Article Figures & Data
Figures
- Figure 1.
ChR2-EYFP expression in M1 projection neurons and retrograde identification of distinct L5 projection neurons in the mouse S1. A, AAV-encoding ChR2-EYFP was injected into the ipsilateral M1 to label their axonal projections in S1. Retrograde fluorescent tracers (CTB-647 or rAAV-Cre) were also injected into M1 or S2 to label subclasses of IT neurons or SC, SP5, or POM to label different subclasses of PT neurons. EYFP expression in M1 axons/terminals (far left image) and retrogradely labeled neurons in L5 were evident 3 weeks postinjection. B, C, Whole-cell recordings (B) and neurobiotin filling (C) of L5 pyramidal neurons not only revealed that labeled cells were healthy and had apical dendrites that terminated in tufts near the pia but also confirmed the physiological and morphological differences between L5 projection classes. See Table 1 for physiological differences. See also Extended Data Figure 1-1.
- Figure 2.
L5 projection neurons in S1 are excited differently by long-range M1 inputs. A–E, Top, Average EPSCs evoked by two optical stimuli (LED, purple arrows, 0.5 ms pulse duration) delivered at a 50 ms interval (20 Hz). Data are shown from L2/3 neurons paired to various L5 IT and PT subtypes, including L5 M1p (A), L5 S2p (B), L5 SCp (C), L5 SP5p (D), and L5 POMp (E). Vertical scale bars, 50 pA (A–E). Bottom, Summary of paired-pulse ratios for a 50 ms interstimulus interval. Asterisks denote a significant difference from responses evoked in control L2/3 RS neurons (M1p: 1.05 ± 0.06, L2/3: 1.33 ± 0.05, n = 14 pairs, 9 mice; p = 0.0073, paired t test; S2p: 0.91 ± 0.07, L2/3: 1.20 ± 0.10, n = 6 pairs, 5 mice; p = 0.00329, paired t test; SCp: 1.32 ± 0.12, L2/3: 1.45 ± 0.06, n = 8 pairs, 4 mice; p = 0.34802, paired t test; SP5p: 1.30 ± 0.10, L2/3: 1.51 ± 0.11, n = 8 pairs, 4 mice; p = 0.18343, Wilcoxon paired signed-rank test; POMp: 1.36 ± 0.05, L2/3: 1.30 ± 0.0.11, n = 11 pairs, 7 mice; p = 1.0, Wilcoxon paired signed-rank test). F–J, Top, Average EPSCs evoked by a 20 Hz train of optical stimuli recorded in a single L5 M1p neuron (F), a L5 S2p neuron (G), a L5 SCp neuron (H), a L5 SP5p neuron (I), and a L5 POMp neuron (J). Vertical scale bars, 100 pA (F–J). Bottom, Summary of EPSC amplitudes plotted as a function of stimulus number within 20 Hz trains for all L2/3–L5 pairs (normalized to first responses; M1p vs L2/3: p = 2.88 × 10−20; S2p vs L2/3: p = 4.37 × 10−10; SCp vs L2/3: p = 0.18018; SP5p vs L2/3: p = 0.24162; POMp vs L2/3: p = 0.09125; two-way ANOVA, stim. 2–10). EPSCs were recorded at −94 mV in voltage clamp, near the reversal for inhibition, and the light intensity for each cell was set to obtain an initial peak of 100–200 pA. K, Comparison of initial EPSC amplitude (normalized to L2/3 response), paired-pulse ratio, and EPSC ratio for the tenth pulse in a 20 Hz train for the IT subclasses M1p and SC2 (EPSP amplitude, p = 0.74689, Mann–Whitney U test; PPR, p = 0.20283, two-sample t test; stim10/stim1, p = 0.96999, two-sample t test). L, Same as K but for the PT subclasses SCp, SP5p, and POMp (EPSP amplitude, p = 0.0.86112, one-way ANOVA; PPR, p = 0.90595, one-way ANOVA; stim10/stim1, p = 0.38276, one-way ANOVA). Values are represented as mean ± SEM.
- Figure 3.
Synaptic responses during repetitive M1 activation are less facilitating for L5 IT than L5 PT neurons. A, L5 IT and PT neurons receive similar strength M1 inputs. EPSCs of both cells were normalized to control L2/3 RS neurons (IT: 14 cells, 11 mice; PT: 14 cells, 10 mice; p = 0.66247, Mann–Whitney U test). Bars represent the mean (A). We normalized the evoked response in a given L5 cell type to the response in the L2/3 neuron to control for the variability in the level of ChR2 expression in different animals. B, Summary of short-term plasticity for all L5 IT (M1p and S2p) and PT neurons (SCp, SP5p, and POMp) to a pair of 20 Hz optical stimuli (left) and a 20 Hz optical stimulus train (right; data combined from Fig. 2; PPR: L5 IT: 1.01 ± 0.05, n = 20 cells, 15 mice; L5 PT: 1.33 ± 0.06, n = 27 cells, 15 mice; p = 1.89 × 10−4, Mann–Whitney U test; train: p = 1.53 × 10−30, two-way ANOVA, stim. 2–10). C, Average EPSCs evoked by a 20 Hz train of optical stimuli recorded in a single L2/3 and L6 M1p neuron labeled with CTB. D, Summary of EPSC amplitudes plotted as a function of stimulus number within 20 Hz trains for CTB-labeled M1p neurons in L2/3 (n = 8 cells, 3 mice), L5 (n = 14 cells, 9 mice), and L6 (n = 9 cells, 4 mice; normalized to first responses; * indicates p < 2.23 × 10−18, two-way ANOVA, stim. 2–10). L5 M1p data from Figure 2. Values are represented as mean ± SEM (B, D).
- Figure 4.
Time course of M1 synaptic responses at L5 PT and IT neurons. A, Left, Average M1-evoked EPSP, scaled to match amplitude, recorded in L5 IT and PT neurons. Right, Summary plots showing the kinetics of M1-evoked EPSPs for both cell types, as measured by the 20–80% rise time (IT: 2.1 ± 0.3 ms; PT: 1.1 ± 0.6 ms; p = 0.00113, Mann–Whitney U test), half-width (IT: 21.1 ± 1.4 ms; PT: 13.2 ± 0.7 ms; p = 1.02 × 10−5, two-sample t test), and decay tau (IT: 24.6 ± 2.4 ms; PT: 15.1 ± 0.9 ms; p = 1.0 × 10−4, two-sample t test, n = 6 IT cells from 3 mice; n = 20 PT cells from 11 mice). B, Left, Average M1-evoked EPSP recorded under control conditions and in the presence of ZD7288 (10 µM) for L5 IT and PT neurons. Right, Summary plots showing the change in decay tau for both cell types (IT control: 18.3 ± 4.2 ms, IT + ZD: 22.2 ± 5.01 ms, n = 5 cells from 3 mice, p = 0.2268, paired t test; PT control: 9.7 ± 0.9 ms, PT + ZD: 20.4 ± 2.4 ms, n = 6 cells from 3 mice, p = 0.00267, paired t test). Values are represented as mean ± SEM.
- Figure 5.
Trains of M1 input excite L5 PT neurons more strongly than IT cells due to short-term facilitation. A, Representative M1 responses and the methods used to calculate the EPSP peak, peak depolarization, and subtracted difference. B, Average EPSPs evoked by a 20 Hz train of optical stimuli recorded in a single L5 IT (left) and PT neuron (right). M1 responses were recorded at −94 mV in current clamp, near the reversal for inhibition, and the light intensity was set to obtain an initial subthreshold EPSP of ∼3 mV. C, Summary plot showing the average EPSP peak and peak depolarization ratio as a function of stimulus number within trains (normalized to the first response) for IT neurons. Note the EPSP peaks decreased during repetitive stimulation, whereas the peak depolarization increased through the train (p = 1.79 × 10−6, two-way ANOVA, stim. 2–10, n = 5 cells from 3 mice). D, Summary plot showing the average EPSP peak and peak depolarization ratio as a function of stimulus number within trains (normalized to the first response) for PT neurons. There is no significant difference between the EPSP peak and the peak depolarization (p = 0.54533, two-way ANOVA, stim. 2–10, n = 24 cells from 11 mice). E, Summary plot showing how much the subtracted difference accounts for the peak depolarization as a function of stimulus number within trains for IT and PT populations (p = 3.91 × 10−64, two-way ANOVA, stim. 2–10, n = 5 IT cells from 3 mice and 24 PT cells from 11 mice). F, Summary plot showing the average peak depolarization in millivolts as a function of stimulus number within trains for IT and PT populations (p = 2.08 × 10−4, two-way ANOVA, stim. 2–10, n = 5 IT cells from 3 mice and 24 PT cells from 11 mice). Values are represented as mean ± SEM.
- Figure 6.
Coupling a short train of M1 synaptic inputs with a single spike at the soma increases the spike output and burst probability in L5 PT neurons. A, B, Subthreshold EPSPs evoked by a single LED stimulus (A: LEDs) or the 5th pulse of a 20 Hz train of stimuli (B: LEDT) for an example L5 PT neuron. The LEDs and the first response in an LEDT only produced a voltage response of 2–3 mV at the soma and never reached the threshold for either an action potential or calcium-mediate action potential. C, A short threshold current injection (typically 5 ms) at the soma (Isoma) evoked a single action potential. D, E, Voltage responses to combining somatic action potential (used in C) with a single EPSP (used in A) or the EPSP evoked by the 5th pulse of a 20 Hz train (used in B) separated by an interval of 3–4 ms between the start of the somatic current injection and that of the light pulse. The mean synaptic latency from the onset of the light was 2.3 ± 0.1 ms (n = 10 cells; 7 mice). F, Group data summarizing the action potential output at the soma during coupling for L5 PT neurons. Coupling a somatic action potential with the 5th pulse of an optical train produced significantly more spikes (* indicates p < 0.006, one-way ANOVA). G–L, Same as A–F for an example L5 IT neuron. There was no difference in spike output at the soma during coupling for L5 IT neurons (p = 0.258, one-way ANOVA) or the number of cells bursting for the M1p cells. The bars in F and L represent the means.
- Figure 7.
Coupling a train of M1 synaptic inputs with a somatic action potential requires precise timing in PT neurons. A, Voltage responses of a POMp neuron to combining a somatic action potential with the EPSP evoked by the 5th pulse of a 20 Hz train separated by different time intervals (−4, −2, 0, 2, 4, 6–8, 9–13, and 13+ ms). The actual timing was based on the peak of the AP and not the onset of the current injection. Optically evoked synapse responses had synaptic delays with short onset latencies (∼2 ms). B, C, Plots show that this cell produced more spikes and had a higher incidence of bursting when the onset of the 5th pulse was 0–6 ms after the somatic action potential.
Tables
- Table 1.
Electrophysiological properties of excitatory projection neurons in L5 of mouse primary somatosensory cortex
L5 intratelencephalic (IT) neurons L5 pyramidal tract (PT) neurons All ITs All PTs p value Test M1p S2p SP5p SCp POMp Cell depth normalized to pia 0.52 ± 0.01 0.54 ± 0.01 0.65 ± 0.00 0.63 ± 0.01 0.62 ± 0.01 0.53 ± 0.01 0.63 ± 0 1.69 × 10−12 MW test RMP (mV) −83.3 ± 1.2 −84.7 ± 1.3 −84.9 ± 0.7 −82.8 ± 1.1 −79.0 ± 0.5 −83.8 ± 0.9 −81.5 ± 0.5 0.02156 MW test Rm (MΩ) 114.0 ± 10.1 87.4 ± 8.4 73.5 ± 4.8 65.4 ± 4.4 78.8 ± 5.5 105.2 ± 7.6 74.0 ± 3.1 1.38 × 10−4 MW test τm (ms) 17.5 ± 1.1 15.2 ± 1.7 13.0 ± 0.6 14.7 ± 1.1 15.4 ± 0.9 16.7 ± 0.9 14.5 ± 0.5 0.31490 t test Cm (pF) 166.0 ± 13.1 174.2 ± 12.9 184.0 ± 10.2 227.2 ± 13.4 212.4 ± 15.4 168.7 ± 9.6 207.3 ± 8.4 0.01140 MW test Rheobase (pA) 100.9 ± 8.1 111.1 ± 11.1 121.1 ± 11.0 89.5 ± 8.2 72.2 ± 6.1 103.7 ± 6.5 89.2 ± 5.1 0.04767 MW test Spike threshold (mV) −58.1 ± 0.8 −59.3 ± 1.1 −61.0 ± 0.5 −62.5 ± 1.0 −63.5 ± 0.5 −58.5 ± 0.6 −62.6 ± 0.4 3.64 × 10−6 MW test Spike amplitude (mV) 90.0 ± 1.5 91.9 ± 1.6 92.5 ± 1.1 91.6 ± 1.3 96.5 ± 1.2 90.6 ± 1.1 94.2 ± 0.7 0.01243 t test Spike half-width (mV) 0.61 ± 0.02 0.69 ± 0.04 0.54 ± 0.01 0.51 ± 0.01 0.57 ± 0.03 0.63 ± 0.02 0.55 ± 0.01 7.69 × 10−6 MW test Max rate of rise (mV*ms) 591.5 ± 20.5 588.5 ± 17.4 632.8 ± 16.3 618.9 ± 19.0 660.2 ± 22.0 590.5 ± 14.7 642.6 ± 12.5 0.02296 t test Max rate of decay (mV*ms) −132.6 ± 5.1 −116.2 ± 11.3 −152.0 ± 5.3 −164.4 ± 4.0 −167.4 ± 5.1 −127.1 ± 5.2 −162.7 ± 3.1 2.73 × 10−7 MW test RMP - threshold (mV) 24.5 ± 0.9 25.4 ± 1.5 23.9 ± 0.8 20.1 ± 1.1 15.6 ± 0.6 24.8 ± 0.8 18.9 ± 0.6 9.55 × 10−7 t test Voltage sag amplitude (mV) −3.3 ± 0.4 −3.4 ± 0.4 −3.4 ± 0.3 −3.4 ± 0.3 4.3 ± 0.3 −3.4 ± 0.3 −3.8 ± 0.2 0.14805 MW test Percent with bursts 0.0 (0/19) 11.1 (1/9) 26.3 (5/19) 36.8 (7/19) 50.0 (18/36) 3.6 (1/28) 40.5 (30/74) (n = 18–19 cells) (n = 9 cells) (n = 19 cells) (n = 16–19 cells) (n = 28–36 cells) (n = 26–28 cells) (n = 62–74 cells) See Materials and Methods for an explanation of how the electrophysiological parameters were defined/measured. Data shown as mean ± SEM. All membrane potentials were corrected for a −14 mV liquid junction potential. Two-sample t test (t test); Mann–Whitney test (MW test).
Figure 1-1
Anatomical characterization of L5 IT neurons labeled with AAVretro. A, Injection schematic showing AAVretro carrying genes for Cre and mCherry was injected unilaterally into M1 of Ai14 mice in vivo at ∼3 weeks of age. B, Two example fluorescent images of live coronal slices (300 μm thick) through S1 from two different Ai14 mice injected in M1 ∼21 days prior with AAVretro.EF1a-mCherry-IRES-Cre. Images show tdTomato expressing S1 neurons following AAVretro injection. Higher magnification images show retrogradely labeled neurons in L2/3 and L6, but very few in L5 (n = 5 mice). Download Figure 1-1, TIF file.
In this issue
