Figure 6.
High-frequency release from a single synapse can cause both local and spillover desensitization. A, The 400 AMPARs (using the Raman and Trussell Markov scheme) were randomly tiled on the postsynaptic membranes of each of the 5 synapses in our reconstructed fiber. B, A zoomed view of the AMPAR tiling on synapse #4 of the reconstruction. C, The reliability of neurotransmission at synapse #4 was measured in response to the release of 3000 (univesicular release; left) or 9000 (multivesicular release; right) glutamate molecules from the center of the cleft at t = 0 ms across 20 different trials (i.e., different random seeds). The dramatic improvement in the reliability of the responses to multivesicular release is attributable, in part, to the stimuli probing a nonlinear region of the glutamate-receptor activation curve. The CV (SD/mean) was calculated from the peak channel openings across trials. D–F, Four hundred Hertz neurotransmission was simulated at a single synapse by releasing either 3000 (left) or 9000 (right) glutamate molecules from synapse #4 every 2.5 ms while monitoring activation (i.e., the fraction of AMPARs in open states, PO) and desensitization (i.e., the fraction of AMPARs in desensitized states, PDES) at each of the 5 synapses in the cluster. D, In response to 400 Hz signaling, the gain of the local response (i.e., activation) was reduced for events occurring late in the stimulus train, an effect that was exacerbated by a larger quantal content (i.e., MVR) and/or a reduction in the local ECVF (i.e., smooth/modified cluster). Activation traces are normalized to the peak response of the first event in the sequence. E, Underlying this persistent gain change was a buildup of local receptor (i.e., at synapse #4) desensitization that reduced the number of receptors that are available to be activated by subsequent release events. Desensitization is reported here as the fraction of synaptic receptors in desensitized states at synapse #4 over time. F, Although release from a single synapse was rarely sufficient to trigger activation at neighboring synapses in the cluster, transmitter spillover did give rise to significant desensitization at quiescent synapses. Spillover desensitization is reported here as the fraction of all synaptic receptors in desensitized states in the 4 quiescent synapses in the cluster (i.e., synapses 1–3, 5) over time. D–F, Data are averages of 20 trials.