Layer-specific sensory processing impairment in the primary somatosensory cortex after motor cortex infarction

Abstract

Primary motor cortex (M1) infarctions sometimes cause sensory impairment. Because sensory signals play a vital role in motor control, sensory impairment compromises the recovery and rehabilitation of motor disability. However, the neural mechanism of the sensory impairment is poorly understood. We show that sensory processing in mouse primary somatosensory cortex (S1) was impaired in the acute phase of M1 infarctions and recovered in a layer-specific manner in the subacute phase. This layer-dependent recovery process and the anatomical connection pattern from M1 to S1 suggested that functional connectivity from M1 to S1 plays a key role in the sensory processing impairment. A simulation study demonstrated that the loss of inhibition from M1 to S1 in the acute phase of M1 infarctions could impair sensory processing in S1, and compensation for the inhibition could recover the temporal coding. Consistently, the optogenetic activation of M1 suppressed the sustained response in S1. Taken together, we revealed how focal stroke in M1 alters the cortical network activity of sensory processing, in which inhibitory input from M1 to S1 may be involved.

Figure 1

Laminar patterns of vM1 axonal innervations to vS1 and experimental timeline of the vM1 photothrombotic infarction model. (A), Anterograde tracer injection into the vibrissa primary motor cortex (vM1). Scale bar, 1 mm. D, dorsal; L, lateral. (B), Labelled vM1 axons in the vibrissa primary somatosensory cortex (vS1). The pixel intensity of the axon signals was normalised to the peak value (green line). The signals of retrogradely labelled L2/3 and L5 neurons (arrowheads) were extracted from this measurement. Scale bar, 100 μm. (C), Laminar distribution of vM1 axons in vS1. Axons intensely innervated L1 and L5b, but sparsely innervated L4 compared to the mean of all layers (*P < 0.05, one sample t-test; 3 mice). (D), vM1 axons selectively innervate L1 and L5b of vS1. (E), vM1 infarction model made by the local irradiation of green light. Electrophysiological recordings from vS1 were performed at POD3 and POD14. Right, The infarction site was identified by cytochrome oxidase staining (1.4 mm anterior to the bregma, POD3). Scale bar, 1 mm; D, dorsal; L, lateral. (F), The means of the largest areas of infarction were 1.52 ± 0.26 mm² at POD3 and 0.74 ± 0.08 mm² at POD14 (P = 0.02, two sample t-test). Error bars are defined as SEM.

Figure 2

vM1 infarction disturbed temporal coding in vS1. (A), The whisker stimulation and 16-channel extracellular recording setup. Left, Directions of the whisker deflection (double-edged arrow) and a trace of the whisker position. Right, Electrolytic lesions at both ends of the recording sites. Scale bar, 100 μm. (B), Examples of peristimulus time histograms (PSTHs) of multiunit activity (MUA) evoked by whisker deflections in L2/3 (upper) and L5b (lower) from sham, POD3, and POD14 mice. MUA was classified into onset (blue area, 0–30 ms after the deflection onset) and sustained (red area, 30–180 ms after the deflection onset) responses. 5 ms/bin. C, Temporal coding index (TCI, left, black dots; see Methods) and spontaneous MUA (right, green dots) in L2/3 (upper) and L5b (lower) from sham (3 mice), POD3 (5 mice) and POD14 (4 mice). ***P < 0.001; **P < 0.01; Tukey’s honestly significant difference test. ns, not significant.

Figure 3

Simulation and CSD analysis indicate inhibition from vM1 to vS1. (A), Sensory-evoked responses simultaneously recorded from both vS1 and vM1. Recorded vM1 responses (upper, grey bars) and recorded vS1 L5b responses (lower, black bars) to whisker deflections (onset was set to zero). (B), Left, A schematic diagram for the simulation of vM1 and vS1 responses using an integrate-and-fire model from recorded vS1 responses. The simulated vM1 responses (blue line) positively correlated with the recorded vM1 responses (grey, the same as in A) (r = 0.68, P < 0.001). The simulated S1 responses (red line) negatively correlated with the recorded vS1 responses (black bars, the same as in A) (r = −0.52, P < 0.001). (C), The excitatory synaptic inputs in vS1 were revealed by CSD analysis as current sinks (red in the colourmap). Note, L5b of the CSD (dotted area in the colourmap and black line in the voltage graph) positively correlated with the simulated S1 responses in B (red line, r = 0.67, P < 0.001). (D), A schematic diagram of a network model from the simulation and CSD analysis. The excitatory synaptic input to pyramidal neurons in vS1 (black arrow) was observed as a current sink in the CSD. The input from inhibitory interneurons in vS1 was the simulated vS1 response (red arrow).

Figure 4

Application of inhibitory input to vM1 infarction can mimic temporal coding in sham animals. (A), The loss of excitatory inputs to vS1 L5b (dotted area) was visualised by CSD analysis at POD3. (B), Upper, An example of the recorded vS1 L2/3 responses at POD3. Lower, Simulated inhibition from vM1 (cyan line) suppressed the sustained responses. (C), TCI was recovered to the sham level in simulated vS1 responses: in L2/3, TCI at POD3 of recorded (0.85 ± 0.03) and simulated (1.00 ± 0.03); in L5b, TCI at POD3 of recorded (0.87 ± 0.03) and simulated (1.06 ± 0.01); in L5b, TCI at POD14 of recorded (0.94 ± 0.03) and simulated (1.05 ± 0.01). Each value was normalised to the level of sham mice. *P < 0.05; **P < 0.01; Dunnett’s test was used to compare with the sham level. ns, not significant. To calculate normalised TCI, the response to the first stimulus (from the stimulus onset until 180 ms) was used.

Figure 5

Optogenetic vM1 activation suppresses the sustained response to whisker stimulus in vS1. (A), Upper, schematic of the optogenetic activation. ChR2 was expressed in vM1 by virus-vector injection. The neurons in vM1 were activated by 473 nm light stimulus. Lower, an example of ChR2 expression in vM1 based on ChR2-EYFP (green) and vGluT2 (grey) staining. Scale bar, 1 mm. (B), An example of MUA in vS1 in response to whisker stimulus (piezo trace, orange) and 473 nm ChR2 activation in vM1 for 5 ms. The sustained responses at points (a–c) were used in the graphs seen in C and D. (C), The sustained responses at (b) (shown in B) were significantly suppressed compared with those at (a,c) (P < 0.001, N = 28 from four mice; Wilcoxon signed-rank test). (D), The sustained responses at (b) without ChR2 activation were not different from those at (a,c) (P = 0.48, N = 29 from four mice; Wilcoxon signed-rank test). The MUA of zero spikes at (a–c) was excluded from the analysis (N = 4 from C, N = 3 from D).