Table 1.
Score assessment scale for scoring swim ability in mice.
Fig 1.
Swim motor training accelerates the acquisition of the four-limb motor pattern in P3 trained mice.
A. Schematic representation of the experimental protocol. For each training (1 to 4), trained mice realized five successive swimming sessions (S1 to S5) of 15 seconds separated by 45 second breaks twice a day (9 AM and 5 pm) on post-natal day 1 (P1) and P2. Untrained mice were tested only once either at P1 9 AM or P1 5 pm or P2 9 AM or P2 5 pm during two swimming sessions of 7 seconds each (S1 green box and S2 red box). The greatest number of limbs used during the first 7 seconds of each session was scored (motor score, Table 1 P1-P3) B. Histograms of the percentage of untrained (trial, black filles bars) and trained (training, purple filled bars) animals according to the motor score assessed during the first (bars with green border; B1) and second swimming session (bars with red border; B2). The number of animals included in the analysis appears on histogram bars. C. Percentage of trained mice with a motor score of 4 (4 four limbs used to swim) during the different swim trainings. The total number of animals with a motor score of 4 appears on histogram bars. D. Violin plots of the time to right (left panel) during the righting test and time until abandon (right panel) in untrained (black, n = 36 pups) and trained (purple, n = 39 pups) mice. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 two-way ANOVA analysis, followed by uncorrected Fisher’s LSD post-tests in B and C and Mann-Whitney tests in D. Underlying data can be found in the S1 Data Sheet.
Fig 2.
Differential gene expression in the LMC of P3 untrained and trained mice.
A. Volcano plot showing RNA-seq data of downregulated (purple open circles) and upregulated (purple filled circles) genes, using a 2-fold change (FC) and a p-value (< 0.05) as selection criteria, indicated by dashed lines. Bar plots showing significant fold changes (log2) in the expression profiles of transcripts encoding transcription factors (B), homeobox and apoptosis-related proteins (C), as well as ion channels, synaptic, and neurotransmission-related proteins (D). Kcna3: potassium voltage-gated channel, shaker-related subfamily, member 3. Kcnc3: potassium voltage gated channel, Shaw-related subfamily, member 3. Kcne3: potassium voltage-gated channel, Isk-related subfamily, gene 3. Kcnh3: potassium voltage-gated channel, subfamily H (eag-related), member 3. Kcng4: potassium voltage-gated channel, subfamily G, member 4. Kcnj12: potassium inwardly-rectifying channel, subfamily J, member 12. Kcnk12: potassium channel, subfamily K, member 12. Kctd12b: potassium channel tetramerization domain containing 12b. Hcn2: hyperpolarization-activated, cyclic nucleotide-gated K+ 2. Scn1b: sodium channel, voltage-gated, type I, beta. Piezo1: piezo-type mechanosensitive ion channel component 1. Pkd2l1: polycystic kidney disease 2-like 1. Trpv4: transient receptor potential cation channel, subfamily V, member 4. Camkv: CaM kinase-like vesicle-associated. Sntb2: syntrophin, basic 2. Syndig1l: synapse differentiation inducing 1 like. Rapsn: receptor-associated protein of the synapse. Grik5: glutamate receptor, ionotropic, kainate 5 (gamma 2). Nos1ap: nitric oxide synthase 1 (neuronal) adaptor protein. Adra1d: adrenergic receptor, alpha 1d. Cckar: cholecystokinin A receptor. Sstr3: somatostatin receptor 3. Trh: thyrotropin releasing hormone. Npy: neuropeptide Y. Gad2: glutamic acid decarboxylase 2. Ascl2: achaete-scute family bHLH transcription factor 2. Atf5: activating transcription factor 5. Bex4: brain expressed X-linked 4. Brf1: BRF1, RNA polymerase III transcription initiation factor 90 kDa subunit. Ddx4: DEAD box helicase 4. E2f1: E2F transcription factor 1. E2f8: E2F transcription factor 8. Foxp2: forkhead box P2. Foxq1: forkhead box Q1. Gdf5: growth differentiation factor 5. Gfra4: glial cell line derived neurotrophic factor family receptor alpha 4. Igfbp2: insulin-like growth factor binding protein 2. Igfbp7: insulin-like growth factor binding protein 7. Klf14: Kruppel-like transcription factor 14. Neurod6: neurogenic differentiation 6. Notch4: notch 4. Nupr1: nuclear protein transcription regulator 1. Reg3b: regenerating islet-derived 3 beta. Rgmb: repulsive guidance molecule family member B. Sp8: trans-acting transcription factor 8. Spdef: SAM pointed domain containing ets transcription factor. Tcfl5: transcription factor-like 5 (basic helix-loop-helix). Tfap2b: transcription factor AP-2 beta. Wif1: Wnt inhibitory factor 1. Barhl1: BarH like homeobox 1. Barhl2: BarH like homeobox 2. Dlx3: distal-less homeobox 3. Hhex: hematopoietically expressed homeobox. Hdx: highly divergent homeobox. Hoxb6: homeobox B6. Hoxb7: homeobox B7. Hoxc5: homeobox C5. Hoxc11: homeobox C11. Hoxd11: homeobox D11. Hopx: HOP homeobox. Lhx5: LIM homeobox protein 5. Otp: orthopedia homeobox. Vsx2: visual system homeobox 2. Aven: apoptosis, caspase activation inhibitor. Casp6: caspase 6. Pawr: PRKC, apoptosis, WT1, regulator. Raw data are available under the following link: https://doi.org/10.57745/WE0VK8.
Table 2.
Training impact on electrical properties of P3 lumbar MNs. Rin: input resistance. Rheobase: lowest intensity of current injected in MNs to elicit an action potential (AP). AP threshold: voltage measured at the foot of the AP. AP amplitude: measured between the resting membrane potential and the AP peak. AP half width: time spent by the potential > 50% of the AP maximum amplitude. AP rise time: time spent by the potential between 10% and 90% of the AP maximum amplitude. AP half decay time: time spent by the potential between the AP maximum amplitude and the 50% decreasing amplitude. ADP: after-depolarization potential. All values are means ± SEM. The number of MNs recorded is indicated between brackets. ns: no significantly different. P-values are obtained from Mann-Whitney test or T-test statistical analysis, depending of the normal distribution of data sets. Underlying data can be found in the S1 Data Sheet.
Fig 3.
Training impact on intrinsic membrane properties of lumbar MNs.
A. Representative traces of the after depolarization (ADP) and after hyperpolarization (AHP) expressed after an action potential (A1 upper trace, truncated onto the trace) induced by a short current pulse (bottom trace) in an untrained (black trace) MN and a trained MN (purple trace) held at −60 mV in current clamp condition. Percentage of untrained (black) and trained (purple) MNs expressing ADP or not. *p < 0.05, Chi square test (A2). Violin plots of the absolute values of AHP amplitude (AHP abs ampl) measured in untrained (black) and trained (purple) MNs (A3). Same data representation as in A3 for the AHP half width (A4), the AHP rise time (A5) and the AHP half decay time (AHP HT Decay; A6). N = 58 untrained MNs and n = 52 trained MNs; * significantly different; ***p < 0.001, and ****p < 0.0001, Mann-Whitney tests. B. Representative traces of the firing activity induced in a MN during the application of a 300 and a 400 pA depolarizing current pulse injection (B1). Violin plot of the ƒ/I slope computed in the different MNs tested in untrained (black, n = 20 MNs) and trained (purple, n = 10 MNs) animals (B2). Plot of the normalized AHP amplitude as a function of the mean frequency of spike activity obtained during series of depolarizing current pulses in untrained (black dots) and trained (purple dots) animals. The dashed lines correspond to the linear fitting. * Significantly different; **p < 0.01, Pearson test (B3). C. Representative traces of the three different types of discharge recorded from MNs during triangular ramp current injection (upper panels) and corresponding plots (bottom panels) of the instantaneous firing frequencies as a function of the current injected during the depolarization (filled circles) and the repolarization (open circles) phase (C1). Percentage of MNs expressing the Type 1, 2 or 3 profile of discharge in untrained (black bars, n = 27) and trained (purple bars, n = 16) MNs (C2). Underlying data can be found in the S1 Data Sheet.
Table 3.
Training impact of the lumbar MN discharge properties. ISI: initial interspike interval corresponding of the instantaneous discharge between the first two AP. ISI slope: ISI frequency-current relationship. ESFF: early-state firing frequency representing the mean 3 first ISI. ESFF slope: ESFF frequency-current relationships. Current min: minimum current injected in MNs used for slope analysis of spike-frequency adaptation (SFA) parameters. Current max: maximum current injected in MNs used for slope analysis of SFA parameters. All values are means ± SEM. The number of MNs recorded is indicated between brackets. ns: no significantly different. P-values are obtained from Mann-Whitney test or T test statistical analysis, depending of the normal distribution of data sets. Underlying data can be found in the S1 Data Sheet.
Fig 4.
Training impact on the Kir and IH currents in P3 lumbar MNs.
A. Representative traces of the current recorded during the application of a long voltage ramp from −40 mV to −150 mV in MNs held at −40 mV in control condition (black trace) and in the presence of 500 μm Ba2+ (blue trace) (A1). The mean cord conductance of the current was measured at membrane potentials equally distant from the reversal potential (Erev), Erev+40 and Erev-40. Violin plots of the chord conductances measured in untrained (black, n = 40) and trained (purple, n = 36) MNs (A2). Violin plots of the chord conductance of the current suppressed by Ba2+ in untrained (black, n = 15) and trained (purple, n = 17) MNs (A3). * Significantly different; *p < 0.05, **p < 0.01, ***p < 0.001, two-way ANOVA analysis, followed by uncorrected Fisher’s LSD post-tests. B. Representative Kir2.2 (green) and GAPDH (red) bands from ventral lumbar spinal cords of untrained and trained P3 mice (top panel), with the quantitative results of western blotting analysis (bottom panel). C. Sample membrane current traces obtained in response to negative voltage steps in a MN held at −60 mV in voltage clamp conditions. The IH current was computed by subtracting the instantaneous current (■) from the steady state (☐) (E1). Mean instantaneous I-V curves derived from the current responses generated by a series of voltage steps in untrained (black dots, n = 14) and trained (purple dots, n = 17) MNs (E2). Underlying data can be found in the S1 Data Sheet.
Fig 5.
Training impact on the ADSP expression at VLF-MNs synapses.
A. Schematic diagram of the experimental protocol and sample traces of EPSCs elicited during paired-pulse stimulations of VLF axons (VLF stim, 50 ms interval, black dots) in lumbar MNs held at −60mV before and after VLF-HFS (50 Hz, 2 s, middle trace). B. Pooled data average time courses of normalized VLF-EPSC amplitudes in VLF-MNs synapses expressing LTD (green dots) or LTD (blue dots) in untrained mice. C. Pooled data average time courses of normalized VLF-EPSC amplitudes in VLF-MNs synapses expressing LTD (green dots), STD (blue dots) or no plasticity (red dots) after VLF-HFS in trained mice. D. Percentage of different ADSPs expressed by untrained (bars with black border, n = 25) and trained (bars with purple border, n = 37) MNs. * Significantly different. *p < 0.05, Chi square test. Underlying data can be found in the S1 Data Sheet.
Fig 6.
Training impact on spontaneous network activity in spinal motor networks of P3 mice.
A. Representative traces of spontaneous excitatory post-synaptic currents (sEPSCs) recorded in the presence of strychnine and gabazine in an untrained (black trace) and a trained (purple trace) MN held at −60 mV (VH −60 mV) (A1). Cumulative distributions of sEPSC frequency (A2) and amplitude (A3) recorded in untrained (black dots, n = 20) and trained (purple dots, n = 14) MNs. B. Representative traces of spontaneous inhibitory post-synaptic currents (sIPSCs) recorded in the presence of DNQX and AP5 in an untrained (black trace) and a trained (purple trace) MN held at −90 mV (VH −90 mV). Cumulative distributions of sIPSC frequency (B2) and amplitude (B3) recorded in untrained (black dots, n = 24) and trained (purple dots, n = 11) MNs. * Significantly different. ***p < 0.001, and ****p < 0.0001, Kolmogorov–Smirnov test. Underlying data can be found in the S1 Data Sheet.
Fig 7.
HPLC analysis of noradrenaline (NA), dopamine (DA) and serotonin (5-HT) contents in the ventral lumbar spinal cord of P3 (n = 16 and 17) and P10 (n = 20 and 17) untrained and trained mice, respectively.
A. Representative chromatogram of the monoamine oxidation over time. B. Violin plots of dopamine content in the ventral lumbar spinal cords from P3 and P10 untrained (black) and trained (purple) mice. C. Same as in B for 5-HT content analysis. D. Same as in B for NA content analysis. *p < 0.05, **p < 0.01, two-way ANOVA analysis, followed by uncorrected Fisher’s LSD post-tests. Underlying data can be found in the S1 Data Sheet.
Table 4.
Statistical table of two-way ANOVA analysis for axonal myelination and area, neuromuscular junction (Acetylcholine plaque) parameters, muscle parameters and motor activity quantification considering anatomical position or age and training as main factors. Number of P3 animals used: Myelin analysis: 4 trained and 4 untrained mice; Ach Plaques: 5 trained and 4 untrained mice; Muscles: 4 trained and 5 untrained and Motor activity: 14 trained and 11 untrained mice.
Fig 8.
Impact of short motor training on axonal myelination in the lumbar spinal cord.
A. Representative fluorescence microscopy image of a transverse spinal cord section depicting fluoromyelin labeling (green channel). Magnified views of the white matter in the dorsal commissure (a, red hatched box), dorsal horn (b, white hatched box), ventral commissure (c and d, green hatched boxes), and ventral horn (e and f, blue hatched boxes) (A1). Representative image of raw data and after the detection of fluoromyelin labeling using Fiji (A2). Calibration bars: 20 μm and 5 μm in magnifications B. Violin plots illustrating the thickness of the myelin sheath (B1) and axon area (B2) in the different regions analyzed for trained (purple filled violins) and untrained (unfilled violins) mice. *p < 0.05, **p < 0.01, ****p < 0.0001, two-way ANOVA analysis, followed by uncorrected Fisher’s LSD post-tests. Graph plot of the myelin sheath thickness as a function of the axon area in trained (purple dots) and untrained (black dots) mice (B3). The lines represent the linear fitting. Four trained and 4 untrained mice were used for this analyze. Underlying data can be found in the S1 Data Sheet.
Fig 9.
Neuromuscular junctions after training in P3 hindlimb muscles.
A. Representative confocal images of transverse tibialis anterior section from a P3 mouse, showing neurofilament (NF), α-bungatoxin (α-BTX,) and laminin labeling (A1). Representative image of NMJ detection (upper panel) and 3D reconstruction (lower panel) with Fiji (A2). Calibration bars: 10 μm. B. Violin plots present the volume (B1), area (B2), area/volume ratio (B3), compactness coefficient (B4) and sphericity coefficient (B5) of AchR clusters in the tibialis anterior and posterior muscles for trained (purple) and untrained (black) mice. C. Violin plots of the surface contact between motoneuron terminals and NMJs in trained (purple) and untrained (black) mice. Analyzed muscles were from 5 trained mice and 4 untrained mice. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 two-way ANOVA analysis, followed by uncorrected Fisher’s LSD post-tests. Underlying data can be found in the S1 Data Sheet.
Fig 10.
Impact of training on myosin heavy chain subtypes in P3 hindlimb muscles
A. Cross-section through the hindlimb of a P3 mouse labeled by immunofluorescence for laminin (A1). The tibialis anterior is outlined in blue, the lateral gastrocnemius (Gastroc Lat) in orange, and the medial Gastrocnemius (Gastroc Med) in green. Representative images of the immunofluorescence labeling obtained in the tibialis (upper panel) and lateral gastrocnemius muscles (lower panel) for Embryo Myosin heavy chain (MyHC), MyHC type I BAD5 antibody, MyHC IIA SC71 antibody, and MyHC IIB BF-F3 antibody (A2). Calibration bar: 200 μm. B. Violin plots of the muscle area (B1), fiber area (B2), and fiber density (B3) in the three muscles analyzed in untrained (unfilled violins) and trained (purple-filled violins) P3 mice. C. Violin plots of the percentage of MyHC Embryo (C1), MyHC I (C2), and MyHC IIB in the three muscles analyzed in untrained (unfilled violins) and trained (purple-filled violins) P3 mice. Tibialis and gastrocnemius muscles were isolated from 4 trained mice and 5 untrained mice. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, two-way ANOVA analysis, followed by uncorrected Fisher’s LSD post-tests. Purple asterisks correspond to significant effects associated with training, and black asterisks to significant effects linked to muscle subtype. Underlying data can be found in the S1 Data Sheet.
Fig 11.
Motor and postural development in trained pups during the first two postnatal weeks.
A. Histogram showing the percentage of untrained (black filled bars) and trained (purple filled bars) animals based on the motor score (Table 1 P5-P12) assessed during swimming between postnatal day 5 (P5) and P14 (A1). Percentage of untrained (black filled bars) and trained (purple filled bars) mice able to raise their heads above water during swimming as a function of age (A2). B. Percentage of untrained (black filled bars) and trained (purple filled bars) mouse pups able to raise their heads (B1), shoulders (B2), and pelvis (B3) on a solid surface as a function of age. **p < 0.05 and ****p < 0.0001, Chi-square test. Violin plots showing the time to right during the righting test in untrained (black, n = 11) and trained (purple, n = 14) P7 mice. **p < 0.01, Mann-Whitney test (B4). C. Violin plots showing the duration of motor activity per minute in untrained (black, n = 11) and trained (purple, n = 14) mice as a function of age. **p < 0.01, two-way ANOVA analysis, followed by uncorrected Fisher’s LSD post-tests. Underlying data can be found in the S1 Data Sheet.
Table 5.
Primary antibodies, molecular probes and conjugates used for IHC and WB experiments. MHC: Myosin Heavy Chain. AChR: Acetylcholine Receptor. GAPDH: glyceraldehyde-3-phosphate dehydrogenase.