Channelrhodopsin reveals its dark secrets

See allHide authors and affiliations

Science  24 Nov 2017:
Vol. 358, Issue 6366, pp. 1000-1001
DOI: 10.1126/science.aar2299

A key element of information processing in the brain is the flow of ions across membranes through ion channels. For optogenetic applications, scientists replace these ion channels with the light-gated channels such as the cation channels channelrhodopsin 1 and 2 (ChR1 and ChR2) (1, 2) or more recently discovered anion channels (3). These channels allow remote light-controlled activation and deactivation of excitable cells with high spatiotemporal resolution; desired biological signals such as action potentials can thereby be switched on and off with light (4). On page 10.1126/science.eaan8862 of this issue, Volkov et al. (5) report the high-resolution structures of ChR2 wild type in its dark-adapted state and of a ChR mutant. The structures provide a detailed understanding of site-specific mutations used in optogenetics and will enable the rational design of optimized optogenetic tools.

ChRs belong to the large family of microbial rhodopsins. They span the membrane with seven-transmembrane α-helices and contain a deeply embedded retinal chromophore, which is covalently attached to a lysine residue via a protonated Schiff base. The first-discovered member of this family, the light-driven proton pump bacteriorhodopsin (6), inspired researchers worldwide to develop new biophysical tools that resolve structure and function, especially of membrane proteins, with high spatiotemporal resolution. A prominent example is cryo–electron microscopy, which Henderson et al. developed and used to determine the first high-resolution structural model of bacteriorhodopsin in 1990 (7). Around the same time, we used time-resolved Fourier transform infrared spectroscopy to resolve the detailed proton transfer steps through bacteriorhodopsin (8). The precise timing of the proton transfer steps in bacteriorhodopsin is reminiscent of a mechanical clock. The steps are controlled by the de- and reprotonation of the Schiff base after light-induced retinal isomerization (see the figure). Proton transfer via precisely arranged protein-bound water molecules is a crucial aspect of this mechanism (9).

Fundamental research on bacteriorhodopsins paved the way to the discovery of microbial ChRs, which rapidly led to biological applications. However, the lack of a ChR crystal structure has limited further advances. In 2012, Kato et al. provided a first insight into the ChR structure with the C1C2 chimera structural model, consisting of five helices of ChR1 and two helices of ChR2 (10). However, the light-induced responses of the C1C2 chimera are different from those of the ChR2 wild type. Volkov et al. now report the wild-type ChR2 structure at a resolution of 2.4 Å and compare the structural models for C1C2 and wild-type ChR in detail. They also discuss the potential molecular mechanism of the light-gated ion channel opening, taking into account the wealth of information provided by previous time-resolved biophysical studies. Their proposed mechanism resembles that of bacteriorhodopsin but also has important differences (see the figure).

Channel gating in microbial rhodopsins

Several gates and water cavities are known to play key roles in ion transfer through the light-driven proton pump bacteriorhodopsin. Volkov et al.'s structure of the light-gated ion channel channelrhodopsin 2 (ChR2) highlights similarities and differences between the two proteins.


The ChR2 wild-type structure has more and larger water-filled cavities than C1C2: two on the extracellular site (EC1 and EC2) and two on the intracellular site (IC1 and IC2). The authors propose three gates for the ion flux: a central, an extracellular, and an intracellular gate. The ion-conducting water-filled cavities are separated by the three gates and are arranged in a way similar to the cavities in bacteriorhodopsin (see the figure), pointing to a common precursor protein. But in contrast to consecutive gate switching, which is needed for pumping in bacteriorhodopsin, all gates must be opened simultaneously in ChR to allow ion flux through the protein. As in bacteriorhodopsin, ChR opening and closing seems to be controlled by the de- and reprotonation of the Schiff base.

The most striking difference between bacteriorhodopsin and wild-type ChR involves helix 2, which allows water influx by its opening movement. E90 (glutamic acid 90) on helix 2 is part of the central gate and determines the ion selectivity (11). E90 is differently oriented in wild-type ChR than in C1C2. The closing of the channel seems to be controlled by the reprotonation of the Schiff base. An additional, newly identified water molecule between C128 (cysteine 128) and D156 (aspartic acid 156), not observed in C1C2, may be involved in reprotonation of the Schiff base by the DC gate; this gate is not located in the ion pore but seems to play an important role in the lifetime of the open channel (5).

Volkov et al.'s wild-type structure of ChR2 allows for more precise biomolecular simulations of the ChR opening mechanism. Further insights would come from the structure of the light-adapted state of the protein. A detailed mechanistic understanding is a prerequisite for rationally designed, optimized tools for advanced optogenetic applications, far exceeding most applications in the brain today. Such optogenetic tools may not only allow for the restoration of vision, but may be applied to improve other physiological processes such as hearing (12, 13).


View Abstract

Stay Connected to Science

Navigate This Article