Structure-based design and synthesis of inhibitors of the mitotic kinases Nek2 and CDK2

Abstract

Cyclin-dependent kinase 2 (CDK2) and Nek2 (Never-In-Mitosis A related kinases) are cell-cycle associated serine-threonine kinases that play an important role in the regulation of cellular proliferation and mitosis. Aberrant CDK2 and Nek2 activity are strongly associated with cancer, and inhibitors of these protein kinases are of potential therapeutic use as antitumour agents. Inhibitors of Nek2 –Previous studies identified a series of 2-aminoaryl-6-ethynylpurines (e.g. 42), as potent and selective irreversible Nek2 inhibitors with good antitumour activity in vitro and in vivo. 6-Ethynylpurine 42 binds within the ATP domain of Nek2 via a triplet of hydrogen bond interactions with the hinge region, enabling a covalent reaction between Cys-22 and the 6-ethynyl substituent. However, subsequent SAR studies indicated possible off-target activity for this series, which has been investigated through the synthesis and evaluation of control compounds engineered to be inactive against Nek2. With a view to abolishing Nek2-inhibitory activity without imposing dramatic structural changes, the effect of methylation at the purine N-7 (64) and N-9 (63) positions was studied. The corresponding isosteric 6-cyanopurine derivatives (65, 66, and 67) were also synthesised, which were expected to resemble closely 6-ethynylpurines without the capacity to react covalently. Evaluation of these derivatives in cell-based assays confirmed the presence of a growth-inhibitory activity unrelated to Nek2 inhibition. Regioselective N-7 methylation of 65 proved challenging, and a novel approach was developed whereby initial N-9 protection enabled selective methylation at the purine N-7 position, with concomitant loss of the N-9 protecting group giving the target purine 67. Kinetic studies have also been conducted with 42, 63 and 64 to assess the impact of N-7/N-9 methylation on the chemical reactivity of the 6-ethynyl ‘warhead’. A model system was developed to investigate a possible correlation between the chemical and biological reactivity of the 6-ethynyl group of the purine derivatives with thiols. Inhibitors of CDK2 –Previous studies have resulted in the identification of the purine CDK2 inhibitor 60, found to inhibit CDK2 in a time-dependent manner (IC50 = 63 nM) via conjugate addition of a lysine residue (Lys89), located in the ‘hinge region’ of the v ATP-binding domain, to the vinyl sulfone functionality of 60. This is thought to represent the first example of irreversible CDK2 inhibition, and prompted more detailed investigations with 60. Compound 60 exhibited a short half-life in both plasma (44 min) and medium (19 min), ascribed to hydration of the vinyl sulfone. Therefore, a priority was to improve the chemical stability of 60 without compromising inhibitory potency and time-dependent inhibition. Using the crystal structure of 60 in complex with CDK2 to guide inhibitor design, a range of derivatives (162-168) have been synthesised bearing α-substituents (R) on the vinyl sulfone group. In addition, synthetic methodology was developed for the synthesis of the corresponding 2-hydroxyalkyl products which are putative ATPcompetitive inhibitors. From the biological studies conducted on purine derivatives 162- 168, only the α-chlorovinyl sulfone compound 167 (IC50 (4 h) = 14 nM) has emerged as of particular interest, whereas others have shown competitive CDK2 inhibition, confirmed by X-ray crystallography. The β-position of the vinyl sulfone moiety has been briefly explored with purines (323, 333 and 339). In this series, one inhibitor (323) has been shown to bind covalently to CDK2 by structural biology studies. Alternative ‘warheads’ to the vinyl sulfone group have been investigated and exhibited competitive CDK2 inhibitory activity in a time-dependent inhibition assay. Finally, insertion of the vinyl sulfone warhead into known CDK2 inhibitors was investigated in preliminary studies.