Extracellular stimulation to myelinated axon

[arkent]

Hi Ted,

I am trying to develop a simple axon model that represents nodes of Ranvier connected by "perfectly insulating" myelinated internodal segments. Sweeney channel mechanics are used at the nodes (sodium and leakage currents). I would like to apply stimulation to this model using the extracellular mechanism, with the extracellular potentials at the nodes of Ranvier calculated through interpolation within a FEM program (ANSYS Ansoft). Since the internodal myelinated segments are perfectly insulating, I did not add any channel mechanics or transmembrane capacitance to these segments, nor did I apply a non-zero extracellular voltage. My procedure code for creating such an axon is below:

Code: Select all

proc create_axon(){
	
	for i=0, Nnodes-1 {
	node[i]{
		nseg=1
		diam=Dnode
		L=Lnode
		Ra=rhoa
		cm=Cm
		insert sw //sweeney channel mechanics//
		ena = 35.64
		insert extracellular xraxial=1e9 xg=1e9 xc=0
		}
	}	
	
	for i=0, Nnodes-2 {
		internode[i] {
		nseg=1
		diam=Dinter
		L=Linter
		Ra=rhoa
		cm=0
		insert extracellular xraxial=1e9 xg=1e9 xc=0 e_extracellular=0
		}
	}
	
	
	//connect nodes//	
	for i=0, Nnodes-2 {
		connect node[i](1), internode[i](0)
		connect internode[i](1), node[i+1](0)
	}
	
}

The code is executing properly, but is it not clear whether the output is within reason (I am looking at whether action potentials are generated within a population of fibers near a stimulating electrode). Could you please review my methodology, and let me know if you think there are theoretical error(s) with this approach?

Much appreciated.

[ted]

Ah, yes, the Sweeney variation on the Chiu et al model.

The only real problems I see with your code are:
1. it doesn't declare a default section.
2. it violates the strong recommendation that the 0 end of a child section should be connected to the 1 end of its parent. That leads to all kinds of confusion because it produces models in which increasing range along a section does NOT mean increasing anatomical distance from the root node.

These are easy to take care of.

Code: Select all

create node[Nnodes], inter[Nnodes-1]
access node[0]
for i=0,Nnodes-2 connect inter[i](0), node[i](1)
for i=1,Nnodes-1 connect node[i](0), inter[i-1](1)

A very minor tip to reduce the amount of code you have to write: iterate over section names, instead of using indices.

Code: Select all

forsec "node" {
  ... assign node biophys and anat properties ...
}
forsec "inter" {
  ... assign internode biophys and anat properties ...
}

extracellular potentials at the nodes of Ranvier calculated through interpolation within a FEM program

If the extracellular medium is assumed to be nondispersive, and stimulation is monopolar or dipolar, the extracellular potential at each node will be the product of a stimulus current and a transfer resistance. If the stimulus is more complex (e.g. multiple electrodes driven by two or more currents with different time courses), the extracellular potential at each node will be the linear sum of such products (a different transfer resistance for each stimulus current waveform). There are relatively convenient ways for implementing this.