All experiments were approved by Ethics Committee of the Second Affiliated Hospital of Soochow University (no. 2018012), and strictly followed the guidelines of the National Institutes of Health Guide for the Care and Use of Laboratory Animals (8th edition). A total of 98 Sprague-Dawley male rats (8–10 weeks old), supplied by the Experimental Animal Center of Soochow University, were housed in a 22–24°C, 40–60% humidity, specific pathogen-free, 12-h light/dark cycle room, with free access to a standard rat diet and water. The health and behavior of each group of rats were observed daily. The CCI NP model was prepared as previously described (14). Briefly, after being anesthetized with isoflurane (3.5% induction, 2–2.5% maintenance), both sides of the sciatic nerves of rats were exposed and ligated with non-absorbable sutures at 4 sites, with an interval of 1 mm. Sciatic nerves in sham rats were only isolated without ligation, serving as a control. Zero, three and seven days after surgery, the dorsal horn of spinal cords (L4-L6) were harvested under narcosis (isoflurane, 2–2.5% maintenance) for further analysis. Severely inert rats and those with the experiments completed were euthanized immediately by 99.9% carbon dioxide (the air replacement rate was 25% of the container volume per minute) to avoid or alleviate unnecessary suffering or distress. The absence of a heartbeat was used to verify the death of animals.

Rats were anesthetized with isoflurane (3.5% induction, 1.5–2% maintenance), and fixed in the prone position. Following conventional disinfection, atlantooccipital membranes were exposed and a 1-cm length incision was made, then PE-10 polyethylene catheters were slowly implanted into the lumbar interspace through it. Rats with motor function impairment were excluded. Whether the operation was a success was confirmed by the occurrence of bilateral hind limb paralysis after 2% lidocaine injection. Recombinant Lenti-SLC24A2 (10 µl; Shanghai GeneChem Co., Ltd.), miR-135a-5p agomir or antagomir (10 µM; Guanzhou RiboBio Co., Ltd.) were injected through a microinjection syringe, which linked with the intrathecal catheter, 72 h or 24 h before surgery. NCKX2 antisense oligo-deoxyribonucleotide (AS-ODN) (5′-CCAATGACTCGAATTAGCTT-3′, 140 µg/kg) was infused continuously with an osmotic minipump (1 µl/h; RWD Life Science Co., Ltd.) connected to an intrathecal catheter 24 h before surgery (15,16).

Mechanical pain was determined by paw withdrawal threshold (PWT) in response to increasing mechanical pressure created by the calibrated Electronic von Frey filament as previously described (17). The pressure for bilateral hind paw withdrawal was recorded and defined as PWT. The cut-off was set at 50 g. The withdrawal threshold was calculated as the average of three continuous tests, with at least 10 min between them. Thermal pain was evaluated by paw withdrawal latency (PWL) in response to light radiant heat. Light was focused on the plantar surface of the bilateral hind paws. The interval between light onset and paw withdrawal was recorded and defined as PWL. The cut-off time was set at 20 sec. The test was repeated at least three times, with a 5-min interval between them, and the average PWLs was considered as the threshold of thermal-evoked pain.

Bioinformatics analyses were performed to determine the miRNAs that directly bind to SLC24A2 using programs in TargetScan (http://www.targetscan.org/vert_72/), miRanda (http://www.microrna.org/microrna/getGeneForm.do), miRDB (http://mirdb.org), PITA (https://genie.weizmann.ac.il/pubs/mir07/index.html) and miRWalk (http://mirwalk.umm.uni-heidelberg.de).

The 3′UTR of SLC24A2 carrying the wild-type binding sites or mutant binding sites of the miR-135a-5p were cloned into the pmirGLO dual-luciferase vector (Promega Corporation). The vectors were transfected into 293T cells by Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) with miR-135a-5p as well as agomir administration (Guanzhou RiboBio Co., Ltd.). After 48 h, cells were harvested and luciferase activity was detected using the Dual-Luciferase assay system (Promega Corporation) according to the manufacturer's instructions. The results were normalized to Renilla luciferase activity and expressed as relative luciferase activity.

RNA in lumbar dorsal spinal cord tissues were extracted by TRIzol (Invitrogen; Thermo Fisher Scientific, Inc.). Reverse transcription was performed using the PrimeScript RT reagent kit for mRNA or MiScript Reverse Transcription kit for miRNA (Qiagen, Inc.). An ABI PRISM 7500 real-time PCR system and SYBR Premix Ex Taq II (Takara Biotechnology Co., Ltd.) for mRNA or a MiScript SYBR-Green PCR kit (Qiagen, Inc.) for miRNA were used in the PCR reactions. The following thermocycling conditions were used for RT-qPCR: Initial denaturation at 95°C for 5 min; 40 cycles of 95°C for 15 sec and 60°C for 60 sec. GAPDH or U6 served as internal controls, respectively. Primers used were as follows: SLC24A2 forward (F), 5′-CTGGAGGAGCGAAGGAAAGG-3′ and reverse (R), 5′-TGTGAAAGTTCTGGGGCTGAC-3′; GAPDH F, 5′-CAACTTTGGCATCGTGGAAGGG-3′ and R, 5′-CAACGGATACATTGGGGGTAGG-3′; miR-135a-5p F, 5′-CTCCTAGGTATGGCTTTTTATTC-3′ and R, 5′-TCAACTGGTGTCGTGGAGTC-3′; U6 F, 5′-CTCGCTTCGGCAGCACA-3′ and R, 5′-AACGCTTCACGAATTTGCGT-3′. The relative expression of target genes was calculated by the 2^−∆∆Cq method (18).

Total protein of lumbar dorsal spinal cord tissues was isolated by N-PER™ Neuronal Protein Extraction reagent (Thermo Fisher Scientific, Inc.) with a protease inhibitor. The protein concentration was determined by the Pierce™ BCA Protein assay kit (Thermo Fisher Scientific, Inc.). Then, 20 µg protein was loaded into each well and separated by 10% SDS-PAGE. Subsequently, the protein blots were transferred onto PVDF membranes. After being blocked by 5% bovine serum albumin (Beijing Solarbio Science & Technology Co., Ltd.) for 2 h at room temperature, the membranes were incubated overnight at 4°C with the primary antibodies: Rabbit anti-NCKX2 antibody (1:200; ab192419; Abcam) and GAPDH (1:4,000; sc-25778; Santa Cruz Biotechnology, Inc.). The membranes were washed and a horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (1:2,000; ab205718; Abcam) was added followed by incubation at room temperature for 1 h. Finally, the membranes were color developed with ECL (EMD Millipore). The gray value was calculated by Quantity One software (version 4.5.0; Bio-Rad Laboratories, Inc.) normalized to GAPDH.

The protein expression of interleukin (IL)-1β, tumor necrosis factor (TNF)-α and IL-6 in lumbar dorsal spinal cord tissue homogenates were measured using commercially available ELISA kits (SEKM-0002, SEKM-0034 and SEKM-0007; Solarbio Life Science), according to the manufacturer's protocols. The experiment was repeated three times, and the mean value was used.

SPSS (version 18.0; SPSS, Inc.) was used for data analysis. Data were presented as the mean ± standard deviation. Comparisons between two groups were estimated by a Student's t-test, whereas multiple group comparisons were carried out by one-way ANOVA analysis with a Bonferroni's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

To investigate the association between SLC24A2 and NP, RT-qPCR was used to examine the expression levels of SLC24A2 in the dorsal spinal cord of rats after CCI surgery. The results revealed that SLC24A2 expression was significantly decreased in CCI rats compared with sham rats 3 and 7 days after surgery (P<0.05 and P<0.01, respectively, Fig. 1A). Moreover, the expression levels of the encoded protein (NCKX2) of the SLC24A2 gene were determined 7 days after surgery by western blotting. It was revealed that NCKX2 expression was also significantly decreased in the lumbar dorsal spinal cord of CCI rats (P<0.05, Fig. 1B). These results indicated a negative association between SLC24A2 expression and the progression of NP.

To determine the biological effect of SLC24A2 on NP, SLC24A2 was overexpressed in the dorsal spinal cord of rats by intrathecal injection of lentivirus carrying SLC24A2 (Lenti-SLC24A2). First, it was demonstrated that the expression levels of SLC24A2 and its encoded protein NCKX2 were both significantly increased in CCI rats treated with Lenti-SLC24A2 (both P<0.01, Fig. 2A and B). Then, mechanical and thermal pain responses were evaluated using behavioral experiments. It was determined that mechanical allodynia and thermal hyperalgesia were significantly enhanced in rats subjected to CCI but alleviated by overexpressing SLC24A2 (P<0.05, Fig. 2C and D). These results indicated a suppressive role of SLC24A2 in NP.

A previous study revealed that changes in calcium concentration in cells could effectively regulate the inflammatory response (19); therefore, the biological effect of SLC24A2 on the expression of a number of key inflammatory cytokines, which increase during the development of NP, was subsequently investigated. It was revealed that IL-1β, IL-6 and TNF-α were significantly increased in the dorsal spinal cord of CCI rats, however their expression was reversed in the Lenti-SLC24A2-treated group (P<0.05, Fig. 3). The results indicated a suppressive effect of SLC24A2 on neural inflammation.

To determine whether the decrease in SLC24A2 after CCI was caused by miRNAs and identify the key regulator, bioinformatics analyses were performed to predict the miRNAs that directly bind to SLC24A2. All programs predicted that the seed regions of miR-135a-5p could bind conserved sites in the 3′UTR of SLC24A2 (Fig. 4B). Moreover, it was revealed that the expression of miR-135a-5p was increased in the CCI rats (P<0.01, Fig. 4A). Therefore, it was next investigated whether miR-135a-5p directly interacts with the 3′UTR of SLC24A2. The results revealed that co-transfection of miR-135a-5p agomir with the luciferase vector containing the SLC24A2 3′UTR with the wild-type miRNA binding site led to a significant decrease in relative luciferase activity in 293T cells (P<0.01, Fig. 4B and C). However, when the binding site was mutated, the inhibitory effect of overexpressed miR-135a-5p on relative luciferase activity was blocked (P>0.05, Fig. 4B and C). This result indicated that miR-135a-5p directly targeted the 3′UTR of SLC24A2. Furthermore, the effect of up- or downregulation of miR-135a-5p on NCKX2 expression was determined by western blotting. The results revealed that miR-135a-5p agomir treatment significantly decreased the protein levels of NCKX2 in the dorsal spinal cord (P<0.01, Fig. 4D), and the miR-135a-5p antagomir had the opposite effect on NCKX2 expression (P<0.01, Fig. 4E). In short, these results indicated that SLC24A2 is a target gene of miR-135a-5p.

To determine whether miR-135a-5p regulates NP by targeting SLC24A2, rescue experiments were performed by co-infecting rats with miR-135a-5p antagomir and NCKX2 AS-ODN. The results revealed that the suppressive effects of the miR-135a-5p antagomir on mechanical allodynia (P<0.01, Fig. 5A) and thermal hyperalgesia (P<0.01, Fig. 5B) were significantly reversed by NCKX2 AS-ODN treatment. Moreover, the decreased expression of the inflammatory cytokines IL-1β (P<0.01, Fig. 6A), IL-6 (P<0.01, Fig. 6B) and TNF-α (P<0.01, Fig. 6C) caused by the miR-135a-5p antagomir was also reversed by NCKX2 AS-ODN. Overall, these results indicated that miR-135a-5p exerted its effect on NP by targeting SLC24A2 and its encoded protein.