Abstract
Patients with chronic pain often develop comorbid depressive symptoms, which makes the pain symptoms more complicated and refractory. However, the underlying mechanisms are poorly known. Here, in a repeated complete Freund's adjuvant (CFA) male mouse model, we reported a specific regulatory role of the paraventricular thalamic nucleus (PVT) glutamatergic neurons, particularly the anterior PVT (PVA) neurons, in mediating chronic pain and depression comorbidity (CDC). Our c-Fos protein staining observed increased PVA neuronal activity in CFA-CDC mice. In wild-type mice, chemogenetic activation of PVA glutamatergic neurons was sufficient to decrease the 50% paw withdrawal thresholds (50% PWTs), while depressive-like behaviors evaluated with immobile time in tail suspension test (TST) and forced swim test (FST) could only be achieved by repeated chemogenetic activation. Chemogenetic inhibition of PVA glutamatergic neurons reversed the decreased 50% PWTs in CFA mice without depressive-like symptoms and the increased TST and FST immobility in CFA-CDC mice. Surprisingly, in CFA-CDC mice, chemogenetically inhibiting PVA glutamatergic neurons failed to reverse the decrease of 50% PWTs, which could be restored by rapid-onset antidepressant S-ketamine. Further behavioral tests in chronic restraint stress mice and CFA pain mice indicated that PVA glutamatergic neuron inhibition and S-ketamine independently alleviate sensory and affective pain. Molecular profiling and pharmacological studies revealed the 5-hydroxytryptamine receptor 1D (Htr1d) in CFA pain-related PVT engram neurons as a potential target for treating CDC. These findings identified novel CDC neuronal and molecular mechanisms in the PVT and provided insight into the complicated pain neuropathology under a comorbid state with depression and related drug development.
- 5-hydroxytryptamine receptor 1D
- affective pain
- pain and depression comorbidity
- paraventricular thalamic nucleus
- sensory pain
Significance Statement
Patients with chronic pain frequently experience concurrent depressive symptoms, which complicates the treatment of the pain symptoms. Recently, the paraventricular thalamic nucleus (PVT) has attracted growing interest in modulating animal behaviors of pain sensation and stress-related affective disorders. However, whether and how PVT mediates chronic pain and depression comorbidity (CDC) remains poorly understood. Here, utilizing chemogenetics, viral labeling, neuronal cell type-based TRAP (translating ribosome affinity purification) technology, and pharmacological studies, our current study revealed that PVT is a central hub for regulating sensory pain and depressive-like behaviors but not affective pain by functioning through the 5-hydroxytryptamine receptor 1D (Htr1d). These findings provide new insight into the regulatory role of the PVT and a potential molecular target for future analgesic development.
Introduction
Chronic pain has been recognized as pain that persists beyond normal tissue healing time, usually lasting longer than 3–6 months with heavy socioeconomic burdens (Mills et al., 2019; Bonilla-Jaime et al., 2022; Meda et al., 2022; Patel and Dickenson, 2022; Johnston and Huckins, 2023; Yang et al., 2023). Compared with patients with acute pain, those with persistent chronic pain usually have poor treatment outcomes with standard biomedical measures (Patel, 2023). This is primarily due to the development of comorbid states with mood disorders in these patients, including depression (Bair et al., 2003; Cohen et al., 2021; Meda et al., 2022; Patel, 2023). Epidemiological studies indicate that about 20–90% of chronic pain patients in different clinics are identified with depression, and the lifetime prevalence of depression is more than doubled in patients with chronic pain compared with those without (Bair et al., 2003; Meda et al., 2022; Mullins et al., 2023). An increasingly growing body of evidence suggests that chronic pain-induced sensory and affective pain symptoms involve different neural adaptations (Zhu et al., 2021; Llorca-Torralba et al., 2022). Mutual reinforcement between maladaptations underlying the sensory and affective processing complicates pain symptoms under comorbidity, which makes the treatment more challenging (Patel, 2023). The neurophysiological mechanisms of chronic pain and depression comorbidity (CDC) remain poorly understood, though recently receiving increasing interest with the revolution of modern neuroscience targeting, observing, and manipulating strategies (Zhou et al., 2019; Zhu et al., 2019; Nectow and Nestler, 2020; Zhu et al., 2021; Patel, 2023). Understanding the neural and molecular mechanisms underlying CDC has long been a challenge for preclinical studies and will benefit future analgesic strategy development and guide clinical practices.
The paraventricular thalamic nucleus (PVT), an almost 100% glutamatergic midline thalamic nucleus, is a crucial hub for information processing and has attracted growing interest in modulating animal behaviors, including pain sensation and stress-related affective disorders (Hsu et al., 2014; Cheng and Chen, 2018; Barson et al., 2020; Kooiker et al., 2021; Deng et al., 2023; Zhang et al., 2023). For example, we recently reported that PVT glutamatergic neurons exhibited elevated neuronal activity in mice with inflammatory pain and identified nucleus accumbens (NAc)-projecting PVT glutamatergic neurons as a critical neural circuit for pain sensation under physiological and pathophysiological states (Zhang et al., 2023). In a well-established chronic social defeat stress model of depression (Krishnan et al., 2007; Zhang et al., 2019), PVT-derived enhanced excitatory neurotransmission was observed in the NAc shell of the depressed susceptible mice, and projection-specific chemogenetic inhibition ameliorates chronic stress-induced depressive-like behaviors (Deng et al., 2023). However, whether and how PVT glutamatergic neurons regulate CDC induced by chronic pain remains poorly understood.
In the present study, utilizing a combination of cell-type-specific chemogenetics, molecular profiling, and pharmacological and behavioral approaches, we identified the PVT, particularly the PVA glutamatergic neurons as a novel cellular target in mediating sensory pain, but not affective pain, and depressive-like symptoms in CDC mice. Interestingly, the affective component of pain symptoms in CDC mice could be selectively alleviated by systematically administering S-ketamine, a rapid-onset antidepressant (Berman et al., 2000; Zarate et al., 2006; Li et al., 2010; Qin et al., 2023). Our molecular profiling data revealed 5-hydroxytryptamine receptor 1D (Htr1d) in the PVA as a potential target for treating CDC, which was confirmed by following pharmacological experiments. Thus, our findings identified novel neural and molecular targets for CDC and provided helpful information for understanding its neuropathology and developing future medications.
Materials and Methods
Animals
C57BL/6J (purchased from Cavens) male mice at the age of 8–12 weeks were used. Animals were housed under standard laboratory conditions (12 h light/dark cycle, lights on from 08:00 A.M. to 20:00 P.M., temperature of 23 ± 2°C, and humidity of 55–60%) with ad libitum access to standard lab mouse pellet food and water (Xia et al., 2020; Zhang et al., 2023). Efforts were made to minimize animal suffering and reduce the number of animals used. All experiments were reviewed and approved by the Animal Care and Use Committee of Xuzhou Medical University (approval number: 202209S054).
Adeno-associated virus (AAV) vectors
All viruses were purchased from BrainVTA Technology or BrainCase : rAAV-Ef1α-DIO-hM3D(Gq)-mCherry (AAV2/9, 5.14 × 1012 genomic copies per ml, 200 nl, BrainVTA), rAAV-CaMKIIa-CRE-WPRE-hGH-pA (AAV2/9, 5.36 × 1012 genomic copies per ml, 200 nl, BrainVTA), rAAV-Ef1α-DIO-hM4D(Gi)-EGFP-WPREs-pA (AAV2/9, 5.09 × 1012 genomic copies per ml, 200 nl, BrainVTA), rAAV-cFos-tTA-NLS-FLAG (AAV2/9, 2.07 × 1012 genomic copies per ml, 200 nl, BrainCase), and rAAV-TRE3G-HA-NBL10 (AAV2/9, 2.45 × 1012 genomic copies per ml, 200 nl, BrainCase).
Complete Freund's adjuvant-induced model of inflammatory pain
Mice were anesthetized with pentobarbital sodium (50 mg/kg, i.p.). Complete Freund's adjuvant (CFA, 10 µl, InvivoGen) was subcutaneously injected into the left hindpaw plantar surface with microliter syringes to induce inflammatory pain (Zhou et al., 2019; Zhu et al., 2021; Zhang et al., 2023). The chronic inflammatory pain was established by a booster injection of CFA every other 7 d four times in total. Sham animals receive injections of saline (0.9% NaCl).
Spared nerve injury model of neuropathic pain
Spared nerve injury (SNI) and sham surgeries were performed under anesthesia with pentobarbital sodium (50 mg/kg, i.p.). The sciatic nerve consisting of the common peroneal, tibial, and sural nerves was exposed by blunt dissection of the skin and muscles of the left thigh. The common peroneal and tibial branches were ligated with 5-0 silk-braided nonabsorbent suture, and a 2 mm section from the ligature was removed, leaving the sural nerve intact (Zhou et al., 2019; Zhu et al., 2021). The skin was stitched and disinfected with iodophor. The sciatic nerve was exposed for the sham animals.
Chronic restraint stress
Mice were subjected to chronic restraint stress, as previously reported (Zhu et al., 2021). Briefly, mice were restrained in a 50 ml plastic centrifuge tube 6 h/d for 3 weeks. At the same time, the control mice were deprived of food and water and allowed to move freely in their home cages.
Fifty percent paw withdrawal thresholds (50% PWTs)
The mechanical paw withdrawal threshold (PWT) measurement was adapted from and carried out with the up-down paradigm as previously described by Chaplan et al. (1994) and our previous studies (Zhao et al., 2021; Zhang et al., 2023). Mice were acclimatized for 1 h in transparent acrylic enclosures (10 cm × 10 cm × 20 cm) on a wire mesh platform in a temperature-controlled, quiet room. A sequence of calibrated von Frey filaments (North Coast, 0.02, 0.04, 0.07, 0.16, 0.4, 1.0, 2, and 6 g) was chosen for the measurements, starting with the 0.16 g hair. Each hair was applied perpendicularly to the plantar surface of the left hindpaw, with sufficient force to bend the filament. Lifting, shaking, or licking the paw indicated a positive response and prompted the next weaker filament, and the absence of a paw withdrawal response prompted the next stronger filament. This paradigm continued until a total of six measurements or until four consecutive positive or four consecutive negative reactions occurred. The 50% mechanical withdrawal thresholds were calculated as
Assessment of depressive-like behaviors
For all the following behavioral tests, mice were habituated for 24 h in the test room before test. The behavior was recorded using a tracking system (SMART 3.0, Panlab), and the light condition was kept ∼20 lux. For the chemogenetic experiment, the mice were injected with clozapine N-oxide (CNO, hM3Dq DREADDs, 1 mg/kg; hM4Di DREADDs, 3 mg/kg) (Clarke et al., 2023), and the depressive-like behaviors (tail suspension test, TST, and forced swimming test, FST) were measured 0.5 h after injection, or on the next day of the last injection (for repeated manipulation experiments).
TST
Mice were suspended about 50 cm above the surface of an experimental table with adhesive tape, which was placed 1 cm from the tip of the tail to prevent the mouse from climbing upwards. Each mouse was tested for 6 min and the immobility time of the animal was measured in the last 5 min. The mouse was considered immobile when there were only small movements of the forefeet or no movement at all (Zhang et al., 2018).
FST
Mice were placed in a Plexiglas cylinder, 30 cm in height and 15 cm in diameter, containing water to a depth of 15 cm so that the mice could not touch the bottom with their hindpaws. The water temperature was kept at 25 ± 1°C. Each mouse was placed in the cylinder for 6 min, and the last 4 min was defined as the test stage. The duration of immobility time of the animal was recorded. The mouse was considered immobile when floating in the water or swimming with only one leg to slide slightly. The mouse nose should be kept above the water, and related movement was also defined as immobility (Can et al., 2012; Zhang et al., 2019; Qin et al., 2023).
Immunohistochemistry
Mice were deeply anesthetized with pentobarbital sodium (50 mg/kg, i.p.) and subsequently perfused with phosphate-buffered saline (0.01 M PBS) and 4% (w/v) paraformaldehyde. The mouse brains were carefully dissected and post-fixed in 4% paraformaldehyde at 4°C for 12 h. The mouse brains were then transferred to 30% (w/v) sucrose solution at 4°C until the brains saturate. The brains were sectioned at a thickness of 30 µm using a cryostat (Leica Biosystems). The sections were washed in 0.01 M PBS three times. Brain sections were subsequently blocked in 1% bovine serum albumin for 2 h at room temperature and incubated with the primary rabbit anti-c-Fos (1:800, 2250, Cell Signaling Technology) overnight at 4°C in TBS (0.1% Triton X-100 in PBS), followed by the secondary anti-rabbit Alexa-488 (1:500, A-21206, Thermo Fisher Scientific) or the secondary anti-rabbit Alexa-594 (1:500, A-21207, Thermo Fisher Scientific) in TBS for 1 h at room temperature. Fluorescence images were captured by a Zeiss confocal microscope (LSM880) (Ma et al., 2023; Zhang et al., 2023).
Stereotaxic surgeries and microinjections
Stereotaxic surgeries were conducted using a stereotaxic apparatus (RWD Life Technology) after general anesthesia with pentobarbital sodium (50 mg/kg, i.p.). For virus injection, a volume of 200 nl viruses was injected into the PVT with a microliter syringe (unilateral coordinates, AP, −0.30 mm from Bregma; ML, 0.00 mm lateral from midline; and DV, −4.3 mm vertical from the cortical surface) (Gao et al., 2020) at a speed of 100 nl/min with a 10 min delay before the needle was withdrawn. For local infusion, the cannula was implanted above the PVT of the mouse brain. Agonists and antagonists of Galr1 (M617 TFA, GC61607, GlpBio; M35, GC36102, GlpBio) and Htr1d receptors (GR 46611, GC14373, GlpBio; BRL 15572 hydrochloride, GC50012, GlpBio) or saline–vehicle were infused once or once a day for 7 d. The injector cannula was removed with a 5 min delay to prevent backflow.
Electrophysiology
To verify the function of the chemogenetic viruses (AAV-hM4Di or AAV-hM3Dq), the mice were anesthetized with isoflurane for 30 min. As we previously reported (Zhang et al., 2019; Ma et al., 2023; Zhang et al., 2023), brains were quickly removed and chilled in a high-concentration sucrose buffer [254 mM sucrose, 3 mM KCl, 1.25 mM NaH2PO4, 10 mM D-glucose, 24 mM NaHCO3, 2 mM CaCl2, and 2 mM MgSO4 (pH 7.35, 295–305 mOsm) at 0–4°C for about 2 min]. Coronal sections (250 µm thick) containing the PVT were cut in the dissection buffer using an automated vibrating-blade microtome (DTK-1000) and were subsequently transferred to an incubation chamber containing artificial cerebrospinal fluid (ACSF): 128 mM NaCl, 3 mM KCl, 1.25 mM NaH2PO4, 10 mM D-glucose, 24 mM NaHCO3, 2 mM CaCl2, and 2 mM MgSO4 (pH 7.35, 295–305 mOsm), at 34°C, gassed with 95% O2 and 5% CO2. After recovery for at least 1 h, slices were transferred to room temperature (20–24°C) and were constantly perfused with ACSF. Recordings were made in the ACSF and gassed with 95% O2 and 5% CO2. Recordings were conducted under visual guidance by a microscope with transmitted light illumination. The recording glass microelectrodes (5–8 mΩ) slowly approach the appropriate cell surface and release the negative pressure of the electrode. The cells were tightly combined with the electrode. The slices were perfused with CNO (10 µM). Firing of neurons can be recorded after approximately 1–2 min (Gao et al., 2020).
c-Fos-tTA strategy and molecular profiling
C57BL/6J mice began to be exposed to the Dox diet (200 mg/L in drinking water) 7 d before stereotaxic surgery. Stereotaxic surgeries were carried out by injecting a mixture of rAAV-cFos-tTA-NLS-FLAG (BrainCase) and rAAV-TRE3G-HA-NBL10 (BrainCase) into the PVT for selectively labeling the ribosomes of activated neurons. Three weeks after the surgery, mice injected with the mixed viruses were injected with CFA to establish inflammatory pain. Dox-containing water was replaced with normal drinking water from the third day for 3 d to allow sufficient labeling of CFA sensory pain-related engram neurons in the PVT. After the tagging procedure, Dox was reintroduced to the drinking water. Ten days after CFA injection when their mechanical withdrawal thresholds returned to the baseline level, the mice were then sacrificed for molecular profiling as described previously (Nectow et al., 2017a,b; Zhang et al., 2019), and the PVT was rapidly dissected on ice and removed to the dissection buffer. Brains were divided into three groups of 15 mice per group, immunoprecipitated in the presence of anti-HA magnetic beads (Thermo Fisher). The collected RNA was purified using RNeasy Mini Kit (QIAGEN) and quantified by Quant-iT dsDNA HS Assay Kit, followed by RNA sequencing library construction and sequencing with Illumina.
Statistics
All data were analyzed using GraphPad Prism 7.0 (GraphPad Software) and were presented as mean ± SEM unless otherwise stated. Comparisons between the two groups were analyzed using the two-tailed unpaired Student's t test. Comparisons between multiple groups were analyzed using a one-way or two-way analysis of variance (ANOVA) followed by post hoc Tukey's multiple-comparisons test when appropriate. Statistical significance was defined as p value <0.05.
Results
Hyperactivity of PVT neurons in CDC mice
To examine PVT neuronal response under CDC, an inflammatory pain-related CDC mouse model was established by repeated intraplantar injection of CFA (Fig. 1A) (Zhou et al., 2019). As previously reported, the CFA mice developed a long-lasting, reliable decrease of 50% PWTs (Fig. 1B) (Zhou et al., 2019). Multiple time-point behavioral tests demonstrated that, when compared with the saline control mice, the CFA mice displayed comparable immobile time in TST and FST during the second week post-CFA injection and developed a remarkable increase of immobility from the fourth week which could last at least to the seventh week (Fig. 1C–E). Yet, the CFA mice did not show a significant change in the sucrose preference test (SPT) (Fig. 1C–E). Consistent behavioral phenotypes were also observed in mice subjected to SNI in their sciatic nerve branches (the common peroneal and tibial nerves): SNI induced a persistent decrease in 50% PWTs lasting at least to the 10th week, an increase of immobility in the TST and FST, and a decreased sucrose preference was observed from the sixth week which could be prolonged to the 10th week (Fig. 2A–E). These behavioral data indicate that CFA-induced chronic inflammatory pain and SNI-induced neuropathic pain were sufficient to establish stable CDC behaviors with a workable time window for the following functional investigations.
PVT neurons are activated in CFA mice with comorbid depression. A, Experimental timeline. B, 50% PWTs at different time points after multiple CFA injections (saline, n = 10–11 mice; CFA, n = 10 mice). C–E, Immobile time in the TST and FST and sucrose preference at 2 W (C), 4 W (D), and 7 W (E) after CFA injection (saline, n = 10–11 mice; CFA, n = 10 mice). F, Representative images of c-Fos protein expression at different Bregma levels of PVT in an anterior–posterior manner in CFA mice with comorbid depression. Scale bar, 500 µm. G, Quantitative data of PVT c-Fos protein expression in an anterior–posterior manner (saline, n = 4 mice; CFA, n = 4 mice). Data were analyzed by two-way ANOVA with post hoc Tukey’s multiple-comparisons test between groups (B) or two-tailed unpaired Student's t test (C–E, G). All data are presented as the mean ± SEM *p < 0.05, ***p < 0.001, ns, not significant.
Chronic neuropathic pain and comorbid depression increased the activity of PVT neurons. A, Schematic diagram of SNI surgical and the experimental procedure. B, 50% PWTs at different time points after SNI (Sham, n = 11 mice; SNI, n = 10 mice). C–E, The immobile time in the TST and FST and sucrose preference at 2 W (C), 6 W (D), and 10 W (E) after SNI (Sham, n = 11 mice; SNI, n = 10–14 mice). F, Representative images of c-Fos protein expression at different Bregma levels of PVT in an anterior–posterior manner in SNI mice with comorbid depression. Scale bar, 100 µm. G, Quantitative data of PVT c-Fos protein expression in an anterior–posterior manner (Sham, n = 4 mice; SNI, n = 4 mice). Data were analyzed by two-way ANOVA with post hoc Tukey’s multiple-comparisons test between groups (B) or two-tailed unpaired Student’s t test (C–E, G). All data are presented as the mean ± SEM *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant.
Following immunofluorescent staining for c-Fos protein of the PVT in an anterior–posterior manner in CFA-CDC mice, an immediate c-fos gene-coded protein indicating neuronal activity revealed an increased expression in the PVT, especially the anterior portion of the PVT (PVA), but not in the posterior PVT (PVP) or the middle part (PV) (Fig. 1F,G) when compared with their controls. Similarly, in SNI-CDC mice, an increased expression of c-Fos was observed in the PVA and PV, but not in the PVP (Fig. 2F,G) when compared with their controls. These results consistently demonstrate an enhanced PVA neuronal activity in CDC mice.
Chemogenetic activation of PVA glutamatergic neurons induces pain- and depressive-like behaviors in naive mice
Given the increased c-Fos protein expression in the PVA of CDC mice, we next assessed the causal link between enhanced PVA neuronal activity and the comorbid behaviors. To target PVA glutamatergic neurons, an AAV Cre virus encoding CaMKII (AAV-CaMKIIa-CRE) and a Cre-inducible AAV carrying hM3Dq (AAV-DIO-hM3Dq-mCherry) were mixed to inject into the PVA (Fig. 3A). Immunofluorescent staining confirmed a robust mCherry expression in the PVA, but not in the PV and PVP (Fig. 3B). In vitro electrophysiological recording performed in acutely isolated PVA slice found an elevated firing activity in mCherry-expressing PVA neurons with CNO perfusion (Fig. 3C). The von Frey mechanical allodynia behavioral test revealed that a single dose of intraperitoneal CNO (30 min before the behavioral test) was sufficient to induce a significant decrease in 50% PWTs but did not affect the TST or FST immobile time (Fig. 3D–F). As expected, 7 d of repeated CNO administration (once a day) not only decreased the 50% PWTs but also increased the TST and FST immobile time in mice injected with AAV-DIO-hM3Dq-mCherry when compared with those receiving the control virus (Fig. 3G–I). These results suggest that chemogenetic activation of PVA glutamatergic neurons is sufficient to induce real-time pain-like behavior and that establishing depressive-like behaviors requires a long-term repeated activation of the PVA glutamatergic neurons.
Chemogenetic activation of PVA glutamatergic neurons induces pain- and depressive-like behaviors in naive mice. A, Experimental timeline. B, Schematic illustration for virus injection and representative confocal images for virus expression in an anterior–posterior manner. Scale bar, 500 µm. C, A sample electrophysiological recording trace showing the excitatory effect of CNO on an hM3Dq-mCherry-expressing PVT neuron. D, 50% PWTs 30 min after a single CNO administration (control, n = 8 mice; hM3Dq, n = 8 mice). E, F, Immobile time in the TST (E) and FST (F) 30 min after a single CNO administration (control, n = 8 mice; hM3Dq, n = 8 mice). G, 50% PWTs after a 7 d repeated CNO treatment (control, n = 8 mice; hM3Dq, n = 8 mice). H, I, Immobile time in the TST (H) and FST (I) after a 7 d repeated CNO treatment (control, n = 8 mice; hM3Dq, n = 8 mice). Data were analyzed by two-tailed unpaired Student’s t test. All data are presented as the mean ± SEM *p < 0.05, **p < 0.01, ns, not significant.
Chemogenetic inhibition of PVA glutamatergic neurons attenuates CFA-inflammatory pain but has no analgesic effect in CFA-CDC mice
To determine the necessity of the PVA neuronal hyperactivity in mediating pain and depression comorbidity, we repeated the above chemogenetic virus surgery by replacing AAV-DIO-hM3Dq-mCherry with AAV-DIO-hM4Di-EGFP 3 weeks before the first CFA injection or SNI surgery (Fig. 4A–C,H,I). The von Frey behavioral test performed 1 week after the first CFA injection observed that a single dose of CNO (30 min before the behavioral test) led to an increase of 50% PWTs in CFA mice injected with AAV-DIO-hM4Di-EGFP when compared with those receiving the control virus (Fig. 4D). This effect was not observed in the sham control mice (Fig. 4D). Surprisingly, during the fourth week after repeated CFA injections, acute administration of CNO showed no analgesic effect in CFA-CDC mice with comorbid pain- and depressive-like behaviors (Fig. 4E). Interestingly, chemogenetic inhibition of PVA glutamatergic neurons normalized the TST and FST immobile time in these CFA-CDC mice (Fig. 4F,G). Consistently, behavioral measurements in SNI mice demonstrated similar findings: 2 weeks after SNI surgery, inhibition of PVA glutamatergic neurons was analgesic (Fig. 4J), an effect which was not observed in the same cohort of mice 6 weeks after SNI when they developed pain and depression comorbidity (Fig. 4K); chemogenetic inhibition of PVA glutamatergic neurons also exhibited antidepressant effect in SNI-CDC mice (Fig. 4L,M). These results strongly suggest that chemogenetic inhibition of PVA glutamatergic neurons alleviates pain behaviors in CFA and SNI mice without depressive-like symptoms and only has an antidepressant effect, but not an analgesic effect, in CDC mice with pain and depression comorbidity.
Chemogenetic inhibition of PVA glutamatergic neurons has no analgesic effect in CFA or SNI mice with comorbid depression. A, Experimental timeline. B, Schematic illustration for virus injection and representative confocal images for virus expression in an anterior–posterior manner. Scale bar, 500 µm. C, A sample electrophysiological recording trace showing the inhibitory effect of CNO on an hM4Di-EGFP-expressing PVT neuron. D, 50% PWTs upon chemogenetic inhibition of PVT glutamatergic neurons in CFA mice 1 week after CFA injection (Sham-EGFP, n = 8 mice; Sham-hM4Di, n = 8 mice; CFA-EGFP, n = 9 mice; CFA-hM4Di, n = 8 mice). E–G, 50% PWTs (E), immobile time in the TST (F) and FST (G) upon chemogenetic inhibition of PVT glutamatergic neurons in CFA mice with comorbid depression 4 weeks after CFA injection. Behavioral tests were performed 30 min after CNO administration (Sham-EGFP, n = 8 mice; Sham-hM4Di, n = 8 mice; CFA-EGFP, n = 8 mice; CFA-hM4Di, n = 8 mice). H, Schematic diagram of the experimental procedures of chemogenetic inhibition of PVT glutamatergic neurons. I, Schematic illustration of virus injection and a representative confocal image showing virus expression. Scale bar, 100 µm. J, 50% PWTs of mice with pain 30 min after chemogenetic inhibition (Sham-EGFP, n = 6 mice; Sham-hM4Di, n = 6 mice; SNI-EGFP, n = 6 mice; SNI-hM4Di, n = 7 mice). K–M, 50% PWT (K), immobile time in the TST (L) and FST (M) of mice with pain and comorbid depression 30 min after chemogenetic inhibition (Sham-EGFP, n = 6 mice; Sham-hM4Di, n = 5 mice; SNI-EGFP, n = 6 mice; SNI-hM4Di, n = 7 mice). Data were analyzed by one-way ANOVA with post hoc Tukey’s multiple-comparisons test between groups. All data are presented as the mean ± SEM *p < 0.05, ***p < 0.001, ns, not significant.
Chemogenetic inhibition of PVA glutamatergic neurons is analgesic in S-ketamine-treated CDC mice
Next, we investigate the potential impact of depressive-like behaviors on the analgesic effect by inhibiting PVA glutamatergic neurons. We treated CFA-CDC mice with a sub-anesthetic dose of S-ketamine, a noncompetitive NMDA receptor antagonist with a rapid-onset antidepressant effect (Berman et al., 2000; Zarate et al., 2006; Li et al., 2010). To do this, an intraperitoneal injection of S-ketamine was integrated into the chemogenetic inhibition experiment (Fig. 5A,B). As reported above, from the fourth week, mice subjected to repeated CFA injection developed a comorbid state with pain- and depressive-like symptoms (Fig. 5C,D). Behavioral tests performed 12 h after S-ketamine injection indicated that S-ketamine did not affect the 50% PWTs in both saline control and CFA-CDC mice (Fig. 5E) but reversed the increased immobile time in the TST and FST of CFA-CDC mice (Fig. 5F,G). Interestingly, chemogenetic inhibition of PVT glutamatergic neurons robustly increased the 50% PWTs in S-ketamine-treated CFA-CDC mice (Fig. 5H). These findings support the notion that long-term chronic pain symptoms under the comorbid state were composed of sensory and affective components (Huang et al., 2020). Together with the data in Figures 3 and 4, these results led us to hypothesize that inhibition of PVA glutamatergic neurons only had an analgesic effect on sensory pain, while S-ketamine could only alleviate affective pain.
PVA glutamatergic neuron inhibition combined with S-ketamine alleviates pain-like behavior in CFA mice with comorbid depression. A, Experimental timeline. B, Schematic illustration for virus injection and representative confocal images for virus expression in an anterior–posterior manner. Scale bar, 250 µm. C, 50% PWTs in CFA mice with comorbid depression (saline, n = 12 mice; CFA, n = 13 mice). D, Immobile time in the TST (left) and FST (right) in CFA mice with comorbid depression (saline, n = 12 mice; CFA, n = 13 mice). E–G, 50% PWTs (E), immobile time in the TST (F) and FST (G) in CFA mice with comorbid depression 12 h after intraperitoneal injection of S-ketamine (saline–saline, n = 6 mice; saline–S-ketamine, n = 6 mice; CFA–saline, n = 6 mice; CFA–S-ketamine, n = 7 mice). H, 50% PWTs upon chemogenetic inhibition of PVT glutamatergic neurons in S-ketamine-treated CFA mice with depression (saline–saline, n = 6 mice; saline–S-ketamine, n = 6 mice; CFA–saline, n = 6 mice; CFA–S-ketamine, n = 7 mice). Data were analyzed by two-tailed unpaired Student’s t test (C, D) or one-way ANOVA with post hoc Tukey’s multiple-comparisons test between groups (E–H). All data are presented as the mean ± SEM *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant.
PVA glutamatergic neuron inhibition and S-ketamine distinctly regulate sensory pain and affective pain
To test the hypothesis above, we established a 21 day chronic restraint stress (CRS) model of depression, which could reliably induce affective pain-like behavior, showing as a decrease in 50% PWTs in the von Frey test (Zhu et al., 2021). Our behavioral tests demonstrated that S-ketamine pretreatment but not real-time chemogenetic inhibition of the PVA glutamatergic neurons reversed the reduction of 50% PWTs and normalized the increase of immobile time in the FST in CRS mice (Fig. 6A,B). In contrast, S-ketamine treatment 12 h before behavioral tests failed to reverse the reduction of 50% PWTs in both CFA and SNI mice only with sensory pain symptoms (Fig. 6C,D). Combined with the results in Figure 5, these data demonstrate that S-ketamine can alleviate depressive-like behaviors and depression-related affective pain in CDC mice. These data also further support the sensory pain-specific regulatory role of the PVA glutamatergic neurons.
PVA glutamatergic neuron inhibition and S-ketamine distinctly regulate sensory pain and affective pain. A, Experimental timeline and 50% PWTs and immobile time in the FST of CRS mice with affective pain 12 h after intraperitoneal injection of S-ketamine (control–saline, n = 8 mice; control–S-ketamine, n = 8 mice; CRS–saline, n = 7 mice; CRS–S-ketamine, n = 7 mice). B, Experimental timeline and 50% PWTs and immobile time in the FST of CRS mice 30 min after CNO injection for chemogenetic inhibition of PVT glutamatergic neurons (control–saline, n = 8 mice; control–CNO, n = 8 mice; CRS–saline, n = 8 mice; CRS–CNO, n = 8 mice). C, Experimental timeline and 50% PWTs in CFA-inflammatory pain mice 12 h after intraperitoneal injection of S-ketamine (saline–saline, n = 5 mice; saline–S-ketamine, n = 6 mice; CFA–saline, n = 7 mice; CFA–S-ketamine, n = 7 mice). D, Experimental timeline and 50% PWTs of mice with sensory neuropathic pain induced by SNI 12 h after intraperitoneal injection of S-ketamine (Sham–saline, n = 6 mice; Sham–S-ketamine, n = 6 mice; SNI–saline, n = 6 mice; SNI–S-ketamine, n = 6 mice). Data were analyzed by one-way ANOVA with post hoc Tukey’s multiple-comparisons test between groups. All data are presented as the mean ± SEM *p < 0.05, ns, not significant.
Neuronal activity-dependent identification of Htr1d for CDC modulation
Cell-activity-related molecular profiling can be conducted in specific brain regions, unveiling more specific molecular targets for developing novel target-based medications. To do this, we first investigate whether reactivate pain-activated PVT neurons could induce depressive-like behaviors. We employed cell-activity-related labeling and reactivation of the pain-responding PVA neurons by injecting rAAV-cFos-tTA-NLS-FLAG and AAV-TRE3G-hM3Dq-mCherry into the PVA of C57BL/6J mice. PVA neurons that respond to CFA-inflammatory pain will be labeled when Dox-containing drinking water is withdrawn. We observed a larger number of PVA neurons captured through activity tagging by CFA-induced pain compared with the saline control (Fig. 7A,B). When the CFA mice recovered from the inflammatory pain (Fig. 7C), we chemogenetically reactivated these labeled PVA pain engram neurons and found that a single chemogenetic reactivation of PVA pain-associated neurons was sufficient to induce mechanical allodynia but did not elicit depressive-like behaviors in mice that previously experienced CFA-inflammatory pain (Fig. 7D–F). As expected, 7 d of repeated activation of these neurons was sufficient to provoke mechanical allodynia and depression-like behaviors in these mice (Fig. 7G–I). These results indicated that reactivation of PVA CFA-inflammatory pain engram neurons can induce pain- and depressive-like behaviors.
Reactivation of PVT CFA-inflammatory pain engram neurons induces pain- and depressive-like behaviors in mice that previously experienced CFA-inflammatory pain. A, Schematic illustration for virus injections and representative confocal images showing the expression of virus. Scale bar, 100 µm. B, C, 50% PWTs of mice on day 3 and day 10 after CFA injection (saline, n = 8 mice; CFA, n = 11 mice). D–F, 50% PWTs, TST and FST immobile time upon a single chemogenetic activation of PVT CFA-inflammatory pain engram neurons (saline, n = 7 mice; CFA, n = 11 mice). G–I, 50% PWTs, TST and FST immobile time after repeated chemogenetic activation of PVT CFA-inflammatory pain engram neurons (saline, n = 7 mice; CFA, n = 11 mice). Data were analyzed by two-tailed unpaired Student’s t test. All data are presented as the mean ± SEM *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant.
Next, to molecularly profile PVA pain-related neurons, a mixture of rAAV-c-Fos-tTA-NLS-FLAG and rAAV-TRE3G-HA-NBL10 was injected into the PVA of C57BL/6J mice, which enabled the labeling of inflammatory pain-activated neurons upon Dox withdrawal and the following profiling procedures (Fig. 8A,B). On the 10th day after CFA injection, animals recovered from the CFA-inflammatory pain and were sacrificed for molecular profiling (Fig. 7C, 8A). RNA sequencing (RNA-seq) was employed to analyze the ribosome immunoprecipitation (IP) and total/input RNA (control) samples to identify transcripts expressed explicitly in CFA pain engram neurons. The fold enrichment for each RNA was calculated by dividing the IP RNA abundance by the total/input (control) RNA abundance. By comparison of IP/total RNA (control), this experiment revealed enriched transcripts with high translation levels. Remarkably, we observed statistically significant enrichments (q < 0.05) of more than 1.5-fold for 879 transcripts upregulated across various functional subfamilies, including those encoding neuroactive ligand–receptor interactions (Fig. 8C,D). Among the enriched transcripts, Htr1d (encoding 5-hydroxytryptamine receptor 1D) and Galr1 (encoding galanin receptor 1) were selected as potential candidates for further functional investigations (Fig. 8E).
Identification of Galr1 and Htr1d in CFA-inflammatory pain-labeled PVT engram neurons as potential receptor targets for regulating pain and depression comorbidity. A, Schematic diagram of the experimental procedures. B, Schematic illustration for virus injection and a representative confocal image for virus expression. Scale bar, 100 µm. C, RNA-seq scatterplots for CFA-inflammatory pain engram neurons in the PVT with normalized expression. Significantly different genes (q < 0.05) are displayed in red (enriched) and green (reduced), and all other genes are displayed in gray. Black lines represent unity and 1.5-fold change in enrichment. D, KEGG enrichment analysis showing top 10 pathways. E, FPKM of Galr1 and Htr1d (input, n = 15 mice/repeat, 3 repeats in total; IP, n = 15 mice/repeat, 3 repeats in total).
Subsequently, we investigated the effects of pharmacological agonists and antagonists on pain and depression-like behaviors by local infusion into the PVA (Fig. 9A,B). Behavioral results indicated that a single infusion of galanin receptor 1 agonist M617 [0.15 nM, 1 µl (Fu et al., 2016)] (Fig. 9C–E) or 5-HT1D receptor agonist GR-46611 [0.6 nM, 1 µl (Da Silva et al., 2007)] (Fig. 9F–H) in PVA both significantly decreased the 50% PWTs in naive mice, but did not affect the behavioral performance in the TST and FST. Seven days of repeated infusion experiments (once a day) revealed that multiple galanin receptor 1 agonist M617 infusions failed to induce depressive-like behaviors (Fig. 9I–K), while in contrast, the 5-HT1D receptor agonist GR-46611 induced a remarkable decrease of 50% PWTs in the von Frey test and prolonged immobile time in both TST and FST in naive control mice (Fig. 9L–N).
Effect of pharmacological modulation of galanin receptor 1 (Galr1) and 5-hydroxytryptamine receptor 1D (5-HT1D) on pain- and depressive-like behaviors. A, Schematic diagram of the experimental procedures of pharmacological administration. B, Schematic illustration for cannula placement and a representative confocal image for cannula placement. Scale bar, 1 mm. C–E, 50% PWTs (C), immobile time in the TST (D) and FST (E) of naive mice following a single dose of intra-PVA infusion of galanin receptor 1 agonist M617 (vehicle, n = 6–8 mice; M617, n = 6–8 mice). F–H, 50% PWTs (F), immobile time in the TST (G) and FST (H) of naive mice following a single dose of intra-PVA infusion of 5-HT1D receptor agonist GR-46611 (vehicle, n = 6 mice; GR-46611, n = 6 mice). I–K, 50% PWTs (I), immobile time in the TST (J) and FST (K) of naive mice following a 7 d repeated intra-PVA infusion of galanin receptor 1 agonist M617 (vehicle, n = 6–8 mice; M617, n = 6–8 mice). L–N, 50% PWTs (L), immobile time in the TST (M) and FST (N) in naive mice after a 7 d repeated intra-PVA injection of 5-HT1D receptor agonist GR-46611 (vehicle, n = 6 mice; GR-46611, n = 6 mice). O–Q, 50% PWTs (O), immobile time in the TST (P) and FST (Q) of pain and comorbid depressive mice following a single dose of intra-PVA infusion of galanin receptor 1 antagonist M35 (saline–vehicle, n = 6 mice; saline–M35, n = 6 mice; CFA–vehicle, n = 6 mice; CFA–M35, n = 6 mice). R–T, 50% PWTs (R), immobile time in the TST (S) and FST (T) of pain and comorbid depressive mice following a 7 d repeated intra-PVA infusion of galanin receptor 1 antagonist M35 (saline–vehicle, n = 6 mice; saline–M35, n = 6 mice; CFA–vehicle, n = 6 mice; CFA–M35, n = 6 mice). U–W, 50% PWTs (U), immobile time in the TST (V) and FST (W) of CFA mice with comorbid depression following a single dose of intra-PVA infusion of 5-HT1D receptor antagonist BRL-15572 (saline–vehicle, n = 6 mice; saline–BRL-15572, n = 6 mice; CFA–vehicle, n = 6 mice; CFA–BRL-15572, n = 6 mice). X–Z, 50% PWTs (X), immobile time in the TST (Y) and FST (Z) in CFA mice with comorbid depression after a 7 d repeated intra-PVA injection of 5-HT1D receptor antagonist BRL-15572 (saline–vehicle, n = 6 mice; saline–BRL-15572, n = 6 mice; CFA–vehicle, n = 6 mice; CFA–BRL-15572, n = 6 mice). Data were analyzed by two-tailed unpaired Student’s t test (C–N) or by one-way ANOVA with post hoc Tukey’s multiple-comparisons test between groups (O–Z). All data are presented as the mean ± SEM *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant.
Then, we examined how pharmacologically antagonizing both receptors affects the pain- and depressive-like behaviors in CDC mice. Behavioral tests revealed that either a single dose or multiple infusions of galanin receptor 1 antagonist M35 [0.15 nM, 1 µl (Duan et al., 2015)] did not affect the animal's performance in the von Frey test, TST, and FST (Fig. 9O–T). Acutely single dose of 5-HT1D receptor antagonist BRL-15572 [0.15 nM, 1 µl (Vidal-Cantú et al., 2016)] did not affect the animal's performance in the von Frey test and TST (Fig. 9U,V) but reversed the increased immobility in the FST (Fig. 9W). Upon repeated administration, 5-HT1D receptor antagonist BRL-15572 reversed the reduction of 50% PWTs and normalized the increase of immobile time in both the TST and FST (Fig. 9X–Z). The molecular profiling analysis and pharmacological results presented above suggest that the 5-HT1D receptor in the PVT neurons may serve as a molecular target for developing novel drugs for pain and depression comorbidity.
Discussion
Unlike acute pain, which carries danger-warning function and survival values, chronic pain is usually a symptom or disease involving complicated, multidimensional, and dynamic interactions between sensory and affective modalities (Grichnik and Ferrante, 1991). Tissue damage or inflammation-related primary sensory pain induces negative emotions, such as depression, which will reciprocally deteriorate with pain sensation to develop chronic pain and mental disorder comorbidity, thus complicating its clinical treatment (Cohen et al., 2021; Patel, 2023; Yang et al., 2023). Animal studies also found that long-lasting neuropathic and inflammatory pain induce CDC (Zhou et al., 2019; Zhu et al., 2021). In the present study, we further demonstrated that CFA and SNI mice experienced two different states: early state only with sensory pain and late state with CDC (Fig. 10), which could last for weeks to provide a workable time window for the following functional investigations.
PVA glutamatergic neurons are activated in mice with sensory pain or with pain and depression comorbidity. Comorbid pain symptoms in CDC mice are composed of sensory and affective components that could be independently relieved by PVT inhibition and S-ketamine, thus providing important guiding information for future treatment of chronic pain and depression comorbidity. When treating chronic pain in the future, both sensory and affective factors should be taken into consideration as these may affect treatment outcomes.
The PVT has been implicated in mediating pain sensation and stress-related negative emotions in our and others' recent animal studies (Hsu et al., 2014; Cheng and Chen, 2018; Barson et al., 2020; Kooiker et al., 2021; Deng et al., 2023; Zhang et al., 2023). In the present study, we further demonstrated the essential role of PVA in mediating CDC based on its neuronal response to a comorbid state. Single chemogenetic activation of PVA glutamatergic neurons in wild-type control mice induced pain-like behaviors, while the development of depressive-like symptoms required repeated neuronal activation. This is consistent with the naturally developing process of CDC. Interestingly, inhibition of these neurons alleviated pain symptoms in CFA and SNI mice with sensory pain but not in CRS mice with affective pain. And acute inhibition of PVA glutamatergic neurons or sub-anesthetic S-ketamine treatment could reverse depressive-like behaviors without affecting the comorbid pain behaviors in CDC mice. Interestingly and surprisingly, in CDC mice receiving pretreatment with S-ketamine, chemogenetic inhibition of PVA glutamatergic neurons re-gained the analgesic effect. This evidence highly suggested that the comorbid pain symptoms in CDC mice were possibly composed of sensory and affective components and that PVA inhibition only had an analgesic effect on sensory pain but not on affective pain. Indeed, our following pharmacological experiments confirmed that S-ketamine only displayed analgesic effect in CRS mice with affective pain but not in CFA or SNI mice only with sensory pain. These data suggested that pain symptoms in CDC mice contained sensory and affective components that could be independently relieved by PVA inhibition and S-ketamine (Fig. 10). In this case, in CDC mice, the analgesic effect of PVA neuronal inhibition was concealed by the affective component of pain symptom. When the affective component of pain was alleviated by ketamine treatment, the effect of PVA neuronal inhibition was observed. Thus, these findings provided crucial guiding information for future treatment of CDC: both sensory and affective factors should be taken into consideration when treating chronic pain in the future, which may affect treatment outcomes.
Defining the neuronal adaptations underlying pathological processes or treatment effects with the armamentarium of modern neuroscience tools is one of the best ways to pursue the mechanisms of action for novel therapeutics (Monteggia et al., 2014). Once critical cellular candidates are identified, molecular profiling could be performed in specific cell types, yielding novel molecular targets for developing novel drugs (Ekstrand et al., 2014; Monteggia et al., 2014; Nectow et al., 2017a,b; Heifets and Malenka, 2019; Zhang et al., 2019; Zhang et al., 2023). In this study, we combined the neuronal activity labeling approach based on the expression of immediate early gene c-fos to profile the molecular expression characterization for screening possible druggable targets. Our profiling data showed a relatively enriched Htr1d mRNA expression, which encoded 5-hydroxytryptamine receptor 1D (5-HT1D receptor) in CFA-inflammatory pain-related engram neurons. There is little evidence to suggest that the 5-HT1D receptor had a potential role in regulating depression- and pain-like behaviors. For example, a postmortem study of untreated suicidal victims with a confirmed history of depression showed a much higher density of 5-HT1D receptors in the globus pallidus (Lowther et al., 1997). Consistently, a clinical trial indicated that the sensitivity of postsynaptic 5-HT1D receptors in patients after treatment with SSRIs has been found to be impaired (Whale et al., 2001). In addition, the 5-HT1D receptor was also reported to be involved in pain regulation at the peripheral and spinal levels (Wu et al., 2001; Potrebic et al., 2003; Liu et al., 2005; Classey et al., 2010; Cervantes-Durán et al., 2013; Avila-Rojas et al., 2015). However, how the 5-HT1D receptor in the brain regulates chronic pain and its comorbid depression remained not well known. Based on the profiling data, our pharmacological experiments further found that PVA local infusion of its agonist could induce pain-like (either single or repeated infusions) and depressive-like behaviors (repeated infusions) in wild-type mice; a single infusion of its antagonist showed an analgesic effect in CFA pain mice but not in CFA-CDC mice, and repeated infusions of its antagonist was able to alleviate both the pain- and depression-like behaviors.
Interestingly, chemogenetic or pharmacological activation of PVT neurons takes less time to induce depressive-like behaviors than repeated CFA injections in the FST and TST. However, the underlying mechanisms remain unknown, possibly because that chemogenetically or pharmacologically activated PVT neurons represent larger subpopulations than that directly involving in the development of CDC. Future more specific studies are needed to address this interesting and crucial scientific question. Moreover, we found that chemogenetic inhibition of PVA glutamatergic neurons did not affect 50% PWTs of control mice, while intra-PVA infusion of 5-HT1D receptor antagonist increased 50% PWTs in saline control mice. This divergence in behavioral results was possibly due to the difference of affected PVT neuronal populations by the chemogenetic and pharmacological strategies.
The PVT plays a critical role in regulating many behavioral outcomes with a robust functional heterogeneity due to distinct sub-cell types and downstream projections (Gao et al., 2023). For instance, while early studies support that pharmacological inactivation of the PVT increases food intake (Stratford and Wirtshafter, 2013), others show that activation of the PVT increases food-seeking behaviors (Labouèbe et al., 2016; Meffre et al., 2019; Sofia Beas et al., 2020). In the behavioral tests of Figures 4 and 5, all of the PVT CamKII-Cre-expressing neurons were affected and regulated by intraperitoneal administration of CNO. The behavioral outcome observed was due to a universal inhibition of PVT glutamatergic sub-populations with distinct functions. We recently demonstrated that inhibition of PVT → NAc projection was sufficient to increase the 50% PWTs in naive mice, a nonopioid receptor-dependent analgesic effect. We also profiled the molecular signatures of these projecting glutamatergic neurons to find a relatively enriched expression of 5-HT1D receptors as a potential druggable target for nonopioid analgesia (Zhang et al., 2023). Our data in Figure 9 further proved that pharmacologically antagonizing 5-HT1D receptors in the PVT was sufficient to increase the 50% PWTs in naive control mice, possibly by inhibiting more NAc-projecting PVT glutamatergic neurons (Zhang et al., 2023). These findings supported that 5-hydroxytryptamine receptor 1D in the PVA is a molecular target critical for the development and maintenance of CDC, holding a possibility for developing future medications.
In conclusion, our study established an essential role of the PVA glutamatergic neurons in mediating CDC, providing a deeper understanding of the complexity of comorbid pain and a molecular target with translational potential.
Availability of data and materials
All data are available in the main text or the supplementary materials. Further access to the original data was available by contacting the corresponding author at hongxing.zhang{at}xzhmu.edu.cn. Supplementary information is available on the journal's website.
Footnotes
This work was supported by the STI2030-Major Projects (2021ZD0203100), National Natural Science Foundation of China (31970937, 82271255, 82101315, 82301413, 82171227 and 82204081), Jiangsu Province Innovative and Entrepreneurial Team Program, Jiangsu Province Key R&D Program Social Development Project (BE2023690), Science and Technology Department of Jiangsu Province (BK20220665), the Natural Science Foundation of Zhejiang Province (LY22H090019), the Priority Academic Program Development of Jiangsu Higher Education Institutions (21KJB320001), China Postdoctoral Science Foundation (2022M722676, 2023M732973, and 2022M722675), Xuzhou Medical University start-up grant for excellent scientist (D2020053, D2020033, and D2022005), and the Postgraduate Research and Practice Innovation Program of Jiangsu Province (KYCX22_2932 and KYCX23_2952).
↵* M.C., R.J., L.S., and X.W. contributed equally to this work.
All authors declare no competing financial interests. H.Z., M.C., R.J., and J.-L.C. are named on a patent pending for 5-hydroxytryptamine receptor 1D as a molecular target for treating chronic pain and depression comorbidity.
- Correspondence should be addressed to He Liu at lh121061{at}163.com, Jun-Li Cao at caojl0310{at}aliyun.com, or Hongxing Zhang at hongxing.zhang{at}xzhmu.edu.cn.
