Animals
Male Sprague–Dawley rats (Chiyoda Kaihatsu Co., Tokyo, Japan) and adult male Thy1-eYFP (H-line) mice (8-12 weeks of age) were housed in individual plastic cages lined with wood chips (length: 25 cm; width: 40 cm; height: 25 cm). The animals were maintained at a constant temperature of 23°C ± 1°C under a 12-h light/dark cycle (lights on: 8:00 am to 8:00 pm), with water and food provided ad libitum. We used 75 rats for electrophysiology (AMPA/NMDA ratio analysis, N = 16; miniature recordings, N = 24; paired-pulse ratio analysis, N = 20; current-clamp recordings, N = 15). Twelve transgenic mice were used for live imaging studies. All animal housing and surgical procedures were approved by the Institutional Animal Care and Use Committee of Yamaguchi University Graduate School of Medicine and comply with the Guide for the Care and Use of Laboratory Animals published by the National Institute of Health (NIH Publication No. 85-23, revised 1996).
Motor Learning Test
To evaluate the changes in motor skills, we conducted the rotor rod test (ENV577; Med Associates Inc., St. Albans, VT, USA) with 4-week-old rats and 8-10-week-old mice. Animals were assigned to either the naive (untrained), 1-day training (1-day trained), or 2-days training (2-days trained) group (Fig. 1). For each test, the rats were allowed 10 attempts with a 30-s time interval between the trials. The rotor rod was set to increase from 4 to 40 rpm over 5 min, and the rod-riding duration was recorded (25, 26).
Electrophysiology
Thirty minutes after their final rotor rod test, the 1-day or 2-days trained rats were anesthetized with overdose of pentobarbital (400 mg/kg, ip) or ketamine (750 mg/kg, ip) / xylazine (60 mg/kg, ip). After confirming that the animal has no pain reaction, acute brain slices were prepared, and whole-cell recordings were performed as previously described (27). Briefly, the brains were quickly perfused with ice-cold dissection buffer (25.0 mM NaHCO3, 1.25 mM NaH2PO4, 2.5 mM KCl, 0.5 mM CaCl2, 7.0 mM MgCl2, 25.0 mM glucose, 110.0 mM choline chloride, 3.10 mM pyruvic acid, and 11.6 mM ascorbic acid) and gassed with 5% CO2/95% O2. The coronal brain samples were cut into 350-μm slices with a Leica vibratome (VT-1200; Leica Biosystems, Nussloch, Germany) in dissection buffer and transferred into a physiological solution (22°C–25°C; 118 mM NaCl, 2.5 mM KCl, 26 mM NaHCO3, 1 mM NaH2PO4, 10 mM glucose, 4 mM MgCl2, and 4 mM CaCl2, pH 7.4, gassed with 5% CO2/95% O2). We maintained three to four slices from each rat and then selected one to two brain slices for patch recordings based on the brain atlas (28). Glass electrodes were made with a horizontal puller (Model P97; Sutter Instrument, Novato, CA, USA). Whole-cell recordings were obtained from the pyramidal neurons of the M1 layer V using an Axopatch 1D amplifier (Molecular Devices Inc., San Jose, CA, USA). The recordings were digitalized using a Digidata 1440 AD board, recorded at 5 kHz, and analyzed offline using the pCLAMP 10 software (Molecular Devices).
The cell recording coordinates were approximately 1.2 mm anterior to the bregma, 2.0 mm lateral to the midline, and 0.2–0.3 mm below the dura surface, the region corresponding to the M1 forelimb representation (27).
Current-Clamp Recordings
The recording chamber was perfused with the physiological solution of 22°C–25°C. For current-clamp recordings, the pipettes were filled with 130 mM K-Gluconate, 5 mM KCl, 10 mM HEPES, 2.5 mM MgCl2, 4 mM Na2ATP, 0.4 mM Na3GTP, 10 mM Na-phosphocreatine, and 0.6 mM EGTA at pH 7.25. The liquid junction potential was not corrected. During the 300-ms current injections from -100 to +550 pA, we counted the numbers of evoked spikes. To determine the threshold for each neuron, we identified the minimum voltage for the action potential induction. The afterhyperpolarization (AHP) amplitude was evaluated by measuring the voltage at the spike initiation and the lowest voltage during AHP (29). We determined a transient component following the action potential as fast AHP (fAHP), while the late component within 100 ms as medium AHP (mAHP).
Voltage-Clamp Recordings
For the AMPA/NMDA ratio analysis, we added 0.1 mM picrotoxin to block GABAA receptor-mediated inhibition and 4 μM 2-chloroadenosine to stabilize the evoked neuronal responses (30). Patch recording pipettes (4–7 MΩ) were filled with an intracellular solution (115 mM cesium methanesulfonate, 20 mM CsCl, 10 mM HEPES, 2.5 mM MgCl2, 4 mM Na2ATP, 0.4 mM Na3GTP, 10 mM sodium phosphocreatine, and 0.6 mM EGTA, pH 7.25). A bipolar tungsten stimulating electrode (Unique Medical Co., Ltd., Tokyo, Japan) was positioned in the M1 region 200–300 µm laterally from the recorded neurons. The stimulus intensity was increased from 0.4 to 0.7 mA until a synaptic response with an amplitude of > 10 pA was recorded. The AMPA/NMDA ratio was calculated as the ratio of the peak current at -60 mV to the current at +40 mV 150 ms after the stimulus onset (50–100 traces were averaged for each holding potential).
Paired-Pulse Stimulation
To analyze the presynaptic plasticity at excitatory synapses, we added 0.1 mM picrotoxin and 4 μM 2-chloroadenosine to the perfusate and performed paired-pulse stimulation at -60 mV. To analyze the presynaptic plasticity at inhibitory synapse, we added, 0.1 mM APV, and 4 μM 2-chloroadenosine to the perfuse and performed paired-pulse stimulation at +15 mV. For the paired-pulse ratio evaluation, we recorded 50 to 100 sweeps with paired stimuli at 100 ms intervals, and calculated the second-to-the-first amplitude ratio (31-33).
Miniature Recordings
We used modified intracellular solution for the miniature recordings to adjust the reversal potential of the GABAA receptor response (127.5 mM cesium methanesulfonate, 7.5 mM CsCl, 10 mM HEPES, 2.5 mM MgCl2, 4 mM Na2ATP, 0.4 mM Na3GTP, 10 mM sodium phosphocreatine, 0.6 mM EGTA, pH 7.25). Moreover, we added 0.5 µM tetrodotoxin (TTX; Wako Pure Chemical Industries Ltd., Osaka, Japan) and 0.1 mM APV (Sigma-Aldrich Co., Tokyo, Japan) to the perfusate to block both the action potentials and the NMDA receptor-mediated excitatory postsynaptic currents (EPSCs). The voltage was clamped at -60 mV for mEPSC recording and at +15 mV for the miniature IPSC (mIPSC) recording. We analyzed mean amplitude and the frequency of the mEPSCs/mIPSCs above 10 pA (31, 34). To calculate the excitation–inhibition (E/I) balance, mean mEPSC amplitude or the frequency was divided by the corresponding mean mIPSC amplitude or the frequency in each neuron (25, 32). After the recording, we confirmed that mEPSCs and mIPSCs were completely abolished by 10 µM CNQX (Sigma) and 10 µM bicuculline methiodide (Sigma), respectively.
Two-Photon Microscopy
Adult, 8–12-week-old male Thy1-eYFP (H-line) mice were used for the in vivo imaging. Animal preparation was conducted as previously described (35). For the high-resolution imaging of the synaptic structures in the cortex of living mice, the overlying opaque skull bone should be partially removed to create a cranial window, namely, the “open-skull” glass window. In the open-skull preparation, a piece of the cranial bone is removed (approximately 4 mm in diameter), leaving the dura intact, and then, the exposed brain is covered with a circular glass coverslip (f = 4 mm; Matsunami Glass Ind., Ltd, Osaka, Japan). To fix the head of the mice, we attached the square stainless-steel frame (CF-10, Narishige) on the bone. After the motor training, the mice were anesthetized with ketamine (130 mg/kg, ip) / xylazine (10 mg/kg, ip). These image stacks were then obtained using two-photon laser microscopy, customized for in vivo imaging (A1 MP+/A1R MP+ multiphoton confocal microscope, Nikon, Tokyo, Japan) with a 25× water objective lens (NA 1.10, Nikon). These stacks were taken from the brain surface, an area covering 500 × 500 μm (2048 × 2048 pixels, 0.12 μm/pixel) and comprised over 100 optical sections of layer I with 0.7 μm Z-steps at a zoom of 2. All fluorescence signals under the wavelength of 480 nm were detected via non-descanned detector. During taking the images, we maintained the illumination does not saturate.
We used the Nikon image analysis software (NIS elements AR) to evaluate the spine volumes manually. A median filter with a kernel size of 3 was applied to reduce noise when drawing the ROI for each spine. Image contrast was adjusted, but care was taken to avoid contrast saturation. Images were selected and analyzed at the depth at which each spine was maximally visible. Although not blinded, clear spine images were randomly selected for consecutive and stable analysis. After selecting a single image plane, a ROI polygon were manually drawn around each spine. The spine head radius (R) was calculated from the spine area (square root of area/π). Then, the spine head volume (V) was calculated as a spherical volume according to the following formula:
V = 4/3 × π × R3
We compared the volume of traced spines among the untrained, 1-day trained, and 2-days trained mice. Control mice were monitored the spines for 3 successive days without training. Based on the temporal change, the spines were categorized into four groups: type 1 (consistent increase), type 2 (consistent decrease), type 3 (decrease after increase), and type 4 (increase after decrease). Spine formation/elimination was defined as the percentage of formation/eliminated spines compared with the observed untrained ones. In the control group, they were compared with the first observation.
Self-entropy analysis
As we previously reported, we calculated the diversity of spine volume by measuring the distribution of the appearance probability (26, 32). We used a standard spreadsheet software (Excel 2010, Microsoft Co., Redmond, WA, USA) to calculate the self-entropy per single spine. First, we identified the distribution of the appearance probability before the task, followed by the analysis of appearance probability of all tracked spines. Then, we determined the distribution of the appearance probability of spine volumes using one-dimensional kernel density analysis. Let X1, X2,…, Xn denote a sample of size n from real observations. The kernel density estimates of P at the point x is given by the following equation:
where K is a smooth function called the Gaussian kernel function and h > 0 is the smoothing bandwidth that controls the amount of smoothing. We chose Silverman’s reference bandwidth or Silverman’s rule of thumb (36, 37),
which is given by the following equation:
h = 0.9 An−1/5
where A = min (standard deviation, interquartile range/1.34). By using the distribution of spine volumes before training, we identified the distribution of the appearance probability at any point. Subsequently, we calculated the appearance probability at selected points. All data points for probability in untrained and trained mice were converted to self-entropy (bit) using the Shannon entropy concept, as defined in the Information Theory (38). Each self-entropy of a single spine was plotted: e.g., a point with a high appearance probability (around the mean level of spine volume) indicated a low self-entropy, whereas a point with a very rare probability (deviated extra-large spine) indicated a high self-entropy (25).
Statistical Analysis
To determine motor skill development, data on the latency using one-way ANOVA with repeated measures, followed by post hoc analysis with Fisher’s protected least significant difference (PLSD) test, where the variable was trial.
To evaluate the correlations among the motor performance and mean cell parameters (both current and voltage clamp data), we averaged the latency in each training day to calculate Pearson’s correlation coefficient. Because the variance differs for each parameter, the data were transformed by log (1 + x) prior to the statistical analysis (39). For the correlations among the motor performance and spine parameters, we averaged the latency in each training day to calculate Spearman’s correlation coefficient. P-values < 0.05 were considered statistically significant.