experimentally showed the first example of an acetylated histone interaction
with a bromodomain using a short peptide of H4 with the PCAF bromodomain
and determined a 346 μM K d via NMR titration. Such weak affinities would
subsequently be found to be a general characteristic of histone-bromodomain
interactions [20].
What would later pave the way for many biophysical and drug discovery studies,
the Zhou lab also characterized the first bromodomain small-molecule complex
with acetylated histamine [22]. Canonically, a conserved asparagine residue in
bromodomains typically forms a hydrogen bond with the acetyl group of histones
along with a water-mediated hydrogen bond with tyrosine. In this case, using
NOE-derived distance restraints, they localized binding to a hydrophobic cavity
between the BC and ZA loops. The mode of binding was distinct, as the N-acetyl
group on histamine did not engage N803. However, alanine mutagenesis subsequently showed that Y809 was essential for binding interactions.
The following year the first bromodomain crystal structures were solved by
Jacobsen et al. of a TAFII250 (also known as TAF1) tandem bromodomain [23]
and Owen et al. [18] of GCN5p, co-crystallized with a peptide of histone H4
containing residues 15–29 acetylated at lysine 16. In contrast to the results of
Zhou and co-workers, the GCN5p bromodomain structure showed the hydrogen
bond-mediated interactions of the acetyl group with the conserved asparagine side
chain on the BC loop (N407). A network of conserved water residues helped form a
water-mediated hydrogen bond with the phenolic hydroxyl of Y364. In addition to
hydrogen bonding, van Der Waals interactions form between the acetyl group and
F352 of the WPF shelf at the base of the pocket. Although an unacetylated histone
was shown to interact with GCN5p, HSQC NMR results disputed data from Ornaghi
et al. Their data favored a secondary low-affinity site with an arginine i + 3 residues
away [24] over a primary binding interaction with the unacetylated peptide. Wright
and co-workers would complement the GCN5p studies with a solution NMR
analysis of the protein alone and calculated affinities of acetylated H4 and H2A
histone peptides with dissociation constants of ~0.9 mM [25]. Further structural
analysis pointed out a negative electrostatic potential near the hydrophobic binding
cavity, consistent with binding interactions with the highly cationic peptides.
A network of five structural waters also identified by Owen et al. is conserved in
bromodomains. These water molecules need to be considered as targetable groups
within the protein complex rather than displaceable for achieving potent inhibition
when small molecules are designed to engage this pocket [18, 26]. These waters are
shown in Fig. 2 alongside additional bromodomain structural elements. One example demonstrating the importance of structured water interactions was shown by
Crawford et al. for tuning selectivity and affinity of a series of new bromodomain
inhibitors based on the pyrrolopyridone scaffold (Fig. 3a) [29]. Liu et al. showed a
structured water could also be displaced by a polar acetyl lysine mimic, using a
hydrazide [27]. Further, in a computational analysis of 24 bromodomains by Vidler
et al. [26], they determined the importance of considering 5 structural water molecules present in the acetylated lysine binding site for evaluating druggability.
Applied Biophysics for Bromodomain Drug Discovery
293
with a bromodomain using a short peptide of H4 with the PCAF bromodomain
and determined a 346 μM K d via NMR titration. Such weak affinities would
subsequently be found to be a general characteristic of histone-bromodomain
interactions [20].
What would later pave the way for many biophysical and drug discovery studies,
the Zhou lab also characterized the first bromodomain small-molecule complex
with acetylated histamine [22]. Canonically, a conserved asparagine residue in
bromodomains typically forms a hydrogen bond with the acetyl group of histones
along with a water-mediated hydrogen bond with tyrosine. In this case, using
NOE-derived distance restraints, they localized binding to a hydrophobic cavity
between the BC and ZA loops. The mode of binding was distinct, as the N-acetyl
group on histamine did not engage N803. However, alanine mutagenesis subsequently showed that Y809 was essential for binding interactions.
The following year the first bromodomain crystal structures were solved by
Jacobsen et al. of a TAFII250 (also known as TAF1) tandem bromodomain [23]
and Owen et al. [18] of GCN5p, co-crystallized with a peptide of histone H4
containing residues 15–29 acetylated at lysine 16. In contrast to the results of
Zhou and co-workers, the GCN5p bromodomain structure showed the hydrogen
bond-mediated interactions of the acetyl group with the conserved asparagine side
chain on the BC loop (N407). A network of conserved water residues helped form a
water-mediated hydrogen bond with the phenolic hydroxyl of Y364. In addition to
hydrogen bonding, van Der Waals interactions form between the acetyl group and
F352 of the WPF shelf at the base of the pocket. Although an unacetylated histone
was shown to interact with GCN5p, HSQC NMR results disputed data from Ornaghi
et al. Their data favored a secondary low-affinity site with an arginine i + 3 residues
away [24] over a primary binding interaction with the unacetylated peptide. Wright
and co-workers would complement the GCN5p studies with a solution NMR
analysis of the protein alone and calculated affinities of acetylated H4 and H2A
histone peptides with dissociation constants of ~0.9 mM [25]. Further structural
analysis pointed out a negative electrostatic potential near the hydrophobic binding
cavity, consistent with binding interactions with the highly cationic peptides.
A network of five structural waters also identified by Owen et al. is conserved in
bromodomains. These water molecules need to be considered as targetable groups
within the protein complex rather than displaceable for achieving potent inhibition
when small molecules are designed to engage this pocket [18, 26]. These waters are
shown in Fig. 2 alongside additional bromodomain structural elements. One example demonstrating the importance of structured water interactions was shown by
Crawford et al. for tuning selectivity and affinity of a series of new bromodomain
inhibitors based on the pyrrolopyridone scaffold (Fig. 3a) [29]. Liu et al. showed a
structured water could also be displaced by a polar acetyl lysine mimic, using a
hydrazide [27]. Further, in a computational analysis of 24 bromodomains by Vidler
et al. [26], they determined the importance of considering 5 structural water molecules present in the acetylated lysine binding site for evaluating druggability.
Applied Biophysics for Bromodomain Drug Discovery
293