C57BL/6j mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The mice were placed in the laboratory for 2 h for adaptation and then housed in separate cages. The mice were placed in an environment with a constant humidity and temperature of 50% ± 10% and 21°C ± 2°C, respectively. The mice were also subjected to a day: night cycle of 12:12 h and ad/lib feed and water for the duration of the study. The C57/Bl6j male mice aged 15–20 days were used for the preparation of the SNc brain slices. Additionally, C57BL/6j male mice aged 7–8 weeks were used for double immunofluorescence staining. The animal feeding and experimental procedures in this experiment were conducted strictly in accordance with the Qingdao University Laboratory Animal Use Regulations.

To avoid estrogen interference, four male mice were selected for immunofluorescence experiments (Kim et al., 2022; Kim et al., 2023). Male mice were anesthetized with pentobarbital sodium (70 mg/kg, i. p., dissolved in saline) and administered physiological saline and 4% paraformaldehyde solution through the heart. The brains were removed and fixed in a 4% paraformaldehyde solution, and then dehydrated in 20% and 30% sucrose solutions at 4°C sequentially. Subsequently, the brains were frozen and sectioned into 30 μm slices for immunofluorescence staining. The slices were incubated with 5% bovine serum albumin (BSA) for 1 h to block non-specific binding, before incubation with primary antibodies containing mouse anti-tyrosine hydroxylase antibodies (CST, 1:1,000) and rabbit anti–CB2R antibodies (Abcam, 1:250) at 4°C for 12 h. Secondary antibodies (goat anti-mouse immunoglobulin G (IgG) (Abcam, 1:500); goat anti-rabbit IgG (Abcam, 1:500)) were then applied at room temperature for 1 h. Following incubation, the sections were washed, dehydrated, mounted on polylysine-coated glass slides, and examined using a fluorescence microscope (BX 53, Olympus, Tokyo, Japan). Sections were also observed using an Olympus fluorescence microscope (VS120, Olympus, Tokyo, Japan) and 20x or 40x objective lens micrographs obtained.

In the present study, postnatal day 15 (P15) to P20 male mice were used as no significant changes in CNR2 mRNA levels are observed in male mice during P15–P40 (Dufour et al., 2014). The slice patch-clamp recordings were the same as reported previously (Chang et al., 2020). Briefly, the mice were anesthetized before decapitation, and slices were cut in an ice-cold, oxygenated cutting solution containing (in mM) 124 NaCl, 3 KCl, 0.5 CaCl₂, 2 MgCl₂, 1.3 NaH₂PO₄, 26 NaHCO₃, and 10 glucose. Coronal slices (250 μm) containing SNc were cut using a vibratome (LeicaVT1000S, Germany) and transferred to a holding chamber. The slices were the incubated at 36 °C in artificial cerebrospinal fluid (ACSF) containing (in mM) 124 NaCl, 3 KCl, 1.3 NaH₂PO₄, 1.3 MgCl₂, 2.4 CaCl₂, 26 NaHCO₃, and 10 Glucose (osmolarity, 300–310 mM, pH 7.4), and continuously bubbled with 95% O₂ and 5% CO₂. After 1 h of recovery, the slices obtained were then transferred to an imaging chamber on the stage of an upright microscope (BX51WI; Olympus Tokyo, Japan) and continuously perfused with oxygenated standard ACSF at a flow rate of 2 mL/min at RT.

Whole-cell patch-clamp recordings were conducted using glass pipettes with a resistance of 3–7 MΩ. The internal solution (295 mOsm) contained (in mM): 120 K-gluconate, 10 HEPES, 10 EGTA, 20 KCl, 2 MgCl₂, 2 Na₂ATP, 0.3 Na₃GTP, and 10 Na₂-phosphocreatine; pH 7.2 with KOH. Cells were visualized under infrared differential interference contrast (DIC) microscopy, and electrodes were positioned using a micromanipulator. After a tight seal (resulting in electrode resistance >1 GΩ) was formed between the electrode tip and the cell surface, suction was briefly applied until a whole-cell patch-clamp recording configuration was obtained (access resistance lower than 20 MΩ). Recordings started at least 5 min after establishing whole-cell configuration to allow the proper wash-in of the intra-pipette solution. Whole-cell patch clamp recordings of SNc neurons was conducted to record spikes under current-clamp mode. Step-current injections (of 800 milliseconds) were delivered from 0 pA to 50 pA increments. The number of action potentials (AP), AP initiation, AP duration, after-hyperpolarization (AHP), and inter-spike interval (ISI) between the first and second firings were measured in current mode with the cell at −70 mV and were analyzed with a custom Matlab program (R2018b).

For assessments of mEPSCs, electrodes were filled with an intracellular solution containing (mM): 133 K-gluconate, 8 NaCl, 0.6 EGTA, 2 MgATP, 0.3 Na₃·GTP, and 10 HEPES. To allow verification of the identity of recorded neurons, sodium channel blocker tetrodotoxin (TTX 1 μM), D-APV (a NMDA receptor antagonist, 50 μM) and picrotoxin (a selective GABA_A receptor antagonist) 100 μM were included in the solution. The membrane potential was clamped at −70 mV and the mEPSCs was recorded for 15 min. For mIPSCs recordings, electrodes were filled with an intracellular solution containing (mM): 130 KCl, 1.0 MgCl₂, 5 EGTA, 5 Na₂·ATP, and 5 HEPES. To allow for verification of the identity of recorded neurons, TTX, NBQX (an AMPA receptor antagonist, 10 μM) and D-APV (50 μM) were included in the solution. The membrane potential was clamped at −60 mV, recording time were the same as above.

Total 53 mice were used in electrophysiological experiments, we maintained two slices from each mice and then selected one neuron for patch recordings based on the brain slice. Only the neurons with the morphology and electrophysiological properties of dopaminergic neurons in SNc were included for further analysis. Series resistance was automatically compensated using a patch-clamp amplifier (Multiclamp 700 B, Molecular Devices). Data acquisition and analysis were performed using a digitizer (DigiData 1550B, Molecular Devices) and a pClamp10 analysis software (Molecular Devices). Signals were filtered at 2 kHz and sampled at 10 kHz. Offline analyses of the data obtained from electrophysiological recordings were conducted using the Clampfit software version 10.7 (Axon Instruments, Inc., United States). A semi-automated sliding template protocol was utilized to analyze both classes of miniature data. The detection criteria were established by optimizing the scaling factor and ensuring an optimal fit quality. Events were confirmed when the criterion exceeded a predefined threshold level. Furthermore, the algorithm dynamically adjusted for variations in recording noise, considering only amplitudes that were at least three times the standard deviation of the noise (3σ). Each event detected by the template was individually evaluated and accepted for analysis based on two criteria: non-overlapping events and stable baseline recordings (2.5 ms), both before and after the rising and decay phases of mIPSCs/mEPSCs. Once validated, the mIPSCs/mEPSCs were aligned at their onset and averaged.

JWH133 and AM630 were purchased from APE x Bio (United States). Other chemical drugs used for the electrophysiology experiments were obtained from Sigma. The Graphpad prism 5.0 software was used to statistically analyze the experimental results and to make the corresponding results analysis and trend graphs. The experimental data are expressed as mean ± standard error (Mean ± S.E.M.), and paired t-test of P < 0.05 was considered statistically significant.

Firstly, we observed the expression of CB2Rs on dopaminergic neurons in the SNc of normal mice. Double-labeling immunofluorescence showed that dopaminergic neurons in the substantia nigra expressed CB2Rs. As shown in Figure 1A, some dopamine neurons in the substantia nigra labeled with TH (red) also exhibit CB2R immunoreactivity (green).

Based on our previous studies, we identified SNc dopaminergic neurons based on their morphological and electrophysiological characteristics: (1) SNc dopaminergic neurons are mainly located at the ventral aspect of the midbrain (Figure 1B); (2) they are densely packed, with large neuronal cell bodies, and are generally spindle, triangular, or multipole shaped (Figure 1C); (3) when dopamine neurons are stimulated with a hyperpolarized current of −150 pA (after whole-cell recordings in the current-clamp mode), significant inward rectification is observed in the neuronal membrane potential (Figure 1D); (4) In Gap free mode, SNc dopaminergic neurons exhibit slow and regular spontaneous firing activity (Figure 1E); and (5) the duration of the AP is relatively long, with a half width of about 3 ms (Figure 1F). Based on the abovementioned morphological and electrophysiological characteristics, neurons identified as dopaminergic were used in subsequent experiments.

As mentioned above, most SNc dopaminergic neurons exhibit regular tonic discharges. To observe the effect of CB2R activation on SNc dopaminergic neuronal excitability, we first selected 1 μmol/L CB2R agonist JWH133 perfusion brain slices according to previous literature reports. As demonstrated in Figure 2A, among the recorded seven dopaminergic neurons, the average firing frequency was 1.08 ± 0.13 Hz. After perfusion with JWH133, the firing frequency decreased to 0.66 ± 0.09 Hz (P < 0.05). The original frequency could be restored following washing with normal ACSF.

To further confirm whether the inhibitory effect of JWH133 on dopaminergic neuronal excitability was dose-dependent, brain slices perfused at four different concentrations (0.01, 0.1, 10, and 30 μmol/L) were selected. As displayed in Figure 2B: 0.01 μmol/L JWH133 demonstrated no significant effect on neuronal discharge frequency (P > 0.05) in the recorded display among the five neurons. In the recorded seven neurons: 0.1 μmol/L JWH133 had no obvious effect on neuron firing frequency (P > 0.05). However, 10 μmol/L JWH133 significantly reduced the neuron firing frequency, from the basic 1.18 ± 0.07 Hz to 0.80 ± 0.69 Hz (P < 0.01). Similarly, 30 μmol/LJWH133 significantly reduced neuron firing frequency, from the basic 1.11 ± 0.13 Hz to 0.69 ± 0.09 Hz (P < 0.01). These results indicate that the effect of JWH133 on the firing frequency of dopaminergic neurons has a significant dose correlation. Figure 2C illustrates that perfusing brain slices with 1 μmol/L of the CB2R blocker AM630 alone had no statistically significant effect on the firing frequency of dopaminergic neurons in the SNc region compared to the control group. When brain slices were perfused with both JWH133 (1 μmol/L) and AM630 (1 μmol/L), the inhibitory effect of JWH133 was blocked.

In the current-clamp mode, we administered a depolarization current stimulation of 10–50 pA to dopaminergic neurons and recorded the corresponding indicators for evoked discharges. The results demonstrated that among the seven recorded neurons, the number of AP bursts were positively correlated with increasing depolarization current amplitude (Figure 3A). In contrast, the number of APs induced by the depolarization current decreased significantly after perfusion with 10 μmol/L JWH133 (Figure 3A). To explore whether activating CB2R in the SNc brain region affects the excitability of neurons, we prepared acute SNc slices. We found that activating CB2R on dopaminergic neurons in the SNc significantly reduced neuronal excitability. Specifically, depolarizing current (50 pA, duration 800 ms) triggered larger inter-spike interval (ISI) (Figure 3B), larger afterhyperpolarization potentials (AHP), longer action potential initiation times, and fewer action potential numbers compared to the control group, indicating reduced excitability (Figures 3C–F; paired t-test, ISI: t = 4.055, df = 6, P < 0.01; AP initiation: t = 3.392, df = 6, P < 0.05; AHP: t = 3.187, df = 6, P < 0.05; AP duration: t = 2.403, df = 6, P = 0.0531; AP number: t = 5.303, df = 6, P < 0.01). Therefore, we concluded that activating CB2R in the SNc region reduces the intrinsic excitability of SNc dopaminergic neurons.

As CB2R is also expressed at pre-synapses in the SNc area, we also investigated whether JWH133 affects synaptic currents. We first examined the regulatory effects of JWH133 on mEPSCs. After whole-cell recording on mouse SNc brain slices was conducted, a set of spontaneous inward currents was obtained in the Gap free recording mode. After incorporating TTX to block the sEPSCs caused by the AP, mEPSCs were obtained. This inward current was completely blocked by NBQX and AP-5, confirming that the EPSCs were mediated by excitatory amino acid receptors (Figure 4A). As shown in Figure 4C, compared to the control group, the frequency of mEPSCs slightly decreased after the application of 10 μmol/L JWH133 (n = 10, t = 3.146, df = 9, P < 0.05). The half decay time of mEPSC slightly increased (n = 10, t = 2.370, df = 9, P < 0.05, Figure 4D). However, there was no significant difference in amplitude between mEPSC and the control group (n = 10, t = 0.09425, df = 9, P > 0.05, Figure 4B). This result suggests that the activation of CB2R in the SNc region alters the frequency of mEPSCs, which may be related to changes in the probability of pre-synaptic release and/or the number of synapses. We also observed an effect of JWH133 on GABA_A receptor-mediated mIPSCs. After the addition of NBQX (10 μmol/L) and AP-5 (50 μmol/L) to the perfused ACSF to block EPSCs caused by excitatory amino acids, IPSCs were isolated. After adding TTX (1 μmol/L) to block spontaneous IPSCs (sIPSCs) induced by APs, mIPSCs were obtained (Figure 4E). In 10 SNc slices prepared from 10 wild-type mice, 10 μM JWH133 bath perfusion for 10 min did not change the frequency, amplitude, or half decay time of mIPSCs (Amplitude: t = 0.1849, df = 9, P > 0.05; Frequency: t = 0.8929, df = 9, P > 0.05; Half decay time: t = 0.2911, df = 9, P > 0.05) These results suggest that the activation of CB2Rs has l effect on presynaptic GABA release or postsynaptic GABA_A receptor function. (Figures 4F–H).