Slow GABA(A) mediated synaptic transmission in rat visual cortex.

BACKGROUND

Previous reports of inhibition in the neocortex suggest that inhibition is mediated predominantly through GABA(A) receptors exhibiting fast kinetics. Within the hippocampus, it has been shown that GABA(A) responses can take the form of either fast or slow response kinetics. Our findings indicate, for the first time, that the neocortex displays synaptic responses with slow GABA(A) receptor mediated inhibitory postsynaptic currents (IPSCs). These IPSCs are kinetically and pharmacologically similar to responses found in the hippocampus, although the anatomical specificity of evoked responses is unique from hippocampus. Spontaneous slow GABA(A) IPSCs were recorded from both pyramidal and inhibitory neurons in rat visual cortex.

RESULTS

GABA(A) slow IPSCs were significantly different from fast responses with respect to rise times and decay time constants, but not amplitudes. Spontaneously occurring GABA(A) slow IPSCs were nearly 100 times less frequent than fast sIPSCs and both were completely abolished by the chloride channel blocker, picrotoxin. The GABA(A) subunit-specific antagonist, furosemide, depressed spontaneous and evoked GABA(A) fast IPSCs, but not slow GABA(A)-mediated IPSCs. Anatomical specificity was evident using minimal stimulation: IPSCs with slow kinetics were evoked predominantly through stimulation of layer 1/2 apical dendritic zones of layer 4 pyramidal neurons and across their basal dendrites, while GABA(A) fast IPSCs were evoked through stimulation throughout the dendritic arborization. Many evoked IPSCs were also composed of a combination of fast and slow IPSC components.

CONCLUSION

GABA(A) slow IPSCs displayed durations that were approximately 4 fold longer than typical GABA(A)fast IPSCs, but shorter than GABA(B)-mediated inhibition. The anatomical and pharmacological specificity of evoked slow IPSCs suggests a unique origin of synaptic input. Incorporating GABA(A) slow IPSCs into computational models of cortical function will help improve our understanding of cortical information processing.