Impact of the circadian nuclear receptor REV-ERBα in dorsal raphe 5-HT neurons on social interaction behavior, especially social preference

Abstract

Social interaction among conspecifics is essential for maintaining adaptive, cooperative, and social behaviors, along with survival among mammals. The 5-hydroxytryptamine (5-HT) neuronal system is an important neurotransmitter system for regulating social behaviors; however, the circadian role of 5-HT in social interaction behaviors is unclear. To investigate whether the circadian nuclear receptor REV-ERBα, a transcriptional repressor of the rate-limiting enzyme tryptophan hydroxylase 2 (Tph2) gene in 5-HT biosynthesis, may affect social interaction behaviors, we generated a conditional knockout (cKO) mouse by targeting Rev-Erbα in dorsal raphe (DR) 5-HT neurons (5-HT^DR-specific REV-ERBα cKO) using the CRISPR/Cas9 gene editing system and assayed social behaviors, including social preference and social recognition, with a three-chamber social interaction test at two circadian time (CT) points, i.e., at dawn (CT00) and dusk (CT12). The genetic ablation of Rev-Erbα in DR 5-HTergic neurons caused impaired social interaction behaviors, particularly social preference but not social recognition, with no difference between the two CT points. This deficit of social preference induced by Rev-Erbα in 5-HT^DR-specific mice is functionally associated with real-time elevated neuron activity and 5-HT levels at dusk, as determined by fiber-photometry imaging sensors. Moreover, optogenetic inhibition of DR to nucleus accumbens (NAc) 5-HTergic circuit restored the impairment of social preference in 5-HT^DR-specific REV-ERBα cKO mice. These results suggest the significance of the circadian regulation of 5-HT levels by REV-ERBα in regulating social interaction behaviors.

Fig. 1. 5-HT^DR-specific REV-ERBα cKO mice exhibit alterations in daily mRNA expression and 5-HT levels in the DR.

a Design of AAV-PHP.eB-FLEX-SaCas9-U6-sgRNA for 5-HT^DR-specific REV-ERBα cKO and stereotactic injection site into the DR of control (SERT-cre) mice. b Histological image of DR in control and 5-HT^DR-specific REV-ERBα cKO mice. The scale bar indicates 20 μm. The right upper boxes represent a high-magnification image of a white arrow. The scale bar indicates 10 μm. c mRNA expression profiles of 5-HT-related genes (Tph2, Pet-1, and Rev-erbα) and circadian clock genes (Bmal1, Per2, and Cry2) in the DR were examined at two circadian time points, CT00 and CT12, in control and 5-HT^DR-specific REV-ERBα cKO mice using real-time qPCR. mRNA levels were normalized using the expression level of a housekeeping gene, TATA-box-binding protein (Tbp). d The amount of 5-HT in the DR was quantified at CT00 and CT12 in the control mouse and 5-HT^DR-specific REV-ERBα cKO mouse using HPLC-ECD. e The amount of 5-HT in the NAc was quantified at CT00 and CT12 in control mice and 5-HT^DR-specific REV-ERBα cKO mice using HPLC-ECD. In this and subsequent figures, all data are presented as the mean ± s.e.m. n = 4 for mRNA studies, and n = 3 for 5-HT level in DR and NAc, respectively. Statistical differences in control versus 5-HT^DR-specific REV-ERBα cKO mice at two circadian time points were evaluated using two-way ANOVA followed by a post hoc comparison using Tukey’s test. ns, not significant, *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 2. 5-HT^DR-specific REV-ERBα cKO mice exhibit impaired social interaction behavior, especially in social preference, but not social recognition, as revealed by the three-chamber behavior test.

a Experimental schedule for AAV injection with Rev-erbα sgRNA in the DR of the SERT-cre mouse at 8 w and social interaction behavior test at two circadian time points, CT00 vs. CT12. b Experimental scheme of the three-chamber behavior test consisting of habituation, social preference, and social recognition tests. E, empty cup; N, novel mouse; F, familiar mouse c Quantification of the social preference test in control mice at CT00 and CT12. Note that the control mouse spent significantly more time interacting with a novel mouse than an empty cup at both CT00 and CT12. In this and subsequent figures, a representative heatmap image and dotted white circle indicate the location of the cup. d Quantification of the social preference test in 5-HT^DR REV-ERBα cKO mice at CT00 and CT12. Note that the DR 5-HT REV-ERBα cKO mice showed no difference in investigation time between the empty cup and the novel mouse at both CT00 and CT12. e Quantification of the social recognition test results in control mice at CT00 and CT12. Note that the control mice spent significantly more time interacting with the novel mouse than the familiar mouse at both CT00 and CT12. f Quantification of the social recognition test in 5-HT^DR REV-ERBα cKO mice at CT00 and CT12. Note that the 5-HT^DR REV-ERBα cKO mice spent significantly more time interacting with the cup containing the novel mouse than the cup containing the familiar mouse at both CT00 and CT12. Data are presented as the mean ± s.e.m. (n = 6–8). Statistical differences between an empty cup and novel mice (c and d) and between novel mice and familiar mice (e and f) at two circadian time points were evaluated using two-way ANOVA followed by a post hoc comparison using Tukey’s test. ns, not significant, *P < 0.05, **P < 0.01 ***P < 0.001.

Fig. 3. Pharmacological manipulation of REV-ERBα activity with its antagonist (SR8278) and agonist (GSK4112).

a Experimental schedule for microinjection in the DR and social interaction behavior test at CT00 and CT12 and representative confocal image of the location of cannulation for antagonist or agonist treatment. b Quantification of the three-chamber social preference test in SR8278-treated control mice at CT00 and CT12. (left panel) The control mouse spent significantly more time interacting with the novel mouse than the empty cup at both CT00 and CT12 in the pretest. (right panel) Microinjection of SR8278-treated control mice exhibited decreased investigation time with the novel mouse at both CT00 and CT12. c Quantification of the social recognition test in SR8278-treated control mice at CT00 and CT12. The control mice with pre- and posttreatment of SR8278 spent significantly more time interacting with the cup with a novel mouse than the cup with the familiar mouse at both CT00 and CT12. d Quantification of the social preference test in GSK4112-treated 5-HT^DR REV-ERBα cKO mice at CT00 and CT12. The mice in the pretest spent significantly less time interacting with the cup containing the novel mouse than the empty cup at both CT00 and CT12. Mice receiving microinjection of GSK4112 interacted more with the novel mouse than the pretest mice at both CT00 and CT12. e Quantification of the three-chamber social recognition test in GSK4112-treated 5-HT^DR REV-ERBα cKO mice at CT00 and CT12. The WT mice pre- and posttreatment with GSK4112 spent significantly more time interacting with the cup containing the novel mouse than with the cup containing the familiar mouse at both CT00 and CT12. Data are presented as the mean ± s.e.m. (n = 4–8). Statistical differences between an empty cup and novel mouse (b, d) and between a novel mouse and familiar mouse (c and d) at two circadian time points were evaluated using two-way ANOVA followed by a post hoc comparison using Tukey’s test. ns, not significant, *P < 0.05, **P < 0.01.

Fig. 4. 5-HT^DR-specific REV-ERBα cKO mice show high neural activity in response to GCaMP7s and real-time 5-HT levels in the DR determined by iSeroSnFR.

a Experimental schedule for fiber photometry for the measurement of GCaMP7s and iSeroSnFR activity in control and 5-HTDR-specific REV-ERBα cKO mice at 6-h intervals per day (24 h). b Schematic of stereotactic injection of AAV-mediated GCaMP7s and optic fibers in the DR of control and 5-HT^DR-specific REV-ERBα cKO mice by fiber photometry recording and representative image of GCaMP7s expression and fiber location. c Freely moving 5-HT^DR-specific REV-ERBα cKO mice exhibited significantly increased neural activity at CT12 compared to control mice. The peak numbers of GCaMP7s fluorescence activity per min. d, e Ca²⁺ signals of 5-HT^DR-specific REV-ERBα cKO mice during the social preference test were increased both at CT00 (left) and CT12 (middle), as revealed by the area under the curve (AUC) analysis of each mouse (right). f Schematic of stereotactic injection of iSeroSnFR and implantation of optic fiber in the DR of control and 5-HT^DR-specific REV-ERBα cKO mice for in vivo fiber photometry recording and representative image of GCaMP7s expression and fiber location. g Freely moving 5-HT^DR-specific REV-ERBα cKO mice exhibited significantly increased 5-HT levels at CT12 compared to control mice, as determined by the peak fluorescence activity per min. h, i The 5-HT levels of 5-HT^DR-specific REV-ERBα cKO mice during the social preference test were increased at both CT00 (left) and CT12 (middle), as revealed by AUC analysis of each mouse (right). Data are presented as the mean ± s.e.m. (n = 5 for (c and g), and n = 6 for d, e, h, and i). Statistical differences in groups were evaluated using two-way ANOVA followed by a post hoc comparison using Sidak’s test for the number of peaks and Tukey’s test for AUC analysis. *P < 0.05, **P < 0.01, ***P < 0.001 for control versus cKO and ^##P < 0.01, ^####P < 0.0001 for comparing each time point to the lowest value in control.

Fig. 5. 5-HT^DR-specific REV-ERBα cKO mice show high neural activity in response to GCaMP7s and real-time 5-HT levels in the NAc.

a Experimental schedule for fiber photometry for the measurement of GCaMP7s and iSeroSnFR activity in control and 5-HT^DR-specific REV-ERBα cKO mice at 6-h intervals per day (24 h). b Schematic of stereotactic injection of AAV-mediated GCaMP7s and optic fibers in the NAc of control and 5-HT^DR-specific REV-ERBα cKO mice by fiber photometry recording. c Freely moving 5-HT^DR-specific REV-ERBα cKO mice showed significantly increased neural activity at CT12 in the NAc compared to control mice. The peak numbers of GCaMP7s fluorescence activity per min. d, e Ca²⁺ signals of 5-HT^DR-specific REV-ERBα cKO mice during interaction with a novel mouse on the social preference test were increased both at CT00 (left) and CT12 (middle), revealed by AUC analysis of each mouse (right). f Schematic of stereotactic injection of iSeroSnFR and implantation of optic fibers in the NAc of control and 5-HT^DR-specific REV-ERBα cKO mice for in vivo fiber photometry recording g Freely moving 5-HT^DR-specific REV-ERBα cKO mice exhibited significantly increased 5-HT levels at CT12 compared to control mice, as determined by the peak fluorescence activity per min. h, i The 5-HT levels of 5-HT^DR-specific REV-ERBα cKO mice during the social preference test were increased at both CT00 (left) and CT12 (middle), as revealed by AUC analysis of each mouse (right). Data are presented as the mean ± s.e.m. (n = 5 for (c and g), and n = 6 for d, e, h, and i). Statistical differences in groups were evaluated using two-way ANOVA followed by a post hoc comparison using Sidak’s test for the analysis of the number of peaks and Tukey’s test for AUC analysis. *P < 0.05, ***P < 0.001, ****P < 0.0001 for control versus cKO and ^####P < 0.0001 for each time point compared to the lowest value in control.

Fig. 6. DR-NAc circuit-specific ablation of REV-ERBα in 5-HT neurons resulted in deficits in social preference.

a Experimental schedule for AAV_retro injection at 8 w and social interaction behavior test performed at two circadian time points, CT00 vs. CT12. b Schematic of stereotactic injection of Rev-erbα sgRNA in the NAc of SERT-cre mice. c Representative image of the DR region after 4 weeks of recovery of AAV_retro-sgRNA injection. d Quantification of the social preference test in a control mouse. The control mouse spent significantly more time interacting with the novel mouse than the empty cup at both CT00 and CT12. e Quantification of the social preference test in 5-HT^DR-NAc-specific REV-ERBα cKO mice. 5- HT^DR-NAc REV-ERBα cKO mice also spent less time interacting with the novel mouse at both CT00 and CT12. f Quantification of the social recognition test in control mice at CT00 and CT12. The control mice spent significantly more time interacting with the novel mouse than the familiar mouse at both CT00 and CT12. g Quantification of the social recognition test in 5-HT^DR-NAc REV-ERBα cKO mice at CT00 and CT12. Notably, the 5-HT^DR-NAc REV-ERBα cKO mice spent significantly more time interacting with the novel mouse than the familiar mouse at both CT00 and CT12. (n = 5). Data are presented as the mean ± s.e.m. (n = 5). Statistical differences between an empty cup and novel mouse (c, d) and between a novel mouse and familiar mouse (e, f) at two circadian time points were evaluated using two-way ANOVA followed by a post hoc comparison using Tukey’s test. ns, not significant, *P < 0.05, **P < 0.01, ****P < 0.0001.

Fig. 7. Optogenetic inhibition of DR 5-HTergic projections into the NAc rescued impaired social preference in DR 5-HT REV-ERBα cKO mice.

a Schematic of AAV-DIO-NpHR and bilateral optic fiber implantation targeting the NAc. b Schematic of AAV-DIO-SSFO and bilateral optic fiber implantation targeting the NAc. c. Optogenetic inhibition of DR 5-HT in the NAc circuit in control mice significantly impaired social preference, as revealed by a decrease in the social preference index. d Optogenetic activation of DR 5-HT in the NAc circuit in control mice significantly enhanced social preference, as revealed by an increase in the social preference index. e Optogenetic inhibition of DR 5-HT in the NAc circuit in REV-ERBα cKO mice significantly rescued the impairment of social preference, as revealed by an increase in the social preference index. f Optogenetic activation of the DR 5-HT-to-NAc circuit in REV-ERBα cKO mice was not different before activation and after activation. Data are presented as the mean ± s.e.m. (n = 4). Statistical differences in groups were evaluated using two-way ANOVA followed by a post hoc comparison using Sidak’s test between lights off and lights on at each time point. n.s., not significant, *P < 0.05, ***P < 0.001.