Using a socially enriched environment test to evaluate the effects of different therapeutic approaches on social behavior in a mouse model of social anxiety disorder

SFA-induced social fear

Numerous animal models, including the social defeat paradigm and exposure to foot shocks, have been employed to investigate the mechanisms of SAD due to its widespread prevalence and severe symptoms. However, these paradigms lack precision in inducing specific behavioral changes, such as generalized anxiety or depression (Denmark et al. 2010; Hollis et al. 2010). In light of these limitations, we adopted the paradigm developed by Toth et al. (2012). This method allows for the elicitation of targeted responses to social fear without introducing additional confounding behavioral changes. As described by Toth et al. (2012), this paradigm is grounded in operant conditioning, in which experimental mice learn to associate social interactions with punishment in the form of electrical shocks, thus inducing fear and avoidance of social stimuli.

In this first part of our study, our aim was to investigate alterations in the social behavior of male mice conditioned to social fear using the SEE test. Our results demonstrate that SFC led to a reduction in IPZ object exploration among conditioned mice compared to control mice. Specifically, conditioned mice showed a tendency to withdraw when unfamiliar mice approached an object. In contrast, their OPZ exploration remained unchanged compared to nonconditioned mice. This suggests that SFC did not alter their exploratory activity, but the presence of stimulus mice nearby interfered with it. This observation is reinforced by the preference index for OPZ exploration, which was higher in the SFC+ group compared to the SFC− group. These findings suggest that SFC promotes an increased preference for OPZ exploration and results in reduced social interaction in conditioned mice. These observations may reflect the social avoidance behaviors seen in individuals with anxiety disorders, where individual exploration is favored while social interaction is limited (American Psychiatric Association 2013). To further explore aspects of social fear, it would be relevant to investigate other components of this behavior. Indeed, the results regarding interaction time further reflect the effectiveness of SFC in inducing SAD. Thus, our conditioned mice exhibited fewer social interactions than nonconditioned mice. These results are consistent with those of Toth et al. (2012). Similarly, in humans, anxiety reduces the motivation to interact with others, leading highly anxious individuals to avoid social interactions even more (Turner 1988). In this regard, the study conducted by Duronto et al. (2005) demonstrates that social interaction avoidance is a reliable indicator of anxiety.

It is important to note that social avoidance can also reflect disinterest in social stimuli, which could bias result interpretation. Therefore, it is crucial to evaluate other fear-related responses, such as immobility (Blanchard et al. 2005). Indeed, data from the SEE test reveal a higher immobility duration in the group subjected to SFC compared to the control group. This suggests that these mice have memorized social interactions associated with electrical shocks during the approach of a conspecific. We posit that this immobility behavior specifically represents avoidance of social stimulus interactions. These observations are consistent with the findings of Toth et al. (2013). In other words, our data confirm that the reduction in social interactions is specifically attributed to social fear and not a lack of interest.

Treatment with acute oxytocin

The pivotal role of oxytocin in social behavior has garnered substantial interest among researchers due to its profound influence on various affiliative behaviors, including mating, parental behavior, and attachment (Bartz and Hollander 2006). Such social involvement suggests that this hormone plays a significant role in the functioning of our “social brain” (Feldman 2012). According to Buchheim et al. (2009), individuals experiencing relational insecurity exhibit an increase in their feelings of attachment and emotional security following oxytocin administration. Similarly, a study by De Oliveira et al. (2012) suggests that oxytocin reduces anticipatory anxiety in healthy subjects when speaking in public.

Using C57BL/6n mice, Zhang et al. (2015) have demonstrated that subcutaneous oxytocin administration (100 μg/kg, 10 mL/kg, administered 30 min before behavioral tests) increased social interaction time and modulated both social and nonsocial behaviors.

Given the above-mentioned effects, we sought to explore the influence of this neuropeptide on social fear. To do this, we induced social fear through SFC in male mice, and, after a 24 h interval, we administered oxytocin (100 µg/kg, 10 mL/kg) subcutaneously. Thirty minutes later, we subjected them to the SEE test.

The results revealed that oxytocin administration in the conditioned group led to a significant increase in IPZ exploration compared to OPZ exploration of objects, in comparison with the conditioned vehicle group. These results suggest that oxytocin may promote collective exploration in conditioned mice. Furthermore, the results of the preference index for OPZ exploration showed a highly significant decrease in this index in the SFC+/OXT group compared to the SFC+/Veh group. Collectively, these findings indicate that oxytocin reduces social fear induced by SFC. Regarding social interaction, the results showed a significant increase in this parameter in the conditioned group treated with oxytocin compared to the conditioned vehicle group. This implies that oxytocin facilitated an increase in social interaction behaviors in conditioned mice. Additionally, we also examined immobility, which was significantly reduced in the conditioned group treated with oxytocin compared to the conditioned group treated with the vehicle. Although this reduction did not reach the level of the nonconditioned groups, it suggests that oxytocin contributed to a reduced fear response in conditioned mice. In summary, based on the analysis of the parameters mentioned above, it appears that oxytocin has an opposing effect on SFC.

In previous studies in rodents, oxytocin has been identified as a neuropeptide that promotes social approach behavior and assists rodents in overcoming social proximity avoidance. This aligns perfectly with our results. Indeed, it has been reported that acute oxytocin treatment increases sociability and social affiliation, extending beyond reproductive-related behaviors (Lukas et al. 2011; Ramos et al. 2013; Holley et al. 2015). Furthermore, mice with mutations in the oxytocin receptor gene exhibit abnormalities in social behavior (Takayanagi et al. 2005). Additionally, according to De Oliveira et al. (2012), both male and female rats and mice respond to oxytocin in the same way as anxiolytics. Just as in animals, oxytocin administration also promotes social interactions in humans. A single intranasal administration of oxytocin has been shown to increase the subjective sensation of attachment security, which is associated with reduced stress reactivity and improved social interaction capacity (Buchheim et al. 2009). Thus, oxytocin diffuses to several brain structures involved in social behavior and fear regulation, including the amygdala (Meyer-Lindenberg et al. 2011). A study in rats has identified a pathway through which oxytocin inhibits amygdala activation, thus modulating its communication with the autonomic nervous system (Huber et al. 2005). This action of oxytocin contributes, in part, to its anxiolytic properties by reducing the hypothalamo-hypophyseal axis reactivity to social stressors (Meyer-Lindenberg et al. 2011). Finally, it has been demonstrated that the amygdala is connected to prefrontal and striatal regions involved in affiliative or avoidance behaviors among individuals of the same species (McDonald 1991). Therefore, the modulation of the activity of these brain structures, as well as others, by oxytocin could play a crucial role in reducing social fear induced by fear conditioning in our study.

Treatment with brain stimulation

The prefrontal cortex plays a crucial role in inhibiting inappropriate responses, particularly the IL region of the medial prefrontal cortex, which is vital for the extinction of conditioned fear (Quirk et al. 2006). In a prior study, the GABA_A receptor inhibitor (bicuculline methiodide) was infused into the IL cortex, resulting in an anxiogenic effect (Bi et al. 2013). Furthermore, in the same study, these researchers suggested that imbalances in afferents to pyramidal neurons in the IL cortex could lead to anxiety disorders. Additionally, activation of this region increases motivation, sociability, and reduces stress in rodents (Myers et al. 2019).

Given the above-mentioned effects, we aimed to investigate the efficacy of stimulating this brain structure in reducing social fear responses. To do this, we induced social fear in male mice. Subsequently, the mice underwent two sessions of the SEE test: one before stimulation and one after.

The results demonstrated an increase in social interaction and IPZ object exploration following HFS in both the conditioned and unconditioned groups. We also observed that the value of the OPZ exploration preference index significantly decreased after stimulation in both groups. However, SFC+ mice maintained a higher OPZ index than SFC− mice during the second session of the SEE test. Furthermore, HFS significantly reduced immobility duration in conditioned mice compared to the prestimulation period. In summary, these results suggest that the IL cortex plays a substantial role in the modulation of social anxiety. In line with this last finding, prior research has indicated that microstimulation of the IL cortex reverses the effect of SFC, resulting in decreased freezing (Milad and Quirk 2002). Moreover, the IL cortex influences the hypothalamo-pituitary stress axis response (Radley et al. 2006). This explanation could be extrapolated to our results, indicating a reduction in stress responses and an improvement in social interaction among conditioned mice. On the other hand, the IL cortex projects to numerous sites involved in fear response control, notably the amygdala. More specifically, intercalated neurons receive a significant projection from the IL cortex (McDonald et al. 1996), and stimulation of this connection promotes rapid extinction and inhibition of central amygdala neurons, leading to reduced fear expression (Milad and Quirk 2002).

This study suggests that HFS in the IL cortex may have a potential therapeutic role in alleviating social anxiety and improving social behavior.

Treatment with extinction training

Within the realm of psychotherapy, one approach employed is exposure therapy, which aims to gradually diminish fear by inducing the extinction of fear responses through repeated exposure to aversive conditioned stimuli. This method is applicable, especially in cases of acquired social fear (Fedoroff and Taylor 2001). This therapeutic approach has proven effective in treating SAD in humans (Clark et al. 2003) and in certain animal models of conditioned fear (Toth et al. 2012). Studies suggest that cognitive-behavioral interventions that incorporate exposure techniques yield slightly more favorable results than interventions solely focused on cognitive restructuring (i.e., cognitive therapy focused on anxiety and/or patient's thoughts) (Gould et al. 1997).

In this context, our objective was to explore the effect of exposure therapy on the extinction of social fear using a model of SFC. To achieve this, male mice underwent four extinction sessions with a 48 h interval between each session.

The results confirmed the extinction of social fear based on various measured parameters. Indeed, during the fourth session of the test, a significant increase in IPZ exploration was observed in the conditioned mice compared to their first session, although this level remained lower than that of unconditioned mice. Conversely, for the SFC+ group, the OPZ exploration preference index and immobility duration showed a decrease during the last session compared to the first. Nevertheless, the OPZ exploration preference index and immobility duration remained higher in the SFC+ group compared to the CPS− group. Furthermore, a slight increase in social interaction was observed between the first and fourth session of the test in SFC+ mice. In conclusion, the repetition of the SEE test led to a decrease in conditioned fear.

The obtained results can be attributed to the fact that exposure promotes corrective learning, involving unlearning or deconditioning phobic reactions (Marks and Toben^a 1990). It is worth noting that extinction is a learning process, and the reduction of fear results from the inhibition of the response rather than the complete erasure of the initial fear memory (Garcia 2002; Bouton et al. 2006; Ji and Maren 2007). From a neurobiological perspective, extinction can be considered as a form of learning that does not affect previous plasticity but rather reflects an increase in plasticity in GABAergic inhibitory interneurons. Therefore, the greater the plasticity of GABAergic interneurons, the more they inhibit the central amygdala, and the less fear is expressed (Royer and Paré 2002).

Beyond the amygdala, other neural structures may be involved in extinction, such as the medial prefrontal cortex and the hippocampus, both of which have interconnections with the amygdala (Quirk and Mueller 2008). These structures appear to modulate the inhibition of conditioned responses (Bruchey et al. 2007; Ji and Maren 2007).

This study suggests that repeated exposure tests may have potential therapeutic implications for promoting the extinction of social fear, offering insights into the underlying mechanisms of this process.

Comparison of the three therapeutic approaches

In the final segment of our study, we conducted a comparative assessment of three therapeutic approaches to determine their relative efficacy. Our findings indicate that there is no significant difference in the preference index for OPZ exploration between the CONT group and the EXT group. This suggests that a second exposure to the SEE test does not mitigate the impact of SFC on the behavior of conditioned mice. In contrast, both the OXT and HFS groups exhibited a comparable reduction in OPZ exploration preference, which was significantly different from the CONT and EXT groups. Therefore, it can be inferred that OXT and HFS are more likely to influence the tendency of mice to collectively explore their environment.

Additionally, when examining interaction time, the results show no significant differences between the CONT and EXT groups, indicating that the second exposure to the SEE test and the absence of treatment have a similar effect on social interaction, resulting in reduced interaction time. On the contrary, both the OXT and HFS groups demonstrated a significantly higher interaction time. Notably, the EXT group, while not significantly different, spent less time in social interaction compared to the OXT and HFS groups. These outcomes suggest that both oxytocin administration and HFS treatments are more effective in enhancing social interaction behaviors that were disrupted by SFC, in contrast to the absence of treatment and the second exposure.

Furthermore, the CONT group exhibited significantly longer freezing durations compared to the EXT, OXT, and HFS groups. However, there were no significant differences between the EXT group and the OXT and HFS groups, even though the graphical representation might imply longer freezing durations in the EXT group. This indicates that the second exposure to the SEE test, oxytocin administration, and HFS treatments were equally effective in reducing freezing durations when compared to the absence of treatment.

The obtained results unequivocally demonstrate that our comparative evaluation of three therapeutic approaches in the context of mitigating the effects of SFC in mice provides valuable insights. The findings highlight that both oxytocin administration and HFS treatments show more promise in reversing the effects of SFC on mice. Conversely, it becomes evident, upon close examination of the results, that achieving comparable positive outcomes to those of the OXT and SFC groups would require multiple exposure sessions.

This suggests that all three therapeutic approaches have a positive impact when compared to the absence of treatment, offering potential strategies for addressing the behavioral consequences of SFC in preclinical models.

The choice to conduct behavioral tests, such as the SEE test, 30 min after oxytocin administration was informed by both previous studies and the pharmacokinetic profile of oxytocin. Subcutaneous injections have been shown to result in peak or effective oxytocin activity within a 30 min window, as demonstrated in behavioral tests like the social interaction test (Zhang et al. 2015). This timing aligns with findings from studies on nasal administration, which similarly report that oxytocin reaches its peak concentration ∼30 min post-administration (Neumann et al. 2013). This consistency across different routes of delivery highlights a robust pharmacokinetic profile, making 30 min post-administration the optimal window for assessing oxytocin's effects on behavior.