Abstract

The GluD2 receptor, a member of the ionotropic glutamate receptor (iGluRs) superfamily, is prominently expressed in excitatory synapses in the cerebellum. GluD2 does not bind glutamate, but D-serine (D-ser) which is involved in regulation of synaptic plasticity in the cerebellum. However, the signaling functions of GluD2 appear to be non-ionotropic as D-ser does not activate the GluD2 ion channel. However, the molecular mechanism by which D-ser binding to GluD2 can participate in cerebellar signaling functions is elusive, but is known to require the intracellular C-terminal domain (CTD).
In the first project, we studied the organization of the CTD relative to the transmembrane domain (TMD) in GluD2 by measuring Förster resonance energy transfer (FRET) between fluorescent proteins (FPs) inserted in the TMD and CTD regions of GluD2 subunits expressed in HEK293 cells. Specifically, by screening of multiple GluD2 subunit constructs that were dual-tagged with cyan (CFP) and yellow (YFP) FP variants in various positions, we found that FRET can occur between CFP and YFP inserted in the intracellular region of the TMD and various positions in the CTD. By determination of FRET efficiency (EFRET) from fluorescence lifetime imaging microscopy (FLIM), we found that EFRET for CFP in the M1-M2 loop varied dependent on the position of the acceptor YFP in the CTD. As EFRET is dependent on the distance between donor and acceptor FP, these results provide insight into the spatial arrangement of the CTD relative to the TMD in GluD2. We furthermore used selected FRET-enabled GluD2 constructs which were expressed in live cells and perfused with different ligands to study potential ligand-induced conformational changes in the CTD using FLIM experiments. We found that application of 7-CKA, which binds to the extracellular orthosteric binding site and captures the ligand-binding domain in an open conformation, increases FRET; indicating that extracellular ligand binding to GluD2 can transmit conformational changes across the membrane to rearrange the CTD.
Next, we studied the protein-protein interactions (PPIs) between the CTD and known interaction partners of GluD2 using bioluminescence resonance energy transfer (BRET). This included generation of both full-length and truncated interaction partners fused with green FP (GFP) and a series of full-length GluD2 and isolated CTD constructs fused with Renilla luciferase 8 (Rluc8). BRET analysis confirmed that either the full-length protein or an isolated domain from seven out of eight tested GluD2 interaction partners was interacting with the isolated GluD2 CTD.
Based on previously suggested phosphorylation sites within the CTD of GluD2, the dephospho- and phosphomimic mutations Ala and Asp were introduced within of the CTD of a full-length GluD2 construct tagged with Rluc8 in the CTD. BRET experiments using these constructs revealed that the interaction with post-synaptic density protein-95 (PSD-95) can be modulated by insertion of four phosphomimic Asp mutations sitting in a cluster of residues (aa 997-984).
To study movements within the extracellular part of GluD2 with FRET-FLIM using site-specific labeling of cysteines (Cys), removal of inherently accessible Cys by mutation to serine (Ser) was explored. Membrane-expressed FP-tagged GluD2 constructs containing single and multiple Cys-to-Ser mutations within the ATD and TMD revealed that FRET can occur between small-molecule fluorophore labeled inherent Cys residues in the ATD and CFP located in the ATD/LBD linker region. However, further work is required to minimize sources of background FRET effects.