27 May 2005: Inhibitors for specific protein kinases

The individual members of a large and important family of enzymes, the protein kinases, are very similar, which makes developing inhibitors for specific protein kinases a challenge. Now, US chemists have developed a bioinformatics-based approach and successfully designed inhibitors that target only two protein kinases.

Human cells contain 491 different protein kinases, which are responsible for phosphorylating certain amino acids. The enzymes are critically involved in a wide variety of cellular processes, including metabolism, transcription, and cell movement. They are linked to a number of diseases, including cancer, making them important drug targets.

Nearly all kinase inhibitors developed so far target the binding site for the nucleotide adenosine 5?-triphopshate (ATP). This site has a very similar structure in every protein kinase, so kinase inhibitors tend to inhibit a variety of protein kinases, rather than a select few.

Chemists at the University of California, San Francisco, led by Jack Taunton, have designed specific inhibitors by taking advantage of a selectivity filter in the ATP binding site known as the gatekeeper. The gatekeeper in most protein kinases is quite large and prevents large molecules from entering the site. But gatekeepers are smaller in around 20 per cent of kinases, and let in large compounds, such as inhibitors.

Taunton’s team realised that if they could find a protein kinase with both a small gatekeeper and a second selectivity filter in the ATP binding site, then they might be able to design an inhibitor that only targeted that specific kinase. For the second selectivity filter, they decided to look for the amino acid cysteine, which is commonly targeted by electrophilic inhibitors.

Using bioinformatics techniques, they investigated the structure of the ATP binding site in all 491 protein kinases, until they found two kinases, RSK1 and RSK2, that contained both selectivity filters. They then designed two inhibitors, based on pyrrolopyrimidines with attached halomethylketone compounds, and discovered that they only bound to and inhibited RSK1 and RSK2.

The chemists are now using a slightly different approach to develop specific inhibitors for other protein kinases. ’In order to target these kinases, which lack the threonine gatekeeper, we need to use alternative scaffolds,’ said Taunton. ’These compounds form covalent bonds with the reactive cysteine and exploit subtle differences in the hinge region and sugar binding pockets, rather than targeting a new selectivity filter.’ Jon Evans