The unique disulfide linked activation loop of DYRK kinases and possible redox activity control

Abstract

The cysteine of HCD (C286) in DYRK1A is involved in disulfide bridge formation with a cysteine (C312) in the DFGSSC sequence. The purpose of this project was to investigate how the state of the disulfide bridge would affect enzyme catalytic and ligand binding properties of the protein kinase. A mutant, DYRK1A C312A, was thus designed to eliminate the disulfide bridge. The mutant was expressed and purified following the same protocol as for DYRK1A wt, including HisTrap purification, TEV cleavage and size exclusion chromatography. Crystallization trials were performed for both the wt and the mutant with the kinase inhibitor Staurosporine. DYRK1A wt with STU crystallized and diffracted with at a resolution of 2.33 Å. The DYRK1A C312A mutant with STU crystallized and diffracted with a resolution of 2.59 Å. The structure was solved by molecular replacement in Molrep (CCP4) and refined by Refmac5 and Phenix. Molecular dynamics (MD) simulations (SCHRODINGER) were performed with the intent to compare diverse disulfide bridge states. Ligand binding and enzyme catalytic properties were analyzed using a combination of techniques, including activity assays, microscale thermophoresis, and isothermal calorimetry. The Thermofluor assay confirmed that both the wt and the mutant bind tightly to STU and AZ-191. It also showed that the mutant consistently has a slightly lower melting temperature than the wt, which would indicate that it is less stable. Solvent accessible surface area (SASA) analysis support the theory of accessibility to conserved cysteine residues.