The Otop channels form proton-selective ion channels and are candidates for the sour taste receptor (1). To test how these channels function, structural information is crucial. Cryogenic electron microscopy (CryoEM) has had a recent renaissance (2), due to improved hardware and software and so is a good technique to characterise membrane proteins that typically are difficult to crystallise. In this preprint the structure determination of zebrafish Otop1 and chicken Otop3 by cryoEM was accomplished.This was complemented by functional data, simulation and mutagenesis to probe the molecular function of these channels.
I have summarised below some of the results that I found most interesting. According to these structures, Otop1 and Otop2 are both dimers made up of monomers of 12 transmembrane helices, which may have arisen due to gene duplication of a bundle of 6 helices. The authors suggest this because the six helical bundles can be aligned very closely with each other.
Figure 1. Otop1 structure coloured by conservation. A and B – side views C – top view
In each of the structures several lipids were modelled into the density. Lipids found in the interface between monomers appear to be important as they remain during simulations and, when removed, cause significant changes in conformation. A cholesterol molecule is modelled in the structure of Otop1 between the N- and C-terminal bundles, which is much more mobile during simulation, suggesting it is less tightly bound.
The structures were used in simulations to identify the position of the proton permeation pore. Several possible channel positions are discussed in the preprint. Firstly, the possible site between monomers is unlikely to allow proton movement, as it is not well conserved and cholesterol molecules are present, which would prevent the passage of water and protons. Secondly, there are possible permeation pathways in both the N- and C-terminal bundles, that appear to have two half channels. The third possible proton pathway between the N- and C-terminal bundles is probably the most likely, as, in simulations, water molecules can be seen to hop over the narrow constriction in the pore; however these results are not conclusive at this stage. Mutation of several sites was also used to test which site might be responsible for proton transport. Several positions were important for transport; however the authors point out the highly conserved QNY triad that might be of importance.
Fluorescence-detection Size Exclusion Chromatography (FSEC) is a very useful technique to rapidly examine constructs for stability and monodispersity. This was developed by the Gouaux lab (3) initially as a pre-crystallisation screen to identify membrane protein variants that were more likely to form well-diffracting crystals. Briefly, it involves solubilisation of a membrane protein fluorescent protein fusion (usually GFP) in one or more detergents followed by Size Exclusion Chromatography (SEC) of these unpurified samples: high quality proteins will appear as symmetric peaks at the predicted molecular weight (supplementary figures 3 and 4 panel a). The authors used this methodology to identify the orthologues of the Otop channels that expressed well and appeared monodisperse. FSEC is also used to very clearly show the difference in oligomeric state when certain amino acids at the interface are mutated (supplementary figure 7). The FSEC traces also show the importance of CHS addition, as the chromatograms are more symmetric when CHS is added in addition to DDM. CHS is a very common additive in membrane protein purification, as the cholesterol analogue can often help compensate for the loss of lipids upon extraction from the membrane. The CHS is likely to be important as the cryoEM densities of each channel could be fit with several cholesterol or CHS molecules.
Conclusions and comments
- Bearing in mind that the final buffer used in the purification of these channels was at pH8, would it be expected that the channel were open? Further to this, what pH were the simulations performed at?
- The rational design of monomerisation mutants was a really nice touch. It would be interesting to see if these monomers are functional, as this may give us evidence to support a central pore in each monomer, rather than between monomers.
- Multiple mutations were made in the N- and C-terminal domains and tested for activity; however mutation of the QNY motif was not tested, was this because mutants could not be purified due to lowered yield or stability?
- In this work it was stated that a dozen orthologues were tested: it would be really interesting to see the FSEC traces and rationale used to choose the final orthologues.
- There are several different possibilities for solubilising membrane proteins, including detergents, SMA, amphipols or nanodiscs. How was the nanodisc approach decided upon?
Work still remains to figure out where the natural proton channel resides in these Otop channels. Future work may focus on solving CryoEM structures at a lower pH, which may show an open channel. This is a really nice body of work that I look forward to hearing more about in the future.
Why I chose this article
This is a really nice article covering the structure determination of two channels that have only recently been identified. I really like the mutagenesis work, both trying to find the gating residues and rational design of monomer mutants.
- Tu Y-H, Cooper AJ, Teng B, Chang RB, Artiga DJ, Turner HN, et al. An evolutionarily conserved gene family encodes proton-selective ion channels. Science. 2018;359(6379):1047.
- Kühlbrandt W. The Resolution Revolution. Science. 2014;343(6178):1443.
- Kawate T, Gouaux E. Fluorescence-Detection Size-Exclusion Chromatography for Precrystallization Screening of Integral Membrane Proteins. Structure. 2006 2006/04/01/;14(4):673-81.
Posted on: 12th February 2019