All experimental procedures were reviewed and approved by the Animal Care and Use Committee of Xuzhou Medical University (Approval No. 202207S034) and conducted in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and the ethical standards of the International Association for the Study of Pain. Mice were housed in groups of up to five per cage under a 12 h light/dark cycle (lights on from 08:00 to 20:00), with food and water available ad libitum. Ambient temperature was maintained at 23 °C ± 2 °C with relative humidity at 55%–60%. Only male C57BL/6J mice aged 8–14 weeks (20–30 g) without signs of illness, injury, or abnormal grooming/locomotion were used for the experiments. Food and water were withheld during the modeling and testing sessions (1–2 h). All behavioral tests were conducted during the light phase (09:00–16:00), and experimenters were blinded to the treatment conditions throughout testing. All efforts were made to minimize animal suffering and to reduce the number of animals used.

All AAV vectors were purchased from BrainVTA Technology (Wuhan, China): rAAV-D₂-mCherry-WPREs-bGH (AAV2/9, BrainVTA Technology, PT-0367), rAAV-D₂-hM3D(Gq)-mCherry-WPREs (AAV2/9, BrainVTA Technology, PT-8500), and rAAV-D₂-hM4D(Gi)-mCherry-WPRE-hGH pA (AAV2/9, BrainVTA Technology, PT-8119). The D₂-promoter viral approach was supported by published co-labeling validation in the NAcS, confirming the virus’s high specificity for targeting D₂R-positive neurons (Xia et al., 2025). Viruses were bilaterally injected into the NAcS (AP = +1.65 mm, ML = ±0.55 mm, DV = −4.65 mm) at a volume of 200 nL per side using a microsyringe pump. After injection, the needle was left in place for 10 min to minimize reflux. Experiments were performed 3 weeks after viral injection to allow for stable transgene expression. For chemogenetic experiments, control mice received rAAV-D₂-mCherry-WPREs-bGH to control for viral infection and fluorophore expression without DREADD activation. Clozapine-N-oxide (CNO; MedChemExpress, 34233–69–7) was dissolved in sterile saline containing 0.1% DMSO and administered intraperitoneally (3 mg/kg) 30 min before behavioral testing. To control for potential off-target effects of CNO and its back-metabolism to clozapine, mice in both DREADD-expressing and control-virus groups received the same dose and injection volume of CNO. Vehicle controls (when applicable) received an equivalent volume of saline containing 0.1% DMSO.

Mice were anesthetized with 1% pentobarbital sodium (40 mg/kg, i.p.) via intraperitoneal injection and positioned on a small animal stereotaxic frame (RWD Life Technology Co., Ltd., Shenzhen, China). Ophthalmic ointment was applied to the mice’s eyes throughout the surgery to maintain moisture. A 1 cm scalp incision was made to expose the skull, and the skull was adjusted to align the bregma and lambda horizontally. Small holes were drilled above the target brain region using a dental drill, and bilateral cannula (RWD Life Technology Co., Ltd.) were fixed to the skull with acrylic cement and inserted into the NAcS (AP = +1.65 mm, ML = ±0.55 mm, DV = −4.65 mm). Stainless-steel obturators were inserted into each cannula to prevent occlusion. After the surgery, the mice were placed in a cage with a heating pad to maintain body temperature and returned to their home cage once they were fully awake. Mice were allowed to recover for at least 1 week before behavioral tests. The A_2AR agonist CGS 21680 (MedChemExpress, HY-13201A) and antagonist SCH 58261 (MedChemExpress, HY-19533) were initially dissolved in 100% dimethyl sulfoxide (DMSO) to prepare 10 mM stock solutions and stored at −20 °C. On the day of the experiment, stock solutions were diluted with sterile saline to the desired working concentrations, with the final DMSO concentration kept below 0.1%. The doses used were selected based on our previous studies and preliminary experiments. Control mice received equivalent volumes of saline vehicle. Drugs or vehicle solution (saline containing the same concentration of DMSO) were bilaterally microinjected into the NAcS at a volume of 0.5 μL per side and a rate of 0.2 μL/min. The injection cannula was left in place for an additional 5 min to prevent backflow before removal. After the microinjection, the mice were returned to their home cages for 30 min before behavioral tests.

The SNI surgery was performed in accordance with established protocols (Richner et al., 2011; Cichon et al., 2018). Mice were anaesthetized using sodium pentobarbital (40 mg/kg, i.p.). Clean skin on the lateral surface of the left thigh and then dissect the skin and muscles bluntly to reveal the three branches of the sciatic nerve: the common peroneal nerve, the tibial nerve, and the sural nerve. At the bifurcation of the sciatic nerve, the common peroneal and tibial nerve branches were tightly ligated with 6–0 non-absorbable sutures and transected distal to the ligation site, while the sural nerve was left intact. In the sham group, nerves were exposed and dissected without any sectioning. After the skin was closed, the surgical site was disinfected with povidone-iodine.

Thermal nociception was assessed using the Hargreaves method (Hargreaves et al., 1988). Mice were individually placed on an elevated glass platform within a transparent acrylic container (8 cm long × 8 cm wide × 5.5 cm high) for 1 h in a temperature-controlled, quiet room. A focused beam of light was directed onto the plantar surface of the left hindlimb. The time from the onset of the thermal stimulus to hindlimb withdrawal is recorded as the paw withdrawal latency (PWL). This latency reflects the mouse’s pain threshold. The thermal stimulus duration was limited to 20 s to prevent tissue damage. Testing was repeated 5 times with a 5 min interval between trials, and the average time was recorded as the PWL. All tests were conducted blind.

The forced swimming test was performed as previously described (Porsolt et al., 1977; Cryan et al., 2002). The mouse was individually placed in a transparent acrylic cylinder (25 cm in height, 12.5 cm in diameter) filled with water to a depth of 15 cm, maintained at 25 °C ± 1 °C. After a 1 min habituation period in the water, immobility time was recorded for the following 5 min.

The open field test was performed as previously described (Seibenhener and Wooten, 2015). The mouse was placed individually in a transparent, white acrylic arena (45 cm × 45 cm). After a 1 min habituation period, the mouse’s movement was recorded for the following 10 min. Total distance traveled, velocity, and the amount of time spent in the center versus the periphery of the arena were measured using a video tracking system.

Immunohistochemistry was performed as previously described (Cao et al., 2024). Mice were anesthetized with 1% sodium pentobarbital (40 mg/kg, i.p.) and secured in a supine position on a perfusion table. A midline incision was made along the lower costal margin to expose the thoracic cavity. The diaphragm and both sides of the sternum were sequentially incised to open the thoracic cavity and expose the heart. The pericardium was removed with forceps. A perfusion needle was inserted along the longitudinal axis of the heart into the left ventricle and secured in place. The right atrial appendage was incised, and mice were perfused with 20 mL of cold 0.01 M phosphate-buffered saline (PBS, pH 7.4), followed by 20 mL of PBS containing 4% paraformaldehyde (PFA). Successful perfusion was indicated by pallor of the extremities and whole-body convulsions. After perfusion, the head was severed using tissue scissors. The occipital, parietal, and frontal bones were removed sequentially with care to avoid damaging the brain. All twelve cranial nerves were transected using forceps. The intact brain was carefully extracted and post-fixed in 4% PFA at 4 °C for 12 h. The next day, the PFA solution was replaced with 30% sucrose for cryoprotection. Once the brain sank to the bottom of the sucrose solution, sectioning and staining were performed. The brain was embedded in tissue embedding medium on an ice-cold platform at −20 °C. After freezing, coronal sections (30 μm thick) were cut using a Leica cryostat and collected in PBS.

Sections were incubated overnight at 4 °C in primary antibody dilution buffer (PBST; PBS containing 0.1% Tween-20) containing one of the following primary antibodies: mouse anti-A2AR (1:100, Santa Cruz, sc-32261), rabbit anti-Iba1 (1:500, Wako, 016–20001), rabbit anti-GFAP (1:500, Proteintech, 16825-1-AP), rabbit anti-NeuN (1:500, Proteintech, 26975-1-AP), rabbit anti-DRD1 (1:100, HUABIO, HA751324), or rabbit anti-DRD2 (1:100, HUABIO, ET1703-45). Each primary antibody was applied to separate sections. The next day, sections were washed three times with PBST (5 min each) and then incubated with the corresponding secondary antibody diluted in PBST (anti-mouse Alexa 488, 1:500, Thermo Fisher Scientific, A28175; or anti-rabbit Alexa 594, 1:500, Thermo Fisher Scientific, A11012) for 2 h at room temperature in the dark. Sections were then washed three times with PBST (5 min each) to remove residual secondary antibody, mounted onto slides, and coverslipped with an anti-fade mounting medium. Slides were stored at −20 °C in the dark. Images were acquired using a Zeiss laser scanning confocal microscope (LSM880).

Whole-cell patch-clamp recordings were performed in acute NAcS-containing brain slices as previously described (Zhao et al., 2024). Electrophysiological recordings were performed 6–10 weeks after SNI surgery, at the time when stable PDC was established. The mice were anesthetized with isoflurane, and brains were rapidly removed and immersed in an ice-cold sucrose-based cutting solution (0 °C–4 °C) for approximately 2 min. Brains were then sectioned into 300-μm-thick slices containing the NAcS using a vibratome (DTK-1000). The sucrose-based cutting solution contained (in mM): 254 sucrose, 3 KCl, 1.25 NaH₂PO₄, 10 D-glucose, 24 NaHCO₃, 2 CaCl₂, and 2 MgSO₄ (pH 7.35, 295–305 mOsm). The brain slices were incubated in artificial cerebrospinal fluid (ACSF) equilibrated with 95% O₂ and 5% CO₂ and containing (in mM): 128 NaCl, 3 KCl, 1.25 NaH₂PO₄, 10 D-glucose, 24 NaHCO₃, 2 CaCl₂, and 2 MgSO₄ (pH 7.35, 295–305 mOsm)] for 50 min, then equilibrated at room temperature for approximately 1 h before use. For recording, the brain slices were placed in the recording chamber and continuously perfused with ACSF equilibrated with 95% O₂ and 5% CO₂. Suitable cells were located under a microscope, and a glass pipette electrode (5–8 MΩ) was slowly approached to the cell surface. After achieving a gigaohm(GΩ) seal, membrane rupture was induced by suction to establish the whole-cell recording configuration. Patch pipettes were filled with an internal solution containing (in mM): 130 K-gluconate, 10 KCl, 10 HEPES, 0.5 EGTA, 2 MgCl₂, 2 ATP-Na₂, and 0.3 GTP-Na, pH 7.3, 285 mOsm. To minimize pseudoreplication, we followed standardized recording practices: (1) we recorded from 2-3 brain slices per animal; (2) we recorded no more than 3 cells per slice, with recorded cells separated by at least 100 μm.

Data normality was assessed using the Shapiro–Wilk test. Non-normally distributed data were analyzed with nonparametric methods. For normally distributed data with equal variances, two-group comparisons were performed using unpaired two-tailed t tests. For multiple groups, one- or two-way ANOVA was applied as appropriate, followed by Tukey’s post hoc test. When normality assumptions were not met, two-group comparisons used the Mann–Whitney U test; comparisons among more than two independent groups used the Kruskal–Wallis test; and repeated-measures analyses across related groups used the Friedman test, with Dunn’s multiple comparisons test for post hoc analysis. For electrophysiological experiments, data were analyzed at the level of individual cells (n), and the number of animals (N) from which cells were recorded is consistently reported throughout the manuscript. To address potential concerns about pseudoreplication arising from recording multiple cells from the same animal, we performed complementary analyses at the animal level for all key findings. Specifically, we calculated the mean value for all cells recorded from each animal and then performed statistical tests using these animal-level means. These animal-level analyses confirmed that all major findings remained statistically significant, demonstrating that our conclusions are robust regardless of whether cells or animals are treated as the unit of analysis. Data were presented as mean ± SEM. Statistical significance was set at P < 0.05 (*P < 0.05, **P < 0.01, and ***P < 0.001). Statistical analyses were conducted in GraphPad Prism 8.0.

Previous studies have shown that D₂-MSNs play an essential regulatory role in pain and negative emotions (Ren et al., 2016; Wu et al., 2018). To explore how NAcS D₂-MSNs respond under PDC, we established a neuropathic pain-related SNI mouse model by ligating and transecting the common peroneal and tibial nerves while the sural nerve was left intact (Figure 1A). As previously reported, SNI mice exhibited a sustained decrease in 50%PWTs and PWL (Li et al., 2021). Behavioral tests conducted at multiple time points revealed that, compared to Sham mice, SNI mice showed a significant decline in 50%PWTs (Figure 1B) and PWL (Figure 1C) that persisted for at least 10 weeks. In addition, at week 6 post-surgery, SNI mice displayed a significant increase in immobility time in the forced swimming test (FST), which persisted at least until the 10 th week (Figure 1D). These findings suggest that SNI-induced neuropathic pain induces stable comorbid behaviors, providing a reliable time frame for subsequent functional studies.

Next, we investigated alterations in the excitability of D₂-MSNs in the NAcS under PDC. First, we injected AAV2/9-D₂-mCherry into the NAcS to label D₂-MSNs (Figures 1E,F). At 6–10 weeks post-SNI surgery, we performed whole-cell patch-clamp recordings in acutely prepared NAcS slices and quantified neuronal excitability by measuring depolarizing current-evoked action potentials (eAPs). (Figure 1E). Compared with Sham mice, SNI-PDC mice exhibited a significant increase in evoked action potential firing and a reduced action potential threshold (Figures 1G,H). These results collectively indicate that the intrinsic excitability of NAcS D₂-MSNs is markedly upregulated in PDC mice.

Previous studies have shown that D₂-MSNs play an essential regulatory role in pain and negative emotions. To further explore the potential role of D₂-MSNs in regulating pain and negative emotions in the NAcS, we used chemogenetics to selectively activate D₂-MSNs and assess their effects on pain and despair-like behaviors. First, we injected hM3Dq (AAV2/9-D₂-hM3Dq-mCherry) into the NAcS, while control mice received an equivalent volume of control virus (AAV2/9-D₂-mCherry) (Figures 2A,B). Both experimental and control groups were subjected to identical CNO administration protocols to control for potential off-target effects. Ex vivo whole-cell recordings in acute NAcS slices showed that CNO perfusion increased neuronal excitability in hM3Dq-mCherry-expressing NAcS neurons (Figure 2C). Von Frey and Hargreaves behavioral tests indicated that a single intraperitoneal dose of CNO (3 mg/kg, i.p., administered 30 min before testing) sufficiently reduced 50%PWTs and PWL, whereas no significant changes were observed in control virus-injected mice receiving the same CNO treatment, with no effect on FST immobility time (Figures 2D–F). In contrast, repeated CNO administration for 7 days (3 mg/kg, i.p., once daily) not only reduced 50%PWTs and PWL in mice injected with hM3Dq but also increased FST immobility time compared to mice receiving control virus, while control virus-injected mice receiving repeated CNO did not exhibit these behavioral alterations (Figures 2G,H). These results indicate that chemogenetic activation of NAcS D₂-MSNs is sufficient to induce real-time hyperalgesia, whereas the emergence of despair-like behavior requires prolonged/repeated activation of NAcS D₂-MSNs.

To assess the necessity of activating D₂-MSNs in PDC, we established a neuropathic pain-related SNI mouse model (Figure 3A). We replaced AAV2/9-D₂-hM3Dq-mCherry with AAV2/9-D₂-hM4Di-mCherry, with SNI-control mice receiving AAV2/9-D₂-mCherry, and performed the same chemogenetic viral surgery (Figures 3A–C). Both the hM4Di and SNI-control groups received identical CNO administration protocols. One week after SNI surgery, behavioral tests using von Frey and Hargreaves showed that a single dose of CNO (3 mg/kg, i.p., administered 30 min before testing) increased the 50%PWTs and PWL in SNI mice injected with AAV2/9-D₂-hM4Di-mCherry, whereas no significant effects were observed in control virus-injected SNI mice receiving CNO (Figures 3D,E). Similarly, 6 weeks after SNI surgery, acute CNO administration (3 mg/kg, i.p., administered 30 min before testing) increased the 50%PWTs and PWL in SNI mice injected with AAV2/9-D₂-hM4Di-mCherry (Figures 3F,G) and decreased immobility time in the FST (Figure 3H), indicating that chemogenetic inhibition of NAcS D₂-MSNs in SNI-PDC mice also exhibited antidepressant effects in the FST. These findings strongly suggest that inhibiting NAcS D₂-MSNs alleviates neuropathic pain behavior and attenuates despair-like behavior in SNI-PDC mice.

To characterize the cellular localization of A_2ARs in the NAcS, we performed immunofluorescence co-staining with the neuronal marker Neuronal Nuclei (NeuN). A_2AR immunoreactivity was readily detected in NeuN-positive cells, indicating neuronal expression of A_2ARs in the NAcS (Figures 4A,B). Similar immunofluorescent co-staining was performed using the microglial marker ionized calcium-binding adapter molecule 1 (Iba1, Figures 4C,D) and the astrocytic marker glial fibrillary acidic protein (GFAP, Figures 4E,F). A_2AR signals showed prominent colocalization with Iba1 (Figures 4C,D), whereas little to no overlap was observed with GFAP (Figures 4E,F). These data indicate that A_2ARs exhibit cell-type-specific expression in the NAcS, primarily localizing to neurons and microglia rather than astrocytes.

In the NAcS, D₁ and D₂ receptors are predominantly distributed (Conrad et al., 2010), both of which play crucial roles in regulating neurocircuitry associated with emotion and pain (Ikemoto et al., 1997; Domingues et al., 2025; Wang et al., 2023). Previous studies have confirmed the widespread expression of A_2ARs in the brain and have further revealed their specific distribution in the NAcS (Zhang et al., 2013). Immunofluorescence analysis revealed substantial cellular-level colocalization of A_2ARs with D₂Rs (Figures 4G–J), suggesting that A_2ARs are enriched in D₂R-positive cell populations in the NAcS.

Next, we explored how A_2AR signaling in the NAcS modulates the excitability of D₂-MSNs. We injected the AAV2/9-D₂-mCherry virus into the NAcS (Figures 5A,B) to specifically label D₂-MSNs. Subsequently, we assessed neuronal excitability by quantifying eAPs. In isolated NAcS slices prepared from naïve mice, perfusion with the A_2AR agonist CGS 21680 (0.1 μM) significantly increased the number of eAPs in D₂-MSNs and reduced the rheobase (the minimal current required to evoke an action potential) (Figures 5C,D), indicating an enhancement of intrinsic excitability. To further validate the regulatory role of A_2ARs, we tested the selective A_2AR antagonist SCH 58261 (0.1 μM) under the same conditions. In isolated NAcS slices from naïve mice, perfusion with the A_2AR antagonist SCH 58261 (0.1 μM) significantly decreased the number of eAPs, and increased the rheobase (Figures 5E,F) in NAcS D₂-MSNs, showing that the A_2AR antagonist downregulated NAcS D₂-MSNs neuronal excitability. Collectively, these electrophysiological data indicate that A_2ARs bidirectionally modulate the excitability of D₂-MSNs in the NAcS, with agonist activation enhancing and antagonist blockade suppressing neuronal activity.

To determine whether local pharmacological manipulation of NAcS A_2AR activity modulates pain and despair-like behaviors in naïve and SNI mice, we bilaterally implanted guide cannulas targeting the NAcS 2 weeks before behavioral tests for intra-NAcS microinjections of the A_2AR agonist or antagonist (Figures 6A,B). Our behavioral tests demonstrated that a single microinjection of the A_2AR agonist CGS 21680 (2 ng/side) significantly decreased the 50%PWTs (Figure 6C) and PWL (Figure 6D) in naïve mice under physiological conditions, with no significant effect on immobility time in the FST (Figure 6E). No significant effect on motor activity was observed in the OPT (Supplementary Figure S1C). However, five times microinjections of CGS 21680 (2 ng/side) increased immobility time in the FST (Figure 6F). We next examined the behavioral consequences of intra-NAcS A_2AR antagonism using the selective antagonist SCH 58261 (4 ng/side). We found that a single microinjection of SCH 58261 increased 50%PWTs and PWL in naïve mice (Figures 6G,H) and decreased immobility time in the FST (Figure 6I). No significant effect on motor activity was observed in the OPT (Supplementary Figure S1D). These pharmacological findings suggest that A_2ARs exert bidirectional regulatory effects on pain and depression-like behaviors in the NAcS.

To further validate the analgesic effects of NAcS A_2AR antagonism in SNI comorbid mice, SCH 58261 (4 ng/side) was injected into the NAcS 30 min before behavioral tests (Figure 7A). Compared to the saline-injected SNI group, a single microinjection of the A_2AR antagonist SCH 58261 (4 ng/side) significantly increased 50%PWTs (Figures 7B,D) and PWL (Figures 7C,E) in mice at week 1 and week 6 after SNI, with no significant effect on immobility time in the FST (Figure 7F). Repeated administration of SCH 58261 for 5 consecutive days (once per day) not only increased the 50%PWTs and PWL (Figures 7G,H) in SNI-PDC mice but also decreased immobility time in the FST (Figure 7I). These results suggest that NAcS A_2AR antagonism attenuates neuropathic pain during both early and later post-SNI phases, and that chronic treatment is required to produce robust antidepressant-like efficacy in the FST in the established PDC state.