Optogenetic Stimulation Recruits Cortical Neurons in a Morphology-Dependent Manner

This is an impressive work that reused neuronal morphologies from Mainen and Sejnowski's 1996 study of the effect of cell shape on firing pattern. The original morphologies available to Mainen and Sejnowski did not include spines, but those authors knew that spines could account for a very large fraction (up to about 50%, depending on cell type) of total cell surface area, and that this could have major effects on the generation and propagation of electrical signals. They chose to represent the spine surface area contribution by applying scale factors to neurite lengths and diameters, following the "folding factor" approach (Major et al. 1994). That worked fine for them, but it distorts the entire cell in a complex, nonuniform way. Here's relevant source code from Mainen and Sejnowski (available from modeldb.science/2488; see cells/demofig1.hoc):

spine_dens = 1
// just using a simple spine density model due to lack of data on some
// neuron types.

spine_area = 0.83 // um^2 -- K Harris

proc add_spines() { local a
forsec $o1 {
a =0
for(x) a=a+area(x)

F = (L*spine_area*spine_dens + a)/a

L = L * F^(2/3)
for(x) diam(x) = diam(x) * F^(1/3)
}
}

Clearly the "folding factor" approach is inappropriate for any modeling study in which the anatomical coordinates, lengths, or diameters of a neuron's branches are important. I am not aware of anyone deliberately using it since the late 1990s. Instead, modelers either ignore spines or apply a scale factor to specific membrane capacitance and ion channel densities that reflects the contribution of spines to surface area.

So how did Berling et al. deal with spines? Were spines ignored, were they explicitly represented, were their contributions to surface area implemented by scaling specific membrane capacitance and channel densities, or was some other strategy used (hopefully, not the folding factor approach)? They didn't say, and the reviewers didn't ask, but this is an important question for a paper that addresses stimulation of neurons with quantitative methods. The authors did provide their source code, but there's a lot of it (30 megabytes) and it is inaccessible to anyone who is not somewhat expert in Python--and that includes many if not most experimentalists who should be the primary audience for this paper. The authors could easily answer this question, and that answer could enhance the scientific value of this paper.

Reference

Major G, Larkman AU, Jonas P, Sakmann B, Jack JJB. Detailed passive cable models of whole-cell recorded CA3 pyramidal neurons in rat hippocampal slices. Journal of Neuroscience 14: 4613–4638, 1994.