The Shoelson Lab at the Joslin Diabetes Center has been instrumental in showing that a subacute inflammatory process links obesity to the development of insulin resistance and type 2 diabetes. In terms of the biochemical events that mediate the inflammatory response, two intracellular pathways are activated in parallel, one mediated by the c-Jun N-terminal kinase (JNK) and the other by IKK beta and NF-kappaB. Studies of NF-kappaB activation in insulin resistance have centered mostly on the transcriptional regulation of proinflammatory cytokines and chemokines, such as TNF-alpha, MCP-1, IL-6, IL-1 beta and resistin. In contrast, studies with JNK have focused on the direct phosphorylation of insulin receptor substrate 1 (IRS-1). Insulin receptor (IR) is a tyrosine kinase that is activated by insulin binding. Activated IR binds to the IRS-1 phosphotyrosine-binding (PTB) domain and sequentially phosphorylates IRS-1 at numerous tyrosine sites dispersed throughout its primary sequence. The phosphorylated tyrosine residues engage and activate SH2 domain containing proteins that propagate the signal into the cell. JNK phosphorylates human IRS-1 at two distinct sites, Ser307 and Ser312, and when these sites are phosphorylated; IRS-1 neither binds IR nor activates cellular signaling. Phosphorylation at Ser312 has been studied in greatest detail and shown to correlate with the development of insulin resistance in rodent models and diabetic, insulin-resistant patients. While these findings link inflammation to the inhibition of insulin signaling, they do not provide a molecular mechanism. Given what is known about IR and IRS-1 structure, and the fact that Ser307 and Ser312 are located approximately 50 residues C-terminal to the PTB domain (IRS-1 residue 156-260) in an otherwise disordered region of the protein, it is not at all obvious how phosphorylation at these sites causes insulin resistance. I have hypothesized that JNK phosphorylation of IRS-1 inhibits PTB domain binding to IR as follows. Since Ser307 and Ser312 are in a disordered and apparently flexible region of the protein, these sites are free to loop around the PTB domain and interact with its phosphotyrosine binding site. Phosphoserine residues have never been shown to bind PTB domains, but I hypothesize that this occurs because they are sufficiently close that the intermolecular interaction is energetically (entropically) favored. Engagement of its PTB domain by pSer307 or pSer312 would auto-inhibit IRS-1 by preventing it from docking as usual to pTyr960 of the insulin receptor. We will test this hypothesis by generating phosphoserine extended IRS-1 PTB domain by expressed protein ligation (EPL) and by measuring binding affinities with isothermal titration calorimetry. We will also solve the crystal structure of the self-inhibited form of the IRS-1 PTB domain. ? ? ?
| Park, Sang-Youn; Jin, Wanzhu; Woo, Ju Rang et al. (2011) Crystal structures of human TBC1D1 and TBC1D4 (AS160) RabGTPase-activating protein (RabGAP) domains reveal critical elements for GLUT4 translocation. J Biol Chem 286:18130-8 |
