Multi-scale Modeling of a Canonical Signaling Transduction Domain: From Conformational Ensembles to Protein Function in Protein Kinase A


Biological Significances:  

The immediate biomedical impact pertains to developmental disorders, but implications are much broader, including neurodegenerative disease and aging.  The goal of our project is to understand how protein structure translates into the subcellular mechanism of action within macromolecular complexes. We are exploring these fundamental questions in the context of cyclic-adenosine mono-phosphate (cAMP) - dependent protein kinase A's (PKA) regulatory subunits (R) in cardiomyocytes. Our research has an implication on glycolysis regulation in cardiac muscles. 

Project Summary:

This project generates multi-scale computational models of PKA, linking atomic-level structures through free-energy landscapes and association rates to models of subcellular regulation. These models provide insight into the functional mechanism of the β-adrenergic pathway PKA in the cardiomyocyte and accelerate the development of key technologies and methods for integrating dispersed biological and chemical information into a coherent model that support prediction and study of small molecules and mutations in the function of the cell.

This project represents a new paradigm for multi-scale modeling in which association rates and free-energy landscapes derived from molecular-scale simulations are integrated into mathematical models representing subcellular-scale events and macromolecular complexes. Technology advances interconnect established modeling methods, Molecular Dynamics (MD) simulations, Brownian Dynamics (BD) simulations, and Virtual Cell modeling, using Markov State Models (MSM). MSMs are powerful, scalable mathematical tools for describing the dynamics and equilibrium states of complex interconnected systems of discrete states.

This project advances the application of MSM to protein function, expands the integration of BD and MD to determine association rates, and provides for novel integration of MD and BD data into Continuity. Specifically, it drives the development of workflows to automate job preparation and management of MD and BD; novel analysis methods for MD and BD leveraging of MSM; development of BrownDye, Continuity, and z-stack; and integration of outside utilities of MSMBuilder2 and Virtual Cell.


Key Outcomes:

  • Malmstrom RD, Kornev AP, Taylor SS, Amaro RE. Protein-Ligand Interactions through the Computational Microscope: Allostery in a Canonical Signaling Domain. Biophysical Journal. 2016;110(3):52a-a
  • Malmstrom RD, Kornev AP, Taylor SS, Amaro RE. Allostery through the computational microscope: cAMP activation of a canonical signalling domain. Nature Communications. 2015;6. doi: ARTN 7588 10.1038/ncomms8588.
  • Malmstrom RD, Lee CT, Van Wart A, Amaro RE. On the Application of Molecular-Dynamics Based Markov State Models to Functional Proteins. J Chem Theory Comput. 2014;10(7):2648-57. doi: 10.1021/ct5002363.