Repolarization of the action potential in varicose axons in mammalian brain

Abstract

Most synapses in mammalian brain’s grey matter are formed en passant by varicosities, also called boutons, located on very thin unmyelinated axons. The boutons are usually found along the full path of typical grey matter axons (GMAs). They constitute ~50% of the GMA membrane and contain the machinery for transmitter release. The small diameters of GMAs (on average 0.17 um) has prevented intracellular electrical recordings and knowledge about electrical events in GMAs is therefore very limited. We have explored repolarization of the action potential (spike) in cerebellar parallel fibers in adult rat brain slices by utilizing a greasegap technique with which we can distinguish between fast (few ms) and slow (tens of ms) repolarization. The fast repolarization stops abruptly at a potential more positive than rest (’brake point’), and transits to a depolarizing after-potential (DAP) with a decay time constant of 65 ms on average. Our SFN 2012 abstract showed that TEA (0.3 - 1 mM) preferentially reduced the fast repolarization, while margatoxin (MgTX) reduced currents that controlled the DAP amplitude without influencing the fast repolarization. This suggested that two K+ channel families served separate functions in the repolarization of parallel fiber spikes. We have continued to study channels that determine the ‘break point’. The reason is that we observed, with intracellular granule cell recordings and electrical activation of the axon, that the most obvious change in the spike just before the axon started to burst during 4-AP wash-in was a change in the ‘break point’ in depolarizing direction. The mechanisms controlling bursts are interesting because bursts increase metabolic demands and transmitter release. With grease-gap recordings we found that the combination of TEA and MgTX moved the ‘brake point’ much more in depolarizing direction than the sum of the effects of the two compounds given alone. This means that TEA- as well as MgTX-sensitive currents could repolarize the spike, but repolarization failed when both of these currents were blocked. There was a small, fast repolarization left even with the two blockers, which was insensitive to increases in blocker concentration, and insensitive to the Ca2+ channel blocker Cd2+ (200 uM). This residual was blocked by 100 uM quinine, which blocks Kv2 channel, but also other ion channels. We conclude that MgTX-, TEA- and quinine-sensitive potassium channels can contribute to fast repolarization of the spike in cerebellar parallel fibers in the slice preparation.