Loss of mGluR5 in D1 Receptor-Expressing Neurons Improves Stress Coping

Abstract

The metabotropic glutamate receptor type 5 (mGluR5) has been proposed to play a crucial role in the selection and regulation of cognitive, affective, and emotional behaviors. However, the mechanisms by which these receptors mediate these effects remain largely unexplored. Here, we studied the role of mGluR5 located in D1 receptor-expressing (D1) neurons in the manifestation of different behavioral expressions. Mice with conditional knockout (cKO) of mGluR5 in D1 neurons (mGluR5^D1 cKO) and littermate controls displayed similar phenotypical profiles in relation to memory expression, anxiety, and social behaviors. However, mGluR5^D1 cKO mice presented different coping mechanisms in response to acute escapable or inescapable stress. mGluR5^D1 cKO mice adopted an enhanced active stress coping strategy upon exposure to escapable stress in the two-way active avoidance (TWA) task and a greater passive strategy upon exposure to inescapable stress in the forced swim test (FST). In summary, this work provides evidence for a functional integration of the dopaminergic and glutamatergic system to mediate control over internal states upon stress exposure and directly implicates D1 neurons and mGluR5 as crucial mediators of behavioral stress responses.

Keywords: D1; anxiety; mGluR5; memory; social behavior; stress.

Figure 1

(A) Breeding strategy for the generation of mGluR5^D1 WT and cKO mice. (B) Example image of mGluR5 expression in the cortex, striatum, and amygdala of mGluR5^D1 WT (left panel) and cKO mice (right panel). Scale bar: 500 μm. (C) Example image of D1 receptor expression (green) in the vmITC of mGluR5^D1 WT (top panels) and cKO mice (bottom panels), (D) mGluR5 expression (red), and (E) overlapped images. Scale bar: 40 μm. (F) Example image of mGluR5 expression throughout the striatum and NAc of mGluR5^D1 WT (left panel) and cKO mice (right panel). Scale bar: 500 μm. Area abbreviations: dorsal striatum (dStr), striatum (Str), lateral septum (LS), nucleus accumbens core (NAc core) and shell (NAc Sh), central amygdala (CeA), main nucleus of intercalated cells of the amygdala (vmITC), and basolateral amygdala (BLA).

Figure 2

(A) Scheme of the forced swim test (FST). (B) Time spent immobile during the test (%). mGluR5^D1 cKO mice showed higher time spent immobile than their WT littermates. Unpaired T-test, T = 3.49, p = 0.002. n = 12 mGluR5^D1 WT, n = 13 mGluR5^D1 cKO mice. ** p ≤ 0.01.

Figure 3

(A) Scheme of the open field test. (B) Time spent in the center of the open field (s). mGluR5^D1 WT and cKO mice showed similar time spent in the center of the open field. Unpaired T-test, T = 1.20, ns (p = 0.24). (C) mGluR5^D1 cKO mice travelled less distance during a 20 min test in an open field as compared to their WT littermates. Unpaired T-test, T = 2.086, p = 0.047. (D) Scheme of the elevated plus maze. (E) Time spent in the open arms of the maze (s). mGluR5^D1 WT and cKO mice showed similar anxiety-like behavior as measured with time spent in the open arms: Unpaired T-test, T = 1.22, ns (p = 0.23). (F) Visits to the open arms of the maze. mGluR5^D1 WT and cKO mice showed similar anxiety levels as measured with time spent in the open arms: Mann–Whitney test, U = 71.5, ns (p = 0.38). n = 12 mGluR5^D1 WT, n = 15 mGluR5^D1 cKO mice. * p ≤ 0.05.

Figure 4

(A) Scheme of the novel object recognition behavioral paradigm. (B) mGluR5^D1 WT and cKO mice showed similar recognition memory as measured with time spent interacting with the objects (s): Two-way ANOVA, main effect genotype: F(1, 25) = 0.001, ns (p = 0.97); main effect object: F(1, 25) = 11.41, p = 0.002; interaction effect: F(1, 25) = 0.94, ns (p = 0.34). (C) mGluR5^D1 WT and cKO mice showed similar recognition memory, as measured with the discrimination ratio. Unpaired T-test, T = 0.69, ns (p = 0.49). n = 12 mGluR5^D1 WT, n = 15 mGluR5^D1 cKO mice. ** p ≤ 0.01.

Figure 5

(A) Scheme of the social preference behavioral paradigm. (B) mGluR5^D1 WT and cKO mice showed similar social preference as measured with time spent interacting with the object vs the mouse (s): Two-way ANOVA, main effect genotype: F(1, 25) = 3.28, ns (p = 0.08); main effect object/mouse: F(1, 25) = 10.86, p = 0.002; interaction effect: F(1, 25) = 0.003, ns (p = 0.96). (C) mGluR5^D1 WT and cKO mice showed similar social preference, as measured with the discrimination ratio. Unpaired T-test, T = 0.38, ns (p = 0.70). (D) Scheme of the social novelty paradigm. (E) mGluR5^D1 WT and cKO mice showed similar social novelty as measured with time spent interacting with the familiar vs the novel mouse (s): Two-way ANOVA, main effect genotype: F(1, 25) = 1.71, ns (p = 0.20); main effect novel/familiar: F(1, 25) = 27.73, p = 0.001; interaction effect: F(1, 25) = 0.16, ns (p = 0.69). (F) mGluR5^D1 WT and cKO mice showed similar social novelty, as measured with the discrimination ratio. Unpaired T-test, T = 0.26, ns (p = 0.79); n = 12 mGluR5^D1 WT, n = 15 mGluR5^D1 cKO mice. ** p ≤ 0.01; *** p ≤ 0.001.

Figure 6

(A) Scheme of the two-way active avoidance (TWA) test. (B) mGluR5^D1 cKO mice avoided more foot shocks during the first exposure (Day 1) to the TWA test than their WT littermates. Mann–Whitney, U = 18, p = 0.014. (C) mGluR5^D1 cKO mice received less punished responses during the first exposure (Day 1) to the TWA test than their WT littermates. Mann–Whitney, U = 16.5, p = 0.010. (D) Both mGluR5^D1 WT and cKO mice increased the number of avoided foot shocks with repeated testing in the TWA (Day 2–6) to a similar extent. Two-way ANOVA, main effect genotype: F(1, 25) = 0.081, ns (p = 0.77); main effect day: F(1, 25) = 8.97, p = 0.0001; interaction effect: F(1, 25) = 0.37, ns (p = 0.82); n = 9 mGluR5^D1 WT, n = 11 mGluR5^D1 cKO mice. * p ≤ 0.05; ** p ≤ 0.01.