Adolescent CD1 male mice (Envigo, Netherlands), weighing 22–27 g, were used from postnatal day (PND) 25 to 80 for intravenous self-administration and subsequent behavioral and neurochemical studies. Mice were housed in groups of three to six per cage with standard conditions of temperature (21 ± 1 °C) and humidity (60%) under a 12-h light/dark cycle (lights on at 7.00 AM) and with ad libitum access to water and food (Mucedola, Italy) until the intravenous self-administration (IVSA) experiments. All animal care procedures and experiments were carried out in accordance with European Council directives (609/86 and 63/2010) and in compliance with the animal policies issued by the Italian Ministry of Health and the Ethical Committee for Animal Experiments (CESA, University of Cagliari). We made all efforts to minimize pain and suffering and to reduce the number of animals used, according to the 3Rs principles.
JWH-018, AM251, and AM630 were purchased from Tocris (Bristol, UK). Drugs were solubilized in 0.5% EtOH, 0.5% Tween 20, and 99% saline. AM251 and AM630 (CBR1- and CBR2-antagonist/inverse agonists, respectively) were administered intraperitoneally (i.p.) 30 min prior to JWH-018 IVSA at different doses depending on the group of animals; AM251: 0.3–1.0 mg/kg (volume 10 mL/kg); AM630: 0.5 and 1.0 mg/kg (volume 10 mL/kg).
Overall experimental design and study group
This study was designed to characterize the behavioral and neurobiological consequences of adolescence JWH-018 IVSA at adulthood (Fig. 1). First, we performed IVSA experiments in adolescent mice (PND 30–55) to assess (i) the JWH-018 dose–response curve (Experiment I); (ii) the response rates in different reinforcement IVSA schedules (FR1-FR3-PR; Experiment II); (iii) the response rates in different JWH-018 IVSA phases (Acquisition, Extinction, Reacquisition; Experiment III), and (iv) the effect of CBRs blockade on JWH-018 IVSA (Experiment IV). Afterwards, mice from Experiment II (IVSA from PND 34 to 53) were used for behavioral and neurochemical characterization at adulthood. In particular, after 3 days of recovery from surgery (PND 31–33), IVSA experiment started under different schedules of reinforcement from PND 34 to 53 for 7 days/week. At adulthood, after a drug-free washout period (~ 3 weeks), from PND 77 to 79 animals were randomly divided into different groups to perform behavioral tests, with a sub-group of the mice of each treatment group assigned to repetitive/compulsive-like behaviors’ evaluation (marble burying and nestlet shredding tests) and another sub-group assigned to risk-taking behaviors’ evaluation (wire-beam bridge). At PND 80, mice were sacrificed for subsequent neurochemical assays. In particular, a sub-group of the mice of each group were randomly assigned to immunohistochemical evaluations (GFAP and IBA-1 immunoreactivity), and another sub-group to western immunoblotting (COX-2, EAAT2, CB1 expression) and cytokines measurements. The experimental group sizes (n ≥ 5) were chosen based on our previous experimental protocols (Pintori et al. 2021; Castelli et al. 2014) and are shown in the figure legends. Due to experimental protocol criteria (e.g., acquisition criteria during IVSA phase) and/or technical issues (e.g., catheter obstruction or damage), some animals were excluded from statistical analysis, thus reducing group sizes in few cases.
Intravenous self-administration studies during adolescence
Detailed descriptions of surgery and IVSA methods are included in the Supplementary Information (SI).
Characterization of JWH-018 dose–response curve in the IVSA experimental paradigm (Experiment I)
The first experiment was aimed at characterizing a dose–response curve of JWH-018 (2.5–15 µg/kg/25 µl infusion) in adolescent mice. Mice IVSA experiment was carried out under Fixed Ratio (FR) 1 schedule of reinforcement (1 right lever press: 1 injection; 20-s time-out with light off, session duration 2 h), from PND 30 until PND 53 (24 sessions). During time-out (TO) period, right (active) lever presses had no programmed consequences, while left (inactive) lever presses were never associated with programmed consequences.
Characterization of response rates in different reinforcement IVSA schedules (FR1, FR3, PR; Experiment II)
Once the dose at which adolescent mice acquired operant behavior was established (7.5 µg/kg/25 µl infusion), we used different FR protocols to evaluate the reinforcing properties of JWH-018. After a first acquisition phase under FR1 schedule of reinforcement, in which the number of lever pressing achieved 80% of stability for at least three consecutive sessions (7 sessions in total), mice underwent FR3 (3:1) for 12 sessions. The same protocol of IVSA was used for the control group, which was exposed to Vehicle solution (0.5% EtOH, 0.5% Tween 20, and 99% saline). During the last session of the JWH-018 IVSA, after at least 75% responding on the active lever in three consecutive FR3 sessions, the experiment was performed under a Progressive Ratio (PR) schedule of reinforcement in which the number of active lever presses required to obtain each subsequent injection was based on the adapted exponential sequence: 1, 2, 4, 6, 9, 12, 15… (Valentini et al. 2013). PR sessions lasted for 2 h or until mice did not complete the ratio for the delivery of at least one injection within 1 h.
Characterization of response rates in different JWH-018 IVSA phases (Acquisition, Extinction, Reacquisition; Experiment III)
After the acquisition phase under FR1-FR3 schedules (see acquisition criteria), a different group of mice underwent extinction and reacquisition phase protocols to assess seeking behavior and reacquisition of operant behavior after the extinction phase to JWH-018. Extinction session schedule was identical to the acquisition schedule, except for the absence of any delivery of JWH-018 infusion. After 6 sessions performed in extinction phase, the protocol was switched to reacquisition phase for 7 sessions in the same condition of acquisition.
Effect of CBRs blockade on JWH-018 IVSA (Experiment IV)
To investigate the role of CBRs on operant behavior, in two separate groups of animals subjected to the same protocol of the Experiment II (except for PR), we evaluated the effects of the CB1R antagonist/inverse agonist AM251 (0.3–1.0 mg/kg i.p.), and the CB2R antagonist/inverse agonist AM630 (0.5, 1.0 mg/kg i.p.) injections on IVSA behavior. AM251, AM630, or vehicle has been administered 30 min before starting one of the IVSA FR3 session that reached the established criteria of stability (see acquisition criteria); before each antagonist pretreatment, responsiveness for JWH-108 was established to the same stability criteria.
Behavioral studies at adulthood
For the behavioral studies, after a drug-free period (~ 3 weeks) from the last session of IVSA, adult mice (PND 77–79) from Experiment II were tested in marble burying, nestlet shredding and wire-beam bridge tests to assess potential repetitive/compulsive-like and risk-taking phenotypes, respectively (Fig. 1). To minimize stress carry-over effects, mice were subjected to these behavioral tasks in the same experimental room (light intensity between 20 and 30 lx), with an inter-test interval of 24 h and performed in the following order: marble burying, nestlet shredding, and wire-beam bridge tests.
Marble burying test
Marble burying test was performed in a transparent plastic cage (50 cm L, 30 cm W, 20 cm H) containing 5 cm of fresh hardwood chip bedding. Twenty standard glass marbles (1.5 cm in diameter, arranged in five rows of four marbles each) were placed uniformly over the surface of bedding. Mice were individually placed in the cage and their activity was recorded for 20 min. At the end of the session, animals were gently removed, and the number of marbles totally (≥ 95%) buried was counted. In order to avoid the presence of olfactory cues, bedding was replaced and marbles were cleaned between each mouse test.
Nestlet shredding test was performed in the same type of cage used for marble burying, but with 3 cm of fresh hardwood chip bedding containing a 5 × 5 cm packed cotton nestlet (Ancare Corp; Bellmore, NY) laid on the top of the bedding material. Mice were individually placed in the cage and their activity was monitored for 75 min. Then, the remaining cotton nestlet was dried overnight and the amount (i.e., percentage) of the nestlet shredded was determined by weighing the day after.
This test was performed as previously described, with slight modifications (Frau et al. 2017). The apparatus consisted of 50-cm high Plexiglas platform and a 100-cm high Plexiglas wall, oppositely placed at 50 cm distance. Platforms were connected by a horizontal, flexible wire-mesh metallic grid. Mice were individually placed in the proximity of the edge (3 cm from the edge) of one platform, to make the starting position uncomfortable and promote movement. Behavioral measures, as index of behavioral disinhibition, were scored for 5 min and included the latency to access the bridge (with all 4 paws on it) and to first movement; movement duration and frequency; stretch attend postures and head dipping phenotypes.
At the end of the behavioral tests (PND 80), a sub-group of the animals from the Experiment II were selected for immunostaining analyses, while another sub-group for cytokines measurements and western immunoblotting (see Fig. 1).
Brain tissue preparation and GFAP and IBA-1 immunofluorescent staining
Procedures were carried out as previously described (Castelli et al. 2014; Pintori et al. 2021) (for details, see SI).
Imaging and quantitative analysis of GFAP and IBA-1 immunofluorescent staining
An Olympus IX 61 microscope and an Olympus 12-bit cooled F View II camera (Hamburg, Germany) were used for observations and for capturing the images, respectively (For details, see SI).
Cytokine measurements and Western immunoblotting
Mice were sacrificed by cervical dislocation; striatum and cortex were quickly collected and snap frozen in liquid nitrogen and stored at − 80 °C. For cytokine measurements and immunoblotting for excitatory amino acid transporter [(EAAT2) or glutamate transporter (GLT-1)] and cyclooxygenase 2 (COX-2), tissues were prepared following the instruction of Bio-Plex Cell Lysis Kit Product Insert (for details, see SI).
Brain cytokine concentrations were measured using a fully quantitative ELISA-based chemiluminescent assay (BioRad 23-plex mouse kit), which simultaneously detects cytokines, chemokines, and growth factors. All samples were run in duplicate and assayed with the BioRad reagent kit and cell lysis for tissue samples, according to the manufacturer’s instructions. Lyophilized cytokines standards containing cytokines: [(IL)1α, IL1β, IL2, IL3, IL4, IL5, IL6, IL9, IL10, IL12(p40), IL12(p70), IL13, IL17, interferon (IFN)γ, TNFα; chemokines: eotaxin, monocyte chemotactic protein-1 (MCP-1), macrophage inflammatory protein (MIP)1α, MIP1β, regulated on activation normal T cells expressed and secreted (RANTES), keratinocyte derived chemokine (KC/GRO/CXCL1), and growth factors: granulocytes macrophage colony-stimulating factor (GM-CSF) and granulocyte (G)-CSF)] were reconstituted to a master standard stock (for details, see SI).
EAAT2, COX-2, and CB1R immunoblotting
Samples (cortex and striatum) were thawed on ice and diluted to final concentration of 12 µg/µl protein.
Samples with equal amounts of protein (60 µg) mixed with Laemmli loading buffer were denatured at 70 °C for 10 min and loaded on gradient gel (4–12% NuPAGE Bis–Tris mini gels 15 well 1 mm gels, Life Technologies, CA, USA). To identify the specific bands, internal molecular weight (MW) standards (Precision Plus Protein Western C Standards, Bio-Rad, Hercules, CA, USA) were run in parallel. Then, proteins were transferred onto polyvinylidene difluoride (PVDF) membranes following the company’s protocol (Amersham GE Healthcare, UK). The protein transfer was checked by the ponceau red coloring of the membrane. Membranes were blocked for 1 h at room temperature using a mixture of 20 mM Tris base, 137 mM sodium chloride, and 0.1% Tween 20 (TBS-T) containing 5% BSA or dry powder milk before incubation overnight at 4 °C with the primary antibodies. The following primary antibodies were used: mouse monoclonal anti-EAAT2 (1:500; SC-365634; Santa Cruz, Dallas, TX), mouse monoclonal anti-COX-2 (1:100; SC-376861; Santa Cruz, Dallas, Texas, USA), and mouse monoclonal anti-β-Actin (1:2000; mAbcam 8226; Abcam, Cambridge, UK). For detection of COX-2, the blot was stripped with Restore Western Blot Stripping Buffer (Thermo Scientific, Rockford, IL) and re-blotted with the anti-COX-2 overnight at 4 °C. Blots were incubated with goat horseradish peroxidase (HRP) conjugated secondary anti-mouse antibody (1:5000; Vector, CA, USA) for 1 h at RT, and after TBS-T wash, incubated with the chemiluminescent detection solution Clarity Western ECL Substrate (Bio-Rad, Hercules, CA, USA) according to the protocol provided by the company, and visualized by ImageQuant LAS-4000 (GE Healthcare, Little Chalfont, UK). The signals of the specific bands were normalized with the densities of the corresponding band of endogenous protein β-Actin. Therefore, the normalized data were expressed as a protein/β-actin ratio. Densitometric analysis of the acquired signals was carried out, using the Image Studio Lite Software (LI-COR Biosciences (RRID:SCR_014211, Li-Cor, http://www.licor.com/bio/products/software/image_studio_lite/). With regard to CB1R immunoblotting, protein extracts were mixed with denaturing 4 × Laemmli loading buffer and warmed for 30 min at 37 °C. Samples (18–24 µg per lane) were analyzed on 4–20% precast polyacrylamide gels (Bio-Rad, Hercules, California) and transferred onto PVDF membranes 0.45 µm (Merk Millipore, Billerica, MA). Membranes were blocked in a mixture of Tris-buffered saline and polysorbate 20 (20 mM Tris–HCl pH 7.6, 150 mM NaCl, 0.05% Tween 20) containing 5% of non-fat milk for 1 h at RT. The membrane was incubated at RT for 1 h using antibodies rabbit anti-CB1R (CB1, ab23703; 1:200, Abcam, Cambridge, UK), mouse anti-alpha tubulin (used as loading control) (sc69969; 1:5000, Santa Cruz, Dallas, US). After the transfer step, the membranes were incubated for 1 h with the primary antibody anti-CB1R. After stripping with Restore Western Blot Stripping Buffer, membranes were re-blotted with a mouse monoclonal anti-alpha-tubulin antibody for normalization. Bound primary antibodies were detected with HRP-linked antibodies (1:2000, Cell Signaling Technology, Danvers, MA) and visualized by enhanced chemiluminescence detection (Clarity Western ECL Substrate, Bio-Rad, Hercules, California). The optical densities of immunoreactive bands were quantified by the Image Lab software (Bio-Rad, Hercules, California) after acquisition on ChemiDoc Touch (Bio-Rad, Hercules, California, US).
All data are presented as mean ± SEM. Data were tested for normal distribution using Shapiro–Wilk’s test. Non-parametric test (i.e., Mann–Whitney U-test) was chosen when data were found not to be normally distributed. For IVSA studies, the response rates exhibited during each IVSA phase (Acquisition, Extinction, Reacquisition) schedule (FR1-FR3), and JWH-018 dose tested (2.5–15 µg/kg/25 µl infusion) were analyzed separately by two-way repeated measures (RM) ANOVA with response (i.e., active or inactive lever presses), session and/or schedule (mean of the last 3 FR3 sessions, PR) as factors, followed by Sidak’s multiple comparisons. For RM tests, whenever we could not assume sphericity, a Geisser-Greenhouse correction was carried out by GraphPad Prism 8 software (GraphPad Prism). Considering JWH-018 intake, the data (mean ± SEM of JWH-018 consumption at each dose during 2 h of IVSA sessions) were analyzed by RM one-way ANOVA followed by Sidak’s post hoc test. The same statistical analysis was used to assess the effects of CBRs blockade on IVSA. For neurochemical assays, to assess the effect of JWH-018 IVSA on cytokine levels and on GFAP and IBA-IR, data were analyzed by using two-way ANOVA, with treatment and brain areas as factors, followed by Tukey’s multiple comparison. Then, the effect of IVSA on cytokines, GFAP, and IBA-1-IR within each brain area was analyzed by Student’s t-test. For behavioral and Western Blot experiments, the data were analyzed by using Student’s t-test. Differences were considered significant at p < 0.05. Effect sizes were calculated by using Cohen’d or Hedges’g when sample sizes were equal or not equal, respectively. Post hoc tests were conducted only when a significant main effect and/or interaction were detected. All analyses were performed using the GraphPad software package (Prism, version 8; GraphPad, San Diego, California, USA).