Introduction

PTSD is classified into the following models based on pathophysiological research: abnormal fear learning model (FL), exaggerated threat detection model (TD), and emotion regulation/functional execution impairment model (EF). Fear extinction dysfunction, a typical manifestation of PTSD, has garnered increasing attention in recent years [1]. In Pavlovian fear conditioning experiments, fear extinction is defined as the inhibition of a conditioned fear response, which initially elicits a fear reflex (freezing) but gradually decreases with prolonged and repeated exposure to a conditioned stimulus (CS) in the absence of the unconditioned stimulus (US). When fear extinction dysfunction occurs, this conditioned fear suppression fails to manifest [2]. Extinction learning involves the acquisition of a memory that antagonizes the original fear memory, rather than forgetting it. This process competes with the previously learned fear memory and forms the basis of the widely used exposure therapy for fear memory [3, 4]. Context is broadly defined as a set of circumstances surrounding an event [5]. Increasing research focuses on the neural mechanisms that encode context in the brain. Neural circuits, including the hippocampus, medial prefrontal cortex, and amygdala, are involved in contextual learning and memory [5, 6]. The hippocampus plays a crucial role in contextual fear conditioning, with regions such as the dorsal hippocampus (dHPC) [7], hippocampal dentate gyrus (DG) [6], CA1 [8], and CA3 [9] being involved in the modulation of contextual fear memory extinction. The hippocampal CA1 area is considered a key region for accurate contextual memory [10]. Therefore, understanding the neurobiomolecular mechanisms related to fear extinction in the hippocampus is crucial for the treatment of PTSD.

Chronic pain affects 30% of the global population, significantly impairing patients’ quality of life [11]. According to the biopsychosocial model, the development of chronic pain is associated with depression [12], anxiety [13], PTSD [14], and other psychological disorders, exhibiting significant comorbidity [11, 15, 16]. The comorbidity of chronic pain and mental illness is becoming a research hotspot in the field of pain [17, 18]. Clinical studies have shown that the incidence of PTSD among patients with chronic pain is higher than in the general population, and patients with PTSD often experience chronic pain [18]. Clinical studies found that 3.5 to 4.7% of people experienced PTSD in the USA; however, the rate of PTSD was much higher in the chronic pain population (9 to 10%) [18]. Numerous clinical studies have found that chronic pain is comorbid with PTSD, but the mechanisms underlying this phenomenon are still unclear due to the lack of animal studies. Postoperative persistent pain, a type of chronic pain, is common and harmful in clinical practice. The SMIR model is a commonly used animal model for postoperative persistent pain [19, 20]. In this study, we used the SMIR model in mice to investigate the molecular mechanisms underlying the comorbidity between chronic pain and PTSD.

NMDARs are seven-transmembrane proteins, composed of four subunits, including NR1 and NR2A/NR2B. NMDARs are widely expressed in the nervous system and play crucial roles in synaptic plasticity, long-term potentiation (LTP), neuronal survival and death, and neurological diseases [21,22,23]. Activated NMDARs open channels to allow Ca2+ influx selectively activating downstream signaling pathways to produce biological functions [24]. Numerous studies have shown that NMDARs-related signaling pathways are important for pain perception [25], but their role in pain-related comorbidities requires further study. Rodent studies have found that injecting the NMDA receptor antagonist MK-801 into the ventral hippocampus of healthy mice impairs consolidation after fear extinction learning [26]. Additionally, the NMDA receptor partial agonist D-cycloserine (DCS) promotes fear extinction and reduces fear recovery [27,28,29,30]. Similarly, NMDA receptor activation has been shown to promote fear extinction in human studies [31]. Consequently, some researchers have proposed treating PTSD by activating NMDARs [21]. However, whether the contextual fear extinction dysfunction in the SMIR model is caused by NMDARs requires further exploration.

Brain-derived neurotrophic factor (BDNF), a member of the neurotrophic factor family, is widely expressed in the hippocampus and participates in neuroplasticity by activating tyrosine kinase receptor B (TrkB) in the postsynaptic density. Many studies have shown that the BDNF-related signaling pathway is vital for pain [32], but its role in pain-related comorbidities remains unclear. Additionally, BDNF plays a key role in fear extinction [33]. Studies have found that increasing BDNF protein content in the hippocampus can enhance contextual fear extinction learning [34]. Reduced BDNF expression, although not affecting the learning of contextual fear memory, altered hippocampus-dependent learning and retrieval of contextual fear extinction [33]. Additionally, reduced BDNF expression interfered with BDNF/TrkB signal transduction and induced dendritic spine loss [35]. Some studies have indicated that the BDNF/TrkB signaling pathway is indispensable for the generation and maturation of dendritic spines [36]. Postsynaptic density protein 95 (PSD-95) is the major scaffolding protein of glutamatergic synapses and interacts with NMDARs to regulate synaptic plasticity [37]. Overexpression of BDNF and TrkB can increase PSD-95 expression and enhance synaptic plasticity [38]. Synaptic plasticity also plays an important role in fear extinction [21]. Activating NMDA receptors can upregulate BDNF expression and enhance synaptic activity, thereby promoting contextual fear extinction [9]. However, the functions of the BDNF/TrkB signaling pathway in fear extinction in SMIR model mice are not clear.

Extracellular signal-regulated kinase (ERK), a downstream effector of NMDARs, can enhance cAMP response element-binding protein (CREB) activity, thereby promoting synaptic plasticity and BDNF synthesis [39, 40]. Upon activation of NMDA receptors, ERK phosphorylation increases [9], which regulates protein expression in the hippocampus under fear conditioning and enhances extinction learning [41]. Therefore, BDNF expression may be regulated through the NMDARs/ERK/CREB signaling pathway.

Overall, we hypothesize that chronic postoperative pain causes fear extinction dysfunction by inhibiting the NMDARs/ERK/CREB/BDNF signaling pathway in the hippocampus. NMDARs and BDNF/TrkB signaling pathways play crucial roles in contextual fear extinction and might serve as target molecules for PTSD. Many clinical studies have revealed the comorbidity between chronic pain and PTSD, but the molecular mechanisms remain poorly understood. We hope that our study may partially explain the molecular mechanisms underlying this comorbidity.

Materials and methods

Animals

Eight-week-old C57BL/6 male mice (25–30 g) were purchased from Jiangsu Jicui Yaokang Biotechnology Co., Ltd., with certificates of qualification for all mice. All animal operations in this study strictly followed relevant regulations and guidelines and were approved by the Animal Care Committee of Anhui Medical University. The animals were kept under standard conditions (12-h light/dark cycle, temperature 22 ± 1 °C, relative humidity 50% ± 10%) with adequate food and water. All efforts were made to minimize the number and suffering of experimental animals.

Surgery and drug injections

Following previous studies [19, 42], we imitated and modified the surgery procedure. The mice were placed in the supine position under 3% sevoflurane anesthesia, and the skin of the right inner thigh was prepared. The shaved skin is then scrubbed repeatedly with a sterile alcohol swab to disinfect the area and expose the great saphenous vein. A 1.0–1.3 cm incision was made on the inner thigh skin about 3 mm medial to the great saphenous vein to expose the thigh muscles. An incision about 1 cm long was then made in the superficial thigh (gracilis) muscle layer, about 3 mm medial to the saphenous nerve. By spreading blunt scissors over the muscle incision, the superficial muscle was further separated to allow insertion of a custom dissecting retractor. The skin and superficial thigh muscles were then retracted by 1 cm to expose the inferior adductor fascia, and the stretching continues for 1 h. Mice in the Sham group received the same anesthesia and surgery, but no skin or muscle stretching was performed. After recovery from anesthesia, all animals were able to walk normally and had normal access to food and water.

The mice were placed in a stereotaxic instrument, and guide cannulae (26 gauge; Plastics One, RWD Life Science) were implanted in 1.4 mm above the dorsal side of the bilateral hippocampus, using CA1 coordinates: −1.18 mm AP; +2.18 mm ML; −1.9 mm DV (Fig. S1). The cannula was fixed on the skull with dental cement, and a stainless steel core was inserted into each casing to keep it unobstructed and prevent infection. The mice were allowed to recover for at least one week after operation.

The NMDARs agonist NMDA (N-methyl-D aspartic acid) and the TrkB receptor inhibitor ANA-12 were both purchased from Sigma-Aldrich (USA) and dissolved in 0.9% normal saline. Following previous studies, each mouse was bilaterally co-infused with 0.005 µg of NMDA [43] and ANA-12 at a concentration of 10 µM [9]. To avoid damage to the hippocampus due to drug injection, ANA-12 was injected immediately after NMDA. Each side of the hippocampus was injected with 1 µl at a rate of 0.06 µl/min, followed by a 2-min diffusion time. The animals were returned to their original cages after the intracranial injection of drugs.

Behavioral experiments

Contextual fear conditioning and extinction

As mentioned earlier [44], the behavioral environments were conducted in a conditioned reflex chamber (Harvard Apparatus, Holliston, MA, USA). The mice were habituated by gentle handling for 5 min per day for three days before fear conditioned. On the first day, the mice adapted to the behavioral environment for 3 min and then received three foot shocks (0.60 mA, 2 s) at 180 s, 250 s, and 320 s. After the third shock, they remained in the chamber for 80 s before being returned to their cage. Twenty-four hours after fear conditioning, the mice underwent 45 min of extinction training without foot shock in the same chamber. After training, the mice were returned to their original cages for 24 h. On the third day, the mice were placed in the same chamber for a 5-min extinction memory retrieval experiment. During the training and testing, the movement of the mice on the grid floor was recorded by the Panlab Startle and Fear combined system with Packwin 2.0 software (Harvard Apparatus, Holliston, MA, USA) as shock amplitude. The degree of fear was assessed by freezing behavior, defined as complete immobility except for breathing, and was recorded using Panlab Startle and Fear combined system with Packwin 2.0 software (Freeze Frame Actimetrics software).

Mechanical allodynia test

The mechanical pain threshold was measured at various time points after establishing the mouse model [42]. The electronic Von Frey apparatus (IITC Life Science, US) was used to measure the withdrawal threshold of the foot under mechanical stimulation. The force causing the withdrawal reflex of each right hind paw was recorded and averaged.

Animal experiment protocol

Mice were randomly divided into the following six groups: Control、Sham、SMIR、SMIR+Vehicle、SMIR + NMDA、SMIR + ANA-12. After SMIR or Sham operation, mechanical pain at the bottom of the right hindfoot was measured using an electronic Von Frey apparatus on the 1st, 3rd, 5th, 7th, 10th, 14th, 21st and 28th days. Contextual fear conditioned, extinction and retrieval experiments were conducted on the 8th-10th days. The SMIR operation was performed after intracranial catheterization in the mice. Contextual fear conditioning was conducted on the 8th day after the operation. After the panic state of the mice subsided, Vehicle or NMDA or ANA-12 was injected into the brain. Extinction training and retrieval experiments were carried out on the 9th and 10th days. After the behavioral tests, the mice were euthanized, and brain tissue was collected for the follow-up experiments.

Western blot analysis

Hippocampal tissue was collected on day 9 post-surgery. The samples were lysed to extract proteins as described previously [45, 46]. The sample were heated at 95 °C for 15 min to denature the proteins. Proteins in the lysed tissue were separated using 10% and 13.5% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes (PVDF) (Millipore, Bedford, MA, USA). The membrane was blocked using TBST containing 0.1% Tween-20 and 5% skim milk, followed by incubation with the following antibodies: anti-NR2A(1:1000, #4205S, Cell Signaling Technology, USA); anti-NR2B(1:1000, #14544S, Cell Signaling Technology, USA); anti-ERK1/2(1:1000, #4695 T, Cell Signaling Technology, USA); anti-P-p44/42MAPK(1:1000, #9101S, Cell Signaling Technology, USA); anti-CREB(1:1000, #9197 T, Cell Signaling Technology, USA); anti-P-CREB(1:1000, #9198S, Cell Signaling Technology, USA); anti-BDNF(1:1000, #EPR1292, Abcam, USA); anti-β3-Tubllin(1:1000, #5666S, Cell Signaling Technology, USA); anti-PSD-95(1:1000, #2507S, Cell Signaling Technology, USA); anti-TrkB(1:1000, #4603, Cell Signaling Technology, USA); anti-TrkB(B-3) (1:500, #sc-7268, Santa Cruz Biotechnology, USA). The membranes were then incubated with HRP-conjugated anti-rabbit and anti-mouse IgG (Cell Signaling Technology, USA) for 1 h at room temperature and washed. The membranes were then incubated with chemiluminescence detection reagents (ECL, BL520A, Bioshap, Hefei, China) and visualized using a chemiluminescence apparatus (ImageQuant LAS 4000, GE Healthcare, Pittsburgh, PA, USA). Protein content was analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

RNA extraction and real-time qPCR

The mRNA from hippocampal tissue was extracted using the HiPure Total RNA Mini Kit (Magen Biotech Co., Ltd). Single-stranded cDNA was synthesized using a reverse transcription kit (#RR036A, TakaRa). The expression of mRNA was detected by DNA amplification kit (#RR820A, TakaRa) and fast real-time qPCR system. The following primers are used: BDNF-R GAGCATCACCCGGGAAGTGT; BDNF-F CAAGGCAACTTGGCCTACCC; β-actin-R CACGATGGAGGGGAATACAG; β-actin-F CAGCTTCTTTGCAGCTCCTT. The mRNA levels of BDNF and β-actin were normalized to β-actin as an internal control.

Golgi-cox

After extraction, the mouse brain tissue was immediately placed in fixative (Servicebio, G1101) and fixed at room temperature for over 48 h. The brain tissue was cut into 2–3 mm thick sections, rinsed gently several times with normal saline, then completely immersed in Golgi-Cox staining solution (Servicebio, G1069), and placed in a cool, ventilated, and light-avoided area for 14 days. The staining solution was changed after the initial 48 h and then every 3 days for a total of 14 days. Subsequently, the tissue was washed three times with distilled water and placed in 80% glacial acetic acid solution overnight. After softening, the tissue was washed with distilled water and placed in 30% sucrose solution. The tissue was cut into 100 µm thick sections using an oscillating microtome, attached to gelatin slides, and dried overnight in the dark. The tissue slides were then treated with concentrated ammonia for 15 min, washed with distilled water for 1 min, treated with acidic film fixative for 15 min, washed with distilled water for 3 min, dried, and sealed with glycerin gelatin (Servicebio, G1402). A digital slide scanner was used to perform panoramic multi-slice scanning to obtain a comprehensive image of the brain tissue. Image acquisition and analysis were performed using a microscope.

Transcriptome sequencing

Five mice were randomly selected from each of the Sham and SMIR groups, and hippocampus samples were sent to OEBiotech (Shanghai, China) for RNA extraction and sequencing. The thresholds for significantly different genes (DEGs) were set at a corrected p-value of 0.05 and a log2(fold change) of 1.0. Differential gene expression was analyzed using KEGG and WikiPathways enrichment, followed by Gene Set Enrichment Analysis (GSEA).

Statistical analysis

All sample sizes were determined based on paradigms used in previous similar studies [44]. Statistical analyses were performed using GraphPad Prism version 9.5.1 and ImageJ. F-tests were used to confirm the mean square error of the data. Normally distributed data were expressed as mean ± SEM, with n representing the number of animals or samples. Each point represents one sample. Two-way repeated measures (RM) ANOVA followed by Tukey’s multiple comparisons test was used to detect significant differences among multiple groups of behavioral data. Other data from multiple groups were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test. Significant was set at P < 0.05.

Results

SMIR mice develop contextual fear extinction dysfunction

As a previous study reported that SMIR mice did not exhibit thermal allodynia [19], we only performed mechanical allodynia measurements in this study. The mechanical pain threshold in the right hind paw was measured using an electronic Von Frey device on days 1, 3, 5, 7, 10, 14, 21, and 28 after SMIR surgery (Fig. 1A). Compared with the Control group, SMIR mice exhibited obvious pain behaviors, while the Sham group did not. Mechanical allodynia in SMIR mice persisted from the first to the 28th postoperative day (Fig. 1B).

Fig. 1: SMIR Mice Develop Contextual Fear extinction dysfunction.
figure 1

A Experimental Flow Chart. B Mechanical allodynia was observed with PWT (n = 6). Differences between groups were adopted two-way RM (Repeated-measures) ANOVA followed by Tukey’s multiple comparisons test. ***p < 0.001, ****p < 0.0001 vs Control; ###p < 0.001, ####p < 0.0001 vs Sham. C Contextual Fear Conditioning Experiment (n = 12). Differences between groups were adopted two-way RM (Repeated-measures) ANOVA followed by Tukey’s multiple comparisons test. D Contextual Fear Extinction Experiment (n = 12). Differences between groups were adopted two-way RM (Repeated-measures) ANOVA followed by Tukey’s multiple comparisons test. E Contextual Fear Retrieval Experiment (n = 12). Differences between groups were adopted one-way ANOVA followed by Tukey’s multiple comparisons test. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001.

Full size image

Mice underwent a 3-day contextual fear extinction protocol as previously described (Fig. 1A). The 8th and 21st post-surgery days were selected for the contextual fear conditioning (CFC) experiment. Compared with the Control group, the SMIR group did not exhibit abnormal fear memory learning (Fig. 1C, S4A). Similarly, in the fear memory retrieval experiment, the SMIR group exhibited similar fear memory to the Control group (Fig. S2). In our experiments, during the extinction learning process, the freezing behavior of control mice decreased rapidly and maintained at a low level. However, the SMIR group exhibited a clear fear extinction dysfunction, manifested as impairment in extinction learning and extinction memory retrieval (Fig. 1D, E, S4B, S4C). To exclude the influence of surgery and anesthesia on fear extinction, we compared the Control group with the Sham group and found that the Sham group did not exhibit fear extinction dysfunction (Fig. 1C, D, E). Additionally, we evaluated locomotor activity and found that the operation had no effect on the locomotor activity of mice (Fig. S5). We also evaluated spatial memory using the Y-maze and found that the operation had no effect on the spatial memory of mice (Fig. S6). The data indicate that SMIR mice exhibit obvious fear extinction dysfunction.

The BDNF/TrkB signaling pathway was inhibited in SMIR mice

To further understand the gene expression in the hippocampus of SMIR mice, we performed transcriptomic analysis. We identified 370 differentially expressed genes (DEGs) in hippocampal samples, including 255 up-regulated and 115 down-regulated genes (Fig. 2A, B). KEGG and WikiPathways enrichment analyses of the differential genes revealed potential signaling pathways, including Alzheimer’s disease, the PI3K-AKT signaling pathway, the MAPK signaling pathway, and the cGMP-PKG signaling pathway (Fig. S3), which were partially consistent with published PTSD transcriptome results [47]. GSEA analysis further revealed the changes in the expression of genes related to BDNF signaling pathway, including the down-regulation of BDNF gene in SMIR mice (Fig. 2C, S3).

Fig. 2: Transcriptomic Analysis of Hippocampus in SMIR Mice.
figure 2

A The heatmap plot (Red represents relatively highly expressed protein-coding genes, and blue represents relatively low-expressed protein-coding genes) and B the volcano plot of differentially expressed genes in the hippocampus between the Sham and SMIR mice. C BDNF Pathway gene set enrichment analysis map (GSEA). D Representative Western blot analysis of BDNF and TrkB-fl (TrkB-full-length) in hippocampus. β-Tublin was included as loading Control. E Quantitative analysis of BDNF levels (n = 5). F Quantitative analysis of TrkB-fl (TrkB-full-length) levels (n = 4).

Full size image

Based on the transcriptomic analysis results and the recognized importance of BDNF in fear extinction [33], we measured the expression of BDNF signaling pathway-related proteins, including BDNF and TrkB-full-length (TrkB-fl) proteins, on the 8th day post-surgery. Compared with the Control group and the Sham group, the expression of BDNF [F (2, 12) = 6.992, p < 0.05; Fig. 2D, E] and TrkB-fl [F (2, 9) = 9.245, p < 0.01; Fig. 2D, F] were reduced in the SMIR group, while there were no significant changes of BDNF and TrkB-fl between the Control group and the Sham group.

The NMDARs/ERK/CREB signaling pathway was inhibited in SMIR mice

As reported by previous studies, BDNF expression is regulated by the NMDARs/ERK/CREB signaling pathway [9]; therefore, we also measured the expression of NMDARs/ERK/CREB signaling pathway-related proteins. The hippocampus tissue of mice was removed on the 9th day post-surgery (Fig. 3H). Compared with the Control group and the Sham group, the expression of NR2B [F (2, 9) = 8.438, p < 0.05; Fig. 3A, B] and NR2A [F (2, 9) = 7.924, p < 0.05; Fig. 3A, C] were decreased in the SMIR group, while there were no significant changes of NR2B and NR2A between the Control group and the Sham group. Compared with the Control group and the Sham group, the expression of p-ERK/ERK [F (2, 9) = 12.98, p < 0.01; Fig. 3A, D] and p-CREB/CREB [F (2, 9) = 32.80, p < 0.001; Fig. 3A, E] were down-regulated in the SMIR group, while there were no significant changes of p-ERK/ERK and p-CREB/CREB between the Control group and the Sham group.

Fig. 3: The BDNF/TrkB and NMDARs/ERK/CREB Signaling Pathways were Inhibited in SMIR Mice.
figure 3

A Representative Western blot analysis of NR2B、NR2A、P-ERK、ERK、P-CREB and CREB in hippocampus. β-Tublin was included as loading control. B Quantitative analysis of NR2B levels (n = 4). C Quantitative analysis of NR2A levels (n = 4). D Quantitative analysis of P-Erk1/2 levels (n = 4). E Quantitative analysis of P-CREB levels (n = 4). F Representative Western blot analysis of PSD-95 in hippocampus. β-Tublin was included as loading Control. G Quantitative analysis of PSD-95 levels (n = 4). H Experimental flow chart. I Representative map of dendritic spine density of neurons in hippocampus. J Quantitative statistics of dendritic spine density (n = 6). Data are presented as the mean ± SEM. *p < 0.05, **p < 0.01.

Full size image

PSD-95 is a crucial scaffolding protein of the postsynaptic zone that interacts with NMDARs and plays a significant role in memory and synaptic plasticity [38]. Therefore, we examined the expression of PSD-95 proteins. Compared with the Control group and the Sham group, the expression of PSD-95 [F (2, 9) = 6.024, p < 0.05; Fig. 3F, G] was decreased in the SMIR group, while there was no significant change of PSD-95 between the Control group and the Sham group. Dendritic spines are also closely related to synaptic plasticity. Therefore, we examined the changes in dendritic spines in hippocampus. Compared with the Control group and Sham group, dendritic spines were missing in the SMIR group, while no significant difference was observed between the Control group and Sham group (Fig. 3I, J).

NMDARs agonist upregulates the expression of BDNF through NMDARs/ERK/CREB signaling pathway

To verify the upstream and downstream relationships among NMDARs, ERK/CREB, and BDNF, we injected NMDA [43], a specific agonist of NMDARs, into the hippocampus on the 8th day after surgery (Fig. 4H). It was confirmed that the mice did not exhibit epileptic behavior after drug injection, and hippocampal tissue samples were collected on the 9th day to detect protein expression. Compared with Sham group, the expressions of BDNF [F (3, 12) = 7.161, p < 0.05; Fig. 4A, B]、TrkB-fl [F (3, 12) = 9.103, p < 0.05; Fig. 4A, C]、p-ERK/ERK [F (3, 12) = 8.480, p < 0.05; Fig. 4D, E] and p-CREB/CREB [F (3, 12) = 522.8, p < 0.01; Fig. 4D, F] were down-regulated in the SMIR and SMIR+Vehicle groups. Compared with SMIR group, there were no differences in the expressions of BDNF [F (3, 12) = 7.161, p > 0.05; Fig. 4A, B]、TrkB-fl [F (3, 12) = 9.103, p > 0.05; Fig. 4A, C]、p-ERK/ERK [F (3, 12) = 8.480, p > 0.05; Fig. 4D, E] and p-CREB/CREB [F (3, 12) = 522.8, p > 0.05; Fig. 4D, F] in the SMIR+Vehicle group, while the expressions of BDNF [F (3, 12) = 7.161, p < 0.05; Fig. 4A, B]、TrkB-fl [F (3, 12) = 9.103, p < 0.05; Fig. 4A, C]、p-ERK/ERK [F (3, 12) = 8.480, p < 0.05; Fig. 4D, E] and P-CREB/CREB [F (3, 12) = 522.8, p < 0.0001; Fig. 4D, F] were up-regulated in the SMIR + NMDA group. Compared with the SMIR+Vehicle group, the expressions of BDNF [F (3, 12) = 7.161, p < 0.01; Fig. 4A, B]、TrkB-fl [F (3, 12) = 9.103, p < 0.01; Fig. 4A, C]、p-ERK/ERK [F (3, 12) = 8.480, p < 0.05; Fig. 4D, E] and p-CREB/CREB [F (3, 12) = 522.8, p < 0.0001; Fig. 4D, F] were up-regulated in the SMIR + NMDA group.

Fig. 4: NMDARs Agonist Upregulates the Expression of BDNF through NMDARs/ERK/CREB Signaling Pathway.
figure 4

A Representative Western blot analysis of hippocampal BDNF and TrkB-fl (TrkB-full-length) after NMDA treatment. β-Tubulin was included as a loading Control. B Quantitative analysis of BDNF levels (n = 4). C Quantitative analysis of TrkB-fl (TrkB-full-length) levels (n = 4). D Representative Western blot analysis of hippocampal P-Erk1/2 and P-CREB after NMDA treatment. Erk1/2 and CREB were included as loading Control. E Quantitative analysis of P-Erk1/2 levels (n = 4). F Quantitative analysis of P-CREB levels (n = 4). G Quantitative analysis of mRNA levels of hippocampus BDNF gene after NMDA treatment (n = 4). H Experimental flow chart. Data are presented as the mean ± SEM. *p < 0.05, **p < 0.01, ****p < 0.0001.

Full size image

Meanwhile, we examined the mRNA expression levels of BDNF after NMDA administration to further validate the regulation of BDNF expression by NMDARs at the transcriptional level. Compared with the Sham group, mRNA [F (3, 12) = 8.144, p < 0.05; Fig. 4G] expression was reduced in both the SMIR and SMIR+Vehicle groups. There was no difference in mRNA [F (3, 12) = 8.144, p > 0.05; Fig. 4G] expression in the SMIR+Vehicle group compared with the SMIR group, whereas mRNA [F (3, 12) = 8.144, p < 0.05; Fig. 4G] expression was increased in the SMIR + NMDA group. Compared with the SMIR+Vehicle group, mRNA [F (3, 12) = 8.144, p < 0.05; Fig. 4G] expression was increased in the SMIR + NMDA group. These results indicated that activating NMDARs could activate the NMDARs/ERK/CREB signaling pathway, thereby upregulating the transcription of BDNF and increasing the expression level of BDNF protein.

NMDARs agonist improves contextual fear extinction dysfunction in SMIR mice by inducing PSD-95 expression and increasing dendritic spines in the hippocampus

Many studies have demonstrated that activating NMDARs can promote fear extinction. We performed CFC training on the 8th day post-SMIR surgery, followed by intracranial NMDA injection into the hippocampus, and subsequent extinction and retrieval experiments on the 9th and 10th days (Fig. 5A). Our results indicated that, compared with the Sham group, mice in the SMIR group and SMIR+Vehicle group exhibited impaired extinction learning and reduced retrieval of extinction memory (Fig. 5B, C, D). No significant difference was observed between the SMIR group and the SMIR+Vehicle group (Fig. 5B, C, D). Compared with the SMIR group and SMIR+Vehicle group, NMDA treatment enhanced extinction learning and increased the retrieval of extinction memory in SMIR mice (Fig. 5B, C, D), suggesting that the NMDARs agonist ameliorated contextual fear extinction dysfunction in SMIR mice.

Fig. 5: NMDARs Agonist Improves Contextual Fear extinction dysfunction in SMIR Mice by Inducing PSD-95 Expression and Increasing Dendritic Spines in the Hippocampus.
figure 5

A Experimental flow chart. B Contextual Fear Conditioning Experiment (n = 11). Differences between groups were adopted two-way RM (Repeated-measures) ANOVA followed by Tukey’s multiple comparisons test. C Contextual Fear Extinction Experiment (n = 11). Differences between groups were adopted two-way RM (Repeated-measures) ANOVA followed by Tukey’s multiple comparisons test. D Contextual Fear Retrieval Experiment (n = 11). Differences between groups were adopted one-way ANOVA followed by Tukey’s multiple comparisons test. E Representative Western blot analysis of hippocampal PSD-95 after NMDA treatment. β-Tubulin was included as a loading Control. F Quantitative analysis of PSD-95 levels (n = 5). G Quantitative statistics of dendritic spine density (n = 6). H Representative map of dendritic spine density of neurons in hippocampus. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Full size image

Studies have demonstrated that BDNF can promote the expression of PSD-95 [38], therefore, we analyzed the protein expression of PSD-95 after NMDA treatment. We found that hippocampal PSD-95 [F (3, 16) = 9.958, p < 0.01; Fig. 5E, F] expression was reduced in the SMIR and SMIR+Vehicle groups compared to the Sham group. No significant difference in hippocampal PSD-95 [F (3, 16) = 9.958, p > 0.05; Fig. 5E, F] was observed between the SMIR group and the SMIR+Vehicle group. Hippocampal PSD-95 [F (3, 16) = 9.958, p < 0.05; Fig. 5E, F] expression was elevated in the SMIR + NMDA group compared to the SMIR and SMIR+Vehicle groups. These results indicated that activation of NMDARs could upregulate PSD-95 expression.

Studies have found that downregulation of BDNF expression can induce dendritic spine deletion [35], therefore, we examined the dendritic spines in the hippocampus. Compared with the Sham group, the SMIR and SMIR+Vehicle groups exhibited dendritic spines deletion (Fig. 5G, H). There was no significant difference in dendritic spine deletion between the SMIR group and the SMIR+Vehicle group (Fig. 5G, H). Dendritic spine density increased in the SMIR + NMDA group compared to the SMIR and SMIR+Vehicle groups. This suggested that SMIR surgery induced dendritic spine deletion in the mouse hippocampus (Fig. 5G, H), and activation of hippocampal NMDARs restored the deleted dendritic spines in SMIR mice.

TrkB receptor inhibitor (ANA-12) reverses NMDA-ameliorated contextual fear extinction dysfunction in SMIR mice by re-inducing hippocampal dendritic spine loss

To further confirm the upstream-downstream relationship between NMDARs and the BDNF signaling pathway, we administered ANA-12 immediately following intracranial injection of NMDA in SMIR mice (Fig. 6A). We found that mice in the Sham group, SMIR + NMDA group, and SMIR + NMDA + ANA-12 group all exhibited normal extinction learning, but mice in the SMIR + NMDA + ANA-12 group and SMIR group showed increased freezing time during memory retrieval (Fig. 6B–D). The results indicated that simultaneous treatment with NMDA and ANA-12 did not impair the process of extinction learning, but impaired the retrieval of extinction memory. Activating NMDARs can promote fear extinction, but the consolidation of extinction memory also requires the transduction of the BDNF/TrkB signaling pathway.

Fig. 6: TrkB Receptor Inhibitor (ANA-12) Reverses NMDA-ameliorated Contextual Fear extinction dysfunction in SMIR Mice by Inducing Hippocampal Dendritic Spine Loss.
figure 6

A Experimental flow chart. B Contextual Fear Conditioning Experiment (n = 11). Differences between groups were adopted two-way RM (Repeated-measures) ANOVA followed by Tukey’s multiple comparisons test. C Contextual Fear Extinction Experiment (n = 11). Differences between groups were adopted two-way RM (Repeated-measures) ANOVA followed by Tukey’s multiple comparisons test. D Contextual Fear Retrieval Experiment (n = 11). Differences between groups were adopted one-way ANOVA followed by Tukey’s multiple comparisons test. E Representative map of dendritic spine density of neurons in hippocampus. F Quantitative statistics of dendritic spine density (n = 9). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Full size image

Consistent with previous studies [35], the loss of dendritic spines was induced by inhibiting the BDNF/TrkB signaling pathway. Our results indicated that the number of dendritic spines was similar in the Sham and SMIR + NMDA groups; however, the SMIR + NMDA + ANA-12 and SMIR groups exhibited dendritic spine loss (Fig. 6E, F), suggesting that ANA-12 reversed NMDA-ameliorated contextual fear extinction dysfunction in SMIR mice by re-inducing hippocampal dendritic spine loss.

Discussion

Multiple clinical studies have found that chronic pain is comorbid with PTSD; however, the molecular mechanisms of this comorbidity remain unclear due to a lack of experimental animal studies. In the present study, we observed that SMIR mice exhibited impaired contextual fear extinction. Additionally, SMIR operation might inhibit the NMDARs/ERK/CREB signaling pathway in the hippocampus, leading to the down-regulation of the BDNF/TrkB signaling pathway, thereby reducing PSD-95 expression and inducing dendritic spine loss in the hippocampus. Notably, activation of NMDARs promotes extinction learning in SMIR mice; however, the consolidation of extinction memory also requires the activation of the BDNF/TrkB signaling pathway. The comorbidity between chronic pain and PTSD is a common challenge in clinical practice. This study may partially elucidate the molecular mechanisms underlying the comorbidity between chronic pain and PTSD, and also validates the crucial role of NMDARs and the BDNF/TrkB signaling pathway in treating fear extinction dysfunctions in SMIR mice (Fig. 7).

Fig. 7
figure 7

Chronic pain causes inhibition of the NMDARs/ERK/CREB signaling pathway, resulting in inhibition of the BDNF/TrkB signaling pathway, down-regulation of PSD-95 expression and reduction of synaptic activity, thereby inducing fear extinction dysfunction.

Full size image

Nowadays, the molecular mechanism underlying the comorbidity between chronic pain and PTSD is still unclear, and specific drugs for the treatment of patients with both chronic pain and PTSD are still lacking [11, 15, 16]. The incidence of PTSD in patients with chronic pain is higher than that in the general population [18]. One explanation is that chronic pain may induce PTSD-like symptoms in patients. Fear extinction dysfunction is considered one of the core features of PTSD, suggesting that chronic pain may affect fear extinction. A recent clinical study demonstrated that patients with chronic back pain (CBP) exhibited impaired extinction memory retrieval compared to healthy controls [48]. These clinical studies suggest that chronic pain can induce fear extinction dysfunction; however, there is currently no research on the mechanism through which chronic pain induces this disorder. According to previous study [19], chronic pain in mice in the current research was induced using a modified SMIR operation method. The SMIR mice exhibited hippocampus-dependent contextual fear extinction dysfunction, manifested by impaired extinction learning and memory retrieval. Our findings are consistent with the aforementioned clinical studies, where chronic pain induced by SMIR surgery led to fear extinction dysfunction in mice.

Many signaling pathways are implicated in the induction and maintenance of PTSD. Wang et al. conducted transcriptomic analysis on the hippocampus of PTSD mice induced by the SPS&S model and found that differentially expressed genes (DEGs) were involved in the regulation of genes associated with Alzheimer’s disease, mTOR signaling, long-term potentiation, PI3K-AKT signaling, and other pathways [47]. To investigate the molecular mechanisms of fear extinction dysfunction in SMIR mice, we also conducted transcriptomic analysis. Similar to the findings of Wang et al., DEGs were also involved in the regulation of genes related to Alzheimer’s disease and the PI3K-AKT signaling pathway. Through gene set enrichment analysis (GSEA), we identified the down-regulation of genes related to BDNF signaling pathways in SMIR mice (Fig. 2). Considering the important role of the BDNF signaling pathway in fear extinction, we further examined the transduction and expression of BDNF signaling pathway-related genes and found down-regulation of this pathway in SMIR mice (Fig. 2).

As one of the most widely distributed neurotrophins in the mammalian brain, BDNF plays a critical role in development, neuronal survival, and particularly activity-dependent neuronal plasticity. Clinical studies have shown that humans with BDNF Val66Met alleles exhibit extinction learning disorders [33, 49]. In animal studies, BDNFMet/Met mice also exhibited fear extinction dysfunction, and after administration of exogenous BDNF, BDNFMet/Met mice showed restored extinction learning [50]. Generally, when the function or expression of BDNF was impaired, fear extinction was facilitated by administering exogenous BDNF to the brain, suggesting that BDNF is crucial for extinction learning [34, 51, 52]. Our study also found that the transcription and translation levels of BDNF were decreased in the hippocampus of SMIR mice. Moreover, the expression level of TrkB was also decreased, and SMIR mice exhibited fear extinction dysfunction. These results suggeste that BDNF/TrkB plays a crucial role in fear extinction.

Numerous studies have demonstrated the crucial role of NMDARs in fear extinction and proposed strategies to treat PTSD by activating NMDARs [21]. Our study demonstrated decreased expression of NR2A and NR2B in the hippocampus and contextual fear extinction impairment in SMIR mice, consistent with the view that inhibition of NMDARs causes fear extinction impairment [9, 26, 28]. However, studies have found that in a rat model of chronic visceral pain, the expression of hippocampal NR2B is increased [53], and in a mouse model of chronic neuropathic pain, the expression of hippocampal NR2A is also increased [54]. The paradoxical expression of NMDARs may result from differences in animal species or types of chronic pain. Subsequently, we used NMDA, a specific agonist of NMDARs, to activate NMDARs. We found that treatment with NMDA promoted fear extinction learning in SMIR mice, similar to the results of extinction training immediately following DCS treatment [55]. Previous studies reported that NMDARs agonist treatment after extinction learning could enhance the consolidation of extinction memory [56, 57]. We also found that SMIR mice exhibited enhanced extinction memory retrieval following NMDA treatment. These studies collectively verified that the strategy of treating PTSD by activating NMDARs was reliable.

Studies have shown that activating NMDARs can induce an increase in BDNF gene expression in hippocampal neurons [58, 59]. Consistent with these studies, we found that activation of NMDARs increased the transcription and translation levels of BDNF. Additionally, we found that the expression of TrkB receptors in the hippocampus of SMIR mice was down-regulated, and this expression was up-regulated after treatment with NMDA. Wang et al. found that activating NMDARs can increase BDNF/TrkB signaling transduction, thereby promoting fear extinction [38], and one report indicated that blocking TrkB can impair memory consolidation [60]. Therefore, we speculated that SMIR mice treated with NMDA promoted the consolidation of extinction memory through the BDNF/TrkB signaling pathway. To test this hypothesis, we simultaneously injected NMDA and the TrkB receptor antagonist ANA-12 into the hippocampus. The results showed that the process of NMDA-restored extinction learning was not blocked by ANA-12 in SMIR mice, but NMDA-restored consolidation of extinction memory was impaired by ANA-12 in SMIR mice. These results indicated that the consolidation of extinction memory requires BDNF/TrkB signal transduction, and the BDNF/TrkB signaling pathway is downstream of NMDARs.

Generally, synaptic NMDARs transmit information through the activation of the survival signaling protein ERK, which drives the transcription of cAMP response element binding protein (CREB), promoting the synthesis of BDNF [39, 40]. We found that the activation of the ERK/CREB signaling pathway in the hippocampus of SMIR mice was inhibited, which may explain the reduction in BDNF expression. Following NMDA treatment, the ERK/CREB signaling pathway was activated, consistent with previous studies. These results suggest that BDNF expression may be regulated by the NMDAR/ERK/CREB signaling pathway in SMIR mice.

The role of synaptic plasticity in fear extinction has garnered widespread attention [4, 21]. The postsynaptic density protein PSD-95 and dendritic spines are closely related to synaptic plasticity [61, 62]. We found that the expression of PSD-95 was reduced in the hippocampus of SMIR mice, suggesting the destruction of synaptic plasticity in SMIR mice. Studies have shown that activating NMDARs or increasing BDNF can promote PSD-95 expression and enhance synaptic plasticity [38, 63]. Additionally, we found that the expression of PSD-95 in the hippocampus of SMIR mice increased after treatment with NMDA. Giza et al.‘s study indicated [59] that BDNFVal66Met, a mutant gene of BDNF, could lead to a decrease in the expression of mature BDNF, inducing the loss of dendritic spines and thus impaired fear extinction This is consistent with our experimental results. Missing dendritic spines were observed in the hippocampus of SMIR mice. Furthermore, activation of the BDNF/TrkB signaling pathway promoted an increase in dendritic spine density [36]. As predicted, we found that lost dendritic spines were restored after treatment with NMDA, and dendritic spines were lost again in the hippocampus of SMIR mice treated with both NMDA and ANA-12. From this, we conclude that in the hippocampus of SMIR mice, activating NMDARs promotes fear extinction by activating BDNF/TrkB signal transduction, increasing synaptic PSD-95, and promoting the increase in dendritic spine density.

Multiple brain areas are involved in the regulation of fear extinction, including the medial prefrontal cortex (mPFC), basolateral amygdala (BLA), and ventral hippocampus (vHPC) [64]. The CA1 region of the hippocampus, dominated by pyramidal cells, is the main information output region of the hippocampus [65]. It receives and processes information from various brain regions inside and outside the hippocampus [66], and outputs information to regions such as the entorhinal cortex (EC) [67]. Research has shown that the anterior cingulate cortex (ACC) and insular cortex (IC) play important roles in brain networks involved in pain and emotion [68, 69]. A clinical study using MRI showed that patients with PTSD had reduced volumes of the hippocampus and ACC [70]. Recent studies have found that neurocircuits from the ACC to the ventral hippocampus mediate contextual fear generalization [71], and there are also bidirectional neural projections between the ACC and the dorsal hippocampus [71]. However, whether the neural circuit between the ACC and hippocampus mediates fear extinction dysfunction remains unclear. It has also been suggested that the thalamic nucleus reuniens (NR)-CA1 pathway is necessary for normal fear extinction [66]. Our research also has many shortcomings. Although the hippocampus is a classic brain region for fear extinction dysfunction, it is not typically associated with chronic pain. However, our study shows significant molecular changes in the hippocampus of SMIR mice. A possible explanation is that pain-related brain areas, such as the anterior cingulate cortex, regulate the activity of the hippocampus. For example, the neurocircuits from the anterior cingulate cortex (ACC) to the ventral hippocampus, identified in previous studies, may be activated in chronic pain mice, based on the fact that the ACC is activated in chronic pain [10, 71, 72]. Additionally, the neurons in the hippocampus may be regulated by pain-related brain areas, potentially inducing molecular and plasticity changes in the hippocampus of chronic pain mice. The upstream brain areas of the hippocampus regulate fear extinction impairment still needs to be studied.