Electroacupuncture Ameliorates Mechanical Allodynia of a Rat Model of CRPS-I via Suppressing NLRP3 Inflammasome Activation in Spinal Cord Dorsal Horn Neurons

Abstract

Complex regional pain syndrome type-I (CRPS-I) is a chronic neurological disorder that results in severe pain and affects patients' life quality. Conventional therapies usually lack effectiveness. Electroacupuncture (EA) is an effective physical therapy for relieving CRPS-I pain. However, the mechanism underlying EA-induced analgesia on CRPS-I still remain unknown. Spinal NLRP3 inflammasome was recently identified to contribute to pain and neuroinflammation in a rat model of CRPS-I by our group. Here, we aimed to study whether EA could inhibit spinal NLRP3 inflammasome activation, thus resulting in pain relief and attenuation of spinal neuroinflammation in the rat model of CRPS-I. We established the rat chronic post-ischemic pain (CPIP) model to mimic CRPS-I. CPIP rats developed remarkable mechanical allodynia that could be relieved by daily EA intervention. NLRP3 inflammasome was activated in spinal cord dorsal horn (SCDH) of CPIP rats, accompanied with over-production of pro-inflammatory cytokine IL-1β. Immunostaining revealed that the cellular distribution of NLRP3 was predominantly located in SCDH neurons. Pharmacological activation of NLRP3 inflammasome per se is sufficient to produce persistent mechanical allodynia in naïve animals, whereas blocking NLRP3 inflammasome attenuates mechanical allodynia of CPIP rats. EA exclusively reduced NLRP3 overexpression in SCDH neurons and attenuated spinal glial cell over-activation in CPIP rats. EA-induced anti-allodynia with attenuation of spinal glial cell over-activation were all mimicked by intrathecal blocking NLRP3 inflammasome and reversed by activating NLRP3 inflammasome, respectively, through pharmacological methods. Finally, spinal blocking IL-1β attenuated mechanical allodynia and spinal glial cell over-activation in CPIP rats, resembling the effects of EA. In all, these results demonstrate that spinal NLRP3 inflammasome activation contributes to mechanical allodynia of the rat model of CRPS-I and EA ameliorates mechanical allodynia through inhibiting NLRP3 inflammasome activation in SCDH neurons. Our study further supports EA can be used as an effective treatment for CRPS-I.

Keywords: allodynia; complex regional pain syndrome; electroacupuncture; glial cell; inflammasome; spinal cord.

Figure 1

The establishment of the rat CPIP model to mimic CRPS-I and nocifensive behavior evaluation. (A) Representative pictures showing the changes of the hind paw after the O-ring ligation at different time points as indicated. The red arrow indicates the position of the O-ring. (B,C) Ipsilateral (B) or contralateral (C) hind paw swelling after model establishment. (D,E) 50% paw withdraw threshold of ipsilateral (D) or contralateral (E) hind paw from sham and CPIP model rats after model establishment. (F,G) Normalized area under the curve calculated from (D) and (E), respectively. n = 6 rats/group. **p < 0.01 vs. Sham group. Two-way ANOVA with Tukey's post-hoc test was applied in (B–E). One-way ANOVA with Tukey's post-hoc test was applied in (F,G).

Figure 2

EA intervention attenuated mechanical allodynia of CPIP model rats and reduced NLRP3 inflammasome activation in spinal cord. (A) Experimental protocol showing time points for model establishment and EA intervention. (B,C) 50% PWT changes in ipsilateral (B) or contralateral (C) hind paw following EA/sham EA intervention. (D,E) Normalized AUC calculation from panel B&C. (F,G) Western blot showing protein expression of NLRP3 (F), IL-1β (G) in ipsilateral spinal cord tissues. The upper panel shows representative Western blot images and the lower panel shows the summarized data. β-actin was used as a reference control. n = 6 rats/group. *p < 0.05 vs. Sham group. ^##p < 0.01, ^#p < 0.05 vs. CPIP group. NS, no significance vs. CPIP group. Two-way ANOVA with Tukey's post-hoc test was applied in panels B&C. One-way ANOVA with Tukey's post-hoc test was applied in (D–G).

Figure 3

The cellular distribution of NLRP3 in spinal cord dorsal horn of CPIP model rats. (A–C) Double immunostaining showing NLRP3 (in green) co-expression with the neuronal marker NeuN (A, in red), astrocyte marker GFAP (B, in red) and microglial cell marker OX42 (C, in red) in ipsilateral spinal cord dorsal horn of CPIP rats. DAPI (in purple) was used as counterstain. Quantification is illustrated on the right of the panels. Scale bar indicates 100 μm. Data were obtained from 9 rats/group.

Figure 4

EA intervention attenuates NLRP3 overexpression in spinal cord dorsal horn neurons. (A) Double immunostaining showing NLRP3 (in green) co-expression with the neuronal marker NeuN (in red) in ipsilateral spinal cord dorsal horn of CPIP rats under different treatment conditions. DAPI (in purple) was used as counterstain. Quantification is illustrated on the right of the panels. (B) Summary of the mean fluorescence intensity of NLRP3. All intensities were normalized with the value of sham group. (C) Summary of the percentage of NLRP3+ cells among all NeuN+ cells. **p < 0.01 vs. Sham group. ^#p < 0.05 vs. CPIP group. NS, no significance vs. CPIP group. Scale bar indicates 100 μm. Data were obtained from 9 rats/group. One-way ANOVA with Tukey's post-hoc test was applied in (B,C).

Figure 5

EA intervention reduces glial cell overactivation in spinal cord dorsal horn of CPIP rats. (A) Immunostaining pictures showing the effect of EA/sham EA intervention on astrocyte (Upper panels, marked with GFAP) and microglia (lower panels, marked with OX42) overactivation in SCDH. (B) Summary of the mean fluorescence intensity of GFAP, which were normalized with the value of sham group. (C) Summary of the total number of GFAP⁺ cells/observation field. (D) Summary of the normalized mean fluorescence intensity of OX42, which were normalized with the value of sham group. (E) Summary of the total number of OX42⁺ cells/observation field. Scale bar indicates 100 μm. Data were obtained from 9 rats/group. **p < 0.01 vs. Sham group. ^##p < 0.01 vs. CPIP group. NS, no significance vs. CPIP group. One-way ANOVA with Tukey's post-hoc test was applied in (B–E).

Figure 6

EA's anti-allodynic effect on CPIP rats is mimicked by pharmacological blocking NLRP3 inflammasome. (A) Experimental protocol. (B) 50% PWT following EA or MCC950 intervention (30 μg/rat/day, i.t.). (C) Normalized AUC deduced from (B). (D) Immunostaining showing the effect of EA/MCC950 intervention on astrocyte (Upper panels, marked with GFAP) and microglia overactivation in SCDH. (E,F) Summary of the mean fluorescence intensity of GFAP (E) and OX42 (F), normalized with the value of sham group. n = 5 rats/group. Scale bar indicates 100 μm. **p < 0.01 ^#p < 0.05, ^##p < 0.01. Two-way ANOVA with Tukey's post-hoc test was applied in panel B. One-way ANOVA with Tukey's post-hoc test was applied in (C,E,F). NS, no significance.

Figure 7

NLRP3 inflammasome activation per se causes mechanical allodynia and spinal glial activation in naïve animals. (A) Experimental protocol indicating treatment time points. (B) 50% PWT of rats after intrathecal vehicle (0.1% ethanol in PBS) or nigericin (5 μg/rat/day) injection. (C) Normalized AUC deduced from (B). n = 6 rats/group. (D–F) Western blot showing protein expression of NLRP3 and IL-1β in spinal cord. (D) shows the representative images. (E,F) shows the summarized data of NLRP3 and IL-1β protein expression. (G) Immunostaining of GFAP after intrathecal nigericin in SCDH. (H) Summary of the mean fluorescence intensity of GFAP. (I) Summary of the total number of GFAP⁺ cells/observation field. (J) Immunostaining of OX42 after intrathecal nigericin in SCDH. (K) Summary of the mean fluorescence intensity of OX42. (L) Summary of the total number of OX42⁺ cells/observation field. Scale bar indicates 100 μm. *p < 0.05, **p < 0.01 vs. +Veh group. n = 4–6 rats/group. Two-way ANOVA with Tukey's post-hoc test was applied in panel B. Student's t-test was used in other panels.

Figure 8

EA-induced anti-allodynic effect in CPIP rats is reversed by NLRP3 inflammasome activator nigericin. (A) Experimental protocol. (B) Time course showing the effect of intrathecal nigericin on EA-induced anti-allodynia in CPIP rats. (C) Normalized AUC deduced from (B). (D) Immunostaining of GFAP in ipsilateral SCDH after intrathecal nigericin/Veh injection. (E) Summary of the mean fluorescence intensity of GFAP. (F) Summary of the total number of GFAP⁺ cells/observation field. (G) Immunostaining of OX42 in ipsilateral SCDH after intrathecal nigericin/Veh injection. (H) Summary of the mean fluorescence intensity of OX42. (I) Summary of the total number of OX42⁺ cells/observation field. Scale bar indicates 100 μm. *p < 0.05, **p < 0.01 vs. CPIP+EA+Veh group. n = 6 rats/group. Two-way ANOVA with Tukey's post-hoc test was applied in panel B. Student's t-test was used in other panels.

Figure 9

Spinal blocking IL-1β prevents mechanical allodynia and attenuates spinal glial over-activation in CPIP model rats. (A) Experimental protocol. (B) Time course showing the effect of intrathecal IL-1Ra (100 ng/rat/day) or vehicle (Veh 0.1% BSA in PBS) on mechanical allodynia of CPIP rats. (C) Normalized AUC deduced from panel B. **p < 0.01 vs. Sham group. ^##p < 0.01 vs. CPIP+Veh group. (D) Immunostaining of GFAP in ipsilateral SCDH after intrathecal IL-1Ra/Vehicle injection. (E) Summary of mean fluorescence intensity of GFAP (normalized to CPIP+Veh group). (F) Summary of total number of GFAP⁺ cells/observation field. (G) Immunostaining of OX42 in ipsilateral SCDH after intrathecal IL-1Ra/Veh injection. (H) Summary of mean fluorescence intensity of OX42 (normalized to CPIP+Veh group). (I) Summary of total number of OX42⁺ cells/observation field. *p < 0.05, **p < 0.01 vs. CPIP +Veh group. n = 5 rats/group. Scale bar indicates 100 μm. Two-way ANOVA with Tukey's post-hoc test was applied in panel B. Student's t test was used in other panels.