Research Article: Mechanistic insights into the phosphoryl transfer reaction in cyclin-dependent kinase 2: A QM/MM study

Date Published: September 4, 2019

Publisher: Public Library of Science

Author(s): Rodrigo Recabarren, Edison H. Osorio, Julio Caballero, Iñaki Tuñón, Jans H. Alzate-Morales, Matthias Stein.


Cyclin-dependent kinase 2 (CDK2) is an important member of the CDK family exerting its most important function in the regulation of the cell cycle. It catalyzes the transfer of the gamma phosphate group from an ATP (adenosine triphosphate) molecule to a Serine/Threonine residue of a peptide substrate. Due to the importance of this enzyme, and protein kinases in general, a detailed understanding of the reaction mechanism is desired. Thus, in this work the phosphoryl transfer reaction catalyzed by CDK2 was revisited and studied by means of hybrid quantum mechanics/molecular mechanics (QM/MM) calculations. Our results suggest that the base-assisted mechanism is preferred over the substrate-assisted pathway when one Mg2+ is present in the active site, in agreement with a previous theoretical study. The base-assisted mechanism resulted to be dissociative, with a potential energy barrier of 14.3 kcal/mol, very close to the experimental derived value. An interesting feature of the mechanism is the proton transfer from Lys129 to the phosphoryl group at the second transition state, event that could be helping in neutralizing the charge on the phosphoryl group upon the absence of a second Mg2+ ion. Furthermore, important insights into the mechanisms in terms of bond order and charge analysis were provided. These descriptors helped to characterize the synchronicity of bond forming and breaking events, and to characterize charge transfer effects. Local interactions at the active site are key to modulate the charge distribution on the phosphoryl group and therefore alter its reactivity.

Partial Text

Cyclin-dependent kinases is a family of Serine/Threonine kinases that phosphorylate peptide substrates using adenosine triphosphate (ATP) as phosphate source with a unique function in the regulation of the cell cycle [1]. As their names state, they depend on the binding of a cyclin protein in order to be fully activated [2–4] and also on the phosphorylation of specific residues [5–9]. In particular, cyclin-dependent kinase 2 (CDK2), which can bind Cyclin E or A, needs to be phosphorylated at Thr160 [10]. CDK2 helps in the progression from G1 to S phase during the cell cycle and its malfunctioning, e.g. by mutations, has been related to different human cancers [11]. For this reason, many CDK inhibitors have been proposed, which in the majority of the cases bind in the ATP binding site. Also, and most recently, new inhibitors have been designed to tackle protein-protein interactions and allosteric sites [12]; however, the development of potent and selective inhibitors has been very challenging and in many cases disappointing, without getting through clinical trials [13]. In this context, the precise knowledge of the phosphoryl transfer mechanism in CDKs, and kinases in general, could lead to new strategies for the development of more potent and selective drugs. Moreover, CDK2 is a very interesting system to be used as a model of study within the CDK family due to the availability of many crystallographic structures and kinetic data [10,14–18].

CDK2 is a very important kinase that is considered a study model for the cyclin-dependent kinase family. As other kinases, this enzyme catalyzes the phosphoryl transfer reaction from an ATP molecule to a peptide substrate containing a Ser o Thr residue to be phosphorylated. Though this system has been subject of experimental and computational studies, the two proposed reaction mechanisms, substrate-assisted and base-assisted, had not been studied until now within the same computational approach, which is a requisite in order to have a proper comparison between them. In this context, our results suggest that the base-assisted mechanism is the preferred path, at least when one Mg2+ ion is used as cofactor, with an estimated potential energy barrier of 14.3 kcal/mol, which is in good agreement with the experimentally derived value. This mechanism is stepwise, in contrast to the substrate-assisted mechanism that is concerted. Both mechanisms show TSs that have a high dissociative character. Interestingly, a new feature in the base-assisted mechanism has been observed: a spontaneous proton transfer from Lys129 to one of the oxygen atoms of the transferred phosphoryl group. This event takes place late in the reaction progress, at the second TS, with the proposed effect of neutralizing the negative charge on the phosphoryl group. We recognize that a major limitation in the present work is the absence of a second Mg2+ ion in the active site, which is expected to modulate the energy barrier and possibly the mechanism. These aspects are expected to be covered in future work, and therefore, the prevalence of the base-assisted mechanism remains to be confirmed. Despite this, the calculated energy barrier seems to be reasonable, and agrees with the experimental data, what would be suggesting that the reaction could still be carried out with only one Mg2+ ion in the active site.