Slide 19

Figure 17A. Naundorf et al. plotted the AP onset rapidness versus AP onset span and stated that single compartment Hodgkin-Huxley models could not fit these parameters. However, we have found that Hodgkin-Huxley models that take into account the axonal spike initiation properties of cortical neurons do replicate these two features of spike generation in real neurons. Shown here is a plot of AP onset span (a measure of variability of membrane potentials for spike onset) versus the AP onset rapidness (A measure of the "kink" in the spike at onset). Solid blue dots are from real axons, while solid red dots are from real somata. Note that the somata have a faster onset rapidness than do axons (recorded at 30-60 microns from the soma, since this is where spikes are initiated). In addition, the onset span is on average larger for the soma than axon, during spike generation in response to a noisy current injection in the soma. Open circles are obtained from several of the models presented here in this seminar. Note that their values are within the range of real cells.

So far our results indicate that the following Hodgkin Huxley models predict the results of Naundorf et al:

• Mainen and Sejnowski (Neuron 15: 1427; Nature 382: 363).
• Johnston-Colbert-Pan-Migliore (Hoffman, Magee, Colbert, Johnston Nature 387: 869).
• McCormick et al. full model
• McCormick et al. simple model

These two Hodgkin-Huxley models also predict the "kink" in somatic spikes as a result of spike initiation in the axon.
Somatic spike threshold variability in these models has not yet been tested.

• Pare, Lang and Destexhe (Neuroscience 84: 377)
• Baranauskas and Martina (J. Neurosci. 26: 671-684)
  

We conclude Hodgkin and Huxley models readily replicate the results of Naundorf et al. without exotic Na+ channel cooperativity.