The lipid kinase phosphoinositide 3-kinase γ (PI3Kγ) has attracted attention as a potential target to treat a variety of autoimmune disorders, including multiple sclerosis, due to its role in immune modulation and microglial activation. By minimizing the number of hydrogen bond donors while targeting a previously uncovered selectivity pocket adjacent to the ATP binding site of PI3Kγ, we discovered a series of azaisoindolinones as selective, brain penetrant inhibitors of PI3Kγ. This ultimately led to the discovery of 16, an orally bioavailable compound that showed efficacy in murine experimental autoimmune encephalomyelitis (EAE), a preclinical model of multiple sclerosis.

In acute myelogenous leukemia (AML), the FLT3 receptor tyrosine kinase (RTK) is highly expressed with 30% of patients expressing a mutated, constitutively active form of this protein. To inhibit this receptor, VX-322 was developed and found to be very potent against both the FLT3 and c-KIT RTKs with enzyme K(i) values of <1 nM and a cellular IC(50) between 1 and 5 nM. It was efficacious in a FLT3-ITD dependent myeloproliferative mouse model, doubling survival compared to other FLT3 inhibitors, with 25% of the mice cured. Upon treatment of primary AML patient blast cells, the dual inhibition of FLT3 and c-KIT was superior to inhibitors targeting a single RTK. Thus, this compound may represent an improved pharmacologic and selectivity profile that could be effective in the treatment of AML.

A high-throughput screen of our compound archive revealed a novel class of dual FMS-like tyrosine kinase 3 (FLT3)/c-KIT inhibitors. With the help of molecular modeling, this class was rapidly optimized for both potency against FLT3 and FLT3/c-KIT and excellent potency in cell-based assays, leading to dose-dependent cell death in acute myelogenous leukemia (AML) patient blast samples. Ultimately, the AML patient blast data defined the preferred target profile as we designed and evaluated a set of FLT3 selective and FLT3/c-KIT dual molecules. Further optimization for pharmacokinetic properties resulted in the selection of the dual FLT3/c-KIT inhibitor, N(3)-(4-(trans-4-morpholinocyclohexyl)phenyl)-1-(pyridin-2-yl)-1H-1,2,4-triazole-3,5-diamine, VX-322 (compound 37), to move forward to preclinical evaluation.

A series of substituted 3-aryl-6-amino-triazolo[4,3-b]pyridazines were identified as highly selective inhibitors of Pim-1 kinase. Initial exploration identified compound 24 as a potent, selective inhibitor, limited in its utility by poor solubility and permeability. Understanding the unusual ATP-binding site of the Pim kinases and X-ray crystallographic data on compound 24 led to design improvements in this class of inhibitor. This resulted in compound 29, a selective, soluble and permeable inhibitor of Pim-1.

A method is described which eliminates the spectral interference currently attributed to iron, during the determination of selenium. Carbon monoxide was used as an auxiliary gas at a flow-rate of 0.1 l min–1 to remove oxygen from the furnace. This prevented the formation of iron oxide which had been previously shown to be implicated in causing spectral interference. Matrix modification was performed by pre-treating the pyrolytic graphite platform with a 1%m/V platinum solution. Studies were carried out with aqueous solutions containing 70 ng ml–1 of Se, with Fe concentrations varying from 50 to 2000 µg ml–1, Recoveries ranged from 98.0 to 104.7%. Further studies were carried out on solutions containing 100 ng ml–1 and 200 ng ml–1 of Se and varying Fe concentrations. The method of standard additions was used to analyse these solutions. The procedure is rapid, simple and allows the use of deuterium-arc background correction during the determination of Se in water. Previously it was thought that a Zeeman-effect or Smith-Hieftje background correction system was required.