Determination of the Dynamic Gain Function of Cortical Interneurons with distinct Electrical Types
by Ricardo Martins Merino
Date of Examination:2016-12-21
Date of issue:2017-04-25
Advisor:Prof. Dr. Fred Wolf
Referee:Prof. Dr. Fred Wolf
Referee:Prof. Dr. Walter Stühmer
Referee:Dr. Andreas Neef
Referee:Prof. Dr. Jochen Staiger
Referee:Prof. Dr. Siegrid Löwel
Referee:Dr. Dr. Oliver Schlüter
Files in this item
Name:MartinsMerino_Dissertation.pdf
Size:3.64Mb
Format:PDF
Abstract
English
In the living brain, individual neurons are constantly bombarded by thousands of synaptic inputs, which results in a fluctuating membrane potential. Neurons under such conditions are said to operate in a “fluctuation-driven regime”, in which stochastic incursions of the membrane potential to suprathreshold values result in the emission of action potentials. The process of transforming inputs, i.e., the fluctuating membrane potential, into an output, the spikes, is called “information encoding”. The dynamic gain function is a way to identify how this encoding takes place, by identifying the relationship between input frequencies and neuronal output. In this thesis, I sought to advance our understanding of how nerve cells encode information by means of two different approaches. In the first approach, a technical one, I characterized optogenetic tools that can be used to facilitate the traditionally laborious and time consuming determination of the neuronal gain function. Among the fast light sensitive channels available to date, chronos was the most promising. However, while it fulfilled all the basic requirements for a noninvasive fluctuating light stimulation, issues with respect to its level of expression in neurons hinder its applicability. In the second front, I used electrophysiological tools to effectively characterize the dynamic gain function of distinct electrical types of interneurons. I showed that fast spiking and adapting interneurons exhibit different frequency preferences, and that the correlation time of the noise input differently affects the gain curves of these cells. In the fast noise regime, adapting neurons exhibited a low-pass filter-like behavior, with peak gain in the theta range (1-10 Hz), while fast spiking cells showed a band-pass filter behavior with strong resonance in the 100-200 Hz band. Interestingly, in the slow noise regime, while fast spiking gain behavior qualitatively did not change, adapting interneurons exhibited a band-pass-like behavior, with peak at 100 Hz. In order to further characterize the gain of fast spiking neurons, these cells were subdivided into three categories: continuous, delayed, and stuttering. The gain calculation of each of these three subtypes showed that, while in the fast regime their responses were considerably similar, in the slow regime they exhibited distinct resonance peaks, with a considerable variation of the gain at the peak. To the best of my knowledge, this is the first time that the gain of inhibitory interneurons is characterized in the noise-driven regime.
Keywords: Interneurons; Frequency-response function; Gain function; Optogenetics; Electrophysiology