Stochastic resonance and bifurcations in a heterogeneous neuronal population explain intrinsic oscillatory patterns in entorhinal cortical stellate cells

Abstract

Stellate cells in the medial entorhinal cortex manifest peri-threshold oscillatory patterns in their membrane voltage. Although different theoretical frameworks have been proposed to explain these activity patterns, these frameworks do not jointly account for heterogeneities in intrinsic properties of stellate cells and stochasticity in ion-channel and synaptic physiology. In this study, we first performed in vitro patch-clamp electrophysiological recordings from rat MEC stellate cells and found pronounced cell-to-cell variability in their characteristic physiological properties. We demonstrate that stochasticity introduced into either a simple nonlinear dynamical system (Hopf bifurcation) or into heterogeneous biophysical models yielded activity patterns that were qualitatively similar to peri-threshold oscillations in stellate cells. We developed five quantitative metrics for identification of valid oscillatory traces and confirmed that these metrics reliably captured the variable amplitude and non-rhythmic oscillatory patterns observed in our electrophysiological recordings. We analyzed traces from a stochastic Hopf bifurcation system for further confirmation on the reliability of these quantitative metrics in detecting oscillatory patterns. Apart from providing confirmation, these analyses provided a key insight about the manifestation of stochastic resonance in the stochastic bifurcation system, but not with theta-filtered noise. We then validated peri-threshold activity patterns obtained from a heterogeneous population of stellate cell models, with each model assessed with multiple trials of different levels and forms of noise (ion-channel, synaptic, and additive) at different membrane depolarizations. Strikingly, the validation process confirmed the manifestation of heterogeneous stochastic bifurcations across all models and revealed the detection of the highest number of valid oscillatory traces at an optimal level of each form of noise. The manifestation of stochastic resonance in this heterogeneous model population explained why intrinsic theta oscillations might not be observed under in vivo conditions where noise levels are typically high. Together, we provide several lines of evidence that demonstrate heterogeneous stochastic bifurcations as a unifying framework that fully explains peri-threshold oscillatory patterns in stellate cells and argue for state-dependence in the manifestation of these oscillations.