SNF - Electron Microscopy of Membrane Proteins

Membrane proteins are central to health and disease, and represent the major focus of intensive research efforts. Structure determination of membrane proteins faces hurdles, mostly in expression, purification, and 3D crystallization. Electron crystallography represents a valuable alternative for the structure determination of membrane proteins.

We propose to structurally study three regulated ion channel membrane proteins by electron crystallography of 2D membrane crystals: The E. coli chloride channel ClC-ec1, now identified as chloride-proton antiporter, the M. loti cyclic nucleotide gated potassium channel MloK1, and the Ca2+ regulated potassium channel MthK from Methanobacterium autotrophicum.

A crystallography structure of ClC-ec1 exists, but the available data do not allow establishing a conclusive model about the functioning and pH-dependent inhibition of the antiporter. For MloK1 only a structure of the cyclic nucleotide binding domain (CNBD) and the transmembrane part alone is available, but no data exist for the full-length molecule that demonstrate the action mechanism of cAMP binding to the CNBDs and their interaction with the putative voltage sensors. For MthK the mechanism of Ca2+ binding to the tetrameric or octameric RCK domains on channel opening/closing has not directly been demonstrated.

We have obtained excellently ordered 2D crystals of ClC-ec1, with which we want to determine the membrane-embedded 3D structure at neutral pH by electron crystallography. Using electron diffraction and molecular replacement, we will then determine the structure at acidic pH. These data should allow determining the conformational changes associated with pH-dependent activation.

We also have 2D crystals of MloK1. We can grow these crystals now also in the presence and absence of cAMP, which then show structurally different projection maps. We will elaborate the membrane-embedded 3D structure of this gated potassium channel, and study the mechanism of its regulation through cAMP, and the channel opening/closing effect on the orientation of the putative voltage sensor domains.

We also have reconstituted MthK in lipid membranes, which always are stacked together. We will use electron tomography and single particle sub-volume averaging to study the intermediate-resolution structure of MthK in function of Ca2+ binding, and characterize the structural modifications for the channel dimensions.