Computational methods are fully integrated into the drug discovery process and can help identity bioactive molecules, ‘scaffold-hop’ into new IP-space, optimise a small molecule’s physicochemical properties. Our research group uses a range of these computational methods in the discovery of novel ligands for target-based drug discovery. We have pioneered a fragment-based computational design tool (SPROUT), coupled with synthetic chemistry, to identify bespoke small molecules for a plethora of biological targets. We collaborate with colleagues at the University of Leeds, and internationally, in the fields of anti-infectives, cardiovascular medicine and parasitology. Our primary aim is to identify new modulators for challenging protein targets (such as membrane proteins in conjugation with cryo-EM), for neglected tropical diseases (malaria and toxoplasmosis), and for drug-resistant bacteria (ESKAPE pathogens). In addition to their use as selective probes (e.g. via incorporation of fluorescent moieties), a number of these systems are also being progressed as therapeutic drug leads.
Current major projects
- Discovering new lead compounds to combat antimicrobial resistance (AMR)
- Lead optimisation of anti-parasitic compounds to treat malaria and toxoplasmosis
- Identification of chemical modulators to restore insulin signalling and restore cardiovascular repair
- Computational strategies for the prediction of solid-form properties
Detailed research programme
Discovering new lead compounds to combat antimicrobial resistance (AMR)
The UK Government report by Jim O’Neill suggested that AMR would be a leading cause of worldwide deaths by 2050, overtaking cancer. By combining computational methods and traditional drug discovery approaches, we are designing new compounds to fight drug-resistant bacteria. We are actively working in three areas: 1) Developing new anti-Gram-negative compounds through optimizing the physicochemical properties of approved anti-Gram-positive antibiotics (such as linezolid), guided by structure-based drug design. 2) Dual targeting of DNA gyrase and topoisomerase IV to overcome fluoroquinolone-resistant bacteria, targeting a recently discovered allosteric site; 3) Expanding the toolbox of metallo-beta-lactamases inhibitors (MBLis), a class of drug-resistance conferring enzymes. These have unmet clinical need and are top priority, as highlighted in a 2019 WHO report.
Lead optimisation of anti-parasitic compounds to treat malaria and toxoplasmosis
Membrane-integral pyrophosphatases (mPPases), dihydroorotate dehydrogenase (DHODH) and cytochrome bc1 represent drug targets against protozoan parasites, such as malaria, toxoplasmosis and trypanosomiasis. These diseases are increasingly resistant to current treatments and affect large proportions of the global population. Studying the structure and function of mPPases is a multidisciplinary project involving computational biology and chemistry approaches in combination with laboratory-based techniques with the aim of uncovering and designing novel mPPase inhibitors. We have a well-established DHODH research programme which has developed a potent and selective DHODH ‘late-lead’ chemical series which is in development with Medicines for Malaria Venture (MMV) for malaria chemoprevention. We are also part of an international collaboration developing tetrahydroquinolone (THQ) compounds targeting cytochrome bc1.
Identification of chemical modulators to restore insulin signalling and restore cardiovascular repair
In collaboration with the Leeds Institute of Cardiovascular and Molecular Medicine (LICAMM), we have active projects targeting the protease BACE1, the insulin receptor and endothelial nitric oxide. Utilising a combination of electron microscopy, computational drug design and synthetic chemistry, we aim to identify small molecules able to restore insulin sensitivity to insulin resistant cells. This would deliver a novel therapeutic strategy for type II diabetes, in turn providing protection from the associated cardiovascular complications of ischaemic heart disease and stroke.
Computational strategies for the prediction of solid-form properties
Desirable drug-like properties such as aqueous solubility and oral bioavailability are often ‘fixed’ in the later stages of drug discovery. Prediction of these properties during the early design phase would be of great interest to the pharmaceutical and agrochemical industries. Through investigations into a highly polymorphic material (5-methyl-2-(2-nitrophenyl)amino-3-thiophenecarbonitrile)) we aim to develop a set of chemical descriptors which accurately capture the nature of the solid state of drug forms. These descriptors will be used to develop computational models for the prediction of properties related to a solid form’s bioavailability, such as aspect ratios, crystal morphology and aqueous solubility.