Modelling of Spontaneous Activity and its Developmental Role in the Immature Vertebrate Retina

Lead Research Organisation: University of Edinburgh
Department Name: Sch of Informatics

Abstract

It has been shown that the development of the brain before birth, when stimulation from the environment has little or no influence, relies heavily on spontaneous patterned activity in nerve cells. In the retina, which is the first processing stage for visual stimuli, spontaneous activity is visible as slowly propagating waves of activity, called retinal waves. Experimental studies suggest that retinal waves influence the connectivity of the retina and its projections to other brain areas that process visual information. The proposed study will investigate, using mathematical modelling and computer simulations, the mechanisms that generate retinal waves and their influence on retinal development. A key question will be whether the properties of retinal waves such as velocity, size and frequency, which change in the course of development, provide developmental cues. This research could contribute to the development of treatments of retinal degeneration, a disease where retinal cell death leads to a gradual loss of vision. By inducing developmental processes after the insertion of healthy tissue, an induced network restructuring could restore normal vision. The retina could further be a model for the understanding and treatment of developmental disorders resulting from a disruption of early spontaneous activity.

Technical Summary

A common feature of developing structures in the central nervous system is spontaneous, patterned activity. Experimental and theoretical evidence suggests an important role in the refinement of the neural circuitry, but the current models are still controversial. The immature retina spontaneously generates propagating waves of activity (retinal waves), which appear before the onset of normal vision. The aim of the proposed study is to describe retinal waves by means of mathematical models and to assess their influence on retinal development. Crucially, it has been shown that wave properties change in the course of development, which appears to be a consequence of the maturation of the retinal circuitry.

As a first step, a model of retinal waves will be developed to account for these distinct developmental stages. The model will consist of single neurons and their connectivity, and simulation of spontaneous activity patterns will be used to assess the influence of different local factors, such as properties of individual neurons, synaptic transmission and connectivity, on the global wave properties. In a second step, the model will be used to study the influence of waves on retinal development. Experimental data suggests a developmental role of retinal waves in dendritic remodelling of retinal ganglion cells, the output neurons of the retina. These processes, which are yet to be understood, are thought to lead to a refinement of retinal circuits responsible for the distinct properties of ganglion cell receptive fields such as segregation into On- and Off-centre cells or possibly direction selectivity. Specifically, it will be investigated whether distinct wave properties are required for these processes and how in turn maturation influences wave properties. A main goal of the modelling studies is to obtain experimentally testable predictions. Therefore, the variables in the model will be chosen to reflect known physiological and anatomical properties of individual neurons.

The understanding of activity-dependent developmental processes in the retina is important to understand and control the structural reorganisation observed in degenerative retinal diseases. Possible treatments may be developed by reintroducing healthy tissue into the degenerated retina. Subsequent induction of developmental processes may lead to restructuring of the network to enable it to carry out its normal function. Finally, the retina can act as a model system for the understanding of central nervous system development. Therefore, results of this study may be of general relevance in the context of pathological disruption of spontaneous activity and developmental processes.
 
Description MRC Career Development Award
Amount £556,000 (GBP)
Organisation Medical Research Council (MRC) 
Sector Public
Country United Kingdom
Start 10/2009 
End 09/2014
 
Description Analysis of high density multielectrode array recordings 
Organisation 3Brain GmbH
Country Switzerland 
Sector Private 
PI Contribution Development of analysis tools for data from large scale, high-density 4,096 channel multielectrode array (HD-MEA) recordings. All methods are published as open source projects.
Collaborator Contribution IIT provided existing and new experimental data for projects in my group. This in kind contribution includes hardware and reagents required for experiments, as well as investigator time to support the project. 3Brain GmbH provided support with software development, software implementations of algorithms developed in my group, and sponsored a conference visit for a student.
Impact Multi-disciplinary nature: linking theory, experiment and bioengineering, outputs (as of 02/2016): - publication DOIs: 10.1113/jphysiol.2013.262840, 10.1016/B978-0-12-397266-8.00151-4, 10.3389/fninf.2015.00028, 10.1523/JNEUROSCI.4421-14.2015 - data sets published online: http://www.carmen.org.uk/?page_id=13 - analysis algorithms published on the CARMEN web platform for electrophysiological data: https://portal.carmen.org.uk/ - open source spike sorting toolkit available: https://github.com/martinosorb/herding-spikes - algorithms (spike detection) developed in this collaboration were implemented in the free Brainwave Software, which is used by several groups worldwide to analyse HD-MEA data, http://www.3brain.com/brainwave-software - spike sorting methods are currently adopted by other research groups using HD-MEA
Start Year 2009
 
Description Analysis of high density multielectrode array recordings 
Organisation Italian Institute of Technology (Istituto Italiano di Tecnologia IIT)
Department Neuroscience and Brain Technologies IIT
Country Italy 
Sector Academic/University 
PI Contribution Development of analysis tools for data from large scale, high-density 4,096 channel multielectrode array (HD-MEA) recordings. All methods are published as open source projects.
Collaborator Contribution IIT provided existing and new experimental data for projects in my group. This in kind contribution includes hardware and reagents required for experiments, as well as investigator time to support the project. 3Brain GmbH provided support with software development, software implementations of algorithms developed in my group, and sponsored a conference visit for a student.
Impact Multi-disciplinary nature: linking theory, experiment and bioengineering, outputs (as of 02/2016): - publication DOIs: 10.1113/jphysiol.2013.262840, 10.1016/B978-0-12-397266-8.00151-4, 10.3389/fninf.2015.00028, 10.1523/JNEUROSCI.4421-14.2015 - data sets published online: http://www.carmen.org.uk/?page_id=13 - analysis algorithms published on the CARMEN web platform for electrophysiological data: https://portal.carmen.org.uk/ - open source spike sorting toolkit available: https://github.com/martinosorb/herding-spikes - algorithms (spike detection) developed in this collaboration were implemented in the free Brainwave Software, which is used by several groups worldwide to analyse HD-MEA data, http://www.3brain.com/brainwave-software - spike sorting methods are currently adopted by other research groups using HD-MEA
Start Year 2009
 
Description Nature and role of retinal waves during retinal development 
Organisation Newcastle University
Department Institute of Neuroscience
Country United Kingdom 
Sector Academic/University 
PI Contribution Development of theoretical models of spontaneous neural activity in the developing retina, and analysis of experimental data from the collaborating lab. Students in my and the collaborating lab have contributed data and carried out experiments.
Collaborator Contribution Experimental data set, including experiments specifically designed to test hypotheses derived from models.
Impact Multi-disciplinary: computational neuroscience and modelling, data analysis and experimental electrophysiology. Outputs are the following publications: 10.1016/B978-0-12-397266-8.00151-4 24366261 19176816 21865458 Impact comes from advances in fundamental understanding of visual system development, including developmental differences in models of receptor dystrophy.
Start Year 2006
 
Description Physiology of auditory brainstem neurons 
Organisation University of Leicester
Country United Kingdom 
Sector Academic/University 
PI Contribution Analysis of theoretical and computational models of synaptic transmission, neural excitability and neural homeostasis, and analysis of experimental data provided by the collaborating lab.
Collaborator Contribution Provided experimental data and expertise, and carried out new experiments based on model predictions.
Impact Multi-disciplinary: linking computational and theoretical neuroscience and experimental physiology Outputs are the following publications: 19018705 18450780 17138605 21224222 21903083 Outputs are advances in basic research that help to understand the neural basis of hearing and related conditions such as tinnitus.
Start Year 2006