Positive N-Methyl-D-Aspartate Receptor Modulation by Rapastinel Promotes Rapid and Sustained Antidepressant-Like Effects

Abstract

Background: Modulation of glutamatergic synaptic transmission by N-methyl-D-aspartate receptors can produce rapid and sustained antidepressant effects. Rapastinel (GLYX-13), initially described as a N-methyl-D-aspartate receptor partial glycine site agonist, exhibits rapid antidepressant effect in rodents without the accompanying dissociative effects of N-methyl-D-aspartate receptor antagonists.

Methods: The relationship between rapastinel's in vitro N-methyl-D-aspartate receptor pharmacology and antidepressant efficacy was determined by brain microdialysis and subsequent pharmacological characterization of therapeutic rapastinel concentrations in N-methyl-D-aspartate receptor-specific radioligand displacement, calcium mobilization, and medial prefrontal cortex electrophysiology assays.

Results: Brain rapastinel concentrations of 30 to 100 nM were associated with its antidepressant-like efficacy and enhancement of N-methyl-D-aspartate receptor-dependent neuronal intracellular calcium mobilization. Modulation of N-methyl-D-aspartate receptors by rapastinel was independent of D-serine concentrations, and glycine site antagonists did not block rapastinel's effect. In rat medial prefrontal cortex slices, 100 nM rapastinel increased N-methyl-D-aspartate receptor-mediated excitatory postsynaptic currents and enhanced the magnitude of long-term potentiation without any effect on miniature EPSCs or paired-pulse facilitation responses, indicating postsynaptic action of rapastinel. A critical amino acid within the NR2 subunit was identified as necessary for rapastinel's modulatory effect.

Conclusion: Rapastinel brain concentrations associated with antidepressant-like activity directly enhance medial prefrontal cortex N-methyl-D-aspartate receptor activity and N-methyl-D-aspartate receptor-mediated synaptic plasticity in vitro. At therapeutic concentrations, rapastinel directly enhances N-methyl-D-aspartate receptor activity through a novel site independent of the glycine coagonist site. While both rapastinel and ketamine physically target N-methyl-D-aspartate receptors, the 2 molecules have opposing actions on N-methyl-D-aspartate receptors. Modest positive modulation of N-methyl-D-aspartate receptors by rapastinel represents a novel pharmacological approach to promote well-tolerated, rapid, and sustained improvements in mood disorders.

Keywords: NMDA receptor; depression; rapastinel.

Figure 1.

Rapastinel produces an acute and sustained antidepressant-like effect. (A) Rapastinel and S-ketamine produce dose-dependent antidepressant-like responses in the rat forced swim test (FST; n = 6 rats per group, *P < .05, **P < .01, ***P < .001 vs vehicle, 1-way ANOVA, Bonferroni’s multiple comparison test). (B) Dose-dependent increase in extracellular levels of rapastinel in medial prefrontal cortex (mPFC) of awake rats (n = 3 rats). Dashed line indicates lower limit of quantitation. Data are mean ± SEM.

Figure 2.

Rapastinel enhances N-methyl-D-aspartate receptor (NMDAR) function in NMDAR subtype-expressing HEK and cultured rat cortical neurons. (A–D) Rapastinel enhances [³H] MK-801 binding in HEK cells expressing NR2A (A; n = 25 glycine, n = 7 rapastinel), NR2B (B; n = 25 glycine, n = 11 rapastinel), NR2C (C; n = 14 glycine, n = 8 rapastinel), and NR2D (D; n = 13 glycine, n = 8 rapastinel). (E) Rapastinel enhances the NMDA-induced intracellular calcium ([Ca²⁺]_i) increase in primary rat cortical neurons. The NMDAR-dependent functional impact of rapastinel and S-ketamine was determined by sequentially infusing cortical neurons with 10 μM NMDA, 10 μM NMDA + rapastinel (or S-ketamine), 10 μM NMDA, and 10 μM NMDA + 3 μM D-serine. (F) Enhancement and inhibition of 10 μM NMDA-induced [Ca²⁺]_i increase is concentration specific with rapastinel (n = 3–8), whereas S-ketamine blocks the response in a concentration-dependent fashion (n = 4–8). Data are mean ± SEM.

Figure 3.

Enhancement of N-methyl-D-aspartate receptor (NMDA)-activated currents by 100 nM rapastinel with or without magnesium. (A) Representative traces from cultured cortical neurons voltage-clamped at +40 mV or at –60 mV. (B) Quantification of current amplitude in A (Vm –60 mV: n = 7; Vm +40 mV: n = 3, *P < .05, paired t test). NMDA current amplitudes measured in magnesium-free solution were increased at both –60 mV and +40 mV following application of 100 nM rapastinel. (C) With magnesium, NMDA current amplitudes were increased following application of 100 nM rapastinel when held at +40 mV (n = 5, P = .03). Data are mean ± SEM.

Figure 4.

Rapastinel acts independent of the glycine coagonist site as a weak N-methyl-D-aspartate receptor (NMDA) coagonist or inhibitor. (A–D) The effect of 100 nM (A–B) and 1 μM (C–D) rapastinel on NMDA-induced calcium response in the presence of 10 μM NMDA and varying concentrations of D-serine (A,C; n = 5–10, *P < .05, **P < .01, paired t test), and 3 μM D-serine and varying concentrations of NMDA (B,D; n = 8–13, **P < .01, paired t test). Data are normalized to 10 μM NMDA+3 μM D-serine, which is considered maximum response (100%). (E) Dose-dependent effect of rapastinel on [Ca²⁺]_i changes elicited by 3 μM NMDA+3 μM D-serine in cortical neurons or by 100 nM glutamate+3 μM D-serine in NR2A- or NR2B-expressing HEK cells (n = 5–12). (F) Representative trace of NMDA (10 μM)-induced calcium response with 100 nM rapastinel in the presence of 300 μM 7-CK with or without 100 μM (2R)-amino-5-phosphonopentanoate. (G) Rapastinel (100 nM)’s effect on NMDA (10 μM)-induced calcium response is not abolished by 300 μM 7-CK or 10 μM MDL 105,519 (n = 5–12, P < .01 paired t test for NMDA+rapastinel+7-CK vs NMDA+7-CK or for NMDA+rapastinel+MDL vs NMDA+MDL) but is completely blocked by (2R)-amino-5-phosphonopentanoate. Data are mean ± SEM.

Figure 5.

Rapastinel functional site is different from the glycine binding site. (A) Glycine increased [³H] MK-801 binding in wild-type NR1-NR2B controls but not in NR1-NR2B-expressing HEK cells containing a loss of function mutation in the glycine binding site (F484A/T518L) (n = 18). (B) Rapastinel increased [³H] MK-801 binding in both wild-type and glycine mutant cells (n = 18, 2-tailed t test). The modulation of N-methyl-D-aspartate receptor (NMDA)-induced intracellular calcium response of rapastinel was abolished in HEK293 cells transiently transfected with a (C) R392E mutation in NR2A and (D) R393E mutation in NR2B. Data are mean ± SEM.

Figure 6.

Rapastinel potentiates N-methyl-D-aspartate receptor (NMDAR)-mediated excitatory postsynaptic currents (EPSCs) and acutely enhances long-term potentiation (LTP) in medial prefrontal cortex (mPFC) pyramidal neurons. (A–C) Increase in the amplitude of pharmacologically-isolated NMDAR-mediated EPSCs with 100 nM or 1 μM rapastinel (A; n = 6, P < .05, paired t test) with no change in NMDAR current kinetics (inset), paired-pulse response profiles in the mPFC measured as ratio of the second population spike response to the first (PS2/PS1) (B: n = 9, P > .20, paired t test), or frequency or amplitude of spontaneous miniature EPSCs recorded in mPFC pyramidal neurons (C; n = 6, P > .20, paired t test). (D) Enhancement of the magnitude of LTP induction in mPFC slices by theta burst stimulus trains (TBS) and treated with 100 nM rapastinel starting 20 minutes prior to TBS (n = 8, P < .01, Fisher’s least significant difference [LSD] test; fEPSP, field excitatory postsynaptic currents) compared with slices treated with 3 μM S-ketamine (n = 7) and untreated control slices (n = 8). Inserts are representative signal averages of 4 fEPSPs before (dotted grey) and 47 to 50 minutes after TBS with 100 nM rapastinel (blue), control (black), or 3 μM S-ketamine (green). (E) Effects of varying concentrations of rapastinel and S-ketamine on LTP of excitatory postsynaptic potentials (n = 6–9 slices, **P < .01, ***P < .001 vs untreated control slice LTP, Fisher’s LSD test). Data are mean ± SEM.