NIST only participates in the February and August reviews.
Exploring the molecular details of the ERK2 enzyme
The MAP kinases, ERK1 and ERK2, are central regulators of the RAS-RAF signaling pathway. Oncogenic mutations in enzymes upstream in this pathway act as drivers for many cancers, by activating ERK1/2. Inhibitors targeting two of these enzymes, BRAF and MEK, have achieved dramatic clinical success for metastatic melanoma, but most patients develop resistance. Because of this, ERKs are important therapeutic targets, and high-affinity inhibitors are in clinical trials. Therefore, understanding the molecular details of regulatory mechanisms for ERK1/2 is a timely and important goal.
Structural, biochemical, and biophysical experiments carried out by Natalie Ahn's lab and others have established key aspects of ERK1/2 function and regulation. However, the molecular mechanisms for the regulation of these kinases by phosphorylation are still incompletely understood. Key observations from hydrogen-deuterium exchange mass spectrometry (HDX-MS) and NMR experiments have revealed that phosphorylation of ERK2 leads to the regulation of dynamics, in a manner not obvious by X-ray crystallography [1,2]. In collaboration with the Ahn lab, we carried out molecular dynamics (MD) simulations of inactive and active states of ERK2, each extended out to 360 µs of total sampling [3,4]. The findings revealed differential dynamics between inactive vs active states. Unexpected conformational variants of active ERK2 were observed within the activation loop (A-loop). Our MD analyses demonstrated that the variable A-loop conformers altered distances between catalytic residues within the active site. Conversely, HDX-MS measurements by Ahn's lab showed that ERK inhibitor binding to the active site altered the conformation of the A-loop. These findings reveal novel properties of ERK2, unusual among protein kinases. Here, phosphorylation at the A-loop controls the dynamics at the active site, while inhibitor binding to the active site leads to biased selection of conformational states of the A-loop. Together these findings demonstrate allosteric coupling between the active site and A-loop, in a manner that will almost certainly impact the design of inhibitors towards ERK1/2.
We have only scratched the surface of this important problem, and our discoveries to date now set the stage for deeper examination. Needed are further MD simulations using enhanced sampling and metadynamics methods that have the potential to interrogate the conformational space and electrostatics driving forces controlling ERK2 dynamics and its allosteric regulation by inhibitory ligands. Multiscale modeling that combines quantum mechanical and molecular mechanical potentials will allow us to determine the atomistic details of phosphoryl transfer and catalytic turnover. Integrative modeling and machine learning have the promise of establishing new tools for combining computational and experimental data from HDX-MS and NMR to explain the dynamics and allosteric properties of this unique model system. In addressing these goals, our work will promote cross-fertilization between the NIST Thermodynamics Research group and the Ahn research lab at CU-Boulder. Our collaboration will develop new computational strategies and tools for measuring protein dynamics and allosteric regulation and the means for iteratively testing and validating them experimentally.
1. Pegram L, et al., and Ahn NG. Activation loop dynamics are controlled by conformation-selective inhibitors of ERK2, PNAS, 2019 (DOI: 10.1073/pnas.1906824116)
2. Anderson JW, et al., and Ahn NG. Conformation selection by ATP-competitive inhibitors and allosteric communication in ERK2, eLife, 2024 (DOI: 10.7554/eLife.91507.3)
3. Pegram L, Riccardi D, and Ahn NG. Activation loop plasticity and active site coupling in the MAP kinase, ERK2, JMB, 2023 (DOI: DOI: 10.1016/j.jmb.2023.168309)
4. Riccardi D, Pegram Laurel, Ahn N (2023), Activation loop plasticity and active site coupling in the MAP kinase, ERK2: Supplementary Data, Trajectories, and Scripts, National Institute of Standards and Technology, (DOI: 10.18434/mds2-2988)
Molecular dynamics; QM/MM; Electrostatics; Machine Learning; HDX-MS; Drug Design; Enzyme; Kinase; Allostery; Cancer