Cross-modality integration of sensory signals leading to initiation of locomotion

Lead Research Organisation: University of Plymouth
Department Name: Sch of Computing, Electronics & Maths

Abstract

An animal that fails to respond appropriately to its sensory environment jeopardises its survival. If we try to catch a bird it will fly or run off. Most animals, like ourselves, can walk, run, swim or fly when stimulated. Even though this response seems very simple compared to our ability to think, talk and learn, the details of the way nervous circuits in the brain and spinal cord initiate locomotion remain poorly understood. In mammals, which have been studied most intensively, we have a broad knowledge of the areas of the brain and types of nerve cells which are involved, but the nervous system is astonishingly complex. To simplify the problem, we chose a very small animal, the newly hatched frog tadpole whose spinal cord is the thickness of a human hair. At this stage, around 2000 nerve cells in the brain and spinal cord may be sufficient to allow the tadpole to swim when it is touched. Over many years of study, we have classified these nerve cells into less than 20 types and defined their anatomy, properties, connections and the networks they form. Critically, we have discovered a small population of one type of nerve cell in the brain which is responsible for producing the rhythmic nervous system activity driving swimming. Experimental stimulation of these cells can lead to swimming and silencing them can stop it. Using our detailed knowledge about the tadpole nervous system, we have completed a first generation computer model of how the tadpole's nerve cells grow to make the connections and assemble the networks controlling swimming. Compared to other vertebrate animals, we therefore have remarkably detailed knowledge about this simple animal's behaviour and the nervous system controlling it.

Our aim now is to exploit this knowledge about the tadpole to ask how and where its brain makes the decision to start to swim. This is possible because, uniquely, we have identified the brain nerve cells which drive swimming. We will test the hypothesis that these are also the nerve cells which make the decision to swim. In experiments in Bristol we will test whether all the signals about touch, light and other senses which the tadpole must take into account, converge on these cells. We can do this directly by recording their electrical activity and stimulating the tadpole while monitoring whether or not it swims. Using additional imaging methods where active cells are made to emit light, we will trace how signals pass cell-by-cell along the pathways in the nervous system connecting different kinds of stimulus to the nerve cells controlling swimming. Meanwhile in St Andrews we will use sophisticated electrical recording methods and pharmacology to examine the detailed properties of the nerve cells driving swimming. This will help us understand how they respond to sensory signals by switching from silence at rest to acting as pacemakers during swimming, rather like the cells which drive the heartbeat. In parallel, the Plymouth team will use computer models of the nerve cells and networks in the tadpole brain and spinal cord, combining the new findings from Bristol and St Andrews to understand the basic requirements to start and generate swimming activity. They will also allow us to extend theoretical models of decision making down to the level of nerve cells, which is where decisions are ultimately made.

A shared evolutionary origin means that although the tadpole nervous system is small and young, it is built on the same principles as all vertebrates. Our findings should therefore provide broader insights into how brain networks controlling locomotion are organised. In the brain of mammals, like ourselves, these networks are remarkably complex, and when they go wrong, they cause severe problems in initiating locomotion, like Parkinsonism. Our hope is that study of a much simpler system may uncover core principles which lie concealed in the adult brain.

Technical Summary

The brain and spinal cord networks controlling and initiating locomotion in adult vertebrates, especially mammals, are remarkably complex. We will exploit a simple system, the hatchling frog tadpole, where we have defined the neurons and networks generating swimming locomotion. In particular, we have identified the "dIN" reticulospinal neurons which drive swimming. In contrast to other model systems like worm, fly and fish we can study detailed neuron and synapse function using in-situ whole-cell recording. We will show: how different modalities of sensory input converge on the key dIN neuron population; how sensory integration determines the "decision" to swim, taking the state of the tadpole into account; and how the tadpole selects and correctly implements the way swimming starts, from different possible directions and strengths. A 'systems biology' approach will operate across 3 labs at 3 levels: (1) Ca imaging and whole-cell recording will trace sensory pathways to dINs neuron-by-neuron, showing how inputs interact (excitation by skin touch or light dimming, inhibition by head pressure) to control dIN firing and so initiation of swimming; optogenetic silencing will test the role of neuron populations; (2) in situ voltage-clamp will show precisely how membrane currents determine dIN responsiveness, single spiking at rest but pacemaking when NMDARs become activated during swim initiation; (3) our detailed model of a small, coupled dIN population will test the contributions of different currents to spike threshold and synaptically induced firing; our axon growth model (lengthened to include the full range of sensory input pathways) will generate a full network connection map (of ~3000 neurons) which we will map onto a functional model to evaluate our understanding of the swim initiation process. We will extend "decision" theory down to the neuronal level where noisy sensory inputs are integrated and compared to a threshold to choose a suitable course of action.

Planned Impact

The aim of this project is uncover basic principles about how sensory stimuli lead to the initiation of locomotion in animals. If it is successful, the results will benefit the range of scientific communities around the world studying this problem in a wide range of animals from nematode worms and fruit flies to man. The main academic beneficiaries have already been detailed.

If the fundamental principles about the initiation of locomotion and how nervous systems make decisions are revealed the project should additionally provide insights for:
1) medical profession in relation to movement disorders like Parkinsonism and spinal injury restoration.
2) drug companies and charities interested in such disorders.
3) research policy makers who will see the value of choosing the simplest and most appropriate system to investigate each problem. This also helps to reduce the use of adult mammals in research.

Tadpoles are some of the most familiar of animals to the wider public and attract attention across the whole age spectrum. They therefore provide a valuable, highly accessible entry point to explaining specific issues about brain and behaviour, what nervous systems are and how they work to make animals behave. They also provide a context for broader explanations about how scientific research is carried out, the significance of the 'systems biology' approach, and the importance and relevance of projects like ours which use simple model animals. The research will generate good imagery of tadpole behaviour, the growth processes of model nerve cells as they form circuits in the brain, and the activity patterns of all the neurons in the tadpole's nervous system as it responds to stimulation and swims away. These will be suitable to popularise the study of simple model animals and their brains in websites and museums and science centres like '@ Bristol' (where we have already had some participation).

In terms of training, the physiology RAs will learn specific techniques and broad experimental approaches which will prepare them for research careers in any part of academic, medical or pharmaceutical neuroscience. It is critical that the UK continues to train highly skilled electrophysiologists. These techniques remain essential for providing detailed information on the properties and connectivity of neuronal networks at all levels; however, researchers with suitable skills are becoming scarce. All RAs will gain in experience of working as a group, research organisation, data analysis and presenting their results orally and by writing papers. The computational RA will gain experience working and communicating with biologists, while the physiologists will gain closer insight into the use of computer models and communicating with mathematicians, increasing the effectiveness of the systems biology approach.

Since our research is at a fundamental level we would expect that the true impact of our research outputs on understanding of adult animals will be slow and act cumulatively over a period of many years, as our findings encourage targeted studies into equivalent processes in progressively more complex systems. The impacts of researcher training will be more immediate, contributing to a skill base relevant to all sectors.

Lastly, the tadpole research on which this project is based is recognised internationally but is carried out almost entirely in, and identified with, the UK. The two UK labs with the necessary expertise, Bristol and St Andrews, will collaborate on this project. The contribution that the success of research output from this model system makes to international neuroscience raises the profile of UK research in this area.

Publications


10 25 50
 
Description The findings from the project have inspired various visualisation projects that at different stages of development. For example, we developed a 3D visualisation of the axon growth and neuronal activity in the tadpole spinal cord. This has been used to explain the basic principles of neuronal development and function to the general public. A new virtual reality version of this visualisation is currently under development, which we plan to demonstrate to a much wider audience, beginning at a conference in April.
First Year Of Impact 2015
Sector Education
Impact Types Societal
 
Title Improved developmental approach 
Description We improved our existing model of neuron development in the tadpole spinal cord. The new model allows all neurons to grow in parallel, and can include the effects of axon-axon interactions. This allows the model to produce results which more accurately reflect biology. 
Type Of Material Computer model/algorithm 
Provided To Others? No  
Impact A paper describing this work is in preparation. In future we anticipate that there will be a huge impact for this kind of modelling, as it allows us to produce virtual organisms on which experiments can be performed without the use of real animals. 
 
Description Bristol & St Andrews 
Organisation University of Bristol
Department School of Veterinary Science
Country United Kingdom of Great Britain & Northern Ireland (UK) 
Sector Academic/University 
PI Contribution This BBSRC Project is a collaboration between Plymouth University, University of Bristol, and University of St Andrews. Our role in Plymouth involves computational modelling of the tadpole nervous system. Our results have led to new insights into the structure and function of this system, and have provided ideas for further biological experiments that our collaborators can perform.
Collaborator Contribution Our partners provide experimental neurobiological data that we can use to refine our computational models. They also participate directly in our modelling and adjustment of the model, and provide the questions that need to be answered using modelling.
Impact Multi-disciplinary: Neurobiology, Computing and Mathematics. Most of the papers and publications listed in this submission are a result of this collaboration.
Start Year 2006
 
Description Bristol & St Andrews 
Organisation University of St Andrews
Department School of Medicine St Andrews
Country United Kingdom of Great Britain & Northern Ireland (UK) 
Sector Academic/University 
PI Contribution This BBSRC Project is a collaboration between Plymouth University, University of Bristol, and University of St Andrews. Our role in Plymouth involves computational modelling of the tadpole nervous system. Our results have led to new insights into the structure and function of this system, and have provided ideas for further biological experiments that our collaborators can perform.
Collaborator Contribution Our partners provide experimental neurobiological data that we can use to refine our computational models. They also participate directly in our modelling and adjustment of the model, and provide the questions that need to be answered using modelling.
Impact Multi-disciplinary: Neurobiology, Computing and Mathematics. Most of the papers and publications listed in this submission are a result of this collaboration.
Start Year 2006
 
Title Fireflies: Software for interactively exploring dynamical systems using GPU computing 
Description Fireflies exploits the power of graphical processing unit (GPU) computing to produce spectacular interactive visualizations of arbitrary systems of ordinary differential equations. In contrast to typical phase portraits, Fireflies draws the current position of trajectories (projected onto 2D or 3D space) as single points of light, which move as the system is simulated. Due to the massively parallel nature of GPU hardware, Fireflies is able to simulate millions of trajectories in parallel (even on standard desktop computer hardware), producing "swarms" of particles that move around the screen in real-time according to the equations of the system. 
Type Of Technology Software 
Year Produced 2015 
Impact Visualisations produced by this software have been used as part of our undergraduate teaching. 
URL https://bitbucket.org/rmerrison/fireflies
 
Title Simulation of Neural Activity Patterns with Python (SNAPPy) 
Description This software was developed by us for simulation of the tadpole spinal cord. However, the software was designed to be very flexible and allow modelling of other neuronal circuits beyond the tadpole. 
Type Of Technology Software 
Year Produced 2014 
Impact The software has been used (by Dr Merrison-Hort) to develop a model of the mammalian basal ganglia as part of his PhD work (not based on any funding body award). The results obtained from this model were published in Frontiers in Computational Neuroscience. 
 
Description BBC News on line publication about new BBSRC grant 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact BBC news England: Tadpole brains studied to help understand Parkinson's


http://www.bbc.co.uk/news/uk-england-25076279

Public engagement
I was interviewed by local media
Year(s) Of Engagement Activity 2013
 
Description PyPlym 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Industry/Business
Results and Impact The postdoctoral fellow on this grant presented our work at a local "PyPlym" event, about the use of the Python programming language. The audience ranged from those just learning to program to professional developers from industry. As well as technical details on how our research uses Python, the talk included a brief overview of computational neuroscience.
Year(s) Of Engagement Activity 2016
URL https://www.meetup.com/PyPlym/
 
Description School Visit (Clyst Vale) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Schools
Results and Impact Approximately 15 sixth form students (members of the college's "Science Club") attended my talk about our work modelling the tadpole nervous system. The talk described the basics of neuroscience and computational modelling, and (according to their teacher) fit in very well with the subjects they were currently learning about in lessons. The students engaged well with the talk, and my subsequent demo of an interactive virtual reality "tour" of the tadpole spinal cord.
Year(s) Of Engagement Activity 2016
 
Description Talk about computational modelling of the tadpole at SciBar event 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Public/other audiences
Results and Impact Dr Merrison-Hort gave a talk to the general public at the Plymouth "SciBar" event. The talk sparked a long discussion amongst the audience about the possibilities of computational neuroscience, and about neuroscience in general.

Not aware of any.
Year(s) Of Engagement Activity 2014
 
Description Virtual reality tadpole demonstrations 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact In collaboration with the university's Interactive Systems Studio we have developed a virtual reality tour of the tadpole nervous system that teaches users the basics of neuroscience and how this simple animal behaves. We have demonstrated this system at numerous events, from international scientific conferences (e.g INCF World Congress in Neuroinformatics 2016) to local science events for the general public (e.g. Plymouth University Robotics Open Evening). Many people, of all ages, have tried out the system and learnt something about our research and neuroscience in general.
Year(s) Of Engagement Activity 2015,2016