Structural basis for recruitment of the adapter proteins APS and SH2-B to the insulin receptor

Insulin is the principal hormone that regulates blood glucose levels. Insulin action is initiated with the activation of the insulin receptor through autophosphorylation. The resulting phosphotyrosines serve as recruitment sites for downstream proteins. APS and SH2-B, members of an adapter protein family, have been identified as substrates of the insulin receptor. These proteins are composed of an N-terminal dimerization region, a central pleckstrin homology domain and a C-terminal Src homology-2 (SH2) domain. The interaction with the insulin receptor is mediated by the SH2 domain of these adapters. To understand the molecular basis for recruitment of APS to the activated insulin receptor, we determined the crystal structure of the APS SH2 domain in complex with the phosphorylated insulin receptor kinase. The structure reveals a novel dimeric configuration of the APS SH2 domain in which the C-terminal half of the domain diverges from canonical SH2 domain architecture. In addition, each APS SH2 monomer engages two phosphotyrosines in the activation loop of the kinase. Once recruited by the insulin receptor, APS is phosphorylated by the insulin receptor on a C-terminal tyrosine residue. We characterized this phosphorylation site as a novel Cbl recognition motif, in which residues N-terminal to the phosphotyrosine interact with the four-helix bundle of the Cbl TKB (tyrosine kinase binding) domain. The recruitment of Cbl by APS may function to potentiate insulin-stimulated glucose uptake in adipose cells.; We also determined the crystal structure of the SH2-B SH2 domain in complex with the Jak2 pTyr-813 phosphopeptide. The structure reveals a canonical SH2 domain-phosphotyrosine interaction with recognition of a glutamate at the +1 position and a leucine at the +3 position relative to phosphotyrosine. Biochemical studies demonstrate that, although the SH2 domains of SH2-13 and APS share ∼80% sequence identity, their binding specificities differ and are dependent on their oligomeric state.