Central amygdala circuits modulate food consumption through a positive-valence mechanism

Abstract

The complex behaviors underlying reward seeking and consumption are integral to organism survival. The hypothalamus and mesolimbic dopamine system are key mediators of these behaviors, yet regulation of appetitive and consummatory behaviors outside of these regions is poorly understood. The central nucleus of the amygdala (CeA) has been implicated in feeding and reward, but the neurons and circuit mechanisms that positively regulate these behaviors remain unclear. Here, we defined the neuronal mechanisms by which CeA neurons promote food consumption. Using in vivo activity manipulations and Ca2+ imaging in mice, we found that GABAergic serotonin receptor 2a (Htr2a)-expressing CeA neurons modulate food consumption, promote positive reinforcement and are active in vivo during eating. We demonstrated electrophysiologically, anatomically and behaviorally that intra-CeA and long-range circuit mechanisms underlie these behaviors. Finally, we showed that CeAHtr2a neurons receive inputs from feeding-relevant brain regions. Our results illustrate how defined CeA neural circuits positively regulate food consumption.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Figure 1: CeAHtr2a neurons increase food consumption.
Figure 2: CeAHtr2a neurons are required for normal food consumption.
Figure 3: CeAHtr2a neurons increase in activity during food consumption.
Figure 4: CeAHtr2a-neuron activity is positively reinforcing and modulates reward consumption.
Figure 5: CeA neuronal modulators of feeding behavior form reciprocal inhibitory connections.
Figure 6: CeAHtr2a neuron inhibition of the PBN is rewarding and modulates consummatory behavior.
Figure 7: Inputs to CeAHtr2a neurons come from feeding-relevant brain regions.

References

  1. 1

    Grace, A.A., Floresco, S.B., Goto, Y. & Lodge, D.J. Regulation of firing of dopaminergic neurons and control of goal-directed behaviors. Trends Neurosci. 30, 220–227 (2007).

    CAS  PubMed  PubMed Central  Google Scholar 

  2. 2

    Berridge, K.C. & Robinson, T.E. What is the role of dopamine in reward: hedonic impact, reward learning, or incentive salience? Brain Res. Brain Res. Rev. 28, 309–369 (1998).

    CAS  PubMed  Google Scholar 

  3. 3

    Wise, R.A. Role of brain dopamine in food reward and reinforcement. Phil. Trans. R. Soc. Lond. B 361, 1149–1158 (2006).

    CAS  Google Scholar 

  4. 4

    Ciocchi, S. et al. Encoding of conditioned fear in central amygdala inhibitory circuits. Nature 468, 277–282 (2010).

    CAS  Google Scholar 

  5. 5

    Haubensak, W. et al. Genetic dissection of an amygdala microcircuit that gates conditioned fear. Nature 468, 270–276 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  6. 6

    Li, H. et al. Experience-dependent modification of a central amygdala fear circuit. Nat. Neurosci. 16, 332–339 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  7. 7

    Isosaka, T. et al. Htr2a-expressing cells in the central amygdala control the hierarchy between innate and learned fear. Cell 163, 1153–1164 (2015).

    CAS  PubMed  Google Scholar 

  8. 8

    Botta, P. et al. Regulating anxiety with extrasynaptic inhibition. Nat. Neurosci. 18, 1493–1500 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  9. 9

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

    CAS  PubMed  PubMed Central  Google Scholar 

  10. 10

    Robinson, M.J., Warlow, S.M. & Berridge, K.C. Optogenetic excitation of central amygdala amplifies and narrows incentive motivation to pursue one reward above another. J. Neurosci. 34, 16567–16580 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  11. 11

    Seo, D.O. et al. A GABAergic projection from the centromedial nuclei of the amygdala to ventromedial prefrontal cortex modulates reward behavior. J. Neurosci. 36, 10831–10842 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  12. 12

    Mahler, S.V. & Berridge, K.C. Which cue to “want?” Central amygdala opioid activation enhances and focuses incentive salience on a prepotent reward cue. J. Neurosci. 29, 6500–6513 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  13. 13

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

    CAS  PubMed  PubMed Central  Google Scholar 

  14. 14

    Carter, M.E., Soden, M.E., Zweifel, L.S. & Palmiter, R.D. Genetic identification of a neural circuit that suppresses appetite. Nature 503, 111–114 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  15. 15

    Han, S., Soleiman, M.T., Soden, M.E., Zweifel, L.S. & Palmiter, R.D. Elucidating an affective pain circuit that creates a threat memory. Cell 162, 363–374 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  16. 16

    Armbruster, B.N., Li, X., Pausch, M.H., Herlitze, S. & Roth, B.L. Evolving the lock to fit the key to create a family of G protein-coupled receptors potently activated by an inert ligand. Proc. Natl. Acad. Sci. USA 104, 5163–5168 (2007).

    PubMed  PubMed Central  Google Scholar 

  17. 17

    Fortin, S.M., Chartoff, E.H. & Roitman, M.F. The aversive agent lithium chloride suppresses phasic dopamine release through central GLP-1 receptors. Neuropsychopharmacology 41, 906–915 (2016).

    CAS  PubMed  Google Scholar 

  18. 18

    Bret-Dibat, J.L., Bluthé, R.M., Kent, S., Kelley, K.W. & Dantzer, R. Lipopolysaccharide and interleukin-1 depress food-motivated behavior in mice by a vagal-mediated mechanism. Brain Behav. Immun. 9, 242–246 (1995).

    CAS  PubMed  Google Scholar 

  19. 19

    Dantzer, R. Cytokine-induced sickness behavior: mechanisms and implications. Ann. NY Acad. Sci. 933, 222–234 (2001).

    CAS  PubMed  PubMed Central  Google Scholar 

  20. 20

    Levitsky, D.A. Feeding conditions and intermeal relationships. Physiol. Behav. 12, 779–787 (1974).

    CAS  PubMed  Google Scholar 

  21. 21

    Wu, Z., Autry, A.E., Bergan, J.F., Watabe-Uchida, M. & Dulac, C.G. Galanin neurons in the medial preoptic area govern parental behaviour. Nature 509, 325–330 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  22. 22

    Betley, J.N. et al. Neurons for hunger and thirst transmit a negative-valence teaching signal. Nature 521, 180–185 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  23. 23

    Chen, Y., Lin, Y.C., Kuo, T.W. & Knight, Z.A. Sensory detection of food rapidly modulates arcuate feeding circuits. Cell 160, 829–841 (2015).

    PubMed  PubMed Central  Google Scholar 

  24. 24

    Mandelblat-Cerf, Y. et al. Arcuate hypothalamic AgRP and putative POMC neurons show opposite changes in spiking across multiple timescales. eLife 4 (2015).

  25. 25

    Jennings, J.H. et al. Visualizing hypothalamic network dynamics for appetitive and consummatory behaviors. Cell 160, 516–527 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  26. 26

    Ghosh, K.K. et al. Miniaturized integration of a fluorescence microscope. Nat. Methods 8, 871–878 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  27. 27

    Stamatakis, A.M. et al. A unique population of ventral tegmental area neurons inhibits the lateral habenula to promote reward. Neuron 80, 1039–1053 (2013).

    CAS  Google Scholar 

  28. 28

    Wickersham, I.R. et al. Monosynaptic restriction of transsynaptic tracing from single, genetically targeted neurons. Neuron 53, 639–647 (2007).

    CAS  PubMed  PubMed Central  Google Scholar 

  29. 29

    Elmquist, J.K., Scammell, T.E., Jacobson, C.D. & Saper, C.B. Distribution of Fos-like immunoreactivity in the rat brain following intravenous lipopolysaccharide administration. J. Comp. Neurol. 371, 85–103 (1996).

    CAS  PubMed  Google Scholar 

  30. 30

    Yamamoto, T. et al. C-fos expression in the rat brain after intraperitoneal injection of lithium chloride. Neuroreport 3, 1049–1052 (1992).

    CAS  PubMed  Google Scholar 

  31. 31

    Schwarz, L.A. et al. Viral-genetic tracing of the input-output organization of a central noradrenaline circuit. Nature 524, 88–92 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  32. 32

    Carleton, A., Accolla, R. & Simon, S.A. Coding in the mammalian gustatory system. Trends Neurosci. 33, 326–334 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  33. 33

    Morton, G.J., Meek, T.H. & Schwartz, M.W. Neurobiology of food intake in health and disease. Nat. Rev. Neurosci. 15, 367–378 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  34. 34

    Chometton, S. et al. A premammillary lateral hypothalamic nuclear complex responds to hedonic but not aversive tastes in the male rat. Brain Struct. Funct. 221, 2183–2208 (2016).

    PubMed  Google Scholar 

  35. 35

    Dahlstroem, A. & Fuxe, K. Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies of brain stem neurons. Acta Physiol. Scand. Suppl. 232 (Suppl.), 1–55 (1964).

    Google Scholar 

  36. 36

    Liang, C.L., Sinton, C.M. & German, D.C. Midbrain dopaminergic neurons in the mouse: co-localization with Calbindin-D28K and calretinin. Neuroscience 75, 523–533 (1996).

    CAS  PubMed  PubMed Central  Google Scholar 

  37. 37

    Grewe, B.F. et al. Neural ensemble dynamics underlying a long-term associative memory. Nature 543, 670–675 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

  38. 38

    de Araujo, I.E. in Neurobiology of Sensation and Reward (ed. Gottfried, J.A.) (CRC Press/Taylor & Francis, 2011).

  39. 39

    Han, W. et al. Integrated control of predatory hunting by the central nucleus of the amygdala. Cell 168, 311–324.e18 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

  40. 40

    Fadok, J.P. et al. A competitive inhibitory circuit for selection of active and passive fear responses. Nature 542, 96–100 (2017).

    CAS  Google Scholar 

  41. 41

    Nieh, E.H. et al. Inhibitory input from the lateral hypothalamus to the ventral tegmental area disinhibits dopamine neurons and promotes behavioral activation. Neuron 90, 1286–1298 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  42. 42

    Navarro, M. et al. Lateral hypothalamus GABAergic neurons modulate consummatory behaviors regardless of the caloric content or biological relevance of the consumed stimuli. Neuropsychopharmacology 41, 1505–1512 (2016).

    CAS  PubMed  Google Scholar 

  43. 43

    Madisen, L. et al. A toolbox of Cre-dependent optogenetic transgenic mice for light-induced activation and silencing. Nat. Neurosci. 15, 793–802 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  44. 44

    Soriano, P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat. Genet. 21, 70–71 (1999).

    CAS  PubMed  PubMed Central  Google Scholar 

  45. 45

    Takeoka, A., Vollenweider, I., Courtine, G. & Arber, S. Muscle spindle feedback directs locomotor recovery and circuit reorganization after spinal cord injury. Cell 159, 1626–1639 (2014).

    CAS  PubMed  Google Scholar 

  46. 46

    Lopes, G. et al. Bonsai: an event-based framework for processing and controlling data streams. Front. Neuroinform. 9, 7 (2015).

    PubMed  PubMed Central  Google Scholar 

  47. 47

    Slotnick, B. A simple 2-transistor touch or lick detector circuit. J. Exp. Anal. Beh. 91, 253–255 (2009).

    Google Scholar 

  48. 48

    Mukamel, E.A., Nimmerjahn, A. & Schnitzer, M.J. Automated analysis of cellular signals from large-scale calcium imaging data. Neuron 63, 747–760 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  49. 49

    Paixão, S. et al. EphrinB3/EphA4-mediated guidance of ascending and descending spinal tracts. Neuron 80, 1407–1420 (2013).

    PubMed  PubMed Central  Google Scholar 

  50. 50

    Qi, J. et al. A glutamatergic reward input from the dorsal raphe to ventral tegmental area dopamine neurons. Nat. Commun. 5, 5390 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  51. 51

    Sciolino, N.R. et al. Recombinase-dependent mouse lines for chemogenetic activation of genetically defined cell types. Cell Rep. 15, 2563–2573 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  52. 52

    Lein, E.S. et al. Genome-wide atlas of gene expression in the adult mouse brain. Nature 445, 168–176 (2007).

    CAS  Google Scholar 

  53. 53

    Paxinos, G. & Franklin, K.B.J. Paxinos and Franklin's the Mouse Brain in Stereotaxic Coordinates (Elsevier Academic Press, 2013).

  54. 54

    Chung, K. et al. Structural and molecular interrogation of intact biological systems. Nature 497, 332–337 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank T. Caudullo, E. Fejzulahi and B. Hoisl for their help with management of the animal colony; R. Kasper for technical help with the in vivo optogenetics setup; T. Ruff and D. del Toro for help with tissue clearing; M. Fischer for help with histology; M. Schwarz (Max Planck Institute for Medical Research) for providing Cre-dependent AAV-mCherry virus; S. Arber (Friedrich Miescher Institute for Biomedical Research) for providing Cre-dependent AAV-Synaptophysin-myc virus; and A. Ghanem for initial help with rabies viruses. We thank P. Sah and N. Gogolla for critical reading of this manuscript. Modified rabies virus was provided by the GT3 Core Facility of the Salk Institute with funding from the NIH–NCI CCSG (P30 014195), an NINDS R24 Core Grant and funding from the NEI. This study was supported by the Max-Planck Society and the Deutsche Forschungsgemeinschaft (Synergy to R.K. and SPP1665 to K.K.C. and R.K.).

Author information

Affiliations

Authors

Contributions

A.M.D., H.K., M.P., M.M., J.G. and C.S. designed and analyzed experiments. A.M.D. performed behavior experiments. H.K. performed electrophysiology and assisted with behavior experiments. M.P. performed rabies tracing. H.K. and M.P. performed other tracing experiments. C.S. performed paired recordings and hM3D ex vivo verification. A.M.D., H.K., M.P. and P.L.A.M. performed histology. A.M.D., H.K., M.M. and J.G. performed Ca2+ imaging experiments. K.-K.C. provided rabies virus and shared expertise in its use for monosynaptic tracing. R.K. and A.L. supervised experiments. A.M.D., H.K., M.P. and R.K. wrote the manuscript, and all other authors provided input.

Corresponding author

Correspondence to Rüdiger Klein.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Validation of CeA Htr2a-Cre expression.

a, Representative coronal section through the CeA of an Htr2a-Cre mouse expressing a floxed-lacZ reporter and labeled for LacZ and Htr2a mRNAs (representative of 3 mice). White box indicates the location of the high-magnification panel on the right. White arrowheads indicate LacZ+ cells that co-express Htr2a. Approximately 86% of CeALacZ neurons in the CeA express Htr2a (n = 3 mice). b-d, Representative images of the CeA from Htr2a-Cre; lacZ (b) and Htr2a-Cre; tdTomato (c, d) mice, labeled for SOM and β-Gal proteins (b), mCherry protein and TAC2 mRNA (c) and mCherry protein and CRH mRNA (d) (representative of 3 mice). White box indicates the location of the high-magnification panel on the right. White arrowheads indicate double positive cells. e, Percentage of Htr2a+ neurons in the CeA that co-express SOM, TAC2 and CRH (n = 3 sections per mouse, 3 mice per marker).Bar graphs show mean ± SEM. Scale bars: 50 μm.

Supplementary Figure 2 Validation of hM3Dq in CeAHtr2a neurons and supporting data from CeAHtr2a-neuron chemogenetic activation experiments.

a, Representative image of hM3Dq-mCherry expression in the CeA of an Htr2a-Cre mouse. b, Whole-cell recording in an acute slice from a CeAHtr2a;tdTomato neuron expressing hM3Dq-mCherry after bath application of 1 μM CNO (representative of 3 mice). c,d, IP administration of CNO (2 mg/kg) induced cfos in the CeA of CeAHtr2a::hM3Dq mice compared to controls (n = 3 sections per mouse, 3 mice, two-tailed unpaired t test: t(4) = 6.95, P = 0.0023). e-h, Free feeding assay. e, Raster plot of feeding bouts of example individual mice. f, CeAHtr2a::hM3Dq mice treated with CNO spent significantly more time feeding than CeAHtr2a::mCherry controls (n = 5 mice per group, two-tailed unpaired t test: t(8) = 2.45, P = 0.0399). There was no difference in the average feeding bout duration (g; n = 5 mice per group, two-tailed unpaired t test: t(8) = 2.23, P = 0.0563) and the number of bouts (h; n = 5 mice per group, two-tailed unpaired t test: t(8) = 0.90, P = 0.3944) between the groups. i-k, Food versus clay pellet assay. i, CNO-treated CeAHtr2a::hM3Dq mice spent more time eating food compared to a clay pellet (n = 6 mice per group, two-way ANOVA: Virus: F(1,20) = 16.28, P = 0.0006, Food: F(1,20) = 19.49, P = 0.0003, Interaction: F(1,20) = 5.47, P = 0.0299, Bonferroni post-hoc analysis: *** P < 0.001). The average duration of food consumption bouts was increased compared to clay bouts (j: n = 6 mice per group, Two-way ANOVA: Virus: F(1,20) = 7.46, P = 0.0129, Food: F(1,20) = 10.36, P = 0.0043, Interaction: F(1,20) = 3.37, P = 0.0812, Bonferroni post-hoc analysis: ** P < 0.05). There was no difference in the number of bouts (k; n = 6 mice per group, Two-way ANOVA: Virus: F(1,20) = 5.57, P = 0.0263, Food: F(1,20) = 3.62, P = 0.0716, Interaction: F(1,20) = 0.01, P = 0.9349). l-n, Bitter food assay. l, Chemogenetic activation of CeAHtr2a neurons overcame the suppression of feeding by bitter food (n = 8 mCherry mice (Normal), n = 9 hM3D mice (Normal), n = 10 mCherry mice (Bitter), n = 8 hM3D mice (Bitter), two-way ANOVA: Virus: F(1,31) = 4.30, P = 0.0465, Food: F(1,31) = 2.85, P = 0.1017, Interaction: F(1,31) = 2.54, P = 0.1215, Bonferroni post-hoc analysis: * P < 0.05). m, Scheme of two-bottle lickometer assay to determine sensitivity to bitter taste. n, CeAHtr2a::hM3Dq mice and mCherry expressing controls strongly prefer water compared to 1 mM Quinine solution (n = 5 mice, two-tailed unpaired t test: t(8) = 0.52, P = 0.6178). o, Progressive ratio-2 (PR2) task. Chemogenetic activation of CeAHtr2a neurons did not increase the number of active nose-pokes for food reward (n = 8 mCherry mice, n = 9 hM3D mice, two-tailed paired t test: mCherry; t(7) = 0.86, P = 0.4184, hM3D; t(8) = 0.91, P = 0.3917). p-s, Open field. p, Example locomotor traces of CeAHtr2a::hM3Dq and mCherry mice in the open field test (representative of 5 mice per group). Shaded area = center zone. q, CeAHtr2a::hM3Dq and mCherry control mice showed comparable anxiety-like behavior in the open field test (n = 5 mice per group, two-tailed unpaired t test: t(8) = 0.68, P = 0.518). r,s, Locomotor behavior [distance travelled (r; n = 5 mice per group, two-tailed unpaired t test: t(8) = 0.03, P = 0.9793) and velocity (s; n = 5 mice per group; two-tailed t test: t(8) = 0.01, P = 0.9904)] was not significantly different.Box–whisker plots display median, interquartile range and fifth to ninety-fifth percentiles of the distribution. * P < 0.05, ** P < 0.01, *** P < 0.001. Scale bars: 100 μm.

Supplementary Figure 3 In vivo validation of ChR2 in CeAHtr2a neurons and supporting data from CeAHtr2a-neuron optogenetic activation experiments.

a, Representative expression of ChR2-eYFP in the CeA of an Htr2a-Cre mouse. b,c, Intracranial photostimulation (20 Hz, 10 ms pulse length, 20 minutes) induced cfos in the CeA of CeAHtr2a::ChR2 mice compared to controls (n = 3 sections per mouse, 3 mice, two-tailed unpaired t test: t(2) = 7.05, P = 0.0196). d, Approximate optic fiber locations for photostimulation experiments (Fig. 1k-p, Fig. 4a-h, Supplementary Fig. 3, Supplementary Fig. 6a-d). e, Photostimulation of CeAHtr2a neurons at 10 Hz and 20 Hz increased food intake (n = 8 eYFP mice, n =9 ChR2 mice. For eYFP vs ChR2 ON 20 Hz, two-tailed unpaired t test: t(15) = 2.87, P = 0.0118. For ChR2 OFF vs ChR2 ON 20 Hz, two-tailed paired t test: t(15) = 3.45, P = 0.0087. For eYFP vs ChR2 ON 10 Hz, two-tailed unpaired t test: t(15) = 2.29, P = 0.0366. For ChR2 OFF vs ChR2 ON 10 Hz, two-tailed paired t test: t(8) = 2.10, P = 0.0688. For eYFP vs ChR2 ON 5 Hz, two-tailed unpaired t test: t(14) = 1.24, P = 0.2364. For ChR2 OFF vs ChR2 ON 5 Hz, two-tailed paired t test: t(7) = 1.27, P = 0.2443). f, Photostimulation of CeAHtr2a neurons at 20 Hz increased the total percentage of time spent feeding (n = 7 mice per group, 20 Hz, two-tailed unpaired t test: t(12) = 2.72, P = 0.0186. n = 5 eYFP mice, n= 5 ChR2 mice, 10 Hz, two-tailed unpaired t test: t(9) = 1.43, P = 0.1873. n = 5 eYFP mice, n = 6 ChR2 mice, 5 Hz, two-tailed unpaired t test: t(9) = 0.50, P = 0.6277). g, CeAHtr2a::ChR2 mice did not exhibit a change in latency to eat after the onset of photostimulation (n = 6 eYFP mice, n= 5 ChR2 mice, two-tailed unpaired t test: t(12) = 0.80, P = 0.4369). h, CeAHtr2a::ChR2 mice spent significantly more time engaging in appetitive motor sequences (n = 7 mice per group, 20 Hz, Mann Whitney test: P = 0.0209. n = 5 eYFP mice, n = 6 ChR2 mice, 10 Hz, two-tailed unpaired t test: t(9) = 2.83, P = 0.0196. n = 5 eYFP mice, n = 6 ChR2 mice, 5 Hz, two-tailed unpaired t test: t(9) = 0.10, P = 0.2999).Line graph values are mean ± SEM. Box–whisker plots display median, interquartile range and fifth to ninety-fifth percentiles of the distribution. * P < 0.05, ** P < 0.01. Scale bars: 100 μm.

Supplementary Figure 4 Supporting data from loss-of-function behavior experiments.

a, Representative images of the CeA of Htr2a-Cre;tdTomato mice expressing control eYFP or dtA virus and immunostained for PKCδ two months after virus injection (PKCδ channel only is shown). b, Food intake normalized to body weight (n = 4 mice per group, two-way repeated measures ANOVA: Virus: F(1,6) = 0.33, P = 0.5870, Time: F(23,138) = 4.05, P = <0.0001, Interaction: F(23,138) = 1.09, P = 0.3627) and c, body weight (n= 4 mice per group, two-way repeated measures ANOVA: Virus: F(1,6) = 0.01, P = 0.9318, Time: F(23,138) = 1.72, P = 0.0297, Interaction: F(23,138) = 0.46, P = 0.9829) of CeAHtr2a::dtA and CeAHtr2a::eYFP controls maintained on a chow diet. d, Example locomotor traces of CeAHtr2a::dtA and CeAHtr2a::eYFP control mice in the open field test (representative of 5 mice per group). Shaded area = center zone. e, CeAHtr2a::dtA and CeAHtr2a::eYFP mice displayed comparable anxiety-like behavior in the open field test (n = 5 mice per group, two-tailed unpaired t test: t(8) = 0.61, P = 0.5584). f,g, Locomotor behavior [distance travelled (f; n = 5 mice per group, two-tailed unpaired t test: t(8) = 0.96, P = 0.3646) and velocity (g; n = 5 mice per group, two-tailed unpaired t test: t(8) = 1.01, P = 0.3434)] was not significantly different. h, Representative expression of eNpHR-mCherry in the CeA of an Htr2a-Cre mouse. i, Approximate optic fiber locations for photoinhibition experiments (Fig. 2g-l, Fig. 4i-k, Supplementary Fig. 4, and Supplementary Fig. 6e-j). j. Photoinhibition of CeAHtr2a neurons did not significantly affect the latency to eat (n = 7 mice per group, two-tailed unpaired t test: t(12) = 1.05, P = 0.3137).Line graph values are mean ± SEM. Box–whisker plots display median, interquartile range and fifth to ninety-fifth percentiles of the distribution. Scale bars: 100 μm.

Supplementary Figure 5 Supporting data from in vivo Ca2+ imaging experiments from CeAHtr2a neurons.

a, Cre-dependent GCaMP6s expression in the CeA of an Htr2a-Cre;tdTomato mouse (representative of 3 mice). b, Activity of CeAHtr2a::GCaMP6s neurons increased during eating compared to baseline period (n = 69 cells across 4 mice, two-tailed unpaired t test: t(68) = 2.52, P = 0.0141). Connected points are responses of individual neurons. c, Histogram of preference indices of individual neurons for feeding during all food eating bouts. d, Example cells from two mice that reduced activity during eating. Grey bars= eating bouts. e, Mean Ca2+ responses of cells classified as Down cells (n = 9/69 cells) before and upon eating onset of all recorded feeding bouts. f, Mean Ca2+ activity of cells that reduced activity during feeding compared to baseline (n = 9 cells across 4 mice, two-tailed unpaired t test: t(8) = 3.91 P = 0.0045). Connected points are the average responses of all classified cells during individual bouts. g, Ca2+ responses of representative cells averaged across 12 FR1 trials (Upper) (total n = 36 cells from 3 mice). Lower= Ca2+ responses of the individual trials. h, Population Ca2+ responses of all recorded neurons (n = 36 cells from 3 mice) aligned to eating bout end. Upper = mean. Lower = heat map of normalized Ca2+ responses from all neurons. Each row is one trial.Bar graph values are mean ± SEM. * P < 0.05, ** P < 0.01. Scale bar: 50 μm.

Supplementary Figure 6 CeAHtr2a-neuron activity promotes place preference while silencing is not intrinsically aversive.

a, Representative locomotor traces of CeAHtr2a::ChR2 mice in a RTPP experiment (representative of 7 mice). b, CeAHtr2a::ChR2 mice spend significantly more time on the side of the RTPP chamber paired with photostimulation (n = 6 eYFP mice, n= 7 ChR2 mice, 10 Hz, two-tailed unpaired t test: t(11) = 2.50, P = 0.0293); 5 Hz, two-tailed unpaired t test: t(11) = 2.66, P = 0.0224). c,d, The distance travelled by CeAHtr2a::ChR2 mice (c; n = 5 mice per group, 20 Hz, two-tailed unpaired t test: t(8) = 1.72, P = 0.1243. 10 Hz, two-tailed unpaired t test: t(8) = 0.23, P = 0.8257. 5 Hz, two-tailed unpaired t test: t(8) = 1.42, P = 0.1933) and velocity (d; n = 5 mice per group, 20 Hz, two-tailed unpaired t test: t(8) = 0.77, P = 0.4660. 10 Hz, two-tailed unpaired t test: t(8) = 1.25, P = 0.2451. 5 Hz, two-tailed unpaired t test: t(8) = 0.51, P = 0.6218) in the stimulated side of the RTPP chamber was comparable to controls. e, Photoinhibition of CeAHtr2a neurons in food restricted mice did not affect licking of a spout that delivered a palatable reward (n = 7 mCherry mice, n = 8 NpHR mice, two-tailed unpaired t test: t(13) = 1.89, P = 0.8531). f, Scheme depicting RTPP paradigm where CeAHtr2a::NpHR mice were tested. g, Representative locomotor trace of a CeAHtr2a::NpHR mouse during the RTPP experiment (representative of 7 mice). h, CeAHtr2a::NpHR mice did not show any preference for one side of the RTPP chamber (n = 7 mice per group, two-tailed unpaired t test: t(12) = 0.05, P = 0.9618). i,j, The distance travelled by CeAHtr2a::NpHR mice (i; n = 7 mice per group, two-tailed unpaired t test: t(12) = 1.40, P = 0.1872) and velocity (j; n = 7 mice per group, two-tailed unpaired t test: t(12) = 0.78, P = 0.4504) in the stimulated side of the RTPP chamber was comparable to controls.Box–whisker plots display median, interquartile range and fifth to ninety-fifth percentiles of the distribution. * P < 0.05.

Supplementary Figure 7 Local circuit connectivity of CeAHtr2a and CeAPKCδ neurons.

a, Paired patch clamp recordings in acute slices of Htr2a-Cre;tdTomato mice of CeAHtr2a+ cells (blue) and/or CeAHtr2a‒ cells (grey). Action potentials evoked in current clamp (top trace) elicited responses recorded in voltage-clamp at holding potentials of -20 mV (black), -40 mV (grey) and -80 mV (light grey) (bottom trace). The graph below each connection shows mean amplitudes ± SEM of synaptic responses of all neurons with the same connections at holding potentials between -80mV and -20mV. b, Example picture of a paired recording between a CeAHtr2a-Cre;tdTomato+ neuron (left arrow) and a CeAHtr2a-Cre;tdTomato- neuron (right arrow). Dotted lines denote the recording electrodes. c, Mean amplitudes of synaptic responses in individual connections (grey dots) at holding potentials of -20mV. Connection rates are shown above the graph. d, Negative control for monosynaptic rabies tracing experiment omitting Cre transgenes shows only local non-specific infection of (EnvA)SAD∆G-EGFP rabies virus likely due to leaky expression of a small amount of the TVA receptor too low for mCherry to be detected but still capable of permitting infection of the pseudotyped rabies virus. e, Schematic representation of Cre-dependent rabies-based trans-synaptic tracing to reveal monosynaptic inputs to CeAHtr2a and CeAPKCδ neurons. f, Numbers of Htr2a and PKCδ starter neurons in CeL and CeM per mouse quantified in Fig.5h. Each dot is a separate experiment. g-h, Identification of local monosynaptic inputs to CeAHtr2a (g) and CeAPKCδ (h) using Cre-dependent, rabies virus–based monosynaptic tracing (images representative of 3 Htr2a-Cre and 4 Prkcd-Cre mice). White boxes indicate the location of the high-magnifications on the right. Arrows indicate starter cells (co-labeled with eGFP and mCherry). Arrowheads indicate input neurons (RABV-eGFP+ cells) immunolabeled for β-Gal (g) or PKCδ antibody (h). Asterisks denote RABV-eGFP-labeled input neurons only.Bar and line graph values are mean ± SEM. Scale bar: 50 μm.

Supplementary Figure 8 Long-range projections of CeAHtr2a neurons and supporting data from optogenetic activation of CeAHtr2a-neuron terminals in the PBN.

a, Targeting of CeAHtr2a neurons with an AAV-Flex-synaptophysin-myc. Immunostaining against myc highlights the regions where CeAHtr2a axons terminate. Synaptophysin-myc positive terminals are seen in the BNST, LH, ventrolateral periaqueductal gray (vlPAG), medial reticular nucleus (MRN), and NTS. Similar results were obtained in 3 animals. aco: anterior commissure; mt: mammilothalamic tract; fx: fornix; ic: internal capsule; mlf: medial longitudinal fascicle; sp5: spinal tract of the trigeminal nerve; icp: inferior cerebellar peduncle. b, Three-Dimensional visualization (dorsal and lateral) of CeAHtr2a axonal projections labeled by an AAV-Flex-mCherry virus injected into the central amygdala of an Htr2a-Cre mouse. High (red) and low (blue) mCherry fluorescence intensity is seen distinctly in the BNST, LH, vlPAG and PBN. c, Photostimulation of CeAHtr2a terminals by 473nm light pulses suppressed induced firing of PBN neurons by current injection in a PTX-dependent manner (n = 5 cells, ACSF+, one-way repeated measures ANOVA: Time-point: F(2,4) = 18.34, P = 0.001, Bonferroni post-hoc analysis: ** P < 0.01; PTX+, one-way repeated measures ANOVA: Time-point: F(2,4) = 0.29, P = 0.7538). d, Post-hoc characterization of neurobiotin-filled PBN neurons after whole-cell recording. Recovered neurons were identified as CGRP+ (filled arrowhead) (2/14 cells) or CGRP- (empty arrowhead). White box indicates the location of the higher-magnification on the right. e, Representative expression of CeAHtr2a::ChR2-eYFP presynaptic terminals in the PBN. f, Approximate optic fiber tip locations for photostimulation experiments (Fig 6. k-q and Supplementary Fig. 8). g, Raster plot of feeding bouts of ad libitum fed mice expressing the indicated proteins. h,i, Photostimulation of CeAHtr2a::ChR2 terminals in the PBN at 20 Hz did not significantly affect the number of feeding bouts (h; n = 7 mice per group; two-tailed unpaired t test: t(12) = 0.71, P = 0.4910) nor the latency to eat after onset of photostimulation (i; n= 7 mice per group, two-tailed unpaired t test: t(8) = 0.6965, P = 0.5058).Line graph values are mean ± SEM. Box–whisker plots display median, interquartile range and fifth to ninety-fifth percentiles of the distribution. ** P < 0.01. Scale bars: 1 mm in panel a and b, 100 μm in panel e, 50 μm in other panels.

Supplementary Figure 9 PBN-projecting CeA neurons receive input from CeAPKCδ neurons.

a, TRIO method to reveal monosynaptic inputs to CeA neurons projecting to PBN making use of retrogradely transported CAV-Cre and HSV-Cre viruses. b, Numbers of PBN-projecting starter neurons in CeL and CeM per mouse quantified in panel d. Each dot is a separate experiment. c, Identification of local monosynaptic inputs to PBN-projecting CeA neurons using TRIO method (representative of 4 mice). White boxes indicate the location of the high-magnifications on the right. Arrowheads indicate input neurons (RABV-eGFP+ cells) immunolabeled for PKCδ antibody. Asterisks denote RABV-eGFP-labeled input neurons only. d, Quantification of the relative abundance of PKCδ+ input cells to PBN-projecting CeA neurons (n = 4 mice). (Number of RABV-EGFP+ counted input cells for PKCδ IHC,67-200). e, Scheme of approach to probe CeAPKCδ neuron-mediated inhibition of PBN-projecting CeAPKCδ- neurons. f, Photoactivation of CeAPKCδ neurons strongly inhibited retrobead-positive (left panels) CeAPKCδ- neurons that project to the PBN.Bar graph values are mean ± SEM. Scale bars: 50 μm.

Supplementary Figure 10 Long-range monosynaptic inputs to CeAHtr2a and CeAPKCδ neurons.

a, Numbers of PKCδ, Htr2a, PBN-projecting and PBN-projecting Htr2a starter neurons in CeL and CeM per mouse quantified in Fig.7a-b and panels c-e below. Each dot is a separate experiment. b, Relationship between numbers of starter cells in CeA and number of long range input neurons. Each dot is a separate experiment. c, Proportion of monosynaptic inputs to CeAHtr2a (n = 6 mice), PBN-projecting CeAHtr2a (n = 4 mice), PBN-projecting CeA (n = 5 mice) and CeAPKCδ (n =5 mice) neurons, normalized against the total number of inputs cells in each animal. IC, insula cortex; Isocx, isocortex; Pir, piriform cortex; LA, lateral amygdala; BLA, basolateral amygdala, anterior part; BLP, basolateral amygdala, posterior part; Hipp, hippocampal formation; Enth, enthorinal cortex; BNST, bed nucleus of the stria terminalis; VP, ventral pallidum; aPVT, anterior paraventricular thalamus; pPVT, posterior paraventricular thalamus; Multimod. T., multimodal thalamus; VPMpc, ventral posteromedial nucleus of the thalamus, parvicellular part; PO H., preoptic hypothalamus; PVH, paraventricular hypothalamus; Tub. p. H., tuberal posterior hypothalamus; Post. H., posterior hypothalamus; PSTN, parasubthalamic nucleus; SNL, substantia nigra, lateral part; MRN, midbrain reticular nucleus; CUN, cuneiform nucleus; RR, midbrain reticular nucleus, retrorubral area; DR, dorsal raphe; lPBN, lateral parabrachial nucleus. d, Coefficient of variation for each region giving more than 1% of the total inputs to CeAHtr2a and, or CeAPKCδ neurons. e, Example graphs showing a strong positive correlation between inputs from the IC and VPMpc, PSTN and Tub. p. H. and no correlation between inputs from the Tub.p.H. and IC. Each dot is a separate experiment. Values shown are Pearson correlation coefficients with p values from a two-tailed unpaired t test: VPMPc vs IC: r(4) = 0.98, P = 0.0006; Tub. p. H. vs PSTN: r(4) = 0.83, P = 0.0418; IC vs Tub. p. H.: r(4) = 0.30, P = 0.5678.Bar graph values are mean ± SEM. Box–whisker plots display median, interquartile range and fifth to ninety-fifth percentiles of the distribution. * P < 0.05, *** P <0.001.

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–10. (PDF 2286 kb)

Life Sciences Reporting Summary (PDF 171 kb)

Photostimulation of CeAHtr2a::ChR2 neurons at 20Hz evokes feeding-related motor behaviors.

These include, licking of the walls and floor and chewing and gnawing motor sequences. (MP4 29294 kb)

Representative move of Ca2+ activity (post-processing movie) (1x speed) in CeAHtr2a::GCaMP6s neurons synchronized with the behavior video during the free-feeding experiment.

Upon the onset of eating after an overnight fast, increased Ca2+ activity is observed in CeAHtr2a:: GCaMP6s neurons. (MOV 670 kb)

Representative move of Ca2+ activity (post-processing movie) (1x speed) in CeAHtr2a::GCaMP6s neurons synchronized with the behavior video during the FR1 experiment.

Key behavior events are denoted by text. Ca2+ activity is observed in CeAHtr2a:: GCaMP6s neurons during eating but not during the nose-poke that triggered release of the food pellet. (MP4 3111 kb)

CeAHtr2a::ChR2 mice with bilateral optic fibres over the CeA strongly nose-poke for intracranial delivery of a 20Hz 473nm light pulse train. (MP4 27195 kb)

Three-dimensional visualization (dorsal and lateral) of CeAHtr2a axonal projections labeled by an AAV-Flex-mCherry virus injected into the central amygdala of an Htr2a-Cre brain.

High (red) and low (blue) mCherry fluorescence intensity is seen distinctly in the BNST, LH, vlPAG and PBN. (MP4 15696 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Douglass, A., Kucukdereli, H., Ponserre, M. et al. Central amygdala circuits modulate food consumption through a positive-valence mechanism. Nat Neurosci 20, 1384–1394 (2017). https://doi.org/10.1038/nn.4623

Download citation

Further reading

Search

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing