AMBER-DYES: Characterization of Charge Fluctuations and Force Field Parameterization of Fluorescent Dyes for Molecular Dynamics Simulations

Abstract

Recent advances in single molecule fluorescence experiments and theory allow a direct comparison and improved interpretation of experiment and simulation. To this end, force fields for a larger number of dyes are required which are compatible with and can be integrated into existing biomolecular force fields. Here, we developed, characterized, and implemented AMBER-DYES, a modular fluorescent label force field, for a set of 22 fluorescent dyes and their linkers from the Alexa, Atto, and Cy families, which are in common use for single molecule spectroscopy experiments. The force field is compatible with the AMBER protein force fields and the GROMACS molecular dynamics simulation program. The high electronic polarizability of the delocalized π-electron orbitals, as found in many fluorescent dyes, poses a particular challenge to point charge based force fields such as AMBER. To quantify the charge fluctuations due to the electronic polarizability, we simulated the 22 dyes in explicit solvent and sampled the charge fluctuations using QM/MM simulations at the B3LYP/6-31G*//TIP3P level of theory. The analysis of the simulations enabled us to derive ensemble fitted RESP charges from the solvated charge distributions of multiple trajectories. We observed broad, single peaked charge distributions for the conjugated ring atoms with well-defined mean values. The charge fitting procedure was validated against published charges of the dyelike amino acid tryptophan, which showed good agreement with existing tryptophan parameters from the AMBER, CHARMM, and OPLS force field families. A principal component analysis of the charge fluctuations revealed that a small number of collective coordinates suffices to describe most of the in-plane dye polarizability. The AMBER-DYES force field allows the rapid preparation of all atom molecular dynamics simulations of fluorescent systems for state of the art multi microsecond trajectories.

Probing the Action of Chemical Denaturant on an Intrinsically Disordered Protein by Simulation and Experiment

“…The second is to use the distances between the “C 1 ” atoms of each chromophore as described previously 64 without a correction factor. A more sophisticated approach, which assumes only that Förster theory is sufficiently accurate, can also be applied to the simulations including explicit donor and acceptor chromophores 64, 72–74 . In this case, the transfer rate k ET ( x ) for configuration x in the simulation trajectory is given by $$k_{\text{ET}}false(xfalse)=\frac{3}{2}{k}_{\mathrm{D}}{R}_0^6\frac{{\kappa }^{2}false(xfalse)}{R^{6}false(xfalse)}$$where k D is the donor fluorescence decay rate in the absence of an acceptor, and the orientational factor κ is given by $$\kappa ={\hat{\mu }}_{\mathrm{D}}·{\hat{\mu }}_{\mathrm{A}}-3false(\hat{R}·{\hat{\mu }}_{\mathrm{A}}false)false(\hat{R}·{\hat{\mu }}_{\mathrm{D}}false),$$where μ̂ D and μ̂ A are unit vectors in the direction of the donor and acceptor transition dipoles, respectively, and R̂ is a unit vector pointing between donor and acceptor.…”

Quantitative Interpretation of FRET Experiments via Molecular Simulation: Force Field and Validation

“…To date, a number of groups have proposed parameters for modeling chromophores in solution, in most cases with specific applications in mind (22,(25)(26)(27)(28). However, these parameters have, for the most part, not been quantitatively validated against experiment; part of the reason may be a lack of suitable experimental data, or that good agreement of dynamical properties is not expected due to the viscosity of commonly used water models such as TIP3P (29) being too low.…”

Genetic Algorithm Optimization of Point Charges in Force Field Development: Challenges and Insights

“…Although several alternative methods of charge derivation have been proposed, 47,53,60,62,63 restraining the charges of buried atoms to prevent the optimization from converging toward unreasonable values and/or to reduce conformational dependence of the charges became the most popular in force field development. [64][65][66][67][68][69][70][71][72][73][74][75][76][77] In most of these methods, besides a constraint on the total charge of the molecule, an additional restraining function is added to the LS sum (eq 1) to keep the buried atom charges close to some predefined values, despite its possible negative effect on the dipole moment values and the overall quality of MEP. 53,78 Considering the challenges presented by the relatively straightforward single-objective point charge fitting against the MEP, simultaneous optimization of point charges along with other force field parameters against a diverse training set could be expected to present even more pitfalls.…”