Acupotomy Alleviates Pain and Attenuates Spinal Neuronal Hyperexcitability in Knee Osteoarthritis Rats: Implication of the CaMKII/GluA1 Axis

Abstract

Objective

To investigate the spinal mechanisms underlying the analgesic effects of acupotomy on chronic pain and central sensitization in rats with monosodium iodoacetate (MIA)-induced knee osteoarthritis (KOA).

Methods

Following MIA-induced KOA and a three-week acupotomy intervention in rats, pain behaviors and joint pathology were assessed. Spinal RNA-sequencing was performed to identify differentially expressed genes. Key targets (CaMKII and GluA1) were validated at both mRNA and protein levels using RT-qPCR and Western blotting. Furthermore, immunofluorescence and whole-cell patch-clamp electrophysiology were employed to evaluate the activation and excitability of spinal substantia gelatinosa (SG) neurons.

Results

Acupotomy significantly alleviated mechanical and thermal hyperalgesia and reduced joint degeneration in KOA rats. RNA-seq analysis revealed the modulation of 220 genes in the spinal cord. These DEGs were significantly enriched in calcium signaling and synaptic transmission pathways, highlighting CaMKII and Gria1 as potential key molecules. Molecular validations confirmed that acupotomy downregulated CaMKII and Gria1 mRNA expression and the phosphorylation ratios of CaMKII and GluA1. Furthermore, acupotomy reduced the activation of p-CaMKII-positive neurons in the spinal SG region and dampened their abnormally high-frequency action potential discharges.

Conclusion

Acupotomy intervention ameliorates peripheral joint degeneration and exerts significant analgesia in KOA, which is accompanied by a reduction in the intrinsic hyperexcitability of spinal SG neurons. Although direct causal validation utilizing targeted inhibitors is warranted, our findings suggest that this analgesic effect is closely associated with the downregulation of the CaMKII/GluA1 signaling axis, providing a plausible central mechanistic framework.

Graphical Abstract

Introduction

Knee osteoarthritis (KOA) is a degenerative joint disease characterized by the progressive destruction of articular cartilage, subchondral bone sclerosis or cyst formation, synovial inflammation, and periarticular muscle atrophy.¹ Its hallmark clinical manifestations are refractory chronic pain and severe functional impairment.² As a major contributor to the global burden of musculoskeletal pain—the second leading cause of disability worldwide—KOA not only severely diminishes patients’ quality of life but also drives substantial productivity losses and escalating healthcare costs.^3,4 Currently, non-steroidal anti-inflammatory drugs (NSAIDs) and opioids serve as the standard pharmacological interventions. Although these agents offer temporary symptom relief, their long-term use is profoundly limited by systemic adverse effects, particularly gastrointestinal, cardiovascular, and renal toxicities.⁵ Furthermore, patients with advanced-stage KOA frequently require total joint arthroplasty due to severe cartilage deterioration and intolerable pain.⁶ Consequently, exploring safe, effective, and targeted intervention strategies capable of providing long-lasting analgesia remains a critical unmet clinical need in the management of KOA.

The clinical manifestation of KOA-associated pain evolves dynamically throughout the disease course. Early-stage KOA is typically characterized by intermittent, activity-induced sharp pain, which gradually transitions into spontaneous, persistent dull pain and hyperalgesia in advanced stages.⁶ Nociceptive transmission initiates with the transduction of pathological stimuli by peripheral nociceptors. These nerve impulses travel along primary afferent fibers to the spinal dorsal horn for initial integration, and subsequently project to the brainstem, cerebral cortex, and limbic system, ultimately generating pain perception.⁷ Emerging evidence suggests that sustained aberrant nociceptive input from the periphery can induce and maintain “central sensitization.” Defined operationally as an amplified responsiveness of nociceptive neurons in the central nervous system to their normal or subthreshold afferent input,⁸ central sensitization is increasingly recognized as a crucial contributor to the refractory chronic pain in KOA, acting in concert with peripheral nociceptive, inflammatory, neuropathic-like, and psychological factors.³ As the primary relay and integration center for peripheral nociceptive signals entering the central nervous system, the spinal dorsal horn plays a pivotal role in this process.⁹ Numerous studies have demonstrated that in chronic KOA pain models, the spinal dorsal horn exhibits profound synaptic plasticity and altered synaptic transmission efficacy, accompanied by specific aberrant gene expression profiles.^10,11 Consequently, shifting the mechanistic focus from peripheral joint pathology to spinal central sensitization provides a critical avenue for elucidating the complex pathogenesis of KOA-associated chronic pain.

Acupotomy is a minimally invasive intervention that utilizes a micro-scalpel to release pathological tissues, aiming to relieve mechanical compression and restore joint biomechanics. Although widely used for KOA management, systematic reviews indicate that the current clinical evidence is limited by methodological constraints, including reliance on subjective indicators, inadequate allocation concealment, small sample sizes, and the lack of standardized sham controls.¹² Therefore, acupotomy is considered a therapeutic option requiring further biological validation. However, existing mechanistic studies are predominantly confined to peripheral soft tissue repair and local inflammatory modulation.^13,14 This lack of clarity regarding central mechanisms underscores the need for preclinical investigations to bridge the gap between clinical observations and underlying molecular rationales.

Despite the recognized role of spinal central sensitization in KOA pain, the precise molecular mechanisms underlying the analgesic effects of acupotomy remain largely unexplored. While synaptic plasticity and glutamatergic neurotransmission are known to be broad contributors to central sensitization, the specific transcriptomic alterations and molecular axes modulated by acupotomy are not fully understood. To address this gap and avoid a priori mechanistic assumptions, we employed an unbiased RNA-seq to screen for differentially expressed genes in the spinal cord of KOA rats following acupotomy intervention. Guided by these transcriptomic findings, we subsequently focused on the CaMKII/GluA1 signaling pathway. Through molecular validations and whole-cell patch-clamp electrophysiology, we aimed to explore whether the modulation of this specific pathway contributes to the dampened intrinsic excitability of substantia SG neurons, thereby providing a potential biological rationale for acupotomy-induced analgesia.

Methods

Animals and Groups

Male SPF Sprague-Dawley (SD) rats (weighing 230–250 g) were purchased from SPF Beijing Huafukang Co. Ltd. All animals were housed at the Animal Experiment Center of Beijing University of Chinese Medicine under standard environmental conditions (temperature: 22 ± 2°C, humidity: 50%–60%, 12-hour light/dark cycle) with ad libitum access to food and water. All experimental procedures complied with the National Institutes of Health (NIH) guidelines and were approved by the Experimental Animal Ethics Committee of Beijing University of Chinese Medicine (Approval No. BUCM2024091102-3197). Following a one-week acclimatization period, rats underwent baseline pain threshold screening to minimize intrinsic behavioral variance. Animals exhibiting extreme baseline hypersensitivity or hyposensitivity were excluded prior to modeling. The final cohort size was justified by an a priori power analysis (G*Power 3.1; effect size f = 0.6, α = 0.05, power = 0.90) for the primary in vivo outcomes, yielding a target of exactly 39 rats with consistent baseline thresholds for the formal study. These 39 rats were randomly assigned to three groups (n = 13 per group): the sham operation group (Sham), the knee osteoarthritis group (KOA), and the acupotomy intervention group (Apo). Randomization was performed via a computer-generated sequence. Investigators responsible for behavioral testing, histological scoring, image quantification, electrophysiological recording, and statistical analysis were blinded to group allocation. Samples for molecular assays and RNA-seq were processed using coded identifiers, and group identities were revealed only after data processing was completed. Following behavioral evaluations, the animals in each group were further allocated into distinct sub-cohorts (n = 3–6) due to mutually exclusive tissue harvesting requirements.

Induction of Osteoarthritis

Anesthesia was induced with 4% isoflurane in oxygen (1 L/min) and maintained with 2% isoflurane in oxygen (1 L/min). Following routine skin preparation and disinfection, the lower pole of the right patella and the tibial tuberosity were identified by palpation. A 22G needle (0.7 mm × 30 mm) was slowly inserted through the lateral or medial edge of the patellar ligament to a depth of approximately 3–5 mm. Upon confirmation of intra-articular placement, 50 μL of MIA solution (2 mg/50 μL) was slowly injected into the joint cavity. The sham-operation group received an intra-articular injection of an equal volume of normal saline. Post-injection, the knee joint was passively flexed and extended five times to ensure even distribution of the solution within the joint cavity. Finally, the rats were returned to their home cages for recovery.

Acupotomy Intervention

Two weeks post-modeling, the animals were anesthetized using the identical isoflurane protocol described above, and the right hindlimb was positioned with the knee joint flexed at a 45° angle. Nodules, tissue adhesions, and high-stress points along the medial and lateral collateral ligaments were identified via palpation and marked with gentian violet. Following standard skin preparation and disinfection, acupotomy was performed. The acupotomy needle was inserted perpendicularly through the skin, maintaining the blade parallel to the longitudinal axis of the ligament until it reached the bone surface. At the medial aspect of the knee (the medial collateral ligament), longitudinal loosening and transverse peeling were executed using three strokes. The longitudinal incision length was restricted to approximately 1.5 mm, with a transverse peeling angle from 60° to 90°. At the lateral aspect of the knee (the lateral collateral ligament), the same technique was applied using a single stroke, with a longitudinal incision of approximately 1.0 mm and a transverse peeling angle from 30° to 45°. Immediately following the procedure, the instrument was withdrawn, and brief manual compression was applied to achieve hemostasis. This acupotomy intervention was repeated every three days over a consecutive three-week period. The remaining two control groups received only anesthesia without any surgical manipulation.

Measurement of Knee Joint Diameter

Calipers were used to measure the transverse diameters of the bilateral knee joints before modeling, as well as on days 7, 14, 21, 28, and 35 following the procedure. To ensure accuracy, measurements for each knee joint were taken in triplicate and averaged.

Paw Withdrawal Latency (PWL)

All rats were allowed to acclimate to the testing environment on the apparatus for 5 min prior to the experiment. During the test, the surface of the hot plate was maintained at a constant temperature of 55.0 ± 0.1°C. Timing commenced the moment the right hind paw contacted the heated surface and was terminated upon the first display of nociceptive behavior, such as paw licking, jumping, or rapid paw withdrawal. Each rat underwent three consecutive trials with a 5-minute inter-trial interval. The mean of the three trials was calculated to determine the PWL for each animal.

Mechanical Withdrawal Threshold (MWT)

In a quiet environment maintained at 24 ± 2°C, the rats were placed in transparent Plexiglas chambers on an elevated stainless steel wire mesh platform. The plantar surface of the right hind paw was stimulated perpendicularly using the 0.8 mm diameter rigid tip of an electronic von Frey anesthesiometer. An increasing force was applied at a uniform rate until a brisk paw withdrawal response was elicited. Each rat was tested in triplicate with a 10-minute inter-trial interval, and the mean value was recorded as the MWT.

Sample Collection

Following the experiments, three rats per group were randomly selected. In strict accordance with AVMA guidelines, the animals were deeply anesthetized via an intraperitoneal injection of a cocktail containing ketamine (100 mg/kg), atropine (0.1 mg/kg), and diazepam (5 mg/kg). Upon confirming the complete loss of the pedal withdrawal reflex, the rats were transcardially perfused with 4% paraformaldehyde (PFA). To isolate the spinal lumbar enlargement, the 13th thoracic vertebra (T13) was identified by palpating and tracing the 13th rib. A laminectomy spanning T12 to L2 was then performed to extract the L4–L6 spinal segments. The harvested knee joints and spinal cords were post-fixed in 4% PFA for 24 hours. Knee joints were decalcified in EDTA solution (replaced every 4 days for 8 weeks until penetrable by a fine needle) prior to paraffin embedding. Subsequently, knee joints and spinal cords were cut into 4 µm coronal and transverse sections, respectively, for hematoxylin and eosin (H&E) and immunofluorescence staining. For the remaining rats, deep anesthesia and reflex loss were induced and confirmed using the identical protocol. Following rapid euthanasia via cervical dislocation, the lumbar enlargement was freshly harvested using the same anatomical landmarks. These fresh tissues were immediately allocated for downstream analyses: 4 rats per group for patch-clamp recordings and transcriptomic analysis, and 6 rats per group for polymerase chain reaction (PCR) and Western blotting (WB).

HE Staining

Morphological changes of the knee joints were assessed using hematoxylin and eosin (H&E) staining. Briefly, tissue sections were deparaffinized in xylene (2 × 10 min) and rehydrated in descending concentrations of ethanol (100% to 70%). Following hematoxylin staining, differentiation with 1% acid alcohol, and bluing with 1% ammonia, the sections were counterstained with eosin. After dehydration in an ascending ethanol series and clearing in three changes of xylene, the slides were mounted with neutral resin and examined under a light microscope for image acquisition.

Transcriptome Sequencing and Analysis

Total RNA was extracted from four randomly selected lumbar spinal cord tissue samples per group using TRIzol reagent. RNA concentration and purity were assessed using a NanoDrop 2000 spectrophotometer (concentration ≥ 20 ng/µL), while RNA integrity was evaluated via agarose gel electrophoresis. The RNA Quality Number (RQN) was determined using an Agilent 5300 Fragment Analyzer, with an inclusion criterion of RQN > 4.5. Poly(A) mRNA was enriched from the total RNA using oligo(dT) magnetic beads. The isolated mRNA was then randomly fragmented into approximately 300 bp pieces, which served as templates for first- and second-strand cDNA synthesis via reverse transcription. Following end repair, A-tailing, and the ligation of sequencing adapters, the cDNA fragments were purified, size-selected, and PCR-amplified to construct the sequencing library. Paired-end sequencing was performed on an Illumina HiSeq platform with the assistance of Majorbio Bio-Pharm Technology Co., Ltd.

Raw sequencing reads were filtered to remove adapter sequences and low-quality reads, yielding high-quality clean data for downstream analysis. The clean reads were aligned to the reference genome using HISAT2 software, and gene expression levels were quantified using RSEM. Differential expression analysis between the groups was conducted using the DESeq2 package. Differentially expressed genes (DEGs) were identified based on a threshold of |log2(fold change)| > 1.5 and a statistical significance of Pajust < 0.05. Venn diagrams were generated to identify overlapping key DEGs. Finally, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed to elucidate the primary biological functions and signaling pathways associated with the DEGs, focusing on terms and pathways with Pajust < 0.05.

RT-qPCR

The lumbar enlargement tissues of the spinal cord (50 mg/sample) were placed into pre-cooled RNase-free Eppendorf tubes. An appropriate volume of lysis buffer and grinding beads were added for thorough tissue homogenization. Total RNA was subsequently extracted using the TRIzol method. A 20 μL reverse transcription reaction was prepared according to the manufacturer’s instructions for the cDNA synthesis kit. The reverse transcription thermocycling conditions were as follows: incubation at 25°C for 5 min, 42°C for 30 min, and enzyme inactivation at 85°C for 5 s.

For quantitative real-time PCR (RT-qPCR), the amplification mixture was prepared and subjected to the following cycling program: initial denaturation at 95°C for 30s, followed by 40 cycles of denaturation at 95°C for 15s, and a combined annealing/extension step at 60°C for 30s. All primers were designed and synthesized by Wuhan Servicebio Technology Co., Ltd. (Wuhan, China). Using GAPDH as the endogenous reference gene, the relative mRNA expression levels of the target genes (CaMKII and Gria1) were calculated employing the 2^−ΔΔCt method. The primer sequences and expected product lengths were as follows: CaMKII (202 bp): Forward 5’-AAGATGTGCGACCCTGGAA-3’, Reverse 5’-GGTACTGAGTGATGCGGATGTAG-3’; Gria1 (141 bp): Forward 5’-GAATCAGAACGCCTCAACGC-3’, Reverse 5’-TGTCACATTGGCTCCGCTCT-3’; GAPDH (138 bp): Forward 5’-CTGGAGAAACCTGCCAAGTATG-3’, Reverse 5’-GGTGGAAGAATGGGAGTTGCT-3’.

Western Blotting (WB)

Western blotting was performed to detect the protein expression levels of p-CaMKII and p-GluA1 in the spinal cord dorsal horn. The tissue homogenates were centrifuged at 12,000 rpm (approximately 15,400 × g) for 10 min at 4°C, and the total protein concentration of the supernatant was determined using a BCA protein assay kit. Subsequently, 5× loading buffer was added to the protein samples, which were then boiled for 15 min to ensure complete denaturation. Equal amounts of protein were separated by SDS-PAGE and transferred onto membranes. The membranes were blocked with skim milk for 1 h at room temperature and then incubated overnight at 4°C with primary antibodies against CaMKII (1:2000), p-CaMKII (1:2,000), GluA1 (1:2,000), p-GluA1 (1:2,000), and βactin (1:2,000). Following standard washing steps, the membranes were incubated with the corresponding secondary antibodies (1:500 for p-CaMKII and p-GluA1; 1:5,000 for β-actin) for 2 h. Protein bands were visualized using an enhanced chemiluminescence (ECL) solution and captured with a gel imaging system. The optical density of each band was quantified using ImageJ software. The relative expression level of each target protein was calculated as the ratio of its densitometric value to that of the internal control (β-actin).

Immunofluorescence Detection

Double immunofluorescence staining was performed to detect the co-expression of p-CaMKII and c-Fos in neurons within the substantia gelatinosa (SG) region of the spinal cord. Briefly, tissue sections were deparaffinized, subjected to microwave antigen retrieval for 20 min, and blocked with 5% bovine serum albumin (BSA) for 30 min. After removing the blocking buffer, the sections were incubated with the first primary antibody (anti-p-CaMKII, 1:1000) overnight at 4°C in a humidified chamber. Following washing, the sections were incubated with the corresponding secondary antibody for 50 min at room temperature. After another washing step, Tyramide Signal Amplification (TSA) dye was applied and incubated for 10 min at room temperature in the dark. To strip the previously bound antibodies, an immunofluorescence antibody elution buffer was applied for 5 min and discarded, followed by a second incubation with the same elution buffer for 30 min at room temperature. Subsequently, the sections were washed and incubated with the second primary antibody (anti-c-Fos, 1:3000) overnight at 4°C in a humidified chamber. After washing, the sections were incubated with the appropriate secondary antibody for 50 min at room temperature. Nuclei were counterstained with DAPI for 10 min at room temperature. After draining the excess liquid, an autofluorescence quenching agent was applied for 5 min at room temperature. Finally, the sections were rinsed under running tap water for 10 min and coverslipped using an anti-fade mounting medium. Images were acquired using a fluorescence microscope. For the quantitative analysis of immunofluorescence, 3 rats were used per group (n = 3). Three spinal cord sections were randomly selected from each rat. The number of cells showing c-Fos co-expression in p-CaMKII-positive neurons within the superficial dorsal horn was manually counted using ImageJ software. The average cell count from the three sections was calculated to represent a single biological replicate for that animal.

Patch-Clamp Electrophysiology of Spinal Cord Slices

The rats were deeply anesthetized with 2% isoflurane and decapitated. The spinal cord was rapidly excised, and the lumbar enlargement segments were sectioned at a thickness of 300 μm using a vibrating microtome in ice-cold artificial cerebrospinal fluid (ACSF). The slices were then incubated in ACSF continuously aerated with carbogen (95% O2 and 5% CO2) for at least 1 h prior to recording. The ACSF contained (in mM): 124 NaCl, 25 NaHCO3, 2.5 KCl, 1 NaH2PO4, 2 CaCl2, 1 MgSO4, and 10 glucose. Whole-cell recordings were performed using a Multiclamp 700B amplifier (Molecular Devices). Data acquisition and analysis were conducted using pCLAMP 10.2 software (Clampex and Clampfit). Recording microelectrodes were pulled from borosilicate glass capillaries with a tip resistance of 4–6 MΩ when filled with the intracellular solution. For the recording of evoked action potentials, the intracellular solution contained (in mM): 124 K-gluconate, 5 NaCl, 1 MgCl2, 0.2 EGTA, 2 MgATP, 0.1 Na3GTP, 10 HEPES, and 10 phosphocreatine disodium (pH adjusted to 7.2 with KOH, osmolarity 290 mOsm). To evaluate neuronal excitability, cells were recorded in current-clamp mode and held at 0 pA. Action potentials were evoked by a series of 500 ms depolarizing current steps ranging from 0 to 200 pA in 20 pA increments, and the number of action potentials was quantified. Recordings were strictly accepted only if the seal resistance was > 1 GΩ and the initial series resistance was < 35 MΩ. Data were discarded if the series resistance fluctuated by more than 15% during the recording period.

Statistical Analysis

Statistical analyses were performed using SPSS, with visualizations generated in GraphPad Prism. Two-tailed P < 0.05 was considered statistically significant. For longitudinal measures (knee diameter, MWT, PWL) and electrophysiological data (firing frequency), repeated-measures ANOVA was used to evaluate main effects and Time/Current × Group interactions, with Greenhouse-Geisser correction applied when Mauchly’s test indicated a violation of sphericity (p < 0.05). Significant interactions were followed by simple main effects analysis (multivariate general linear model) to compare groups at specific time points or current steps, with Bonferroni correction applied to all pairwise comparisons. Non-parametric histological data (modified Mankin scores), expressed as medians, were compared using the Independent-Samples Median Test with Bonferroni-corrected post-hoc comparisons. Normally distributed variables (WB, RT-qPCR, IF staining), expressed as mean ± SD or SEM, were compared using one-way ANOVA with Bonferroni post-hoc correction.

Results

Acupotomy Alleviates Pain-Like Behaviors and Improves Knee Joint Histopathology in KOA Rats

As depicted in Figure 1A, MIA-induced KOA rats received a 3-week acupotomy intervention. Compared to the Sham group, KOA rats exhibited significantly increased knee joint transverse diameters (Figure 1B), which were markedly reduced by acupotomy during days 28–35. Furthermore, acupotomy significantly reversed the KOA-induced reductions in MWT and PWL from day 21 to 35 (Figure 1C and D), effectively mitigating mechanical and thermal sensitization.

Figure 1.

Acupotomy alleviates pain-related behaviors and cartilage degeneration in rats with MIA-induced KOA. (A) Schematic diagram of the experimental timeline. (B) Changes in the ipsilateral knee joint diameter. (C) MWT. (D) PWL. (E) Representative images of H&E staining in the ipsilateral knee joints. (F) Mankin scores of the cartilage degeneration. ***P < 0.001 vs the Sham group; #P < 0.05, ##P < 0.01, ###P < 0.001 vs the KOA group; *P < 0.05 for the indicated group comparisons in (F). Data in (B–D) are presented as mean ± SEM, and data in (F) are presented as medians. n = 13 per group for behavioral and joint diameter assessments (B–D), and n = 3 per group for histological analysis (E and F).

Histological evaluation via H&E staining further corroborated these findings (Figure 1E). While the Sham group maintained intact articular cartilage with well-aligned chondrocytes, the KOA group showed severe cartilage erosion, massive chondrocyte loss, and structural disorganization. Acupotomy notably attenuated these degenerative changes, preserving cartilage smoothness and cellularity. Consistent with these morphological observations, quantitative analysis revealed that Mankin scores were significantly elevated in the KOA group, confirming severe osteoarthritic damage, whereas acupotomy intervention significantly reduced these scores (Figure 1F), thereby ameliorating histopathological joint damage.

Transcriptomic Profiling of the Spinal Cord: Identification of Key Signaling Pathways for Acupotomy Analgesia

To further investigate the potential central mechanisms underlying acupotomy-induced analgesia, we performed RNA-seq analysis on the spinal cord lumbar enlargement tissues across all experimental groups.

Screening and Identification of Differentially Expressed DEGs

Transcriptomic profiling revealed 1,060 DEGs (591 upregulated, 469 downregulated) between the KOA and Sham groups, and 397 DEGs (139 upregulated, 258 downregulated) between the Apo and KOA groups (Figure 2A–C). Venn diagram analysis identified 220 counter-regulated DEGs (177 KOA-up/Apo-down and 43 KOA-down/Apo-up) as the core targets of acupotomy (Figure 2D and E). Hierarchical clustering of these core DEGs (Figure 2F) showed that acupotomy significantly reversed the KOA-induced transcriptional changes, driving the overall expression profile back toward the Sham group.

Figure 2.

Acupotomy modulates gene expression profiles in the lumbar enlargement of the spinal cord. (A) Overview of differentially expressed genes (DEGs) across the three experimental groups, n = 4 per group. (B) Volcano plot showing the DEGs between the KOA and Sham groups. (C) Volcano plot showing the DEGs between the Apo (Acupotomy) and KOA groups. (D) Venn diagram illustrating the overlap between genes significantly upregulated following KOA induction and those significantly downregulated by acupotomy intervention. (E) Venn diagram illustrating the overlap between genes significantly downregulated following KOA induction and those significantly upregulated by acupotomy intervention. (F) Hierarchical clustering heatmap of the core DEGs. (G) Gene Ontology (GO) enrichment analysis of the core DEGs. (H) KEGG pathway enrichment analysis of the core DEGs. (I) Heatmap displaying the relative expression levels of CaMKII across individual samples based on RNA-seq data. (J) Heatmap displaying the relative expression levels of Gria1 across individual samples based on RNA-seq data. (K) Relative mRNA expression levels of CaMKII in the lumbar enlargement tissues, as assessed by RT-qPCR. (L) Relative mRNA expression levels of Gria1 in the lumbar enlargement tissues, as assessed by RT-qPCR. ***P < 0.001. Data are presented as mean ± SEM, n = 6 per group.

GO and KEGG Enrichment Analysis of Core DEGs

GO analysis (Figure 2G) revealed that the 220 core DEGs were primarily enriched in nervous system function and signal transduction. KEGG analysis (Figure 2H) showed significant enrichment in environmental information processing (EIP) pathways, including neuroactive ligand-receptor interaction, calcium signaling, and cAMP signaling, as well as cellular processes (CP) related to synaptic transmission, such as GABAergic, glutamatergic, and dopaminergic synapses. Within these calcium and synaptic pathways, transcriptomic data indicated that CaMKII and Gria1 (GluA1) were significantly upregulated in KOA rats but downregulated by acupotomy (Figure 2I and J). Given their critical roles in spinal central nociceptive transmission, the CaMKII/AMPA pathway was selected as a key candidate pathway for subsequent experimental validation.

CaMKII and Gria1 mRNA Expression in the Spinal Cord

As shown in Figure 2K and L, the mRNA expression levels of CaMKII and Gria1 in the spinal cords of the KOA group were significantly higher than those in the Sham group (P < 0.001). However, following acupotomy intervention, the mRNA levels of both target genes in the Apo group were significantly lower than those in the KOA group (P < 0.001).

Acupotomy Is Associated with Reduced Phosphorylation of CaMKII and GluA1 in the Spinal Cord

As shown in Figure 3A, to assess the activation status of the CaMKII/GluA1 signaling axis, we quantified the relative phosphorylation levels by calculating the ratios of phosphorylated to total proteins. Compared with the Sham group, the MIA-induced KOA model led to a significant increase in the ratios of p-CaMKII/CaMKII (Figure 3B) and p-GluA1/GluA1 (Figure 3C) in the rat spinal cord (P < 0.001). Conversely, acupotomy treatment significantly suppressed these elevated phosphorylation ratios in the Apo group (P < 0.05).

Figure 3.

Acupotomy suppresses the phosphorylation levels of CaMKII and GluA1 in the lumbar enlargement of the spinal cord. (A) Representative Western blot images of p-CaMKII, total CaMKII, p-GluA1, and total GluA1. (B) Quantitative analysis of the relative p-CaMKII/CaMKII ratio. (C) Quantitative analysis of the relative p-GluA1/GluA1 ratio. ***P < 0.001, *P < 0.05. Data are presented as mean ± SEM, n = 6 per group.

Acupotomy Reduces the Abnormal Activation of p-CaMKII-Positive Neurons in the Spinal Dorsal Horn of KOA Rats

As illustrated in Figure 4, C-Fos co-expression in p-CaMKII-positive neurons of the spinal dorsal horn was significantly increased in KOA rats versus the Sham group (P < 0.001), reflecting pathological neuronal activation. Notably, acupotomy effectively reversed this increase, with the Apo group showing a significantly reduced co-expression rate relative to the KOA group (P < 0.05).

Figure 4.

Activation of p-CaMKII in the dorsal horn of the spinal cord lumbar enlargement across experimental groups. (A) Representative immunofluorescence images of the spinal dorsal horn stained for DAPI (blue), p-CaMKII (red), c-Fos (green), and their merged overlay in the Sham, KOA, and Apo groups. Scale bar = 100 μm. (B) Quantification of the percentage of c-Fos⁺ cells among p-CaMKII⁺ cells in the Sham, KOA, and Apo groups. ***P < 0.001, *P < 0.05. Data are presented as mean ± SEM, n = 3 per group.

Acupotomy Reduces the Excitability of Neurons in the Spinal SG Region

As depicted in Figure 5, repeated-measures ANOVA with Greenhouse-Geisser correction demonstrated a significant main effect of current on the AP firing frequency (P < 0.001), indicating that AP frequency increased with rising depolarizing currents across all groups. Importantly, a significant Current × Group interaction was observed (P = 0.003), suggesting that the response trajectories to increasing currents differed significantly among the groups. Further post-hoc Bonferroni analysis revealed no significant differences in AP frequency among the groups at lower current injections (0–80 pA, P > 0.05). At 100 pA, the KOA group exhibited a significantly higher firing rate compared to the Sham group (P < 0.05). At higher depolarizing currents (120–200 pA), the AP frequency in the KOA group remained significantly elevated compared to the Sham group (P < 0.01). However, acupotomy intervention (Apo group) significantly decreased the AP firing frequency compared to the KOA group across these high-current steps (P < 0.05). These findings demonstrate that acupotomy intervention effectively suppresses the abnormally elevated intrinsic excitability of SG neurons induced by KOA under high-intensity stimulation.

Figure 5.

Effect of acupotomy on the action potentials of neurons in the spinal SG region. (A) Representative traces of SG neurons. (B) Quantitative analysis of the number of APs elicited by incremental depolarizing current injections (0–200 pA). *P < 0.05, **P < 0.01 vs the Sham group; #P < 0.05, ##P < 0.01 vs the KOA group. Data are presented as mean ± SEM, n = 4 per group.

Discussion

In this study, we combined transcriptomic profiling, molecular validation, immunofluorescence, and electrophysiological recording to characterize the spinal effects of acupotomy in a MIA-induced KOA rat model. Acupotomy reduced pain-related behaviors and attenuated several markers of nociceptive neuron hyperexcitability — findings that converge on, though do not by themselves establish, a reduction in central sensitization, with a possible contribution from the spinal CaMKII/GluA1 signaling axis. We discuss these findings below in relation to existing mechanistic understanding of KOA pain, noting throughout where our data support a specific interpretation and where they remain associative.

KOA is a multifactorial chronic degenerative disease primarily characterized by joint pain and functional impairment. Central sensitization serves as a crucial mechanism underlying pain chronification and spatial spread in KOA.^15–17 Under pathological conditions, the persistent input of peripheral nociceptive signals into the central nervous system drives adaptive functional and structural changes within the spinal cord and higher pain centers.³ The MIA model, characterized by chondrocyte apoptosis and synovial inflammation, closely mimics the pathological progression of human KOA through its progressive histological degeneration and nociceptive abnormalities.^18,19 Following MIA injection, sustained inflammatory infiltration and abnormal mechanical stress activate peripheral nociceptors, which subsequently project nociceptive afferent signals to the spinal cord. Accordingly, the MIA-induced KOA model was selected for the present study. Our behavioral and morphological findings demonstrate that acupotomy intervention significantly elevates the thresholds for mechanical and thermal stimuli, exerting analgesic effects while concurrently ameliorating cartilage lesions. These findings are consistent with previous reports.²⁰ Existing literature establishes that acupotomy delivers localized therapeutic benefits by dissecting focal pathological tissues, releasing deep fascial adhesions, reducing intra-articular pressure, and restoring local biomechanical balance.^13,21,22 What remained unclear was whether acupotomy’s analgesic effect also involves changes in nociceptive processing at the spinal level, which we investigated next.

To address this, we profiled gene expression in the spinal dorsal horn by RNA sequencing. KOA altered expression of over 1,000 genes, of which acupotomy reversed 220. Subsequent KEGG pathway analysis demonstrated that these reversed genes were significantly enriched in calcium signaling and synaptic transmission pathways. Building upon these findings, we performed an in-depth exploration of key DEGs within these pathways, specifically identifying CaMKII, which encodes calcium/calmodulin-dependent protein kinase II, and Gria1, which encodes the core subunit of the AMPA receptor. Subsequent RT-qPCR validation confirmed that the relative mRNA expression levels of CaMKII and Gria1 in the spinal tissues across all groups were highly consistent with our sequencing data. Collectively, these results indicate that acupotomy effectively suppresses the upregulation of the spinal CaMKII/Gria1 signaling axis at the transcriptional level.

Although alterations in gene transcription establish the molecular framework for nociceptive transmission, shifts in synaptic efficacy during central sensitization are more directly dictated by the functional states of key proteins. Following sustained peripheral nociceptive stimulation, the influx of calcium ions (Ca²⁺) into the postsynaptic membrane not only drives the transcriptional upregulation of the aforementioned genes but also triggers the rapid autophosphorylation and activation of CaMKII (p-CaMKII) within the synapse.^23–26 Activated p-CaMKII specifically phosphorylates the GluA1 subunit, a critical component of AMPA receptors characterized by high Ca²⁺ permeability, thereby generating p-GluA1.^26,27 Previous studies have demonstrated that this CaMKII-mediated phosphorylation of GluA1 significantly enhances the conductance of AMPA receptor channels and promotes their trafficking and insertion into the postsynaptic membrane, ultimately intensifying excitatory postsynaptic responses.^28–30 This sustained enhancement of excitatory drive depolarizes the resting membrane potential of neurons, effectively lowering the threshold for action potential generation and culminating in a pathological increase in overall neuronal firing frequency.³¹ Western blotting showed that acupotomy reduced p-CaMKII and p-GluA1 levels in the spinal cord of KOA rats, consistent with this pathway. We emphasize, however, that we did not measure AMPA receptor surface expression or trafficking, record excitatory postsynaptic currents, or perform loss-of-function experiments such as pharmacological CaMKII inhibition or spinal knockdown. The link between CaMKII activation, AMPA-receptor-mediated transmission, and firing should therefore be read as a hypothesis our data are compatible with, not one they demonstrate.

The superficial dorsal horn of the spinal cord, particularly the substantia gelatinosa (SG, lamina II), serves as a crucial relay station for integrating afferent signals from peripheral unmyelinated C fibers and thinly myelinated Aδ fibers. It plays a pivotal role in the primary processing, filtering, and central sensitization of nociceptive information.^32–35 Consequently, both the histomorphological and cellular functional assessments in our study were precisely targeted at this region. Double immunofluorescence staining revealed that KOA induced robust co-expression of c-Fos, a marker of recent neuronal activation, within p-CaMKII-positive neurons in the SG region. Acupotomy intervention significantly reduced this co-expression, consistent with an inhibitory effect on the hyperactive state of nociceptive neurons. As an immediate early gene, c-fos is rapidly and transiently expressed in spinal dorsal horn neurons following peripheral noxious stimulation.³⁶ Its protein product, c-Fos, has long been used as a surrogate marker for tracking nociceptive pathway activation. In chronic pain conditions like KOA, sustained afferent barrage is associated with persistent accumulation of c-Fos in the SG region, reflecting pathological hyperexcitability of the local pain circuitry. We note that c-Fos expression, pain-related behaviors, and phosphorylation markers each provide partial, indirect evidence rather than direct proof of central sensitization as a discrete physiological state.

To further evaluate neuronal functional output, we employed the whole-cell patch-clamp technique. Electrophysiological recordings demonstrated that, in response to incremental depolarizing current steps, SG neurons in the KOA model group exhibited a significantly increased AP firing frequency, indicating pathologically enhanced intrinsic excitability. In the context of chronic pain conditions such as KOA, sustained nociceptive input from the degenerating joint drives maladaptive neuroplasticity in the spinal dorsal horn.³⁷ Because AP firing frequency serves as a fundamental neural code for stimulus intensity, this shift toward high-frequency bursting may contribute to amplified transmission of ascending pain signals.³⁸ Conversely, acupotomy effectively suppressed this high-frequency firing. While our current-clamp recordings directly demonstrated attenuation of intrinsic neuronal excitability, we did not directly examine AMPA-mediated synaptic currents, and pathway-specific causality was not established. We therefore hypothesize, without direct evidence in this study, that downregulation of the CaMKII/GluA1 signaling axis may contribute to the reduced excitatory postsynaptic drive underlying this attenuated firing.^39,40 Collectively, our findings indicate that acupotomy reduces the intrinsic excitability of primary nociceptive neurons in the spinal SG region, a result compatible with reduced central nociceptive transmission, though the underlying causal pathway remains to be directly tested.

However, the present study has several limitations that must be acknowledged. First, mechanistic validation is incomplete. Establishing a causal role for CaMKII/GluA1 signaling will require specific inhibitors or viral knockdown, together with voltage-clamp recordings and surface-protein fractionation to map AMPA receptor trafficking directly.

Second, we emphasize that the behavioral, immunofluorescence, phosphorylation, and electrophysiological measures used in this study each capture only a partial aspect of nociceptive processing. Central sensitization is a complex, multidimensional functional construct, and our findings should be interpreted as evidence of reduced central nociceptive excitability.

Third, because acupotomy profoundly improved peripheral joint swelling and histopathology, the reduced spinal activation and excitability observed here may largely be secondary to a reduction in peripheral nociceptive input, rather than representing a direct neuromodulatory effect on the central nervous system. To definitively differentiate whether acupotomy passively reduces central sensitization by merely eliminating peripheral inflammatory input or actively modulates spinal neuroplasticity, future studies should evaluate contralateral or secondary hyperalgesia. Furthermore, whether acupotomy exerts further modulatory effects on higher-order pain centers, such as the anterior cingulate cortex (ACC) or periaqueductal gray (PAG), via ascending pathways remains to be elucidated.

Fourth, the current study lacks a sham-acupotomy control group. Although the Sham and KOA groups underwent identical anesthesia and handling to control for procedural stress, we cannot completely rule out the non-specific physiological effects induced by local tissue micro-injury. Given that our primary aim was to elucidate the downstream molecular mechanisms of standard clinical acupotomy, the classical intervention model was utilized. Future studies incorporating a standardized sham-acupotomy group are warranted to definitively distinguish specific biomechanical release effects from non-specific needle penetration responses.

Fifth, this study exclusively utilized male rats. While this approach was deliberately chosen to minimize baseline variability associated with the estrous cycle and cyclic estrogen fluctuations during initial transcriptomic and electrophysiological profiling, we acknowledge that it restricts the generalizability of our findings. Given the well-recognized sex-related dimorphism in osteoarthritis pain mechanisms, validating the specific molecular targets of acupotomy in female KOA models represents a critical avenue for future research.

Finally, translating short-term preclinical findings into durable clinical benefits requires caution. A recent clinical study reported that intra-articular carboxymethylchitosan produced early improvements in pain and function in patients with advanced knee osteoarthritis, but this benefit attenuated at later follow-up.⁴¹ Although the intervention type and study design differ substantially from the present preclinical model, this evidence underscores the general principle that short-term analgesic gains in KOA interventions do not necessarily predict durable clinical benefit, highlighting the necessity of extended post-treatment follow-up in future studies of acupotomy.

Conclusion

Taken together, these findings indicate that acupotomy reduces pain-related behaviors and attenuates markers of nociceptive neuron hyperexcitability in the spinal SG region of KOA rats, results compatible with, though not sufficient to fully establish, a reduction in central sensitization. Downregulation of the spinal CaMKII/GluA1 axis offers a plausible molecular correlate of these changes, but the proposed causal sequence from CaMKII activation to AMPA-mediated transmission and AP firing requires direct validation, and the present design cannot distinguish a direct central effect from one secondary to reduced peripheral input. This work therefore offers a correlative framework, rather than a demonstrated causal mechanism, for the central effects of acupotomy in KOA.

Funding Statement

This work was supported by the National Key R&D Program of China (Grant No. 2023YFC3502700), the Key Project of the Hubei Provincial Natural Science Foundation Innovation and Development Joint Fund (Grant No. 2024AFD271), the Hubei Provincial Natural Science Foundation Innovation and Development Joint Fund (Grant No. 2023AFD159), and 2026 Beijing University of Chinese Medicine Graduate Independent Research Project (Grant NO. ZJKT2026024).

Data Sharing Statement

Data and materials will be made available by the corresponding author, Zhigang Li, upon reasonable request. The raw genome sequencing data generated in this study have been deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1454231.

Ethics Approval

All experimental procedures were conducted in accordance with the National Institutes of Health (NIH) guidelines and approved by the Experimental Animal Ethics Committee of Beijing University of Chinese Medicine (Approval No. BUCM2024091102-3197). Consent to Participate: Not applicable.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare no competing interests.