Social Experiences

This phase comprised five encounters with a conspecific at an interval of 6 days (PND 73, 79, 85, 91, 97). Specifically, subjects were confronted each time either with an unfamiliar dominant male (adverse experience, AA group, see “Repeated Interaction with Dominant Opponent” section) or an unfamiliar female in pro-estrus or estrus (beneficial experience, BA and BB groups, see “Repeated Interaction with a Receptive Female” section). Encounters took place in the animals’ housing room during the dark phase under red light conditions. A transparent plastic cover was placed on the cage in which the interaction took place, allowing observation and at the same time preventing the mice from jumping out. While females and dominant males were re-used as interaction partners (maximum once per day), each experimental animal met each female/dominant male only once during the five encounters.

The experimental male was placed in the home cage (Makrolon type III) of an unfamiliar adult male of the aggressive NMRI strain (Navarro, 1997). A “losing experience” (see Jansen et al., 2010) was confirmed by an experienced observer (NK). The confrontation lasted for a maximum of 10 min, but was stopped early in cases of high aggression to prevent the animals from injury. The average number of attacks by the NMRI male during these confrontations was 4.3 ± 0.9 per minute (mean ± SD).

An unfamiliar adult C57BL/6J female in pro-estrus or estrus was placed in the home cage of the experimental male and the animals could interact for 10 min. A socio-positive or sexual experience was confirmed by an experienced observer (NK). Estrous states of the females were assessed by microscopically examining vaginal smears, and the decision was made on the basis of the cytology (as described in Allen, 1922; Nelson et al., 1982; Byers et al., 2012; Kästner et al., 2017).

Throughout this phase (PND 107–PND 114), subjects were housed in special custom-made cage systems that comprised an interaction cage (Makrolon type III, with bedding and enrichment; food and water ad libitum), a refuge cage (Makrolon type II, with bedding and enrichment; food and water ad libitum) and a water basin (Makrolon type II, filled with water to a height of app. 3 cm; Meyer et al., 2016; see Figure ​Figure2).2). The water basin was connected to the interaction cage as well as to the refuge cage via plastic tunnels. Thus, to get from the interaction cage into the refuge cage, the water basin had to be crossed. Handling (as described in “Escapable Social Defeat” and “Living with a Female” sections) took place on PND 107, 108, 111, 112 and 113 between 8:30 a.m and 11:00 a.m, once per day for each experimental animal.

Cage system used in phase 2. The custom-made cage systems comprised an interaction cage (Makrolon type III with bedding and enrichment), a refuge cage (Makrolon type II with bedding and enrichment) and a water basin (Makrolon type II, filled with water to a height of approximately 3 cm; Figure modified after Meyer et al., 2016).

One day before the beginning of this phase, mice of all three groups were placed individually into the interaction cage of one cage system, while there was no water in the water basin. It was made sure that each individual had explored the whole set-up, including the refuge cage, before on the next day the first handling took place. Before the first handling, bedding and enrichment were renewed in each set up and either an aggressive NMRI male (adverse experience, AA and BA groups, “Escapable Social Defeat” section) or a female in pro-estrus or estrus (beneficial experience, BB group, “Living with a Female” section) were placed into the interaction cage of the setup, where they stayed throughout this phase.

The experimental animal was placed into the interaction cage, i.e., to the NMRI male. As a response to the attacks of the NMRI mouse, it escaped via the water basin into the refuge cage. To prevent the NMRI mouse from following in any case, the tunnels were sealed by plastic inserts. These were only removed before the next handling on the next day, when the experimental animal was taken out of the refuge cage and again introduced into the interaction cage. Escape latencies were measured in minute intervals. Median escape latencies were below 3 min on the first day (PND 107) and below 1 min on the following days.

The experimental animal was placed into the interaction cage, i.e., to the C57BL/6J female, where it stayed until the end of this phase. To correct for a possible handling effect, experimental animals of this condition were sham-handled on handling days. During these occasions they were given the possibility to escape via the water basin, what, however, never occurred.