Analgesic Effect of Perineural Injection of BoNT/A on Neuropathic Pain Induced by Chronic Constriction Injury of Sciatic Nerve in Rats

This study was designed to investigate the analgesic effect of perineural injection of BoNT/A on neuropathic pain induced by sciatic nerve chronic constriction injury (CCI) and possible mechanisms. SD rats were randomly divided into Sham group, CCI group and BoNT/A group. Paw mechanical withdrawal threshold (pMWT) and paw thermal withdrawal latency (pTWL) of each group were detected at different time points after surgery. The expression of myelin markers, autophagy markers and NLRP3 inflammasome-related molecules in injured sciatic nerves were examined at 12 days after surgery. Moreover, C-fiber evoked potential in spinal dorsal horn was recorded. The expression of SNAP-25, neuroinflammation and synaptic plasticity in spinal dorsal horn of each group were examined. Then rats treated with BoNT/A were randomly divided into DMSO group and Wnt agonist group to further explore the regulatory effect of BoNT/A on Wnt pathway. We found that pMWT and pTWL of ipsilateral paw were significantly decreased in CCI group compared with Sham group, which could be improved by perineural injection of BoNT/A at days 7, 9 and 12 after surgery. The peripheral analgesic mechanisms of perineural injection of BoNT/A might be related to the protective effect on myelin sheath by inhibiting NLRP3 inflammasome and promoting autophagy flow, while the central analgesic mechanisms might be associated with inhibition of neuroinflammation and synaptic plasticity in spinal dorsal horn due to inhibiting SNAP-25 and Wnt pathway. As a new route of administration, perineural injection of BoNT/A can relieve CCI induced neuropathic pain probably via both peripheral and central analgesic mechanisms.


Introduction
Peripheral nerve entrapment is a common clinical disease, inducing neuropathic pain and motor dysfunction of innervated areas, for example, median nerve entrapment in the carpal canal, with a prevalence ranging from 3.72 to 6.80% Juan-juan He and Xiao-mei Wei have contributed equally to this work.
in the United State [1]. Currently, the main treatment for peripheral neuropathic pain is pharmacotherapy, but it is hard to achieve sufficient pain relief, and the side effects of analgesics are frequent [2].
Recently, clinical trials have shown that botulinum toxin A (BoNT/A) is safe and effective in relieving peripheral neuropathic pain, such as postherpetic neuralgia [3], diabetic peripheral neuropathy [4] and trigeminal neuralgia [5]. Compared with other analgesic drugs, the greatest advantage of BoNT/A is a sustained effect after a single application and low risk for adverse effects [6]. Animal studies have documented the analgesic effect of BoNT/A in peripheral nerve injury, such as anti-inflammatory effects, regenerative effects in the injured nerve, effects on astroglia and microglia, effects on the ascending pain processing pathway, and indirect central actions on the endogenous opioid and GABA neurotransmission [7]. However, in those animal studies, intranerve injection [8], intrathecal injection [9] or intraplanter injection [10] of BoNT/A are usually adopted for administration, which was not suitable for peripheral nerve entrapment in clinical practice.
Perineural injection, also known as hydrodissection, is a new method to treat peripheral nerve entrapment, providing not only a mechanical effect to release and decompress the entrapped nerve, but also a pharmacological effect relieving pain through numerous mechanisms [11]. However, perineural injection of BoNT/A is seldom performed through animal experiments and clinical practical tests. Recent cases reported that perineural BoNT/A injection was an effective and safe method, and perineural BoNT/A injection showed a particularly enhanced effect in patients without signs of hyperalgesia, which contrasted the results noted after subcutaneous BoNT/A injection [12]. However, it is not clear about the mechanisms by which this new route of administration is effective, especially in model of peripheral nerve entrapment.
Therefore, we intend to study the analgesic effect of perineuronal injection of BoNT/A in the animal model of peripheral nerve constriction injury, and explore the analgesic mechanism from the peripheral and central perspectives, in order to provide a suitable administration method of BoNT/A for the treatment of neuropathic pain induced by peripheral nerve entrapment.

Animals and Groups
A total of 51 Male Sprague Dawley (SD) rats were used for this study. Rats were housed in standard cages at a constant temperature under a 12 h light/dark cycle, with free access to water and food. The rats weighed 200-220 g at the time of surgery. This study was approved by Institutional Animal Care and Use Committee, Sun Yat-Sen University (Approval NO. SYSU-IACUC-2022-000798). An observer who was blinded to the groups of animals performed the behavioral tests.
The rats were randomly divided into Sham group (sham operated rats treated with saline, n = 13), CCI group (chronic constrictive injury rats treated with saline, n = 13) and BoNT/A group (CCI rats treated with BoNT/A, n = 13). In order to further study the regulatory effect of BoNT/A on Wnt pathway, rats treated with BoNT/A were randomly divided into DMSO group (CCI rats were treated with DMSO after administration of BoNT/A, n = 6) and Wnt agonist group (CCI rats were treated with Wnt agonist after administration of BoNT/A, n = 6).

Surgery
CCI surgery was performed according to the method described in the literature [13]. Briefly, rats were anesthetized with an intraperitoneal injection of pentobarbital sodium (30 mg/kg). After making a skin incision along the lateral surface of the biceps femoris, blunt forceps were inserted into the muscle belly to split the biceps femoris muscle and expose the left sciatic nerve. Next, three ligatures (3-0 chromic gut suture) were loosely tied around the nerve at 1-mm spacing. The sciatic nerves in the Sham group were only exposed without ligation.

Drugs Administration
The BoNT/A (Lanzhou Biotechnique Development Co., LTD) was dissolved in normal saline (100U/5 ml). Perineural injection of BoNT/A (15 U/kg) or an equivalent volume of normal saline was administrated under the guidance of ultrasound 4 days after surgery. The Wnt agonist BML-284 (AbMole, America) was dissolved in DMSO accordance with the instructions. BML-284 (40 uM in 20ul) or an equivalent volume of DMSO was injected intrathecally at the L5 and L6 intervertebral space in rats by a direct transcutaneous intrathecal injection method as described in the literature [14] at days 5, 8 and 11 after CCI surgery. The drug doses were chosen based on previous research [15].

Ultrasound Guided Perineural Injection
Perineural injection was performed 4 days after surgery under the high-resolution musculoskeletal ultrasound device (Ultrasound System, M-Turbo, Sonosite, America; transducer, L25x/13-6 MHz, Sonosite, America). Rats were positioned on their right side under anesthesia. The hind portion was shaved and gel was placed, then the ultrasound transducer probe was placed on the lateral side of hind limb at the mid-thigh, close to the skin incisions for CCI surgery. Small adjustments were usually required to provide optimal image quality. Following this technique, the sciatic nerve was identified as an oval hypoechoic structure underneath the gluteal fascia (Fig. 1A). Once the nerve was identified, moving the probe as close to the distal knot as possible, then insert a 26 G needle in an in-plane approach (Fig. 1B). When the tip of the needle was considered as close as possible to the nerve without touching it, BoNT/A was injected to separate the nerve from surrounding soft tissues. The ultrasound-guided injection technique referred to the literature [16] and our previous study [17].

Behavioral Studies
pMWT was assessed with the up-down method and a 50% paw withdrawal threshold was determined according to the method described in the literature [13]. A set of von-Frey hair (Aesthesio, USA) of different forces was applied perpendicularly to the ipsilateral hind paws until bent. A positive response was recorded when the rats withdrew, flicked, or licked their paw. A stronger force was used in case of no response and a weaker force was used in case of a positive response. Every hind paw was tested alternately at 5 min intervals. The largest force used was 15 g to reduce animal suffering. At least 4 valid readings were obtained after the first positive response.
A BL-200 radiant heat apparatus (Chengdu Taimeng Technology & Market Corporation, China) was applied to assess pTWL according to the method described in the literature [13]. Rats were placed in a clear plastic chamber with a glass floor, and a radiant heat source beneath the floor was aimed at the plantar surface of the ipsilateral hind paw. The time rats needed to withdraw from the heat stimulus was automatically recorded as withdrawal latency. The intensity of the light source was adjusted to produce a withdrawal latency of 12-15 seconds (s) in the sham group, and the maximum cutoff time was set at 20 s to reduce animal suffering.
The baseline threshold of mechanical allodynia and thermal hyperalgesia of rats were tested before surgery, and no differences was observed among the rats.

C-fiber Evoked Field Potentials Recording
C-fiber evoked field potentials in the spinal dorsal horn were recorded, according to the method described in the literature [18]. Briefly, C-fiber evoked field potentials in response to electrical stimulation of the sciatic nerve were recorded in the spinal dorsal horn (L4 and L5 segments) with a glass microelectrode, which was driven by an electronically controlled microstepping motor (Narishige Scientific Instrument Laboratory, Japan). The amplitudes of C-fiber evoked field potentials were recorded and analyzed by the LTP program. Test stimuli delivered to the sciatic nerve were single square pulses (0.5 ms duration, in 1-min intervals). The intensity of stimulation was 1 V, 2.5 V, 5 V, 7.5 V, 10 V, 12.5 V, 15 V, 20 V, 25 V, respectively. The place of stimulation was the proximal portion of the lesion (the segment of sciatic nerve tied with three ligatures), and the distance from the stimulation site at the sciatic nerve to recording site in the lumbar spinal dorsal horn was about 10 cm.

Tissue Harvesting
12 days after surgery, rats were deeply anesthetized with pentobarbital sodium (50 mg/kg) by intraperitoneal injection, after which, rats were perfused with 0.9% saline. For immunofluorescence studies, rats were transcardially perfused with 0.9% saline followed by 4% paraformaldehyde. Ipsilateral sciatic nerves and spinal cord were collected for immunofluorescence staining or western blotting analyses. The samples for western blotting were snap frozen in liquid Fig. 1 Ultrasound imaging of sciatic nerve and ultrasound-guided perineural injection. A Ultrasound imaging of sciatic nerve. The thin white line represents the gluteus fascia, within which is the sciatic nerve appeared as a hypoechoic structure. B Ultrasound-guided perineural injection of BoNT/A. Needle was inserted in an in-plane approach adjacent to the sciatic nerve. When the tip of the needle was considered as close as possible to the nerve without touching it, BoNT/A was injected to separate the nerve from surrounding soft tissues. Short blue arrow: sciatic nerve, long white arrow: puncture needle nitrogen immediately after collection and stored at − 80 °C. The samples for immunofluorescence staining were fixed in 4% paraformaldehyde at 4 ℃ for 12 h.

Immunofluorescence
Immunofluorescence staining was performed as literature described [19]. Samples were dehydrated using graded sucrose (20% and 30% sucrose in 0.1 M PBS) at 4℃ for 3-4 days. Transverse sections of the lumbar spinal cord were cut at 10 um thickness, and the transverse sections of sciatic nerve 0.5 cm distal to the lesion (the segment of sciatic nerve tied with three ligatures) were cut at 14 um thickness using a cryostat (Leica, Germany). Sections were treated with citrate buffer at 95 °C for 5 min and cooled to room temperature. Then sections were blocked in immunofluorescence blocking solution containing 0.3% Triton X-100 at room temperature for 1 h. The sections of sciatic nerve were stained with primary antibodies against S100 (1:200, Abcam, UK) and NF200 (1:200, Abcam, UK), while the sections of spinal cord were stained with primary antibodies against β-catenin (1:200, Abcam, UK), SNAP-25 (1:100, Thermo Fisher Scientific, MA) and Iba-1 (1:200, Wako, Germany) for 12-14 h at 4℃, then washed in PBS and followed by incubation with the corresponding secondary antibodies (1:500, Cell Signaling Technology, USA). After washing with PBS, tissues were mounted with antifade fluorescence mounting medium and cover-slipped. Images were captured using a fluorescence microscope (Nikon, Japan). The observation site was determined under a high-power field (10 × 40) and 3 randomly chosen field was selected to take pictures. The fluorescence images were analyzed with Image J software (NIH, USA).

Western Blot Analysis
Western blot analysis was performed as literature described [13]. The sciatic nerve segments from 0.5 cm proximal to 0.5 cm distal to the lesion were removed and used for Western blotting. Tissue samples (injured sciatic nerve or lumbar spinal dorsal horn) were lysed in RIPA buffer containing phosphatase and protease inhibitors, then centrifuged at 12,000 g for 30 min at 4℃. After centrifugation, the supernatant solution was collected, and the protein concentrations were determined by a BCA assay kit (Invitrogen, USA). The proteins were subsequently boiled and equal concentrations of total protein were loaded on 12.5% SDS- . The membranes were then washed with TBST buffer and incubated with HRP-conjugated anti-rabbit secondary antibodies. Protein bands were visualized with enhanced chemiluminescence Western blot detection reagents (Millipore, USA). Individual protein bands were quantified by densitometry using ImageJ software.

Data Analysis
Data are presented as the mean ± SD. Statistical analysis of all data was conducted with SPSS 17.0. The results from the behavioral tests and C-fiber evoked field potentials recordings were analyzed with repeated-measure analysis of variance (ANOVA). The comparisons of multiple groups were analyzed by one-way ANOVA with post hoc least significant difference (LSD) test. Student's t test was used for comparisons between two groups. A two-tailed p < 0.05 was regarded as statistically significant.

Perineural Injection of BoNT/A Significantly Alleviated CCI-Induced Neuropathic Pain
Compared with the Sham group, the CCI group exhibited a significant decrease in paw mechanical withdrawal threshold (pMWT) and paw thermal withdrawal latency (pTWL) at different time points after surgery, suggesting that neuropathic pain was produced by chronic construction injury of the sciatic nerve in rats (Fig. 2). After perineural injection of BoNT/A, both pMWT and pTWL were elevated significantly compared with the CCI group at 7, 9 and 12 days after surgery, respectively, showing an analgesic effect of perineural injection of BoNT/A against CCI induced neuropathic pain (Fig. 2).

Perineural Injection of BoNT/A Showed a Protective Effect on Myelin Sheath in CCI Rats
The Schwann cells that comprise the myelin sheath in the distal end of the injured sciatic nerve was observed by immunofluorescence staining. In the Sham group, intact myelin sheaths surround axons was visible, and the expression of S100 protein (Schwann cell marker) and NF200 protein (axonal marker) were abundantly detected in sham operated sciatic nerves. Compared with the Sham group, CCI group exhibited a significant decrease in the expression of S100, and axons with thinly or absent myelinated sheaths was observed, suggesting demyelination change of axons after sciatic nerve injury. However, perineural injection of BoNT/A significantly increased S100 immunoreactivity, which demonstrated a protective effect of BoNT/A on myelin sheath (Fig. 3). The results of western blotting showed that compared with Sham group, the expression of MBP (a marker of myelinating Schwann cells) in injured sciatic nerve decreased greatly in CCI rats. After perineural injection of BoNT/A, the expression of MBP increased after CCI surgery and pain behavioral measurements were performed on day 0 (before surgery), day 4 (before injection of BoNT/A), day 7, day 9, and day 12 after surgery. Compared with Sham group, ipsilateral pMWT and pTWL were markedly decreased in CCI group at different time points after surgery. However, perineural injection of BoNT/A significantly increased pMWT and pTWL on days 7, 9, and 12 after surgery when compared with CCI group. Curves showed quantified results as mean ± SD (n = 6 in each group). ***p < 0.001 compared with Sham group, ###p < 0.001 compared with CCI group Fig. 3 Representative photographs showed the expression of S100 (Schwann cell marker) and NF200 (axonal marker) in the distal portion of injured sciatic nerves (A), and the histograms showed summary data of S100 expression (B) (n = 4 in each group). 12 days after surgery, the CCI group exhibited a significant decrease in the expression of S100 than Sham group, and axons with thinly or absent myelinated sheaths were observed, which was improved by perineural injection of BoNT/A. S100 and NF200 were detected by immunofluorescence under 400 × magnification, Scale bar = 50 μm. Data were presented as mean ± SD. ***p < 0.001 compared with Sham group, ###p < 0.001 compared with CCI group significantly, suggesting that perineural injection of BoNT/A could promote myelin repair. (Fig. 4) However, significant differences of S100 and MBP protein expression were observed between the Sham group and BoNT/A group, indicating CCI model exhibited severe damage to the myelin sheath, which could not be completely reversed by perineural injection of BoNT/A. (Figs. 3, 4).

Perineural Injection of BoNT/A Augmented Autophagic Flux in the Injured Sciatic Nerve by CCI
Autophagy plays important roles in remyelination by Schwann cells after sciatic nerve injury. We observed an increased autophagosome marker microtubule-associated protein-1 light chain 3 (LC3)-II level and autophagy flux marker p62 level in the ipsilateral sciatic nerves of the CCI group compared with Sham group, demonstrating induction of autophagy and prohibition of autophagic flux after sciatic nerve injury. After perineural injection of BoNT/A, an increase in LC3-I to LC3-II conversion was observed, as well as a significantly decrease in the expression of p62, indicating enhanced autophagy flux. (Fig. 5).

Perineural Injection of BoNT/A Inhibited the Activation of NLRP3 Inflammasome
It was reported that the autophagy dysfunction can lead to excessive activation of NOD-like receptor family pyrin domain-containing 3 (NLRP3) inflammasome [20]. We observed the expression levels of NLRP3 inflammasome components (NLRP3, ASC and caspase 1) as well as IL-18 were significantly increased in the ipsilateral sciatic nerves greatly in CCI rats, which was rescued by perineural injection of BoNT/A. Data were presented as mean ± SD. ***p < 0.001 compared with Sham group, #p < 0.05 compared with CCI group Fig. 5 Perineural injection of BoNT/A augmented autophagic flux in the injured sciatic nerve. 12 days after surgery, the expressions of autophagosome marker LC3-II level and autophagy flux marker p62 level in the ipsilateral sciatic nerves of the CCI group were increased compared with Sham group. After perineural injection of BoNT/A, the expression of LC3-II increased, accompanied by a significantly decrease in the expression of p62. n = 4 in each group. Data were presented as mean ± SD. *p < 0.05, ***p < 0.001 compared with Sham group, ##p < 0.01 and ###p < 0.001 compared with CCI group of the CCI group compared with Sham group at 12 days after surgery, demonstrating activation of NLRP3 inflammasome after sciatic nerve injury. After perineural injection of BoNT/A, the expression of NLRP3, ASC and caspase 1 decreased significantly, which was accompanied by a decline of IL-18 production, indicating inhibition of NLRP3 inflammasome activation (Fig. 6).

Perineural Injection of BoNT/A Reduced the Expression of SNAP-25 in Spinal Dorsal Horn
BoNT/A can cleave synaptosomal-associated protein 25 kDa (SNAP-25) and block the release of some neurotransmitters (e.g., acetylcholine) from pre-junctional nerves at the neuromuscular junction. It was reported under some circumstances BoNT/A can reach the central nervous system by axonal retrograde transport and inhibit central sensitization [7]. However, whether perineural injection of BoNT/A can reach spinal cord and prevent central sensitization induced by CCI is unknown. As shown in Fig. 7a, b, 12 days after surgery, the expression level of SNAP-25 was significantly increased in CCI group compare with the Sham group, and it is abundantly expressed in the superficial layer of spinal dorsal horn detected by immunofluorescence. After perineural injection of BoNT/A, there was a significant reduction in the expression level of SNAP-25, indicating perineural injection of BoNT/A exerted a central effect by downregulating the expression of SNAP-25, and maybe perineural injection of BoNT/A could enter the axonal retrograde transport pathway.

Perineural Injection of BoNT/A Attenuated CCI Induced LTP and pNR2B Expression in Ipsilateral Spinal Dorsal Horn
The central sensitization is underlined primarily by longlasting enhancement of synaptic transmission, called longterm potentiation (LTP) [21]. Peripheral nerve injury can induce LTP, manifesting as the increased amplitude of C-fiber evoked field potential in spinal dorsal horn. NR2B, a subunit of N-methyl-d-aspartate receptors (NMDAR), is implicated in neuroplasticity, and the phosphorylation of NMDAR at Tyr1472 (pNR2B) has repetitively been reported to be important for central sensitization in pain models [14]. As shown in Figs. 8 and 9, the amplitude of C-fiber evoked field potential was significantly increased when induced by electrical stimulation (5-12.5 V) of the sciatic nerve in the CCI group compared with Sham group, accompanied by the up-regulation of pNR2B in spinal dorsal horn 12 days after surgery, suggesting enhanced synaptic plasticity after sciatic nerve injury. After perineural injection of BoNT/A, amplitude of C-fiber evoked field potential decreased as well as the expression of pNR2B down-regulated, indicating perineural injection of BoNT/A played an important role in regulating synaptic plasticity. Fig. 6 Perineural injection of BoNT/A inhibited the activation of NLRP3 inflammasome in the ipsilateral sciatic nerves. The expression of NLRP3 inflammasome components (NLRP3, ASC and caspase 1) and IL-18 in the ipsilateral nerves were markedly increased 12 days after CCI. After perineural injection of BoNT/A, the expres-sion of NLRP3 inflammasome components (NLRP3, ASC and caspase 1) decreased significantly, accompanied by a decline in IL-18. n = 4 in each group. Data were presented as mean ± SD. *p < 0.05, **p < 0.01 and ***p < 0.001 compared with Sham group, #p < 0.05, ##p < 0.01 and ###p < 0.001 compared with CCI group

Perineural Injection of BoNT/A Inhibited the Activation of Microglia and the Expression of IL-18 in Spinal Dorsal Horn
Activation of microglia in the spinal dorsal horn following peripheral nerve injury is well characterized, and it exerts a crucial role in pathologic process of neuroinflammation. Iba-1 is a marker of microglia, which is markedly up-regulated in activated microglia, and Iba-1 immunofluorescence staining is often used to detect microglia activation in the spinal cord. Interleukin-18 (IL-18) is a proinflammatory cytokine produced by activated microglia. As . Data were presented as mean ± SD. **p < 0.01, ***p < 0.001 compared with Sham group, #p < 0.05, compared with CCI group. SNAP-25 was detected by immunofluorescence under 400 × magnification, Scale bar = 50 μm Fig. 8 Perineural injection of BoNT/A attenuated the CCIinduced LTP at spinal C-fiber synapses. Stimulationresponse curve of C-fiber evoked field potentials in different groups were shown (n = 5 in each group). The electrophysiological recordings were performed at 12 days after surgery. When the intensity of electrical stimulation gradually increased from 5 V to 12.5 V, the amplitude of C-fiber evoked field potential in CCI group was significantly increased than that in Sham group, but the effect was attenuated by perineural injection of BoNT/A. Curves showed quantified results as mean ± SD. *p < 0.05, **p < 0.01 and ***p < 0.001 compared with sham group, #p < 0.05, ##p < 0.01 and ###p < 0.001 compared with CCI group Fig. 9 Perineural injection of BoNT/A attenuated CCI-induced upregulation of pNR2B. The expression of pNR2B was detected by western blotting in different groups (n = 4 in each group). 12 days after surgery, the expression of pNR2B in spinal dorsal horn was greatly increased in CCI group compared with Sham group. After perineural injection of BoNT/A the expression of pNR2B decreased significantly. **p < 0.01 and ***p < 0.001 compared with sham group, #p < 0.05, compared with CCI group  Figs. 10 and 11, 12 days after surgery, activation of microglia was observed the spinal dorsal horn in CCI group, accompanied by increased IL-18 level compared with the Sham group. After perineural injection of BoNT/A, the activation of microglia was inhibited, and the IL-18 level was downregulated.

Perineural Injection of BoNT/A May Regulate Neuroinflammation and Synaptic Plasticity Through the Wnt Pathway
Evidence showed that the activation of the Wnt pathway promoted neuroinflammation in the spinal cord and affected neuronal synaptic plasticity. Whether perineural injection of BoNT/A affects the Wnt pathway is still unknown. As shown in Fig. 12a, b, 12 days after surgery, the Wnt pathway was activated in the spinal dorsal horn in CCI group, compared with Sham group, and β-catenin was mainly expressed in the superficial layers of the spinal dorsal horn. After perineural injection of BoNT/A, the activation of Wnt pathway was inhibited. As shown in Figs. 13 and 14, When Wnt agonist was used to active the Wnt pathway after perineural injection of BoNT/A, the mechanical allodynia and thermal hyperalgesia ameliorated by BoNT/A was reversed, accompanied by the expression of pNR2B and IL-18 increased in spinal dorsal horn, suggesting perineural injection of BoNT/A regulated neuroinflammation and synaptic plasticity possibly through the Wnt signaling pathway.

Discussion
In this study, we found that perineural injection of BoNT/A could relieve CCI induced neuropathic pain, the peripheral mechanism might be related to the protective effect on myelin sheath by promoting autophagy flow and inhibiting NLRP3 inflammasome, while the central mechanism might be related to inhibition of inflammatory response and synaptic plasticity in spinal dorsal horn due to its ability to inhibit SNAP-25 and Wnt pathway.
Chronic constriction injury of the sciatic nerve is a widely used peripheral nerve entrapment animal model of Data were presented as mean ± SD. **p < 0.01 compared with Sham group, #p < 0.05, ##p < 0.01 compared with CCI group. β-catenin was detected by immunofluorescence under 400X magnification, Scale bar = 50 μm neuropathic pain, which involves partial nerve injury [22]. In this study, a rat CCI model was chosen to explore the effect of perineural injection of BoNT/A in the treatment of peripheral nerve injury induced-neuropathic pain, aimed at providing a suitable administration method of BoNT/A for clinical practice. Ultrasound-guided perineural injection, also known as hydrodissection, is a new method to treat peripheral nerve entrapment in clinical practice. Injectates commonly included normal saline, 5% dextrose, corticosteroids, local anesthetics, hyaluronidase, and platelet-rich plasma [23], and it is speculated that larger injected volumes yield better results due to the conjoined effect of hydrodissection and better injection distribution in the treatment of peripheral nerve compression [24]. Although no study has  BoNT/A + DMSO group, the expressions of pNR2B and IL-18 in spinal dorsal horn were significantly increased in BoNT/A + Wnt agonist group. Data were presented as mean ± SD. **p < 0.01 compared with BoNT/A + DMSO group reported the use of BoNT/A for hydrodissection, ultrasoundguided nerve block with BoNT/A for intractable neuropathic pain has been reported [25]. In addition, it was found that a higher dilution of BoNT/A resulted in better therapeutic effects than a lower dilution volume from a clinical case report [26]. Thus, we believed that BoNT/A was a suitable injectate for hydrodissection, and a high dilution of BoNT/A was used in this study, which proved to be a feasible and effective administration method in the treatment of peripheral nerve injury induced-neuropathic pain.
In this study, we found that perineuronal administration of BoNT/A showed protective effects on Schwann cells in injured nerve. In a previous study, it was reported that intranerve delivery of BoNT/A can induce sensorimotor recovery by directly stimulating axonal regeneration and upregulating the expression of S100 in sciatic nerve crush model, suggesting an enhancing effect of BoNT/A on migration or proliferation of Schwann cells [27]. In another study conducted by Seo, a higher expression of S100β was detected in injured nerves treated with a single session of intranerve BoNT/A, suggesting BoNT/A induced the activation of Schwann cells and functional improvement in the injured nerve [8]. In this study, we found that perineural injection of BoNT/A increased the expression of S100 and MBP in the injured sciatic nerve, which was consistent with the results of previous studies, suggesting that perineural administration of BoNT/A could effectively protect Schwann cells against peripheral nerve injury. However, it could not completely reverse the damage, a possible reason might be the short observation time, and a longer experimental time may make the improvement more obvious. Another possible reason might be the multiple mechanisms implicated in the damage of Schwann cells, while perineural administration of BoNT/A might act only via partial mechanisms.
As we know, after peripheral nerve injury, myelin is degraded to myelin debris, then cleared by the upregulation of Schwann cells autophagy [28]. It has been confirmed that Schwann cells autophagy is involved in peripheral nerve injury induced neuropathic pain. Moreover, it was reported that autophagy inducer could alleviate pain and facilitate nerve regeneration [29]. In this study, activation of autophagy was observed in injured nerve, and we found that perineural administration of BoNT/A could increase the expression of LC3-II and reduce the expression of p62 in injured nerve, indicating the autophagy flux was promoted by BoNT/A, which might be a way to protect Schwann cells against peripheral nerve degeneration. However, further research is still needed to explore the mechanisms of BoNT/A in promoting Schwan cells autophagy.
It is reported that the autophagy dysfunction can lead to excessive activation of NLRP3 inflammasome [20], thus promotes neuroinflammatory injury and pain, and inhibits the regeneration and repair of peripheral nerve injury [30]. However, the role of NLRP3 inflammasome in peripheral nerves is less researched, especially in injured sciatic nerve. The study conducted by Cui suggested that NLRP3 inflammasome was activated after sciatic nerve injury in wild-type mice, which upregulated the expression of IL-1β and IL-18, contributing to the inflammatory response. Moreover, NLRP3 knockout mice exhibited an increased sciatic functional index and attenuated inflammatory response, suggesting that NLRP3 deficiency was beneficial to sciatic nerve recovery after injury [31]. In this study, we found that NLRP3 inflammasome was activated in injured sciatic nerve accompanied by the upregulation of IL-18, which was basically consistent with results of Cui's study [31]. Moreover, we found that perineural administration of BoNT/A could inhibit the activation of NLRP3 inflammasome and downregulate the expression of IL-18 in injured nerve. As mentioned above, perineural administration of BoNT/A showed a protective effect on myelin sheath, probably through promoting Schwwan cells autophagy and thus inhibiting activation of NLRP3 inflammasome. However, further study is needed to clarify the precise mechanisms of this action.
Except for peripheral mechanisms, we also explored the central mechanisms of perineural injection of BoNT/Ainduced analgesia. It is well established that indirect central effects can be produced by peripherally injected BoNT/A through peripheral mechanisms of alteration of central sensorimotor integration [32]. Therefore, the results associated with peripheral protective effects of BoNT/A in this study suggested the indirect central effect of perineural injection of BoNT/A. Alternatively, the direct central effect may be a consequence of retrograde axonal transport of BoNT/A from the injection site to central nervous system, leading to cleavage of SNAP-25 which results in the inhibition of pain neurotransmitter release in central nervous system. In this study, we found that the CCI induced upregulation of SNAP-25 in the spinal dorsal horn could be downregulated by perineural injection of BoNT/A, which suggested perineural injection of BoNT/A might exert a direct central analgesic effect. It might be involved in the retrograde axonal transport of BoNT/A. Above all, it is possible that both indirect and direct central effects contributed to the analgesic effect of perineural injection of BoNT/A. LTP in spinal cord, resulted from the activation of pNR2B, is an important form of synaptic plasticity and central sensitization [18]. In this study, enhanced LTP at spinal C-fiber synapses accompanied by upregulation of pNR2B in spinal dorsal horn were observed in CCI rats, as a means of enhanced spinal synaptic plasticity, which was consistent with the findings of a previous study [33]. Moreover, we found perineural injection of BoNT/A inhibited LTP at spinal C-fiber synapses and the expression of pNR2B in spinal dorsal horn in this study, indicating perineural injection of BoNT/A played a role in regulating spinal synaptic plasticity and central sensitization.
Emerging evidence has demonstrated that glial activation and overproduction of proinflammatory cytokines induce LTP and persistent pain [21]. In this study, we found that microglia were activated and proinflammatory cytokine IL-18 was upregulated in spinal cord after CCI surgery, which could be attenuated by perineural injection of BoNT/A. It was reported that intraplantar BoNT/A injection could attenuate pain-related behavior, microglial activation [34] and restore the neuroimmune balance by decreasing the levels of pronociceptive factors (IL-1β and IL-18) and increasing the levels of antinociceptive factors (IL-10 and IL-1RA) in the spinal cord [10]. Although the route of administration in this study differed from the previous studies, our findings were basically consistent with the previous studies, indicating that perineural injection of BoNT/A might be an effective method to inhibit neuroinflammation and LTP.
It was reported that Wnt signaling plays an important role in regulating synaptic plasticity and proinflammatory cytokines release, such as TNF-a and IL-18 [35]. Therefore, whether BoNT/A played an analgesic role through inhibiting Wnt pathway was worth discussing. Wnt signaling pathways include the canonical Wnt/β-catenin pathway and noncanonical β-catenin-independent pathways, both of which are involved in the regulation of neuropathic pain [33]. In this study, we found that Wnt/β-catenin pathway was activated in spinal cord dorsal after CCI surgery, accompanied by the enhanced neuroinflammation and synaptic plasticity, which could be attenuated by perineural injection of BoNT/A. However, Wnt agonist abolished the suppressive effects of BoNT/A on neuropathic pain, Wnt/β-catenin pathway and the expression of IL-18 and pNR2B, indicating that perineural injection of BoNT/A probably inhibited spinal neuroinflammation and synaptic plasticity through regulating Wnt pathway.

Conclusion
In conclusion, we found that as a new route of administration, perineural injection of BoNT/A could relieve CCI induced neuropathic pain, and both peripheral and central mechanisms might involve in the analgesic effects. More future researches are needed to explore the underlying mechanisms further. The results of this study may provide a suitable administration route of BoNT/A for clinical treatment of neuropathic pain induced by peripheral nerve entrapment.