Cancer is one of the leading cause of mortality in the developed countries accounting for 21% of all mortalities. The use of standard chemotherapeutic agents in clinic during the last decades have highlighted the high cytotoxicity of such compounds. The research for non-toxic drugs, together with the need of cancer-specific therapies, has led to the development of molecular targeted therapies.
The present project is based on the hypothesis that an effective chemotherapeutic strategy would be that of interfering with the PI3K/Gab2 interaction.
The family of lipid kinases termed phosphoinositide 3-kinases (PI3Ks) are key regulators involved in many cellular processes including cell survival, proliferation and differentiation.
PI3Ks can transduce signals from various growth factors and cytokines into intracellular messages by generating the phospholipid Phosphatidylinositol (3,4,5)-trisphosphate (PIP3). The lipid PIP3 can then activate different downstream effector pathways.
The activation of PI3K require the interaction with some receptors that generally are tyrosine kinase receptors (RTKs). This interaction can occur directly or indirectly through the binding with scaffolding proteins such as Gab2.
Gab2, a key member of the Gab family of proteins, is involved in the amplification and integration of signal transduction, evoked by a variety of extracellular stimuli, including growth factors, cytokines and antigen receptors. This protein, moreover, is overexpressed in breast, gastric, and lung cancers.
Upon activation of some receptor tyrosine kinases, the scaffold Gab2 becomes tyrosyl phosphorylated and interacts with SH2 domain containing signal relay molecules, including the p85 subunit of PI3K.
In this project we plan to address the mechanism of interaction between the disordered region of Gab2 and the SH2 domains of PI3K, with the final goal of designing specific inhibitors to hijack such interactions.
To define an interaction between two proteins it is fundamental to use the appropriate techniques. We wish to gain a complete understanding of the mechanism and the structural characteristics of PI3K/Gab2 binding process.
To achieve this goal we propose a combination of kinetic binding and NMR experiments.
The kinetic binding experiments are performed using a stopped-flow apparatus. With this instrument we can define the mechanism of the interaction, whether the system populates intermediate states or not during the process.
At the same time these experiments allow us to determine the residues fundamental for the interaction and any eventual allosteric mechanisms that govern the process.
Proteins that are able to interact with different partners, usually control their activity through conformational changes generally induced by covalent modifications or by the binding with allosteric effectors. However, in the literature, some examples are reported that show how the cis/trans isomerization of a single peptidyl-prolyl imide bond, can represent a mechanism by which proteins regulate their activity (Mallis R. J. et al. 2002; Lummis S. C. R. et al. 2005; Sarkar P. et al. 2007; Severin A. et al. 2009). In the work of Mallis et al., for example, it has been observed that the cis/trans isomerization of a single peptidyl-prolyl imide bond of the SH2 domain of the Itk protein, drastically affects the function of this domain. More precisely, they have demonstrated how this SH2 domain interacts with different partners according to whether the peptidyl-prolyl imide bond is in a cis or in a trans conformation.
From preliminary studies conducted analyzing the structures of the two SH2 domains of PI3K, we observed the presence of a cis peptidyl-prolyl imide bond conserved in both domains. For this reason, one of our goal is to understand if the eventual isomerization of such bonds can represent a mechanism of control with which PI3K regulates the binding with Gab2. Such evidence could be very important both for the design of a drug that prevents this interaction, and in further defining a mechanism of control of the protein function still poorly characterized.
A complete understanding of the recognition between Gab2 and PI3K can only occur with the use of NMR techniques. With these experiments is possible to define the structural basis of the process and to elucidate the dynamic and the allosteric nature of the reaction.
As mentioned above, the SH2 domains of PI3K specifically recognize the sequence pY-x-x-M of the binding partner. The specificity for the methionine is still poorly understood. For this reason one of the aim of this project is to define the residues of the SH2 domains that are energetically coupled with the methionine. In other words we want to define which are the residues of the SH2 domains that define the specificity for the methionine and so for the binding partners of PI3K. After this we can design molecules able to interfere with such residues and so hijack the interaction.
The experimental plan is therefore to:
i) determine by NMR the structural features of the disordered regions of Gab2 implicated in the recognition of PI3K
ii) determine by NMR the structure of the complexes between the two SH2 domains of PI3K and the interacting regions of Gab2
iii) address the binding mechanism of isolated SH2 domains from PI3K and Gab2
iv) infer, by site directed mutagenesis, the interactions critical for the molecular recognition between Gab2 and its PI3K
v) exploit structural and mechanistic knowledge to derive an in silico pharmacophore model and screen the ZINC database of purchasable drug-like and lead-like molecules, in order to find potential inhibitors of PI3K/Gab2 interaction.
vi) test in vitro, by fluorescence monitored equilibrium and kinetics, the inhibitors selected by virtual screening