Animals

C57BL/6 female mice 7–8 weeks old (Guangdong Medical Laboratory Animal Center, China) were pre-habituated to animal facilities (a 12/12 light dark cycle, 23 ± 1 ℃, and ad libitum food and water) for one week prior to experimental and behavioral testing. Previous evidences have shown that ketamine exerts enhanced antidepressant actions in female rodents compared to males [19, 20]. In addition, 2R, 6R-HNK in the brains of female mice is approximately three times higher compared to that of males when the levels of ketamine and norketamine were equivalent [10]. Therefore, only female mice were chosen in this study. Excluding cell experiments, a total of 220 mice were used in the in vivo experiments. The animals used in different experimental processes are shown in Table 1. Some animals that violated the guidelines of tests were excluded from results (e.g., showed no interest in any object during the New-object recognition test). All the behavioral tests were performed by two experimenters, one blinded to drugs and/or surgical treatments. The experimental protocols and animal handling procedures were approved by the Institutional Animal and Use Committee (IASUC), Sun Yat-sen University (Nos. SYXK (yue) 2017–0081 and 2022–0081).

Table 1 The animals used in different experimental processesDrug administration

The drugs included 2R, 6R-HNK (SML1873, Sigma, USA), S-Ketamine (Esketamine, Jiangsu Hengrui Medicine Co., Ltd, China), formaldehyde (252,549, Sigma-Aldrich, USA), menthol (M2772, Sigma), Allyl-isothiocyanate (HE-3277, Beijing Henghui Bio Co., China), capsaicin (21,750, Sigma), capsazepine (GC17918, Glpbio, USA), KCL (P816354, Macklin, China) or vehicle (saline in vivo or HHBS in vitro).

It has been reported that i.p. injection (10 mg·kg−1) of 2R, 6R-HNK also turn out to be more effectively analgesic than ketamine in nerve-injury neuropathic pain, postoperative pain and CRPS pain models in preclinic study via a μ-opioid receptor-independent pathway [18], making it a putative anodyne candidate either. Besides, studies have shown that 2R, 6R-HNK concentrations in plasma and brain tissue rapidly peak within 30 min after intravenous (i.v.)/per os (p.o.) administration and then decline fast, which is similar to the liver metabolic pharmacokinetic properties of Ketamine and its high lipid solubility [14, 21]. Based on the previous study, 7 ~ 21 μM i.t. injection of 2R, 6R-HNK is equivalent to i.p. injection of 10 mg·kg−1 [22], which is also the concentration for antidepressants [10, 23]. Therefore, to figure out the inhibitory effect of 2R, 6R-HNK on spinal nociception on CPP mouse model, we not only examined the antinociceptive effects of i.t. (7, 21 and 42 μM) and traditional i.p. (10 mg·kg−1) administrations, but also compared their actions with S-Ketamine. As for the calcium imaging and cell cultures, our medicine timing and dosage referred to previous literature [24, 25] and were described in detail below.

Adult female mice (20 ± 2 g) were anesthetized with 1.5% isoflurane and injected intrathecally with different drugs in a volume of 10 μl with a Luer-Tip Hamilton syringe at the level of the pelvic girdle. Given that cerebrospinal fluid volume was about 200 μl in each mouse, intrathecal drugs were formulated to 20 times the working concentration in saline, divided and stored at -20 °C before injection. Menthol (420 μM i.t.; 300 μM in vitro) was conserved in equal proportion as effective 2R, 6R-HNK concentration (42 μM i.t.; 30 μM in vitro) from vitro to vivo.

HFS-induced acute and chronic pain model

Protocols of surgical preparation for HFS-induced acute (within 7 days) and chronic (more than 2 weeks) pain model in mice have been described in our previous study [6]. Briefly, mice were anesthetized with 2% isoflurane anesthesia and the left sciatic nerves were dissected for electrical stimulation with Pt electrodes. HFS at 10 V (HFS: 100 Hz, 0.5 ms, 100 pulses given in 4 trains of 1 s duration at 10 s intervals) was delivered to the left sciatic nerve. In the sham group, only the sciatic nerve was exposed. At the end, the muscle and skin were sutured in two layers.

LF-PENS-induced acute and chronic pain model

To improve HFS-induced acute and chronic pain model without skin and muscle injuries, we chosen low-frequency percutaneous electric nerve stimulation (2 Hz, 0.5 ms, 10 V, a total of 120 pulses within 1 min) to induce acute and chronic pain model for CPP. Under 2% isoflurane anesthesia, the back skins of mouse left thigh were shaved and LF-PENS was percutaneously delivered to the popliteal fossae with Pt electrodes. Saline controls were only shaved after anesthesia.

Mechanical sensitivity (von Frey) test

Mice had been continuously acclimated to the test environment and the experimenter for 0.5 h since 3 d prior to formal test. To test mechanical allodynia, mice were placed in the 10 × 10 × 10 cm3 isolation rooms on elevated wire grid for 0.5 h to be peaceful but not sleepy. A set of von Frey filaments (0.04–1.4 g; North Coast medical, USA) was used to mechanically stimulate the plantar surface. Each filament was applied vertically to the lateral and medial plantar surface of the paw, and withdrawal response evoked in up to 5 s/at least one of five repetitive stimuli was positive. Mechanical paw withdrawal threshold (PWT) was determined using the up-down method. Mice with pre-surgery/treatment basal threshold (Bas) less than 0.6 g were excluded [26]. We used analgesic efficiency to evaluate the inhibition efficiency of the drug. Rate of change in mouse PWT = (ln2ln(PWT))/(ln2 - ln0.02) × 100, analgesic efficiency = (PWT change rate in model group—PWT change rate in treated group) / PWT change rate in model group × 100.

Open field test (OFT)

The mice had been given pre-adaption for 0.5 h per day in a 40 × 40 × 40 cm3 open acrylic box in the last 3 d before test. During the formal experiment, mice were put in the center of the open field or facing the wall. Their tracks, average speeds and times both in the central (30 × 30 cm2) and peripheral area within 10 min were recorded and analyzed with the tracking software.

Hargreaves test

Hargreaves test was measured by the Plantar Test Apparatus (390G, IITC Life Science, USA). Mice had been given pre-adaption in single home cages about 15 × 15 × 15 cm3 for 1 h per day since 3 d before test. Radiation intensity was adjusted to 30, so that the heating temperature reaches about 52 ~ 55 °C within 20-s-trial [27]. In the formal experiment, the mice were placed in the isolation rooms on glass for 0.5 h. With the aiming spot of the transmitter aiming at the middle of foot, the irradiation was turned on, and the latency of shrinking, lifting, licking feet, or jumping was recorded (no response during trials was recorded as 20 s). The test would be repeated 3 times on each rodent foot after no-less-than-5-min interval.

Tail-flick test

To further evaluate CPP thermal pain, tail-flick test, a commonly used measure of nociception in animals [28], was also performed in this study. The positive tail withdrawal is not only a spinal reflex, but also involved higher brain centers [29]. According to a previous study [26], the cylinder limiter (50 ml) was used to limit mice in size with their tails exposed. Mice were acclimated to the limiter prior to formal test until not be obviously resisting to enter and struggling inside. In the formal experiment, the limiter was held in hand and closed by thumb, leaving the tail hang down naturally. 1/3 to the distal of the tail was drown into the thermostatic water bath (50 °C), and the latency to evoked tail flick reaction within 15 s was recorded (not-reaction was counted as 15 s). The experiments were repeated for 3 times at each temperature with an interval of more-than-1-min, and the tails were wiped until dry at the end of each test.

Acetone test

After 3 d of continuous acclimatization, mice were placed in the 10 × 10 × 10 cm3 isolation rooms on elevated wire grid for 0.5 h to be peaceful but not sleepy. Then, acetone in syringe was sprayed into the plantar surface of each mouse through blunt-plastic-needle. The latency of reaction and corresponding behavioral score were recorded within 3 repeats between no-less-than-1-min interval (foot lifting/shaking/shaving counted 1 while foot licking counted 2, and 1 additionally counted for behavioral continuity) [26, 30].

Spontaneous pain behavior test

After acclimatization, mice were placed in single transparent home cages about 15 × 15 × 15 cm3 to record the spontaneous behaviors during 20-min-exploration/40-min-resting period. According to a previous study [31], spontaneous pain related behaviors of mice were recorded and analyzed (foot retraction/ lifting counted 1 and licking counted 2).

Sucrose preference test (SPT)

Before SPT, mice were habituated to water deprivation but food ad libitium for 24 h (from 8:00 PM on the 7th/20th d after LF-PENS to 8:00 PM on 8th/21th d). The purpose of water deprivation is to increase the times and total amount of liquid intake within 24 h of the test, improve the difference among groups, and try to avoid the error caused by insufficient intake. For the next 24 h, the mice were placed in individual compartments (15 × 15 × 15 cm3) separated by 0.5 × 0.5 cm2 wire in plastic cages with adequate food and bedding materials, so that the mice could sense and communicate with their companions. Distilled water and 2% sucrose solution were loaded in pairs of water feeder separately. The bottles of sucrose solution/water were weighed before and after the later 24-h test to calculate liquid consumption. According to the previous study [27], preference was defined as a percentage of sucrose intake to total volume consumed during trials.

Tail suspension test (TST)

Mice were hanging over with tail stuck by medical tape at 1 cm to the distal and body naturally hanged down, and the distance from the fall protection is not less than 50 cm. Within 5 min their struggling and stable time were recorded on a white background. Mice were also recorded as stationary due to inertial swing. If a mouse successfully escaped by grasping its tail, the animal will be excluded [32, 33].

Forced swimming test (FST)

The FST was performed in a transparent glass beaker with inner diameter of 30 cm, containing water about 25 °C. Water was changed between rodents and water level was no less than 20 cm for bottom contactless [27]. Animals would have 10 min swim-adaption 3 d before test, and be forced swimming for 5 min in formal trials. All mice would dry by electric fire and their behaviors in test were recorded directly above the containers. The videos were then analyzed by tracking software.

New-object recognition test (NORT)

We performed NORT based on a previous study [34]. Since 3 d before test, the mice had been given pre-adaption for 0.5 h per day in a 40 × 40 × 40 cm3 open acrylic box, and objects A and B with limited edges and corners (balls and cylinders) were selected as new objects for identification. In the formal experiment, two identical objects A were fixed in the central area of the box about 10 cm apart, and mice were put into the box in turn facing the wall. Combined with the trajectory tracking software, the movement of mice within 5 min and their interacting time with two objects were recorded (included sniffing, pushing, climbing and other behaviors, and single interaction time less than 1 s was counted as 1 s). After taking out the mice, the ones who spent more time on were reserved, while the other were replaced with new objects B. The mice were put into the box once again facing the box wall, and so did the recorder work. The box would be wiped with 70% ethanol solution after each trial. Interacting time to objects A and B within the latter 5 min were analyzed to calculate the new object preference, which was defined as a percentage of B-interacting time to total interaction during trials.

Elevated plus maze test (EPMT)

In this experiment, pre-adaption was abandoned to get rid of bias caused by memory. Mice were put in the center of a plus maze (each arm 30 × 5 cm2), elevated one meter above the floor with two face-to-face open arms and two closed arms (with 20 cm-tall walls on both sides) at less than 100 lx. Each mouse was heading to open arms at first and recorded with videotapes in 5 min. The plus maze would be wiped with 70% ethanol solution after each trial. The times of rodents’ entrance of different arena [35] were evaluated and statistically analyzed.

Immunofluorescent staining (IF)

Mice were perfused intracardially with 20 ml PBS (4 °C, pH = 7.4) followed by 20 ml 4% paraformaldehyde (PFA, 4 °C; Sigma) in PBS. Brains, L4 DRGs and L4 ~ 5 spinal cord were harvested and post-fixed for 4 ~ 6 h. After dehydration with 30% sucrose in PBS at 4 °C, all the tissues were sliced into 25 μm (brain),16 μm (DRG),18 μm (spinal cord) sections using a cryotome (CM3050S, Leica, Germany), transferred on to Superfrost Plus Microscope slides (FD Neuro Technologies, Inc, USA). After 3 washes in PBS, the cultured DRG neurons on coverslips were fixed with 4% PFA for 30 min and then washed in PBS for another 3 times without dehydration. Next, the slices/neurons were blocked with 5% donkey serum in 0.3% Triton X-100 (Sigma) for 60/30 min at room temperature (RT), and then incubated for 18 h on cradle at 4 °C with a mixture of primary antibodies: rabbit anti-c-Fos antibody (1:500, 2250S, Cell Signaling Technology (CST), USA), rabbit anti-p-ERK antibody (1:500, 4370, CST), mouse anti-GFAP antibody (1:500, 3670, CST), goat anti-Iba1 antibody (1:1000, ab5076, Abcam, USA), goat anti-CGRP antibody (1:1000, NBP3-00520, Novus, USA), mouse anti-CGRP antibody (1:2000, ab81887, Abcam), rabbit anti-TRPA1 antibody (1:500, SAB2105082, Sigma), rabbit anti-TRPV1 antibody (1:1000, GTX54762, Genetex, USA), mouse anti-VGLUT2 antibody (1:100, MA5-27,613, Invitrogen, USA). Following 3 washes, the sections were then incubated with secondary antibodies (Alexa Fluor 488, 555, 647; Life Technologies, USA) at RT for 60 min and rinsed for another 60 min. The slices were protected by Antifade Mounting Medium with DAPI (P0131, Beyotime, China) under coverslips and fluorescent images were obtained with a fluorescence microscope (EVOS FL, Thermo Fisher Scientific, USA).

The ranges of nuclei of interest were determined by mouse brain atlas (for example, the ACC was 0.62 ~ 0.14 mm before bregma and other brain nuclei were 1.46 ~ 1.94 after bregma). 5 mice were used for analysis in each group, and 4 slices with strong positive signal within the limited range of the nuclei (Fig. 4a, red boxes) were blindly selected from the 6–10 stained ones in each group for statistics. The total number of c-Fos, p-CREB, p-ERK and other positive immunoreactive cells were calculated in each section by ImageJ software (National Institutes of Health, Bethesda, MD). According to the previous study [6], the intensity of the positive immunoreactivity was quantified using relative optical density (RelOD) by the ImageJ image processing. The intensity in sample from sham, saline or control tissues was set as 1 or 100% baseline. The IF quantitative statistical methods in L4 DRGs and L4 ~ 5 SDHs were the same as above.

RT-qPCR

L4 DRGs, L4 ~ 5 SDHs or cultured DRG neurons were extracted and homogenized in RNAzol® RT (RN190-200, MRC, China). RNA was isolated using RNAzol/double-free water extraction and cDNA was prepared from total RNA by reverse transcription reaction with Evo M-MLV RT Premix (AG11706, Accurate Biology, China). qPCR was performed with CFX 96 touch (C1000™, Bio-rad, USA) using 2 × Master qPCR Mix SYBR Green I (TSE201, TSINGKE, China) in following conditions: 95 °C for 30 s; 40 cycles of 95 °C for 5 s, 60 °C for 30 s and melting analysis at last. Primers sequences were as follow (Table 2), and relative mRNA expressions were normalized to Actb in each group and the ratios from saline/sham/control group were set as baseline 1.

Table 2 Primers for RT-qPCR in tissues and cellsWestern blot (WB)

Mice were subjected to deep anesthesia by urethane, followed by transcardial perfusion with cold PBS. Ipsilateral L4 ~ L5 DRGs and SDHs were carefully dissected, homogenized, and sonicated in a lysis buffer containing a protease inhibitor cocktail (Cat# P1045, Beyotime Biotechnology). For protein analysis, 15 μg total proteins were loaded into each well, separated using SDS-PAGE, and subsequently transferred to PVDF membranes. The membranes were blocked with a solution of 5% bovine serum albumin in TBST (Tris-buffered saline, 0.1% Tween 20) at RT for 1 h. Next, they were incubated overnight at 4 °C with primary antibodies including Rb anti-p-ERK (#4370, Cell Signaling Technology, 1:1000), Rb anti-TRPA1 (SAB2105082, Sigma, 1:1000), Rb anti-TRPV1 (GTX54762, Genetex, 1:200, and Ms anti-β-actin (ab170325, Abcam, 1:1000). Subsequent steps involved incubating the blots with horseradish peroxidase-conjugated IgG at RT for 1 h. Protein bands were detected by Omni-ECL™Femto Light Chemiluminescence Kit (SQ201, Epizyme) and then captured by Tanon-5200 Chemiluminescent Imaging System (Tanon Science and Technology, China). The integrated optical density of each immunoreactive band was measured Software ImageJ 1.51j8, and then normalized to β-actin. The protein X/β-actin ratio in the saline group was established as the baseline for comparison.

DRG neuronal culture

3-week C57BL/6 female mice (Guangdong Medical Laboratory Animal Center, China) were made unconscious under isoflurane anesthesia (5%) and then decapitated immediately. All the DRGs were dissected out, cut into pieces and transferred to DMEM/F12 (C11330500BT, GIBCO, USA). The samples were digested with 5 mL DMEM/F12 containing collagenase (3 mg, C9891, Sigma) and trypsin (2 mg, T9201, Sigma) for 20 ~ 25 min and then the same volume of complete medium (DMED/F12 + 10% FBS + 1% penicillin and streptomycin (100 × , GIBCO)) was added to stop the digestion. Next, the samples were filtrated (100 μM, BS-100-CS, Biosharp, China) and centrifuged for twice (1000 g × 5 min, 20 °C), and then resuspended into single cells.

For calcium imaging, these cells were cultured in 0.01% PLL-coated (Poly-L-Lysine, P4832, Sigma) 96-well plates in complete medium in humid conditions and 5% CO2 at 37 °C for 6 h. For TRPA1 agitation test in vitro, DRG neurons were incubated in coated 24-well plates for 72 h in complete medium containing 2% B-27 Supplement 50 × (17,504–044, GIBCO), which were renewed every day. 5 μg /mL Ara-C (C2035, TCI, China) was added at 24 h to remove irrelevant cells [36]. The time course and concentration of 2R, 6R-HNK (30 μM, at 48 h), menthol (300 μM, at 48 h), capsazepine (10 μM, at 48 h + 30 min) and formaldehyde (10 μM, at 48 h + 30 min) referred to our results or previous research [36].

Calcium imaging

DRG cells were incubated with Fluo-8 Calcium Flux Assay Kit (ab112129, Abcam) in 37 °C CO2 incubator and RT for 30 min without light exposure, and then fluorescence dye was rinsed and replaced with 50 μl HHBS (CB1048, G-CLONE, China) [24, 25]. Then, calcium imaging was carried on Automatic High-pass Living Cell Imaging Analysis System (lionheart FX, BioTek, USA) in the following conditions: 37 °C, 5% CO2 in dark, with cylinders of Auto Sampler filled with 10 × drugs, containing 1 mM Allyl-isothiocyanate (AITC, an agonist of TRPA1 receptor, 100 μM, 120 or 30 s), 10 μM capsaicin (Cap, an agonist of TRPV1 receptor, 1 μM, 30 s) and 1.44 M KCL (a broad agonist of neurons, 144 mM, 30 s). Cells turned out to be overlapping or losing of integrity, as well as floating during experiments were excluded. To avoid interference from inactive/dead cells, the DRG neurons with calcium activity in response to KCL were used for statistical analysis. Besides, those turned out to be inactive (ascending average fluorescence intensity/baseline < 10%) to both AITC/Cap and KCL were classified as meaningless.

In details, blank control was added with HHBS, and the experimental group was pre-incubated by 2R, 6R-HNK (3, 10, 30 μM in HHBS) for 3 h before being put on the machine. Ex/Em = 490/525 nm, and the baseline was recorded for 15 s at 3 s/sheet, followed by AITC or caspaicin stimulation, and recorded for 15 s at 30 ms/sheet. Finally, cells were recorded for another 30 s at 30 ms/sheet after 2nd stimulation from KCL or capsaicin. The interval between administrations is 2 min/80 s. Fluorescence intensity was analyzed and statistically analyzed using Graphpad Prism.

Statistical analysis

All data were presented as mean ± SEM. Statistical analysis was calculated using GraphPad Prism Version 8.3.0 (GraphPad Software, LLC, CA, USA). Level of significance was set at p < 0.05. Kruskal–Wallis non-parametric test (followed by Dunn's multiple comparisons test to establish significance) was applied to the data that do not pass the homogeneity of variance test or not follow the normal distribution. When the data in groups were normally distributed with equal variances, t-test (2 tailed) was performed between 2 groups or ANOVA (followed by Tukey’s or Bonferroni's) for more groups. Behaviors results containing time and group two factors were tested by two-way repeated ANOVA (followed by Tukey’s). Detailed statistical analysis is summarized in Supplementary Table 1 (Table S1).