Skip to main content

Structural and mechanistic study of ion transport in Channelrhodopsin-2

Periodic Reporting for period 1 - LIIT-ChR2 (Structural and mechanistic study of ion transport in Channelrhodopsin-2)

Reporting period: 2015-04-01 to 2017-03-31

Channelrhodopsin-2 (ChR2) is a light sensitive ion channel found in green algae. ChR2 is now widely used in optogenetics. Neurons expressing ChR2 can be depolarized rapidly and reversibly by illumination, hence allowing control of the activation/inactivation of neurons in specific locations of the brain. Despite its importance, very little is known about the ChR2 structure, light cycle and mechanism of action. In this project, we studied the mechanism of light activation by elucidating the structure and properties of each of the intermediates of the photocycle by means of theoretical methods.

The project objectives were designed to understand the photo-chemical cycle of ChR2 and to target the current technical limitations of the protein. Understanding the structure and biophysical properties of the ChR2 will help in the engineering of new mutants or chemically modified variants with desirable properties. The aim is to expand the range of applicability of ChR2 in optogenetics and in medical devices, as well as to contribute to the basic knowledge of ion transfer in membrane channels.

The complete photocycle is defined by at least five photochemically different species. Despite some structural and chemical characteristics of each of the states in the photocycle are known, there is no atomistic structure of any of them. Using classical and quantum mechanics molecular dynamics simulations, we intended to model the intermediates and the transition mechanisms between them, validate the structures using the experimental data and aid our experimental collaborators in designing new variants of ChR2.

We have successfully modelled an atomistic structure of three photo-intermediates, as well as the transition mechanism between them. We have generated a new set of force-field parameters to simulate the retinal chromophore in ChR2 based on hybrid quantum-mechanics/molecular-mechanics (QM/MM) calculations. We have characterized the intermediates by calculating their spectroscopic properties, chemical shift, pore hydration and conductance. We have validated our structures by comparison with experimental observables such as FTIR spectroscopy, UV-Vis or EM. We have also aided our experimental collaborators in the structural characterization of non-natural chromophores in the active site of ChR2.

Given the lack of structural experimental data, our theoretical approach required a careful cycle of modelling and validation of each of the photo-intermediates as well as the engineered ChR2. Starting from the crystal structure of a chimeric channelrhodopsin, we modeled the closed (inactive) state of the channel. We validated the structure by calculating the spectroscopic properties and comparing them to the experimental observations. We found that the active has some flexibility that was not observed in the crystal structure (because it gives a static picture only of the protein) but that can explain the features of the absorption spectra. From this structure and using enhanced sampling methods, we modeled and validated two other photo-intermediates. We found the formation of a water pore and the mechanism of proton transfer. Our simulations also provided an atomistic explanation for FTIR experiments.

The classical force-field we developed reproduces accurately the same structure of the chromophore and the same protein-ligand interactions as the QM/MM trajectory, and is a remarkable improvement in reproducing the experimental data compared to the standard set of parameters. We have shown that we can further exploit these parameters to simulate all photo-intermediates and even different ligands with little modifications. In a multidisciplinary collaboration with the group of Prof. Bamberg, we aid in the design of a modified chromophore that shows interesting spectroscopic properties. We are currently using the methodological setup, the conformational ensembles and the force-field parameters to design new mutants with tuned biophysical properties.
The state-of-the-art force-fields can describe very accurately many biophysical properties of proteins, lipids, sugars and nucleic acids. However, new organic molecules are often poorly described, especially when they are embedded in a protein environment that is very different to a solvent or “bulk” environment. This is the case of the retinal Schiff base in the active site of ChR2: Validation of a classical model based on a standard force-field (gaff) parameters by calculating the spectroscopic properties resulted in a poor agreement with experiments. To solve this problem, we developed an ad hoc set of force-fields parameters for the specific case of ChR2 that showed great improvement in reproducing the experimental data. All scientific community can benefit from these parameters to perform their studies on ChR2 for bioengineering or biomedical applications.

Modelling ChR2 is a challenge: There is little structural data it’s a membrane protein with a photo-active adduct that undergoes several local and global conformational changes. In this project, we applied equilibrium and biased molecular dynamics simulations and enhanced sampling methods combined with classical, quantum mechanics and QM/MM levels of theory. We have provided atomistic structures and reaction mechanisms that can explain experimental observations.
Structure of the transmembrane Channelrhodopsin-2 in its resting state.