The interplay between acute and long-term effects of the genetic makeup related to the social environment and social peers was assessed using rough-and-tumble play and concomitant emission of 50-kHz USV in a group of juvenile female Cacna1c^+/– (HET) rats (N = 40) and Cacna1c^+/+ (WT) littermate controls (N = 40) by systematically manipulating the genotype of cage mates and the genotype of play partners, respectively.

As described in preceding studies (Kisko et al., 2018, 2020), HET rats were generated through zinc finger technology (Geurts et al., 2009) by SAGE Labs (now Envigo, Indianapolis, IN, USA) on a Sprague-Dawley (SD) background. HET rats carry a 4 base pair deletion at position 460 649-460 652 in genomic sequence, resulting in an early stop codon in exon 6. Genotyping was performed using DNA obtained from tail samples on a 3130xl Genetic Analyzer (Thermo Fisher Scientific, Waltham, MA, USA) with the following primers:

50-GCTGCTGAGCCTTTTATTGG-30 (Cacna1c Cel-1 F).

50-CCTCCTGGATAGCTGCTGAC-30 (Cacna1c Cel-1 R).

An established breeding protocol was applied to obtain HET and WT offspring, pairing SD females (Charles River, Sulzfeld, Germany) and male HET rats (Kisko et al., 2018). The day of birth was defined as postnatal day (PND) 0. Breeding was performed at the Faculty of Psychology, Philipps-University of Marburg, Germany.

Rats were identified by paw tattoo, using non-toxic animal tattoo ink (Ketchum permanent tattoo inks green paste, Ketchum Manufacturing Inc., Brockville, Canada). The ink was inserted subcutaneously through a 30-gauge hypodermic needle tip into the center of the paw on PND 3 ± 1. All procedures were conducted in strict accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and the relevant local or national rules and regulations of Germany and were subject to prior authorization by the local government (Tierschutzbehörde, Regierungspräsidium Gießen, Germany).

For our design we manipulated the genotype of the cage mate and the genotype of the play partner to fully assess the interplay between acute and long-term effects of the genetic makeup related to the social environment and social peers on rough-and-tumble play behavior and concomitant 50-kHz USV emission in juvenile Cacna1c haploinsufficient females.

Testing took place between PND 31 and 36. All rats were isolated the day prior to their first play session in a Makrolon type III cage (265 mm × 150 mm × 425 mm, plus high stainless-steel covers; Techniplast Deutschland GmbH) to enhance social motivation. In line with previous experiments (Kisko et al., 2018, 2020), play sessions were initialized by a 2-min anticipation phase, during which the focus rat habituated to the observation arena (35 cm × 35 cm, with Plexiglas walls; floor covered with 1 cm of fresh bedding). The anticipation phase was followed by a 5-min play phase, where focus and partner rats were allowed to socially interact. Before each play pair, the observation arena was cleaned thoroughly with an acetic acid solution (0.1%) and new bedding was added. Testing was conducted under red light conditions (∼28 lux). A digital camera (TK-1281 Color Video Camera, JVC, Yokohama, Japan) was used to record rats for behavioral analyses. Focus rats were marked with a non-toxic commercial marker, to enable the observer to distinguish the rats on the video.

Ultrasonic vocalizations recordings were taken during the play phase using UltraSoundGate Condenser CM16 Microphones connected to an UltraSoundGate 416H USB audio device (Avisoft Bioacoustics) placed 35 cm above the floor of the center of the observation arena. Briefly, acoustic data was recorded with a sampling rate of 250 000 Hz (recording range: 0–125 kHz; 16 bit) by Avisoft RECORDER USGH and transferred to Avisoft SASLab Pro for acoustical analysis. Start of behavioral recording was acoustically labeled by an audible beep signal from a stopwatch to synchronize audio and video recordings.

For comparing rough-and-tumble play behaviors and pro-social 50-kHz USV between genotype of the cage mates and genotype of the partner, analysis of variances (ANOVA) for repeated measurements were calculated with the between-subject factor genotype of the cage mates (H), genotype of the focus rat (G), genotype sequence of the play partner (i.e., WT or HET first) (SQ) and the within-subject factors genotype of the play partner (i.e., same or opposite) (P), play session (SE), rough-and-tumble play component (i.e., pinning, wrestling, chasing and nape attack) (C). Post hoc assessment of significant main effects was calculated using independent or paired t-tests, where appropriate. A significance threshold was set at p < 0.050.

For juvenile WT and HET females, the amount of time spent engaged in rough-and-tumble play averaged across all play sessions was not dependent on the genotype of the cage mates, i.e., housing [PxH: F_(1,32) = 0.000, p = 0.990]. Because the genotype of the cage mate had no effect on play, the following results (Figures 2, 3) depict pooled housing conditions. However, the duration of rough-and-tumble play was dependent on the genotype of the play partner [PxG: F_(1,32) = 4.316, p = 0.046; Figure 2A and Table 2A]. Importantly, the effect of the play partner’s genotype on the time spent playing was strongly modulated by the sequence of the play partner (i.e., WT or HET play partner first).

Within play sessions, the duration of non-play social interactions in WT and HET females depended on the play partner’s sequence [PxSQ: F_(1,32) = 7.909, p = 0.008]. However, with a closer look at the duration of non-play social behavior between pairs playing first with a same- or an opposite-genotype partner, no differences were found and are therefore not reported (Figures 3C–C”). As there were no significant differences in the duration of non-play social behaviors in same-and opposite-genotype pairs apart from being added to the total duration of non-play social interactions, individual components (Table 1) were not analyzed or discussed further. Additionally, the genotype of the cage mate had no effect on non-play social interactions and therefore the following results and representative figures (Figures 3C–C”’) are pooled for housing composition.

In slight contrast to rough-and-tumble play behavior, the total 50-kHz USV emission averaged across all play sessions in WT and HET females was not dependent on the genotype of the play partner [PxG: F_(1,31) = 0.717, p = 0.404, Figure 3A]. Instead, the emission of 50-kHz USV during rough-and-tumble play in same- and opposite-genotype pairs was dependent on the play partner’s sequence [PxSQ: F_(1,31) = 20.460, p < 0.001]. The genetic composition of the cage mates had no effect on the total averaged number of 50-kHz USV emissions across all play sessions. Accordingly, the post-hoc t-test results (Table 3A) and representative figures (Figures 4, 5B–B”’) for 50-kHz USV are pooled housing conditions.

Play behavior is essential to development (Pellis and Pellis, 2009; Vanderschuren and Trezza, 2013; Vanderschuren et al., 2016). Juvenile rats are highly motivated to play and in the present study all pairs engaged in rough-and-tumble play, suggesting that motivation to play was present both in Cacna1c haploinsufficient females and wildtype littermate controls, irrespective of genetic makeup related to the social environment and social peers. However, we saw a reduced amount of time playing in same-genotype HET pairs, particularly if their first play partner was a HET. Interestingly, the time spent playing was rescued to levels equivalent to the same-genotype WT control pairs when a HET female played with a WT play partner regardless of sequence. Notably, this was evident not only in the average duration of play across all play sessions but also within each individual play session and relevant play components, i.e., pinning, wrestling, chasing and nape attacks. The WT females, therefore, can enhance the playful motivation of their HET play partners acutely and long-term, as the effect is persistent when the subsequent play partner is another HET.

The influence of the WT female rats on the HET contrasts our initial hypothesis that the hyper-playful HET females were responsible for increasing the playful motivation of their WT cage mates (Kisko et al., 2020). In rats, play partners and early social environment can affect the overall playful and social characteristics established during the critical period (Varlinskaya et al., 1999; Burgdorf et al., 2005; Himmler S. et al., 2014; Kisko et al., 2015a,b; Burke et al., 2017; Redecker et al., 2019; Stark and Pellis, 2020). Devocalization studies have shown that control rats living with devocalized cage mates spent less time playing than controls housed with other controls (Kisko et al., 2015b). Additionally, when rats are reared with adults or non-playful cage mates as juveniles, they show impairments in socio-cognitive functions (Bell et al., 2010; Baarendse et al., 2013; Himmler S. et al., 2014; Schneider et al., 2016; Pellis et al., 2017; Stark and Pellis, 2020) and are less socially competent when navigating adult social interactions (Kisko et al., 2015a; Stark and Pellis, 2020). In the current study, manipulating the post-weaning social environment had no significant main effects on the time spent playing for pairs of juvenile female rats. While it is important to acknowledge that the post-weaning social environment could have potentially influenced the individual play partners, it would be remiss not to consider its impact. During rough-and-tumble play in situations where it is possible, participants adapt their play behavior to their partner (Varlinskaya et al., 1999; Burgdorf et al., 2005; Himmler S. et al., 2014; Kisko et al., 2015b). In this manner, each partner benefits from the interaction. In adult Cacna1c haploinsufficient rats, we have found that the HET females will adapt their social behavior based on their partner’s genotype (Redecker et al., 2019). Our results indicate that the juvenile females adjust their behavior to suit their play partner’s genotype. However, it would be important to examine each playmate closely, particularly in opposite-genotype pairs. Additionally, the sequence in which the partners genotypes are presented plays a role in determining how playful the pair will be. For instance, HET females appear to be more playful when the first play partner was a WT, while WT females show more subtle effects when playing first with a HET play partner, i.e., spending less time playing only in certain play sessions and within select play components.

In same-genotype HET and opposite-genotype WT pairs, there is an increase in non-play social behaviors when the genotype of the first play partner was a HET. In a study by Varlinskaya et al. (1999), the authors showed that in juvenile rats, the isolated, hyper-playful partner would adapt their behavior if their partner was reluctant to play and instead engage in more non-play-directed social behaviors (Varlinskaya et al., 1999). Therefore, it may be that the HET females are not as motivated to engage in rough-and-tumble play behaviors but rather prefer more non-play social interactions. If this were the case, we would also expect to see no difference in 50-kHz USV emissions, especially during periods of play, compared to non-play social interactions. At this point, however, we cannot say whether the HET individual drives the duration of non-play social interactions. We also cannot say during which behaviors 50-kHz USV are emitted, suggesting a more in-depth look and a detailed assessment of the individual play partners in combination with 50-kHz USV emissions.

Like play behavior, the sequence of play partner affected the emission of 50-kHz USV in same-genotype HET pairs. When the first play partner was a WT, the same-genotype HET pairs emitted more 50-kHz USV and consequently showed no difference to the same-genotype WT control pairs. For the opposite-genotype pairs, the sequence of play partner had no effect on 50-kHz USV emissions and as such were comparable to same-genotype WT control pairs. Thus, for juvenile female rats the WT play partner appears to increase the emission of 50-kHz USV to control levels and appears to rescue deficits in 50-kHz USV emission in same-genotype HET pairs. Higher 50-kHz USV emissions in the same-genotype HET pairs after playing first with a WT or in opposite-genotype pairs are not altogether surprising, as they also spend more time playing. It is well-known that high rates of social play are associated with high rates of 50-kHz USV emissions in juvenile rats (Knutson et al., 1998; Burgdorf et al., 2005).

For same-genotype WT pairs, we again see mild effects of the first play partner. In fact, sequence effects are only evident within individual play sessions for 50-kHz USV emissions in WT females playing with a same- or opposite-genotype partner. Specifically, during the first play session same-genotype WT pairs emit more 50-kHz USV when the first play partner was a HET compared to a WT. In opposite-genotype WT pairs 50-kHz USV are reduced when the first play partner was a HET compared to WT, particularly within the second play session. The findings in WT pairs with the same and opposite-genotype partners are not too surprising because they can be attributed to the influence of the first play partner and the lasting effects it may have on play behavior. For instance, in half of the same-genotype WT pairs, the first play partner is another WT, while in the other half, they initially played with a partner of different genotype (HET). It is understandable, therefore, that they might be accustomed to playing with the HET partner in such cases. The reason for this is that the difference in behavior only occurs during the first play session. We can speculate that WT females may produce more 50-kHz USV initially to create a playful atmosphere possibly out of habitual tendency, even when they are paired with a new play partner. However, in subsequent sessions, they learn that this behavior is not necessary. Regarding the opposite-genotype WT pairs, the difference in 50-kHz USV emissions occurs only in the second play session. This might also be attributed to the influence of the previous play partner. In half of the pairs, the first play interaction involves a HET partner, whereas in the other half, they have already played with a WT partner for 3 days. As a result, they may still be highly motivated to play and, therefore, emit more 50-kHz USV to encourage their new HET play partner to engage in higher levels of playful behavior.

We know from devocalization studies that a vocal rat playing with a devocalized partner will increase the emission of 50-kHz USV to be comparable to pairs of two vocal rats leading to an increase in the playful mood and higher rates of play (Burke et al., 2017). Although, there is no way to say if the WT females are calling more than the HET females within the opposite-genotype pairs. However, it does not seem so unlikely based on the lower emission rates in same-genotype HET pairs.

Interestingly, while the genotype of cage mates did not affect play behavior, we found an interaction between the genotype of the cage mate and the sequence of the first play partner within 50-kHz USV emissions. In the females from SAME-genotype cages, the emission of 50-kHz USV was higher when the first play partner was the opposite genotype. Sequence effects in SAME-genotype cages are interesting as they reinforce the impression that WT females emit more 50-kHz USV when paired with HET play partners and that the HET females may adapt not only their behavior but also their social communication to complement the play partner leading to an increase in play and 50-kHz USV emissions in the subsequent play partners. In the MIXED-genotype housing condition, we see only an effect in the WT-HET pairs with higher 50-kHz USV when playing with a same-genotype partner first, which is likely driven by the increase in 50-kHz USV during the second play session. Thus, while not a primary effect, the genotype of the cage mate is not wholly unimportant as it provides the initial play experiences with and for both genotypes. For HET females, this is particularly important as it may rescue deficits that would be more prominent if housed with only other SAME-genotype HET cage mates.

There is strong evidence for social learning from conspecifics in various species, including rats (Carcea and Froemke, 2019). This includes social learning of food preferences (Galef Bennet, 2012) and observational fear learning (Kim et al., 2019). Moreover, we recently obtained evidence suggesting that juvenile rats may learn the appropriate use of USV within social contexts from their cage mates and play peers (Kisko et al., 2015a). By removing the ability to produce USV through devocalization, we found that devocalized juvenile rats reared with only other devocalized cage mates are not able to appropriately navigate ambiguous social interactions as adults, leading to increased aggression (Kisko et al., 2015a). In the present study, it is possible that the HET females may have learned from their WT cage mates, and vice versa.

Studies in mice have shown that social deficits in an ASD mouse model can be rescued by rearing them with a highly social mouse strain (Yang et al., 2011). From our study, similar effects might also occur in genetic rat models. The early social environment, through social peers and genotype of the cage mates, influences social communication in our Cacna1c juvenile HET and WT females. It provides further evidence to support findings in mice (Yang et al., 2015) that elements from the social environment, particularly the home cage, can interact with genotype to impact particular aspects of a disease model. Our findings reinforce the importance of the early social environment as a factor in phenotypic changes modulating genetic rat models of neuropsychiatric disorders.

In contrast to the current study, Yang et al. (2015) showed that mutant mice reared in mixed-genotype cages had fewer social interaction-induced USV than the wildtype controls, yet the mutant mice and wildtype controls reared in same-genotype cages did not differ in social interaction-induced USV emissions. The authors suggest that the genetic mutation is not necessarily causing the results but may be due to the mutant mouse’s subordinate status, which is eliminated in the same-genotype cage conditions (Yang et al., 2015). In Cacna1c adult females reared in mixed-genotype cages, we found reduced 50-kHz USV in same-genotype HET compared to same-genotype WT control and opposite-genotype pairs (Redecker et al., 2019). Although, in the social dominance tube test, the HET females were dominant over the WT females (Redecker et al., 2019), speaking somewhat against the results of our study being due to dominant-and subordinate roles within MIXED-genotype cages.

While we did not see any main effect of manipulating the cage mate’s genotype on the play behaviors, we did see an effect on 50-kHz USV emissions. Thus, the effects on behavior may be more subtle and become prominent when looking at the individual rats in each pair and how they respond to one another when initiating or responding to a playful attack. Social dominance in juvenile rats is typically measured through pinning behavior, with the more dominant rat pinning more (Panksepp et al., 1985). In the current study, the duration of play and play components are expressed as a value for the pair. Consequently, we cannot exclusively say at this moment what each specific partner is doing and if one is more dominant and thus pinning more than the other, but the analysis is on-going to establish the behavior of each rat. Furthermore, evidence from Pellis et al. (2022) shows that high and low-playing rats emit the same number of 50-kHz calls yet show differences in the call subtypes and associated behaviors. Thus, detailed investigations for the emission of 50-kHz USV during play would also provide more insight into the playful motivation and how they may use different communication signals depending on the partner’s genotype and the genotype of their cage mates.

In humans, the CACNA1C gene has been linked through GWAS to an increased risk of ASD (Li et al., 2015), partly characterized by social and communication impairments. In the current study, the same-genotype HET pairs show social behavior and communication deficits reflected by reduced social play and 50-kHz USV emissions. However, there is a prominent rescue effect for the HET females by the WT play partners in behavior and 50-kHz USV. Our previous findings indicated that juvenile HET females have a hyper-playful repertoire and had male typical 50-kHz USV emission patterns (Kisko et al., 2020, 2021) yet still showed social communication impairments, particularly in aspects related to social incentive (Kisko et al., 2020). As adults, the HET females displayed further ASD-like social deficits and more repetitive behaviors than the WT controls (Redecker et al., 2019). Late emerging ASD-like phenotypes suggest first that rearing WT and HET juveniles together masks social deficits due to the rescue effect by WT cage mates and the innate drive for juveniles to engage in social play during this critical development period. Secondly, like humans (Baron-Cohen, 2002), ASD-like social impairments in Cacna1c haploinsufficient rats appears to manifest differently in males and females. It has been suggested that in humans autistic females may have fewer social impairments and show higher levels of social motivation than males (Head et al., 2014; Hiller et al., 2014), which is similar to what we have observed in the juvenile Cacna1c haploinsufficient rats.

Additionally, in female children a diagnosis of attention-deficit-hyperactivity disorder is often diagnosed years before a correct or accompanying diagnosis of ASD is recognized, due to the masking of social deficits (Rommelse et al., 2010). Thus, the alteration in Ca_v1.2 protein levels may exert similar sex differences regarding ASD-like phenotypes in Cacna1c haploinsufficient juvenile rats, which may be supplemented by housing them with WT and HET cage mates in females. Furthermore, mixed-genotype housing in males may mask behavioral social deficits but lead to more noticeable deficits in social communication through USV, resulting in decreased social motivation and reduced 50-kHz USV emissions (Kisko et al., 2018). The housing arrangement with HET and WT cages mates, therefore, may have inadvertently rescued and potentially concealed social play and communication deficits in the HET females.