Our research interests centre on protein-protein interactions, particularly helical, coiled coil and amyloid-based proteins, and understanding how key interactions achieve stability and specificity. We use the Protein-fragment Complementation Assay, PCA, to screen libraries and select peptides or proteins with significant improvements in stability over wild-type interactions. Specificity of protein-protein interactions can be further improved by using a Competitive And Negative Design Initiative, CANDI, which I have developed to maximise the energy barrier between desired and non-desired protein pairs. Work currently being undertaken includes:

Design and screening of peptide-based libraries, particularly for oncogenic transcriptional regulator proteins, to investigate the role of key residues in stability and specificity.
Monitoring the folding of natural and designed variants of protein-protein interactions to understand how stability and specificity is manifested.
Generating peptides to bind, reverse, and prevent amyloid fibril formation, particularly for beta-amyloid, the primary factor implicated in Alzheimers disease.
Development of our Wellcome Trust funded game 'Cascade' - being designed in collaboration with Fayju. Cascade seeks to increase Alzheimer's disease awareness, understanding, empathy, and to highlight the desperate need for a cure.

Transcription Block Survival, TBS, is a peptide-library screening platform to derive functional transcription factor antagonists within a complex intracellular environment. Our exemplar for the approach is the oncogenic transcriptional regulator cJun, a key component of the Activator Protein-1, AP-1, system. cJun drives key genes responsible for cell cycle modulation, via TPA-response element, TRE, consensus sequence binding, and is upregulated in numerous cancers.

Developing antagonists that bind cJun, but that more importantly ablate function, remains a primary challenge hampering the search for effective molecules. During TBS we exploit this process by introducing fifteen TRE consensus sites directly into the coding region of the essential gene dihydrofolate reductase, DHFR. Introduction of cJun leads to TRE binding, preventing DHFR expression by directly blocking RNAP transcription to abrogate cell growth under selective conditions.

TBS proof-of-concept is further demonstrated by an absolute requirement for i) the cJun basic domain and ii) the presence of TRE sites within DHFR for transcriptional block to occur.

Screening a 130,000 member library next identified a sequence that both binds cJun and antagonises function as demonstrated by restored cell growth by DHFR transcription. In vitro hit validation using CD, ITC and EMSA experiments confirmed high target affinity, KD = 14.4 ± 3.7 nM, importantly demonstrating effective antagonism of the cJun/TRE interaction.

TBS is an entirely tag-free genotype-to-phenotype approach, selecting desirable attributes such as high solubility, target specificity and low toxicity within the complex cellular environment. It facilitates rapid library screening to accelerate the identification of therapeutically valuable sequences.