Wild-type (WT) male C57BL/6J mice (8–12 weeks old) were purchased from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). Tau^-/- knockout (KO) mice were obtained from Jackson Laboratories (#007251, United States), and male Tau^-/- mice were used in our study. All mice were housed in an animal facility on a standard 12-h light/12-h dark cycle (lights on from 7:00 A.M. to 7:00 P.M.) with ad libitum access to food and water at room temperature (22–28°C). All experimental procedures and animal protocols were performed in accordance with guidelines from the Institutional Animal Care and Use Committee of West China Hospital of Sichuan University (20211398A). All efforts were made to minimize the suffering of the mice.

Cocaine was obtained from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China) and dissolved in saline. PP2 (S7008, Selleck) was dissolved in vehicle (saline containing 1% dimethylsulfoxide).

The CPP paradigm was performed using a standard three-chambered apparatus comprising two large conditioning compartments (black and white) with different floors (bar and grid floors) and a small middle chamber (grey, smooth PVC floor) connecting the two large compartments. Before each session, the animals were habituated to the chambers for at least 10 min on 2 consecutive days. During the pretest session, mice were placed in the chamber and provided free access to all chambers for 15 min, and the time spent in each chamber was recorded. After the pretest, animals that represented a strong unconditioned preference for either chamber (chamber bias, >300 s) were excluded, and about 50 animals were excluded from the CPP experiments in our study. The remain animals were randomly assigned to two groups, one group that received alternating injections of cocaine [20 mg/kg, intraperitoneal (i.p.)] and saline and the other group that received saline injections in both compartments, and trained for 6 days. After the injection, animals were trained in the conditioning or nonconditioning chambers for 30 min and then returned to their home cages. On the test day, the mice were placed in the neutral chamber and allowed to explore both compartments for 15 min. The time spent in each chamber was measured to calculate the CPP score. CPP scores are recorded as the time spent in the cocaine-paired chamber minus the time spent in the saline-paired chamber.

Mice were anaesthetized with sodium pentobarbital (60 mg/kg), and a single sterilized silastic catheter (0.51 ID x 0.94 mm OD, BB518-20, Scientific Commodities) was implanted into the right jugular vein. The skin on the animal’s back was attached to the distal end of the catheter via a stainless steel guide cannula (RWD Life Science). After surgery, mice received 0.1 ml of a saline solution containing penicillin (160000/ml) and heparin (30 U/ml) via the catheters daily. Following 1 week of recovery, the animals were trained to intravenously self-administer cocaine (0.75 mg/kg/infusion) or saline in daily 2 h sessions over 10 days on an FR1 schedule in an operant chamber. Active pokes resulted in the injection of cocaine accompanied by the presentation of blue light for 20 s and an audible tone for a 5 s timeout period. Inactive pokes failed to elicit drug injection and presentation of the conditioning stimuli.

Locomotor activity was recorded as the distance travelled. The animals were acclimated to chambers (48 cm × 48 cm) for 10 min on 2 consecutive days. Baseline locomotor activity was not significantly different between groups. During the injection course, animals were administered cocaine (20 mg/kg, i.p.) or an equal volume of saline and immediately placed in the chamber for 15 min. The distance travelled was recorded daily for 7 consecutive days, and automated tracking and recording were performed with EthoVision 7.0 software (EthoVision 7.0; Noldus Information Technology, Leesburg, VA).

Y-mazes are often used to assess spatial memory. The experiment was performed using a maze shaped like the letter “Y”, with a starting arm and two target arms. The luminance of overhead white lighting is 40 lux and numerous visual cues were placed on the wall of the testing room. On the test day, mice were put at the end of one arm (designated as the start arm), and were trained to leave the start arm and enter the right arm (familiar arm), not the left arm (novel arm) for 10 min. After 120 min inter-trial interval, the novel arm was opened, the mice were allowed to explore all three arms. The choices of novel and familiar arms were alternate between tests of different subjects, and the time spent in the novel and familiar arms during the 5-min test was automatically recorded by the software (ANYmaze, United States).

Animals were placed in a test chamber (48 cm × 48 cm) for 10 min on the first day (without objects). On the second day, mice were placed in the test chamber for 10 min (with two identical objects). On the third day, the familiar object was shifted to a diagonal position (different position but the same size and colour as the familiar object). The animals were placed again in the area and allowed to explore the object at the novel position for 10 min. The time spent exploring the object was recorded by EthoVision 7.0 software (EthoVision 7.0; Noldus Information Technology, Leesburg, VA).

The mice were placed in a test chamber (48 cm × 48 cm) for 10 min on the first day (without objects). On the second day, mice were placed in the test chamber for 10 min (with two identical objects). On the third day, one of the familiar objects was replaced with a novel object (different size and colour from the familiar object). Animals were again placed in the area and allowed to explore the novel object for 10 min. The time spent exploring the object was recorded by EthoVision 7.0 software (EthoVision 7.0; Noldus Information Technology, Leesburg, VA).

Animals were anaesthetized with sodium pentobarbital (60 mg/kg) and mounted on a standard stereotaxic frame (RWD Life Science). The hair of each mouse was shaved, the incision site was cleaned with medical alcohol, the scalp was incised, the skull was exposed, and the hippocampus (AP, −1.7 mm; ML, ±1.2 mm; DV, −1.5 mm) was bilaterally implanted with permanent guide cannulas (RWD Life Science) using a stereotaxic instrument. The guide cannula was anchored with dental cement, and a stainless-steel stylet blocker was inserted into each cannula to prevent blockage and infection. All mice were subjected to training after 1 week of recovery.

PP2, an ATP-competitive Src kinase inhibitor, was dissolved in saline containing 1% DMSO at a final concentration of 62.5 ng/μL. PP2 was injected bilaterally (1 μL/side, 0.5 μL/min) with a microinjector 15 min before cocaine or saline administration.

LV-pClenti-hSyn-EGFP-3xFLAG-WPRE was purchased from Obio Technology Co., Ltd. (Shanghai, China). The Fyn shRNA was cloned into pClenti-hSyn-EGFP-3xFLAG-WPRE and confirmed by sequencing. The sequences of the scrambled control shRNA and Fyn shRNA were 5′-TTC​TCC​GAA​CGT​GTC​ACG​T-3′ and 5′-CCC​AAG​AGG​TAC​CTT​TCT​T-3′, respectively. Recombinant lentiviruses were produced by transient transfection in HEK293T cells, and then the level of Fyn was analysed using western blot.

Mice (8–12 weeks old) were anaesthetized with sodium pentobarbital (60 mg/kg) and placed on a stereotaxic apparatus (RWD Life Science) to inject the virus into the dorsal hippocampus (AP, −1.7 mm; ML, ±1.2 mm; DV, −1.5 mm). After shaving the hair and cleaning the incision site with medical grade alcohol, the scalp was incised to expose the skull, and the connective tissue was gently removed from the skull surface with cotton swabs. Small craniotomy holes were drilled with a skull rotor (RWD Life Science) for virus injection. Microsyringe needles were used to bilaterally infuse the hippocampal tissue with 1 μL of virus at a rate of 0.1 μL/min. After each injection, the syringe was left in place for an additional 5 min and then slowly withdrawn to allow the virus to diffuse. Mice recovered for at least 1 week before behavioural testing.

Hippocampal tissues were lysed, and proteins were extracted using a protein extraction kit (K269-500, Biovision). A Bradford assay (P0006, Beyotime) was used to determine the total protein concentration. Twenty micrograms of protein were loaded and separated on a 10% sodium dodecyl sulfate-polyacrylamide gel and then transferred to a polyvinylidene difluoride (PVDF) membrane (IPVH00010, Millipore) in a mixture of Tris-glycine buffer and 20% (v/v) methanol. The membrane was blocked with 5% non-fat dry milk for 1 h and then incubated with the primary antibody with gentle shaking overnight at 4°C. The next day, after three washes, the blots were incubated with the corresponding secondary antibody for 2 h at room temperature. Protein levels were visualized using a chemiluminescence substrate (WBKLS0500, Millipore) and a chemiluminescence imaging system (CLINX, Shanghai, China). Chemi Analysis software was used to quantify the optical density of each band (CLINX, Shanghai, China). The following antibodies were used for Western blot: mouse anti-Tau (1:1,000, Thermo Scientific), rabbit anti-Fyn (1:1,000, Abcam), rabbit anti-p-Fyn (T 416) (1:1,000, Cell Signaling Technology), rabbit anti-β-actin (1:1,000, Cell Signaling Technology), HRP-conjugated goat anti-mouse (SAB), and HRP-conjugated goat anti-rabbit (SAB).

The cannula implantation site was confirmed by performing HE staining of 5-μm thick coronal sections, and images were captured using a light microscope (Figure 1A). Mice with misplaced cannulas were excluded from the statistical analysis.

Single-cell fluorescent labelling was performed as previously described with slight modifications (Luo et al., 2016). Briefly, mice (8–12 weeks old) were anaesthetized with sodium pentobarbital (60 mg/kg), placed on a stereotaxic apparatus (RWD Life Science), and the dentate gyrus received a virus injection (AP, −2.0 mm; ML, ±1.4 mm; DV, −2.2 mm). After their hair was shaved, medical-grade alcohol was used to clean the incision site, the skull was exposed via a scalp incision, and cotton swabs were used to gently remove the connective tissue from the skull surface. Small craniotomy holes were drilled with a skull drill (RWD Life Science) for virus injection. Microsyringes were used to bilaterally infuse a mixture of adeno-associated viruses (AAVs) [pAAV-Syn-DIO-(tTA-P2A-mNeonGreen)-WPRE (Obio Technology Co., Ltd., Shanghai, China) and pAAV-PTRE-tight-NLS-Cre (Obio Technology Co., Ltd., Shanghai, China)] into the brain at a rate of 0.05 μL/min for 10 min (0.5 μL per side). After each injection, the syringe was left in place for an additional 5 min and then slowly withdrawn to allow diffusion of the virus. The mice were allowed to recover for at least 3 weeks before analysing hippocampal neuron morphology.

After 3 weeks, the animals that had received the AAV injection were deeply anaesthetized with sodium pentobarbital (60 mg/kg) and perfused transcardially with phosphate-buffered saline (PBS) followed by ice-cold 4% paraformaldehyde in 0.1 M PBS (pH 7.4). We carefully extracted the brains from the skull, postfixed them with 4% PFA overnight, and then dehydrated them in 30% sucrose at 4°C. We sectioned the brains into 50 μm coronal slices using a freezing microtome (Leica, Germany), mounted them on slides, dried them and stored them at −80°C until processing for immunohistochemistry.

Sections were washed with PBS 3 times (10 min each time) and covered with anti-fade mounting medium with DAPI (H-1200, Vector). A laser confocal microscope (Nikon, Japan) was used to acquire brain images. Each neuron was scanned at high magnification (100X, oil immersion lens) to ensure that all parts of the dendrites were intact. A minimum of 3 neurons per slice from each group were examined, and at least 40 neurons were selected from each group. Confocal microscopy was performed for the 3D reconstruction of neurons. The total dendritic length and dendritic spine density were measured using ImageJ software (US National Institutes of Health, Bethesda, MD, United States).

Mice were deeply anaesthetized with sodium pentobarbital (60 mg/kg) by i.p. injection and 50 ml of freshly prepared fixative containing 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M PBS (pH 7.4) were perfused intracardially. Whole brains were removed, postfixed with 2.5% glutaraldehyde overnight at 4°C, impregnated with 1% osmium tetroxide for 1 h, dehydrated in graded alcohol solutions, flat embedded in Durcupan ACM (Fluka) and cured for 48 h at 60°C. Small pieces containing the hippocampal dentate gyrus were removed from the specimens and glued on a plastic block with cyanoacrylate. Ultrathin sections were cut and mounted on Formvar-coated single-slot grids. A transmission electron microscope (JEOL, Japan) was used to observe the synapse structure. We obtained 5 images of each section, yielding at least 50–70 synapses from each mouse. Image-Pro Plus 6.0 software was used for the morphometric analysis.

All data were analysed with GraphPad Prism 7 software and are presented as the means ± SEMs. The normality of the data distribution was analysed using the Kolmogorov–Smirnov test. An unpaired two-tailed Student’s t test was used for simple comparisons. One-way or two-way ANOVA followed by the Bonferroni post hoc test were utilized for multiple comparisons. In all results, n refers to the animal number. For all results, statistical significance was set to p < 0.05.

Previous studies have suggested that PP2, a Src kinase Fyn inhibitor, is able to inhibit Fyn activity and alter Tau expression in individuals with Alzheimer’s disease (AD) (Chen et al., 2013; Li and Gotz, 2017). We explored whether inhibition of Fyn activity suppresses cocaine-associated memory formation and modulates Tau expression. The CPP paradigm, a widely used protocol for assessing drug-associated memory formation (Sanchis-Segura and Spanagel, 2006), was used to determine the role of Fyn in cocaine-associated memory formation. The timeline of the cocaine CPP procedure is shown in Figure 1B. During habituation to the CPP apparatus, neither group showed a side preference. In the training phase, cocaine induced a significant increase in the CPP score, and pre-administration of PP2 into the hippocampus significantly decreased the cocaine-CPP score [F _{(3, 32)} = 4.795, p < 0.01, two-way ANOVA, Figure 1C]. Injection of PP2 significantly reduced the level of phosphorylated Fyn, accompanied by an increase in Tau expression after cocaine-CPP training [p-Fyn: F _{(3, 20)} = 10.05, p < 0.01, one-way ANOVA; Tau: F _{(3, 20)} = 4.354, p < 0.05, one-way ANOVA, Figure 1D]. In addition, to assess whether PP2 altered the working and spatial memory of mice after cocaine-associated memory formation, we also assessed short-term working memory and short-term spatial and recognition memory using the Y-maze, novel object position and recognition test after cocaine-CPP training, respectively. No significant PP2-induced impairment in the spontaneous alternation behaviour was observed in the Y-maze tests (Figure 1E). Compared with control groups, PP2-treated mice displayed no preference for the novel location of an object during the novel object position test, indicating no deficit in the spatial memory after PP2 administration (Figure 1F). Similarly, in the novel object recognition test, PP2 did not reduce the exploratory behaviour or preference for novel objects (Figure 1G). Based on these results, pharmacological blockade of Fyn activity impairs cocaine-associated memory and is accompanied by increased Tau expression, rather than affecting short-term working memory and short-term spatial and recognition memory.

We continued to examine the role of Fyn in cocaine self-administration, a paradigm that incorporates memories of the rewarding effects of the drug and drug exposure context. Cocaine-associated memory is formed when an instrumental action (an active poke) results in cocaine injection (the unconditioned stimulus) and is paired with an audiovisual cue (the conditioned stimulus). The mice were trained on an FR1 schedule of reinforcement, in which a single active poke induced an infusion of cocaine or saline. The timeline of the cocaine self-administration procedure is shown in Figure 2A. Compared with saline-treated mice, the cocaine-treated mice showed a marked increase in the number of active pokes. Thus, the cocaine-treated mice exhibited strong and reliable cocaine-associated memory formation in the self-administration paradigm (Figure 2B). Moreover, we also found that the number of active pokes decreased after the administration of PP2 (Figure 2B). In contrast, the numbers of active pokes and inactive pokes were not different between vehicle- and PP2-pretreated mice during saline treatment [active pokes: F _{(27, 360)} = 3.497, p < 0.0001, repeated measures two-way ANOVA, Figures 2B,C]. In addition, the number of cocaine infusions increased during cocaine-associated memory formation, and PP2 pre-treatment decreased the number of cocaine infusions [F _{(27, 324)} = 5.800, p < 0.0001, repeated measures two-way ANOVA, Figure 2D]. Finally, PP2 also obviously decreased the levels of phosphorylated Fyn and reversed the downregulation of Tau expression in the hippocampus of the self-administration paradigm [p-Fyn: F _{(3, 20)} = 5.604, p < 0.01, one-way ANOVA; Tau: F _{(3, 20)} = 4.814, p < 0.05, one-way ANOVA, Figure 2E]. Collectively, our results show that PP2 may reduce cocaine-induced seeking behaviours in a self-administration paradigm.

A locomotor activity assay in which cocaine-associated contextual stimuli were absent was used to measure the psychomotor and psychostimulant effects of cocaine by assessing the distance travelled after drug exposure. The timeline of the cocaine locomotor activity procedure is shown in Figure 3A. Compared to the cocaine group, the PP2 pre-treatment group did not exhibit a significant decrease in distance travelled (Figure 3B). In addition, PP2 did not alter the hippocampal levels of phosphorylated Fyn or Tau during cocaine-induced hyperlocomotion (Figure 3C). Therefore, cocaine-induced hyperlocomotion is not attenuated by PP2, further indicating that the Src kinase Fyn is involved in cocaine-associated memory formation through regulation of Tau levels in the hippocampus.

Previous studies have shown that cocaine-induced structural changes in neurons are involved in cocaine-mediated behavioural adaptations (Cahill et al., 2016). We explored whether PP2 suppresses cocaine memory formation by remodelling dendritic structures and altering synapse numbers by assessing the effect of PP2 on the dendritic structure of hippocampal granule neurons in mice subjected to cocaine-CPP training. We analysed the dendritic complexity of granule neurons in the dorsal dentate gyrus, which generally exhibits adult hippocampal neurogenesis, and promotes memory formation (Aimone et al., 2006), using ImageJ software and the Sholl analysis plugin, as described previously (Xu et al., 2020). In our study, cocaine significantly increased the dendritic complexity, including increasing the dendritic length and spine density (Figures 4A–D). PP2 administration inhibited the cocaine-induced increase in dendritic length and obviously reversed the cocaine-induced increase in spine density [Dendritic length: F _{(3, 8)} = 10.29, p < 0.05, one-way ANOVA; Spine density: F _{(3, 8)} = 7.429, p < 0.05, one-way ANOVA, Figures 4B,C]. In addition, PP2 reduced cocaine-induced dendritic arborization after cocaine-CPP training [F _{(75, 208)} = 2.393, p < 0.0001, repeated measures two-way ANOVA, Figure 4D]. We also performed electron microscopy to analyse synaptic alterations during cocaine-associated memory formation. An analysis of dorsal dentate gyrus synapses in cocaine-treated mice revealed significant increases in the density and length of the postsynaptic density (PSD) after cocaine-CPP training (Figures 4E–G), whereas no alteration in the thickness (a measure of maturation) of the PSD was observed between the cocaine-treated group and the control group (Figure 4H). However, this increase was obviously attenuated by the PP2 injection [PSD density: F _{(3, 8)} = 18.55, p < 0.01, one-way ANOVA; PSD length: F _{(3, 8)} = 15.72, p < 0.01, one-way ANOVA, Figures 4E–G]. Thus, our studies suggest that Fyn activation contributes to cocaine-induced dendritic remodelling of granule neurons and increases the number of synapses in the hippocampus after cocaine-CPP training.

We injected PP2 into the hippocampus of Tau KO mice during CPP training to further confirm whether the role of Fyn in regulating cocaine-associated memory formation depends on Tau expression. The timeline of the cocaine CPP procedure is shown in (Figure 5A). Compared with saline-injected Tau KO mice, cocaine-injected Tau KO mice showed a marked increase in the CPP score (Figure 5B). PP2 pre-treatment failed to affect the cocaine-CPP score of Tau KO mice, although the level of phosphorylated Fyn was decreased [CPP Score: F _{(3, 32)} = 5.762, p < 0.01, two-way ANOVA; p-Fyn: F _{(3, 20)} = 9.356, p < 0.01, one-way ANOVA, Figures 5B,C]. Collectively, these results provide strong evidence that Fyn modulates cocaine memory formation, possibly by depending on Tau expression.

As PP2 is a nonspecific inhibitor of the Src kinase Fyn, we used a genetic approach to further assess the direct contribution of Fyn to cocaine-associated memory formation. We performed shRNA-mediated knockdown of Fyn in the dorsal hippocampus, and this approach ensured specificity for Fyn and limited the manipulation to the dorsal hippocampus. The timeline of the cocaine CPP procedure is shown in Figure 6A. Notably, the Fyn level was significantly decreased after the injection of LV-sh-Fyn [F _{(3, 20)} = 6.663, p < 0.01, one-way ANOVA, Figure 6B]. Compared with the control groups, Fyn knockdown in the dorsal hippocampus substantially attenuated the cocaine-CPP score [F _{(3, 36)} = 8.367, p < 0.01, two-way ANOVA, Figure 6C]. We continued to administrate of PP2 into the hippocampus of Fyn knock-down mice (Figure 6D). Notably, no significant difference was observed between Fyn knockdown and control mice in performance on the Y-maze, novel object position, and recognition test (Figures 6E–G), suggesting that neither PP2 nor shRNA-mediated knockdown of Fyn affected short-term working memory and short-term spatial and recognition memory.

Taken together, our data indicate that systemic pharmacological and dorsal hippocampal genetic inhibition of Fyn strongly and selectively suppresses cocaine-associated memory formation and thus represents a promising option to explore for the development of targeted therapies for cocaine addiction.