(A*STAR) Protein solubility in biology and biotechnology

Protein aggregation is an issue in biology and biotechnology. A wealth of studies have focussed on proteinopathies and neurodegenerative diseases in particular. Here, the b-sheet propensity model has become prominent, largely because in many instances the aggregated state comprises a large fraction of b-structure. However, generally several properties contribute to the relevant weak interactions for intermediates on the pathway to aggregation, including charge, hydrophobicity, b-sheet propensity, and flexibility. Our recent work (Warwicker and Curtis groups in Manchester) has highlighted further complexity, with specific amino acid contributions from non-native states [Austerberry JI et al 2019 Biochemistry 58:3413-3421]. This computational project will use bioinformatics and molecular simulations to develop improved predictive models for protein solubility and aggregation, whilst at the same time increasing our understanding of the key molecular mechanisms and underlying physical chemistry. Current work in the Manchester grouping [https://www.biorxiv.org/content/10.1101/625830v2] highlights the use of machine learning alongside a physical chemical picture. However, this current understanding still falls short. First, more data are becoming available and incorporation will improve performance of the learning algorithms. Second, and equally important, molecular simulation will be added to probe which transitions between folded, partially folded, and unfolded forms are most crucial in promoting protein aggregation. Simulations will be carried out in Chandra Verma’s group at the Bioinformatics Institute (BII) in Singapore. This collaboration is part of an ongoing successful set of projects between these UoM groups and the BII. Whilst informatics work will cover many proteins, simulations require a focus on a few systems due to the increased computational resource. One such will be the tumour suppressor protein p53, for which aggregation of mutant variants has been associated with cancer. The Verma group has substantial experience with p53 simulations and experimental collaborations, for example developing stapled peptide inhibitors of protein-protein interactions [Yuen TY et al 2019 Chem Sci 10:6457-6466], as well as studies of misfolding [Pradhan MR et al 2019 Nucl Acids Res 47:1637–1652]. A second system is monoclonal antibodies (mAbs) and in particular the complementarity determining regions (CDRs). Sequence variation at the CDRs mediates high affinity binding to antigen, but also weak affinity cross-interactions between proteins. Since CDRs are typically relatively non-polar, and less rigid in the absence of cognate antigen, their role in protein aggregation is likely to be representative of partial unfolding more generally in proteins. Models developed during the project will be made available as web tools on our server www.protein-sol.manchester.ac.uk and as installable
software. Collaborations will be sought with experimental groups to further test the models, such as those with synthetic biology expertise in the Manchester Institute of Biotechnology (MIB).

Entry Requirements:
Applications should be submitted online and candidates should make direct contact with the Manchester supervisor to discuss their application directly. Applicants must have obtained, or be about to obtain, at least an upper second class honours degree (or equivalent) in a relevant subject.