Hormones and Behavior

The impact of variation in parental caregiving has lasting implications for the development of offspring. However, the ways in which parents impact each other in the context of caregiving is comparatively less understood, but can account for much of the variation observed in the postnatal environment. Prairie voles (Microtus ochrogaster) demonstrate a range of postnatal social groups, including biparental pairs and pups raised by their mothers alone. In addition to the challenges of providing parental care, prairie vole parents often experience acute natural stressors (e.g., predation, foraging demands, thermoregulation) that could alter the way co-parents interact. We investigated how variation in the experience of raising offspring impacts parental behavior and neurobiology by administering an acute handling stressor on prairie vole families of single mothers and biparental parents over the course of offspring postnatal development. Mothers and fathers exhibited robust behavioral plasticity in response to the age of their pups, but in sex-dependent ways. Pup-directed care from mothers did not vary as a function of their partners presence, but did covary with the number of hypothalamic vasopressin neurons in experience-dependent ways. The relationship between vasopressin neuron numbers and fathers behaviors was also contingent upon the stress handling manipulation, suggesting that brain-behavior associations exhibit stress-induced plasticity. These results demonstrate that the behavioral and neuroendocrine profiles of adults are sensitive to distinct and interacting experiences as a parent, and extend our knowledge of the neural mechanisms that may facilitate parental behavioral plasticity.

This study examined the long-term effects of low social status on reproductive success and seasonal changes in reproductive and stress hormones of adult female rhesus macaques. Rhesus are a matriarchal, matrilineal species that form hierarchical social hierarchies maintained by aggression. Thus, animals in the bottom of the hierarchy (low social status: subordinate) experience high levels of aggression from those in higher social ranks and therefore must remain vigilant. Social subordination generates chronic psychosocial stress and could negatively impact reproductive function. Twenty-seven adult female rhesus monkeys (13 dominant -DOM-, 14 subordinate -SUB-; 8-11 years old) were studied for stress neuroendocrine function -measuring basal plasma levels of cortisol-, as well as for reproductive endocrine function -measuring basal levels of estradiol (E2). In addition, we examined the reproductive success rates of these females (defined as number of live births/years with male access), their success rates raising infants that survived to 1 year, and the characteristics and effectiveness of their maternal care. Our findings show significantly lower reproductive success rates in SUB than DOM females, blunted E2 seasonal changes, driven by higher E2 levels in the anovulatory season and more attentive maternal care of offspring early in life than dominant females. Age at first birth was negatively associated with infant survival rate across ranks. Interestingly, higher levels of reproductive hormones (E2) during the anovulatory season predicted lower reproductive success, although the effect was driven by animals lactating at that point. Overall, these findings suggest that subordinate female macaques show a phenotype consistent with impaired reproductive function. Simple SummaryThis study examined the long-term effects of low social status on reproductive success and seasonality of reproductive and stress physiology of female rhesus monkeys. The despotic rhesus social hierarchy is matrilineal and maintained by aggression. Males leave their natal group in bachelor groups around puberty, but females remain, so social status of group members is passed down the female line. Rhesus with low social status (subordinate) experience significantly more aggression and harassment than their higher-ranking counter parts (dominant). This makes rhesus macaques an ideal model for studying chronic psychosocial stress and its long-term effects. Our findings show significantly lower reproductive success rates (number of live births based on access to males; lower infant survival rates) in subordinate than dominant females. Subordinate females also show blunted seasonal changes of the gonadal hormone estradiol (E2), driven by higher E2 levels in the anovulatory season than dominant animals, as well as more attentive maternal care of offspring early in life than dominant animals. Age at first birth was negatively associated with infant survival rate across ranks. Interestingly, higher levels of reproductive hormones (E2) during the anovulatory season were associated with lower reproductive success, although the effect was driven by animals lactating at that point. Overall, these findings suggest that female macaques with lower social status show a phenotype consistent with impaired reproductive function, though the biological mechanisms involved need to be further explored.

The socially monogamous prairie vole (Microtus ochrogaster) and promiscuous meadow vole (Microtus pennsylvanicus) are closely related, but only prairie voles display long-lasting pair bonds, biparental care, and selective aggression towards unfamiliar individuals after pair bonding. These social behaviors in mammals are largely mediated by steroid hormone signaling in the social behavior network (SBN) of the brain. Hormone receptors are reproducible markers of sex differences that can provide more information than anatomy alone, and can even be at odds with anatomical dimorphisms. We reasoned that behaviors associated with social monogamy in prairie voles may emerge in part from unique expression patterns of steroid hormone receptors in this species, and that these expression patterns would be more similar across males and females in prairie than in meadow voles or the laboratory mouse. To obtain insight into steroid hormone signaling in the developing prairie vole brain, we assessed expression of estrogen receptor alpha (Esr1), estrogen receptor beta (Esr2), and androgen receptor (Ar) within the SBN, using in situ hybridization at postnatal day 14 in mice, meadow, and prairie voles. We found species-specific patterns of hormone receptor expression in the hippocampus and ventromedial hypothalamus, as well as species differences in the sex bias of these markers in the principal nucleus of the bed nucleus of the stria terminalis. These findings suggest the observed differences in gonadal hormone receptor expression may underlie species differences in the display of social behaviors.

Paternal experience improves memory and reduces anxiety-like behavior in males, but it is unclear whether these changes are due to mating, siring offspring, or caregiving behavior. Likewise, paternal males have larger testes, a measure of sperm production, but again, the effects of siring and caregiving experience are difficult to disentangle. We examined behavioral and physiological outcomes in three groups of male mice: 1) virgins, 2) males paired with sterile females ( non-fathers), and 3) experienced fathers (Experiment 1). Compared to virgins and non-fathers, experienced fathers exhibited increased recognition memory (novel object recognition) and decreased anxiety-like behavior (elevated plus maze). Virgin males, however, had smaller testes and fewer sperm compared to non-fathers. We then compared the same traits in three additional groups of male mice: 1) non-fathers, 2) non-fathers with experience caring for unrelated pups ( pup-sensitized non-fathers), and 3) first-time fathers, to determine whether the behavioral and physiological observations in Experiment 1 were due to siring offspring or caregiving, and how rapidly these changes occur (Experiment 2). Recognition memory and anxiety-like behavior did not differ among these three groups, suggesting that caring for a single litter does not recapitulate the behavioral changes observed in experienced fathers (Experiment 1). Despite equal mating opportunity, we observed larger testes in first-time fathers compared to non-fathers, suggesting that investment in sperm production may be more plastic than behavioral changes. Finally, we compared pup interactions in pup-sensitized non-fathers and first-time fathers. While pup-sensitized non-fathers were slower to approach pups than first-time fathers, they spent more time grooming pups, whereas first-time fathers invested more time in nest building, suggesting different caregiving behaviors in pup-sensitized males and biological fathers. Taken together, our study revealed that mating, siring, and caregiving experience contributes to changes in memory, anxiety-like behavior, and reproductive investment in males of a biparental species.

Social connectedness strongly influences health and longevity, and adult pair bonds provide psychological benefits distinct from other social relationships. Oxytocin (OT), corticotropin-releasing hormone (CRH), and opioids, play an important role in the formation and maintenance of pair bonds. Evidence suggests that OT modulates the stress response via the hypothalamic-pituitary-adrenal (HPA) axis, while the kappa ({kappa}) opioid system interacts with and may modulate OT signaling in contexts of stress and separation. In this study 20 coppery titi monkeys were exposed to a physical stressor under three social conditions: baseline (no stressor, partner present), stress (stressor, no partner present) and buffering (stressor, partner present). We predicted stress-induced dynorphin release would reduce {kappa}-opioid receptor availability measured via [{superscript 1}{superscript 1}C]GR103545 Positron Emission Tomography (PET) and lower cerebrospinal fluid (CSF) OT, whereas partner presence would mitigate dynorphin release and increase CSF OT, with reduced dynorphin inferred from higher {kappa}-opioid receptor radioligand binding. Our results show condition-dependent differences in [{superscript 1}{superscript 1}C]GR103545 binding in several brain regions, including the amygdala and hippocampus, with altered binding in both the stress and social buffering conditions. Cortisol levels were elevated in the stress condition compared to baseline. Females exhibited lower CSF OT levels during stress than at baseline, whereas plasma OT levels did not differ across conditions or between sexes. Spearman correlations revealed no significant associations between plasma and CSF OT. Together, these findings highlight the complex interaction between {kappa}-opioid signaling, OT, and HPA axis activity in the context of social relationships and highlight neuroendocrine mechanisms underlying stress regulation in pair-bonded species.

The nigrostriatal and mesoaccumbal dopamine systems are thought to contribute to changes in behavior and learning during adolescence, yet it is unclear how the rise in gonadal hormones at puberty impacts the function of these systems. We studied the impact of prepubertal gonadectomy on evoked dopamine release in male Mus spicilegus, a mouse whose adolescent life history has been carefully characterized in the wild and laboratory. To examine how puberty impacts the dopamine systems in M. spicilegus males, we removed the gonads prepubertally at P25 and then examined evoked dopamine release in the dorsomedial, dorsolateral, and nucleus accumbens core regions of striatal slices at P60-70. To measure dopamine release, we used near-infrared catecholamine nanosensors (nIRCats) to enable study of spatial distribution of dopamine release sites in each striatal region. We found that prepubertal gonadectomy led to a significantly reduced density of dopamine release sites and reduced dopamine release at each site in the dorsolateral nigrostriatal system compared to sham controls. By contrast, mesoaccumbal dopamine release was comparable between sham and gonadectomized groups. Our data suggest that during adolescence the development of the nigrostriatal dopamine system is significantly affected by the rise in gonadal hormones in males, while the mesoaccumbal system shows no detectable sensitivity at this time point. These data are consistent with molecular studies in rodents that suggest nigrostriatal neurons are sensitive to androgens at puberty, and extend our understanding of how gonadal hormones could impact the spatial distribution and release potential of dopamine terminals in the striatum. Significance StatementHere we use a wild-derived species, Mus spicilegus, to study adolescent development. This wild-derived species has value over standard lab mice because it is more likely to exhibit evolved developmental programs relevant to dispersal and other natural behaviors. By using this wild-derived species and metrics of evoked dopamine release with spatial resolution, we can test if the rise in gonadal hormones at puberty plays a role in maturation of dopamine terminal function in the striatum. These findings may help us better understand developmental programs in humans that orchestrate changes in behavior at adolescent milestones in contexts of both health and disease.

Female-female mounting is widespread in mammals. Although the behavioral functions of female-female mounting likely vary according to species and to behavioral context, a large body of work has investigated the relationship of female-female mounting to female sexual receptivity and hormonal status. While this relationship has been extensively explored in female rats, fewer instances of female-female mounting have been described in mice, and correspondingly, less is known about the potential relationship in mice between female-female mounting and estrous state. Recently, we found that short-term social isolation robustly promotes same-sex mounting in C57BL/6J female mice. In the current study, we tested whether displays of female-female mounting by naturally cycling 3-days-single-housed females during interactions with naturally cycling, group-housed stimulus females are related to estrous state. We found no evidence that same-sex mounting by single-housed females is related to the estrous state of either female in the pair. These findings suggest that same-sex mounting displayed by single-housed female mice is not related to female sexual behavior, and future work remains to determine the functions of female-female mounting in this species and behavioral context.

In placental mammals, estradiol levels are chronically elevated during pregnancy, but quickly drop to prepartum levels following birth. This may produce an "estrogen withdrawal" state that has been linked to changes in affective states in humans and rodents during the postpartum period. The neural mechanisms underlying these affective changes, however, are understudied. We used a hormone-simulated pseudopregnancy (HSP), a model of postpartum estrogen withdrawal, in adult female C57BL/6 mice to test the impact of postpartum estrogen withdrawal on several behavioral measures of anxiety and motivation. We found that estrogen withdrawal following HSP increased anxiety-like behavior in the elevated plus maze, but not in the open field or marble burying tests. Although hormone treatment during HSP consistently increased sucrose consumption, sucrose preference was generally not impacted by hormone treatment or subsequent estrogen withdrawal. In the social motivation test, estrogen withdrawal decreased the amount of time spent in proximity to a social stimulus animal. These behavioral changes were accompanied by changes in the expression of {Delta}FosB, a transcription factor correlated with stable long-term plasticity, in the nucleus accumbens (NAc). Specifically, estrogen-withdrawn females had higher {Delta}FosB expression in the nucleus accumbens core. Using transgenic reporter mice, we found that this increase in {Delta}FosB occurred in both D1- and D2-expressing cells in the NAc core. Together, these results suggest that postpartum estrogen withdrawal impacts anxiety and motivation and increases {Delta}FosB in the NAc core.

Peri-adolescence is a critical developmental stage marked by profound changes in the valence of social interactions with parents and peers. We hypothesized that the oxytocin (OXT) and vasopressin (AVP) systems, known for influencing social behavior, would be involved in the maintenance and breaking of bonding behavior expressed by peri-adolescent males and females. In rodents, OXT is associated with mother-pup bonding and may promote social attachment to members of the natal territory. AVP, on the other hand, can act in contrasting ways to OXT and has been associated with aggression and territoriality. Specifically, we predicted that in peri-adolescent male and female juveniles of the biparental and territorial California mouse (Peromyscus californicus), a) OXT would increase the social preferences for the parents over unfamiliar age-matched peers (one male and one female), and b) AVP would break the parent-offspring bond and either increase time in the neutral chamber and/or approach to their unfamiliar and novel peers. We examined anxiety and exploratory behavior using an elevated plus maze and a novel object task as a control. Peri-adolescent mice were administered an acute intranasal (IN) treatment of 0.5 IU/kg IN AVP, 0.5 IU/kg IN OXT, or saline control; five minutes later, the behavioral tests were conducted. As predicted, we found that IN OXT enhanced social preference for parents; however, this was only in male and not female peri-adolescent mice. IN AVP did not influence social preference in either sex. These effects appear specific to social behavior and not anxiety, as neither IN OXT nor AVP influenced behavior during the elevated plus maze or novel object tasks. To our knowledge, this is the first evidence indicating that OXT may play a role in promoting peri-adolescent social preferences for parents and delaying weaning in males. HIGHLIGHTSO_LIIn a 3-chambered choice test, peri-adolescent female and male California mice prefer their parents over peers or an empty chamber C_LIO_LIIntranasal oxytocin (IN OXT) enhances male but not female peri-adolescent social preference for their parents C_LIO_LIIntranasal arginine vasopressin (IN AVP) did not influence social preference in either sex C_LIO_LINeither IN OXT nor AVP alter peri-adolescent behavior in an elevated plus maze or novel object task C_LIO_LIOXT may play a role in delaying weaning in males C_LI

Abstract/SummaryFriendships, or selective peer relationships, are a vital component of healthy social functioning in humans, while deficits in these relationships are associated with negative physical and mental health consequences. Like humans, prairie voles are among the few mammalian species that form selective social bonds with both peers and mates, making them an excellent model for the mechanistic investigation of selective social attachment. Here, we explored the role of oxytocin receptors in selective peer attachment using prairie voles lacking a functional oxytocin receptor gene (Oxtr1-/-). We found that Oxtr1-/- animals exhibited significant delays in peer relationship formation compared to wildtype animals. Oxytocin receptor function also contributed to the maintenance of peer bonds, as Oxtr1-/- voles displayed reduced relationship stability and lost selective attachments rapidly in a multi-chamber, group-living habitat. Oxtr1-/- voles also showed deficits in both general social reward as well as selective social reward for a peer partner over an unfamiliar conspecific. Evoked oxytocin release in the nucleus accumbens was reduced in Oxtr1-/- animals compared to their wildtype counterparts, indicating that these voles do not have a compensatory increase in oxytocinergic signaling. Together, these data indicate that oxytocin receptors influence the formation, persistence, and reward value of peer relationships.

Oxytocin (Oxt) signaling via its receptor, the Oxt receptor (Oxtr), is important to the onset of mammalian maternal care. Specifically, evidence suggests that Oxt signaling around the time of parturition underlies the critical shift in how pups are perceived, i.e. from aversive stimuli to rewarding stimuli. Previous work from our lab has found that both Oxtr knockout (-/-) mice and forebrain-specific Oxtr knockout (FB/FB) are more likely than controls to abandon their first litters. Based on these data, we hypothesized that this observed pup abandonment phenotype was due to a failure of the brain to "switch" to a more maternal state. In order to identify where in the brain Oxt signaling contributes to the onset of maternal care we performed three experiments. In Experiment 1, virgin Oxtr FB/FB females were assessed for genotypic differences in maternal behavior and c-Fos expression following maternal sensitization was quantified. In Experiment 2, c-Fos expression was quantified in Oxtr -/- and Oxtr FB/FB females following parturition. In Experiment 3, based on our findings from Experiment 2, the Oxtr in the nucleus accumbens shell (NAcc) was genetically deleted in female Oxtr floxed mice (Oxtr Flox/Flox) mice using a Cre recombinase expressing adeno-associated virus. In Experiment 1, sensitized virgin Oxtr FB/FB females had significantly lower retrieval latencies on the first day of testing and reduced c-Fos expression in the dorsal lateral septum compared to controls. In Experiment 2, increased c-Fos expression was observed in the NAcc shell of both Oxtr -/- and Oxtr FB/FB dams as compared to controls. In Experiment 3, virally mediated knockout of the Oxtr in the NAcc shell completely disrupted the onset of maternal care. Thus, by genetically deleting Oxtr expression in the NAcc the pup abandonment phenotype previously observed in Oxtr -/- and Oxtr FB/FB dams was recreated. Taken together, these data suggest that in post-parturient mice, Oxtr expression in the NAcc shell is critical to the onset of maternal behavior.

Historic bias toward study of male sex hormones and sexual signals currently constrains our perspective of hormone--behavior--phenotype relationships. Resolving how ornamented female phenotypes evolve is particularly important for understanding the diversity of social signals across taxa. Studies of both males and females in taxa with variable female phenotypes are needed to establish whether sexes share mechanisms underlying expression of signaling phenotypes and behavior. White-shouldered Fairywren (Malurus alboscapulatus) subspecies vary in female ornamentation, baseline circulating androgens, and response to territorial intrusion. The moretoni ornamented female subspecies show higher female, but lower male androgens, and a stronger pair territorial response relative to pairs from the lorentzi unornamented female subspecies. Here we address whether subspecific differences in female ornamentation, baseline androgens, and pair territoriality are associated with ability to elevate androgens following gonadotropin releasing hormone (GnRH) challenge and in response to simulated territorial intrusion. We find that subspecies do not differ in their capacity to circulate androgens in either sex following GnRH or territorial intrusion challenges. Whereas pre-GnRH androgens were somewhat predictive of degree of response to territorial intrusions, higher androgens were associated with lower territorial aggression. Post-GnRH androgens were not predictive of response to simulated intruders, nor did females sampled during intrusion elevate androgens relative to flushed controls, suggesting that increased androgens are not necessary for the expression of territorial defense behaviors. Collectively, our results suggest that capacity to produce and circulate androgens does not underlie subspecific patterns of female ornamentation, territoriality, and baseline androgens.

Uncovering the genetic, physiological, and developmental mechanisms underlying phenotypic variation is necessary for understanding how genetic and genomic variation shape phenotypic variation and for discovering possible targets of selection. Although the neural and endocrine mechanisms underlying social behavior are evolutionarily ancient, we lack an understanding of the proximate causes and evolutionary consequences of variation in these mechanisms. Here, we examine in the natural environment the behavioral, neuromolecular, and fitness consequences of a morpholino-mediated knockdown of the mineralocorticoid receptor (MR) in the brain of nesting males of the ocellated wrasse, Symphodus ocellatus, a species with male alternative reproductive tactics. Even though MR knockdown did not significantly change male behavior directly, this experimental manipulation strongly altered glucocorticoid signaling and neuroplasticity in the preoptic area, the putative hippocampus homolog, and the putative basolateral amygdala homolog. We also found that individual variation in stress axis gene expression and neuroplasticity is strongly associated with variation in male behavior and fitness-related traits. The brain region-specific effects of MR knockdown on phenotypic integration in the wild reported here suggest specific neuroendocrine and neuroplasticity pathways that may be targets of selection.

Aggression is a common behavioral response to limited environmental resources. Most research on the neural basis of aggression in vertebrates focuses on adult males, where sex steroid hormones and the ventromedial hypothalamus are important regulators of aggressive behavior. However, the young of many species also display aggression, although the neural basis of juvenile aggression is not well understood. Here we examine juvenile aggression in Mimic poison frog (Ranitomeya imitator) tadpoles, who live in small isolated pools with limited resources and display aggressive behaviors towards intruder tadpoles. We first conducted a longitudinal study of dyadic behavior and found aggressive behavior increases with tadpole age. We next evaluated which brain regions may be important for tadpole aggression by quantifying pS6-positive cells as a proxy for neural activity. We found more pS6-positive cells in the ventral hypothalamus of aggressive tadpoles compared to controls, similar to findings in adult mammals. We then quantified colocalization of pS6 with vasopressin and oxytocin cells and found no difference across behavior groups. Based on this information, we hypothesized that similar brain regions, but different neuromodulators, may promote aggression in juvenile tadpoles compared to the literature in adult animals. We then used an untargeted approach to molecularly profile pS6-positive neurons and found enrichment of the proopiomelanocortin (POMC) gene in aggressive tadpoles. As POMC is cleaved into several signaling peptides, we used pharmacology to target each pathway and discovered that blocking opioid receptors increases aggressive behavior. Together, this work suggests that POMC-derived B-endorphin is a negative regulator of juvenile aggression through the opioid receptor signaling. More broadly, this work suggests that similar brain regions, but different signaling molecules may be used to regulate aggression in adult and juvenile animals.

Prolactin, a hormone involved in vertebrate parental care, is hypothesized to inhibit reproductive hypothalamic-pituitary-gonadal (HPG) axis activity during parenting, thus maintaining investment in the current brood as opposed to new reproductive efforts. While prolactin underlies many parental behaviors in birds, its effects on other reproductive behaviors, such as courtship, remain unstudied. How prolactin affects neuropeptide and hormone receptor expression across the avian HPG axis also remains unknown. To address these questions, we administered ovine prolactin (oPRL) or a vehicle control to both sexes in experienced pairs of the biparental rock dove (Columba livia), after nest removal at the end of incubation. We found that oPRL promoted parental responses to novel chicks and stimulated crop growth compared to controls, consistent with other studies. However, we found that neither courtship behaviors, copulation rates nor pair maintenance differed with oPRL treatment. Across the HPG, we found oPRL had little effect on gene expression in hypothalamic nuclei, but increased expression of FSHB and hypothalamic hormone receptor genes in the pituitary. In the gonads, oPRL increased testes size and gonadotropin receptor expression, but did not affect ovarian state or small white follicle gene expression. However, the oviducts of oPRL-treated females were smaller and had lower estrogen receptor expression compared with controls. Our results highlight that some species, especially those that show multiple brooding, may be able to maintain mating behavior despite elevated prolactin. Thus, mechanisms may exist for prolactin to promote investment in parental care without concurrent inhibition of reproductive function or HPG axis activity.

A prominent body of research spanning disciplines has been focused on the potential underlying role for oxytocin in the social signatures of monogamous mating bonds. Behavioral differences between monogamous and non-monogamous vole species, putatively mediated by oxytocinergic function, constitute a key source of support for this mechanism, but it is unclear to what extent this hormone-behavior linkage extends to the primate order. In a preregistered experiment, we test if oxytocin receptor blockade affects affiliative behavior in mixed-sex pairs of Eulemur, a genus of strepsirrhine primate containing both monogamous and non-monogamous species. Inconsistent with past studies in monogamous voles or monkeys, we do not find confirmatory evidence in Eulemur that monogamous pairs affiliate more than non-monogamous pairs, nor that oxytocin receptor blockade of one pair member selectively corresponds to reduced affiliative or scent-marking behavior in monogamous species. We do, however, find exploratory evidence of a pattern not previously investigated: simultaneously blocking oxytocin receptors in both members of a monogamous pair predicts lower rates of affiliative behavior relative to controls. Our study demonstrates the value of non-traditional animal models in challenging generalizations based on model organisms, and of methodological reform in providing a potential path forward for behavioral oxytocin research.

Physical activity offers myriad benefits to health and well-being, in humans and other animals as well. In rodents, voluntary wheel running can attenuate the effects of both physical and social stressors on rodent social behavior. Whether wheel running affects rodent social behaviors per se remains less well understood. We conducted the current study to test whether home cage access to running wheels impacts the social behaviors of adult, group-housed C57BL/6J female mice during same-sex interactions with novel females. Group-housed females were either given continuous home cage running wheel access or a standard paper hut starting at weaning, and as adults, social behaviors were measured during interactions with novel females. In two cohorts, we found that 5 weeks of running wheel access during adolescence reduced the time that subject females spent investigating a novel female and also tended to reduce total ultrasonic vocalizations produced during interactions. These effects were not reversed by a 2-week period of running wheel removal but were recapitulated in a different cohort by 2 weeks of running wheel access in adulthood. Unexpectedly, we found that these effects on female social behavior were not due to wheel running per se, because females raised from weaning with immobile running wheels also showed low rates of social behaviors during same-sex interactions in adulthood. Overall, we find that the presence of a running wheel in the home cage has an enduring inhibitory influence on female social behavior during same-sex interactions, a finding that has implications for the design of studies that include same-sex interactions between female mice.

The natural alignment of animals into social dominance hierarchies produces adaptive, and potentially maladaptive, changes in the brain that influence health and behavior. Aggressive and submissive behaviors assumed by animals through dominance interactions engage stress-dependent neural and hormonal systems that have been shown to correspond with social rank. Here, we examined the impact of social dominance hierarchies established within cages of group-housed laboratory mice on expression of the stress peptide pituitary adenylate cyclase-activating polypeptide (PACAP) in areas of the extended amygdala comprising the bed nucleus of the stria terminalis (BNST) and central nucleus of the amygdala (CeA). We also quantified the impact of dominance rank on corticosterone (CORT), body weight, and behavior including rotorod and acoustic startle response. Weight-matched male C57BL/6 mice, group-housed (4/cage) starting at 3 weeks of age, were ranked as either most-dominant (Dominant), least-dominant (Submissive) or in-between rank (Intermediate) based on counts of aggressive and submissive encounters assessed at 12 weeks-old following a change in homecage conditions. We found that PACAP expression was significantly higher in the BNST, but not the CeA, of Submissive mice compared to the other two groups. CORT levels were lowest in Submissive mice and appeared to reflect a blunted response following social dominance interactions. Body weight, motor coordination, and acoustic startle were not significantly different between the groups. Together, these data reveal changes in specific neural/neuroendocrine systems that are predominant in animals of lowest social dominance rank, and implicate PACAP in brain adaptations that occur through the development of social dominance hierarchies.

Aggression is an evolutionarily conserved, adaptive component of social behavior. Studies in male mice illustrate that aggression is influenced by numerous factors including the degree to which an individual finds aggression rewarding and will work for access to attack and subordinate mice. While such studies have expanded our understanding of the molecular and circuit mechanisms of male aggression very little is known about female aggression, owed in part to limited availability of valid mouse models in females. Here we use an ethologically relevant model of male vs. female aggression by pair housing adult male and female outbred CFW mice with opposite sex cage mates. We assess reactive (defensive) aggression in the resident intruder (RI) test and appetitive (rewarding) aggression in the aggression conditioned place preference (CPP) and operant self-administration (SA) tests. Our results show dramatic sex differences in both qualitative and quantitative aspects of reactive vs. appetitive aggression. Males exhibit more wrestling and less investigative behavior during RI, find aggression rewarding and will work for access to a subordinate to attack. Females exhibit more bites, alternate between aggressive behaviors and investigative behaviors more readily during RI, however, they do not find aggression to be rewarding or reinforcing. These results establish sex differences in aggression in mice, providing an important resource for the field to better understand the circuit and molecular mechanisms of aggression in both sexes.

Same-sex affiliative relationships are common in humans and some social animals, forming one of the bases of group living. The neuropeptide vasopressin (VP) and its receptors mediate these relationships and behaviours in mammals and birds with gregarious and colonial social structures. In some species, affiliative relationships between dominant and subordinate individuals can be maintained while still retaining strict dominance hierarchies where three or more individuals interact. However, it is unclear whether triadic interaction promotes these relationships, and whether the VP system is also involved in such affiliations due to the lack of suitable animal models and experimental settings. This study addresses these questions with two experiments. In Experiment 1, two-week cohabitation among three male crows facilitated affiliative relationships in particular dyads within each triad. In Experiment 2, vasopressin 1a receptor (V1aR) antagonism disrupted affiliative behaviours and led to the resurgence of agonistic behaviours in affiliated males but not in unaffiliated ones by peripherally administering a V1aR antagonist. These findings suggest that the VP system might universally mediate same-sex affiliative relationships, despite differences in inherent aggression levels among individuals. The triadic cohabitation paradigm established here could advance our understanding of animal societies and be applied across various species, sexes, and social structures. Impact statementThis study provides evidence that a triadic interactive environment facilitates the formation of affiliative relationship between specific dominant and subordinate males in large-billed crows, with crucial involvement of vasopressin 1a receptor in maintaining this affiliation.