Ventromedial Thalamus-Projecting DCN Neurons Modulate Associative Sensorimotor Responses in Mice

Abstract

The deep cerebellar nuclei (DCN) integrate various inputs to the cerebellum and form the final cerebellar outputs critical for associative sensorimotor learning. However, the functional relevance of distinct neuronal subpopulations within the DCN remains poorly understood. Here, we examined a subpopulation of mouse DCN neurons whose axons specifically project to the ventromedial (Vm) thalamus (DCN^Vm neurons), and found that these neurons represent a specific subset of DCN units whose activity varies with trace eyeblink conditioning (tEBC), a classical associative sensorimotor learning task. Upon conditioning, the activity of DCN^Vm neurons signaled the performance of conditioned eyeblink responses (CRs). Optogenetic activation and inhibition of the DCN^Vm neurons in well-trained mice amplified and diminished the CRs, respectively. Chemogenetic manipulation of the DCN^Vm neurons had no effects on non-associative motor coordination. Furthermore, optogenetic activation of the DCN^Vm neurons caused rapid elevated firing activity in the cingulate cortex, a brain area critical for bridging the time gap between sensory stimuli and motor execution during tEBC. Together, our data highlights DCN^Vm neurons' function and delineates their kinematic parameters that modulate the strength of associative sensorimotor responses.

Keywords: Deep cerebellar nuclei; Sensorimotor learning; Trace eyeblink conditioning; Ventromedial thalamus.

Fig. 1

The DCN projects to the ventral thalamus in mice. A Schematic of viral injections for anterograde tracing of DCN axons in the ventral thalamus. B Representative images showing the distribution of mCherry⁺ neurons (red) in the interpositus (Int) and lateral (Lat) cerebellar nuclei (scale bar, 500 μm). Med, medial cerebellar nucleus. Inset, magnified view of the DCN neurons in the same panel (scale bar, 100 μm). C The distribution of DCN axons (red) in the Vm and VL. Inset, magnified view of the DCN axon terminals in the same panel (scale bar, 100 μm). Vm, ventromedial thalamus; VL, ventrolateral thalamus. D Schematic of viral injections for retrograde tracing the Vm-projecting DCN neurons (DCN^Vm). E Left, representative fluorescence image of showing retrograde tracer retro-beads in the Vm; right, distribution of DCN^Vm neurons (red) in the interpositus (Int) and lateral (Lat) cerebellar nuclei (scale bar, 500 μm or 200 μm). Inset, magnified view of the labeled DCN^Vm neurons (scale bar, 100 μm).

Fig. 2

Tracing neurons in the DCN projecting to the Vm. A Schematic of viral injections for retrograde tracing of DCN neurons projecting to the Vm (DCN^Vm). B Left, representative fluorescence image showing DCN^Vm neurons expressing mCherry in the medial (Med), interpositus (Int), and lateral (Lat) DCN; right, representative fluorescence image of showing the axon terminals of DCN^Vm neurons in the Vm (scale bar, 100 μm). VL, ventrolateral thalamus; Vm, ventromedial thalamus.

Fig. 3

In vivo optogenetic identification of DCN^Vm neurons. A Upper, schematic of optogenetic identification of DCN^Vm neurons in vivo; lower, representative fluorescence image of showing a recording site in the DCN (scale bar, 200 μm). Inset, magnified view of the recording site in the same panel (scale bar, 100 μm). B Upper, example recording from an optogenetically-identified DCN^Vm neuron; lower, averaged blue laser-evoked firing activity in optogenetically-identified DCN^Vm neurons (averaged from 81 units in 10 mice). C Distribution of latency of laser-evoked spiking for all optogenetically-identified DCN^Vm neurons (n = 81 units from 10 mice). D Distribution of Pearson correlation coefficients between blue laser-evoked and spontaneous spike waveforms for all optogenetically-identified DCN^Vm neurons (n = 81 units from 10 mice) (scale bars, 0.5 ms and 0.1 mV).

Fig. 4

CS-evoked DCN^Vm neuronal activity is modulated by tEBC training. A Overview of the experimental design. Before daily training, 450-nm blue laser pulses are presented to optogenetically identify DCN^Vm neurons in vivo (Epoch #1). Then, the mice (n = 10) are subjected to trace eyeblink conditioning (tEBC) training (Epoch #2). B Averaged eyelid responses in a representative mouse across 5 consecutive training days [arrowheads, conditioned eyeblink response (CR); scale bar, 250 ms]. C CR incidence measured from 10 mice across 5 consecutive training days. The mice show a clear increase in CR incidence across tEBC training (F _{(4, 36)} = 5.267, P = 0.002, one-way ANOVA with repeated measures). The CR incidence at the late-learning stage (days 4 and 5) is significantly higher than that at the early-learning stage (t ₍₁₈₎ = −2.709, P = 0.014, independent t test). D Pseudo-color map showing the Z-score-transformed firing activity of DCN^Vm neurons across distinct learning stages (early, n = 28 vs late, n = 34). E Stacked bar graphs illustrating the proportions of DCN^Vm neurons with significantly altered CS-evoked activity (both increased and decreased) and those with no response. A greater proportion of DCN^Vm neurons show CS-evoked elevated activity at the late-learning stage than at the early-learning stage [late, 44.1% (15/34) vs early, 28.6% (8/28)]. F Average CS-evoked DCN^Vm activity in the early-learning (n = 28, blue trace) and late-learning (n = 34, red trace) stages. G CS-evoked activity of DCN^Vm neurons (n = 34, red circles) at the late-learning stage is significantly greater than at the early-learning stage (n = 28, blue circles; Z = −2.1288, P = 0.0333, Wilcoxon rank sum test). H US-evoked activity of DCN^Vm neurons (n = 34, red circles) at the late-learning stage are comparable to that at the early-learning stage (n = 28, blue circles; Z = 0.4314, P = 0.6662, Wilcoxon rank sum test). Data are expressed as the mean ± SEM; *P <0.05, n.s., not significant. CS, conditioned stimulus (blue bar); US, unconditioned stimulus (gray bar); LD, laser diode.

Fig. 5

DCN^Vm neuronal activity in the CR vs no-CR trials. A Averaged CS-evoked activity of DCN^Vm neurons (n = 30) in the CR vs no-CR trials. CR, conditioned eyelid response; CS, conditioned stimulus (blue); US, unconditioned stimulus (gray). B CS-evoked DCN^Vm firing activity in CR trials is significantly greater than in no-CR trials (t ₍₂₉₎ = 2.8739, P = 0.0075, paired t test). C US-evoked DCN^Vm firing activity in CR trials is comparable to that in no-CR trials (t ₍₂₉₎ = −0.1062, P = 0.9161, paired t test). For B and C, 4 DCN^Vm neurons were excluded from analysis because they generated too few spikes to normalize their firing in the no-CR state. Data are expressed as the mean ± SEM; **P <0.01, n.s., not significant.

Fig. 6

Optogenetic activation of DCN^Vm neurons improves CR performance. A Left, coronal section of a mouse brain showing ChR2-mCherry expression (red) stained with DAPI (blue) in DCN^Vm neurons (scale bar, 200 μm); right, overview of the experimental design. Two groups of mice (n = 9 for ChR2-expressing and n = 7 for mCherry-expression in DCN^Vm neurons) received CS–US paired presentations for 5 consecutive days. B Averaged eyeblink responses illustrating the effect of optogenetic DCN^Vm activation on the performance of learned CRs. The optogenetic stimulation is trigged by the CS, and is presented 1 ms–150 ms before the onset of the US. C Left, smoothed orbicularis oculi electromyography (O.O.EMG) recorded from the left upper eyelid during tEBC. Each line represents O.O.EMG activity from individual trials (1–100, with Trial 1 at the bottom and Trial 100 at the top). Right, smoothed O.O.M. EMG recorded from the left upper eyelid. During tEBC, the optogenetic stimulation was trigged by the CS, and are presented 150 ms before the onset of the US. For B and C, CS, conditioned stimulus (blue); US, unconditioned stimulus (gray). D–F CR incidence (D), CR amplitude (E), and latency to the CR peak (F) measured from DCN^Vm-ChR2 (n = 9, blue bars) and DCN^Vm-mCherry (n = 7, open bars) mice. Optogenetic activation of DCN^Vm neurons increases the CR incidence and peak amplitude in DCN^Vm-ChR2 mice (CR incidence, t ₍₈₎ = −3.419, P = 0.009; increase in CR peak amplitude: t ₍₈₎ = −3.013, P = 0.017, paired t tests). Data are expressed as the mean ± SEM; **P <0.01, *P <0.05, n.s., not significant.

Fig 7

Optogenetic inhibition of DCN^Vm neurons impairs CR performance. A Left, coronal section of a mouse brain showing ArchT-GFP expression (green) stained with DAPI (blue) in DCN^Vm neurons (scale bar, 200 μm); right, overview of the experimental design. Two groups of mice (n = 6 for ArchT-expression and n = 5 for GFP-expression in DCN^Vm neurons) receive CS–US paired presentations for 5 consecutive days. B Averaged eyeblink responses illustrating the effect of optogenetic DCN^Vm inhibition on CR performance. An optogenetic stimulus is trigged by the CS, and is presented 1 ms–250 ms before the onset of the US. C, D CR amplitude (C) and CR incidence (D) measured from DCN^Vm-ArchT (n = 6, green bars) and DCN^Vm-GFP (n = 5, open bars) mice. Optogenetic inhibition of the DCN^Vm neurons significantly diminishes the CR peak amplitude in DCN^Vm-ArchT mice (t ₍₅₎ = 2.663, P = 0.037, paired t test). In contrast, optogenetic inhibition of DCN^Vm neurons has no effect on the CR incidence in DCN^Vm-ArchT mice (t ₍₅₎ = 0.858, P = 0.430, paired t test). Data are expressed as the mean ± SEM; *P <0.05, n.s., not significant.

Fig. 8

Optogenetic activation of DCN^Vm neurons alters firing activity in cingulate cortical neurons. A Upper, schematic of tetrode recording in the cingulate cortex (Cg) and the optogenetic activation of DCN^Vm neurons; lower, representative DAPI image showing a tetrode recording site in the Cg (scale bar, 500 μm). B Upper, heatmap rows represent Z-score-transformed average peristimulus time histograms for individual Cg units (n = 157 units from 3 mice), and columns represent time bins relative to the onset of optogenetic stimulation (10-ms bin width); lower, plot shows the average firing responses of Cg units activated (n = 50, red trace) or suppressed (n = 32, blue trace) by DCN^Vm. The average firing rates of Cg units activated by DCN^Vm activation were significantly higher than those suppressed by DCN^Vm activation (Z = 3.7190, P = 0.0002, Wilcoxon rank sum test). C Proportions of increased firing (31.8%, red), decreased firing (20.4%, blue) and not significantly modulated (47.8%, gray) evoked in Cg units by DCN^Vm activation. D Plot showing CS-evoked firing responses of Cg units activated (n = 50, red trace) or suppressed (n = 32, blue trace) by DCN^Vm activation. The CS responsiveness of Cg units with increased firing was significantly higher than those with decreased firing evoked by DCN^Vm activation (Z = 4.2056; P = 2.6045 × 10^-5, Wilcoxon rank sum test). Data are expressed as the mean ± SEM; shaded areas indicate SEM; ***P <0.001.

Fig. 9

Chemogenetic manipulation of DCN^Vm neurons has no effects on motor coordination. A Upper, schematic for the chemogenetic manipulation of DCN^Vm neurons; lower, coronal section of a representative mouse brain showing hM4Di (or hM3Dq)-mCherry expression (red) stained with DAPI (blue) in DCN^Vm neurons (scale bar, 1 mm). B Mean latency to fall from the rod in DCN^Vm-hM4Di mice increased from day 1 to day 4 (F _{(3, 15)} = 3.413, P = 0.045, n = 6 mice, one-way ANOVA with repeated measures). C Effects of CNO and saline injection on motor coordination in DCN^Vm-hM4Di mice. There is no significant difference in the mean latency to fall between the CNO (red) and saline (open) injections (n = 6, t ₍₅₎ = −1.041, P = 0.346, paired t test). D Mean latency to fall from the rod in DCN^Vm-hM3Dq mice likewise, increased from day 1 to day 4 (F _{(3, 15)} = 4.242, P = 0.023, n = 6 mice, one-way ANOVA with repeated measures). E Effects of CNO and saline injections on motor coordination in DCN^Vm-hM3Dq mice. There is no significant difference in the mean latency to fall between the CNO (red bar) and saline (open bar) injections (t ₍₅₎ = −0.622, P = 0.561, paired t test). Data are expressed as the mean ± SEM; n.s., not significant.