Microinjection of valproic acid into the ventrolateral orbital cortex exerts an antinociceptive effect in a rat of neuropathic pain.

Rationale: Ventrolateral orbital cortex (VLO) has been found to play an important role in the regulation of neuropathic pain (NPP). As a traditional mood stabilizer, valproic acid (VPA) is currently employed in the treatment of NPP. However, whether VPA plays an analgesic role in VLO is still unknown. Objectives: To elucidate the underlying analgesic mechanism of microinjection of VPA into the VLO on spared nerve injury (SNI), an animal model of NPP. Methods: We firstly examined the role of VPA by intraperitoneal and intral-VLO injection. Then, we accessed its role as a histone deacetylase inhibitor by intral-VLO microinjection of sodium butyrate. Finally, the GABAergic mechanism was measured through the intra-VLO microinjection of several agonists and antagonists of various GABAergic receptor subtypes. Results: Both intraperitoneal and intral-VLO injection of VPA attenuated SNI-induced mechanical allodynia. Microinjection of sodium butyrate, one of the histone deacetylase inhibitors, into the VLO attenuated the mechanical allodynia. Besides, microinjection of valpromide, a derivative of VPA which is a GABAergic agonist, into the VLO also attenuated allodynia. Furthermore, microinjection of picrotoxin, a GABAA receptor antagonist, into the VLO attenuated mechanical allodynia; microinjection of picrotoxin before VPA into the VLO increased VPA-induced anti-allodynia. Besides, microinjection of CGP 35348, a GABAB receptor antagonist, into the VLO attenuated allodynia; microinjection of CGP 35348 before VPA into the VLO also increased VPA-induced anti-allodynia. What is more, microinjection of imidazole-4-acetic acid (I4AA), a GABAC receptor antagonist, into the VLO enhanced allodynia; microinjection of I4AA before VPA into the VLO decreased VPA-induced anti-allodynia. Conclusions: These results suggest that both the histone acetylation mechanism and GABAergic system are involved in mediating VLO-induced anti-hypersensitivity.

Keywords: Ventrolateral orbital cortex; Neuropathic pain; Valproic acid; Histone deacetylase inhibitor; GABAergic modulation

Neuropathic pain (NPP), which is mainly caused by damage or dysfunction of the central or peripheral nervous system, is recognized as one of the most difficult therapeutic pain syndromes in the world (Colloca et al. [7]; Torrance et al. [28]). It has a substantive negative impact on quality of life and causes a heavy financial burden for patients(Lapane et al. [18]). Epidemiological studies suggested that the prevalence of NPP in the general population may be as high as 7 to 8%, representing 20 to 25% of patients with chronic pain (Bouhassira [2]). Therefore, the study of the pathogenesis of NPP is of great significance to find effective treatment methods.

Valproic acid (VPA), a traditional mood stabilizer which usually used in bipolar disorder, is currently employed in the treatment of NPP by exerting the role of histone deacetylase inhibitor (HDACI) and GABAergic agonist (Bialer [1]; Cohen et al. [6]). According to previous studies in our laboratory, microinjection of VPA into the ventrolateral orbital cortex (VLO) exerts an antidepressant effect (Xing et al. [31]). However, whether VPA plays an analgesic role in VLO is still unknown. As a part of the endogenous analgesic system, neuroanatomical and functional studies have demonstrated that VLO receives information from an ascending pathway from the spinal cord via the thalamic nucleus submedius (Sm) and emits signals via a descending pathway to the spinal cord through the periaqueductal gray (PAG) (Tang et al. [27]). Although many studies have revealed that the VLO plays a key role in the analgesia of NPP (Shao et al. [25]; Wei et al. [30]), whether the histone acetylation mechanism is involved in VLO-induced pain modulation is still unknown (Lin et al. [19]; Wei et al. [29]). In addition, only GABAA receptors have been studied in VLO (Huo et al. [13]; Tang et al. [27]). Whether GABAB and GABAC receptors play a role is unknown.

So in the present study, we aimed to examine the role of VPA's analgesic effect in VLO on spared nerve injury (SNI)-induced neuropathic pain model and explore the underlying mechanism.

The following experiments were carried out on male Sprague-Dawley rats (200–250 g) provided by the Medical Experimental Animal Center of Xi'an Jiaotong University (Shaanxi, PR China). Four rats were housed in a cage and freely to food and tap water. The animal room is in a regulation environment (22 ± 1 °C, 50 ± 5% humidity) with 12–12 h light-dark cycle (lights turned on at 07:00 am). The experimental protocol was approved by the Institutional Animal Care Committee of the Xi'an Jiaotong University. All efforts were made to minimize the number of animal used and their misery.

SNI was performed as previously described (Dang et al. [8]). Briefly, the rats were intraperitoneally (i.p.) anesthetized with sodium pentobarbital (50 mg/kg, SCRC, Shanghai, China), and an incision was preformed along the skin of the thigh. After that, biceps femoris muscle was separated so the sciatic nerve and its three branches were exposed. The tibial and common personal nerves were doubly ligated using no. 0 silk and cut in the center of double ligation with removal of 2–3 mm of nerve stump. The sural nerve remained intact and incisions were sutured. Following surgery, the rats were administrated with sodium penicillin (0.2 million units/day for 5 days, i.p.) and allowed to recover for a week.

The rats were i.p. anesthetized with sodium pentobarbital (50 mg/kg, SCRC, Shanghai, China), and the head was fixed in stereotaxic apparatus. A small craniotomy was performed above the VLO, contralateral to the SNI. A stainless steel guide cannula (0.8 diameter) was stereotaxically inserted with the tip 2.0 mm dorsal to the VLO at the following coordinates: 3.2 mm anterior to bregma, 2.0 mm lateral, and 4.6 mm below the cortical surface. Then, the cannula was immobilized with three microscrews and dental cement. Following surgery, the rats were also administrated with sodium penicillin (0.2 million units/day for 5 days, i.p.) and allowed to recover for a week.

One week after the operation of craniotomy, paw withdrawal threshold (PWT) was measured using the up-down method (Chaplan et al. [3]). The rats were placed into a plexiglass box (280 × 250 × 210 mm3) with a stainless steel mesh floor to acclimatize until cage exploration and grooming behavior had ceased for approximately 20 min. Ten Von Frey filaments (Stoelting Company, Wood Dale, IL, USA), with approximately equal logarithmic incremental (0.17) bending force, were chosen (Von Frey numbers: 3.61, 3.84, 4.08, 4.17, 4.31, 4.56, 4.74, 4.93, 5.07, and 5.18, equivalent to 0.4, 0.6, 1.0, 1.4, 2.0, 4.0, 6.0, 8.0, 10, and 15.0 g, respectively). Starting with filament 4.31 (2.0 g), the middle filament in the series, Von Frey filaments with different intensities were repeatedly applied over a 2-s time interval from below and perpendicular to the fourth and fifth toes of the hind paw with sufficient force to cause slight bending against the paw for approximately 6–8 s. If withdrawal response to filament stimulation was positive, the next lower force was delivered. If the response was negative, the next higher force was delivered. Positive and negative responses were recorded and converted to a 50% threshold using a formula provided by Chaplan et al. ([3])). PWT measurements were performed in 10-min intervals over a 60-min observation period. If PWT was decreased to < 4.0 g, the mechanical allodynia was considered to be successfully established.

Before injection, the rat was taken out of the cage and a veil of gauze was put on its head while the guide cannula was left on the outside. A 1.0-μl microsyringe with the tip extending 2.0 mm beyond the end of the guide cannula was inserted into VLO through the guide cannula. Drugs dissolved in normal saline (0.5 μl) were slowly infused into the VLO over a 60-s period with a constant speed. The PWT on SNI-induced allodynia was measured blindly by an experimenter before and 10, 20, 30, 40, 50, and 60 min after drug administration.

The following drugs were used in this experiment (freshly prepared in saline): VPA, valpromide (VPM), sodium butyrate (SB), GABAA receptor antagonist picrotoxin, GABAB receptor antagonist CGP 35348, GABAC receptor antagonist imidazole-4-acetic acid (I4AA) (RBI/Sigma, St. Louis, MO, USA, for every drug).

At the end of the behavioral experiment, all rats were anesthetized with sodium pentobarbital (50 mg/kg), perfused with transcardially 100 ml 0.01 M phosphate-buffered saline (PBS, pH 7.4) followed by 10% formalin. Then, the brain was removed and post-fixed with 10% formalin solution for 7 days. After that, the coronal section of the injection site was obtained at 30 μm by cryomicrotomy. The brain slices were stained with cresol violet according to standard Nissl-staining procedure. Only data from rats with the correct injection location were included in the data analysis (Fig. 1).

Graph: Fig. 1 Histological verification of microinjection representative probe placements in the VLO of rat. a Diagram of rat brain sagittal section showing the position of VLO. b Photomicrographs of a coronal brain section showing the representative positions of rats which were included in the statistical analysis

All data were expressed as the means ± SEM. Linear regression was used to analyze the correlation between drug dose and effect. The differences during the observation time as well as each time point among different groups were tested statistically using two-way analysis of variance (two-way ANOVA), followed by a post hoc multiple comparison. Unpaired Student's t test was also used to analyze between two groups. P < 0.05 was considered to be statistically significant.

Intraperitoneal injection of 200 mg/kg VPA had a significant analgesic effect (p < 0.001, Fig. 2a, b). VPA (10, 20, 50 μg, in 0.5 μl, respectively), administrated into the contralateral VLO, could significantly alleviate SNI-induced hyperalgesia in a dose-dependent manner (r = 0.998, p = 0.018, Fig. 2d). As shown in Fig. 2c, there was a significant difference between treatments (F = 53.96, p < 0.001), across times (F = 13.001, p < 0.001), and between interactions (F = 4.168, p < 0.001) in the time-course curves (i.e., saline and different doses of VPA-treated groups).

Graph: Fig. 2 Effect of VPA intraperitoneal injection and microinjection into the VLO on SNI-induced allodynia. a, b Effect of VPA (200 mg/kg) intraperitoneal injection at the time points and during the entire observation time. c, d Effect of VPA (10, 20, 50 μg) microinjection into the VLO at the time points and during the entire observation time. *p < 0.05; **p < 0.01; ***p < 0.001, compared to saline group; #p < 0.05; ##p < 0.01, compared to 10 μg VPA; +p < 0.05; ++p < 0.01, compared to 20 μg VPA (ANOVA followed by a post hoc multiple comparison). Time point "0" represents baseline score prior to drug injection

VPM (1, 2, 3 μg, in 0.5 μl, respectively), administrated into the contralateral VLO, significantly relieved SNI-induced mechanical hyperalgesia in a dose-dependent manner (r = 0.991, p = 0.087, Fig. 3b). As shown in Fig. 3a, time-course curves were significantly different between treatments (F = 27.261, p < 0.001), across times (F = 24.420, p < 0.001), and between interactions (F = 7.096, p < 0.001).

Graph: Fig. 3 Effect of VPM and SB microinjection into the VLO on SNI-induced allodynia. a, b Effect of VPM (1, 2, 3 μg) microinjection into the VLO at the time points and during the entire observation time. c, d Effect of SB (20, 50, 100 μg) microinjection into the VLO at the time points and during the entire observation time. *p < 0.05; **p < 0.01; and ***p < 0.001, compared to saline group; #p < 0.05, compared to 1 μg VPM; +p < 0.05, compared to 2 μg VPM; (ANOVA followed by a post hoc multiple comparison). Time point "0" represents baseline score prior to drug injection

SB (1, 2, 3 μg, in 0.5 μl, respectively), administrated into the contralateral VLO, significantly relieved SNI-induced mechanical hyperalgesia in a dose-dependent manner (r = 0.999, p = 0.026, Fig. 3d). As shown in Fig. 3c, time-course curves were significantly different between treatments (F = 33.570, p < 0.001), across times (F = 25.652, p < 0.001), and between interactions (F = 7.129, p < 0.001).

GABAA receptor antagonist picrotoxin (100 ng) microinjection into the VLO cannot significantly attenuated SNI-induced mechanical allodynia (p = 0.107, Fig. 4a). But, picrotoxin (100 ng) microinjection into the VLO, 5 min prior to 20 μg VPA, significantly enhanced VPA-induced attenuation of mechanical allodynia (p = 0.015., Fig. 4b). Mechanical PWT scores in picrotoxin + VPA group were significantly increased compared with the picrotoxin group during the entire 60-min observation period.

Graph: Fig. 4 Effect of GABAA and GABAc receptor antagonist on VPA-induced attenuation of SNI-induced allodynia. a, b Effect of GABAA receptor antagonist picrotoxin (100 ng) on VPA (20 μg)-induced attenuation of SNI-induced allodynia at the time points and during the entire observation time. c, d Effect of GABAc receptor antagonist I4AA (100 pg) on VPA (20 μg)-induced attenuation of SNI-induced allodynia at the time points and during the entire observation time. *p < 0.05; **p < 0.01, compared to saline group; #p < 0.05; ###p < 0.001, compared to 100 ng picrotoxin or 100 pg I4AA + 20 μg VPA (ANOVA followed by a post hoc multiple comparison). Time point "0" represents baseline score prior to drug injection

Liking GABAA receptor antagonist picrotoxin, GABAB receptor antagonist CGP 35348 (5, 10, 100 ng, in 0.5 μl, respectively), administration into the contralateral VLO, significantly relieved SNI-induced mechanical hyperalgesia in a dose-dependent manner (Fig. 5b). As shown in Fig. 5a, time-course curves were significantly different between treatments (F = 5.401, p < 0.001), across times (F = 11.18, p < 0.001), and between interactions (F = 83.68, p < 0.001). Furthermore, CGP 35348 microinjection into the VLO, 5 min prior to 20 μg VPA, significantly enhanced VPA-induced attenuation of mechanical allodynia (p = 0.018, Fig. 5c). The PWT scores in CGP 35348 + VPA group were significantly increased compared with the CGP 35348 group during the entire 60-min observation period.

Graph: Fig. 5 Effect of GABAB receptor antagonist CGP 35348 on VPA-induced attenuation of SNI-induced allodynia. a, b Effect of different CGP 35348 doses (5, 10, 100 ng) microinjection into the VLO at the time points and during the entire observation time. c, d Effect of CGP 35348 (5 ng) on VPA (20 μg)-induced attenuation of SNI-induced allodynia at the time points and during the entire observation time. *p < 0.05; **p < 0.01; ***p < 0.001, compared to saline group; #p < 0.05; ##p < 0.01; ###p < 0.001, compared to 5 ng CGP 35348; +p < 0.05; ++p < 0.01; +++p < 0.001, compared to 10 ng CGP 35348 or 20 μg VPA (ANOVA followed by a post hoc multiple comparison). Time point "0" represents baseline score prior to drug injection

Unlike GABAA receptor and GABAB receptor antagonist, GABAc receptor antagonist I4AA (100 pg, in 0.5 μl), administration into the contralateral VLO, significantly enhanced SNI-induced mechanical hyperalgesia (p = 0.031, Fig. 4d). What is more, I4AA microinjection into the VLO, 5 min prior to 20 μg VPA, significantly decreased VPA-induced attenuation of mechanical allodynia (p = 0.048, Fig. 4d).

Results from the present study indicated that not only VPA intraperitoneal injection attenuated spared nerve injury (SNI)-induced mechanical allodynia but also microinjection of VPA into the VLO resulted in SNI-induced mechanical allodynia. Combined with the founding that both microinjection of SB and VPM attenuated allodynia, both the HDACI role and GABAergic agonist effects of VPA are involved in its analgesic effect. Furthermore, microinjection of antagonists and agonists of GABA receptor subtypes into the VLO showed that different GABA receptor subtypes play different roles in VPA-induced anti-hypersensitivity. Inhibition of GABAA and B receptors increased VPA-induced anti-allodynia. Inhibition of GABAC receptor enhanced allodynia.

VPA, an anticonvulsant which has been used in clinical practice for more than four decades, has several mechanisms including the ability to act as an HDACi and the role that increases GABA concentrations in the brain via inhibition of GABA transaminase (Cincarova et al. [5]; Dufour-Rainfray et al. [10]; Rosenberg [23]). More and more studies have proved that VPA is employed in the treatment of NPP, e.g., a Cochrane report reviewed 3 studies (129 participants) that examined VPA in NPP hinting that VPA may reduce pain in postherpetic neuralgia and diabetic neuropathy(Gill et al. [11]). Some other studies approved the efficacy of VPA in trigeminal neuralgia and cancer-related NPP (Hardy et al. [12]; Ross [24]). However, more studies are needed to explore the underlying mechanism of VPA in analgesia. The VLO is an important part of an endogenous analgesic system which is composed of the spinal cord-Sm-VLO-PAG-spinal cord loop (Tang et al. [27]).

Histone deacetylation modification is an important epigenetic regulation method, which is regulated by histone acetylase and histone deacetylase. Many studies have revealed that histone deacetylase inhibitors are documented to attenuate NPP by increasing histone acetylation (Khangura et al. [16]). For example, administration of SB (200 and 400 mg/kg, oral) in chronic constriction injury(CCI)-subjected rats significantly attenuated behavior related to injury-induced pain (Kukkar et al. [17]). In addition, delivery of HDACi MS-275 and MGCD0103 into rat spinal cord in models of traumatic nerve injury and antiretroviral drug-induced peripheral neuropathy relieved the mechanical and thermal hypersensitivity (Denk et al. [9]). In line with these results, our study found that the SNI-induced mechanical hyperalgesia was suppressed by intraperitoneal injection of VPA. Furthermore, microinjection of VPA or SB into the VLO both attenuated the hyperalgesia, indicating that the histone acetylation of the neurons in the VLO may take part in the modulation of NPP.

GABA is the main inhibitory neurotransmitter in the central nervous system (CNS), and its receptors can be divided into three types according to their different pharmacological characteristics: GABAA, GABAB, and GABAC receptors. Different from that GABAA and GABAC receptors can form ligand-gated chloride channels, GABAB receptors belong to the family C of G protein-coupled receptor family and are accompanied by K+ and Ca2 + channels (Chebib and Johnston [4]). Previous studies have demonstrated that GABAergic neurons and GABAergic terminals are widely distributed in VLO (Huo et al. [13]). As an internuncial neuron, GABAergic neurons are involved in various neurotransmitter-mediated analgesic effects in VLO, such as dopamine, 5-HT, and glutamate (Dang et al. [8]; Huo et al. [14]; Zhang et al. [32]). However, only the role of GABAA receptors has been explored in previous studies. Therefore, it is of great significance to study the role of three subtypes of GABA receptors in VLO. In accordance with previous results that microinjection of the GABAA receptor antagonists bicuculline and picrotoxin into the VLO dose-dependently depress the TF reflex and SNI-induced mechanical hyperalgesia (Dang et al. [8]; Qu et al. [21]), the present study found that microinjection of picrotoxin prior to VPA into the VLO increased VPA-induced abirritation.

Besides, our data showed that microinjection of GABAB receptor antagonist CGP 35348 into the VLO exerted dose-dependently demulcent effects similar to GABAA receptor antagonists, and administration of CGP 35348 prior to VPA also enhanced VPA-induced abirritation. In the early 1990s, 10 years after the discovery of GABAB receptors, the GABAB receptor agonist baclofen was already employed as an analgesic, and both systemic and intrathecal injections of baclofen in spinal cord can induce analgesia (Malcangio [20]). However, few studies have been done about the effect of GABAB receptor in CNS. Most studies focused on its role in the regulation of the activity of both peptidergic primary afferent terminals and dorsal horn neurons in the spinal cord. This study fills the gap in this field.

Nevertheless, unlike the abovementioned subtypes, the GABAC receptor played an adverse role. Microinjection of GABAC receptor antagonist I4AA increased the SNI-induced mechanical hyperalgesia. Prior to VPA, microinjection of I4AA into the VLO also lowered the PWT. GABAC receptor, as a traditional retinal-related receptor, has been found to exert effects in many aspects including myopia, memory, sleep, and pain (Johnston et al. [15]). Injection of GABAC receptor agonist CACA (cis-4-aminocrotonic acid) into paws decreased the mechanical hyperalgesia induced by an intraplantar injection of prostaglandin E2 (Reis and Duarte [22]). As for the CNS, activation of GABAC receptor significantly increased the tail-withdrawal latency (Tadavarty et al. [26]). These studies suggest that GABAC receptor is involved in pain modulation in both peripheral and central nervous systems.

In conclusion, the present study suggests that VPA plays an analgesic in VLO-modulated NPP. Both the histone acetylation mechanism and GABAergic system participate in mediating VPA-induced anti-hypersensitivity. Histone acetylation of VLO neurons increases the analgesic effect. Suppression of GABAA and GABAB receptors in the VLO also enhances the PWT, and however, the GABAC receptors may exert the opposite effect.

This research was supported by the National Natural Science Foundation of China (NSFC No. 81771435 to Yong-hui Dang), the Natural Science Basic Research Plan in Shaanxi Province of China (No. 2016JM8078 to Yong-hui Dang), and the research project of State Key Laboratory for Manufacturing Systems Engineering (No. SKLMS 2017002).

The authors declare that they have no conflicts of interest.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.