Imprecision in Precision Medicine: Differential Response of a Disease-Linked GluN2A Mutant to NMDA Channel Blockers

Mutations in N-methyl-d-aspartate receptors (NMDAR) subunits have been implicated in a growing number of human neurodevelopmental disorders. Previously, a de novo mutation in GRIN2A, encoding the GluN2A subunit, was identified in a patient with severe epilepsy and developmental delay. This missense mutation, which leads to GluN2A-P552R, produces significant dendrotoxicity in transfected rodent cortical neurons, as evidenced by pronounced dendritic blebbing. This injurious process can be prevented by treatment with the NMDA antagonist memantine. Given the increasing use of FDA approved NMDA antagonists to treat patients with GRIN mutations, who may have seizures refractory to traditional anti-epileptic drugs, we investigated whether additional NMDA antagonists were effective in attenuating neurotoxicity associated with GluN2A-P552R expression. Intriguingly, we found that while treatment with memantine can effectively block GluN2A-P552R-mediated dendrotoxicity, treatment with ketamine does not, despite the fact that both drugs work as open NMDAR channel blockers. Interestingly, we found that neurons expressing GluN2A-P552R were more vulnerable to an excitotoxic insult—an effect that, in this case, could be equally rescued by both memantine and ketamine. These findings suggest that GluN2A-P552R induced dendrotoxicity and increased vulnerability to excitotoxic stress are mediated through two distinct mechanisms. The differences between memantine and ketamine in halting GluN2A-P552R dendrotoxicity could not be explained by NMDA antagonist induced changes in MAP or Src kinase activation, previously shown to participate in NMDA-induced excitotoxicity. Our findings strongly suggest that not all NMDA antagonists may be of equal clinical utility in treating GRIN2A-mediated neurological disorders, despite a shared mechanism of action.

Neuronal signature of an antipsychotic response

D2 receptor blockade has been cited as a principal mechanism of action of all antipsychotic medications, but is poorly predictive of symptom improvement or neurophysiological responses recorded using human brain imaging. A potential hurdle in interpreting such human imaging studies arises from the inability to distinguish activity within neuronal subcircuits. We used single cell resolution imaging to record activity in distinct populations of medium spiny neurons in vivo within the mouse ventral striatum, a structure associated with schizophrenia symptoms and antipsychotic therapeutic efficacy. While we expected the antipsychotic haloperidol to excite D2 receptor expressing neurons, we report a strong cellular depression mediated by the hypofunctional NMDA channel, which may be mediated in part by the action of haloperidol on the sigma1 receptor. Altogether, the impact of haloperidol on Ca2+ events in D2 receptor expressing neurons predicted psychomotor inhibition. Our results elucidate mechanisms by which antipsychotics act rapidly in the brain to impact psychomotor outputs.