Subcortical serotonin 5HT2c receptor-containing neurons sex-specifically regulate binge-like alcohol consumption, social, and arousal behaviors in mice

Abstract

Binge alcohol consumption induces discrete social and arousal disturbances in human populations that promote increased drinking and accelerate the progression of Alcohol Use Disorder. Here, we show in a mouse model that binge alcohol consumption disrupts social recognition in females and potentiates sensorimotor arousal in males. These negative behavioral outcomes were associated with sex-specific adaptations in serotonergic signaling systems within the lateral habenula (LHb) and the bed nucleus of the stria terminalis (BNST), particularly those related to the receptor 5HT_2c. While both BNST and LHb neurons expressing this receptor display potentiated activation following binge alcohol consumption, the primary causal mechanism underlying the effects of alcohol on social and arousal behaviors appears to be excessive activation of LHb_5HT2c neurons. These findings may have valuable implications for the development of sex-specific treatments for mood and alcohol use disorders targeting the brain's serotonin system.

Fig. 1. In-vivo LHb_5HT2c and BNST_5HT2c calcium signals are modulated by exposure to affective stimuli.

A, Surgical schematic and experimental timeline for GCaMP fiber photometry experiments. B Representative viral infections and fiber placements in LHb (left) and BNST (right). Similar viral placement was reproduced in n = 37 mice. C Representative traces for water drinking session (individual mice). D Representative traces for alcohol drinking session (individual mice). E Peri-event plot of LHb_5HT2c GCaMP activity during the first interaction with a novel juvenile social target (Females n = 20 mice/1 trial; Males: n = 15 mice/1 trial). F Average z-score of LHb_5HT2c GCaMP signal for 0–5 s post interaction (Females n = 20 mice/1 trial, Males: n = 15 mice/1 trial). G Peri-event plot of LHb_5HT2c GCaMP activity during the acoustic startle test (Females: n = 21 mice/10 trials; Males: n = 15 mice/10 trials). H Average z-score of LHb_5HT2c GCaMP signal for 0–5 s post startle stimulus (Females: n = 21 mice/10 trials; Males: n = 15 mice/10 trials). I Peri-event plot of LHb_5HT2c GCaMP activity during voluntary water consumption (Females: n = 21 mice/1-4 bouts; Males: n = 15 mice/1–4 bouts). J Average z-score of LHb_5HT2c GCaMP signal for 0–5 s post water drinking bout start (Females: n = 21 mice/1–4 bouts; Males: n = 15 mice/1–4 bouts). K Peri-event plot of LHb_5HT2c GCaMP activity during voluntary alcohol consumption (Females: n = 21 mice/1-4 bouts; Males: n = 15 mice/1–4 bouts). L Average z-score of LHb_5HT2c GCaMP signal for 0-5 s post alcohol drinking bout start, p = 0.0038 (Females: n = 21 mice/1-4 bouts; Males: n = 15 mice/1–4 bouts). M Peri-event plot of BNST_5HT2c GCaMP activity during the first interaction with a novel juvenile social target (Females: n = 20 mice/1 trial; Males: n = 15 mice/1 trial). N Average z-score of BNST_5HT2c GCaMP signal for 0–5 s post interaction, p = 0.0073 (Females: n = 20 mice/1 trial; Males: n = 15 mice/1 trial). O Peri-event plot of BNST_5HT2c GCaMP activity during the acoustic startle test (Females: n = 22 mice/10 trials; Males: n = 15 mice/10 trials). P Peak z-score of BNST_5HT2c GCaMP signal for 0–1 s post startle stimulus, p = 0.0012 (Females: n = 22 mice/10 trials; Males: n = 15 mice/10 trials). Q Peri-event plot of BNST_5HT2c GCaMP activity during voluntary water consumption (Females: n = 20 mice/1–4 bouts; Males: n = 15 mice/1–4 bouts). R Average z-score of BNST_5HT2c GCaMP signal for 0–5 s post water drinking bout start, p = 0.0296 (Females: n = 20 mice/1–4 bouts; Males: n = 15 mice/1–4 bouts). S Peri-event plot of BNST_5HT2c GCaMP activity during voluntary alcohol consumption (Females: n = 21 mice/1–4 bouts; Males: n = 15 mice/1–4 bouts). T Average z-score of BNST_5HT2c GCaMP signal for 0–5 s post water drinking bout start, p = 0.0103 (Females: n = 21 mice/1-4 bouts; Males: n = 15 mice/1–4 bouts). Unless otherwise stated, statistical comparisons were performed using a two-tailed Student’s unpaired t test. All data are represented as mean ± SEM. *p < 0.05, **p < 0.01. Source data are provided as a Source Data file. Created with Biorender.com.

Fig. 2. In-vivo LHb_5HT2c and BNST_5HT2c serotonin signals are modulated by exposure to affective stimuli.

A Surgical schematic and experimental timeline for GRAB-5HT fiber photometry experiments. B Representative viral infections and fiber placements in LHb (left) and BNST (right). Similar viral placement was reproduced in n = 39 mice. C, Representative traces for water drinking session. D Representative traces for alcohol drinking session. E Peri-event plot of LHB_5HT2c GRAB-5HT activity during the first interaction with a novel juvenile social target (Females: n = 14 mice/1 trial; Males: n = 18 mice/1 trial). F Average z-score of LHb_5HT2c GRAB-5HT signal for 0-5 s post interaction (Females: n = 14 mice/1 trial; Males: n = 18 mice/1 trial). G Peri-event plot of LHb_5HT2c GRAB-5HT activity during the acoustic startle test (Females: n = 14 mice/10 trials; Males: n = 18 mice/10 trials). H Tau (time constant) of LHb_5HT2c GRAB-5HT signal for 0-5 s post startle stimulus (fit with one phase exponential decay curve, bars depict 95% confidence interval) (Females: n = 14 mice/10 trials; Males: n = 18 mice/10 trials). I Peri-event plot of LHb_5HT2c GRAB-5HT activity during voluntary water consumption (Females: n = 14 mice/1-4 bouts; Males: n = 19 mice/1–4 bouts). J Average z-score of LHb_5HT2c GRAB-5HT signal for 0–5 s post water drinking bout start (Females: n = 14 mice/1–4 bouts; Males: n = 19 mice/1-4 bouts). K Peri-event plot of LHb_5HT2c GRAB-5HT activity during voluntary alcohol consumption (Females: n = 14 mice/1–4 bouts; Males: n = 16 mice/1–4 bouts). L Average z-score of LHb_5HT2c GRAB-5HT signal for 0–5 s post alcohol drinking bout start (Females: n = 14 mice/1-4 bouts; Males: n = 16 mice/1–4 bouts). M Peri-event plot of BNST_5HT2c GRAB-5HT activity during the first interaction with a novel juvenile social target (Females: n = 16 mice/1 trial; Males: n = 21 mice/1 trial). N Average z-score of BNST_5HT2c GRAB-5HT signal for 0–5 s post interaction (Females: n = 16 mice/1 trial; Males: n = 21 mice/1 trial). O Peri-event plot of BNST_5HT2c GRAB-5HT activity during the acoustic startle test (Females: n = 16 mice/10 trials; Males: n = 21 mice/10 trials). P Tau (time constant) of BNST_5HT2c GRAB-5HT signal for 0-5 s post startle stimulus (fit with one phase exponential decay curve, bars depict 95% confidence interval) (Females: n = 16 mice/10 trials; Males: n = 21 mice/10 trials). Q Peri-event plot of BNST_5HT2c GRAB-5HT activity during voluntary water consumption (Females: n = 16 mice/1–4 bouts; Males: n = 23 mice/1–4 bouts). R Average z-score of BNST_5HT2c GRAB-5HT signal for 0–5 s post water drinking bout start, p = 0.0331 (Females: n = 16 mice/1–4 bouts; Males: n = 23 mice/1–4 bouts). S Peri-event plot of BNST_5HT2c GRAB-5HT activity during voluntary alcohol consumption (Females: n = 16 mice/1-4 bouts; Males: n = 21 mice/1–4 bouts). T Average z-score of BNST_5HT2c GRAB-5HT signal for 0–5 s post water drinking bout start (Females: n = 16 mice/1–4 bouts; Males: n = 21 mice/1–4 bouts). Unless otherwise stated, statistical comparisons were performed using a two-tailed Student’s unpaired t test. All data excluding panels 2H and 2P are represented as mean ± SEM. *p < 0.05, **p < 0.01. Source data are provided as a Source Data file. Created with Biorender.com.

Fig. 3. Anatomical and neurochemical characterization of DRN-LHb and DRN-BNST circuits.

A Representative CtB 647 infection in BNST, females. Similar images were obtained in n = 3 mice B Representative CtB 555 infection, females. Similar images were obtained in n = 3 mice. C Representative CtB 647 infection in BNST, males. Similar images were obtained in n = 4 mice. D Representative CtB 555 infection in LHb, males. Similar images were obtained in n = 4 mice. E Representative CtB 647 labeling in DRN, females. Similar images were obtained in n = 3 mice. F Representative CtB 555 labeling in DRN, females. Similar images were obtained in n = 3 mice. G Representative CtB 647 infection in BNST, males. Similar images were obtained in n = 4 mice. H Representative CtB 555 infection, males. Similar images were obtained in n = 4 mice. I Representative 5HT labeling in DRN, females. Similar images were obtained in n = 3 mice. J Representative labeling in DRN for CtB 647, CtB 555, and 5HT, females. Similar images were obtained in n = 3 mice. K Representative 5HT labeling in DRN, males. Similar images were obtained in n = 4 mice. L Representative labeling in DRN for CtB 647, CtB 555, and 5HT, males. Similar images were obtained in n = 4 mice. M Percent of DRN-LHb neurons positive for 5HT N Percent of DRN-BNST neurons positive for 5HT. O Percent of CtB 555 neurons that co-express CtB 647 in DRN. P Percent of CtB 647 neurons that co-express CtB 555 in DRN. Q In-situ hybridization for 5HT_2c, vGAT, and vGlut2 in BNST, females. Similar images were obtained in n = 3 mice. R In-situ hybridization for 5HT_2c, vGAT, and vGlut2 in BNST, males. Similar images were obtained in n = 4 mice. S, In-situ hybridization for 5HT_2c, vGAT, and vGlut2 in LHb, females. Similar images were obtained in n = 3 mice. T In-situ hybridization for 5HT_2c, vGAT, and vGlut2 in LHb, males. Similar images were obtained in n = 4 mice. U Dorsal BNST 5HT2c neuron overlaps with vGAT or vGlut2. V Ventral BNST 5HT2c neuron overlaps with vGAT or vGlut2. W LHb 5HT2c neuron overlap with vGAT or vGlut2. n = 3 males, n = 4 females, 2 slices/mouse. No statistical comparisons were performed on this data as they are intended to be descriptive. All data are represented as mean ± SEM. Source data are provided as a Source Data file. Created with Biorender.com.

Fig. 4. Chemogenetic activation of LHb_5HT2c or BNST_5HT2c neurons modulates affective behaviors and binge alcohol consumption.

A Surgical schematic and experimental timeline for BNST_5HT2c chemogenetic activation experiments. B Representative viral infection in BNST. Similar viral infection was reproduced in n = 35 mice. C Social preference in the 3-chamber sociability test, p_virusXsex = 0.0192. D Social novelty preference in the 3-chamber sociability test. E Acoustic startle behavior in females (10 trials/dB), two-way repeated-measures ANOVA followed by Holm-Sidak’s post-hoc, p_virus = 0.0814, p_virusXdB = 0.0105. 3. F Acoustic startle behavior in males (10 trials/dB), two-way repeated measures ANOVA followed by Holm-Sidak’s post-hoc, p_virus = 0.0246, p_virusXdB = 0.0111. G Acoustic startle behavior 120 dB only (10 trials), p_virus = 0.0058, p_sex = 0.037. H Sucrose consumption, p_virus = 0.0387, p_sex = 0.0001. I 2 h Alcohol intake in DiD, p_virus = 0.0303, p_sex = 0.0007, p_sexXvirus = 0.0003. J 2 h Alcohol preference in DiD. K Surgical schematic and experimental timeline for LHb_5HT2c chemogenetic activation experiments. L Representative viral infection in the LHb. Similar viral infection was reproduced in n = 31 mice. M Social preference in the 3-chamber sociability test. N Social novelty preference in the 3-chamber sociability test, p_virus = 0.0186. O Acoustic startle behavior, females (10 trials/dB), two-way repeated measures ANOVA followed by Holm-Sidak’s post-hoc, p_virus = 0.0084, p_virusXdB = 0.0051. P Acoustic startle behavior, males (10 trials/dB), two-way repeated measures ANOVA followed by Holm-Sidak’s post-hoc, p_virusXdB = 0.012. Q Acoustic startle behavior, 120 dB only (10 trials), p_sex = 0.0221, p_virus = 0.0031. R Sucrose consumption, p_virus < 0.0001, p_sex < 0.0001, p_{virus × sex} < 0.0001. S 2 h Alcohol consumption in DiD, p_virus = 0.0026, p_sex = 0.0301. T, 2 h Alcohol preference in DiD, p_virus = 0.0099, p_sex = 0.0696. BNST, Females: mcherry n = 8 mice, hM3Dq n = 9 mice; Males: mCherry n = 10 mice, hM3Dq n = 8 mice. LHb, Females: mCherry n = 9 mice, hM3Dq n = 7 mice; Males: mCherry n = 9 mice, hM3Dq n = 8 mice. Unless otherwise stated, statistical comparisons were performed using a two-way ANOVA followed by Holm-Sidak’s post-hoc. All data are presented as mean ± SEM. $ denotes effect of virus, @ denotes effect of sex, # denotes interaction (sex × virus or virus × stimulus strength), * denotes post-hoc p < 0.05, ** denotes post-hoc p < 0.01, *** denotes post-hoc p < 0.001, **** denotes post-hoc p < 0.0001. CNO: clozapine-n-oxide. Source data are provided as a Source Data file. Created with Biorender.com.

Fig. 5. DiD induces unique affective disturbances in males and females.

A Experimental timeline for DiD behavioral studies. B Average alcohol intake in DiD, two-tailed Student’s unpaired t-tests, p_2h = 0.0042, p_4h = 0.0069C Cumulative alcohol intake in DiD, two-tailed Student’s unpaired t-test, p = 0.0017. D Average alcohol preference in DiD, two-tailed Student’s unpaired t-tests, p_2h = 0.0841, p_4h = 0.0685. E Blood alcohol concentration immediately following 4h of DiD, two-tailed Student’s unpaired t-test. F Social preference in the 3-chamber sociability test. G Social novelty preference in the 3-chamber sociability test, p_sexXDiD = 0.008. H Percent time in center of open field, p_DiD = 0.0019, p_sex < 0.0001. I Center entries open field, p_DiD < 0.0001. J Total distance in open field, p_DiD = 0.0013, p_sex = 0.0366. K Mean velocity in open field, p_DiD = 0.0013, p_sex = 0.0366. L Acoustic startle behavior, females (10 trials/dB), two-way repeated measures ANOVA. M Acoustic startle behavior, males (10 trials/dB), two-way repeated measures ANOVA followed by Holm-Sidak’s post-hoc, p_DiD = 0.0821, p_DiDXdB = 0.0055. N Acoustic startle behavior, 120 dB only (10 trials), p_DiD = 0.0164, p_sex = 0.0034. O Experimental timeline for DiD qPCR studies. P Average alcohol intake in DiD, two-tailed Student’s unpaired t tests, p_2h = 0.1589, p_4h = 0.1012. Q Cumulative alcohol intake in DiD, two-tailed Student’s unpaired t-test, p = 0.0448. R Average alcohol preference in DiD, two-tailed Student’s unpaired t tests, p_2h = 0.0054, p_4h = 0.1964. S Relative expression of Htr2c mRNA in Hb, p_DiD = 0.0005, p_sex = 0.0902, p_DiDXsex = 0.1012. T Relative expression of Htr2c mRNA in BNST. For panels 3B-3N, Females: water n = 8 mice, DiD n = 8 mice; Males: water n = 8 mice, DiD n = 8 mice. For panels 3P–3R, Females: n = 10 mice; Males: n = 10 mice. For panels 3S and 3T, Females: water n = 10 mice/3 replicates, DiD n = 10 mice/3 replicates; Males: water n = 10 mice/3 replicates, DiD n = 10 mice/3 replicates. Unless otherwise stated, statistical comparisons were performed using two-way ANOVA followed by Holm-Sidak’s post-hoc. All data are presented as mean ± SEM. & denotes effect of DiD (&p < 0.05, &&p < 0.01, &&&p < 0.001, &&&&p < 0.0001), @ denotes effect of sex (@p < 0.05, @@p < 0.01, @@@p < 0.001, @@@@p < 0.0001), # denotes interaction (sex × DiD or DiD × stimulus strength, #p < 0.01, ##p < 0.01), and * denotes post-hoc p < 0.05, ** denotes post-hoc p < 0.01. Source data are provided as a Source Data file. Created with Biorender.com.

Fig. 6. DiD induces physiological adaptations in LHb_5HT2c and BNST_5HT2c.

A Experimental timeline for DiD electrophysiology studies. B Average alcohol intake in DiD, two-way Student’s unpaired t-tests. C Cumulative alcohol intake in DiD, two-way Student’s unpaired t test. D Average alcohol preference in DiD, two-way Student’s unpaired t tests. E Resting membrane potential (RMP) in LHb_5HT2c neurons. F LHb_5HT2c rheobase. G LHb_5HT2c action potential (AP) threshold. H LHb_5HT2c elicited APs, females, two-way repeated measures ANOVA. I LHb_5HT2c elicited APs, males, two-way repeated measures ANOVA followed by Holm-Sidak’s post-hoc, p_DiD = 0.0236, p_DiDXcurrent = 0.0512. J Average RMP in BNST_5HT2c neurons, p_sex = 0.0190. K BNST_5HT2c rheobase. L BNST_5HT2c AP threshold. M BNST_5HT2c elicited APs, females, two-way repeated measures ANOVA. N BNST_5HT2c elicited APs, males, two-way repeated measures ANOVA. O Voltage–Current plot for LHb_5HT2c females, two-way repeated measures ANOVA. P Voltage–Current plot for LHb_5HT2c males, two-way repeated measures ANOVA. Q Voltage–current plot for BNST_5HT2c females, two-way repeated measures ANOVA, p_DiD = 0.1032, p_DiDXvoltage = 0.0701. R Voltage–Current plot for BNST_5HT2c males, two-way repeated measures ANOVA. S Representative trace in medial LHb_5HT2c during rheobase test. T Quantification of rheobase values in medial LHb_5HT2c in females, two-way Student’s unpaired t test, p = 0.0157 (water n = 5 mice/7 cells; DiD n = 5 mice/10 cells). U Breakdown of cell properties at RMP, BNST females. V Breakdown of cell properties at RMP, LHb females. W Breakdown of cell properties at RMP, BNST males. X Breakdown of cell properties at RMP, LHb males. Depol block = silent at greater than −40 mV. BNST Females: water n = 6 mice/10 cells, DiD n = 6 mice/19 cells; BNST Males: water n = 5 mice/16 cells, DiD n = 5 mice/16 cells. LHb Females: water n = 7 mice/14 cells, DiD n = 6 mice/19 cells; LHb Males: water n = 6 mice/19 cells, DiD n = 6 mice/19 cells. Unless otherwise stated, statistical comparisons were performed using two-way ANOVA followed by Holm-Sidak’s post-hoc. All data except for box plots are represented as mean ± SEM. For box plots, center line is the mean, box limits are 25th–75th percentiles, whiskers are min to max. & denotes effect of DiD, @ denotes effect of sex, # denotes interaction (sex × DiD or DiD × current), and * denotes post-hoc p < 0.05. Source data are provided as a Source Data file. Created with Biorender.com.

Fig. 7. DiD modulates the calcium responses of LHb_5HT2c and BNST_5HT2c neurons to affective stimuli.

A Surgical schematic and experimental timeline for LHb_5HT2c and BNST_5HT2c GCaMP fiber photometry experiments in DiD-exposed mice. B Average alcohol consumption in DiD (n = 12 females, n = 8 males). C Cumulative alcohol intake in DiD (n = 12 females, n = 8 males). D Average alcohol preference in DiD (n = 12 females, n = 8 males). E Peri-event plot of LHb_5HT2c GCaMP activity during the first interaction with a novel social target, females (water n = 10 mice/1 trial, DiD n = 11 mice/1 trial). F Peak z-score of LHb_5HT2c GCaMP signal for 0–1 s post interaction, females, Mann–Whitney U test (water n = 10 mice/1 trial, DiD n = 11 mice/1 trial). G Peri-event plot of LHb_5HT2c GCaMP activity during the first interaction with a novel social target, males (water n = 7 mice/1 trial, DiD n = 8 mice/1 trial). H Average z-score of LHb_5HT2c GCaMP signal for 0–5 s post interaction, males (water n = 7 mice/1 trial, DiD n = 8 mice/1 trial). I Peri-event plot of BNST_5HT2c GCaMP activity during the first interaction with a novel social target, females (water n = 10 mice/1 trial, DiD n = 12 mice/1 trial). J Average z-score of BNST_5HT2c GCaMP signal for 0–5 s post interaction, females (water n = 10 mice/1 trial, DiD n = 12 mice/1 trial). K Peri-event plot of BNST_5HT2c GCaMP activity during the first interaction with a novel social target, males (water n = 7 mice/1 trial, DiD n = 8 mice/1 trial). L Average z-score of BNST_5HT2c GCaMP signal for 0–5 s post interaction, males (water n = 7 mice/1 trial, DiD n = 8 mice/1 trial). M Peri-event plot of LHb_5HT2c GCaMP activity during the acoustic startle test, females (water n = 10 mice/10 trials, DiD n = 11 mice/10 trials). N Average z-score of LHb_5HT2c GCaMP signal for 0–5 s post acoustic startle stimulus, females (water n = 10 mice/10 trials, DiD n = 11 mice/10 trials). O Peri-event plot of LHb_5HT2c GCaMP activity during the acoustic startle test, males (water n = 7 mice/10 trials, DiD n = 8 mice/10 trials). P Average z-score of LHb_5HT2c GCaMP signal for 0–5 s post acoustic startle stimulus, males, p = 0.0008 (water n = 7 mice/10 trials, DiD n = 8 mice/10 trials). Q Peri-event plot of BNST_5HT2c GCaMP activity during the acoustic startle test, females (water n = 10 mice/10 trials, DiD n = 12 mice/10 trials). R Average z-score of BNST_5HT2c GCaMP signal for 0–5 s post acoustic startle stimulus, females (water n = 10 mice/10 trials, DiD n = 12 mice/10 trials). S Peri-event plot of BNST_5HT2c GCaMP activity during the acoustic startle test, males (water n = 7 mice/10 trials, DiD n = 8 mice/10 trials). T Average z-score of BNST_5HT2c GCaMP signal for 0–5 s post acoustic startle stimulus, males, p = 0.0157 (water n = 7 mice/10 trials, DiD n = 8 mice/10 trials). Unless otherwise stated, statistical comparisons were performed using two-way Student’s unpaired t tests. All data represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.0001. Source data are provided as a Source Data file. Created with Biorender.com.

Fig. 8. DiD modulates serotonin release onto LHb_5HT2c and BNST_5HT2c neurons during exposure to affective stimuli.

A Surgical schematic and experimental timeline for LHb_5HT2c and BNST_5HT2c GRAB-5HT fiber photometry experiments in DiD-exposed mice. B Average alcohol consumption in DiD, p_2h = 0.0532 (Females n = 8 mice, Males n = 12 mice). C Cumulative alcohol intake in DiD, p = 0.2897 (Females n = 8 mice, Males n = 12 mice). D Average alcohol preference in DiD p_4h = 0.0039 (Females n = 8 mice, Males n = 12 mice). E Peri-event plot of LHb_5HT2c GRAB-5HT activity during the first interaction with a novel social target, females, (water n = 6 mice/1 trial, DiD n = 8 mice/1 trial). F Average z-score of LHb_5HT2c GRAB-5HT signal for 0–5 s post interaction, females, p = 0.0699 (water n = 6 mice/1 trial, DiD n = 8 mice/1 trial). G Peri-event plot of LHb_5HT2c GRAB-5HT activity during the first interaction with a novel social target, males (water n = 9 mice/1 trial, DiD n = 10 mice/1 trial). H Average z-score of LHb_5HT2c GRAB-5HT signal for 0–5 s post interaction, males (water n = 9 mice/1 trial, DiD n = 10 mice/1 trial). I Peri-event plot of BNST_5HT2c GRAB-5HT activity during the first interaction with a novel social target, females (water n = 8 mice/1 trial, DiD n = 7 mice/1 trial). J Average z-score of BNST_5HT2c GRAB-5HT signal for 0–5 s post interaction, females, p = 0.0008 (water n = 8 mice/1 trial, DiD n = 7 mice/1 trial). K Peri-event plot of BNST_5HT2c GRAB-5HT activity during the first interaction with a novel social target, males (water n = 11 mice/1 trial, DiD n = 12 mice/1 trial). L Average z-score of BNST_5HT2c GRAB-5HT signal for 0–5 s post interaction, males (water n = 11 mice/1 trial, DiD n = 12 mice/1 trial). M Peri-event plot of LHb_5HT2c GRAB-5HT activity during the acoustic startle test, females (water n = 6 mice/10 trials, DiD n = 8 mice/10 trials). N Peak z-score of LHb_5HT2c GRAB-5HT signal for 0–1 s post acoustic startle stimulus, females (water n = 6 mice/10 trials, DiD n = 8 mice/10 trials). O Peri-event plot of LHb_5HT2c GRAB-5HT activity during the acoustic startle test, males (water n = 9 mice/10 trials, DiD n = 10 mice/10 trials). P Average z-score of LHb_5HT2c GRAB-5HT signal for 0–5 s post acoustic startle stimulus, males (water n = 9 mice/10 trials, DiD n = 10 mice/10 trials). Q Peri-event plot of BNST_5HT2c GRAB-5HT activity during the acoustic startle test, females (water n = 8 mice/10 trials, DiD n = 8 mice/10 trials). R Average z-score of BNST_5HT2c GRAB-5HT signal for 0–5 s post acoustic startle stimulus, females (water n = 8 mice/10 trials, DiD n = 8 mice/10 trials). S Peri-event plot of BNST_5HT2c GRAB-5HT activity during the acoustic startle test, males (water n = 11 mice/10 trials, DiD n = 12 mice/10 trials). T Average z-score of BNST_5HT2c GRAB-5HT signal for 0–5 s post acoustic startle stimulus (water n = 11 mice/10 trials, DiD n = 12 mice/10 trials). Unless otherwise stated, statistical comparisons were performed using two-way Student’s unpaired t tests. All data represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.0001. Source data are provided as a Source Data file. Created with Biorender.com.

Fig. 9. Deletion of 5HT_2c in the BNST or LHb partially normalizes alcohol-induced social and startle disturbances.

A Surgical schematic and experimental timeline for LHb 5HT_2c deletion. B Representative viral infection for LHb 5HT_2c deletion experiments. Similar viral infection was reproduced in n = 66 mice. C Social preference in the 3-chamber sociability test, females. D Social novelty preference in the 3-chamber sociability test, females, p_DiDXvirus = 0.022, p_DiD = 0.0670, p_virus = 0.1352. E Social preference in 3-chamber sociability test, males, p_virus = 0.0334. F Social novelty preference in 3-chamber sociability test, males. G Acoustic startle behavior, females (10 trials/dB), three-way repeated measures ANOVA followed Holm-Sidak’s post-hoc, p_virusXdB = 0.0112, p_DiD = 0.0908, p_virusXdBXDiD = 0.087. H Acoustic startle behavior, 120 dB only, females (10 trials), p_virus = 0.0417. I Acoustic startle behavior, males (10 trials/dB), three-way repeated measures ANOVA followed by Holm-Sidak’s post-hoc, p_virusXDiD = 0.0104, p_virusXdBXDiD = 0.006. J Acoustic startle behavior 120 dB only, males (10 trials), p_virusXDiD = 0.0059. K Surgical schematic and experimental timeline for BNST 5HT_2c deletion. L Representative viral infection for BNST 5HT_2c deletion experiments. Similar viral infection was reproduced in n = 61 mice. M Social preference in the 3-chamber sociability test, females, p_virus = 0.0329. N Social novelty preference in the 3-chamber sociability test, females, p_DiDXvirus = 0.0019. O Social preference in 3-chamber sociability test, males. P Social novelty preference in 3-chamber sociability test, males, p_DiD = 0.062. Q Acoustic startle behavior, females (10 trials/dB), three-way repeated measures ANOVA, Holm-Sidak’s post-hoc. R Acoustic startle behavior, 120 dB only, females (10 trials), three-way repeated measures ANOVA, Holm-Sidak’s post-hoc. S Acoustic startle behavior, males (10 trials/dB), three-way repeated measures ANOVA followed by Holm-Sidak’s post-hoc, p_virus = 0.07, p_virusXdB = 0.1753. T Acoustic startle behavior 120 dB only, males (10 trials). LHb females, GFP: water n = 8 mice, DiD n = 8 mice; Cre: water n = 9 mice, DiD n = 6 mice. LHb males, GFP: water n = 8 mice, DiD n = 9 mice; Cre: water n = 9 mice, DiD n = 9 mice. BNST females, GFP: water n = 7 mice, DiD n = 9 mice; DiD water n = 7 mice, DiD n = 8 mice. BNST males, GFP: water n = 7 mice, DiD n = 9 mice; Cre: water n = 8 mice, DiD n = 6 mice. Unless otherwise stated, statistical comparisons were performed using two-way ANOVA followed by Holm-Sidak’s post-hoc. All data are presented as mean ± SEM. & denotes effect of DiD, # denotes interaction (DiD × virus), * denotes post-hoc effects. Source data are provided as a Source Data file. Created with Biorender.com.

Fig. 10. Chemogenetic inhibition of LHb_5HT2c normalizes DiD-induced social and arousal disturbances.

A Surgical schematic and experimental timeline. B Representative image of viral infection in LHb. Similar viral infection was reproduced in n = 59 mice. C Social preference in the 3-chamber sociability test, p_DiD = 0.0317 D Social novelty preference in the 3-chamber sociability test, LHb females, p_DiD = 0.2023, p_virus = 0.2472, p_virusXDiD = 0.0016. E Social preference in 3-chamber sociability test, LHb males. F Social novelty preference in the 3-chamber sociability test, LHb males, p_virus = 0.0298. G Acoustic startle behavior, LHb females (10 trials/dB). H Acoustic startle behavior 120 dB only, LHb females (10 trials). I Acoustic startle behavior, LHb males (10 trials/dB), p_virus = 0.1033, p_virusXDiD = 0.0028, p_virusXdBXDiD = 0.0043. J Acoustic startle behavior 120 dB only, LHb males (10 trials), p_virusXDiD = 0.0044. K Surgical schematic and experimental timeline. L Representative image of viral infection in BNST. Similar viral infection was reproduced in n = 64 mice. M Social preference in 3-chamber sociability test, BNST females. N Social novelty preference in 3-chamber sociability test, BNST females, p_DiD = 0.0253. O Social preference in the 3-chamber sociability test, BNST males. P Social novelty preference in 3-chamber sociability test, BNST males. Q Acoustic startle behavior, BNST females (10 trials/dB), p_virus = 0.1188, p_virusXdB = 0.0294, p_DiDXdB = 0.0459, p_virusXdBXDiD = 0.045. R Acoustic startle behavior 120 dB only, BNST females (10 trials), p_DiD = 0.0130, p_virusXDiD = 0.2662. S Acoustic startle behavior, BNST males (10 trials/dB). T Acoustic startle behavior 120 dB only, BNST males, p_virus = 0.1282 (10 trials). LHb females, mCherry: water n = 7 mice, DiD n = 8 mice; hM4Di: water n = 6 mice, DiD n = 6 mice. LHb males, mCherry: water n = 8 mice, DiD n = 8 mice; hM4Di: water n = 9 mice, DiD n = 7 mice. BNST females, mCherry: water n = 8 mice, DiD n = 7 mice; hM4Di: water n = 8 mice, DiD n = 9 mice. BNST males, mCherry: water n = 9 mice, DID n = 8 mice; hM4Di water n = 6 mice, DiD n = 9 mice. All data are presented as mean ± SEM. & denotes effect of DiD, # denotes interaction (DiD x virus, #p < 0.05, ##p < 0.01), $ denotes effect of virus ($p < 0.05, $$p < 0.01), * denotes post-hoc p < 0.05, ** denotes p < 0.01. CNO: clozapine-n-oxide. Source data are provided as a Source Data file. Created with Biorender.com.