Mice
All experimental procedures were conducted in accordance with the Institutional Animal Care and Use Committee at Shandong University. Male C57BL/6j mice (6-8weeks old) were provided by the Department of Laboratory Animal Sciences, Shandong University. Mice were maintained under a 12-hourlight/dark cycle with food and water ad libitum.
Viral vectors
pAAV-EF1a-DIO-hM4Di-mCherry-WPRE, pAAV-Ef1a-DIO-mCherry-WPRE, pAAV-hSyn-EGFP-P2A-NLS-Cre-WPRE ( Retro), pAAV-CMV-DIO-KCNQ3-WPRE, pAAV-CMV-DIO-mCherry-WPRE, pAAV-CMV-GCaMP6s-P2A-nls-dtomato, pscAAV-Hsyn-Cre-tWPA, and pAAV-hSyn-mCherry-P2A-NLS-Cre-WPRE (Retro) viruses were purchased from Shanghai SunBio Biomedical technology Co. Ltd. (Shanghai). rAAV-NCSP-YFP-2E5 viruses were purchased from Brain Case Biotechnology Co., Ltd (Shenzhen). All viruses were aliquoted and stored at –80 °C and the viral titers were more than 1012 viral particles per mL. Red retrobeads (Lumafluor, USA) was diluted by PBS into a 4-fold solution for injection.
pAAV-CMV-DIO-KCNQ3-WPRE were made by sub-cloning their sequences into pAAV-CMV-DIO-EGFP-WPRE vector with respective titers of 4.7×1012 virus molecules per mL (Obio Technology, Shanghai, China).
Stereotaxic surgery for viral injections
Mice (6-8 weeks old) were anesthetized with 5% chloral hydrate (10mL/kg) intraperitoneal (ip) and then placed on a stereotaxic instrument (Model 900, Kopf, USA). A craniotomy was performed with a drill at the bilateral stereotaxic coordinates of the NAcLat (from bregma: +1.2 mm A/P, ± 1.7 mm M/L) or the VTA (from bregma: -3.4 mm A/P, ±0.5 mm M/L). An Hamilton syringe (10 μL) was then slowly lowered into the NAcLat (4.7 mm D/V) or VTA (4.5 mm D/V) and viruses or red retrobeads (0.2 μl) were injected bilaterally at a speed of 0.04 μl/min. The syringe was slowly withdrawn 10 min after the injection, the scalp skin sutured over the holes, and mice were placed back into their original cages for recovery. AAV-injected mice were subjected to the experiments two weeks after surgery whereas red retrobeads-injected mice 4 to 7 days after surgery.
Cannula implantation and microinfusion
Mice were placed on the stereotaxic apparatus after anesthesia. Mice were bilaterally implanted with a stainless 26-gauge cannula directly above the VTA (from bregma: -3.4 mm A/P, ±0.5mm M/L and 3.9mm D/V) or NAcLat (from bregma: +1.2 mm A/P, ± 1.7 mm M/L and 4.0 mm D/V). The cannula was fixed with dental cement. After the surgery, mice were housed individually, and the experiments carried out after 5 days of recovery 57.
After recovery from surgery, the cannulated mice received an intra-VTA microinfusion of ICA069673 (10μM, 200 nl, Med Chem Express) and intra-NAcLat microinfusion of clozapine N-oxide (CNO) (3 μM, 200 nl, Med Chem Express) or filtered 1× PBS as vehicle control. We performed bilateral microinfusion 0.2 μl per side through the cannula with an Hamilton syringe (5 μL) at a rate of 0.04 μL/min, removing slowly the syringe 5 min after the injection.
Conditioned place preference test
A custom-made two-compartment apparatus (15 × 15 × 30 cm) was used to perform conditioned place preference (CPP) test. One chamber had white walls and the other one black walls, and the two chambers were separated by a doorway (5 × 5 cm). The behavioral activity of the mice was recorded with a camera on the chamber lid.
We placed the mice in the CPP apparatus and left free to explore it for 15 min to assess their baseline place preference. On the first day (day1), control mice were injected with saline (10 ml/kg, i.p.) and the MA-administered mice were injected with MA (2mg/kg, i.p.). ICA069673-administered mice received a microinfusion of ICA069673 (10μM, 200 nl) through the cannula into the VTA 15min before MA (2mg/kg,i.p.) injection. Mice injected with AAV-hM4Di-mCherry or AAV-mCherry received a microinfusion of CNO (3 μM, 200 nl) into the NAcLat 15min before MA (2mg/kg,i.p.) injection. Then, mice were confined in the white chamber for 30 min before returning to their home cage. This procedure was repeated on day 3 and 5. On day 2, 4 and 6, control mice and MA-administered mice were injected with saline (10 ml/kg, i.p.), ICA069673-administered mice received a microinfusion of saline (200 nl) into the VTA 15min before saline (10 ml/kg, i.p.) injection. Mice injected with AAV-hM4Di-mCherry or AAV-mCherry received a microinfusion of saline (200 nl) into the NAcLat 15min before saline (10 ml/kg, i.p.) injection. Then, mice were confined in the black chamber for 30 min before return to home cage. After each trial, the apparatus was cleaned with 75% alcohol. On day 7, we left the mice to freely explore both sides of the CPP apparatus for 15 min. We got CPP score by subtracting the time spent in the black chamber from the time spent in the white chamber. The movement of animals in the two chambers was monitored and analyzed by a video tracking system (DigBehav, Jiliang Software Technology, China)67. For a scheme of the experimental protocol, please see Fig. 1A.
Brain slice preparation and electrophysiology
Brain slice preparation
We quickly removed the brains of the anesthetized mice after undergoing the CPP paradigm, and immediately placed them in the N-Methyl-D-glucamine (NMDG)-containing cutting solution containing the following (in mM): 2.5 KCl, 93 NMDG, 30 NaHCO3, 25 glucose, 1.2 NaH2PO4, 5 sodium ascorbate, 20 HEPES, 2 thiourea, 3 sodium pyruvate, 0.5 CaCl2, 10 MgSO4, pH 7.4, 295-305 mOsm. We used a vibratome (VT1200, Leica Microsystems, USA) to cutbrain slices with 300 μm thickness containing the NAcLat or VTA. We put the slices in the cutting solution to recover for 30 min at 36ºC saturated with 95% O2 and 5% CO2. Then, we left the brain slices at room temperature for 1 hour before use. We placed the NAcLat or VTA-containing slices in the perfusion chamber and continuously fed into a flowing oxygenated recording solution(in mM: 25 NaHCO3, 25 glucose, 125 NaCl, 2.5 KCl, 2 CaCl2, 1.25 NaH2PO4 and 1 MgCl2, pH 7.4, 295-305 mOsm) 68.
Currents and action potentials recording
We observed neurons under a 40x water immersion lens and used a laser to excite fluorescence. Glass capillaries (World Precision Instruments, USA) with a resistance of 3-5 mΩ were pulled on a micropipette puller (P-1000, Sutter Instrument Co., USA). Evoked and spontaneous currents were recorded at -65 mV. The recordings all used a K-gluconate-based internal solution (in mM: 126 K-gluconate, 10 KCl, 2 MgSO4, 4 NaCl, 0.3 Na-GTP, 4 Mg-ATP, 0.2 EGTA, 10 HEPES and 10 phosphocreatine, pH 7.3, 290 mOsm).
EPSC recording
EPSC was recorded at -65 mV with a pipette solution (in mM: 130 CsCl, 2 MgCl2, 2 Mg-ATP, 10 HEPES and 0.2 EGTA, pH 7.2, 290 mOsm). When recording EPSCs, 100 μM Picrotoxin and 5 μM GP52432 were added to the recording solution to block GABA-A and GABA-B receptors, respectively.
Patch-seq
Single DA cell isolation
Brain slices were prepared as described above for electrophysiology. We placed VTA-containing slices in the perfusion chamber and continuously fed into a flowing oxygenated recording solution. We found the retrobeads-labeled VTA cells under the microscope, performed whole-cell recording with the pipette filled with internal solution, and gave -300pA current stimulation to determine whether they were DA neurons. We then used a new pipette with a tip that was about a quarter to a third the diameter of the cell body. We applied negative pressure to the pipette and could see the entire cell body entering the pipette. The pipette was discarded if extracellular contents entered the pipette. Otherwise, we used a positive pressure to eject the contents into a PCR tube which contained 4 µl of RNase-free lysis buffer.
Library construction and sequencing
The samples in the PCR tube were directly amplified using the Smart-Seq2 method, and the amplified product cDNA with a length of about 1-2 kb was obtained by the reaction. Qubit® 3.0 Flurometer was used to measure the cDNA concentration of amplified products. The Agilent 2100 Bioanalyzer was used to detect the fragment distribution of amplified cDNA samples to ensure the quality. Library construction was performed using the amplified cDNA. The sample cDNA was disrupted into small fragments of approximately 350 bp by sonication using the Bioruptor® Sonication System (Diagenode Inc.). We perform end repair, 3’ ends A tailing and adapter ligation on the samples. After each step, Beckman Ampure XP magnetic beads were used for purification. The adaptor products were taken for PCR amplification, and each sample was introduced with different Index tags for distinguishing from each other during on-machine sequencing. The PCR amplification products were passed through the Pippin HT to construct the final library. After library construction, the library was tested for fragment length distribution and effective concentration using the Agilent 2100/LabChip GX Touch and Q-PCR. Eligible libraries were loaded on the HiSeq sequencing platform for PE150 sequencing program 23.
Immunofluorescence staining and confocal imaging
Mice were perfused transcardially with 4% paraformaldehyde after anesthesia 30 minutes after the CPP test. Mouse brains were removed and fixed overnight at 4ºC before equilibration in 30% sucrose solution. We used a cryostat (Leica Biosystems, Germany) to prepare the sections (40 μm). Free-floating immunohistochemistry was performed on the obtained sections that were blocked with 0.3% Triton X-100 and 5% bovine serum albumin (BSA) for 1 h. Then, we incubated sections with primary antibodies (mouse anti-TH (1:500, Santa cruz, USA), mouse anti-Arc (1:500, Santa cruz, USA), rabbit anti-Kv7.3 (1:500, Alomone, Israel)) overnight at 4ºC. We incubated sections with secondary antibody (donkey anti-rabbit IgG Alexa 647 (1:200, Proteintech, USA) and FITC goat anti-mouse IgG (1:200, Proteintech, USA)) for 2 hours. The slides were mounted after washing the sections with PBS. We obtained confocal images with a LSM880 confocal microscope (Zeiss, Germany).
Brains of the mice infected with rAAV-NCSP-YFP-2E5 were sectioned at 70 μm. Slices were slide mounted and images were obtained with a 63× oil objective on the LSM880. We measured spine density from YFP-expressing VTA neurons and used the software ImageJ to analyze spine types 69. The spine morphology was divided into 3 types: mushroom spine, thin spine and stubby spine 70, 71.
RT-PCR
The brains of the anesthetized mice were removed and we used a stereomicroscope to dissect the VTA and stored the tissue at -80 ºC. We extracted total RNA using RNA fast 200 kit (Fastagen, China). We detected the quality and quantity of RNA samples by using Nanodrop 2000 Spectrophotometer, and used PrimerScript RT reagent kit with gDNA Eraser (TAKARA, Japan) for reverse transcription of total RNA on the basis of the manufacture’sprotocol. We performed qRT-PCR with FastSYBR Mixture (CWbio, China), and run the PCR program as follows: denaturing at 95 ºC for 120s, 40 cycles at 95 ºC for 35s, 62 ºC for 30s, and 72 ºC for 40s, and a final extension at 72 ºC for 300s.
Local field potential
Mice were anesthetized with 5% chloral hydrate (10mL/kg), fixed on a stereotaxic apparatus and after performing a craniotomy implanted with a single tungsten wire (diameter 50 μm) for recording local field potential (LFP) in the VTA (-3.4 mm A/P, ±0.5mm M/L, 4.5 mm D/V) or in the NAcLat (1.2 mm A/P, ±1.7mm M/L, 4.7 mm D/V). The reference electrode (miniature stainless-steel screw) was implanted over the cerebellum. The electrodes were fixed on the skull using dental acrylic and then connected to the amplifiers (AD Instruments, Australia) while the animal still anesthetized to process a recording for 5 min. We used LabChart software to record and analyze data for getting the power spectral density and spectrogram. We divide neuronal oscillations into five frequency bands: delta (1–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), beta (13–30 Hz) and gamma (30–100 Hz).
Electrochemistry
Carbon fiber electrode preparation
5 μm diameter carbon fiber (Tokai Carbon Co., Japan) with a length of about 10 cm was inhaled into a glass capillary (coreless, outer diameter: 1.5 mm, inner diameter: 0.89 mm, length: 8 cm), and the pipette was drawn by the P-1000 micropipette puller. Under the microscope, the exposed carbon fiber was cut to 100−300 μm with a surgery scalpel. We used silicone rubber to fill the tip of the glass capillary.
DA release detection
Brain slices containing the NAcLat were prepared as described above for electrophysiology. KCl (3M solution was added to the prepared carbon fiber electrode, and then the electrode was fixed on the micromanipulator arm of the patch clamp, and the silver wire was immersed in the solution in the electrode tube. The NAcLat-containing slices were placed in the perfusion chamber and continuously fed into a flowing oxygenated recording solution. The NAcLat was found under the microscope, and the carbon fiber electrodes were inserted. In the voltage-clamp mode, a voltage of 780 mV was added, and the spike generated by the release of secretory DA could be observed. When the voltage returned to 0 mV, the spike disappeared. The addition of 70mM high K+ solution to the extracellular fluid caused the depolarization of cells to release DA. The carbon fiber electrode can oxidize DA to dopamine quinone, and the current generated by the oxidation, which is proportional to the amount of DA content, can be detected by the carbon fiber electrode. A calibration curve with different standard concentrations of DA was performed to calculate the concentration of DA in the NAcLat 72, 73.
Statistical analysis
We used GraphPad Prism 8.0.2 version software for the statistical analysis of the data. Data were checked for normality using Kolmogorov–Smirnov test. Depending on the experimental conditions, we used t-test (unpaired, two-tailed) and one-way or two-way ANOVA followed by Bonferroni test post-hoc comparisons when allowed. Cumulative frequency and amplitude plot were analyzed with the two-sample Kolmogorov-Smirnov test. All data are presented as mean ± SEM. A p-value <0.05 was considered to indicate statistical significance.