File(s) under permanent embargo

Reason: Restricted by author. A copy can be supplied under Section 51(2) of the Australian Copyright Act 1968 by submitting a document delivery request through your library or by emailing

PI3 kinase and inhibitors: targeting isoform selectivity

posted on 22.03.2017, 01:43 authored by Pinson, Jo-Anne
Phosphatidylinositol 3-kinases (PI3K) are enzymes that play important roles in many cell signalling pathways, including second messenger systems. These processes help regulate cell growth, proliferation, motility and survival. In some diseases, these key functions become deregulated, which may result in tumour growth, excessive platelet aggregation or inflammatory responses. Thus PI3K inhibitors have the potential to target cancer, cardiovascular disease and immuno-inflammatory diseases. In this thesis, investigation of several chemical classes for identification of isoform selective inhibitors of PI3K was undertaken. Utilizing reported X-ray structures of PI3K isoforms in the context of molecular modeling, a number of potent and selective inhibitors of PI3K isoforms were identified. While detection of inhibitors of the α-isoform for treatment of cancer is desirable, selective inhibitors can provide information in determining the nature of the binding interactions between PI3K isoforms. Therefore discovery of inhibitors of other isoforms are equally important. Biochemical and virtual screening methods were compared to assess virtual screening as a tool for discovery of isoform selective “hit” compounds. While assay screening identified a series of sub-micromolar compounds most active against PI3Kγ, no selective inhibitors of PI3Kα were identified. It was found that virtual screening experiments were heavily dependent upon the use of higher resolution liganded structures and were further improved by use of induced fit docking. In subsequent chapters, a strategy where selectivity is induced by progressive structural modification of non-selective lead compounds was examined. Specifically, targeted elaboration of the non-selective benzodioxol thiazolidinedione (Z)-5-(Benzo[d][1,3]dioxol-5-ylmethylene)thiazolidine-2,4-dione (3) yielded compounds most active against PI3Kγ or PI3Kβ. However, the observed selectivity was frequently detrimental to the compound potency. Utilizing the more potent, non-selective lead molecule ZSTK474, yielded a greater diversity of observed inhibitory responses across isoforms. Replacement of benzimidazole and morpholine fragments showed remarkable changes in selectivity or potency. For example, exchange of morpholine for a modified piperazine (4-(4-(2-(Difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-yl)piperazin-1-yl)(furan-2-yl)methanone (253) achieved preferential selectivity for PI3Kδ. Of great significance, the isonipecotic acid derivative 1-(4-(2-(Difluoromethyl)-1H-benzo[d]imidazol-1-yl)-6-morpholino-1,3,5-triazin-2-yl)piperidine-4-carboxylic acid (252) was a very potent pan-PI3K inhibitor, with sub-nanomolar potency for PI3Kδ. Replacement of benzimidazole for aryl moieties via Suzuki coupling achieved improved selectivity for PI3Kα. Introduction of the methoxy substituent at the 5 or 6 position of the 2-difluoromethylbenzimidazole moiety of ZSTK474 showed a ten-fold improvement in activity. In relation to this series, we have developed a methodology for the separation of regioisomers. Additionally, application of computational methods provided an explanation for the improved activity observed in some cases. In the synthesis of over one-hundred and ninety compounds, we have elucidated a series of inhibitors exhibiting a range of selectivities for PI3K isoforms. In addition, we have preserved or, in some examples, improved upon the activity of our non-selective lead molecules. This has required the design of additional synthetic and purification techniques imperative to final product integrity. Furthermore, utilization of virtual screening experiments has provided a means for recognition of active from decoy compounds and a structural explanation of observed biological activity in some instances. This project has successfully achieved all aims outlined and found the systematic modification of a potent, non-selective lead molecule can induce isoform selectivity providing a rational approach to drug design.


Campus location


Principal supervisor

Philip E Thompson

Additional supervisor 1

David Chalmers

Year of Award


Department, School or Centre

Medicinal Chemistry and Drug Action


Doctor of Philosophy

Degree Type



Faculty of Pharmacy and Pharmaceutical Sciences