ITQB Auditorium
STRUCTURE-GUIDED FRAGMENT-BASED DRUG DISCOVERY FOR CANCER AND TUBERCULOSIS: FIGHTING THE EMERGENCE OF RESISTANCE
Tom L Blundell, PhD, FRS, FMedSci
Cambridge University
Knowledge from sequences of genomes of humans and pathogens has the potential to accelerate diagnosis, prognosis and cure of disease. We are moving quickly into an era of precision medicine, not only in familial diseases where a mutation in a human gene is important, but also for understanding somatic mutations in cancer. Equally important, the genome sequences of pathogens, for example in tuberculosis or leprosy, can give clues about the choice of existing drugs, repurposing of others, and the design of new ones to combat the increasing occurrence of drug resistance.
One approach is to exploit state-of-the-art methods to bring new drugs for different targets to the market, but this will be difficult to finance if patient populations are small. Structure-guided fragment-based screening techniques have proved effective in lead discovery not only for classical enzyme targets but also for less “druggable” targets such as protein-protein interfaces. Initial screening involves small fragments with very low, often millimolar affinities, and biophysical methods such as isothermal calorimetry (ITC), analytical ultracentrifugation (AUC), thermal shift, surface plasmon resonance (SPR), nuclear magnetic resonance (NMR) and X-ray crystallography are used to explore chemical space of potential ligands. The approach involves a fast initial screening of a library of around 1000 compounds, followed by a validation step involving more rigorous use of related methods to define three-dimensional structure, kinetics and thermodynamics of fragment binding. The use of high throughput approaches does not end there, as it becomes a rapid technique to guide the elaboration of the fragments into larger molecular weight lead compounds. I will discuss progress in using these approaches for targets in cancer and in mycobacteria tuberculosis, abscessus and leprae infections.
I will also review our computational approaches using both statistical potentials (SDM) and machine learning methods (mCSM) for understanding mechanisms of drug resistance. These have demonstrated that resistance does not only arise from direct interference of the resistance mutation to drug binding but can also result allosteric mechanisms, often modifying target interactions with other proteins. These lead to new ideas about repurposing and redesigning drugs.
Host: Maria Arménia Carrondo