hide
Free keywords:
-
Abstract:
Interneurons (INs) are thought to play a large role in the shaping of neuronal activity during sensory input as well as during spontaneous activity. Despite this important function, almost all of electrophysiological recordings have been performed in vitro and functional data has mainly stemmed from paired recordings in the acute slice. Using 2-photon microscopy in combination with whole-cell patch-clamp recordings we were able to reliably target INs in layers 1 and 2 of the rat somatosensory cortex, in vivo. Intrinsic optical signal imaging was used to find the cortical column primary activated by a particular whisker. Post hoc staining and reconstruction was used to confirm the cell type and the position of the recorded neurons.
We recorded from both INs (n = 30) and pyramidal neurons (PNs) (n = 15) in layers 1 and 2 of barrel cortex (average depth +/- SD: -62 +/- 35 μm and 144 +/- 50 μm). INs showed significantly higher rates of spontaneously occurring action potentials than PNs (1.12 +/- 0.06 Hz vs. 0.25 +/- 0.08 Hz, for INs (n=20) and PNs (n=15); p<0.0015) while the frequency of Up states was the same in both cell types (1.92 +/-0.09 Hz and 1.97 +/-0.09 Hz). Sensory stimulation caused subthreshold membrane potential deflections in all neurons recorded. The latency from whisker deflection onset to 10% EPSP amplitude was 8.89+/-0.33ms for INs and 10.02 +/- 0.58ms for PNs. The time course of the subthreshold responses differed strongly between cells. ~60% of the INs showed an early PSP peak, on average at 17.4 +/- 2.4 ms, and ~40% showed a later PSP peak at 77 +/- 33ms. Subthreshold responses in INs often also showed hyperpolarizing components that became visible at depolarized potentials around -55 to -50mV. Suprathreshold responses ranged from 0 to 1.21 APs/stim/100ms for INs and from 0 to 0.45 APs/stim/100ms for PNs. All INs together responded on average with 0.2+/-0.06 APs/stim/100ms and PNs with 0.09 +/-0.14 APs/stim/100ms. Although INs on average responded with more than twice as many APs than PNs, this was mainly due to a subpopulation of ~15% of the INs responding with very high reliability (0.62 +/- 0.13 APs/stim/100ms). This shows that a small group of INs is reliably activated during sensory input, suggesting that prevailing inhibition rather than a lack of excitation prevents PNs in L2/3 from responding stronger.
