Social dominance emerges as a consequence of agonistic behavior, (i.e., any combative behavior involving struggle among individuals of the same species over limited resources) (Lorenz, 1963). The establishment of the dominant-subordinate status implies the recognition of the fighting ability and motivation among opponents and usually requires aggression (Nelson, 2005; Summers and Winberg, 2006). Once dominance is established, a stable hierarchy can suppress further aggression and unwanted fights among group members (Holekamp and Strauss, 2016). As a result, a clear status-dependent asymmetry in the behavior of contenders is observed immediately after the resolution of the agonistic contest, which can also consolidate in enduring social hierarchies (Milewski et al., 2022). These behavioral asymmetries between dominants and subordinates are regulated by distinctive neuroendocrine mechanisms occurring in the vertebrate social behavior network (SBN) (Goodson, 2005; Kelly, 2022; Newman, 1999; O'Connell and Hofmann, 2011). Multiple neuromodulators, acting both via fast wired circuits and slow diffusive ways, shape the spatio-temporal pattern of activity of the SBN, thus providing the emergence of status-dependent behaviors and the maintenance of stable hierarchies (Goodson and Kabelik, 2009; Newman, 1999; O'Connell and Hofmann, 2011).

A widely used strategy for studying the short-term activation of the SBN is to identify the expression of immediate-early genes (IEGs) among its nodes using strictly controlled social behavioral experiments (Fischer et al., 2018; Friesen et al., 2022; Ghahramani et al., 2022; Goodson et al., 2005; Kabelik et al., 2018; Williamson et al., 2018). As a general rule, the expression of IEGs in the SBN is higher in animals in social interaction with respect to isolated animals (Cabrera-Álvarez et al., 2017; Delville et al., 2000; Loveland and Fernald, 2017). In particular, it has been found that the expression of IEGs in the SBN is also higher in dominants than in non-interacting animals (Cabrera-Álvarez et al., 2017; Delville et al., 2000; Kollack-Walker and Newman, 1995; Loveland and Fernald, 2017) and can show status-dependent spatial patterns across the SBN (Delville et al., 2000). Yet, IEGs, as nonspecific markers of activity, may also be enhanced in both dominants and subordinates and show no difference between them immediately after an agonistic encounter (Otsuka et al., 2020).

Hypothalamic nonapeptides of the vasopressin (AVP) and oxytocin (OT) family are well known to modulate the activity of the SBN in a context-dependent manner and are therefore involved in social status-dependent behaviors across vertebrates (Bales and Carter, 2003; Gobrogge et al., 2009; Goodson and Bass, 2000; Kelly and Goodson, 2014; Sokołowska et al., 2020). Within the enormous diversity of the neuropeptidergic control of the dominant-subordinate status, there are at least three general aspects that account for its complexity. First, AVP and the OT systems usually have complementary or antagonistic roles. AVP is generally related to aggression and dominance (Goodson et al., 2009; Klomberg and Marler, 2000; Santangelo and Bass, 2006; Stribley and Carter, 1999), while OT is associated with enhanced submissive behaviors and decreased aggression (Hellmann et al., 2015; Oliveira et al., 2022; Reddon et al., 2012). Second, the same neuropeptide systems have distinctive actions depending on social status. For example, several reports have shown a differential pattern of activation of AVP neurons between dominants and subordinates in different vertebrates (Ho et al., 2010; Kabelik et al., 2013; Terranova et al., 2016), and that the same pharmacological manipulations of the AVP system induce different actions in dominants and subordinates (Goodson et al., 2009; Huffman et al., 2015; Perrone and Silva, 2018; Semsar et al., 1998). Third, while these pharmacological actions and status-dependent neuronal activation patterns occur during the emergence of dominance, long-term status-dependent changes are frequently observed in AVP cellular traits (Ferris et al., 1989; Larson et al., 2006) and less often in OT ones (Iwata et al., 2010).

In teleost fish, the localization of AVP-like (vasotocin, AVT) and OT-like (isotocin, IT) somata and fibers follow a general common pattern with two main cell groups (magnocellular and parvocellular) located in the nucleus preopticus ventricularis anterior (PPa) and the nucleus preopticus ventricularis posterior (PPp) located between the anterior commissure and the optic chiasm in the region usually named preoptic area (POA; Moore and Lowry, 1998; Pouso et al., 2017, Pouso et al., 2021; Ramallo et al., 2012; Tripp et al., 2020). Three populations of nonapeptidergic neurons are described in the APO: parvocells, magnocells, and gigantocells, recognized for playing different roles in regulating social dominance (Greenwood et al., 2008). Several teleost model systems (e.g., Astatotilapia burtoni; Danio rerio) and approaches (e.g., immunohistochemistry, in situ hybridization) have been used to identify neuropeptidergic cellular correlates of the dominant-subordinate status (Fatsini et al., 2017; Greenwood et al., 2008; Iwata et al., 2010; Larson et al., 2006; Pavlidis et al., 2011; Ramallo et al., 2012). However, most previous studies have only focused a) on the status-dependent asymmetries in the number and size of AVTergic neurons (Greenwood et al., 2008; Iwata et al., 2010; Larson et al., 2006; Ramallo et al., 2012) but not of IT neurons; and b) on the cellular traits of long-term stable social hierarchies (Greenwood et al., 2008; Iwata et al., 2010; Larson et al., 2006; Ramallo et al., 2012) but not of the emergence of the dominant-subordinate status.

Gymnotus omarorum (Gymnotiformes, Gymnotidae; Richer-de-forges et al., 2009), is a neotropical weakly electric fish that displays a clear-cut example of pure territorial aggression during the non-breeding season (Batista et al., 2012; Jalabert et al., 2015; Perrone and Silva, 2018; Quintana et al., 2016; Silva et al., 2013; Zubizarreta et al., 2012). While dominants are highly aggressive even after the conflict is solved, subordinates signal submission in a precise sequence of locomotor and electric traits (Batista et al., 2012; Perrone and Silva, 2018; Quintana et al., 2016). The presence and neuroanatomical distribution of AVT and IT neurons have been previously described in this species as well as their wide projections to brain areas including those involved in the control of aggression and electromotor behaviors (Pouso et al., 2017, Pouso et al., 2021). The administration of IT does not modify either locomotor or electromotor activity in isolated individuals (Pouso et al., 2021), while the effect of IT on the agonistic encounter of G. omarorum has not been previously evaluated. In contrast, the behavioral asymmetry between dominants and subordinates in G. omarorum is paralleled by status-dependent AVT modulations (Perrone and Silva, 2018). While in dominant individuals' aggression requires an endogenous release of AVT, the administration of AVT to subordinates induce them to increase their electric signaling of submission (Perrone and Silva, 2018). Although these status-dependent actions of AVT suggest distinctive activation patterns of the AVT system between dominants and subordinates during the emergence of dominance, no short-term plastic changes of neuropeptidergic neurons measured by IEGs have been evaluated in the agonistic behavior of G. omarorum so far.

An elegant way of evaluating the involvement of neuropeptidergic neurons during the establishment of the dominant-subordinate status is to search for short-term plastic changes of AVP and OT cells associated with the expression of IEGS. In this study, we took advantage of the non-breeding aggressive behavior of male Gymnotus omarorum to focus on the cellular changes of POA neuropeptidergic neurons of both dominants and subordinates following an agonistic encounter. To do this, we conducted careful behavioral experiments including two conditions of social interaction (fighting and non-fighting male dyads) that implied a similar global activation of the POA measured by the expression of the IEGS. In these similarly socially activated brains, we were thus able to search for short-term plastic status-dependent changes of AVT+ and IT+ cells by immunohistochemistry.