Effects of resocialization on post-weaning social isolation-induced abnormal aggression and social deficits in rats.

As previously shown, rats isolated from weaning develop abnormal social and aggressive behavior characterized by biting attacks targeting vulnerable body parts of opponents, reduced attack signaling, and increased defensive behavior despite increased attack counts. Here we studied whether this form of violent aggression could be reversed by resocialization in adulthood. During the first weak of resocialization, isolation-reared rats showed multiple social deficits including increased defensiveness and decreased huddling during sleep. Deficits were markedly attenuated in the second and third weeks. Despite improved social functioning in groups, isolated rats readily showed abnormal features of aggression in a resident-intruder test performed after the 3-week-long resocialization. Thus, post-weaning social isolation-induced deficits in prosocial behavior were eliminated by resocialization during adulthood, but abnormal aggression was resilient to this treatment. Findings are compared to those obtained in humans who suffered early social maltreatment, and who also show social deficits and dysfunctional aggression in adulthood.

The temporal dynamics of the effects of monoacylglycerol lipase blockade on locomotion, anxiety, and body temperature.

Studies with the monoacylglycerol lipase blocker JZL184 have suggested that enhanced 2-arachidonoylglycerol signaling suppresses locomotion, lowers body temperature, and decreases anxiety. Although the neurochemical effects of JZL184 develop within 30 min, its behavioral and autonomic effects have been studied much later. To clarify temporal dynamics, we studied the effects of intraperitoneal injections of JZL184 in mice on home-cage locomotion and body temperature for 120 min using in-vivo biotelemetry. We also studied the effects of 4, 8, and 16 mg/kg JZL184 in the open field and elevated plus maze at various time points. In the home cage, JZL184 blunted injection-induced body temperature increases but exerted no long-term effects. Vehicle injections increased the duration of rapid movements whereas the duration of motionless periods was decreased, a pattern also abolished by JZL184. Although the highest dose exerted a mild long-term effect on the relative duration of motionless periods, JZL184 seemed to have phasic rather than tonic effects in the home cage. By contrast, open field and plus maze behavior was affected 80 and 120 min but not 40 min after treatments, which may indicate tonic rather than phasic effects in these tests. Our findings confirm earlier reports of a mild anxiolytic effect of JZL184, but surprisingly, the compound markedly and dose dependently increased locomotion in the open field in both CD1 and C57BL/6J mice. These findings are difficult to reconcile at present, but suggest that the effects of monoacylglycerol lipase inhibition are more complex than previously believed and may depend strongly on as yet unidentified factors such as environmental conditions, the time of testing, species/strains, etc.

Post-weaning social isolation induces abnormal forms of aggression in conjunction with increased glucocorticoid and autonomic stress responses.

We showed earlier that social isolation from weaning (a paradigm frequently used to model social neglect in children) induces abnormal forms of attack in rats, and assumed that these are associated with hyperarousal. To investigate this hypothesis, we deprived rats of social contacts from weaning and studied their behavior, glucocorticoid and autonomic stress responses in the resident-intruder paradigm at the age of 82 days. Social isolation resulted in abnormal attack patterns characterized by attacks on vulnerable targets, deficient social communication and increased defensive behaviors (defensive upright, flight, freezing). During aggressive encounters, socially deprived rats rapidly switched from one behavior to another, i.e. showed an increased number of behavioral transitions as compared to controls. We tentatively term this behavioral feature "behavioral fragmentation" and considered it a form of behavioral arousal. Basal levels of plasma corticosterone regularly assessed by radioimmunoassay between 27 and 78 days of age were not affected. In contrast, aggression-induced glucocorticoid responses were approximately doubled by socially isolation. Diurnal oscillations in heart rate assessed by in vivo biotelemetry were not affected by social isolation. In contrast, the aggression-induced increase in heart rate was higher in socially isolated than in socially housed rats. Thus, post-weaning social isolation induced abnormal forms of aggression that developed on the background of increased behavioral, endocrine and autonomic arousal. We suggest that this paradigm may be used to model aggression-related psychopathologies associated with hyperarousal, particularly those that are triggered by adverse rearing conditions.

Brain mechanisms involved in predatory aggression are activated in a laboratory model of violent intra-specific aggression.

Callous-unemotional violence associated with antisocial personality disorder is often called 'predatory' because it involves restricted intention signaling and low emotional/physiological arousal, including decreased glucocorticoid production. This epithet may be a mere metaphor, but may also cover a structural similarity at the level of the hypothalamus where the control of affective and predatory aggression diverges. We investigated this hypothesis in a laboratory model where glucocorticoid production is chronically limited by adrenalectomy with glucocorticoid replacement (ADXr). This procedure was proposed to model important aspects of antisocial violence. Sham and ADXr rats were submitted to resident/intruder conflicts, and the resulting neuronal activation patterns were investigated by c-Fos immunocytochemistry. In line with earlier findings, the share of attacks aimed at vulnerable targets (head, throat and belly) was dramatically increased by ADXr, while intention signaling by offensive threats was restricted. Aggressive encounters activated the mediobasal hypothalamus, a region involved in intra-specific aggression, but sham and ADXr rats did not differ in this respect. In contrast, the activation of the lateral hypothalamus that is tightly involved in predatory aggression was markedly larger in ADXr rats; moreover, c-Fos counts correlated positively with the share of vulnerable attacks and negatively with social signaling. Glucocorticoid deficiency increased c-Fos activation in the central amygdala, a region also involved in predatory aggression. In addition, activation patterns in the periaqueductal gray - involved in autonomic control - also resembled those seen in predatory aggression. These findings suggest that antisocial and predatory aggression are not only similar but are controlled by overlapping neural mechanisms.

Structural and functional alterations in the prefrontal cortex after post-weaning social isolation: relationship with species-typical and deviant aggression.

Although the inhibitory control of aggression by the prefrontal cortex (PFC) is the cornerstone of current theories of aggression control, a number of human and laboratory studies showed that the execution of aggression increases PFC activity; moreover, enhanced activation was observed in aggression-related psychopathologies and laboratory models of abnormal aggression. Here, we investigated these apparently contradictory findings in the post-weaning social isolation paradigm (PWSI), an established laboratory model of abnormal aggression. When studied in the resident-intruder test as adults, rats submitted to PWSI showed increased attack counts, increased share of bites directed towards vulnerable body parts of opponents (head, throat, and belly) and reduced social signaling of attacks. These deviations from species-typical behavioral characteristics were associated with a specific reduction in the thickness of the right medial PFC (mPFC), a bilateral decrease in dendritic and glial density, and reduced vascularization on the right-hand side of the mPFC. Thus, the early stressor interfered with mPFC development. Despite these structural deficits, aggressive encounters enhanced the activation of the mPFC in PWSI rats as compared to controls. A voxel-like functional analysis revealed that overactivation was restricted to a circumscribed sub-region, which contributed to the activation of hypothalamic centers involved in the initiation of biting attacks as shown by structural equation modeling. These findings demonstrate that structural alterations and functional hyperactivity can coexist in the mPFC of rats exposed to early stressors, and suggest that the role of the mPFC in aggression control is more complex than suggested by the inhibitory control theory.

Neural mechanisms of predatory aggression in rats-implications for abnormal intraspecific aggression.

Our recent studies showed that brain areas that are activated in a model of escalated aggression overlap with those that promote predatory aggression in cats. This finding raised the interesting possibility that the brain mechanisms that control certain types of abnormal aggression include those involved in predation. However, the mechanisms of predatory aggression are poorly known in rats, a species that is in many respects different from cats. To get more insights into such mechanisms, here we studied the brain activation patterns associated with spontaneous muricide in rats. Subjects not exposed to mice, and those which did not show muricide were used as controls. We found that muricide increased the activation of the central and basolateral amygdala, and lateral hypothalamus as compared to both controls; in addition, a ventral shift in periaqueductal gray activation was observed. Interestingly, these are the brain regions from where predatory aggression can be elicited, or enhanced by electrical stimulation in cats. The analysis of more than 10 other brain regions showed that brain areas that inhibited (or were neutral to) cat predatory aggression were not affected by muricide. Brain activation patterns partly overlapped with those seen earlier in the cockroach hunting model of rat predatory aggression, and were highly similar with those observed in the glucocorticoid dysfunction model of escalated aggression. These findings show that the brain mechanisms underlying predation are evolutionarily conservative, and indirectly support our earlier assumption regarding the involvement of predation-related brain mechanisms in certain forms of escalated social aggression in rats.

The neural background of hyper-emotional aggression induced by post-weaning social isolation.

Post-weaning social isolation in rats is believed to model symptoms of early social neglect-induced externalizing problems including aggression-related problems. We showed earlier that rats reared in social isolation were hyper-aroused during aggressive contacts, delivered substantially more attacks that were poorly signaled and were preferentially aimed at vulnerable body parts of opponents (head, throat and belly). Here we studied the neural background of this type of aggression by assessing the expression of the activation marker c-Fos in 22 brain areas of male Wistar rats submitted to resident-intruder conflicts. Post-weaning social isolation readily produced the behavioral alterations noticed earlier. Social isolation significantly increased the activation of brain areas that are known to directly or indirectly control inter-male aggression. Particularly, the medial and lateral orbitofrontal cortices, anterior cingulate cortex, bed nucleus of the stria terminalis, medial and basolateral amygdala, hypothalamic attack area, hypothalamic paraventricular nucleus and locus coeruleus showed increased activations. This contrasts our earlier findings obtained in rats with experimentally induced hypoarousal, where abnormal attack patterns were associated with over-activated central amygdala, lateral hypothalamus, and ventrolateral periaqueductal gray that are believed to control predatory attacks. We have observed no similar activation patterns in rats socially isolated from weaning. In summary, these findings suggest that despite some phenotypic similarities, the neuronal background of hypo and hyperarousal-associated abnormal forms of aggression are markedly different. While the neuronal activation patterns induced by normal rivalry and hypoarousal-driven aggression are qualitative different, hyperarousal-associated aggression appears to be an exaggerated form of rivalry aggression.

Temporal changes in c-Fos activation patterns induced by conditioned fear.

Mechanisms underlying shock-induced conditioned fear - a paradigm frequently used to model posttraumatic stress disorder, PTSD - are usually studied shortly after shocks. Some of the brain regions relevant to conditioned fear were activated in all the c-Fos studies published so far, but the overlap between the activated regions was small across studies. We hypothesized that discrepant findings were due to dynamic neural changes that followed shocks, and a more consistent picture would emerge if consequences were studied after a longer interval. Therefore, we exposed rats to a single session of footshocks and studied their behavioral and neural responses one and 28 days later. The neuronal activation marker c-Fos was studied in 24 brain regions relevant for conditioned fear, e.g. in subdivisions of the prefrontal cortex, hippocampus, amygdala, hypothalamic defensive system, brainstem monoaminergic nuclei and periaqueductal gray. The intensity of conditioned fear (as shown by the duration of contextual freezing) was similar at the two time-points, but the associated neuronal changes were qualitatively different. Surprisingly, however, Multiple Regression Analyses suggested that conditioned fear-induced changes in neuronal activation patterns predicted the duration of freezing with high accuracy at both time points. We suggest that exposure to electric shocks is followed by a period of plasticity where the mechanisms that sustain conditioned fear undergo qualitative changes. Neuronal changes observed 28 days but not 1 day after shocks were consistent with those observed in human studies performed in PTSD patients.

Neural inputs of the hypothalamic "aggression area" in the rat.

A part of the mediobasal hypothalamus (known as hypothalamic attack area) plays a central role in the control of aggressive behavior as its electrical stimulation reliably and rapidly elicits biting attacks in cats and rodents. The efferent connections of this brain region were described in rats, but afferent pathways were not investigated so far. We injected the retrograde tracer cholera toxin B subunit into the mediobasal hypothalamus of male Wistar rats and studied the distribution of labeled cells by immunohistochemical method. The retrograde labeling outlined three continuous, distinct afferent cell populations: (i) a telencephalic midline "plate" containing the orbitofrontal - medial prefrontal - septal regions which ends in the bed nucleus of stria terminalis; (ii) a temporal column including the medial amygdala, amigdalohippocampal area and subiculum; (iii) a diffuse column along the medial hypothalamus which ends in the posterior hypothalamic nucleus. Sparse labeling was present in brainstem nuclei, except for the lateral parabrachial nucleus that provides a significant input. The projections of the medial prefrontal cortex to the hypothalamic attack area indicate a direct, earlier undescribed pathway with marked importance in the control of aggressive behavior. Similarly, we identified several brain regions which send very significant projections to the hypothalamic attack area but their importance in the control of aggressive behavior are nearly unknown. The comparison of the present and earlier findings shows that efferent and afferent connections overlap in many regions to a significant extent, suggesting that reverberating circuits are important in the control of aggressive behavior.

Substance P neurotransmission and violent aggression: the role of tachykinin NK(1) receptors in the hypothalamic attack area.

Substance P and its tachykinin NK(1) receptors are highly expressed in brain regions involved in emotional control. We recently showed that NK(1)-mediated substance P neurotransmission is deeply involved in the control of aggressiveness. To get further insights into the NK(1) receptor/aggression relationship, we studied the role of NK(1) receptor-expressing neurons of the hypothalamic attack area, the only brain region in rats from which biting attacks can reliably be elicited by both electrical and neurochemical stimulation. We show here that the hypothalamic attack area preferentially expresses the NK(1) type of tachykinin receptors. When such neurons were lesioned by substance P-conjugated saporin (SP-sap) infused into the hypothalamic attack area, violent attacks were dramatically reduced, whereas milder forms of aggression (soft bites and offensive threats) remained unaltered. The lesions were neuron type-specific as SP-sap lesions markedly reduced NK(1) staining without significantly affecting total cell counts. NK(1) staining in the neighboring lateral hypothalamus was not affected, which confirms the spatial specificity of the lesion. Surprisingly, the lesions also reduced anxiety-like behavior in the elevated plus-maze. This effect is likely explained by the extensive connections of the hypothalamic attack area with brain regions involved in the control of anxiety. The present findings suggest that violent and milder forms of attack are differentially controlled. NK(1) receptor-expressing neurons of the hypothalamic attack area are tightly and specifically involved in the former but not in the latter. Our data also raise the possibility of a coordinated control of violent attacks and anxiety by the same NK(1)-expressing neurons.