Downregulation of cortical inhibition mediates ocular dominance plasticity during the critical period

Abstract

Monocular deprivation (MD) during the critical period (CP) shifts ocular dominance (OD) of cortical responsiveness toward the nondeprived eye. The synaptic mechanisms underlying MD-induced OD plasticity, in particular the contribution of cortical inhibition to the plasticity, have remained unsolved. In this study, using in vivo whole-cell voltage-clamp recordings, we revealed eye-specific excitatory and inhibitory synaptic inputs to layer 4 excitatory neurons in mouse primary visual cortex (V1) at a developmental stage close to the end of CP. We found in normally reared mice that ocular preference is primarily determined by the contralateral bias of excitatory input and that inhibition does not play an active role in shaping OD. MD results in a parallel reduction of excitation and inhibition driven by the deprived eye, while reducing the inhibition but preserving the excitation driven by the nondeprived eye. MD of longer periods causes larger changes in synaptic amplitude than MD of shorter periods. Furthermore, MD resulted in a shortening of onset latencies of synaptic inputs activated by both contralateral and ipsilateral eye stimulation, while the relative temporal relationship between excitation and inhibition driven by the same eye was not significantly affected. Our results suggest that OD plasticity is largely attributed to a reduction of feedforward input representing the deprived eye, and that an unexpected weakening of cortical inhibitory connections accounts for the increased responsiveness to the nondeprived eye.

Figure 1.

Both excitation and inhibition have a contralateral bias in normally reared mice. A, Schematic drawing of the experimental setup. The computer monitor was placed in front of the mouse. One of two eyes was blocked alternately by a sliding eye shutter. Recording was made in the binocular area of V1. B, Left, Synaptic currents evoked by a flash stimulus recorded under different membrane potentials in a neuron. Scale bars: 20 pA, 20 ms. Right, Mean ± SD of currents within a 5 ms time window after the earliest response onset (circle) and around the response peak (triangle), plotted as a function of membrane voltage. Best-fit linear regression lines are shown. The intercept of x-axis indicates the reversal potential. C, Top, Average excitatory and inhibitory currents evoked by contralateral and ipsilateral stimulation in an example cell. Scale bars: excitation (Exc), 0.1 nA; inhibition (Inh), 0.21 nA; 200 ms. A sample noise pattern for flash stimulation is shown. Right inset, The reconstructed morphology of the cell showing that it was a pyramidal neuron located in layer 4 (L4). Scale bar, 50 μm. Bottom, Synaptic conductances derived from the currents. Shading indicates 95% confidence level. Scale bars: 2 nA, 200 ms. D, Synaptic responses to moving sinusoidal gratings in another cell. The three stimulus cycles were marked on top. Scale bars: Exc, 0.1 nA; Inh, 0.14 nA; 2 nS; 300 ms. E, ODI of inhibition versus that of excitation plotted for individual cells. Dash line is the best-fit linear regression line. N = 13 cells (from 13 mice) for noise stimulation and 9 cells (9 mice) for moving grating (MG) stimulation. F, E/I ratio for contralateral and ipsilateral responses. Error bar, SE. G, ODI of excitation, inhibition, as well as VEP response. There is no significant difference between Exc and Inh (p > 0.5, paired t test). Error bar, SE. N = 22 for VEP. H, Onset latencies of excitatory (white) and inhibitory (gray) responses to flash noise stimuli. Error bar, SE. N = 13. ***p < 0.001, paired t test between excitation and inhibition, and between contralateral and ipsilateral responses. I, Onset latency difference between excitation and inhibition (white), or between contralateral and ipsilateral responses (gray).

Figure 2.

MD induced decreases of inhibition for both deprived and nondeprived eyes. A, ODI of VEP and spike responses in normally reared mice (NR, n = 22) and mice following MD3D (n = 13), MD6D (n = 28), and MD2W (n = 22). Error bar, SE. *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA and post hoc test (same for the rest of figures). B, Average VEP amplitudes in four groups of mice. C, Average spike rates. D, Top, Average excitatory and inhibitory currents to flash noise stimuli in an example cell after MD6D. Bottom, Synaptic conductances derived from the currents. Scale bars: top, 0.1 nA; bottom, 1 nS; 200 ms. E, ODI of excitation and inhibition in control mice (white, n = 13) and mice following MD3D (gray, n = 11 cells from 11 mice), MD6D (black, n = 15 cells, 14 mice) and MD2W (red, n = 15 cells, 15 mice). F, Average peak excitatory and inhibitory conductances evoked by flash noise stimuli. G, E/I ratio for contralateral and ipsilateral stimulation. H, Onset latencies of excitatory and inhibitory responses driven contralaterally and ipsilaterally. Gray columns represent combined data of MD3D and MD6D.