Differential Roles of the Two Raphe Nuclei in Amiable Social Behavior and Aggression - An Optogenetic Study

Abstract

Serotonergic mechanisms hosted by raphe nuclei have important roles in affiliative and agonistic behaviors but the separate roles of the two nuclei are poorly understood. Here we studied the roles of the dorsal (DR) and median raphe region (MRR) in aggression by optogenetically stimulating the two nuclei. Mice received three 3 min-long stimulations, which were separated by non-stimulation periods of 3 min. The stimulation of the MRR decreased aggression in a phasic-like manner. Effects were rapidly expressed during stimulations, and vanished similarly fast when stimulations were halted. No carryover effects were observed in the subsequent three trials performed at 2-day intervals. No effects on social behaviors were observed. By contrast, DR stimulation rapidly and tonically promoted social behaviors: effects were present during both the stimulation and non-stimulation periods of intermittent stimulations. Aggressive behaviors were marginally diminished by acute DR stimulations, but repeated stimulations administered over 8 days considerably decreased aggression even in the absence of concurrent stimulations, indicating the emergence of carryover effects. No such effects were observed in the case of social behaviors. We also investigated stimulation-induced neurotransmitter release in the prefrontal cortex, a major site of aggression control. MRR stimulation rapidly but transiently increased serotonin release, and induced a lasting increase in glutamate levels. DR stimulation had no effect on glutamate, but elicited a lasting increase of serotonin release. Prefrontal serotonin levels remained elevated for at least 2 h subsequent to DR stimulations. The stimulation of both nuclei increased GABA release rapidly and transiently. Thus, differential behavioral effects of the two raphe nuclei were associated with differences in their neurotransmission profiles. These findings reveal a surprisingly strong behavioral task division between the two raphe nuclei, which was associated with a nucleus-specific neurotransmitter release in the prefrontal cortex.

Keywords: GABA; aggression; dorsal raphe; glutamate; median raphe; serotonin.

FIGURE 1

Schematic representation of behavioral studies. (A) Representative photomicrographs of the placement of optic fibers and the extent of ChR2 expression in the brainstem. (B) Schematic representation of optic fibers and ChR2 expression on Paxinos (2001) plates in two mice; (C) the cage of behavioral studies, with an experimental (black, Bl6), and a stimulus mouse (white, CD1); (D) the timing of optogenetic stimulations. AQ, aqueduct; blue bars (B), the tip of optic fibres; DR, dorsal raphe; green area (A,B), virus labeling; MRR, median raphe region; OP, optic fibers; red immunochemical labeling (A), serotonergic neurons.

FIGURE 2

The behavioral effects of the optogenetic stimulation of the MRR on day 1. Mice implanted with optic fibers encountered an unfamiliar opponent in a neutral arena (see Figure 1C). (A–C) Left-hand panels: the duration of behaviors in 3-min bins; right-hand panels: the average duration of behaviors (0–21 min). (D) left-hand panel: the duration of offensive behaviors shown on a minute-by-minute basis; right-hand panel: effect of stimulation on the duration of offense. Values represent the average of the three stimulations shown on the left-hand graph of the same panel. N3, the average of the last (3rd) minutes of the non-stimulated phase that preceded stimulations; S1-3, averages of the 1st, 2nd, and 3rd min of the stimulation periods. The timing of stimulation was indicated by the color code; circles indicate controls, squares and bold lines indicate stimulated mice. Sample sizes: non-stimulated n = 8; stimulated n = 9. ^∗Significant effect of optic stimulation in post hoc tests (p < 0.05 at least). Note that there were multiple significant differences between the time-points of left-hand panels; for clarity, the significance of such differences was shown on right-hand panels only.

FIGURE 3

The behavioral effects of the optogenetic stimulation of the DR on day 1. Mice implanted with optic fibers encountered an unfamiliar opponent in a neutral arena (see Figure 1C). (A–C) left-hand panels: the duration of behaviors in 3-min bins; right-hand panels: the average duration of behaviors (min 0–21). Insert of (A), values expressed as percent of min 0–3. See text for explanations. Insert of (B), min-by-min representation of the duration of offense in controls (see text). The timing of stimulation was indicated by the color code; circles indicate controls, squares and bold lines indicate stimulated mice. Sample sizes: non-stimulated n = 7; stimulated n = 7. ^∗Significant difference between sham and real stimulation in post hoc tests; ^#Significant main effect (p < 0.05 at least in both cases); ^oTrend-level main effect (0.1 < p < 0.05).

FIGURE 4

Carryover effects of stimulations. Findings presented here show behavior observed in trials 2, 3, and 4 when all mice had a history of stimulation. The aim of this study was to investigate carryover effects (see Experimental design). (A,B) Differences in the duration of behaviors as compared to trial 1. Values show differences in the time devoted to a particular behavior expressed as the percentage of total test time. (C) The duration of offense in trials 1–4. Here data were shown separately for mice stimulated or non-stimulated within the particular trials. Each cohort of mice (indicated by roman numbers) was submitted to alternating trials of stimulation and non-stimulation. Horizontal line in (A,B), the average duration of behaviors in trial 1; Horizontal bars in (A,B) standard errors of the average duration of behaviors in trial 1; DR, dorsal raphe; MRR, median raphe region; ^∗Significant differences between trials in post hoc tests (p < 0.05 at least).

FIGURE 5

In vivo release of serotonin (A), glutamate (B), and GABA (C) in the prefrontal cortex of mice stimulated optogenetically in their raphes (median raphe region, MRR; dorsal raphe, DR). The stimulation protocol was identical with that employed for behavioral studies. Vertical blue lines, the timing of stimulations. Note that the first stimulation was started 90 min after the last basal sampling and 15 min before the fourth sampling, whereas the third stimulations started right at the beginning of the fifth fraction. Sample sizes: control n = 6; MRR stimulation n = 9; DR stimulation n = 5. Vertical columns at the right-hand side of graphs, neurotransmitter responses to the infusion of KCl into the raphes. DR, dorsal raphe; MRR, median raphe region; ^∗significant effect of stimulations compared to control levels, same time-point; ^#significant effect of KCl infusion as compared to baseline levels (the first three time points of each curve).