Cyclin dependent kinases (CDKs) belong to a family of serine/threonine protein kinases that play a key role in cell cycle and transcriptional regulation. CDK9, in complex with its regulatory partner cyclin T1, is involved in RNA transcriptional regulation through phosphorylating the RNA polymerase II. Inhibition of CDK9 can target multiple cancer-relevant pathways by downregulating the transcriptionally inducible genes of cancer cells, such as cell cycle regulators and antiapoptotic factors. Although many CDK inhibitors have been in clinical trials, there is only one CDK9-selective compound (BAY1143572) in trials. Therefore, it is still necessary to study other highly selective CDK9 inhibitors. Through the analysis of the cocrystal structure of lead compound 5-(2-((3-(1,4-diazepan-1-yl)phenyl)amino)-pyrimidin-4-yl)-N,4-dimethyl-thiazol-2-amine (5) bound to CDK2 and CDK9, compound (5) favours the “inward” conformation in CDK9, while it adopts both the “inward” and “outward” conformations in CDK2. Aurora A and aurora B (ARKs) inhibitors are also found to adopt the similar “outward” conformation” in ATP binding site of ARKs. Therefore, an “inward” conformation macrocyclic structure was designed to improve the selectivity against CDK9 over CDK2 and ARKs. Three series of macrocyclic compounds were designed, synthesised and tested against CDK9, CDK2, and ARKs. The first series macrocyclic compounds with amide linker shows poor inhibitory activity against CDK9, and the molecular docking study shows these macrocyclic compounds fail to reside in the ATP binding site of CDK9. Following that, ring closing metathesis (RCM) was exploited as an alternative cyclisation strategy. Although the synthesis work was unsuccessful at the first onset, compound 6-methyl-12-oxa-3,6-diaza-2(4,2)-pyrimidina-1,4(1,3)-dibenzenacyclodo-decaphan-8-ene (43) with a similar structure as designed compounds was introduced as the model compound to optimise the RCM reaction. The thiazole group was then demonstrated to be the interfering factor in RCM and compound 11-oxa-3-aza-2(4,2)-pyrimidina-5(1,4)-piperazina-1,4(1,3)-dibenzenacyclo-undecaphan-7-ene (44) with a replacing phenyl group was successfully synthesised and showed poor activity against CDK2 and ARKs (>219 nM for these three kinases). According to the published paper regarding lead compound (5), extensive substituents modification (-CH3, -CN, -F) of compound (44) afforded 11-oxa-3-aza-2(4,2)-pyrimidina-5(1,4)-piperazina-1,4(1,3)-dibenzenacyclo-undecaphan-7-ene-1-carbonitrile (79) as the most selective CDK9 inhibitor. It shows 60-fold selectivity for CDK9 over CDK2 and 90-fold selectivity for CDK9 over ARKs. A further Diversity Kinase Profile Screening demonstrates the high selectivity of compound (79) that only tyrosine-protein kinase Lyn(h) appeared as the off-target. To improve the inhibitory activity against CDK9, alternative macrocyclic pyrimidine systems were designed, and compound 1-methoxy-4-(4-methylpiperazine-1-carbonyl)-3,8-diaza-2(4,2)-pyrimidina-1(1,3),4(1,2)-dibenzenacyclononaphan-7-one (89) showed an attractive Glide score based on molecular docking studies. Because of the time limit, the synthetic work for compound (89) has not been finished.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:748363 |
| Date | January 2018 |
| Creators | Gao, Rui |
| Publisher | University of Nottingham |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://eprints.nottingham.ac.uk/50294/ |
Page generated in 0.0128 seconds