Animals

Adult male (n = 4) and female (n = 8) Wistar Kyoto rats were purchased from Charles River (Charles River Laboratories Italia s.r.l., Milan, Italy). Rats were housed in standard conditions at 22 ± 2 °C, under a 12:12 light-dark cycle (lights on at 08:00 AM). Food (standard chow diet, VRF1, SDS -Charles River Laboratories) and water were provided ad libitum throughout the study. One male and two female rats (3-month-old) were housed together to achieve a successful mating. Obtained pups (37 males and 41 females) were sexed and weaned at postnatal day 28 (PND28) and were then housed into separated cages containing 4 same-sex rats.

Experimental groups

After the selection phase (described below), selected rats (n = 30 males / 34 females) at PND40 were randomly divided into the following experimental groups:

Control males (CON-M, n = 12);

Control females (CON-F, n = 13);

Male rats subjected to GD protocol (GD-M, n = 18);

Female rats subjected to GD protocol (GD-F, n = 21).

The control groups underwent a 10 min/day period of adaptation to the apparatus, one week before the beginning of the testing phase. The GD group underwent the pre-training and training phase as described below.

Animal care and handling were according to the European Union Council Directive of 22nd September 2010 (2010/63/UE); all the procedures reported in the present study were approved by the Italian Ministry of Health (authorization n°1035/2020-PR). The experimental design conforms to the ARRIVE guidelines originally published by Kilkenny et al. in 2010 (Kilkenny et al., 2010).

GD protocol

A schematic representation of the experimental timeline is reported in Fig. 1A.

Fig. 1

Experimental timeline. Schematic representations of ((A) experimental procedures’ timeline, (B) training and different stimuli during (C-F) Test 1 and Test 5 (videogame alone), (D) Test 2 (videogame vs. new object), and (E) Test 3 (videogame vs. sexual stimulus) and Test 4 (videogame vs. social stimulus). PND = Postnatal day

Behavioral apparatus

The GD protocol took place in a rectangular planned apparatus (50 × 55 × 50 cm) with a plastic base and plywood walls (Fig. 2). The apparatus was positioned 50 cm above the floor in a room illuminated with medium-low light using a house light, tempered to 22 ± 2 °C, and with a camera (Basler GenICam, acA 1300-60 gm) fixed on the roof to register (Media Recorder, Noldus, Wageningen, the Netherlands) the sessions.

Fig. 2

Behavioral apparatus. Photos of (A) the apparatus with (B-E) a graphic representation of the different zones (play zone, other zone, and social zone). The apparatus has (B) a movable panel that was replaced with (D) an open one during tests where social interaction is planned (Test 3–4). The red rectangle represents the play zone, the blue rectangle represents the social zone, and finally, the white rectangle represents the remaining zone of the arena

In the center of the apparatus, there was a removable panel that allowed it to be divided into two areas: on the one hand, the play area (50 × 55 × 50 cm), on the other one (20 × 50 × 50 cm) an area in which different competitive stimuli were placed during the test 2, 3 and 4 (detailed below).

In the play area, a 50 × 30 cm touchscreen tablet (developed by UX experts from SCAI DOO.IT Group, Turin, Italy) was fixed on the shorter wall of the apparatus, with which the rats could interact freely. The area in front of the touchscreen tablet was defined as the play area (5 × 30 cm) (Fig. 2).

The game involved the fixed or random appearance of a white dot (4.5 cm in diameter) on the black screen. When the animal touched the dot correctly, it got a reward of 0.5 g of yogurt (Strawberry yogurt – no added sugar). Yogurt delivery followed a fixed ratio (FR) for the first 3 weeks of the training phase, and then it moved to a random ratio (RR) for the last 2 weeks of the training and for the Test phase.

Selection of animals and pre-training phase

Starting from PND40, rats underwent the selection, pre-training, and training phases.

All animals participated in the selection phase. From PND40 to PND45, the spontaneous behavior of the animals was observed for 20 min in their home cages after 1 h of habituation in the behavior room. Exploratory parameters were used to select rats having showed high levels of anxiety toward a new environment (room in which they would perform the training period and the test period) or toward the operator. The assessment consisted of evaluating exploratory ability (movement inside the cage), fear of the operator, behaviors implemented to get the yogurt (leaning out of the cage) and freezing (Palanza 2001). The daily score value was assigned according to the percentage of time during which the rat performed that type of behavior (0 (< 20%); 1 (> 20/<40); 2 (> 40/<60); 3 (> 60/<80); 4 (> 80/<100). 17 was defined as the mean score obtained from the weekly average to all subjects. Animals with a five-day mean score of less than 17 were excluded from the study (M = 2, F = 3). All the other animals were randomly divided into the following groups: controls (M = 12, F = 13), which were not subjected to the next pre-training and training phases, and animals subjected to GD protocol (M = 18, F = 21).

In the pre-training phase, from PND48 to PND52, the animals underwent a period of adaptation to the apparatus of 5 min per day. The spontaneous behaviors of the animals were also observed in this phase. As before, the assessment consisted of evaluating exploratory ability, fear of the operator, behaviors implemented to get the yogurt, and freezing, assigning to each behavior a daily score (0 = never, 1 = little, 2 = enough; 3 = a lot; 4 = always). Animals that were frightened by the apparatus explored little or nothing and did not come close to the reward and displaying a five-day mean score of less than 17 were excluded from the experiment (M = 5, F = 4).

Training phase

The training phase lasted 5 weeks (from PND55 to PND87), 5 days per week. The sessions took place always between 8:00 a.m. and 1:00 p.m. The rats were placed in the room with soft light (20 lx) 1 h before the beginning of the test.

In the first training week, during the first 3 days, the tablet was in OFF mode. On the last two days of the first week, the tablet was switched to ON mode with the videogame running, and the reward was dispensed with FR1 1:1 (1 correct touch = 1 reward). In the second week, the reward was dispensed with FR2 2:1, whereas for week 3 dispensing was with FR3 3:1. In the 4th and 5th week of training, the reward was dispensed in RR (random ratio).

Each day, the rats were evaluated for the different game-related parameters: time spent in front of the screen, interaction with the screen (both correct touch or not), following the dot, undertraining the connection between correct touch and reward, the number of touches made, and reward obtained.

A form was filled daily with reached scores in all evaluated parameters for each rat. In this form were indicated the different behaviors related to the game. Depending on the number of correct touches, a different score was assigned to each game-related behavior performed by the animal (time spent in front of the screen, interaction with the screen, following the dot, understanding the dot-reward link). The score assigned to each behavior (from the minimum of 1 to the maximum of 5) was related to the number of correct touches (0 = 0 correct touches; 1 = 1–5 correct touches; 2 = 6–10 correct touches; 3 = 11–15 correct touches; 4 = 16–20 correct touches; 5 ≥ 20 correct touches). The daily score was the sum of the scores obtained in each behavior plus the sum of the correct touches. In this way, it was also possible to evaluate the daily progression of the performance. Last, each week, a mean weekly score was assigned to each animal.

Test

The test phase, lasting 6 consecutive days, from PND90 to PND95, aimed to understand if the animals have developed game addiction, testing their ability and interest in the videogame even in the absence of reward and in the presence of new stimuli (exploratory, sexual, or social). Thus, the tests were performed as follows (Fig. 1A):

- PND90: Test 1

The tablet was in ON mode with the videogame running (Fig. 1C). The animal was allowed to choose whether to interact with the videogame to obtain the reward or to explore the apparatus. Tested animal was assessed for attachment (time spent in the play zone), duration (time spent interacting with the video game), and loss of control (number of correct touches, speed, and distance traveled in the play zone) shown during the play session.

- PND91: Day of adaptation to the new object (wheel)

The animal could explore the apparatus in the presence of the wheel, a hitherto unknown stimulus, with the tablet in OFF mode.

- PND92: Test 2

The condition was similar to the previous day (the day of adaptation to the wheel), but the tablet was switched ON with the videogame running. The animal was free to choose whether to play or to interact with the stimulus, and the same parameters of Test 1 were evaluated with the addition of all those behaviors expressed towards the wheel (exploration time and duration of the run on the wheel) (Fig. 1D).

- PND93: Test 3

The tablet was in ON mode with the videogame running. In addition, an unknown animal of the opposite sex (socio-sexual stimulus) was present in the apparatus (Fig. 1E). The animal was able to choose whether to interact with the video game or with the sexual stimulus. The time spent in the sexual zone (zone adjacent to the sexual stimulus), number, and total duration of sniffing, in addition to the same parameters evaluated in Test 1, were recorded.

PND94: Test 4

The test condition was similar to test 3, but a co-specific of the same sex (social stimulus) was present in the apparatus (Fig. 1E). The parameters assessed were the same as in Test 3.

PND95: Test 5

Differently from the previous tests, the animal was evaluated as in Test 1 but did not receive a reward, when it made the correct touches (Fig. 1F).

All females (both tester and no-tester ones) were tested during the estrus phase of the ovarian cycle in Test 3 or 4, when social interaction with a conspecific was planned. Thus, if the animal was not in the estrus phase, the test was postponed until the next day. The estrus phase was assessed by vaginal smear (Cora et al. 2015).

All tests lasted 10 min and were recorded using the camera attached to the roof at a distance of 2.5 m from the apparatus. The reward was dispensed in RR.

The following parameters were evaluated during the tests and were subsequently analyzed with EthoVision 8 (Noldus Information Technology; Noldus, Spink, & Tegelenbosch, 2001):

Time (s) spent in the different zones of the arena (play zone and arena);

Distance (cm) traveled (in different zones or total arena);

Speed (cm/s) achieved (in different zones or total arena);

Game-related behaviors: interaction with games (cumulative duration), correct touches (i.e., number of goals; frequency);

Non-game-related behaviors: grooming (cumulative duration), protected rearing (frequency), and unprotected rearing (frequency);

Wheel-related behaviors: time (s) spent exploring the wheel, time (s) spent running on the wheel;

Behaviors related to socio-sexual or social interaction: time (s) spent sniffing.

Fixation and tissue sampling

At PND115, a subset of rats, randomly selected (5 control males, 5 control females, 5 GD males, and 5 GD females), underwent a 10-minute play session were sacrificed 90 min later by deep irreversible anesthesia (intraperitoneal injection of Zoletil 100 mg/kg - Rompum 20 mg/kg) and transcardially perfused with 0.9% NaCl and then with 4% paraformaldehyde (PFA) solution (Tronel and Sara 2002). Females were sacrificed in the estrus phase, assessed by vaginal smears (Alboni et al. 2017; Cora et al. 2015; Di Micioni et al. 2017).

Brains were removed and stored in 4% PFA solution for 24 h, followed by several washes in 0.01 M phosphate-buffered saline (PBS). Finally, they were stored in 30% sucrose solution in 0.01 M PBS at + 4 °C, frozen in pre-cooled isopentane on dry ice at -35 °C and stored in a deep freezer at -80 °C until sectioning.

Brains (n = 5/group) were serially cut in the coronal plane at 30 μm thickness using a cryostat in three series. The sectioning plane was oriented to match the corresponding patterns to the coronal sections of the rat brain atlas (Paxinos and Watson 1998). Sections were collected in a cryoprotective solution and stored at -20 °C.

c-Fos immunohistochemistry

The presence of c-Fos was detected by immunohistochemistry performed on free-floating sections from one series. Briefly, the sections were washed overnight in 0.01 M PBS at pH 7.3. The following day, sections were first incubated with a citrate buffer (citric acid 10 mM, 0.05% Tween, pH 6.0) previously heated at 95 °C for antigen retrieval and then washed three times in 0.01 M PBS. Next, the sections were washed in 0.01 M PBS containing 0.5% Triton X-100 for 30 min and then treated to inhibit endogenous peroxidase activity with a solution of 0.01 M PBS containing methanol/hydrogen peroxide for 20 min. Sections were incubated for 30 min with blocking solution containing normal goat serum (Vector Laboratories, Burlingame, CA, USA) and bovine serum albumin (Sigma-Aldrich, St. Louis, Missouri, USA) diluted in 0.01 M PBS containing 0.2% Triton, and then incubated two overnight at + 4 °C with polyclonal anti-c-Fos antibody (Cell Signaling Technology, Danvers, Massachusetts, USA; 9F6, Cat. #2250; Rabbit, 1:3.000) diluted in blocking solution. A biotinylated goat anti-rabbit secondary antibody (Vector Laboratories, Burlingame, CA, USA) diluted in 0.01 M PBS, pH 7.3–7.4, containing 0.2% Triton X-100 was then used at a dilution of 1:200 for 60 min at room temperature. The antigen-antibody reaction was revealed by 60 min incubation with avidin–peroxidase complex (Vectastain ABC Kit Elite, Vector Laboratories, Burlingame, CA, USA). The peroxidase activity was visualized with a solution containing 0.400 mg/ml 3,3-diamino-benzidine (Sigma-Aldrich, Milan, Italy) and 0.004% hydrogen peroxide in 0.05 M Tris–HCl buffer at pH 7.6. Sections were mounted on chromallum-coated slides, air-dried, cleared in xylene, and cover-slipped with New-Entellan mounting medium (Merck, Milano, Italy). This antibody was successfully used in previous studies (Cho et al. 2020; Netser et al. 2020; Zhou and Jia 2021). The specificity of this antiserum was previously assessed (Kovary and Bravo 1991), but, as a further control, we omitted the primary antiserum or the secondary biotinylated one and replaced it with 0.01 M PBS. Positive cell bodies were totally absent.

c-Fos quantitative analysis

For quantitative analysis, selected standardized sections of brain areas summarized in Supplementary Table 1, were chosen according to the rat brain atlas (Paxinos and Watson 1998). Sections of comparable levels (see Supplementary Table 1) for each nucleus were acquired with Slide-Scanner Axioscan Z1 (ZEISS, Oberkochen, DE) both at low and high magnification (5x and 20x, respectively). Digital images were processed and analyzed by ImageJ (version 2.10/1.53c; Wayne Rasband, NIH, Bethesda, MD, USA). Measurements were performed within predetermined fields (region of interest, ROI), boxes of fixed size and shape that are inserted inside each labeled considered nucleus (see Supplementary Table 1). In particular, we counted the number of c-Fos-positive cells in all analyzed nuclei.

Statistical analysis

Quantitative behavioral data was analyzed with SPSS 27 statistic software (SPSS Inc, Chicago, IL, USA) via two-way analysis of variance (ANOVA), with sex and condition (GD vs. CON) considered as independent variables. To understand whether the behavior of the GD groups was maintained across the different tests, we compared the behavioral parameters performed in Test 1 with the other tests, using three-way ANOVA (test, sex, and condition as independent variables) for repeated measures.

For the data from the last week of training, we used two-way ANOVA (day and sex as independent variables) for repeated measures to compare the behaviors of the GD groups on different days.

Either two-way ANOVA (as described for behaviors) or nested one-way (nested ANOVA) was used to quantify immunohistochemical data, when there were multiple data from the same brain area of the same animal. Using the nested ANOVA allowed us to account for within-sample variability (between observations of the same animal) and within- and between-group variability: group (CON vs. GD) was considered a fixed factor, while animals and observations (cells/sections) were considered random factors; sample sizes (N of animals) in the groups and N of observations per sample were kept the same. If the ANOVA was significant, the post hoc analysis was performed using Tuckey’s HSD (honestly significant difference) test.

Finally, to find out whether there was a significant correlation between time spent in the play zone and other behaviors (the speed and distance traveled in the play zone, duration of interaction with the tablet, and number of correct touches) during training and in Test 1, we used Person R Correlation.

The data are presented as mean ± SEM, and the differences between groups are considered significant for values of p < 0.05.