Activation of HDAC4 and GR signaling contributes to stress-induced hyperalgesia in the medial prefrontal cortex of rats

"Stress-induced hyperalgesia (SIH)" is a phenomenon that stress can lead to an increase in pain sensitivity. Epigenetic mechanisms have been known to play fundamental roles in stress and pain. Histone acetylation is an epigenetic feature that is changed in numerous stress-related disease situations. However, epigenetic mechanism for SIH is not well known. We investigated the effect of histone acetylation on pain hypersensitivity using SPS (single-prolonged stress) + CFA (complete Freund's adjuvant) model. We showed that the glucocorticoid receptor (GR)-pERK-pCREB-Fos signaling pathway was upregulated on stress-induced hyperalgesia and the paw withdrawal threshold in the SPS+CFA group dropped significantly compared with the SPS or CFA group. Histone deacetylases 4 (HDAC4)-expressing neurons in the medial prefrontal cortex (mPFC) were increased in the SPS+CFA-exposed group compared with CFA-exposed or SPS-exposed group. And we showed that the effects of stress-induced hyperalgesia were critically regulated via reversible acetylation (HDAC4) of the GR. Inhibiting HDAC4 by microinjection of sodium butyrate into the mPFC could disrupt glucocorticoid receptor (GR) signaling pathway, which lowered SPS+CFA-caused mechanical allodynia and alleviated anxiety-like behavior. Together, our studies suggest that HDAC inhibitors might involve in the process of stress-induced hyperalgesia.


Introduction
Pain experience is not only consisting of sensory-discriminative dimension, cognitive and also affectional processing in the brain Chen, 2014, 2009). Chronic pain is often endured alongside affective disturbance such as anxiety and depression. Meanwhile abnormal cognition and emotion disorder could influence pain processing. Clinical observations suggest that stress have been shown to exacerbate pain conditions, recognized as stress-induced hyperalgesia (SIH) (Delvaux, 1999;Herrmann et al., 2000;Nash and Thebarge, 2006). The inter-relationship between pain and anxiety has made great challenge for treatmeat of pain. The neurobiological basis for this effect is poorly understood. Stress-induced genetic and epigenetic modifications might be thought to be the underlying cellular mechanism (Schmidt et al., 2013). play a major role in controlling cognition and emotion such as stress induced-anxiety disorders (Wang et al., 2015;Ji and Neugebauer, 2012). Previous studies have showed that stress and pain is under regulation by epigenetic mechanisms such as histone modifications (Stankiewicz et al., 2013;Ligon et al., 2016;Descalzi et al., 2015). Histone deacetylases (HDACs) by removing acetyl groups from lysine side chains regulate protein functions. HDAC4, which belongs to class IIa histone deacetylases family, plays a key role in stress in vitro (Chu et al., 2008) and inflammation-associated thermal hypersensitivity in vivo (Crow et al., 2015). The translocation of HDAC4 from the cytoplasm to the nucleus is induced during stress. Previous studies have showed that the effect of HDAC inhibitor MS-275 (a highly selective inhibitor of Class I HDAC) within the mPFC involves in depression-related symptoms (Covington et al., 2015). HDAC inhibitors are found to improve stress-induced memory impairments, and suppression of HDAC4 by sodium butyrate in the hippocampus could abolish stressinduced effects (Sailaja et al., 2012).
Stress involves in the hypothalamic-pituitary-adrenal (HPA) axis and influence neurotransmission and synaptic plasticity in the prefrontal cortex. Stress could activate the HPA axis, leading to downstream adrenocorticotropic hormone (ACTH) secretion and subsequent secretion of Glucocorticoid (GC) hormones into the circulation (Mifsud et al., 2011). During periods of stress, elevated circulating levels of glucocorticoids are binding to glucocorticoid receptor (GR) (Schmidt et al., 2013), which cause ERK1/2/MSK1-Elk-1 pathway activation in order to enhance the impact on epigenetic and gene expression mechanisms (Mifsud et al., 2011). GRs seem to facilitate the activation of MSK1 and Elk-1 by phosphorylated ERK1/2 (pERK1/2) as scaffold. Our previous studies have also showed that stress-induced hyperalgesia involves activation of ERK1/2 (Qi et al., 2014). However, it is unknown whether chromatin modifications such as histone acetylation regulate GR in the mPFC or GRs act like scaffolds recruiting HDAC in stressinduced hyperalgesia.
In this study, we use a SPS (single-prolonged stress) model (Mifsud et al., 2011) or a modified version of SPS (Wang et al., 2008) to mimic posttraumatic stress disorder (PTSD), injection of complete Freund's adjuvant (CFA) for the persistent pain and a model of SPS + CFA (Qi et al., 2014). SPS included 2 h restraint stress followed immediately by forced swim test (usually 20 min). The associated rats were allowed to recover for a short period (15 min) and were then exposed to diethyl ether until unconsciousness was achieved. Modified version of SPS is aslo a model for PTSD. In this model, a single inescapable electric foot shock was given to rats immediately after SPS procedures. We examined the interaction between HDAC4 and GR in the mPFC during stress-induced hyperalgesia. Our previous studies have demonstrated SPS exacerbates chronic pain. However, whether HDAC4 in mPFC is involved in regulation of GR in the PTSD-pain comorbidity is yet to know. Thus, we inhibited the level of HDAC4 by administering sodium butyrate into the mPFC to determine the contribution of HDAC4 through regulation of GR in the PTSD-pain comorbidity.

Results
2.1. SPS + CFA increased more GRs expression, and GRs facilitated the activation ERK1/2 (pERK1/2), CREB (pCREB) and Fos GR-expressing neurons were observed in the mPFC of rats. Immunofluorescent staining showed that GR-ir was expressed in the nucleus and cytoplasm of the mPFC neurons (Fig. 2). There were significant differences between groups in relation to GR/NeuN-double labeled neurons in NeuN-labeled neurons (F 3, 12 = 58.64, P < 0.001). Morphological analysis showed that there was more GR/NeuN-double labeled neurons in NeuN-labeled neurons in both SPS exposure group and SPS + CFA exposure group than control group (Students-Newman-Keuls test, P < 0.05). There were higher numbers of GR/NeuN-double labeled neurons in NeuN-labeled neurons in SPS + CFA exposure rats compared with CFA or SPS exposure rats (P < 0.05). We also found that the numbers of GR/NeuN-double labeled neurons in NeuN-labeled neurons were higher in SPS exposure rats than CFA exposure rats (P < 0.05). There were significant differences between groups in relation to numbers of GR/pERK1/2-ir neurons (F 3, 12 = 68.37, P < 0.001). Double immunofluorescent staining showed that the colocalization of GR-ir and pERK1/2-ir in the nuclear of mPFC neurons in SPS and SPS + CFA group increased in comparison with the control group (p < 0.05) (Fig. 3). There were higher numbers of GR/pERK1/ 2-ir neurons in SPS + CFA exposure rats compared with CFA or SPS exposure rats (P < 0.05). We also found that the numbers of GR/ pERK1/2-ir neurons were higher in SPS exposure rats than CFA exposure rats (P < 0.05). Additionally, pERK/Fos-ir neurons, pCREB/ Fos-ir neurons were also observed in mPFC neurons in CFA, SPS and SPS + CFA group (Figs. 4,5). There were significant differences between groups in relation to numbers of pERK/Fos-ir neurons (F 3, 12 = 65.18, P < 0.001) and pCREB/Fos-ir neurons (F 3, 12 = 74. 15, P < 0.001). We counted the number of pERK/Fos-ir neurons, pCREB/ Fos-ir neurons and found that the numbers of pERK/Fos-ir neurons, pCREB/Fos-ir neurons were higher in rats of SPS or SPS + CFA exposure than control rats (P < 0.05). There were higher numbers of pERK/Fos-ir neurons, pCREB/Fos-ir neurons in SPS + CFA exposure rats compared with CFA or SPS exposure rats (P < 0.05). We also found that the numbers of pERK/Fos-ir neurons, pCREB/Fos-ir were higher in SPS exposure rats than CFA exposure rats (P < 0.05).

SPS + CFA increased HDAC4 and HDAC4 was enriched in GR-ir neurons
We examined HDAC4 levels in rats induced by CFA, SPS and CFA + SPS. There were significant differences in HDAC4 protein levels between groups (F 3, 12 = 98.36, P < 0.001). Rats receiving SPS or SPS + CFA showed more HDAC4 levels in contrast to the control group (Students-Newman-Keuls test, P < 0.05). There was no change in total H3 expression. We investigated the protein levels of HDAC4. HDAC4 protein expression was increased in the SPS + CFA exposure rats compared with CFA (P < 0.05) or SPS (P < 0.05) exposure rats (Fig. 7).
HDAC4-expressing neurons were observed in the mPFC using IHC. Immunofluorescent staining showed that HDAC4-ir was expressed in the nucleus and cytoplasm of the mPFC neurons. IHC labeling result indicated that 95% HDAC4-ir neurons were positive for GR in SPS + CFA group. There were significant differences in HDAC4/GR double labeled neurons between groups (F 3, 12 = 89.23, P < 0.001). HDAC4/GR double labeled neurons in SPS + CFA exposed significantly increased in comparison with CFA (P < 0.05) or SPS (P < 0.05) group (Fig. 8).
Lysates from control, SPS, CFA and SPS + CFA tissues were immunoprecipitated for GR and blotted for HDAC4 or GR. While total levels of H3 were not different between control, SPS, CFA and SPS + CFA. There were significant differences in GR (F 3, 12 = 76.19, P < 0.001) or HDAC4 (F 3, 12 = 67.19, P < 0.001) protein levels between groups. The levels of GR or HDAC4 were increased in SPS and SPS + CFA exposure rats compared with control rats (P < 0.05). Under these conditions the association between HDAC4 and GR was dramatically increased in the SPS + CFA-exposed rats compared with CFA rats (P < 0.05) or SPS-exposed rats (P < 0.05) ( Fig. 9).

Effects of sodium butyrate on anxiety-like behavior induced by SPS and SPS + CFA by EPM
EPM assessed anxiety-like behavior induced by SPS, CFA and SPS + CFA. The time in open arms (OA time%) and the number of times entry into open arms (OA entries%) were shown. One-way ANOVA showed significant differences in percentage of time spent in open arms (OA time%) (F 11, 84 = 66.28, P < 0.001) and percentage of the number of times of entry into open arms (OA entries%) (F 11, 84 = 99.16, P < 0.001). Rats exposed to SPS + CFA or SPS spent significantly less time in the OA time% and less number in the OA entries% in contrast to the control or CFA group (Students-Newman-Keuls test, P < 0.05) (Fig. 10). Observed reduction in OA time% and OA entries% was obviously changed by Sodium Butyrate chronic treatment in the mPFC of SPS + CFA-(P < 0.05) and SPS-exposed rats (P < 0.05).

Sodium butyrate treatment attenuated SPS + CFA-induced mechanical hyperalgesia
In the present study, there was a significant effect (F 6, 323 = 98.62, P < 0.001) on the paw withdrawal thresholds (PWTs). The SPS + CFA exposure rats obviously decreased the pain threshold of the hindpaw from day 7. Our present and previous analysis revealed that PWTs were significantly reduced in the SPS exposure rats (Students-Newman-Keuls test, P < 0.05), CFA rats after CFA injection (P < 0.05), and SPS + CFA exposure rats (P < 0.01) from day 7 compared to the control rats ( Fig. 11 A). The SPS + CFA exposure rats from day 9 showed significantly lower PWT than SPS exposure rats (P < 0.05) or CFA exposure rats (P < 0.05). Sodium Butyrate chronic treatment in the mPFC increased the PWT in SPS + CFA exposure rats and SPS exposure rats (P < 0.05), but not in CFA exposure rats (P < 0.05) (Fig. 11 A). The data were not showed that the control group treated with Sodium Butyrate altered the baseline mechanical threshold compared to both the control and the control + vehicle group (P > 0.05) and vehicle injection was not significantly different from control group (P > 0.05) (Fig. 11 B).

Discussion
Our studies suggested that (1) The PTSD-pain model increased more serum corticosterone levels. The PTSD-pain model enhanced the mechanical hypersensitivity, and administration of sodium butyrate into the mPFC blocked the hyperalgesia. (2) the PTSD-pain rats showed anxiety-like behaviors, and administration of sodium butyrate into the mPFC attenuated anxiety-like behaviors. (3) SPS-treated rats induced more expression of GRs, pERK, pCREB and Fos in the mPFC. It suggests that SPS-induced changes might underline stress-induced hyperalgesia.
(4) the PTSD-pain rats produced more HDAC4 changes. And, HDAC4/ GR double labeled cells in rats of PTSD-pain model significantly increased in comparison with CFA or SPS group. The protein levels of acetyl-Histone H3 were decreased in rats of SPS and SPS + CFA exposure than control rats. The association between HDAC4 and GR was dramatically increased in the SPS + CFA-exposed rats compared with CFA rats or SPS-exposed rats.
Stress is a state of disharmony or threatened homeostasis, which could modulate pain perception, resulting in either analgesia caused by stressor hyperalgesia caused by stress (Ahmad and Zakaria, 2015). Acute stress could produce antinociception (Costa et al., 2005), while chronic stressful stimuli could produce an increase in pain sensitivity (Imbe et al., 2006). Stressful stimuli could also produce an increase in different type of chronic pain disorder (Delvaux, 1999;Herrmann et al., 2000;Nash and Thebarge, 2006). Epidemiological and clinical studies have been shown that PTSD could exacerbate the chronic pain disorder (Moeller-Bertram et al., 2012). However, less is known about the mechanic of SIH. Activation of the stress system induces various changes in body systems, including activating the hypothalamic-pituitaryadrenal (HPA) axis, causing the glucocorticoids (GCs) release (Carrasco and Van de Kar, 2003). Glucocorticoids enter the brain and bind to glucocorticoid receptor (GR), which is expressed in the prefrontal cortex (de Kloet et al., 2005). Our results showed that SPS could induce more GRs expression in the PFC. Glucocorticoid acting via GRs might lead to stimulation of NMDARs, then through Ca 2+ -CAMKII-ERK1/2 signaling. Our previous and present studies showed that PTSD upregulated ERK1/ 2. Meanwhile, our present studies showed that SPS induce more expression of pCREB and Fos. These results suggest that SPS-induced GC-mediated response might through GRs-ERK-CREB-Fos pathway.
Epigenetics is a heritable phenomenon in which the environment can cause lasting changes in gene transcription without alterations in the sequence of the gene. Recent findings have showed that epigenetics is known to play a key role in pain and chronic stress-induced behavior (Mifsud et al., 2011), whether epigenetics is involved in the stress-induced hyperalgesia remains unknown. Thus, in the present study, we determined the mechanisms of PTSD-pain comorbidity.
Recent findings have showed that during stress gene expression in the central nervous system is regulated by epigenetics (Chmielewska et al., 2019;Zannas et al., 2016). The HPA is important and necessary in coordinating both rapid and long-term behavioral, physiological and molecular responses to psychogenic stressors in the brain system, leading to disorder of the neural, endocrine, and immune system and the secretion of glucocorticoids (GCs) (Stankiewicz et al., 2013). During stress, long-term and rapid, dynamic gene expression is regulated by epigenetic mechanisms such as histone modifications leading to the gene on or off, which results in changes in protein production. These changes might lead to mechanical hyperalgesia. Epigenetic mechanism consists of histone modifications, DNA methylation and microRNA activity. Histone modification is the most important in stress and pain. Histone which regulates gene expression includes histone acetylationby histone acetyl transferases (HATs) and histone deacetylation by histone deacetylases (HDACs). Histone is in charge of organizing DNA into chromat in which can exist in a two forms (closed or open state) (Mifsud et al., 2011). Histone acetylation which is associated with active gene transcription for numerous genes studied is the modification most likely to decondense chromatin and expose previously silent genes for transcription (Kouzarides, 2007). In this study, our morphological data showed that PTSD-pain animals up regulated HDAC4/GR in the mPFC. The protein levels of acetyl-Histone H3 were decreased in rats of SPS and SPS + CFA exposure than control rats. It suggest that HDAC4 might regulate GR in the mPFC contributes to stress-induced hyperalgesia. Furthermore, co-immunoprecipitation studies showed that the association between HDAC4 and GR was dramatically increased in the SPS + CFA-exposed rats compared with CFA rats or SPSexposed rats. Meanwhile, behavior results showed that PTSD-pain animals induced anxiety-like behaviors and hyperalgesia, which could be partly reversed by injection with a highly non-selective HDAC inhibitor, Sodium butyrate. Through the above results, we could figure out stressinduced disorder of the neural, endocrine, and immune system might be associated with histone modification. More HDAC4/GR activation producing emotional disorder might be related to"limbically augmented pain syndrome", which might be the neurobehavioral mechanisms underlying stress-induced hyperalgesia.
Previous results have demonstrated that the epigenetic signal cascade transmission starts with an "epigenator", such as REST and CREB (Berger et al., 2009). Epigenator is a concept, referred to all signals including environmental cues and intrinsic processes which are end up with the recruitment of an "epigenetic initiator". Furthermore, the initiator could cause modifications without signals and persist within a neuron with an "epigenetic maintainer". The epigenetic initiator recruits epigenetic maintainer, and they might act through histone modification and DNA methylation to establish epigenetic patterns (Tsankova et al., 2006). In this study, our data revealed that GRs-ERK-CREB-Fos signaling was upregulated in SPS group compared to control group and GRs-ERK-CREB-Fos was significantly upregulated in SPS + CFA group compared to SPS group. Our results suggest that GRs-ERK-CREB-Fos might serve as epigenetic initiator. HDAC4 are recruited by GR in SPS + CFA, which might be "epigenetic maintainer". Anyway, the epigenetic pattern is still to be known and more studies should be explored in elucidating the phenomenon.
Previous studies showed that there was a relationship between emotional disorder and allodynia (Liu and Chen, 2014). PTSD is with increased cortisol levels. In the present study, SPS could lead to emotional disorder by behavior tests, and significantly reduced paw withdrawal threshold was found in SPS + CFA compared with SPS and CFA. Our results further confirmed that emotional disorder induced by SPS decreased the pain threshold and exaggerated nociceptive sensitivity.
In generally, our present results demonstrate that SPS exposure could exacerbate chronic pain induced by CFA. PTSD-induced emotional impairments could cause an increase in mechanical allodynia. Enhanced HDAC4/GR activation might lead to the hyperalgesia in SPS + CFA, and GR-pERK-pCREB-Fos pathway may be involved in this processing.

Animals
Male Sprague-Dawley rats (weighing 250-300 g) were obtained from laboratory animal center of the 960 Hospital. Animals were grouphoused across different treatment in standard cages, and given access to food and water ad libitum under conditions of 22-25°C ambient temperature and humidity conditions of 40~70% on a 12/12 h dark-light cycle (light on at 6:00 and off at 18:00). All animal procedures were Fig. 6. A Immunoblots of GR, pERK1/2/tERK1/2, pCREB/tCREB and Fos in the mPFC. B, C, D, E Densitometry analysis of western blot bands of GR, pERK1/2/ tERK1/2, pCREB/tCREB and Fos. Compared to the control group, the protein of GR, pERK1/2/tERK1/2, pCREB/tCREB and Fos was up in SPS and SPS + CFA exposure rats (*P < 0.05, vs the control group; # P < 0.05, vs the CFA group; & P < 0.05, vs the SPS group; ◆ P < 0.05, vs the CFA group). The data are presented as mean ± SEM.

Models
SPS procedures were conducted as previously reported for the PTSD model (Antelman et al., 1988). This included 2 h restraint stress followed immediately by forced swim test (usually 20 min). The associated rats were allowed to recover for a short period (15 min) and were then exposed to diethyl ether until unconsciousness was achieved. CFA injection was conducted as previously in our previous studies (Qi et al., 2014) for the chronic inflammatory pain model. Fifty microliters of 50% CFA (Sigma, St Louis, MO, USA; 1 mgM. tuberculosis in 0.85 ml mineral oil and 0.15 ml mannide mono-oleate) in 0.9% saline was injected to subcutaneous plantar of the left hindpaw. SPS + CFA model was also conduced as previously in our previous studies (Qi et al., 2014). Rats were exposed to SPS and then on day 8, CFA injection was given once. A total of 144 animals were used in the study. 72 animals were used and randomly assigned to one of 4 groups (control, SPS, CFA,   9. A The mPFC tissues were immunoprecipitated for GR and blotted for H3 and HDAC4. B Densitometry analysis of bands of HDAC4/ H3. C Densitometry analysis of bands of GR/ H3 (*P < 0.05, vs the control group; # P < 0.05, vs the CFA group; & P < 0.05, vs the SPS group; ◆ P < 0.05, vs the CFA group). The data are presented as mean ± SEM. SPS + CFA). Each group included 18 rats, 9 for behavioral tests, 5 for immunohistochemical staining, and 4 for Western blot analysis and coimmunoprecipitation. For behavioral tests, there were 12 groups. Each group included 9 for behavioral tests in the other eight groups. Behavioral experiments were done double blindly and started at a fixed time during testing days, and rats were always habituated in the testing room for 15 min before behavioral tests.

Antibodies and drugs
The following antibodies were used: mouse-anti-HDAC4 (SAB5300452; Sigma, St Louis, MO, USA); rabbit-anti-pERK1/2 (pERK, 12185; Cell Signaling, Beverly, MA); mouse-anti-pERK1/2 (pERK, 4696; Cell Signaling, Beverly, MA); rabbit anti-total-ERK 1/2 (tERK, 4695; Cell Signaling); mouse-anti-Fos (AB11959; Abcam, Cambridge, MA); rabbit-anti-GR (AB3578, Abcam); rabbit anti-CREB (AB32515, was showed in this histogram in different groups in the EPM test (* P < 0.05, vs the no Sodium Butytate SPS group; ** P < 0.05, vs the no Sodium Butytate SPS + CFA group; # P < 0.05, compared to the no Sodium Butytate control group). Results are expressed as the mean ± SEM. N = 9 rats/group. Fig. 11. A Effect of CFA + SPS and Sodium Butytate injection on mechanical hyperalgesia. Von Frey tests showed that SPS + CFA exposure rats had significantly lower mechanical hyperalgesia. Compared with the control rats, the PWT was decreased in the injured hindpaw of the SPS + CFA exposure rats. Sodium Butytate reversed the PWT reduction (*P < 0.05, compared to the control group; #P < 0.05, compared to the CFA group; &P < 0.01, compared to the control group; □P < 0.05, compared to the SPS group; ◆P < 0.05, compared to no Sodium Butytate-treated SPS group; ** P < 0.01 compared to no Sodium Butytate-treated SPS + CFA group. comparison with Students-Newman-Keuls test). B Effect of CFA, SPS, SPS + CFA, and vehicle chronic injection on mechanical hyperalgesia. CFA, SPS, and SPS + CFA rats had significantly induced mechanical hyperalgesia as shown by the Von Frey tests. The effect of vehicle groups was not significantly different from no Sodium Butytate -treated groups on mechanical hyperalgesia (*P < 0.05, compared to the control group; #P < 0.05, compared to the CFA group; @P < 0.01, compared to the control group; &P < 0.05, compared to the SPS group; ◆ P < 0.05, compared to the SPS group. Comparison with Students-Newman-Keuls test). Results are expressed as the mean ± SEM. N = 9 rats/group. Abcam); rabbit anti-pCREB (AB32096, Abcam). Rats were injected with 100 mg/kg sodium butyrate or saline daily from 2nd −8th day by microinjections into the mPFC. Sodium butyrate, a highlynon-selective HDAC inhibitor.

Microinjections
Rats were anesthetized by intraperitoneal injection of sodium pentobarbital (50 mg/kg) and placed in a stereotaxic apparatus (Kopf Instruments, Tujunga, CA, USA) on the second day. The head was fixed using the ear and bite bars, and an incision exposed the skull. The reference points were as follows: dorsal + 2.9 mm AP, ± 1.9 ML, and − 0.9 DV and ventral + 2.9 AP, ± 1.9 ML, and − 2.8 DV. The cannulae were inserted into the dorsal or ventral mPFC with four screws and dental cement. A dummy cannula was inserted into a guide cannula to keep the cannula tract clear. Rats regained consciousness and were allowed to recover in their home cages before the start of behavioral testing. On the 2nd to 8th day, rats were administered with sodium butyrate or vehicle daily.
A mixture of normal rabbit or mouse sera was used to replace the first specific rabbit and mouse primary antibodies to incubate the sections from the dishes. The following staining procedures used the staining methodology outlined in this section. Immunopositive staining was not observed for these samples.

HDAC 4 co-immunoprecipitation studies
The mPFC tissues collected from control, CFA, SPS and CFA + SPS rats were immediately homogenized in 1X RIPA buffer (Millipore) plus protease inhibitor cocktail (Roche). 250 ug of lysate was diluted to a concentration of 1 ug/ul in RIPA buffer and incubated with 1 ug H3 antibody tumbling overnight at 4°C. Samples were then incubated with 40 ul Protein G beads (Santa Cruz Biotechnology) for 2 h at 4°C while tumbling. Beads were washed twice with RIPA buffer and once with 1X PBS, and eluted in 2X loading buffer (Invitrogen). Proteins were separated using electrophoresis, transferred to a nitrocellulose membrane, and probed for HDAC4 or GR.

Behavioral tests
We used Von Frey filament (Stoelting, Kiel, WI, USA) to evaluate the mechanical allodynia of rats. The rats were covered under inverted plastic boxes (35 × 35 × 55 cm) on an elevated mesh floor to test the withdrawal responses threshold of the left hindpaw evoked by mechanical stimuli. The left hindpaw was touched by one of a series of Von Frey filaments with gradually increasing stiffness (2, 4, 6, 8, 10, 15, and 26 g), which were applied to the plantar surface for 5-6 s. Each filament was applied six times and the paw withdrawal threshold (PWT) was considered the minimal value causing at least six responses. Acute withdrawal, biting, licking, or shaking of the ipsilateral hind limb and vocalization were considered to be positive signs of withdrawal.
The apparatus used for the elevated plus maze test comprises four equally illuminated plastic arms located 50 cm above the floor, including both open arms and closed arms with tall walls. The rats were released from the central area (10 × 10 cm), facing a closed arm (Shanghai Mobiledatum Technology Co., Ltd, China). The behavior was recorded with a video camera. The

Statistical analysis
Behavioral results (EPM) were analyzed by one-way analysis of variance (ANOVA). The number of GR/NeuN, pERK1/2/GR, pERK1/2/ Fos, pCREB/Fos and HDAC4/GR immunoreactive neurons and the expression of protein GR, HDAC4, pERK1/2, pCREB and Fos were analyzed using two-way analysis of variance (ANOVA) across groups. Mechanical allodynia results were analyzed using two-way ANOVA (groups by days). Students-Newman-Keuls method was used as a post hoc test to detect differences between groups. The data were all expressed as means ± S.E.M. The tests were two sided. P < 0.05 was defined as statistical significance.

Funding
This study was funded by the Military Medical Science and Technology Youth Training Program (20QNPY072) and the Military Medical Service support capability innovation and generation special plan (20WQ021).

Author contributions
Jian Qi wrote the main manuscript text. Zhang Li prepared Figs. 1-5.Chen Chen prepared Figs. 6-10. All authors reviewed the manuscript.

Compliance with ethical standards
Ethical approval: All animal procedures were carried out in conformity with Guidelines for the Care and Use of Mammals in Neuroscience and Behavioral Research and approved by the Ethics Committee of Animal protection of the 960 Hospital of the PLA (Jinan, P. R. China). This article does not contain any studies with human participants performed by any of the authors.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.