PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Science. Author manuscript; available in PMC 2017 September 24.
Published in final edited form as:
PMCID: PMC5568803
NIHMSID: NIHMS856429

Cryo-EM structures of the triheteromeric NMDA receptor and its allosteric modulation

Summary

N-methyl-D-aspartate receptors (NMDARs) are heterotetrameric ion channels assembled as diheteromeric or triheteromeric complexes. Here we report structures of the triheteromeric GluN1/GluN2A/GluN2B receptor in the absence or presence of the GluN2B-specific allosteric modulator Ro 25-6981 (Ro), determined by cryo-EM. In the absence of Ro, the GluN2A and GluN2B amino terminal domains (ATD) adopt “closed” and “open” clefts, respectively. Upon binding Ro, the GluN2B ATD clamshell transitions from an open to a closed conformation. Consistent with a predominance of the GluN2A subunit in ion channel gating, the GluN2A subunit interacts more extensively with GluN1 subunits throughout the receptor, in comparison to the GluN2B subunit. Differences in the conformation of the pseudo 2-fold related GluN1 subunits further reflect receptor asymmetry. The triheteromeric NMDAR structures provide the first view of the most common NMDA receptor assembly and show how incorporation of two different GluN2 subunits modifies receptor symmetry and subunit interactions, allowing each subunit to uniquely influence receptor structure and function, thus increasing receptor complexity.

Introduction

The NMDAR (1) is a molecular coincidence “detector” that transduces binding of glycine (2) and glutamate (3), together with the voltage-dependent unblock of magnesium (4, 5), into the opening of a transmembrane ion channel, resulting in depolarization of the postsynaptic membrane potential and entry of calcium, thereby initiating both electrical and chemical signals in the postsynaptic cell (6). Spread throughout the central nervous system (7, 8), NMDARs are integral for fast excitatory signal transmission, essential for normal brain development, and function and are implicated in multiple neurological injuries, diseases and disorders (9). NMDARs play particularly crucial roles in learning and memory and are the targets of clinically relevant drugs for treatment of Alzheimer’s disease (10), schizophrenia (11), depression (12) and epilepsy (13).

Diversity in NMDAR function arises in part from a spectrum of NMDAR subunits that can assemble within the heterotetrameric complex and that include the glycine-binding GluN1 and GluN3 subunits and the glutamate-binding GluN2A-D subunits (9). The prototypical NMDAR harbors two GluN1 subunits and two GluN2 subunits, whereby the identity and count of the GluN2 subunit are dictated by cell-specific conditions. The GluN1/GluN2A or GluN1/GluN2B receptors are canonical representatives of diheteromeric receptors (14) and the GluN1/GluN2A/GluN2B receptor is the paradigm triheteromer, the most common NMDAR receptor spread throughout the hippocampus and cortex (1519). While a great deal is known about the physiology, pharmacology and structure of the GluN1/GluN2B diheteromeric receptor, there is a relative dearth of knowledge about the GluN1/GluN2A/GluN2B triheteromer. This receptor is uniquely modulated by the GluN2A and GluN2B allosteric antagonists divalent zinc (Zn) (20) and the phenylethanolamines ifenprodil or Ro 25-6981 (Ro) (21), respectively, and exhibits ion channel gating kinetics, as well as pharmacology, that are distinct from either the GluN2A or GluN2B diheteromeric receptors (2224). While there are multiple structures of the diheteromeric GluN1/GluN2B receptor (2527), there is no experimental structure of a full-length GluN2A-containing NMDAR, and thus the conformation and interactions of the GluN2A subunit remain unknown.

To define how two different GluN2 subunits are incorporated into the NMDAR heterotetrameric assembly and to elaborate the structure of the full-length GluN2A subunit, we elucidated the structure of the GluN1/GluN2A/GluN2B triheteromeric receptor by single particle cryo-electron microscopy (cryo-EM), using a high affinity GluN2B-specific Fab to unambiguously distinguish the GluN2B subunit from the GluN2A subunit. We further carried out single particle cryo-EM studies in the presence or absence of the GluN2B-specific allosteric antagonist, Ro, in order to understand the structural basis for the action of GluN2B specific antagonists in the context of a triheteromeric receptor. Our structural studies illustrate the architecture of the triheteromeric receptor complex, define the molecular action of Ro, and show how the GluN2 subunits participate in distinct interactions throughout the receptor assembly.

Receptor isolation and structure determination

The wild-type triheteromeric GluN1/GluN2A/GluN2B receptor from Xenopus laevis, deemed triNMDARwt, expresses poorly and is biochemically unstable, thus hindering single particle cryo-EM studies. To improve expression level and stability we exploited two sets of constructs derived from previously published studies, the triNMDAEM construct and the triNMDAEM-G610 construct (Fig. S1 A–C), where the latter construct has the G610 to R substitution in the GluN1 subunit reverted to the wild-type glycine residue (26, 27). In comparison to triNMDAwt (Fig. S1D), the triNMDAEM-G610 preserves small but measureable glycine/glutamate-induced conductances as well as inhibition by Zn or Ro. We carried out additional studies using constructs where we reverted an additional mutation in GluN1 M4 helix to its wild-type identity (GluN1EM-G610-M816) or where we simply used wild-type GluN1 subunit, in combination with GluN2AEM and GluN2BEM constructs (Fig. S1 E–H; see Methods). These two closely related constructs showed larger currents that were also inhibited by Zn or Ro, thus supporting the conclusion that the functional receptor is a triheteromeric assembly. Agonist-induced currents of the triNMDAEM construct are not measurable, either due to low conductance or to a low open probability or both.

Receptor expression in mammalian cells was monitored by fluorescence-detection size-exclusion chromatography (FSEC) (28) and enhanced by utilization of a bicistronic Bacmam virus harboring the GluN1 and GluN2A subunits, together with a monocistronic GluN2B virus (29). A systematic screening of fusion partners showed that a SRC homology 3 (SH3) domain fused to the C-terminus of the GluN2B subunit further increases receptor expression (Fig. S1 I). Coexpression of the GluN1, GluN2A and GluN2B subunits yields triheteromeric GluN1/GluN2A/GluN2B receptors in addition to diheteromeric GluN1/GluN2A and GluN1/GluN2B receptors. To isolate the triheteromeric complex, we exploited Strep-II tagged GluN2A, His8 tagged GluN2B and untagged GluN1 subunits (Fig. S1 I) combined with a two-step affinity purification strategy. Dual-affinity purified triheteromeric receptor eluted from a size-exclusion chromatography column (SEC) as a single sharp peak and showed three protein species on a SDS-polyacrylamide gel (Fig. S1 J).

To probe the biochemical integrity of the triheteromeric receptor preparation and to label the GluN2B subunit for cryo-EM reconstruction, we developed the 10B11 and 11D1 monoclonal antibodies to the GluN1 and GluN2B subunits, respectively (Fig. S1 K). These antibodies recognize three-dimensional epitopes and do not bind under denaturing conditions, thus allowing us to selectively distinguish subunits within the assembled complex. To evaluate the extent to which the triheteromeric receptor preparation was a homogenous population and not a mixture of diheteromeric receptors, we analyzed the shifts of the receptor upon binding of GluN1 and GluN2B-specific Fabs using FSEC. The 10B11 (GluN1) Fab shifts both the diheteromeric GluN1/GluN2A and triheteromeric GluN1/GluN2A/GluN2B receptors to earlier elution volumes, whereas 11D1 (GluN2B) only shifts the triheteromeric receptor (Fig. S1 L–M). In addition, the shift of the triheteromeric receptor by 11D1 is about half of that promoted by 10B11, consistent with the presence of two GluN1 and one GluN2B subunits in the purified receptor, as opposed to the preparation being a mixture of diheteromeric receptors (Fig. S1 M).

To define the arrangement of subunits in the triheteromeric NMDAR and to understand the structural basis for receptor modulation by GluN2B-specific modulators, we elucidated structures of the glycine and glutamate-bound triNMDAREM in complex with the 11D1 Fab in the absence or presence of Ro, termed the non-Ro-bound and Ro-bound states, respectively (Fig. S2, S3, S4). Ambient zinc, a nanomolar-affinity allosteric inhibitor acting at the GluN2A subunit (30), was chelated by EDTA in the receptor preparation. We also carried out single particle cryo-EM studies on the functional triNMDAREM-G610 receptor-11D1 Fab complex to compare to the triNMDAREM (Fig. S5, Fig. S6). To determine the effects of Fab binding, we determined a low-resolution structure of the triNMDAREM in the absence of the Fab (Fig. S7), finding that the triNMDAREM-G610 structure is indistinguishable, at the present resolution, from that of triNMDAREM (Fig. S6, Fig. S8). Because the triNMDAREM yielded the highest resolution reconstructions, we use it as the basis for the overall structural analysis.

The GluN2A and GluN2B subunits share 73% sequence identity and have similar predicted secondary and tertiary structure. We thus employed the GluN2B-specific Fab, 11D1, to distinguish the GluN2A and GluN2B subunits for particle alignment in image processing (Fig. S9 A–C). After reference-free two-dimensional (2D) classification we observed a strong signal for a single Fab on one side of the receptor in most classes of the non-Ro-bound and Ro-bound receptor (Fig. 1 A, Fig. S3 B, Fig. S4 B). In some classes, however, we observed a second Fab with variable intensity or apparent occupancy. Nevertheless, in subsequent 3D classification, a Fab-free reference model resulted in five 3D classes showing an unambiguous signal for only a single Fab (Fig. S3 C, Fig. S4 C). Therefore, the apparent second Fab signal in some of the 2D classes was due to the pseudo 2-fold symmetry of the complex and an imperfect alignment of particles. The 3D classes each occupy a similar percentage of particles and share similar overall shape; the differences in the reconstructions are mainly due to the flexibility of the Fab and the intrinsic flexibility between the extracellular domains and transmembrane domain (TMD) of the receptor (Fig. S3 C, Fig. S4 C). Three-dimensional refinement of individual classes yielded low-resolution reconstructions. By combining all of the particles and masking out the mobile Fab, we obtained reconstructions of the non-Ro-bound and Ro-bound forms of the triNMDAREM at nominal resolutions of 4.5 and 6.0 Å, respectively (Fig. S3 D–F, Fig. S4 D–F).

Figure 1
Architecture of triheteromeric NMDARs

In the non-Ro-bound form, the densities for the ATDs, LBDs, ATD-LBD linkers, and TMD including M2, P loop and M3 are continuous and mostly well defined, with some bulky side chains visible (Fig. 1 B, Fig. S9, Movie S1). By contrast, the densities of the LBD-TMD linkers, M1 and M4, although less well defined, are of sufficient strength and connectivity to trace the main chain. The loops connecting M1 and M2 are not visible in the density map. The position of the GluN2B subunit is determined by the density of Fab bound primarily to the R1 lobe of GluN2B ATD, along with a minor interface to the R2 lobe of the adjacent GluN1 ATD (Fig. 1B–E, Fig. S9 A–C). The quality of the density maps and the validity of the structure is supported by the observation of 9 and 7 N-linked glycans in the non-Ro and Ro-bound structures, respectively (Fig. S9 D–G). Despite a lower resolution of the Ro-bound structure, we visualized the densities of ATD, LBD, and ATD-LBD linkers, along with the M3 helix, and were able to fit molecular models derived from the non-Ro-bound structure into these density features (Fig. 1 C, Fig. S4). We defined the structure of the GluN2A ATD (Fig. S10) by exploiting a homology model derived from high-resolution GluN2B ATD structures (25, 26, 31) in combination with flexible fitting (32). In comparison to the unliganded GluN2B ATD, the GluN2A ATD ‘clamshell’ possesses more pronounced cleft closure and a more extensive interface − ~30% larger − with its paired GluN1 ATD, as discussed below.

Overall architecture

The triheteromeric NMDAR adopts a bouquet-like shape, with ATD, LBD and TMD arranged in layers from top to the bottom, assembled as a GluN1/GluN2A/GluN1/GluN2B heterotetramer (Fig. 1 B–D) (25, 26). The receptor displays a “dimer of dimers” arrangement (3335), first seen in the GluA2 AMPA receptor structure (36), with a swapping of domains between the ATD and LBD layers. The two GluN1 subunits occupy the A/C positions and the GluN2A and GluN2B subunits occupy the B/D positions, respectively. Within the ATD layer, the GluN1 (A)/GluN2A (B), and GluN1 (C)/GluN2B (D) subunits associate as ‘local’ ATD heterodimers, respectively, whereas in the LBD layer, GluN1 (C)/GluN2A (B), and GluN1 (A)/GluN2B (D) interact to form local LBD heterodimers (Fig. 1 E).

There are extensive subunit-subunit interactions within and between GluN1/GluN2A and GluN1/GluN2B heterodimers that meld the tetrameric assembly together. Within the ATD layer the most intensive subunit-subunit interactions occur within each ‘local’ heterodimer, involving interactions between subunits at the A/B and C/D positions. Inter-heterodimer interactions between the R2 interface of GluN2A and GluN2B ATDs (Fig. 1 E) suggest a structural basis for transduction of conformational movements between the GluN2A- and GluN2B-containing ATD heterodimers. Even though the Ro-bound structure exhibits a similar overall subunit arrangement to the non-Ro-bound structure, there is a “compression” between the R1 lobes of the two GluN1 ATDs along with an increase in the separation between the R2 lobes of the GluN2 ATDs (Fig. 1 C–D), demonstrating that binding of Ro to the GluN1/GluN2B ATD heterodimer promotes conformational rearrangement throughout the entire ATD layer.

In the transition from the ATD layer to the LBD layer, the distance between ATD and LBD in GluN1/GluN2A is shorter, resulting in a more compact ATD-LBD interface than in GluN1/GluN2B (Fig. 1 F). At the level of the LBD, GluN2A (B) and GluN2B (D) represent the distal subunits, while GluN1 A and C are in proximal positions (Fig. 1 E), consistent with the greater role of the GluN2 subunit in receptor gating (1, 3739). As a consequence of domain swapping, the LBD heterodimers are formed by A-D and B-C subunits with extensive heterodimer contacts, termed the ‘major’ interface. By contrast, the ‘minor’ interfaces are formed by fewer, yet still significant, interactions between LBD heterodimers (Fig. 1 E). The TMDs of four subunits show a similar arrangement as observed in AMPA, kainate and NMDA receptors (25, 26, 36, 4043). However, the organization of the four transmembrane helices are distinct between the two GluN1 subunits, particularly the orientation of M4, a helix that participates in extensive interactions with an adjacent GluN2 subunit, relative to the rest of the TMD (Fig. 1 F). The differences within the TMD layer, along with the distinct arrangements throughout the extracellular domains, demonstrate that triheteromeric NMDARs harbor an overall asymmetric architecture distinct from diheteromeric NMDARs which, by contrast, have an approximate overall 2-fold axis of symmetry in their resting and activated states (2527, 44). The asymmetry of the triheteromeric NMDAR is thus consistent with non-equivalent functional roles of the GluN2A and GluN2B subunits (22, 45, 46).

Asymmetry of ATD and ATD-LBD interfaces

NMDA receptors harbor binding sites for subunit-specific allosteric modulators on the GluN2 subunits (1, 47, 48), and conformational changes are conveyed to the LBD layer via the ATD-LBD interface and the ATD-LBD linker (38). In the triheteromeric NMDAR, the GluN1/GluN2A ATD makes multiple interactions with the LBD layer, whereas the GluN1/GluN2B ATD is separated from the LBD layer by a solvent-filled gap (Fig. 2 A–B). We hypothesize that these differences are largely the consequence of distinct conformations of the ATD-LBD linkers. Whereas the GluN2A ATD and LBD interface is occupied by a contracted linker, which mediates extensive ATD-LBD interactions, the GluN2B linker adopts an extended conformation and does not ‘fill’ the space between the ATD and LBD, resulting in a contact area that is ~40% of that of GluN2A.

Figure 2
Asymmetric ATD and ATD-LBD interfaces

Inspection of the ATD-LBD interface shows that the GluN2A ATD not only interacts with its own LBD, but also with its neighboring GluN1 subunit. The protruding α5-α6 loop on the “bottom” of the GluN2A ATD is embedded within an ATD-LBD crevice formed by loop 1 of the GluN2A LBD together with loop 2 and helix F of an adjacent GluN1 LBD (Fig. 2 C). By contrast, the α5-α6 loop of the GluN2B ATD is positioned to make only a few interactions with loop 2 of the GluN1 LBD (Fig. 2 D). More extensive GluN2A ATD to LBD contacts are consistent with studies that show triheteromeric NMDARs are less sensitive to GluN2B-specific modulators while retaining substantial modulation by GluN2A-specific small molecules (22, 23).

To further explore the conformations of the GluN2A and GluN2B subunits in the triheteromeric receptor, we compared the ATD heterodimers and individual subunits with each other (Fig. 2 E–G) and also with the isolated non-Ro-bound GluN1/GluN2B ATD heterodimer and the Ro-bound diheteromeric GluN1/GluN2B receptor structures (Fig. S11 A–D) (26, 44). First, upon superposition of the R1 lobes of the two GluN1 subunits from the triheteromeric NMDAR, we see that the GluN1/GluN2A ATDs exhibit a clockwise rotation relative to the GluN1/GluN2B ATDs. Second, when the individual GluN2A and the GluN2B ATDs are superimposed via their R1 lobes, we observe that the GluN2A ATD adopts a more closed-clamshell conformation in comparison to the apo GluN2B ATD. Indeed, the conformation of the GluN2A ATD is more similar to that of the Ro-bound GluN2B ATD structure in the intact diheteromeric GluN1/GluN2B receptor (Fig. S11 A–D), where the GluN2B ATD adopts a closed-clamshell conformation due to the binding of phenylethanolamine allosteric modulators at the interface of the GluN1/GluN2B ATDs (26, 27, 44, 46).

The triheteromeric receptor, prepared with zinc chelator, should not have zinc bound to the receptor and thus it is striking that the ‘apo’ GluN2A ATD adopts a closed cleft conformation (38, 49). We thus compared the conformation of the GluN2A ATD from the triheteromeric receptor structure with the structure of the isolated GluN2A ATD in its zinc-bound form by superposition of their respective R1 lobes. This analysis shows that the isolated, zinc-bound GluN2A ATD adopts a different conformation with an “outward” twisted R2 lobe (Fig. S11 E–F) (50). We speculate that in the intact receptor, zinc might induce a conformation of the GluN2A ATD that is different from what is seen in the current triheteromeric receptor structure, perhaps a conformation similar to that of the isolated, zinc-bound ATD structure, with a ‘twisted’ conformation of the R1 and R2 lobe. The zinc-bound ATD, in turn, could stabilize the LBD layer in a conformation that is not compatible with an open ion channel gate. We note, nevertheless, that protons also inhibit NMDAR activity (51) and it is possible that the pH of the triheteromeric receptor preparation (6.5) stabilizes the GluN2A ATD in a closed-cleft conformation, in the absence of zinc. Most importantly, further structural studies are required to unambiguously resolve the effects of protons and zinc on the structure of the triheteromeric receptor.

The closed clamshell of the GluN2A ATD leads to more extensive interactions with its paired GluN1 ATD through two loops − the upper part of the α5-α6 loop in the R2 lobe and post-α8 loop in the R1 lobe (Fig. 2 F). This suggests an important role of the α5-α6 loop that connects the ATD heterodimer interface with the ATD-LBD interface. A portion of the long post-α8 loop in the GluN2A R1 lobe adopts an α-helix conformation in GluN2B, which effectively compresses its length and reduces the extent to which the GluN2B subunit can interact with the GluN1 subunit. Lastly, the GluN2B ATD of the triheteromeric NMDAR exhibits an open clamshell conformation, similar to that of the diheteromeric non-Ro-bound GluN1/GluN2B (Fig. S11), thus indicating that the GluN2A subunit does not profoundly alter the GluN1/GluN2B ATD conformation and thus modulator binding (22).

The conformational differences between GluN2A and GluN2B subunits at the ATD-LBD interfaces are reminiscent of the difference between NMDAR and non-NMDAR glutamate receptors. In non-NMDARs, direct interactions between the ATD and LBD layers are minimal. Therefore, even if the ATDs undergo ligand-induced movements, there are few molecular routes for transduction of conformational changes and thus the ATDs of non-NMDARs do not markedly modulate ion channel gating (36, 4143, 52). By contrast, there are multiple distinct contacts between the ATD and LBD layers in NMDARs, and thus a number of mechanisms by which movements of the ATDs can be transduced to the LBD layer and, subsequently, to the TMD to modulate ion channel activity (38, 53, 54). More extensive contacts of the GluN2A ATD with both the GluN1 ATD and LBD layers suggests that the GluN2A subunit is positioned to more profoundly influence the activity of triheteromeric NMDARs in comparison to the GluN2B subunit, a structural observation consistent with analysis by electrophysiological studies of triheteromeric NMDARs (22).

Organization of LBD layer

Comparison with the agonist-bound LBD crystal structures reveals that all four of the LBD clamshells in the triheteromer adopt a “closed” conformation, consistent with complete occupancy by the full agonists (Fig. S12) (34). Indeed, we visualized density for glutamate in the GluN2A LBD (Fig. S12 A), but not in the GluN2B subunit, likely due to a lower local resolution of the map for the GluN2B LBD. Overall, the LBD layer shares a similar organization with diheteromeric GluN1/GluN2B receptors. The GluN2 LBDs form two interfaces with their adjacent GluN1 subunits: a major interface within the heterodimer formed by the D1-D1 interface, and a minor interface between the heterodimer formed by the D2 of the GluN1 subunit and D1 of the GluN2 subunit (Fig. 1 E, Fig. 3 A). While the major interfaces are similar, the minor interface of the GluN2A subunit buries ~2.6-fold larger solvent accessible surface than that of the GluN2B subunit, thus disrupting the pseudo two-fold symmetry at the LBD layer (Fig. 3 B–C). Specifically, in the GluN1/GluN2A minor interface, loop 1 of GluN2A is coupled to the loop connecting helix F and G in the GluN1 subunit; helix K and the downstream S1-M4 linker in the GluN2A subunit are also in closer contact with helix E and the M3-S2 linker in the GluN1 subunit (Fig. 3 C). Thus, the more extensive interactions of GluN2A with both GluN1 subunits within the LBD layer are consistent with the dominant role of the GluN2A subunit in the gating of triheteromeric NMDARs (22).

Figure 3
Asymmetry in the LBD layer

Subunit rearrangement upon allosteric modulation

The presence of a GluN2A subunit markedly decreases the inhibition rate, efficacy and binding affinity of phenylethanolamine allosteric antagonists, such as ifenprodil or Ro, at triheteromeric GluN1/GluN2A/GluN2B receptors, in comparison to diheteromeric GluN1/GluN2B receptors (22, 23). However, the molecular basis by which the Ro-insensitive GluN2A subunit perturbs phenethylamine modulation is unclear. The triNMDAREM constructs bind Ro with submicromolar affinity (Fig. 4 A), ~3-fold weaker than that of the diheteromeric GluN1/GluN2B receptor (26). To probe the structural underpinnings for differences in phenethylamine binding and modulation between diheteromeric and triheteromeric NMDARs, we solved the structure of the triheteromeric NMDAR in the presence of Ro (Fig. 1 C, Fig. S4). In the Ro-bound structure, the ATD layer of the GluN2B subunit undergoes a substantial rearrangement, leading to a more compact ATD-LBD interface (Fig. 4 B). The GluN1/GluN2B ATD undergoes a clockwise rotation (Fig. 4 C) due to an identical clamshell closure of the GluN2B ATD as in the Ro-bound GluN1/GluN2B diheteromer (Fig. 4 D, Fig. S13, Movie S1). By contrast, the GluN1/GluN2A ATD heterodimer remains largely unaffected as does the Ro binding pocket at the GluN1/GluN2B ATD interface (Fig. S13). However, the binding of Ro does remodel the ATD layer from asymmetric to pseudo two-fold symmetric by repositioning the two ATD heterodimers (Fig. 4 E–F) which, in turn, diminishes the contacts between the two GluN2 ATDs (Fig. 4 G–H). Our results suggest that phenylethanolamine modulators perturb the overall receptor structure, at least in part, by alterations in the interface between GluN2 ATDs. This interface is less extensive in the diheteromeric GluN1/GluN2B receptor (27, 44), explaining reduced Ro affinity on triheteromeric NMDAR, because the Ro-induced clamshell closure requires overcoming the interactions at the GluN2 ATD interface. In the Ro-bound state, the diminished GluN2 ATD interface precludes the GluN2B ATD from affecting the GluN2A ATD, consistent with an increase of zinc inhibition via the GluN2A subunit in the presence of ifenprodil, akin to the enhanced ifenprodil inhibition in the presence of divalent zinc (45).

Figure 4
Consequences of Ro 25-6981 binding

In addition to the ATD layer, contacts of the GluN2B ATD with the LBD layer are altered upon Ro binding. In particular, the lower part of the α5-α6 loop moves apart from loop 2 of the GluN1 LBD, resulting in few interactions with the GluN2B LBD (Fig. 4 I). We speculate that, in the absence of Ro, the lower part of the α5-α6 loop is in a position that hinders the GluN1/GluN2A LBD heterodimer from restoring its “resting” position after channel activation. Binding of Ro decouples the α5-α6 loop from GluN1 LBD, facilitating the GluN1/GluN2A LBD heterodimer movement toward the pore center, promoting channel closure.

Asymmetric transmembrane domain

The M3 helix is a crucial structural determinant linking agonist binding to channel gating. In the non-Ro-bound structure, M3, M2 and the pore loop (P loop) have well defined densities. The M3 helix and the tip of the P loop form a pyramidal vestibule harboring an elliptical density feature representing a trapped MK-801 molecule, with its two aromatic rings positioned toward the M3 helices of GluN2A and GluN2B (Fig. 5 A–B), consistent with experimental measurements of MK-801 binding to the triNMDAEM construct (Fig. 5 C) despite a low ion channel activity. Based on the observation that the M3 bundle crossing occludes the ion channel pore, the non-Ro-bound structure is an agonist-bound, closed-blocked state.

Figure 5
MK-801 binding and asymmetric TMD organization

By comparing the M3 helix of the triheteromeric GluN1/GluN2A/GluN2B and diheteromeric GluN1/GluN2B receptor, we observed that, unlike diheteromeric NMDAR with pseudo 4-fold symmetry (25, 26), the TMD of triNMDAR deviates substantially from pseudo 4-fold symmetry due to an extension of the GluN1 (C) subunit from the pore axis (Fig. 5 D). To understand why the two GluN1 subunits are asymmetric, we traced their interactions through the LBD to the ATD-LBD interface and the ATD. The GluN1 M3 helix is connected to its D2 LBD lobe, which in turn interacts with the GluN2 R2 ATD lobe at the ATD-LBD interface (Fig. 5 E). Thus, we suggest that the arrangement of the GluN1 M3 helix is indirectly altered by the asymmetric GluN2 ATD and ATD-LBD interface, specifically, the interactions between GluN1 (A) and GluN2A, and GluN1 (C) and GluN2B. Compared to the compact interaction between GluN2A ATD and GluN1 (A), the contacts between GluN2B ATD and GluN1 (C) are nearly ruptured (Fig. 2 A–D, Fig. 5 E), which promotes movement of the D2 lobe of the GluN1 (C) subunit toward the TMD, thus “pushing” the M3 helix away from the central pore axis.

Mechanism

The presence of GluN2A and GluN2B subunits in triheteromeric NMDAR disrupts the 2-fold symmetry in the ATD and LBD layers and the pseudo 4-fold symmetry in the TMD layer of diheteromeric NMDARs. The GluN2A ATD adopts a “closed” cleft in the absence of the allosteric inhibitor zinc and at low pH, while the GluN2B ATD cleft transitions from “open” to “closed” upon binding of the allosteric modulator Ro. The GluN2A and GluN2B subunits have similar structures yet they make distinct contributions to receptor structure and function (Fig. 6 A–B). First, the GluN2A ATD interacts extensively with the GluN1 R1 lobe within the ATD heterodimer, and thus is poised to modify the conformational properties of its GluN1 ATD partner to a greater extent than GluN2B. Second, the GluN2A ATD resides on top of its cognate LBD, participating in extensive interactions that include insertion of the α5-α6 loop into a pocket located on top of the GluN2A/GluN1 LBD minor interface. By contrast, the GluN2B ATD is loosely connected to its LBD and is not positioned to make as extensive of contacts with the corresponding LBD layer. Lastly, the GluN2A LBD interacts extensively with the GluN1 LBD at both minor and major interfaces, whereas the GluN2B LBD is primarily coupled to the GluN1 subunit at the major interface. Taken together, the presence of two different GluN2 subunits endows the NMDA receptor with enhanced diversity in structure and function, enriching the complexity of the receptor for signaling at chemical synapses and increasing its effectiveness as a potential target for therapeutic agents.

Figure 6
Conformational asymmetry and allosteric modulation

Methods

Construct design

GluN1, GluN2A and GluN2B constructs used here, deemed GluN1EM, GluN2AEM and GluN2BEM, were designed based on the previously reported Xenopus laevis NMDA GluN1-Δ1 and GluN2B-Δ2 constructs (26, 27). GluN1EM is generated by removal of the C-terminal of GluN1-Δ2, including the enhanced green fluorescent protein (eGFP), 3C cleavage site (Leu-Glu-Val-Leu-Phe-Gln-Gly-Pro) and an octa-histidine tag. GluN2BEM reverts the K216C mutation in GluN2-Δ2 to a wild-type lysine residue, with a GFP11, SH3 fusion protein and an octa-histidine tag placed at its C-terminus. GluN2AEM was designed based on the GluN2BEM construct, but including the C-terminal 3C cleavage site, eGFP and strepII tag. Due to the coexistence of diheteromeric GluN1/GluN2A and GluN1/GluN2B, we used split GFP to monitor the expression level of triheteromeric NMDAR over diheteromeric NMDAR. Briefly, the GFP can be cleaved into two parts including GFP1-10 and GFP11 and these can be reassembled into a functional intact GFP by fusing them into interacting protein subunits (55). We fused GFP1-10 and GFP11 into the C-terminal of bicistronic GluN1-GluN2A and GluN2B, respectively. As a result, only triheteromeric GluN1/GluN2A/GluN2B containing both GFP1-10 and GFP11 will fluoresce, and the diheteromeric GluN1/GluN2A or GluN1/GluN2B containing either GFP1-10 or GFP11 will not, which was useful for initial construct screening. We replaced the GFP1-10 in the bicistronic GluN1-GluN2A construct with an intact GFP for large-scale expression because monitoring the GFP fluorescence was important to control the quality of virus production for the bicistronic construct. The final constructs we used for EM experiments are listed in Fig. S1 I. To boost expression levels, GluN1EM and GluN2AEM constructs were cloned into the same pEG BacMam vector, with the result deemed the GluN1-GluN2AEM construct.

Purification of tri-NMDARs

Bacmid and baculovirus of GluN1-GluN2AEM and GluN2BEM in BacMam vector were generated (29) and P2 viruses were used to infect suspension HEK293 GnTI- cells at a multiplicity of infection (M.O.I.) of 1:1 (GluN1-GluN2AEM:GluN2BEM) and then incubated at 37 °C. At 12 h post-transduction, 10 mM sodium butyrate and 2.5 μM MK-801 were added to the culture and the temperature was set to 30 °C. The cells were collected and resuspended in a buffer containing 150 mM NaCl, 20 mM Tris 8.0 in the presence of 1 mM PMSF, 0.8 μM aprotinin, 2 μg/ml leupeptin, and 2 mM pepstatin A (protease inhibitors). The receptor was extracted from whole cell with a buffer containing 150 mM NaCl, 20 mM Tris 8.0, 1% MNG-3, protease inhibitors and 2 mM cholesteryl hemisuccinate (CHS) for 2 h at 4 °C. The solubilized receptors containing GluN1/GluN2A, GluN1/GluN2B and GluN1/GluN2A/GluN2B were incubated with TALON resin to remove strepII-tagged GluN1/GluN2A. The bound GluN1/GluN2B and GluN1/GluN2A/GluN2B receptors were then eluted with 250 mM imidazole at pH 8.0 to streptactin resin. His-tagged GluN1/GluN2B passed through the column and only GluN1/GluN2A/GluN2B was bound to streptactin resin and eluted with buffer containing 5mM desthiobiotin. The receptor was concentrated, mixed with Fab 11D1 at a molar ratio 1:1.2 and was further purified by size-exclusion chromatography in the buffer containing 400 mM NaCl, 20 mM MES pH 6.5, 0.5 mM n-dodecyl β-D-maltoside (DDM), and 0.2 mM CHS. Peak fractions containing the receptor were pooled and concentrated to 4 mg/ml.

Antibody production

Monoclonal antibodies against GluN1 and GluN2B (10B11 and 11D1, respectively) were raised by Dan Cawley (Vector and Gene Therapy Insititute, OHSU) using standard methods. GluN1/GluN2 was purified as described previously (26) in 1 mM DDM and in the presence of 1 mM glutamate, 1 mM glycine, and 0.2 mM CHS. Purified GluN1/GluN2 was reconstituted into liposomes for immunization as described previously for SERT (56) (57), except 400 mM NaCl and 0.8% Na deoxycholate was used for reconstitution of GluN1/GluN2 and excess salt and detergent was removed using PD-10 desalting columns. Mice were immunized with 30 μg of reconstituted GluN1/GluN2B to generate hybridoma cell lines. Antibodies were screened by fluorescence-based size-exclusion chromatography (28) and western blot to select clones that recognized natively folded GluN1/GluN2B and GluN1/GluN2A protein. The 10B11 and 11D1 monoclonal antibodies were purified from hybridoma supernatants using 4-mercapto-ethyl-pyridine chromatography resin. Fab was generated by papain cleavage of 10 mg mAb at 1 mg/ml final concentration for 2 h at 37 °C in 50 mM NaPi, pH 7.0, 1 mM EDTA, 10 mM cysteine and 1:100 w/w papain. The digest was quenched with 30 mM iodoacetamide at 25 °C for 10 min and Fc was removed from the MAb digest using Protein A. Fab 10B11 was purified by anion exchange using a Hi Trap Q HP column in 20 mM Tris pH 8 and 200 mM NaCl. Fab 11D1 was purified by cation exchange using a Hi Trap SP HP column in 20 mM NaOAc pH 5 and 200 mM NaCl.

EM data acquisition and processing

Purified GluN1/GluN2A/GluN2B was mixed with 2 mM glycine, 2 mM glutamate, 20 μM MK-801, 0.5 mM EDTA and/or 1 mM Ro (in DMSO) a few hours before grid preparation. Double blotting of a 1.3+2.5 μl sample at a concentration of 4 mg/ml was applied to a glow-discharged Quantifoil holey carbon grid (gold, 1.2/1.3 μm size/hole space, 300 mesh), blotted using a Vitrobot Mark III using 3s blotting time with 100% humidity, and then plunge-frozen in liquid ethane cooled by liquid nitrogen. Images were taken by an FEI Titan Krios electron microscope operating at 300 kV with a nominal magnification of 85k. Images were recorded by a Gatan K2 Summit direct electron detector operated in super-resolution counting mode with a binned pixel size of 1.70 Å or 1.33 Å for Fab-bound data or non-Fab-bound data, respectively. For the non-Ro-bound data, each image was dose-fractionated to 70 frames with a total exposure time of 21 s with 0.3 s per frame. For the Ro-bound data, each image was dose-fractionated to 50 frames with a total exposure time of 20 s with 0.4 s per frame. The images were recorded using the automated acquisition program SerialEM. Nominal defocus values varied from 1.3 to 2.5 μm.

For the Ro-bound data, super-resolution counting images were 2x2 binned in Fourier space, motion corrected and summed using MotionCor2 (58). Defocus values were estimated using Gctf (59). Approximately 3000 particles were manually picked and subjected to an initial reference-free 2D classification using Relion (60). Eight representative 2D class averages were selected as templates for automated particle picking for all the Fab-bound data using Gautomatch (http://www.mrc-lmb.cam.ac.uk/kzhang/Gautomatch/). The auto-picked particles were visually checked and false positives were removed. The particles were further cleaned-up by two rounds of 2D classification using Relion. The CTF values of individual particles from selected 2D class averages were estimated using Gctf (59). For 3D classification in Relion, a Fab-free reference model was generated from the GluN1/GluN2B crystal structure (PDB code: 4TLM) and low-pass filtered to 50 Å using EMAN2 (61). The 3D classes (5 for Fab-bound data or 6 for non-Fab-bound data) each occupy a similar percentage of particles and share similar overall shape with differences mainly due to the flexibility of the Fab and the intrinsic flexibility between the extracellular domains and transmembrane domain of the receptor. Three-dimensional refinement of individual classes yielded low-resolution reconstructions. By combining particles from all the classes, we obtained a reconstruction at higher resolution. Initial 3D refinement was carried out using Relion. Particles were further refined using Frealign’s local refinement (62). Subsequently, for all the Fab-bound data, a soft mask around the receptor was calculated and supplied for final refinement using Frealign. The final resolutions reported in Table S1 are based on the gold standard FSC 0.143 criteria (60). No symmetry was applied during the image processing. A similar procedure was used for the other three datasets with the exception that no additional soft mask was used during 3D refinement of non-Fab-bound data.

Model building

A homology model of the triheteromeric GluN1/GluN2A/GluN2B receptor was generated with the crystal structure of the diheteromeric GluN1/GluN2B receptor (26) (PDB code: 4TLM) as a template using the SWISS-MODEL online server (63). A homology model for the Fab was made using the Fab from an existing crystal structure (PDB code: 4M48) by mutating all the residues to alanine. Both homology models were first rigid-body fitted into the non-Ro EM density map using Chimera (64), followed by molecular dynamics flexible fitting (MDFF) (32), which improves the model to map correlation coefficients (CC, without Fab, the same below) from 0.803 to 0.911 (backbone, 0.775 to 0.876). This model was then subjected to Rosetta refinement and a total number of 100 models were generated. The top 10 models with best geometry statistics were inspected for map agreement using Coot (65). The chosen model has slightly improved CC (0.919 and 0.883 for all atoms and backbone, respectively). In addition, the positioning of loops/linkers is improved and some errors related to secondary structure assignments and geometry are corrected. The model from Rosetta was further manually adjusted in Coot, guided by the crystal structures of intact diheteromeric GluN1/GluN2B (PDB code: 4TLM) (26) and GluN1/GluN2A LBD domains (PDB code: 2A5T) (34), as well as by the densities of bulky side chains. The final model has a CC of 0.929 and 0.892 for all atoms and backbone, respectively. For the Ro-bound data, the structure of the non-Ro-bound state was first rigid-body fitted into the EM density map, followed by MDFF. Subsequently, each subdomain of the non-Ro-bound structure, including the R1 and R2 lobes of ATD, D1 and D2 lobes of LBD, and TMD, were aligned to the Ro-bound model and combined to a new Ro-bound model. This model, and the linkers between subdomains in particular, was inspected and manually adjusted in Coot, guided by the non-Ro-bound structure, the crystal structure of intact diheteromeric GluN1/GluN2B (PDB code: 4TLM) and the crystal structure of the LBD of diheteromeric GluN1/GluN2A (PDB code: 2A5T). For validation, FSC curves were calculated between the final models and EM maps. The geometries of the atomic models were evaluated using MolProbity (66). All figures were prepared using UCSF Chimera and Pymol (Schrödinger) (67) .

Saturation binding experiments

MK-801 and Ro binding affinity was determined by scintillation-proximity assay (SPA). SPA experiments were set up in triplicate wells of a 96-well plate. Affinity-purified triheteromeric NMDARs (20 nM) saturated with 2 mM glyine and 2 mM glutamate was incubated with 1 mg/ml copper yttrium silicate (Cu-YSi) beads (Perkin Elmer) and 3H-labelled MK-801 or Ro (1:9 3H:1H) in SEC buffer (20 mM Tris pH 8, 150 mM NaCl, 1 mM MNG-3 and 0.2 mM CHS) with a final volume of 100 μl. Non-specific binding was determined by the addition of 1 mM ifenprofil (for 3H-Ro) or 1 mM PCP (for MK-801). Plates were incubated at room temperature until the counting was stable and were read using a MicroBeta TriLux 1450 LSC and luminescece counter. Data were analyzed using GraphPad Prism.

Two-electrode voltage clamp electrophysiology (TEVC)

The GluN1, GluN2A, and GluN2B constructs for TEVC experiments in the pGEM vector are engineered with the C-terminal tags according to methods developed by the Hansen lab for selective cell-surface expression of recombinant triheteromeric NMDAR (22). The RNAs were transcribed using the mMessage mMachine T7 Ultra Kit (Ambion). Xenopus laevis oocytes were injected with a total 30–200 ng of mRNA with a ratio of GluN1:GluN2A:GluN2B 2:1:1 and were incubated at 16 °C for 2–3 days in the presence of 50 μM competitive antagonist D-APV and 10 μg/ml gentamicin. Borosilicate pipettes were filled with 3 M KCl. The recordings were performed in a buffer containing 100 mM NaCl, 0.3 mM BaCl2, 5 mM HEPES 7.3 and 0.05 mM heavy-metal chelator ethylenediaminetetraacetic acid (EDTA) at −60 mV. All recording experiments were carried out at least 3 times independently.

Supplementary Material

Movie S1

Movie S1. Cryo-EM map and model of non-Ro-bound triheteromeric NMDAR and its conformational transition upon binding of Ro. The movie first shows the overall density map and the fitted atomic model of the non-Ro-bound triheteromeric NMDAR. The movie then shows the local densities for the ATD-LBD linkers, LBD-TMD linkers and TMD. Subsequently the movie shows the differences between GluN1/GluN2A and GluN1/GluN2B heterodimers. At the end the movie shows a conformational transition induced upon binding of Ro.

Supplementary information

Figure S1. Construct design, detection and preparation of the triheteromeric GluN1/GluN2A/GluN2B complex. (A-C) Cartoon representations of the triheteromeric GluN1/GluN2A/GluN2B construct. GluN1, GluN2A and GluN2B are shown in orange, red, and blue, respectively. Deletions in the ATD linker of GluN2A and GluN2B are highlighted with a yellow wedge. The point mutations are highlighted in white circles. Mutations of glycosylation sites in GluN1 (N300Q, N350Q, N368D, N440D, N469D, and N769E), in GluN2A (N67Q, N372A, N431A, N529A, and N675A) and GluN2B (N69Q, N343D and N486V) are not shown. Also not shown are mutations in signal peptide for GluN1 (C22A) and GluN2B (M20S, G21R and C22A), and mutation of a potentially reactive cysteine in GluN2B (C581A). Where noted, some constructs contain mutations in the TMD of NR1 (G610R and M816Y). (D) Activation of triNMDARwt current by 0.2 mM/0.2 mM glutamate/glycine at pH 7.3 by TEVC. (E-H) Inhibition of triNMDAEM-G610/GluN2AEM/GluN2BEM (E), GluN1EM-G610-M816/GluN2AEM/GluN2BEM (F) or GluN1WT/GluN2AEM/GluN2BEM (G) current by 1μM Zn or 3μM Ro, respectively, and the extents of inhibition (H). Current responses were activated by 0.2mM/0.2mM glutamate/glycine at pH 7.3. The extents of Zn or Ro inhibition were determined from measurements on 5 oocytes and calculated using the ratio of current following Zn or Ro (asterisk) application and the glutamate/glycine-induced ‘peak’ current. The error bars represent s.e.m. (I) Summary of Xenopus laevis NMDA cryo-EM construct design. (J) SDS gel of purified triheteromeric NMDARs. To differentiate the three subunits, receptors were treated using EndoH and 3C protease prior to loading on the SDS gel. A band at molecular weight 60 kD marked with an asterisk corresponds to EndoH and two bands at lower molecular weights represent 3C protease, cleaved GFP and GFP11-SH3, respectively. Note that the receptors for cryo-EM experiments were not treated by EndoH and 3C protease. (K) Schematic cartoon showing two binding sites for GluN1-specific Fab 10B11 and one binding site for GluN2B-specific Fab 11D1 in the triheteromeric GluN1/GluN2A/GluN2B complex. (L) FSEC traces show a shift of the diheteromeric GluN1/GluN2A complex upon binding to the GluN1-specific Fab 10B11, but no shift with the GluN2B-specific Fab 11D1. (M) The triheteromeric GluN1/GluN2A/GluN2B complex binds to both GluN1-specific Fab 10B11 and GluN2B-specific Fab 11D1; notably, the shift with the GluN1-specific Fab 10B11 is twice that of the GluN2B-specific Fab 11D1.

Figure S2. The work-flow of cryo-EM data processing, using the non-Ro-bound data as an example. Particles were auto-picked from 2751 micrographs using Gautomatch and visually checked in Relion. After removing false positives, a total number of 343485 particles were subjected to two rounds of 2D classification in Relion. The CTF values of individual particles (302052) from selected 2D class averages were estimated using Gctf. For 3D classification in Relion, a Fab-free reference model was generated from the GluN1/GluN2B crystal structure (PDB code: 4TLM) and low-pass filtered to 50 Å using EMAN2. The 3D classes each occupy a similar percentage of particles and share similar overall shape with differences mainly due to the flexibility of the Fab and the intrinsic flexibility between the extracellular domains and transmembrane domain of the receptor. Three-dimensional refinement of individual classes yielded low-resolution reconstructions. By combining particles from all the classes, we obtained reconstruction of higher resolution. Initial 3D refinement was carried out using Relion. Particles were further refined using Frealign’s local refinement. Subsequently, for all the Fab-bound data, a soft mask around the receptor was calculated and supplied for the final refinement using Frealign. A similar procedure was used for the other three datasets with the exception that no additional soft mask was used during 3D refinement of non-Fab-bound data.

Figure S3. Cryo-EM analysis of non-Ro-bound triNMDAREM in complex with Fab 11D1. (A) Representative electron micrograph. (B) Selected two-dimensional class averages of the electron micrographs. (C) Reference map generated from the GluN1/GluN2B crystal structure (PDB-code: 4TLM) (left), 3D classification with the percentage of particles in each class from 1 to 5, overlay of the five 3D classes (marked with *) with the total number of particles (right). (D) The gold-standard Fourier shell correlation curves for the EM maps with (red) and without (black) masking the Fab. The FSC curve between the atomic model and the final EM map is shown in blue. (E) Angular distribution of particles used for refinement. (F) The three-dimensional map is colored according to local resolution estimation.

Figure S4. Cryo-EM analysis of Ro-bound triNMDAREM in complex with Fab 11D1. (A) Representative electron micrograph. (B) Selected two-dimensional class averages of the electron micrographs. (C) Reference map generated from the GluN1/GluN2B crystal structure (PDB-code: 4TLM), 3D classification with percentage of particles in each class, overlay of the five 3D classes (marked with *) with the total number of particles. (D) The FSC curves for the EM maps with (red) and without (black) masking the Fab. The FSC curve between atomic model and the final EM map is shown in blue. (E) Angular distribution of particles used for refinement. (F) The three-dimensional map is colored according to local resolution estimation.

Figure S5. Cryo-EM analysis of non-Ro-bound triNMDAREM–G610 in complex with Fab 11D1. (A) Representative electron micrograph. (B) Selected two-dimensional class averages of the electron micrographs. (C) Reference map generated from the GluN1/GluN2B crystal structure (PDB-code: 4TLM), 3D classification with percentage of particles in each class, overlay of the five 3D classes (marked with *) with the total number of particles. (D) The FSC curves for the EM maps with (red) and without (black) masking the Fab. The FSC curve between the atomic model (non-Ro-bound triNMDAREM) and the final EM map is shown in blue. (E) Angular distribution of particles used for refinement. (F) The three-dimensional map is colored according to local resolution estimation.

Figure S6. Comparison of non-Ro-bound triNMDAREM-G610 with non-Ro-bound triNMDAREM. (A) Density map of non-Ro-bound triNMDAREM−G610. (B) Overlay of the triNMDAREM−G610 map with the atomic model of non-Ro-bound triNMDAREM. The map is shown as a transparent surface and the model is shown in cartoon representation. (C) Overlay of the same map and model as in (B) for each subunit. The non-Ro-bound triNMDAREM model was rigid-body fitted into the non-Ro-bound triNMDAREM-G610 density using Chimera.

Figure S7. Cryo-EM analysis of Ro-bound triNMDAREM in the absence of Fab (Ro-bound/non-Fab-bound triNMDAREM). (A) Representative electron micrograph. (B) Selected two-dimensional class averages of the electron micrographs. (C) Reference map generated from the GluN1/GluN2B crystal structure (PDB-code: 4TLM), 3D classification with percentage of particles in each class, overlay of the five 3D classes (marked with *) with the total number of particles. (D) The FSC curve for the EM map is shown in black and the FSC curve between the atomic model (Ro-bound triNMDAREM) and the final EM map is shown in blue. (E) Angular distribution of particles used for refinement. (F) The three-dimensional map is colored according to local resolution estimation.

Figure S8. Overlay of the EM map of the Ro-bound/non-Fab-bound triNMDAREM with the atomic model of Ro-bound triNMDAREM. (A-D) Superimposition of the map with the model viewed from different orientations. The Ro-bound triNMDAREM model was rigid-body fitted into the Ro-bound/non-Fab-bound triNMDAREM density in Chimera.

Figure S9. Representative densities of the non-Ro-bound triNMDAREM. (A–B) Cryo-EM map viewed parallel to the membrane (A) and from the extracellular side (B), respectively. The Fab was included during refinement. (C) The interface between the receptor and the Fab. The atomic model is shown in cartoon representation and EM density is shown as transparent surface. The positions of two NAG (N-linked glycans) sites interacting with the Fab are indicated. (D-E) The densities of the two NAG moieties interacting with the Fab. (F–G) Densities of the helix α6 in GluN1. (H) Densities of the helix α5 in GluN2A. (I) Densities of the helix αK in GluN1. (J) Densities of the M2 in GluN2A. (K) Densities of the M3 in GluN2A.

Figure S10. The ATD of the GluN2A subunit. (A) Sequence and secondary structure prediction of GluN2A ATD. (B–C) The densities and atomic model of GluN2A ATD.

Figure S11. Comparison of the ATD heterodimers of the non-Ro-bound triNMDREM with the non-Ro-bound, Ro-bound diheteromeric GluN1/GluN2B and isolated zinc-bound GluN1/GluN2A ATD, respectively. For comparisons in A-D, the structures were superimposed using the R1 lobe of the GluN1 ATD. The distances (Å) between the R2 lobes of the GluN1 and GluN2 subunits and the rotation angles between the two GluN2 R2 lobes are indicated. (A) Comparison of the GluN1/GluN2A ATD heterodimer with non-Ro-bound diheteromeric GluN1/GluN2B. (B) Comparison of the GluN1/GluN2A ATD heterodimer with Ro-bound diheteromeric GluN1/GluN2B. (C) Comparison of the GluN1/GluN2B ATD heterodimer with non-Ro-bound diheteromeric GluN1/GluN2B. (D) Comparison of the GluN1/GluN2B ATD heterodimer with Ro-bound diheteromeric GluN1/GluN2B. (E-F) Comparison of the GluN1/GluN2A ATD heterodimer with the isolated zinc-bound GluN1/GluN2A ATD heterodimer, viewed from the extracellular side of the membrane (E) or in parallel to the membrane (F). The structures were superimposed using the R1 lobe of the GluN2A ATD.

Figure S12. Comparison of the LBD heterodimer of non-Ro-bound triNMDAREM with the soluble LBD heterodimer of diheteromeric GluN1/GluN2A, all bound with glycine and glutamate. The structures are superimposed using the GluN1 LBD. (A) Putative density for bound glutamate in the GluN2A LBD clamshell. (B–D) The positions of the GluN1/GluN2A LBD heterodimer in the triheteromeric NMDAR (A) and its comparison with soluble diheteromeric GluN1/GluN2A (B-C). (E–G) The positions of the GluN1/GluN2B LBD heterodimer in the triheteromeric NMDAR (D) and its comparison with the soluble diheteromeric GluN1/GluN2A (E–F).

Figure S13. Comparison of the ATD heterodimer of Ro-bound triNMDAREM with the non-Ro-bound and Ro-bound diheteromeric GluN1/GluN2B, respectively. The structures were superimposed using the R1 lobe of the GluN1 ATD. The distances (Å) between the R2 lobes of the GluN1 and GluN2 subunits and the rotation angles between the two GluN2 R2 lobes are indicated. (A) Comparison of the GluN1/GluN2A ATD heterodimer with non-Ro-bound diheteromeric GluN1/GluN2B. (B) Comparison of the GluN1/GluN2A ATD heterodimer with Ro-bound diheteromeric GluN1/GluN2B. (C) Comparison of the GluN1/GluN2B ATD heterodimer with non-Ro-bound diheteromeric GluN1/GluN2B. (D) Comparison of the GluN1/GluN2B ATD heterodimer with Ro-bound diheteromeric GluN1/GluN2B.

Table S1. Statistics of data collection, 3D reconstruction and models.

Acknowledgments

We thank the Multiscale Microscopy Core (OHSU) and FEI for the support with microscopy and the Advanced Computing Center (OHSU) and Intel for computational support. We are grateful to L. Vaskalis for help with illustrations and H. Owen for proofreading. We thank all Gouaux and Baconguis laboratory members, and C. Yoshioka, for helpful discussions. The work was supported by the National Institute of Neurological Disorders and Stroke of the National Institutes of Health under award number R01NS038631 (E.G.). E.G. is an investigator with the Howard Hughes Medical Institute.

References

1. Traynelis SF, et al. Glutamate receptor ion channels: structure, regulation, and function. Pharmacol Rev. 2010;62:405. [PubMed]
2. Johnson JW, Ascher P. Glycine potentiates the NMDA response in cultured mouse brain neurons. Nature. 1987;325:529. [PubMed]
3. Watkins JC, Evans RH. Excitatory amino acid transmitters. Annu Rev Pharmacol Toxicol. 1981;21:165. [PubMed]
4. Nowak L, Bregestovski P, Ascher P, Herbet A, Prochiantz A. Magnesium gates glutamate-activated channels in mouse central neurones. Nature. 1984;307:462. [PubMed]
5. Mayer ML, Westbrook GL, Guthrie PB. Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature. 1984;309:261. [PubMed]
6. MacDermott AB, Mayer ML, Westbook GL, Smit AB, Barker JL. NMDA-receptor activation increases cytoplasmic calcium concentration in cultured spinal cord neurones. Nature. 1986;321:519. [PubMed]
7. Moriyoshi K, et al. Molecular cloning and characterization of the rat NMDA receptor. Nature. 1991;354:31. [PubMed]
8. Monyer H, et al. Heteromeric NMDA receptors: molecular and functional distinction of subtypes. Science. 1992;256:1217. [PubMed]
9. Paoletti P, Bellone C, Zhou Q. NMDA receptor subunit diversity: impact on receptor properties, synaptic plasticity and disease. Nat Rev Neurosci. 2013;14:383. [PubMed]
10. McKeage K. Memantine: a review of its use in moderate to severe Alzheimer’s disease. CNS Drugs. 2009;23:881. [PubMed]
11. Balu DT. The NMDA receptor and schizophrenia: From pathophysiology to treatment. Adv Pharmacol. 2016;76:351. [PMC free article] [PubMed]
12. Moskal JR, et al. GLYX-13, an NMDA receptor gycine site functional partial agonist enhances cognition and produces antidepressant effects without the psychotomimetic side effects of NMDA receptor antagonists. Expert Opin Investig Drugs. 2014;23:243. [PMC free article] [PubMed]
13. Hu C, Chen W, Meyers SJ, Yuan H, Traynelis SF. Human GRIN2B variants in neurodevelopmental disorders. J Pharmacol Sci. 2016 doi: 10.1016/j.jphs.2016.10.002. [PMC free article] [PubMed] [Cross Ref]
14. Al-Hallaq RA, Conrads TP, Veenstra TD, Wenthold RJ. NMDA di-heteromeric receptor populations and associated proteins in rat hippocampus. J Neurosci. 2007;27:8334. [PMC free article] [PubMed]
15. Chazot PL, Stephenson FA. Molecular dissection of native mammalian forebrain NMDA receptors containing the NR1 C2 exon: Direct demonstration of NMDA receptors comprising NR1, NR2A, and NR2B subunits within the same complex. J Neurochem. 1997;69:2138. [PubMed]
16. Luo J, Wang Y, Yasuda RP, Dunah AW, Wolfe BB. The majority of N-methyl-D-aspartate receptor complexes in adult rat cerebral cortex contain at least three different subunits (NR1/NR2A/NR2B) Mol Pharmacol. 1997;51:79. [PubMed]
17. Vicini S, et al. Functional and pharmacological differences between recombinant N-methyl-D-aspartate receptors. J Neurophysiol. 1998;79:555. [PubMed]
18. Rauner C, Köhr G. Triheteromeric NR1/NR2A/NR2B receptors constitute the major N-methyl-D-asparate receptor population in adult hippocampal synapses. J Biol Chem. 2011;286:7558. [PMC free article] [PubMed]
19. Tovar KR, McGinley MJ, Westbrook GL. Triheteromeric NMDA receptors at hippocampal synapses. J Neurosci. 2013;33:9150. [PMC free article] [PubMed]
20. Williams K. Separating dual effects of zinc at recombinant N-methyl-D-asparate receptors. Neurosci Lett. 1996;215:9. [PubMed]
21. Williams K. Ifenprodil discriminates subtypes of the N-methyl-D-asparate receptor: selectivity and mechanisms at recombinant heteromeric receptors. Mol Pharmacol. 1993;44:851. [PubMed]
22. Hansen KB, Ogden KK, Yuan H, Traynelis SF. Distinct functional and pharmacological properties of triheteromeric GluN1/GluN2A/GluN2B NMDA receptors. Neuron. 2014;81:1084. [PMC free article] [PubMed]
23. Stroebel D, Carvalho S, Grand T, Zhu S, Paoletti P. Controlling NMDA receptor subunit composition using ectopic retention signals. J Neurosci. 2014;34:16630. [PubMed]
24. Cheriyan J, Balsara RD, Hansen KB, Castellino FJ. Pharmacology of triheteromeric N-methyl-D-asparate receptors. Neurosci Lett. 2016;617:240. [PMC free article] [PubMed]
25. Karakas E, Furukawa H. Crystal structure of a heteromeric NMDA receptor ion channel. Science. 2014;344:992. [PMC free article] [PubMed]
26. Lee CH, et al. NMDA receptor structures reveal subunit arrangement and pore architecture. Nature. 2014;511:191. [PMC free article] [PubMed]
27. Zhu S, et al. Mechanism of NMDA receptor inhibition and activation. Cell. 2016;165:704. [PMC free article] [PubMed]
28. Kawate T, Gouaux E. Fluorescence-detection size-exclusion chromatography for precrystallization screening of integral membrane proteins. Structure. 2006;14:673. [PubMed]
29. Goehring A, et al. Screening and large-scale expression of membrane proteins in mammalian cells for structural studies. Nat Protoc. 2014;9:2574. [PMC free article] [PubMed]
30. Paoletti P, Ascher P, Neyton J. High-affinity zinc inhibition of NMDA NR1-NR2A receptors. J Neurosci. 1997;17:5711. [PubMed]
31. Karakas E, Simorowski N, Furukawa H. Subunit arrangement and phenylethanolamine binding in GluN1/GluN2 NMDA receptors. Nature. 2011;475:249. [PMC free article] [PubMed]
32. Trabuco LG, Villa E, Schreiner E, Harrison CB, Schulten K. Molecular dynamics flexible fitting: A practical guide to combine cryo-electron microscopy and x-ray crystallography. Methods. 2009;49:174. [PMC free article] [PubMed]
33. Sun Y, et al. Mechanism of glutamate receptor desensitization. Nature. 2002;417:245. [PubMed]
34. Furukawa H, Singh S, Mancusso R, Gouaux E. Subunit arrangement and function in NMDA receptors. Nature. 2005;438:185. [PubMed]
35. Jin R, et al. Crystal structure and association behavior of the GluR2 amino-terminal domain. EMBO J. 2009;28:1812. [PubMed]
36. Sobolevsky AI, Rosconi MP, Gouaux E. X-ray structure, symmetry and mechanism of an AMPA-subtype glutamate receptor. Nature. 2009;462:745. [PMC free article] [PubMed]
37. Wyllie DJ, Livesey MR, Hardingham GE. Influence of GluN2 subunit identity on NMDA receptor function. Neuropharmacology. 2013;74:4. [PMC free article] [PubMed]
38. Gielen M, Siegler Retchless B, Mony L, Johnson JW, Paoletti P. Mechanism of differential control of NMDA receptor activity by NR2 subunits. Nature. 2009;459:703. [PMC free article] [PubMed]
39. Kazi R, Dai J, Sweeney C, Zhou HX, Wollmuth LP. Mechanical coupling maintains the fidelity of NMDA receptor-mediated currents. Nat Neurosci. 2014;17:914. [PMC free article] [PubMed]
40. Chen L, Dürr K, Gouaux E. X-ray structures of AMPA receptor-cone snail toxin complexes illuminate activation mechanisms. Science. 2014;345:1021. [PMC free article] [PubMed]
41. Dürr K, et al. Structure and dynamics of AMPA receptor GluA2 in resting, pre-open and desensitized states. Cell. 2014;158:778. [PMC free article] [PubMed]
42. Meyerson JR, et al. Structural mechanism of glutamate receptor activation and desensitization. Nature. 2014;514:328. [PMC free article] [PubMed]
43. Herguedas B, et al. Structure and organization of heteromeric AMPA-type glutamate receptors. Science. 2016;352:aad3873. [PMC free article] [PubMed]
44. Tajima N, et al. Activation of NMDA receptors and the mechanism of inhibition by ifenprodil. Nature. 2016;534:63. [PMC free article] [PubMed]
45. Hatton CJ, Paoletti P. Modulation of triheteromeric NMDA receptors by N-terminal domain ligands. Neuron. 2005;46:261. [PubMed]
46. Zhu S, Stroebel D, Yao CA, Taly A, Paoletti P. Allosteric signaling and dynamics of the clamshell-like NMDA receptor GluN1 N-terminal domain. Nat Struct Mol Biol. 2013;20:477. [PubMed]
47. Westbook GL, Mayer ML. Micromolar concentrations of Zn2+ antagonize NMDA and GABA responses of hippocampal neurons. Nature. 1987;328:640. [PubMed]
48. Paoletti P. Molecular basis of NMDA receptor functional diversity. Eur J Neurosci. 2011;33:1351. [PubMed]
49. Sirrieh RE, MacLean DM, Jayaraman V. Amino-terminal domain tetramer organization and structural effects of zinc binding in the N-methyl-D-aspartate (NMDA) receptor. J Biol Chem. 2013;288:22555. [PMC free article] [PubMed]
50. Romero-Hernandez A, Simorowski N, Karakas E, Furukawa H. Molecular basis for subtype-specificity and high-affinity zinc inhbition in the GluN1-GluN2A NMDA receptor amino-terminal domain. Neuron. 2016 doi: 10.1016/j.neuron.2016.11.006. [PubMed] [Cross Ref]
51. Traynelis SF, Cull-Candy SG. Proton inhibition of N-methyl-D-aspartate receptors in cerebellar neurons. Nature. 1990;345:347. [PubMed]
52. Meyerson JR, et al. Structural basis of kainate subtype glutamate receptor desensitization. Nature. 2016;537:567. [PMC free article] [PubMed]
53. Hansen KB, Furukawa H, Traynelis SF. Control of assembly and function of glutamate receptors by the amino terminal domain. Mol Pharmacol. 2010;78:535. [PubMed]
54. Furukawa H. Structure and function of glutamate receptor amino terminal domains. J Physiol. 2012;590:63. [PubMed]
55. Cabantous S, Terwilliger TC, Waldo GS. Protein tagging and detection with engineered self-assembling fragments of green fluorescent protein. Nat Biotechnol. 2005;23:102. [PubMed]
56. Coleman JA, Green EM, Gouaux E. X-ray structures and mechanism of the human serotonin transporter. Nature. 2016;532:334. [PMC free article] [PubMed]
57. Coleman JA, Green EM, Gouaux E. Thermostabilization, expression, purification, and crystallization of the human serotonin transporter bound to S-citalopram. J Vis Exp. 2016 doi: 10.3791/54792. [PubMed] [Cross Ref]
58. Zheng S, Palowcak E, Armache J-P, Cheng Y, Agard D. Anisotropic correction of beam-induced motion for improved single-particle electron cryo-microscopy. bioRxiv. 2016 doi: http://dx.doi.org/10.1101/061960.
59. Zhang K. Gctf: Real-time CTF determination and correction. J Struct Biol. 2016;193:1. [PMC free article] [PubMed]
60. Scheres SH. RELION: implementation of a Bayesian approach to cryo-EM structure determination. J Struct Biol. 2012;180:519. [PMC free article] [PubMed]
61. Tang G, et al. EMAN2: an extensible image processing suite for electron microscopy. J Struct Biol. 2007;157:38. [PubMed]
62. Grigorieff N. Frealign: An exploratory tool for single-particle cryo-EM. Methods Enzymol. 2016;579:191. [PubMed]
63. Arnold K, Bordoli L, Kopp J, Schwede T. The SWISS-MODEL workspace: a web-based environment for protein structure and homology modelling. Bioinformatics. 2006;22:195. [PubMed]
64. Pettersen EF, et al. UCSF Chimera–a visualization system for exploratory research and analysis. J Comput Chem. 2004;25:1605. [PubMed]
65. Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development of Coot. Acta Crystallogr D Biol Crystallogr. 2010;66:486. [PMC free article] [PubMed]
66. Chen VB, et al. MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr D Biol Crystallogr. 2010;66:12. [PMC free article] [PubMed]
67. DeLano WL. DeLano Scientific. San Carlos, CA, USA: 2002.