Skip to main content

Bump and Hole approach to elucidate function of individual bromodomains of BET proteins

Periodic Reporting for period 1 - Bump-BET (Bump and Hole approach to elucidate function of individual bromodomains of BET proteins)

Reporting period: 2015-10-01 to 2017-09-30

Small molecules can be used to probe the biological function of individual proteins, but achieving high selectivity against structurally similar paralogues remains a challenging task. The Bromo and Extra-Terminal (BET) proteins brd2, brd3, brd4 and brdt play important roles in transcriptional regulation by controlling networks of genes involved in cellular proliferation and cell cycle regulation. Key to the functions of BET proteins is a pair of highly homologous bromodomains (BD) in tandem, which bind to chromatin and recognize histone acetylation on its tail. Misregulation of BET proteins activity has been linked to disease states. Elucidation of the processes controlled by BET proteins would benefit greatly from chemical probes that perturb individual BDs with exquisite control and selectivity. Potent small molecule inhibitors have been developed against BET BD, however their lack of selectivity against individual BD within the family renders these compounds unsuitable to interrogate the function of individual proteins or their individual domains. A “bump-and-hole” approach is employed to untangle the function of each single BD. A mutation is introduced in the BD of interest making a “hole” in the ligand binding pocket while retaining its biological function. On the other hand, a “bump” modification is made to the inhibitor scaffold so that the “bump” is tolerated by the “hole” on the mutant BD which the derivative is selective for the mutant over WT. A discriminative and fine-tuned pair of BD mutant / BET-inhibitor (I-BET) derivative was obtained with the best compound demonstrated >100 fold selectivity towards BD mutant over WT, as measured by isothermal titration calorimetry.
At the same time, the researcher pursued a closely related project using a proteolysis targeting chimera (PROTAC) approach to target BET proteins. The researcher contributed to the report of the first-ever PROTAC-induced E3-target ternary complex and led an SAR study to understand E3 ligand selectivity in the context of ternary complex formation. These works provide structural basis for PROTAC-mediated target protein degradation.
The host lab has developed a bump-and-hole system for the BET BD, pairing a leucine/alanine mutation with an ethyl-derived analogue of an established benzodiazepine scaffold based on I-BET762 [1]. Building on top of this, the system is optimized with the introduction of a more conservative and less disruptive leucine/valine (L/V) mutation. The L/V BD mutant retains good binding to chromatin peptide comparable to the WT protein. An extensive SAR study of diverse benzodiazepine analogues is performed, which identifies potent, mutant-selective inhibitors with desirable physiochemical properties. Along with the SAR study, eleven crystal structures are solved for the L/V mutant alone and in complex with the representative benzodiazepine analogues providing precious information for structure-guided analogue design (figure 1B). The best compound shows about 200 nM potency, with >100-fold selectivity for the L/V mutant over WT. Through a variety of in vitro and cellular assays the capabilities of the optimized system is validated. Work for the bump-and-hole until this point is written up in an article which is submitted to a peer-reviewed journal and is currently under revision. Generation of cell line incorporating these mutation by CRISPR is on the way which will provide a setting closer to physiological condition.
In the PROTAC approach project, BET proteins are targeted by linking BET-inhibitor, JQ1, to E3 ligase VHL (Von Hippel–Lindau) ligand [2]. The researcher characterized and profiled the activity of these PROTAC molecules in cell and in vitro. The first PROTAC molecule reported by the host lab, MZ1, although based on a pan-selective BET-inhibitor scaffold, demonstrates a selective substrate degradation preference against BRD4 over BRD2 and BRD3. In a collaborative work within the Ciulli lab, first crystal structure of a PROTAC induced complex of E3-ligase with its target protein is reported [3]. A model explaining the observed selectivity is proposed and leads to the development structure-designed PROTAC, e.g. AT1, which shows exquisite selectivity for BRD4.
The researcher also leads a SAR study of another series of BET-targeting PROTAC molecules [4]. The researcher showed that by substituting the targeting recruitment warhead would result in a completely different selectivity profile against BET proteins. Extensive work on in vitro E3-PROTAC-BD complex formation was performed and the results were backed up by cellular protein degradation study.

Overview of the results
During the project period, an optimized bump-and-hole system on the BET BD has been established. It provides selective chemical tools for future cellular and in vivo target validation studies. The PROTAC approach demonstrates a strategy to achieve selective and efficient protein knock-down by small molecule. Structural basis for efficient PROTAC activity is established, supported by extensive SAR study both in vitro and in cell. This provides valuable knowledge for PROTAC design in the developing field.
The researcher’s work contributes to two peer-reviewed research papers on PROTAC [3,4] and one paper under revision on bump-and-hole . His contribution is also recognized by the field as he is invited as speaker at the 12th Annual Drug Discovery Chemistry Conference (2017) held in San Diego on Targeted Protein Degradation by Small Molecules. Knowledge gains in the project also enables him to co-author a review paper on the topic of chemical genetics in epigenetics [5].

1. Baud M.G.J. et al., Science 2014, 346(6209), 638-641.
2. Zengerle M. et al., ACS Chemical Biology 2015, 10(8):1770-7
3. Gadd M.S. et al. Nature Chemical Biology 2016, 13, 514–521.
4. Chan K.-H. et al. Journal of Medicinal Chemistry 2017, DOI: 10.1021/acs.jmedchem.6b01912
5. Runcie A. C. et al. Current Opinion in Chemical Biology 2016, 33, 186–194.
Although there are already BET inhibitors in clinical trials and numerous studies at preclinical stage, function of each member in the BET family is still not fully understood [1]. This has already raised concern in the community that there is not enough scientific basis to use pan-selective inhibitor in treatment [2]. The optimized chemical probe/ mutant pair we develop will provide tools for the science community to better understanding of BET proteins and eventually lead to more promising and safer treatment.
PROTAC is an approach very different from the conventional inhibition model. In theory, this can be applied to any target when a specific ligand is available, no matter where it binds on the target. This is a promising approach to targets previously regarded as non-druggable as an active site is not a pre-requisite. Work done by the researcher and his colleagues provides structural basis of PROTAC activity. This set an example for structure-based PROTAC development in the field. Work on PROTAC also starts to gain recognition in the board scientific community, for example the researcher was invited to the 12th Annual Drug Discovery Chemistry Conference (2017) and our work is recommended by F1000Prime [3].

1. French C.A. Adv. Cancer Res. 2016, 131, 21-58.
2. Andrieu G. et al., Drug Discov. Today Technol. 2016, 19, 45-50.
3. Murray B., 2017 Retrieved from
Schematic diagram of the project.