Given that we planned to

Given that we planned to isolate the GluN1/GluN3A receptor by immunoaffinity chromatography using a Myc epitope tag on GluN3A, we first checked that the tag does not affect the functional properties of the receptor. To do this we co-transfected HEK-293 cells with DNA encoding WT GluN1 and either WT or Myc-tagged GluN3A, and measured the Ca2+ responses of the cells to addition of glycine, using the fluorescent Ca2+ indicator, Fluo-4. As shown in Fig. 2A, addition of glycine (200μM) to the co-transfected cells elicited a small but clear Ca2+ signal, as previously shown with GluN1/GluN3 receptors endogenously expressed in CNS myelin [9]. Importantly, the presence of the N-terminal Myc tag on GluN3A did not affect the Ca2+ response to glycine. As expected, cells expressing GluN1/WT GluN3A or GluN1/Myc-GluN3A did not respond to glycine in the presence of the glycine site antagonist DCKA (1mM; Fig. 2B), and glycine did not elicit a Ca2+ signal in cells expressing RFP only (Fig. 2C). On the basis of these results, we conclude that Myc-tagged GluN3A is fully functional. tsA 201 cells were co-transfected with DNA encoding WT GluN1 and Myc-GluN3A, and protein was isolated from the transfected cells by anti-Myc affinity chromatography. Analysis of an isolated protein sample on a silver stained gel revealed two bands of almost equal intensity at ∼120 and ∼140kDa (Fig. 3A, arrows). Immunoblotting using appropriate anti-subunit 3902 indicated that these two bands were GluN1 and GluN3A, respectively. Hence, GluN1 was robustly co-isolated with GluN3A. Contaminant bands were also seen on the silver stained gel. Isolated receptors were integrated into liposomes (PC:PS, 3:1). The proteoliposomes were deposited onto a mica disk to produce a supported proteolipid bilayer. AFM imaging of this bilayer under fluid revealed a smooth layer containing protein particles (Fig. 3B). Sections through the breaks in the bilayer indicated a height above the mica support of approximately 4nm, as seen previously [5]. Imaging of a number of the bilayer-integrated particles revealed two distinct height populations at around 4nm and around 7nm; for example, Fig. 2B shows two particles of heights 4.1nm (arrowhead) and 7.0nm (arrow). To further characterize the two particle populations, the proteoliposomes were incubated with an antibody against the ABD of the GluN1 subunit (described above). Fig. 3C shows a gallery of particles that were either singly (top panels) or doubly (bottom panels) decorated by antibodies. The frequency distribution of heights of the antibody-decorated particles (determined as illustrated in Fig. 2B) had a peak at 7.2±0.3nm (n=45; Fig. 3D), demonstrating that the taller particles represent the extracellular region of the receptor. The smaller particles likely represent a mixture of incompletely assembled receptors, receptors integrated intracellular face up, and impurities. Fig. 4A shows a frequency distribution of the heights of particles in an untreated proteolipid bilayer. As mentioned above, the distribution had two peaks, at 4.0±0.2nm (n=51) and 6.8±0.2nm (n=49). When bilayer-integrated GluN1/GluN3A receptors were incubated with 10μM glycine, the frequency distribution of heights of the particles above the bilayer again showed two peaks, at 3.9±0.2 (n=44) and 6.0±0.1nm (n=56; Fig. 4B). Importantly, the higher peak was significantly lower than that in the control sample (P=0.004), whereas the lower peak did not change, indicating that the extracellular region had undergone a conformational change upon addition of glycine. A similar reduction in the height of the extracellular domain was seen after addition of 10μM d-serine (height peaks 3.7±0.2nm (n=59) and 5.9±0.1nm (n=49; P=0.006); Fig. 4C). In contrast, glycine did not cause a reduction in height of the extracellular region when added along with DCKA, at either 1mM (height peaks 3.7±0.2nm (n=55) and 6.9±0.2nm (n=47; P=0.639); Fig. 4D) or 10μM (height peaks 4.0±0.3nm (n=63) and 7.1±0.2nm (n=55; P=0.363); Fig. 4E). Further, the height of the extracellular domain was unchanged after addition of 10μM glutamate (height peaks 4.2±0.2nm (n=51) and 6.8±0.1nm (n=50; P=0.120); Fig. 4F).