Molecular dynamics simulation of the heart type fatty acid binding protein in a crystal environment

Abstract

Crystallographic data comes from a space-time average over all the unit cells within the crystal, so dynamic phenomena do not contribute significantly to the diffraction data. Many efforts have been made to reconstitute the movement of the macromolecules and explore the microstates that the confined proteins can adopt in the crystalline network. We explored different strategies to simulate a heart fatty acid binding protein (H-FABP) crystal by means of Molecular Dynamics (MD) simulations. We evaluate the effect of introducing restraints according to experimental isotropic B-factors and we analyzed the H-FABP motions in the crystal using Principal Component Analysis (PCA), isotropic and anisotropic B-factors. We compared the behavior of the protein simulated in the crystal confinement versus in solution, and we observed the effect of that confinement in the mobility of the protein residues. Restraining one-third of Cα atoms based on experimental B-factors produce lower B-factors than simulations without restraints, showing that the position restraint of the atoms with the lowest experimental B-factor is a good strategy to maintain the geometry of the crystal with an obvious decrease in the degrees of motion of the protein. PCA shows that, as position restraint reduces the conformational space explored by the system, the motion of the crystal is better recovered, for an essential subspace of the same size, in the simulations without restraints. Restraining only one Cα seems to be a good balance between giving flexibility to the system and preserving its structure.

Communicated by Ramaswamy H. Sarma

KEYOWRDS:

• Molecular dynamics
• H-FABP-fatty acid complex
• protein crystal

Disclosure statement

The authors declare no conflicts of interest.

Additional information

Funding

Support of this work by Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) and Universidad Nacional de La Plata (UNLP) of Argentina is greatly appreciated. We thank the “Nodo de Cálculo de la Universidad Nacional de Tierra del Fuego, Antártida e Islas del Atlántico Sur” (COFECYT 2018-240-APN-MCT) for the facilities to perform part of the calculations here presented. E.I.H. and C.M.C. are members of CONICET - Argentina. H.A.A. is teaching researcher from UNLP and Y.R.E. was supported by the CONICET.