At least seven different CDKs have been reported to be implicated in the cell cycle regulation in vertebrates. Among these, CDK2 functions during the progression of cell cycle from the G1 to S phase,. CDK2, like most of the other CDKs, follows a two-step process to become fully functional: the association with the regulatory subunit �C cyclin A or cyclin E, phosphorylation of residue Thr160 located in the so-called activation loop,. However, certain CDKs, e.g. CDK5 do not follow this mode of activation. The activity of CDK5 is restricted to nervous system by the localization of its activators p25/p35/p39, the binding of which makes CDK5 fully active without the subsequent requirement of phosphorylation of the activation loop residue,. While aberrant activity of CDK2 has been identified in a number of diseases Y-27632 including cancer, embryonic lethality, male sterility etc., the deregulation of CDK5 causes serious neurodegenerative disorders, e.g. Alzheimer��s disease, lateral sclerosis, stroke etc. CDKs are highly homologous and contain a conserved catalytic core. For example, CDK2 and CDK5 share a sequence homology of 60%, with the substrate binding pocket alone showing nearly 93% sequence similarity,. The 3D structures of CDKs are mainly composed of two domains, the N and the C-terminal domains,. The catalytic cleft that binds ATP is located at the interface of these two domains. A glycine rich loop, commonly known as G-loop, lies above the ATP binding pocket and is conserved in many kinases. The primary function of this loop is to align the substrate and ATP correctly, for a smooth transfer of the c-phosphate. The N-terminal domain is primarily composed of a b-sheet, containing five antiparallel bstrands, and one a-helix. This helix with the ����PSxAxRE���� motif is a signature of this class of proteins and constitutes the main point of interaction with activator proteins. The loop which precedes the PSxAxRE helix, known as the 40s loop, also interacts with the activator protein. The C-terminal domain is predominantly ahelical and contains the so-called T-loop, the residue Thr160 of which becomes phosphorylated by CAK for CDK2 activation. However, CAK does not phosphorylate CDK5 on the analogous Ser159,. The catalytic pockets of CDK2 and CDK5 are primarily comprised of 20 LY2109761 residues, three of which differ from CDK2 to CDK5 as follows: Lys83 to Cys83, His84 to Asp84 and Asp145 to Asn144. The respective partner proteins, Cyclin E and p25, though have less sequence homology, are structurally similar with both possessing the typical cyclin box fold. Due to their key regulatory roles, CDKs have become important pharmaceutical targets for inhibitor design,. There is a particular demand for CDK5 specific inhibitors to treat various neurodegenerative diseases. However, it is difficult to design the inhibitor specific to a particular CDK due to the structural homology among number of CDKs. Very recently, Helal et al. have identified novel cis-substituted cyclobutyl-4- aminoimidazole inhibitors that gave improved enzyme and cellular potency against CDK5/p25 with up to 30-fold selectivity over CDK2/Cyclin E. To understand the molecular basis of higher potency of these inhibitors, here we carry out all-atom molecular dynamics simulations of active CDK5/p25 and CDK2/ CyclinE bound to a series of cyclobutyl-substituted imidazole inhibitors. The atomic details of the stereospecificity and selectivity of these inhibitors are obtained from energetics and binding characteristics to the CDKs. The N-acetyl analogue of cis-OH, cis-N-acetyl has shown a tenfold improved potency over cis-OH against CDK5/p25 in vitro. Moreover, it showed a sevenfold better selectivity for CDK5 over CDK2.