A systems biology approach to studying skeletal muscle ageing

Ageing is associated with loss of muscle strength and endurance (muscle quality) which severely impairs mobility of people in everyday life. Our understanding of what determines muscle quality in humans, and how this is altered as we age is not sufficient to design specific interventions that may substantially protect people from loss of function. These interventions could be specialised clinical physiotherapy which includes devices that enhance the process of exercise and they could involve new drug therapies that promote improved muscle function or accelerate the gains made with physiotherapy by accelerating the biological effects of exercise.

The starting point for our project is to first generate detailed knowledge of the age related process by studying muscle samples obtained from younger and older human subjects. We have already produced a lot of data, using a new technology called a gene chip . Gene chips provide a snap-shot of the genes switched on and off in the muscle tissue and we can then relate these profiles to the age (of physical fitness) of each subject. The analysis of all this data requires establishment of novel statistical models to cope with its complexity. The studies of each human subject can produce 60,000 data points and a data set of this size produces new challenges when trying to summarise the conclusions. We will develop new methods for this project which will be of broad use to the biomedical community, where application of similar technology is used to study cancer and other major diseases.

Skeletal muscle is negatively affected by the ageing process whereas its functionality is a profound determinant of the quality of life and of metabolic health. Until the generation of the data presented in this application, there was a complete lack of large scale human molecular data to drive hypothesis-led experiments in the field of muscle ageing. The aim of the application is to utilise informatics analysis and statistical modelling to determine if bio-bank materials can permit hypothesis driven analysis (e.g. reverse genetics experiments in cells or animal physiological models). We present analysis of a novel bio-bank that has allowed us to produce the first robust and cross-validated classifier of age for human skeletal muscle using cross-sectional cohort design yielding a 13-gene signature that was 92% accurate in determining muscle age. Expansion of the bio-bank will allow us to determine how this gene expression signature varies with physical capacity and lean body mass in two longitudinal studies. We will establish if our ageing signature represents an underlying biological clock . We aim to understand the function of the 13-gene signature. Eleven genes are under-expressed in older compared with younger muscle, while two genes are over-expressed in older subjects. One gene is a validated age-related gene, triosephosphate isomerase 1 (TPI1) from the glycolysis pathway that is 50% lower in aged muscle and TPI1 appears to be a normal protection strategy, where increased expression reduces reactive oxygen species. The function of the remaining genes in skeletal muscle is unknown and we will manipulate the level of expression of each gene in isolation and together to determine if the genes function as a single transcriptional unit or whether they represent marker genes of several independent gene networks. Direct measurements of muscle function will complement these transcriptomic studies.