Altered regulation of PTEN by mutagenesis and p85 binding

Growth and proliferation are normal functions of cells mediated in part via receptor tyrosine kinases such as the epidermal growth factor (EGF) receptor (EGFR). The EGFR binds the extracellular signaling ligand EGF and transduces the signal into the cell. Phosphatidylinositol 3'-kinase (PI3K) responds to EGFR activation and initiates downstream signaling cascades responsible for cell cycle entry, proliferation and inhibition of apoptosis. Cell cycle arrest is required to stop cell growth and proliferation as well as allow apoptosis, if required. The phosphatase and tensin homologue deleted on chromosome ten (PTEN) directly opposes PI3K signaling since its substrate is the PI3K product phosphatidylinositol 3,4,5-trisphosphate. PI3K is a heterodimer composed of a p85 regulatory subunit and a p110 catalytic subunit. PTEN is an essential tumor suppressor protein. Absence of PTEN has been associated with several types of cancer. Our laboratory has characterized new specialized functions for the p85 protein. One function discovered was the ability of p85 to enhance PTEN lipid phosphatase activity. In this thesis PTEN activity is shown to be enhanced at least 3.5 fold in vitro by an equimolar amount of p85. We performed an analysis of PTEN using seven PTEN mutants. Two types of mutants were created: i) regulatory or possible regulatory phosphorylation sites were substituted to mimic both phosphorylated and non-phosphorylated states and ii) alanine substitution of basic amino acid residues. The phosphorylation sites altered were the casein kinase 2 phosphorylation sites in the regulatory domain and tyrosine 336, a proposed regulatory phosphorylation site. Three mutants involving alanine substitution for basic amino acid residues included one mutant in the PASE domain and two more mutants in the C2 domain. It was observed that GFP-PTEN translocates to the plasma membrane upon EGF stimulation. The mimic of constitutive phosphorylation of the Casein kinase 2 sites resulted in cytoplasmic localization whereas the non-phosphorylated mimic was plasma membrane localized regardless of EGFR activation status. Neutralization of positive charge in the PASE and C2 domains seriously impeded the ability of PTEN to bind to phosphorylated phosphatidylinositol lipids and abolished the ability of the protein to translocate to the plasma membrane in response to receptor activation. Located within a cluster of positively charged lysine residues in the C2 domain is a potential phosphorylation site at tyrosine 336. The phosphorylation mimic showed decreased binding to some membrane lipids compared to the non-phosphorylated mimic. The results we generated are consistent with a current model for PTEN regulation that proposes PTEN is localized to the cytoplasm in quiescent cells and dephosphorylation of the regulatory domain occurs upon EGF stimulation allowing translocation to the plasma membrane. The model proposes that dephosphorylation of the casein kinase 2 sites unmasks regions of positive charge that interact with the anionic plasma membrane. Furthermore, the results suggested that at the plasma membrane p85 interacts with PTEN to increase lipid phosphatase activity and may be involved in targeting PTEN to the activated receptor where PI3,4,5P3 lipids are being produced