Abstract
The lateral habenula encodes aversive stimuli contributing to negative emotional states during drug withdrawal. Here we report that morphine withdrawal in mice leads to microglia adaptations and diminishes glutamatergic transmission onto raphe-projecting lateral habenula neurons. Chemogenetic inhibition of this circuit promotes morphine withdrawal-like social deficits. Morphine withdrawal-driven synaptic plasticity and reduced sociability require tumor necrosis factor-α (TNF-α) release and neuronal TNF receptor 1 activation. Hence, habenular cytokines control synaptic and behavioral adaptations during drug withdrawal.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.
References
Goeldner, C. et al. Biol. Psychiatry 69, 236–244 (2011).
Lutz, P. E. et al. Neuropsychopharmacology 39, 2694–2705 (2014).
Meye, F. J., Trusel, M., Soiza-Reilly, M. & Mameli, M. Pharm. Biochem. Behav. 162, 87–93 (2017).
Margolis, E. B. & Fields, H. L. PLoS One 11, e0159097 (2016).
Wang, J. et al. Neurosci. Lett. 653, 64–70 (2017).
Valentinova, K. & Mameli, M. Cell Rep. 16, 2298–2307 (2016).
Pollak Dorocic, I. et al. Neuron 83, 663–678 (2014).
Lewitus, G. M. et al. Neuron 90, 483–491 (2016).
Campbell, L. A., Avdoshina, V., Rozzi, S. & Mocchetti, I. Brain Behav. Immun. 34, 130–140 (2013).
Michaud, M. et al. J. Am. Med. Dir. Assoc. 14, 877–882 (2013).
Probert, L. Neuroscience 302, 2–22 (2015).
Nie, X. et al. Neuron 99, 464–479.e7 (2018).
Hutchinson, M. R. et al. Brain Behav. Immun. 22, 1178–1189 (2008).
Patel, A., Siegel, A. & Zalcman, S. S. Brain Behav. Immun. 24, 1276–1280 (2010).
He, P., Liu, Q., Wu, J. & Shen, Y. FASEB J. 26, 334–345 (2012).
Stellwagen, D., Beattie, E. C., Seo, J. Y. & Malenka, R. C. J. Neurosci. 25, 3219–3228 (2005).
Kane, M. J. et al. PLoS One 7, e48975 (2012).
Lecca, S. et al. eLife 6, e30697 (2017).
Matthews, G. A. et al. Cell 164, 617–631 (2016).
Acknowledgements
The authors thank F.J. Meye and the Mameli Laboratory for comments on the manuscript. This work was supported by funds from the ERC StG SalienSy 335333 and the SNSF (31003A) to M.M. They also thank D. Szymkowski for the donation of XENP1595, and G. Kollias and H. Strubbe for the use and breeding of the Tnfr1fl/fl mouse line.
Author information
Authors and Affiliations
Contributions
K.V., A.T. and M.M. performed and analyzed the ex vivo recordings and behavior experiments. A.L.L. and J.A.C. contributed to the ex vivo recordings. M.T., I.M. and L.M. performed the molecular biology experiments. C.B. and S.T. provided support for the behavioral experiments. A.M. and R.C.P. analyzed the microglia morphology. A.V. provided conceptual and experimental input related to TNF-α signaling and the Tnfr1fl/fl mice. K.V. and M.M. conceptualized, designed the study and wrote the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Journal peer review information: Nature Neuroscience thanks David Stellwagen and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Integrated supplementary information
Supplementary Figure 1 MORwd induces postsynaptic depression of AMPAR-mediated neurotransmission in medial LHb.
(a) Left: sample traces, box and scatter plots of sEPSCs amplitudes recorded in LatLHb (saline+naloxone (ncells/mice = 9/3; gray) versus NP-MORwd (ncells/mice = 9/4; orange), two-sided t-test,, t16 = 0.098, P = 0.923). Right: same but sEPSCs were recorded in MedLHb (saline+naloxone (ncells/mice = 9/4; black) versus NP- MORwd (ncells/mice = 9/5; red), two-sided t-test, t16 = 3.493, **P = 0.003). (b) Left: sample traces, box and scatter plots of sEPSCs frequencies recorded in LatLHb (saline+naloxone (ncells/mice = 9/4, gray) versus NP-MORwd (ncells/mice = 9/4, orange), two-sided t-test,, t16 = 1.331, P = 0.202). Right: same but sEPSCs were recorded in MedLHb (saline+naloxone (ncells/mice = 9/5, black) versus NP-MORwd (ncells/mice = 9/4,red), two-sided t-test, t16 = 0.161, P = 0.874). (c) Recording map color-coded for the value of AMPAR:NMDAR ratios recorded throughout the LHb. Lighter colors indicate smaller AMPAR:NMDAR ratios, while darker colors represent high AMPAR:NMDAR ratio. (d) Top: Sample traces and normalized EPSC versus pulse number plots recorded at 5, 10 and 20 Hz in LatLHb (saline+naloxone (ncells/mice = 10/2, gray) versus NP-MORwd (ncells/mice = 10/3, orange), 5 Hz interaction factor F(4, 36) = 0.227, P = 0.921; 10 Hz interaction factor F(4,36) = 0.251, P = 0.907; 20Hz interaction factor F(4,36) = 0.573, P = 0.683 two-way ANOVA Repeated Measures). Bottom: same but in MedLHb (ncells/mice = 10/3, saline+naloxone (black) versus NP- MORwd (red), 5Hz interaction factor F(4,36) = 0.1.183, P = 0.334; 10Hz interaction factor F(4,36) = 1.171, P=0.34; 20Hz interaction factor F(4,36) = 0.88, P = 0.485 two-way ANOVA Repeated Measures). (e) Spontaneous withdrawal timeline. AMPAR:NMDAR ratios from MedLHb 1 h, 10, 20 or 30 days post-saline or MOR (saline 1 h and 10 days pooled (nmice/cells = 6/22; black) versus MORwd 1 h (nmice/cells = 5/11; open red) and MOR 10 (nmice/cells = 3/12), 20 (nmice/cells = 3/11) and 30 days withdrawal (nmice/cells = 3/11; red), F(4, 62) = 3.90 one-way ANOVA, **P = 0.007). (f) Example of peak-scaled NSFA of MedLHb neurons in the saline- and NP-MORwd group. Pooled data for conductance (γ) and number of channels (N) open at the peak together with amplitude versus N of channels and conductance plots (Saline+naloxone, ncells/mice = 5/4; MORwd, ncells/mice = 8/5; N of channels, two-sided t-test, t11 = 5.67, ***P = 0.0001, r2(N-Channels) = 0.416; *P = 0.017; Conductance, t11 = 0.006, P = 0.99, r2(Conductance) = 0.03, P = 0.55). (g) Left: sample traces, box and scatter plots for rectification index calculated from AMPAR EPSCs recorded at -70, 0 and 40 mV in LatLH (saline+naloxone (ncells/mice = 9/7, gray) versus NP- MORwd (ncells/mice = 10/7, orange), two-sided t-test, t17 = 0.210, P = 0.836). Right same but recordings in MedLHb (saline+naloxone (ncells/mice = 12/8, black) versus NP- MORwd (ncells/mice = 9/5, red), two-sided t-test, t19 = 1.292, P = 0.212). (h) Sample traces, box and scatter plots of AMPAR:NMDAR ratios recorded in MedLHb via 405 nm laser-assisted uncaging of MNI-glutamate, 500 μM (saline+naloxone (ncells/mice = 8/2, black) versus NP-MORwd (ncells/mice = 10/3, red), two-sided t-test, t16 = 3.521, **P = 0.003). Bottom right: Absolute AMPAR versus absolute NMDAR uncaging-evoked current plots from saline+naloxone (open black circles) or NP-MORwd mice (open red circles). The mean with S.E.M. AMPA and NMDA currents are shown with black and red filled circles for saline versus MORwd respectively (saline+naloxone versus NP-MORwd: AMPA, two-sided t-test, t16 = 3.536, **P = 0.003; NMDA, two-sided t-test, t16 = 0.195, P = 0.848). Data are presented as box plots 10–90 percentiles and scatter.
Supplementary Figure 2 TNFα levels in the LHb increase following spontaneous MORwd.
(a) TNFα (cyan) and DAPI (magenta) immunostaining in slices from saline-treated, (b) MOR-treated (sacrificed 1 h after the last MOR injection) and (c) animals in spontaneous MORwd (10–13 days post last MOR injection). (d) Normalized LHb TNFα optical density in saline (black), MOR (open red) and spontaneous MORwd (red) (nmice = 8, saline (black) versus MOR (open red) versus spontaneous MORwd (red), F(2,20) = 7.7 one-way ANOVA, **P = 0.003). Data are presented as box plots 10–90 percentiles with median and scatter.
Supplementary Figure 3 TNFα signaling is necessary and sufficient for MORwd-induced plasticity.
(a) Spontaneous MORwd protocol, sample traces, box and scatter plots for AMPAR:NMDAR ratios recorded in MedLHb slices incubated with (+) or without (–) exogenous TNFα from spontaneous MORwd mice (10 days). (MORwd -TNFα (red) versus +TNFα (pink), ncells/mice = 9/4, two-sided t-test, t16 = 0.986, P = 0.339). (b) MedLHb AMPAR:NMDAR ratios from saline or MPLA-injected MOR-treated mice (MOR/saline (nmice/cells = 3/10; shaded blue) versus MOR/MPLA (nmice/cells = 4/11; dark blue), two-sided t-test, t19 = 3.070, **P = 0.006). (c) Sample traces, time versus amplitude plot and bar graphs showing the effect of MPLA (1 μg/ml) on AMPAR-EPSCs (MOR (nmice/cells = 4/11 ; 63.93 ± 7.06%; open red), NP-MORwd (nmice/cells = 3/10 cells; 91.63 ± 8.86%; filled red), MOR versus NP-MORwd, two-sided t-test, t19 = 2.419, *P = 0.026). Data of this panel are represented as mean and sem. (d) Sample traces, time versus amplitude plot and bar graphs showing the effect of MPLA (1 μg/ml) on evoked AMPAR-EPSCs (baseline (1) vs 30 min post-MPLA (2)) recorded in LatLHb (open orange) or MedLHb (open red) in slices obtained from morphine-treated animals (ncells/mice=8/4, morphine LatLHb 103.98 ± 10.13; ncells/mice = 11/5, morphine MedLHb 63.31 ± 7.06%; morphine LatLHb versus morphine MedLHb, two-sided t-test, t17 = 3.406, **P = 0.003). Note that the data set for MedLHb is the same as in c and is used for comparison. Data are presented as mean and SEM. (e) Sample traces, time versus amplitude plot and bar graph showing the effect of MPLA (1 μg/ml) on evoked AMPAR-EPSCs (baseline (1) vs 30 min post-MPLA (2)) in the presence of TNFα dominant negative peptide (XENP1595, 6mg/1ml) recorded in MedLHb in slices obtained from morphine-treated animals (ncells/mice = 7/2, 98.88 ± 8.23 %, two-sided t-test, t-test, t6 = 0.073, P = 0.944). Data are presented with mean and SEM. (f) NP-MORwd protocol with dominant-negative TNFα (XENP1595, 30 mg/kg) pretreatment, sample traces, box and scatter plots for AMPAR:NMDAR ratios recorded in MedLHb (saline+naloxone (ncells/mice = 10/3, black) versus NP-MORwd (ncells/mice = 12/3, green), two-sided t-test, t-test, t20 = 0.165, P = 0.871). Data are presented as box plots 10–90 percentiles and scatter.
Supplementary Figure 4 Behavioral assessment of MORwd.
(a) Box and scatter plot showing the percent time spent in the compartment containing the social stimulus for C57Bl6 mice (N = 22 mice/group, saline+naloxone (black) versus NP-MORwd (red), two-sided t-test, t42 = 2.401, *P = 0.021). (b) Box and scatter plot showing the percent time spent in the compartment containing the object stimulus for C57Bl6 mice (N = 22 mice/group, saline+naloxone (black) versus NP-MORwd (red), two-sided t-test, t42 = 2.465, *P = 0.02). (c) Box and scatter plot showing the percent time spent in the central compartment for C57Bl6 mice (N = 22 mice/group, saline+naloxone (black) versus NP-MORwd (red), two-sided t-test, t42 = 1.186, P = 0.242). (d) Box and scatter plot showing locomotor activity during social preference test for C57Bl6 mice (N = 22 mice/group, saline+naloxone (black) versus NP-MORwd (red), two-sided t-test, t42 = 2.621, *P = 0.012). (e) Box and scatter plot showing the percent time spent in the compartment containing the social stimulus for TNF-R1fl/fl mice (AAV-Control: 58.02 ± 2.96% saline+naloxone (Nmice = 20, black) versus 40.45 ± 5.08% NP-MORwd (Nmice = 23, red); AAV-Cre: 54.41 ± 3.15% saline (Nmice = 13, open gray) versus 64.52 ± 5.94 % NP-MORwd (Nmice = 13, open pink), interaction factor F(1,65) = 8.591 two-way ANOVA, **P = 0.005). (f) Box and scatter plot showing the percent time spent in the compartment containing the object stimulus for TNF-R1fl/fl mice (N of mice same as panel e. AAV-Control: 26.42 ± 2.47 % saline+naloxone (black) versus 30.9 ± 4.79% NP-MORwd (red); AAV-Cre: 31.19 ± 2.07% saline+naloxone (open gray) versus 20.7 ± 4.13 % NP-MORwd (open pink), interaction factor F(1,65) = 4.136 two-way ANOVA, *P = 0.046). (g) Box and scatter plot showing the percent time spent in the central compartment for TNF-R1fl/fl mice (N of mice same as panel e. AAV-Control: 15.38 ± 3.08% saline+naloxone (black) versus 27.82 ± 4.98% NP-MORwd (red); AAV-Cre: 14.23 ± 2.9 % saline+naloxone (open gray) versus 14.62 ± 3.08% NP-MORwd (open pink), interaction factor F(1,65) = 1.748 two-way ANOVA, P = 0.191). (h) Box and scatter plot showing number of exploration bouts of TNF-R1fl/fl with the juvenile (N of mice same as panel e. AAV-Control: 56 ± 3.51 saline+naloxone (black) versus 39.30 ± 4.8 NP-MORwd (red); AAV-Cre: 50.38 ± 3.78 saline+naloxone (open gray) versus 57.15 ± 7.01 NP-MORwd (open pink), interaction factor F(1.65) = 5.519 two-way ANOVA, *P = 0.022). (i) Box and scatter plot showing number of exploration bouts of TNF-R1fl/fl with the object (N of mice is the same as panel e. AAV-Control: 26.9 ± 1.88 saline+naloxone (black) versus 23.35 ± 2.13 NP-MORwd (red); AAV-Cre: 28.08 ± 2.01 saline+naloxone (open gray) versus 21.69 ± 3.16 NP-MORwd (open pink), interaction factor F(1.65) = 0.361, P = 0.55). (j) Box and scatter plot showing locomotor activity during social preference test for TNF-R1fl/fl mice (N of mice is the same as panel e. AAV-Control: 28.13 ± 1.35 m saline+naloxone (black) versus 33.8 ± 5.59 m NP-MORwd (red); AAV-Cre: 23.78 ± 1.47m saline+naloxone (open gray) versus 40.17 ± 7.11 m NP-MORwd (open pink), interaction factor F(1,65) = 1.224 two-way ANOVA, P = 0.273). Data are presented as box plots 10–90 percentiles with median and scatter.
Supplementary information
Rights and permissions
About this article
Cite this article
Valentinova, K., Tchenio, A., Trusel, M. et al. Morphine withdrawal recruits lateral habenula cytokine signaling to reduce synaptic excitation and sociability. Nat Neurosci 22, 1053–1056 (2019). https://doi.org/10.1038/s41593-019-0421-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41593-019-0421-4
This article is cited by
-
Glutamatergic neurons in ventral pallidum modulate heroin addiction via epithalamic innervation in rats
Acta Pharmacologica Sinica (2024)
-
Neural circuits regulating prosocial behaviors
Neuropsychopharmacology (2023)
-
Blockade of the Dopamine D3 Receptor Attenuates Opioids-Induced Addictive Behaviours Associated with Inhibiting the Mesolimbic Dopamine System
Neuroscience Bulletin (2023)
-
TNFR1 signaling converging on FGF14 controls neuronal hyperactivity and sickness behavior in experimental cerebral malaria
Journal of Neuroinflammation (2023)
-
Reward and aversion encoding in the lateral habenula for innate and learned behaviours
Translational Psychiatry (2022)