Development of Second-Generation Indole-Based Dynamin GTPase Inhibitors Christopher P. Gordon, † Barbara Venn-Brown, † Mark J. Robertson, † Kelly A. Young, † Ngoc Chau, ‡ Anna Mariana, ‡ Ainslie Whiting, ‡ Megan Chircop, ‡ Phillip J. Robinson, ‡ and Adam McCluskey* ,† † Chemistry, Centre for Chemical Biology, School of Environmental and Life Sciences, The University of Newcastle, University Drive, Callaghan, NSW 2308, Australia ‡ Cell Signaling Unit and Cell Cycle Unit, Children’s Medical Research Institute, The University of Sydney, 214 Hawkesbury Road, Westmead, NSW 2145, Australia * S Supporting Information ABSTRACT: Focused library development of our lead 2-cyano-3-(1-(3-(dimethylamino)- propyl)-2-methyl-1H-indol-3-yl)-N-octylacrylamide (2) confirmed the tertiary dimethyla- mino-propyl moiety as critical for inhibition of dynamin GTPase. The cyanoamide moiety could be replaced with a thiazole-4(5H)-one isostere (19, IC 50(dyn I) = 7.7 μM), reduced under flow chemistry conditions (20, IC 50(dyn I) = 5.2 μM) or replaced by a simple amine. The latter provided a basis for a high yield library of compounds via a reductive amination by flow hydrogenation. Two compounds, 24 (IC 50 (dyn I) = 0.56 μM) and 25 (IC 50(dyn I) = 0.76 μM), stood out. Indole 24 is nontoxic and showed increased potency against dynamin I and II in vitro and in cells (IC 50(CME) = 1.9 μM). It also showed 4.4-fold selectivity for dynamin I. The indole 24 compound has improved isoform selectivity and is the most active in-cell inhibitor of clathrin-mediated endocytosis reported to date. ■ INTRODUCTION Dynamin is a large GTPase known to play a crucial role in membrane remodelling, notably during endocytosis. 1 Three dynamin genes exist, and while the expressed proteins have over 80% homology, they show differential tissue distribution, suggesting they may have distinct biological roles. Common to all are four domains, a GTPase domain (required for vesicle fission), 2,3 a pleckstrin homology domain (targeting domain and potentially a GTPase inhibitory module), 4 a bundle signaling element (which controls dynamin self-assembly into rings), 5,6 and a proline-rich domain (which interacts with proteins containing an SH3 domain 2,4 and is the site for dynamin phosphorylation in vivo). 7,8 Two crystal structures of mammalian dynamin I have been solved. 9,10 Each lack the PRD domain but provided insights into the complex dynamics involved in oligomerization and how dynamin may act as a scission protein in cells. Endocytic mechanisms serve a variety of cellular functions including the uptake of cellular nutrients, regulation of cell- surface receptor expression and signaling, antigen presentation, and maintenance of synaptic transmission. Clathrin mediated endocytosis (CME) is one such mechanism and in mammalian cells its usual function is internalization of extracellular materials. In neuronal cells, this is a specialized pathway termed synaptic vesicle endocytosis (SVE) and its role is internalization of membrane to allow vesicle recycling. 2,4 SVE and CME thus perform distinct functional roles in different cell types but share the same underlying protein machinery. The differences in SVE and CME mostly arise from which dynamin gene product is present. Dynamin II (dynII) is found throughout the body and is the major scission protein for CME, while dynamin III (dynIII) is limited primarily to the brain and testes and at low levels in a few other tissues. The dynamin I (dynI) gene product is only found in neuronal cells where it coexists with the other two forms. DynI is expressed in these cells at about 50-fold higher levels compared with either dynIII or dynII. This high level of dynI was expected to be critical for neuronal cell function, however knockout studies have shown both dynII and especially dynIII are able to facilitate SVE with limited success, hinting to a range of redundancies within the endocytic network. 11 A crucial role has thus been suggested for dynI to allow SVE to occur across a broad range of neuronal activities and has been implicated as a potential target for epilepsy. 1 Control of endocytic pathways presents immense therapeutic potential with defects in such pathways being involved in multiple disease states such as Alzheimer’s disease, Hunting- ton’s disease, Stiff-person syndrome, Lewy body dementias, and Niemann-Pick type C disease. 12-15 Endocytic pathways are also utilized by viruses, including HIV, 16 and bacteria to gain entry into cells. 17 Links to genetic disease are also emerging with mutations in dynII causing Charcot-Marie-Tooth disease and centronuclear myopathy, and dynII mutations are involved in T-cell precursor acute lymphoblastic leukemia. 18-20 Recently attention has also turned toward nonendocytic roles that dynamin plays in the body including those involving the Received: June 15, 2012 Article pubs.acs.org/jmc © XXXX American Chemical Society A dx.doi.org/10.1021/jm300844m | J. Med. Chem. XXXX, XXX, XXX-XXX