Figure 3.

EPSCs can be subthreshold in NM under physiological conditions. A, Current and conductance clamp were used to stimulate NM neurons with realistic inputs. A1, Current-clamp “EPSCs” (top traces) were varied in amplitude until a spike was evoked (bottom traces) to measure EPSC threshold. A2, A conductance clamp input was used to mimic depolarizing inhibition. Square conductance steps had an onset 5 ms before the EPSC. The reversal potential for the conductance clamp input was −37 mV. Spikes in response to just-threshold inputs show accommodation because of depolarizing inhibition. Voltage traces are offset because of depolarization by the conductance input. B, EPSC threshold was increased by depolarizing conductances in neurons (mean ± SEM; n = 13). C, Bars show actual EPSC amplitude as a function of input frequency [from Brenowitz and Trussell (2001); amplitudes calculated for a membrane potential of −65 mV]. EPSC thresholds from B are indicated by dashed lines. The shaded regions below each threshold indicate EPSC amplitudes that will be subthreshold during such inhibition. At high frequencies (333 Hz), EPSC amplitude was subthreshold during inhibition. D, EPSG (reversal potential, 0 mV) threshold was measured in the multicompartment model under varying levels of depolarizing inhibitory conductance. EPSG threshold increased to a greater extent than EPSC threshold in B. E, EPSC amplitudes shown in C were converted to conductance and illustrated as a function of input frequency. The dashed lines indicate the EPSG thresholds of the multicompartment model from D. EPSGs become subthreshold at lower firing rates and lower levels of inhibition than EPSCs.