Revealing Hidden Conformational Space of LOV Protein VIVID Through Rigid Residue Scan Simulations

Abstract

VIVID(VVD) protein is a Light-Oxygen-Voltage(LOV) domain in circadian clock system. Upon blue light activation, a covalent bond is formed between VVD residue Cys108 and its cofactor flavin adenine dinucleotide(FAD), and prompts VVD switching from Dark state to Light state with significant conformational deviation. However, the mechanism of this local environment initiated global protein conformational change remains elusive. We employed a recently developed computational approach, rigid residue scan(RRS), to systematically probe the impact of the internal degrees of freedom in each amino acid residue of VVD on its overall dynamics by applying rigid body constraint on each residue in molecular dynamics simulations. Key residues were identified with distinctive impacts on Dark and Light states, respectively. All the simulations display wide range of distribution on a two-dimensional(2D) plot upon structural root-mean-square deviations(RMSD) from either Dark or Light state. Clustering analysis of the 2D RMSD distribution leads to 15 representative structures with drastically different conformation of N-terminus, which is also a key difference between Dark and Light states of VVD. Further principle component analyses(PCA) of RRS simulations agree with the observation of distinctive impact from individual residues on Dark and Light states.

Allosteric mechanism of the circadian protein Vivid resolved through Markov state model and machine learning analysis

“…The sampled conformational space in the 12μs MD simulations has a remarkable agreement with our previous study of VVD through perturbed MD simulations. [36] Our analyses show that the MD simulations of transient light and dark configurations sampled larger conformational space than the native light and dark configurations. These results conform that the covalent bond could facilitate the conversion of dark state to the light state, agreeing with the experimental observations.…”

Recognition of protein allosteric states and residues: Machine learning approaches

“…The theory for allosteric information transduction within the proteins has evolved from single pathway formed by residues into allosteric information transduction network model . Numerous methods for identifying key allosteric residues from simulations have been developed recently . These computational methods focus on correlation analysis related to protein dynamics.…”

Steric Interactions at Gln154 in ZEITLUPE Induce Reorganization of the LOV Domain Dimer Interface

“…We propose that in ZTL, steric interactions due to C4a hybridization contribute to ordering Gln154 and thereby stabilizing the buried conformation. In VVD, light-activation also leads to increased stability of a C-terminal salt-bridge 38 . Further, residues capable of forming the Arg125 -Glu158 salt bridge are retained in existing LOV structures, suggesting the signaling motif may be conserved across the LOV superfamily.…”