A high‐resolution solid‐state ¹³C‐NMR study on crystalline bovine heart cytochrome‐c oxidase and lysozyme

Abstract

We have recorded 100.6-MHz high-resolution solid-state ' 3C-NMR spectra of crystalline cytochrome-c oxidase from bovine heart muscle and hen egg-white lysozyme, to compare conformation and dynamics of a typical membrane-protein complex with those of lysozyme. The absence of severe interference with the solid-state 13C-NMR spectra, from both the line broadenings from paramagnetic centers and overlapping of intense detergent signals, provided spectral resolution of 3C-NMR feature of cytochrome-c oxidase crystals comparable to that of lysozyme crystal and better than that of dissolved or lyophilized samples. In fact, the observed peak intensities of the polar heads of the detergents BL8SY and Brij 35 were only about 10% and 3% of the anticipated values, respectively.The dynamic behavior of the backbone and side chains of cytochrome-c oxidase was compared with that of lysozyme on the basis of the 13C spin-lattice relaxation times (Tl): the backbone of the cytochrome-c oxidase turned out to be more flexible than that of lysozyme. Molecular motions of the detergent molecules attached to the proteins are found to be highly heterogeneous. Detergent molecules undergo rapid tumbling motions in the crystals in about 10 ns as detected by T l . In addition to rapid motions, slow motions were detected by 'H spin-lattice relaxation time in the rotating frame ( TyQ) and cross-polarization time ( TCH), together with data from static spectra, indicating that the aliphatic portion of the detergent interacts more strongly with hydrophobic protein surfaces than do the polar heads.Knowledge of three-dimensional structure of atomic resolution is indispensable for understanding protein function. This goal will be, in principle, achieved by X-ray diffraction or multi-dimensional NMR spectroscopy combined with suitable molecular modelling (Wiithrich et al., 1986(Wiithrich et al., , 1989Clore and Gronenborn, 1991). This approach, however, is not always straightforward for transmembrane proteins, because crystallization of membrane proteins is extremely difficult compared with that of water-soluble proteins. Furthermore, many NMR peaks are masked by strong signals from the sohbilization detergent. In fact, only limited examples of three-dimensional structures of transmembrane proteins have been revealed by X-ray diffraction, such as the photosynthetic reaction center from Rhodopseudomonas viridis (Deisenhofer et al., 1985) and bacterial porin (Garavito and Rorsenbasch, 1980).

Resonance Raman studies on the CuA site of cytochrome c oxidase using a multichannel scanning Raman spectrometer with a CCD detector

“…The weak amide III intensity in the frequency region between 1235and 1240 cm™1 and the stronger intensity around 1300 cm™1 suggest a small population of the /3-sheet and a larger population of -helix, respectively (Lippert et al, 1976). The large content of -helix is consistent with the results from circular dichroism (Urry et al, 1967), solidstate NMR (Tuzi et al, 1992), and the primary structure (Frey et al, 1985) as well as the amide I Raman band described above. Lippert et al (1976) proposed a method for estimating the secondary structure distribution of proteins from Raman spectra.…”

“…(1) All the observed protein modes exhibited neither frequency shifts nor intensity changes upon the redox changes of the metal centers. This is in consequence with the results from solid-state NMR studies which detected no difference in protein signals between the resting and fully reduced states (Tuzi et al, 1992). Presumably, the redox-dependent conformational change of the protein is localized, and its range is considerably small compared with a large molecular size of this enzyme.…”

“…Although the -helix content is in agreement with that of the CD study (Urry et al, 1967), this may be underestimated since it did not undergo a deuteration shift, and the content of nonregular structure may be overestimated. Recent results from solid-state NMR estimate the -helix content to be as high as 0.6 (Tuzi et al, 1992).…”

Hydration-Coupled Dynamics in Proteins Studied by Neutron Scattering and NMR: The Case of the Typical EF-Hand Calcium-Binding Parvalbumin

“…The resemblance between the selective narrowing of the lysyl 13 C ⑀ and 13 C ␦ resonances in the solid-state 13 C-NMR spectra of parvalbumin and the narrowing of the same carbon resonances in poly(Llysine), at increasing hydration, suggests that in both cases these specific effects involve an increase in the mobility of the lysyl side chain as a whole. A solid-state 13 C-NMR study of crystalline (fully hydrated) lysozyme showed that the lysyl 13 C ⑀ resonances in the 41 ppm region of the spectrum display a relatively short longitudinal relaxation time of 0.4 s (mean value), indicating rapid tumbling in the 10 ns time range (Tuzi et al, 1992). Similarly, the T 1 values of the 13 C ⑀ resonances of hydrated poly(L-lysine) powders (h up to 0.5) are in the 0.2-0.4 s range (at h ϭ 0.5, T 1 ϭ 0.36 s), indicating motions of the lysyl polar heads in the 10 ns.…”

Dynamics of Reassembled Thioredoxin Studied by Magic Angle Spinning NMR: Snapshots from Different Time Scales

“…[15][16][17][18][19][20][21][22][23][24][25][26][27] Solidstate NMR methods can probe internal dynamics in the absence of the overall molecular tumbling, using anisotropic line shape analysis and relaxation measurements. [28][29][30][31] In solid-state NMR, the anisotropic resonance frequency depends on the orientation of the molecular frame with respect to the static magnetic field.…”