPSCs through immunophenotyping of mesenchymal stromal cell markers and HLADR using flow cytometry and tri-lineage differentiation to osteocytes, adipocytes and chondrocytes. Cells were harvested using 0.25% trypsin-ethylenediaminetetraacetic acid (EDTA; GIBCO-BRL) and resuspended in wash buffer containing 0.01% sodium azide in ice-cold phosphate-buffered saline (PBS). The cells were blocked with 2% FBS and incubated at room temperature for 10 min. Immunophenotyping for MSC specific CD markers was performed using indirect staining method and HLADR by direct staining methods [30].

For direct staining, cells were incubated with mouse anti-human HLADR conjugated with Allophycocyanin (APC) for 45 min in the dark at 4 °C, followed by two washes with wash buffer. For indirect staining, the primary antibodies CD73, CD90, and CD105 (Thermo Fisher, n = 5 biological replicates) were used, followed by the appropriate secondary antibodies, such as anti-mouse IgG or anti-rabbit IgG conjugated to Alexa 488 (Abcam). Staining with primary antibodies was conducted in 3% bovine serum albumin (BSA, Sigma-Aldrich, Saint Louis, MO) overnight at 4 °C, followed by secondary antibody incubation at room temperature for 30 min. Each incubation step was followed by washing with a buffer solution containing PBS, 0.01% sodium azide, and 1% FBS. Isotype controls, including Mouse/Rabbit IgG1-FITC and APC (Abcam), were used for verification. Flow cytometry analysis was conducted on a FACS Verse system (BD Biosciences), with data collected from 10,000 gated events and analyzed using BD FACSuite software.

To assess the multilineage differentiation potential of DPSCs, the cells were cultured in adipogenic and osteogenic induction medium (Invitrogen) for 21 days. The protocol was adapted from our previously published report [32]. On the 22nd day, the cells were fixed in 4% paraformaldehyde (PFA) and stained with Oil Red-O for adipocytes and Alizarin Red for osteoblasts. For chondrogenic differentiation, DPSCs were harvested using trypsinization and subsequently pelleted. After cell counting, 5 × 10⁵ cells were used for chondrogenic induction via the pellet culture method. The cells were cultured in chondrogenic medium composed of KODMEM supplemented with 10% FBS, 1% Penicillin-Streptomycin, GlutamaX, and 10 ng/ml TGF-β. The 3D pellets were maintained in this medium for 21 days [33]. Differentiated cells was then fixed with 4% PFA at RT for 45 min. After fixation, Safranin staining was performed to assess the presence of proteoglycans. Images were taken using EVOS M5000 microscope with a standard RGB filter.

Small extracellular vesicles (sEV) isolation and characterisation

As per Shekari et al., (2023) [31], the details of culture media composition, culture condition and the cell density obtained for collection of sEV are as follows: DPSCs were cultured in KnockOut™ DMEM supplemented with 10% FBS, 1.25% GlutaMAX, and 1% Penicillin-Streptomycin in 100 × 15 mm diameter tissue culture plates within a CO₂ incubator maintained at 37 °C, 5% CO₂, and 95% humidity until cells reached 75% confluency. At this point, the medium was replaced with serum-free KnockOut™ DMEM containing 1.25% GlutaMAX, and incubated for 72 h before collecting the conditioned media. Cell viability was assessed using trypan blue exclusion, which confirmed an approximate cell density of 1.27 × 10⁴ cells/cm². The conditioned medium was collected and proceeded with isolation using Optima XPN-90 Ultracentrifuge (Beckmann Coulter). The isolation was performed using Sucrose-gradient Ultracentrifugation method [34, 35] and the sEV were characterized as per MISEV23 [36]. The culture supernatants were cleared of the floating unadhered cells and cellular debris by centrifugation at 300 g for 10 min at 4 °C, followed by removal of large vesicles by centrifugation at 10,000 g for 30 min at 4 °C. The supernatant thus collected was subjected to filtration using 0.2 μm syringe filters so as to exclude the > 0.2 μm EV. For the density gradient-based ultracentrifugation, the filtered supernatant thus collected from the previous centrifugation was loaded slowly over 3 mL of 30% sucrose solution prepared in 1x phosphate-buffered saline, (PBS), forming a layer, and centrifuged at 1,00,000 g, 4 °C for 70 min. The supernatant was then discarded and the sucrose layer (~ 5 mL) was resuspended in 5 ml of PBS and ultracentrifuged at 1,00,000 g at 4 °C for 70 min to pellet down the sEV. After this, the sEV were resuspended in 500 µL sterile PBS and stored at − 80 °C for further use. The recovery of sEV was estimated by measuring the protein concentration using the Nanodrop and stored at -80 °C for the experiment.

Characterisation of expression of sEV CD markers via flow cytometry

Flow cytometry analysis was performed using BD FACS Verse (BD Biosciences). sEV were identified by light scatter for 10,000 gated events, at a low flow rate and analysed using BD FACSuite software (BD Biosciences). sEV (100 µg) after the isolation was fixed in 2% PFA. Then it was pelleted down at 1,00,000 g using PBS with 0.01% sodium azide (wash buffer) for 70 min at 4 °C and resuspended using 200 µl of PBS + 3% BSA (blocking reagent), followed by overnight incubation with primary antibodies (1:50) for the CD markers (CD9, CD81 and CD63; Thermo Fischer Scientific; n = 5 biological replicates). This was followed by incubation with appropriate secondary antibodies anti-mouse IgG or anti-rabbit IgG coupled to AlexaFluor488 (1:100) (Abcam, Cambridge, UK) for 2 h at RT. Each incubation step was followed up by a washing step to remove the excess antibodies with a wash buffer. To avoid the possibility of detecting artifacts and background noises, the wash buffers and sheath fluid were filtered for the use in the experiment. Unstained sEV, along with Mouse/rabbit IgG1 coupled to AlexaFluor488 (Abcam, Cambridge, UK) was used as an Isotype control [30].

Transmission electron microscopy

The samples were prepared following a previously published protocol [37], with some minor modifications. The formvar/carbon-coated grids were coated with 1% alcian blue for 5 s and dried. These grids were then coated with 20 µl of sEV washed distilled water and allowed to dry for 30 min. Further, the grids were stained with aqueous 2% uranyl acetate for 5 s for negative staining. Grids were scanned at 80 kV using JEM-1400Plus, JEOL (Tokyo Japan) Transmission Electron Microscope (TEM) & images were acquired with inbuilt Gatan SC1000B camera.

Dynamic light scattering for size distribution

Dynamic light scattering was performed using DELSA Max Pro (Beckman Coulter) using sEV resuspended in filtered PBS against only PBS as the blank. 50 µl of the sample of 1 mg/ml protein equivalent concentration of sEV was used for the experiment. The readings were acquired in Delsa Max software, for each trial 5 acquisitions were taken where percent polydispersity (Pd) < 20%.

Encapsulation of phloroglucinol in sEV via sonication: (sEV-Phl formulation)

Phloroglucinol encapsulation into sEV was performed by sonication. Pre-aliquoted sterile 500 µl of 1x PBS containing sEV at a concentration of 550 µg/ml was sonicated with 1 mg/ml phloroglucinol at a saturation concentration. The sEV-Phl mixture was sonicated using a Vibra Sonics instrument with the following settings: 30% amplitude, 3 cycles of Pulse 30s on and 2 min off at RT as a cooling period between each cycle. This allows the sEV to maintain their structural integrity. To remove unencapsulated drug, the mixture was diluted in 15 ml of PBS and centrifuged at 100,000 × g for 70 min. The supernatant containing excess drug was completely discarded, and the drug-loaded sEV pellet was resuspended in 500 µl of 1X PBS and sEV protein content was estimated using the nanodrop. The size distribution and image characterisation of the formulation was done following the same procedure as for the sEV using DLS and TEM.

Total phenol content estimation

The sEV-Phl formulation was re-sonicated under the same conditions as incorporation, undergoing 6 cycles of sonication without the cooling period, to rupture the sEV and release the phenol into the solution mixture. The total polyphenol content in sEV-Phl was detected using Total polyphenols Quantification Assay Kit (My BioSource) as per manufacturer’s protocol. The underlying principle of the assay states, tungsten-molybdenum acid can be reduced by phenols under alkaline conditions and produce blue compounds, which has a characteristic absorption peak at 760 nm. The total phenol content in any sample can be calculated by plotting the absorbance against the standard and blank. Standards were prepared from a stock of 1 mg/ml (O-dihydroxy benzene) solution to a serial concentration ranging from 20 to 150 µg. Reagent A -was added to each tube followed by the addition of Reagent B (Alkaline Working Solution) after 2 min. The volume was made up with double distilled water and allowed to stand for 10 min at RT followed by the measurement of absorbance at 760 nm. The concentration of the samples (n = 8 biological replicates) was calculated against the standard curve.

sEV and sEV-Phl uptake in vitro

SH-SY5Y neuronal cells is a common model for in vitro evaluation of neuroprotective effects of drugs. sEV-Phl formulation was stained with PKH26 (2 µmol) in 200 µl of Diluent C for 5 min and neutralised with equal quantity of FBS, diluted 10 times with PBS and subjected to three rounds of sucrose gradient ultracentrifugation until the supernatant appeared completely clear, ensuring that only membrane-integrated dye remained, and any free dye was eliminated. Also, as a negative control, we used the supernatant liquid (volume same as used for sEV) into the culture media of cells to ensure no free dye remained to label the cells. SH-SY5Y cells were seeded into 24 -well plate (50,000 cell/well), cultured for 48 h, and then incubated with different concentrations of PKH26-labeled sEV-Phl ranging from 10 to 100 µg of sEV equivalent protein 3 h. For longer time point of 48 h sEV-Phl of 70 µg sEV equivalent protein was used. Following the incubation, the cells were washed three times with ice-cold PBS and fixed in 4% PFA, followed by staining with β tubulin III (Abcam, 1:200 dilution), followed by a fluorescent tagged secondary antibody anti-mouse IgG coupled to AlexaFluor488 (Thermo Fischer Scientific) and nuclei was counterstained using 4’,6-diamidino-2-phenylindole (DAPI). Cells were immunostained using two different combinations: CD63 with PKH-26 labelled sEV, and PKH-26 labelled sEV with β tubulin III. For the CD63 and β tubulin III/PKH-26 visualization, CD63 and β tubulin III was detected using AlexaFluor488. Accumulation of fluorescently-labelled sEV at different concentrations was visualized by a fluorescence microscopic system Leica 348. Digital images were obtained using the EMCCD camera (Andor) and processed in Image J software. The images for 48 h incubation of sEV-Phl at 70 µg was acquired in Zeiss Confocal Microscope using the laser scanning technology and processed in Image J software.

Assessment of ROS levels upon addition of sEV and sEV-Phl in SH-SY5Y cells under 6-OHDA stress

Cells were plated at a density of 1.5 × 10⁵ cells per well in a 24-well plate. To induce stress, a 6-OHDA treatment at a concentration of 70 µM was applied for 48 h. Basal ROS levels were measured in SH-SY5Y cells as a control, while ROS levels were also assessed in SH-SY5Y cells treated with 6-OHDA and in cells under 6-OHDA stress treated with 70 µg sEV-equivalent protein of sEV and sEV-Phl. After washing, equal numbers of live cells from all experimental groups were collected and incubated with 10 µM H₂DCF-DA (molecular probes, Invitrogen) at 37 °C for 15 min. H₂DCF-DA diffuses across the cell membrane and is hydrolyzed by intracellular esterases to the non-fluorescent dichlorofluorescein (H₂DCF). Cellular ROS oxidizes H₂DCF to form a highly fluorescent compound, 2,7-dichlorofluorescein (DCF). The fluorescence intensity of DCF (excitation/emission: 485 nm/535 nm) was measured using a TECAN Spark spectrophotometer with Magellan software, and results are presented as mean ± SD for n = 4 biological replicates.

Induction of PD, intranasal delivery and confirmation by behavioural and electrophysiological studies

The work has been reported in line with the ARRIVE guidelines 2.0. The rats were divided, using a simple randomization method. Sample size was selected using ‘resource equation method’. Adult male Wistar acquired from NIMHANS CARF was administered intranasally with a single bilateral dose of 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine (MPTP) (Sigma Aldrich, Merck- Germany) 0.1 mg/nostril per rats weighing 280–300 g of 2.5 months age [38]. Age and weight-matched rats were selected for each experimental group before MPTP or saline nasal administration in control animals. Prior to intranasal administration, rats were anesthetized and positioned with their necks extended to orient the nostrils upward. We then delivered 10 µl of the experimental solution (MPTP, vehicle control, sEV, or sEV-Phl) into each nostril using a micropipette. To ensure maximum absorption and prevent loss of solution, rats were maintained in this position for 2 min before returning them to their cages. This technique leverages gravity to facilitate solution flow into the nasal passages while minimizing solution loss through sneezing or external spillage.

The animals were divided into the following groups:

1. Control (Bi-lateral intra-nasal saline administration).

2. MPTP (induced PD rats; Bi-lateral intra-nasal 0.1 mg/nostril MPTP administration).

3. MPTP + sEV [Bi-lateral intra-nasal 0.1 mg/nostril MPTP administration, followed by sEV administration at a dose of 250 µg/kg body weight of animals at Day 3 post MPTP treatment (PMT)]

4. MPTP + sEV-Phl [Bi-lateral intra-nasal 0.1 mg/nostril MPTP administration, followed by sEV loaded Phloroglucinol administration at a dose of 250 µg/kg body weight of animals at Day 3 post MPTP treatment (PMT)]

A total of 258 animals were used in this study. For behavioural studies 8 animals/group and ELISA 7 animals/group were assigned separately for each time points within the groups. For IHC and biodistribution studies, 6 animals were allocated to each timepoints of the groups. No experimental units were excluded.

Rats were anesthetized using ketamine and xylazine (80:5 mg/kg, intraperitoneally) for the experiments and euthanized with an overdose of isoflurane, followed by decapitation for terminal procedures. This included the collection of blood serum and brain samples for analyses such as midbrain neurotransmitter measurement, TNF-α levels, and immunohistochemistry. For all animal experiments multiple investigators were involved. Random numbers were generated by the first investigator (ID) who was only aware of the specific groups. Experiments and analysis were conducted by KM, RG, AM, GW, RS and NB. All protocols were reviewed and approved by the institutional animal ethics committee of the National Institute of Mental Health and Neurosciences (Bengaluru, India).

Behavioural studies

Prodromal symptoms

1. Olfactory discrimination: The olfactory discrimination ability of rats was evaluated following the method outlined by [39]. In this task, each rat was placed in a cage divided into two identical compartments (30 × 30 × 20 cm) with an open door between them, allowing the rat to choose between a compartment with fresh husk and another familiar compartment containing unchanged husk that the rat had occupied for 48 h prior. Each trial lasted 5 min, and the assessments were conducted at Day 3, week 1, week 2 and week 4 PMT and Time spent in familial cage (minutes) was plotted as mean ± SD for n = 8 biological replicates.

2. Nerve conduction velocity (NCV): NCV was measured non-invasively as described in our previous study [40]. iWorx data acquisition and analysis system and LabScribe software (iWorx Systems, Inc., USA) was used to obtain the readings and analyse the data. Rats were anesthetized with ketamine: xylazine (80 mg/kg:5 mg/kg; i.p.) and placed on a warm bed. The sciatic nerve was stimulated at the sciatic notch with a supramaximal stimulus of 8 V at 20 Hz. Recording electrodes was placed at first interosseous muscle of the hind paw. Latency was calculated from the stimulus artifact to the onset of the negative M-wave deflection. NCV was calculated by subtracting the distal latency (t2) from the proximal latency (t1) and the results was divided by the distance between the stimulating and recording electrodes, which is represented by the formula: NCV = distance (mm) / latency. NCV measurement was performed at Day 3, week 1, week 2 and week 4 PMT, and plotted was plotted as mean ± SD for n = 8 biological replicates.

Motor symptoms

1. Rotarod test: Each rat was trained to perform the task with three training trials for 30 min each day at a speed of 15 rpm for two days followed by 25 rpm on the 3rd day for acclimatization. In the test session on 4th day, the rats were placed on the rotarod and their performance time was recorded [30] Motor coordination was assessed using Rotamex-5 (Columbus Instruments). Measurements were performed at Day 3, week 1, week 2 and week 4 PMT. All readings were taken in triplicate. The rotarod performance time (s) was plotted as mean ± SD for n = 8 biological replicates.

2. IR actimeter: The digital IR actimeter was used to test the locomotor activity of PD rats in which a continuous beam of light falls on photoelectric cells. The apparatus comprises of a frame provided with 16 IR source on X axis and 16 IR source on Y axis creating a 16 × 16 grid. The instrument control panel displays the number of beam brakes by the animal on all axis and total of all in the actimeter. Since the instrument comes with the facility of Automatic Reading averaging in experiment report, the result obtained are free of human error and bias. Any interruption in the continuity of light by the animal was recorded on a digital counter in the form of counts which corresponds to the locomotor activity. The animals from each group were individually placed in the apparatus and allowed to move freely for 5 min while the beam breaks were recorded automatically and report generated digitally. Measurements were performed at Day 3, week 1, week 2 and week 4 PMT and number of beam breaks in 5 min was plotted as mean ± SD for n = 8 biological replicates.

3. Open field test: To assess the anxiety like behaviour of rats upon MPTP injection in comparison to the control, sEV treated and sEV-Phl treated groups the rats were subjected to an open field test. The apparatus consisted of a square cardboard open field arena with a white floor 100 × 100 cm wide, divided into 25 squares of 20 × 20 cm measurement. The whole apparatus was covered by a wall 25 cm high. Each rat was placed in the centre of the field and the number of squares crossed in a runtime of five minutes was eventually recorded. Usually, the exploratory nature of the rats reduces with the progression of the diseases. Measurements were performed at Day 3, week 1, week 2 and week 4 PMT and the number of squares covered in 5 min was plotted as mean ± SD for n = 8 biological replicates.

Assessment of neurotransmitter levels and lipid peroxidation in the midbrain

Midbrain lysate preparation

Rats from each group (n = 7 biological replicates) were sacrificed after each time point and midbrain was dissected out, weighed and homogenized in 1x PBS at 4 °C as per the protocol provided by the Kit instruction manual. The sample was then centrifuged at 3000 rpm and supernatant was carefully removed and stored in -20 °C as separate aliquots of 50 µl for the following neurotransmitter assays. To maintain the uniformity of the tissue homogenate across the samples at each time points, the concentration of the tissue lysate was kept constant at 200 mg/ml.

Assay of dopamine, norepinephrine, acetylcholine, GABA and TNF-α using ELISA kit

The assay for dopamine, GABA, norepinephrine, and acetylcholine was conducted according to the manufacturer’s protocol (Sunlong Rat ELISA kit). Samples were mixed with the sample dilution buffer and added to the pre-coated wells provided in the kit. The plate was incubated for 30 min at 37 °C. Following incubation, the wells were washed with diluted wash buffer as specified in the protocol. HRP-conjugate was then added to each well, except the blank, and incubated for another 30 min at 37 °C. Subsequently, the wells were washed again with wash buffer, and chromogen solutions A and B were added to each well, followed by a 15-minute incubation at 37 °C. A stop solution was then added, and the optical density (OD) was measured at 450 nm using a Microtiter Plate Reader within 15 min of adding the stop solution.

For quantification, a standard curve was generated by plotting the known concentrations of rat dopamine, norepinephrine, acetylcholine, GABA, and TNF-α standards against their corresponding OD values on a log-log scale (x-axis for concentrations and y-axis for OD). The concentrations of neurotransmitters and the immune marker in the samples were determined by plotting the sample’s OD (subtracting the blank OD) on the Y-axis using the standard curve.

Estimation of lipid peroxidation (MDA levels)

In vivo ROS generation in the brain leads to the formation of lipid peroxides, which rapidly decompose into a more stable compound, malondialdehyde (MDA). MDA levels in midbrain lysate (n = 6 biological replicates) were measured using the TBARS (Thiobarbituric Acid Reactive Substances) assay, combined with Butylated Hydroxy Toluene (BHT) as described by [41]. The assay is based on the colorimetric detection of the MDA-TBA adduct, which forms under acidic conditions in the presence of HCl when heated to 80–100 °C. The MDA-TBA adduct is then quantified by measuring absorbance at 532 nm [42].

Immunohistochemistry of SNpc region of rat brain via colorimetric method

The rat brains for each group was collected at weeks 1, 2, and 4 post-MPTP administration (n = 6 biological replicates) were processed for paraffin-embedded immunohistochemistry as follows. First, the brains were manually dissected to obtain coronal sections of the midbrain and fixed in 10% formalin for 24 h. The samples were then dehydrated using a graded series of alcohol (60%, 70%, 80%, 90%) followed by three rinses in fresh absolute alcohol, two rinses in chloroform, and two rinses in paraffin. Each step involved 1.5 h of incubation. After processing, the brain tissues were embedded in paraffin blocks, and 4–5-micron-thick sections were cut using Leica’s Histo-core Multicut Microtome. These sections were mounted on glass slides and subjected to overnight incubation at 4 °C with the primary antibody, TH (PAB438Ra01), raised in rabbit. After incubation, the sections were washed twice in Tris buffer, treated with Secondary PathnSitu-target binder for 15 min, washed twice again with Tris buffer, and then treated with Polymer Plus for 15 min. Following two more Tris buffer washes, the sections were incubated with DAB substrate solution for 2–5 min to develop colour and immediately placed under running tap water to stop the reaction. The sections were then counterstained with haematoxylin for 30–60 s. They were then cleared, mounted with coverslip, dried and taken for visualisation and image acquisition.

Immunohistochemistry of SNpc region of rat brain for confocal imaging

Rats underwent transcardial perfusion prior to brain tissue isolation for immunohistochemistry (IHC). The procedure involved perfusing 350–400 ml of 0.9% saline to flush out blood from the circulatory system. Once the outflow ran clear, indicating effective blood removal, 350 ml of 4% paraformaldehyde (PFA) was perfused to fix the tissues [43, 44]. Following perfusion, the brains were carefully dissected and immersed in 4% PFA at 4 °C for 7 days to allow tissue hardening before sectioning. Coronal sections, 50 μm thick, of the midbrain were prepared using a Leica 1200 S vibratome. IHC was performed on free-floating sections using a 24-well plate. For antigen retrieval, the sections were incubated at 65 °C in a water bath for 2.5 h, immersed in 200 µl of antigen retrieval buffer, consisting of an equal mix of saline-sodium citrate (SSC) buffer and formamide. Permeabilization and blocking were carried out using a solution containing 5% BSA, 5% goat serum, and 0.1% Triton-X for 2 h at room temperature. After each step, the sections were washed twice for 5 min with 1X PBS on a rocker. Following blocking, the brain sections were incubated overnight at 4 °C in a primary antibody solution (1:200 dilution) prepared in 5% BSA. The next day, after two PBS washes, the sections were incubated with a goat-derived secondary antibody (1:300 dilution) for 6 h at room temperature, followed by a counterstaining with 5 µg/ml 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) for 1 min. The RRID numbers for the antibodies used are given in supplementary Table 1. The slices were then mounted using DABCO (Sigma) as the mounting reagent. Following mounting, tissue sections were visualized using the ZEISS Axio Observer.Z1/7 and LSM 980 confocal microscope, equipped with a Plan-Apochromat 40X/1.3 Oil DIC (UV) VIS-IR M27 objective. To investigate the association of sEV and sEV-Phl with dopaminergic neurons, microglia, and astrocytes, sEV and phloroglucinol loaded sEV (sEV-phl) were labelled with PKH-26 (red), as previously described, and administered intranasally to MPTP-induced PD rats on Day 3 post-treatment. The brain samples from these rats (n = 6 biological replicates) were then processed for IHC as outlined. Image acquisition was performed using ZEISS ZEN 3.7 software (RRID: SCR_013672), employing a fluorescence contrast method. Detector gain was set to 650 V, and the pinhole was adjusted as follows: 1 AU/30 µm for DAPI, 1 AU/36 µm for FITC, 1 AU/45µm for AF647, and 1 AU/45 µm for PKH26. The excitation/emission settings were 405/495 for DAPI, 488/517 for FITC, 650/665 for AF647, and 551/567 for PKH26. Images were captured at a resolution of 2791 × 2791 pixels with an 8-bit depth and an effective NA of 1.3, then analyzed using ImageJ.

BrdU administration in rats and immunohistochemistry of SNpc region of rat brain for fluorescence imaging

To evaluate cell proliferation in the SNpc region, control, MPTP-treated PD, sEV-treated, and sEV + Phl-treated rats were administered BrdU (5’-bromo-2’-deoxyuridine; Sigma) at a dose of 50 mg/kg once daily for four consecutive days before sacrifice at the 2-week post-MPTP (PMT) timepoint [45]. Rats administered with BrdU were perfused as previously described, and brain tissues were sectioned using a Leica 1200 S vibratome to obtain 50 μm thick slices. The sections were stained and then visualized, and imaged using a Leica DMi8 fluorescence microscope equipped with a Semi-Apochromats 40×/0.80 PL FLUOTAR objective and a Leica DFC3000G camera. Six animals were included in each group (n = 6 biological replicates). Image processing was carried out using Leica Application Suite X (LasX) software (RRID: SCR_013673).

NIR Dye labelling and tracking sEV-Phl in SH-SY5Y cells and MPTP induced PD rats

Ex vivo organ imaging

Approximately 2000 µg of protein equivalent of sEV was first pelleted down using Ultracentrifuge at 100,000 g at 4 °C for 70 min and then resuspended in 500 µl (400 µg/ml) of Diluent C provided in the CellVueNIR-815 cell labelling kit, and 0.5 µL of the Dye was added to the suspension. The suspension was incubated for 5 min at RT with occasional stirring to ensure that sEV are not forming aggregates and the sEV membrane is exposed to the dye for binding. Knock-out serum replacement (Gibco) was used at 1:1 ratio with the diluent C to neutralise the reaction. The suspension was then diluted in 10 ml PBS and centrifuged at 100,000 g at 4 °C for 70 min to remove the excess unbound dye. sEV-Phl was labelled with NIR dye CellVue 815 (LICOR Biosciences, USA) as per the manufacturer’s instruction. Then NIR dye labelled sEV-Phl were administered through intranasal route to MPTP induced PD rats (n = 6 Biological replicates) at Day 3 PMT. Rats were sacrificed with an overdose of isoflurane and all the vital organs- heart, lungs, liver, spleen, pancreas, kidneys and brain were dissected. We administered NIR dye-labelled sEV-Phl intranasally to both control and MPTP-treated rats, then measured signal intensity in the brain 24 h after administration (Day 4 post-MPTP treatment). This timing allowed us to evaluate the targeted delivery and retention of the therapeutic sEV in the brain tissue under both normal and neurotoxin-induced pathological conditions. Images were acquired in 800-nm channel and white light and represented as heatmaps of the same. Signal intensity emitted by each of the organs were measured using ImageStudio5.2 software (LI-COR Bio-sciences). Relative fluorescence intensity was calculated by specifying a region of interest (ROI) around the fluorescence signal, normalized with its area and represented as signal intensity.

Detection of chemokine marker-CXCR4 in DPSC and DPSC- sEV via flow cytometry

DPSCs from P11 were trypsinized and fixed in 2% paraformaldehyde (PFA), blocked with 3% BSA and incubated with primary antibody against CXCR4 (Thermo Fisher Scientific) (1:100 dilution) overnight at 4 °C. After the incubation, cells were washed once to remove the excess antibodies, and incubated again with Fluorescein isothiocyanate (FITC)-tagged secondary antibody (1:200 dilution) at RT for 45 min. Further cells were washed PBS and resuspended in sheath fluid for flow cytometry analysis. Unstained cells incubated only with the secondary antibody were used as isotype control for gating. Flow cytometry was performed using FACS Verse (BD Biosciences) and the data was analysed using FACS Suite software (BD Biosciences).

sEV were collected from same batch of cells were and then fixed in 2% PFA for this procedure. After incubation with primary antibody CXCR4, the same protocol was followed as mentioned for the CD marker expression of sEV in the characterisation panel. Flow cytometry was performed using FACS Verse (BD Biosciences) and the data was analysed using FACS Suite software (BD Biosciences).

Statistical analysis

Data are represented as mean ± standard deviation. One-way or two-way analysis of variance (ANOVA) was used, as per requirements, followed by Bonferroni post-hoc analysis using R software (R Foundation, Vienna, Austria) or Sigma Plot12.5 (Systat Software, Inc, San Jose, CA, USA). Sigma Plot12.5 or Prism5 (GraphPad Software Inc, San Diego, CA, USA) was used to prepare all graphs. For behavioural studies, we employed one-way ANOVA followed by Bonferroni post-hoc analysis, while neurotransmitter levels were analysed using two-way ANOVA with Bonferroni post-hoc testing. Statistical significance between experimental groups is denoted by symbols, with a single symbol indicating P < 0.05, double symbols representing P < 0.005, and triple symbols signifying P < 0.001.

Results

Characterisation of human DPSCs and DPSC-derived sEV

As per the position paper [31] Shekari et al., (2023) the identity of the cells from which the sEV were isolated was performed. Commercially purchased DPSCs were characterized through immunophenotyping for mesenchymal markers and assessed for multipotency via osteogenic, adipogenic and chondrogenic differentiation. Supplementary Fig. 2A shows a phase contrast image of a confluent DPSC culture. As depicted in supplementary Fig. 2B, FACS analysis revealed that DPSCs were immunopositive for mesenchymal surface antigens CD73 (97.75 ± 0.05%), CD90 (99.00 ± 0.05%), and CD105 (98.49 ± 0.74%), and were not immunopositive for HLADR. The DPSCs also demonstrated the potential to differentiate into osteogenic and adipogenic lineages. By the 21st day of induction, the induced cells were positive for Alizarin Red and Oil Red O staining, indicating osteogenic and adipogenic differentiation respectively, as shown in supplementary Figs. 2C&D. Safranin O staining showed the presence of proteoglycans (Red) indicating successful differentiation to chondrocytes (supplementary Fig. 2E). In contrast, control DPSCs without induction cues were negative for all the three stains (supplementary Fig. 2C, D&E).

sEV were isolated from the conditioned medium of DPSCs through sucrose gradient ultracentrifugation [35]. In TEM, cup-shaped vesicles were observed (Fig. 1A), which are in line with the classical cup-shaped morphology of sEV. Additionally, the size distribution of the sEV analysed using DLS revealed that the radii ranged from 45 nm to 90 nm, and that the peak percentage intensity of scattered light occurred at 50.80 nm radius, showing the uniformity of the sEV with respect to size (Fig. 1B). An average yield of 1230.80 ± 53.55 µg/ml of protein was obtained for sEV isolated from 18 million cells in each isolation showed the consistency of the yield of sEV from DPSCs (Fig. 1C). FACS analysis showed that the sEV were immunopositive for CD 81 (98.07 ± 0.07%), CD 63 (97.05 ± 0.22%) and CD 9 (98.62 ± 0.18%) (Fig. 1D).

Fig. 1.

Characterisation of DPSC derived sEV. (A) TEM image of sEV isolated from DPSCs. (B) Size distribution of the sEV measured by dynamic light scattering with the peak intensity of scattered light at radii 52.8 nm. (C) Table represents the protein concentration of sEV from 10 isolations with a mean of 1230.80 ± 53.55 µg/ml from 18 million cells. (D) Flow cytometry analysis of isolated sEV for the surface expression of tetraspanins classical exosome markers CD81, CD63 and CD9. Percent positive for each marker is mentioned in the figure as mean ± SD (n = 5 biological replicates) and isotype control was used for gating

Characterization of phloroglucinol (Phl) loaded sEV

The amount of phloroglucinol incorporated in the sEV were estimated by in vitro release through sonication using the total phenol assay kit. An 8.53 ± 0.32 µg/ml of polyglycol content was obtained from 3 µg/ml protein equivalent of Phl loaded sEV (sEV-Phl, Fig. 2A); approximately a ratio of 1:3. The protein concentration of sEV is indicated for all experiments. Further, the size of sEV after loading with Phl by sonication was measured by DLS and visualized using TEM. Fig. 2B&C represents the TEM images of sEV before and after phloroglucinol incorporation through sonication. Similar size range of sEV were observed between sEV and sEV-Phl. The distribution of sEV size before and after phloroglucinol incorporation detected by DLS is represented in Fig. 2D&E. The peak percentage intensity of scattered light for only sEV was at 50.80 nm, and for sEV-Phl it was at 51.80 nm. The TEM images and DLS results confirm that sonication and incorporation of phloroglucinol did not change the size of the sEV.

Fig. 2.

Characterization of phloroglucinol (Phl) loaded sEV (sEV-Phl). (A) The table represents the phloroglucinol concentration released from the sEV-Phl through sonication. A mean of 8.53 ± 0.32 µg/ml was obtained from 3 µg/ml equivalent protein of Phl loaded sEV (sEV-Phl); approximately a ratio of 1:3 (n = 8 biological replicates). (B, C) TEM image of sEV (sEV) and sEV-Phl. (D, E) Size distribution of sEV and sEV-Phl respectively measured by dynamic light scattering

sEV-Phl associates with SH-SY5Y cells and reduces ROS levels in these cells under 6-OHDA stress

Exosomal uptake by the neuronal cells was evaluated for different protein concentrations of sEV-Phl in SH-SY5Y cells under control culture conditions. As represented in Fig. 3A, after 3 h of exposure to PKH26 labelled (red) sEV-Phl, a distinct association of sEV-Phl was observed at the dosing concentration of 10 µg/ml of sEV-Phl. In the presence of 70 µg/ml and 100 µg/ml of sEV-Phl a similar amount of association was observed, showing a saturation in the exosomal uptake in the SH-SY5Y cells. After 48 h at 70 µg/ml concentration, the confocal images showed distinct cytosolic localization of the PKH-26 labelled sEV-Phl in ß tubulin III immunostained SH-SY5Y cells (Fig. 3B, C). PKH-26 labelled sEV appear as punctate and also as clusters in the cytosol. EV aggregation or “clumping” is a well-documented phenomenon, attributed to their lipophilic nature, particularly when suspended in polar solvents or within the cytosol (which is not lipid-based) and a similar pattern has been reported in earlier studies too [46–50]. The Pearson’s correlation coefficient analysis confirmed the cytosolic localization of the sEV-Phl (Fig. 3D) with 0.507 ± 0.05 (n = 6 biological replicates with each 3 technical replicates) as Pearson’s colocalization coefficient of PKH-26 labelled sEV-Phl with ß tubulin III. CD63 is also expressed by SH-SY5Y; however, PKH-26 labelled- sEV show colocalization in some ROIs with exosome marker CD63 (supplementary Fig. 3). Next, the effect of the anti-oxidant Phl in sEV was assessed in vitro by measuring the ROS levels in SH-SY5Y cells treated with 6-OHDA stress. The 6-OHDA treated cells showed significant increase in ROS levels over control (Fig. 3E; P < 0.001), and a significant reduction in the ROS level was observed in the sEV treated group over 6-OHDA treated cells (P < 0.001). The 6-OHDA treated cells in the presence of sEV-Phl showed further significant reduction in ROS levels than that in the presence of only sEV (P < 0.001). In sum, a significantly reduced ROS level was noted in the SH-SY5Y cells under 6-OHDA stress that were co-treated with sEV and sEV encapsulated with Phl. This confirms that sEV-Phl is effectively internalized by cells in vitro, and that the antioxidant properties of the polyphenolic compound remain preserved within the sEV encasement.

Fig. 3.

Association of sEV-Phl with SH-SY5Y cells in vitro and its impact on ROS generation under 6-OHDA-induced stress: (A) Representative immunofluorescence images of sEV-Phl labelled with PKH-26 (red) added in different concentrations for 3 h to SH-SY5Y cells immunostained with ß tubulin III (green). (B) Representative confocal images of PKH-26 labelled sEV-Phl (red; 70 µg/ml protein equivalent) localization in SH-SY5Y cells immunostained with ßtubulin III (green) at 48 h. (C) Magnified images of the insets. Nuclei is counterstained with DAPI. (D) Pearson colocalization coefficient of PKH-26 labelled sEV-Phl and ß tubulin III represented as mean ± SD (n = 6 biological replicates with 3 technical replicates) (E) Plot showing the ROS generation in SH-SY5Y cells under control and 6-OHDA stress condition and when treated with sEV, sEV-Phl along with 6-OHDA stress (n = 4 biological replicates). Statistical significance between control and 6-OHDA and 6-OHDA cells treated with sEV groups is represented by *(***P < 0.001). ^$Represents a significant difference between 6-OHDA and sEV and sEV-Phl-treated groups (^$$$P < 0.001). ^#Represents a significant difference between sEV and sEV-Phl groups (^#P < 0.001)

Association of sEV-Phl with dopaminergic neurons, astrocytes and microglial cells after intranasal administration in MPTP PD rats

To evaluate the association of sEV and sEV-Phl with cells in the substantia nigra pars compacta (SNpc) region of the midbrain, PKH-26 labelled sEV and sEV-Phl was administered intranasally to MPTP-induced PD rats on Day 3 post-MPTP treatment (PMT). IHC was performed to detect dopaminergic neurons (TH), astrocytes (GFAP), and microglia (IBA1) on week 1 and 2 PMT (Fig. 4A, B), corresponding to Days 3 and 11 post- sEV and sEV-Phl administration respectively (Fig. 4). The colocalization of PKH-26 with dopaminergic neurons, astrocytes, and microglia was quantified using Pearson’s coefficient (Fig. 4C, D). In the first week PMT, the colocalization of PKH-26-labeled sEV with IBA-1-positive microglial cells was moderately higher in the MPTP + sEV group compared to colocalization with TH-positive neurons and GFAP-positive astrocytes. However, in the MPTP + sEV-Phl group, the Pearson colocalization coefficient was similar across all three cell types (Fig. 4A, C). The colocalization of sEV-Phl with TH was higher than the sEV alone at week 1 PMT (Fig. 4C; P < 0.05). By the second week PMT, the colocalization of sEV with TH-positive neurons in the MPTP + sEV group significantly increased compared to the first week PMT (Fig. 4C; *P < 0.001). During this period, higher colocalization with TH-positive cells was observed compared to GFAP and IBA-1-positive cells (*P < 0.001). In contrast, in the MPTP + sEV-Phl group at week two PMT, the colocalization coefficient remained consistent among TH, GFAP, and IBA-1-positive cells. For GFAP-immunopositive cells, PKH-26 localization was observed in the nucleus rather than the cytosol, resulting in a negligible Pearson’s coefficient for PKH-26 colocalization with GFAP (Fig. 4A, B). This indicates that both sEV and phloroglucinol-encased sEV were capable of colocalizing with TH, GFAP, and IBA-1-positive cells in the SNpc.

Fig. 4.

Association of sEV-Phl with dopaminergic neurons, astrocytes and microglial cells after intranasal administration in MPTP PD rats: (A, B) Represent the confocal images of SNpc region immunostained with TH, IBA-1 and GFAP along with PKH-26 labelled sEV and sEV-Phl at week 1 PMT (A) and week 2 PMT (B) respectively. The magnified images of two insets each are indicated in the corresponding images. Nuclei is counterstained with DAPI (blue). (C) Represents the plot of Pearson’s colocalization coefficient of PKH-26 labelled sEV and sEV-Phl with TH, IBA-1 and GFAP at week 1 and week 2 PMT (n = 6 biological replicates). Statistical significance between sEV and sEV-Phl for week 1 and 2 PMT are represented by ^# (^###P < 0.001). ^¥Represents a significant difference between week 1 and week 2 PMT for the sEV group (^¥¥¥P < 0.001)

Chemokine receptor expression and effect of sEV-Phl on neuroinflammation and lipid peroxidation

sEV isolated from MSCs and DPSCs are known to express chemokine receptors similar to their parent cells [51]. In this study, FACS analysis revealed that over 96% of our DPSC population was immunopositive for CXCR4, and nearly 68% of isolated sEV also expressed CXCR4 (Fig. 5A). CXCR4 is crucial for homing of MSCs and sEV to injured tissues or cells by detecting pro-inflammatory signals such as TNF-α [52–55]. We observed a significant expression of TNF-α in the MPTP-treated groups at weeks 1, 2, and 4 PMT (Fig. 5B, C). At week 1 PMT, a few cells in the MPTP + sEV and MPTP + sEV-Phl groups expressed TNF-α, which colocalized with PKH-26-labeled sEV and sEV-Phl (Fig. 5B). Confocal images visually confirmed a noticeable reduction in TNF-α expression in the MPTP + sEV and MPTP + sEV-Phl groups, suggesting decreased neuroinflammation in these sEV-treated groups. To further confirm the inflammatory signature under MPTP toxicity and the effects of sEV and sEV-Phl treatment, TNF-α levels were measured in the midbrain and blood serum using ELISA (Fig. 5C). There was a significant increase in TNF-α levels in the MPTP-treated groups across all three time points compared to the control (Fig. 5C; P < 0.001). However, the sEV and sEV-Phl treated groups exhibited a marked reduction in TNF-α levels compared to their respective MPTP groups (Fig. 5C; P < 0.001). As phloroglucinol is a flavonoid with antioxidant properties, we also measured lipid peroxidation levels in midbrain tissues across all experimental groups. A distinct increase in malondialdehyde (MDA) levels was observed in the MPTP-treated groups compared to the control (Fig. 5D; P < 0.001). Notably, the sEV and sEV-Phl treated groups showed significant reductions in MDA levels compared to the MPTP rats (P < 0.001). Furthermore, at weeks 2 and 4 PMT, the sEV-Phl group exhibited a significantly greater reduction in MDA levels than the sEV group alone (P < 0.001). sEV and sEV-Phl thus contributed to reducing the pro-inflammatory state with sEV-Phl exerting a comparatively higher antioxidant effect.

Fig. 5.

Chemokine receptor expression and effect of sEV-Phl on neuroinflammation and lipid peroxidation: (A) Represents the FACS histogram of immunopositive population of CXCR4 in DPSCs and sEV isolated from DPSCs. Percent positive for each marker is mentioned in the figure as mean ± SD (n = 6 biological replicates) and isotype control was used for gating. (B) Representative confocal images of SNpc region of control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups immunostained with TNF-a at week 1, 2 and 4 PMT. PKH-26 labelled sEV are in red and nuclei was counterstained with DAPI (blue) (n = 6 biological replicates). (C) Represents the TNF- α levels from midbrain lysate measured by ELISA for control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups at week 1, 2 and 4 PMT (n = 7 biological replicates). (D) Represents the malondialdehyde (MDA) levels measured for control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups at week 1, 2 and 4 PMT (n = 6 biological replicates). Statistical significance between control versus MPTP or MPTP + sEV or MPTP + sEV-Phl is denoted by *(**P < 0.005, ***P < 0.001). ^$Represents significant difference between MPTP and MPTP + sEV or MPTP + sEV-Phl (^$$P < 0.005, ^$$$P < 0.001). ^#Represents significant difference between MPTP + sEV and MPTP + sEV-Phl (^##P < 0.005)

Effect of sEV-Phl on behavioural parameters

After Day 3 PMT, a marked decrease in time spent in the familiar compartment was observed in the MPTP group compared to the control (Fig. 6A; P < 0.001). This impairment persisted from week 1 to week 4 (P < 0.001). The sEV and sEV-Phl treated rats demonstrated a significant increase in time spent in the familiar compartment starting from week 1 compared to their respective MPTP-treated groups (Fig. 6A; P < 0.001). In the sEV-Phl group, this effect was sustained through week 4 and was comparable to the control group. By week 4, there was an improvement in olfactory discrimination observed in the sEV group as well; however, this was significantly less pronounced compared to the sEV-Phl treated rats (P < 0.05).

Fig. 6.

Effect of sEV-Phl on behavioural parameters: (A) Longitudinal measurement of olfactory discrimination (A), nerve conduction velocity (NCV) (B), rotarod performance (C), open-field test (D) and locomotor activity through actimeter (E) for control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups at week 1, 2 and 4 PMT (n = 8 biological replicates). Statistical significance between control versus MPTP or MPTP + sEV or MPTP + sEV-Phl is denoted by *(**P < 0.005, ***P < 0.001). ^$Represents significant difference between MPTP and MPTP + sEV or MPTP + sEV-Phl (^$$P < 0.005, ^$$$P < 0.001). ^#Represents significant difference between MPTP + sEV and MPTP + sEV-Phl (^##P < 0.005)

A significant decline in Nerve Conduction Velocity (NCV) was observed from Day 3 PMT (Fig. 6B; P < 0.001). This decrease in NCV in MPTP groups was observed throughout, at all the time points. sEV and sEV-Phl administration showed increase in NCV of MPTP-treated rats from week 1 and it retained till week 4 PMT (Fig. 6B; P < 0.001). The NCV levels measured were comparable between sEV and sEV-Phl groups (P > 0.05).

A decrease in rotarod performance time was observed in MPTP-treated rats starting from week 1 PMT, with a significant difference noted from week 2 onwards (Fig. 6C; P < 0.001). The sEV-Phl treatment showed a trend of improvement in rotarod performance time starting from week 1 PMT, with significant differences observed at week 2 (P < 0.001) and week 4 (P < 0.005) compared to the corresponding MPTP group (Fig. 6C). Similarly, the sEV treatment also exhibited improvement in rotarod performance time starting from week 2, which was sustained through week 4 (P < 0.005). At week 2, the rotarod performance time in the sEV-Phl group was significantly higher than in the sEV group (P < 0.001).

In the locomotor activity assessment using the Open Field Test (OFT) and actimeter, a significant decrease in the number of squares covered was observed at week 2 PMT in the MPTP-treated group, which persisted through week 4 compared to control (Fig. 6D). At week 2 PMT, the sEV-Phl treatment demonstrated an improvement in OFT performance (P < 0.05), which further enhanced by week 4 (P < 0.005). In contrast, the sEV treatment alone showed a moderate improvement in OFT performance; however, this improvement did not reach significance compared to the corresponding MPTP-treated groups.

A significant decrease in the number of beam breaks was observed in MPTP-treated rats starting from week 2 PMT compared to the control (Fig. 6E). This decline in the actimeter parameter persisted through week 4 (P < 0.001). The sEV-Phl treatment exhibited a moderate improvement at week 2 PMT, but the improvement was only significant compared to its corresponding MPTP group by week 4 PMT (P < 0.001). Similarly, the sEV group showed a significant improvement in actimeter performance at week 4 PMT (P < 0.005).

Overall, both the sEV and sEV-Phl treatments showed significant protective effects for MPTP-treated rats, with the sEV-Phl group exhibiting higher improvements with earlier onset in motor and locomotor activities compared to the sEV group.

Effect of sEV-Phl on dopaminergic neurons in SNpc

As represented in the micrographs of immunohistochemistry slides of week 1,2 and 4 PMT, a distinct loss in TH immunopositive neurons was observed in the SNpc region. Both the sEV and the sEV-Phl treatment in the MPTP rats showed recovery of the TH immunopositive neurons (Fig. 7A). The higher magnification images in Fig. 7B show the intensity of TH expression in the dopaminergic neurons. The intensity (CTCF values) of TH immunopositive neurons all the time points (Fig. 7C) in MPTP group was significantly lesser than control (P < 0.001). The sEV-Phl treatment shows higher TH expression at week 1 (P < 0.05), week 2 and week 4 (P < 0.001) PMT in comparison to its corresponding MPTP group. At week 1 PMT the sEV-Phl showed higher TH intensity than even the sEV group (P < 0.05). The significant increase of TH intensity was observed from week 2 PMT in the sEV treated group in comparison to MPTP group (P < 0.001). This indicated that sEV and sEV-Phl treatment resulted in protection and recovery of TH immunopositive neurons.

Fig. 7.

Effect of sEV-Phl on dopaminergic neurons in substantia nigra: (A) Representative bright field images of TH immunostained neurons (2.5x) in substantia nigra pars compacta (SNpc) for control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups at week 1, 2 and 4 PMT. (B) Higher magnification images (20x) of the TH immunostained neurons in SNpc for all the experimental groups. (C) Plots showing CTCF values of TH intensity obtained for the experimental groups for all the three time points (n = 54 neurons from 6 biological replicates). Statistical significance between Control versus MPTP or MPTP + sEV or MPTP + sEV-Phl is denoted by *(*P < 0.05. **P < 0.005, ***P < 0.001). ^$Represents significant difference between MPTP and MPTP + sEV or MPTP + sEV-Phl (^$P < 0.05, ^$$$P < 0.001)

Effect of sEV-Phl on neurogenesis in SNpc region

Since an increase in number of TH immunopositive neurons was distinctly observed in the sEV and the sEV-Phl treated groups from week 2 PMT and given that recent studies [9, 56] have shown endogenous neurogenesis in adult mice SNpc upon treatment with small molecules (antioxidants), we too assessed the same through IHC and confocal imaging at this time point. The signature of newly formed dopaminergic neurons was identified with the expressions of the proliferation marker Ki67 and the floor plate cell marker FOXA2 [9, 57]. As represented in the Fig. 8A, Ki67 expression and its nuclear localization was distinct in the sEV-Phl treated rat SNpc region and the Ki67 expressing cells were co-immunopositive for TH (Fig. 8A). Ki67 expression in the nucleus was also observed in the TH immunopositive neurons in the sEV group as well; however, the Pearson’s coefficient values make it evident that the nuclear expression of Ki67 is significantly higher in the sEV-Phl group than the sEV group at week 2 PMT (Fig. 8B; P < 0.001). The control and the MPTP group rat SNpc region of the brain slices showed negligible Ki67 staining. Similarly, co-immunopositive expression of FOXA2 with TH was vivid in the sEV and the sEV-Phl treated groups in comparison to MPTP (Fig. 8C). Quantification of the number of FOXA2 immunopositive cells further indicated that in the sEV-Phl group the number of FOXA2 immunopositive cells was significantly higher than the sEV group (Fig. 8D; P < 0.001). In the sEV-Phl group, co-immunopositive expression of Ki67 and FOXA2 with TH was observed in the SNpc even at week 4 PMT (supplementary Fig. 4). This was further confirmed by BrdU staining for all the experimental groups (Fig. 8E, F). Unlike in the control and MPTP-treated groups, BrdU⁺ TH⁺ double immunopositive cells was distinctly visible in the sEV and sEV-Phl groups. The number of BrdU-immunopositive dopaminergic (TH⁺) neurons was significantly higher in the sEV + Phl group compared to the sEV group (Fig. 8E&F; P value < 0.001). Brain sections from animals that did not receive BrdU administration showed an absence of BrdU expression (green) (supplementary Fig. 4C). We conclude that both sEV-Phl and sEV treatments successfully induced neurogenesis in the SNpc region, with neurogenesis occurring earlier and more prominently in the sEV-Phl group.

Fig. 8.

Effect of sEV-Phl on neurogenesis in SNpc region: (A) Representative confocal images of SNpc region of control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups immunostained with Ki67 with TH at week 2 PMT. (B) Represents the Pearson’s colocalization coefficient of the Ki67 with DAPI for TH immunopositive cells for all the groups at week 2 PMT (n = 12 taken from 6 biological replicates with each 2 technical replicates). (C) Representative confocal images of SNpc region for all the groups immunostained with FOXA2 with TH at week 2 PMT. Nuclei was counterstained with DAPI (blue). (D) Represents the number of FOXA2 immunopositive cells per field (n = 6 biological replicates with each 3 technical replicates). (E) Representative fluorescence images of SNpc region for all the groups immunostained with BrdU (green, Alexa Fluor 488) and TH (red, Alexa Fluor 647), with nuclei counterstained using DAPI (blue). Insets display a higher magnification of the cell marked by a square. (F) Represents the ratio of BrdU-positive dopaminergic neurons (BrdU⁺TH⁺) to the total number of DA neurons (TH⁺) per field (n = 6 biological replicates with each 3 technical replicates). Statistical significance between control versus MPTP or MPTP + sEV or MPTP + sEV-Phl is denoted by *(*P < 0.05. **P < 0.005, ***P < 0.001). ^$Represents significant difference between MPTP and MPTP + sEV or MPTP + sEV-Phl (^$P < 0.05, ^$$$P < 0.001)

Effect of sEV-Phl on midbrain dopamine, GABA, norepinephrine (NE) and acetylcholine (Ach) content

As the IHC results showed a distinct reappearance of TH immunopositive neurons in the SNpc region in both sEV and sEV-Phl groups, we reconfirmed these findings by measuring midbrain dopamine levels using dopamine ELISA (Fig. 9A). MPTP administration resulted in a gradual and significant decrease in midbrain dopamine levels compared to the control group, and at weeks 2 (P < 0.05) and 4 (P < 0.005) dopamine levels were significantly lower than week 1 PMT. Significantly higher dopamine levels were observed in the sEV-Phl treatment groups compared to the MPTP group at weeks 1, 2 and 4 PMT (P < 0.001), while the sEV group showed higher dopamine content than MPTP group in week 2 and 4 PMT (P < 0.001). However, the sEV group still had significantly lower dopamine levels than both the control and sEV-Phl groups at week 2 PMT (P < 0.001). At week 4, the levels were comparable between the sEV and sEV-Phl groups. Notably, at week 4, the sEV-Phl group had dopamine levels comparable to the control group, whereas the sEV group still showed lower dopamine levels than the control group (P < 0.05). These results indicate that sEV-PHL treatment is more effective in restoring midbrain dopamine levels.

Fig. 9.

Effect of sEV-Phl on midbrain dopamine, GABA, norepinephrine and acetylcholine content: (A) Represents the dopamine levels from midbrain lysate measured by ELISA for control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups at week 1, 2 and 4 PMT (n = 7 biological replicates). (B) Represents the GABA levels measured for control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups at week 1, 2 and 4 PMT (n = 7 biological replicates). (C) Represents the norepinephrine levels measured for control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups at week 1, 2 and 4 PMT (n = 7 biological replicates). (D) Represents the acetylcholine levels measured for control, MPTP, MPTP + sEV and MPTP + sEV-Phl groups at week 1, 2 and 4 PMT (n = 7 biological replicates). Statistical significance between Control versus MPTP or MPTP + sEV or MPTP + sEV-Phl is denoted by *(*P < 0.05, ***P < 0.001). ^$Represents significant difference between MPTP and MPTP + sEV or MPTP + sEV-Phl (^$$$P < 0.001). ^#Represents significant difference between MPTP + sEV and MPTP + sEV-Phl (^###P < 0.001). ^¥Represents a significant difference between week 1 and week 2 or week 4 PMT for the MPTP group (^¥P < 0.05, ^¥¥P < 0.005)

The GABA levels were significantly higher at all time points in the MPTP groups compared to the control group (P < 0.001; Fig. 9B). Treatment with sEV and sEV-Phl resulted in a significant reduction in GABA levels compared to their corresponding MPTP groups (P < 0.001). The GABA levels in the sEV and sEV-Phl treated groups were comparable to those in the control group, indicating that sEV and sEV-Phl treatments did not have any detrimental effects on GABA levels.

The relationship between dopamine and norepinephrine (NE) is well-established, with NE known to regulate attention, decision-making, and working memory [58, 59], all of which can influence locomotion as assessed by the Open Field Test (OFT). Observing the significant recovery in MPTP rats treated with sEV-Phl during the OFT prompted us to measure NE levels (Fig. 9C) as it plays a critical role in mood, exploratory behaviour, and stress response. Significantly reduced NE levels were observed in the MPTP group at weeks 1, 2, and 4 post-MPTP treatment (P < 0.001). No significant recovery was noted in the sEV group compared to the MPTP group, with NE levels remaining significantly lower than in the control group throughout weeks 1, 2, and 4 post-MPTP treatment (P < 0.001). In contrast, the sEV-Phl group demonstrated significantly higher NE levels in the midbrain region compared to the MPTP group across the entire study period, including week 1 (P < 0.001), week 2 (P < 0.001), and week 4 (P < 0.001) post-MPTP treatment, indicating the therapeutic potential of the sEV-Phl formulation.

PD is also characterized by an imbalance between acetylcholine (Ach) and dopamine, often resulting from the degeneration of the dopaminergic nigrostriatal pathway. In the MPTP group, Ach levels significantly increased at week 1 (P < 0.05), week 2 (P < 0.05), and week 4 PMT (P < 0.001) over the control group (Fig. 9D). The sEV group showed a significant decrease in Ach levels compared to the MPTP group at week 2 (P < 0.05) and week 4 PMT (P < 0.001). In fact, while the sEV group did show an initial rise in Ach levels at week 1 PMT (P < 0.05) compared to control, it came back to control levels at weeks 2 and 4. The sEV-Phl treated group demonstrated significantly lower Ach levels at week 1 PMT (P < 0.001) over the MPTP group, a reduction that persisted through week 4 (P < 0.001). Additionally, at week 1 post-MPTP treatment, a significant difference in Ach levels was observed between the sEV and sEV-Phl treatments (P < 0.001). These findings further confirm that sEV-Phl treatment helps maintain the balance of neurotransmitters, working in coordination with dopamine.

Biodistribution of sEV-Phl in MPTP rats

NIR dye-tagged sEV-Phl was administered intranasally, and its biodistribution was tracked in the brain and other peripheral organs at weeks 1, 2, and 4 PMT (Fig. 10). A significantly higher signal intensity was detected in the midbrain starting at week 1 PMT (P < 0.05). At week 2 PMT, the midbrain continued to show higher NIR signal intensity compared to the forebrain (P < 0.005) and hindbrain (P < 0.05). By week 4 PMT, a reduction in NIR signal intensity was observed for all the three brain regions. Here too, signal intensity in the midbrain was higher than the hindbrain and the forebrain (P < 0.005). The NIR signal intensity in the midbrain of control rats was significantly lower compared to MPTP-treated rats at 24 hours post-sEV administration (Day 4 PMT; Supplementary Fig. 5). This finding confirms that the enhanced sEV docking observed in the diseased condition is likely driven by pro-inflammatory signals present in MPTP treated rat midbrain. NIR signals were also observed in vital organs such as the heart, lungs, liver, spleen, pancreas, and kidneys, with the transplanted cells being present in all tissues throughout the study period (Fig. 10C, D). Among these organs, the highest NIR signal intensity was recorded in the kidneys (P < 0.001), while the lowest was found in the lungs. This suggests that sEV administered intranasally do enter the systemic circulation but effectively bypass the “first-pass effect”.

Fig. 10.

Biodistribution of sEV-Phl in MPTP rats: (A) Representative heatmap images of ventral side of brain showing the forebrain, midbrain and hindbrain region to track the docking of sEV-Phl administered in MPTP rats at week 1, 2 and 4 PMT. (B) Bar graph representing quantification of signal intensities normalized w.r.t area obtained from these three-brain region at week 1, 2 and 4 PMT (n = 6 biological replicates). (C) Representative heatmap images of NIR organ and tissue imaging of the peripheral organs in the MPTP-sEV-Phl rats at week 1, 2 and 4 PMT (n = 6 biological replicates). (D) Bar graph representing quantification of signal intensities normalized w.r.t area obtained from these three-brain region at week 1, 2 and 4 PMT. Statistical significance between midbrain versus forebrain and hindbrain is denoted by ^¶(^¶P < 0.05, ^¶¶P < 0.005). ^ØRepresents significant difference between forebrain and hindbrain (^ØP < 0.05, ^ØØP < 0.001). ^ϒRepresents significant difference between kidney and the other organs (^ϒϒP < 0.005)

Discussion

While various studies have documented the neurotrophin-enhancing effects of polyphenols [60], the blood-brain barrier (BBB) remains a significant hurdle for the direct administration of neurotrophin-based treatments [61, 62]. Clinical trials involving neurotrophic factors such as GDNF and neurturin have highlighted issues relating to BBB permeability and molecular specificity [63, 64]. While sEV derived from MSCs have shown promise as drug-delivery vehicles, there are yet no studies investigating their ability to deliver polyphenols in models of neurodegenerative diseases. This study aimed to explore the potential of DPSC-derived sEV as a novel drug delivery system, in the context of neurodegenerative diseases like PD.

DPSCs were the chosen parent cells for sEV isolation because DPSC-derived sEV have been shown to exhibit similar neuromodulatory and neuroprotective effects as the DPSCs themselves [65–67]. In previous studies, we have demonstrated that DPSC-derived conditioned media protected dopaminergic (DA) neurons from 6-OHDA toxicity in non-contact scenarios, indicating that these benefits are mediated by the mesenchymal stromal cell (MSC) secretome. DPSCs indeed displayed typical MSC characteristics as reported in our previous study [30].

sEV isolated using sucrose gradient ultracentrifugation exhibited: (i) a consistent yield, as indicated by protein concentration, (ii) uniform size, as demonstrated by dynamic light scattering (DLS) detection, and (iii) high purity, with flow cytometry analysis showing approximately 98% of sEV expressing all CD markers characteristic of exosomes. This is in line with previous studies that have reported high purity and homogeneity of sEV isolated through density gradient ultracentrifugation [68], and the position paper MISEV23 [36].

Notably, the incorporation of the polyphenol phloroglucinol into these sEV did not compromise their size or structural integrity, as validated by transmission electron microscopy (TEM) and DLS analysis. This indicates that the sEV membrane retains its stability and functionality post-drug loading, which is critical for ensuring effective drug delivery. While sonication has been previously employed for loading small-molecule drugs like paclitaxel and proteins into sEV derived from carcinoma cell lines and MSCs [69–71], this study is the first to demonstrate the successful encapsulation of a phenolic compound such as phloroglucinol into DPSC-derived sEV. Previous studies have shown DPSC sEV to be effective carriers for anticancer drugs by incubating or priming DPSCs with the drugs of interest, targeting the inhibition of cancer cell growth [72, 73].

In in vitro experiments, sEV, and more so those loaded with phloroglucinol, significantly reduced ROS levels in SH-SY5Y cells subjected to 6-OHDA-induced stress, indicating successful cellular uptake and retention of antioxidant activity of the formulation. The protective effect of phloroglucinol against oxidative stress-induced cell damage in SH-SY5Y cells has been previously reported [74]. Phloroglucinol being a polyphenol is known to have high affinity for estrogenic (ERβ) receptors [12, 13], which are abundantly expressed in nasal mucosa [14]. This interaction significantly reduces the bioavailability of these compounds for DA neurons in the midbrain. This interaction limits their bioavailability to dopaminergic neurons in the midbrain. Traditional drug delivery methods lack organ specificity, leading to indiscriminate uptake by healthy and diseased cells alike. Additionally, the nasal mucosa harbours metabolic enzymes that degrade xenobiotics [75, 76], potentially altering the structure and function of intranasally delivered compounds [77]. Hence, protective effects observed with phloroglucinol in vitro may not translate effectively under in vivo disease conditions and would require a delivery vehicle to overcome metabolic degradation, non-specific absorption and deliver at the site of DA neuron degeneration in the midbrain. Our findings confirm that the sEV-Phl showed anti-oxidant potential. Moreover, MSC-derived sEV have earlier been shown to protect hippocampal neurons from oxidative stress [78], and here we demonstrate that DPSC-derived sEV alone similarly mitigate ROS generation in SH-SY5Y cells exposed to 6-OHDA. Given that both sEV and phloroglucinol possess antioxidant properties, the sEV-Phl formulation exhibited an enhanced protective effect as anticipated. The observed saturation of sEV uptake at higher concentrations also corroborates findings that sEV-mediated drug-delivery is dose-dependent, with a plateau effect in cellular internalization. Confocal imaging revealed a predominantly cytosolic localization of the sEV. Consistent with previous reports, we also observed that PKH-26-labeled sEV frequently appeared as intracellular clusters [46–50]. The co-localization of PKH-26 with CD63 immunostaining further confirmed the presence of DPSC sEV in the cells.

Next, in a chronic MPTP-induced rat model of PD where non-motor symptoms were present but motor symptoms had not yet appeared, sEV and sEV-Phl were administered intranasally. The intranasal route was chosen due to its non-invasive nature and its well-established ability to deliver substances directly to the midbrain via the rostral migratory stream (RMS) and olfactory tract, as supported by preclinical and clinical findings [79–81]. This method of CNS delivery has been previously utilized in multiple preclinical studies for the administration of MSC-derived and engineered sEV in models of Alzheimer’s disease, PD, and brain inflammation [82–84]. Consistent with these studies, NIR organ imaging in our study revealed that sEV-Phl successfully crossed the BBB and docked preferentially in the midbrain region. While NIR signals were also detected in the forebrain and hindbrain, the strongest and most sustained signal was observed in the midbrain throughout the study. In comparison to control rat brain the docking of sEV were higher in MPTP rat brain because it’s well established that docking and homing of transplanted cells or their secreted products, such as sEV, are highly influenced by the tissue microenvironment—particularly by cytokines and chemokines released in response to tissue damage and inflammation [24, 85–90]. These findings corroborate previous observations where transplanted cells demonstrated significantly higher docking rates in MPTP-treated rats compared to controls [91]. Similarly, a primate study reported minimal sEV docking in non-diseased control animals, further supporting the disease-specific targeting phenomenon [92].

sEV distribution in the body that we observed following intranasal administration has also been previously reported [80], indicating that sEV do enter systemic circulation. This likely occurs through nasal blood vessels and/or by traversing the cerebrospinal fluid via olfactory ensheathing cells, ultimately reaching blood vessels within the central nervous system.

Additionally, we observed the presence of sEV-Phl in peripheral organs at corresponding time points, with the highest NIR signal detected in the kidneys, suggesting that sEV-Phl is excreted via the renal pathway rather than accumulating in other organs over time. The nanometer-scale size of sEV necessitates an evaluation of the “first-pass effect,” and our NIR organ imaging data suggests that intranasal delivery is safe and does not cause sEV accumulation in the lungs, thereby clearing the “first-pass effect.”

Notably, at week two post-MPTP treatment (PMT), a re-emergence of the NIR signal was observed in the midbrain region. This re-entry of sEV into the midbrain is likely due to the heightened inflammatory cues in this region. The docking of sEV and sEV-Phl in the SNpc region was further confirmed by confocal imaging of PKH-26 labelled sEV and sEV-Phl at weeks one and two PMT. At week one PMT, PKH-26 labelled sEV and sEV-Phl were found to associate preferentially with microglia in the SNpc region of the MPTP rat brain, rather than with DA neurons or astrocytes. This observation aligns with findings by [93], who reported a similar microglial preference following intranasal administration of MSC-derived exosomes in a murine model of epilepsy. This preferential association may be attributed to the migration and docking capabilities of MSC- and DPSC-derived sEV, which are facilitated by their expression of chemokine receptors [52, 94–97]. The DPSC-derived sEV exhibit CXCR4 immunopositivity similar to their parent cells, and the enhanced neuroinflammation in the MPTP-induced rat brain is evidenced by elevated midbrain TNF-α levels. In addition, the consistent association of sEV and sEV-Phl with TNF-α-expressing cells in the SNpc region at all three time points further supports the notion that sEV recognize and dock in response to pro-inflammatory signals. Previous studies have also demonstrated that sEV tend to localize preferentially with microglial cells and induce their polarization from proinflammatory M1 type to anti-inflammatory/neuroprotective M2 type in models of neurodegenerative diseases and brain injuries [98, 99].

By week two (PMT), the association of sEV with tyrosine hydroxylase (TH)-positive neurons increased, likely due to the more extensive degeneration of TH neurons observed at this stage, as indicated by the reduced TH content in colorimetric IHC results. It is now well-established that degenerating or apoptotic cells release signals that attract sEV docking [100, 101].

The MPTP-induced PD rat model is especially useful as it replicates key pathological features of human PD, including the progressive degeneration and death of DA neurons in the SNpc and the development of characteristic motor dysfunction [102]. In our study, the MPTP group exhibited progressive degeneration of TH-positive DA neurons, decreased midbrain dopamine levels, and impaired motor function.

Most importantly, the intranasal administration of sEV and sEV-Phl in MPTP-induced PD rats resulted in significant neuroprotection, as demonstrated by the recovery of TH-positive neurons, restoration of dopamine levels, and reduction in neuroinflammatory markers like TNF-α. Notably, sEV alone exhibited a similar protective effect as sEV-Phl in reducing TNF-α levels in the midbrain, highlighting their inherent ability to mitigate neuroinflammation. The sEV derived from DPSCs are known for their strong immunomodulatory properties [28, 67], which effectively counteracted the pro-inflammatory response triggered by MPTP.

The group treated with sEV-Phl showed an earlier recovery of TH-positive neurons as early as week 1 post-treatment, compared to the group that received sEV alone. While sEV from DPSCs eventually led to the recovery of TH neurons, dopamine levels, and behavioural parameters, this effect occurred later than in the sEV-Phl group. This delayed response is likely due to the fact that while DPSC-derived sEV possess intrinsic antioxidant properties and contain neurotrophic factors [103, 104], the additional antioxidant effect provided by phloroglucinol encapsulated within the sEV likely contributes to the accelerated neuroprotection observed in the sEV-Phl group, as evidenced by the reduced lipid peroxidation in the midbrain. Our previous study demonstrated that DPSC-conditioned medium, rich in neurotrophic factors, effectively protected dopaminergic neurons from 6-OHDA-induced stress in vitro [29]. Other research groups have similarly reported the repair and regeneration of neuronal cells in neurodegenerative rodent models following the administration of sEV derived from DPSCs and MSCs, highlighting the therapeutic potential of these sEV in neuroprotection and recovery.

The early restoration of norepinephrine levels in the sEV-Phl group further supports the observed recovery of dopamine neurotransmitter levels. The relationship between dopamine and norepinephrine is well-established, with both playing crucial roles in attention, mood regulation, exploratory behaviour, decision-making, and working memory [58]. The restoration of norepinephrine following sEV-Phl and sEV treatment correlated with improved performance in the Open Field Test (OFT) for MPTP-treated rats, with the sEV-Phl group showing more pronounced recovery.

MPTP-induced PD in rats can lead to typical non-motor symptoms, such as olfactory dysfunction, a condition found in about 90% of early-stage PD cases and often preceding motor symptoms by several years [105]. This dysfunction has been linked to an increased risk of developing dementia 2–6 years after a decline in olfactory performance, a condition not influenced by dopamine replacement therapy [106]. While the underlying mechanism of olfactory dysfunction in PD remains unclear, and we lack effective prevention or treatment methods, our study demonstrates that intranasal administration of DPSC-derived sEV and phloroglucinol-loaded sEV (sEV-Phl) significantly improved olfactory function in PD rats. Notably, the sEV-Phl group showed a more sustained improvement at week 4 PMT.

Additionally, a decline in nerve conduction velocity (NCV) of the sciatic nerve, a symptom associated with neuropathic pain in PD patients [107, 108], was observed alongside the onset of olfactory dysfunction. Previous studies have indicated that NCV impairment is linked to sciatic nerve demyelination and systemic inflammation, as well as a decline in the function and migration of endogenous bone marrow-derived mesenchymal stem cells (BM-MSCs) [40]. Our earlier research had demonstrated that DPSC transplantation leads to significant improvements in myelin diameter and NCV. In the present study, we observed a recovery in NCV following the administration of DPSC sEV and sEV-Phl, with the improvement coinciding with a reduction in pro-inflammatory TNF-α levels as early as week 1 post-treatment. The similar effects observed between the sEV and sEV-Phl groups suggest that DPSC-sEV alone are sufficient to contribute to the enhancement of NCV, through their presence in the circulation and their impact on systemic inflammation.

In PD, striatal dopamine levels typically decrease while striatal acetylcholine levels increase [109, 110], a pattern that was replicated in the MPTP-induced PD rat model. The sEV-Phl treatment led to an early normalization of these neurotransmitter levels in the midbrain. This early biochemical recovery was mirrored by corresponding behavioural improvements, particularly in motor functions, which are primarily driven by dopamine signalling in coordination with acetylcholine and GABA. Acetylcholine, in conjunction with dopamine, plays a critical role in initiating and maintaining body movements through decision-making and motor-related action selection [111]. Additionally, recent studies have highlighted the significant role of GABA system alterations in the pathogenesis of PD [112]. In an earlier study using an MPTP-induced primate PD model, increased GABA concentration in the striatum was observed, in correspondence with behavioural changes [113]. In this study, an increase in GABA levels was noted as early as week one PMT, which may also have contributed to impaired locomotor activities when coupled with decreased dopamine and increased acetylcholine levels. The sEV-Phl group showed an early restoration of both acetylcholine and GABA, with the sEV-only group demonstrating similar recovery over time. Therefore, sEV-Phl treatment resulted in a more rapid and earlier improvement in motor coordination, a modest enhancement in locomotor activity, and a notably sustained recovery of olfactory function.

Current clinical treatments for PD typically offer only short-term improvements in motor symptoms and fail to prevent the degeneration of dopaminergic neurons. In our study, however, treatment with DPSC sEV and sEV-Phl significantly improved locomotor abilities in a PD rat model and notably increased the number of dopaminergic neurons in the SNpc. This suggests that DPSC sEV can indeed protect and rescue dopaminergic neurons from degeneration. Notably, the presence of phloroglucinol in the sEV facilitated an earlier rescue of these neurons, evident from as early as week one PMT.

Previous research has provided evidence supporting neurogenesis in the SNpc region of rodent brain [114]. Additionally, a prior study demonstrated that the microneurotrophin BNN-20 can promote neurogenesis within this region [9]. In human, a study using post-mortem brains from PD patients demonstrated that adult human neural progenitors can be isolated from the substantia nigra and cultured in vitro under specific growth conditions [115]. The study suggested that multipotent neural stem/progenitor cells reside within the substantia nigra and other regions but under PD appear to lack the essential factors required for neural differentiation when cultured independently. When co-cultured with human embryonic stem cell derived neural progenitors, they were able to differentiate both to neurons and glia. In alignment with [9], we confirmed the presence of newly generated DA (TH-immunopositive) neurons by co-staining with Ki67 and the floor plate cell marker FOXA2. This indicates that both sEV-Phl and DPSC sEV can initiate endogenous neurogenesis in the SNpc region. By week two PMT, we observed a significant increase in the colocalization of the proliferative marker Ki67 with TH-positive neurons in the sEV-Phl group over sEV, corresponding with differences in TH immunopositivity and dopamine content, suggesting that phloroglucinol enhances the neurogenesis and self-repair processes initiated by the sEV. Additionally, newly-generated neural progenitor cells in the subventricular zone migrate along the rostral migratory stream into the olfactory bulb, where they differentiate into granule cells and interneurons [116, 117]. This migration and differentiation may be a key factor in the enhanced olfactory recovery observed in PD rats treated with sEV and sEV-Phl.

The sEV-Phl treatment group demonstrated superior neuroprotective effects compared to the sEV-only group, particularly in rapid and earlier improvement in motor coordination and sustained recovery of olfactory function. These findings suggest that phloroglucinol, when delivered via sEV, not only retains its antioxidant properties but also amplifies the therapeutic benefits of sEV. Previous studies have underscored the significance of antioxidant-loaded sEV in mitigating oxidative stress in neurodegenerative diseases [71, 118], making this discovery particularly promising for advancing PD treatment strategies.

Furthermore, the restoration of neurotransmitter balance—including the normalization of dopamine, norepinephrine, GABA, and acetylcholine levels—also indicates the potential of sEV-Phl to address the complex biochemical dysregulation characteristic of PD. The ability of sEV to preserve the stability and bioactivity of encapsulated drugs like phloroglucinol is essential for achieving such comprehensive therapeutic outcomes.

A limitation of this study is that the therapeutic intervention was administered during the non-motor stage, before the onset of motor impairments. In clinical practice, however, patients typically seek medical attention only after motor symptoms have appeared. Therefore, it is crucial to assess the efficacy of the sEV-Phl formulation administered after motor impairments have developed. The study also lacks comprehensive data on drug release kinetics, which must be characterized through dialysis-based assays to establish the formulation’s pharmacokinetic profile before clinical translation. While PKH-26 co-labeling with CD63 was successfully performed in vitro, this approach remains unvalidated in our in vivo IHC experiments. PKH-26 can generate non-specific fluorescent particles that may confound interpretation of sEV uptake in SNpc cells, necessitating future IHC analysis of PKH-26-labeled sEV with CD63 (we did use NIR imaging to evaluate tissue biodistribution). Moreover, our use of pooled DPSCs from multiple donors to generate sEV prevents assessment of donor-specific therapeutic efficacy. As DPSCs from different donors may exhibit variable therapeutic potential, future studies should investigate donor-to-donor variability in sEV-mediated neuroprotection to optimize patient selection and treatment protocols. Additionally, evaluating the sEV-Phl formulation in other neurodegenerative disease models will further validate and strengthen its therapeutic potential.

Conclusion

This study adds to the growing evidence that sEV are a highly effective drug delivery system, particularly for neuroprotective and regenerative therapies in neurodegenerative diseases such as PD. The successful encapsulation and delivery of phloroglucinol via DPSC-derived sEV, which led to significant neuroprotection, neurogenesis, and functional recovery in a PD model, underscores the potential of this innovative approach. Future research should aim to optimize the sEV loading process, elucidate the mechanisms underlying sEV-mediated drug delivery, and assess the long-term efficacy and safety of sEV-based therapies in clinical settings.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: I.D; methodology: K.M, R.G, A.M, G.W, R.S, N.B and I.D; formal analysis and investigation: K.M, R.G, A.M, R.S, N.B and I.D; writing - original draft preparation: I.D; writing-review and editing: K.M, R.G, A.M, R.S, N.B and I.D; funding acquisition: I.D; resources: I.D; supervision: I.D, A.M. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

This work is supported by a grant obtained from Indian Council for Medical Research (ICMR), Government of India, New Delhi; contract grant No. 2019–2697/SCR/ADHOC-BMS and Department of Science & Technology (DST – TDP), Government of India, New Delhi; contract grant No. TDP DST/TDT/TDP-68/2022. K.M is supported by DHR-YSS fellowship, R.G is supported by CSIR-SRF Ph.D fellowship.

Data availability

All data supporting the conclusions of this article are included within the article and its supplementary files. All additional files are included in the manuscript.

Declarations

Ethics approval and consent to participate

For the animal experiments ethics approval was taken from the Institutional Animal Ethics Committee (IAEC) of NIMHANS, Bengaluru, India for the project title “Cell-free therapeutic approach to Parkinson’s disease using exosomes of human dental pulp stem cells as drug delivery tool through intranasal route in in vivo PD model targeting efficacy and bio-distribution” of approval number AEC/69/446/BP dated May 10, 2019 and Institutional Bio-safety Committee (IBSC) of approval number NIMHANS/DO/IBSC MEETING/2018 dated April 18, 2018. Approval for use of commercially available human dental pulp stem cells was obtained from Institutional Committee for Stem Cell and Research (IC-SCR) for the same project title on May 27, 2019 of approval number SEC/03/018/BP. For the commercially procured DPSCs, the original source (HiMedia Laboratories) has confirmed that all procedures related to this product have been reviewed and approved by IEC and IC-SCR at HiMedia Laboratories Private Limited and the donors had signed informed consent.

Competing interest

The authors declare that they have no competing interests.

Declarations

The authors declare that they have not use AI-generated work in this manuscript.