Staying active = staying alive

Overactivation of NMDA (N-methyl-D-aspartate) receptors (NMDARs) can lead to excitotoxicity, possibly through the production of free radicals. However, Hardingham and colleagues now show that NMDAR activity in synapses, but not at extra-synaptic sites, in fact has neuroprotective effects through stimulation of an antioxidant enzyme system.

Initial experiments showed that inhibiting NMDAR activity in mice increased protein carbonyl levels, a marker of oxidative damage, and induced apoptosis in the cortex; this led the authors to study the relationship between NMDAR activation, oxidative stress and cell death in more detail in neuronal cultures. In these experiments they used hydrogen peroxide (H₂O₂, a highly reactive oxygen species (ROS)) to induce neuronal death, tetrodotoxin (TTX) to block action potentials, and the NMDAR antagonist MK-801 to block NMDAR activity.

First, the authors showed that both TTX and MK-801 increased H₂O₂-induced cell death. By contrast, stimulating synaptic activity by adding BiC/4-AP (a combination of a GABA_A receptor antagonist and a K⁺ channel antagonist) decreased H₂O₂-induced neuron death; this effect could be blocked by TTX and by MK-801, indicating that NMDARs mediate the neuroprotective effect of synaptic activity. Moreover, H₂O₂-treated neurons in which synaptic activity was enhanced showed less accumulation of ROS compared with neurons in which the NMDARs were blocked. Thus, synaptic activity, through activation of NMDARs, might protect against H₂O₂-induced neuronal death by preventing the accumulation of ROS.

The authors next investigated how synaptic NMDAR activation might decrease ROS levels. H₂O₂ can be reduced by the thioredoxin–peroxiredoxin (TRX–PRX) system; when this system is overwhelmed, it produces over-oxidized PRX, which interferes with the normal function of the TRX–PRX system and so increases free radical damage. Stimulating synaptic activity in H₂O₂-treated neurons with BiC/4-AP reduced over-oxidized PRX levels, and this effect was decreased by addition of MK-801, indicating that NMDA-mediated synaptic activity increases the activity of the TRX–PRX system in response to an oxidative insult.

Subsequently, the authors showed that synaptic NMDAR activation regulates the TRX–PRX system by downregulating the expression of a novel Forkhead box O target gene, Txnip . This gene encodes thioredoxin-interacting protein, which inhibits TRX. Indeed, overexpression of Txnip increased H₂O₂-induced neuron death. They also found another way by which synaptic activity affects the TRX–PRX system: stimulating synaptic activity with BiC/4-AP could reverse over-oxidation of PRX. Indeed, synaptic activity induced the expression of the genes for SRXN1 and SESN2 (through AP-1 and C/EBPβ, respectively), which are involved in the reduction of over-oxidized PRX. To summarize, by increasing SRXNB1 and SESN2 levels and decreasing TXNIP levels, synaptic NMDAR activity promotes the activity of the TRX–PRX system, resulting in neuroprotection against oxidative insults. The potential in vivo relevance of these findings was revealed in a mouse ischaemia-reperfusion model of neuron death: here, over-oxidized PRX was found around the occlusion site, indicating that the TRX–PRX system was overwhelmed.

Oxidative stress resulting from the accumulation of free radicals causes neuronal damage and death in several neurodegenerative diseases and in normal aging. The finding that NMDAR-mediated synaptic activity has neuroprotective effects against free-radical-induced damage evokes the question of whether this neuroprotective mechanism is disturbed in these conditions and whether it might be harnessed for therapeutic purposes. In addition, further research is required to determine why overactivation of extrasynaptic NMDARs results in the production of free radicals whereas activation of synaptic NMDARs protects against them.