Objective: Dopamine neurons (DAs) of the substantia nigra degenerate in Parkinson’s disease and are disrupted in Huntington’s disease. Here we propose single cell computational modeling methods that can capture the variety of electrophysiology within this neuronal population, providing a method to understand the transition to pathology.

Background: Degeneration caused by disease may derive from altered electrophysiology that pushes DAs beyond their normal phenotype. For example, sustained engagement of calcium channels due to low-threshold L-type channel activity results in relatively high basal calcium flux and oxidative stress, constituting a risk if normal neuronal behavior is disrupted in disease.

Methods: In vitro data, which are variable due to differential channel regulation, can be used to understand the balance of ion channel properties required for normal DA function. Here we demonstrate a new evolutionary algorithm that generates a population of DA models with the full range of observed phenotypes. Parameters that generate these phenotypes vary widely. We tuned models of ion channels unaffected by suprathreshold channel blockers and fitted voltage to subthreshold oscillations. We performed a principal component (PC) analysis on the parameter sets in the population after this optimization, which identified combinations of parameters that covary and explain the range of acceptable models.

Results: Certain PCs were highly correlated with subthreshold oscillation amplitude and frequency, and we were able to identify separable, low-dimensional parameter combinations capable of fully controlling each oscillation feature. We term these axes in parameter space “functional regulatory units”. These units provide a specific axis through parameter space which efficiently controls key features of the model. Specifically, subthreshold oscillation features in our dopamine neuron model can be fully controlled by varying several parameters related to L-type channels, calcium buffering and extrusion, and calcium-dependent potassium channels.

Conclusions: By identifying how to combine changes to multiple parameters simultaneously to produce a desired functional change, we propose a method to rescue pathological phenotypes. We propose that this method can provide great insight into neuronal function, dysfunction, and therapeutic design by identifying functional regulatory units for control and rescue of behavioral features of neuron models.

« Back to 2017 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/efficient-control-of-dopamine-neuron-physiology-for-rescuing-disease-phenotypes/