Chemical Biology of Oncogenic Gene Regulation
Targeted Protein Degradation
Concepts in the design of small-molecule drugs continue to follow a predictable path. Scientists typically aim to identify high-affinity binders of hydrophobic pockets to inhibit the activity of a target protein. While successful for enzymes or receptors, this inhibitor-centric paradigm forces scientists to seek innovation where it is easiest to look for. Consequently, 80% of the proteome, including some of the most compelling disease-relevant targets such as MYC, RAS, or b-catenin, remain beyond the reach of therapeutic development.
Our approach towards blocking otherwise “undruggable” proteins is to hijack the ubiquitin-proteasome system via small-molecules in order to chemically “earmark” these proteins for degradation by the proteasome. Targeted protein degradation is based on small molecules, generally called “degraders”, that induce the degradation of proteins by modulating the ubiquitination activity of E3 ligases. Overall, there are two types of small-molecule degraders, both of which are of interest to us: (i) non-chimeric compounds that act as molecular glues, or (ii) hetero-bifunctional molecules, often referred to as “proteolysis-targeting chimeras” (PROTACs). Both function by inducing associations of a target protein of interest (POI) and an E3 ubiquitin ligase. Induced molecular proximity consequently prompts ubiquitin transfer to the POI and its ensuing proteasomal degradation.
- Molecular Glues | MG Examples of molecular glues include thalidomide, lenalidomide and the related family of “immunomodulatory drugs” (IMiDs), certain aryl sulfonamides, as well as several plant hormones, including auxin. All of these small molecules bind substrate receptors (SR) of the cullin RING (CRL) family of E3 ligases. Binding induces or complements a neomorphic interaction surface that leads to the cooperative binding of a given neo-substrate, ubiquitin transfer, and its subsequent degradation by the proteasome. Intriguingly, many of these MGs induce the degradation of transcriptional regulators that would otherwise be considered as “undruggable”. Unfortunately, the discovery of MGs has been a process entirely driven by serendipity, and no rational discovery strategies exist. We are convinced that MGs could emerge as a key therapeutic strategy towards disrupting gene-regulatory circuits in cancer. We have identified scalable tools and phenotypic screens to identify novel MGs. In the lab, we are using functional genetic screens and quantitative proteomics approaches to further characterize novel MGs. Upon mechanistic characterization, the goal is to further optimize identified MGs by the means of medicinal chemistry to develop innovative, well-characterized chemical probes that might represent a vector for ensuing drug-development programs outside of academia.
- Heterobifunctional Degraders | PROTACs PROTACs are based on a modular design where a targeting warhead and an E3-ligase binder are connected via a flexible linker. PROTACs can thus simultaneously bind to an E3 ligase and a POI, enforce their molecular proximity, and induce POI ubiquitination and degradation. Due to this modular architecture, PROTACs can in principle be adapted to different POIs simply by exchanging the targeting warhead. This potential versatility rendered PROTACs particularly interesting for the pharmaceutical development and led to a recent increase in the number of probes for preclinical target exploration.
We have recently reported the first in vivo compatible heterobifunctional degrader (Winter et al. Science 2015). Conjugation of a competitive BET-bromodomain antagonist to an IMiD-like phthalimide moiety afforded the heterobifunctional compound dBET1. dBET1 induced fast, potent and selective degradation of BET proteins BRD2, BRD3 and BRD4 in cell lines and in mouse xenograft experiments, and continuous degrader treatment outperformed competitive BET inhibition in disseminated AML xenografts. We furthermore showed that phthalimide conjugation is a generalizable approach by additionally reporting heterobifunctional degraders of the prolyl isomerase FKBP12. More recently, it has emerged that converting a small-molecule binder into a heterobifunctional degrader can offer means to engineering selectivity in otherwise promiscuous chemical scaffolds. In a recent manuscript, we teamed up with researchers at the Dana Farber Cancer Institute to describe novel degraders that allow the homolog-selective degradation of the kinase CDK6, a prolific target for the treatment of Acute Myeloid Leukemia (Brand et al., Cell Chem Bio 2018).
Gene Control in Cancer
Aberrant gene-control is a multifaceted key feature of human cancer. Dysregulated transcriptional circuits can arrest cancer cells in an undifferentiated, proliferative state by preventing definitive lineage commitment. In other instances, aberrant transcriptional networks might prompt the continuous overexpression of an oncogene, cause a dormant, highly drug-resistant cell state, or cause a primary tumor cells to colonize distant organs. Despite the paramount importance of transcription for arguably all key aspects of cancer, the spectrum of drugs capable to block or remodel aberrant gene control is very limited.
Our lab is interested in understanding the molecular basis of aberrant gene-regulation in cancer, particularly focusing on blood cancers, but also on malignancies driven by mutant Ras proteins. Towards that end, we combine acute protein ablation systems (“degradation Tag”/ “dTAG”), or direct small-molecule inhibitors and degraders (PROTACs or MGs; see above) with global and unbiased measurements of gene activity and genome structure. To capitalize on the high kinetic resolution of protein degradation systems, our focus is on nascent RNA-sequencing methods. We believe that a quantitative understanding of fundamental aspects of human gene-control is key to the innovation of novel therapeutic strategies. Recent work has elucidated the role of BET proteins as master regulators of pause release (Winter et al., Mol Cell 2017), and identified the YEATS domain containing protein ENL as a therapeutic target to inhibit AML-specific transcriptional circuits (Erb et al., Nature 2017).
Chemical Genetics and Chemoproteomics
As a chemical-biology lab, the identification of protein targets of (anti-cancer) small-molecules, as well as the elucidation of their mechanism-of-action is our great passion. In our efforts to find and characterize novel molecular glues that degrade non-drugged cancer targets, we employ two sub-disciplines of chemical biology:
- Chemical Genetics To understand cellular effectors required for drug-action, or to identify the biologically relevant target of small-molecule drugs, we apply a combination of the various functional-genomics strategies, including genome-wide, pooled CRIPSR/Cas9 screens, NGS-based profiling of drug-resistant sub-clones, and cellular barcoding studies to monitor the evolution of drug-resistant clones.
- Chemical Proteomics For the vast majority of small-molecules of synthetic or natural origin, we don’t understand their associated spectrum of target proteins. This is of particular relevance for small molecules that are identified through chemical screens with a phenotypic, cellular readout. Moreover, most small-molecules are much more promiscuous than anticipated. This means that they don’t only bind one protein in a cellular context, but might modulate a range of closely related, but also completely dissimilar protein targets. Towards functionally and mechanistically annotating small-molecules with a non-understood anti-cancer phenotype, we are employing target-identification assays that are based on chemical proteomics. In chemical proteomics, small-molecules of interest are immobilized on beads and exposed lysates of cells of interest. Proteins bound to the drug-affinity resin are subsequently identified via unbiased, quantitative proteomics, and validated via orthogonal strategies.
Georg Winter obtained his PhD from the Medical University of Vienna, working on elucidating the mechanism of action of anti-neoplastic drugs under the supervision of Prof. Giulio Superti-Furga at CeMM. He specialized in proteomics as well as chemical genetic approaches to identifying drug resistance mechanisms and on mechanistically elucidating synergistic drug combinations. He continued his training in chemical biology, working as a postdoctoral fellow with Dr. James Bradner at the Dana Farber Cancer Institute/Harvard Medical School. There, he innovated a generalizable pharmacological solution to in vivo target protein degradation and applied this strategy to the study of leukemic gene regulation. Georg Winter was recruited as a CeMM Principal Investigator in June 2016. His lab develops and applies methods for target protein degradation with the ultimate goal of understanding and disrupting oncogenic transcriptional circuits. To that end, the Winter laboratory combines phenotypic drug screens, chemical genetics and drug-target identification approaches with holistic measurements of global gene activity and genome structure. The ultimate goal of the research conducted in the Winter laboratory is to connect basic research in gene regulation and the ubiquitin-proteasome system with functional genomics and chemical probe development to develop novel and personalized therapeutic paradigms.
Mayor-Ruiz C, Jaeger MG, Bauer S, Brand M, Sin C, Hanzl A, Mueller AC, Menche J, Winter GE. Plasticity of the Cullin-RING Ligase Repertoire Shapes Sensitivity to Ligand-Induced Protein Degradation. Molecular Cell. 2019 Aug 22;75(4):849-858.e8. DOI: 10.1016/j.molcel.2019.07.013. (abstract)
Brand M, Jiang B, Bauer S, Donovan KA, Liang Y, Wang ES, Nowak RP, Yuan JC, Zhang T, Kwiatowski N, Mueller AC, Fischer ES, Gray NS, Winter GE. Homolog-Selective Degradation as a Strategy to Probe the Function of CDK6 in AML. Cell Chem Biol. 2018 Nov (Epub ahead of print) (abstract)
Winter GE, Mayer A, Buckley DL, Erb MA, Roderick JE, Vittori S, Reyes JM, di Iulio J, Souza A, Ott CJ, Roberts JM, Zeid R, Scott TG, Paulk J, Lachance K, Olson CM, Dastjerdi S, Bauer S, Lin CY, Gray NS, Kelliher MA, Churchman LS, Bradner JE. BET Bromodomain Proteins Function as Master Transcription Elongation Factors Independent of CDK9 Recruitment. Mol Cell. 2017 Jul 6 67, 5–18. (abstract)
Erb MA, Scott TG, Li BE, Xie H, Paulk J, Seo HS, Souza A, Roberts JM, Dastjerdi S, Buckley DL, Sanjana NE, Shalem O, Nabet B, Zeid R, Offei-Addo NK, Dhe-Paganon S, Zhang F, Orkin SH, Winter GE, Bradner JE. Transcription control by the ENL YEATS domain in acute leukaemia. Nature. 2017 Mar 9;543(7644):270-274. (abstract)
Winter GE, Buckley DL, Paulk J, Roberts JM, Souza A, Dhe-Paganon S, Bradner JE. DRUG DEVELOPMENT. Phthalimide conjugation as a strategy for in vivo target protein degradation. Science 2015. Science 348, 1376-81. (abstract)