Neurexin‐2 is a potential regulator of inflammatory pain in the spinal dorsal horn of rats

Abstract Chronic pain is one of the serious conditions that affect human health and remains cure still remains a serious challenge as the molecular mechanism remains largely unclear. Here, we used label‐free proteomics to identify potential target proteins that regulate peripheral inflammatory pain and reveal its mechanism of action. Inflammation in peripheral tissue was induced by injecting complete Freund's adjuvant (CFA) into rat hind paw. A proteomic method was adopted to compare the spinal dorsal horn (SDH) in peripheral inflammatory pain (PIP) model rats with controls. Differential proteins were identified in SDH proteome by label‐free quantification. The role of screened target proteins in the PIP was verified by small interfering RNA (siRNA). A total of 3072 and 3049 proteins were identified in CFA and normal saline (NS) groups, respectively, and 13 proteins were identified as differentially expressed in the CFA group. One of them, neurexin‐2, was validated for its role in the inflammatory pain. Neurexin‐2 was up‐regulated in the CFA group, which was confirmed by quantitative PCR. Besides, intrathecal siRNA‐mediated knock‐down of neurexin‐2 attenuated CFA‐induced mechanical and thermal hyperalgesia and reduced the expression of SDH membrane glutamate receptors (eg mGlu receptor 1, AMPA receptor) and postsynaptic density (eg PSD‐95, DLG2). These findings increased the understanding of the role of neurexin‐2 in the inflammatory pain, implicating that neurexin‐2 acts as a potential regulatory protein of inflammatory pain through affecting synaptic plasticity in the SDH of rats.

symptoms of inflammatory pain were developed as hyperalgesia and allodynia. 4 The alteration in the proteome that assists in identifying proteins was studied. It is critical to understand the mechanisms of inflammatory pain and the selection of novel drug targets.
With the rise and development of high-throughput genomic and proteomic technologies, many new molecules have been developed and many disease-related key targets and pathways have been revealed, providing an opportunity for developing new drugs. 5 Proteomics becomes a research hotspot for studying various disease mechanisms. 6 Gene function is realized by proteins.
Proteins are regarded as real-life activity performers and can better reflect the current functional state of the body. If the molecular mechanism that is associated with pain can be directly studied from the protein level, the essential act of chronic inflammatory pain will be revealed more comprehensively and systematically.
Studying proteomics involves a study based on the proteome, analyses the changes in protein composition, level and modification status from the overall level, understands the interaction between proteins and reveals the function of proteins and the laws of cell activities. 7 In this study, a proteomic-based strategy was used to compare the spinal dorsal horn (SDH) in a peripheral inflammatory pain (PIP) model of a rat with a control group. PIP was induced by the injection of CFA unilaterally into the rat hind paw. Differential proteins were founded in the SDH proteome using label-free quantification. Further investigation was conducted by bioinformatics analysis, and the changes of target proteins were verified by Western blotting and quantitative PCR (qPCR) methods.
Furthermore, small interfering RNA (siRNA) was used to study the effect of target protein on inflammatory pain and its mechanism.
Here, we demonstrate that neurexin-2 is a regulator of inflammatory pain.

| Reagents and animals
The antibody information was listed in Table S1. Neurexin-2-siRNA for in vivo transfection and its mismatch control (MC-siRNA) were purchased from RiboBio Co. 8

| Pain model establishment
The pain model was established according to the previously published method. 4 Briefly, CFA was injected subcutaneously into rat left hind paw.

| Drug administration
Intrathecal tubing and drug administration were conducted as reported previously. 9 Neurexin-2-siRNA and MC-siRNA were dissolved in saline at a final concentration of 2 nmol/μL (10 μL, n = 8) for subarachnoid administration before and on days 1 and 3 after the CFA injection.

| Thermal paw-withdrawal latency test
Thermal paw-withdrawal latency (TWL) test was performed as described previously. 4 The paw-withdrawal latencies were recorded. F I G U R E 1 Complete Freund's adjuvant (CFA) produces inflammatory hyperalgesia in rats. Thermal withdrawal latency (TWL; A) and mechanical withdrawal threshold (MWT; B) were significantly decreased in CFA-treated rats when compared with normal saline (NS)-treated rats. Data are expressed as means ± SE (n = 8). *P < .05, **P < .01 vs NS-treated rats

| Mechanical paw-withdrawal threshold test
Mechanical paw-withdrawal threshold (MWT) test was performed as described previously. 4 MWT was recorded.

| Protein preparation
After treatment, L4-L6 enlargement of left SDH was collected. The proteins were obtained from tissue homogenates in RIPA buffer (MO, Sigma). Homogenates were centrifuged at 13,000 g at 4°C for 10 min, and then, the supernatants were collected. Protein concentration was measured using the BCA protein assay kit (Pierce, Thermo Scientific). Two hundred fifty microgram protein of each sample was taken and digested according to previously described method 10 for proteomic determination.

| Proteomic determination and sequence database searching and data analysis
The protein samples were analysed by HPLC-MS/MS according to the method in Appendix S1.

| Western blot analysis
Western blot analysis was described in Appendix S1. Briefly, proteins were separated by SDS-PAGE, incubation of primary and second antibodies were conducted, and protein band was visualized using the ECL method.

| Real-time quantitative PCR
The qPCR was performed as in Appendix S1. ΔΔ C t method was used to evaluate the differential expression.
F I G U R E 2 Identification of proteins that show significant changes in abundance after complete Freund's adjuvant (CFA) injection. Proteins with more than twofold differential expression data were analysed within PERSEUS using three-samplettests and a false discovery threshold of 0.05 to identify candidates that significantly change in abundance. Five proteins were downregulated, and eight proteins were up-regulated in abundance 3 d post-CFA injection relative to NS injection. "↙" marked as neurexin-2 TA B L E 1 Differentially expressed proteins with more than twofold differential expression and P-value < .05 in CFA rats vs NS rats identified from proteomics analysis The significance of italic formating implies that it is the target protein for further study in this article.

F I G U R E 3
The functional annotation of dysregulated proteins was analysed by protein analysis. A, Biological process, molecular function and cellular component. B, Pathway analysis of 13 dysregulated proteins was indicated by PANTHER, DAVID, STRING and Reactome. "*," involved neurexin-2

| Statistical analysis
All data are presented as mean ± SE. The differences in the molecular expression or behavioural scores among groups were analysed by one-or two-way repeated-measures analysis of variance, respectively, followed by the Bonferroni test. P values < .05 were considered to be statistically significant.

| CFA induces hyperalgesia
In the current study, a rat CFA model that was used in several studies was adopted for investigation. Figure 1 showed the behavioural effects of CFA and NS when injected subcutaneously into the left hind paw. MWT testing and TWL testing were performed on rats before and every 2 days (last for 13 days) after CFA administration. Before injection, the rats in the NS and CFA groups had similar values to the pain stimuli. The CFA rats had a significant reduction in the pain responses from the first day. This suggests behavioural signs of the inflammatory pain. The NS rats showed slight changes with no significant increase in mechanical and thermal responses ( Figure 1A,B).

| Overall protein changes identified by proteomics
Using proteomics, a total of 3072 and 3049 proteins were identified in CFA and normal saline (NS) groups, respectively. Of them, 2992 proteins were identified in both groups. Thirteen proteins F I G U R E 4 Immunofluorescence double staining showed that neurexin-2 (green) was colocalized with neuronal marker, NeuN (red), but neither with astrocyte marker GFAP (red) nor with microglial marker Iba-1 (red). Original magnification: 100×, scale bar: 100 μm were found to be dysregulated between two groups (CFA/NS ratio ≥twofold or ≤0.5-fold, P-value < .05). Of these proteins, five were down-regulated, and eight proteins were up-regulated ( Figure 2, Table 1).

| Bioinformatics analysis
Biological function analysis (GO analysis and pathway analysis) was performed on 13 different proteins obtained in this experiment by the DAVID enrichment analysis system. GO analysis of 13 differential proteins was performed to study their function, and biological process (BP) analysis results revealed that differentially expressed proteins are mostly involved in the cellular process, single-organism cellular process, biological regulation and metabolic processes. Molecular function (MF) analysis revealed that most of these proteins are related to molecular binding. Cellular component (CC) analysis revealed that the differential proteins were mainly distributed in the cell membrane and organelles, and extracellular regions ( Figure 3A). Among the 13 differential proteins, KEGG pathway showed two proteins each related to complement and coagulation cascades and cell adhesion molecules (CAMs), and one protein each related to choline metabolism in cancer, Huntington's disease, Alzheimer's disease, ribosome, cardiac muscle contraction, sulphur relay system, Parkinson's disease and oxidative phosphorylation ( Figure 3B).
Immunofluorescence images showed that neurexin-2 was colocalized with neuron, but neither with astrocyte nor with microglia ( Figure 4). This suggests that neurexin-2 is located in the neuron cell. In the BP, neurexin-2 participates in biological regulation, biological adhesion and development process. Based on the MF, neurexin-2 is mainly involved in molecular transducer activity and binding. In the CC, neurexin-2 was distributed in the cell membrane and synaptic sites ( Figure 3A). Furthermore, previous studies have reported that neurexin is mainly distributed in the pre-synapses of neurons and is very important for the formation and differentiation of synapse by binding neuroligin that is expressed in the postsynaptic neurons. 11 One of the main functions of neurexin is to regulate the growth of synapses. 12 The changes of synaptic plasticity in the formation and maintenance of hyperalgesia have long been agreed 13 ; however, the role of neurexin in hyperalgesia has not been reported so far, and hence, this study selected neurexin-2 for further research.

| Validation of neurexin-2 by qPCR and Western blot
The differential expression of neurexin-2 in the SDH tissue from NS and inflammatory rats was validated by qPCR and Western blotting.
Both have demonstrated a significant up-regulation of neurexin-2 in the SDH tissues from CFA-treated rats when compared to NS rats ( Figure 5A-C). These results all showed that neurexin-2 was up-regulated in the SDH tissue from inflammatory rats by qPCR, Western blotting and proteomics approach.
F I G U R E 5 CFA induces a time-dependent increase in spinal neurexin-2 mRNA and protein expression. The expression of neurexin-2 was assayed on days 1, 3 and 7 after CFA injection. A, qPCR results showed that the mRNA expression of Nrxn2 was a time-dependent up-expression in SDH tissues from CFA-treated rats (n = 6). B, Representative Western blots for neurexin-2 and β-actin at different time-points after CFA injection; β-Actin was used as the loading control. C, The bar graph showed that the relative level of neurexin-2 protein was increased in SDH tissues of CFA-treated rats in a time-dependent manner (n = 6). *P < .05, **P < .001 vs NS rats

| Silencing of the neurexin-2 transgene with siRNA in neurons separated from SDH tissue
Three siRNA oligonucleotides (siRNA1, siRNA2 and siRNA3 sequences) targeting the rat neurexin-2 were used to cotransfect in neurons from SDH tissue in vitro ( Figure 6A) and then the level of neurexin-2 transgene expression was analysed by qPCR ( Figure 6B).

| Intrathecal injection of selected neurexin-2-siRNA specifically decreases neurexin-2 expression levels in the SDH tissue
The effects of neurexin-2-siRNA on expression levels of neurexin-2 in SDH tissues were also analysed via qPCR and Western blotting. As outlined in Figure 7A, following intrathecal injection of neurexin-2-siRNA, the expression of Nrxn2 mRNA in SDH neurons was decreased when compared to the MC-siRNA group ( Figure 7A).

| Intrathecal injection of neurexin-2-siRNA reduces MWT and TWL in peripheral inflammatory pain rats
To

| D ISCUSS I ON
Complete Freund's adjuvant is a potent inflammatory agent that causes localized and long-lasting inflammatory pain. In this study, the rat model of chronic inflammatory pain was prepared by injection of CFA 0.1ml in the hind paw as described in the literature. 14 The results showed that MWT was decreased on days 1, 3 and 7 after CFA injection. The shortening of TWL suggests that the rat model of chronic inflammatory pain was successfully prepared and is regarded as an ideal research object in comparative proteomics research of chronic inflammatory pain. SDH is the gateway to sensory information and an integrated primary centre. 15 Peripheral noxious stimulation can be directly transmitted to SDH neurons for integration through dorsal root ganglion axons. 16,17 Therefore, in our study, the global proteomic changes in SDH of a CFA rat model using quantitative proteomics method coupled to bioinformatics analysis was

F I G U R E 7 Complete Freund's adjuvant (CFA) increases
Nrxn2 mRNA and neurexin-2 proteins in the dorsal horn of spinal cord L4-6 segments and selected neurexin-2-siRNA that reverses this effect by intrathecal injection once in 2 d for 0-4 d as described in the section 2. A, qPCR analyses of Nrxn2 mRNA expression in rat lumbar SDH tissues and dorsal root ganglions in four groups. B, Representative Western blots of neurexin-2 and β-actin at different administration reagents; β-Actin was used as the loading control. C, Bar graph showed that the relative level of neurexin-2 protein was increased in SDH tissues of CFA-treated rats, and increased neurexin-2 expression in the selected neurexin-2-siRNA treatment group was significantly lower than the MC-siRNA compared with MC-siRNA rats (n = 6). # P < .05 vs NS group; **P < .001 vs MC-siRNA group Our findings revealed that neurexin-2 expression showed a significant increase in SDH of CFA rats when compared with the control group. While massive efforts have been made in recent years to investigate the molecular mechanisms of synaptic levels within the CNS, very little has been reported on neurexin-2 in SDH of CFA rat models. 18,19 Neurexin-2 is one of the neurexins that cleave isomers, and are presynaptic proteins that help to connect neurons at the synapse. 20 In conclusion, our present study suggests that injecting CFA into rat hind paw induced peripheral inflammation and pain hypersensitivity. In association with these changes, a total of 3072 and 3049 proteins were identified in CFA and NS rats, respectively, and 13

CO N FLI C T O F I NTE R E S T S
The authors confirm that they have no competing interests. F I G U R E 9 Complete Freund's adjuvant (CFA) increases the expression of mGlu1 receptor, AMPA receptor, PSD-95, DLG2 membrane proteins in the dorsal horn of spinal cord L4-6 segments, while neurexin-2-siRNA reverses this effect by intrathecal injection as described in the section 2. A, Representative Western blots of mGlu1 receptor, AMPA receptor, PSD-95, DLG2 membrane proteins and β-actin with different administration reagents; β-Actin was used as the loading control. B-E, Bar graph showed that the relative level of mGlu1 receptor, AMPA receptor, PSD-95 and DLG2 membrane proteins was increased in SDH tissues of CFA-treated rats, and increased proteins in the neurexin-2-siRNA treatment group were significantly lower than the MC-siRNA (n = 6). **P < .001 vs NS group; # P < .05, ## P < .001 vs MC-siRNA group

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available in the supplementary material of this article.