Abstract
A necessity for cell survival is finely tuned signaling inputs and the propagation of those signals. Protein kinases that phosphorylate signaling proteins are known as the “writers” of these signals, whereas phosphatases are known as the “erasers.” The interplay between these two proteins is crucial for signal propagation. When these signals are not tightly regulated, they often lead to a wide variety of cancers. A commonly studied protein kinase family is the Src family kinases that are crucial to cell regulation and implicated in many types of cancer. Using a suite of biochemical and biophysical techniques, we show activation loop phosphorylation (activating) wins the tug of war over C-terminal tail phosphorylation (inactivating) by modulating the open and closed equilibrium. We uncover a novel mechanism of activation of Src kinase dependent on phosphorylation of the activation loop in the closed conformation. We show dephosphorylation of the inhibitory tyrosine by the protein tyrosine phosphatase SHP2 can only occur in the open state, and activation loop phosphorylation can drive dephosphorylation of the C-terminal tail through a
global opening of the kinase. Intriguingly, our 13C-methyl NMR data revealed why Src evolved an inhibitory C-terminal phosphorylation site. We found nucleotide binding stabilizes the open a-C helix-“In” active state, and therefore Src would be constitutively active under in vivo concentrations of 4-5 mM ATP. Lastly in Chapter 2, we use our experimental setup to show how
vi small-molecule inhibitors can hijack these built-in dynamics and promote different conformational states of Src. We describe a second experimental setup to dissect the kinetics of opening and closing of all Src phosphorylated states. This Fluorescence Resonance Energy Transfer (FRET) experimental setup would have the advantage of looking at protein dynamics on the single-molecule level. A FRET dye pair was designed to report on distance changes when the protein undergoes exchange between conformational states. Sortase ligation reactions were optimized to ligate separately fluorophore-labeled regulatory and catalytic domains of Src. Sortase ligation was found not to alter Src catalytic activity. However, biotinylation and fluorophore labeling did alter the catalytic activity of Src, measured by substrate peptide phosphorylation levels. We observed very minimal EFRET transitions in our data, indicating we were not observing conformational transitions in the protein. We reasoned the altered catalytic activity of our Src constructs was from modifying the protein with labels and they were disrupting the conformational dynamics of Src. Further, minimal to-no changes in EFRET populations were observed when looking at the various phosphorylation states of Src. This was a concerning result that contradicted the NMR experimental evidence in Chapter 2. To determine if the Src FRET construct could conformationally exchange, data was recorded of Src bound to conformationally selective inhibitors Src kinase inhibitor 1 (SKI) and imatinib (Gleevec). We again found very few differences in the raw and processed data that led us to conclude our FRET construct design was not optimal. Taken together, we provide an in-depth
study on Src activation and inactivation through its regulatory tyrosine sites and its open and closed conformational equilibrium.