Molecular Dynamics of Proteorhodopsin in Lipid Bilayers by Solid-State NMR

Abstract

Environmental factors such as temperature, hydration, and lipid bilayer properties are tightly coupled to the dynamics of membrane proteins. So far, site-resolved data visualizing the protein's response to alterations in these factors are rare, and conclusions had to be drawn from dynamic data averaged over the whole protein structure. In the current study, high-resolution solid-state NMR at high magnetic field was used to investigate their effects on the molecular dynamics of green proteorhodopsin, a bacterial light-driven proton pump. Through-space and through-bond correlation experiments were employed to identify and characterize highly mobile and motionally restricted regions of proteorhodopsin. Our data show that hydration water plays an essential role for enhancing molecular dynamics of residues in tails and interhelical loops, while it is found less important for residues in transmembrane domains or rigid, structured loop segments. In contrast, switching the lipids from the gel to their liquid crystalline phase enhances molecular fluctuations mainly in transmembrane helices on a time scale of 10(-6) s, but has little effect on loop and tail residues. Increased mobility is especially observed in helices C, F, and G, but also in the EF loop. Fluctuations in those regions are relevant to structural dynamics during the photocycle of proteorhodopsin. Our data are important for the functional understanding of proteorhodopsin, but also offer an important contribution to the general understanding of site-resolved effects of water and lipid bilayers onto the dynamic properties of membrane proteins.

Recent Solid-State NMR Studies of Membrane-Bound Peptides and Proteins

“…Adequate resolution of C-C correlations is key to successful structure determination experiment, because the 5-70 ppm aliphatic region is the fingerprint region for residue types. Well-resolved 13 C-13 C correlations have recently been reported for a number of membrane proteins, including the trimeric autotransporter adhesion protein YadA [170], the potassium channel KcsA-Kv1.3 [178], human aquaporin-1 [179], DsbB [172], proteorhodopsin [180,181], and sensory rhodopsin II (SRII) [182]. NCOCX (for inter-residue along protein backbone), NCACX (for intraresidue), and CONCA (for inter-residue) with 13 C-13 C correlations construct spin systems for individual amino acid residues and connect the spin systems along the amino acid sequence.…”

Isotope Labeling of Eukaryotic Membrane Proteins in Yeast for Solid-State NMR

“…Once the purification is complete, it is important to precisely quantify the purified isotope-labeled protein before attempting lipid reconstitution, as protein-to-lipid ratio may critically affect the quality of ssNMR spectra (Shi et al, 2009;Yang, Aslimovska, & Glaubitz, 2011). While quantification of purified LR and other microbial rhodopsins is trivial, as it can be done spectrophotometrically, we used several concurrent methods to determine an exact amount of hAQP1.…”

“…In principle, the higher protein-to-lipid ratios can lead to both better spectral resolution and higher signal amplitudes of ssNMR, which in some cases can be related to formation of 2D crystals, e.g., in the case of Anabaena sensory rhodopsin (Ward, Wang, et al, 2015). On the other hand, for some proteins, such as green proteorhodopsin, having somewhat lower protein-to-lipid ratios not leading to 2D crystallization produces more homogeneous samples with better spectral resolution (Shi et al, 2009;Yang et al, 2011). It should be also taken into the account that the final actual protein-to-lipid ratio can be somewhat different from the intended one, depending on how much of endogenous lipids is retained by the protein.…”

Full-length Vpu and human CD4(372–433) in phospholipid bilayers as seen by magic angle spinning NMR

“…Dipolar coupling-based CP (Hartmann and Hahn, 1962;Pines et al, 1973) is effective if the vector connecting the interacting nuclei is restricted in mobility and correlation times of residual motions are above the millisecond range. In contrast, 1 H-13 C polarization transfer by insensitive nuclei enhanced by polarization transfer (INEPT) (Morris and Freeman, 1979;Burum and Ernst, 1980) is effective only if dipolar couplings are averaged out by large amplitude motions on time scales of < 10 -5 s (Andronesi et al, 2005;Yang et al, 2011). Slower motions in the range between milliseconds and microseconds on the other hand can affect T 1ρ relaxation and thus decrease the CP efficiency even if the amplitude of the motion is not large enough for isotropic averaging of the dipolar coupling (Fares et al, 2005;Sharpe et al, 2006).…”

“…The resulting proteo liposomes are too large for solution-state NMR studies. Solid-state NMR spectroscopy offers a viable alternative for investigations of membrane proteins reconstituted in native phospholipid bilayers (McDermott, 2009;Marassi et al, 2011;Yang et al, 2011;Hong et al, 2012;Miao et al, 2012).…”