NMDA receptor subunits and associated signaling molecules mediating antidepressant-related effects of NMDA-GluN2B antagonism

Abstract

Drugs targeting the glutamate N-methyl-d-aspartate receptor (NMDAR) may be efficacious for treating mood disorders, as exemplified by the rapid antidepressant effects produced by single administration of the NMDAR antagonist ketamine. Though the precise mechanisms underlying the antidepressant-related effects of NMDAR antagonism remain unclear, recent studies implicate specific NMDAR subunits, including GluN2A and GluN2B, as well as the alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptor (AMPAR) subunit glutamate receptor interacting molecule, PSD-95. Here, integrating mutant and pharmacological in mice, we investigated the contribution of these subunits and molecules to antidepressant-related behaviors and the antidepressant-related effects of the GluN2B blocker, Ro 25-6981. We found that global deletion of GluA1 or PSD-95 reduced forced swim test (FST) immobility, mimicking the antidepressant-related effect produced by systemically administered Ro 25-6981 in C57BL/6J mice. Moreover, the FST antidepressant-like effects of systemic Ro 25-6981 were intact in mutants with global GluA1 deletion or GluN1 deletion in forebrain interneurons, but were absent in mutants constitutively lacking GluN2A or PSD-95. Next, we found that microinfusing Ro 25-6981 into the medial prefrontal cortex (mPFC), but not basolateral amygdala, of C57BL/6J mice was sufficient to produce an antidepressant-like effect. Together, these findings extend and refine current understanding of the mechanisms mediating antidepressant-like effects produced by NMDAR-GluN2B antagonists, and may inform the development of a novel class of medications for treating depression that target the GluN2B subtype of NMDAR.

Keywords: Depression; GluA1; GluN2B; Glutamate; PSD-95; Prefrontal cortex.

Figure 1. FST phenotype of loss-of-function NMDAR, AMPAR and PSD-95 mutations

(A) GluN1^INTER-KO mutants showed similar immobility to floxed controls (n=7–18 per genotype). (B) GluA1^KO mutants showed less immobility than WT controls (n=7–8). (C) PSD-95 mutants showed significantly less than WT controls (n=10–11 per genotype). Data are means ± SEM. *P<.05 WT versus KO

Figure 2. FST antidepressant-like effects of Ro 25-6981 in NMDAR, AMPAR and PSD-95 loss-of-function mutants

(A) Systemic Ro 25-6981 reduced FST immobility in C57BL/6J mice, relative to vehicle (n=6 per treatment). (B) In GluN1^INTER-KO mutant mice, Ro 25-6981 decreased FST immobility regardless of genotype (n=12 per genotype/treatment). (C) Ro 25-6981 decreased FST immobility in WT controls but not GluN2A^KO mutant mice, and vehicle-treated mutants were less immobile than vehicle-treated WT controls (n=10–14 per genotype/treatment). (D) In GluA1^KO mutant mice, Ro 25-6981 decreased FST immobility regardless of genotype, and mutants were less immobile than WT controls (n=6 per genotype/treatment). (E) Ro 25-6981 decreased FST immobility in WT controls but not PSD-95^KO mutant mice, and vehicle-treated mutants were less immobile than vehicle-treated WT controls (n=5–6 per genotype/treatment). Veh=vehicle, Ro=Ro 25-6981. Data are means ± SEM. *P<.05 Ro versus Veh, #P<.05 KO versus WT

Figure 3. Regionally localized FST antidepressant-like effects of Ro 25-6981

(A) Estimated cannula placements in the mPFC. (B) Microinfusion of Ro 25-6981 into the mPFC reduced FST immobility, as compared to vehicle-infusion (n=8–9 per treatment). (C) Intra-mPFC infusion of Ro 25-6981 did not affect entries or time spent in the light compartment of the light/dark exploration test (n=7–9 per treatment). (D) Estimated cannula placements in the BLA. (E) Microinfusion of Ro 25-6981 into the BLA did not alter FST immobility (n=7–9 per treatment). (F) Intra-BLA infusion of Ro 25-6981 did not affect entries or time spent in the light compartment of the light/dark exploration test (n=8–9 per treatment). mPFC=medial prefrontal cortex, BLA=basolateral amygdala, Veh=vehicle, Ro=Ro 25-6981. Data are means ± SEM. *P<.05 Ro versus Veh