Exploring novel therapeutic modalities. Targeting relevant proteins in cancer

Abstract

[eng] Only a small portion of the human proteome is pharmaceutically accessible based on traditional pharmacology approaches. Most proteins, for example transcription factors, scaffolding proteins and non-enzymatic proteins, are still nowadays considered undruggable. This obeys to several major reasons: the lack of functional binding sites on these proteins, the assumption of the occupancy-driven paradigm, whereby therapeutic efficacy and pharmacologically relevant inhibition is mostly achieved upon >90 % target engagement and the lack of tools to identify novel hit molecules. Novel pharmacological modalities are needed to address the plethora of still unmet clinical needs. In this thesis, we present three different projects aiming to increase the druggable proteome. Since drug discovery is a long, tedious and expensive process, it is crucial to develop initial hits with high probabilities to success. The combination of computational and biophysical approaches enables the identification of a high number of hits (computationally) and a proper characterisation (biophysically) of the selected candidates to advance them through the drug discovery process. E3 ligases have a crucial role in the ubiquitin-proteasome system, by recruiting specific substrates for proteasomal degradation. They have been described to be relevant in many diseases. The development of targeted protein degradation (TPD) approaches has also raised E3 ligases as attractive candidates for drug discovery. Even though there are more than 600 E3 ligases in human, there are very few examples of E3 ligases targeted for both TPD strategies or its own modulation. Following a computational structure-based approach, our lab previously identified several ligands binding allosterically to the FBXW7 E3 ligase. In this thesis, we aim to evaluate the biological effect and to disentangle the mechanism of action of the previously identified allosteric ligands of the FBXW7 E3 ligase. Pharmacological modulation of the epigenetic TET2 enzyme has a huge potential in cancer treatment. On one hand, activation and inhibition of TET2 could impede cancer progression and tumour relapse respectively. On the other hand, downregulation of TET2 could increase the efficacy of CAR-based cell therapies. Our lab previously identified novel small-molecules able to bind allosterically to the TET2 epigenetic enzyme. In this scenario, we aim (i) to biophysically characterise novel series of TET2 allosteric ligands, and (ii) to develop TET2-based PROTACs that could be beneficial for the development of CAR-based therapies using some of these previously identified allosteric TET2 ligands. Exhaustive drug discovery programs are needed to obtain a promising drug candidate. Therefore, there is a huge need of more and better initial hits, to increase the probabilities to success. On-demand chemical collections have emerged as a transformative resource for the drug discovery pipeline, facilitating the access to the chemical space and expecting to have available trillions of molecules in the following years. Despite the huge potential of these collections, adequate tools are needed to navigate these databases. Our lab has developed a pipeline to explore and navigate the massive collections of chemical entities. To validate the method, the lab has applied it to identify ligands for a highly studied test target, BRD4 (BD1). The aim of this work is to validate the hits obtained through the computational platform using a set of orthogonal biophysical techniques.