Patterns of violent aggression-induced brain c-fos expression in male mice selected for aggressiveness

Neural circuitry of play fighting in golden hamsters

In hamsters, play fighting matures gradually into adult aggression. As these two behaviors share many similarities in this species, we predicted that a single neural circuitry controls their offensive component. The goal of the present study was to identify neural systems associated with offensive play fighting in male juvenile golden hamsters. The neural circuitry related to this behavior was identified through quantification of c-Fos immunolabeling. We also looked for vasopressin cells possibly associated with play fighting. We found that areas previously associated with offensive aggression in adult hamsters, including the ventrolateral hypothalamus, the medial amygdala, and the bed nucleus of the stria terminalis, also showed enhanced c-Fos expression after play fighting. In addition, vasopressin neurons in the nucleus circularis and the medial division of the supraoptic nucleus expressed enhanced c-Fos immunolabeling in juveniles after play fighting, as previously reported in adult hamsters after aggression. Finally, enhanced c-Fos expression associated with play fighting was also found in areas previously unexplored in adult hamsters, such as the prefrontal cortex. Together, our results support the hypothesis of a single core neural circuitry controlling the offensive components of play fighting and adult aggression throughout puberty in hamsters.

Photoperiod alters central distribution of estrogen receptor alpha in brain regions that regulate aggression

Testosterone or its metabolite, estrogen, regulates aggression in males of many mammalian species. Because plasma testosterone levels are typically positively correlated with both aggression and reproduction, aggression is expected to be higher when males are in reproductive condition. However, in some photoperiodic species such as Siberian hamsters (Phodopus sungorus), males are significantly more aggressive in short day lengths when the testes are regressed and circulating testosterone concentrations are reduced. These results led to the formation of the hypothesis that aggression is modulated independently of circulating steroids in Siberian hamsters. Thus, recent studies have been designed to characterize the role of other neuroendocrine factors in modulating aggression. However, aggression may be mediated by testosterone or estrogen despite basal concentrations of these steroids by increasing sensitivity to steroids in specific brain regions. Consistent with this hypothesis, we found that males housed under short days have increased expression of estrogen receptor alpha in the bed nucleus of the stria terminalis, medial amygdala, and central amygdala. Neural activation in response to an aggressive encounter was also examined across photoperiod.

Development of violence in mice through repeated victory along with changes in prefrontal cortex neurochemistry

Recent reviews on the validity of rodent aggression models for human violence have addressed the dimension of pathological, maladaptive, violent forms of aggression in male rodent aggressive behaviour. Among the neurobiological mechanisms proposed for the regulation of aggressive behaviour in its normal and pathological forms, serotonin plays a major role. However, the results on the detailed mechanism are still confusing and controversial, mainly because of difficulties in extrapolating from rodent to human psychopathological behaviour. Our aim was to investigate the involvement of serotonin in pathological aggression. We subjected mice genetically selected for high (SAL, TA, NC900 lines) and low (LAL, TNA, NC100) aggression levels to a repeated resident-intruder experience (RRI mice) or to handling as a control procedure (CTR mice). Pathological aggression parameters we recorded were aggression towards females and lack of communication between the resident and its opponent. In the same mice, we measured the monoamine levels in the prefrontal cortex, a brain region strongly involved in the regulation of motivated behaviour. Our results show that SAL mice augmented their proneness to attack and showed the most pathological phenotype, with disregard of the opponent's sex, high territorial behavioural patterns, and low sensitivity to signals of subordination. In contrast, TA and NC900 augmented their proneness to attack and low discrimination of the opponent's signals, without showing offence towards females. After repeated resident-intruder experience, serotonin levels in the prefrontal cortex were significantly lower in SAL than in LAL whereas dopamine turnover was significantly higher, compared to CTR mice. Serotonin turnover was significantly reduced in all RRI mice, with no strain differences. Noradrenaline was significantly lower in aggressive mice of the TA and NC900 lines compared to their low-aggressive counterparts, with no effect of the repeated resident-intruder experience. We conclude that social experience changes prefrontal cortex neurochemistry and elicits pathologically aggressive phenotypes.

Neurobiological Mechanisms of Aggression and Stress Coping: A Comparative Study in Mouse and Rat Selection Lines

Aggression causes major health and social problems and constitutes a central problem in several psychiatric disorders. There is a close relationship between the display of aggression and stress coping strategies. In order to gain more insight into biochemical pathways associated with aggression and stress coping, we assessed behavioral and neurobiological responses in two genetically selected rodent models, namely wild house mice selectively bred for a short (SAL) and long (LAL) attack latency and Wistar rats bred for high (HAB) or low (LAB) anxiety-related behavior. Compared to their line counterparts, the SAL mice and the LAB rats display a high level of intermale aggression associated with a proactive coping style. Both the SAL mice and the LAB rats show a reduced hypothalamic-pituitary-adrenal (HPA) axis response to non-social stressors. However, when exposed to social stressors (resident-intruder, sensory contact), SAL mice show an attenuated HPA response, whereas LAB rats show an elevated HPA response. In both rodent lines, the display of aggression is associated with high neuronal activation in the central amygdala, but reduced neuronal activation in the lateral septum. Furthermore, in the lateral septum, SAL mice have a reduced vasopressinergic fiber network, and LAB rats show a decreased vasopressin release during the display of aggression. Moreover, the two lines show several indications of an increased serotonergic neurotransmission. The relevance of these findings in relation to high aggression and stress coping is discussed. In conclusion, exploring neurobiological systems in animals sharing relevant behavioral characteristics might be a useful approach to identify general mechanisms of action, which in turn can improve our understanding of specific behavioral symptoms in human psychiatric disorders.

Chronic intracerebral prolactin attenuates neuronal stress circuitries in virgin rats

Prolactin (PRL) has been shown to promote maternal behaviour, and to regulate neuroendocrine and emotional stress responses. These effects appear more important in the peripartum period, when the brain PRL system is highly activated. Here, we studied the mechanisms that underlie the anti-stress effects of PRL. Ovariectomized, estradiol-substituted Wistar rats were implanted with an intracerebroventricular cannula and treated with ovine PRL (0.01, 0.1 or 1 microg/h; 5 days via osmotic minipumps) or vehicle, and their responses to acute restraint stress was assessed. Chronic PRL treatment exerted an anxiolytic effect on the elevated plus-maze, and attenuated the acute restraint-induced rise in plasma adrenocorticotropin, corticosterone and noradrenaline. At the neuronal level, in situ hybridization revealed PRL effects on the expression patterns of the immediate-early gene c-fos and corticotropin-releasing factor (CRF). Under basal conditions, PRL significantly reduced c-fos mRNA expression within the central amygdala. In response to restraint, the expression of both c-fos mRNA and protein and of CRF mRNA was decreased in the parvocellular part of the paraventricular nucleus (PVN) of PRL-treated compared with vehicle-treated animals. In conclusion, our data demonstrate that chronic elevation of PRL levels within the brain results in reduced neuronal activation within the hypothalamus, specifically within the PVN, in response to an acute stressor. Thus, PRL acting at various relevant brain regions exerts profound anxiolytic and anti-stress effects, and is likely to contribute to the attenuated stress responsiveness found in the peripartum period, when brain PRL levels are physiologically upregulated.

The activation of prefrontal cortical neurons in aggression—A double labeling study

Violence is associated with prefrontal deficits in humans, suggesting that this brain area inhibits aggressiveness. Its role, however, remains controversial, as certain subdivisions of the prefrontal cortex become activated by fights in rodents. Disparate human findings also show that this area is acutely activated by aggression under certain conditions. We explored prefrontal neuronal activation patterns in resident rats exposed to psychosocial (sensory contact with the intruder) and aggressive encounters. Both psychosocial and aggressive encounters increased c-Fos activation in the prelimbic (PrL), anterior cingular (Cg1), agranular insular (AI), ventral (VO) and lateral orbital (LO) cortices. The infralimbic (IL) and medial orbital (MO) cortices were activated significantly by aggressive encounters only. No other prefrontal regions were activated by psychosocial or aggressive encounters. The overwhelming majority of activated cells were pyramidal (glutamatergic) cells in the Cg1, IL, PrL, MO, and VO, whereas interneuron and pyramidal cell activation was similar in AI and LO. When rats showed violent aggression, the activation of GABAergic inhibitory cells decreased in these two, and two other areas (IL and MO). Notably, the latter two areas appeared to be specifically involved in aggressive behavior. The change occurred in a recently developed model of violent aggression. In this model, pyramidal cell activation in the above mentioned four areas (IL, MO, AI, and LO) predicted over 95% of variation in attack counts in general and violent attacks in particular. Based on these data, we present a tentative hypothesis on the involvement of the prefrontal cortex in the control of aggression.