Functional selectivity of insulin receptor revealed by aptamer-trapped receptor structures

Abstract

Insulin receptor (IR) is a key regulator of energy metabolism. It is the first signal transducer that transports glucose present in the blood into the cell. IR in response to extracellular signals activates various cell secondary messengers that cause various reactions. Representatively, glucose metabolism, protein and lipid synthesis, etc. Therefore, IR regulates blood glucose level and plays a key role in supplying an intracellular energy source. IR malfunction causes severe diseases such as diabetes and cancer. Despite its importance, the mechanism of the insulin receptor (IR) activation has not been elucidated yet. When insulin is bound to IR, insulin induces conformational change of IR, and phosphorylates tyrosine residues in the intracellular domains. Phosphorylated tyrosine residues recruit secondary messengers that participate in two different signaling pathways, metabolic and mitogenic pathways. Metabolic pathway is related with glucose uptake from blood vessel to cell. Mitogenic pathway mediate cell growth and proliferation. Because of the mitogenic pathway, it causes side effects in insulin therapy in diabetic patients. While insulin binding activates both major signaling pathways, the discovery of agonists with functional selectivity has revealed that IR activation by some ligands leads to selective downstream signaling and cellular functions of IR in a manner distinct from that of insulin. These ligands include antibodies, peptides, and nucleotides that selectively activate the AKT pathway and induce metabolic effects, but exert much weaker MAPK pathway signaling and mitogenic effects than insulin. It suggests that the IR has a conformation with functional selectivity like some GPCRs, and that the selective agonists capture the conformation. Current models for IR activation cannot explain how the agonists selectively regulate receptor autophosphorylation and intracellular signaling in a manner distinct from insulin. The mechanism of action of selective agonists remains unclear due to a lack of agonist bound structures. IR has negative cooperativity. In a high-concentration insulin environment, IR has low affinity to insulin binding. Therefore, the structure of the insulin-saturated IR is not the fully activated state. However, characterizing the fully activated state of IR is difficult because IR is rapidly saturated when insulin is treated in vitro. A DNA aptamer with positive allosteric modulator activity has been reported. It works with insulin to induce full activity of IR. Here, I captured IR in three different conformations by DNA aptamers and determined structure in the range of 3.62 to 4.27 Å resolution using cryo-electron microscopy. IR-A62 (A62) is an agonistic DNA aptamer which phosphorylates only one specific tyrosine residue among seven phosphorylation sites of IR. Monophosphorylated IR selectively activates only the metabolic pathway, not the mitogenic pathway. In the IR, A62 and insulin complex, A62 captures the intermediate state between mono-phosphorylated IR to fully active state. IR-A43 (A43) has a synergic property that acts with insulin and increases phosphorylation more than 25-fold than treatment with insulin alone. In a complex of IR, A43 and insulin, A43 captures the fully activated conformation of IR. Using two DNA aptamers (A62 and A43), I determined three different structures of the IR in the absence or presence of insulin. In this dissertation, I will describe the structures of IR and A62 complex structure at first. In this complex, two A62 molecules are symmetrically bound to an IR dimer at the same position, resulting in an arrow-like shaped IR structure. Each A62 aptamer interacts with L1 and FnIII-1′ simultaneously which are known to be the insulin binding sites 1 and 2, respectively. The two membrane-proximal regions of FnIII-3 are brought in the 34~41 Å distance using hinge motion without changing the internal protomer conformation. Through this structure, I suggest that the arrow-like conformation, in which two membrane proximal regions are in the 34~41 Å distance resulting in the mono-phosphorylation of IR and selectively activates the metabolic pathway. Secondly, I present the structure of the IR in complex with an insulin and one A62 aptamer. The A62 aptamer is located in the same site within the arrow-like conformation, but one of A62 was replaced by insulin. In this structure, one side of IR head domains were lifted by insulin, but A62 captures the other side, which makes tilted-T shape. The distance between the two membrane proximal regions of tilted T shape is around 37~45 Å, and tilted T exhibited also mono-phosphorylated state. In third, I present the structure of the IR in complex with an insulin and one A43 aptamer. Insulin lifts up one side head domains of IR like tilted T, but the other side of head is down-shifted and graps both stalks simultaneously, resulting in the Г –shape conformation. The interactions between downside head and stalks brings the two membrane proximal regions closer together. By binding to the downside head part, A43 stabilizes the fully activated IR conformation. These structures provide important insights into the activation mechanism of IR and provide good frameworks for development of antidiabetic drugs without side effects.