Björn M Kampa & Greg J Stuart
Division of Neuroscience, John Curtin School of Medical Research, Australian National University, Canberra, Australia

https://doi.org/10.36866/pn.58.29

Our current ideas suggest that memories are stored in the neural circuits of the brain via changes in the strength of their connections (Hebb, 1949). Long-term increases in synaptic strength are called long-term potentiation (LTP), whereas long-term decreases are called long-term depression (LTD). Interestingly, these two forms of plasticity depend on the precise timing of single action potentials in pre- and postsynaptic cells (Linden, 1999). Like many forms of synaptic plasticity, this so-called spike­timing dependent plasticity (STDP) is dependent on activation of N-methyl­D-aspartate (NMDA) receptors. These ionotropic glutamate receptors are blocked by external Mg2+ at resting membrane potentials, and require depolarisation to open (Mayer et al. 1984; Nowak et al. 1984). During STDP this depolarisation is thought to be supplied by postsynaptic action potentials that ‘backpropagate’ from the soma to the site of synaptic input in the dendritic tree.

Exactly how quickly action potentials can remove Mg2+ from NMDA receptor channels is unclear. Early studies investigating the kinetics of short interruptions in single channel openings during steady state changes in membrane potential concluded that Mg2+ block and unblock of NMDA receptors was extremely rapid (sub­millisecond). Recent studies, however, have shown a slow component of Mg2+ unblock during depolarising voltage steps (Spruston et al. 1995; Vargas-Caballero & Robinson, 2003; Kampa et al. 2004; Vargas-Caballero & Robinson, 2004). In addition, we have found that the relative amplitude of fast and slow components of Mg2+ unblock depends on the timing of depolarising voltage steps relative to the onset of glutamate applications (Kampa et al. 2004). Fitting a kinetic model to this data indicates that Mg2+ binding increases desensitisation of NMDA receptors, and reduces both the open channel probability and affinity for glutamate. These findings are consistent with a recent study by Vargas-Caballero and Robinson (2004) who also report that binding of Mg2+ enhances the rate for NMDA channel closure. Our kinetic model may also explain the original observations of Nowak et al. (1984)that single channel burst duration, as well as frequency, are decreased at negative holding potentials by external Mg2+ .

As a result, Mg2+ unblock becomes slower the longer the NMDA receptor has been blocked by Mg2+ after glutamate has bound. The physiological consequence of this is that depolarising voltage steps, like action potentials, that occur later in time will have a smaller effect on NMDA receptor activation than depolarisations that occur just after the release of glutamate (Fig. 1).

This finding is likely to have important implications for STDP. To test this, we decided to have a closer look at what happens to Mg2+ unblock of NMDA receptors during STDP. As it is not possible to record directly from synaptic NMDA receptors, we used brief (1 ms) applications of glutamate to nucleated patches to mimic synaptic activation of NMDA receptors. Dendritic recordings of membrane potential during pairing of excitatory postsynaptic potentials (EPSPs) and backpropagating action potentials provided realistic postsynaptic voltage waveforms. By combining the two, we were able to measure NMDA receptor currents similar to what would be expected to occur during STDP. During STDP-type protocols, we found that the time window for NMDA receptor activation by backpropagating action potentials was narrower than expected assuming Mg2+ unblock occurred instantaneously (Fig. 2). These results indicate that slow magnesium unblock of NMDA receptor channels increases the precision of the STDP timing window. This finding may help explain the short time window required for LTP induction (~10 ms) compared to the long time glutamate can stay bound to the NMDA receptor (~100 ms).

In summary, we have shown through the use of realistic dendritic voltage waveforms that backpropagating action potentials can deliver sufficient depolarisation to postsynaptic NMDA receptors to release the Mg2+ block inside the NMDA receptor channel.

Importantly, we find that the ability of backpropagating action potentials to activate NMDA receptors is increased if the timing of presynaptic glutamate release and postsynaptic firing is more synchronised.

References

Hebb DO (1949). The organization of behavior. Wiley, New York.

Kampa BM, Clements J, Jonas P & Stuart GJ (2004). Kinetics of Mg2+ unblock of NMDA receptors: implications for spike-timing dependent synaptic plasticity. J Physiol 556, 337-345.

Linden DJ (1999). The return of the spike: postsynaptic action potentials and the induction of LTP and LTD. Neuron 22, 661-666.

Mayer ML, Westbrook GL & Guthrie PB (1984). Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature 309, 261-263.

Nowak L, Bregestovski P, Ascher P, Herbet A & Prochiantz A (1984). Magnesium gates glutamate-activated channels in mouse central neurones. Nature 307, 462-465.

Spruston N, Jonas P & Sakmann B (1995). Dendritic glutamate receptor channels in rat hippocampal CA3 and CA1 pyramidal neurons. J. Physiol 482, 325-352.

Vargas-Caballero M & Robinson HP (2003). A slow fraction of Mg2+ unblock of NMDA receptors limits their contribution to spike generation in cortical pyramidal neurons. J. Neurophysiol. 89, 2778­2783.

Vargas-Caballero M & Robinson HPC (2004). Fast and slow voltage­dependent dynamcs of magnesium block in the NMDA receptor: the asymmetric trapping block model. J Neurosci 24, 6171-6180.