Spike Train Analysis
Neuronal network cultures on microelectrode (sometimes called multielectrodes) arrays are an excellent tool for the study
of the mode of action of neuronal networks
as they occur in the central nervous system (CNS) of human beings or animals. Since this technology allows the simultaneous
recording of action potentials from numerous individual neurons over an extended period of time, new insights into the
functionality of such networks are possible. Therefore the timestamps of all action potentials, or so-called spikes, are recorded
during an experiment for each neuron over time as the 'spike train'. It is possible to study multiple neurons, up to 256, of
a neuronal network in parallel.
|
|
|
Characteristic drug-induced network patterns of 15 simultaneously recorded spinal cord neurons over a 30s period.
|
Methods of spike train analysis are used to assess the activity patterns of neuronal networks. With the aid of the program
PATTERN EXPERT squid we can calculate more than 200 different spike train parameters.
Based on these 200 spike train features for an experimental episode, the changes of the activity patterns caused by a substance
can be clearly identified, i.e. classified for every substance. This documents the reproducibility of the observed activity patterns.
Feature categories
Based on spike train analysis, we describe five fundamental categories: general activity, general activity in the bursts, synchronicity,
oscillations and the internal structure of the bursts, and the bursts shape.
General activity
The spike rate, i.e. the number of spikes per unit of time, for a neuron or the average spike rate for all neurons, is one of the most
important features. The average spike rate usually correlates very well with the applied concentration of a substance.
Curve fitting yields EC10 (10% effective concentration), EC50, EC90 and the Hill coefficients, nH.
The dose-response curve for the average spike rate can show a biphasic course, clearly indicating the effects of several
mechanisms of actions of the applied substance.
General activity in the bursts
Spike trains of action potentials are clearly structured over time into phases of rest and of activity, so-called bursts. The mean
frequency of the spikes in bursts describes the strength of the bursts; a further parameter in this category is the ratio of the
spikes contained in burst periods to all spikes. The latter parameter describes very well how clearly the bursts can be
separated. Different tissue cultures, e.g. from the frontal cortex and from the spinal cord, differ significantly in this feature.
|
|
|
30 sec of spontaneous network activity recorded from 20 single neurons;
left: spinal cord; right: frontal cortex
|
Synchronicity
The synchronicity of activity patterns in neuronal networks is subject of intense investigation. The influence of substances on
the synchronicity is a very important characteristic phenomenon. These effects can be ideally studied with neuronal
networks on microelectrodes arrays. Since the action potentials can be recorded from several neurons simultaneously with
high precision in the microsecond range over a
longer time, the analysis of the synchronicity is possible with a high resolution
in time and per neuron. Synchronicity can now be analyzed and characterised at different time scales, which was not
previously possible.
Oscillations
Action potentials occur with clearly periodic patterns. The regularity of these periodic patterns is an additional characteristic of
the
effects of neuro-active substances.
Burst Shape
The burst structure, i.e. the temporal distribution of the spikes in a burst, is an essential feature in our analyses for describing
the changes of the activity patterns induced by substances.
Profiling
In connection with our pattern recognition methods, we select those
characteristic spike train features for individual substances or groups of substances that deliver dose-response curves with a high significance and informative value.
