Structure-Based Models of NO Diffusion in the Nervous System - Computational Neuroscience: A Comprehensive Approach - page 97

The existence of a freely diffusing modulatory transmitter suggests a radically
different form of signalling in which the transmitter acts four-dimensionally in space
and time, affecting volumes of the brain containing many neurons and synapses. NO
cannot be contained by biological membranes, hence its release must be coupled
directly to its synthesis. Because the synthetic enzyme nNOS can be distributed
throughout the neuron, NO can be generated and released by the whole neuron. NO
is therefore best regarded as a 'non-synaptic' transmitter whose actions moreover
cannot be confined to neighbouring neurons [18, 33]. So not only can NO operate
over a large region, it can also mediate long-lasting changes in the chemical and
electrical properties of neurons within that volume [2, 41].
Because nNOS is a soluble enzyme and thus likely to be distributed throughout
a neuron's cytoplasm, the whole neuron surface is a potential release site for NO.
Thus the morphology of NO sources, as well as the presence of structured sinks
(such as blood vessels), will have a major influence on the dynamics of NO spread.
Understanding this dynamics is clearly a very important part of a more general un-
derstanding of volume signalling processes. However, because the NO molecule is
so small and non-polar it is very difficult to gather accurate empirical data in this
area. Therefore it is natural to turn to computational modelling to shed light on
volume signalling.
Somewhat ironically, many of NO's characteristics that complicate its empirical
investigation, make it much easier to model than many conventional neurotransmit-
ters whose large size and polarity make them impermeable to cell membranes. Thus
while these molecules also diffuse, their movement is restricted to the extracellu-
lar space near their release site and to model their spread would therefore require
accurate modelling of the morphology of the extracellular space and any local in-
homogeneities. In contrast, because of NO's minute size and non-polarity it can be
assumed as a good first approximation to diffuse isotropically through most brain
tissue and so the morphology of the synaptic cleft and other surrounding matter need
not be modelled. This means that complex factors such as tortuosity and viscosity,
which affect the movement of larger molecules, do not need to be included in the
governing equations.
This chapter demonstrates how to model NO diffusion from continuous structures
of biologically realistic dimensions. The central part of the chapter describes and
justifies in some detail the methods used to build such models. It then goes on to
show how these models provide insights into a number of salient functional questions
that arise in the context of volume signalling. Chief among these is how large a
volume can be affected, and for how long, from various NO generating neuronal
structures. Finally, work on more abstract computational models of neural networks
incorporating functionally active diffusing neuromodulators is introduced. These
networks serve as the nervous system of autonomous robots, generating sensorimotor
behaviours in these devices, and thus help to give insights into possible functional
roles of gaseous diffusing modulators in real nervous systems.