Leptin Contributes to Neuropathic Pain via Extrasynaptic NMDAR-nNOS Activation

Leptin is an adipocytokine that is primarily secreted by white adipose tissue, and it contributes to the pathogenesis of neuropathic pain in collaboration with N-methyl-D-aspartate receptors (NMDARs). Functional NMDARs are a heteromeric complex that primarily comprise two NR1 subunits and two NR2 subunits. NR2A is preferentially located at synaptic sites, and NR2B is enriched at extrasynaptic sites. The roles of synaptic and extrasynaptic NMDARs in the contribution of leptin to neuropathic pain are not clear. The present study examined whether the important role of leptin in neuropathic pain was related to synaptic or extrasynaptic NMDARs. We used a rat model of spared nerve injury (SNI) and demonstrated that the intrathecal administration of the NR2A-selective antagonist NVP-AAM077 and the NR2B-selective antagonist Ro25-6981 prevented and reversed mechanical allodynia following SNI. Administration of exogenous leptin mimicked SNI-induced behavioral allodynia, which was also prevented by NVP-AAM077 and Ro25-6981. Mechanistic studies showed that leptin enhanced NR2B- but not NR2A-mediated currents in spinal lamina II neurons of naïve rats. Leptin also upregulated the expression of NR2B, which was blocked by the NR2B-selective antagonist Ro25-6981, in cultured dorsal root ganglion (DRG) neurons. Leptin enhanced neuronal nitric oxide synthase (nNOS) expression, which was also blocked by Ro25-6981, in cultured DRG cells. However, leptin did not change NR2A expression, and the NR2A-selective antagonist NVP-AAM077 had no effect on leptin-enhanced nNOS expression. Our data suggest an important cellular link between the spinal effects of leptin and the extrasynaptic NMDAR-nNOS-mediated cellular mechanism of neuropathic pain.

Leptin is a 16-kDa adipocytokine that is primarily secreted by white adipose tissue. It is well known for its role in metabolic regulation and obesity, which are mediated via a long-form leptin receptor (Ob-Rb) [28][29][30][31]. Leptin and Ob-Rb were recently found in the CNS, which indicates that leptin has a broad role in the regulation of neuronal functions [32,33]. Recent studies showed that leptin played a significant role in nerve injury-induced neuropathic pain in rats [34,35]. The peripheral effects of leptin on neuropathic pain following nerve injury were mediated via macrophage stimulation [36], and our previous results indicated that the central effect of leptin was likely related to the activation of NMDARs [34,35].
Synaptic and extrasynaptic NMDARs contain different subunits, and leptin contributes to the pathogenesis of neuropathic pain via activation of NMDARs. Therefore, the present study further investigated whether leptin contributed to neuropathic pain via synaptic or extrasynaptic NMDARs. Synaptic NMDARs in the spinal cord and DRG primarily comprise NR1/NR2A, and extrasynaptic NMDARs consist of NR1/NR2B and NR1/NR2D [7][8][9][10]. However, NR2D does not play a role in the occurrence or development of pain sensation [19,37], Therefore, we focused on the role of NR2A and NR2B in the contribution of leptin to neuropathic pain.
Neuronal nitric oxide synthase (nNOS) plays an important role in the development and maintenance of neuropathic pain [22,[38][39][40], and it is a downstream factor of NR2B [19,22,41,42]. The present study also investigated whether leptin contributed to neuropathic pain via synaptic or extrasynaptic NMDAR-nNOS activation. Our results indicated that spinal leptin played an important role in the pathogenesis of neuropathic pain via activation of the extrasynaptic NMDAR-nNOS pathway.

Experimental Animals
Male Sprague-Dawley rats, 4-6 weeks old, were used for electrophysiological experiments and primary culture of DRG neurons. Rats weighing 250-280 g were used for behavioral experiments. All SD rats were purchased from Southern Medical University Animal Center (Guangzhou, Guangdong, China). The rats were given sufficient food and water and maintained on a 12:12-h light/dark cycle in an SPF animal room at 22-25°C with 45 ± 10% humidity. All animal protocols were approved by the Animal Experimentation Ethics Committee of Southern Medical University and performed in accordance with the principles of the 'Administrative Measures for Laboratory Animals of Southern Medical University' and the 'Guidelines for Ethical Examination of Animal Experiments of Southern Medical University'.

Spared Nerve Injury And Drug Delivery
Spared nerve injury (SNI) was produced according to the method of Decosterd and Woolf [43]. Briefly, rats weighing 280-300 g were anesthetized using pentobarbital sodium (50 mg/kg, intraperitoneal). The skin of the lateral left thigh was incised, and the biceps femoris muscle was separated to expose the sciatic nerve and its three terminal branches: the sural, common peroneal, and tibial nerves. The common peroneal and tibial nerves were tightly ligated, and a 2-mm length segment close to the ligation was removed. The intact sural nerve was untouched. The wound was stitched in two layers.

Behavioral Test for Allodynia
Rats were habituated to the test environment for 1 h over 2 consecutive days before baseline testing. Mechanical allodynia was examined using von Frey filaments applied to the plantar surface of each hind paw [45]. A positive response was defined as hind paw withdrawal from at least 2 of 5 applications of a von Frey filament. The paw withdrawal threshold (PWT) was determined using the "up-and-down" method [46]. Food intake and body weight were monitored to exclude the effects of these two factors. All behavioral experiments were carried out at 9 AM, and the investigators were blinded to treatment conditions.

Immunocytochemistry
Cultured DRG neurons in 96-well plates were fixed with 4% paraformaldehyde for 2 h, blocked with bovine serum albumin (3% v/v, Sigma, USA) and Triton-X100 (0.3% v/v, Sigma, USA) for 1.5 h at room temperature and incubated overnight at 4°C with one of the following primary antibodies: NR2A (Novus, USA) at 1:500, mouse monoclonal; NR2B (Abcam, USA) at 1:1000, rabbit polyclonal; and nNOS (BD Biosciences, USA) at 1:1000, mouse monoclonal. Cells were rinsed with PBS and incubated for 1.5 h at room temperature with an FITC-or cyanine 3-conjugated secondary antibody (1:300; Jackson ImmunoResearch, USA). For controls, primary antibodies were omitted in the process. Four to 6 wells were randomly selected for examination using a fluorescence microscope attached to a CCD spot camera (Olympus IX71, Japan) and processed using Adobe Photoshop v7.

Statistics
Data from mechanical allodynia (threshold bending force in grams) tests were analyzed using repeated-measure 2-way analysis of variance (ANOVA) as described previously [35]. Testing time points detected overall differences between treatment groups. The data were also examined using 2-way ANOVA across treatment groups to examine overall differences between testing time points. When significant main effects were observed, post hoc Newman-Keuls tests were performed to determine the source(s) of differences. The experimenters were blinded to the treatment conditions. All results are expressed as the means ± SE. Differences were considered statistically significant at the level of p = 0.05. For all other experiments, differences were compared using one-way ANOVA followed by post hoc Newman-Keuls tests in SPSS 17.0 (SPSS Inc., Chicago, IL, USA).

NR2A and NR2B Antagonists Prevented and Reversed SNI-Induced Allodynia
To determine whether blockade of synaptic and extrasynaptic NMDARs prevented the development of pain behaviors after SNI, the NR2A-selective antagonist NVP-AAM077 (4 nmol) and the NR2B-selective antagonist Ro25-6981 (20 nmol) were administered intrathecally once daily for 14 days beginning immediately after SNI or sham operation. Behavioral tests on the hind paws were performed 30 min after administration on days 1, 3, 5, 7, 10, and 14. NVP-AAM077 and Ro25-6981 prevented the development of mechanical allodynia of the hind paw ipsilateral to SNI compared to the vehicle group (Fig. 1a n = 6; P < 0.05). NVP-AAM077 and Ro25-6981 had no effect on mechanical allodynia on the contralateral side in the same rat (Fig. 1b). NVP-AAM077 and Ro25-6981 also reversed established neuropathic pain behaviors in SNI rats. A single intrathecal administration of NVP-AAM077 (4 nmol) and Ro25-6981 (20 nmol) attenuated mechanical allodynia on day 14 after SNI 30 min after treatment, and the effect of NVP-AAM077 was maintained for 24 h ( Fig.  1c; n = 5; P < 0.05). NVP-AAM077 and Ro25-6981 had no effect on the contralateral side (Fig. 1d). The attenuation effect of NVP-AAM077 and Ro25-6981 was the same as treatment with 10 nmol MK-801, which is a noncompetitive NMDAR antagonist (Fig. 1e, n = 6). These data demonstrated that NR2A and NR2B antagonists prevented and reversed SNIinduced allodynia.

NR2A and NR2B Antagonists Prevented Exogenous Leptin-Mimicked SNI-Induced Behavioral Changes
Our previous research found that spinal leptin contributed to the pathogenesis of neuropathic pain, and the central effect of leptin was likely related to the activation of NMDARs [34,35]. Leptin upregulated the expression of spinal NMDAR subunit NR1 and enhanced NMDAR-induced currents in spinal cord lamina II neurons [34,35]. The present study further examined whether leptin contributed to the pathogenesis of neuropathic pain by acting on synaptic or extrasynaptic NMDARs.

Leptin Enhanced NR2B-But Not NR2A-Mediated Currents in Dissociated Lamina II Neurons in Naïve Rats
Our previous results showed that leptin enhanced NMDARmediated currents in dissociated substantia gelatinosa (lamina II) neurons of the spinal cord dorsal horn [34,35]. We chose lamina II neurons because finely myelinated (Aδ) and unmyelinated (C) primary nociceptive afferent fibers project to this spinal region [50,51]. The present study further examined whether leptin enhanced synaptic or extrasynaptic NMDARmediated currents in lamina II neurons in naïve rats.
After treatment with 1 μM Ro25-6981 to block NR2Bmediated currents, the application of 100 nM leptin to ACSF for 5 min did not change the NR2A-mediated currents (Fig.  3b, d). In contrast, the application of 100 nM leptin to ACSF for 5 min significantly enhanced the currents after blockade with 0.4 μM NVP-AAM077 (Fig. 3c, d P < 0.05). These results suggested that leptin enhanced extrasynaptic, but not synaptic, NMDAR-mediated currents in dissociated lamina II neurons.

Leptin Enhanced NR2B, But Not NR2A, Expression in Cultured DRG Neurons
The DRG convey sensory information from the periphery to the CNS and may be damaged in peripheral sensory neuropathic pain [52]. To confirm the relationship of leptin with synaptic or extrasynaptic NMDARs in neuropathic pain, we used an in vitro dissociated DRG neuron culture from adult (postnatal 4-to 6-week-old) naïve rats [47]. DRG neuron culture was established for 72 h, and the culture medium was replaced with a medium containing leptin or vehicle for 72 h [34]. Immunohistochemistry results showed that exposure to exogenous leptin upregulated the expression of NR2B in DRG neuron culture in a dose-dependent manner, with the maximal enhancement occurring at 2 ng/ml leptin (2 ng/ml > 20 ng/ml > 0.2 ng/ml = vehicle. Figure 4a, n = 3, P < 0.05). This concentration (2 ng/ml) of leptin was used in subsequent culture experiments. Immunohistochemistry results and western blot results showed that the addition of 1 μM Ro25-6981 to the leptin (2 ng/ml) culture medium for 72 h significantly diminished the upregulation of NR2B expression by leptin, whereas 1 μM Ro25-6981 alone did not change the baseline expression of NR2B (Fig. 4a, c, d, n = 3). However, 2 ng/ml leptin did not significantly upregulate NR2A expression (Fig. 4b, e, f, n = 3). These data showed that leptin significantly upregulated NR2B, but not NR2A, expression in cultured DRG neurons.

Leptin Enhanced nNOS Expression in Cultured DRG Neurons
To examine the cellular mechanisms underlying the spinal leptin effect on neuropathic pain, we determined whether the expression of nNOS was changed after exogenous leptin treatment because nNOS is a downstream target following NMDAR activation [19,22,41,42]. The inclusion of leptin (2 ng/ml) in culture medium for 72 h significantly upregulated nNOS expression in cultured DRG neurons, as revealed by immunohistochemistry (Fig. 5a) and Western blot (Fig. 5b, c).

Discussion
The present study demonstrated the following results: (1) the intrathecal administration of the NR2A-selective antagonist NVP-AAM077 and the NR2B-selective antagonist Ro25-6981 prevented and reversed SNI-induced allodynia; (2) administration of exogenous leptin mimicked SNI-induced allodynia, which was also prevented by NR2A and NR2B antagonists; (3) leptin enhanced spinal NR2B-but not NR2A-mediated currents in naïve lamina II neurons; and (4) leptin upregulated the expression of NR2B, but not NR2A, and nNOS in cultured DRG cells, which were blocked by the NR2B antagonist Ro25-6981. These data suggested that the contribution of leptin to the pathogenesis of neuropathic pain was related to NR2B-based extrasynaptic NMDARs and activation of nNOS, as the downstream factor of NMDAR. However, the relationship between NR2A-based synaptic NMDARs and the effect of leptin on neuropathic pain needs further investigation.
Leptin is a 16-kDa adipocytokine that is primarily produced by white adipose tissue, and it is well-known for its role in metabolic regulation and obesity [28][29][30][31]. Mounting evidence indicates that leptin has broad roles in the regulation of neuronal functions [32,33]. Leptin was recently demonstrated to contribute to the pathogenesis of neuropathic pain [34-36, 53, 54]. Leptin and leptin receptors (Ob-Rb) are expressed in the spinal cord dorsal horn and DRG and are upregulated after nerve injury [34,[54][55][56][57]. Leptin-deficient animals or leptin receptor (Ob-Rb)-deficient animals showed an absence of nerve injury-induced neuropathic pain behaviors, which were reversed by exogenous leptin administration [34,36,53,58]. Chronic administration of exogenous leptin induced thermal hyperalgesia and mechanical allodynia in naïve rats [34,35]. Treatment with a leptin antibody abolished nerve injury or exogenous leptin administration-induced neuropathic pain behaviors [34,36]. Our data demonstrated that exogenous leptin administration mimicked SNI-induced mechanical allodynia (Fig. 2a), which further supports the contribution of leptin to neuropathic pain.
NMDARs are one of three ligand-gated ion channels activated by the excitatory transmitter glutamate. Functional NMDARs are a heteromeric complex primarily comprising two NR1 subunits and two NR2 subunits. The different NR2 subunits endow NMDARs with different roles. NR2A predominates at synaptic sites and plays a role in LTP. NR2B is mostly expressed at extrasynaptic sites and is involved in LTD [5,6,13,14]. In ischemia-and brain trauma-induced neuronal injury, NR2A activation promotes neuronal survival, and NR2B activation results in neuronal death [15][16][17][18].
NMDARs play a crucial role in the mechanisms of peripheral and central sensitization [1][2][3][4]. Our previous studies indicated a possible functional link between leptin and NMDARs in neuropathic pain [34,35]. The NMDAR antagonist MK-801 attenuated the upregulation of NMDAR subunit NR1 expression and neuropathic pain behavior induced by the spinal administration of exogenous leptin. Leptin enhanced NMDAR-mediated currents in rat spinal cord slices and enhanced spinal NMDA-induced spontaneous BSL behaviors, which are a behavioral indication of spinal excitation.
The important role of NR2B in the development of neuropathic pain may be due to the impairment of glutamate uptake. Peripheral nerve injury decreased glutamate transporter protein expression and glutamate uptake activity in the spinal dorsal horn [64][65][66]. Impaired glutamate uptake caused glutamate spillover outside active synapses and increased the open probability of NMDA channels, which led to activation of extrasynaptic NMDA receptors [65,67,68]. Nie and Weng demonstrated that blockade of glutamate transporters resulted in an increased proportion of NR2B subunit activation induced by peripheral input, and stronger afferent input further augmented this increase [69]. These data suggested that impairment of glutamate uptake led to the excessive activation of extrasynaptic glutamate receptors, which is a crucial process related to the initiation and maintenance of neuropathic pain.
Our results support a critical relationship between leptin and NR2B-based extrasynaptic NMDARs in neuropathic pain based on the following results: (a) intrathecal administration of the NR2B-selective antagonist Ro25-6981 prevented and reversed SNI-induced allodynia; (b) spinal administration of leptin mimicked SNI-induced allodynia, which was also prevented by Ro25-6981; (c) mechanistic studies showed that leptin upregulated NR2B expression, which was blocked by Ro25-6981, in cultured DRG cells; and (d) leptin enhanced NR2B-mediated currents in spinal cord lamina II neurons. These results indicated that leptin contributed to neuropathic pain via extrasynaptic NMDAR activation. We will further investigate whether leptin directly impairs glutamate uptake and causes glutamate spillover outside of the active synapses, leading to the excessive activation of extrasynaptic NMDA receptors in neuropathic pain.
Whether NR2A is involved in neuropathic pain is controversial. Some studies showed that NR2A expression decreased or did not change in the superficial dorsal horn after nerve injury [7,24,70,71]. Abe et al. found that NR2Adeficient mice developed neuropathic pain similar to wildtype mice after transection of spinal nerve L5 [19]. These data suggest that NR2A is not involved in the process of neuropathic pain. However, other studies demonstrated that the expression level of NR2A increased significantly in the dorsal horn and DRG after nerve injury [25,27]. Nozaki et al. found that zinc alleviated pain via high-affinity binding to NR2A [26]. These results suggest that the role of NR2A in neuropathic pain is complicated.
The present study found that (a) intrathecal administration of the NR2A-selective antagonist NVP-AAM077 prevented and reversed SNI-induced allodynia, and (b) treatment with NVP-AAM077 also prevented spinal leptin-induced allodynia. However, leptin did not upregulate NR2A expression in cultured DRG cells, and leptin did not enhance NR2Amediated currents in spinal cord slices. According to these results, whether NR2A is involved in the contribution of leptin to neuropathic pain is not clear.
Three types of NOS, neuronal NOS (nNOS), endothelial NOS (eNOS), and inducible NOS (iNOS), exist in the central nervous system. nNOS plays an important role in the development and maintenance of neuropathic pain. The expression and activation of nNOS in the spinal cord dorsal horn and DRG are increased after nerve injury, and intrathecal injection of a nonselective NOS inhibitor or selective nNOS inhibitor significantly alleviated pain behavior [22,39,72,73]. Overexpression of nNOS decreased the pain threshold in mice with partial sciatic nerve ligation [40], and nerve injuryinduced mechanical allodynia was significantly reduced in nNOS gene knockout mice [38].
nNOS is involved in neuropathic pain as a downstream target of NR2B. NR2B is concentrated in the postsynaptic density (PSD), where it integrates dynamically with cytoskeletal proteins and signal transduction factors, such as nNOS, and hence it participates in nociceptive signal transduction [22,41]. In some pain models, not only nNOS activity and expression was increased, NR2B expression or NR2B phosphorylation at Tyr1472 in the spinal cord dorsal horn or DRG was also increased [19,39,42,74,75], and NR2B antagonists reversibly reduced nNOS activation to basal levels by [19,22,76].
Our data showed that leptin upregulated the expression of NR2B and nNOS in cultured DRG cells, and the NR2Bselective antagonist Ro25-6981 blocked this upregulation (Fig. 4a, Fig. 5). These results and the aforementioned results suggested, for the first time, that the contribution of leptin to the pathogenesis of neuropathic pain was related to the NR2Bbased extrasynaptic NMDAR-nNOS pathway. We hypothesized that spinal leptin impairs glutamate uptake after nerve injury. The impairment of glutamate uptake causes glutamate spillover outside of the active synapses, which leads to excessive activation of extrasynaptic NMDAR. Extrasynaptic NMDARs directly interact with nNOS in PSD via the PSD95 scaffold protein, and the activation of extrasynaptic NMDARs increases nNOS activation and nitric oxide (NO) production, which drive nociceptive signal transduction and central sensitization. Our data support extrasynaptic NMDARs as a new target for new pharmacological interventions for neuropathic pain.

Compliance with Ethical Standards
Conflict of Interest The authors declare that they have no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.