Animals
Male Sprague Dawley rats (275-325 g) (Charles River, Calco, Italy; San Diego, California, US) had access to water and food ad libitum and were handled in accordance with the guidelines for care and use of experimental animals of the European Community Council (2010/63/UE L 276 20/10/2010), with Italian law (DL 04.03.2014, N° 26; Authorization n° 371/2020-PR), with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the Institutional Animal Care and Use Committee of The Scripps Research Institute. Every effort was made to minimize suffering and reduce the number of animals used.
Drugs
Alcohol 1 g/kg (5.8 mL/kg) (Sigma-Aldrich, Milan, Italy) 20% (v/v) in water was administered intragastrically (i.g.). Caffeine (3 and 15 mg/kg) (Sigma-Aldrich, Milan, Italy) was dissolved in saline (3 mL/kg) and administered intraperitoneally (i.p.) 20 min before water or alcohol or dissolved in normal Ringer (see below) to 10 µM to be delivered by reverse dialysis in the pVTA, starting 30 min before water or alcohol. DPCPX and SCH 58261 (Tocris, Bristol, UK) were suspended in saline with 0.3% Tween-80 and in 0.5% methyl cellulose, respectively. Both drugs were administered i.p., at the dose of 2 mg/kg, 20 min before water or alcohol. (±)-Salsolinol (Santa Cruz Biotechnology Inc., Dallas, TX, United States) was dissolved in normal Ringer (see below) to 10 nM and delivered by reverse dialysis in the pVTA. The doses and the concentrations of alcohol (40, 41, 46-49), caffeine (10, 13, 24), DPCPX (50), SCH 58261 (51), salsolinol (37, 40, 52) and morphine (40) were selected based on previous literature.
Microdialysis experiments
Vertical probes were stereotaxically implanted in the pVTA and AcbSh according to the rat brain atlas of Paxinos and Watson (1998) (53): AP: -5.8 mm and ML: ±0.5 mm from bregma and DV: -8.0 mm from dura, for the pVTA; AP: 1.8 mm and ML: ±1 mm from bregma and DV: -7.6 mm from dura, for the AcbSh. Probes were implanted ipsilaterally, at random distribution between left and right brain sides. The location of the probes was reconstructed and referred to the rat brain atlas plates (53) through histological analysis. On the experiment day, pVTA and AcbSh probes were connected to an infusion pump and perfused with normal Ringer (in mM: 147 NaCl, 4 KCl, 2.2 CaCl2) at flow-rate of 1 μl/min. Dialysate samples (10 μL) were injected without purification into a high-performance liquid chromatograph (HPLC) to simultaneously quantify salsolinol and DA as previously described (41). Sensitivity of the assay was 5 femtomoles/sample.
Electrophysiological experiments
Rat brain slices were prepared as previously described (54). In brief, animals were decapitated under 5% isoflurane anesthesia. Brains were harvested and transferred to a modified ice-cold artificial cerebrospinal fluid (aCSF) solution containing (in mM): 220 sucrose, 2 KCl, 0.2 CaCl2, 6 MgSO4, 26 NaHCO3, 1.3 NaH2PO4, and 10 D-glucose (pH 7.4, adjusted by aeration with 95% O2 and 5% CO2). Horizontal brain slices containing the pVTA were sectioned (260 μm) in ice-cold modified aCSF using a Leica VT1200S vibratome (Leica, Heidelberg, Germany). Slices were transferred to a nylon mesh immersed in standard aCSF containing (in mM): 126 NaCl, 3 KCl, 2 CaCl2, 1 MgCl2, 26 NaHCO3, 1.25 NaH2PO4, and 10 D-glucose (pH 7.4, adjusted by aeration with 95% O2 and 5% CO2). After incubation for at least 40 min at 35°C, followed by at least 1 h at room temperature, the hemi-slices were transferred to the recording chamber and continuously perfused with standard aCSF at a flow rate of ~2 mL/min. The bath temperature was maintained at 33°C for all recordings.
Patch-clamp recordings from pVTA dopaminergic neurons were performed as previously described (54). Recording pipettes were prepared from borosilicate capillaries with an internal filament using a P-97 Flaming Brown micropipette puller (Sutter Instruments, Novato, CA, USA). The resistance of the pipettes ranged from 4.5 to 6.0 MΩ when they were filled with the following solution (in mM): 135 potassium gluconate, 10 MgCl2, 0.1 CaCl2, 1 EGTA, 10 Hepes-KOH (pH 7.3), and 2 ATP (disodium salt). Signals were recorded with an Axopatch 200-B amplifier (Axon Instruments Inc., San Jose, CA, USA), filtered at 2 kHz, and digitized at 5 kHz. The pClamp 9.2 software (Molecular Devices, Union City, CA, USA) was used to measure and analyze the firing rate and other membrane kinetic parameters of pVTA neurons and the occurrence of HCN-mediated Ih currents (see below). The cell-attached configuration was used to monitor the spontaneous and pharmacologically conditioned firing rate of DA neurons. After obtaining a pipette-membrane seal with a GΩ resistance, at least 10 min were allowed before recording to obtain a stable and regular spontaneous firing rate. In addition, the whole-cell configuration was obtained at the end of each recording to determine the presence of Ih currents, to confirm the identity of pVTA DA neurons (55). Accordingly, in our experimental conditions, identified pVTA DA neurons showed both a robust Ih (mean amplitude: -134.4 ± 15 pA, n = 60) in response to a single hyperpolarizing voltage step, from -65 to -115 mV, and a spontaneous regular firing rate of action potentials (4.26 ± 1.3 Hz n = 30). In each recording, after 3 min of recording a stable basal firing rate, different drugs were perfused: 60 mM alcohol (5 min), 10 μM caffeine (10 min), 10 nM salsolinol (10 min), 10 μM SCH 58261 (10 min), 10 mM DPCPX (10 min), and 1 μM morphine (10 min).
In-vitro synthesis of salsolinol
The protocol followed to synthesize salsolinol was an adaptation of Akbayeva et al., 2023 (56) to obtain a catalase-mediated oxidation of alcohol into acetaldehyde and the production of salsolinol in presence of DA via Pictet-Spenlger reaction. The blank consisted of bovine catalase (Sigma Aldrich, Italy) at 0.33 mg/mL (666.67-1666.67 units/mL) and DA hydrochloride (Sigma Aldrich, CAS No. 62-31-7) at 1.5 mM dissolved in PBS. Triplicates of blank, blank + 0.05 M catalase inhibitor 3AT (Sigma Aldrich, Italy), and blank + 0.05 M caffeine (Sigma Aldrich, Italy) were kept in an agitator at 37°C for 20 min. After that, PBS or 1 mM alcohol in PBS + 0.06 M hydrogen peroxide (Sigma Aldrich, Italy) in PBS were added to the solutions and the samples were placed back in an agitator at 37°C for additional 30 min. The reactions were then quenched with formic acid (FA, final concentration 1% v/v). The same steps were also followed using a more diluted catalase solution (0.0033 mg/mL or 6.67-16.67 units/mL). Samples were centrifuged at 4 °C for 15 min at 14 000 g, and the supernatant was collected and diluted 1:1000 in LC graded H2O before untargeted metabolomics analysis.
pVTA harvesting and sample preparation
Rats were randomly divided into 4 experimental groups: saline-water, saline-alcohol, caffeine-water, caffeine-alcohol. Subjects received pre-treatment with saline or caffeine i.p. (15 mg/kg). Twenty minutes after pre-treatment, rats were treated with water or alcohol (1 g/kg) i.g. and returned to their home cages. After 30 min from alcohol treatment, rats were decapitated under 5% isoflurane deep anesthesia, brains were removed and pVTA from both hemispheres harvested, weighted, and immediately frozen in dry ice. Pre-chilled LC graded 50% MeOH:H2O containing 1 µM sulfadimethoxine, as an internal standard, was added to each pVTA sample to obtain a final 1:20 w/v ratio. One 5 mm stainless steel bead was added to each sample before homogenization at 25 Hz for 5 min (TissueLyser II, Qiagen). Samples were left to incubate for 1 h at 4 °C before centrifugation at 14 000 g for 15 min at 4 °C. In separate Eppendorf’s tubes, 900 µL of supernatant was collected and added to 180 µL of formic acid (500 nM). Samples were then centrifuged again for 10 min at 14 000 g and 4 °C. The collected supernatant (1 mL) was then dried overnight in a speed vacuum concentrator. Samples were stored at -80°C and on the day of the untargeted metabolomics experiments were reconstituted in 200 µL of 50% acetonitrile (ACN) and vortexed.
Untargeted metabolomics and in-vitro synthesis of salsolinol experiments
For the metabolomics experiments, a Vanquish ultra-high performance liquid chromatography (UHPLC) system coupled to a Q Exactive quadrupole orbitrap mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) was used. Samples (5 μL) were injected into a Kinetex C18 column (50 × 2.1 mm, 1.7 µM particle size, 100 A pore size; Phenomenex, Cat#00B-4475-AN) at 30 °C column temperature. A flow rate of 0.5 mL/min was used for both the in-vitro synthesis of salsolinol and pVTA experiments with elution carried out using LC grade H2O (A) and 100% ACN (B), both acidified with 0.1% FA. Different elution gradients were used. For the in-vitro synthesis of salsolinol experiment: 0-1 min 0.1 % B, 1-3 min 0.1-40 % B, 3-3.5 min 40-100 % B, 3.5-5 min 100% B, 5-5.1 min 100-0.1% B, 5.1-6.5 min 0.1 % B; for the pVTA experiment: 0-1 min 5 % B, 1-7 min 5-100 % B, 7-7.5 min 100 % B, 7.5-8 min 100-5% B, 8-10 min 5 % B.
The mass spectrometer was operated in data-dependent acquisition (DDA) mode, and it was used in an m/z range from 100 to 1500 Da in the pVTA experiments and 50 to 750 Da in the in-vitro synthesis of salsolinol experiment, operating in positive ionization mode. Full scan MS1 was performed at 1e6 with a resolution of 35 000 and 70 000 for the pVTA and in-vitro synthesis of salsolinol experiment respectively, with a maximum ion injection time (IT) of 100 ms. MS2 experiments were performed at a resolution of 17 500 with maximum IT of 100 ms for pVTA and 50 ms for catalase, and TopN was used for the 5 most abundant precursor ions per MS2. The MS2 precursor isolation window was set to 1 m/z with no offset. The step collision energy was set to 20 eV, 30 eV, and 40 eV.
Metabolomics data processing
Acquired .raw files were converted into open-access .mzML format using MSConvert 3.0.23 (57). Both .raw and .mzML files have been deposited and can be downloaded from public metabolomics repository GNPS/MassIVE (https://massive.ucsd.edu/) under the accession codes MSV000094216 (pVTA experiment) and MSV000094218 (in-vitro synthesis of salsolinol experiment). Feature detection and extraction was performed using MZmine 3.9 (58). Briefly, mass detection noise for MS1 and MS2 was set at 5e4 and 1e3 respectively. ADAP chromatogram builder parameters were set as 4 minimum consecutive scans, 8e4 minimum absolute height, and 10 ppm m/z tolerance. Local minimum feature resolver module was set at 85% chromatographic threshold, 0.05 minimum search range RT, and 1.70 minimum ratio of peak top/edge. The 13C isotope filter was applied with an m/z tolerance of 5 ppm and a retention time tolerance of 0.03 minutes. Features were aligned using a m/z tolerance of 5 ppm and retention time tolerance of 0.2 minutes, with weight for m/z over RT was set to 3:1. Features not present in at least two samples and without MS2 acquisition were discarded. Finally, a feature list and two .mgf files, one for molecular networking (59) and one for SIRIUS (60), were exported for downstream analysis.
Metabolomics data analysis
Feature-based molecular networking analyses (61) were performed on GNPS (https://gnps.ucsd.edu/) and can be accessed for both pVTA (https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=abb23428a158496b8bd0c689a43d2940) and catalase (https://gnps.ucsd.edu/ProteoSAFe/status.jsp?task=60b61623c3874081a9b263371b03d49a) experiments. Briefly, tolerances for both precursor ion and fragment ions were set at 0.02 Da. For networking, a minimum modified cosine score of 0.7 and minimum number of matching peaks of 3 were set. Same parameters were set for library search. Generated annotation table was used for subsequent analysis and network were visualized using Cytoscape 3.10 (62). Compound classes were predicted using CANOPUS (63) in SIRIUS 5.8.5. For the in-vitro synthesis of salsolinol experiment, a targeted peak extraction was also performed using Skyline v23.1 (64). Feature list was imported in R 4.2.2 (The R Foundation for Statistical Computing, Vienna, Austria) for univariate and multivariate analyses. Feature list was first cleaned though blank filtering, only features with peak area ratio > 5 compared to blanks were kept. Data quality was assessed calculating coefficient of variance of internal standard in the samples and of the 6 standards present in the quality control samples (QCmix). Principal component analysis (PCA), via `mixOmics` v 6.22 package, was used to inspect data and visualize possible outliers. Before ordination, data was robust center log ratio transformed using `vegan` v 2.6. Batch effects were corrected using the removeBatchEffect function of `limma` v 3.54. Supervised multivariate partial least square discriminant analysis (PLS-DA) models were generated using `mixOmics` and performance (classification error rate) was assessed using a 4-folds cross validation.
Statistical analysis
Statistical analysis was carried out either via Statistica 8.0 (StatsSoft Inc., Tulsa, OK, USA) or PRISM, GraphPad 8 Software (San Diego, CA, USA) with significance set for all the experiments at p < 0.05. For microdialysis experiments, basal dialysate salsolinol and DA were calculated as the average ± SEM of the last three consecutive samples differing by no more than 10%, collected during the time preceding each treatment. Changes in dialysate salsolinol and DA were expressed as fmol/10 μl of dialysate and were analyzed by two- or three-way Analysis of Variance (ANOVA) with repeated measures over time. For electrophysiology experiments, all data are reported as mean ± SEM. Before ANOVA analyses, the normal distribution of data was evaluated by skewness and kurtosis, and homoscedasticity via the Bartlett test. Comparisons among experimental conditions were obtained using at least n=4 rats/group and was performed by one-way ANOVA followed by Tukey’s post-hoc test. Detailed statistical analysis for microdialysis and electrophysiology experiments is available in Supplementary Tables 1 and 2.