17β-Estradiol Attenuates Neuropathic Pain Caused by Spared Nerve Injury by Upregulating CIC-3 in the Dorsal Root Ganglion of Ovariectomized Rats

17β-estradiol plays a role in pain sensitivity, analgesic drug efficacy, and neuropathic pain prevalence, but the underlying mechanisms remain unclear. Here, we investigated whether voltage-gated chloride channel-3 (ClC-3) impacts the effects of 17β-estradiol (E2) on spared nerve injury (SNI)-induced neuropathic pain in ovariectomized (OVX) female Sprague Dawley rats that were divided into OVX, OVX + SNI, OVX + SNI + E2, OVX + SNI + E2 + DMSO (vehicle, dimethyl sulfoxide), or OVX + SNI + E2+Cltx (ClC-3-blocker chlorotoxin) groups. Changes in ClC-3 protein expression were monitored by western blot analysis. Behavioral testing used the paw withdrawal threshold to acetone irritation and paw withdrawal thermal latency (PWTL) to thermal stimulation. Immunofluorescence indicated the localization and protein expression levels of ClC-3. OVX + SNI + E2 rats were subcutaneously injected with 17β-estradiol once daily for 7 days; a sheathed tube was implanted, and chlorotoxin was injected for 4 days. Intrathecal Cltx to OVX and OVX + SNI rats was administered for 4 consecutive days (days 7–10 after SNI) to further determine the contribution of ClC-3 to neuropathic pain. Patch clamp technology in current clamp mode was used to measure the current threshold (rheobase) dorsal root ganglion (DRG) neurons and the minimal current that evoked action potentials (APs) as excitability parameters. The mean number of APs at double-strength rheobase verified neuronal excitability. There was no difference in behaviors and ClC-3 expression after OVX. Compared with OVX + SNI rats, OVX + SNI + E2 rats showed a lower paw withdrawal threshold to the acetone stimulus, but the PWTL was not significantly different, indicating increased sensitivity to cold but not to thermal pain. Co-immunofluorescent data revealed that ClC-3 was mainly distributed in A- and C-type nociceptive neurons, especially in medium/small-sized neurons. 17β-estradiol administration was associated with increased expression of ClC-3. 17β-estradiol-induced increase in ClC-3 expression was blocked by co-administration of Cltx. Cltx causes hyperalgesia and decreased expression of ClC-3 in OVX rats. Patch clamp results suggested that 17β-estradiol attenuated the excitability of neurons induced by SNI by up-regulating the expression of ClC-3 in the DRG of OVX rats. 17β-estradiol administration significantly improved cold allodynia thresholds in OVX rats with SNI. The mechanism for this decreased sensitivity may be related to the upregulation of ClC-3 expression in the DRG.

17β-estradiol plays a role in pain sensitivity, analgesic drug efficacy, and neuropathic pain prevalence, but the underlying mechanisms remain unclear. Here, we investigated whether voltage-gated chloride channel-3 (ClC-3) impacts the effects of 17β-estradiol (E2) on spared nerve injury (SNI)-induced neuropathic pain in ovariectomized (OVX) female Sprague Dawley rats that were divided into OVX, OVX + SNI, OVX + SNI + E2, OVX + SNI + E2 + DMSO (vehicle, dimethyl sulfoxide), or OVX + SNI + E2+Cltx (ClC-3-blocker chlorotoxin) groups. Changes in ClC-3 protein expression were monitored by western blot analysis. Behavioral testing used the paw withdrawal threshold to acetone irritation and paw withdrawal thermal latency (PWTL) to thermal stimulation. Immunofluorescence indicated the localization and protein expression levels of ClC-3. OVX + SNI + E2 rats were subcutaneously injected with 17β-estradiol once daily for 7 days; a sheathed tube was implanted, and chlorotoxin was injected for 4 days. Intrathecal Cltx to OVX and OVX + SNI rats was administered for 4 consecutive days (days 7-10 after SNI) to further determine the contribution of ClC-3 to neuropathic pain. Patch clamp technology in current clamp mode was used to measure the current threshold (rheobase) dorsal root ganglion (DRG) neurons and the minimal current that evoked action potentials (APs) as excitability parameters. The mean number of APs at double-strength rheobase verified neuronal excitability. There was no difference in behaviors and ClC-3 expression after OVX. Compared with OVX + SNI rats, OVX + SNI + E2 rats showed a lower paw withdrawal threshold to the acetone stimulus, but the PWTL was not significantly different, indicating increased sensitivity to cold but not to thermal pain. Co-immunofluorescent data revealed that ClC-3 was mainly distributed in A-and C-type nociceptive neurons, especially in medium/smallsized neurons. 17β-estradiol administration was associated with increased expression INTRODUCTION Neuropathic pain, a form of allodynia or hyperalgesic spontaneous pain, remains a major challenge for pain researchers and clinicians (Fukuda et al., 2017;Xu et al., 2017;Zhang et al., 2018;Ouyang et al., 2019). Inflammatory-mediator release at the site of injury triggers alterations in the properties of primary afferent neurons and increases their excitability leading to ectopic, stimulus-independent activity (Colloca et al., 2017;Alles and Smith, 2018). Changes in ion channels are responsible for development of abnormal discharge (Amescua-Garcia et al., 2018;Xu et al., 2018). Recent research has shown that intracellular Cl − concentration in DRG neurons increased after sciatic nerve section or inflammation (Funk et al., 2008;Si et al., 2019). Studies have focused on chloride channels in primary sensory neurons (PSN), as activation of chloride channels in sensory neurons may cause chloride efflux and depolarization because of high intracellular chloride concentrations (Mao et al., 2012;Bonin and De Koninck, 2013). Numerous studies have shown that anion channels, and particularly chloride channels, may be involved in the pathogenesis of neuropathic pain Si et al., 2019). Actually, downregulation of ClC-3 in DRG neurons contributes to mechanical hypersensitivity following peripheral nerve injury (Riazanski et al., 2011). Thus, modulation of ClC-3 function may be a novel therapeutic avenue for the treatment of neuropathic pain (Pang et al., 2016). Many studies examining the pathogenesis of neuropathic pain as well as its prevention and treatment strategies have suggested that the pain threshold is sex-specific (Ramirez- Barrantes et al., 2016;Vacca et al., 2016). Estrogen receptors are distributed in many pain-related regions in the central and peripheral nervous systems, and 17β-estradiol can affect the generation and transmission of pain on many levels (Amandusson and Blomqvist, 2013). It has been reported that estrogen has a palliative effect on neuropathic pain, but the underlying mechanisms are complex (Lu et al., 2013;Ma et al., 2016;Lee et al., 2018). Estrogen can activate ClC-3 via ERα in the cell membrane of osteoblasts (Deng Z. et al., 2017), promote proliferation of ER + breast cancer MCF-7 cells through the ClC-3 Cl − channel pathway (Yang et al., 2018), and regulate ion channels in pain modulation, but its effects on analgesia Abbreviations: ClC-3, chloride channel-3; DMSO, dimethyl sulfoxide; DRG, dorsal root ganglion; E2, estradiol; ER, estrogen receptor; OVX, ovariectomized; PBS, phosphate-buffered saline; PWCL, paw withdrawal cold latency; PWTL, paw withdrawal thermal latency; ShamOVX, sham ovariectomized; SNI, spared nerve injury. and promotion of pain are inconsistent (Berman et al., 2017;Ren et al., 2018). Numerous studies have reported that estrogen can provide pain relief in females (Vacca et al., 2016;Lee et al., 2018). However, there have been no systematic studies on the effects of estrogen replacement therapy on neuropathic pain in menopausal women. The present work aimed to identify whether ClC-3 plays a role in the effects of estrogen on neuropathic pain in ovariectomized (OVX) rats.

Animals
Adult female Sprague Dawley rats (10-12 weeks old, 200-250 g, n = 180) were purchased from the Animal Center of the Xinjiang Medical University (Ürümqi, China). Animal use was approved by the Committee of Animal Experimental Ethics of the First Affiliated Hospital of Medical College, Shihezi University, China. Animals were housed in plastic boxes with controlled temperature (24 ± 2 • C), humidity (40-50%), and a 12:12 h light:dark cycle. We selected rats with relatively uniform and stable baseline responses to cold and hot stimuli for the experiment. Rats were OVX bilaterally, and the sham OVX (ShamOVX) group underwent operations as previously described Chang et al., 2019). All protocols were approved by the Animal Ethics Committee of the First Affiliated Hospital of Shihezi University School of Medicine (approval No. A2018-165-01) on February 26, 2018, and were consistent with the Guidelines for the Care and Use of Laboratory Animals, published by the United States National Institutes of Health.

Surgical Procedure to Induce a Neuropathic Pain Model by Spared Nerve Injury
We used SNI to prepare a model of neuropathic pain as previously reported (Xu et al., 2017). Experimental procedures were performed on animals under anesthesia with sodium pentobarbital (40 mg/kg, intraperitoneal, Sigma-Aldrich, St. Louis, MO, United States). Care was exercised to prevent infection and reduce the impact of inflammation. After the skin was cut, the sciatic nerve and its three terminal branches were exposed directly through the part formed by the biceps muscle: the lateral side, common fibular nerve, and tibial nerves. The tibial and common peroneal nerves were cut and ligated by SNI, and the sural nerve was preserved. As the common peroneal and tibial nerves are closely connected, followed by removing the distal nerve ends about 3-5 mm. Care was taken not to damage the nearby sural nerve. After surgery, all wounds were irrigated with sterile saline and closed in layers.

Groups and Drug Intervention
All OVX rats were randomly divided into five groups: OVX, OVX + SNI, OVX + SNI + estradiol (E2), OVX + SNI + E2 + DMSO, and OVX + SNI + E2 + chlorotoxin (Cltx). For intrathecal delivery (1 µM/L, 20 µl/day, Sigma-Aldrich) (Thompson and Sontheimer, 2016), Cltx was dissolved in 30% DMSO and injected through a catheter for 4 days. Intrathecal catheters were implanted on SNI day 7 as previously described (Pogatzki et al., 2000). Briefly, a sterile catheter filled with saline was inserted through the intervertebral space at L 5 /L 6 , and the tip of the tube was positioned at the lumbosacral spinal level. Animals with hindlimb paralysis or paresis after surgery were excluded. Animals without movement disorders received lidocaine (2%) through the catheter to verify the intraspinal location. Immediate bilateral hindlimb paralysis (within 15 s) lasting 20-30 min confirmed correct catheterization. Animals without these features were excluded from subsequent experiments. DRGs for patch clamps were incubated with Cltx in vitro. The 7-day procedure of 17β-estradiol (30 µg/kg/day, subcutaneous, Sigma-Aldrich) administration was performed as previously described (Vacca et al., 2016).

Measurement of Serum 17β-Estradiol Levels
Rats were deprived of food overnight, and serum 17β-estradiol levels were assessed according to a previously described protocol (Homberg et al., 2018). Briefly, blood samples were collected from the abdominal aorta under anesthesia, and serum was separated by centrifugation at 15,000 r for 5 min. Serum corticosterone levels were measured with a corticosterone enzyme immunoassay kit (Cayman Chemical, Ann Arbor, ML, United States). Analyses were conducted in duplicate. The intra-assay coefficients of variation were lower than 10% for each analysis.

Behavioral Assays
Heat Hyperalgesia (Hot Plate Test) Thermal hyperalgesia was assessed according to a previously described protocol (Ouyang et al., 2019;Si et al., 2019). The thermal withdrawal latency in response to radiant heat stimulation was measured with an analgesia meter (Ugo Basile, Stoelting, IL, United States). Animals were placed in the chamber and allowed to acclimatize for 30 min before testing. A radiant heat source was focused under the glass floor beneath the hind paws. Thermal-stimulus intensity was adjusted to obtain a baseline thermal withdrawal latency of approximately 20 s. The digital timer automatically recorded the duration between stimulus initiation and thermal withdrawal latency, and a 30 s cutoff was used to prevent tissue damage. Each rat was tested every 5 min, and the average of six trials was used as the PWTL.

Cold Allodynia (Acetone Drop Method)
Cold sensitivity was measured by applying a drop of acetone to the plantar surface of the hind paw as previously described (Deng et al., 2015;Bergeson et al., 2016). Rats were housed and habituated for 30 min in transparent plastic boxes with a wire-mesh floor. After the adaptation period, acetone was gently applied against the plantar skin of the left hind paws with an acetone bubble formed with a 0.1-ml syringe, alternately three times to hind paw at intervals of 5 min, and the duration of licking or biting and remaining in the air was recorded. Each rat was tested every 5 min, and the average of six trials was used as the PWCL.

Sample Preparation
At the predetermined time points, the animals were deeply anesthetized with sodium pentobarbital (40 mg/kg, intraperitoneal; Sigma). Rats were sacrificed after behavioral testing was performed, and ipsilateral L 4−6 DRGs tissues were collected. Samples for RT polymerase chain reaction (RT-PCR) and western blot experiments were snap-frozen in liquid nitrogen and stored at −80 • C. Samples used for immunofluorescence imaging were perfused through the ascending aorta with saline, followed by 4% paraformaldehyde in 0.1 M phosphate buffer (4 • C, pH 7.4) as previously reported .

Immunofluorescence
The L 4−6 DRG on the surgical side was removed and fixed in 4% paraformaldehyde overnight, followed by dehydration in 20% or 30% sucrose in phosphate buffer at 4 • C. The tissue was cut into 5-µm thick sections with a cryostat (Leica CM1950, Nußloch, Germany). The sections were blocked with 20% bovine serum albumin (BSA) for 1 h in a 37 • C incubator (303-0S; Beijing Ever Bright Medical Treatment Instrument Co., Ltd., Beijing, China), washed with phosphate-buffered saline (PBS), and incubated with primary antibody (rabbit anti-ClC-3 polyclonal antibody; 1:100, 13359S, CST) overnight at 4 • C. After washing with PBS, the sections were incubated with secondary antibody (TRITC-conjugated anti-rabbit secondary antibody; 1:100; Santa Cruz Biotechnology, Heidelberg, Germany) for 1 h at 37 • C. For double immunofluorescence staining, tissue sections were incubated with a mixture of anti-ClC-3 antibody and antibodies against neurofilament-200 (NF-200; a marker for myelinated A-fibers, 1:100; ab82259; Abcam, Cambridge, United Kingdom), calcitonin gene related peptide (CGRP, a marker of peptidergic C-type neurons, 1:100; ab81887; Abcam) for 2 nights at 4 • C, or IB4 (FITC-conjugated; a marker for non-peptidergic C-type neurons, 5 µg/ml; L2895; Sigma). Except for IB4-treated tissue sections, the other sections were treated with a mixture of FITC-and TRITC-conjugated secondary antibodies at a 1:100 dilution for 1 h at 37 • C. IB4 was 1:750 mixed with TRITC-conjugated secondary antibody. The sections were rinsed with 0.01 M PBS three times, mounted on gelatincoated slides, and air dried. Immunoreactivity was visualized by fluorescence microscopy, and a negative control was used by omission of the primary antibody to confirm the specificity of the immunoreaction. Sections were observed at 200× magnification using a confocal laser scanning microscope (LSM710; Carl Zeiss AG, Oberkochen, Germany). Optical density measurements and data analysis of CLCN3-positive cells for the two types of DRG neurons were performed using Image-Pro Plus 6.0 (Media Cybernetics, Rockville, MD, United States). The percentage fluorescence results of positive neurons of three independent experiments were recorded.

Western Blot Analysis
Frozen tissues were homogenized, and proteins were extracted using a nucleoprotein and cytoplasmic protein extraction kit (Keygen Biotech, Nanjing, China) and 30 µg of protein was mixed with sodium dodecyl sulfate sample buffer. Proteins were separated on standard sodium dodecyl sulfate-polyacrylamide gel electrophoresis (8-10% gels) and transferred onto 0.45µm nitrocellulose membranes (Invitrogen, Carlsbad, CA, United States). Membranes were blocked in 5% milk for 1 h and incubated overnight at 4 • C with the following primary antibodies: mouse anti-ClC-3 (1:750 dilution; ab134285; Abcam) and anti-β-actin (1:1000 dilution, ab8226, Abcam). The next day, the membranes were rinsed with tris-buffered saline Twenty three times for 10 min and incubated with the secondary antibodies (anti-mouse immunoglobulin G against the primary antibodies). Staining was visualized using enhanced chemiluminescence (GE Healthcare, Chicago, IL, United States). Band intensities were quantified by ImageJ software (Rawak Software Inc., Germany).

Quantitative RT-PCR Analysis
Total RNA was extracted from the ipsilateral L 4−6 DRGs of rats using Trizol (Invitrogen) and reverse-transcribed to cDNA using a qRT-PCR kit (Invitrogen) according to the manufacturer's instructions (Sang et al., 2018). For each cDNA target, 2 µL aliquots of each completed reverse transcriptase reaction were amplified in a 20 µL reaction volume using SYBR Green Real Time PCR Master Mix (Toyobo Co., Ltd., Osaka, Japan) in 45 cycles of 95 • C and 60 • C for 12 s and 35 s, respectively. The following primers were used for amplification: ClC-3, 5 -ATGCTTGGTCAGGATGGCTTGTAG-3 (forward) and 5 -AGT CATCCAGTCAGCAGCAATGTC-3 (reverse); β-action, 5 -AGCAGA TGT GGATCAGCAAG-3 (forward) and 5 -AACAGTCCGCCTAGAAGCAT-3 (reverse). We used the mRNA level of β-actin as an internal control, and we ran a standard curve to determine the relative levels of each cDNA target. Relative gene expression levels were calculated using the 2 −( Ct) method. The expression level of each gene was analyzed in triplicate.

Isolation of DRG Neurons
L 4−6 DRG neurons from the ipsilateral side of the operation were dissociated using enzyme digestion as previously described . The drug-intervention group DRGs were treated with 17β-estradiol and Cltx. Briefly, the excised ganglia were freed from their connective tissue sheaths and cut into pieces with a pair of sclerotic scissors in DMEM/F12 medium (GIBCO; Thermo Fisher Scientific, Waltham, MA, United States) under low temperature on ice. The fragments were transferred into 5 mL of DMEM/F12 medium containing trypsin (0.4 mg/mL, Sigma) and collagenase (type IA, 0.6 mg/mL, Sigma) and incubated for 5 min at 37 • C. The ganglia were then gently triturated using fine fire-polished Pasteur pipettes. The suspension was dissociated in DMEM/F12 medium, supplemented with 10% fetal bovine serum, and DRG neurons were plated on glass cover slips coated with Poly-L-Lysine (Sigma). Cells were maintained in a humidified atmosphere (5% CO 2 , 37 • C) and used for electrophysiological recordings 6-24 h after plating.
FIGURE 1 | Experimental model and schedule of drug intervention. (A) Ovarian removal (OVX) and spared nerve injury (SNI) model protocol. Normal female rats underwent ovariectomy 2 weeks before SNI. Two weeks later, OVX rats underwent SNI and behavioral testing at different time points of SNI; the ipsilateral L 4-6 dorsal root ganglion was obtained as tissue sample after behavioral testing. (B) After the SNI model was established, rats were treated with 17β-estradiol for 7 days (from day 0 to day 6, 30 µg/kg/day) subcutaneously. On the 7th day of SNI, intrathecal Cltx or DMSO (1 µM/day, 20 µL) was administered for 4 days. The L 4-6 dorsal root ganglia of rats were collected on the 7th day and 10th day of SNI after behavioral testing. OVX, ovariectomy; SNI, spared nerve injury; E2, 17β-estradiol; Cltx, Chlorotoxin; DMSO, vehicle, dimethyl sulfoxide.

Electrophysiological Recordings
All recordings were performed on small and medium diameter (20-35 µm) neurons as previously described (Chen et al., 2011). Coverslips with DRG neurons were mounted in a small flow-through chamber positioned on the stage of an inverted microscope (Nikon Eclipse Ti, Tokyo, Japan) to select DRG cells with smooth membrane surfaces and good translucency for experiments. Coverslips were continuously perfused with gravity-driven bath solution. Standard whole-cell patch-clamp recordings from isolated DRG neurons were performed at room temperature (22 • C) using an EPC-10 amplifier and the PULSE program (HKA Electronics, Lambrecht, Germany). The membrane capacitance was read from the amplifier by PULSE to measure the size of cells and current densities. Glass pipettes (3-5 M ) were prepared with a Sutter P-87 puller (Sutter Instruments, Novato, CA, United States). Action potentials were elicited by a series of depolarizing currents from 0 to 500 pA (150 ms) in 50-pA step increments under the current clamp mode to measure the current threshold (rheobase) in the vicinity of the explosive action potential current. The current was altered by 10 pA per step, i.e., the minimal current that evoked an action potential, as a parameter for excitability. The recorded signal was amplified by a MultiClamp 700B amplifier (Molecular Devices, LLC, Sunnyvale, CA, United States), filtered at 10 kHz, and converted by an Axon Digidata 1550A D/A converter (Molecular Devices) at a sampling frequency of 10 to 20 kHz. Voltage errors were minimized by using 80-90% series resistance compensation, and linear leak subtraction was used for all recordings. For the current clamp experiments, the bath solution contained (in mM): 140 NaCl, 5 KCl, 2 CaCl 2 , 2 MgCl 2 , 10 D-glucose, 10 HEPES; the pH was adjusted to 7.4 with NaOH. The pipette solution contained (in mM): 30 KCl, 100 K-aspartate, 5 MgCl 2 , 2 Mg-ATP, 0.1 Na-GTP, 40 HEPES; the pH was adjusted to 7.2 with KOH. All chemicals were obtained from Sigma.

Statistical Analysis
All data are expressed as mean ± SEM of three independent experiments. The normal distribution hypothesis of the test data and the homogeneity of variance were examined before further statistical analysis. Statistical analysis was performed using SPSS 10.0 (SPSS Inc., Chicago, IL, United States). PWCL and PWTL were analyzed using repeated-measures analysis of variance, and multiple comparisons between groups at each time point were conducted using Bonferroni's post hoc tests. Regarding the western blot, PCR, and patch-clamp data, analysis among multiple groups was carried out by one-way analysis of variance followed by Tukey's post hoc tests. Student's t-test was used for two-group comparisons. P < 0.05 was considered statistically significant.

The Established OVX Model Had No Effect on Cold and Thermal Hypersensitivity
Normal female rats underwent OVX 2 weeks before SNI ( Figure 1A). Behavioral tests showed that the sensitivity to cold and heat stimulation had remained unchanged 2 weeks after OVX (Figures 2A,B), and ClC-3 expression in DRG neurons did not change significantly within these 2 weeks ( Figure 2E). Estrogen levels were measured in rat blood samples collected from the abdominal aorta under anesthesia before and after ovarian resection. The results showed that 17β-estradiol levels were significantly lower in the OVX group compared to pre-OVX ( Figure 2F; 11060 ± 1113 in the pre-OVX vs. 240.1 ± 38.07 in the OVX group, P < 0.001; n = 6 in each group).

Development of Cold and Thermal Hypersensitivity After SNI Treatment in OVX Rats
An OVX + SNI model was used to stimulate neuropathic pain in menopausal female rats. These rats showed painsensitizing behaviors such as paw protection, paw licking, and dorsiflexion (data not shown). Behavioral tests showed that OVX + SNI rats developed significant cold hyperalgesia. The increased sensitivity to cold stimulation started on the 3rd day after SNI and lasted until the end of behavioral testing ( Figure 2C and Supplementary 2.117 ± 0.6256, P < 0.001; n = 6 in each group). There was no significant change in thermal stimulation ( Figure 2D).

ClC-3 Was Mainly Expressed in Medium/Small-Sized DRG Neurons of OVX Rats
Immunofluorescent double staining experiments showed that ClC-3 protein colocalized with IB4, CGRP, and NF-200 ( Figure 3A). The percentages of IB4-, CGRP-, and NF-200positive neurons relative to the percentage of ClC-3-positive cells were 34.47 ± 1.602%, 25.43 ± 1.267%, and 35.41 ± 1.552%, respectively (n = 6 in each group; Figure 3B). These results showed that ClC-3 was mainly located in A-and C-type neurons in the DRG. The neuronal diameter size ranges of IB4, CGRP, and NF-200 were 31.00 ± 1.13, 17.75 ± 0.87, and 42.75 ± 1.917, respectively ( Figure 3C; n = 10 in each group). ClC-3 expression, mainly in medium/small-sized as well as in large DRG neurons, indicated that ClC-3 may be involved in the regulation of superficial sensations such as pain.

17β-Estradiol Decreased the Excitability of DRG Neurons Caused by SNI in OVX Rats When Blocked by Cltx
To examine why 17β-estradiol decreased the excitability for cold sensitivity caused by SNI in OVX rats, we examined the characteristics of the APs of DRG neurons. APs were elicited by a series of depolarizing currents from 0 to 500 pA (150 ms) in 50-pA step increments under the current clamp mode to measure the current threshold (rheobase), i.e., the minimal current that evoked an action potential, which was used as a parameter for excitability (Figures 7A-F). All DRG neurons from OVX rats were harvested on day 10 of SNI, with or without 17β-estradiol administration; DRGs for patch clamps were incubated with Cltx in vitro. The data suggested increased excitability of DRG neurons after SNI. Similarly, the voltage threshold of the APs in the OVX + SNI group was significantly lower than that in the OVX group. 17β-estradiol decreased excitability as it was blocked by Cltx ( Figure 8A; OVX + SNI group vs. OVX group, 91.67 ± 15.37 vs. 300 ± 18.26, P < 0.001; OVX + SNI + E2 group vs. OVX + SNI group, 250 ± 18.26 vs. 91.67 ± 15.37, P < 0.001; OVX + SNI + E2 + Cltx group vs. OVX + SNI + E2 group, 100 ± 12.91 vs. 250 ± 18.26, P < 0.01; n = 6 in each group). The mean number of APs at double-strength rheobase (2 rheobase) was higher in the OVX + SNI group (Figure 8B). When 17β-estradiol was administered, the number of APs decreased under double-strength rheobase stimulation, and increased after intrathecal Cltx administration (Figure 8E; OVX + SNI group vs. OVX group, 17.5 ± 0.4282 vs. 2.167 ± 0.4773, P < 0.001; OVX + SNI + E2 group vs. OVX + SNI group, 4.333 ± 0.4944 vs. 17.5 ± 0.4282, P < 0.001; OVX + SNI + E2 + Cltx group FIGURE 7 | 17β-estradiol attenuated increased excitability of DRG neurons in spared nerve injury ovariectomized rats and was inhibited by Cltx. Current threshold (rheobase) was determined as the current required for activating the first action potential. (A-F) On the right, representative traces of action potentials (APs) evoked by current injections into DRG neurons from OVX, OVX + SNI, OVX + SNI + E2, OVX + SNI + E2 + DMSO, and OVX + SNI + E2 + Cltx groups; n = 6 per group; On the right, twice in the figure, the number of action potentials produced at the corresponding 2 × rheobase.
vs. OVX + SNI + E2 group, 18.83 ± 0.4773 vs. 4.333 ± 0.4944, P < 0.01; n = 6 in each group). Other action potential parameters such as membrane capacitance, resting membrane potential, and magnitude of APs were not significantly different between the groups (Figures 8C,D). Furthermore, the size of all neurons was between 20-35 µm ( Figure 8B). Administration of control solution had no effect on the rheobase and APs.

DISCUSSION
This study reported that ClC-3 expression in DRG neurons was not significantly changed 2 weeks after OVX. However, according to the literature, mechanical pain was observed 5 weeks after simple OVX and there were also observed changes in pain-related proteins (Amandusson and Blomqvist, 2013;Jiang et al., 2017). We can confirm that OVX has no effect on ClC-3 expression before SNI in this study. ClC-3 is distributed in the central nervous system (Riazanski et al., 2011) and, in this study, its expression decreased following SNI in OVX rats. Notably, ClC-3 was expressed at high levels in DRG cells, especially in medium/small-sized neurons. It was reported that in C57BL/6J mouse DRG neurons, ClC-3 is expressed at a high level especially in small size neurons (Pang et al., 2016). An SNI model was established 2 weeks after OVX; the induced neuropathic pain tended to begin on the 3rd day of SNI and to persist until the 21st day. It was reported that, in male rats, SNI-caused neuropathic pain lasted longer (Vacca et al., 2016). This indicates that OVX may affect SNI-induced neuropathic pain to some degree. However, hyperalgesia and decreased ClC-3 expression in OVX SNI-treated rats were reversed by 17β-estradiol replacement.
DRG cells were used for all patch clamp experiments. After the establishment of the SNI model, the reduction in ClC-3 expression decreased the activation rheobase of APs and increased the membrane input resistance in DRG neurons. Therefore, the same current injection induced more APs in the DRG neurons of the OVX + SNI group. Decreased ClC-3 expression did not affect cell membrane capacitance, resting membrane potential, or the amplitude of APs in DRG neurons. These findings indicate that increase in the excitability of DRG neurons contributes to hypersensitivity of primary afferent neurons to cold stimulation in OVX + SNI rats.
When 17β-estradiol was administered, the increase in excitability was attenuated. Conversely, excitability increased after administration of both 17β-estradiol and Cltx, a ClC-3 specific blocker. The most likely ion channel internalization by Cltx in gliomas is ClC-3 ( Thompson and Sontheimer, 2016). Cltx, which binds to ClC-3 with MMP-2/MT1-MMP, forms a macromolecular protein complex on the cell membrane surface that indirectly affects the action of the chloride channel (Deshane et al., 2003;Thompson and Sontheimer, 2016). It differs from NPPB in inhibiting ClC-3 ion channels, as NPPB blocks the function of the ClC-3 ion channel, while Cltx reduces the number of functional chloride channels on the cell membrane surface. Regardless, Cltx was found to cause internalization of ClC-3 into caveolar rafts 15 min after its application (McFerrin and Sontheimer, 2006;Thompson and Sontheimer, 2016;Wang D. et al., 2017). In this regard, 17β-estradiol upregulated ClC-3 in DRG neurons of SNImodel rats at both the gene and protein levels; however, after 17β-estradiol and Cltx were administered, ClC-3 mRNA levels were not significantly decreased compared to those in the 17β-estradiol-administered group (Figure 9). This suggests that 17β-estradiol may affect the expression of ClC-3 at the gene level, thus increasing the sensitivity to cold stimulation by FIGURE 9 | Schematic of potential 17β-estradiol-mediated mechanisms of neuropathic pain regulation in ovariectomized female rats. 17β-estradiol may regulate the expression of ClC-3 mRNA through the 17β-estradiol receptor, including nuclear receptors α and β or membrane receptor GPER, via certain signaling pathways in the cell, thereby affecting the expression of ClC-3 mRNA and protein and relieving neuropathic pain. However, this effect may be blocked by Cltx, a specific blocker that affects the function of ClC-3 channel protein but does not affect the expression of mRNA. GPER, G protein-coupled 17β-estradiol receptor. ER-α and ER-β: nuclear 17β-estradiol receptors α and β.
affecting the excitability of DRG neurons. Interestingly, when Cltx was used in the control OVX group, there were observed behavioral changes in cold allergy, and the allergic reaction increased. That was further verified that estrogen likely regulates neuropathic pain in OVX rats through ClC-3. It is valuable to note that in OVX + SNI rats, Cltx showed limited effects on ClC-3 protein expression and hyperalgesia, this phenomenon indicates that there are other regulatory mechanisms to be studied. The existing literature on the role of 17β-estradiol is inconsistent; both nociceptive and anti-nociceptive 17β-estradiol effects have been reported (Vacca et al., 2016;Sorge and Totsch, 2017;Stinson et al., 2019). Furthermore, the results may also depend on 17β-estradiol levels and the structures and systems involved (Craft, 2007;Vacca et al., 2016). Pathological pain can be divided into inflammatory, cancer, and neuropathic (Amandusson and Blomqvist, 2013). Evidence suggests that 17β-estradiol may promote inflammatory pain but has a therapeutic effect on sexual pain (Ma et al., 2016;Vacca et al., 2016); it can also alleviate neuropathic pain caused by chemotherapy through different ERs (Ma et al., 2016;Kramer et al., 2018). Many studies have previously reported that 17β-estradiol can regulate the expression of pain-related proteins in the central nervous system and peripheral neurons such as DRG cells, thereby alleviating SNI-induced neuropathic pain and associated anxiety (Lu et al., 2013;Small et al., 2013;Liu et al., 2015;Ramirez-Barrantes et al., 2016;Vacca et al., 2016;Xu et al., 2017;Lee et al., 2018). Further, 17β-estradiol reduces pain thresholds in neuropathic rats by increasing the expression of NMDAR1 (Deng C. et al., 2017). The pathogenesis of neuropathic pain is mainly underpinned by changes in ion channels that influence APs (Scholz et al., 2019). A previous study reported that altered activity resulted in changes in the properties and/or expression of various types of ion channels, such as voltagegated Na + , K + , and Ca 2+ channels (Waxman and Zamponi, 2014;Daou et al., 2016); however, the role of anion channels remains unclear.
A recent report indicated that ClC-3 is a member of the voltage-gated chloride channel family; its deletion caused increased excitability of DRG cells and decrease in the mechanical pain threshold in rats and mice (Pang et al., 2016). ClC-3 belongs to the ClC voltage-gated chloride channel Superfamily and includes two different functional groups: voltage-gated chloride channels and Cl − /H + reverse transporters (Deshane et al., 2003;Riazanski et al., 2011;Liu et al., 2013;Hong et al., 2015). According to previous reports, estrogen may alleviate neuropathic pain (Vacca et al., 2016;Lee et al., 2018). It has been reported that estrogen reduces the pain threshold in males, likely due to its sexually dimorphic actions (Alabas et al., 2012;Bereiter et al., 2019). In neutered females, estrogen has analgesic effects that may be mediated by ClC-3. Previous investigations reported that pain involves two effects that may occur at different times. There are no reports of estrogen increasing pain sensitivity; however, when estrogen levels increase during pregnancy, pain sensitivity is known to decrease, and oophorectomy results in hyperalgesia in mice subjected to mechanical and thermal tests (Amandusson and Blomqvist, 2013;Berman et al., 2017;Ren et al., 2018;Yousuf et al., 2019). However, more studies have favored the antagonistic effect of estrogen on pain (Gintzler and Liu, 2012;Bálint et al., 2016;Kramer et al., 2018). Future studies, performed with OVX female rats or mice, should investigate the role of ClC-3 in the 17β-estradiol-mediated effects on SNIinduced neuropathic pain in OVX animals. These investigations will provide more evidence for the multifarious effects of estrogen on pain. Indeed, this study did not assess compensatory mechanisms caused by dysfunctional hormonal conditions. ERs are widely distributed in the nervous system (Tang et al., 2014;Bálint et al., 2016;Nourbakhsh et al., 2018;Li L. et al., 2019;Liu et al., 2019). Reportedly, estrogen could influence the expression of P2X3 via ERα and GPR30 to affect neuropathic pain, which may be mediated through the ERK pathway (Lu et al., 2013). Future studies may confirm the mechanisms by which ClC-3 regulates ERs. The results of this study provide a new direction for new treatments in the clinical treatment of neuropathic pain in menopausal women.

CONCLUSION
In conclusion, our results showed the complex interactions involved in estrogen-induced pain regulation and revealed the potent role of 17β-estradiol in neuropathic pain, which was altered in female OVX rats. Estrogen may decrease sensitivity to cold stimulation through increased ClC-3 expression in rats experiencing chronic neuropathic pain 2 weeks after OVX.

DATA AVAILABILITY STATEMENT
The datasets generated for this study are available on request to the corresponding author.

ETHICS STATEMENT
The animal study was reviewed and approved by Institutional Animal Care and Use Committee of the Medical College of Shihezi University.

AUTHOR CONTRIBUTIONS
J-QS, Z-ZX, and L-CZ conceived and designed the experiments. Z-ZX conducted the experiments. Q-YC, S-YD, MZ, and C-YT helped with the experiments. Z-ZX and YW analyzed the data. Z-ZX and J-QS wrote the manuscript. All authors discussed and commented on the manuscript.

FUNDING
This work was supported by the National Natural Science Foundation of China (Grant Nos. 30160026 and 81960188). The funding sources had no role in study design, conception, analysis, or interpretation of data, writing, and deciding to submit this paper for publication.

ACKNOWLEDGMENTS
This study was performed at the Key Laboratory of Xinjiang Endemic and Ethnic Diseases of Xinjiang Provincial Department of Physiology, Shihezi University School of Medicine.