Structural and kinetic characterization of DUSP5 with a Di-phosphorylated tripeptide substrate from the ERK activation loop.

Introduction: Dual specific phosphatases (DUSPs) are mitogen-activated protein kinase (MAPK) regulators, which also serve as drug targets for treating various vascular diseases. Previously, we have presented mechanistic characterizations of DUSP5 and its interaction with pERK, proposing a dual active site.

Methods: Herein, we characterize the interactions between the DUSP5 phosphatase domain and the pT-E-pY activation loop of ERK2, with specific active site assignments. We also report the full NMR chemical shift assignments of DUSP5 that now enable chemical shift perturbation and dynamics studies.

Results And Discussion: Both phosphates of the pT-E-pY tripeptide are dephosphorylated, based on P NMR; but, steady state kinetic studies of the tripeptide both as a substrate and as an inhibitor indicate a preference for binding and dephosphorylation of the phospho-tyrosine before the phospho-threonine. Catalytic efficiency (k/K) is 3.7 MS for T-E-pY vs 1.3 MS for pT-E-Y, although the diphosphorylated peptide (pT-E-pY) is a better substrate than both, with k/K = 18.2 MS. Steady state inhibition studies with the pNPP substrate yields K values for the peptide inhibitors of: 15.82 mM (pT-E-Y), 4.932 mM (T-E-pY), 1.672 mM (pT-E-pY). Steady state inhibition studies with pNPP substrate and with vanadate or phosphate inhibitors indicated competitive inhibition with Kis values of 0.0006122 mM (sodium vanadate) and 17.32 mM (sodium phosphate), similar to other Protein Tyrosine Phosphatases with an active site cysteine nucleophile that go through a five-coordinate high energy transition state or intermediate. Molecular dynamics (MD) studies confirm preferential binding of the diphosphorylated peptide, but with preference for binding the pY over the pT reside in the catalytic site proximal to the Cys263 nucleophile. Based on MD, the monophosphorylated peptide binds tighter if phosphorylated on the Tyr vs the Thr. And, if the starting pose of the docked diphosphorylated peptide has pT in the catalytic site, it will adjust to have the pY in the catalytic site, suggesting a dynamic shifting of the peptide orientation. 2D H-N HSQC chemical shift perturbation studies confirm that DUSP5 with tripeptide bound is in a dynamic state, with extensive exchange broadening observed-especially of catalytic site residues. The availability of NMR chemical shift assignments enables additional future studies of DUSP5 binding to the ERK2 diphosphorylated activation loop.

Summary: These studies indicate a preference for pY before pT binding, but with ability to bind and dephosphorylate both residues, and with a dynamic active site pocket that accommodates multiple tripeptide orientations.