Protein engineering and rational mutagenesis in the crystallization and structure determination of human Cyclin B1

Cyclin B binds and activates the Cyclin Dependent Kinase, CDK1, forming a heterodimeric complex known as the Mitosis Promoting Factor, which drives mitosis in all eukaryotes from yeast to humans. Using rational mutagenesis and protein engineering, we have identified an active, soluble construct of Human Cyclin B1, and report here the crystal structure to 2.9 A. Optimization of well-ordered, X-ray diffracting crystals required construction of point mutants substituting contiguous charged surface glutamate residues with alanine, resulting in improved crystal lattice contacts. Experimentally this resulted in both widening of the range of Cyclin B crystallization conditions, as well as improvements in crystal morphology, reproducibility and growth kinetics. In addition, extension of the observable diffraction limit from 4-5 A to 2.9 A was achieved through crystal dehydration by post-crystallization soaking in saturating concentrations of Sodium Bromide. Cyclin B is an all alpha-helical protein composed of tandem helical repeat domains, a hallmark of the cyclin fold. Although core elements of the 100-residue conserved N-terminal domain known as the cyclin box are nearly identical among the cell cycle cyclins, Cyclins A, B, and E, despite low sequence similarity, contextual residues lining interaction surfaces vary markedly. Superposition of Cyclin B onto the Cyclin A-Cdk2 structure shows a conserved interface, however, surface chemistry changes and clashes with the Cdk2 T-loop and 'PSTAIRE' helix indicate differences in how the cyclins activate Cdk. Examination of the putative substrate recognition or 'RxL' binding domain using the Cyclin A-Cdk2-p27 structure as a template reveals that although Cyclin B contacts the core P27 'RxL' motif, adjacent steric clashes and loss of contacts might disrupt binding, providing a route for Cyclin B mediated Cdk1 specificity built upon a common module. Based on the structure we postulate that cyclins are rigid scaffolds supporting multiple interaction surfaces fine tuned for unique functions. Furthermore, the availability of large quantities of active, soluble Cyclin B1 will enable detailed biochemical and biophysical analysis of its cell cycle function and may aid in the development of peptide based Cyclin B specific inhibitors for use in cancer therapeutics.