New Horizons in Structural Biology of Membrane Proteins: Experimental Evaluation of the Role of Conformational Dynamics and Intrinsic Flexibility

Abstract

A plethora of membrane proteins are found along the cell surface and on the convoluted labyrinth of membranes surrounding organelles. Since the advent of various structural biology techniques, a sub-population of these proteins has become accessible to investigation at near-atomic resolutions. The predominant bona fide methods for structure solution, X-ray crystallography and cryo-EM, provide high resolution in three-dimensional space at the cost of neglecting protein motions through time. Though structures provide various rigid snapshots, only an amorphous mechanistic understanding can be inferred from interpolations between these different static states. In this review, we discuss various techniques that have been utilized in observing dynamic conformational intermediaries that remain elusive from rigid structures. More specifically we discuss the application of structural techniques such as NMR, cryo-EM and X-ray crystallography in studying protein dynamics along with complementation by conformational trapping by specific binders such as antibodies. We finally showcase the strength of various biophysical techniques including FRET, EPR and computational approaches using a multitude of succinct examples from GPCRs, transporters and ion channels.

Keywords: EPR; FRET; HDX-MS; NMR; X-ray; antibody fragments; cryo-EM; membrane protein dynamics; membrane protein structure; membrane proteins.

Figure 1

(A) Example of ¹⁹F-NMR spectrum, shown for NCX channel in nanodisc with ¹⁹F-probe attached to the side chain of residue 220. Signals are fitted by Lorentzian lines. (B) Dynamic energy landscape encompassing key functional states of adenosine A_2AR [9]. The conformational ensemble is represented by five key functional states—two inactive states (S1 and S2) differentiated by the switching of a conserved ionic lock and three active states (A1, A2, and A3) associated with G protein coupling. A3, an intermediate or precoupled state, plays a role in the recognition and binding of the G protein. A1 and A2, on the other hand, are responsible for GDP release and stabilization of the nucleotide-free complex. While A1 is more efficacious (thicker downward arrow) and stabilized to a larger extent by the full agonist, A2 is less efficacious (thinner downward arrow) and is preferentially stabilized by a partial agonist. Reproduced with permission from Elsevier.

Figure 2

(A) 3D class averages from a combined dataset of Slo2.2 vitrified in the presence of 20 mM, 40 mM, 80 mM, and 160 mM Na⁺ concentrations. Classes that resemble closed Slo2.2 are colored red and open colored blue. (B) A fraction of particles in each class is shown for the 20–160 mM Na⁺ concentrations. Reproduced with permission from Elsevier. (C) Structures illustrate a twisting elevator movement to effect substrate traversal across the bilayer. Transport domain is depicted as colored cylinders and scaffold domain as dark grey surface. For clarity, only one protomer of the obligate homotrimer is shown. Models were resolved by X-ray crystallography, cryo-EM, and aided by engineered disulfide trapping. Left panel: outward-open Glt_Ph (PDB 2NWW); middle panel: intermediate, putative Cl⁻ conducting state of Glt_Ph (PDB 6WYK); right panel: inward-open (TBOA-inhibited) state of Glt_Tk (PDB 6XWR).

Figure 3

bR retinal trans-to-cis isomerization observed by ultrafast TR-SFX. Dark grey indicates the dark state, red and orange intermediates (49–406 and 457–646 fs), and blue the 13-cis isomer (10 ps) prior to relaxation. Illustration derived from PDB 6G7H, 6G7I, 6G7J, 6G7K.

Figure 4

(A) Schematic showing the overlap of the donor emission and acceptor excitation spectra leading to the observation of FRET. Additionally shown in inset is the effect of R₀ on FRET intensity. (B) Different membrane mimicking systems available for in vitro reconstitution of membrane protein for FRET studies. Nanodisc image was generated using BioRender. (C) The different conformational intermediates occupied by MsbA (i) and DtpA (ii) during the transport of substrate molecule. Nanobody against the closed periplasmic conformer of DtpA is shown in purple. The position of F259 residue in TM helix 6b of LeuT (PDB 3GJD) is shown that acts as a volumetric sensor (iii).

Figure 5

(A) Energy level diagram of MTSL. Interaction of the nuclear and electron spins causes further splitting of the spin system with different energy levels known as hyperfine interaction. EPR spectrum of the free spin label (blue) overlaid onto the spectrum of spin label conjugated to a protein (red), bottom panel. (B) The DEER four-pulse sequence. For interacting spins, the modulation observed from the refocused echo in the time domain is directly used to measure distances using model-independent analysis. (C) (i) The various states of TM6 in β₂AR. Inactive states (S1, S2); active intermediate state S3 in the presence of agonist (green); and active state S4 with Nb80. Positions of spin-label, IA-PROXYL on TM4 and TM6 for EPR measurements are shown in red. (ii) Structure of GLIC (PDB 5J0Z) highlighting the outermost helix M4 (blue) that interacts with DHA (magenta). Top view without the extracellular domain showing the outward movement of M4. (iii) Structure of BmrCD (PDB 7M33) with protomers BrmD (in yellow) and BmrC (in orange). Arrows indicate position of spin labels. Schematic showing the consensus (C, solid circle), and degenerate NBS (D, broken circle) along with a schematic of DEER distance measurements indicating the presence of HES (H), nucleotide bound (N) and apo (A) conformational states for each of the NBSs.

Figure 6

(A) Schematic showing the extent of deuterium exchange using a color gradient from red (decreased)—white (unchanged)—blue (increased) in the TMs of NhaA. The yellow star denotes the binding site for Li⁺. (B) The structure of F₀F₁-ATP synthase (PDB 3OAA) highlights the C-terminal helix in the γ subunit which shows maximum exchange.

Figure 7

(A) Conventional monoclonal antibody (mAb). Various fragments obtain from chemical or enzymatic cleavage. (B) Heavy chain antibody (hcAb) found in camelids. The small (~15 kDa) variable region (VHH) from hcAb is called a nanobody. (C) (i) Ab fragment stabilized conformers. Structure of β₂AR and A_2AR stabilized by antibody fragments Fab5 (PDB 2R4R) and Fab2838 (PDB 3VG9). (ii) Alignment of β₂AR structures in the presence of different conformational Nbs 60 (5JQH), 71 (6MXT), 80 (3P0G), mimicking a transition from inactive to active state, indicated the outward movement of intracellular TM6. (iii) Overlay of outward-open structures of LacY in the apo and substrate NPG (blue) bound state, conformationally trapped by Nb 9039 (5GXB) and 9047 (6C9W) respectively.