Ion channels are pore-forming integral membrane proteins that control the diffusion of ions across biological membranes by gating, or transforming between open (conducting) and closed (non-conducting) conformations in response to various triggers, such as changes in the voltage across the cell membrane. In many ion channels, a water-filled, hydrophobic cavity in the center of the pore domain, i.e., the central cavity, is at the core of ion conduction, but the properties of these cavities are largely unknown. It has been shown experimentally that the gating of several types of ion channels includes a solvent-dependent component . In addition, molecular dynamics simulations have predicted that the solvent-dependent component involves a dehydration transition that drives the central cavity to be emptied and collapsed upon channel closing [2, 3], a process referred to as hydrophobic gating .
Water in ion channel pores is difficult to detect, and whether the central cavity is empty in the closed state has been an open question. While X-ray crystallography has traditionally been used for determining the structure of soluble proteins, membrane proteins are more challenging to characterize by this method. For one, membrane proteins are usually crystallized in detergents, and not their natural lipid membrane environment. Moreover, only water molecules that are sufficiently ordered may be visible in crystallographic maps. Therefore, the conformation, hydration state, and interactions with lipid membranes of ion channels have remained largely unexplored experimentally, limiting our knowledge of the roles of water and the native membrane in ion channel structure and function.
Researchers at IBBR, the NIST Center for Neutron Research (NCNR), and the National Institutes of Health (NIH) have used neutron diffraction and solid-state NMR to determine and quantify water distributions along the conduction pores in two tetrameric channel proteins embedded in lipid membranes: the potassium channel KcsA of Streptomyces lividans, and the trans-membrane domain of the M2 proton channel of influenza A virus. Neutron diffraction is the key technique for such studies because it takes advantage of contrast variation between hydrogen (H) and deuterium (2H). Replacing H with 2H in the solvent, protein, or lipid fatty acid chains does not compromise the original structures of these components within the system, and is a highly sensitive probe, due to the large difference in scattering length between the two isotopes. Ultimately, by varying the H-to-2H ratio of the components within the system, it is possible to change which component(s) is highlighted in a given experiment and, therefore, selectively gain information.
For the KcsA channel, the pore of the protein is filled with water in the closed state and therefore not collapsed, in contrast with the results from the molecular dynamics simulations discussed above. The water distribution is peaked in the middle of the membrane, consistent with water being present in the central cavity. For the trans-membrane domain of the M2 channel, which was chosen as a comparative example, the pore of the protein is filled with water in the closed state, and the water distribution has an hourglass shape such that it peaks near the exterior of the membrane and dips near the center of the membrane. The water distributions for both proteins studied are consistent with the shapes of the pores observed in the available high-resolution crystal structures [5-9], suggesting that the water profiles report on the structures and functional states of the channels. It should also be noted that the degree of hydration observed for the KcsA pore is higher than that inferred from its crystal structures [5-7], consistent with water being clustered at the exits of the pore in addition to the central cavity. Such water clusters could form hydrogen-bonded networks that help to stabilize water in the pore.
The results of this study provide the missing link between molecular dynamics simulations of ion channels in lipid membranes and crystal structures in detergents, increasing our understanding of the role that pore hydration plays in channel gating.
This work was published as: J.R. Blasic, D.L. Worcester, K. Gawrisch, P. Gurnev, and M. Mihailescu, “Pore Hydration States of KcsA Potassium Channels in Membranes.” J. Biol. Chem. (2015) 290, 26765-26775. doi: 10.1074/jbc.M115.661819
- Zimmerberg, J., Bezanilla, F., and Parsegian, V. A. (1990) Biophys. J. 57, 1049–1064.
- Jensen, M. Ø., et al. (2010) PNAS 107, 5833–5838.
- Jensen, M. Ø., et al. (2012) Science 336, 229–233.
- Aryal, P., Sansom, M. S., and Tucker, S. J. (2015) J. Mol. Biol. 427, 121–130.
- Doyle, D. A., et al. (1998) Science 280, 69–77.
- Zhou, Y., et al. (2001) Nature 414, 43–48.
- Uysal, S., et al. (2009) PNAS 106, 6644–6649.
- Stouffer, A. L., et al. (2008) Nature 451, 596–599.
- Acharya, R., et al. (2010) PNAS 107, 15075–15080.