Details of the activity of promising anticancer drugs known as BET inhibitors remain elusive. An approach called click chemistry enables in-depth analysis of how these drugs modulate the function of a crucial target protein, BRD4.
The transition from the discovery of bioactive small molecules to their use as effective therapeutic agents remains a considerable hurdle in drug development1. To this end, in a paper published in Science, Tyler et al.2 describe an interesting application of an approach called click chemistry that enables detailed characterization of the mechanisms of action and therapeutic potential of small molecules. The authors demonstrate that this approach can be used to derive valuable insights into the tissue, cellular and subcellular distribution of such molecules, as well as the molecular function of their protein targets.
Click chemistry involves the addition of a chemical group called a click linker3,4,5 to a compound of interest, enabling that compound to be covalently linked to useful molecules (for example, fluorescent tags) in an efficient and chemically selective manner. Importantly, the click linker is introduced in a way that does not alter the biological properties of the compound under study. Tyler et al. added click linkers to two compounds, JQ1 and IBET-762, that are closely related to drugs that show promise as treatments for cancer and other diseases. This enabled the authors to use click chemistry to investigate the properties of these compounds (Fig. 1).
Known as BET inhibitors, JQ1 and IBET-762 bind to the bromodomain regions of BET proteins, which typically interact with histone proteins around which DNA in the cell is packaged as chromatin. BET inhibitors prevent this interaction and therefore block the regulation of chromatin by BET proteins. A key target of JQ1 and IBET-762 is the BET protein BRD4, which contains two bromodomains (BD1 and BD2) and plays a central part in activating gene transcription in chromatin. Inhibition of BRD4 by these drugs causes downregulation of some, but not all, BRD4-associated genes. Yet the reason for this selectivity has been unclear.
The authors demonstrated using the 'clickable' drugs that JQ1 and IBET-762 act to displace BRD4 from chromatin more effectively at the regulatory DNA sequences that control transcription from a distance than at the sequences at which transcription actually starts. In addition, they found that binding between the BET inhibitors and BRD4 was markedly higher across the sequences of genes that are immediately downregulated following BET inhibition than across those that are downregulated later. The genes that are immediately downregulated are probably directly affected by the drugs, with at least some of the later effects of BET inhibition on gene expression representing secondary changes caused by the effect of modulating these directly targeted genes.
Tyler and colleagues showed that, for genes that are responsive to BET inhibitors, BRD4 binds histones through BD1, leaving BD2 free to be bound by the drugs. This highlights the potential of inhibitors that are specific to BD1, which might displace BRD4 from chromatin more effectively than inhibitors that bind to both of its bromodomains. By contrast, the binding of BRD4 to inhibitor-unresponsive genes was independent of the bromodomains, providing an explanation for the selective inhibition of a particular set of genes by BET inhibitors.
As well as using click chemistry to study drug–chromatin interactions, Tyler et al. investigated the range of proteins to which clickable JQ1 and IBET-762 could bind. Their analysis revealed that, despite belonging to different classes of chemical, the two drugs bind to the same protein complexes. This helps to explain why they have identical effects, and why resistance to one leads to resistance to the other6.
Next, Tyler and co-workers tested whether they could use click chemistry to discern differing intracellular drug concentrations. They incubated cultured cells with different concentrations of clickable BET inhibitors, 'clicked' the drugs onto fluorescent tags and mixed the resulting cell populations. Using a technique called flow cytometry, they could indeed identify and isolate cells exposed to different concentrations of drug by their level of fluorescence.
The researchers then used this approach to investigate drug uptake by cells from different tissues in a mouse model of acute myeloid leukaemia (a cancer for which BET inhibitors are currently in clinical trials). Following systemic administration in mice, intracellular concentrations of the drug JQ1 were considerably higher in leukaemic cells of the spleen than in those of the bone marrow. This provides a plausible explanation for the previous observation7 that treatment with BET inhibitors leads to the rapid clearance of acute myeloid leukaemic cells from the spleen, but not from the bone marrow.
Last, the investigators showed that the differences in drug concentration were influenced not only by the anatomical origin of the cells being studied, but also by cell type. In particular, levels of inhibitor were substantially higher in leukaemic blood cells than in normal blood cells, with the higher levels correlating with suppression of the target genes of BET proteins. This, according to the authors, might be why doses of BET inhibitor that have good anti-leukaemic effects in mice do not have a major impact on normal-blood-cell counts.
Collectively, Tyler and colleagues' experiments showcase the potential of click chemistry for characterizing the molecular mechanisms by which putative drugs act, thereby enhancing their preclinical evaluation and guiding their subsequent development as therapeutics. The study also demonstrates how small molecules can be used as molecular probes to investigate their own targets, be they proteins or other biomolecules. So far, methods to assess the biological interactions of small molecules such as BET inhibitors have been largely limited to using the small molecules as bait in vitro to isolate and identify their binding partners, along with any associated proteins8 or DNA9,10. As well as this type of analysis, click chemistry facilitates the investigation of drug distribution in vivo, in combination with an evaluation of the effects of a drug on the whole organism and its molecular targets.
It is probable that click chemistry will, in the future, be used to study diverse compounds beyond those that regulate chromatin. Its potential to quantify the tissue distribution of molecules of interest makes the technology relevant to many types of drug, irrespective of their target. In addition, the approach could enable the development of drugs that preferentially inhibit one subcellular function of a protein over another — crucial for proteins that have distinct roles in several compartments of the cell and that are differentially bound by the same molecules at various subcellular locations. Finally, notwithstanding these practical applications, the ability to define a drug's journey into an organism and then its cells and their compartments is, for the scientific enthusiast, a fantastic voyage.