Protein kinase R (PKR)-like endoplasmic reticulum kinase (PERK) is activated in response to a variety of endoplasmic reticulum stresses implicated in numerous disease states. Evidence that PERK is implicated in tumorigenesis and cancer cell survival stimulated our search for small molecule inhibitors. Through screening and lead optimization using the human PERK crystal structure, we discovered compound 38 (GSK2606414), an orally available, potent, and selective PERK inhibitor. Compound 38 inhibits PERK activation in cells and inhibits the growth of a human tumor xenograft in mice.

We recently reported the discovery of GSK2606414 (1), a selective first in class inhibitor of protein kinase R (PKR)-like endoplasmic reticulum kinase (PERK), which inhibited PERK activation in cells and demonstrated tumor growth inhibition in a human tumor xenograft in mice. In continuation of our drug discovery program, we applied a strategy to decrease inhibitor lipophilicity as a means to improve physical properties and pharmacokinetics. This report describes our medicinal chemistry optimization culminating in the discovery of the PERK inhibitor GSK2656157 (6), which was selected for advancement to preclinical development.

Phosphoinositide-dependent protein kinase-1(PDK1) is a master regulator of the AGC family of kinases and an integral component of the PI3K/AKT/mTOR pathway. As this pathway is among the most commonly deregulated across all cancers, a selective inhibitor of PDK1 might have utility as an anticancer agent. Herein we describe our lead optimization of compound 1 toward highly potent and selective PDK1 inhibitors via a structure-based design strategy. The most potent and selective inhibitors demonstrated submicromolar activity as measured by inhibition of phosphorylation of PDK1 substrates as well as antiproliferative activity against a subset of AML cell lines. In addition, reduction of phosphorylation of PDK1 substrates was demonstrated in vivo in mice bearing OCl-AML2 xenografts. These observations demonstrate the utility of these molecules as tools to further delineate the biology of PDK1 and the potential pharmacological uses of a PDK1 inhibitor.

The syntheses of the optically pure asymmetric hydroborating agents 1 (a, R = Ph; b, R = TMS) in both enantiomeric forms are reported. These reagents are effective for the hydroboration of cis-, trans- and trisubstituted alkenes. More significantly, they exhibit unprecedented levels of selectivity in the asymmetric hydroboration of 1,1-disubstituted alkenes (28-92% ee), a previously unanswered challenge in the nearly 50 year history of this reagent-controlled process. For example, the hydroboration of alpha-methylstyrene with 1a produces the corresponding alcohol 6f in 78% ee (cf., Ipc2BH, 5% ee). Suzuki coupling of the intermediate adducts 5 produces the nonracemic products 7 very effectively (50-84%) without loss of optical purity.

A multigram synthesis of the C29-C51 subunit of altohyrtin C (spongistatin 2) has been accomplished. Union of this intermediate with the C1-C28 fragment and further elaboration furnished the natural product. Completion of the C29-C51 subunit began with the aldol coupling of the boron enolate derived from methyl ketone 8 and aldehyde 9. Acid-catalyzed deprotection/cyclization of the resulting diastereomeric mixture of addition products was conducted in a single operation to afford the E-ring of altohyrtin C. The diastereomer obtained through cyclization of the unwanted aldol product was subjected to an oxidation/reduction sequence to rectify the C35 stereocenter. The C45-C48 segment of the eventual triene side chain was introduced by addition of a functionalized Grignard reagent derived from (R)-glycidol to a C44 aldehyde. Palladium-mediated deoxygenation of the resulting allylic alcohol was followed by adjustment of protecting groups to provide reactivity suitable for the later stages of the synthesis. The diene functionality comprising the remainder of the C44-C51 side chain was constructed by addition of an allylzinc reagent to the unmasked C48 aldehyde and subsequent dehydration of the resulting alcohol. Completion of the synthesis of the C29-C51 subunit was achieved through conversion of the protected C29 alcohol into a primary iodide. The synthesis of the C29-C51 iodide required 44 steps with a longest linear sequence of 33 steps. From commercially available tri-O-acetyl-d-glucal, the overall yield was 6.8%, and 2 g of the iodide was prepared. The C29-C51 primary iodide was amenable to phosphonium salt formation, and the ensuing Wittig coupling with a C1-C28 intermediate provided a fully functionalized, protected seco-acid. Selective deprotection of the required silicon groups afforded an intermediate appropriate for macrolactonization, and, finally, global deprotection furnished altohyrtin C (spongistatin 2). This synthetic approach required 113 steps with a longest linear sequence of 37 steps starting from either tri-O-acetyl-d-glucal or (S)-malic acid.