Brain 5-HT deficiency increases stress vulnerability and impairs antidepressant responses following psychosocial stress

Abstract

Brain serotonin (5-HT) deficiency and exposure to psychosocial stress have both been implicated in the etiology of depression and anxiety disorders, but whether 5-HT deficiency influences susceptibility to depression- and anxiety-like phenotypes induced by psychosocial stress has not been formally established. Most clinically effective antidepressants increase the extracellular levels of 5-HT, and thus it has been hypothesized that antidepressant responses result from the reversal of endogenous 5-HT deficiency, but this hypothesis remains highly controversial. Here we evaluated the impact of brain 5-HT deficiency on stress susceptibility and antidepressant-like responses using tryptophan hydroxylase 2 knockin (Tph2KI) mice, which display 60-80% reductions in brain 5-HT. Our results demonstrate that 5-HT deficiency leads to increased susceptibility to social defeat stress (SDS), a model of psychosocial stress, and prevents the fluoxetine (FLX)-induced reversal of SDS-induced social avoidance, suggesting that 5-HT deficiency may impair antidepressant responses. In light of recent clinical and preclinical studies highlighting the potential of inhibiting the lateral habenula (LHb) to achieve antidepressant and antidepressant-like responses, we also examined whether LHb inhibition could achieve antidepressant-like responses in FLX-insensitive Tph2KI mice subjected to SDS. Our data reveal that using designer receptors exclusively activated by designer drugs (DREADDs) to inhibit LHb activity leads to reduced SDS-induced social avoidance behavior in both WT and Tph2KI mice. This observation provides additional preclinical evidence that inhibiting the LHb might represent a promising alternative therapeutic approach under conditions in which selective 5-HT reuptake inhibitors are ineffective.

Keywords: antidepressants; habenula; serotonin; stress.

Fig. 1.

Susceptibility of WT and Tph2KI mice to social defeat stress. Social preference scores of WT and Tph2KI mice following either 7 d (A) or 10 d (B) of SDS. *P < 0.05 by Tukey’s post hoc analysis compared with the Tph2KI control. ** denotes the main effect of SDS by two-way ANOVA, P < 0.05. n = 19–21 per group in A and n = 10–15 per group for B. The error bars represent the SEM.

Fig. 2.

Signaling consequences of social defeat in the nucleus accumbens of WT and Tph2KI mice. Quantification and representative images of Western blots showing the levels of pGSK3β (A) or pERK (B) in the nucleus accumbens following 7 d of SDS in WT and Tph2KI mice. Quantification and representative images of Western blots from cell fractionation experiments documenting the levels of β-catenin in the nucleus (C) and cytosol (D) in the nucleus accumbens of WT and Tph2KI mice following 7 d of SDS. n = 5 per group. *P < 0.05 vs. WT control by Tukey’s post hoc test. The error bars represent the SEM.

Fig. 3.

Effects of 7 d of social defeat on gene expression in the amygdala, frontal cortex, and nucleus accumbens of WT and Tph2KI mice. Quantification of the mRNA levels of DVL-1 (A), DVL-2 (B), and DVL-3 (C) in the medial prefrontal cortex of control WT and Tph2KI mice and in animals exposed to 7 d of SDS (SDS-7). Quantification of the mRNA levels of DVL-1 (D), DVL-2 (E), and DVL-3 (F) in the amygdala of control WT and Tph2KI mice and in animals exposed to 7 d of SDS. Quantification of the mRNA levels of DVL-1 (G), DVL-2 (H), and DVL-3 (I) in the nucleus accumbens of control WT and Tph2KI mice and in animals exposed to 7 d of SDS. n = 5 per group. ^$P < 0.05 vs. all other groups by Tukey’s post hoc tests. ^P < 0.05 vs. WT SDS and Tph2KI control groups by Tukey’s post hoc tests. *P < 0.05 vs. WT control by Tukey’s post hoc test. ** indicates the significant main effect of SDS by two-way ANOVA. X indicates significant genotype-by-stress interaction by two-way ANOVA (P < 0.05). The error bars represent the SEM.

Fig. 4.

Effects of 10 d of social defeat on gene expression in the amygdala, frontal cortex, and nucleus accumbens of WT and Tph2KI mice. Quantification of the mRNA levels of DVL-1 (A), DVL-2 (B), and DVL-3 (C) in the medial prefrontal cortex of control WT and Tph2KI mice and in animals exposed to 10 d of SDS (SDS-10). Quantification of the mRNA levels of DVL-1 (D), DVL-2 (E), and DVL-3 (F) in the amygdala of control WT and Tph2KI mice and in animals exposed to 10 d of SDS. Quantification of the mRNA levels of DVL-1 (G), DVL-2 (H), and DVL-3 (I) in the nucleus accumbens of control WT and Tph2KI mice and in animals exposed to 10 d of SDS. n = 10 per group. # indicates the significant main effect of genotype by two-way ANOVA (P < 0.05). ** indicates the significant main effect of SDS by two-way ANOVA (P < 0.05). The error bars represent SEM.

Fig. 5.

Antidepressant-like responses of WT and Tph2KI mice to fluoxetine following psychosocial stress. (A) Quantification of social preference behavior following chronic fluoxetine treatment in SDS-exposed WT and Tph2KI animals. (B) Quantification of social preference behavior following chronic fluoxetine treatment in control WT and Tph2KI animals. n = 13–20 per group in A and n = 9–11 per group in B. *P < 0.05 compared with WT control by Tukey’s post hoc test. The error bars represent the SEM.

Fig. 6.

Antidepressant-like effects of lateral habenula inhibition. (A) A representative image showing a stereotactic injection of AAV8-hSyn-M4D(Gi)DREADD-mCherry into the lateral habenula. (B) CNO administration to M4D(Gi)DREADD-expressing animals leads to a significant increase in social preference in both WT and Tph2KI mice. (C) A representative image showing a missed stereotactic injection, which resulted in M4D(Gi)DREADD expression in the hippocampus, not the lateral habenula. (D) CNO administration to improperly localized M4D(Gi)DREADD-expressing animals did not significantly affect social preference. **P < 0.05 by two-way ANOVA (significant main effect of CNO). n = 6–11 per group for B and n = 13 saline and 8 CNO for D. The error bars represent the SEM.