Directing negative emotional states through parallel genetically-distinct basolateral amygdala pathways to ventral striatum subregions

Abstract

Distinct basolateral amygdala (BLA) cell populations influence emotions in manners thought important for anxiety and anxiety disorders. The BLA contains numerous cell types which can broadcast information into structures that may elicit changes in emotional states and behaviors. BLA excitatory neurons can be divided into two main classes, one of which expresses Ppp1r1b (encoding protein phosphatase 1 regulatory inhibitor subunit 1B) which is downstream of the genes encoding the D1 and D2 dopamine receptors (Drd1 and Drd2 respectively). The role of Drd1+ or Drd2+ BLA neurons in learned and unlearned emotional responses is unknown. Here, we identified that the Drd1+ and Drd2+ BLA neuron populations form two parallel pathways for communication with the ventral striatum. These neurons arise from the basal nucleus of the BLA, innervate the entire space of the ventral striatum, and are capable of exciting ventral striatum neurons. Further, through two separate behavioral assays, we found that the Drd1+ and Drd2+ parallel pathways distinctly influence both learned and unlearned emotional states when they are activated or suppressed and do so depending upon where they synapse in the ventral striatum - with unique contributions of Drd1+ and Drd2+ circuitry on negative emotional states. Overall, these results contribute to a model whereby parallel, genetically-distinct BLA to ventral striatum circuits inform emotional states in a projection-specific manner.

Fig. 1. Ventral striatum projecting BLA neurons arise from the BA and are comprised of Drd1+ and Drd2+ neurons.

A Schematic of approach for identifying BLA Drd1+ and Drd2+ inputs to the NAc. B Example of a NAc injection in Drd1-Cre (top) and Drd2-Cre (bottom) mice (ac = anterior commissure, NAcC & NAcSh = nucleus accumbens core & shell, respectively), and C example images of NAc projecting Drd1+ (top) or Drd2+ (bottom) neurons along the anterior-posterior axis of the BLA. Scale bars = 500 µm. Di Anti-NeuN and anti-DsRed immunofluorescence images to identify the size of the NAc projecting BLA neural population. Scale bars = 40 µm. Dii Quantification of the Drd1 (**p < 0.01, n = 3, 2/1/sex) or Drd2 (*p < 0.05, n = 3, 2/1/sex) expressing BLA cells along the entire AP axis that project to the NAc (Two-way ANOVA, ROI main effect F(1,8) = 26.65, p = 0.001). E Schematic of approach for identifying BLA Drd1+ and Drd2+ inputs to the TuS (TuS = tubular striatum). F Example of a TuS injection in Drd1-Cre (top) and Drd2-Cre (bottom) mice, and G example images of TuS projecting Drd1+ (top) or Drd2+ (bottom) neurons along the anterior-posterior axis of the BLA. Scale bars = 500 µm. Hi Anti-NeuN and anti-DsRed immunofluorescence images to identify the size of the NAc projecting BLA neural population. Scale bars = 40 µm. Hii Quantification of the Drd1 (****p < 0.0001, n = 3, 2/1/sex) or Drd2 (****p < 0.0001, n = 3, 2/1/sex) expressing TuS projecting BLA cells along the entire AP axis (Two-way ANOVA, ROI main effect F(1,8) = 295.3, p < 0.001). Mean ± SEM.

Fig. 2. Drd1+ and Drd2+ BLA neurons innervate the entire span of the ventral striatum.

A Schematic of approach for identifying BLA Drd1+ and Drd2+ synaptic innervation of the ventral striatum. B Maps of BLA injection sites for Drd1-Cre (left) and A2a-Cre mice (right), where each color represents individual mice. C Representative images showing direct innervation of BLA neurons into the ventral striatum in both Drd1- and A2a-Cre mice. Inset locations denoted by square white box, inset sale bar = 10 µm. D Quantification of synaptophysin puncta in Drd1-Cre (n = 3, 2/1/sex) and A2a-Cre (n = 3, 2/1/sex) mice. Mean ± SEM. E Approach for defining whether, and if so, BLA neurons send contralateral projections to the ventral striatum using an AAV encoding synaptophysin. GFP in the BLA in one hemisphere and synaptophysin.mRuby in the BLA in the opposite hemisphere. F Representative images from one mouse of ipsilateral and contralateral projections from the BLA to the ventral striatum following injection as in E. These example images show that while some BLA neurons transverse into the contralateral ventral striatum, qualitatively, the bulk of input is ipsilateral. G Diagram of approach for identifying BLA Drd1+ and Drd2+ projecting neurons to the NAc and TuS in the ventral striatum. H Percentage of the Drd1+ and Drd2+ neurons within the BA that project to the NAc (red), TuS (green), or both (yellow) (n = 3 mice/genotype, 2/1/sex, 5–6 sections/mouse). I Representative image of a retrograde injection in a Drd1-Cre mouse and J the anterior to posterior span of VS projecting BLA neurons. Scale bars = 500 µm. PCx piriform cortex, NAcC and NAcSh nucleus accumbens core and shell, respectively, BLA basolateral amygdala, BA basal amygdala, LA lateral amygdala.

Fig. 3. Synaptic properties of Drd1 and Drd2 expressing ventral striatum neurons receiving BLA neuronal projections.

Ai Schematic indicating Cre-dependent expression of ChR2 in Drd1+ BLA neurons of Drd1-Cre;Ai9 mice. In these mice, tdTomato + neurons are presumably Drd1+ and tdTomato- neurons are presumably Drd1Ø. During whole-cell patch clamp recordings, ChR2 expressing BLA terminals were activated by 470 nm light. Aii Schematic indicating Cre-dependent expression of ChR2 in Drd2+ BLA neurons of A2a-Cre;Ai9 mice. tdTomato + neurons are presumably Drd2+, and tdTomato- neurons are presumably Drd2Ø. B Example light-evoked monosynaptic EPSCs (top) and light-evoked polysynaptic EPSCs (bottom) from Drd1+ TuS neurons under voltage clamp mode. C Neurons organized by response type upon stimulation of Drd1+ TuS projecting BLA terminals. D Example evoked EPSCs from Drd1Ø TuS neurons. E Neurons organized by response type upon stimulation of Drd1Ø TuS projecting BLA terminals. F Example evoked EPSCs from Drd2+ TuS neurons. G Neurons organized by response type upon stimulation of Drd2+ Tus projecting BLA terminals. H Example evoked EPSCs from Drd2Ø TuS neurons. I Neurons organized by response type upon stimulation of Drd2+ Tus projecting BLA terminals. The holding potential was −70 mV. Drd1-Cre;Ai9 n = 2M/1F, A2a-Cre;Ai9 n = 4F/4M.

Fig. 4. BLA Drd1+ and Drd2+ neurons innervating the ventral striatum promote aversive states depending upon projection target.

A Paradigm for optic activation of NAc or TuS projecting Drd1+ or Drd2+ BA neurons and B 3-chamber real-time place preference/aversion assay where optic stimulation occurs in only one side of the chamber (chamber A, blue glow). Ci Optical stimulation of Drd1+ BLA→NAc neurons resulted in less time spent in the light-paired chamber (*p < 0.05, One-way ANOVA, F(2,18) = 5.04, p = 0.018, EYFP controls n = 4M/4F, Drd1+ ChR2 n = 3M/4F, Drd2+ ChR2 n = 3M/3F). Cii Stimulation of Drd1+ BLA→NAc neurons results in avoidance of the light-paired chamber (upper, t(6) = 2.981, *p = 0.025), demonstrated by representative heat map of chamber preference from one mouse (lower). Di Optical stimulation of Drd2+ BLA→TuS neurons resulted in less time spent in the light-paired chamber compared to optical stimulation of EYFP controls (*p < 0.05, Welch’s ANOVA W(2.00,8.31) = 6.02, p = 0.024, EYFP controls n = 6M/2F, Drd1+ ChR2 n = 2M/4F, Drd2+ ChR2 n = 2M/4F). Dii Stimulation of Drd2+ BLA→TuS neurons results in avoidance of the light-paired chamber (upper, t(5) = 2.916, *p = 0.033), demonstrated by representative heat map of chamber preference from one mouse (lower). EYFP control groups were collapsed across genotypes, containing both Drd1-Cre and A2a-Cre mice, since these behavioral results were not different from each other (percent of time spent in non-paired side for Drd1 + vs Drd2+ NAc EYFP controls unpaired t-test, t(6) = 1.85, p = 0.1134; and Drd1+ vs Drd2+ TuS EYFP controls unpaired t-test, t(6) = 0.305, p = 0.771). Behaviors did not differ by sex (all unpaired t-tests: NAc YFP t(6) = 1.29, p = 0.245; NAc drd1+ ChR2 t(5) = 0.933, p = 0.394; NAc Drd2+ ChR2 t(4) = 0.309, p = 0.772; TuS YFP t(6) = 1.10, p = 0.315; TuS Drd1+ ChR2 t(4) = 0.818, p = 0.459; TuS Drd2+ ChR2 t(4) = 0.253, p = 0.813; Mean ± SEM.

Fig. 5. BLA Drd1+ and Drd2+ neurons innervating the ventral striatum support pavlovian fear learning depending upon projection target.

A Paradigm for DREADD induced silencing of NAc or TuS projecting Drd1+ or Drd2+ BLA cells. Some aspects created in https://BioRender.com. B Influence of DREADD agonist J60 (100 nL, 10 nM) on fear learning in all NAc injected mice, C left NAc injected mCherry controls (****p < 0.0001, Two-way RM ANOVA, Trial main effect F(1,27) = 86.2, p < 0.001, n = 10M/19F); D middle Drd1+ hM4D(Gi) mice (****p < 0.0001, **p < 0.01, Two-way RM ANOVA, Trial main effect F(1,13) = 48.9, p < 0.001, n = 6M/9F); and E right Drd2+ hM4D(Gi) mice (****p < 0.0001, Two-way RM ANOVA, Trial main effect F(1,13) = 301, p < 0.001, n = 8M/7F). F Influence of DREADD agonist J60 (100 nL, 10 nM) on fear learning in all TuS injected mice, G left mCherry TuS injected controls (****p < 0.0001, Two-way RM ANOVA, Trial main effect F(1,29) = 151, p < 0.001, n = 15M/16F); H middle Drd1+ hM4D(Gi) mice (****p < 0.0001, Two-way RM ANOVA, Trial main effect F(1,13) = 85.6, p < 0.001, n = 7M/8F); and I right Drd2+ hM4D(Gi) mice (***p < 0.001, ****p < 0.0001, Two-way RM ANOVA, F(1,12) = 7.71, p = 0.017, n = 7M/8F). mCherry control groups were collapsed across genotypes, containing both Drd1-Cre and Drd2-Cre mice since these behavioral results were not different from each other (All Two-way RM ANOVAs genotype main effect: vehicle treated Drd1-Cre vs Drd2-Cre NAc mCherry controls, F(1,13) = 0.359, p = 0.559; J60 treated Drd1-Cre vs Drd2-Cre NAc mCherry controls, F(1,12) = 0.323, p = 0.580; vehicle treated Drd1-Cre vs Drd2-Cre TuS mCherry controls, F(1,14) = 0.181, p = 0.677; and J60 treated Drd1-Cre vs Drd2-Cre TuS mCherry controls, F(1,13) = 3.58, p = 0.081). Behaviors differed by sex for Drd1+ hM4D(Gi) mice suggesting sex differences in output to NAc among Drd1+ neurons (all Three-way ANOVAs for sex main effect: NAc mCherry controls F(1,25) = 0.0354, p = 0.852; Drd1+ hM4D(Gi) NAc F(1,11) = 14.0, p = 0.003; Drd2+ hM4D(Gi) NAc F(1,11) = 0.215, p = 0.652; TuS mCherry controls F(1,27) = 2.81, p = 0.102; Drd1+ hM4D(Gi) TuS F(1,11) = 0.617, p = 0.449; Drd2+ hM4D(Gi) TuS F(1,10) = 0.474, p = 0.507. Mean ± SEM.

Fig. 6. Overview of findings illustrating the behavioral consequences of activating (left) or suppressing (right) Drd1+ or Drd2+ BLA neuron inputs to the NAc and TuS.

Solid lines annotate pathways found herein to play a role in real-time avoidance and learned aversion. Dashed lines indicate pathways that were not found to support either of those behaviors/states, at least using the approaches applied in the present study.