(489a) Multisite Phosphorylation Underlies Ultrasensitive Response of CDK Inhibitor Sic1 in Yeast

One third of all proteins in eukaryotic cells are phosphorylated at any time and the majority of them are phosphorylated on more than one site. The pattern and/or the degree of phosphorylation affect the functionality or stability of proteins and thereby control the downstream process in signaling pathways. An interesting example is the Sic1 protein, a cyclin dependent kinase inhibitor that plays a crucial role in regulating the G1/S transition in the cell cycle of Saccharomyces cerevisiae.

The Sic1 protein contains nine phosphorylable residues for the Cln2-Cdc28 kinase complex. At the late G1 phase, when the activity of Cln2-Cdc28 increases, Sic1 is recruited by the Cdc4 protein for ubiquitination and subsequent proteolysis after becoming phosphorylated on at least six sites. The cell can enter into the S phase when Sic1 is degraded. Our previous theoretical studies under steady-state and transient conditions predict that the elimination of Sic1 in the G1/S transition of the yeast cell cycle occurs in a switch-like fashion due to multisite phosphorylation. However, no experimental data is available for this statement. In this work, we aim to integrate experiments with previous theoretical work to elucidate the ultrasensitivity property of Sic1.

We first quantify the steady-state stimulus -response curve of Sic in vitro. We measure the fraction of Sic1 that binds to Cdc4 (as system's response) at different ratios of kinase and phosphatase (as system's stimulus). Then we seek to measure two sets of key parameters in our theoretical model: the kinetics of the phosphorylation and dephosphorylation reactions; the binding between various phosphorylated states of Sic and the Cdc4 protein. For kinetic assays, we develop and apply mass spectrometry based methods to quantitate multisite phosphorylation. Due to the nondeterministic order of multi-step phosphorylation, there are a overwhelmingly large number of parameters involved. Besides making simplifications based on system characteristics, we are also exploring the approach of utilizing various Sic1 mutants to deconvolve the complex network. Finally, measured parameters are plugged into the model; the correctness and accuracy of the model can be evaluated by comparisons between experimental observations and model predictions on both wild type and mutants.

In addition to in vitro steady-state experiments, in vivo temporal profile of Sic1 during the cell cycle is being examined at single-cell resolution to determine to what degree the degradation of Sic1 is switch-like.

The fundamental knowledge gained from the Sic1 system can be used to understand the behavior of other multisite proteins such as p27, a functional homologue of Sic1 in mammalian cells, of which the misfunction has been correlated with a number of human tumors and cancers.