AMPA receptor plasticity of On α RGCs revealed by optogenetic stimulation of inputs

Abstract

Purpose :
We have previously demonstrated a form of NMDA receptor (NMDAR)-dependent plasticity in which light drives rapid insertion of Ca²⁺-permeable AMPA receptors (CP-AMPARs) into synapses between On bipolar and On a retinal ganglion cells (Jones et al, 2012; PMID:22884330). Here we sought to explore how this change in AMPAR expression alters synaptic performance using an optogenetic approach.

Methods :
We crossed a mouse line containing a floxed channelrhodopsin (Ai32) with a Cck^cre mouse line, resulting in the expression of channelrhodopsin in type 6 CBCs. 1 ms illumination with a 460 nm LED drove synaptic currents in On a RGCs that persisted when photoreceptor output was blocked with L-AP4 and ACET. The intensity of the LED was modulated to generate an intensity response (IR) function for each cell. Current voltage (IV) relations were measured with spermine in the pipet solution (500 µM) to block CP-AMPARs in a voltage-dependent manner. IV relations were used to calculate a rectification index (RI) to quantify relative CP-AMPAR expression.

Results :
To activate NMDARs, a series of 30 flashes (1 ms duration, 1 Hz) was presented. IVs measured following the NMDAR activating protocol rectified strongly (RI of 0.38±0.04, n=17), indicating insertion of CP-AMPARs into the synapse. However, if D-APV was present at all times during the experiment, including the NMDAR activating protocol, the IV relation was more linear (RI=0.72±0.07, n=7) and this difference was highly significant (p=0.0008; Wilcoxon rank sum test). Thus activation of a single type of CBC, type 6, is sufficient to induce rapid insertion of CP-AMPAR into the synapse. We next measured the IR function for each cell and fit each with the Hill equation to obtain the light intensity required to evoke a half-maximal response (I_1/2). Plotting the I_1/2 against the RI for each cell revealed a strong correlation between an increased (I_1/2) and CP-AMPAR expression (R=0.70; p=0.005; n=14). Furthermore, preliminary experiments using a paired pulse paradigm demonstrated a decrease in paired pulse depression associated with higher expression of CP-AMPARs, suggesting a decrease in presynaptic release probability.

Conclusions :
We demonstrate a form of NMDAR-dependent AMPAR plasticity that requires only activation of type 6 CBCs. Insertion of CP-AMPARs appears to decrease synaptic performance, perhaps by decreasing release probability via an undetermined mechanism.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.