Reliable neuromodulation from adaptive control of ion channel expression

Abstract

Neuromodulation significantly alters neuronal activity and the responsiveness of both neurons and circuits to external inputs by adapting ion channel expression. However, basal ion channel expression is highly variable in neurons, even those with similar functions, which poses the question of how neuromodulation can act reliably. In this paper, we exploit the biophysical structure of neurons and the properties of neuromodulation-induced intracellular signaling to test whether reliable neuromodulation could be achieved by an intracellular control system adapting ion channel expression. The proposed controller has the typical structure of a linear adaptive loop that tunes cellular feedback gains determining neuronal excitability. The feedforward block transforms the neuronal feedback gain reference trajectory into an ion channel expression reference trajectory, while the feedback block, in the form of a simple PI controller, tracks the reference. Both blocks are biologically grounded, yet simple and mathematically tractable. We show that such a simple and biologically grounded control scheme can explain how reliable neuromodulation could be achieved in highly variable neurons. These results illustrate how a complex and highly nonlinear control problem can be tackled by a simple, biologically plausible control loop involving only a few variables.