Electroacupuncture Alleviates Cerebral Ischemia–Reperfusion Injury by Downregulating IL-17 A and Inhibiting Neurotoxic Astrocyte Activation

Abstract

Cerebral ischemia-reperfusion injury (CIRI) represents a critical pathological mechanism underlying ischemic stroke, yet effective therapeutic interventions remain limited. Neurotoxic astrocytes, activated by inflammatory mediators such as interleukin-17 A (IL-17 A), exacerbate neuronal damage. Although electroacupuncture (EA) has demonstrated neuroprotective properties, its influence on IL-17 A signaling and subsequent astrocyte-mediated neurotoxicity in CIRI remains unclear. This study aims to investigate whether EA mitigates CIRI by downregulating IL-17 A to suppress the activation of neurotoxic astrocyte. A mouse model of middle cerebral artery occlusion and reperfusion (MCAO/R) was established employing the Zea-Longa modified ligation method. EA was applied to the Baihui (GV20) and Fengfu (GV16) acupoints. Neurological and behavioral evaluations were performed using the Modified Neurological Severity Score (mNSS), foot fault test, and balance beam test. Cerebral infarction volume was quantified via TTC staining, and neuronal ultrastructure was examined by transmission electron microscopy. Laser speckle imaging was employed to monitor cerebral blood flow before and after modeling and EA treatment. Western blotting was used to analyze protein expression levels of IL-17 A, IL-17RA, NF-κB p65, Bax, Bcl-2, and cleaved-Caspase-3/Caspase-3. Co-localization of IL-17 A with GFAP and C3, as well as IL-17RA with GFAP, was assessed via immunofluorescence staining. qPCR was performed to quantify IL-17 A mRNA levels, while TUNEL staining assessed neuronal apoptosis. ELISA was used to determine the concentrations of IL-17 A, TNF-α, and IL-1β in brain tissue. EA significantly improved neurological function, reduced cerebral infarct size, and alleviated neuronal apoptosis. Compared to the MCAO/R group, EA markedly downregulated IL-17 A expression and its related signaling proteins, inhibited neurotoxic astrocyte activation (C3⁺/GFAP⁺), and suppressed the release of proinflammatory cytokines. Notably, administration of recombinant IL-17 A reversed the neuroprotective effects of EA. These findings suggest that EA mitigates ischemic brain injury by inhibiting IL-17 A-mediated neurotoxic astrocyte activation and neuroinflammation, highlighting its potential as a therapeutic strategy for CIRI.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s11064-025-04603-8.

Introduction

Cerebral stroke ranks as the second leading cause of death and the third leading cause of disability globally, with nearly 7 million annual fatalities attributed to stroke, representing a major global healthcare challenge worldwide [1]. Ischemic stroke (IS), which is the most prevalent form, accounted for 62.4% of new stroke cases in 2019. This condition arises from disrupted cerebral blood flow due to various factors, which leads to ischemia, hypoxia, and subsequent tissue damage [2, 3]. Given its rapid onset, narrow therapeutic window for thrombolysis, and severe complications, options for effective clinical treatments remain limited [4, 5]. Thus, investigating the biological mechanisms of post-ischemic recovery, developing proactive therapeutic strategies, and identifying novel treatment targets for IS are of both practical and theoretical significance.

Neuronal apoptosis remains a well-established mechanism in CIRI pathogenesis. Emerging evidence highlights the crucial involvement of neuro-immune interactions, particularly involving reactive astrocytes and their secreted inflammatory cytokines, in exacerbating secondary brain injury [6, 7]. As a key inflammatory mediator in Th17 responses, IL-17 A has been increasingly implicated in CIRI progression [8, 9]. Clinical observations have correlated elevated IL-17 A levels in serum and cerebrospinal fluid with larger infarct volumes and poorer outcomes in IS patients [10]. Mechanistically, IL-17 A worsens blood-brain barrier dysfunction through NF-κB and MAPK pathway activation while directly inducing microglial polarization toward the M1 phenotype [11, 12]. Importantly, the high expression of IL-17 receptors on astrocytes suggests their potential role as primary responders to CNS IL-17 A signaling [13, 14].

Astrocytes play a complex dual role in maintaining homeostasis within the central nervous system. During ischaemic injury, this duality is reflected in as two distinct functional phenotypes: A1 and A2. A1-type astrocytes exacerbate tissue damage within the ischaemic core by releasing inflammatory mediators and reactive oxygen species, thereby hindering neural repair. Conversely, A2-type astrocytes exert neuroprotective effects on neurons within the ischaemic penumbra by secreting multiple neurotrophic and protective factors, while also promoting neuroregeneration [15]. Notably, during the acute ischemic phase, astrocytes demonstrate a tendency to transition towards the neurotoxic A1 phenotype. This phenotype is characterised by the upregulation of complement component C3 and increased secretion of pro-inflammatory factors such as TNF-α and IL-1β [16]. This transformation further intensifies the local inflammatory response, leading to widespread neuronal death and damage to synaptic structure and function [17, 18]. Moreover, excessive activation results in the formation of glial scarring, which impedes axonal regeneration and repair [19]. Research indicates that IL-17 A may contribute to the process of neuronal apoptosis following CIRI by mediating the activation of type A1 astrocytes [20, 21]. Consequently, therapeutic targeting of IL-17 A-driven A1-type astrocyte formation may represent a promising approach to alleviate neuroinflammation and enhance functional recovery in IS.

Acupuncture serves as an effective preventive and therapeutic intervention for IS, demonstrating holistic regulatory effects and multi-targeted comprehensive actions [22, 23]. Studies indicate its ability to modulate diverse immune response processes, particularly in context of neuroimmunomodulation [24, 25]. As crucial participants in neuroimmune responses, astrocytes have emerged as key therapeutic targets for IS [26], though most research has focused on downstream effects rather than acupuncture’s mechanisms underlying the alteration of astrocyte structural and functional phenotypes [27, 28]. Current evidence shows that acupuncture can regulate plasma IL-17 A levels [29], and given IL-17 A’s close association with astrocyte phenotype polarization and their shared role in IS-related neuroinflammation, this study aims to examines whether EA mitigates inflammatory brain damage in MCAO/R mice by suppressing neurotoxic A1-type astrocyte activation through the IL-17 A downregulation.

Materials and Methods

Animals

Six- to eight-week-old male C57BL/6J mice (weighing 24–26 g) of specific pathogen-free (SPF) grade were obtained from the Shandong Experimental Animal Center (production license: SCXK [Zhejiang] 2019-0004). After a one-week acclimatization period under a 12-hour light/dark cycle at 20–25 °C and 40–75% relative humidity, the mice underwent balance beam training to exclude those exhibiting excessive sensitivity or sluggishness. Using the random number table method: In Experiment 1, 45 mice were randomly assigned to three groups (Sham, MCAO/R, and EA; n = 15/group), while Experiment 2 divided 40 mice into four groups (MCAO/R + PBS, MCAO/R + IL-17 A, EA + PBS, and EA + IL-17 A; n = 10/group). The experimental protocol received approval from the Animal Ethics Committee of Anhui University of Traditional Chinese Medicine (approval number: AHUCM-mouse-2024135).

MCAO/R Mouse Model

Mice were fasted for 12 h prior to the experiment with free access to water. The left middle cerebral artery occlusion/reperfusion (MCAO/R) model was established using the classical suture embolization method. The detailed surgical procedure was as follows: anesthesia was initially induced with isoflurane inhalation (flow rate: 0.41 ml/min, maintenance concentration: 1%). After ligating the common carotid artery and the proximal external carotid artery, a small V-shaped incision was made on the common carotid artery. A silicone-coated embolization filament (tip diameter: 0.24 mm) was then inserted from the carotid bifurcation into the internal carotid artery and advanced approximately 8–10 mm to occlude the origin of the middle cerebral artery. After 90 min of occlusion, the filament was carefully withdrawn to restore cerebral blood flow. Sham-operated mice underwent identical surgical procedures (including vessel dissection) but without embolus insertion. Detailed procedures for establishing the MCAO/R model are described in the Supplementary Materials.

EA Treatment

EA treatment was initiated 24 h after mold establishment, targeting the Baihui (GV20) and Fengfu (GV16) acupoints as per “Experimental Acupuncture” guidelines. Under isoflurane anesthesia (0.41 ml/min, 1% concentration), mice were secured to a board and treated with 0.25 × 15 mm sterile needles connected to an SDZ-V EA device (Hwato Electronic Acupuncture Device SDZ-III). The proximal and distal ends of each acupoint were linked to the positive and negative electrodes, respectively. The treatment parameters included continuous wave at a frequency 2 Hz/10 Hz and a current of 0.5 mA, which was gradually increased until mild local trembling (without vocalization or struggle) was observed. Each 20-minute session was administered daily for 7 consecutive days.

Administration of Recombinant Mouse Interleukin-17 A (rmIL-17 A) Protein

RmIL-17 A (HY-P70753, MCE) was dissolved in phosphate-buffered saline (PBS) and administered intranasally at 4 µg/kg body weight daily for seven consecutive days, while control mice received PBS alone under identical conditions. This delivery method enables direct CNS access via the olfactory pathway, effectively circumventing the blood-brain barrier [30], with the selected dosage (4 µg/kg) having been previously established to reliably induce neuroinflammatory responses [31].

Modified Neurological Severity Score (mNSS)

The neurological function of mice in each group was evaluated at 4 h post-surgery and 7 days post-treatment using the mNSS, which assesses motor, sensory, balance beam, and reflex functions on an 18-point scale (0–18), with higher scores indicating more severe neurological impairment. The specific criteria are outlined in Supplementary Material (Table 1).

Behavioral Test

On the 7th day following the intervention, mice from each group underwent behavioral assessments designed to evaluate motor coordination and integration. The balance beam test utilized a 100 cm × 1 cm beam positioned 50 cm above the ground, with mice having been pre-trained preoperatively for normal beam traversal. Performance was scored according to the Feeney 5-point scale. The specific criteria are outlined in Supplementary Material (Table 2).

For the foot fault assessment, mice were allowed to walk freely for 1 min on a 32 cm × 20 cm steel grid (with openings measuring1.2 cm × 1.2 cm) while being recorded from below. Right forelimb missteps and total steps were quantified, with error rate calculated as: (missteps/[missteps + correct steps])×100%.

2,3,5-Triphenyltetrazolium Chloride (TTC) Staining

The extracted brain tissue was immediately frozen at −20 °C for 20 min and then serially sectioned in the coronal plane to generate five slices, each with a thickness of 2 mm. These slices were incubated in a 2% solution of 2,3,5-triphenyltetrazolium chloride (TTC; T8877, Sigma-Aldrich) for 30 min at 37 °C under light-protected conditions, followed by fixation in 4% paraformaldehyde for 24 h. Digital images were acquired using ImageJ software (NIH, MD), and infarct area was quantified by calculating the percentage of total infarct area relative to total brain area.

Transmission Electron Microscopy (TEM)

The brain tissue isolated from the infarct region was fixed in 2.5% glutaraldehyde, rinsed in phosphate buffer, post-fixed with osmium tetroxide, then washed twice with distilled water, dehydrated through an acetone gradient, embedded in epoxy resin 812, and ultrathin sectioned at 60 nm thickness. The sections were double-stained with 3% uranyl acetate and lead citrate before being examined under a transmission electron microscope for ultrastructural analysis and imaging.

Laser Speckle Contrast Imaging (LSCI)

Anesthesia was maintained with inhaled isoflurane (0.41 ml/min, 1% concentration). The mouse was positioned prone on the operating table with the skull leveled in all directions. Following scalp shaving and disinfection with iodine solution, a 2 cm mid-sagittal scalp incision was made using scissors to expose the skull, which was continuously moistened with physiological saline. Cerebral blood flow perfusion images were captured under laser indicator light, with quantitative observations conducted using consistent coordinate-based regions of interest (ROI). Data analysis was performed using moorFLPIReviewV50 software, comparing cerebral brain blood flow perfusion volumes before and after modeling, as well as before and after treatment in the same mouse, with statistical analysis employed to evaluate differences.

Western Blotting (WB)

Mouse cerebral infarction tissue was homogenized in RIPA lysis buffer for total protein extraction. The proteins were separated by 10% SDS-PAGE and transferred to a PVDF membrane (Millipore, USA). After blocking with milk, the membranes were incubated overnight at 4 °C with primary antibodies against IL-17 A (1:1000, ab79056, Abcam, UK), IL-17RA (1:1000, CSB-PA12929A0Rb, Huamei Bio), NF-κB p65 (1:1000, R25149, Zenbio, China), Bax (1:1000, R22708, Zenbio), Bcl-2 (1:1000, R23309, Zenbio), Caspase3 (1:1000, 19677-1-AP, Proteintech, China), and β-actin (1:1000, 200068-8F10, Zenbio). Following PBST washes, the membranes were incubated with the corresponding mouse secondary antibodies on a shaker for 2 h. Protein bands were visualized using ECL chemiluminescence, and band intensities were quantified with ImageJ software.

Immunofluorescence (IF) Staining

Mice were transcardially perfused, after which brains were rapidly extracted and fixed in 4% paraformaldehyde. The subsequent steps included dehydration in 30% sucrose, OCT embedding, antigen retrieval, washing with PBS, and blocking with goat serum. Primary antibodies—IL-17 A (1:200, ab79056, Abcam, UK), GFAP (1:200, bsm-42001 M, Bioss, Beijing), IL-17RA (1:200, CSB-PA12929A0Rb, Huamei Bio), and C3 (1:200, 21337-1-AP, Proteintech, Wuhan, China)—were applied and incubated overnight at 4 °C. Following incubation, fluorescent secondary antibodies (1:100) were added and incubated in the dark. After thorough washing with PBS, samples were mounted with DAPI-containing anti-fade medium and imaged using a laser confocal microscope for analysis.

Quantitative Real-Time PCR(qPCR)

Mouse cerebral infarction tissue was collected, and total RNA was extracted using TRIzol reagent. For RNA isolation, 10–30 mg of tissue was homogenized with 1 mL of TRI Reagent using an electric grinder. To eliminate residual genomic DNA, 1 µL gDNA digester, 2 µL 5× gDNA digester buffer, and RNase-free H₂O were added to the sample, adjusting the total volume to 10 µL, followed by incubation at 42 °C for 2 min. Reverse transcription was performed by adding 2 µL Hifair II Enzyme Mix, 2 µL 5× Hifair II Buffer Plus, 1 µL Oligo (dT) 18 (50 µM), and RNase-free H₂O to the reaction mixture (final volume: 20 µL) under the following conditions: 25 °C for 5 min, 42 °C for 30 min, and 85 °C for 5 min. The resulting cDNA was amplified via qPCR using the following program: initial denaturation at 95 °C for 3 min, followed by 40 cycles of 95 °C for 15 s and 60 °C for 30 s. β-actin served as the internal control to quantify IL-17 A mRNA levels. Relative gene expression was calculated using the 2^−ΔΔCt method, and data analysis was conducted with LightCycler 480 fluorescence quantitative PCR software. The primer sequence list for this research is shown in Supplementary Material (Table 3).

Enzyme-Linked Immunosorbent Assay (ELISA)

The supernatant obtained from homogenized mouse brain infarction tissue is quantitatively analyzed for IL-17 A, IL-1β, and TNF-α levels using ELISA kits (Boster, EK0431, EK0394, and EK0527), with all measurements performed in triplicate.

TUNEL Staining

Following cryosectioning, the samples were stained with a TUNEL assay kit (Beyotime, China) and subsequently mounted for observation, while TUNEL-positive cells were quantified in each group using laser confocal microscopy.

Statistical Analysis

The statistical analyses were conducted using SPSS 26.0 (IBM, USA) and GraphPad Prism 8.0 (Boston, MA), with all normally distributed data presented as mean ± standard deviation (SD). Intergroup comparisons were analyzed by one-way ANOVA, adopting p < 0.05 as the significance threshold.

Results

EA Improves Neurological Dysfunction in MCAO/R Mice

This study investigated the neurofunctional benefits of EA in MCAO/R mice through mNSS scoring, foot fault tests, balance beam test, TTC staining, and laser speckle imaging. The MCAO/R group exhibited significantly elevated neurological deficit scores compared with the Sham group (Fig. 1B), demonstrating substantial impairment, while EA treatment markedly reduced these scores (Fig. 1B), indicating functional recovery. The balance beam test revealed prolonged crossing times and increased slips in MCAO/R mice (Fig. 1C), with EA intervention significantly improving both parameters (Fig. 1C). The foot fault test showed higher error counts in MCAO/R animals (Fig. 1D), which were substantially decreased by EA (Fig. 1C), reflecting enhanced motor coordination. TTC staining demonstrated significantly larger infarct volumes in MCAO/R compared with Sham groups (Fig. 1E, G), while EA treatment led to a reduction in infarction size (Fig. 1E, G), suggesting neuroprotective effects. Laser speckle imaging revealed impaired cerebral perfusion in MCAO/R mice (Fig. 1F, H), which was significantly improved by EA (Fig. 1F, H). These findings collectively demonstrate that EA promotes neurobehavioral recovery in MCAO/R mice through reducing infarct volume and enhancing cerebral perfusion, thereby improving neurological function, motor coordination, and balance, supporting its therapeutic potential for CIRI.

Fig. 1.

EA enhances neurological recovery in MCAO/R mice. A Experiment 1: Experimental flow chart. B mNSS, (n = 8). C Balance Beam Test (Feeney 5-point scale, n = 8). D Foot Fault Rate (n = 8) across sham, model, and EA groups. E Ischemic hemisphere infarct volume (n = 3). F Ischemic-side cerebral blood flow perfusion (n = 3). G–H Quantitative analyses of infarct volume and perfusion, respectively. Data are expressed as mean ± SD. *p < 0.05; **p < 0.01

EA Improves Neuronal Damage in Ischemic Brains of MCAO/R Mice

To investigate the neuroprotective effects of EA against neuronal apoptosis in CIRI, we utilized WB, TUNEL staining, and TEM. WB analysis revealed significantly elevated Bax levels and reduced Bcl-2 expression in MCAO/R compared to Sham groups (Fig. 2A-C), which EA treatment reversed (Fig. 2A-C). The cleaved-Caspase3/Caspase3 ratio showed similar patterns, being markedly higher in MCAO/R (Fig. 2A, D) and attenuated by EA (Fig. 2A, D). TUNEL staining corroborated these findings, demonstrating significantly fewer apoptotic cells in EA-treated compared to MCAO/R groups (Fig. 2E, F) despite increased apoptosis in MCAO/R compared to Sham (Fig. 2E, F). The results demonstrate that EA suppresses MCAO/R-induced neuronal apoptosis by modulating Bax and Bcl-2 expression. TEM analysis revealed that the MCAO/R group exhibited severe ultrastructural damage, including nuclear condensation, chromatin marginalization, mitochondrial swelling, cristae rupture, and vacuolization, compared to the sham group. In contrast, the EA group showed significantly reduced neuronal damage (Fig. 2G), suggesting that EA reduces neuronal apoptosis by inhibiting Caspase-3 activation. Collectively, these findings indicate that EA alleviates MCAO/R-induced neuronal apoptosis and improves ultrastructural integrity by regulating Bax/Bcl-2 expression and suppressing Caspase-3 activation.

Fig. 2.

EA ameliorates neuronal damage in the ischemic brain of MCAO/R mice. A Protein expression profiles of Bax, Bcl-2, Caspase3, and cleaved-Caspase3. B–D Quantitative analyses of Bax/β-actin, Bcl-2/β-actin, and cleaved-Caspase3/Caspase3 levels (n = 3). E–F Representative TUNEL staining images of apoptotic cells and their quantification across sham, MCAO/R, and EA groups (scale bar 20 μm, n = 3). G Ultrastructural neuronal changes in the ischemic penumbra via TEM, with red arrows marking nuclei, yellow arrows indicating rough endoplasmic reticulum, and blue arrows highlighting mitochondria (scale bar 2 μm, 500 nm). Data are expressed as mean ± SD. *p < 0.05; **p < 0.01

EA Inhibits the Release of Ischemic Encephalitis Factor IL-17 A

To examine the impact of EA on IL-17 A expression after IS, we measured IL-17 A protein, mRNA levels, and positive cell counts by WB, qPCR, and IF staining in Sham, MCAO/R, and EA-treated groups. WB revealed significantly elevated IL-17 A protein levels in MCAO/R brain tissue compared with Sham (Fig. 3A), and this increase was attenuated after EA treatment (Fig. 3A). Similarly, qPCR analysis showed markedly increased IL-17 A mRNA levels in MCAO/R mice compared to Sham (Fig. 3E), and EA treatment significantly reduced expression (Fig. 3E). IF staining indicated that IL-17 A was predominantly expressed in neurons and glial cells within the ischemic penumbra, with MCAO/R mice exhibiting more IL-17 A-positive cells than Sham (Fig. 3F, G), while EA decreased this count (Fig. 3F, G). Since IL-17 A binds to receptors on glial/endothelial cells, thereby activating the NF-κB pathway and exacerbating inflammation [32], we assessed IL-17RA and NF-κB p65 by WB. MCAO/R mice showed increased IL-17RA and NF-κB p65 protein levels compared with Sham (Fig. 3C, D), which EA mitigated (Fig. 3C, D). These findings suggest that EA suppresses IL-17 A release following stroke, potentially exerting neuroprotection effects through inhibition of IL-17 A-driven inflammation.

Fig. 3.

EA suppresses the release of the ischemic inflammation factor IL-17 A. A Protein expression profiles of IL-17 A, IL-17RA, and NF-κB p65. B–D Quantitative analysis of IL-17 A/β-actin, IL-17RA/β-actin, and NF-κB p65/β-actin protein expression, respectively (n = 3). E Relative IL-17 A mRNA expression (n = 3). IF staining: F Quantification of IL-17 A-positive cells in the infarct region (scale bar 20 μm, n = 3), and G representative IF images. Data are expressed as mean ± SD. *p < 0.05; **p < 0.01

EA Inhibits Neurotoxic Astrocyte Activation and Related Factor Secretion

To further examine the impact of EA on astrocyte activation, we assessed the expression of GFAP and C3 using IF staining. GFAP serves as a marker for astrocyte activation, whereas C3 indicates a neurotoxic astrocyte phenotype. Our findings revealed significantly higher GFAP/C3 co-expression in the MCAO/R group compared to the Sham group (Fig. 4A, B), demonstrating MCAO/R-induced astrocyte activation. Notably, EA treatment substantially reduced GFAP/C3 co-expression versus the MCAO/R group (Fig. 4A, B), indicating its inhibitory effect on both astrocyte activation and the neurotoxic phenotype transition. Since neurotoxic astrocytes can aggravate inflammatory brain damage through cytokine secretion and the amplification of inflammatory cascades, we subsequently measured tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) secretion levels across all groups (Sham, MCAO/R, EA) using ELISA. The results demonstrated significantly elevated TNF-α and IL-1β levels in the MCAO/R group compared to the Sham controls (Fig. 4C, D), confirming MCAO/R-induced neuroinflammation. EA treatment markedly reduced these inflammatory markers versus MCAO/R (Fig. 4C, D), indicating effective suppression of inflammatory factor secretion. These findings indicate that EA exerts neuroprotective effects by attenuating neuroinflammation through inhibition of astrocyte activation and their transition to neurotoxic phenotypes. As a classical trigger for astrocyte activation, NF-κB activation plays a key role in driving the transition of astrocytes into a reactive state [33]. The NF-κB signaling pathway acts downstream of IL-17 A, and the decreased expression of NF-κB p65 protein following EA treatment indicates suppression of its activation. Based on integrated results from ELISA, immunofluorescence, and Western blot analyses, we found that EA significantly suppressed MCAO/R-induced activation of neurotoxic reactive astrocytes and the release of associated inflammatory factors. This effect is likely mediated through the inhibition of IL-17 A secretion and IL-17RA receptor expression, leading to reduced expression and activation of downstream NF-κB p65 protein.

Fig. 4.

EA suppresses neurotoxic astrocyte activation and associated factor secretion. A, B Representative IF images and quantitative analysis of GFAP and C3 expression in the peri-infarct region (scale bar 20 μm, n = 3). C, D ELISA quantification of inflammatory cytokines IL-1β and TNF-α in the peri-infarct area (n = 3). Data are expressed as mean ± SD. *p < 0.05; **p < 0.01

IL-17 A is One of the Key Targets of EA Treatment

Given the above findings, it remains uncertain whether the observed mitigation of neuroinflammatory damage is specifically mediated through EA-mediated downregulation of IL-17 A and subsequent suppression of neurotoxic astrocytes. To clarify this potential mechanism, recombinant IL-17 A was administered as an intervention (Fig. 5). Phosphate-buffered saline (PBS) was used as a vehicle control to ensure that any observed effects could be specifically attributed to IL-17 A. The recombinant IL-17 A was applied in four experimental groups: MCAO/R + PBS, MCAO/R + IL-17 A, EA + PBS, and EA + IL-17 A. WB, IF staining, and ELISA were employed to assess the expression of key molecules. Initial WB and ELISA analyses confirmed that, compared to MCAO/R + PBS, MCAO/R + IL-17 A significantly increased IL-17 A levels, whereas EA + PBS markedly reduced these levels (Fig. 6A, B, E). Notably, the administration of recombinant IL-17 A effectively counteracted EA’s suppressive effect, resulting in significantly higher IL-17 A levels in the EA + IL-17 A group compared to EA + PBS (Fig. 6A, B, E). A similar trend was observed for the IL-17 A-specific receptor IL-17RA: MCAO/R + IL-17 A upregulated its expression compared to MCAO/R + PBS (Fig. 6A, C), while EA + PBS downregulated it (Fig. 6A, C), an effect fully reversed by IL-17 A co-administration (Fig. 6A, C). Downstream of IL-17RA signaling, the transcription factor NF-κB p65 exhibited parallel changes—MCAO/R + IL-17 A robustly enhanced its expression compared to MCAO/R + PBS (Fig. 6A, D), whereas EA + PBS attenuated this response (Fig. 6A, D). Critically, co-treatment with recombinant IL-17 A completely abolished EA’s inhibitory effect on NF-κB p65 (Fig. 6A, D). IF staining further localized IL-17RA expression, particularly in astrocytes (GFAP-labeled). Consistent with WB data, the MCAO/R + IL-17 A group exhibited significantly increased IL-17RA immunoreactivity, with pronounced colocalization with GFAP-positive astrocytes in the peri-infarct region (Fig. 6F, G). EA + PBS treatment markedly reduced both IL-17RA staining intensity and its colocalization with GFAP (Fig. 6F, G), an effect largely negated in the EA + IL-17 A group (Fig. 6F, G). To evaluate the functional impact of the IL-17 A/IL-17RA/NF-κB axis modulation on neuroinflammation, pro-inflammatory cytokine levels in brain homogenates were measured via ELISA. The MCAO/R + IL-17 A group showed significantly elevated IL-1β and TNF-α levels compared to the MCAO/R + PBS group (Fig. 6H, I), while EA + PBS treatment effectively suppressed these inflammatory factors (Fig. 6H, I). However, recombinant IL-17 A co-administration (EA + IL-17 A group) fully reversed the anti-inflammatory effects of EA, resulting in cytokine levels significantly higher than those in the EA + PBS group (IL-1β, TNF-α; Fig. 6H, I). In summary, recombinant IL-17 A abolished the neuroprotective and anti-inflammatory effects of EA by reactivating the IL-17RA/NF-κB axis in astrocytes and amplifying systemic pro-inflammatory cytokine release, thereby identifying IL-17 A as a critical mediator of post-stroke inflammation and a potential regulatory target for EA therapy.

Fig. 5.

Experiment 2: Experimental flow chart

Fig. 6.

Recombinant IL-17 A administration counteracts the pro-inflammatory effects of EA on neuroinflammation. A Protein expression profiles of IL-17 A, IL-17RA, and NF-κB p65. B–D Quantitative analysis of IL-17 A/β-actin (n = 3), IL-17RA/β-actin (n = 4), and NF-κB p65/β-actin (n = 4), respectively. E ELISA quantification of IL-17 A in the peri-infarct region (n = 3). F, G Representative IF images and quantitative analysis of GFAP/IL-17RA co-staining in the peri-infarct region (scale bar 20 μm, n = 3). H, I ELISA-based quantification of inflammatory cytokines IL-1β and TNF-α in the peri-infarct region (n = 3). Data are expressed as mean ± SD. *p < 0.05; **p < 0.01

Discussion

Our study demonstrates that CIRI significantly upregulates IL-17 A levels and promotes the induction of A1-type astrocytes, which correlate strongly with increased cerebral infarction volume and impaired behavioral performance. Furthermore, we establish that EA treatment downregulates IL-17 A to suppress neurotoxic astrocyte C3 expression, decreases pro-inflammatory cytokines IL-1β and TNF-α release, and mitigates neuronal apoptosis. These neuroprotective effects are mediated through inhibition of Caspase-3 activation and modulation of Bax/Bcl-2 expression.

Extensive research has confirmed that EA exhibits neuroprotective effects in the MCAO/R model through multiple pathways, such as modulating energy metabolism imbalance, counteracting excitotoxicity, suppressing neuroinflammatory cascades, and enhancing neural regeneration [34–37]. Our findings revealed that the EA-treated group showed markedly better outcomes than the model group, characterized by significant improvements in neurological deficit scores, a reduction in cerebral infarction volume, restoration of ischemic-side cerebral blood flow, and increased neuronal survival rates, further supporting EA’s comprehensive therapeutic potential in facilitating post-ischemic stroke neurological recovery.

IL-17 A, a key pro-inflammatory cytokine mainly secreted by Th17 cells, γδ T cells, and innate lymphoid cells (ILCs) [38], exhibits dual functions in stroke progression: exacerbating inflammatory responses acutely while promoting angiogenesis and neurogenesis during recovery phase to support neural repair [39–41]. Various cerebral cells, including neurons, astrocytes, microglia, and brain microvascular endothelial cells, express receptors for IL-17 A [42]. In the acute phase, IL-17 A binding to its receptor complex (IL-17RA/IL-17RC) triggers downstream pathways like NF-κB, MAPK, and C/EBP, upregulating pro-inflammatory factors and chemokines that amplify post-stroke neuroinflammation [43]. Furthermore, IL-17 A can compromise blood-brain barrier integrity either directly through its inflammatory mediators or synergistically with factors like IL-5 by downregulating tight junction proteins [44, 45]. Clinical evidence demonstrates markedly elevated peripheral blood IL-17 A levels in IS patients compared to healthy controls, which are positively correlated with disease severity and unfavorable outcomes [46, 47]. Experimental studies have demonstrated that IL-17 A aggravates CIRI through calpain-mediated TRPC5 hydrolysis [48]. Application of IL-17 A neutralising antibodies in MCAO/R models reduces infarct volume [49], and IL-17 A neutralization significantly suppresses post-tMCAO astrocyte activation in peri-infarct regions at days 3 and 7 [50]. Our findings indicate a substantial increase in IL-17 A protein expression, mRNA levels, and positive cell counts in ischemic infarct areas, all of which were significantly attenuated by EA intervention. Furthermore, EA treatment notably downregulated the expression of the IL-17 receptor IL-17RA and downstream signaling molecule NF-κB p65.

Astrocytes, the most abundant glial cells in the central nervous system, undergo rapid reactive changes after stroke, marked by increased GFAP expression, signifying their shift from quiescent to activated states—a process referred to as “astrocytic reactivity” [51, 52]. Reactive astrocytes exhibit dual roles in post-stroke inflammation, exerting both neuroprotective and neurotoxic effects. Functionally, they can be categorized into two subtypes: A1 and A1. A1 astrocytes amplify inflammation by secreting pro-inflammatory cytokines like IL-1β and TNF-α, while A1 astrocytes support tissue repair and neuroregeneration through the secretion of anti-inflammatory factors such as IL-10 and IL-1α [53–55]. The A1 marker, complement component C3, is highly upregulated in IS, thereby exacerbating neuroinflammation by activating the complement system to increase inflammatory mediators (IL-1β, TNF-α, IL-5), enhancing blood-brain barrier permeability, facilitating immune cell infiltration, and promoting neuronal apoptosis [56, 57]. The caspase-3/Bcl-2/Bax pathway, recognized as a key apoptotic mechanism in cerebral ischemia, is activated during the acute phase of stroke, thereby mediating widespread cell death [58, 59]. Notably, C3 expression is regulated by NF-κB and STAT3 signaling, which are themselves triggered by post-ischemic inflammatory factors [60, 61]. Thus, inhibiting A1 astrocyte activation and mitigating their neurotoxic effects may offer a promising therapeutic approach for post-stroke recovery.

Our study demonstrated marked A1-type astrocyte activation in the ischemic region of MCAO/R model mice, along with elevated C3 protein secretion, increased expression of inflammatory cytokines IL-1β and TNF-α, and activation of the Caspase-3/Bax/Bcl-2 apoptotic pathway, while TUNEL staining confirming enhanced neuronal apoptosis. EA intervention effectively downregulated C3 expression, suppressed inflammatory factor release, and reduced neuronal apoptosis. Astrocyte transition from a resting to a reactive state is primarily mediated by activation of the NF-κB pathway, with upregulation of the p65 subunit playing a pivotal role [62]. Upon binding of IL-17 A to IL-17RA on the astrocytic surface, the NF-κB signaling pathway is activated via phosphorylation, thereby driving phenotypic changes in astrocytes. Consequently, targeting IL-17 A and its downstream signaling has emerged as a potential therapeutic strategy. In this study, we demonstrated that EA significantly suppressed the expression of IL-17 A and IL-17RA in astrocytes within the ischemic region, reduced the activation level of NF-κB p65, decreased the number of neurotoxic astrocytes, and inhibited the secretion of pro-inflammatory cytokines, ultimately mitigating inflammatory injury and promoting neurological recovery.

However, several limitations of our study should be acknowledged. Although our findings provide evidence for the association between IL-17 A and astrocyte reactivity, the mechanistic investigation was confined to NF-κB phosphorylation, and the specific molecular mechanisms underlying NF-κB pathway activation warrant further exploration in future research. Moreover, a more comprehensive understanding of the regulatory role of IL-17 A may require the examination of additional or complementary signaling pathways, including potential synergistic interactions with other inflammatory mediators, which were not investigated in the present study.

Conclusion

In summary, our findings demonstrate that EA at the Baihui (GV10) and Fengfu (GV15) acupoints mitigates neuronal apoptosis, reduces cerebral infarction volume, and enhances neurological recovery following IS by attenuating neuroinflammation. This anti-inflammatory effect is likely mediated through the inhibition of IL-17 A release and downstream IL-17RA/NF-κB signaling, as well as the suppression of neurotoxic astrocyte activation.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Meng-Meng Zhao, Qing Song and Meng-Xing Li: Designed the study, Formal analysis, Wrote the manuscript, Conceptualization; Qian-Yun Xie and Wen-Qiang Sun: Helped with experiments, Collection of data. Yang Zhang, Wei Tang and Meng-Xing Li: Writing and Editing. All authors read and approved the final version of the article accepted for publication.

Funding

This work was supported by Young Scientists Fund of the National Natural Science Foundation of China (82305361), Anhui Provincial Natural Science Foundation(2308085MH298), Natural Science Research Project of Anhui Higher Education Institution (2023AH050845, 2022AH051263), National Natural Science Foundation of China (General Program, 82474617) and Anhui Provincial Association of Traditional Chinese Medicine Research Project Plan for Traditional Chinese Medicine(2024ZYYXH154).

Data Availability

All data or resources used in the paper are available by reasonable requirements to the correspondence authors.

Declarations

Competing Interests

The authors declare no competing interests.