Periodic Reporting for period 1 - MapInDegraders (Closing the gap: studying non-obvious ligand induced degradation events)
Período documentado: 2021-04-01 hasta 2023-03-31
Cancer remains a challenging disease with an urgent need for new therapeutic strategies. One emerging new pharmacology aiming to address this demand is termed targeted protein degradation. Instead of inhibiting the disease-causing protein, the drugs of this class lead to its removal. Despite this strategy coming with multiple advantages, the design of so-called degrader molecules has been challenging. On the one side, the implementation of PROTACs, or proteolysis targeting chimeras, suffers from large unfavorable molecular weights, leading to suboptimal pharmacological properties such as cell penetration. On the other side, smaller degrader molecules do exist, e.g. molecular glue degraders, but their discovery has often been quite serendipitous. We hypothesized that an untapped resource may be hidden in the already developed molecules that fall under this inhibitory-centric paradigm. Indeed, for these inhibitors sporadic reports already exemplified that these can lead to target (here the disease causing protein) destabilization. A quantitative and qualitative understanding of these processes is, however, largely lacking and hampers any translation efforts. In this project we thus explored these non-obvious inhibitor-induced protein destabilization events for one major drug target class: the protein kinases. This work thus aids closing the gap of our understanding how small molecules which were initially designed to inhibit a protein can also lead to their degradation. This may thus provide an alternative route to engineer degrader molecules and could pave the way towards novel, alternative therapeutic strategies in the battle against cancer.
First, we generated a detailed dataset that monitors kinase abundance upon inhibitor perturbation. In total, we profiled ˜100 kinases for ˜1.6k compounds generating thousands of temporal abundance profiles. Next, we developed a computational classifier tool to dissect the data and distinguish global from local perturbations. Global perturbations were excluded as they come with many confounding factors and can largely be attributed to perturbations of major cellular processes such as impairment of transcription and translation, or also general compound toxicity. After trimming down the data, many events remained, importantly spanning a broad range of different kinases. We subjected this dataset to rigorous computational analysis elucidating if chemical or protein properties may drive the phenotype. Importantly, we discovered a sensitivity of mutated kinases particularly in comparison to their non-mutated counterpart for being destabilized upon compound binding. Since many diseases are reliant on mutated kinases, this could provide a first starting point for novel strategies to perturb cancer in a mutant-selective manner. Finally, we focused on four kinases with highly favorable screening results and elucidated their detailed mode of action. The scope of different avenues by which compounds can induce protein destabilization was remarkable and we identified multiple unique mechanisms by which the cell can cope with degrading protein kinases that have been perturbed by small molecule binding.
The generated dataset provides a rich database for inhibitor-induced kinase destabilization and will be made available for the wider research community upon publication of the results. We uncovered many interesting destabilization events within this dataset that all can provide novel insights into how the cellular machineries and processes can be leveraged to implement targeted protein degradation. In addition, we have elucidated multiple of the identified examples which already exemplify the diversity of the mode of actions these small molecules can have. Going beyond inhibition and completely removing a disease-causing protein from the cell provides unique advantages for disease therapy. This study thus provides a first large-scale dataset exploring the untapped potential of small molecules to function as degraders even though they were initially designed as inhibitors. We focused on protein kinases reflective of their biological importance in many cancers. However, potentially unifying features hidden in the generated dataset may provide a rational approach to translate the degrader modalities also to other druggable protein classes.