Phosphorylation of the insulin receptor kinase by phosphocreatine in combination with hydrogen peroxide: the structural basis of redox priming.

Signaling by insulin requires autophosphorylation of the insulin receptor kinase (IRK) at Tyr1158, Tyr1162, and Tyr1163. Earlier experiments with (32)P-gamma-ATP indicated that the nonphosphorylated IRK (IRK-0P) is relatively inactive, and crystallographic data indicated that the ATP binding site of IRK-0P is blocked by its activation loop. We now show that phosphocreatine (PCr) in combination with hydrogen peroxide serves as an alternative phosphate donor and that ATP and PCr use distinct binding sites. Whereas phosphorylation of the IRK by ATP is inhibited by the nonhydrolyzable competitor adenylyl-imidodiphosphate, phosphorylation by PCr is enhanced. The IRK mutant Tyr1158Phe showed no phosphorylation with PCr but almost normal phosphorylation with ATP, whereas Tyr1162Phe was phosphorylated well with PCr but less then normal with ATP. 3-Dimensional models of IRK-0P revealed that the conversion of any of the four cysteine residues 1056, 1138, 1234, and 1245 into sulfenic acid produces structural changes that bring Tyr1158 into close contact with Asp1083 and render the well-known catalytic site at Asp1132 and Tyr1162 accessible from a direction that differs from the known ATP binding site. The mutant Cys1138Ala, in contrast, showed relatively inaccessible catalytic sites and weak catalytic activity in functional experiments. Taken together, these findings indicate that 'redox priming' of the IRK facilitates its autophosphorylation by PCr in the activation loop.