Figure 5
Spines affect short-term ionic plasticity following repetitive activation of GABAergic dendritic inputs.
GABAergic synaptic inputs were inserted into the central part of a dendritic cylinder (diameter 1 μm, length 200 μm) and activated (11 dendritic shaft synapses, density 0.5/μm, rise time 0.5 ms, decay time 6 ms, conductance 1 nS). Activity-dependent Cl− accumulation and EGABA shift were recorded as measures of short-term ionic plasticity. (A,B) Changes in Cl− concentration in the middle (A) and at the distal end (B) of the dendrite brought about by synchronous repetitive activation (10 Hz, 30 pulses) of GABAergic inputs. (C,D) Spatial profile of Cl− accumulation and EGABA shift across the length of the dendrite recorded at the end of GABAergic activity (t = 3000 ms). Note that, in the presence of spines, Cl− accumulation and EGABA shift was decreased both in the central as well as in the distal parts of the dendrite indicating a reduction in homosynaptic as well as heterosynaptic ionic plasticity. Similar effects were observed in simulations of stochastic GABAergic activity (Supplementary Fig. 3) and in reconstructed morphologies with realistic branching (Supplementary Fig. 4). In contrast, following adjustments of the total dendrite volume (using a compensatory increase in dendrite diameters of smooth dendrites), spines decrease heterosynaptic spread of ionic plasticity to distal dendritic areas but increase the local, homosynaptic ionic plasticity. The dendrite with 2 spines/μm (or 5 spines/μm) and 1 μm diameter has the same volume as the one with 0 spines/μm and 1.2 μm diameter (or 1.5 μm diameter). (E) Maximum slope of Cl− transients across different locations along the dendrite for the initial 3000 ms.
