Nucleus accumbens medium spiny neuron subtypes mediate depression-related outcomes to social defeat stress

Abstract

Background: The nucleus accumbens is a critical mediator of depression-related outcomes to social defeat stress. Previous studies demonstrate distinct neuroplasticity adaptations in the two medium spiny neuron (MSN) subtypes, those enriched in dopamine receptor D1 versus dopamine receptor D2, in reward and reinforcement leading to opposing roles for these MSNs in these behaviors. However, the distinct roles of nucleus accumbens MSN subtypes, in depression, remain poorly understood.

Methods: Using whole-cell patch clamp electrophysiology, we examined excitatory input to MSN subtypes and intrinsic excitability measures in D1-green fluorescent protein and D2-green fluorescent protein bacterial artificial chromosome transgenic mice that underwent chronic social defeat stress (CSDS). Optogenetic and pharmacogenetic approaches were used to bidirectionally alter firing of D1-MSNs or D2-MSNs after CSDS or before a subthreshold social defeat stress in D1-Cre or D2-Cre bacterial artificial chromosome transgenic mice.

Results: We demonstrate that the frequency of excitatory synaptic input is decreased in D1-MSNs and increased in D2-MSNs in mice displaying depression-like behaviors after CSDS. Enhancing activity in D1-MSNs results in resilient behavioral outcomes, while inhibition of these MSNs induces depression-like outcomes after CSDS. Bidirectional modulation of D2-MSNs does not alter behavioral responses to CSDS; however, repeated activation of D2-MSNs in stress naïve mice induces social avoidance following subthreshold social defeat stress.

Conclusions: Our studies uncover novel functions of MSN subtypes in depression-like outcomes. Notably, bidirectional alteration of D1-MSN activity promotes opposite behavioral outcomes to chronic social stress. Therefore, targeting D1-MSN activity may provide novel treatment strategies for depression or other affective disorders.

Keywords: Depression; Medium spiny neurons; Nucleus accumbens; Optogenetics; Pharmacogenetics; Social defeat stress.

Figure 1

Excitatory input is altered in medium spiny neuron (MSN) subtypes after chronic social defeat stress. (A) Susceptible mice, used to examine alterations in excitatory input, spend significantly less time interacting with a novel mouse, while resilient mice used in these measures interact the same as control (no defeat) animals (one-way analysis of variance [ANOVA] F_2,19 = 12.00, p < .01, n = 6–8 per group). No differences were observed when a social target was not present (one-way ANOVA F_2,19 = .80, p > .05). (B–C) Excitatory synaptic input measures in dopamine receptor D1-MSNs and dopamine receptor D2-MSNs in susceptible and resilient mice after chronic social defeat stress. (B) Susceptible mice displayed decreased miniature excitatory postsynaptic current (mEPSC) frequency in D1-MSNs (one-way ANOVA F_2,33 = 3.49, p < .05, n = 9–13 cells per group) with no change in mEPSC amplitude (one-way ANOVA F_2,33 = .29, p > .05). (C) D2-MSNs from susceptible animals displayed increased mEPSC frequency in D2-MSNs (one-way ANOVA F_2,23 = 23.87, p < .0001, n = 7–9 cells per group) but not in mEPSC amplitude (one-way ANOVA F_2,23 = .23, p > .05). Groups were compared to No Defeat group: *p < .05, **p < .01, ***p < .001. Error bars represent standard error measure (SEM). n.s., nonsignificant.

Figure 2

Intrinsic excitability is altered in dopamine receptor D1-medium spiny neurons (MSNs) after chronic social defeat stress (CSDS). (A) Susceptible mice, used to examine intrinsic excitability, spend significantly less time in the interaction zone, while resilient mice interact the same as control animals (one-way analysis of variance [ANOVA] F_2,21 = 16.50, p < .0001, n = 6–10 animals per group). No target conditions were similar across groups (one-way ANOVA F_2,21 = .54, p > .05). (B–C) Intrinsic excitability measures of D1-MSNs and D2-MSNs from susceptible and resilient mice after CSDS. (B) Susceptible mice display increased rheobase, the amount of current needed to elicit an action potential (one-way ANOVA F_2,30 = 10.71, p < .001, n = 8–13 cells per group) and current-evoked spikes in D1-MSNs (two-way ANOVA with a main effect of CSDS phenotype F_2,119 = 8.15, p < .001, n = 5–8 cells per group). (C) In D2-MSNs, no changes were observed in rheobase (one-way ANOVA F_{2, 27} = .08, p > .05, n = 8–12 cells per group) or current-evoked spiking (two-way ANOVA no effect of CSDS phenotype F_2,115 = 1.23, p > .05, n = 6–11 cells per group). Groups were compared to No Defeat group: *p < .05, **p < .01, ***p < .001. Error bars represent SEM.

Figure 3

Repeated activation of dopamine receptor D1-medium spiny neurons (MSNs) but not dopamine receptor D2-MSNs following chronic social defeat stress (CSDS) promotes resilience. (A) Sagittal brain images in D1-Cre and D2-Cre animals demonstrating ChETA_A expression in appropriate terminals of each MSN subtype. ChETA_A expressing D1-MSNs project to ventral pallidum (VP), globus pallidus internal (GPi), ventral tegmental area (VTA), and substantia nigra (SN) and D2-MSNs project to VP (scale bar 200 μm). (B) 50-Hz blue light stimulation of ChETA_A elicits 13 Hz to 14 Hz firing frequencies in a representative MSN. (C) Experimental timeline of CSDS, repeated optogenetic activation, and behavioral assays in D1-Cre and D2-Cre mice. (D) Repeated 50-Hz blue light pulses to the nucleus accumbens (NAc) of susceptible D1-Cre mice expressing ChETA_A produces an increase in time spent interacting with a novel CD1 target (two-way repeated measures analysis of variance [ANOVA] significant interaction F_3,20 = 6.26, p < .01, with a main effect of stimulation F_3,20 = 7.88, p < .01, n = 4–8 animals per group) but does not alter interaction time without the social target (two-way repeated measures ANOVA nonsignificant (n.s.) interaction F_3,20 = 1.52, p > .05). Repeated 50-Hz blue light pulses to D1-MSNs also enhanced sucrose preference in susceptible and no defeat control animals (two-way ANOVA main effect of stimulation F_1,24 = 13.96, p = .001 and CSDS F_1,24 = 18.81, p < .001). (E) Repeated 50-Hz blue light pulses to the NAc of susceptible D2-Cre animals does not alter time spent interacting with the target or no target (two-way repeated measures ANOVA nonsignificant interaction: target F_3,19 = .21, p > .05; no target: F_3,19 = 2.28, p > .05) and mice remained anhedonic as measured by sucrose preference (two-way ANOVA nonsignificant interaction F_1,19 = .00003, p > .05; n = 4–8 per group). Groups were compared with no defeat enhanced yellow fluorescent protein (EYFP) group using a Bonferroni post hoc test: *p < .05, **p < .01, ***p < .001. Error bars represent SEM. AAV, adeno-associated virus; SI, social interaction.

Figure 4

Repeated activation of dopamine receptor D2-medium spiny neurons (MSNs) in stress naïve mice induces susceptibility to subthreshold social defeat stress (SSDS). (A) Experimental timeline of D2-Cre optogenetic stimulation and SSDS. (B) D2-Cre mice receiving repeated priming 50-Hz blue light pulses to nucleus accumbens (NAc) before SSDS and during SSDS displayed reduced social interaction (SI) (one-way analysis of variance [ANOVA] F_2,28 = 3.75, p < .05, n = 6–8 animals per group) with no alteration in no target interaction (one-way ANOVA F_2,28 = .26, p > .05). Sucrose preference was unaltered in these conditions (one-way ANOVA F_2,13 = .10, p > .05). Groups were compared with no defeat enhanced yellow fluorescent protein (EYFP) group using a Bonferroni post hoc test: *p < .05. Error bars represent SEM. AAV, adeno-associated virus.

Figure 5

Repeated pharmacogenetic inhibition of dopamine receptor D1-medium spiny neurons (MSNs) but not dopamine receptor D2-MSNs induces a prodepressive phenotype. (A) Experimental timeline of chronic social defeat stress (CSDS) and repeated pharmacogenetic inhibition of D1-Cre and D2-Cre mice. (B) Sagittal confocal images of hM4(Gi)-mCherry expression D1-MSNs and D2-MSNs in the nucleus accumbens (NAc). hM4(Gi) expressing D1-MSNs project to ventral pallidum (VP), globus pallidus internal (GPi), ventral tegmental area (VTA), and substantia nigra (SN) and D2-MSNs project to VP (scale bar 200 μm). (C) Representative traces from a current clamp slice recording of an MSN expressing the hM4(Gi) receptor. A decrease in spiking and input resistance was observed 1 hour following clozapine-N-oxide (CNO) wash on. (D) In resilient D1-Cre mice, repeated CNO injections reduced social interaction (SI) time (two-way repeated measures analysis of variance [ANOVA] significant interaction F_3,21 = 5.40, p < .01 with a main effect of treatment F_3,21 = 6.354, p < .01; n = 4–8 animals per group) and decreased sucrose preference (two-way ANOVA significant interaction F_1,21 = 11.19, p < .01 with a main effect of treatment F_1,21 = 11.65, p < .01) without altering interaction zone times without the target present (two-way repeated measures ANOVA nonsignificant interaction F_3,21 = 1.13, p > .05). (E) No effects were observed in time spent in the interaction zone with the target or no target (two-way repeated measures ANOVA nonsignificant interaction: target F_3,15 = .26, p > .05; no target F_3,15 = 1.28, p > .05) or in sucrose preference (two-way ANOVA nonsignificant interaction F_1,15 = .52, p > .05) in resilient D2-Cre mice. Groups were compared with no defeat saline group using a Bonferroni post hoc test: *p < .05, ***p < .001. Error bars represent SEM. AAV, adeno-associated virus.