SHP2 Signaling from Molecule to Mouse

About This Grant

SHP2 (PTPN11) has attracted much attention, because unlike most protein-tyrosine phosphatases (PTPs), it functions as a “positive” component of receptor tyrosine kinase (RTK) signaling. Germline mutations in PTPN11 are the most frequent cause of the “RASopathies” Noonan Syndrome (NS) and NS with multiple lentigines, somatic mutations contribute to benign and malignant neoplasms, and global SHP2 deficiency in mice leads to trophoblast stem cell (TSC) death/peri-implantation lethality. SHP2 has two N-terminal SH2 domains, a central catalytic (PTP) domain, and two C-terminal tyrosine phosphorylation (pY) sites and acts as an elegant molecular switch, residing in a closed, inactive form in the absence of pY-peptide ligands for its N-SH2 domain. Binding by such ligands, found in receptors and scaffolding adapters (e.g., GAB1, 2), activates SHP2. This mechanism was exploited to develop allosteric inhibitors (SHP2i) now in clinical trials. We know that SHP2 acts upstream of the RAS guanine nucleotide exchange factors SOS1/2 to promote RAS/ERK pathway activation, yet despite much progress, key knowledge gaps persist. How SHP2 signals to SOS1/2 is unclear, as are the precise roles of PTP activity and tyrosine phosphorylation in vitro and in mice. Our new Preliminary Data show that SHP2 is required for the assembly of large SOS1 clusters in response to RTK stimulation. Moreover, the PTP domain and Cterminal pY sites play distinct roles downstream of different RTKs, and neither all PTP-inactive nor tyrosine phosphorylation site mutants phenocopy the effects of global SHP2 deficiency. Most surprisingly, we recently found that PTP activity is dispensable for RAS activation. Instead, a specific conformation of the PTP domain, involving movement of a b-sheet is required. This conformation is mimicked by certain catalytic cysteine (Cyscat) mutants and induced by in vitro oxidation of Cyscat. We propose to (1) use super-resolution microscopy to determine how RTKs and SHP2 regulate SOS1 and GRB2/SOS1 clusters, correlate clusters with RAS activation, and define the involvement and requirement of other early components in RTK signaling, (2) determine whether SHP2 pYs all signal via GRB2, use redox probes/proteomics and SHP2 monobodies (Mb) to define the RTKinduced PTP domain oxidation state, test our “conformational switch” model for RAS activation via a structureinformed genetic approach, use SHP2C459D, which is full capable of RAS activation, and the signaling inactive mutant SHP2C459S in affinity purification/proximity MS proteomics to identify SHP2-interacting proteins critical for RAS activation/inactivation and validate them using reverse genetics, and test the effects of SHP2 sub-domain mutants/Mb on SOS1 and GRB2/SOS1 clusters, and (3) use Ptpn11 mutant knock-in mice to test the effects of SHP2 sub-domains on FGF-induced signaling, transcription, proliferation, differentiation, and survival of TSCs, test whether SHP2C459D mediates “adaptive resistance” to RAS/ERK pathway inhibitors, and ask if it causes “RASopathy” and/or myeloproliferative neoplasm/leukemia. Mice are essential to model these complex defects seen in humans with germline and somatic PTPN11 mutations and to develop new treatments.