Periodic Reporting for period 2 - Glue2Degrade (Therapeutic hijacking of E3 Ligases)
Periodo di rendicontazione: 2021-07-01 al 2022-12-31
To this day, the innovation of new therapeutic concepts is predominantly based on small molecules that bind to the active site of enzymes or transporters to inhibit their underlying biochemical activity. In brief: the community is looking for inhibitors. Because of this concept, our community has delineated the human proteome into a “druggable” and an “undruggable” subset, asking the question which proteins can be functionally inhibited by small-molecule drugs. Based on this delineation, only around 20% of all human proteins are within the reach of small-molecule drugs. In other words: there are proteins that are known as highly valuable and promising targets for therapies since decades. However, the obvious benefits from pharmacologically blocking these proteins could not be realized to this day, simply because of an “inhibitor-centric” concept in the discovery of novel small-molecule drugs.
This highlights the need for new therapeutic concepts to expand the subset of the proteome that is amenable for therapeutic exploration. One promising approach relies on small molecules that induce or stabilize protein-protein interactions. Recent years have seen a resurgence of this concept. In part, this is motivated by the observation that small-molecule-induced recruitment of proteins into the proximity of E3 ubiquitin ligases is sufficient to trigger target ubiquitination and degradation by the proteasome. Frequently, this approach of chemically reprogramming E3 ligases is now referred to as targeted protein degradation (TPD). Current efforts in TPD predominantly focus on heterobifunctional degraders, so-called PROteolysis TArgeting Chimeras (PROTACs). PROTACs consist of two separate warheads that are connected by a flexible linker. Based on this design principle, PROTACs can however only degrade proteins that they can also bind in isolation. This again hampers efforts to induce the targeted and controlled degradation of truly “undruggable” proteins.
This is where the Glue2Degrade proposal sets in, aiming to study general principles of reprogramming ubiquitin ligases via monovalent small-molecule degraders or so-called “molecular glue degraders” (MGDs). MGDs can recruit target proteins into the vicinity of E3 ligases via a highly cooperative mechanism. Put pragmatically, this means that they can induce the degradation of proteins without having to bind to these proteins in isolation. MGDs thus enable the degradation of unligandable and undruggable proteins.
In the Glue2Degrade proposal, we hypothesize that molecular glue degraders might be much more frequent than we are currently anticipating. Via two orthogonal discovery strategies, we aim to use phenotypic profiling in genetically engineered cell systems to find novel degraders. Moreover, we develop assays to report on regulatory ligase dynamics to find MGDs that reprogram specific ligases in a defined manner.
Upon successful completion of Glue2Degrade, we will uncover rational strategies for the scalable discovery of MGDs. These rulesets and strategies will be employed by us and others (academic and commercial sectors) to identify small molecules that induce the targeted and selective degradation (and hence elimination) of disease-causing proteins that are currently outside the reach of traditional pharmacologic approaches. We thus anticipate that work in Glue2Degrade breaks new ground and motivates the development of therapies for live-threatening diseases.
This highlights the need for new therapeutic concepts to expand the subset of the proteome that is amenable for therapeutic exploration. One promising approach relies on small molecules that induce or stabilize protein-protein interactions. Recent years have seen a resurgence of this concept. In part, this is motivated by the observation that small-molecule-induced recruitment of proteins into the proximity of E3 ubiquitin ligases is sufficient to trigger target ubiquitination and degradation by the proteasome. Frequently, this approach of chemically reprogramming E3 ligases is now referred to as targeted protein degradation (TPD). Current efforts in TPD predominantly focus on heterobifunctional degraders, so-called PROteolysis TArgeting Chimeras (PROTACs). PROTACs consist of two separate warheads that are connected by a flexible linker. Based on this design principle, PROTACs can however only degrade proteins that they can also bind in isolation. This again hampers efforts to induce the targeted and controlled degradation of truly “undruggable” proteins.
This is where the Glue2Degrade proposal sets in, aiming to study general principles of reprogramming ubiquitin ligases via monovalent small-molecule degraders or so-called “molecular glue degraders” (MGDs). MGDs can recruit target proteins into the vicinity of E3 ligases via a highly cooperative mechanism. Put pragmatically, this means that they can induce the degradation of proteins without having to bind to these proteins in isolation. MGDs thus enable the degradation of unligandable and undruggable proteins.
In the Glue2Degrade proposal, we hypothesize that molecular glue degraders might be much more frequent than we are currently anticipating. Via two orthogonal discovery strategies, we aim to use phenotypic profiling in genetically engineered cell systems to find novel degraders. Moreover, we develop assays to report on regulatory ligase dynamics to find MGDs that reprogram specific ligases in a defined manner.
Upon successful completion of Glue2Degrade, we will uncover rational strategies for the scalable discovery of MGDs. These rulesets and strategies will be employed by us and others (academic and commercial sectors) to identify small molecules that induce the targeted and selective degradation (and hence elimination) of disease-causing proteins that are currently outside the reach of traditional pharmacologic approaches. We thus anticipate that work in Glue2Degrade breaks new ground and motivates the development of therapies for live-threatening diseases.
In this first period of Glue2Degrade, we have successfully delivered a proof of concept that chemical profiling in engineered cell systems can lead to the identification of novel molecular glue degraders. Based on earlier work (Mayor-Ruiz et al., Mol Cell 2019) we have identified that cells mutant for the gene UBE2M feature impaired activity of up to two hundred different E3 ligases. Based on this finding, we have devised comparative chemical profiling in UBE2Mmut cells to find compounds that require a CRL ligase to convey their mechanism. Coupling this screening paradigm to orthogonal target ID methods, this led to the identification of novel glue degraders that destabilize the protein Cyclin K (Mayor Ruiz, Nat Chem Biol 2020). Based on this concept, we have in the meantime identified a suite of additional small molecules that might act as MGDs, and that we are mechanistically following up in the meantime.
In addition, we have developed an additional discovery approach for novel MGDs. This rests on time-resolved probing of impaired ligase dynamics in intact cells following drug treatment. Based on a variety of different control experiments, we could validate this approach, leading us to identify novel molecular glue degraders that re-program the activity of the E3 ligase DCAF15. The description of this strategy as well as the chemical and mechanistic characterization of the identified, novel glue degraders was published in the Journal of American Chemical Society (JACS) (Hanzl et al., JACS 2022).
In addition, we developed methodologies to characterize functional hotspots on E3 ligases that are hijacked for neosubstrates degradation via molecular glue degraders and PROTACs. In brief, we devised a multi-layered functional genomics approach that allows us to probe, in cellulo, how E3 ligases recognize neosubstrates at a single amino acid resolution. Interestingly, the hotspots identified via our scalable methodology proved to be also of clinical relevance since we could show that patients that relapsed from treatment with approved degraders lenalidomide and pomalidomide presented with mutations in the identified hotspots. These results were published in Nature Chemical Biology (Hanzl, 2022).
Lastly, using FACS-based CRISPR/Cas9 screens, we have identified a novel molecular-glue like modality. In a collaboration with Alessio Ciulli’s lab, we identify and characterize the first of its kind “intramolecular glue degrader”. This bifunctional molecule acts in cis, meaning it binds two adjacent protein domains on the neosubstrate BRD4 and rearranges them. This rearrangement causes the stabilization of a pre-existing interaction between BRD4 and the E3 ligase DCAF16, leading to BRD4 ubiquitination and degradation in a DCAF16-dependent manner. These data have been published in a preprint (Hsia, Hinterndorfer, Cowan et al., biorxiv 2023).
In addition, we have developed an additional discovery approach for novel MGDs. This rests on time-resolved probing of impaired ligase dynamics in intact cells following drug treatment. Based on a variety of different control experiments, we could validate this approach, leading us to identify novel molecular glue degraders that re-program the activity of the E3 ligase DCAF15. The description of this strategy as well as the chemical and mechanistic characterization of the identified, novel glue degraders was published in the Journal of American Chemical Society (JACS) (Hanzl et al., JACS 2022).
In addition, we developed methodologies to characterize functional hotspots on E3 ligases that are hijacked for neosubstrates degradation via molecular glue degraders and PROTACs. In brief, we devised a multi-layered functional genomics approach that allows us to probe, in cellulo, how E3 ligases recognize neosubstrates at a single amino acid resolution. Interestingly, the hotspots identified via our scalable methodology proved to be also of clinical relevance since we could show that patients that relapsed from treatment with approved degraders lenalidomide and pomalidomide presented with mutations in the identified hotspots. These results were published in Nature Chemical Biology (Hanzl, 2022).
Lastly, using FACS-based CRISPR/Cas9 screens, we have identified a novel molecular-glue like modality. In a collaboration with Alessio Ciulli’s lab, we identify and characterize the first of its kind “intramolecular glue degrader”. This bifunctional molecule acts in cis, meaning it binds two adjacent protein domains on the neosubstrate BRD4 and rearranges them. This rearrangement causes the stabilization of a pre-existing interaction between BRD4 and the E3 ligase DCAF16, leading to BRD4 ubiquitination and degradation in a DCAF16-dependent manner. These data have been published in a preprint (Hsia, Hinterndorfer, Cowan et al., biorxiv 2023).
In the further course of the Glue2Degrade project, we continue to find novel mechanisms of how small molecules can rewire E3 ubiquitin ligases. Some of these methods, such as the aforementioned intramolecular gluing, are entirely unexpected, yet might present a rationalizable strategy to develop novel degraders based on the derived molecular insights. Moreover, we expect that the mechanistic insights derived by our work will motivate ensuing translational efforts that are per se outside the scope of this proposal.