From 1 mM to 19 nM (50,000 fold increase)
Aurora kinases are a family of serine/threonine kinases that are essential for cell division via regulating mitosis.
Researchers have proposed them as a novel therapy target for many cancers, with in vivo and in vitro studies supporting this. There are 3 types of Aurora Kinase, A, B and C. Aurora A kinase is activated through its interaction with Targeting protein for Xenopus kinesin-like protein 2 (TPX2) and represents a promising target for cancer therapy.
While several Aurora Kinase inhibitors have been pushed through to clinical trial, they bind competitively with ATP. This presents a challenge in selectively targeting Aurora A. Its ATP-binding site is highly conserved with Aurora B and C. Focusing on the Aurora A-TPX2 protein-protein interaction could allow for selectivity towards Aurora A. This work is the first to report an Aurora A-TPX2 protein-protein interaction inhibitor.
Hit identification started with a fragment screen of 600 compounds in a thermal shift assay. Importantly, they carried out this assay in the presence of an ATP-site inhibitor to focus on fragments binding to other sites. The assay hits were then tested in ligand-based NMR experiments. A TPX2 peptide fragment was used to identify fragments that bind to the same area as the protein-protein interaction. ITC was also used to confirm these hits bound to the kinase.
A representative hit from the initial screening was 3-hydroxybenzoic acid (1), which had an affinity of 1mM as measure by a fluorescence polarization assay. Adding a 4-trifluoromethoxy phenyl group (2) to this hit raised its potency to 63 µM. This compound was key. It allowed for a crystal structure with Aurora A. This enabled a structure-based drug design program. The phenol group could be replaced by the bioisosteric indole group which could make the same hydrogen bond interaction with Glu170.
Interestingly the crystal structure showed the carboxylic acid group twisted out of the plane with respect to the phenyl ring. Their analysis of ligands in the PDB and CSD supports that carboxylic acids are more often in plane with the aromatic ring. I believe this is due to the conjugation between the groups. They proposed that a methyl group placed ortho to the carboxylic acid could minimize the loss of binding due to the energetic penalty of twisting this group. Their hypothesis was correct, adding a methyl at the C-7 position of the indole improved potency 10-fold. Further optimization of the phenyl ring and replacing the carboxylic acid with an acyl sulfamide gave the lead compound CAM2602. The lead compound has an affinity of 19 nM, a 50,000 fold increase from the initial hit molecule. Pretty impressive.
To make this compound, a Bartoli indole synthesis is used to form the indole ring. A Miyaura Borylation Reaction is used to make the boronate species.
The coupling partner is made through an SnAr reaction with the substituted phenol and 2-bromopyridine. The suzuki reaction is then carried out under microwave conditions. Ester hydrolysis and carboxylic acid activation with CDI allow for the acyl sulfamide formation using DBU as a base.
The compound went on to show no activity in the DiscoverX KINOMEscan kinase selectivity panel. Importantly it showed 1000-fold selectivity over Aurora kinase A vs B. The compound went to in vivo studies with the researchers commenting “ In a solid tumor xenograft model, oral delivery of CAM2602 successfully elicited biomarkers of target engagement, increasing PH3-positive cells and decreasing the proportion of those cells positive for P-Thr288 Aurora A; moreover, this compound also reduced tumor growth. These results show that an inhibitor of the Aurora A-TPX2 PPI is a viable route to a therapeutic intervention in cancer.
This article is open access so you can check out the whole paper for free. I recommend taking a look at the crystal structures. https://doi.org/10.1021/acs.jmedchem.4c01165
This is my first newsletter, so I’d love any feedback you have (good or bad). You can let me know by replying to this email.
Also, consider forwarding this email to friends and colleagues who you think may be interested to help the newsletter grow. They can sign up at https://chemistrycapital.co.uk/newsletter
Have a great week :)
Luke.