Activation of lateral habenula inputs to the ventral midbrain promotes behavioral avoidance

Abstract

Lateral habenula (LHb) projections to the ventral midbrain, including the rostromedial tegmental nucleus (RMTg), convey negative reward-related information, but the behavioral ramifications of selective activation of this pathway remain unexplored. We found that exposure to aversive stimuli in mice increased LHb excitatory drive onto RMTg neurons. Furthermore, optogenetic activation of this pathway promoted active, passive and conditioned behavioral avoidance. Thus, activity of LHb efferents to the midbrain is aversive but can also serve to negatively reinforce behavioral responding.

Figure 1. Acute unpredictable foot shock exposure enhances LHb-to-RMTg glutamate release

(a) Sagittal confocal image showing expression of ChR2-EYFP (green) in the LHb-to-midbrain pathway via the fasciculus retroflexus fiber bundle following injection of the viral construct into the LHb. Midbrain TH+ dopamine neurons are shown in blue. A, Anterior; P, Posterior; D, Dorsal, V, Ventral. (b) Horizontal confocal image showing the distribution of LHb terminals in the midbrain. M, medial; L, lateral. (c) Top: representative mEPSC traces recorded from neurons from mice immediately following either 0 or 19 unpredictable foot shocks. Bottom Left: Representative cumulative mEPSC inter-event interval probability plot. Inset: Average mEPSC frequency was significantly increased in neurons from shock exposed mice (t(13) = 2.88, p = 0.01). Bottom Right: Representative cumulative mEPSC amplitude probability plot. Inset: Average mEPSC amplitude was not altered in RMTg neurons from shock exposed mice (t(13) = 0.12, p = 0.91). (d) Left: Representative optically evoked paired-pulse ratios from LHb efferents onto RMTg neurons. Right: Average paired-pulse ratios showing that paired-pulse ratios at LHb-to-RMTg synapses were significantly depressed from mice that received foot shocks (t(14) = 3.56, p = 0.003). n = 8 cells/group. All error bars for all figures correspond to the s.e.m. * indicates p < 0.05 and ** indicates p < 0.01 for all figures.

Figure 2. Activation of LHb inputs to the RMTg produces behavioral avoidance

(a) Left: Real-time place preference location plots from two representative mice showing the animal’s position over the course of the 20-min session. Right: ChR2-EYFP-expressing mice spent significantly less time on the stimulated-paired side (t(10) = 7.90, p < 0.0001). n = 6 mice/group for real-time place preference. (b) ChR2-EYFP-expressing mice spent significantly less time in the stimulation-paired chamber compared to the non-stimulation-paired chamber 24 hrs after the last stimulation conditioning session (t(7) = 3.54, p = 0.01). EYFP-expressing did not show a preference (t(7) = 0.57, p = 0.58). (c) ChR2-EYFP-expressing mice spent significantly less time in the stimulation paired chamber compared to the non-stimulation-paired chamber 7 days after the last stimulation session (t(7) = 3.24, p = 0.01). EYFP-expressing mice did not show a preference (t(7) = 0.17, p = 0.86). n = 8 mice/group for conditioned place preference.

Figure 3. Activation of LHb inputs to the RMTg produces active behavioral avoidance and disrupts positive reinforcement

(a) Example cumulative records of active nose-pokes made by a ChR2-EYFP and an EYFP-expressing mouse to terminate LHb-to-RMTg optical activation. (b) Average number of active nose-pokes from one behavioral session in following training (> 4 days; t(10) = 20.52, p < 0.0001). There was no difference in inactive nose-pokes between the two groups (t(10) = 0.29, p = 0.78). n = 6 mice per group. (c) Example cumulative records of active nose-pokes made by a ChR2-EYFP and an EYFP-expressing mouse when optical stimulation was paired with the nose-poke to receive a sucrose reward. (d) Average number of active and inactive nose-pokes during positive reinforcement (t(14) = 4.01, p < 0.01). There was no difference in inactive nose-pokes between the two groups (t(14) = 1.22, p = 0.24). n = 8 mice per group.