The pEPSP sublinearity could be observed for just under 2 quanta (∼10% sublinearity; Figure 5C, inset, arrow). This sublinearity is less than predicted by simulations (18%, Figure 5C, inset, solid black line), possibly due to the sublinearity of single quantal EPSPs, which simulations predict to be 10%. Voltage-dependent conductances, in particular those mediated by NMDARs and Ca2+ channels, can produce supralinear summation of synaptic inputs (Branco and Häusser, 2011, Cash and Yuste, 1999, Margulis and Tang, 1998 and Urban and Barrionuevo, 1998), whereas K+ channels can produce sublinear summation (Cash and Yuste, 1999, Hu et al., 2010, Margulis

and Tang, 1998 and Urban and Barrionuevo, 1998). In SCs, see more the synaptic input-output relationships remained sublinear in presence of NMDAR, Na+, Ca2+, K+ and HCN channel blockers (Figures 5D and 5E), a condition in which nearly all voltage-dependent conductances are blocked (Figure S5). KU-57788 chemical structure These data and simulations demonstrate that, for synaptic depolarizations induced by up to 15 simultaneously evoked quanta, sublinear dendritic integration in SCs is determined largely by its passive cable properties. Thus far, experimental and modeling results indicate that larger synaptic conductances will produce larger sublinearities (Figure 5C). We therefore tested

the hypothesis that, during paired-pulse facilitation, a sublinear “readout” of the second, potentiated synaptic conductance could underlie the distance-dependent reduction in EPSC PPR (Figure 1). We repeated the PPR stimulation protocol (Figure 1) in presence of submaximal concentrations of a noncompetitive AMPAR antagonist (GYKI 53655 or 53784,

3–9 μM; Paternain et al., 1995). We reasoned that a reduction in synaptic conductance would reduce local depolarization and hence minimize the sublinear report of the facilitated synaptic conductance. Indeed, when EPSC amplitudes were reduced by more than 75% the difference between dendritic and somatic PPRs was no longer observed (Figure 6A). To confirm that synaptic currents were mediated solely by AMPARs, EPSCs were entirely blocked by a saturating concentration of GYKI (40 μM; data not shown). Also, the distance dependence of PPR was not not affected by blockade of voltage-dependent Na+, K+, Ca2+, HCN channels, or mGluRs (Figures S6A–S6D), further supporting a passive cable mechanism. These data show that, although the paired-pulse facilitation is mediated through a presynaptic mechanism, the distance-dependent gradient of short-term plasticity results from a postsynaptic sublinear “readout” of synaptic conductances. These results were confirmed by simulations showing that passive cable properties are sufficient to produce a distance-dependent decrease in PPR (Figure 6B). A conductance ratio of 2.25 produced a simulated EPSC PPR of 2.