Until the 1980s, researchers didn't consider the excitatory amino acid N-methyl-D-aspartate (NMDA) to be a receptor in mammalian brains. That began to change in 1982, when Joerg Lehmann and Bernard Scatton studied the interaction between excitatory amino acids and striatal cholinergic neurons in vitro (Brain Res, 252:77-89, 1982). They suggested that excitatory amino acids such as glutamate act at NMDA receptors on the dendrites of striatal cholinergic interneurons.

Over the years, this line of research eventually demonstrated that in healthy human subjects, NMDA antagonists could cause the negative symptoms and cognitive impairments observed in schizophrenia (Cell Molec Neurobiol, 26:365-84, 2006). That led to the idea that NMDA agonists such as glutamate could combat those symptoms - leaving drug developers with an option other than the dopamine D2 receptors, which essentially all antipsychotics target.

Indeed, researchers say that in the pathophysiology of schizophrenia, NMDA may be a more primary target than dopamine. "When you give animals ketamine or phencyclidine, it causes a disinhibition of pyramidal flow of glutamatergic output; this causes an increase in dopamine release," says Joseph Coyle, a professor of psychiatry and neuroscience at Harvard Medial School. "The dopamine hypothesis is that [an increase in dopamine levels] causes psychosis." But from the effect of ketamine on the NMDA receptor, it seems that "the increase in dopamine release is a downstream effect."

The first compound targeting metabotropic glutamate receptors (mGluR2/3), one kind of NMDA receptor, is currently in Phase II trials and has shown positive results so far. The compound, Eli Lilly's LY2140023, proved as effective as olanzapine, the study's positive control, with fewer extrapyramidal symptoms and less weight gain, according to Lilly's Sandeep Patil, who worked on clinical development and design for the study (Nature Med, 13:1102-7, 2007). Patil notes, however, that the scales for efficacy testing are not linear and lack definitive biomarkers.

Patil, Coyle and others say that researchers must test the long-term effects of this class of compounds to determine its role in the pathophysiology of schizophrenia. One obstacle will be the difficulty in modulating the glutamate system: Overstimulation leads to excitotoxicity, whereas blocking the system can lead to cell death via apoptosis. "It's a highly regulated system," says Coyle. "If it goes out of whack, it can do a ton of damage."