Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Nonlinear recurrent inhibition through facilitating serotonin release in the raphe

Nature Neuroscience volume 28, pages 1024–1037 (2025)Cite this article

Subjects

Abstract

Serotonin (5-HT) neurons in the dorsal raphe nucleus (DRN) receive a constellation of long-range inputs, yet guiding principles of local circuit organization and underlying computations in this nucleus are largely unknown. Using inputs from the lateral habenula to interrogate the processing features of the mouse DRN, we uncovered 5-HT1A receptor-mediated recurrent connections between 5-HT neurons, refuting classical theories of autoinhibition. Cellular electrophysiology and imaging of a genetically encoded 5-HT sensor revealed that these recurrent inhibitory connections spanned the raphe, were slow, stochastic, strongly facilitating and gated spike output. These features collectively conveyed highly nonlinear dynamics to this network, generating excitation-driven inhibition and winner-take-all computations. In vivo optogenetic activation of lateral habenula inputs to DRN, at frequencies where these computations are predicted to ignite, transiently disrupted expression of a reward-conditioned response in an auditory conditioning task. Together, these data identify a core computation supported by an unsuspected slow serotonergic recurrent inhibitory network.

Access through your institution
Buy or subscribe

This is a preview of subscription content, access via your institution

Access options

Access through your institution

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Habenula input to raphe triggers direct excitation and heterosynaptic inhibition in 5-HT neurons.
Fig. 2: 5-HT neurons in raphe are organized in a recurrent inhibitory network.
Fig. 3: Serotonin release is probabilistic and strongly facilitating at inhibitory recurrent connections.
Fig. 4: A quantitative model of short-term release dynamics captures patterned inputs.
Fig. 5: Synaptic facilitation dynamics and spatial structure of release revealed by GRAB5-HT serotonin sensor imaging in DRN.
Fig. 6: Habenula inputs to raphe generate a nonmonotonic network response through recurrent STP.
Fig. 7: Winner-take-all dynamics generated by recurrent STP.
Fig. 8: Habenula inputs to raphe disrupt the expression of learned reward associations in a value- and frequency-dependent manner.

Similar content being viewed by others

Data availability

Source data are provided with this paper. Further source data supporting this Article are available from the authors upon reasonable request.

Code availability

The code required to run and generate the computational simulation figures (Figs. 6 and 7 and Extended Data Figs. 5–7) is available at https://github.com/micllynn/Recurrent5HT_2025/.

References

  1. Dorocic, I. P. et al. A whole-brain atlas of inputs to serotonergic neurons of the dorsal and median raphe nuclei. Neuron 83, 663–678 (2014).

    Google Scholar 

  2. Weissbourd, B. et al. Presynaptic partners of dorsal raphe serotonergic and GABAergic neurons. Neuron 83, 645–662 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  3. Ren, J. et al. Anatomically defined and functionally distinct dorsal raphe 5-HT sub-systems. Cell 175, 472–487 (2018).

    CAS  PubMed  PubMed Central  Google Scholar 

  4. Sharp, T. & Barnes, N. M. Central 5-HT receptors and their function; present and future. Neuropharmacology 177, 108155 (2020).

    CAS  PubMed  Google Scholar 

  5. Cohen, J. Y., Amoroso, M. W. & Uchida, N. Serotonergic neurons signal reward and punishment on multiple timescales. eLife 2015, e06346 (2015).

    Google Scholar 

  6. Li, Y. et al. 5-HT neurons in the dorsal raphe nucleus encode reward signals. Nat. Commun. 7, 10503 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  7. Zhong, W., Li, Y., Feng, Q. & Luo, M. Learning and stress shape the reward response patterns of 5-HT neurons. J. Neurosci. 37, 8863–8875 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

  8. Paquelet, G. E. et al. Single-cell activity and network properties of dorsal raphe nucleus serotonin neurons during emotionally salient behaviors. Neuron 110, 2664–2679 (2022).

    CAS  PubMed  PubMed Central  Google Scholar 

  9. Grossman, C. D., Bari, B. A. & Cohen, J. Y. Serotonin neurons modulate learning rate through uncertainty. Curr. Biol. 32, 1–14 (2022).

    Google Scholar 

  10. Takahashi, A. et al. Glutamate input in the dorsal raphe nucleus as a determinant of escalated aggression in male mice. J. Neurosci. 35, 6452–6436 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  11. Miyazaki, K. et al. Reward probability and timing uncertainty alter the effect of dorsal raphe 5-HT neurons on patience. Nat. Commun. 9, 2048 (2018).

    PubMed  PubMed Central  Google Scholar 

  12. Warden, M. R. et al. A prefrontal cortex-brainstem neuronal projection that controls response to behavioral challenge. Nature 492, 428–432 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  13. Geddes, S. D. et al. Time-dependent modulation of glutamate synapses onto 5-HT neurons by antidepressant treatment. Neuropharmacology 95, 130–143 (2015).

    CAS  PubMed  Google Scholar 

  14. Zhou, L., Liu, M.-Z., Deng, J., Mu, D. & Sun, Y.-G. Organization of functional long-range circuits controlling the activity of serotonergic neurons in the dorsal raphe nucleus. Cell Rep. 18, 3018–3032 (2017).

    CAS  PubMed  Google Scholar 

  15. Matsumoto, M. & Hikosaka, O. Representation of negative motivational value in the primate lateral habenula. Nat. Neurosci. 12, 77–84 (2009).

    CAS  PubMed  Google Scholar 

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

    CAS  Google Scholar 

  17. Bromberg-Martin, E. S. & Hikosaka, O. Lateral habenula neurons signal errors in the prediction of reward information. Nat. Neurosci. 14, 1209–1216 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  18. Amo, R. et al. The habenulo-raphe serotonergic circuit encodes an aversive expectation value essential for adaptive active avoidance of danger. Neuron 84, 1034–1048 (2014).

    CAS  PubMed  Google Scholar 

  19. Yang, Y. et al. Ketamine blocks bursting in the lateral habenula to rapidly relieve depression. Nature 554, 317–322 (2018).

    CAS  PubMed  Google Scholar 

  20. Groos, D. & Helmchen, F. The lateral habenula: a hub for value-guided behavior. Cell Rep. 43, 113968 (2024).

    CAS  PubMed  Google Scholar 

  21. Marquez, J. C., Li, M., Schaak, D., Robson, D. N. & Li, J. M. Internal state dynamics shape brainwide activity and foraging behavior. Nature 577, 239–243 (2020).

    Google Scholar 

  22. Geddes, S. D. et al. Target-specific modulation of the descending prefrontal cortex inputs to the dorsal raphe nucleus by cannabinoids. Proc. Natl Acad. Sci. USA 113, 5429–5434 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  23. Andalman, A. S. et al. Neuronal dynamics regulating brain and behavioral state transitions. Cell 177, 970–985 (2019).

    CAS  PubMed  PubMed Central  Google Scholar 

  24. Newberry, N. R. & Nicoll, R. A. A. A bicuculline-resistant inhibitory post-synaptic potential in rat hippocampal pyramidal cells in vitro. J. Physiol. 348, 239–254 (1984).

    CAS  PubMed  PubMed Central  Google Scholar 

  25. Pazos, A. & Palacios, J. M. Quantitative autoradiographic mapping of serotonin receptors in the rat brain. I. Serotonin-1 receptors. Brain Res. 346, 205–230 (1985).

    CAS  PubMed  Google Scholar 

  26. Bunin, M. A. & Wightman, R. M. Quantitative evaluation of 5-hydroxytryptamine (5-HT) neuronal release and reuptake: an investigation of extrasynaptic transmission. J. Neurosci. 18, 4854–4860 (1998).

    CAS  PubMed  PubMed Central  Google Scholar 

  27. Pineyro, G. & Blier, P. Autoregulation of 5-HT neurons: role in antidepressant drug action. Pharmacol. Rev. 51, 533–591 (1999).

    CAS  PubMed  Google Scholar 

  28. Andrade, R., Huereca, D., Lyons, J. G., Andrade, E. M. & McGregor, K. M. 5‑HT1A receptor-mediated autoinhibition and the control of serotonergic cell firing. ACS Chem. Neurosci. 6, 1110–1115 (2015).

    CAS  PubMed  Google Scholar 

  29. Trinh, A.-T., Harvey-Girard, E., Teixeira, F. & Maler, L. Cryptic laminar and columnar organization in the dorsolateral pallium of a weakly electric fish. J. Comp. Neurol. 524, 408–428 (2016).

    CAS  PubMed  Google Scholar 

  30. Zhang, Y.-P. & Oertner, T. G. Optical induction of synaptic plasticity using a light-sensitive channel. Nat. Methods 4, 139–141 (2007).

    CAS  PubMed  Google Scholar 

  31. Lynn, M. B., Lee, K. F. H., Soares, C., Naud, R. & Béïque, J.-C. A synthetic likelihood solution to the silent synapse estimation problem. Cell Rep. 32, 107916 (2020).

    CAS  PubMed  Google Scholar 

  32. Courtney, N. A. & Ford, C. P. Mechanisms of 5-HT1A receptor-mediated transmission in dorsal raphe 5-HT neurons. J. Physiol. 594, 953–965 (2016).

    CAS  PubMed  Google Scholar 

  33. Rossbroich, J., Trotter, D., Beninger, J., Toth, K. & Naud, R. Linear-nonlinear cascades capture synaptic dynamics. PLOS Comput. Biol. 17, e1008013 (2021).

    CAS  PubMed  PubMed Central  Google Scholar 

  34. Deng, F. et al. Improved green and red GRAB sensors for monitoring spatiotemporal serotonin release in vivo. Nat. Methods 21, 692–702 (2024).

    CAS  PubMed  PubMed Central  Google Scholar 

  35. Harkin, E. F. et al. Temporal derivative computation in the dorsal raphe network revealed by an experimentally driven augmented integrate-and-fire modeling framework. eLife 12, e72951 (2023).

    CAS  PubMed  PubMed Central  Google Scholar 

  36. Huang, K. W. et al. Molecular and anatomical organization of the dorsal raphe nucleus. eLife 8, e46464 (2019).

    PubMed  PubMed Central  Google Scholar 

  37. Matias, S., Lottem, E., Dugué, G. P. & Mainen, Z. F. Activity patterns of serotonin neurons underlying cognitive flexibility. eLife 6, e20552 (2017).

    PubMed  PubMed Central  Google Scholar 

  38. Cohen, J. Y., Haesler, S., Vong, L., Lowell, B. B. & Uchida, N. Neuron-type-specific signals for reward and punishment in the ventral tegmental area. Nature 482, 85–88 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  39. Fonseca, M. S., Murakami, M. & Mainen, Z. F. Activation of dorsal raphe serotonergic neurons promotes waiting but is not reinforcing. Curr. Biol. 25, 306–315 (2015).

    CAS  PubMed  Google Scholar 

  40. Gozlan, H., El Mestikawy, S., Pichat, L., Glowinski, J. & Hamon, M. Identification of presynaptic 5-HT autoreceptors using a new ligand: 3H-PAT. Nature 305, 140–142 (1983).

    CAS  PubMed  Google Scholar 

  41. Descarries, L., Watkins, K. C., Garcia, S. & Beaudet, A. The serotonin neurons in nucleus raphe dorsalis of adult rat: a light and electron microscope radioautographic study. J. Comp. Neurol. 207, 239–254 (1982).

    CAS  PubMed  Google Scholar 

  42. Chazal, G. & Ralston, H. J. Serotonin-containing structures in the nucleus raphe dorsalis of the cat: an ultrastructural analysis of dendrites, presynaptic dendrites, and axon terminals. J. Comp. Neurol. 259, 317–329 (1987).

    CAS  PubMed  Google Scholar 

  43. Colgan, L. A., Cavolo, S. L., Commons, K. G. & Levitan, E. S. Action potential-independent and pharmacologically unique vesicular serotonin release from dendrites. J. Neurosci. 32, 15737–15746 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  44. Liu, C., Kershberg, L., Wang, J., Schneeberger, S. & Kaeser, P. Dopamine secretion is mediated by sparse active zone-like release sites. Cell 172, 706–718 (2018).

    CAS  PubMed  PubMed Central  Google Scholar 

  45. Liu, C., Goel, P. & Kaeser, P. S. Spatial and temporal scales of dopamine transmission. Nat. Neurosci. 22, 345–358 (2021).

    Google Scholar 

  46. Coultrip, R., Granger, R. & Lynch, G. A cortical model of winner-take-all competition via lateral inhibition. Neural Netw. 5, 47–54 (1992).

    Google Scholar 

  47. Douglas, R. J., Koch, C., Mahowald, M., Martin, K. A. C. & Suarez, H. H. Recurrent excitation in neocortical circuits. Science 269, 981–985 (1995).

    CAS  PubMed  Google Scholar 

  48. Echeveste, R., Aitchison, L., Hennequin, G. & Lengyel, M. Cortical-like dynamics in recurrent circuits optimized for sampling-based probabilistic inference. Nat. Neurosci. 23, 1138–1149 (2020).

    CAS  PubMed  PubMed Central  Google Scholar 

  49. Ayaz, A. & Chance, F. S. Gain modulation of neuronal responses by subtractive and divisive mechanisms of inhibition. J. Physiol. 101, 958–968 (2009).

    Google Scholar 

  50. Mejias, J. F., Payeur, A., Selin, E., Maler, L. & Longtin, A. Subtractive, divisive and non-monotonic gain control in feedforward nets linearized by noise and delays. Front. Comp. Neurosci. 8, 19 (2014).

    Google Scholar 

  51. Lewis, J. E., Lindner, B., Laliberté, B. & Groothius, S. Control of neuronal firing by dynamic parallel fiber feedback: implications for electrosensory reafference suppression. J. Exp. Biol. 210, 4437–4447 (2007).

    PubMed  Google Scholar 

  52. Liu, Z., Lin, R. & Luo, M. Reward contributions to serotonergic function. Annu. Rev. Neurosci. 43, 141–162 (2020).

    PubMed  Google Scholar 

  53. Luo, M., Li, Y. & Zhong, W. Do dorsal raphe 5-HT neurons encode ‘beneficialness’? Neurobiol. Learn. Mem. 135, 40–49 (2016).

    CAS  PubMed  Google Scholar 

  54. Harkin, E. F., Grossman, C. D., Cohen, J. Y., Béïque, J.-C. & Naud, R. Serotonin predictively encodes value. Preprint at bioRxiv https://doi.org/10.1101/2023.09.19.558526 (2023).

  55. Eichenbaum, K. D. et al. Minimally invasive method for murine brain fixation. Biotechniques 39, 487–488 (2005).

    CAS  PubMed  Google Scholar 

  56. Jones, E. et al. SciPy: open source scientific tools for Python v.1.8 (2001).

  57. van der Walt, S., Colbert, S. C. & Varoquaux, G. The NumPy array: a structure for efficient numerical computation. Comput. Sci. Eng. 13, 22–30 (2011).

    Google Scholar 

  58. Pachitariu, M. et al. Suite2p: beyond 10,000 neurons with standard two-photon microscopy. Preprint at bioRxiv https://doi.org/10.1101/061507 (2017).

  59. Sigworth, F. J. The variance of sodium current fluctuations at the node of Ranvier. J. Physiol. 307, 97–129 (1980).

    CAS  PubMed  PubMed Central  Google Scholar 

  60. Stimberg, M., Brette, R. & Goodman, D. F. M. Brian 2, an intuitive and efficient neural simulator. eLife 2019, e47314 (2019).

    Google Scholar 

  61. Pozzorini, C. et al. Automated high-throughput characterization of single neurons by means of simplified spiking neurons. PLoS Comput. Biol. 11, e1004275 (2015).

    PubMed  PubMed Central  Google Scholar 

  62. de Hoz, L. & Nelken, I. Frequency tuning in the behaving mouse: different bandwidths for discrimination and generalization. PLoS ONE 9, e91676 (2014).

    PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank members of the Béïque, Naud and Maler laboratories for helpful discussions. We also thank S. El Mestikawy and S. Williams for providing us with the SERT-Cre mouse line and S. Chen for providing us with the SOM-Cre mice. This work was supported by grants from the Canadian Institute for Health Research (PJT- 175319, PJT-175325; J.-C.B. and R.N.), the uOttawa Brain and Mind Institute and the Canadian Heart and Stroke Foundation (G-15-9399; J.-C.B.), Brain Canada (J.-C.B.), the Krembil Foundation (J.-C.B.) and the Canadian Foundation for Innovation (33454; J.-C.B.). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.

Author information

Author notes

  1. Michael B. Lynn

    Present address: Department of Physiology, Anatomy and Genetics, University of Oxford, Oxford, UK

  2. Emerson Harkin

    Present address: Max Planck Institute for Biological Cybernetics, Tübingen, Germany

Authors and Affiliations

  1. Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario, Canada

    Michael B. Lynn, Sean D. Geddes, Mohamad Chahrour, Sébastien Maillé, Léa Caya-Bissonnette, Emerson Harkin, Érik Harvey-Girard, Richard Naud & Jean-Claude Béïque

  2. Department of Pharmacology and Toxicology, University at Buffalo, Buffalo, NY, USA

    Samir Haj-Dahmane

  3. Centre for Neural Dynamics and AI, University of Ottawa, Ottawa, Ontario, Canada

    Richard Naud & Jean-Claude Béïque

  4. Brain and Mind Research Institute, University of Ottawa, Ottawa, Ontario, Canada

    Richard Naud & Jean-Claude Béïque

  5. Department of Physics, STEM Complex, University of Ottawa, Ottawa, Ontario, Canada

    Richard Naud

Authors
  1. Michael B. Lynn
    View author publications

    Search author on:PubMed Google Scholar

  2. Sean D. Geddes
    View author publications

    Search author on:PubMed Google Scholar

  3. Mohamad Chahrour
    View author publications

    Search author on:PubMed Google Scholar

  4. Sébastien Maillé
    View author publications

    Search author on:PubMed Google Scholar

  5. Léa Caya-Bissonnette
    View author publications

    Search author on:PubMed Google Scholar

  6. Emerson Harkin
    View author publications

    Search author on:PubMed Google Scholar

  7. Érik Harvey-Girard
    View author publications

    Search author on:PubMed Google Scholar

  8. Samir Haj-Dahmane
    View author publications

    Search author on:PubMed Google Scholar

  9. Richard Naud
    View author publications

    Search author on:PubMed Google Scholar

  10. Jean-Claude Béïque
    View author publications

    Search author on:PubMed Google Scholar

Contributions

Conceptualization: M.B.L., S.D.G., S.H.-D., R.N. and J.-C.B. Methodology: M.B.L. and S.D.G. Software: M.B.L. Formal analysis: M.B.L. and S.D.G. Investigation: M.B.L., S.D.G., M.C., E.H., L.C.-B., S.M. and E.H.-G. Resources: M.B.L., E.H. and R.N. Writing—original draft: M.B.L. Writing—review and editing: M.B.L., R.N. and J.-C.B. Supervision: J.-C.B. Funding acquisition: J.-C.B. and R.N.

Corresponding author

Correspondence to Jean-Claude Béïque.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Neuroscience thanks Minmin Luo and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Additional information

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

Extended data

Extended Data Fig. 1 Lateral habenula input fails to recruit strong disynaptic inhibition onto 5-HT neurons located in the ventromedial portion of DRN.

a, Experimental setup for panels b-g. Schematic showing viral expression strategy (left), confocal images showing CHETA-YFP and tdTomato expression in DRN (middle), and schematic of whole-cell patch clamp recording setup from SOM neurons, related to panels B and C (right). b, Example trace from CHETA-YFP-expressing 5-HT neuron recorded in voltage clamp, demonstrating photostimulation-evoked inward current. c, Example trace from CHETA-YFP-expressing 5-HT neuron recorded in current clamp, demonstrating photostimulation-evoked spiketrains at 20 Hz. d, Schematic of whole-cell patch-clamp recording setup from 5-HT neurons, related to panels E-G. e, Trial-averaged traces from an example 5-HT neuron, recorded in voltage clamp, illustrating photostimulation-driven currents at two holding potentials. Inset depicts zoomed region. f, Latency to peak outward current response at Vm = −40 mV across all photostimulation events (n = 120 events) for one example neuron. g, I-V curve for photostimulation-driven currents from one example neuron. h, Experimental setup for panels i-o. i, Trial-averaged trace from an example SOM neuron, recorded in voltage clamp, illustrating photostimulation-driven inward currents. Inset depicts zoomed region. j, Latency to peak inward current response across all photostimulation events (n = 615 events, 8 neurons). k, Amplitude of peak current response across all photostimulation events (n = 615 events, 8 neurons). l-m, Current-clamp traces from two example SOM neurons showing subthreshold responses to photostimulation (l) and suprathreshold responses to photostimulation (m). n, Classification of SOM responses to photostimulation across the population of recorded neurons. o, Peak depolarization from baseline for SOM neurons exhibiting subthreshold responses (n = 7 biological replicates). Line and shaded area express mean +- S.D.

Source data

Extended Data Fig. 2 Spatial scale and anatomical features of recurrent connectivity.

a, Experimental schematic demonstrating fluorescent dextran labeling approach. b, Representative confocal images of DRN showing dextran injection site and labeled neurons (left), tdTomato neurons (middle) and overlaid image showing colocalization (right, arrows). c, Representative confocal image (left) depicting dextran injection in dorsal DRN (cyan), tdTomato-positive 5-HT neurons (red), and neurons exhibiting colocalization (arrows). Extracted coordinates are shown for colabeled 5-HT neurons (middle) and all 5-HT neurons (right). d, Locations of all colabeled 5-HT neurons from injection center for dorsal injection (left; n = 17 neurons, 3 injections) and ventral injection (right; n = 21 neurons, 3 injections). Inset depicts anatomical locations of each injection site within DRN. e, Probability density of the distance from colabeled 5-HT neurons to injection center, for dorsal (purple) and ventral (orange) injections (two-sided Kolmogorov-Smirnov test, test statistic=0.588). f, Probability density of colabeled neuron distance, expressed as a difference in probability from the distribution of all 5-HT neurons (unstructured, dotted line). g, Polar plot depicting anatomical location of colabeled 5-HT neurons from injection site, for dorsal (purple) and ventral (orange) injections. h, Polar plot of colabeled neuron location as in g, but expressed as a difference in probability from the distribution of all 5-HT neurons (unstructured, dotted line).

Source data

Extended Data Fig. 3 Frequency-dependent facilitation of serotonin release under low extracellular calcium conditions. Related to Fig. 3.

a,: Experimental setup for panels b-c depicting whole-cell patch clamp recording from genetically identified 5-HT neuron with electrically evoked 5-HT release, under physiological recording conditions (extracellular calcium and temperature). Inset depicts electrical stimulation parameters for frequency-varying experiments. b, Representative averaged traces from a single neuron depicting responses to electrical stimulation at various frequencies. Red dotted lines denote electrical stimulation, and the shaded area denotes standard deviation across trials. c, Charge z-score as a function of electrical stimulation frequency, with 95% noise C.I shown in green (n = 4 neurons). d, Schematic showing electrical stimulation parameters for paired-pulse experiments (e-g). e, Representative averaged traces from a single neuron depicting responses to the paired-pulse electrical stimulation protocol. f, Charge histograms for single-pulse responses and paired-pulse responses at 50 ms for one example neuron. g, Post-stimulus charge as a function of paired-pulse time interval (n = 4 neurons). Green line depicts expected charge from linear summation. Blue line depicts population average and grey lines are responses of individual neurons. h, Experimental setup for panels i-k depicting channelrhodopsin injection in LHb (left), and whole-cell patch clamp recording from genetically identified 5-HT neuron with electrically evoked 5-HT release, under physiological recording conditions (right). i, Representative averaged traces from one example neuron depicting current responses to optogenetic stimulation of LHb afferents at two frequencies. Note that EPSCs are truncated for clarity to focus on direct comparison of the outward GIRK conductance. j, Representative averaged traces from another example neuron depicting current responses to optogenetic stimulation of LHb afferents at four frequencies. k, Total charge evoked by LHb stimulation as a function of optogenetic stimulation frequency. Each line depicts one neuron. Physiological traces are presented as mean values +/− S.D (shaded area).

Source data

Extended Data Fig. 4 Generating a mask for statistically significant autocorrelation regions with a pixel shuffling approach. Related to Fig. 5.

a, Example region showing whole-frame tdTomato fluorescence (left) and binned GRAB responses to 5, 10 and 20 Hz electrical stimulation (8 pulses; right). b, Spatial autocorrelation plots for the frames shown in A. Autocorrelation values are normalized from 0 to 1 for presentation. c, Same as A, but the spatial location of pixels are shuffled. d, Spatial autocorrelation plots for the pixel-shuffled frames shown in c. Autocorrelation values are normalized from 0 to 1 for presentation. e, Probability distribution function of raw autocorrelation values for all pixels, for the spatial autocorrelations performed on pixel-shuffled GRAB responses (grey; computed from D) and for the spatial autocorrelations performed on GRAB responses (colors correspond to labeled electrical stimulation frequencies; computed from b). Pixels to the right of the green threshold have autocorrelation values that statistically exceed those of the shuffled distribution (threshold of p = 0.0001; derived from two-sided empirical comparison of null and experimental distributions; that is 0.01% of null distribution values are past this threshold). f, Spatial mask for statistically significant autocorrelation values in b, where only pixels that exceed the significance threshold compared to the pixel-shuffled noise distribution (p = 0.0001, as computed in e) are shown.

Extended Data Fig. 5 The input-output function of the habenulo-raphe pathway is largely invariant to the fraction of 5-HT neurons innervated by LHb, and this invariance depends on recurrent short-term plasticity. Related to network simulations depicted in Fig. 6.

Panels a-e depict results of simulations where short-term plasticity at 5-HT-5-HT connections is intact. a, Input-output relationship of 5-HT neurons during the transient (0–200 ms) phase of LHb input. Colors denote different fractions of 5-HT neurons receiving LHb input (shown in inset; 10-50%). b, Same as a, but during stationary phase of LHb input (200–400 ms). c, Same as a, but during the post-response period of LHb input (0-1 s after). d, Cumulative distribution function of GIRK conductance across simulated 5-HT neurons, for 5 Hz LHb input (red) and 20 Hz LHb input (blue), and for fractions of 5-HT neurons innervated by LHb at 10% (lighter shade) and 50% (darker shade) (all combinations depicted in inset). Note how GIRK conductance depends primarily on LHb input frequency, instead of on LHb input fraction. e, Relationship between LHb input frequency and GIRK conductance (population mean) as LHb input fraction is varied (colors denote distinct input fractions, shown in inset), demonstrating that population GIRK conductance is largely invariant to LHb input fraction. Panels D-H depict results of simulations where short-term plasticity at 5-HT-5-HT connections has been removed (that is linearized responses). f, Input-output relationship of 5-HT neurons during the transient (0–200 ms) phase of LHb input. Colors denote different fractions of 5-HT neurons receiving LHb input (shown in inset; 10-50%). g, Same as f, but during stationary phase of LHb input (200–400 ms). h, Same as f, but during the post-response period of LHb input (0-1 s after). i, Cumulative distribution function of GIRK conductance across simulated 5-HT neurons, for 5 Hz LHb input (red) and 20 Hz LHb input (blue), and for fractions of 5-HT neurons innervated by LHb at 10% (lighter shade) and 50% (darker shade) (all combinations depicted in inset). Note how without short-term plasticity, GIRK conductance depends on both LHb input frequency and LHb input fraction. j, Relationship between LHb input frequency and GIRK conductance (population mean) as LHb input fraction is varied (colors denote distinct input fractions, shown in inset), demonstrating that without short-term plasticity dynamics, the population GIRK conductance is not invariant to LHb input fraction.

Extended Data Fig. 6 Winner-take-all dynamics in the DRN persist over a wide range of spatial connectivity profiles for recurrent inhibition. Related to network simulations shown in Fig. 7.

a, Schematic of network simulation, with LHb and Area A providing input to 5-HT neurons (top). Plot demonstrating how 5-HT-5-HT recurrent connectivity probability varies with distance, as spatial connectivity profiles are parameterized (bottom). b, Simulation with more global recurrent spatial connectivity of c = 1/2. Shown are mean firing rates of A and LHb inputs (top); mean firing rates of A-recipient 5-HT neurons (middle); and raster plot depicting firing rates of 2500 5-HT neurons receiving either LHb input (purple) or A input (orange) (depicted is the 50% of 5-HT neurons receiving direct input from either LHb or A). c, Simulation with more local recurrent spatial connectivity of c = 1/4. Layout of panels is identical to in b. d-e quantify the strength of winner-take-all dynamics as a function of recurrent spatial connectivity profiles, while short-term plasticity at recurrent connections is intact. d, LHb-evoked change in A-recipient 5-HT neuron responses, as a function of the rate of A input, for 5 Hz (left) and 20 Hz (right) LHb input. Spatial connectivity profile is varied (shown in legend), demonstrating that winner-take-all dynamics are stable across a wide range of spatial recurrent connectivity profiles. e, Strength of winner-take-all effect (quantified as LHb-evoked change in A-recipient 5-HT neuron responses) as A input rate and LHb input rate are varied. Shown are recurrent spatial connectivity values of c = 1 (left) and c = 1/4 (right). f-g quantify the strength of winner-take-all dynamics as a function of recurrent spatial connectivity profiles, while short-term plasticity at recurrent connections has been removed (that is linearized responses). f, LHb-evoked change in A-recipient 5-HT neuron responses, as a function of the rate of A input, for 5 Hz (left) and 20 Hz (right) LHb input. Spatial connectivity profile is varied as depicted in the legend, demonstrating that with short-term plasticity removed, the winner-take-all dynamics vary over the tested range of spatial recurrent connectivity profiles. g, Strength of winner-take-all effect (quantified as LHb-evoked change in A-recipient 5-HT neuron responses) as A input rate and LHb input rate are varied. Shown are recurrent spatial connectivity values of c = 1 (left) and c = 1/4 (right). Note that the winner-take-all effect is weaker than with short-term plasticity intact (e).

Extended Data Fig. 7 Winner-take-all dynamics in the DRN persist over a wide range of overlap fractions between competing inputs, a phenomenon which depends on recurrent short-term plasticity. Related to network simulations shown in Fig. 7.

a, Schematic of network simulation, with LHb and Area A providing input to 5-HT neurons (top). Bottom contains schematic demonstrating how the overlap between LHb and A inputs onto 5-HT neurons is varied from 0-60%. b, Simulation with small input overlap of o = 0.2. Shown are mean firing rates of A and LHb inputs (top); mean firing rates of A-recipient 5-HT neurons (middle); and raster plot depicting firing rates of 2500 5-HT neurons receiving either LHb input (purple) or A input (orange) (depicted is the 50% of 5-HT neurons receiving direct input from either LHb or A). c, Simulation with larger input overlap of o = 0.6. Layout of panels is identical to in b. d-f quantify the strength of winner-take-all dynamics as a function of input overlap fraction, while short-term plasticity at recurrent connections is intact. d, LHb-evoked change in A-recipient 5-HT neuron responses during the transient phase (0–200 ms), as a function of the rate of A input, for 5 Hz (left) and 20 Hz (right) LHb input. Input overlap is varied (shown in legend). e, LHb-evoked change in A-recipient 5-HT neuron responses during the stationary phase (200–500 ms), as a function of the rate of A input, for 5 Hz (left) and 20 Hz (right) LHb input. Input overlap is varied (shown in legend). f, Strength of winner-take-all effect (quantified as LHb-evoked change in A-recipient 5-HT neuron responses) as A input rate and LHb input rate are varied. Shown are input overlap values of o = 0 (left) and o = 0.4 (right), illustrating winner-take-all dynamics (ie negative 5-HT modulation) for these input fractions. g-i quantify the strength of winner-take-all dynamics as a function of input overlap fraction, while short-term plasticity at recurrent connections has been removed (that is linearized responses). g, LHb-evoked change in A-recipient 5-HT neuron responses during the transient phase (0–200 ms), as a function of the rate of A input, for 5 Hz (left) and 20 Hz (right) LHb input. Input overlap is varied (shown in legend). h, LHb-evoked change in A-recipient 5-HT neuron responses during the stationary phase (200–500 ms), as a function of the rate of A input, for 5 Hz (left) and 20 Hz (right) LHb input. Input overlap is varied (shown in legend). i, Strength of winner-take-all effect (quantified as LHb-evoked change in A-recipient 5-HT neuron responses) as A input rate and LHb input rate are varied. Shown are input overlap values of o = 0 (left) and o = 0.4 (right), illustrating that winner-take-all dynamics are no longer present with moderate input overlap of o = 0.4 when short-term plasticity is removed (compare with f).

Extended Data Fig. 8 Recurrent inhibition gates spiking activity of 5-HT neurons.

a, Schematic of experimental design. Left, AAV-DIO-ChETA-YFP is injected into the DRN at a 2:1 ratio with Mannitol. Right, whole-cell recordings from ChETA-negative 5-HT neurons. b, Recurrent 5HT1AR-mediated current in ChETA-negative 5-HT neuron. c, Spiking of a 5-HT neuron in response to current injection. Below, raster plot showing the same neuron’s activity across trials. d, Spiking of a 5-HT neuron in response to current injection and 20 Hz optical activation of ChETA-positive 5-HT neurons. Below, raster plot showing the same neuron’s activity across trials. e, Spiking of a 5-HT neuron in response to current injection in the presence of WAY100-635. Below, raster plot showing the same neuron’s activity across trials. f, Spiking of a 5-HT neuron in response to current injection and 20 Hz optical activation of ChETA-positive 5-HT neurons, in the presence of WAY100-635. Below, raster plot showing the same neuron’s activity across trials. g, Smoothed firing rate (normalized) over time for pre- and post-WAY recordings with LED stimulation (mean +- SE, n = 3 neurons). h, Cumulative spike fraction in pre- and post-WAY recordings with LED stimulation (n = 3 neurons, two-sided Kolmogorov-Smirnov test: p = 7.4*10−15). i, Smoothed firing rate (normalized) over time for pre- and post-WAY recordings without LED stimulation (mean +- SE, n = 3 neurons). j, Cumulative spike fraction in pre- and post-WAY recordings with LED stimulation (3 neurons, two-sided Kolmogorov-Smirnov test: p = 0.85).

Source data

Extended Data Fig. 9 LHb-driven, 5HT1AR mediated, recurrent feedforward modulation of 5-HT firing activity in the DRN.

a, Schematic of experimental design. Left, AAV-ChR2(H134R)-mCherry is injected into the LHb. Right, whole-cell recordings from TdT-positive 5-HT neurons. b, LHb-driven heterosynaptic 5HT1AR-mediated potential in SERT-TdT 5-HT neuron. c, Spiking of a 5-HT neuron in response to a 2.5 s current step (scale, 20 mV, 500 ms). Below, raster plot of the same 5-HT neuron spiking. d, Spiking of a 5-HT neuron in response to a 2.5 s current step with LED activation of ChR2-expressing LHb terminals in the slice for 1 sec in the middle of the current step. Below, raster plot of the same 5-HT neuron spiking. e, Spiking of a 5-HT neuron in response to a 2.5 s current step in the presence of WAY 100-635. Below, raster plot of the same 5-HT neuron spiking. f, In the presence of WAY 100-635, spiking of a 5-HT neuron in response to a 2.5 s current step with LED activation of ChR2-expressing LHb terminals in the slice for 1 sec in the middle of the current step. Below, raster plot of the same 5-HT neuron spiking. g, Instantaneous frequency line (top) and spike count histogram plots (below) from recording in c (mean +- SEM). h, Instantaneous frequency line (top) and spike count histogram plots (below) from recording in d (mean +- SEM). i, Instantaneous frequency line (top) and spike count histogram plots (below) from recording in e (mean +- SEM). j, Instantaneous frequency line (top) and spike count histogram plots (below) from recording in f (mean +- SE). k, Instantaneous frequency population plot in pre- and post-WAY recordings with LED stimulation (n = 3 neurons, mean +- SEM). l, Cumulative frequency population plot in pre- and post-WAY recordings with LED stimulation (3 cells, p = 0.0017, two-sided Kolmogorov-Smirnov test). m, Instantaneous frequency population plot in pre- and post-WAY recordings in the absence of LED stimulation (n = 3 neurons, mean +- SEM). n, Cumulative frequency population plot in pre- and post-WAY recordings in the absence of LED stimulation (3 cells, p = 0.07, two-sided Kolmogorov-Smirnov test).

Source data

Extended Data Fig. 10 The habenulo-raphe pathway disrupts reward-conditioned responses in a value-dependent manner and does not affect unconditioned responses.

a, Confocal image depicting a representative example of LHb injection site. b, Mean licking across all small reward trials for a single mouse depicting control trials (black) or 20 Hz optogenetic stimulation trials (blue). c, Mean licking across all small reward trials for a single mouse depicting control trials (black) or 5 Hz optogenetic stimulation trials (blue). d, Mean licking across all no reward trials for a single mouse depicting control trials (black) or 20 Hz optogenetic stimulation trials (blue). e, Mean licking across all no reward trials for a single mouse depicting control trials (black) or 5 Hz optogenetic stimulation trials (blue). f-g depict post-reward optogenetic stimulation experiments. f, Mean licking across all large reward trials for a single mouse depicting control trials (black) or trials where 20 Hz optogenetic stimulation was delivered during the post-reward period (blue). g, Change in post-reward licking during optogenetic stimulation trials, compared to pre-stimulated baseline period (n = 6 mice, paired two-sided t-test, unstimulated vs stimulated reward licking). h, Correlation between mean trace licking (large reward trials) and optogenetic effect size for 20 Hz stimulation (dark blue) and 5 Hz stimulation (light blue) for trace stimulation during large reward trials. i, Correlation between effect size for 20 Hz stimulation and 5 Hz stimulation for trace stimulation during large reward trials.

Source data

Supplementary information

Source data

Source Data Fig. 1

Statistical Source Data.

Source Data Fig. 2

Statistical Source Data.

Source Data Fig. 3

Statistical Source Data.

Source Data Fig. 4

Statistical Source Data.

Source Data Fig. 5

Statistical Source Data.

Source Data Fig. 8

Statistical Source Data.

Source Data Extended Data Fig. 1

Statistical Source Data.

Source Data Extended Data Fig. 2

Statistical Source Data.

Source Data Extended Data Fig. 3

Statistical Source Data.

Source Data Extended Data Fig. 8

Statistical Source Data.

Source Data Extended Data Fig. 9

Statistical Source Data.

Source Data Extended Data Fig. 10

Statistical Source Data.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lynn, M.B., Geddes, S.D., Chahrour, M. et al. Nonlinear recurrent inhibition through facilitating serotonin release in the raphe. Nat Neurosci 28, 1024–1037 (2025). https://doi.org/10.1038/s41593-025-01912-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue date:

  • DOI: https://doi.org/10.1038/s41593-025-01912-7

Search

Advanced search

Quick links

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