N-Methyl- d -Aspartate Receptors Remain Viable Therapeutic Targets for Stroke

Introduction

Stroke is a major cause of morbidity and mortality in North America and worldwide . The mechanisms underlying neuronal injuries following stroke are still not fully understood, and involve multiple factors. One particularly well-characterized factor is excitotoxicity that results at least partly from the overactivation of N-methyl- d -aspartate subtype glutamate receptors (NMDARs) . However, this idea has been challenged by several failed stroke clinical trials . The reason underlying the discrepancy between basic research and clinical trials remains unknown. A few explanations have been suggested and these include, but are not limited to, (1) the inability to use NMDAR antagonists at protective doses due to the potential blocking of normal brain function and neuronal survival; (2) the inability to administer these drugs within the therapeutic window due to time required for patient transport and diagnosis; and (3) poor trial design and heterogeneity of the patient pool . Although the last factor stresses the need for better designed clinical trials, advancements in our understanding of distinct intracellular pathways linking NMDAR activation to neuronal death through experimental studies may allow scientists to develop more promising NMDAR-based stroke treatments that target specific death signaling pathways without affecting the normal functions of the receptor. This increased specificity not only translates into reduced side effects but also extends the therapeutic window in which the drug may be efficaciously administered.

In this chapter, we will address the roles of NMDARs in regard to the pathophysiology of stroke, and discuss how they can be implicated in early and late consequences. We will also discuss the implications of some existing and future pharmacological agents aimed at NMDARs as potential stroke therapeutics (see Fig. 36.1 for a graphical summary).

N-Methyl- d -Aspartate Subtype Glutamate Receptors at the Acute Phase of Stroke

NMDARs are ionotropic glutamate receptors highly permeable to cations. They mediate rapid excitatory synaptic transmission. The activation of NMDARs requires two concomitant stimuli: binding of glutamate and of the coagonist glycine, and membrane depolarization to remove magnesium cations blocking the channel pore at resting potential. Many neurological processes depend on the physiological activation of NMDA receptors, including synaptic plasticity, learning, memory, and neuronal survival. The NMDAR is a heterotetramer containing subunits from three different families, GluN1, GluN2 (A-D), and GluN3 (A-B). Every tetramer is composed of two GluR1 subunits, and at least two GluN2 subunits. It is now known that the composition of the tetramer influences ion permeability, location of the receptor on the membrane, and downstream effectors coupling . GluN2A-containing NMDARs are mostly located at the synapse, thus playing a physiological role in excitatory neurotransmission. Their activation is associated with many brain functions and neuronal survival. On the contrary, GluN2B-containing NMDARs are preferentially extrasynaptic and are implicated in mediating neuronal death during early stage of neurodevelopment and under pathological conditions.

The physiological activation of NMDARs is mandatory for basal brain functioning, whereas their pathological overactivation has been linked to a specific type of cell death, referred to as excitotoxicity. Excitotoxicity has been described in various neurological diseases; its role in ischemic stroke is prominent and constitutes one of the early steps of the pathophysiological cascade initiated following vascular occlusion. Indeed, downstream from the blocked artery, the drastic reduction in blood flow and subsequent reduction in oxygen and nutrient supplies puts a great stress on cells. One of the initial responses in neurons is an increased release of glutamate. This, associated with the compromised glutamate uptake due to the failure of the energy-dependent glutamate transporters, leads to extracellular glutamate accumulation. The accumulation results in a dual effect contributing to neuronal death: (1) a major intracellular influx of calcium responsible for the activation of proteases, nucleases, and other enzymes that will degrade substrates and produce metabolites in excess like NO or reactive oxygen species; and (2) a strong and prolonged stimulation of receptors, including extrasynaptic NR2B-containing NMDARs, followed by the activation of various downstream effectors .

Given the magnitude of this phenomenon, NMDARs appear to be significant pharmacological targets for acute stroke. Dizocilpine (MK-801) is a noncompetitive antagonist of NMDARs. It was found to be neuroprotective when administered preventively and up to 2 h following experimental stroke. The demonstration of its neuroprotective potential in rodents led to the development of numerous antagonists with different pharmacological profiles: competitive at the glutamate-binding site (selfotel), competitive at the glycine-binding site (gavestinel), ion channel blockers (aptiganel), and NR2B selective (CP-101,606). Promising preclinical results led to several clinical trials in stroke and in traumatic brain injury; however, these trials proved to be disappointing, as no benefit was shown. Furthermore, it appeared that the widespread NMDAR blockade was responsible for serious psychotomimetic and cardiovascular side effects, undetected in rodents. Some of the trials (selfotel and aptiganel) were even stopped prematurely for safety reasons . In addition to these concerns, the use of these molecules is limited by a very narrow time window, as NMDAR activation is a very early event and these interventions become useless once the downstream pathways coupled to the activated receptor are initiated. It is not surprising in that context that the trials of nonspecific NMDAR antagonists were then dropped .

However, one antagonist, memantine, currently used to treat patients with Alzheimer disease (AD), still seems of interest because of its distinct pharmacodynamic profile. Memantine is an NMDAR antagonist. As such, it binds only to the activated receptor, preventing the ion flux to go through the channel, and also dissociates rapidly from the closed channel. Under physiological conditions, the synaptic NMDAR ion channel opens only transiently by the presynaptically released glutamate, and hence is not substantially affected by memantine. Under excitotoxic conditions, elevated extracellular glutamate concentrations keep the NMDA channel open, and so accessible to memantine. Thus, memantine interferes with NMDARs in case of glutamate overload under pathological conditions and does not impair their physiologic activation. The available experimental data are in favor of a neuroprotective effect of memantine at the acute phase of stroke . At the beginning of 2016, there were two clinical trials evaluating memantine in stroke recovery in the clinicaltrials.gov database (NCT02144584; NCT02535611).

With the exception of memantine, several large-scale stroke clinical trials have failed to confirm the efficacy of NMDAR antagonist therapies . Research for effective NMDAR-based stroke therapies has taken a new direction. To be safe, efficient, and usable within a time window allowing potential clinical use, perhaps the target should not be the receptor itself, but the latter steps of its downstream pathways. This new strategy sparked a lot of new research interest in trying to decipher the underlying mechanisms involved and to design new therapeutic approaches either promoting NMDAR-dependent neuronal survival or halting death pathways.