Cerebellar nuclei partial inhibition during and after a motor task impairs learning.

a) Scheme of the implantation and injection. b) Coronal section of the cerebellum showing hM4Di-DREADD expression in the three CN. c) Representative confocal image showing hM4Di-DREADD expression on neuronal membranes. d) Electrode placement close to cells expressing hM4Di-DREADD (red: lesion site, yellow: electrode track). e) Examples of high-pass filtered traces of CN recordings before and after CNO injection. f) Examples of spike shapes obtained from spike sorting in CN (line+shading indicates mean +/- SD). g) Boxplots displaying the percentage of modulation of CN neurons average firing rate induced by CNO or SAL injections during an open field session. CN firing rate was reduced after 1 mg/kg of CNO injection in DREADD injected mice compared to other groups (Wilcoxon test* p<0.05, ***p<0.001. Boxes represent quartiles and whiskers correspond to range; points are singled as outliers if they deviate more than 1.5 x interquartile range from the nearest quartile). h) Latency to fall during fixed speed rotarod (5, 10, 15, 20r.p.m.) for all experimental groups. One way repeated-measure ANOVA on averaged values for all the speed steps in each experimental group followed by Tukey Posthoc pairwise comparison (p>0.05 in all cases). i) Impact of daily of injections of CNO before trial 1 on accelerating rotarod performance. Summary of the performance for each trial/day (repeated measure ANOVA Group effect *p<0.05, ***p<0.001; # p<0.05, ###p<0.001 Tukey pairwise test last trial of each day vs first trial of next day). j) Impact of daily of injections of CNO after the task (30 min after trial 7). All treatment is switched to saline in the Maintenance phase. Same presentation as in i). Data represents mean ± S.E.M. n indicates the number of mice. (See Supplementary Tables 1-5 for detailed statistics)

Changes in learning and consolidation of motor memory induced by cerebellar nuclei inhibition.

a) Example of evolution of latencies to fall for a mouse during the accelerating rotarod protocol. Linear regressions estimated values of trial 1 and 7 during each day are shown using green hollow dots, within-day learning (green arrows), overnight change (red arrows) and day+night learning (brown arrows). b) Evolution of estimated daily learning, overnight change and consolidated learning shows learning mostly in days 1-4 (Learning Phase). c) within-day learning vs daily initial performances (top) and within-day learning vs consolidated learning (bottom) in the Learning Phase. Scatterplot of performance from all control mice; the ellipse contains 50% of a bivariate normal distribution fitted to the values and the dot indicates the center of the distribution. Deming linear regression outcomes are represented with 95% confidence interval in shaded color. d) same as panel c) with superimposition of Controls (black) and mice expressing DREADD in DCN and CNO during the task (orange); only ellipses and regression line are included for clarity of the graph (**p<0.01 ***p<0.001 Wilcoxon test for difference between groups of residuals, i.e. signed distance of performances to Deming regression line of Control mice). e) same as panel d) for CNO administered after the task, (See Supplementary Table 8 for detailed statistics)

Inhibition of cerebellar nuclei (CN) neurons projecting to the centrolateral thalamus (CL) and to the ventral anterior lateral (VAL) thalamus during and after the training sessions differentially impairs motor learning.

a) scheme of the combined viral injections targeting the CN->CL neurons using a retrograde virus expressing the Cre in the thalamus and a virus inducing Cre-dependent expression of inhibitory DREADD in the CN. Top: schematic of the viral injections. Bottom: GFP fluorescence revealing the site of injection of the CAV viruses. b) comparison of the effect of daily injections before trial 1 of CNO (orange) or Saline (black) in mice described in panel a) (Data represents mean ± S.E.M, n indicates the number of mice; *p<0.05, **p<0.01, ***p<0.001; repeated measure ANOVA Group effect). c) Same as b) for daily CNO injections 30 minutes after trial 7 during the Learning Phase. d) scheme of combined viral injections targeting the CN->VAL neurons. e,f) same as (b,c) for CNO-injected (blue) and control (black) mice obtained as described in panel d. g) same as figure 2c for the Saline-treated groups of panels b-f. h-k) same as figure 2d-e for the different groups receiving injections CNO during or after the task compared to Saline-treated mice. Same color coding as panels b,c and e,f. (**p<0.01 ***p<0.001 Wilcoxon test for difference between groups of residuals i.e. signed distance of performances to Deming regression line of Control mice). CN: cerebellar nuclei, VAL: ventral anterior lateral thalamus, CL: central lateral thalamus. (See Supplementary Tables 9,10 for detailed statistics)

Reversal of the inhibition of the cerebello-thalamic neurons in the late Maintenance phase yields lasting impairments.

a) Performance of mice with DREADD expression in CL-projecting CN neurons, which learned the task under respectively CNO and Saline treatment (Fig. 3a) during 7 days and then receive respectively Saline and CNO treatment during a Reversal phase (*: p<0.05 repeated-measure ANOVA group effect; Data represented as mean ± S.E.M, n indicates the number of mice). b) change induced by treatment switch on skill levels for the first trial (daily start) and last trial (daily end); values are estimated as in Fig. 2a, and start and end values for day 8 and 9 are averaged for each animal; *p<0.05, **p<0.01, ***p<0.001 paired Wilcoxon test. Boxes represent quartiles and whiskers correspond to range; points are singled as outliers if they deviate more than 1.5 x interquartile range from the nearest quartile. c,d) same as panels a and b for mice with DREADD expression in VAL-projecting CN neurons (Fig. 3d). CN: cerebellar nuclei, VAL: ventral anterior lateral thalamus, CL: central lateral thalamus. (See Supplementary Table 15 for detailed statistics)

Summary of the behavioral findings.

a) Schematic representation of mouse brain CN: Cerebellar Nuclei, CL: centrolateral thalamus, VAL: ventral and anterior thalamus, DLS: dorsolateral striatum, Motor Ctx: motor cortex. b) Summary of the controls exerted by CN-CL and CN-VAL neurons on the skill learning. The CN-CL neurons contribute to online-learning and retrieval/execution of the learned skill. The CN-VAL neurons contribute to consolidate offline the recent learning into a form of consolidated, readily-available, skill; a defect in consolidation in the early phases (first days) cannot be rescued in the late phase. CN-VAL neurons also participate to retrieval/execution in the late phases of learning.

DAPI positive neurons expressing hM4Di-DREADD-mCherry in cerebellar nuclei.

Cerebellar nuclei inhibition did not affect execution and fatigue, locomotion, motor coordination, balance and strength.

a) Footprint patterns were quantitatively assessed for 3 parameters as shown on representative footprint patterns for all experimental groups. Three parameters are represented graphically: linear movement (bottom left), sigma (bottom middle) and alternation coefficient (bottom right). b) tests of strength and coordination. Grid test: latency reflecting the time before falling from the grid. 30 seconds of cut-off of was established as the maximum latency (dotted line on figure). Horizontal bar: latency to cross the horizontal bar (balance beam test) for all experimental groups. Vertical pole: latency to reach home cage in vertical pole test for all experimental groups. c) Locomotor activity (Velocity) in DREADD and non-DREADD (Sham) injected mice after CNO or SAL injection during open-field sessions before (OF1) and after (OF2) rotarod for experimental days 1, 4 and 7. n indicates the number of mice. Boxes represent quartiles and whiskers correspond to range; points are singled as outliers if they deviate more than 1.5 x interquartile range from the nearest quartile. No posthoc significant difference was found for the data presented in this figure (see Supplementary Tables 6-7 for detailed statistics).

Expression of hM4D(Gi)-mCherry in the cerebellar nuclei following CL and VAL injections.

a) left: schematics of the experiment for CN-CL groups, right: distribution of labeled neurons. Identified soma expressing mCherry are indicated by arrow-heads. b) same as a for CN-VAL groups.

Inhibition of CN-CL or CN-VAL by 1mg/kg CNO does not affect execution and fatigue, locomotion, motor coordination, balance and strength.

a) Footprint patterns were quantitatively assessed for 3 parameters as shown on representative footprint patterns (top) for all experimental groups. Three parameters are represented graphically: linear movement (bottom left), sigma (bottom middle and alternation coefficient (bottom right). b) Latency reflecting the time before falling from the grid. 30 seconds of cut-off of was established as the maximum latency (dotted line on figure). c) Latency to cross the horizontal bar (balance beam test) for all experimental groups. d) Latency to reach home cage in vertical pole test for all experimental groups. e) Locomotor activity (Velocity) in DREADD injected mice after CNO or SAL injection during open-fields sessions before (OF1) and after (OF2) rotarod for day 1, 4 and 7 (**p<0.01 paired t-test OF1 vs OF2). f) Latency to fall during fixed speed rotarod (5, 10, 15, 20 r.p.m.) for all experimental groups. One-way repeated measure ANOVA was performed on averaged values for each speed steps in each experimental group followed by a Tukey Posthoc pairwise comparison (*p<0.05, **p<0.01 for CN->VAL group CNO vs SAL). CN, cerebellar nuclei, CL, centrolateral thalamus; VAL, ventral anterior lateral thalamus. n indicates the number of mice. Boxes represent quartiles and whiskers correspond to range; points are singled as outliers if they deviate more than 1.5 x interquartile range from the nearest quartile. (See Supplementary Tables 11-13 for detailed statistics)