Electroacupuncture therapy improves cognitive dysfunction after ischemic stroke in Sprague-Dawley rats by adjusting the lncRNA-MEG3/miR-4640-3p axis

Abstract

Background:

Ischemic stroke (IS) is a common disease that can cause cognitive dysfunction. Electroacupuncture (EA) is an effective way to alleviate cognitive dysfunction, but its molecular regulatory mechanism is still unclear. Long noncoding RNA-MEG3 (MEG3) is an important factor in the incidence and progression of IS. Herein, we explored the mechanism of EA in IS.

Methods:

A middle cerebral artery occlusion (MCAO) model was established in Sprague-Dawley rats to simulate IS in vivo, followed by electroacupuncture (EA) therapy. AAV-control and AAV-MEG3 were injected into the lateral ventricle of rats. All rats except for the sham group underwent MCAO. EA was performed at Shenting and Baihui points for 30 min, once a day for 14 days. The MEG3 and miR-4640-3p levels in brains were measured by qRT-PCR. Dual-luciferase reporter analysis validated the relationship between MEG3 and miR-4640-3p. The Morris water maze test and the neurological function test were carried out. The pathological morphology of the brain tissue was evaluated by H&E staining. Apoptotic cells in brains were examined utilizing TUNEL staining. The contents of Bax, Bcl-2, caspase-3, and CytC were assessed by western blot.

Results:

EA treatment reduced the content of MEG3 but enhanced miR-4640-3p levels in MCAO rats. MEG3 was a sponge for miR-4640-3p. EA treatment alleviated cognitive dysfunction in MCAO rats by inhibiting MEG3. EA treatment reduced MCAO-induced neural damage and apoptosis by inhibiting MEG3.

Conclusion:

EA improved cognitive dysfunction in IS rats by adjusting the MEG3/miR-4640-3p axis, suggesting that EA may be an effective potential therapeutic strategy for improvement of cognitive dysfunction in IS. This study provided a more reliable experimental basis for clinical EA treatment of IS patients.

Introduction

Stroke is a common sickness with high morbidity, disability, and mortality^[1]. Post-stroke cognitive impairment (PSCI) is the most common complication of stroke^[2]. Ischemic stroke (IS) is more likely to cause brain damage and aggravate cognitive dysfunction after ischemia reperfusion. Substantial evidence has established that neuroinflammation and apoptosis constitute central pathogenic mechanisms in the cascade of ischemic stroke-induced neuronal demise^[3,4]. In addition, the proportion of patients with residual cognitive impairment 6 months after the onset is as high as 37%^[5]. At present, the drugs targeting PSCI are often accompanied by many side effects, making it difficult for patients to persist in taking them for a long time^[6,7]. Therefore, it is crucial to find effective strategies that relieve PSCI with fewer side effects. Currently, transcranial direct current stimulation (tDCS) is considered to be a non-invasive tool to improve PSCI^[8]. However, exploring more effective and safer methods to enrich the treatment of PSCI is necessary and urgent.

HIGHLIGHTS

• Electroacupuncture (EA) treatment reduced the content of MEG3 in middle cerebral artery occlusion (MCAO) rats.

• MEG3 was a sponge for miR-4640-3p.

• EA treatment alleviated cognitive dysfunction in MCAO rats by inhibiting MEG3.

• EA treatment reduced MCAO-induced neuronal damage by inhibiting MEG3.

Acupuncture therapy is a characteristic method based on traditional Chinese medicine (TCM) theory^[9]. Acupuncture on the head acupoints can mediate synaptic plasticity at multiple levels and in multiple ways to improve cognitive dysfunction^[10]. Randomized controlled trials provide evidence that acupuncture has positive effects on PSCI^[11,12]. Baihui (GV20) and Shenting (GV24) are two important acupoints on the Du pulse; acupuncture of these two acupoints can play a role in awakening the brain, body, and spirit^[13]. Xu et al^[14] investigate the effectiveness of theta and gamma electroacupuncture (EA) at Baihui and Shenting acupoints for PSCI. Su et al^[15,16] find that needle retaining after EA at Shenting and Baihui combined with cognitive training has a better therapeutic effect on PSCI than EA alone. A recent clinical trail shows that EA at Baihui and Shenting plus cognitive rehabilitation is a safe and effective treatment for PSCI^[17]. In middle cerebral artery occlusion (MCAO) rats, EA at Baihui or Shenting improves cognitive impairment and reduces neurological deficit scores and apoptosis but enhances the Bcl-2/Bax ratio^[18]. EA at “Baihui, Yintang, and Shuigou” acupoints alleviates PSCI in mice associated with synaptic plasticity of the hippocampus^[19]. Besides, EA at Fengchi, Fengfu, and Dazhui acupoints alleviates PSCI in rats^[20,21]. Nevertheless, the precisely regulatory mechanism of EA under PSCI is still largely uncertain. This will be the focus of this study.

Previously, EA downregulates the m6A methylation of long noncoding RNA (lncRNA) H19 to alleviate cerebral ischemia-reperfusion injury^[22]. lncRNAs are vital to many diseases, like IS^[23]. The abnormal expressions of MEG3, SNHG12, MALAT-1, H19, and other lncRNAs can regulate IS development^[24,25]. MEG3 is a crucial governor in neuronal cell death and angiogenesis in cerebral IS^[26]. MEG3 can induce cerebral ischemia-reperfusion injury by modulating oxidative stress and mitochondrial dysfunction via the hnRNPA1/Sirt2 axis^[27]. MEG3 promotes IS pathogenesis by suppressing miR-424-5p, which targets Sema3A and activates the MAPK pathway, and finally leads to neuronal apoptosis^[28]. Especially, MEG3 regulates the miR-378/GRB2 axis to accelerate neuronal autophagy and neurological dysfunction in IS^[29]. miR-4640-3p is first observed in the blood of dengue patients^[30]. By using bioinformatics analysis, we found the existence of possible binding sites for MEG3 and miR-4640-3p. However, the functions of miR-4640-3p and its regulatory mechanism in IS have not been discussed. Hence, we built an MCAO rat model to simulate the in vivo environment of IS and gave EA intervention. This study explored the relevant mechanism and further clarified the target of EA therapy (Baihui and Shenting acupoints) in the treatment of PSCI. This study was conducted in accordance with the TITAN Guidelines 2025^[31].

Materials and methods

Animals

Eighteen healthy SD rats (male; 8 weeks; weights 220 g) were obtained from Sibeifu Biotechnology Co., LTD. (Beijing, China). All rats were adaptively fed in a specific pathogen-free (SPF) laboratory for one week (22–24 ℃, 50%–60% relative humidity, 12 h light/dark cycle) with free access to water and food. All rat tests were adopted with the Guide for the Care and Use of Laboratory Animals^[32]. The animal research complied with the Animal Ethical and Welfare Committee of People’s Hospital Affiliated to the Fujian University of Traditional Chinese (Approval No. DB20214325). This work has been reported in line with the ARRIVE criteria^[33].

Construction of adeno-associated viral (AAV) vectors

The AAV vector expressing the rat MEG3 (AAV-MEG3) and control (AAV-control) were generated by GeneChem (Shanghai, China). Viral stocks of AAV-MEG3 serotype 1 were produced utilizing the triple-transfection process^[34].

Experiments

Eighteen rats were randomized into six groups: sham operation (sham) group, MCAO group, MCAO + EA (MCAO + EA) group, MCAO + AAV-control group, MCAO + AAV-MEG3 group, and MCAO + EA + AAV-MEG3 (MCAO + EA + AAV-MEG3) group, with three rats/group. The sample size was determined based on pilot data and effect magnitude. Preliminary experiments (n = 2 per group) revealed a large and consistent effect size (Cohen’s d > 2, power > 80% increase cognitive dysfunction) under the experimental conditions, suggesting that even a small sample could detect biologically significant differences. In addition, three animals per group represent the minimum number required to obtain statistically reliable means and standard deviations for the parameters under investigation, based on the experimental design and ethical considerations reducing animal use while maintaining scientific validity. 14 days prior to MCAO surgery, 8 × 10¹⁰ viral genome (v.g.) AAV was injected into the lateral ventricle of rats in the MCAO + AAV-control group, MCAO + AAV-MEG3 group, and MCAO + EA + AAV-MEG3 group, utilizing a 10 μL syringe (Hamilton Company, Reno, Nevada, USA), adapting the method described by Harvey BK et al^[34]. After AAV injection, the needle stayed for 5 min. 14 days later, all rats except for the sham group were subjected to MCAO management. Rats in the MCAO + EA group and MCAO + EA + AAV-MEG3 group underwent EA intervention at 24 h post MCAO induction.

MCAO surgery

An MCAO procedure was performed as described by Harvey BK^[34]. In brief, all rats were intraperitoneally anaesthetized using 30 mg/kg 2% pentobarbital sodium (Sigma-Aldrich, St. Louis, MO, USA). The common carotid artery (CCA), internal carotid artery (ICA), and external carotid artery (ECA) on the left side of rats were cautiously exposed. Subsequently, the CCA and ECA were ligated, followed by a tiny opening, which was created in the ICA at a distance of 2 mm from the common cervical bifurcation. Following this, a nylon embolus (250-280; Maiyue biotechnology; Ningbo, China) was built in the ICA utilizing a 22 mm catheter (Sigma) to block the left middle cerebral artery (MCA) for 1 h. The animal temperature was kept at 37 ℃ during and after surgery. After awakening, the rats were returned to their home cage. Meanwhile, rats in the sham group had their arteries exposed utilizing the same processes, but without ligation.

EA treatment

24 h after MCAO surgery, the MCAO + EA group and MCAO + EA + AAV-MEG3 group rats were subjected to Shenting (in the front center line of the head, at the midpoint of the line connecting the tips of the ears, and stabbed backward obliquely for 2 mm) and Baihui (in the front center line of the head, at the front of the line connecting the frontal parietal sutures, and stabbed upward obliquely for 2 mm) acupoints for EA treatment using a multiple electronic acupunctoscope (SDZ-V; Suzhou Medical Supplies Factory Co. LTD; Suzhou; China). The rats were tied to a homemade rat board to restrict the rat’s movement. The parameters of EA intervention were set as described in the previous study by Chen L^[35]. Density wave was used, frequency 1–20 Hz, intensity 2 mA, 30 min each time, once a day. The intervention lasted 14 days. After 14 days of the intervention, the behavioral and neurological function tests were carried out in line with the modified Zea-Longa score determinations (as described in Table 1). Finally, brain tissues were taken from rats for follow-up studies.

Table 1.

Zea-Longa score

Score	Method
0	No obvious neurological deficit
1	Unable to fully extend the contralateral forepaw
2	Body of the rat circled to one side
3	Circling around and falling to one side
4	Unable to walk spontaneously and loss of consciousness

Morris water maze (MWM) test

An MWM test was carried out in an MWM experimental system (XR-XM101, Xinruan Information Technology Co., LTD, Shanghai, China). A water pool, with a height of 50 cm and a diameter of 160 cm, was filled with water, and a platform (12 cm diameter) was placed 1 cm below the water. A training and testing protocol was followed, consisting of 3 days of training followed by 1 day of testing. During the 3-day training course, eight tests were conducted each day, with four random starting points for entry into the water maze. In each test, if the platform was not found within 1 min, the rat was manually guided to the platform and allowed to stay there for 20 s. If the platform was reached within 1 minute, the subject was allowed to stay on the platform for 20 s as well. The escape latency and the time taken to find the platform were recorded. If a rat failed to complete the task within 1 min, the escape latency was marked as 60 s. The daily average escape latency and the success rate of reaching the platform were calculated for each animal based on eight tests conducted each day. After leaving the water pool, each rat was immediately dried with a towel and exposed to a heat lamp. On the 4th day, a test was conducted. We recorded the ratio of swim path distance in the target quadrant to all quadrants, the time it took to find the hidden target platform, and the distance of the swim path to the hidden target platform.

qRT-PCR

The methods were in accordance with the report of Chen et al^[36]. Total RNAs of brain tissues were separated utilizing a TRIzol reagent (TaKaRa, Dalian, China) and then reverse-transcribed to cDNA. Following this, the qRT-PCR was adopted by employing TB Green® Premix Ex Taq™ (Sigma) in accordance with the procedure. GAPDH and U6 were employed as controls of lncRNA and miRNA. Relative gene contents were computed by the 2^-△△Ct method. The primers are displayed in Table 2.

Table 2.

Primers for qRT-PCR

Name		Primers for PCR (5ʹ-3ʹ)
MEG3	Forward	GGGCTTCTGGAATGAGCATG
	Reverse	TCTATGCCAGATCCTGCCTG
miR-4640-3p	Forward	GCCGAGCACCCCCTGTTTC
	Reverse	CTCAACTGGTGTCGTGGA
U6	Forward	CTCGCTTCGGCAGCACATATACT
	Reverse	ACGCTTCACGAATTTGCGTGTC
GAPDH	Forward	ACAGCAACAGGGTGGTGGAC
	Reverse	TTTGAGGGTGCAGCGAACTT

Luciferase reporter assay

The HEK293T cell line was obtained from Chuan Qiu Biotechnology (Shanghai, China) and cultivated in DMEM (Sigma) accompanied with 10% FBS (Sigma) at 37 ℃ with 5% CO₂. Reporter vectors comprising the wild type (WT) or mutant (MUT) connective site of the MEG3 sequence were constructed, named WT-MEG3 and MUT-MEG3. WT-MEG3 or MUT-MEG3 was co-transfected with miR-4640-3p mimics (GenePharma, Shanghai, China) or miR-NC mimics (GenePharma) into HEK293T. Following this, the luciferase activity was appraised.

Hematoxylin-eosin (HE) staining

The dehydrated brain tissue samples were embedded in paraffin wax (Sigma) and cut into sections 4-μm thickness. The slices were immersed in hematoxylin (Sigma) for 5 min and then washed and immersed in eosin (Sigma) for 1 min. After that, the slices were immersed in 95% ethanol (Sigma) and anhydrous ethanol (Sigma) for 3 s and 2 s. Following this, the slices were immersed in anhydrous ethanol (Sigma) for 10 min. Then, the slices were treated with xylene (Sigma) for 10 min and sealed with neutral gum (Sigma). The staining results were observed and photographed using a microscope (E200MV; Nikon, Tokyo, Japan).

TUNEL staining

The brain tissues were dripped with protease K working solution (Sigma) for 30 min. Then, the samples were exposed to 3% hydrogen peroxide (BOSTER, Wuhan, China) for 20 min. Subsequently, the samples were processed using the TUNEL apoptosis detection kit (C1098; Beyotime, Shanghai, China) as instructed. Finaly, the samples were immersed in hematoxylin (Sigma) for 5 min. After PBS cleaning, the samples were immersed in anhydrous ethanol (Sigma) for 7 min. Then, the slices were treated with xylene (Sigma) for 7 min and sealed with neutral gum (Sigma). The slices were photographed using a microscope (E200MV; Nikon).

Western blot

The methods of western blot were consistent with that described by Hou W^[37]. In brief, after proteins were extracted from brain tissues, SDS-PAGE was performed to separate the proteins. Then, the proteins were electrotransferred onto PVDF membranes. The membranes were blocked by 5% nonfat milk, followed by primary antibodies incubation at 4 ℃ overnight. The primary antibodies include anti-Bax (ab32503; 1:1,000; Abcam, Cambridge, MA, USA), anti-Bcl-2 (ab194583; 1:1000; Abcam), anti-caspase-3 (ab184787; 1:1,000; Abcam), anti-cytochrome C (CytC) (ab133504; 1:1,000; Abcam), and anti-GAPDH (ab8245; 1:2500; Abcam). Subsequently, goat anti-rabbit IgG (ab205718; 1:2500; Abcam) was added to the membranes and incubated at 37 ℃ for 1 h. The protein signals were assessed utilizing an ECL kit (Sigma) and then evaluated using ImageJ software (V1.8.0, NIH, USA).

Statistical analysis

All independent trials were adopted in triplicate and expressed as mean ± standard deviation (SD). Statistical analysis was conducted by using GraphPad Prism 7 (GraphPad Inc., La Jolla, CA, USA). The normality of the data was assessed using the Shapiro-Wilk test, due to the sample size being less than 50, while the homogeneity of variance was tested using the Levene statistics test. For those with normal distribution and homogeneity of variance, one-way ANOVA was used to evaluate the differences among the multiple groups, and Student’s t-test was utilized to compare the statistical differences between two groups. Post hoc analysis was conducted using the Bonferroni test on normally distributed data. In cases where the data were not normally distributed, transformations were applied for confirmation, with the Kruskal-Wallis test used if normality could not be achieved. P < 0.05 was statistically significant.

Results

EA treatment reduced the content of MEG3 in MCAO rats

MEG3 is a critical promotor in cerebral IS^[26]. Thus, we investigated the influence of EA treatment on MEG3 expression in MCAO rats. Figure 1A,B indicates that EA intervention decreased MEG3 levels while increasing miR-4640-3p levels in MCAO rats (P < 0.05). EA intervention reduced MEG3 levels and increased miR-4640-3p levels in MCAO rats. In addition, MEG3 overexpression enhanced MEG3 content, but reduced miR-4640-3p levels, which were lessened by EA therapy in MCAO rats (P < 0.05, Fig. 1A, B). EA intervention reduced MEG3 levels and increased miR-4640-3p levels in MCAO rats. The observed suppression of MEG3 and upregulation of miR-4640-3p by EA uncovered the mechanisms of EA on PSCI improvement and suggested a novel therapeutic strategy for PSCI.

Figure 1.

EA treatment reduced the content of MEG3 in MCAO rats. (A, B) The MEG3 and miR-4640-3p levels were estimated by qRT-PCR. The data were expressed as mean ± SD from three independently repeated experiments. *P < 0.05.

MEG3 was a sponge for miR-4640-3p

In this part, we examined the relationship between MEG3 and miR-4640-3p. Figure 2A displays that MEG3 comprised one miR-4640-3p binding site. Following this, the luciferase assay confirmed that miR-4640-3p suppressed the luciferase activity of the WT-MEG3 group, not MUT-MEG3 group, in HEK293T cells (P < 0.0001, Fig. 2B), suggesting that MEG3 was a sponge for miR-4640-3p.

Figure 2.

MEG3 was a sponge for miR-4640-3p. (A) Target sites between MEG3 and miR-4640-3p. (B) Dual-luciferase reporter analysis validated that MEG3 targeted miR-4640-3p. The data were expressed as mean ± SD from three independently repeated experiments. ****P < 0.0001.

EA treatment alleviated cognitive dysfunction in MCAO rats by inhibiting MEG3

MEG3 can promote oxidative stress, mitochondrial dysfunction, neuronal apoptosis and autophagy, and neurological functional impairment in IS^[27–29]. Besides, decreased MEG3 alleviates cognitive dysfunction caused by ISO by targeting miR-7-5p and plays a neuroprotective effect^[38]. Therefore, by using AAV infection technology to overexpress MEG3 expression, we explored whether the influence of EA on cognitive dysfunction was mediated by the MEG3/miR-4640-3p axis. The effect of EA intervention on the neurological function of MCAO rats was appraised in accordance with the modified Zea-Longa score determinations. MCAO treatment caused certain damage to the neurological function of rats. EA treatment and overexpression of MEG3 did not significantly improve the neurological function of MCAO rats (P > 0.05, Fig. 3A). Besides, we carried out the MWM test to assess the cognitive function of MCAO rats. There was no obvious change in the distance ratio in each group of rats (P > 0.05, Fig. 3B). The time to find the target (P < 0.01, Fig. 3C) and the distance of the swim path to the target (P < 0.01, Fig. 3D) were enhanced in the MCAO group versus the sham group, which were diminished by EA intervention (P < 0.05). Additionally, the time and distance to find the target were heightened by MEG3 overexpression (P < 0.001), while these effects were lessened (P < 0.001) by EA intervention in MCAO rats.

Figure 3.

EA treatment alleviated cognitive dysfunction in MCAO rats by inhibiting MEG3. (A) Zea-Longa score. (B–D) The ratio of swim path distance in the target quadrant to all quadrants (B), the time it takes to find the hidden target platform (C), and the distance of the swim path to the hidden target platform (D) in the MWM test. The data were expressed as mean ± SD from three independently repeated experiments. *P < 0.05, **P < 0.01, ***P < 0.001.

EA treatment reduced MCAO-induced neuronal damage by inhibiting MEG3

In this part, we explore whether EA treatment reduced MCAO-tempted neural injury by inhibiting MEG3. H&E staining results exposed that neurons of the MCAO group were seriously injured, as evidenced by obvious inflammatory infiltration, cell shrinkage, and chaotic cell arrangement. Nevertheless, EA treatment alleviated MCAO-tempted neural injury, while these effects were lessened by AAV-MEG3 co-treatment (Fig. 4A). Meanwhile, TUNEL staining discovered that EA treatment dwindled MCAO-tempted cell apoptosis, but the effect was diminished by AAV-MEG3 co-treatment (Fig. 4B). In addition, the contents of Bax (Fig. 5A, B), caspase-3, and CytC (Fig. 5A, C) were heightened, but the Bcl-2 level (Fig. 5A B) was abridged (P < 0.05) in the brain tissues of the MCAO group versus the sham group. However, EA intervention obviously diminished the levels of Bax, caspase-3, and CytC but increased the Bcl-2 level, while these effects were lessened by AAV-MEG3 co-treatment in MCAO rats (P < 0.05, Fig. 5), demonstrating that EA treatment reduced MCAO-induced neural damage and apoptosis by inhibiting MEG3.

Figure 4.

EA treatment abridged MCAO-induced neuronal damage. (A) The pathological morphology of brain tissue was evaluated by H&E staining. (B) Apoptotic cells in brains were examined utilizing TUNEL staining.

Figure 5.

EA treatment reduced MCAO-induced neural apoptosis. (A–C) The contents of Bax, Bcl-2, caspase-3, and CytC were assessed by western blot. The data were expressed as mean ± SD from three independently repeated experiments. *P < 0.05.

Discussion

In this research, we confirmed that EA treatment reduced the content of MEG3 but enhanced the miR-4640-3p level in MCAO rats. Then, we found that MEG3 was a sponge for miR-4640-3p. Evidence indicates that α-pinene attenuates neurobehavioral deficits in rat models of focal cerebral ischemic stroke^[39]. Similarly, our study demonstrates that electroacupuncture (EA) therapy ameliorates cognitive dysfunction in MCAO rats through suppression of MEG3 expression. Furthermore, studies demonstrate that γ-terpinene (γ-TER) confers neuroprotection against ischemic injury in acute cerebral ischemic rats by reducing infarction volume, cerebral edema, oxidative stress, and neuroinflammation. This mediates the regulation of key apoptosis-associated genes and proteins across distinct brain regions^[40]. In a congruent mechanistic paradigm, our investigation reveals that electroacupuncture (EA) therapy mitigates MCAO-induced neurotoxicity and apoptotic processes through suppression of MEG3 expression. In conclusion, EA therapy improved cognitive dysfunction, neural damage, and apoptosis in MCAO rats by adjusting the MEG3/mir-4640-3p axis. This study demonstrated that electroacupuncture ameliorates cognitive function in IS rats by modulating the MEG3/miR-4640-3p axis, providing direct experimental evidence for the clinical application of electroacupuncture in treating cognitive impairment in IS patients. With its favorable side-effect profile as a traditional medical therapy, electroacupuncture may serve as a valuable complementary approach to current IS rehabilitation strategies, particularly in patients intolerant to pharmacological treatments.

EA therapy has been usually used in clinical management of PSCI^[11]. PSCI is a cognitive disorder belonging to the categories of “dementia,” “dumb disease,” “literary madness,” and “brain pulp elimination” in Chinese medicine, and its disease location belongs to the brain. The clinical manifestations of inattention and slow reaction indicate that pathogenesis is brain dystrophy and deactivation of the spirit machine, and the treatment is to awake the brain and fill the marrow^[41]. The Baihui acupoint is at the top of the human body for the Du vein, foot sun meridian, hand and feet Shaoyang meridian, foot Jueyin meridian, and other meridians following the intersection of the head, which stimulates that the point has the effect of awakening the brain, filling the marrow, and nourishing the brain^[42]. Shenting, for the Du vein, Yangming meridian, foot sun meridian intersection point, and stimulating Shenting point, has a calming effect. Baihui, Shenting, and other acupoints have the functions of replenishing Qi, propagating Yang, dispersing fine, nourishing the brain, and promoting the spirit mechanism, in line with the PSCI of TCM^[42]. Some scholars found that acupuncture at Baihui and Shenting plays a positive role in the recovery of PSCI^[43]. A previous study provides evidence that EA inhibites microglia polarization via regulation of the lnc826-mediated hippo pathway in IS^[44]. In addition, EA alleviates cerebral ischemia-reperfusion injury via m6A methylation of lncRNA H19^[22]. These findings demonstrate that EA has a protective role in IS by regulating lncRNAs. However, whether EA can affect PSCI by regulating lncRNA remains uncovered. In this research, we confirmed that EA intervention reduced the content of MEG3 but enhanced the miR-4640-3p level in MCAO rats. These consequences indicated that MEG3 and miR-4640-3p may play pivotal roles in IS and EA treatment and might modulate the MEG3/miR-4640-3p axis for PSCI therapy. This “electroacupuncture-ncRNA axis” mechanism not only aligns with the core therapeutic principle of “restoring consciousness and unblocking orifices” in traditional Chinese medicine (TCM) – wherein acupoints Shenting and Baihui correspond to prefrontal and limbic cognitive regulatory centers – but also elucidates the scientific basis of acupuncture in modulating neuroplasticity through contemporary molecular biological approaches.

Some studies have found that MEG3 can interfere with P53 to affect the necrotic area and neuronal apoptosis after stroke in the mouse cerebral ischemia model^[45]. In IS patients, MEG3 promotes apoptosis of vascular cells and may serve as an independent prognostic factor for poor outcomes and death^[46]. Besides, MEG3 can exert sponge-like effects and bind to miRNA and play a crucial role in the pathogenesis of IS. For example, MEG3 can adsorb miR-21 through sponge-like effects and inhibit miR-21 levels to promote PDCD4 and apoptosis protein expressions, thereby promoting the cerebral infarction size and neuronal death^[47]. MEG3 also exerted a sponge-like effect to regulate the miR-181b/lipoxygenase 12/15 axis, leading to hypoxia-induced apoptosis of HT22 cells^[48]. In addition, MEG3 positively regulates miR-493-5p and subsequently downregulates MIF expression, thereby suppressing proliferation of neural stem cells after IS^[49]. The MEG3/miR-485/AIM2 axis contributes to pyroptosis via activating caspase-1 signaling during cerebral ischemia/reperfusion, suggesting that this axis is a potent therapeutic target in IS^[50]. Further identification of novel interacting targets miRNAs of MEG3 will help better understand the machinery action of MEG3 and develop corresponding therapeutic strategies. In this research, we first found that MEG3 was a sponge for miR-4640-3p. MEG3 is also associated with cognitive dysfunction. Reportedly, MEG3 alleviates cognitive impairment in Alzheimer’s disease^[51], diabetic-related cognitive impairment^[52], and postoperative cognitive impairment^[53,54]. On the contrary, MEG3 downregulation alleviates an isoflurane-caused increase in the neurological severity score and cognitive dysfunction by targeting miR-7-5p^[38]. Nevertheless, the roles of MEG3 in PSCI are still unclear. In this study, EA improved cognitive dysfunction, but MEG3 overexpression heightened cognitive dysfunction in MCAO rats.

Meanwhile, EA treatment alleviated cognitive dysfunction in MCAO rats by inhibiting MEG3. Moreover, EA inhibited neural damage and apoptosis, but MEG3 overexpression led to abundant neural damage and apoptosis of brain tissues in MCAO rats. EA treatment reduced MCAO-induced neural damage and apoptosis by inhibiting MEG3. Among them, the mitigation effects of EA treatment on MCAO-induced cognitive dysfunction, neural damage, and apoptosis were consistent with the study of Zhang et al^[55] and Liu et al^[18]. In terms of mechanism, this study clarified the interaction target miRNA of MEG3, clarified the regulatory mechanism and action target of MEG3, and provided a more reliable experimental basis for clinical EA treatment of stroke patients.

However, there are still some limitations in this research. Due to the limitation of funding and experimental resources, the sample size selected in this study is small, which may affect the representativeness of the research results. We inspected the molecular mechanism of EA therapy’s modulation of PSCI only in rat models. We also do not perform the rescue experiments by overexpression of miR-4640-3p to verify that EA really improved PSCI by modulating the MEG3/miR-4640-3p axis. We only performed the MWM experiment to evaluate the influence of EA on cognitive dysfunction in MCAO rats. The follow-up study will expand the sample size and include cellular experiments and rescue experiments to substantiate the findings presented herein. In addition, this study revealed the critical regulatory role of MEG3 in IS-associated cognitive impairment. The targeting relationship between MEG3 and miR-4640-3p provides a potential molecular target for novel therapeutic development. Future studies could focus on developing interventions targeting this axis (e.g. MEG3 inhibitors or miR-4640-3p agonists), thereby offering novel therapeutic perspectives for precision treatment of post-IS cognitive impairment. We will further validate the conclusions of this paper in the clinic.

Conclusion

In short, we confirmed that EA therapy improved cognitive dysfunction, neural damage, and apoptosis in IS rats by adjusting the MEG3/mir-4640-3p axis. The conclusion of this research delivers a reliable experimental basis for the further promotion and application of clinical EA therapy in the treatment of cognitive dysfunction in IS.

Ethical approval

The animal research complied with the Animal Ethical and Welfare Committee of People's Hospital affiliated to the Fujian University of Traditional Chinese Medicine (Approval No. DB20214325).

Conflicts of interest disclosure

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Sources of funding

This study was supported by the Fujian Provincial Natural Science Foundation Youth Innovation Project (Project No. 2022J05201) and the Educational Research Project for Young and Middle-aged Teachers in Fujian Province (Project No. JAT191300).

Author contributions

Conception and design: Y.Z. and Y. Z.; method: S.G.; data collection: L.L.; manuscript writing: Y.Z.; manuscript revision: Y.Z.; research supervision: Y.Z. All authors contributed to the article and approved the submitted version.

Research registration unique identifying number (UIN)

Not applicable.

Guarantor

Yongbing Zheng.

Provenance and peer review

Not commissioned, externally peer-reviewed.

Data availability statement

The simulation experiment data used to support the findings of this study are available from the corresponding author upon request.