GluN2D-containing NMDA receptors regulate dentate gyrus function by facilitating granule cell activity and mediating synaptic plasticity

This study demonstrates that GluN2D-containing NMDA receptors are tonically active, extrasynaptic receptors in dentate gyrus granule cells that promote neuronal firing, mediate synaptic plasticity through lateral diffusion and GluD1 facilitation, and are essential for spatial memory.

The Gist

The Big Picture: The Brain's "Hidden Switch"

Imagine your brain is a massive, bustling city. One of the most important neighborhoods in this city is the Hippocampus, specifically a district called the Dentate Gyrus. This district acts like the city's "security checkpoint" or "sorting office." It takes in raw information from the outside world, filters it, and decides what is important enough to send to the "archive" (long-term memory).

Inside this district, there are millions of tiny workers called Granule Cells. Their job is to fire electrical signals (like sending a telegram) to pass information along.

For a long time, scientists knew about two main types of "switches" (receptors) that help these workers fire: the GluN2A and GluN2B switches. They are like the heavy-duty, well-known main power lines of the city. But there was a third, mysterious switch called GluN2D. It was known to exist, but scientists thought it was mostly a "construction worker" that only showed up when the brain was a baby, and then disappeared.

This paper asks a simple question: Does this mysterious GluN2D switch still have a job to do in the adult brain, and if so, what is it?

The Discovery: The "Sleeping Giant" Wakes Up

The researchers found that GluN2D isn't gone; it's just been hiding in the background.

1. It's Always On (Tonic Activity):
   Think of GluN2D receptors as a low-level hum or a background heater in the Granule Cells. Even when the cell isn't being actively stimulated, these receptors are slightly "on," keeping the cell warm and ready to fire.

   • The Experiment: When the scientists used a special "off-switch" (a drug called DQP-1105) to turn off GluN2D, the Granule Cells became sluggish. They stopped firing as often. It was like turning off the background heater in a house; the rooms got cold, and the people inside (the signals) stopped moving around.

2. It Lives Outside the Front Door (Extrasynaptic):
   Most brain switches sit right at the "front door" (the synapse) where messages are exchanged. GluN2D, however, lives in the backyard (extrasynaptic). It doesn't usually help with the standard "hello" messages between cells. Instead, it acts like a safety net or a battery charger, keeping the cell ready for action.

The Magic Trick: Turning a Background Hum into a Superpower

The most exciting part of the paper is what happens when the brain needs to learn something new.

Imagine you are trying to remember a specific route to a new coffee shop. Your brain needs to strengthen the connection between the Granule Cells and the incoming information. This is called Long-Term Potentiation (LTP)—basically, turning a dirt path into a superhighway.

• The Old Way: Scientists used to think you needed to blast the system with high-frequency electricity (like a sledgehammer) to build this highway.
• The New Way: This paper shows that if you use a natural rhythm (a specific pattern of "burst" firing that mimics real-life thinking), you can build the highway.
• The GluN2D Role: When this natural rhythm happens, the "backyard" GluN2D receptors (the ones sitting outside the front door) run over and join the party. They physically move from the backyard to the front door (a process called lateral diffusion).
  • The Analogy: Imagine a construction crew (GluN2D) waiting in the parking lot. When a VIP arrives (the learning signal), the crew rushes to the front door, grabs a ladder, and helps build a bridge. Without them, the bridge never gets built.

The GluD1 Connection: The "Velcro"

Once the GluN2D receptors move to the front door, how do they stay there? They need something to hold them in place.

The researchers found that another protein, called GluD1, acts like Velcro. It grabs the GluN2D receptors and sticks them firmly to the synapse so they can do their job of strengthening the memory. If you remove the Velcro (by deleting the GluD1 gene), the GluN2D receptors fall off, and the memory bridge collapses.

The Real-World Test: Getting Lost in the Maze

To prove this wasn't just a lab trick, the scientists tested mice where they had deleted the GluN2D gene specifically in the Granule Cells.

• The Test: They put the mice in a room with two objects.
  • Test 1 (Object Recognition): "Do you remember seeing this object before?" The mice with no GluN2D passed this easily. They could recognize the object.
  • Test 2 (Object Location): "Do you remember where this object was placed?" The mice with no GluN2D failed miserably. They couldn't remember that the object had moved.

The Takeaway: GluN2D is essential for spatial memory (remembering where things are), which is a key function of the Dentate Gyrus. Without it, the brain's "sorting office" gets confused about locations.

Summary: Why This Matters

This paper changes how we see the adult brain. We thought the GluN2D switch was a relic of childhood. Instead, it turns out to be a critical, active player in how we learn and remember.

• It keeps the lights on: It keeps brain cells ready to fire.
• It builds the bridges: It rushes to the front door to help strengthen memories when we learn something new.
• It holds the memory: It works with GluD1 to make sure those memories stick.

If this system goes wrong, it could explain why people have trouble with spatial memory or why certain brain disorders (like schizophrenia or epilepsy) occur. It's like realizing that the "background hum" in your house was actually the main power generator all along.

Technical Summary

1. Problem Statement

N-Methyl-D-aspartate receptors (NMDARs) are critical for synaptic plasticity, learning, and memory. While the roles of the GluN2A and GluN2B subunits in the adult forebrain are well-characterized, the function of the GluN2D subunit in the mature brain remains poorly understood. Although GluN2D expression declines postnatally, it persists in specific regions like the hippocampus. Previous studies suggested GluN2D is primarily expressed in inhibitory interneurons and mediates tonic currents, but its role in excitatory dentate granule cells (GCs) and its contribution to synaptic plasticity and behavior in the mature dentate gyrus were unknown. Furthermore, the mechanisms by which GluN2D-containing receptors might be recruited to synapses during plasticity were unclear, particularly given that GluN2D lacks the PDZ-binding motifs typically required for synaptic anchoring.

2. Methodology

The authors employed a multi-faceted approach combining pharmacology, genetics, electrophysiology, and behavioral assays in both rats and mice:

• Pharmacological Tools: Used selective GluN2C/GluN2D antagonists (DQP-1105 and QNZ46) to isolate GluN2D-mediated effects. Non-selective NMDAR antagonist D-APV was used as a control.
• Genetic Models:
  • Conditional Knockout (cKO): Generated Grin2d cKO mice by injecting an AAV5-CaMKII-Cre-mCherry virus into the dentate gyrus of Grin2d floxed mice to selectively delete GluN2D in excitatory granule cells.
  • Global Knockout: Used Grid1 (GluD1) KO mice to investigate receptor retention mechanisms.
• Electrophysiology:
  • Whole-cell Patch-clamp: Recorded from dentate granule cells in acute hippocampal slices.
  • Tonic Activity: Measured changes in holding current and action potential firing in the presence of AMPA and GABA blockers.
  • Synaptic Transmission: Recorded NMDAR-mediated excitatory postsynaptic currents (NMDAR-EPSCs) from medial perforant path (MPP) inputs.
  • Plasticity Induction: Applied a Burst Timing-Dependent Plasticity (BTDP) protocol (pre-post pairing of 6 pre-pulses at 50 Hz and 5 post-pulses at 100 Hz) to induce Long-Term Potentiation (LTP) of NMDAR transmission.
  • Lateral Diffusion Assay: Used an antibody cross-linking strategy (injecting anti-GluN2D antibodies into the dentate gyrus) to immobilize surface receptors and test if lateral diffusion is required for LTP.
• Behavioral Assays:
  • Object Location Memory (OLM): Tested spatial memory by moving a familiar object to a new location.
  • Object Recognition Memory (ORM): Tested recognition memory by replacing a familiar object with a novel one.
  • Controls: Open Field Test (locomotion) and Elevated Plus Maze (anxiety).

3. Key Contributions & Results

A. GluN2D Receptors are Tonic and Extrasynaptic in Granule Cells

• Tonic Activation: Bath application of DQP-1105 significantly reduced action potential firing and holding current in dentate granule cells. This effect was abolished by D-APV and was absent in Grin2d cKO mice, proving that GCs express tonically active, extrasynaptic GluN2D-containing NMDARs that facilitate neuronal excitability.
• Basal Transmission: Unlike in interneurons, GluN2D receptors do not contribute significantly to basal MPP-GC synaptic transmission (NMDAR-EPSCs were unaffected by DQP-1105).

B. GluN2D Mediates Burst-Timing-Dependent NMDAR-LTP

• Induction: A physiologically relevant pre-post burst pairing protocol induced robust LTP of NMDAR-mediated transmission at MPP-GC synapses.
• Dependence on GluN2D: This LTP was completely blocked by GluN2D antagonists (DQP-1105, QNZ46) and was absent in Grin2d cKO mice.
• Functional Consequence: Following LTP induction, brief bursts of MPP input elicited significantly more action potentials in granule cells. This "gain of function" was also blocked by GluN2D antagonists, linking receptor plasticity to increased neuronal output.

C. Mechanism: Lateral Diffusion and GluD1 Interaction

• Recruitment Mechanism: Since GluN2D antagonists blocked LTP without affecting basal transmission, the authors hypothesized that extrasynaptic receptors are recruited to the synapse during LTP.
• Lateral Diffusion: Artificially immobilizing surface NMDARs via antibody cross-linking abolished LTP, confirming that lateral diffusion of extrasynaptic GluN2D receptors to the synapse is the mechanism of potentiation.
• Synaptic Retention: Because GluN2D lacks PDZ-binding motifs (unlike GluN2A/B), the authors investigated GluD1 receptors. In Grid1 KO mice, NMDAR-LTP was significantly impaired. This suggests that GluD1 receptors facilitate the retention of GluN2D-containing NMDARs at the synapse, likely via transsynaptic complexes involving neurexin and cerebellin.

D. Behavioral Significance

• Spatial Memory: Grin2d cKO mice showed a significant deficit in Object Location Memory (OLM), failing to distinguish the moved object from the familiar one.
• Specificity: These mice performed normally in Object Recognition Memory (ORM), as well as in tests for locomotion and anxiety. This indicates that GluN2D in granule cells is specifically required for dentate gyrus-dependent spatial memory, not general memory or motor function.

4. Significance

This study fundamentally alters the understanding of NMDAR subunit function in the mature hippocampus:

1. Novel Role in Excitatory Neurons: It establishes that GluN2D is not just an interneuron subunit but is functionally critical in excitatory dentate granule cells for regulating firing thresholds and plasticity.
2. Mechanism of Plasticity: It identifies a unique mechanism for NMDAR-LTP where extrasynaptic receptors diffuse laterally to the synapse, contrasting with the typical exocytosis or synaptic retention models.
3. Molecular Scaffolding: It reveals a non-canonical role for GluD1 in retaining GluN2D receptors at the synapse, bridging the gap between the lack of PDZ motifs in GluN2D and the need for synaptic anchoring.
4. Clinical Relevance: Given the link between GRIN2D mutations and schizophrenia/epilepsy, these findings suggest that dysregulation of tonic GluN2D activity or its plasticity mechanisms could underlie cognitive deficits in these disorders. The specific impairment in spatial memory highlights the dentate gyrus as a critical locus for GluN2D-mediated cognitive processing.