An investigation of NMDA receptor subunit pharmacology
N-Methyl-D-aspartate (NMDA) receptors are critically involved in synaptic transmission, neural development and various forms of neuronal plasticity including long-term potentiation (LTP) and long-term depression (LTD). They are also involved in the production of neuronal damage following excessive activation by glutamate released as a result of hypoxia or ischaemia. Each heteromeric receptor includes one or two NRl subunits, at least two of the four NR2A-D subunits and less usually the NR3AJB subunits. This study demonstrates that the putative NR2B subunit-containing NMDA receptor antagonist Ro 25-6981 potentiates the effects ofNMDA on rat hippocampal slices. The NR2A subunit antagonist PEAQX blocks the effects of NMDA alone and the potentiated response following Ro 25-6981 application. Furthermore, Ro 25-6981 was not neuroprotective as reported previously but unexpectedly precipitated excitotoxicity. The potentiating effect of Ro 25-6981 required around 20 minutes to become apparent, took a further 30 minutes to reach its maximum effect and was irreversible. It was not prevented by staurosporine (a broad-spectrum protein kinase inhibitor), okadaic acid (a potent inhibitor of the serine/threonine protein phosphatases types 1 and 2A) or anisomycin (a protein synthesis inhibitor). However, the potentiation was prevented by cyclosporin A (an inhibitor of Ca2+/calmodulin-dependent phosphatase 2B [calcineurin]). The results indicate that in an intact neuronal network, NR2B subunits tonically gate NR2A subunit-containing receptor function by a negative coupling mechanism involving ca1cineurin activation. NMDA receptor-dependent LTP induced by high frequency stimulation was prevented by PEAQX, an NR2A antagonist. Ro 25-6981 was unable to prevent L TP induction but was associated with a marginal reduction in the magnitude of LTP induced. There is evidence for the binding of homoquinolinic acid to an NMDAinsensitive novel binding site in the brain. This study investigated the pharmacology of homoquinolinate on the evoked field excitatory synaptic potential (fEPSP) recorded from the CAl area of rat hippocampal slices. Two NMDA receptor agonists, quinolinic acid 150/lM and homoquinolinic acid 2.5/lM, caused an approximately 50% inhibition of fEPSP slope. Paired-pulse studies suggested there might be a presynaptic component to this action that is independent of presynaptic adenosine Al receptor activation. The broad-spectrum EAA antagonist kynurenic acid and the NMDA receptor blockers 2-amino-5-phosphonopentanoic acid and dizocilpine could prevent the inhibition of fEPSP slope. None of these antagonists revealed any other NMDA-insensitive activity of homoquinolinic acid. The use of 2-carboxy-3-carboxymethylquinoline (CCMQ) to displace the reported NMDA-insensitive binding had no effect on either baseline fEPSP slope or the depression caused by homoquinolinic acid. It was also apparent that responses to homoquinolinic acid were blocked completely by the NR2A subunit-selective antagonist PEAQX, but not by the NR2B subunit-selective blocker Ro 25-6981. It was concluded that the novel binding site for homoquinolinic acid does not affect synaptic potentials in the hippocampus and that homoquinolinic acid appears to be a selective agonist at NMDA receptors that include the NR2A subunit. Although the NR2B agonist site may be maximally activated under normal conditions and therefore it is not possible to observe any additional effects upon fEPSP slope. This study next investigated the negative coupling between NR2B and NR2A subunit-containing receptors, combining the NR2A1B subunit selective agonist HQA with the NR2B and NR2A selective antagonists Ro 25-6981 and PEAQX. The negative coupling observed previously with applications of NMDA was also seen using HQA and QA. The potentiation of responses to HQA by Ro 25-6981 application was also associated with an enhancement of paired-pulse interactions. The subsequent application of PEAQX was able to block both the depression of fEPSP slope and the associated enhancement of paired-pulse interactions. The presence of a presynaptic element during applications of HQA alone and potentiated responses alike and the blockade of these effects by PEAQX suggests the NR2A subunit-containing NMDA receptor is responsible for the presynaptic effects acting either directly at presynaptic sites or indirectly at postsynaptic sites leading to the raising of a retrograde signal. The NR2B subunit in both its activated and antagonised state was associated with enhancements in paired-pulse interactions which suggest that it is not able to modulate directly the presynaptic element. However, whilst paired-pulse interactions are generally accepted to he presynaptic phenomena, it does not follow that postsynaptic effects cannot influence the appearance of changes in these interactions in field recordings. The absence of any observable difference between HQA, QA and NMDA results suggests that the NR2D subunit is not obviously involved in these processes.