was dampened by the development of resis- tance of the Bcr-Abl kinase to imatinib. The structural basis for this resistance involves the emergence of point mutations in the gene encoding Bcr-Abl, for example leading to the change of Thr315 to isoleucine (T315I), which prevents the drug from binding to the kinase. Studies have shown that a single residue in the ATP binding pocket of kinases, named the gatekeeper residue, controls sensitivity to small-molecule inhibitors. Mutation of the gatekeeper to a larger amino acid confers resistance to the drugs. For example, T315I enzymes are frequently deregulated in human disease. Several multitargeted tyrosine kinase inhibitors have shown impressive clinical activity, and their efficacy has been attributed to a selectivity profile that includes multiple targets 5,6 . Examples are the approved multi- target kinase inhibitors sorafenib, sunitinib and imatinib (Gleevec), which are currently in use for renal cancer, gastrointestinal stromal tumors and chronic myelogenous leukemia (CML), respectively. However, the early suc- cess of imatinib (which inhibits Bcr-Abl, c-Kit and PDGFR tyrosine kinases) in treating CML Kinases remove a phosphate group from ATP and covalently attach it to kinase substrates, thereby altering their function. Protein kinases mainly phosphorylate serine, threonine and tyrosine residues, whereas lipid kinases such as the phosphoinositide 3-kinases (PI3Ks) modify the inositol ring of phosphoinositides. Over the last decade, protein kinases have been an attrac- tive target for therapeutic intervention in many diseases, including inflammation, diabetes and cancer. Most kinase inhibitors target the ATP binding site of protein kinases and show competitive inhibition by binding to the affin- ity pocket, a conserved structural motif located close to the ATP binding pocket. Kinases recog- nize distinct substrates among others through subtle differences in their catalytic structures. These differences permit the development of relatively selective inhibitors. However, the diversity and the size of the kinome (comprising more than 500 kinases 1 ) challenge the develop- ment of truly selective inhibitors 2,3 . Typically drug discovery efforts have focused on develop- ing compounds that are specific for a particular target kinase, but inhibitors with broader speci- ficity may offer an improved therapeutic benefit. In this issue, Apsel et al. 4 identify compounds that inhibit both protein tyrosine kinases and lipid PI3Ks, both of which are well known as potent oncoproteins. Tyrosine kinases (such as c-Kit, Src and Bcr-Abl) are among the most targeted by phar- maceutical companies, mainly because these Killing two kinase families with one stone Benoit Bilanges, Neil Torbett & Bart Vanhaesebroeck Multitargeted protein kinase inhibitors have shown great promise in cancer therapy, but the selectivity profiles of these compounds have largely relied on serendipity or ‘off-target’ activities rather than rational drug design. Purposefully designed compounds with activity against multiple target kinases bring us a step closer to personalized medicine. Benoit Bilanges and Bart Vanhaesebroeck are in the Centre for Cell Signalling, Institute of Cancer, Barts and The London School of Medicine & Dentistry, Charterhouse Square, London EC1M 6BQ, UK. Neil Torbett is at Piramed Pharma, 957 Buckingham Avenue, Slough, Berkshire SL1 4NL, UK. e-mail: bart.vanh@qmul.ac.uk Figure 1 Size and shape selectivity of dual PI3K and tyrosine kinase inhibitors. The variable size and shape of the gatekeeper residue in both lipid and protein kinases create a “filter” to select the binding of the PP121 compound. A catalytic residue conserved between tyrosine kinases and PI3Ks is targeted by the PP121 compound. This interaction stabilizes the enzyme in an ‘active’ conformation, without allowing access to ATP and hence functionally inactivating the kinase. This interaction is responsible for the potency of the drug. H, hydrogen bond. Tyr kinase Ser/Thr kinase PI3K Gatekeeper residue Active ATP Inactive H Differentially oriented gatekeeper residue Active Inactive H PP121 inhibitor Bulky gatekeeper residue Active Active 648 VOLUME 4 NUMBER 11 NOVEMBER 2008 NATURE CHEMICAL BIOLOGY NEWS AND VIEWS © 2008 Nature Publishing Group http://www.nature.com/naturechemicalbiology