by Fred L. Ciske and Thomas G. Brock, Ph.D., Cayman Chemical

A key concept in the field of epigenetics is the generation of persistent changes in gene expression without changing DNA sequence. These persistent changes involve groups of genes regulated in a coordinated manner, resulting in shifts in the ‘epigenetic landscape,’ first proposed by Conrad Waddington in the 1940s. Today, such changes are considered to drive such processes as cellular differentiation from pluripotent embryonic stem cells to specific lineages or the dedifferentiation of mature cells into cancer cells. Changes in the expression of groups of genes may be ascribed to a variety of factors, such as chromatin remodeling. Current research focuses on revealing the molecular details behind these changes. This review touches on lysine acetylation and its role in changing gene expression and highlights a unique protein degradation strategy notably demonstrated on bromodomains (BRDs).

Lysine Acetylation and Bromodomains

Remarkably, ε-N-acetylation of lysine residues on proteins is one of the most frequently occurring post-translational modifications, with more than 3,600 catalogued lysine acetylation sites on 1,750 proteins.¹ The levels of histone acetylation are maintained by two families of enzymes: the histone acetyltransferases (HATs) and histone deacetylases (HDACs). Acetylation of histones regulates gene transcription, DNA repair, and chromatin condensation.² These effects are ultimately determined by the pattern of acetylation marks.

BRDs are the modules on certain proteins that act as the readers of ε-N-lysine acetylation marks placed on histones and other proteins. They are present in diverse nuclear proteins, including BRD and extra-terminal domain (BET) family proteins, HATs (GCN5, PCAF), chromatin-remodeling enzymes (BAZ1B, SMARCA), methyltransferases (MLL, ASH1L), transcription factors (TAF1), and nuclear-scaffolding proteins (PB1). Dysfunction involving BRDs has been implicated in broad categories of diseases, including cancer, obesity, type 2 diabetes, and inflammation.^3,4 Examples of more specific diseases associated with mutations or fusions of genes expressing proteins with BRDs include veno-occlusive disease with immunodeficiency syndrome (SP110), X-linked mental retardation (BRWD3), and infant pro-B acute lymphoblastic leukemia (ALL).^5-7

The BET family protein BRD4 contains tandem BRDs on its N-terminal half (the ET portion resides on the C-terminal portion). The first BRD of the tandem pair displays high affinity for acetylated sites on histone 4 (H4), particularly Lys5, whereas the second BRD binds promiscuously to several acetylated lysines.⁸ Full length BRD4 preferentially binds polyacetylated H4 and the specific BRD4 inhibitor JQ1 abrogates this interaction.⁹ BRD4 plays key roles in cellular proliferation, including binding nucleosomes during M phase when most nuclear regulatory factors are released into the cytoplasm in response to a global stop of transcription.¹⁰ Knockdown of BRD4 in mouse embryonic stem cells suppresses Nanog expression and abolishes self-renewal of stem cells.¹¹ Thus, BRD4 binds H4 in a regulated manner, and binding affects proliferation.

BRD4, bound to acetylated H4, serves both as a docking site and a modulator of proteins that regulate gene expression. For example, the ET portion of BRD4 interacts with the positive transcription elongation factor P-TEFb and causes the release of inhibitory proteins, allowing efficient RNA Pol II-mediated transcription.¹² BRD4 promotes P-TEFb-dependent phosphorylation of Ser2 on the carboxy-terminal domain of Pol II, activating its elongation state. The ET portion of BRD4 can also bind several other proteins, including NSD3, a histone methyltransferase, and JMJD6, an arginine demethylase, that function to activate transcription.¹³

Interest in BRDs as a group is underscored by the extensive development of inhibitors as selective chemical probes (www.thesgc.org) and clinical therapeutics.^14,15 Likely the most studied of these, JQ1 displaces BRD4 from nuclear chromatin at nanomolar concentrations, inducing cell cycle arrest and initiating apoptosis in a variety of cancer cells.¹⁶ While JQ1 is not being tested in clinical trials due to its short half-life in vivo, it remains a valuable tool compound and has recently been used, along with its analogs, in a design strategy that effectively changes its function from an inhibitor to a BRD degrader.

PROTACs: Tagging Proteins for Destruction

As BRD inhibitors are currently under clinical evaluation, a relatively recent protein degradation strategy has demonstrated new potential and may circumvent some of the compensatory mechanisms associated with enzyme inhibition.¹⁷ Proteolysis Targeting Chimeras (PROTACs) recruit the cells’ own housekeeping machinery, the ubiquitin-proteasome system (UPS), to selectively destroy target proteins rather than just inhibit them.^18-20 PROTACs are hetero-bifunctional molecules consisting of two separate but linked structure elements: one binds a target protein of interest while the other engages an E3 ubiquitin ligase for ubiquitin tagging and subsequent proteasomal degradation (Figure 1). This tagging of the protein of interest is an event-driven approach that allows for multiple rounds of binding and, therefore, multiple protein targets may be removed per PROTAC molecule.²¹ Traditionally ‘undruggable’ proteins (scaffolds, weak binders, etc.) can also be targeted since any transient binding interaction with the target protein could be exploited and, depending upon a target’s propensity for ubiquitination, a promiscuous inhibitor could potentially be rendered a selective degrader.

Figure 1. Proteolysis Targeting Chimera (PROTAC) recruitment of an E3 ligase for target protein degradation via the ubiquitin-proteasome pathway.

Thus far, only a few of the more than 600 encoded E3 ligases have been engaged by prepared PROTACs, with von Hippel-Lindau (VHL) and Cereblon (CRBN) E3 ligase components being employed to induce degradation of BRD proteins. These PROTAC molecules were constructed using an alkyl or polyethylene glycol (PEG) linker to join the known BRD2, BRD3, and BRD4 inhibitors JQ1 or OTX015 to peptide-like VHL binding moieties or to CRBN binders like pomalidomide as shown in Figure 2 for ARV-825.^22-25

 

Figure 2. A bifunctional PROTAC molecule ARV-825. The BRD4 protein binder (OTX015) is tethered to a known E3 ligase recognition motif (pomalidomide).

In target protein degradation studies, ARV-825 dramatically knocked down BRD4 levels within 6 hours, and the effect lasted more than 24 hours. Effectiveness of the PROTAC was evidenced by the fact that OTX015 and CRBN moieties separately display binding affinities of 10 nM and 3 µM for their respective targets while the degradation constant (DC₅₀) of ARV-825 was 1 nM, suggesting a catalytic effect.²³ In a separate leukemia study, ARV-825 was synergistic with co-administered JAK inhibitor ruxolitinib and induced high levels of apoptosis in ruxolitinib-resistant cells. Studies on ARV-771, a JQ1-linked, VHL-binding PROTAC, demonstrated cellular activity and delayed leukemia progression in mouse xenografts.^24,26

Both BRD4 and ERK1/2 kinase degradation was demonstrated in HeLa cells using the ‘click’ chemistry approach, wherein PROTAC molecules were generated in situ from separate partner-reactive motifs (Figure 3).²⁷ With the aim of improving drug-like properties of solubility and cell permeability, in-cell generation of these ‘CLIPTACs’ may circumvent potentially difficult linker optimization and expand the inhibitor toolkit available to biologists.

Figure 3. CLIPTAC approach uses 'click' chemistry linkage (blue) to enable in-cell self-assembly of a protein degrader (3).

Key to the success of future PROTAC drug efforts will be the discovery and development of relevant E3 ligase recognition motifs (degron mimics) and their utilization to generate optimal binderlinker-E3 ligase motif combinations with good in vivo properties. PROTAC analog libraries might be generated, for instance, by appending different linker-E3 ligand combinations to a non-critical position on the ligand of the protein of interest (Figure 4).^28,29

 

Figure 4. PROTAC library possibilities. Degree of target protein degradation and drug-like properties might be optimized by ‘mix and match’ of variable linker-E3 ligand combinations for linking to a single functional group on the target protein ligand.

In addition to BRDs and the other proteins mentioned here, PROTAC technology may have broad potential for application over much of the proteome. This will surely be aided by the further elucidation of cell context-dependent ubiquitination processes, E3 ligase activation mechanics, and identification of new E3 ligase binders. PROTACs targeting nuclear hormone receptors, oncoproteins, kinases, and tau proteins have been successfully demonstrated, and new target proteins, including those currently described as ‘undruggable’, offer exciting possibilities for this emerging therapeutic approach.^17,30

References

1. Choudhary, C., Kumar, C., Gnad, F., et al. Science 325(5942), 834-840 (2009).

2. Kouzarides, T. Cell 128, 693-705 (2007).

3. Denis, G.V. Discov. Med. 10(55), 489-499 (2010).

4. Belkina, A.C. and Denis, G.V. Nat. Rev. Cancer 12(7), 465-477 (2012).

5. Bueno, C., Montes, R., Catalina, P., et al. Leukemia 25(3), 400-410 (2011).

6. Cliffe, S.T., Bloch, D.B., Suryani, S., et al. J. Allergy Clin. Immunol. 130(3), 735-742 (2012).

7. Field, M., Tarpey, P.S., Smith, R., et al. Am. J. Hum. Genet. 81(2), 367-374 (2007).

8. Filippakopoulos, P., Picaud, S., Mangos, M., et al. Cell 149(1), 214-231 (2012).

9. Filippakopoulos, P., Qi, J., Picaud, S., et al. Nature 468(7327), 1067-1073 (2010).

10. Dey, A., Ellenberg, J., Farina, A., et al. Mol. Cell Biol. 20(17), 6537-6549 (2000).

11. Liu, W., Stein, P.E., Cheng, X., et al. Cell Death Differ. 21(12), 1950-1960 (2014).

12. Jang, M.K., Mochizuki, K., Zhou, M., et al. Mol. Cell 19(4), 523-534 (2005).

13. Rahman, S., Sowa, M.E., Ottinger, M., et al. Mol. Cell Biol. 31(13), 2641-2652 (2011).

14. Filippakopoulos, P. and Knapp, S. Nat. Rev. Drug Discov. 13(15), 337-356 (2014).

15. Fujisawa, T. and Filippakopoulos, P. Nat. Rev. Mol. Cell Biol. (2017).

16. Wadhwa, E. and Nicolaides, T. Cureus 8(5), e620 (2016).

17. Lai, A.C. and Crews, C.M. Nat. Rev. Drug Discov. 16(2), 101-114 (2017).

18. Sakamoto, K.M., Kim, K.B., Kumagai, A., et al. Proc. Natl. Acad. Sci. USA 98(15), 8554−8559 (2001).

19. Bargagna-Mohan, P., Baek, S.H., Lee, H., et al. Bioorg. Med. Chem. Lett. 15(11), 2724−2727 (2005).

20. Galdeano, C., Gadd, M.S., Soares, P., et al. J. Med. Chem. 57(20), 8657−8663 (2014).

21. Toure, M. and Crews, C.M. Angew. Chem. Int. Ed. Engl. 55(6), 1966–1973 (2016).

22. Winter, G.E., Buckley, D.L., Paulk, J., et al. Science 348(6241), 1376–1381 (2015).

23. Lu, J., Qian, Y., Altieri, M., et al. Chem. Biol. 22(6), 755–763 (2015).

24. Raina, K., Lu, J., Qian, Y., et al. Proc. Natl. Acad. Sci. USA 113(26), 7124–7129 (2016).

25. Zengerle, M., Chan, K.H., and Ciulli, A. ACS Chem. Biol. 10(8), 1770–1777 (2015).

26. Saenz, D.T., Fiskus, W., Qian, Y., et al. Leukemia (2017).

27. Lebraud, H., Wright, D.J., Johnson, C.N., et al. ACS Cent. Sci. 2(12), 927-934 (2016).

28. Lai, A.C., Toure, M., Hellerschmied, D., et al. Angew. Chem. Int. Ed. Engl. 55(2), 807–810 (2016).

29. Bondeson, D.P. and Crews, C.M. Annu. Rev. Pharmacol. Toxicol. 57, 107-123 (2017).

30. Chu, T.-T., Gao, N., Li, Q.Q., et al. Cell Chem. Biol. 23(4), 453-461 (2016).