Male Sprague-Dawley rats (8 weeks), ranging in weight from 250 to 280 g, were used for the study. All rats were kept in cages of 4 under SPF conditions at room temperature (23 $\pm$ 2 °C) under a 12:12 h light/dark cycle. The rats had unrestricted access to standard laboratory chow and tap water.

The experimental timeline and corresponding procedures are shown in Fig. 1A and Fig. 5A. To evaluate the effects of acupuncture on rats with PSSH, four groups were established: Sham-operated group (Sham), MCAO induced PSSH model group (Model), acupuncture at acupoint group (Acup), and acupuncture at control point (non-acupoint) group (CP). To investigate the role of NMDAR, the NMDAR agonist NMDA and antagonist D-(-)-2-Amino-5-phosphonopentanoic acid (D-AP5) were administered. Five groups were established: Model, Model + vehicle, Acup + vehicle, Model + D-AP5, and Acup + NMDA. All rats were randomly assigned to these groups. Excluding the 10 rats that underwent sham surgery, 112 rats were subjected to MCAO procedure for PSSH-model induction, of which 80 were successfully modeled and subsequently used in the experiments (Supplementary Table 1). The sample size was chosen according to previous research and our preliminary data, ensuring adequate statistical power to identify meaningful differences among groups.

Fig. 1.

Acupuncture alleviates spasticity and enhances motor function in MCAO rats with PSSH. (A) Timeline of experimental procedures. (B) Representative laser speckle images (LSI) for monitoring cortical blood flow (CBF) after MCAO surgery. (C) Acupuncture site location. (D) Neurological deficits assessed by the mNSS. (E) Muscle tone assessed by the MAS. (F) Stride length of rats at 3 days and 10 days. (G) Limb coordination index of rats at 3 days and 10 days. (H) Footprints of rats in each group, with blue representing the forelimbs and red representing the hind limbs. N = 10 per group. Data are presented as mean $\pm$ SD. ^{◆⁢◆⁢◆}p 0.001 compared to the Model group; ^#⁢#p 0.01, ^#⁢#⁢#p 0.001 compared to the Acup group; ^*⁣**p 0.001. MAS, modified ashworth scale; MCAO, middle cerebral artery occlusion; mNSS, modified neurological severity score; GB34, Yanglingquan; d, day; PSSH, post-stroke spastic hemiplegia. Sham: Sham-operated group; Model: MCAO induced PSSH model group; Acup: acupuncture at acupoint group; CP: acupuncture at control point (non-acupoint) group. Fig. 1B and Fig. 1C Created with BioRender.com.

The permanent MCAO procedure was performed to establish a PSSH model, following the previously described method with slight modifications [19, 20]. After anesthetization with isoflurane (3% in oxygen; flow rate, 1 L/min; cat. no. R510-22-10; RWD Life Science, Shenzhen, Guangdong, China), a midline skin incision was performed on the ventral neck to expose the right common carotid artery. A silicone-coated nylon monofilament with a diameter of 0.36 mm (Beijing Cinontech Co., Ltd.; Beijing, China) was gently advanced from the right external carotid artery to the internal carotid artery and further to the origin of the middle cerebral artery (MCA). The filament was left in place for the remainder of the experiment. A temperature-controlled heated pad was applied during the ischemic period to maintain the body temperature constantly at 37 $\pm$ 0.5 °C. In sham-operated rats, the identical surgical procedure was carried out but without occlusion of the MCA.

MCAO rats were included in the study if they exhibited a regional cerebral blood flow (CBF) reduction of greater than 70% of baseline levels, and a Zea-Longa score above 1. All behavioral evaluations were performed by an independent investigator who was blind to the interventions of the study.

CBF changes were assessed using a laser speckle imaging system (PeriCam PSI NR; Perimed AB; Stockholm, Sweden). Measurements were taken 5 min before and 5 min after MCAO surgery to determine relative perfusion in the MCA territory. After induction of anesthesia, a midline scalp incision was made to expose the skull. Circular regions of interest (ROIs) with a diameter of 2 mm were placed 3 mm lateral and 2 mm posterior to the bregma on the right cortical surface (Fig. 1B). CBF values were calculated as percentages relative to pre-occlusion baseline levels according to acquired microcirculation images.

Acupuncture treatment was administered once daily to rats in the Acup and CP groups from postoperative day 4 to day 10. Sterile, disposable acupuncture needles (0.25 $\times$ 13 mm; Zhongyantaihe; Tianjin, China) were perpendicularly inserted into the “Yanglingquan” (GB34) acupoint, located in the depression distal to the head of the fibula on the left hind limb (Fig. 1C). The needle was inserted 4–5 mm deep and manipulated with multi-directional lifting and thrusting combined with passive movement of affected hind limb, then retained in place for 30 min. In the CP group, needles were superficially inserted into a non-acupoint at the midpoint of the tail, a site not corresponding to any recognized acupuncture-point location.

NMDA (40 mM, cat. no. 0114; Tocris; Minneapolis, MN, USA) and D-AP5 (15 mM, cat. no. 0106; Tocris) were used for NMDAR modulation. Both compounds were freshly dissolved in sterile distilled water and administered via intrathecal injection at a volume of 30 µL once daily from postoperative day 4 to day 10.

A Bio-Signal Acquisition and Processing System (BL-420N, Chengdu Taimeng Technology Co., Ltd.; Chengdu, Sichuan, China) was used to perform Hoffmann-reflex (H-reflex) recording and analysis. After anesthetization with 3% isoflurane in 97% oxygen, the left sciatic nerve was exposed by blunt dissection and secured on a bipolar hook electrode. To prevent nerve desiccation during the procedure, oleovitamin was applied to the exposed nerve. Electromyographic (EMG) signals were collected using stainless-steel needle electrodes inserted into the interosseous muscles of the left hind paw, and a ground electrode was placed subcutaneously in the tail.

The H-reflex was elicited by applying single biphasic pulses (100 µs width) to the left sciatic nerve through an isolated pulse stimulator. Both M-waves and H-waves were recorded at varying stimulus intensities to determine the thresholds of motor (MT) and H-reflex. To assess frequency-dependent depression (FDD), stimuli of identical intensity (eliciting Hmax) were delivered at frequencies 0.3 Hz, 5 Hz, and 10 Hz [25]. Subsequently, the variations in the amplitude of the H-reflex at 5 Hz and 10 Hz were expressed as a percentage of the response at 0.3 Hz. A repeated 0.3 Hz stimulation was applied at the end of the recording to confirm that the M-wave remained within 95% of baseline; data were excluded if this criterion was not met. All H-reflex and M-wave measurements, including FDD, were obtained 15 min after the final treatment.

The rats were euthanized by cervical dislocation under deep anesthesia induced by intraperitoneal injection of pentobarbital sodium (100 mg/kg body weight, Sigma-Aldrich, St. Louis, MO, USA; Cat. No.: P3761) after the final interventions, and the whole brains were rapidly removed and cooled at –20 °C for approximately 30 min to facilitate sectioning. Each brain was sliced into 2-mm coronal sections and incubated in 0.3% TTC solution (cat. no. SL7140; Coolaber; Beijing, China) at 37 °C in the dark for 20–30 min. The stained slices were then stored in 4% paraformaldehyde (cat. no. SL1830; Coolaber) overnight. Digital images of the brain sections were captured, and infarct areas were analyzed and detected as previously described utilizing ImageJ software (Fiji; National Institutes of Health; Bethesda, MD, USA) [18].

The embedded lumbar-spinal-cord-enlargement tissue was sectioned into 10 µm slices using a motorized cryostat. Donkey serum was used to block first, and then the following primary antibodies were used: anti-c-Fos (1:500, cat. no. 2250T; Cell Signaling Technology; Danvers, MA, USA), anti-NMDAR1 (1:500, cat. no. ab109182; Abcam; Cambridge, Cambs, UK), and anti-KCC2 (1:500, cat. no. ab259969; Abcam) were incubated overnight at 4 °C. The next day, sections were incubated for 2 h at 37 °C in the dark with the fluorescence antibodies (1:500, cat. no. RGAR002, cat. no. RGAR004; Proteintech; Rosemont, IL, USA). After washing, the sections were mounted with anti-fade mounting medium containing 4^′,6-diamidino-2-phenylindole (DAPI) (cat. no. S2110; Solarbio; Beijing, China) and then photographed.

Low-magnification images (4$\times$ objective) were captured using an upright fluorescence microscope (Olympus BX43; Tokyo, Japan), and high-resolution images at 10$\times$, 20$\times$, and 40$\times$ magnification were acquired with a laser scanning confocal microscope (Olympus FV3000). The number of c-Fos-positive cells in the ventral horn was manually counted, and the mean fluorescence intensity of NMDAR1 in the same region was measured. The expression of KCC2 in the ventral horn was quantified using the previously described method [26].

Total RNA was isolated from the lumbar spinal ventral horn of each rat according to the manufacturer’s instructions (cat. no. DP451; Tiangen; Beijing, China). RT-qPCR was performed by mixing cDNA templates with gene-specific primers and PowerUp SYBR Green Master Mix (cat. no. A25742; Thermo Fisher Scientific; Waltham, MA, USA), followed by amplification using a StepOne Plus Real-Time PCR Detection System (CFX Connect; Applied Biosystems; Foster City, CA, USA). The thermal cycling conditions were as follows: initial denaturation at 95 °C for 3 min, followed by 39 cycles of 95 °C for 10 s and 59 °C for 60 s. $\beta$-Actin served as the internal control, and relative gene expression was quantified using the 2^{-Δ⁢Δ⁢Ct} method. Primer sequences are provided in Supplementary Table 2.

Proteins were extracted from the lumbar spinal ventral horn and concentrations were determined with a BCA kit (cat. no. AR0146; Boster; Wuhan, Hubei, China). After SDS-PAGE (8 and 10%) and PVDF membrane transfer (cat. no. IPVH00010; Millipore; Burlington, MA, USA), membranes were blocked and incubated with the following primary antibodies: anti-NMDAR1 (1:6000, cat. no. ab109182; Abcam), anti-KCC2 (1:10,000, cat. no. ab259969; Abcam), anti-phospho-KCC2 (Ser940) (1:3000, cat. no. CAY-29291-100; Cayman Chemical; Ann Arbor, MI, USA), anti-PP1C$\gamma$ (1:2000, cat. no. bs-5422R; Bioss; Beijing, China), anti-PP2A$\alpha$ (1:2000, cat. no. 13482-1-AP; Proteintech), anti-Calpain1 (1:2000; WL03987; Wanleibio; Shenyang, Liaoning, China), anti-SPTAN1 (1:2000, cat. no. 31676-1-AP; Proteintech), and anti-$\beta$-actin (1:10,000, cat. no. 66009-1-Ig; Proteintech). Signals were imaged with the CFX Connect Real-Time System (Bio-Rad; Hercules, CA, USA). Semi-quantitative assessment of protein expression was conducted via densitometry using ImageJ (Fiji; National Institutes of Health).

The data analyses of this study were conducted using SPSS Statistics 25.0 (IBM; Armonk, NY, USA) and GraphPad Prism 10 (GraphPad Software; San Diego, CA, USA). One-way ANOVA, two-way ANOVA, and two-way repeated measures ANOVA were performed as appropriate to assess the significance. Data are presented as mean $\pm$ standard deviation (SD), and differences were considered statistically significant at p 0.05.

Neurobehavioral tests were conducted to assess the therapeutic effects of acupuncture on PSSH. To evaluate the impact of acupuncture on neurological deficits, the mNSS was assessed on day 3 and day 10 post-MCAO. All rats had an mNSS score of 0 prior to surgery. On day 3 after surgery, compared to sham-operated rats, MCAO surgery induced neurological impairments manifested by significantly elevated mNSS scores (all p 0.001) (Fig. 1D). There were effects of time (F[2,108] = 2154, p 0.001), group (F[3,108] = 934.1, p 0.001), and their interaction (F[6,108] = 257.4, p 0.001). After the intervention, the mNSS scores in the Acup group markedly decreased by day 10 and were lower than those in the Model and CP group (all p 0.001; Fig. 1D). In contrast, no meaningful difference was observed between the Model and CP groups on day 10 (p $>$ 0.05; Fig. 1D). These results indicate that acupuncture promoted recovery of neurological function following stroke.

To assess the role of acupuncture on spasticity, the MAS scores of the affected hind limb were measured daily for 10 consecutive days after surgery. All rats had a MAS score of 0 before surgery. Starting from postoperative day 2, the MAS scores of rats in the three MCAO groups markedly increased, but did not in the sham rats (p 0.001) indicating successful induction of spastic hypertonia by MCAO (Fig. 1E). Significant main effects of time (F[10,396] = 82.92, p 0.001), group (F[3,396] = 326.8, p 0.001), and their interaction (F[30,396] = 14.13, p 0.001) were observed. During the treatment period, acupuncture caused a decrease in the MAS scores. From day 7 post-operation onward, the MAS scores in the Acup group were significantly lower than those in the Model group (p 0.001) and the CP group (p 0.01; Fig. 1E). This difference persisted until day 10 post-operation (all p 0.001; Fig. 1E). No significant differences existed between the Model and CP groups at any time (all p $>$ 0.05; Fig. 1E). These results suggested that acupuncture effectively alleviated MCAO induced spasticity.

Building on the results of subjective behavioral assessments, we subsequently performed footprint analysis to objectively evaluate motor function in rats (Fig. 1H). On postoperative day 3, MCAO rats in the three surgery groups showed shorter stride lengths of the hind limb than did the sham-operated rats (all p 0.001) and an increased limb-coordination index (all p 0.001; Fig. 1F,G), indicating significant motor impairment induced by MCAO (Fig. 1F,G). By day 10 post-MCAO, the Acup group showed a longer stride length than did either the Model or CP groups (both p 0.001), with no difference between the latter two groups (p $>$ 0.05; Fig. 1F). Additionally, no meaningful differences in the limb coordination index were detected among the three surgery groups (all p $>$ 0.05; Fig. 1G). These findings suggested that acupuncture intervention promoted motor recovery after stroke, primarily by improving stride characteristics.

We assessed the impact of acupuncture on spinal hyperexcitability and ischemic brain damage. The H-reflex is widely used to assess $\alpha$-motoneuron excitability and the monosynaptic transmission from Ia afferents under spastic conditions [27]. Supplementary Table 3 summarizes the basic characteristics of M-wave and H-wave responses. No significant differences were detected among these groups on most parameters, except for Mmax and Hmax (all p $>$ 0.05). To further quantify reflex excitability, as shown in Fig. 2A, the Hmax/Mmax ratio was larger in the Model group than in the Sham-operated group (p 0.001), indicating enhanced spinal excitability after stroke. Acupuncture treatment at GB34 significantly decreased the Hmax/Mmax ratio (p 0.001), whereas needle manipulation at the control point did not show such an effect (p $>$ 0.05; Fig. 2A). These results suggested that acupuncture effectively suppressed spinal hyperexcitability after stroke.

Fig. 2.

Acupuncture mitigates MCAO-induced spinal hyperexcitability and ischemic brain injury. (A) Hmax/Mmax ratio was measured to assess spinal excitability (N = 6). (B) FDD of the H-reflex (N = 6). (C) Representative H-reflex traces recorded at 0.3 Hz (blue), 5 Hz (purple), and 10 Hz (red) to show changes in the H-reflex (N = 6). (D) Distribution of c-Fos in the lumbar spinal ventral horn neurons. Scale bar, 500 µm. (E) Representative images of c-Fos expression in the lumbar spinal ventral horn of each group. Scale bar, 200 µm. (F) Analysis of c-Fos-positive cell density (cells per mm²) (12 spinal slices from 4 rats per group). (G,H) TTC staining and quantitative analysis of infarction areas in the brain (N = 6). Scale bar, 5 mm. Data are presented as mean $\pm$ SD. ^{◆⁢◆⁢◆}p 0.001 compared to the Model group; ^#⁢#⁢#p 0.001 compared to the Acup group; ***p 0.001. H/M ratio, Hmax/Mmax ratio; FDD, frequency-dependent depression.

Subsequently, FDD was measured to evaluate spinal reflex activity in MCAO rats. Impaired FDD is widely recognized as a hallmark of spinal hyperreflexia [28]. We evoked H-reflexes at stimulation frequencies of 0.3 Hz, 5 Hz, and 10 Hz to evaluate the extent of FDD. The results showed a frequency-dependent decline in H-reflex amplitude across all groups, indicating the presence of FDD to varying degrees (Fig. 2B,C). Significant effects were observed for stimulation frequency (F[2,60] = 588.7, p 0.001), group (F[3,60] = 206.9, p 0.001), and their interaction (F[6,60] = 52.14, p 0.001). Further analysis demonstrated that at both 5 Hz and 10 Hz, the Acup group exhibited greater depression of the H-reflex than did the Model group (p 0.001), whereas the CP group FDD was not different from that of the Model group (p $>$ 0.05; Fig. 2B). These findings suggested that acupuncture at GB34 partially restored FDD and enhanced spinal-reflex modulation after stroke.

As a cellular marker of neuronal activation, c-Fos expression in spinal-ventral-horn neurons was analyzed to further verify the modulatory effect of acupuncture on neuronal excitability (Fig. 2D,E). The c-Fos-positive neuron count was higher in model rats than in sham-operated rats (p 0.001; Fig. 2F). Compared with Model group, acupuncture exerted substantial reduction of the c-Fos-positive neuron count (p 0.001), whereas no similar changes were observed in CP group (p $>$ 0.05; Fig. 2F). These findings suggested that acupuncture at GB34 effectively suppressed hyperexcitability in the spinal ventral horn induced by stroke.

TTC staining was used to evaluate ischemic brain injury. As depicted in Fig. 2G, no obvious infarct area was observed in the Sham group, whereas all three surgery groups exhibited clear ischemic regions, confirming successful induction of cerebral infarction by MCAO. After intervention, the infarct area in the Acup group was smaller than that in the Model group (p 0.001), suggesting that acupuncture mitigated ischemic brain damage (Fig. 2H). No significant difference was observed between the Model and the CP group (p $>$ 0.05; Fig. 2H).

NMDAR enhances neuronal excitability primarily by mediating Ca²⁺ influx, whereas KCC2 maintains low intracellular Cl^– levels to support GABAergic inhibition. Both play critical roles in regulating neuronal excitability. We investigated whether acupuncture treatment alleviated PSSH by modulating NMDAR and KCC2 expression in the spinal ventral horn. NMDAR1 is involved in Ca²⁺ influx and regulating neuronal excitability [29]. We assessed NMDAR1 expression in the spinal ventral horn. The results showed that NMDAR1 mRNA and protein expression levels were higher in the Model group than in the Sham group (all p 0.001; Fig. 3A,E,F; Fig. 4A,C). After acupuncture treatment, NMDAR1 expression was markedly lower in the Acup group than in the Model group (all p 0.001), suggesting that acupuncture effectively inhibited NMDAR1 expression after stroke (Fig. 3A,E,F; Fig. 4A,C).

Fig. 3.

Acupuncture downregulates the expression of NMDAR1 and restores KCC2 expression and activation in spinal cord of MCAO rats with PSSH. (A–D): The mRNA expressions of NMDAR1 (A), KCC2 (B), PP1C$\gamma$ (C) and PP2A (D) were detected by RT-qPCR. (E) Representative Western blot images showing the expression of NMDAR1, S940, KCC2, PP1C$\gamma$, PP2A$\alpha$, calpain1, and SPTAN1. (F–J): Quantitative analysis of expressions of NMDAR1 (F), KCC2 (G), S940/KCC2 (H), PP1C$\gamma$ (I), PP2A$\alpha$ (J), calpain1 (K) and SPTAN1 (L) by using Western blot. N = 6 per group. Data are presented as mean $\pm$ SD. *p 0.05, **p 0.01, ***p 0.001. NMDAR1, N-methyl-D-aspartate receptor 1; KCC2, potassium-chloride cotransporter 2; S940, serine940; SPTAN1, nonerythroid spectrin $\alpha$II; PP2A, protein phosphatase 2A.

Fig. 4.

Acupuncture downregulates the expression of NMDAR1 in spinal ventral horn and upregulates the total and membrane expression of KCC2 in spinal motor neuros. (A,B) Representative immunofluorescence images indicating NMDAR1 (A, green), KCC2 (B, red) co-localization with all nuclei of cells stained by DAPI (blue) in the lumbar spinal ventral horn. Scale bar, 200 µm, 100 µm, 20 µm. The white arrows indicate representative individual neurons, and the white boxes indicate regions of interest, which are enlarged and shown adjacent to the original image. (C–E) Mean fluorescence intensity of NMDAR1 (C), KCC2 total intensity (D) and KCC2 membrane intensity (E) were expressed as the ratio of integrated density to the observation fields. 12 spinal slices from 4 rats per group. Data are presented as mean $\pm$ SD. *p 0.05, ***p 0.001. DAPI, 4^′,6-Diamidino-2-phenylindole.

Then we evaluated KCC2 expression in the spinal ventral horn. The results consistently indicated that both KCC2 mRNA levels and total protein expression were lower in the Model group than in the Sham group (all p 0.001), whereas acupuncture markedly restored KCC2 expression in the Acup group (all p 0.001; Fig. 3B,E,G; Fig. 4B,D). We further examined the phosphorylation level at Ser940 and assessed KCC2 membrane localization. The results showed that both Ser940 phosphorylation and membrane-associated KCC2 expression were lower in the Model group than in the Sham group, and acupuncture treatment significantly reversed these impairments (all p 0.001; Fig. 3E,H; Fig. 4E). These findings suggested that acupuncture upregulated KCC2 total expression and promoted its functional recovery by enhancing Ser940 phosphorylation and increasing membrane localization. Collectively, these results suggested that acupuncture attenuated neuronal hyperexcitability in the spinal ventral horn by downregulating NMDAR1 and restoring KCC2 expression and activity, thereby contributing to its therapeutic effects on PSSH.

We further explored the potential mechanisms by which NMDAR suppresses KCC2 expression and activity. Phosphorylation at the Ser940 site is primarily dephosphorylated by PP1 and protein phosphatase 2A (PP2A). We assessed the expression of PP1C$\gamma$ and PP2A$\alpha$, the catalytic subunits of PP1 and PP2A, in the spinal ventral horn [30, 31]. The results showed that, compared with the Sham group, the Model group exhibited elevated mRNA and protein levels of PP1C$\gamma$ (both p 0.01), whereas acupuncture markedly downregulated its expression (both p 0.01; Fig. 3C,E,I). In contrast, no differences were observed in the mRNA of PP2A and protein levels of PP2A$\alpha$ among the groups (all p $>$ 0.05; Fig. 3D,E,J). These findings suggested that acupuncture promoted KCC2 activity by inhibiting PP1-mediated Ser940 dephosphorylation.

Dephosphorylation of KCC2 at the Ser940 site leads to its inactivation and internalization, followed by degradation by calpain1 in the cytoplasm. The activity of calpain1 is highly dependent on NMDAR-mediated Ca²⁺ influx [32, 33]. Nonerythroid spectrin $\alpha$II (SPTAN1) is a classical substrate of calpain1; it is cleaved into a stable 150-kDa fragment, and is commonly used as a marker of calpain1 activity [32]. We measured the levels of calpain1 and SPTAN1 in the spinal ventral horn. As shown in Fig. 3L, the protein level of SPTAN1 was higher after MCAO (p 0.001) than in sham rats, and acupuncture evidently inhibited this elevation (p 0.001). However, calpain1 protein expression did not differ markedly among all groups (p $>$ 0.05; Fig. 3E,K). These findings suggested that acupuncture may have inhibited calpain1 activity, rather than affecting its expression, thereby reducing KCC2 degradation and promoting its expression. All original WB figures in Fig. 3 are provided in the Supplementary Material-Uncut western bloting images.

NMDAR directly enhances neuronal excitability by mediating Ca²⁺ influx and downregulates the expression and activity of KCC2, thereby indirectly weakening inhibitory regulation. Given that acupuncture has been shown to alleviate PSSH by modulating KCC2, we further investigated whether this effect was associated with its regulation of NMDAR1. To this end, the NMDAR agonist NMDA and antagonist D-AP5 were administered intrathecally. We evaluated mNSS scores, MAS scores, and H-reflex parameters in each group. After treatment, the Acup + NMDA group showed significantly higher mNSS and MAS scores, a higher Hmax/Mmax ratio, and lower FDD suppression than did the Acup + Vehicle group (all p 0.001), indicating that NMDAR activation diminished the therapeutic effects of acupuncture (Fig. 5B–F). In contrast, the Model + D-AP5 group exhibited significant improvements in behavioral and electrophysiological outcomes (all p 0.001), suggesting that NMDAR inhibition alleviated neurological deficits and spasticity, and reduced spinal motor-neuron excitability (Fig. 5B–F). Additionally, no significant intergroup variation was observed between the Model group and the Model + Vehicle group, suggesting that the vehicle did not interfere with the results (all p $>$ 0.05; Fig. 5B–F).

Fig. 5.

The impact of NMDAR antagonist and agonist on acupuncture effects on neurological deficits, spasticity, and H-reflex in MCAO rats with PSSH. (A) Timeline of experimental procedures. (B) Neurological deficits assessed by the mNSS (N = 10). (C) Muscle tone assessed by the MAS (N = 10). (D) Hmax/Mmax ratio was measured to assess spinal excitability (N = 6). (E) FDD of the H-reflex (N = 6). (F) Representative H-reflex traces recorded at 0.3 Hz (blue), 5 Hz (purple), and 10 Hz (red) to show changes in the H-reflex. Data are presented as mean $\pm$ SD. ^△△p 0.01, ^△⁣△△p 0.001 compared to the Model + vehicle group; ^&&p 0.01, ^&⁣&&p 0.001 compared to the Acup + vehicle group; ***p 0.001.

To verify the effectiveness of the pharmacological intervention in modulating NMDAR1, we examined NMDAR1 expression in the spinal ventral horn. Results showed that NMDAR1 mRNA and protein levels were evidently lower in the Model + D-AP5 group than in the Model + Vehicle group (all p 0.01), whereas they were significantly higher in the Acup + NMDA group than in the Acup + Vehicle group (all p 0.01; Fig. 6A,D,E; Fig. 7A,C). No significant intergroup variation was observed between the Model group and the Model + Vehicle group (all p $>$ 0.05; Fig. 6A,D,E; Fig. 7A,C). These findings confirmed that the pharmacological agents effectively modulated NMDAR1 expression and that the vehicle did not interfere with the results. We next assessed the expression and activity of KCC2 in the spinal ventral horn. The Acup + NMDA group exhibited significantly lower KCC2 mRNA levels, total protein expression, Ser940 phosphorylation, and membrane expression than did the Acup + Vehicle group (all p 0.001), indicating that NMDAR activation attenuated the upregulation of KCC2 by acupuncture (Fig. 6B,D,F,G; Fig. 7B,D,E). Conversely, all of these measures were higher in the Model + D-AP5 group than in the Model + Vehicle group (all p 0.01), suggesting that NMDAR inhibition facilitated the restoration of KCC2 expression and activity after stroke (Fig. 6B,D,F,G; Fig. 7B,D,E). Taken together, these findings suggested that NMDAR modulation was closely associated with the effect of acupuncture on PSSH by upregulating KCC2.

Fig. 6.

The effects of NMDAR antagonist and agonist on the NMDAR-PP1/Calpain1-KCC2 pathway. (A–C) The mRNA expressions of NMDAR1 (A), KCC2 (B) and PP1C$\gamma$ (C) were detected by RT-qPCR. (D) Representative Western blot images showing the expression of NMDAR1, S940, KCC2, PP1C$\gamma$ and SPTAN1. (E–I): Quantitative analysis of expressions of NMDAR1 (E), KCC2 (F), S940/KCC2 (G), PP1C$\gamma$ (H) and SPTAN1 (I) by using Western blot. N = 6 per group. Data are presented as mean $\pm$ SD. *p 0.05, **p 0.01, ***p 0.001.

Fig. 7.

The effects of NMDAR antagonist and agonist on the expression of NMDAR1 in spinal ventral horn and the total and membrane expression of KCC2 in spinal motor neuros. (A,B) Representative immunofluorescence images indicating NMDAR1 (A, green), KCC2 (B, red) co-localization with all nuclei of cells stained by DAPI (blue) in the lumbar spinal ventral horn. Scale bar, 200 µm, 100 µm, 20 µm. The white arrows indicate representative individual neurons, and the white boxes indicate regions of interest, which are enlarged and shown adjacent to the original image. (C–E) Mean fluorescence intensity of NMDAR1 (C), KCC2 total intensity (D) and KCC2 membrane intensity (E) were expressed as the ratio of integrated density to the observation fields. 12 spinal slices from 4 rats per group. Data are presented as mean $\pm$ SD. *p 0.05, ***p 0.001. D-AP5, D-(-)-2-Amino-5-phosphonopentanoic acid.

We next determined whether the process by which acupuncture promoted KCC2 activity by inhibiting PP1-mediated Ser940 dephosphorylation was associated with the regulation of NMDAR. The Acup + NMDA group exhibited higher PP1C$\gamma$ mRNA and protein levels than did the Acup + Vehicle group (both p 0.05), whereas the Model + D-AP5 group showed lower levels of expression of PP1C$\gamma$ than did the Model + Vehicle group (both p 0.001; Fig. 6C,D,H). These results suggested that acupuncture maintained KCC2 activity by suppressing NMDAR-induced upregulation of PP1.

We further clarified whether the process by which acupuncture inhibited calpain1 to reduce KCC2 degradation was associated with the regulation of NMDAR1 by acupuncture. SPTAN1 expression was substantially higher in the Acup + NMDA group than in the Acup + Vehicle group (p 0.001), whereas it was lower in the Model + D-AP5 group than in the Model + Vehicle group (p 0.001; Fig. 6D,I). These data causally suggested that acupuncture enhanced the protein stability of KCC2 by suppressing NMDAR-mediated activation of calpain1. All original WB figures in Fig. 6 are provided in the Supplementary Material-Raw data.