MRC DiMeN Doctoral Training Partnership: Predicting dynamic response of SUMOylation to stress: a combined modelling and experimental approach

BACKGROUND
Understanding how cell fate is determined in extreme stress is critical in biomedical systems. It is particularly important in the brain, for example, where neuronal death in the brain is a major cause of disability following ischaemic stroke. Recent experimental work has shown that SUMOylation, a protein post-translational modification where a SUMO (Small Ubiquitin-related MOdifier) peptide is attached to a target protein, is a key determinant of cell fate in response to extreme stress. A better understanding the SUMOylation regulatory machinery will facilitate development of new strategies for therapeutic intervention in cell stress-related diseases. To quantitatively establish the dynamic response of SUMOylation to cell stress, a novel combined modelling and experimental approach will be developed in this project. The new approach is required because testing more than a handful of stress conditions is a considerable logistical challenge.

AIMS AND OBJECTIVES
The aim of the project is to develop novel combined modelling and experimental approach to quantitatively understand the regulation mechanism behind SUMOylation under pathophysiological conditions, and to apply this new method to make predictions. The objectives are: 1) To develop mathematical models of SUMOylation and deSUMOylation (i.e., a cellular process to detach SUMO from a target protein); 2) To validate the developed models by data from literature and experiments ; and 3) To apply the developed mathematical models to simulate or predict how the system response with respect to metabolic stress).

METHODOLOGY
The project will start with a meta-analysis of the various complex regulatory mechanisms identified related to SUMOylation process. Subsequently the identified regulatory mechanisms from meta-analysis will be combined with essential biological knowledge to generate candidate model structures.

A new algorithm, by combining particle filter with nonlinear regression, will be developed to facilitate the model identification. The new algorithm will be able to effectively deal with nonlinear dynamics and non-Gaussian noise. The model will then incorporate a variety of environmental factors. For example, we will identify mechanisms discussed in the experimental literature, such as temperature-dependent transcription, translation, and the role of heat-shock protein/mRNA.

To validate the model, experimental design and data acquisition will be carried out. For examples, samples from human cell lines at different time points following differential cell stresses, such as heat-shock and oxygen and glucose deprivation, will be prepared and will be quantified biochemically and/or immunocytochemically using softwares such as LI-COR Image Studio™ and Image J, respectively.

The experimental data will allow adjustment of the model. In a completely novel step in this domain, the unexplained part of the theoretical model will be constructed using system identification techniques - NARMAX, developed in Sheffield. The rank conditions will be used to evaluate the possible nonlinearities from a library of candidate functions, resulting in a parsimonious model, which will perfectly balance accuracy with model complexity.

EXPECTED OUTCOMES AND FUTURE DIRECTION
The supervisory team consists of the researchers from two departments in two different faculties at the University of Sheffield. Both departments encourage interdisciplinary collaborations. This interdisciplinary project will result in the establishment of a new research network and collaboration between the two departments.

To ensure that it leads to further funding and collaborations in SUMO biology in health and disease, 1) a wider network will be set up with Henley lab at Bristol University (https://research-information.bristol.ac.uk/en/persons/jeremy-m-henley(c934d1ac-68ca-4075-9940-e8d126fe3f03).html) and Hay Lab at Dundee University (http://www.lifesci.dundee.ac.uk/groups/ron_hay/pages/ronhay.html) to further the scope of the project; 2) the PhD student will visit the centres twice during the period of study and the members of these two centres will visit Sheffield to give seminars etc.; 3) we will work closely with two institutions to further validate the predictive models, identify new insights, and develop new ideas for future funding; 4) we will drive the sustainability process by state-of-the-art dissemination.

IMPLICATIONS
Based on the data and the system identification techniques, a predictive model will be developed, which will be validated by experimentally exposing cell populations to the normal and extreme cell stress conditions. This work is crucial for establishing dynamic response of SUMO to cell stress, and it will facilitate development of strategies for therapeutic intervention for stress-related ageing and age-related diseases such as stroke and dementia.

WEBLINKS
https://www.sheffield.ac.uk/bms/research/guo
https://scholar.google.co.uk/citations?user=ruFPXZAAAAAJ&hl=en
https://www.researchgate.net/profile/Chun_Guo2
https://uk.linkedin.com/in/chun-guo-0859492a
https://www.sheffield.ac.uk/acse/staff/lg
https://scholar.google.co.uk/citations?user=3S6DAsAAAAAJ&hl=en