Conical intersection dynamics in solution: The chromophore of Green Fluorescent Protein

Abstract

We use ab initio results to reparameterize a multi-reference semiempirical method to reproduce the ground and excited state potential energy surfaces (PESs) for the chromophore of Green Fluorescent Protein (GFP). The validity of the new parameter set is tested, and the new method is combined with a quantum mechanical/molecular mechanical (QM/MM) treatment so that it can be applied in the solution phase. Solvent effects on the energetics of the relevant conical intersections are explored. We then combine this representation of the ground and excited state PESs with the full multiple spawning (FMS) nonadiabatic wavepacket dynamics method to simulate the photodynamics of the neutral GFP chromophore in both gas and solution phases. In these calculations, the PESs and their nonadiabatic couplings are evaluated simultaneously with the nuclear dynamics, ie. "on-the-fly". The effect of solvation is seen to be quite dramatic, resulting in an order of magnitude decrease in the excited state lifetime. We observe a correlated torsion about a double bond and its adjacent single bond in both gas and solution phases. This is discussed in the context of previous proposals about minimal volume isomerization mechanisms in protein environments.

The electronic nonadiabatic coupling term: Can it be ignored in dynamic calculations?

“…In this respect, it may be interesting to mention another semiclassical approach, the multiple spawning method which not only follows the motion of the wave packet in the MEP region but also incorporates the effects of NACTs along the classical trajectory. 44,45 Consequently, this method seems to be more reliable than the other semiclassical approaches ͑although the details of this approach were not examined by us͒.…”

Optimizing conical intersections of solvated molecules: The combined spin-flip density functional theory/effective fragment potential method

“…[91][92][93][94][95][96][97][98][99][100][101][102][103][104][105] The protein environment plays a crucial role in the fluorescence of the chromophore (HBI). It has been argued that the surrounding residues prevent the chromophore from twisting around the phenoxy ("P-bond": τ P ) or imidazolinone ("I-bond": τ I ) bonds of the bridge (see Fig.…”

Direct dynamics simulations using Hessian-based predictor-corrector integration algorithms

“…20 To calculate a trajectory the potential energy and gradient are obtained directly from this semiempirical-SRP electronic structure theory. For some systems this approach has led to reliable potential energy surfaces, 22,[24][25][26] but for others the semiempirical model used was not sufficiently flexible to accurately represent experiment and high-level ab initio calculations. 21,23 Single reference and minimal basis set semiempirical models, such as AM1 and PM3, may not incorporate sufficient detail of the quantum mechanics to accurately represent the PES.…”

“…21 A semiempirical-SRP model which appears to be substantially more accurate and broadly applicable is one based on a floating-occupation molecular orbital configuration interaction wave function. [25][26][27] A number of different algorithms have been presented for developing analytic PESs by interpolating ab initio data points. [28][29][30][31][32][33][34][35][36][37] Ischtwan and Collins have proposed 28 and used extensively 29 a Shepherd interpolation scheme.…”

“…[20][21][22][23][24][25][26] These new parameters are called specific reaction parameters ͑SRPs͒. 20 To calculate a trajectory the potential energy and gradient are obtained directly from this semiempirical-SRP electronic structure theory.…”