Fig. 4.

Overexpression of the KCC2 co-transporter in cortical principal neurons. a Experimental schematic. b Normalized peak amplitudes of consecutive evoked GABA_A receptor-mediated currents recorded in the transgene-expressing (purple; n = 7) and neighbouring non-transduced (black; n = 6) layer 5 V1 pyramidal neurons in acute brain slices in response to 10 electrical pulses delivered at 10 Hz. c Sample traces show averages of 10 trials from non-transduced and transduced neuron (95% confidence intervals displayed in grey) as well as their recovered immunofluorescence, scale bar: 10 µm. HP holding potential, GFP green fluorescent protein, V-Clamp voltage-clamp. d, e Experimental protocol. The light-sensitive chloride pump halorhodopsin was activated by photostimulation for 1 min while a pyramidal neuron was held at +40 mV to load it with chloride. Scale bar: 20 µm. Raw traces in baseline and after Cl⁻ loading. f Photostimulation to activate halorhodopsin was adjusted to induce photocurrents of a similar peak amplitude in control neurons and in neurons overexpressing KCC2. g Time course of GABA_A receptor-mediated currents recovery from Cl⁻ loading in layer 5 V1 pyramidal neuron expressing halorhodopsin alone (black) and in a cell co-expressing halorhodopsin and KCC2 (purple). Note that the recovery time is three times faster in the KCC2-expressing neuron. h Averaged normalized peak amplitudes of GABA_A receptor-mediated currents evoked by 10 ms GABA puffs (100 µM) in layer 5 V1 pyramidal neurons expressing halorhodopsin alone (black, n = 7) and in cells co-expressing halorhodopsin and KCC2 (purple; n = 7). i Recovery time constants of GABA_A receptor-mediated currents calculated from single exponential fitting (from the first post-loading measure to 2 min recovery). Neurons overexpressing KCC2 show significantly faster recovery than control neurons. Unpaired t-test, *p < 0.05, error bars represent the s.e.m. Source data are provided as a Source Data file