Foundations of Neuromechanical Systems Biology

Lead Research Organisation: Royal Veterinary College
Department Name: Comparative Biomedical Sciences CBS

Abstract

Understanding how animals move is one of the grand challenges of modern science. It has broad impact on society: it affects our ability to explain the biological world, to treat human and animal disease, and to aid those recovering from injury. The more we know about how biological systems control their movement, and how different organs contribute to locomotion, the better we will be able to treat those with neurological disorders or musculoskeletal injury, and to inspire new technologies, such as legged robots.

Locomotion is the signature behavior of animals. In the face of an unpredictable environment, noisy signals from sense organs and noisy forces from muscles, animals are able to move with speed, dexterity and robustness. Yet for one of the most important types of movement, fast terrestrial locomotion on legs, we do not know how sensory information is used to stabilise the body, or how we manage our noisy muscles. Stability may be largely handled by the mechanics of the body; sensory input may still be incorporated, but on longer time-scales; or, rapid locomotion may be constrained by motor noise. This project will test these divergent predictions.

A major obstacle stands in the way of our understanding of how the nervous and musculoskeletal systems work together to produce locomotion. The problem is that locomotion results from the interaction of the brain, spinal cord, musculoskeletal system, and external world. This means that for us to accurately interpret what the role of each of these subsystems is, we need to independently examine and manipulate each subsystem, in an intact, freely behaving animal, in an ethical way. Understanding how each subsystem works in the context of all of them is important not just because history teaches us that linking across subsystems is a reliable way of gaining insight into the whole system, but because disease and injury frequently affect only one of these subsystems, or organs within a subsystem, at a time.

The first major aim of this proposal is to develop the technologies we need to overcome this limitation. In doing so, we believe we are laying the foundation for a new branch of science: neuromechanical systems biology. This field will treat neurons, muscles, skeletons, and the external environment as complex interacting constituents that result in locomotion, in a manner akin to systems biology at the cellular and molecular levels. Considering the function of each organ in the full context of the running animal is important if we are to gain a true picture of what each organ does.

We will couple this integrative approach with powerful new ways to precisely perturb running animals. We will combine optogenetic neural manipulation with real-time tracking and mechanical perturbation to make possible causal, neuromechanical perturbations of freely running mice. By teasing apart the neural and mechanical contributions to locomotion we will gain a clear understanding of the computations performed by the nervous system during locomotion. With this understanding we will confirm, refute, or refine the predictions of optimal feedback control theory, a leading theory of motor coordination. Testing this theory is the second major aim of this proposal.

Optogenetics is an extraordinary new technology that provides unprecedented new ways to study and manipulate the nervous system. Optogenetics allows specific neurons to be turned on and off, extremely quickly, using light. It relies on our knowledge of genetics to place molecular, light-dependent on/off switches in the membranes of specific neurons. It is revolutionizing neuroscience, because it allows us to study the function of parts of the nervous system in a causal manner. Here we propose to combine optogenetics with a neuromechanical approach to locomotion. We firmly believe that this combination will revolutionize our understanding of how biological systems move, and give us important new tools for medicine.

Technical Summary

Locomotion is complex. It emerges from the dynamic interplay of ion channels, neurons, muscles, skeletons, limbs, bodies, and the environment. A century of extraordinary science has studied reflex pathways, muscle function, redundancy in limbs, and whole body mechanics. To build a systems level understanding of locomotion, however, we need to integrate these results with a new class of data from precisely perturbed, intact, freely behaving animals.

The success of systems approaches in molecular and cellular biology has been in part due to: 1) acknowledgment of causation at any level; 2) data and models that span levels of the system, and 3) tools for causal manipulation at multiple levels.

This proposal will bring these key features of a systems biology approach to bear on the problem of locomotion. We will:

1) Build a detailed musculoskeletal model of the mouse hindlimb. This will allow us to predict how muscle output and sensory input are integrated to produce locomotion at the level of the limb.
2) Use optimal feedback control theory to predict how the hindlimb will respond to five crucial perturbations. These are increased motor noise, a loss of sensory feedback, an unexpected external perturbation, and an unexpected external perturbation with increased motor noise, and with a loss of sensory feedback.
3) Use optogenetics to develop an in vivo, freely moving mouse preparation in which a) lower hind-leg muscles can be selectively activated and b) sensory input from the lower hind-leg can be selectively silenced.
4) Compare the simulated predictions of the behaviour of the leg with experimental data from precisely perturbed, freely running mice. This will be achieved using advanced real-time instrumentation built around a treadmill.

Given published data on individual and stride-to-stride variation in the kinematics of running mice, predictions from our model can be tested with mild external perturbations and a reasonable number of subjects (10).

Planned Impact

This grant will impact industry, society, and academe. It pushes academic boundaries in an area with direct economic applications, and strong societal influence. If we are successful, we will have laid the foundation for a new type of systems biology. For the first time, we have the tools to precisely measure and perturb both the nervous system and locomotion of a close mammalian cousin. It is hard to underestimate the impact this could have on an aging society that relies, at its core, on the mobility of its citizens.

Movement is critical to health and quality of life. The total NHS spending on musculoskeletal and neurological disease in 2007 was £7.4bn (Featherstone, 2010; www.policyexchange.co.uk). Circulatory problems and mental health together cost the UK £17.2bn, and a significant fraction of this cost will have its roots in mobility; movement is central to maintaining both good circulation and mental health (Halliwell, 2005; Mental Health Found. London). We hope to lay the foundation for medical advances that improve our ability to treat those facing a lack of mobility, increasingly important as our population ages, and thus have a huge impact on the quality of life of many millions of people.

In other areas of neuroscience, the revolution brought on by optogenetics speaks for itself. While papers in high impact journals are certainly not the final arbiter of societal impact, they are an indicator of subsequent breadth of effect, and since April 2009, 13 Science or Nature papers have used optogenetics to move beyond correlation and causally investigate critical mechanisms for human medicine (anxiety, attention, epilepsy, fear, learning, macular degeneration, Parkinson's disease, respiration, reward, a validation of fMRI). This grant will enable a similar wave of discovery in movement science. With movement disorders affecting 28% of people between 50 and 89, it is of the highest priority.

Industry: This grant will benefit the UK healthcare industry. Pharceutical companies using mouse models for diagnosis and drug development will benefit from the work developing a new mouse line and preparation, in which specific parts of the peripheral nervous system can be perturbed. For diagnosis and drug assays, the feedback treadmill system allows much finer experimental control.

Society: Optogenetics is awe-inspiring. This will contribute to public engagement, both in the discoveries it makes possible and in awareness of science generally. The ethics of optogenetics must be considered by society at large, and this must happen soon, for the UK to keep abreast of international efforts. Discoveries made with optogenetics offer hope to a vast number of people with neurological illness. By demonstrating how it can be used to help those with movement disorders, this proposal will bring positive awareness to UK basic science. The fact that optogenetics was made possible by the discoveries of microbiologists will reinforce the case for basic research.

Academia: skills and trained people. This proposal will provide exceptional training opportunities for its PDRAs. The PDRAs will benefit not only from pushing the frontier within their own disciplines (optogenetics and optimality principles), but from the comparative advantage of learning techniques and modes of thinking from another discipline. The biology PDRA will gain an understanding of engineering control, building systems from first principles, and computer modeling. The control/biomedical PDRA will gain an exposure to genetics and physiology, thinking conceptually about complex biological systems, and learning how to ask questions in biology. We are firm believers that exposure to different modes of thinking is an exceptional stimulus for mental and scientific growth. Their professional development may also be enhanced by involvement in the exploitation of new technologies and through industry collaboration.
 
Description We have developed a new muscle and bone model of the mouse hindlimb which will allow us to predict how the mouse will respond to different changes in terrain and nerve activity. As part of this we have produced an interactive 3D pdf that can be used to dissect the muscles from the hindlimb and this provides the most comprehensive guide to date of the muscle anatomy of the hindlimb of the mouse. We anticipate that this will be very useful to all of those working with mouse musculoskeletal anatomy. These include scientists studying the effects of genetic alterations, mouse models of neuromuscular disease and mouse locomotion (Charles et al., PLoS ONE 2016). As part of this work we show that the adaptations in musculotendon architecture that have been attributable to the development of cursorial locomotion in larger animals are also present in the mouse, suggesting that many of these adaptations may in fact be ancestral for the mammalian lineage. Using this anatomical knowledge we have produced a model of muscle function in the hindlimb (Charles et al., Journal of Anatomy 2016).

We have then gone on to perform high speed video recordings of mice running across a forceplate with markers applied to the centre of each joint. Analysis of the high speed video coupled with the force plate analysis and the previous hindlimb anatomical model has been used to develop a dynamic computer model of mouse fast locomotion. This work is now under review: Charles, J. P., Cappellari, O., Hutchinson, J.R. A dynamic simulation of musculoskeletal function in the mouse hindlimb during trotting locomotion.

In parallel with building the mouse locomotion model we have discovered that peripheral manipulation of nerve activity using optogenetics presented substantial and unexpected challenges compared to using optogenetics in the central nervous system. Most critically, motor and proprioceptive nerves do not travel in precise locations in the sciatic nerve. In some mice the majority of the motor nerves are on the surface layers of the nerve whereas in other inbred littermates the majority of motor nerves run deep within the sciatic nerve. Channelrhodopsin and halorhodopsin require specific wavelengths of light to activate these receptors (blue and yellow light respectively) and these both have poor penetration into the sciatic nerve. Coupled with the relatively long refractory period we have been unable to show the levels of tetanic force that can be achieved with electrical stimulation. While this does not prevent the analysis of the effect of introducing motor noise at different phases of the stride, it has not been possible to develop a system for temporarily shutting off proprioceptive feedback as we had first envisaged.

A second problem is that for good light penetration into the sciatic nerve we have had to use an array of microLEDs surrounding the nerve in order to get adequate stimulation without generating excessive heat. This has proved technically challenging. We performed a number of acute experiments to address these technical challenges and the results are in a penultimate draft for the following paper: Cappellari O, Wilshin S, Charles JP, Spence, AJ and Wells DJ. The dark side of Optogenetics. (in prep).

Finally, our rodent hindlimb model has been compared with a similar human model to assess differences in muscle length changes during locomotion. Mice show much smaller fibre excursions as compared to human and this might explain some of the differences in disease progression when comparing mouse models of muscular dystrophy to human patients (Hu X, Charles JP, Akay T, Hutchinson JR, Blemker SS. Skelet Muscle. 2017).
Exploitation Route Better understanding of how to aid movement disorders or neuromuscular injury.
To build active prosthetic limbs
To inspire the next generation of robots.
Sectors Aerospace, Defence and Marine,Healthcare