Dopamine and serotonin signaling during two sensitive developmental periods differentially impact adult aggressive and affective behaviors in mice

Abstract

Pharmacologic blockade of monoamine oxidase A (MAOA) or serotonin transporter (5-HTT) has antidepressant and anxiolytic efficacy in adulthood. Yet, genetically conferred MAOA or 5-HTT hypoactivity is associated with altered aggression and increased anxiety/depression. Here we test the hypothesis that increased monoamine signaling during development causes these paradoxical aggressive and affective phenotypes. We find that pharmacologic MAOA blockade during early postnatal development (P2-P21) but not during peri-adolescence (P22-41) increases anxiety- and depression-like behavior in adult (>P90) mice, mimicking the effect of P2-21 5-HTT inhibition. Moreover, MAOA blockade during peri-adolescence, but not P2-21 or P182-201, increases adult aggressive behavior, and 5-HTT blockade from P22-P41 reduced adult aggression. Blockade of the dopamine transporter, but not the norepinephrine transporter, during P22-41 also increases adult aggressive behavior. Thus, P2-21 is a sensitive period during which 5-HT modulates adult anxiety/depression-like behavior, and P22-41 is a sensitive period during which DA and 5-HT bi-directionally modulate adult aggression. Permanently altered DAergic function as a consequence of increased P22-P41 monoamine signaling might underlie altered aggression. In support of this hypothesis, we find altered aggression correlating positively with locomotor response to amphetamine challenge in adulthood. Proving that altered DA function and aggression are causally linked, we demonstrate that optogenetic activation of VTA DAergic neurons increases aggression. It therefore appears that genetic and pharmacologic factors impacting dopamine and serotonin signaling during sensitive developmental periods can modulate adult monoaminergic function and thereby alter risk for aggressive and emotional dysfunction.

Fig. 1. Increased adult anxiety/depression-like behavior after early postnatal 5-HTT or MAOA blockade

Anxiety/depression-like behavior was assessed in mice using the open field test (a, b), the novelty suppressed feeding test (c, d), and the shock escape test (e, f). FLX or CLO treatment from P2-P21 reduced the time ambulating (a) and the time rearing (b) when compared to control mice treated with VEH from P2-P21. FLX or CLO treatment from P2-P21 increased the latency to feed when compared to control mice treated with VEH from P2-P21 (c). No effect of treatment, period, or treatment x period interaction was detected for the weight loss after 24 hours of food deprivation (d). FLX or CLO treatment from P2-P21 increased the latency to escape when compared to control mice treated with VEH from P2-P21 (e). No effect of treatment, period, or treatment x period interaction was detected for activity in the shuttle box before the onset of shock (f). (n = 17–32 mice per group). (*p < 0.05; **p < 0.01, ***p < 0.001).

Fig. 2. Altered aggression after developmental 5-HTT or MAOA blockade

Isolation induced aggressive behavior was assessed in mice by scoring the time spent mounting, tail rattling, or biting during a 10-minute encounter. (a) Aggressive behavior was increased in mice treated with CLO from P22-P41 when compared to control mice treated with VEH from P22-P41. Mice treated with FLX displayed reduced aggression when compared to VEH-treated control mice. Of note, FLX treated mice did not display any tail rattling or biting behavior. (n = 7–16 pairs per group). (b) Aggressive behavior was unaffected by transient CLO-treatment from P182-P201 (n = 7 pairs per group). (@ indicates main effect of treatment: p < 0.05; **p < 0.01).

Fig. 3. Altered monoamine and –metabolite levels during and after peri-adolescent MAOA blockade

Mice were injected daily with CLO from P22-P41 and tissue monoamine and monoamine-metabolite levels were assessed on P23 (a, c, e, g, i, k) and P42 (b, d, f, h, j, l) by high performance liquid chromatography. 5-HT (a, b), 5-HIAA (c, d), and NE (e, f) were measured in the brainstem (a–f). DA (g, h), DOPAC (i, j), and HVA (k, l) levels were measured in the striatum (g–l). Monoamine levels were increased and monoamine-metabolite levels were decreased during CLO treatment when compared to VEH treatment as indicated. (n = 6–8 mice per group;). (**p < 0.01; ***p < 0.001).

Fig. 4. Increased adult aggressive behavior after peri-adolescent DAT blockade

Isolation induced aggressive behavior was assessed in mice by scoring the time spent mounting, tail rattling, or biting during a 10 minute encounter (a, b). Aggressive behavior was not altered in mice treated with DMI or DMI+FLX from P22-P41 when compared to VEH treated control mice (a). Aggressive behavior was increased in mice treated with GBR from P22-P41 when compared to VEH treated control mice (b). (n = 5-15 per group) (*p < 0.05).

Fig. 5. Altered behavioral response to amphetamine in adulthood after peri-adolescent MAOA, 5-HTT, or DAT blockade

Behavioral amphetamine response was assessed using the open field. (a) Amphetamine injection (AMP, 3 mg/kg, i.p.) induced locomotor hyperactivity in ADO-VEH and ADO-CLO treated mice. Peri-adolescent CLO-treatment altered the response to AMP (indicated significance refers to peri-adolescent treatment x time interaction). (b) Post-injection activity demonstrates increased behavioral response to AMP in ADO-CLO when compared to ADO-VEH. (c) Transient adult CLO treatment did not impact behavioral response to AMP challenge. (d) ADO-FLX treatment abolished the behavioral response to AMP challenge. (e) ADO-GBR treatment increased the behavioral response to AMP challenge. (n = 5-15 per group) (*p < 0.05, ***p < 0.001).

Fig. 6. Optogenetic stimulation of DAergic VTA neurons increases aggression

DAT^IRESCre:ai32 double mutant mice expressing ChR2-eYFP in VTA neurons as visualized by immunohistochemistry against eYFP (a) and autofluorescence (b). A fiberoptic cable track demarks the position of an implant with its tip just dorsal of the VTA (indicated by a star, b). Tip locations were assessed histologically after behavioral experiments were concluded (c). DAT^IRESCre:ai32 double transgenic mice and single mutant controls were co-housed in mixed pairs and isolation induced aggressive behavior was assessed by scoring the time spent mounting, tail rattling, or biting during a 10 minute encounter (d). Only one mouse in a pair was stimulated (blue). Aggressive behavior was increased in pairs when DAT^IRESCre:ai32 double mutant mice were stimulated when compared to pairs where the single mutant controls were stimulated (d). (n = 11-18 per group) (*p < 0.05).