Article | Published:

Cannabinoid CB1 receptors in the amygdalar cholecystokinin glutamatergic afferents to nucleus accumbens modulate depressive-like behavior

Abstract

Major depressive disorder is a devastating psychiatric disease that afflicts up to 17% of the world’s population. Postmortem brain analyses and imaging studies of patients with depression have implicated basal lateral amygdala (BLA) dysfunction in the pathophysiology of depression. However, the circuit and molecular mechanisms through which BLA neurons modulate depressive behavior are largely uncharacterized. Here, in mice, we identified that BLA cholecystokinin (CCK) glutamatergic neurons mediated negative reinforcement via D2 medium spiny neurons (MSNs) in the nucleus accumbens (NAc) and that chronic social defeat selectively potentiated excitatory transmission of the CCKBLA–D2NAc circuit in susceptible mice via reduction of presynaptic cannabinoid type-1 receptor (CB1R). Knockdown of CB1R in the CCKBLA–D2NAc circuit elevated synaptic activity and promoted stress susceptibility. Notably, selective inhibition of the CCKBLA–D2NAc circuit or administration of synthetic cannabinoids in the NAc was sufficient to produce antidepressant-like effects. Overall, our studies reveal the circuit and molecular mechanisms of depression.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Data availability

The datasets that support the findings of this study are available from the corresponding author upon reasonable request. All requests for raw and analyzed data and materials are promptly reviewed by the Center of Neuroscience of Zhejiang University to verify whether the request is subject to any intellectual property or confidentiality obligations. Any data and materials that can be shared will be released via a Material Transfer Agreement.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Change history

  • 30 January 2019

    In the version of this article originally published, there were several errors in Fig. 4. In Fig. 4a, the title read ‘3D repeated optical inhibition after CSDS.’ It should have read ‘3-day repeated optical inhibition after CSDS.’ In Fig. 4c, two labels that should have been aligned with the time axis appeared in the wrong place in the figure. The ticks labeled ‘SI’ and ‘Fiber implant’ should have also been labeled with ‘10’ and ‘14,’ respectively. Additionally, in Fig. 4j, a label that should have been aligned with the time axis appeared in the wrong place in the figure. The tick labeled ‘Fiber implant’ should have also been labeled with ‘14.’ The errors have been corrected in the print, PDF and HTML versions of the manuscript.

References

  1. 1.

    Pezawas, L. et al. 5-HTTLPR polymorphism impacts human cingulate-amygdala interactions: a genetic susceptibility mechanism for depression. Nat. Neurosci. 8, 828–834 (2005).

  2. 2.

    Ashtari, M. et al. Hippocampal/amygdala volumes in geriatric depression. Psychol. Med. 29, 629–638 (1999).

  3. 3.

    Ressler, K. J. & Mayberg, H. S. Targeting abnormal neural circuits in mood and anxiety disorders: from the laboratory to the clinic. Nat. Neurosci. 10, 1116–1124 (2007).

  4. 4.

    Hu, H. Reward and Aversion. Annu. Rev. Neurosci. 39, 297–324 (2016).

  5. 5.

    Russo, S. J. & Nestler, E. J. The brain reward circuitry in mood disorders. Nat. Rev. Neurosci. 14, 609–625 (2013).

  6. 6.

    Proulx, C. D., Hikosaka, O. & Malinow, R. Reward processing by the lateral habenula in normal and depressive behaviors. Nat. Neurosci. 17, 1146–1152 (2014).

  7. 7.

    Hamilton, J. P., Siemer, M. & Gotlib, I. H. Amygdala volume in major depressive disorder: a meta-analysis of magnetic resonance imaging studies. Mol. Psychiatr. 13, 993–1000 (2008).

  8. 8.

    Drevets, W. C., Bogers, W. & Raichle, M. E. Functional anatomical correlates of antidepressant drug treatment assessed using PET measures of regional glucose metabolism. Eur. Neuropsychopharm. 12, 527–544 (2002).

  9. 9.

    Drevets, W. C. et al. A functional anatomical study of unipolar depression. J. Neurosci. 12, 3628–3641 (1992).

  10. 10.

    Hamilton, J. P. & Gotlib, I. H. Neural substrates of increased memory sensitivity for negative stimuli in major depression. Biol. Psychiatr. 63, 1155–1162 (2008).

  11. 11.

    Stuber, G. D. et al. Excitatory transmission from the amygdala to nucleus accumbens facilitates reward seeking. Nature 475, 377–380 (2011).

  12. 12.

    Ambroggi, F., Ishikawa, A., Fields, H. L. & Nicola, S. M. Basolateral amygdala neurons facilitate reward-seeking behavior by exciting nucleus accumbens neurons. Neuron 59, 648–661 (2008).

  13. 13.

    Kim, J., Pignatelli, M., Xu, S. Y., Itohara, S. & Tonegawa, S. Antagonistic negative and positive neurons of the basolateral amygdala. Nat. Neurosci. 19, 1636–1646 (2016).

  14. 14.

    Valverde, O. & Torrens, M. CB1 receptor-deficient mice as a model for depression. Neuroscience 204, 193–206 (2012).

  15. 15.

    Derks, N. M. et al. Cannabinoid modulation of midbrain urocortin 1 neurones during acute and chronic stress. J. Neuroendocrinol. 24, 1447–1461 (2012).

  16. 16.

    Aso, E. et al. BDNF impairment in the hippocampus is related to enhanced despair behavior in CB1 knockout mice. J. Neurochem. 105, 565–572 (2008).

  17. 17.

    Steiner, M. A., Marsicano, G., Wotjak, C. T. & Lutz, B. Conditional cannabinoid receptor type 1 mutants reveal neuron subpopulation-specific effects on behavioral and neuroendocrine stress responses. Psychoneuroendocrinology 33, 1165–1170 (2008).

  18. 18.

    Martin, M., Ledent, C., Parmentier, M., Maldonado, R. & Valverde, O. Involvement of CB1 cannabinoid receptors in emotional behaviour. Psychopharmacology (Berl.) 159, 379–387 (2002).

  19. 19.

    Sanchis-Segura, C., Cline, B. H., Marsicano, G., Lutz, B. & Spanagel, R. Reduced sensitivity to reward in CB1 knockout mice. Psychopharmacology (Berl.) 176, 223–232 (2004).

  20. 20.

    Beyer, C. E. et al. Depression-like phenotype following chronic CB1 receptor antagonism. Neurobiol. Dis. 39, 148–155 (2010).

  21. 21.

    Christensen, R., Kristensen, P. K., Bartels, E. M., Blidda, H. & Astrup, A. Efficacy and safety of the weight-loss drug rimonabant: a meta-analysis of randomised trials. Lancet 370, 1706–1713 (2007).

  22. 22.

    Jiang, W. et al. Cannabinoids promote embryonic and adult hippocampus neurogenesis and produce anxiolytic- and antidepressant-like effects. J. Clin. Invest. 115, 3104–3116 (2005).

  23. 23.

    Bambico, F. R., Katz, N., Debonnel, G. & Gobbi, G. Cannabinoids elicit antidepressant-like behavior and activate serotonergic neurons through the medial prefrontal cortex. J. Neurosci. 27, 11700–11711 (2007).

  24. 24.

    Sherrin, T. et al. Chronic stimulation of corticotropin-releasing factor receptor 1 enhances the anxiogenic response of the cholecystokinin system. Mol. Psychiatr. 14, 291–307 (2009).

  25. 25.

    Bowers, M. E. & Ressler, K. J. Interaction between the cholecystokinin and endogenous cannabinoidsystems in cued fear expression and extinction retention. Neuropsychopharmacology 40, 688–700 (2015).

  26. 26.

    de la Mora, M. P. et al. Role of the amygdaloid cholecystokinin (CCK)/gastrin-2 receptors and terminal networks in the modulation of anxiety in the rat. Effects of CCK-4 and CCK-8S on anxiety-like behaviour and [H-3]GABA release. Eur. J. Neurosci. 26, 3614–3630 (2007).

  27. 27.

    Savanthrapadian, S. et al. Synaptic properties of SOM- and CCK-expressing cells in dentate gyrus interneuron networks. J. Neurosci. 34, 8197–8209 (2014).

  28. 28.

    Medrihan, L. et al. Initiation of behavioral response to antidepressants by cholecystokinin neurons of the dentate gyrus. Neuron 95, 564–576 (2017).

  29. 29.

    Cai, H. J., Haubensak, W., Anthony, T. E. & Anderson, D. J. Central amygdala PKC-delta+ neurons mediate the influence of multiple anorexigenic signals. Nat. Neurosci. 17, 1240–1248 (2014).

  30. 30.

    Wall, N. R., De La Parra, M., Callaway, E. M. & Kreitzer, A. C. Differential innervation of direct- and indirect-pathway striatal projection neurons. Neuron 79, 347–360 (2013).

  31. 31.

    Kreitzer, A. C. & Malenka, R. C. Endocannabinoid-mediated rescue of striatal LTD and motor deficits in Parkinson’s disease models. Nature 445, 643–647 (2007).

  32. 32.

    Nestler, E. J. & Carlezon, W. A. The mesolimbic dopamine reward circuit in depression. Biol. Psychiatr. 59, 1151–1159 (2006).

  33. 33.

    Francis, T. C. & Lobo, M. K. Emerging role for nucleus accumbens medium spiny neuron subtypes in depression. Biol. Psychiatr. 81, 645–653 (2017).

  34. 34.

    Golden, S. A., Covington, H. E., Berton, O. & Russo, S. J. A standardized protocol for repeated social defeat stress in mice. Nat. Protoc. 6, 1183–1191 (2011).

  35. 35.

    Francis, T. C. et al. Nucleus accumbens medium spiny neuron subtypes mediate depression-related outcomes to social defeat stress. Biol. Psychiatr. 77, 212–222 (2015).

  36. 36.

    Friedman, A. K. et al. Enhancing depression mechanisms in midbrain dopamine neurons achieves homeostatic resilience. Science 344, 313–319 (2014).

  37. 37.

    Beyeler, A. et al. Divergent routing of positive and negative information from the amygdala during memory retrieval. Neuron 90, 348–361 (2016).

  38. 38.

    Lu, H. C. & Mackie, K. An introduction to the endogenous cannabinoid system. Biol. Psychiatr. 79, 516–525 (2016).

  39. 39.

    Araque, A., Castillo, P. E., Manzoni, O. J. & Tonini, R. Synaptic functions of endocannabinoid signaling in health and disease. Neuropharmacology 124, 13–24 (2017).

  40. 40.

    Piomelli, D. The molecular logic of endocannabinoid signalling. Nat. Rev. Neurosci. 4, 873–884 (2003).

  41. 41.

    Vogel, E., Krabbe, S., Grundemann, J., Cusulin, J. I. W. & Luthi, A. Projection-specific dynamic regulation of inhibition in amygdala micro-circuits. Neuron 91, 644–651 (2016).

  42. 42.

    Eggan, S. M., Melchitzky, D. S., Sesack, S. R., Fish, K. N. & Lewis, D. A. Relationship of cannabinoid Cb1 receptor and cholecystokinin immunoreactivity in monkey dorsolateral prefrontal cortex. Neuroscience 169, 1651–1661 (2010).

  43. 43.

    Shin, S. et al. mGluR5 in the nucleus accumbens is critical for promoting resilience to chronic stress. Nat. Neurosci. 18, 1017–1024 (2015).

  44. 44.

    D’Agostino, G. et al. Appetite controlled by a cholecystokinin nucleus of the solitary tract to hypothalamus neurocircuit. eLife 5, e1225 (2016).

  45. 45.

    Wu, Y. E., Pan, L., Zuo, Y., Li, X. & Hong, W. Detecting activated cell populations using single-cell RNA-seq. Neuron 96, 313–329 e316 (2017).

  46. 46.

    Hamilton, P. J. et al. Cell-type-specific epigenetic editing at the Fosb gene controls susceptibility to social defeat stress. Neuropsychopharmacology 43, 272–284 (2018).

  47. 47.

    Kim, J., Zhang, X. Y., Muralidhar, S., LeBlanc, S. A. & Tonegawa, S. Basolateral to central amygdala neural circuits for appetitive behaviors. Neuron 93, 1464–1479 .e5 (2017).

  48. 48.

    Hoffman, A. F. & Lupica, C. R. Direct actions of cannabinoids on synaptic transmission in the nucleus accumbens: a comparison with opioids. J. Neurophysiol. 85, 72–83 (2001).

  49. 49.

    Grueter, B. A., Brasnjo, G. & Malenka, R. C. Postsynaptic TRPV1 triggers cell type-specific long-term depression in the nucleus accumbens. Nat. Neurosci. 13, 1519–1525 (2010).

  50. 50.

    Lutz, B. Endocannabinoid signals in the control of emotion. Curr. Opin. Pharmacol. 9, 46–52 (2009).

  51. 51.

    Lutz, B., Marsicano, G., Maldonado, R. & Hillard, C. J. The endocannabinoid system in guarding against fear, anxiety and stress. Nat. Rev. Neurosci. 16, 705–718 (2015).

  52. 52.

    Haring, M., Guggenhuber, S. & Lutz, B. Neuronal populations mediating the effects of endocannabinoids on stress and emotionality. Neuroscience 204, 145–158 (2012).

  53. 53.

    Chaudhury, D. et al. Rapid regulation of depression-related behaviours by control of midbrain dopamine neurons. Nature 493, 532–536 (2013).

  54. 54.

    Mathur, B. N. & Lovinger, D. M. Endocannabinoid-dopamine interactions in striatal synaptic plasticity. Front. Pharmacol. 3, 66 (2012).

  55. 55.

    Kreitzer, A. C. & Malenka, R. C. Striatal plasticity and basal ganglia circuit function. Neuron 60, 543–554 (2008).

  56. 56.

    Han, J. et al. Acute cannabinoids impair working memory through astroglial CB1 receptor modulation of hippocampal LTD. Cell 148, 1039–1050 (2012).

  57. 57.

    Zhou, T. et al. History of winning remodels thalamo-PFC circuit to reinforce social dominance. Science 357, 162–168 (2017).

  58. 58.

    Xiong, Q., Znamenskiy, P. & Zador, A. M. Selective corticostriatal plasticity during acquisition of an auditory discrimination task. Nature 521, 348–351 (2015).

  59. 59.

    Toledo-Rodriguez, M. & Markram, H. Single-cell RT–PCR, a technique to decipher the electrical, anatomical, and genetic determinants of neuronal diversity. Meth. Mol. Biol. 403, 123–139 (2007).

  60. 60.

    Li, Y. et al. A distinct entorhinal cortex to hippocampal CA1 direct circuit for olfactory associative learning. Nat. Neurosci. 20, 559–570 (2017).

  61. 61.

    Berton, O. et al. Essential role of BDNF in the mesolimbic dopamine pathway in social defeat stress. Science 311, 864–868 (2006).

Download references

Acknowledgements

We thank Z. J. Huang (Cold Spring Harbor Laboratory, USA) and M. He (Fudan University, China) for providing CCK-ires-Cre mice; X. Zhang (Qingdao University, China) and F. Wang (The Fourth Military Medical University, China) for providing CB1R agonist HU210; and Y.-H. Cui (Zhejiang University, China) for stimulating discussions and critical comments. This work was supported by Major Research Plan of the National Natural Science Foundation of China (grant 91432306), Key Project of the National Natural Science Foundation of China (grant 31430034), National Key Research and Development Plan (grant 2016YFA0501000), Non-Profit Central Research Institute Fund of the Chinese Academy of Medical Sciences (grants 2017PT31038 and 2018PT31041), Funds for Creative Research Groups of China (grant 81521062), Program for Changjiang Scholars and Innovative Research Teams in University and Program of Introducing Talents of Discipline to Universities (grant B13026) to X.-M.L.

Author information

C.-J.S. and X.-M.L. designed the research and wrote the manuscript. C.-J.S. performed the in vitro patch–clamp experiments, optogenetic/pharmacogenetic behavioral experiments and immunohistochemical experiments with the help of J.-M.Y., Y.Z., P.S., X.-D.Y., J.-Y.F. and Y.Z. D.Z. performed the virus injections, in vivo recordings and behavioral pharmacological experiments with the help of Q.-X.S., M.-Y.T. and H.-Q.P. K.-X.L. performed the biochemical and in situ hybridization experiments with the help of Y.X. H.H. and S.D. reviewed and edited the manuscript. X.-M.L. supervised all phases of the project.

Competing interests

The authors declare no competing interests.

Correspondence to Xiao-Ming Li.

Extended data

  1. Extended Data Fig. 1 CCK mRNA expression in BLA and the downstream targets of BLA CCK glutamatergic neurons.

    a, Left, coronal representative brain image of CCK mRNA in BLA; scale bar, 500 μm. Right, magnified view of BLA; scale bar, 200 μm. b, Zoomed-out view of BLA CCK-tdTomato neurons (red) from a CCK-ires-Cre::Ai14 mouse co-immunostained for CaMKIIα and GAD67. Scale bar, 200 μm. c, Coronal brain slice (stained with DAPI) from a CCK-ires-Cre mouse with AAV-CaMKIIα-Cre-on-ChR2-mCherry (a) virus injected into BLA (left, injection site) and its axonal terminals in NAcc (right); scale bars, 200 μm for BLA and 300 μm for NAcc. d, Overlay of Cre-on-ChR2-mCherry (n = 8 mice) expression areas in BLA (left, injection site) or NAcc (right). e, Representative image (left) and quantification (right) of axonal terminals in vHPC of Cre-on-ChR2-mCherry or Cre-out-ChR2-eYFP virus-injected mice; scale bar, 200 μm. Two-sided unpaired t-test, t = 9.942, d.f. = 16, P < 0.0001, n = 8, 10 mice for each group, respectively. f, Quantification of in vivo optogenetic activation of CCKBLA-NAcc glutamatergic neurons, with intra-NAcc local infusion of L-365,260 (CCKA receptor antagonist, 1 μg in 200 nl) or L-364,718 (CCKB receptor antagonist, 0.1 μg or 1 μg in 200 nl) in NAcc during real-time place avoidance. One-way ANOVA, F(4, 33) = 12.97, P < 0.0001, n = 6, 8, 8, 8, 8 mice for each group. All data are means ± s.e.m. ****P < 0.0001; n.s., no significance.

  2. Extended Data Fig. 2 CCK, Rspo2, Ppp1r1b mRNA expression in BLA and optogenetic manipulation changes neither locomotion nor anxiety.

    a, Schematic of open-field test (OFT) with intermittent photostimulation. b, Mean total distance per minute across 10 min OFT in Cre-on ChR2 or Cre-out ChR2 mice. c,d, Quantification of total distance per minute (c) and duration in center zone (d) in light ON and OFF periods from Cre-on ChR2 or Cre-out ChR2 mice. Two-way ANOVA in c, F(1, 10) = 0.01743, P = 0.8976. Two-way ANOVA in d, F(1, 10) = 0.002206, P = 0.9635, n = 6, 6 mice for Cre-on ChR2 and Cre-out ChR2 groups, respectively. e, Schematic of elevated plus maze (EPM) with intermittent photostimulation. f,g, Duration spent within open arms in light ON and OFF periods for mice expressing Cre-on-ChR2 (or mCherry control, f) or Cre-out-ChR2 (or eYFP control, g) virus in BLA. Two-way ANOVA in f, F(2, 20) = 0.02805, P = 0.9724, n = 6, 6 mice for mCherry and Cre-on ChR2 groups, respectively. Two-way ANOVA in g, F(2, 20) = 0.005046, P = 0.9950, n = 6, 6 mice for mCherry and Cre-out ChR2 groups, respectively. h, Left, representative images CCK, R-spondin 2 (Rspo2) and protein phosphatase 1 regulatory subunit 1B (Ppp1r1b) mRNA expression in BLA, scale bar 100 μm. Right, representative image showing magnified view of white rectangle; arrows indicate neurons co-labeled with CCK mRNA; scale bar, 50 μm. i, Quantification of CCK mRNA co-labeled with Rspo2 or Ppp1r1b mRNA. Two-way ANOVA, F(2, 24) = 1.371, P = 0.2730, n = 926, 1220, 480 CCK-positive neurons for anterior, intermediate and posterior BLA from 5 mice. All data are means ± s.e.m. ***P < 0.001; ****P < 0.0001; n.s., no significance.

  3. Extended Data Fig. 3 Quantitative analysis of whole-brain inputs to D1 and D2 MSNs in NAcc.

    a, Monosynaptic retrograde rabies tracing from D1 and D2 neurons in NAcc. b, Left, coronal sections through NAcc of D1-Cre (left) and D2-Cre (right) tracing brains showing location of starter cells; scale bar, 1 mm. Right, quantification of starter cells in D1-Cre (n = 5) and D2-Cre (n = 6) mice. Two-sided unpaired t-test, t = 1.4443, d.f. = 9, n = 5, 6 mice for each group. c,d, Top, coronal sections of a D1-Cre (c) and a D2-Cre (d) tracing brain showing distribution of presynaptic partners; scale bar, 1 mm. Bottom, magnified images of rectangular regions in top images; scale bar, 250 μm. e, Inputs to D1 (n = 5) and D2 (n = 6) neurons from whole-brain regions, shown as proportion of total number of cells counted that are located in a region. Striat, striatum;hypoth, hypothalamus; thal, thalamus; amyg, amygdala; HP, hippocampus. Two-way ANOVA, F(28,261) = 3.141, P < 0.0001, n = 5, 6 mice for each group. f,g, Top, coronal sections of D1-Cre (f) and D2-Cre (g) tracing brains showing distribution of presynaptic partners in anterior (f1,g1), intermediate (f2,g2) and posterior (f3,g3) parts of amygdala; scale bar, 1 mm. Bottom, magnified images of rectangular regions in top images; scale bar, 200 μm. h, Comparison between NAcc D1-projecting (D1-P) and D2-projecting (D2-P) neurons in BLA and CeA. Two-way ANOVA, F(1,18) = 1.753, P = 0.2021, n = 5, 6 mice for each group. i, Detection of CCK, CaMKIIα and CB1 transcripts in rabies-labeled BLA neurons by single-cell RT–PCR. j–l, Percentages of cells positive for CCK (j), CaMKIIα (k) and CB1 (l) in D1-P (n = 16 neurons from 5 mice) or D2-P (n = 18 neurons from 5 mice) neurons in BLA (rabies-DsRed identified). Fisher’s exact test, P < 0.0001 in j, P = 1.0 in k, P < 0.0001 in l. All data are means ± s.e.m. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; n.s., no significance.

  4. Extended Data Fig. 4 Evidence of target specificity in D2-GFP and D1-tdTomato mice.

    a,b, Schematic (a) and representative images (b) of NAcc from a D2-GFP mouse, triple-labeled for D2−GFP, D1 and D2 mRNA, scale bar, 200 μm. c, Magnified view from b. Arrowheads indicate D2GFP-positive neurons whereas arrows indicate co-labeled neurons with D2−GFP; scale bar, 50 μm. d, Scaled Venn diagram showing number of D1- and D2-mRNA-positive neurons in NAcc. e, Percentage of NAcc D1−and D2-mRNA-positive neurons co-labeled with D2-GFP-positive neurons (D2-GFP+). Two-sided unpaired t-test in e, t = 133.3, d.f. = 8, P < 0.0001, n = 2658 D2-GFP+ neurons in total from 5 mice. f, Representative image of D1-tdTomato-labeled neurons in NAcc of D1-tdTomato mice; scale bar, 40 μm. g, Representative images of recorded D1-tdTomato-positive (D1+, left) and D1-tdTomato-negative (D1, right) neurons in NAcc using in vitro slice recording. Biocytin was used to indicate recorded neurons; scale bar, 20 μm. h, Representative traces (left) of whole-cell current clamp recordings from D1+ (n = 16 neurons from 3 mice) or D2(n = 16 neurons from 3 mice) MSNs in vitro and current and voltage (IV) curves (right) of D1+ and D1 MSNs. Raw traces show individual voltage responses to a series of 600 ms current pulses from −300 to 300 pA in 200 pA steps. i, Schematic showing tested connections in CCK-ires-Cre::D1-tdTomato mice; non-CCK neurons were transduced by injection of AAV-CaMKIIα-Cre-out-ChR2-eYFP in BLA. j, Left, light responses recorded from two adjacent D1+ or D1 MSNs following 5 ms laser stimulation of non-CCK terminals from BLA. Right, connectivity charts are shown. k, Quantification of amplitude (left) and latency (right) of oEPSCs recorded in NAcc D1+ and D1 MSNs from non-CCK glutamatergic neurons. Two-sided unpaired t-test, t = 9.863, d.f. = 42, P < 0.0001, in amplitude. t = 0.2955, d.f. = 42, P = 0.7691; in latency. n = 38 out of 40 D1+ MSNs (95%) and 6 out of 40 D1 MSNs (15%) from 6 mice. All data are means ± s.e.m. ****P < 0.0001; n.s., no significance.

  5. Extended Data Fig. 5 mEPSC frequencies are increased in D2+ neurons in susceptible mice.

    a,b, Schematic of 10-day chronic social defeat stress (10-d CSDS) procedure (a) and social interaction test (SI, b). c, Left, distribution of interaction ratios. One-way ANOVA, F(2, 39) = 76.55, P < 0.0001, n = 12, 16, 14 mice for each group. Right, quantification in social interaction showing that susceptible mice spent less time interacting with a novel CD1 mouse, whereas resilient mice interacted the same as control mice. Two-way ANOVA, F(2, 39) = 50.58, P < 0.001, n = 12, 16, 14 mice for each group. d,e, Sucrose preference in sucrose preference test (SPT) was decreased and total immobility time in tail-suspension test (TST) was increased in susceptible mice. One-way ANOVA in d, F(2, 39) = 59.02, P < 0.001. One-way ANOVA in e, F(2, 39) = 52.59, P < 0.001, n = 12, 14, 16 mice for each group. f,i, Example mEPSC traces, measured in a whole-cell configuration of D2+ (f) or D2 neurons (i) in NAc. g,j, Cumulative distribution of mEPSC inter-event intervals of D2+ (g) and D2 (j) MSNs recorded in NAcc. h,k, Average mEPSC frequency (left) and amplitude (right) measured in D2+ (h) and D2 (k) MSNs. One-way ANOVA of frequency (h, D2+), F(2, 23) = 135.0, P < 0.0001. One-way ANOVA in amplitude (h, D2+), F(2, 23) = 0.3686, P = 0.6957, n = 8, 10, 8 neurons, each from 4 mice. One-way ANOVA of frequency (k, D2), F(2, 21) = 0.1732, P = 0.8422. One-way ANOVA in amplitude (k, D2), F(2, 21) = 0.6912, P = 0.5120, n = 8, 8, 8 neurons, each from 4 mice, respectively. All data are means ± s.e.m.****P < 0.0001; n.s., no significance.

  6. Extended Data Fig. 6 Photo-inhibition of CCKBLA–D2NAcc circuit prevents the acquisition and expression of social avoidance.

    a, Schematic of virus injection to express Arch3.0 in BLA CCK neurons and optical fiber implantation in NAcc. b, Left, Arch3.0-eYFP expression in BLA; scale bar, 1 mm. Right, magnified image; scale bar, 200 μm. c, Daily photo-inhibition of CCKBLA–D2NAcc circuit during 10 min of sensory contact in non-CSDS control mice (left) and susceptible mice (right, after 5 min physical aggression). d, Social interaction ratio in each group. Two-way ANOVA, F(1,36) = 12.53, P = 0.0011, n = 10, 10, 10, 10 mice for each group. e, Proportions of susceptible, indifferent and resilient mice in eYFP and Arch3.0 group (defeat). f, Sucrose preference in SPT. One-way ANOVA, F(3,36) = 16.72, P < 0.0001, n = 10, 10, 10, 10 mice for each group. g, Total immobility time in TST. One-way ANOVA, F(3,36) = 6.847, P = 0.0009, n = 10, 10, 10, 10 mice for each group. h, Schematic of phasic photo-inhibition of the CCKBLA–D2NAcc circuit during 2.5 min social interaction test. i,j, Quantification of time in social interaction zone in susceptible mice (i) and non-CSDS control mice (j) expressing either eYFP or Arch 3.0. Two-way ANOVA in i, F(1,28) = 32.66, P < 0.0001. Two-way ANOVA in j, F(1,20) = 0.7574, P < 0.0001, n = 8, 8 mice for each group, respectively. Two-way ANOVA in j, F(1,20) = 0.7574, P = 0.3945, n = 6, 6 mice for each group, respectively. All data are means ± s.e.m.****P < 0.0001; n.s., no significance.

  7. Extended Data Fig. 7 Pharmacogenetic bidirectional effects of modulating CCKBLA–D2NAcc circuit on stress susceptibility.

    a,f, Schematic illustrating AAV-CaMKIIα-Cre-on-hM4Di-mCherry (a) and AAV-CaMKIIα-Cre-on-hM3Dq-mCherry (f) viralbilateral injection into BLA of CCK-ires-Cre mice and cannula implantation in NAcc for local infusion of CNO (3 μM, 100 nl). Intra-NAcc infusion of CNO via cannula selectively inhibits/activates synaptic activity via hM3Dq- or hM4Di-mediated activation or inhibition in NAcc, respectively. b,g, Paradigms of 3-day repeated pharmacogenetic inhibition of the CCKBLA–D2NAcc circuit in susceptible mice following 10 days of chronic social defeat stress (CSDS) (b), or acute pharmacogenetic activation of CCKBLA–D2NAcc circuit during social interaction in a two-trial subthreshold social defeat stress (SSDS) paradigm (g). c,h, Social interaction time in the absence or presence of social target. Two-way ANOVA in c, F(1, 40) = 5.316, P = 0.0259, n = 12, 12 mice, respectively. Two-way ANOVA in h, F(1, 36) = 7.301, P = 0.0104, n = 8, 10 mice, respectively. d and i, Sucrose preference in SPT. Two-sided unpaired t-test in d, t = 2.677, d.f. = 22, P = 0.0259, n = 12, 12 mice, respectively. Two-sided unpaired t-test in i, t = 4.002, d.f. = 20, P = 0.0259, n = 8, 14 mice, respectively. e,j, Total immobility time in TST. Two-sided unpaired t-test in e, t = 3.229, d.f. = 22, P = 0.0259, n = 12, 12 mice. Two-sided unpaired t-test in j, t = 4.597, d.f. = 20, P = 0.0259, n = 8, 14 mice. k–m, Representative traces (k) of cell-attached slice recording from BLA CCK neurons expressing hM3Dq-mCherry that was silenced by application of CNO (5 μM, 50 s). CNO induced rapid depolarization of membrane potential (l) and greatly increased firing rate (m), but did not affect membrane potential and firing rates of neurons expressing only mCherry (control). Two-way ANOVA in l (left), F(1, 8) = 53.37, P < 0.0001. Two-sided unpaired t-test in l (right), t = 7.306, d.f. = 8, P < 0.0001. Two-way ANOVA in m, F(1, 8) = 295.2, P < 0.0001. n = 5, 5 neurons, each from 3 mice for mCherry and hM3Dq groups, respectively. All data are mean ± s.e.m. **P < 0.01; ***P < 0.001; ****P < 0.0001; n.s., no significance.

  8. Extended Data Fig. 8 Absence of CB1R in non-CCKBLA–D1NAcc circuit and modulation of CB1R on stress susceptibility.

    a, Schematic of virus injection to express CaMKIIα-Cre-out-ChR2-eYFP in BLA of CCK-ires-Cre::D1-tdTomato mice and in vitro slice recording in D1-tdTomato-positive (D1+) neurons in NAcc. b, Left, normalized oEPSCs following application of CB1 agonist WIN55,212-2 (10 min, 1 μM). Right, representative traces of oEPSCs (top) and 50 ms PPR (bottom) in absence (left) and presence (right) of WIN55,212-2. c, Left, relative (% normalized to baseline) BLA non-CCK to D1 oEPSC amplitude (top) and 50 ms PPR (right) following WIN55,212-2 application (10 min, 1 μM). Two-sided paired t-test in c (left), t = 0.7557, d.f. = 5, P = 0.4839. Two-sided paired t-test in c (right), t = 0.8358, d.f. = 5, P = 0.9366. n = 6 neurons from 3 mice. d, Schematic of local bilateral infusion of AM251 (0.5 μg, 100 μl each side) into NAcc during two-trial SSDS. e–g, Effect of local bilateral infusion of AM251 into NAcc during SSDS on social interaction time (e), SPT (f) and TST (g), Two-way ANOVA in e, F(1,32) = 8.067, P = 0.078; two-sided unpaired t-test in f, t = 6.686, d.f. = 16, P < 0.0001; two-sided unpaired t-test in g, t = 3.810, d.f. = 16, P = 0.0015. n = 8, 10 mice, respectively. h, Paradigms of 3-day repeated intra-NAcc local infusion of WIN55,212-2 (0.5 μg, 100 μl each side) in susceptible mice following 10 days of CSDS. i, Social interaction time in absence or presence of social target. Two-way ANOVA, F(1,20) = 0.06534, P = 0.8009, n = 6, 6 mice for each group, respectively. j, Sucrose preference in SPT. Two-sided unpaired t-test, t = 0.9594, d.f. = 10, P = 0.3600, n = 6, 6 mice for each group, respectively. k, Total immobile time in TST. Two-sided unpaired t-test, t = 1.548, d.f. = 10, P = 0.1527, n = 6, 6 mice for each group, respectively. All data are means ± s.e.m. **P < 0.01; ***P < 0.001; ****P < 0.0001; n.s., no significance.

  9. Extended Data Fig. 9 In vivo fEPSP recordings of synaptic strength in CCKBLA-NAcc circuit and histological verification of optrode/cannula placement.

    a, Recording method used to examine CCKBLA-NAcc synaptic strength in vivo. b, fEPSPs aligned to light stimulation from recordings in NAcc of anaesthetized mice without (left) and with (right) ChR2 expressed in BLA. Each light-evoked fEPSP is shown as a black trace, with the red trace representing the average. c, Left, normalized fEPSP trace evoked by photostimulation, comprising an early component (reflecting light-evoked ChR2 currents in BLA CCK axon terminals) and a late component (reflecting postsynaptic responses in NAcc). Data were normalized to the peak of the first component. Right, rising phase of late component of fEPSPs (10–90% of peak, P) was fitted linearly with the slope of the fit, a good measure of synaptic strength, used for quantification of light-evoked fEPSPs. d, Example placement of optrode in NAcc; circle indicates tip of the optrode. Dashed white lines are boundaries of subregions; scale bar, 1 mm. e–g, Placement of all optrodes and cannulae.

  10. Extended Data Fig. 10 CB1R manipulation in CCKBLA–D2NAcc circuit does not alter anxiety-like behavior.

    a, Representative animal tracks in open-field test (OFT, top) and elevated plus maze (EPM, bottom). b,c, Quantification of total distance, percentage of center duration in OFT (b) and open arms duration in EPM (c), showing that susceptible mice exhibited anxiety-like behavior. One-way ANOVA in b (total distance), F(2, 17) = 0.1236, P = 0.8845. One-way ANOVA in b (% center duration), F(2, 17) = 10.39, P = 0.0011. One-way ANOVA in c, F(2, 17) = 11.79, P = 0.0006. n = 8, 6, 6 mice for each group, respectively. d, Paradigm of 3-day repeated intra-NAcc local infusion of WIN55,212-2 (0.5 μg, 100 μl each side) in susceptible mice following 10-day CSDS. e,f, 3-day repeated intra-NAcc local infusion of WIN55,212-2 (0.5 μg, 100 μl each side) did not rescue anxiety-like behavior in susceptible mice. One-way ANOVA in e (total distance), F(3, 24) = 0.03388, P = 0.9914. One-way ANOVA in e (% center duration), F(3, 24) = 10.52, P = 0.0001. One-way ANOVA in f, F(3, 24) = 9.728, P = 0.0002. n = 6, 6, 8, 8 mice for each group, respectively. g, Representative images of BLA miR30-control-eYFP and miR30-shCB1R-eGFP virus expression in control and CB1-KD mice, co-labeled with CB1 mRNA; scale bar, 200 μm. Inset, magnified view of the rectangular region in BLA; scale bar, 20 μm. h,i, Anxiety-like behavior in mice with expression of various viral constructs in BLA. Two-sided unpaired t-test in h (total distance), t = 0.5039, d.f. = 10, P = 0.6252; Two-sided unpaired t-test in h (% center duration), t = 0.2156, d.f. = 10, P = 0.8329; Two-sided unpaired t-test in i, t = 0.4314, d.f. = 10, P = 0.6738. n = 6, 6 mice, respectively. All data are means ± s.e.m. **P < 0.01; n.s., no significance.

Supplementary information

  1. Reporting Summary

Source data

  1. Source Data Fig. 5

    Full scans of the blots related to Figure 5m

  2. Source Data Fig. 6

    Full scans of the blots related to Figure 6c

Rights and permissions

To obtain permission to re-use content from this article visit RightsLink.

About this article

Fig. 1: NAc-projecting BLA CCK and non-CCK glutamatergic neurons encode opposite emotional valences.
Fig. 2: BLA CCK and non-CCK glutamatergic neurons differentially form synaptic connections with D2 and D1 MSNs, respectively, in NAcc.
Fig. 3: Social stress activates CCKBLA–D2NAcc circuit in susceptible mice.
Fig. 4: Optogenetic inhibition or activation of CCKBLA–D2NAcc circuit bidirectionally regulates susceptibility to social stress.
Fig. 5: Reduced CB1 levels lead to increased synaptic activity in CCKBLA-D2NAcc circuit of susceptible mice.
Fig. 6: Loss- or gain-of-function of CB1 Receptors at CCKBLA–D2NAcc synapses causes or rescues depression-like behaviors.
Extended Data Fig. 1: CCK mRNA expression in BLA and the downstream targets of BLA CCK glutamatergic neurons.
Extended Data Fig. 2: CCK, Rspo2, Ppp1r1b mRNA expression in BLA and optogenetic manipulation changes neither locomotion nor anxiety.
Extended Data Fig. 3: Quantitative analysis of whole-brain inputs to D1 and D2 MSNs in NAcc.
Extended Data Fig. 4: Evidence of target specificity in D2-GFP and D1-tdTomato mice.
Extended Data Fig. 5: mEPSC frequencies are increased in D2+ neurons in susceptible mice.
Extended Data Fig. 6: Photo-inhibition of CCKBLA–D2NAcc circuit prevents the acquisition and expression of social avoidance.
Extended Data Fig. 7: Pharmacogenetic bidirectional effects of modulating CCKBLA–D2NAcc circuit on stress susceptibility.
Extended Data Fig. 8: Absence of CB1R in non-CCKBLA–D1NAcc circuit and modulation of CB1R on stress susceptibility.
Extended Data Fig. 9: In vivo fEPSP recordings of synaptic strength in CCKBLA-NAcc circuit and histological verification of optrode/cannula placement.
Extended Data Fig. 10: CB1R manipulation in CCKBLA–D2NAcc circuit does not alter anxiety-like behavior.