CCES Unicamp

Computational study of tryptophan fluorescence in proteins by molecular dynamics simulations

Date: May 31, 2019, 14:00

Candidate: Alvaro J. Lopez

Advisor: Leandro Martínez

Instituto de Química – Unicamp

Abstract:

Tryptophan fluorescence is highly sensitive to the molecular environment, so it is widely used to investigate the structure and dynamics of proteins. The fluorescence emission wavelength (λ_em ) of tryptophan in proteins is very variable. There are proteins with low λ_em (e.g. Azurine, 308 nm), and others with high λ_em (e.g. Glucagon, 355 nm). The diversity of λ_em motivated many theoretical studies which could explain the Stokes shift phenomena of tryptophan in proteins. In this thesis we developed parametric models to compute the λ_em of tryptophan using molecular dynamics simulations based on two main approaches : i) the solvent accessible surface area of indole and ii) electrostatic interaction between indole and the rest of the system. The performance of the best of our models has a R^2 = 0.81 and a standard deviation of 4.89 nm. These parameterized models represent a simple and fast methodology that can be used in the interpretation of experimental data. A second objective of this thesis was to explain, with the help of our parametric models, the inconsistency found between the reorientational dynamics calculated by molecular dynamics and deduced from time-resolved fluorescence anisotropy experiments of the Trp113 of the enzyme subtilisin Carlsberg. The molecular dynamics sampled preferably conformations with Trp113 partially protected from the solvent (corresponding to the crystallographic structure), while the experiment appears to be consistent with conformations with Trp113 exposed to the Solvent. The calculated reorientational dynamics for the solvent-exposed Trp113 was much faster when compared to the observed for the crystallographic structure. In this way, it was possible to structurally interpret the experimental results, and suggest that the crystallographic conformation is not representative of the set of structures sampled in solution.

Related posts

Improving Compiler-Generated Transactional Code Performance and Programmability via Language-Level Constructs

cces cces

Development of analytical approach for tunneling calculations in double well potentials

cces cces

Stress-based structural topology optimization for design-dependent loads problems using the BESO method.

cces cces
WP Twitter Auto Publish Powered By : XYZScripts.com