Intranasal delivery of hMSC-derived supernatant for treatment of ischemic stroke by inhibiting the pro-inflammatory polarization of neutrophils

Abstract

Background

Stem cells utilized for ischemic stroke treatment often display unstable homing capabilities and diminished activity in vivo, limiting their neuroprotective efficacy. Furthermore, the optimal delivery route for stem cells remains undetermined. While the cytokines secreted by stem cells show promise in modulating post-stroke inflammation, the direct application of these supernatants in ischemic stroke treatment and the underlying mechanisms are still unclear.

Methods

Secretory supernatants (hMSC-L) and cell lysate products (hMSC-M) from primary human umbilical cord mesenchymal stem cells-cultured medium were administered intranasally to mice with cerebral ischemia. The neuroprotective effects of hMSC-L and hMSC-M were assessed with TTC staining, behavioral tests and pathological staining. Flow cytometry and qPCR evaluated the expression of immune cells and cytokines in the CNS and peripheral immune organs. In vitro, flow cytometry and ELISA measured the effects of hMSC-L and hMSC-M on N2 polarization and inflammatory cytokines expression in primary murine neutrophils. Western blot analysis determined the impact of hMSC-L and hMSC-M on the PPAR-γ/STAT6/SOCS1 pathway, which is crucial for N2 neutrophil polarization.

Results

TTC staining, behavioral experiments, and pathological assessments reveal intranasal delivery of hMSC-L and hMSC-M significantly reduces the infarct volume of mice with cerebral ischemia, improves neurological function scores, and promotes motor function recovery. Higher concentrations of hMSC-M contributed a more pronounced effect on neuropathological improvements in ischemic mice. Intranasal delivery of hMSC-L and hMSC-M significantly reduces neutrophil infiltration in the brain post-stroke and increases the proportion of anti-inflammatory N2-subtype neutrophils, boosting the expression levels of IL-10 and TGF-β. In vitro experiments demonstrate that hMSC-L and hMSC-M promote nuclear translocation of PPAR-γ in neutrophils stimulated with PMA, activating the downstream STAT6/SOCS1 signaling pathway to encourage N2-subtype neutrophil polarization.

Conclusions

Intranasal delivery of hMSC-L and hMSC-M effectively ameliorates cerebral ischemic injury in mice, comparable to traditional administration routes like intravenous delivery. Treatment with hMSC-L and hMSC-M enhances the PPAR-γ/STAT6/SOCS1 pathway and improves the neuroinflammatory response post-stroke by increasing N2 neutrophil infiltration. These results provide a theoretical basis for a deeper understanding of the mechanisms of stem cell therapy and for exploring suitable delivery pathways of stem cell treatment.

Introduction

Ischemic stroke (IS) represents a significant global public health challenge due to its severe impact on human health [1, 2]. Clinically, rapid vascular recanalization through intravenous thrombolysis using recombinant tissue plasminogen activator (rtPA) or mechanical thrombectomy is standard; however, these treatments are constrained by stringent indications and a narrow therapeutic time window. Additionally, they address little beyond the immediate restoration of blood flow, failing to counteract other pathophysiological mechanisms triggered by acute ischemia, such as the neuroinflammatory response [3, 4]. This response, mediated by various immune cells, is a critical factor in sustained ischemic brain damage and a recognized target for therapeutic intervention [5].

Despite the development of numerous neuroprotective strategies aimed at modulating inflammatory responses, clinical outcomes have generally been disappointing, often due to limited efficacy or significant side effects. [6] Over the past few decades, stem cell therapies have garnered preliminary approval for their safety and efficacy in treating ischemic stroke. [7] However, the homing efficiency of directly transplanted stem cells remains low. The mechanisms by which stem cells modulate post-stroke neuroinflammation are still not fully understood. Further exploration of optimal delivery routes and underlying mechanisms is necessary to facilitate the clinical application and translation of stem cell therapy.

To fully harness the anti-inflammatory potential of stem cells, prior research has explored the use of derivatives such as protein- and functional RNA-rich exosomes [8], as well as extracellular vesicles [9], for stroke treatment. However, these exosomes may contain only limited biological components, and stem-derived supernatants and cell lysates, which are enriched with bioactive substances from both intracellular and extracellular domains, might offer enhanced efficacy in mitigating the neuroinflammatory response post-ischemic stroke. Additionally, the method of delivery also critically influences the pharmacodynamics and pharmacokinetics, significantly impacting their effectiveness [10]. Commonly employed delivery routes for stem cells—such as intravenous, arterial, and intracranial injections—suffer from low efficiency in transplantation and may jeopardize cell viability. These invasive methods also carry potential risks of secondary injury and brain edema [11]. Conversely, intranasal delivery represents a non-invasive alternative that can bypass the blood–brain barrier, directly introducing therapeutic agents into the brain. Based on these considerations, we sought to explore whether intranasal administration of stem cell secretory supernatant and cell lysate products could effectively regulate the neuroinflammatory response and improve outcomes in IS [12].

Neutrophils are central immune cells in neuroinflammation post-ischemic stroke and are key contributors to subsequent brain injury [13]. Upon recruitment to injured tissue, neutrophils also release reactive oxygen species (ROS), metalloproteinases, perforin, and neutrophil extracellular traps, exacerbating damage to the blood–brain barrier (BBB) [14]. However, clinical trials targeting the inhibition of neutrophil infiltration into ischemic brain areas have largely failed, possibly due to the high heterogeneity of neutrophils. Recent studies have elucidated that neutrophil, like macrophages in the innate immune system, display functional heterogeneity with different phenotypic subtypes. This heterogeneity, preliminarily confirmed in various diseases [15,16,17,18], shows that N1-type neutrophils may exacerbate inflammation and brain damage, whereas N2-type neutrophils could suppress inflammation and aid neural repair [19]. Thus, modulating the balance between these neutrophil subtypes emerges as a potential new strategy for stroke treatment. Mechanically, the expression of Toll-like receptors TLR2 and TLR4 activates the NF-kappa B pathway, which promotes M1 microglia or N1 neutrophil post-stroke. This pathway is also modulated by high-mobility group box 1 protein (HMGB1) [20, 21]. Additionally, peroxisome proliferator-activated receptor gamma (PPAR-γ) and signal transducer and activator of transcription (STAT) pathways, known for their roles in anti-oxidative and anti-inflammatory responses, could foster the suppression of inflammatory M1 or N1 polarization [22, 23]. However, whether stem cell-derived supernatants can potentially influence the neuroinflammatory response post-stroke by regulating neutrophil subset balance via the PPAR-γ/STAT pathway remains unclear.

In the present study, we extracted secretory supernatant and cell lysate from primary cultured human umbilical cord mesenchymal stem cells (hMSCs) and administered them to mice with IS via the intranasal route. Our research primarily investigated the therapeutic effects of stem cell-derived supernatant on IS, specifically its potential to inhibit neuroinflammation. Additionally, we explored its regulatory impact on neutrophil polarization through the PPAR-γ/STAT6/SOCS1 pathway. Understanding the mechanisms underlying stem cell therapies and identifying effective administration methods could provide a new theoretical foundation with significant clinical and societal implications.

Methods and materials

Animal

A total of 115 male C57BL/6 mice, aged 8 weeks and weighing 20–22 g, were obtained from Liaoning Changsheng Biotechnology Co., Ltd. Experimental animals were housed in SPF-grade animal rooms maintained at a temperature of 22–24 °C, relative humidity of 50–60%, and under a 12-h light–dark cycle, with ad libitum access to food and water. All animal experiments were approved by the Ethics Committee of Harbin Medical University. The specific groupings for neurological function scoring and behavioral experiments were as follows:

Group 1: hMSC-L + i.n., hMSC-L + i.v., hMSC-L + i.p., hMSC-M + i.n., hMSC-M + i.v., and hMSC-M + i.p., comprising 10 mice per subgroup.

Group 2: 1 mg/ml hMSC-L + i.n., 2 mg/ml hMSC-L + i.n., 1 mg/ml hMSC-M + i.n., and 2 mg/ml hMSC-M + i.n., each comprising 10 mice.

Group 3: 2 mg/ml hMSC-L + i.n. and 2 mg/ml hMSC-M + i.n., comprising 15 mice per subgroup.

Model and drug intervention

The middle cerebral artery occlusion (MCAO) model in mice was induced as previously described. Briefly, mice were anesthetized with a 2% sodium pentobarbital solution administered via intraperitoneal injection. A midline neck incision was made, and neck muscles were separated to expose the right common carotid artery, external carotid artery, and internal carotid artery. A filament (RWD, Shenzhen, China) was inserted through the external carotid artery and advanced to occlude the origin of the middle cerebral artery, thereby obstructing blood flow. Throughout the surgery, a temperature maintenance device (RWD, Shenzhen, China) was used to maintain the body temperature of experimental animals at 36.5 ± 0.5 °C. After mice recovered consciousness, they received daily intraperitoneal (i.p.), intravenous (i.v.), and intranasal (i.n.) administrations of 100 μl of hMSC-derived supernatant (hMSC-L) and hMSC-derived cell lysate (hMSC-M). Mice were euthanized for subsequent assessments after 3 days of continuous treatment. Euthanasia of experimental animals was performed using cervical dislocation.

Triphenyl tetrazolium chloride (TTC) staining

After perfusion with PBS through the intracardiac route, brain tissues were extracted from the tested mice. The tissues were placed in pre-cooled brain slice molds and sectioned into 2 mm coronal slices. These sections were subsequently incubated at 37 °C for 15 min in a 2% TTC solution dissolved in PBS. Normal tissues displayed a red coloration, whereas infarcted areas appeared white. The extent of brain infarction was quantified using Image J software following photography.

Neurological function scoring

Neurological deficits in mice were assessed using a blinded method. The Longa five-point scale was employed: 0 points signified no neurological deficits; 1 point indicated incomplete extension of the contralateral forelimb; 2 points indicated circling towards the paralyzed side; 3 points indicated leaning towards the paralyzed side; and 4 points indicated an inability to walk spontaneously and loss of consciousness. Higher scores indicated greater severity of neurological impairment in experimental animals.

Behavioral experiments

Corner turn test

This test primarily evaluated unilateral sensory and motor deficits in MCAO mice. The apparatus consisted of two plates forming a 30° angle. Upon reaching the corner and touching the plates’ walls, both whiskers were stimulated. In normal mice, the frequency of turning to the left or right was roughly equivalent, whereas MCAO mice tended to turn more towards the lesioned side. Each mouse underwent 10 trials with 1-min intervals between tests. Calculations involved dividing the number of turns towards the unaffected side by the total number of turns and multiplying by 100%.

Elevated body swing test

This test assessed motor function asymmetry in mice. Each mouse was lifted by its tail to a height where its head was approximately 5 cm above the surface. The body swung to either side along the vertical axis, and the direction of swing was noted. A swing angle > 10° was set as the criterion. Each mouse underwent 20 repetitions with 1-min intervals between repetitions. Calculations involved dividing the number of swings towards the affected side by the total number of swings and multiplying by 100%. All experiments were conducted in a blinded manner, with observers unaware of the group assignments of the mice.

Cell culture and drug preparation of human umbilical cord mesenchymal stem cells (hMSCs)

Primary hMSCs lines were supplied by our laboratory and cultured in DMEM/F12 medium supplemented with 10% fetal bovine serum, 100 IU/ml penicillin, and 100 IU/ml streptomycin at 37 °C, 5% CO2 in a culture chamber (Thermo Fisher, USA), with medium changes every 2–3 days. Upon reaching 70%−80% density, supernatants of the cells were harvested as hMSC-derived supernatant (hMSC-L). Then, cells were further cultured until reaching 90% confluence, then frozen at − 20 °C, subjected to three freeze–thaw cycles, and the collected liquid in the culture vessel was designated as hMSC-derived cell lysate (hMSC-M). hMSC-L contains immunoregulatory substances secreted by stem cells, such as vascular endothelial growth factor, cytokines, and brain-derived neurotrophic factors; hMSC-M also contains intracellular proteins, nucleic acids, and other bioactive substances. The harvested hMSC-L and hMSC-M were diluted twofold, fivefold, and tenfold with medium, and the protein concentration was determined using a BCA protein assay kit (Beyotime, Shanghai, China). A standard curve was generated by serial dilution of protein standards as per the manufacturer’s instructions, and drug concentrations were adjusted to approximately 1 mg/ml and 2 mg/ml for subsequent experiments.

Primary neutrophil culture

Bone marrow was flushed from the tibiae and femurs of C57/BL6 male mice, followed by lysis of red blood cells and density gradient centrifugation at 1500 rpm, 26 °C, acceleration 9, deceleration 0, for 30 min. Neutrophils were isolated using a 1 ml pipette, transferred to a new centrifuge tube, mixed with an equal volume of PBS, and centrifuged at 1200 rpm, 4 °C, for 8 min. Neutrophils were resuspended in 1640 medium, counted, and seeded into 96-well plates at approximately 5 × 10⁵ cells per well. Cells were stimulated with PMA and treated with T0070907 (MCE, USA) to inhibit PPARγ expression, then incubated in a 37 °C, 5% CO2 incubator for 4 h before proceeding to flow cytometry staining and analysis.

Pathological staining

Mouse brain tissues were sectioned into 10 μm thick coronal sections using a cryostat microtome. Hematoxylin and eosin (H&E) staining was conducted with a Hematoxylin–Eosin staining kit (Solarbio, G1120), and Nissl staining utilized Nissl staining solution (Leagene, DK0023).

Immunofluorescence staining

10 μm thick coronal slices of mouse brain sections were blocked with 5% goat serum (Biosharp, BL1097A) for 2 h at 4 °C, followed by overnight incubation at 4 °C with primary antibodies: Rabbit anti-CD31 (1:100, abcam, ab281583), Rat anti-CD31 (1:100, abcam, ab7388), Rat anti-ZO-1 (1:100, eBioscience, 14,977,682), and Rabbit anti-occludin (1:50, Thermo Fisher, 404,700). Subsequently, sections were incubated at room temperature for 1 h with secondary antibodies: TRITC donkey anti-Rat (1:100, Jackson, 712-025-153), FITC Goat Anti-Rabbit IgG (1:100, Jackson, 111-095-003), and then with DAPI (1:1000, Sigma, MBD0015) for 3 min at room temperature. Images were acquired using a confocal microscope (Zeiss, Germany).

Flow cytometry

A mouse MCAO model was established and treated according to the assigned groups. 3 days post-surgery, mice were euthanized, and peripheral blood, spleen, brain, and bone marrow were harvested. Peripheral blood was mixed in a 1.5 ml EP tube containing sodium heparin, from which 100 µl was extracted; anticoagulant was added, and after antibody incubation, red blood cells were lysed. Spleens and bone marrow were minced, and red blood cells were lysed, preparing single-cell suspensions. Brain tissues were digested with type II collagenase at room temperature for 1 h, and leukocytes were separated using a 30% Percoll solution. Following PBS washing, the single-cell suspensions were seeded into 96-well plates at approximately 1 × 10^6 cells per well. Cells were incubated in the dark at 4 °C for 30 min with antibodies including FITC anti-mouse CD45 (Biolegend, 103,108), APC anti-mouse/human CD11b (Biolegend, 101,212), PE anti-mouse Ly-6G Antibody (Biolegend, 127,607), PerCP/Cyanine5.5 anti-mouse Ly-6C (Biolegend, 128,012), PE/Cyanine7 anti-mouse F4/80 (Biolegend, 123,113), and PerCP/Cyanine5.5 anti-mouse CD206 (Biolegend, 141,715). Neutrophils were extracted as described above and incubated in the dark at 4 °C for 30 min with Alexa Fluor® 488 anti-mouse Ly-6G (Biolegend, 127,626), PE anti-mouse CD206 (Biolegend, 141,706), and PE-Cyanine7 anti-mouse iNOS (eBioscience, 25–5920-82). Cells labeled with antibodies were analyzed using a BD flow cytometer (BD Biosciences), and FlowJo V software was utilized for data analysis.

Western blotting

Ischemic brain tissues were sonicated in western and IP cell lysis buffer (Beyotime, Shanghai, China) containing PMSF (Beyotime, Shanghai, China) at 4 °C, followed by centrifugation at 12000 g for 15 min to collect the supernatant. Protein concentration was determined using a BCA protein assay kit (Beyotime, Shanghai, China), and protein samples were prepared accordingly. Nuclear and cytoplasmic proteins were extracted using a nuclear and cytoplasmic protein extraction kit (Beyotime, P0028). Samples were separated on 10% or 15% SDS-PAGE gels and transferred to PVDF membranes (Millipore, USA). Membranes were incubated overnight at 4 °C with primary antibodies: mouse anti-PPARγ (1:500, Proteintech, 66,936–1), Rabbit anti-pPPARγ (1:200, Affinity, AF3284), SOCS1(1:500, Affinity, AF5378), mouse anti-STAT6 (1:500, Proteintech, 66,717–1), Rabbit anti-p-STAT6 (1:500, Proteintech, 51,073–1), mouse anti-Histone H3 (1:1000, Proteintech, 68,345–1), mouse anti-GAPDH (1:1000, Abbkine, KTD101-CN), and mouse anti-β-actin (1:1000, Abbkine, KTD101-CN). Membranes were then incubated at room temperature for 2 h with horseradish peroxidase-conjugated goat anti-mouse IgG (1:1000, Zsbio, ZB2305) or goat anti-rabbit IgG (1:1000, Zsbio, ZB2301). Chemiluminescence detection and image capture were performed using an ECL chemiluminescence imaging system. Image J software was used to analyze grayscale values of protein bands, with Histone 3, GAPDH, and β-actin used as internal references.

Quantitative PCR

RNA was extracted from ischemic brain tissue using Trizol reagent (Takara, Japan) according to the instructions of Reverse Transcriptase M-MLV (Takara, Japan). The extracted RNA was reverse transcribed into cDNA and diluted with DNase/RNase-free water (Thermo Fisher, USA). Real-time quantitative PCR amplification was performed using Hieff qPCR SYBR Green Master Mix (Yeasen, Shanghai, China) and specific primers. The expression of β-actin mRNA was calculated using the 2^−ΔΔCt method as an internal control. The primer sequences used in the study are as follows:

IFNγ-F: GCCACGGCACAGTCATTGA

IFNγ-R: TGCTGATGGCCTGATTGTCTT

TNFα-F: CAGGCGGTGCCTATGTCTC;

TNFα-R: CGATCACCCCGAAGTTCAGTAG

IL1β-F: TTCAGGCAGGCAGTATCACTC

IL1β-R: GAAGGTCCACGGGAAAGACAC

CXCL1-F: ACTGCACCCAAACCGAAGTC

CXCL1-R: TGGGGACACCTTTTAGCATCTT

CXCL2-F: CAGACTCCAGCCACACTTCA

CXCL2-R: AGGTACGATCCAGGCTTCCC

CCL3-F: ACCATGACACTCTGCAACCAA

CCL3-R: CGGTTTCTCTTAGTCAGGAAAATGA

NE-F: TTGCCAGGAATTTCGTCATGT

NE-R: GTTGGCGTTAATGGTAGCGGA

MPO-F: AGGGCCGCTGATTATCTACAT

MPO-R: CTCACGTCCTGATAGGCACA

PADi4-F: TCTGCTCCTAAGGGCTACACA

PADi4-R: GTCCAGAGGCCATTTGGAGG

VEGF-F: TATTCAGCGGACTCACCAGC

VEGF-R: CCTCCTCAAACCGTTGGCA

TGFβ-F: CGTGGAAATCAACGCTCCAC

TGFβ-R: GAAGTTGGCATGGTAGCCCT

IL10-F: GGTGAGAAGCTGAAGACCCTC

IL10-R: GCCTTGTAGACACCTTGGTCTT

β-actin-F: TGGAATCCTGTGGCATCCATGAAAC

β-actin-R: TAAAACGCAGCTCAGTAACAGTCC

Enzyme-linked immunosorbent assay (ELISA)

According to the instructions provided by the ELISA kit, the concentrations of IL-10 (Invitrogen, 88–7105-22), and TGF-β (novous, VAL611) in the supernatant of various neutrophil subtypes were measured.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 9.0 software. Data are presented as mean ± standard deviation. The normality of data distribution was assessed using the Shapiro–Wilk test. For comparisons among multiple groups, one-way analysis of variance (ANOVA) was conducted, followed by Tukey’s multiple comparisons test to identify significant differences between groups. In cases where data did not meet the assumptions of normality, non-parametric tests such as the Kruskal-Walli’s test were utilized, followed by Dunn’s post hoc test. Statistical significance was set at a p-value less than 0.05. Significance levels were denoted as P < 0.05, P < 0.01, and P < 0.001.

Results

Result 1: Intranasal delivery of hMSC-derived secretory supernatant improves infarct volume in mice with cerebral ischemia

A C57BL/6 mouse model of MCAO was employed to evaluate the therapeutic effect of hMSC-derived secretory supernatant (hMSC-L) and hMSC-derived cell lysates (hMSC-M) administered via different routes. Following the successful induction of ischemic stroke, 100 µL of hMSC-L or hMSC-M were administered daily through intranasal (i.n.), intravenous (i.v.), and intraperitoneal (i.p.) routes at consistent time points. TTC staining was performed after 3d of consecutive treatment to evaluate changes in the ischemic volume across the experimental groups. As depicted in Fig. 1A, all administration routes significantly reduced the infarct volume, with the reductions being statistically significant (Fig. 1B and F, ^*P < 0.05, ^**P < 0.01). The neurological function, assessed by Longa score results, improved following treatment with both hMSC-L (Fig. 1C) and hMSC-M (Fig. 1G) across all routes (^*P < 0.05, ^**P < 0.01, ^***P < 0.001). Motor function recovery was further evaluated using the corner turn test and elevated body swing test, which revealed improvements in the frequency of turning towards the affected side (Fig. 1D and H, **P < 0.01, ***P < 0.001) and an increased proportion of right-sided turns (Fig. 1E and I, *P < 0.05, **P < 0.01, ***P < 0.001). Notably, while treatments via intranasal, intravenous, and intraperitoneal routes significantly reduced infarct volumes and enhanced neurological functions, no statistically significant difference were observed in the extent of infarct volume reduction and neurological function recovery between the intranasal and other groups. These results substantiate the efficacy of hMSC-L and hMSC-M in treating cerebral ischemia in mice and suggest that intranasal delivery provides a therapeutic impact comparable to that of conventional routes. However, the noninvasive nature of intranasal delivery offers enhanced convenient and ease of administer.

Fig. 1

Intranasal delivery of hMSC-L and hMSC-M effectively reduced the infarct volume of mice following cerebral ischemia. A: TTC staining to visualize infarct volume post-treatment with human hMSC-L (left panel) and hMSC-M (right panel) using different administration routes. Normal brain tissue stains red, whereas ischemic regions appear white. B: Statistics analysis of infarct volume in mice treated with hMSC-L. C: Neurological function scores for mice post-ischemic stroke treated with hMSC-L. D, E: Recovery of motor function in ischemic stroke mice assessed by the corner test (D) and body elevation test (E) following treatment with hMSC-L. F: Statistics analysis of infarct volume in mice treated with hMSC-M. G: Neurological function scores for mice post-ischemic stroke treated with hMSC-M. H, I: Recovery of motor function in ischemic stroke mice assessed by the corner test (H) and body elevation test (I) following treatment with hMSC-M. n = 5 or 10, versus MCAO ns no significance; ^*P < 0.05, ^**P < 0.01, ^***P < 0.001

Result 2: Intranasal delivery of high concentration of hMSC-derived supernatant is more effective in the treatment of cerebral ischemia

To investigate the influence of different concentrations of therapeutic agents in treating MCAO in mice, we quantified the protein concentrations of hMSC-L and hMSC-M using the bicinchoninic acid (BCA) protein assay. Subsequently, we adjusted the concentrations of these agents to 1 mg/mL and 2 mg/mL using hMSCs medium. MCAO mice were then treated via the nasal mucosal route. TTC staining revealed that higher concentrations of hMSC-L and hMSC-M significantly reduced the cerebral ischemic volume (Fig. 2A, B, and D, ***P < 0.001). Neurological function scores indicated that elevated concentrations of hMSC-L and hMSC-M might more effectively promote neurological recovery in mice (Fig. 2C and E, ^***P < 0.001). Pathological examinations showed that brain tissues in MCAO mice exhibited vacuolization and nuclear atrophy; these pathological changes were ameliorated following treatment with hMSC-L and hMSC-M. Notably, mice treated with 2 mg/mL of either agent exhibited significantly fewer damaged cells compared to the 1 mg/mL group (Fig. 2F and G, ^##P _{1 mg/ml vs 2 mg/ml} < 0.01). Nissl staining highlighted that treatment with hMSC-L and hMSC-M increased the number of Nissl bodies in brain neurons, indicative of functional improvements. Moreover, the higher concentration (2 mg/mL) was more protective towards Nissl bodies (Fig. 2H and I, ^##P _{1 mg/ml vs 2 mg/ml} < 0.01).

Fig. 2

Therapeutic effect of hMSC-derived secretion product on cerebral ischemic via intranasal approach. A: TTC staining illustrating infarct volume post stroke. B: Statistical analysis of infarct volume treated with different concentrations of hMSC-L. C: Neurological function scores for mice treated with different concentrations of hMSC-L. D: Statistical analysis of infarct volume treated with different concentrations of hMSC-M. E: Neurological function scores for mice treated with different concentrations of hMSC-M. F: Representative images and damaged cells statistical analysis (J) of HE stained sections of the cerebral cortex following treatment with hMSC-L(× 200). G: Representative images and damaged cells statistical analysis (L) of HE stained sections of the cerebral cortex following treatment with hMSC-M (× 200). H: Representative images and statistical analysis (K) of Nissl body in the cortex following treatment with hMSC-L (× 200). I: Representative images and statistical analysis (M) of Nissl body in the cortex following treatment with hMSC-M (× 200). n = 5 or 10; versus MCAO ^**P < 0.01, ^***P < 0.001; 1 mg/mL vs 2 mg/mL ns: no significance; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001

These findings further support the neuroprotective effects of hMSC-L and hMSC-M when administered nasally in mice with cerebral ischemia. The more pronounced therapeutic effects at higher concentrations led to the selection of a 2 mg/mL concentration for subsequent experiments.

Result3: Intranasal delivery of hMSC-derived supernatant inhibits neutrophil infiltration after cerebral ischemia and reduce neuroinflammation

Following the onset of ischemic stroke, the disruption of the BBB allows neutrophils, mononuclear macrophages, and other immune cells to infiltrate the CNS, thereby inducing inflammation and parenchymal damage [24,25,26]. Consequently, we further explored the effects of intranasal administration of hMSC-L and hMSC-M on the infiltration of various immune cell subtypes and neuroinflammation post-cerebral ischemia. Flow cytometry analysis revealed that treatment with hMSCs did not alter the infiltration ratio of macrophages (CD4^highCD11b⁺F4/80⁺, Fig. 3A). However, the proportion of immune regulatory CD4^highCD11b⁺Ly6C^high monocytes decreased (Fig. 3B, right panel, ^*P < 0.05), while the proportion of pro-inflammatory CD45^highCD11b⁺Ly6C^int monocytes increased significantly (Fig. 3B, left panel, ^*P < 0.05). Notably, hMSCs treatments particularly reduced the proportion of infiltrating neutrophils in the brain, with a more pronounced decrease observed in the hMSC-M treatment group (Fig. 3C, ^**P _{hMSC-M vs MCAO} < 0.01).

Fig. 3

Intranasal delivery of hMSCs inhibits neutrophil infiltration in the ischemic brain. A: The proportion of CD45^highCD11b⁺ F4/80⁺ macrophages in the brain after hMSC-L and hMSC-M treatment. B: The proportion of CD45^highCD11b⁺Ly6C^high and CD45^highCD11b⁺Ly6C^int monocytes in the brain after hMSC-L and hMSC-M treatment. C: The proportion of CD45^highCD11b⁺ Ly6G⁺ neutrophils in the brain after hMSC-L and hMSC-M treatment. D–F: The proportion of CD45^highCD11b⁺ Ly6G⁺ neutrophils in peripheral blood (D), spleen (E) and bone marrow (F) after hMSC-L and hMSC-M treatment. n = 5, vs MCAO ns: no significance; ^*P < 0.05,.^**P < 0.01

Since these infiltrating immune cells primarily migrate from peripheral immune organs post-stroke, we also assessed changes in these cells within the main peripheral immune organs and peripheral blood. Post-treatment evaluations indicated no significant changes in the proportions of macrophage and monocyte subsets in the peripheral blood (Supplementary Fig. 1), spleen (Supplementary Fig. 2), or bone marrow (Supplementary Fig. 3). However, the proportion of neutrophils decreased in both peripheral blood (Fig. 3D, ^*P < 0.05, ^**P < 0.01) and spleen (Fig. 3E, ^*P < 0.05) but increased in the bone marrow (Fig. 3F, ^*P < 0.05). Given that neutrophils predominantly originate from myeloid progenitor cells in the bone marrow and mature there, we speculate that the increased proportion of neutrophils in the bone marrow may serve to compensate for the diminished peripheral neutrophil levels. These findings suggest that neutrophils could be the primary effector cells influenced by hMSC-derived secretion factors in the treatment of IS.

Subsequently, we examined the impact of secretory factors from hMSCs on neuroinflammatory responses in ischemic brain tissue. Specifically, we analyzed the transcriptional levels of key pro-inflammatory cytokines involved in cerebral ischemia, including IFN-γ (Fig. 4A), TNF-α (Fig. 4B), and IL-1β (Fig. 4C). Treatment with both hMSC-L and hMSC-M resulted in a significant reduction in the transcription of these cytokines. Notably, the decrease in IFN-γ (**P < 0.01) and TNF-α (**P < 0.01). was more pronounced in the hMSC-M treatment group than in the hMSC-L group, suggesting a potentially stronger anti-inflammatory effect of hMSC-M. Furthermore, transcription levels of major neutrophil chemokines, including CXCL1 (Fig. 4D), CXCL2 (Fig. 4E), and CCL3 (Fig. 4F), also decreased significantly in the brains of treated mice (*P < 0.05, **P < 0.01). These findings indicate that treatment with hMSC-L and hMSC-M via the intranasal route can effectively reduce neuroinflammation in MCAO mice. This reduction may be associated with the modulation of neutrophil activity and function.

Fig. 4

Intranasal delivery of hMSCs inhibits the expression of pro-inflammatory cytokines in the ischemic brain. A–C: qPCR analysis of IFN-γ (A), TNF-α (B) and IL-1β (C) mRNA expression in brain tissues after ischemic stroke treated with hMSC-L and hMSC-M. D–E: qPCR analysis of CXCL1 (D), CXCL2 (E) and CCL3 (F) mRNA expression in brain tissues after ischemic stroke treated with hMSC-L and hMSC-M. n = 5, versus MCAO ns: no significance; ^*P < 0.05, ^**P < 0.01,.^***P < 0.001

Result4: Intranasal delivery of hMSC-derived supernatant promotes N2 neutrophils polarization

Neutrophils, under various stimuli, initiate programming systems that lead to the polarization into inflammatory (N1) and immunoregulatory (N2) subtypes [27, 28]. In our study, we investigated the potential of hMSC-L and hMSC-M to modulate neuroinflammation in ischemic brain tissue by influencing neutrophil subset balance. Flow cytometry analysis revealed a significant increase in the proportion of N2 subtype neutrophils (CD45^highCD11b⁺Ly6G⁺CD206⁺) in the brains of mice treated with hMSCs therapies (***P < 0.001), despite an overall reduction in neutrophil infiltration. Notably, the proportion of N2 neutrophils was significantly higher in the hMSC-M group compared to the hMSC-L group (Fig. 5A, ^##P < 0.01). Additionally, the proportion of N2 neutrophils in peripheral immune organs, such as bone marrow and peripheral blood, significantly decreased (Fig. 5B and C, *P < 0.05, **P < 0.01). This shift in neutrophil subpopulations underscores the systemic impact of hMSCs treatment on neutrophil dynamics and suggests a targeted reprogramming mechanism that favors anti-inflammatory responses within ischemic brain tissue.

Fig. 5

Intranasal delivery of hMSCs promotes N2 polarization of neutrophils in the brain after stroke. A: Proportion of CD45^highCD11b⁺Ly6G⁺CD206⁺ N2 subtype neutrophils in the brain treat with hMSC-L and hMSC-M. B, C: CD45⁺CD11b⁺Ly6G⁺CD206⁺ neutrophils in peripheral blood (B) and bone marrow (C) following hMSC-L and hMSC-M treatment. D: qPCR analysis of NE, MPO and PADi4 mRNA expression in brain tissues after ischemic stroke treated with hMSC-L and hMSC-M. E: qPCR analysis of VEGF, TGF-β and IL-10 mRNA expression in brain tissues after ischemic stroke treated with hMSC-L and hMSC-M. n = 5, vs MCAO, ns no significance; ^*P < 0.05, ^**P < 0.01, ^***P < 0.001. hMSC-L vs hMSC-M ns no significance; ^#P < 0.05, ^##P < 0.01

We further investigated the expression levels of secretory factors from different neutrophil subtypes in the brain. Post-treatment with hMSCs, the expression of N1-specific enzymes—neutrophil elastase (NE), myeloperoxidase (MPO), and peptidylarginine deiminase 4 (PADi4)—was notably reduced (Fig. 5D, *P < 0.05, **P < 0.01, ***P < 0.001). Particularly, MPO expression was significantly lower in the hMSC-M group compared to the hMSC-L group (^#P _{hMSC-L vs hMSC-M} < 0.05). Conversely, the expression of TGF-β and IL-10, characteristic factors of the N2 subtype, significantly increased, whereas vascular endothelial growth factor (VEGF) expression did not show a statistically significant change (Fig. 5E, *P < 0.05, **P < 0.01, ***P < 0.001). These results indicate that hMSCs treatment promotes an increase in the N2 neutrophil subtype post-stroke. Notably, this shift appears to occur through the generation and migration of N2 neutrophils from bone marrow to the brain via the peripheral blood, rather than conversion post-infiltration into the brain.

Result 5: hMSC-derived supernatant activates PPAR-γ nuclear translocation to promote N2 neutrophil polarization

We further investigated the molecular mechanisms by which secreted factors from hMSCs regulate neutrophil polarization. PPAR-γ, a crucial transcription factor for cell differentiation, is typically located in the cytoplasm. Upon activation, PPAR-γ undergoes a conformational change, associates with nuclear transport proteins, and translocates to the nucleus to promote transcription of downstream genes [29]. STAT6 and SOCS1, which are signaling molecules related to PPAR-γ function, play significant roles in cell growth, metabolism, apoptosis, and inflammation regulation [30].

To validate the hypothesis, we first investigated the effects of hMSC-L/M intervention on the activation of primary neutrophils extracted from mouse bone marrow tissue. We utilized three different culture medium combinations: the complete lymphoid medium group (CM, control), a 1:1 mixture of CM and hMSC-L/M, and the hMSC-L/M group, stimulating the neutrophils in vitro for 4 h. Previous studies [31, 32] have indicated that the expression of CD11b and CD62L can reflect the inflammatory activation state of neutrophils; therefore, we employed flow cytometry to assess the expression and activity of these markers. The results showed that, compared to the control group, the proportion of Ly6G⁺CD11b^high and Ly6G⁺CD62L⁺ neutrophil subsets was significantly reduced in the hMSC-L/M group, with a marked decrease in the MFI values of CD11b and CD62L (Fig. 6A–F, vs control *P < 0.05, ***P < 0.001). Additionally, the proportion of Ly6G⁺CD62L⁺ neutrophils was also significantly lower than that in the mixed culture medium stimulation group (Fig. 6D–F, vs CM + L/M ^#P < 0.05, ^$P < 0.05). These findings suggest that hMSC-L/M may inhibit the pro-inflammatory activation, migration, and adhesion capabilities of neutrophils. However, our detection at the protein level revealed that stimulation with hMSC-L/M alone did not alter the activation levels of PPAR-γ (Fig. 6G, H).

Fig. 6

hMSCs facilitates the activation of neutrophils in vitro. A–C: Flow cytometry analysis of the proportion of CD11b⁺Ly6G⁺ neutrophils (A), three-dimensional curves (B), and cell percentage expression along with average fluorescence intensity (C). D–F: Flow cytometry analysis of the proportion of Ly6G⁺CD62L⁺ neutrophils, three-dimensional curves (B), and cell percentage expression along with average fluorescence intensity (C). G–H: Western Blot analysis of PPAR-γ and p-PPAR-γ protein expression (G) and statistical analysis (H). n = 4 or 6, versus control ns: no significance; *P < 0.05, **P < 0.01, ***P < 0.001. hMSC(L) versus CM + hMSC(L) ns no significance; ^#P < 0.05. hMSC(M) versus CM + hMSC(M) ns no significance; ^$P < 0.01

To further investigate the mechanism by which hMSC-L/M reprograms neutrophils, we used phorbol 12-myristate 13-acetate (PMA) to activate the neutrophils before introducing the hMSC-L/M intervention. PMA stimulation enhanced the nuclear translocation and expression of cytoplasmic and total PPAR-γ in neutrophils. Intervention with hMSCs further promoted the activation and nuclear translocation of PPAR-γ, reducing its phosphorylation levels. Inhibition of PPAR-γ using T0070907 significantly blocked the nuclear translocation of PPAR-γ induced by hMSCs treatments and decreased the expression levels of cytoplasmic and total PPAR-γ (Figs. 7A–D, hMSC-L vs T0070907 ^##P < 0.01, ^###P < 0.001; hMSC-M vs T0070907 ^$P < 0.05, ^$$$P < 0.001).

Fig. 7

hMSCs activate PPAR-γnuclear translocation to promote N2-type neutrophil polarization in vitro. A, B: Western Blot analysis of nuclear PPAR-γ protein expression and statistical analysis. C, D: Western Blot analysis of cytoplasmic PPAR-γ, total PPAR-γ, and phosphorylated PPAR-γ expression and statistical analysis. E, F: Flow cytometry analysis of the proportion of N2 (Ly6G⁺CD206⁺) and N1 (Ly6G⁺iNOS⁺) neutrophils. G: ELISA analysis of the secretion level of IL-10 and TGF-β in N2-type neutrophils. n = 5, vs PMA ^*P < 0.05, **P < 0.01, ***P < 0.001. PMA + L + inhibitior versus PMA + L ^##P < 0.01, ^###P < 0.001. PMA + M + inhibitior versus PMA + M ^$P < 0.01, ^$$P < 0.01, ^$$$P < 0.001

Under stimulation with PMA, neutrophils cultured in vitro typically exhibited polarization towards either N1 or N2 phenotypes. Upon the addition of human hMSC-L and hMSC-M, there was a significant increase in the proportion of N2-type neutrophils, accompanied by a marked decline in N1-type neutrophils. Conversely, inhibition of PPAR-γ shifted the polarization significantly towards the N1 phenotype (Figs. 7E, F, ^###P_hMSC-_{L vs T0070907} < 0.001, ^$$$P_{hMSC-M vs T0070907} < 0.001). Furthermore, ELISA results indicated that hMSCs stimulation significantly upregulated IL-10 and TGF-β (Figs. 7G), which are characteristic molecules of N2 neutrophils (^***P < 0.001). Inhibition of PPAR-γ expression reversed these trends, underscoring the regulatory role of PPAR-γ in neutrophil polarization and the anti-inflammatory effects mediated by hMSCs.

Subsequently, we evaluated the effects of secreted factors from hMSCs on the PPAR-γ/STAT6/SOCS1 signaling pathways through both in vitro and in vivo experiments. In vivo, compared to the sham group, the MCAO group exhibited increased protein expression levels within the PPAR-γ/p-PPAR-γ/SOCS1/STAT6/p-STAT6 pathway. Treatment with hMSCs further augmented the expression of these pathway components (see Supplementary Fig. 4). In vitro experiments corroborated these findings, demonstrating an increase in SOCS1 and STAT6 protein expression levels following stimulation with hMSCs, with a particularly pronounced activation in the hMSC-M treated cell group. Conversely, inhibiting the expression of PPAR-γ significantly down-regulated the protein expression levels of the SOCS1/STAT6 pathway compared to the hMSCs treatment group (see Supplementary Fig. 5). Furthermore, we discovered that hMSC intervention reversed the alterations in N1 (NE, MPO) and N2 (TGF-β, IL-10) marker expressions induced by PMA (Fig. 8A–F). Notably, the expression of N1 markers was significantly elevated compared to the hMSC-stimulated group following PPAR-γ inhibition. Concurrent inhibition of PPAR-γ and STAT6 (AS1517499) further upregulated N1 marker expression and downregulated N2 marker expression (Fig. 8A–D, vs hMSC-L/M ^$P < 0.05, ^$$P < 0.01, ^$$$P < 0.001). These trends align with the results obtained from ELISA assays on cell culture supernatants (Fig. 8E, F). These results suggest that cytokines secreted by hMSCs may promote the reprogramming of N2 type neutrophils and inhibit N1 type neutrophil polarization through the mediation of the PPAR-γ-driven STAT6/p-STAT6/SOCS1 signaling pathway.

Fig. 8

hMSCs intervention facilitates the reprogramming of N2 type neutrophils through the PPAR-γ/STAT6/SOCS1 signaling pathway. A: qPCR to detect the expression levels of NE and MPO after hMSC-L treatment. B: qPCR to detect the expression levels of TGF-β and IL-10 after hMSC-L treatment. C: qPCR to detect the expression levels of NE and MPO after hMSC-M treatment. D: qPCR to detect the expression levels of TGF-β and IL-10 after hMSC-M treatment. E: ELISA to detect the expression levels of TGF-β and IL-10 after hMSC-L treatment. F: ELISA to detect the expression levels of TGF-β and IL-10 after hMSC-M treatment. n = 4, versus control ^*P < 0.05, **P < 0.01, ***P < 0.001. versus PMA ^#P < 0.05, ^##P < 0.01, ^###P < 0.001. versus hMSC-L/M ^$P < 0.01, ^$$P < 0.01, ^$$$P < 0.001

Result 6: Intranasal delivery of hMSC-derived supernatant improves BBB damage and supports angiogenesis after cerebral ischemia in mice

Finally, we assessed the impact of hMSC-L and hMSC-M on BBB permeability in mice with cerebral ischemia. Immunofluorescence analysis revealed that treatment with hMSC-L and hMSC-M significantly enhanced the expression of tight junction-associated proteins, occludin (Fig. 9A, *P < 0.05, **P < 0.01) and ZO-1 (Fig. 9B, **P < 0.01). Additionally, the expression of the endothelial cell marker CD31 was significantly upregulated in the hMSC-treated groups, with more pronounced angiogenesis observed following hMSC-M treatment (Fig. 9C, **P < 0.01). These results suggest that hMSC-derived secretory supernatants may contribute to the repair of the BBB and promote vascular regeneration following cerebral ischemia in mice.

Fig. 9

hMSCs treatment improves blood–brain barrier integrity and supports angiogenesis after cerebral ischemia in mice. Immunofluorescence staining to detect the expression of occluding (A), ZO-1(B), and CD31(C) after treatment with hMSC-L and hMSC-M. n = 5, versus MCAO *P < 0.05, **P < 0.01

Discussion

Over the past two decades, numerous clinical trials have substantiated the safety and efficacy of stem cell therapy in treating ischemic stroke. Despite these advancements, controversies persist regarding the optimal route of administration, preservation of cell viability, and the precise mechanisms underlying therapeutic effects. In this study, we specifically investigate the efficacy and mechanistic of intranasal delivery using secretomes derived hMSCs for the treatment of IS. Our findings indicate that both hMSC-L and hMSC-M significantly mitigate neural damage in ischemic mice. This may be attributed to the modulation of N2 neutrophil reprogramming mediated by the PPAR-γ/p-PPAR-γ/SOCS1/STAT6/p-STAT6 pathway.

Stem cells derivatives, as bioactive agents, offer more flexible in vivo delivery methods compared to morphologically intact cells. This flexibility is particularly advantageous for treating central nervous system (CNS) diseases, where nasal-to-brain delivery pathways can bypass the BBB. These pathways, including the olfactory and trigeminal nerve routes, facilitate direct drug delivery [33]. Studies have indicated that intranasal drug delivery can achieve CNS drug concentrations up to ten times higher than those achieved through traditional injection, at comparable blood drug concentrations. The efficacy of intranasal delivery of neurotrophic factors, insulin, and erythropoietin has been demonstrated across various CNS diseases [34,35,36]. For example, in a Parkinson’s disease rat model, nasal delivery of mesenchymal stem cells or vesicles resulted in higher transplantation rates compared to injections into the striatum, with stem cells still detectable months later [37]. In this study, we compared the therapeutic effects of nasal delivery, intraperitoneal injection, and tail vein injection of hMSCs derivatives in treating IS. Our findings indicated that all three delivery routes effectively reduced cerebral ischemic injury in mice. Notably, intranasal administration of hMSC-derived products was at least as effective as other conventional routes. Given its comparable efficacy, the intranasal approach offers significant advantages: it is more convenient, non-invasive, and unrestricted by the BBB. Such benefits are particularly meaningful in managing acute IS, potentially offering life-saving advantages.

The therapeutic mechanisms of hMSCs transplantation in IS have been extensively investigated. Beyond supporting angiogenesis and nerve regeneration, a critical role of hMSCs is their participation in immune inflammation regulation through the secretion of numerous cytokines [38]. However, research has revealed that the homing and secretory capacities of transplanted living cells are often less than anticipated. Subsequent studies have highlighted the benefits of structures such as extracellular vesicles and exosomes, which are coated with factors secreted by stem cells, for treating IS [39]. In this study, we collected the secretory supernatant of hMSCs and employed a mild freeze–thaw method to prepare cell lysates, preserving the activity of all cellular components. We observed that these hMSCs derivatives significantly ameliorated cerebral ischemic injury in mice. The beneficial effects included not only reductions in ischemic volume and improvements in neurological scores and motor functions but also enhancements in angiogenesis and increased expression of tight junction proteins. While this study did not specifically identify the macromolecular proteins contained in hMSC-L and hMSC-M, it is recognized that hMSCs release a range of factors including vascular endothelial growth factor, hepatocyte growth factor, brain-derived neurotrophic factor, and glial-derived neurotrophic factor, etc. [40]. Indeed, the therapeutic efficacy of hMSCs cannot be fully explained by the action of single molecules. Rather, the comprehensive effects on nerve regeneration and immune regulation are likely due to the synergistic action of multiple factors that together facilitate systemic repair of the neural function network.

Post-stroke, the ongoing inflammation that occurs both centrally and peripherally presents a unique therapeutic target to potentially delay or prevent secondary neuronal necrosis [41]. The role of stem cell therapy in immune regulation and inflammation inhibition is widely recognized. In this study, we observed that intranasal delivery of hMSC-L and hMSC-M significantly reduced neuroinflammatory levels in the brain. Notably, classic pro-inflammatory cytokines such as IFN-γ, TNF-α, and IL-1β showed significant decreases. While the impact on macrophages and blood-borne monocytes was not marked, we noted a specific reduction in the rate of neutrophil infiltration, particularly with the hMSC-M treatment, where the decrease was substantial. Interestingly, the changes in neutrophils extended beyond the CNS; in the peripheral blood and spleen, the proportion of neutrophils decreased, but in the bone marrow, it increased significantly. These findings suggest that neutrophils may be the primary effector cells through which hMSC-L and hMSC-M exert their therapeutic effects in the treatment of ischemic stroke.

To confirm the specific effects of hMSC-L and hMSC-M on neutrophils, we first investigated their impact on cellular activation. CD11b and CD62L are key markers reflecting the activation state of neutrophils. In the resting state, CD11b is mainly stored in intracellular granules, and upon external stimulation, intracellular CD11b is rapidly translocated to the cell surface [31]. Conversely, the surface expression level of CD62L is high in resting neutrophils but is quickly shed upon inflammatory signals due to the action of ADAM17 [32]. Elevated CD11b expression and CD62L shedding both indicate a transition of neutrophils towards a state of migration and adhesion [31]. Our results demonstrated that, under the influence of hMSC-L/M, the proportion of Ly6G⁺CD11b^high and Ly6G⁺CD62L⁺ neutrophil subsets was significantly reduced, and the signal intensities of CD11b and CD62L markedly decreased, suggesting a change in the neutrophil state. The further research demonstrates that hMSC-L and hMSC-M appear to facilitate the reprogramming of neutrophils towards the N2 phenotype. Notably, while the proportion of N2 neutrophils in peripheral immune organs decreased in treated mice, it significantly increased in the brain. Although the polarization mechanisms of microglia are well-understood, the pathways involved in neutrophil polarization in disease states remain less clear.

Previous studies have shown that PPARγ is involved in the IL-4/IL-13-induced polarization of M2 macrophages. [42] Some traditional Chinese medicinal compounds activate PPARγ and inhibit the STAT1/JAK2 downstream signaling pathways, promoting M2 polarization of microglial cells [43], underscoring PPARγ’s pivotal role in anti-inflammatory pathways. Regulatory proteins downstream of PPARγ, such as SOCS1, can competitively inhibit JAK’s catalytic activity through its SH2 domain, thus impeding pro-inflammatory signaling [44]. Additionally, phosphorylated STAT6 promotes SOCS1 gene transcription [45], which further regulates the generation of phosphorylated STAT6 by interacting with JAK [46]. We found that the PPARγ/SOCS1/STAT6 signaling pathway may also play a regulatory role in the hMSC-L/M-mediated reprogramming of neutrophil subtypes following stroke. In vivo experiments showed that hMSC-L/M treatment significantly enhanced the expression of PPAR-γ, SOCS1, and STAT6 in the brain. In vitro experiments revealed that stimulation with hMSC-L/M alone did not seem to alter PPAR-γ protein levels in neutrophils. However, upon PMA stimulation, hMSC-L and hMSC-M activated PPAR-γ nuclear translocation and upregulated SOCS1/STAT6 signaling in neutrophils.

The inhibition of PPARγ with pharmacological inhibitors led to a significant decrease in N2 polarization and a reduction in the levels of anti-inflammatory cytokines IL-10 and TGF-β. Simultaneous inhibition of PPAR-γ and STAT6 further upregulated the expression of N1 neutrophil markers while significantly downregulating the expression and secretion of N2-type markers. These findings reveal, for the first time, the mechanisms by which hMSC-L and hMSC-M mediate the polarization of neutrophils towards the N2 type through the PPARγ/SOCS1/STAT6 pathway.

In summary, our study demonstrated that intranasal delivery of mesenchymal stem cell secreted supernatant and cell lysates can mitigate neuroinflammation and ischemic brain damage. This effect is achieved by promoting the N2 reprogramming of neutrophils through activation of the PPAR-γ-mediated SOCS1/STAT6/p-STAT6 pathway. However, this study has several limitations. Firstly, we did not thoroughly analyze the major active components within these stem cell derivatives, nor did we investigate their long-term efficacy and safety in stroke models. Furthermore, significant changes were observed in the cellular components of peripheral immune organs in mice post-treatment, yet our study lacks a detailed analysis of the pharmacokinetics and pharmacodynamics of the administered compounds post-intranasal delivery. In addition, the current study has not yet provided a detailed explanation of the effects and mechanisms of intranasally administered hMSC-L/M on other major cell types within the CNS. A deeper understanding of these aspects is crucial to advancing the clinical application of stem cell therapies and enhancing their therapeutic potential.

Conclusions

Our study demonstrated that intranasal delivery of human mesenchymal stem cell supernatant and cell lysates can alleviate ischemic injury by modulating the neuroinflammatory response in mice suffering from stroke. This therapeutic effect is likely mediated through the PPAR-γ/SOCS1/STAT6 pathway, which promotes N2 polarization of neutrophils. This provides a new approach for developing more convenient, safe, and effective stem cell-based stroke therapies. Subsequent research should further validate the clinical potential of hMSC-L/M and optimize delivery strategies to enhance treatment efficacy for stroke.