The neuronal action potential is one of the most important aspects of inter-neuronal communication and brain function. However, relatively little is known about the precise determinants of the action-potential threshold or its site of initiation. Using state-of-the-art electrophysiological recording techniques to study the biophysical mechanisms underlying action potentials, Colbert and Pan now shine new light onto this question [1xIon channel properties underlying axonal action potential initiation in pyramidal neurons. Colbert, C.M. and Pan, E. Nat. Neurosci. 2002; 5: 533–538Crossref | PubMedSee all References][1]. It was previously thought, owing to the high density of Na+ channels in the axon hillock, that action potentials initiate near the neuronal soma. However, experimental evidence for this potential mechanism comes predominantly from theoretical and labeling studies rather than from functional studies. Now Colbert and Pan report direct electrophysiological recordings from neocortical axons and propose that the lowest threshold for action-potential initiation is in the axon itself rather than in the axonal hillock.Using outside-out patch-clamp recordings from somatic, initial axon segment (<30 μm from the soma) and axonal (30–47 μm from the soma) neuronal areas, the authors observed qualitative differences in the currents evoked upon depolarization in these regions. In somatic patches, a combination of Na+ and delayed-rectifier K+ currents, as well as fast A-type K+ currents, could be evoked, whereas fast A-type K+ currents were absent in initial segment and axonal areas. In addition, it was observed that Na+ channels in the axon require significantly less depolarization for activation than do those in the soma or initial segment. This shift did not decrease gradually from soma to axon, but rather, changed sharply at the initial segment. Current densities were uniform throughout the initial segment before increasing two- to threefold in the axon.Using numerical models of neurons, and equivalent shifts in axonal Na+-channel activation, the authors found a decrease in axonal threshold for action-potential initiation with respect to the soma that corresponded well with the experimental data. However, modeling an increase in Na+-channel density in the initial segment only affected the axonal threshold at very high channel densities. In both situations, the site of initiation was biased away from the soma, although the shifted channel activation could better describe the experimental evidence for axonal action-potential initiation.This work provides a biophysical description of the axonal ion channels that might be involved in the initiation of action potentials. Axonal Na+ channels were found to require less depolarization for activation than did those in the soma or initial segment of the axon – this shift in activation kinetics could account for a lower threshold for action-potential initiation in the axon versus the soma. These data suggest that it is not the number of Na+ channels in the axon hillock but, rather, the properties of the Na+ channels in the axon that are responsible for the low axonal action-potential threshold. It is important not to overlook the role of other ion channels with respect to action-potential initiation and the diversity of neuronal physiologies; however, the data reported by Colbert and Pan in this study indicate that the likely determinants of neuronal action-potential initiation are located in the axon.