2.1. Experimental Animals

Adult male C57BL/6J mice (aged 8–12 weeks, weighing 20–25 g) were obtained from Charles River Laboratories (License No. SCXK (Jing) 2021-0006, Beijing, China). The mice were housed in an environment where the temperature was maintained at 25^°C ± 2^°C. They were provided with a 12-h light/dark cycle (lights on at 7:00 a.m. and off at 7:00 p.m.) and had free access to food and water. Upon arrival, the animals were acclimated to the new environment for 1 week before the start of the experiments. Mice in the same cage were randomly assigned to different experimental groups. All experimental procedures were approved by the Animal Care and Use Committee of Tianjin Medical University (approval date: 30 March 2020).

2.2. Systemic Animal Model of CSDS

Depression-related stresses are inherently social, involving environmental stresses, life events, and interpersonal setbacks [25]. Therefore, the CSDS model in mice is a superior model for studying depression, as it more accurately replicates the depression process caused by social factors, leading to long-lasting behavioral and biological changes [26]. In our study, individual C57BL/6J mice were subjected to the CSDS protocol for 14 consecutive days. Each day, mice experienced 5–10 min of social confrontation with CD-1 mice, referred to as direct contact stress. After the confrontation, the C57BL/6J mice were placed on the opposite side of an isolation baffle in the experimental cage for 24 h, representing indirect psychological stress. This procedure was repeated daily for 14 days, with each subject C57BL/6J mouse exposed to a different CD-1 mouse each day [27].

2.3. Behavioral Tests

Open field testing (OFT) is a method used to assess the locomotor abilities and anxiety levels of mice [28]. An open field box of 40 cm in length, width, and height was utilized, with a camera fixed directly above to capture the entire bottom of the box. Mice were placed in the center of the box, and their movements were recorded for 6 min. The first minute allowed mice to acclimate to their surroundings, while the remaining 5 min was used to evaluate their locomotor activity and anxiety levels. Time spent in the central region of the open field was analyzed to gauge anxiety. Before each experiment, the open field box was wiped and dried with 75% alcohol to eliminate olfactory cues from previous mice. After wiping, we will use a hair dryer to dry the test box for 3–5 min to ensure that the test box will not be affected by a smell.

Tail suspension test (TST) was used to assess behavioral despair in mice. Medical tape was applied 1 cm from the tip of the tail, and mice were suspended from a device suspension bar, with the tip of their tails approximately 50 cm above the ground, placing them in a head-down position. The mice were recorded for 6 min with their ventral sides facing the camera in a quiet environment. The first minute of data was discarded, and the immobility time during the final 5 min was recorded.

Sucrose preference test (SPT) is used to evaluate anhedonia, a key symptom of depression characterized by the inability to experience pleasure. Mice were individually housed in cages with two bottles of 1% sucrose solution for adaptive training over 24 h. The SPT was conducted at 9:00 a.m. for 1 h following 24 h of food and water deprivation. Each mouse had access to two bottles, one containing 200 mL of water and the other 200 mL of 1% sucrose solution. The volumes of sucrose solution (V1) and water (V2) consumed were recorded. Sucrose preference was calculated as V1/(V1 + V2) × 100%.

For the social interaction test (SIT), C57BL/6 mice were placed in an open field with a net cage (6 × 8 cm) on one side. In the first stage, C57BL/6 mice were placed in the open field for 5 min without a target in the cage; in the second stage, a CD-1 mouse was placed in the cage, and C57BL/6 mice were returned to the open field for another 5 min. Time spent in the social interaction area (16 × 12 cm) and the avoidance area (8 × 8 cm) was recorded using behavioral analysis software. The social interaction ratio (SIR) was calculated as the time spent in the interaction zone during the second phase divided by the time spent in the interaction zone during the first phase. Mice with SIR < 1 were considered susceptible to social avoidance, while mice with SIR > 1 were considered resistant, showing social preference.

2.4. Stereotaxic Optical Fiber Implantation and Virus Injection

The surgical procedure was conducted following previously described methods [29]. Mice were anaesthetised with isoflurane and placed in a stereotaxic frame (RWD Life Science, Shenzhen, China) with a heating pad to maintain body temperature. The mice were maintained under anaesthesia with 1.5% vol isoflurane/vol oxygen (O₂) throughout the procedure. Ophthalmic ointment was applied to prevent eye drying. Once the scalp fur had been trimmed, a small craniotomy (0.5 × 0.5 mm) was created using a dental drill. The stereotactic injection brain region coordinates are as follows: DRN: anterior posterior (AP): −4.36 mm, medial lateral (ML): 0.00 mm, and dorsal ventral (DV): −3.00 mm; mPFC: AP: +2.23 mm, ML: 0.15 mm, and DV: −1.95 mm. For Ca²⁺ signal recording, 200–300 nL of AAV2/5-GfaABC1D-GCaMP7j solution (approximately 2 × 10¹³ particles/mL) was injected into the tissue at a depth of 2.95 mm using a glass micropipette. The following viruses were used for experimentation: rAAV-GfaABC1D-hM3D (Gq)-P2A-mCherry, rAAV-hSyn-GCaMP6f-WPRE-hGH PA, rAAV-CaMKIIa-GCaMp6f-WPRE-hGH, rAAV-hSyn-CRE-WPRE-hGH, and rAAV-EF1a-DIO-EYFP-WPRE-hGH. Viruses were purchased from Brain Case Biotechnology Co Ltd, OBiO Technology (Shanghai) Corp Ltd. After each injection, the micropipette was held in place for 5 min to allow for diffusion. Upon removal of the micropipette, an optic fiber (1.25-mm outer diameter (OD), 0.37 numerical aperture (NA); Shanghai Fiblaser), threaded through a 3.5-mm ceramic ferrule, was implanted 0.1 mm above the injection site and secured to the skull using dental cement. Recordings were performed at least 3 weeks postsurgery to allow robust virus expression. To confirm the position of GCaMP6f expression, all mice were transcardially perfused with 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS) following the behavioral tests. After overnight fixation in 4% PFA, brain samples were sectioned coronally and imaged using a fluorescence microscope (BX51, Olympus).

2.5. Patch Clamp Recording

Artificial cerebrospinal fluid (ACSF) was frozen at −20°C until it reached a sand-like ice consistency [30]. The ACSF composition was as follows (in millimolar): 120 sodium chloride (NaCl), 1.25 sodium dihydrogen phosphate dihydrate (NaH₂PO₄-2H₂O), 10 glucose hydrate (C₆H₁₂O₆-H₂O), 26 sodium bicarbonate (NaHCO₃), 2 magnesium sulfate heptahydrate (MgSO₄-7H₂O), 2.5 potassium chloride (KCl), and 2 calcium chloride dihydrate (CaCl₂-2H₂O), with a pH between 7.4 and 7.6 and an osmolarity between 290 and 295 mOsm. Additionally, 200 mL of ACSF was prepared and prewarmed to 32°C. The blade was precooled in the frozen ACSF. Mice were deeply anaesthetised with isoflurane (3%–5%), the brain tissues were quickly removed, and their brain tissues were rapidly transferred into the precooled ACSF for 1 min. Brain slices (300 μm) containing the mPFC were secured with glue in the lower slot of a vibrating slicer (Vibratome, Leica VT1200S, Germany). The slicer was then filled with frozen ACSF, and its settings were adjusted (slicing speed: 0.07 μm/s, amplitude: 2.00, and slice thickness: 300 μm) before the brain tissue was cut coronally. Brain slices were incubated in ACSF at 32°C for 30 min, followed by 30 min at room temperature. The slices were then transferred to the recording chamber and allowed to equilibrate for at least 10 min before recording. Whole-cell patch clamp recordings of mPFC pyramidal cells were performed using a 700B amplifier (Molecular Devices, United States) and glass microelectrodes (3–5 MΩ). The brain slice was perfused continuously at a rate of 2 mL/min using a perfusion pump while being aerated with 95% O₂/5% carbon dioxide (CO₂). To record spontaneous excitatory postsynaptic currents (sEPSCs), the intracellular solution consisted of the following components (in millimolar): 130 K-gluconate, 10 KCl, 10 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 5 magnesium ion (Mg²⁺) ATP, 0.25 ethylenebis(oxyethylenenitrilo)tetraacetate (EGTA), pH was adjusted to 7.2 with potassium hydroxide (KOH), and osmolarity was adjusted to 280–290 mOsm. sEPSCs were recorded at holding membrane potentials of −70 mV. Fifty micromolar Picrotoxin was added in the ACSF perfusate. To record paired pulse ratio (PPR), place a concentric circular stimulation electrode in the fifth layer of mPFC and search for cells with a thickness of about 50 μm in the same cell layer for whole-cell recording. Voltage clamp is at −70 mV, electrical stimulation delay 500 ms, stimulation time 0.1 ms, interval 50 ms, restimulation 0.1 ms, recording time 3 s, sweep interval 27 s, and recording 6 sweeps. The stimulation intensity is based on recording to around 100 pA of excitatory postsynaptic current (EPSC). PPR calculation is as follows: EPSC2/EPSC1.

2.6. Calcium Analysis

Start from the time when mice start moving, socializing, and struggling in behavioral tests. The relative fluorescence changes are expressed using ΔF/F, excluding interference from the viral injection dose and basal fluorescence intensity [29]. ΔF/F is defined as the ratio of the fluorescence value (F) to the fluorescence baseline (F baseline), where the F baseline represents the average fluorescence level within 10 s before the recorded target behavior (such as action exploration, social interaction, and struggle) begins. The MultiPhotometry.mlapp program in MATLAB was used to perform signal preprocessing (including baseline correction and behavioral signaling extraction) and to compute ΔF/F. The calcium dynamics of the DRN were evaluated using the ΔF/F metrics of amplitude, duration, and area under the curve (AUC).

2.7. Statistical Methods

The data were expressed as means ± standard deviation (SD). Statistical analysis was conducted using GraphPad Prism software, while Adobe Illustrator 2020 facilitated chart creation. Clampfit, MATLAB (2018a), and MiniAnalysis were utilized to evaluate the frequencies and amplitudes of sEPSCs and PPR. Initially, all data were tested with the Kolmogorov–Smirnov test for normality and confirmed to be normally distributed unless otherwise noted. Paired t-tests were used to analyze the amplitude and the AUC of DRN astrocyte calcium signaling. One-way analysis of variance (ANOVA) was employed to compare data among different groups. Statistically significant differences were established at ∗p < 0.05, ∗∗p < 0.01 and ∗∗∗p < 0.001, two-sided.

3.1. CSDS Decreased DRN Astrocyte Calcium Signaling

In this study, we investigated the effect of CSDS depression modeling on the Ca²⁺ levels of astrocytes in the DRN of mice using a multichannel fiber-optic photometer system in combination with GCAMP7J-labelled astrocytes. T-tests were used to analyze the changes in the area under the relative fluorescence curve (ΔF/F) of Ca²⁺ in the two groups.

We first injected the GCaMP virus AAV2/5-GfaABC1D-GCaMP7j [31], which specifically labels astrocytes, into the DRN and then implanted ceramic inserts. After 3 weeks, stable expression was observed in the corresponding brain regions of the mice. Multichannel fiber-optic photometric detection technology was used to record the calcium signaling of astrocytes in the DRN of mice under normal conditions during OFT, SIT, and TST as the wild-type (WT) group. CSDS modeling was performed for 2 weeks after initial recording, and calcium signaling recordings were repeated postmodeling (Figure 1(a)). After the experiment, the injection site of the virus was confirmed using a fluorescence microscope (Figure 1(b)). The experimental results showed that the peak values of ΔF/F of calcium signaling changes in astrocytes in the DRN during each walking exploration in the OFT under normal conditions were higher than those in mice after CSDS depression modeling. This indicates a decrease calcium signaling in astrocytes in the DRN after depression modeling (Figures 1(c) and 1(d)). Calculation of the area under the ΔF/F curve for calcium signaling revealed a concomitant decrease in the AUC, suggesting that astrocyte activity was reduced in the depressed state (Figure 1(e)). Following CSDS depression modeling, the ΔF/F calcium signaling of astrocytes in the DRN showed a decrease in amplitude during the SIT, with the same rate of rise of calcium signaling but a decrease in rise time (Figures 1(f) and 1(g)). However, the AUC of astrocyte calcium signaling did not show statistical difference (Figure 1(h)). We measured the level of despair in mice using TST and observed changes in calcium signaling in astrocytes during the TST. The ΔF/F calcium signaling of astrocytes in the DRN decreased in amplitude after CSDS depression modeling, and both the rate and rise time of the calcium signaling decreased (Figures 1(i) and 1(j)). This condition resulted in a decrease in the AUC of astrocyte calcium signaling in this brain region (Figure 1(k)).

3.2. Chemogenetic Activation of DRN Astrocyte Activity Affected Depressive-Like Behaviors in Mice

To investigate whether astrocytes in the mouse DRN influence depressive-like behavior in mice, we injected the DRN with either chemogenetically activated virus rAAV-GfaABC1D-hM3D(Gq)-P2A-mCherry (Figure 2(a)). Three weeks postinjection, after stable expression of the virus in the targeted brain regions, we recorded the behavioral characteristics of the mice under normal conditions using the OFT, TST, SIT, and SPT. Following these baseline recordings, the mice underwent a 14-day CSDS depression model. Behavioral changes were recorded after modeling and again 2 h after administration of clozapine-N-oxide (CNO) or saline to the control group (Figure 2(c)). At the end of the experiment, the location of the virus injection was verified by fluorescence microscopy (Figure 2(b)).

The results showed that CSDS modeling reduced the locomotor distance of mice in the OFT. This effect was reversed in the CNO treatment group, where locomotor activity was restored (Figure 2(d)). At the same time, the duration, times, and distance of mice entering the central area in OFT decreased significantly after CSDS modeling, and there was no significant difference in saline injection. After chemogenetic activation of DRN astrocytes, the anxiety levels of mice were improved, and the duration, times and distance of entering the central area were increased (Figures 2(e), 2(f), and 2(g)). In the SIT, chemogenetic activation of astrocytes in the DRN increased the time spent in the social interactive zone (Figure 2(h)) and decreased the time spent in the social escape zone (Figure 2(i)) compared to the control group. In the TST, chemogenetic activation of astrocytes in the DRN decreased immobility time and improved depressive-like behavior in mice (Figure 2(j)). In the SPT, chemogenetic activation of astrocytes of DRN increased the proportion of sucrose–water consumption (Figure 2(k)), indicating a reduction in anhedonia.

3.3. Chemogenetic Activation of DRN Astrocyte Activity Affected Mature and CaMKIIα Neuron Calcium Signaling in DRN

To investigate whether astrocytes in the DRN regulate calcium signaling in mature neurons, we injected chemogenetic viruses into the DRN. For activation, we used rAAV-GfaABC1D-hM3D(Gq)-P2A-mCherry. Simultaneously, we injected the GCaMP6f virus rAAV-hSyn-GCaMP6f-WPRE-hGH PA to label mature neurons (regardless of type) or rAAV-CaMKIIα-GCaMP6f-WPRE-hGH PA, which labels CaMKIIα neurons, into the DRN of mice. After the injections, a ceramic insert was implanted. Three weeks postinjection, after stable virus expression, we recorded calcium signaling changes in mature neurons under normal conditions and 2 h after CNO administration. These recordings were made during the OFT, TST, and SIT using a multichannel fiber optic photometric detection technique. Following these baseline recordings, mice underwent a 14-day CSDS model. Calcium signaling changes were recorded again postmodeling and 2 h after CNO administration (Figure 3(a)).

In the OFT, chemogenetic activation of DRN astrocytes did not significantly alter calcium signaling in mature neurons of WT mice. However, after CSDS modeling, there was a significant difference in the AUC for calcium signaling, with increased peak and duration observed 2 h after CNO administration (Figures 3(b) and 3(c)). In the SIT, chemogenetic activation of DRN astrocytes increased the peak calcium signaling of mature neurons in the DRN (Figures 3(f) and 3(g)). In the TST, activation of DRN astrocytes led to a decreased peak calcium signaling of mature neurons, with a significant difference in the AUC. After CSDS modeling, calcium signaling of mature neurons in the DRN were elevated, and peak signals were further increased 2 h after CNO administration, showing a significant difference in the AUC (Figures 3(f) and 3(g)). Chemogenetic activation of astrocytes following CSDS depression modeling resulted in a significant increase in CaMKIIα neuronal calcium signaling in the DRN during the OFT (Figure 3(d)), SIT (Figure 3(h)), and TST (Figure 3(l)). This increase was also reflected in a significant difference in the AUC for this calcium signaling (Figures 3(e), 3(i), and 3(m)).

3.4. Neural Projections From the DRN to the mPFC

To verify the presence of neuronal projections from the DRN to the mPFC, we used the Cre-loxP system. We injected the DRN of C57BL/6 mice with the Cre virus rAAV-hSyn-CRE-WPRE-hGH, which labels mature neurons, and the mPFC with adenovirus rAAV containing DIO α-DIO-EYFP-WPRE-hGH-Ef1 (Figure 4(a)). The fluorescent proteins carried by Cre and DIO are expressed in neurons only when neurons in the mPFC receive projections from the DRN. Consequently, green fluorescence was found in the mPFCs of mice after perfusion and sectioning of the virus-injected brain, confirming that the mPFC receives neuronal projections from the DRN (Figures 4(b) and 4(c)). After verifying these neural projections, we used immunofluorescence staining to classify and count the neurons in the mPFC receiving DRN projections (Figures 4(e), 4(f), 4(g), 4(h), 4(i), 4(j), 4(k), and 4(l)). The results showed that only 5.28% of EYFP-labelled neurons in the mPFC overlapped with GAD67, a marker for GABAergic neurons (Figures 4(d), 4(e), 4(f), 4(g), and 4(h)). This result suggests that DRN primarily sends neuronal projections to pyramidal neurons in the mPFC (Figures 4(i), 4(j), 4(k), and 4(l)).

3.5. 5-HT Increased EPSCs of Pyramidal Neurons and Presynaptic Transmitter Release in mPFC

Having demonstrated the presence of neuronal projections from the DRN to the mPFC using neural tracing, we sought to verify the modulatory effects of 5-HT from DRN neurons on mPFC neurons. To this end, we employed the patch clamp recording to examine the effects of 5-HT on pyramidal neurons in the mPFC using brain slices. We began by recording sEPSC in mPFC. Upon administering 5-HT (30 μM/mL), we found that both the frequency and amplitude of EPSCs in mPFC pyramidal neurons increased (Figures 5(a), 5(b), and 5(c)). To further investigate whether this increase was caused by enhanced presynaptic transmitter release, we conducted a PPR assay before and after 5-HT administration (Figure 5(d)). The results revealed a decrease in the PPR following 5-HT administration, suggesting that electrophysiological changes caused by 5-HT in the neurons due to an increase in presynaptic transmitter release (Figures 5(e) and 5(f)).

3.6. Chemogenetic Activation of DRN Astrocyte Activity Affected CaMKIIα Neuron Calcium Signaling in mPFC

To investigate whether astrocytes in the DRN regulate calcium signaling in CaMKIIα neurons in the mPFC, we injected the rAAV-CaMKIIα-GCaMP6f -WPRE-hGH PA virus, which labels pyramidal neurons, into the mPFC. Results showed that chemogenetic activation of astrocytes following CSDS resulted in a significant increase in the calcium signaling of pyramidal neuronal in the mPFC during the OFT (Figure 6(a)), SIT (Figure 6(c)), and TST (Figure 6(e)). This increase was also reflected in a significant difference in the AUC for this calcium signaling (Figures 6(b), 6(d), and 6(f)).