Mice with increased Mα2 cell excitability display increased PC activity after training and upon retraining.

A) Timeline of experiments showing viral injections, miniscope lens implantation, base plate fixation and pellet-reaching task. GCaMP6f fluorescence and forepaw movements were recorded in naive, learning, trained and retrained sessions. B-C) Schematics (B) and photograph (C) of a mouse with the miniscope connected during prehension. D) Representative miniscope spatial footprints of L5 pyramidal cells after cell detection. E) Ten representative example fluorescence traces from PCs. F) Average number of successful prehensions per minute, separated by session (Naive; learning, Learn.; training, Tr.; retraining, Retr.) and genotype (Controls, ‘-’, black; Chrna2-Cre+, ‘+’, red). G) Representative traces from a control mouse showing the ΔF detrended (detr.) activity of one neuron -1 to 1 second around the prehension onset, averaging all successful (Suc., blue) and failed prehensions (Fail., orange). H) Average amplitude of PC activity represented as mean ±SEM, separated by session (naive, learning, training, retraining), genotype (Controls, ‘-’, black; Chrna2-Cre+, ‘+’, red) and prehension accuracy (S., successful prehensions; F., failed prehensions).

Increased excitation of Mα2 cells is associated with delayed PC activity during motor learning and premature PC activity on task retraining.

A) Illustration of recorded calcium signals from GCaMP6f-expressing PCs. B) Representative examples of average fluorescence traces around prehension onset for four different neurons, showing their normalized fluorescence curve (F detr. over time) represented in a colormap and latency of detected activity peaks (dashed blue line). C) Normalized average fluorescence matrices (Neuron × Time × F) for successful and failed prehensions, for control (top) and Chrna2-Cre+ (bottom) mice, showing temporal activation patterns for each neuron on the retraining session. Neurons were sorted according to their temporal profile for successful prehensions, and follow the same order for failed prehensions. D) PC peak latency for both genotypes at each session before (bottom) or during (top) successful (left) and failed (right) prehensions. All significance bars refer to significant pairwise differences between sessions for controls; and asterisks represent significant pairwise differences between genotypes for a particular session. E-H) Peak latency with a shaded area indicating the peak width during naive, learning, trained, and retrained sessions for each neuron. At each plot, neurons were sorted according to their temporal profile. I-J) PCs response peak width during successful prehensions separated by genotype and session (I), and during prehensions from the training session separated by genotype and prehension accuracy (J). S., successful; F., failed; B., before; D., during prehension; N, naive; L, learning; T, training; R, retraining.

Mice with increased Mα2 cell excitability display assemblies with less prehension-related activity, decreased spatial area and increased resilience.

A) Flowchart of the analysis pipeline. B-C) Number of assemblies and number of cells per assembly for control mice, detected at each session. D-E) Same as B-C, but for Cre+ animals. F) Activity of one representative assembly. Top, rasterized activity over the full recording time of the six neurons forming the assembly. Mid, the average of neuron activity. Bottom, the estimated rasterized activity of the assembly as a unit. Blue and orange shades show the successful and failed prehension windows, respectively. G) Salience of assemblies during the whole 2s window for each session, grouped by genotype and prehension accuracy. H-I) Representative examples of 4 color-coded assemblies detected at session 6 and traced back to session 1 (H), and spatial distribution of detected assemblies at each session for control mice (I). J-K) Same as H-I, but for Cre+ animals. L) Resilience of assemblies for each genotype, for assemblies detected at retraining and followed back to the naive session. Suc., successful; Fail, failed; N, naive; L, learning; R, retraining.

Mice with increased Mα2 cell excitability display increased prehension success rate, increased LFP low theta and gamma power.

A) Timeline of the experiments. Animals were trained to reach for the pellet and after the 7th session, the hM3Dq was injected and the LFP electrodes implanted. B) Prehension success ratio for control (Left, black) and Chrna2-Cre+ (Right, red) mice in the sessions with saline injection (gray shade) and clozapine injection (purple shade). C) Example of time-frequency domain power processing, with three prehensions highlighted (green dashed squares). Top, raw (blue) and high gamma-filtered (orange) LFP traces. Mid, spectrogram from the filtered signal, with highlights for the high gamma band (red square). Bottom, the time-frequency domain power over time, calculated as the average of the previously highlighted spectrogram region. D-H) Relative power for low theta, high theta, full gamma, low gamma and high gamma, respectively. D and E shows the full dataset due to the observed overall effects, while for F-H only isolated effects for session 13 were found. Shades: blue and orange (successful and failed prehensions); magenta (sessions under CZP treatment). Abbreviations: B. or Bef., before prehensions; D. or Dur., during prehensions; S. or Suc., successful prehensions; F. or Fail, failed prehensions.

Apoptosis in Mα2 cells decreased pasta handling performance.

A) Timeline of the experiments. B) Unilateral injection of Cre-dependent TACASP3 decreased the amount of Chrna2-Cre+ cells in the motor cortex by 80.4%. Enlargement of the injected motor cortex (B’), and contralateral non-injected side (B”). Scalebars: 500µm (B) and 200µm (B’ and B”). C) Animal performing the pasta handling task. D-E) Graph of results on pasta drops and time spent in the task. F) Animal performing the hanging wire task. G-H) Graph of results on falls and time spent in the task. I) Single pellet prehension task before (session 7) and after ablation (sessions 8-12) for WT (black) and Chrna2-Cre+ mice (red).

Summary of findings.

A) Table summarizing the effects of increased Mα2 cells excitation on PCs. B) Schematic outline illustrating plasticity parameters. Increased excitation of Mα2 cells resulted in more narrow tuning of layer 5 PCs, a temporal shift of the average population activity both in the motor planning and execution phases, enhanced resilience, reduced physical assembly coverage and rigid salience upon retraining. Dashed triangles mark PCs that are part of the retrained assembly but not the naive assembly, and vice versa. Blue and yellow depict PCs in the naive assembly that are part of another assembly in the retraining session. Salience values are represented by the degree of color in PCs and assembly distributions. C) Mα2 cells influence fine motor movements, where caspase induced ablation of Mα2 cells resulted in decreased dexterity, whereas increased excitation of Mα2 cells resulted in increased accuracy.