Dynamic Neuronal Excitability

Abstract

Event Abstract Back to Event Dynamic Neuronal Excitability Christoph Kirst1, 2*, Julian Ammer3, 4, Felix Felmy3, 4, Andreas Herz3, 4 and Martin Stemmler3, 4 1 Max Planck Institute for Dynamics and Self-Organization, Germany 2 Bernstein Center for Computational Neuroscience Göttingen, Germany 3 Ludwigs Maximilians University Munich, Biology II, Germany 4 Bernstein Center for Computational Neuroscience Munich, Germany Neurons generally show two types of excitability [1,2]: Type I neurons support arbitrary long inter-spike-intervals, while type II neurons start firing with a non-zero frequency upon current injection. Here we show that a transition from type I to type II can be dynamically controlled in a large number of conductance-based neuron models (including Wang-Buszaki, Morris-Lecar, Connor- Stevens, Erisir et al.), e.g. by an increase in leak conductance. We mathematically prove that the bifurcation structure of this transition is organized by a degenerate Bogdanov-Takens-cusp bifurcation point of co-dimension 3 [3] which implies a switch from type I to type II for the spiking dynamics, a transition from integration to resonance near spike threshold, as well as a region of bistability of resting and regular spiking dynamics. We confirm these predictions experimentally for different neurons using dynamic patch clamp recordings to artificially change the leak conductance. Interestingly, the neuronal excitability type can also be switched dynamically via activation of shunting synapses, which we mimicked experimentally by bath application of GABA. These results imply that inhibitory cells can dynamically control the neuronal excitability type of postsynaptic neurons and as a consequence their synchronization properties. In particular, we show that inhibition can separately synchronize several coexisting sub-populations of excitatory neurons. Moreover, the maximal amount of synchrony in the network can be efficiently regulated by dynamically forcing the neurons into the region of bistability. In conclusion, inhibition-induced dynamic neuronal excitability switching provides a mechanism for flexible and activity controlled dynamic formation of synchronized neuronal cell assemblies.