Bone marrow mesenchymal stem cells attenuate pain and modulate peripheral sodium channel activity in a rat model of complex regional pain syndrome type I

Abstract

Complex Regional Pain Syndrome Type I (CRPS-I) is a chronic neuropathic pain disorder characterized by peripheral nerve hyperexcitability and altered nociceptive signaling. Voltage-gated sodium channels (Nav1.7, Nav1.8, Nav1.9) in dorsal root ganglia (DRG) are key contributors to pain hypersensitivity. This study investigated the analgesic effects and underlying mechanisms of bone marrow mesenchymal stem cell (BMSC) transplantation in a CRPS-I rat model. The model was induced by hind limb ischemia-reperfusion, followed by intrathecal administration of BMSCs. Pain behaviors were assessed using thermal withdrawal latency (TWL), mechanical withdrawal latency (MWL), spontaneous pain scoring, and acetone-evoked cold allodynia. RT-PCR and Western blot analysis were used to evaluate Nav channel expression in DRG tissue, while electrophysiological properties were examined using whole-cell patch clamp to generate current-voltage (I–V) curves. CRPS-I rats exhibited decreased TWL and MWL, elevated expression of Nav1.7, Nav1.8, and Nav1.9, and enhanced sodium current density with delayed inactivation. BMSC transplantation significantly alleviated pain behaviors, downregulated sodium channel expression, and normalized I–V characteristics—marked by increased activation thresholds, reduced peak currents, and faster inactivation kinetics. These findings suggest that BMSCs mitigate neuronal hyperexcitability by modulating peripheral Nav channel activity. This study provides mechanistic evidence supporting the therapeutic potential of BMSC-based interventions for CRPS-I and related neuropathic pain conditions.

Graphical Abstract

Analgesic mechanisms of BMSC transplantation in the CRPS-I rat model and regulation of sodium channel expression.

Introduction

Complex regional pain syndrome (CRPS) is a chronic pain condition that severely impairs quality of life ¹ . It often occurs secondary to localized trauma or systemic disease and is characterized by autonomic dysfunction, trophic changes, microvascular disturbances, and motor impairment^2,3. CRPS is classified into CRPS-I (reflex sympathetic dystrophy) and CRPS-II (causalgia), depending on whether a confirmed nerve injury is present. CRPS is typically accompanied by allodynia and spontaneous pain^4,5, and in severe cases, can lead to limb deformity, dysfunction, and a high disability rate ⁶ . Due to the lack of effective therapies, CRPS-associated neuropathic pain remains a major clinical challenge ² .

In recent years, bone marrow mesenchymal stem cells (BMSCs) have shown promising analgesic effects in various models of chronic pain, including CRPS. For example, in a mouse model of CRPS-I induced by chronic post-ischemic pain (CPIP), the injection of intrathecal and plantar human BMSCs significantly alleviated mechanical allodynia ⁷ . Clinical studies have also reported that intravenous infusion of BMSC-derived exosomes (ExoFlo) in CRPS patients reduced pain scores, improved joint mobility, and did not cause serious adverse events ⁸ . These findings suggest that BMSCs hold therapeutic potential for CRPS. However, studies focusing on the effects of BMSCs in rat models of CRPS-I remain limited. Therefore, this study aims to evaluate the therapeutic effects of BMSCs in a CRPS-I rat model and investigate their influence on pain behavior and inflammatory responses to assess their feasibility as a potential treatment strategy for CRPS.

The exact pathophysiology of CRPS remains unclear, but increasing evidence suggests that ion channel dysfunction, particularly involving sodium, potassium, and calcium channels, plays a critical role in the development of neuropathic pain (NP). Among these, voltage-gated sodium channels (VGSCs) are essential for generating and conducting action potentials in excitable cells such as neurons and skeletal muscle fibers ⁹ . VGSCs consist of an α-subunit and several auxiliary β-subunits^10,11. Abnormalities in VGSC function can lead to neuronal hyperexcitability, a common mechanism underlying NP ¹² . The VGSC subtypes Nav1.7, Nav1.8, and Nav1.9 are highly expressed in the dorsal root ganglia (DRG) and implicated in various NP conditions ¹³ .

Cell-based therapy is emerging as a promising approach for treating neurological disorders. BMSCs, due to their multilineage differentiation potential and immunomodulatory properties, have attracted attention as a potential treatment for NP and other central and peripheral nervous system disorders^14,15. However, their role in modulating VGSC expression in CRPS-I remains poorly understood. In this study, we investigated the analgesic effects of intrathecal BMSC transplantation in a CRPS-I rat model and evaluated changes in the mRNA and protein expression of Nav1.7, Nav1.8, and Nav1.9 in DRG tissue. These findings may provide mechanistic insights into the therapeutic potential of BMSCs in CRPS-I and support the development of stem cell-based strategies for treating neuropathic pain. Clinically, BMSC transplantation holds promise as an effective intervention for CRPS-I and other chronic pain conditions.

Materials and methods

Establishment of the CRPS-I rat model

The Animal Ethics Committee of The Beijing Institute of Biotechnology, Beijing, China (No. AMMS-06-2016-003) approved all animal experiments.

As previously described, a chronic post-ischemia pain model was established to mimic CRPS-I by prolonged ischemia-reperfusion of the right hind limb ¹⁶ . Male Sprague-Dawley rats (250 ± 20 g) were obtained from the Experimental Animal Center of the PLA General Hospital. Animals were fasted for 12 h with free access to water before the experiment. Under anesthesia with 3% pentobarbital sodium (50 mg/kg, i.p.), a 3 ml syringe barrel with both ends removed was fitted around the right hind limb. A rubber O-ring (inner diameter: 5.0 mm) was placed 1.5 cm above the ankle to induce ischemia. After 3 h, the ring was removed to allow reperfusion. Rats recovered spontaneously from anesthesia.

Following reperfusion, the affected limbs exhibited erythema, edema, and plasma extravasation, which lasted approximately 2–4 h. To assess model validity, acute pain responses were measured 8 h post-reperfusion to evaluate early nociceptive changes due to inflammation ¹⁶ . Given the distinct mechanisms of acute and chronic pain ¹⁷ , behavioral assessments for chronic pain hypersensitivity were conducted on postoperative day 7 ¹ . Mechanical allodynia was assessed using von Frey filaments (Stoelting, Chicago, IL, USA) with an ascending stimulus paradigm ¹⁸ . Eight filaments (0.4, 0.6, 1.0, 2.0, 4.0, 6.0, 8.0, and 15.0 g) were applied to the plantar surface of the hind paw through a wire mesh floor (mesh size: 10 × 10 mm) after 30 min of acclimatization. Each stimulus was applied perpendicularly for 2–3 s. Paw withdrawal, shaking, or licking was recorded as a positive response. Each filament was tested five times, and the 50% withdrawal threshold was calculated using Dixon’s up-down method. Cold allodynia was evaluated using a blunt metal probe (diameter: 1.0 mm) gently applied to the plantar surface for 2–3 s without penetrating the skin. Although this stimulus typically does not elicit acute mechanical pain, it may induce cold or mechanical hypersensitivity in the context of CRPS-I due to abnormal activation of C and Aδ fibers. Spontaneous pain behaviors, such as limb shaking, licking, and reduced weight-bearing, were also observed, indicating the successful establishment of the model. The defined time points for acute and chronic pain assessments ensured accurate characterization of pain states and were consistent with previous studies^1,16,17.

Isolation and characterization of primary rat BMSCs

Bone marrow-derived mesenchymal stem cells (BMSCs) were isolated from 4- to 5-week-old male Sprague-Dawley rats (150–200 g; Laboratory Animal Center, Chinese PLA General Hospital) using the whole bone marrow adherence method. Under sterile conditions, rats were euthanized by cervical dislocation, and femurs and tibias were harvested. Bone marrow was flushed from the cavities using sterile PBS (pH 7.4) containing 1% penicillin/streptomycin with a 1 ml syringe and 18G needle. The cell suspension was centrifuged at 1500 rpm for 5 min, and the pellet was resuspended in low-glucose DMEM (LG-DMEM) at 1–3 × 10⁶ cells/ml. Cells were seeded into T-25 flasks and cultured at 37°C with 5% CO₂. Non-adherent cells were removed after 48 h, and the medium was changed every 3 days. Upon reaching 80% confluence, cells were passaged at a 1:3 ratio following digestion with 0.25% trypsin-EDTA.

Third-passage BMSCs were washed and resuspended in PBS containing 0.5% bovine serum albumin (BSA). Surface marker expression was analyzed by flow cytometry using FITC- or PE-conjugated antibodies against CD105, CD90, and CD44 (positive markers), and CD45 and CD11b (negative markers), following the manufacturer’s protocols. Cells were incubated for 30 min at room temperature in the dark and washed before analysis. To assess multipotency, adipogenic and osteogenic differentiation were induced. For adipogenesis, BMSCs were cultured in high-glucose DMEM supplemented with 10% FBS, 0.5 mM isobutylmethylxanthine (IBMX), 1 µM dexamethasone, 10 µg/ml insulin, 100 µM indomethacin, and 1% penicillin/streptomycin for 21 days. Lipid droplet formation was visualized by Oil Red O staining. For osteogenesis, cells were cultured in high-glucose DMEM containing 10% FBS, 0.05 µM ascorbic acid-2-phosphate, 100 nM dexamethasone, 10 mM β-glycerophosphate, and 1% penicillin/streptomycin. After 21 days, calcium deposition was evaluated using Alizarin Red S staining. All media components (LG-DMEM, trypsin, FBS, antibiotics) were obtained from Gibco (New York, USA). Antibodies for flow cytometry included FITC-CD44, PE-CD45, PE-CD90, PE-CD11b (eBioscience, CA, USA), and FITC-CD105 (Bio-Rad, CA, USA).

BMSC transplantation

After weighing, the rats were anesthetized with an intraperitoneal injection of 3% pentobarbital sodium (50 mg/kg) and positioned in a prone posture with a cushion under the abdomen to arch the lower back. The L4/L5 region was shaved and sterilized. A microsyringe needle was inserted at a 30° angle and advanced horizontally along the interspinous ligament to puncture the dura mater and access the subarachnoid space. A total of 20 µL of cell suspension (1 × 10⁷ cells/ml) was injected. The experimental rats were randomly divided into four groups: Blank group (wild-type; WT rats with no intervention), Model group (CRPS-I rats with no intervention), Treatment group (CRPS-I rats with intrathecal BMSC transplantation), and Control group (WT rats with intrathecal injection of cell culture medium). Tissues were harvested on day 14 after BMSC transplantation, coinciding with the final behavioral assessment to align molecular and behavioral outcomes.

Behavioral assessment

To evaluate the analgesic effects of bone marrow mesenchymal stem cell (BMSC) transplantation in a CRPS-I rat model, four behavioral tests were conducted: thermal withdrawal latency (TWL), mechanical withdrawal latency (MWL), spontaneous pain scoring, and cold allodynia testing (acetone evaporation method). For TWL measurement, rats were placed on a transparent glass platform maintained at 37°C and allowed to acclimate for 30 min to ensure a relaxed state. A plantar thermal stimulator was then used to apply a focused heat stimulus to the hind paw. Before testing, the focal distance and light intensity were adjusted to ensure consistent stimulation at the same plantar region. TWL was recorded as the time from stimulus onset to paw withdrawal. Each rat underwent five trials per day with at least 10-min intervals between tests, and the average was calculated as the daily TWL.

For MWL measurement, rats were placed on a preheated metal mesh surface and acclimated for 30 min. Mechanical stimuli were applied to the plantar surface using von Frey filaments with gradually increasing force until a withdrawal response was elicited. A response was considered positive if paw withdrawal occurred during or immediately after filament removal. Each rat was tested five times daily, with a minimum 10-min interval between trials. The average of the five measurements was recorded as the daily MWL.

Spontaneous pain behavior was assessed on postoperative day 14 using video recording. Rats were individually placed in transparent observation chambers and allowed to acclimate for 15 min during a quiet period. Subsequently, spontaneous behaviors were recorded for 10 min. Observed pain-related behaviors included limping, paw lifting, paw licking, and reduced weight-bearing on the affected limb. Pain severity was scored on a 0–3 scale: 0 = normal behavior; 1 = mild pain (e.g. occasional paw lifting); 2 = moderate pain (e.g. sustained licking or weight avoidance); 3 = severe pain (e.g. frequent licking, prolonged lifting, or complete avoidance of weight-bearing) ¹⁶ .

Cold allodynia was evaluated using the acetone evaporation method on postoperative day 14. A 20 μl droplet of room-temperature acetone was gently applied to the mid-plantar surface of the affected hind paw using a micropipette, avoiding direct skin contact. Behavioral responses within 30 s were recorded and scored as follows: 0 = no response; 1 = slight paw withdrawal; 2 = obvious withdrawal or licking; 3 = intense response such as sustained lifting, rapid shaking, or vigorous licking ¹⁹ .

RT-PCR

DRG tissues from rats were collected and lysed in TRIzol reagent to extract total RNA. The integrity of the RNA was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, California, USA). RNA concentration and purity were measured using a NanoDrop® ND-1000 spectrophotometer (NanoDrop, USA). For reverse transcription, the reaction mixture included total RNA (40 μg), 10× DNase I buffer (5 μl), RNase inhibitor (20 U), DNase I (RNase-free, 10 U/2 μl), and RNase-free dH₂O, with a total volume of 50 μl. Reverse transcription was performed using the TIANScript RT Kit (TIANGEN, Beijing, China) at 42°C for 50 min, followed by inactivation of reverse transcriptase at 95°C for 5 min. The RT-PCR reaction mixture (20 μl) included 1 μl of reverse transcription product, 1 μl each of 10 μM forward and reverse primers, 10 μl of 2× master mix, and 7 μl of nuclease-free water. The RT-PCR was conducted under the following conditions: initial denaturation at 96°C for 4 min, followed by 40 cycles of 94°C for 30 s, 58°C for 30 s, and 72°C for 30 s. Fluorescence signals were analyzed using the 2^−ΔΔCT method, with β-actin as the internal control. The primer sequences are listed in Table S1. TRIzol was purchased from Invitrogen (California, USA), the TIANScript RT Kit from TIANGEN (Beijing, China), and the HotStart Fluorescent PCR Core Reagent Kit (SYBR Green I) from BBI (Italy). The RT-PCR instrument (Prism® 7300) was obtained from ABI (USA).

Western blot

DRG tissues were collected from rats and lysed using RIPA buffer (IBL, USA) to extract total protein. Protein concentrations were determined using a BCA assay kit (ThermoFisher, China). Equal amounts of protein were mixed with loading buffer and denatured by boiling for 5 min. Samples were then separated on 10% SDS-PAGE gels (electrophoresis buffer from BOSTER, Wuhan, China) and transferred onto PVDF membranes using a wet transfer system. Membranes were blocked with 5% non-fat milk in TBST at room temperature for 1 h and incubated overnight at 4°C with primary antibodies against Nav1.7, Nav1.8, and Nav1.9 (rabbit polyclonal; ThermoFisher, China). After washing with PBS or TBST, membranes were incubated with HRP-conjugated goat anti-rabbit IgG secondary antibody for 1 h at room temperature. Following additional washes, protein bands were visualized using an ECL detection kit (ThermoFisher, China). Band intensities were analyzed using Image Studio software, and the relative expression levels were calculated as the ratio of the target protein to the internal control (β-actin or GAPDH).

DRG neuron electrophysiology

Male CRPS-I model SD rats (250 ± 20 g) were used. L4-L6 DRGs were dissected under sterile conditions and connective tissue and nerve roots were removed. The tissue was digested with 0.25% trypsin/EDTA (ThermoFisher Scientific, 25200056, USA) at 37°C for 15–20 min, and the reaction was terminated with DMEM (Gibco, 11995065, USA) containing 10% FBS (ThermoFisher Scientific, A5670701, USA). Single-cell suspensions were obtained by gentle trituration and filtration through a 70 µm cell strainer, followed by centrifugation at 1000 rpm for 5 min. The pellet was resuspended in DMEM supplemented with 10% FBS and 1% penicillin/streptomycin (ThermoFisher Scientific, 15140122, USA) ²⁰ . Cells were plated on poly-l-lysine-coated coverslips (0.1 mg/ml; Sigma, P3150-10MG, USA) and maintained at 37°C in 5% CO₂. The culture medium consisted of Neurobasal-A (Gibco, 10888-022, USA) supplemented with 2% B27 (Gibco, 17504-044, USA), 10 ng/ml NGF (CUSABIO, CSB-EP016120HU, China), and 1% penicillin-streptomycin. Whole-cell patch-clamp recordings were performed after 2 h of culture, targeting medium-sized DRG neurons with diameters of 25–40 µm. Previous studies have shown that Nav1.7, Nav1.8, and Nav1.9 are the predominant sodium channel subtypes in small to medium-diameter C-type neurons ²¹ , while Nav1.8 and Nav1.9 are also expressed in subsets of medium to large neurons ²² . By contrast, Nav1.7 expression is relatively limited in medium-diameter A-fiber neurons (~15%) ²³ .

Electrophysiological

Whole-cell patch-clamp recordings were performed on DRG neurons using an EPC-10 USB amplifier (HEKA Elektronik, Lambrecht, Germany) controlled by PatchMaster software at room temperature (25°C ± 2°C). Recording pipettes were pulled from borosilicate glass capillaries (1.5 mm) using a P-97 puller (Sutter, Novato, CA), fire-polished, and had a resistance of 2.0–2.5 MΩ. All salts were purchased from Sigma. The internal solution (mmol/L) contained 140 CsCl, 2 MgCl₂, 10 HEPES, 10 EGTA, and 2 Na₂ATP (pH adjusted to 7.2 with KOH), and the external solution (mmol/L) contained 140 NaCl, 5 KCl, 2 CaCl₂, 2 MgCl₂, 10 HEPES, 1 CdCl₂, 10 TEA-Cl, and 10 glucose (pH adjusted to 7.4 with NaOH). A continuous perfusion system (1–2 ml/min) was used to maintain stable extracellular conditions. To pharmacologically isolate sodium channel subtypes, tetrodotoxin (TTX, 1 μmol/L; MCE, HY-P5868, USA) was applied to block TTX-sensitive currents (Nav1.7), and A-803467 (100 nmol/L; MCE, HY-11079, USA) was used to selectively inhibit Nav1.8 ²⁴ . As no selective blocker is available for Nav1.9, only Nav1.7- and Nav1.8-mediated currents were analyzed. After achieving a gigaohm seal, whole-cell configuration was obtained by gentle suction. Cells were held at −80 mV, and depolarizing steps were applied from −80 to +60 mV in 10 mV increments (200 ms duration) to evoke inward sodium currents. Capacitive transients were canceled, and series resistance was compensated by 80% to minimize voltage errors; liquid junction potentials were corrected prior to seal formation. Currents were sampled at 30 kHz and filtered at 2.9 kHz ²⁵ . For each group, 15 DRG neurons were recorded randomly, and mean values were used to construct current-voltage (I-V) relationships and to analyze activation threshold (defined as the voltage at 10% of peak current), peak current density (pA/pF), and inactivation time constant (τ) ²⁵ .

Data were low-pass filtered at 1 kHz and analyzed using pCLAMP 12.0 (Molecular Devices). I-V relationships were normalized and fitted using a Boltzmann function: I/Imax=1/(1+e(V1/2−V)/k) where V1/2 is the half-activation voltage and k is the slope factor. Inactivation kinetics were fitted with a single-exponential function: I(t)=Ipeak.e−t/τinactivation+Isteady.

Statistical analysis

All data are expressed as mean ± standard deviation (SD). Normality was assessed using the Shapiro–Wilk test, and Levene’s test was used to evaluate homogeneity of variance. Two-group comparisons were performed using independent samples t-tests. For multiple group comparisons, one-way ANOVA was used, followed by Tukey’s post hoc test when significant differences were detected (P < 0.05). The Kruskal–Wallis test was applied for non-normally distributed data, followed by Dunn’s multiple comparison test. Correlation analysis was performed using the Pearson or Spearman methods, depending on the data distribution. Statistical analyses were conducted using GraphPad Prism 9.0 (GraphPad Software, San Diego, CA, USA) or SPSS 19.0 (IBM, USA), and a P-value < 0.05 was considered statistically significant.

Results

Results of isolation and identification of primary rat BMSCs

Primary rat bone marrow-derived mesenchymal stem cells (BMSCs) were isolated and cultured using the whole bone marrow adherence method. Under light microscopy, the cells exhibited a typical spindle-shaped or fibroblast-like morphology and proliferated gradually over time. A confluent monolayer was observed when the cell density reached approximately 80% (Fig. 1a). Third-passage BMSCs were used for surface marker analysis. Flow cytometry revealed that the positive expression rates of CD105, CD90, and CD44 were 98.01%, 99.93%, and 87.10%, respectively (Fig. 1b, c, and d). The hematopoietic and immune cell markers CD45 and CD11b showed low expression levels, with positive rates of 1.46% and 0.28%, respectively (Fig. 1e and f). To assess multipotent differentiation potential, adipogenic and osteogenic induction were performed. After 2–3 weeks of adipogenic induction, Oil Red O staining revealed intracellular lipid droplet formation (Fig. 1g). Following 3 weeks of osteogenic induction, Alizarin Red S staining demonstrated calcium nodule deposition (Fig. 1h).

Figure 1.

Morphological observation, surface marker characterization, and multipotent differentiation potential of primary rat BMSCs. (a) Morphology of primary rat BMSCs observed under an inverted microscope. (b) Flow cytometry analysis of the surface marker CD105 in third-generation BMSCs using FITC-conjugated anti-CD105 antibody. (c) Flow cytometry analysis of the surface marker CD90 in third-generation BMSCs using PE-conjugated anti-CD90 antibody. (d) Flow cytometry analysis of the surface marker CD44 in third-generation BMSCs using FITC-conjugated anti-CD44 antibody. (e) Flow cytometry analysis of the surface marker CD45 in third-generation BMSCs using PE-conjugated anti-CD45 antibody. (f) Flow cytometry analysis of the surface marker CD11b in third-generation BMSCs using PE-conjugated anti-CD11b antibody. (g) Oil Red O staining for adipogenic differentiation of BMSCs, showing lipid droplet formation. (h) Alizarin Red staining for osteogenic differentiation of BMSCs, showing calcium nodule formation (n = 3).

Establishment of the CRPS-I rat model

MWL and TWL were measured to verify the successful induction of the CRPS-I model. One-way ANOVA showed significant differences among time points for both MWL (F(3,56) = 161.23, P < 0.0001) and TWL (F(3,56) = 109.78, P < 0.0001). Compared with baseline (MWL: 11.87 ± 1.31 s; TWL: 3.29 ± 0.36 s), the model group exhibited significantly reduced latencies on postoperative days 7, 10, and 14 (MWL: 2.26 ± 0.30 s, 2.73 v 0.35 s, 2.86 ± 0.36 s; TWL: 0.95 ± 0.34 s, 1.06 ± 0.32 s, 1.10 ± 0.31 s; all P < 0.0001 vs baseline and blank). Tukey’s post hoc analysis confirmed significant reductions at all time points (Fig. 2a, b). These results indicate the successful induction of mechanical and thermal hyperalgesia, validating the CRPS-I rat model.

Figure 2.

Assessment of MWL and TWL in CRPS-I rat models. (a) MWL measurements of rats in different experimental groups at preoperative and on postoperative days 7, 10, and 14. Each group included n = 15. Data are presented as Mean ± SD. ****P < 0.0001 compared to the blank group (b) TWL measurements of rats in different experimental groups at preoperative and postoperative days 7, 10, and 14. Each group included n = 15. Data are presented as Mean ± SD. ****P < 0.0001 compared to the blank group.

Effect of BMSC transplantation on pain behavior in CRPS-I rats

BMSC-induced analgesia was assessed in CRPS-I rats through TWL, MWL, spontaneous pain scoring, and cold allodynia testing. One-way ANOVA showed a significant group effect for TWL on postoperative day 7 (F(3,68) = 5.29, P = 0.0024) while no significant differences were observed on days 10 and 14 (Fig. 3a). In contrast, MWL showed highly significant group differences at all time points (day 7: F(3,65) = 132.66; day 10: F(3,65) = 120.39; day 14: F(3,65) = 109.59; all P < 0.0001; Fig. 3a).

Figure 3.

Effects of BMSC transplantation on TWL and MWL in CRPS-I rat models. (a) Comparison of TWL (in seconds) among different groups of rats at specified time points (Mean ± SD, n = 15). ****P < 0.0001 compared to the blank group; ####P < 0.0001 compared to the model group. (b) Comparison of MWL (in seconds) among different groups of rats at specified time points (Mean ± SD, n = 15). ****P < 0.0001 compared to the blank group; ###P < 0.001, ####P < 0.0001 compared to the model group. (c) Spontaneous pain behavior scores of rats on day 14. (Mean ± SD, n = 15). ****P < 0.0001 compared to the blank group; ##P < 0.0001 compared to the model group. (d) Cold allodynia scores measured by acetone test. ****P < 0.0001 compared to the blank group; ##P < 0.0001 compared to the model group.

Tukey’s post hoc test revealed that TWL and MWL in the model group were significantly reduced compared to baseline and the blank group, indicating the successful induction of thermal and mechanical hyperalgesia (TWL: 0.95 ± 0.10, 1.07 ± 0.13, 1.13 ± 0.09 s; MWL: 2.26 ± 0.12, 2.73 ± 0.19, 2.86 ± 0.12 s; all P < 0.05). In the BMSC-treated group, both TWL and MWL were significantly improved at each time point (TWL: 1.27 ± 0.17, 1.49 ± 0.19, 1.52 ± 0.21 s; MWL: 3.54 ± 0.34, 5.26 ± 0.55, 6.44 ± 0.55 s; all P < 0.05). No significant changes were observed in the control group, and its values were not significantly different from the model group (P > 0.05), confirming that the analgesic effects were specific to BMSC treatment.

During the spontaneous pain behavior assessment conducted during a quiet period, rats in the Model group exhibited pronounced pain-related behaviors such as paw lifting, licking, and reduced weight-bearing, with a mean score of 2.60 ± 0.34. This was significantly higher than that of the Blank group (0.23 ± 0.15, P < 0.0001). The Control group showed a similar score to the Model group (2.51 ± 0.30, P > 0.05), while the BMSC-treated group exhibited a significantly lower score (1.04 ± 0.32, P < 0.01), indicating that BMSC transplantation effectively alleviated spontaneous pain behavior (Fig. 3c).

In the cold allodynia test, rats in the Model group displayed intense responses such as strong paw withdrawal and persistent licking, with a mean acetone response score of 2.82 ± 0.28, which was significantly higher than that of the Blank group (0.41 ± 0.22, P < 0.0001). The Control group (2.75 ± 0.31) showed no significant difference from the Model group (P > 0.05). In contrast, the Treatment group showed a significantly reduced response score (1.18 ± 0.35, P < 0.01), suggesting that BMSC administration effectively attenuated cold hypersensitivity (Fig. 3d).

Regulatory effects of BMSC transplantation on Nav1.7, Nav1.8, and Nav1.9 sodium channel mRNA expression

RT-PCR analysis revealed that mRNA expression levels of Nav1.7, Nav1.8, and Nav1.9 in the DRG of CRPS-I model rats were significantly higher than those in the blank group (P < 0.05, Fig. 4). One-way ANOVA further confirmed significant differences among groups for each sodium channel: Nav1.7 (F(3,8) = 74.14, P < 0.0001), Nav1.8 (F(3,8) = 49.80, P < 0.0001), and Nav1.9 (F(3,8) = 32.67, P < 0.0001) supporting a pronounced upregulation of these channels in the CRPS-I pathological state. Following BMSC transplantation, the expression levels of Nav1.7, Nav1.8, and Nav1.9 were significantly reduced compared to the model group (P < 0.05), suggesting that BMSCs may exert analgesic effects by downregulating sodium channel expression and thereby reducing neuronal hyperexcitability. However, the mRNA levels did not fully return to baseline, indicating that further studies are needed to determine whether such molecular changes are sufficient to translate into behavioral analgesia. Additionally, there were no significant differences between the control and model groups (P > 0.05), indicating that the observed changes in gene expression were specific to the CRPS-I model and not attributable to procedural artifacts.

Figure 4.

Effects of BMSC transplantation on Nav1.7, Nav1.8, and Nav1.9 mRNA levels in rat DRG tissue.

The regulatory effects of BMSC transplantation on the sodium channel proteins Nav1.7, Nav1.8, and Nav1.9

Western blot analysis showed that the protein expression levels of Nav1.7, Nav1.8, and Nav1.9 in the DRG of CRPS-I model rats were significantly higher than those in the blank group (Fig. 5a and b). One-way ANOVA confirmed significant differences among the groups for all three channels: Nav1.7 (F(3,8) = 63.57, P < 0.0001) Nav1.8 (F(3,8) = 47.59, P < 0.0001), and Nav1.9 (F(3,8) = 19.20, P = 0.001). These protein-level changes were consistent with the mRNA expression trends, suggesting that the upregulation of sodium channels may contribute to neuronal hyperexcitability and pain sensitization in CRPS-I. After BMSC transplantation, the expression levels of Nav1.7, Nav1.8, and Nav1.9 were significantly reduced compared to the model group and approached the levels observed in the blank group. It indicates that BMSCs may exert analgesic effects by modulating Nav channel expression and reducing sustained neuronal excitability. No significant differences were observed between the control and model groups (P > 0.05), further supporting that the changes in protein expression were attributable to CRPS-I pathology and BMSC treatment rather than procedural artifacts.

Figure 5.

Effects of BMSC transplantation on Nav1.7, Nav1.8, and Nav1.9 protein levels in rat DRG tissue. (a) The expression levels of Nav1.7, Nav1.8, and Nav1.9 proteins in the blank group, model group, control group, and treatment group; (b) Statistical analysis of the relative expression levels of Nav1.7, Nav1.8, and Nav1.9 proteins in the blank group, model group, control group, and treatment group. (c) Current-voltage (I-V) relationships recorded from DRG neurons in the four groups.

Additionally, sodium I-V relationships were recorded from DRG neurons in all four groups. In the Model group, upregulation of Nav1.7, Nav1.8, and Nav1.9 was associated with increased current density, a leftward shift in the I-V curve, enhanced peak currents, and delayed inactivation, indicating heightened neuronal excitability. The Control group showed similar IV features to the Model group. In contrast, BMSC-treated rats exhibited reduced Nav channel expression, a rightward shift of the IV curve, decreased peak current density, and accelerated inactivation, approaching the profile of the Blank group, suggesting attenuation of DRG hyperexcitability (Fig. 5c).

Discussion

These findings demonstrated that CRPS-I model rats exhibited significantly decreased thermal and mechanical thresholds, as reflected by reduced TWL and MWL values, along with increased spontaneous pain behavior scores and cold allodynia responses, indicating the establishment of a multidimensional pain phenotype. Following intrathecal transplantation of BMSCs, all four behavioral parameters showed marked improvement: TWL and MWL increased, spontaneous pain behaviors were attenuated, and cold hypersensitivity was reduced. Moreover, mRNA and protein levels of Nav1.7, Nav1.8, and Nav1.9 were significantly upregulated in the DRGs of CRPS-I rats, suggesting enhanced neuronal excitability. BMSC transplantation reversed this upregulation. Electrophysiological recordings further revealed that BMSC-treated rats exhibited I-V curve profiles resembling those of the Blank group, characterized by elevated activation thresholds, reduced peak currents, and accelerated inactivation kinetics. These findings indicate that BMSCs alleviate pain partly by modulating Nav channel expression and suppressing DRG hyperexcitability. The significance of this study demonstrates that BMSCs can modulate the expression of VGSC subtypes (Nav1.7, Nav1.8, and Nav1.9), thereby alleviating pain-related behaviors in CRPS-I rats, offering a novel mechanistic perspective and strategy for cell-based therapies in chronic pain conditions.

Although TWL and MWL were significantly improved in the treatment group compared with the model group, they did not fully return to baseline levels, suggesting that the analgesic effect of BMSCs may be limited. This modest behavioral improvement may be attributed to the complexity of CRPS-I pathophysiology, the intervention protocol (e.g. single transplantation), and the relatively short observation period. Additionally, no comparable analgesic effect was observed in the vehicle control group, indicating that the observed effects were specific to BMSC treatment rather than the injection procedure^26,27.

At the molecular level, we observed a marked upregulation of Nav1.7, Nav1.8, and Nav1.9 mRNA and protein expression in the DRG of CRPS-I rats, which are closely associated with neuronal hyperexcitability and pain sensitization. Nav1.7 is primarily involved in initiating pain signaling, Nav1.8 in maintaining chronic pain transmission, and Nav1.9 in sustaining neuronal excitability^12,28. BMSC transplantation significantly downregulated the expression of these sodium channels, potentially attenuating neuronal excitability by inhibiting channel activation. This effect aligns with the known anti-inflammatory and neuroprotective properties of BMSCs in other chronic pain models ²⁹ . However, as expression levels did not fully return to normal, this suggests that the analgesic effects of BMSCs may result from the coordinated regulation of multiple pathways rather than sodium channel modulation alone.

Previous studies have demonstrated that VGSCs are critical regulators of NP ¹² , with their subtypes, Nav1.7, Nav1.8, and Nav1.9, playing key roles in pain signal transmission ³⁰ . Nav1.7 determines the action potential threshold in nociceptive neurons, serving as a central player in chronic pain. Nav1.8 is significantly co-expressed with pain markers and acts as a crucial regulator of inflammatory pain ³¹ , while the electrophysiological properties of Nav1.9 enable it to markedly enhance neuronal excitability in response to inflammatory mediators^32,33. The findings of this study support these observations and further validate the specific roles of these subtypes in the CRPS-I rat model. Unlike previous research, which has primarily focused on directly using VGSC blockers, this study employs a cell therapy approach to downregulate the expression of these channels, offering a more comprehensive regulatory strategy. Combined with existing theories, our results further confirm the central role of VGSC subtypes in abnormal neuronal discharges and hyperalgesia associated with CRPS-I.

In this study, we found that BMSC transplantation may alleviate pain by downregulating Nav channel expression and contributing to the overall recovery of neuronal function in CRPS-I rats ³⁴ . Previous studies have demonstrated that BMSCs exert significant analgesic effects across various chronic pain models, including mouse models of CRPS, as well as in preliminary clinical trials^35,36. The analgesic mechanisms of BMSCs are thought to involve multiple pathways: suppression of inflammation via the secretion of factors such as TGF-β1 ³⁷ , downregulation of Nav1.7, Nav1.8, and Nav1.9 in the DRG, and inhibition of key signaling pathways such as MAPK (including p38 and ERK)^38,39, thereby reducing neuronal hyperexcitability. Additionally, BMSCs may exert neuroprotective effects through the secretion of neurotrophic factors or the delivery of regulatory microRNAs via extracellular vesicles, which modulate the expression of pain-related genes ⁴⁰ . Our findings support and extend these mechanisms in the context of the CRPS-I model, providing further experimental evidence for the potential of BMSCs in neuropathic pain management.

The significant therapeutic effects of BMSC transplantation in the CRPS-I rat model suggest that this strategy could serve as an effective cell therapy approach, particularly for patients with poor response to existing medications or significant side effects. Given the strong association of Nav1.7, Nav1.8, and Nav1.9 with pain, these channels could serve as critical targets for future pain management. In clinical practice, we recommend combining BMSC transplantation with specific VGSC subtype blockers for more precise treatment. Additionally, optimizing the source, dosage, and delivery methods of BMSCs is essential to enhance therapeutic efficacy and minimize adverse effects. Furthermore, clinical trials should focus on evaluating the applicability of BMSC therapy across various pain conditions and ensuring its long-term safety and efficacy.

Despite providing preliminary evidence for the analgesic effects of BMSCs in CRPS-I, this study has several limitations. First, the experiments were conducted solely in a rat model, and the clinical efficacy and safety of BMSCs in humans remain to be further validated. Second, our investigation focused on three VGSC subtypes, Nav1.7, Nav1.8, and Nav1.9, without systematically evaluating inflammatory responses or other pain-related molecular mechanisms. In addition, the study relied mainly on behavioral and molecular analyses, lacking direct electrophysiological evidence regarding changes in neuronal excitability. While TWL and MWL are commonly used in chronic pain models, they may not fully capture the multidimensional nature of pain in CRPS-I. Given the modest behavioral improvements observed with BMSC treatment in this study, future research should incorporate additional behavioral assays, such as acetone evaporation, spontaneous pain behavior, and dynamic mechanical stimulation tests^41–44, to better assess different aspects of nociceptive hypersensitivity. Complementary electrophysiological recordings and analyses of inflammatory markers are also necessary to elucidate the underlying mechanisms and enhance translational relevance.

Electrophysiological recordings indicated that BMSC-treated DRG neurons showed a modest rightward shift of the I-V curve relative to the model group, approaching the control profile, suggestive of improved sodium current properties. Yet this observation remains descriptive, lacking validation through quantitative metrics such as V₁/₂ shifts or peak current density, and thus provides limited support for the conclusion that BMSCs restore sodium current characteristics. The heterogeneity of DRG neurons, susceptibility of voltage-clamp to subtle errors, and the small sample size (10–15 cells/group) likely contribute to variability. Similar challenges have been reported: different injury models induce divergent changes in TTX-R and TTX-S sodium currents ⁴⁵ , and DRG sodium channel activation/inactivation curves are prone to condition-dependent shifts ⁴⁶ . Future studies should increase sample size, refine stimulation protocols, and incorporate key electrophysiological parameters (e.g. V₁/₂, peak current density, inactivation constants) to rigorously evaluate BMSC effects on sodium currents.

Moreover, extended observation periods are warranted to evaluate the long-term efficacy and safety of BMSC therapy, as well as its applicability to other types of chronic pain. Integrating inflammatory biomarker profiling, electrophysiological techniques, and emerging technologies such as gene editing and targeted drug delivery systems may help improve the precision and effectiveness of BMSC-based treatments.

In conclusion, this study provides initial evidence supporting the analgesic potential of BMSCs in a CRPS-I model and highlights the critical role of Nav sodium channels in pain modulation, thereby contributing to a better understanding of chronic pain mechanisms. Combined with existing animal research ⁷ and early clinical observations ⁸ , our findings offer novel experimental support for the future clinical translation of BMSC-based therapies in neuropathic pain management.

This study is the first to systematically evaluate the analgesic effects and molecular mechanisms of BMSCs in a CRPS-I rat model, thereby providing a scientific foundation for the application of BMSCs in chronic pain management. The research highlights the critical roles of Nav1.7, Nav1.8, and Nav1.9 as key channels for pain transmission and demonstrates that BMSC transplantation effectively modulates their expression, thereby alleviating pain responses. These findings offer new insights into the molecular mechanisms of chronic pain and lay the groundwork for the clinical application of BMSCs in CRPS-I and other NP conditions. Moreover, the results can potentially advance the exploration and application of BMSCs in treating other NP disorders, expanding the clinical prospects of cell therapy.

Conclusion

This study demonstrates that BMSC transplantation can effectively improve pain-related behaviors in CRPS-I rats. Specifically, BMSC treatment significantly increased TWL and MWL, indicating reduced pain sensitivity. Additionally, BMSC transplantation downregulated the mRNA and protein expression levels of Nav1.7, Nav1.8, and Nav1.9 in DRG tissue, suggesting that BMSCs may exert analgesic effects by modulating the expression of these voltage-gated sodium channels. These findings provide strong experimental support for BMSC transplantation as a promising cell-based therapeutic strategy for treating CRPS-I.

Supplemental Material

sj-docx-1-cll-10.1177_09636897251383588 – Supplemental material for Bone marrow mesenchymal stem cells attenuate pain and modulate peripheral sodium channel activity in a rat model of complex regional pain syndrome type I

Supplemental material, sj-docx-1-cll-10.1177_09636897251383588 for Bone marrow mesenchymal stem cells attenuate pain and modulate peripheral sodium channel activity in a rat model of complex regional pain syndrome type I by Yuge Jiang, Kaikai Guo, Yi Liu and Longhe Xu in Cell Transplantation