Custom 'ion' (USEION) problems with spatial decay

[tcleland]

Hi folks --

My student Jian Bai Li is developing an intracellular cascade model for which he has defined a GTP variable using the USEION statement in NMODL. The problem he has encountered is that GTP concentrations do not decay in space along multicompartmental sections. I'm new to the definition of custom ion variables, but assume that our GTP needs to be further defined somehow -- e.g. provided with a spatial decay constant and/or other parameters that have built-in defaults in the case of calcium (for example) but not for user-defined ions. But I don't know how to do this. I'm looking, but I thought I'd ask here too. Can anybody point me towards a howto for properly creating custom concentration/charge variables using USEION?

Here's Jian Bai's email to me:

"I wrote a GTP receptor (in MOD) that reads odorant concentration and writes to the concentration of an ion that I defined myself (called GTP). I inserted this receptor into the cilia of my OSN. When the cilia was of only 1 segment, the GTP receptor and the GTP ion concentration behaved normally. However, when the cilia was composed of more than 1 segments, the GTP ion concentration only behaved normally in the segment that the receptor was inserted into. In all other segments, the GTP ion concentration remained at 1. I searched the NEURON book and the internet for answers to my problem, and I still cannot figure out why the GTP receptor doesn't work for cilia with more than 1 segments. It seems as though the GTP ion is lacking a space decay mechanism that specifies how concentration changes as the ion diffuses down the cilium. Yet, NEURON does not have a command that specifies some sort of space decay constant for newly defined ions. Do you know what the problem might be? Thank you very much for your help!"

Thanks a lot for any help & suggestions --

Thom

[ted]

By default the only stuff that spreads down the length of a section is charge. If you want
longitudinal diffusion of any particular solute, the accumulation mechanism for that solute
must be specified with a KINETIC block that contains the LONGITUDINAL_DIFFUSION
keyword. For further discussion of implemeting models of diffusion with kinetic schemes,
and the use of LONGITUDINAL_DIFFUSION, see
9.10 Example 9.8: Calcium diffusion with buffering (pp. 245 et seq. in The NEURON Book).
For slightly different NMODL code see
c:/nrn59/examples/nrniv/nmodl/cadifusl.hoc (MSWin)
or
nrn-x.x/share/examples/nrniv/nmodl/cadifusl.mod
(in NEURON's source code).

Note that chemical signals tend to have a much shorter effective length constant than
electircal signals do. Therefore nseg will probably have to be much larger than if you were
interested only in electrical signals. As always, choose odd values for nseg, and test
for adequate spatial accuracy by following this protocol:

Code: Select all

run a simulation with nseg = 1 and note results
repeat
  triple nseg
  run a new simulation and compare with previous results
until no significant change is seen
divide nseg by 3

[tcleland]

Thanks Ted!
Thom

[ted]

Thanks for using NEURON, citing it in your papers, and asking good questions, Thom.
Good questions help us develop new documentation.

--Ted