The major coat protein (gVIIIp) of bacteriophage M13 solubilized in sodium dodecyl sulfate (SDS) detergent micelles was used as a model system to study this protein in the lipid-bound form. In order to probe the position of gVIIIp relative to the SDS micelles, stearate was added, spin-labeled at the 5- or 16-position with a doxyl group containing a stable nitroxide radical. The average position of the spin-labels in the micelles was derived from the line broadening of the resonances in the 13C spectrum of SDS. Subsequently, we derived a model of the relative position of gVIIIp in the SDS micelle from the effect of the spin-labels on the gVIIIp resonances, monitored via 1H-15N HSQC and TOCSY experiments. The results are consistent with the structure of gVIIIp having two helical strands. One strand is a long hydrophobic helix that spans the micelle, and the other is a shorter amphipathic helix on the surface of the micelle. These results are in good agreement with the structure of gVIIIp in membranes proposed by McDonnell et al. on the basis of solid state NMR data [McDonnell, P. A., Shon, K., Kim, Y., & Opella, S. J. (1993) J. Mol. Biol. 233, 447-463]. This study indicates that high-resolution NMR on this membrane protein, solubilized in detergent micelles, is a very suitable technique for mimicking these proteins in their natural environment. Furthermore, the data indicate that the structure of the micelle near the C-terminus of the major coat protein is distorted.(ABSTRACT TRUNCATED AT 250 WORDS)

The backbone dynamics of the major coat protein (gVIIIp) of the filamentous bacteriophage M13, solubilized in detergent micelles, have been studied using 15N nuclear magnetic resonance spectroscopy at three frequencies. Motional parameters and overall and internal correlation times were derived with the model-free approach. It was also checked whether these parameters had to be modified due to anisotropic motion of the protein/micelle complex. Reduced spectral density mapping was used to calculate the spectral densities at J(O), J(omegaN), and [J(omegaH)]. The spectral densities were interpreted by mapping a linear or scaled linear combination of two Lorentzians onto a J(O)-J(omega) plot. The major coat protein of bacteriophage M13 consists of two alpha-helices, one of which is hydrophobic and located within the micelle, while the other is amphipathic and located on the surface of the micelle. Our results indicate that the motion of the hydrophobic helix is restricted such that it corresponds to the overall tumbling of the protein/micelle complex. The interpretation of the relaxation data of the amphipathic helix by means of the model-free approach and the reduced spectral density mapping indicate that in addition to the overall motion all residues in this helix are subject to motion on the fast nanosecond and picosecond time scales. The motions of the vectors in the low nanosecond range are characterized by similar values of the spectral densities and correlation times and represent the motion of the amphipathic helix on and away from the surface of the micelle. The relaxation data of the residues in the hinge region connecting the helices show that there is an abrupt change from highly restricted to less restricted motion. Both the C-terminal and N-terminal residues are very mobile.

The solution structure of the 18-kDa single-stranded DNA binding protein encoded by the filamentous Pseudomonas bacteriophage Pf3 has been refined using 40 ms 15N- and 13C-edited NOESY spectra and many homo- and heteronuclear J-couplings. The structures are highly precise, but some variation was found in the orientation of the beta-hairpin denoted the DNA binding wing with respect to the core of the protein. Backbone dynamics of the protein was investigated in the presence and absence of DNA by measuring the R1 and R2 relaxation rates of the 15N nuclei and the 15N-1H NOE. It was found that the DNA binding wing is much more flexible than the rest of the protein, but its mobility is largely arrested upon binding of the protein to d(A)6. This confirms earlier hypotheses on the role of this hairpin in the function of the protein, as will be discussed. Furthermore, the complete DNA binding domain of the protein has been mapped by recording two-dimensional TOCSY spectra of the protein in the presence and absence of a small amount of spin-labeled oligonucleotide. The roles of specific residues in DNA binding were assessed by stoichiometric titration of d(A)6, which indicated for instance that Phe43 forms base stacking interactions with the single-stranded DNA. Finally, all results were combined to form a set of experimental restraints, which were subsequently used in restrained molecular dynamics calculations aimed at building a model for the Pf3 nucleoprotein complex. Implying in addition some similarities to the well-studied M13 complex, a plausible model could be constructed that is in accordance with the experimental data.

Full automation of the analysis of spectra is a prerequisite for high-throughput NMR studies in structural or functional genomics. Sequence-specific assignments often form the major bottleneck. Here, we present a procedure that yields nearly complete backbone and side chain resonance assignments starting from a set of heteronuclear three-dimensional spectra. Neither manual intervention, e.g., to correct lists obtained from peak picking before feeding these to an assignment program, nor protein-specific information, e.g., structures of homologous proteins, were required. By combining two earlier published procedures, AUTOPSY [Koradi et al. (1998) J. Magn. Reson., 135, 288-297] and GARANT [Bartels et al. (1996) J. Biomol. NMR, 7, 207-213], with a new program, PICS, all necessary steps from spectra analyses to sequence-specific assignments were performed fully automatically. Characteristic features of the present approach are a flexible design allowing as input almost any combination of NMR spectra, applicability to side chains, robustness with respect to parameter choices (such as noise levels) and reproducibility. In this study, automated resonance assignments were obtained for the 14 kD blue copper protein azurin from P. aeruginosa using five spectra: HNCACB, HNHA, HCCH-TOCSY, 15N-NOESY-HSQC and 13C-NOESY-HSQC. Peaks from these three-dimensional spectra were filtered and calibrated with the help of two two-dimensional spectra: 15N-HSQC and 13C-HSQC. The rate of incorrect assignments is less than 1.5% for backbone nuclei and about 3.5% when side chain protons are also considered.

The membrane-bound form of the major coat protein (gV11Ip) of bacteriophage M I 3 has been studied using nuclear magnetic resonance spectroscopy. As membrane mimetics, we used dodecylphosphocholine (DodPCho) detergent micelles to solubilize the protein. We were able to nearly completely assign all resonances of the protein solubilized in DodPCho micelles by using both homonuclear and heteronuclear multidimensional experiments. Based on the patterns of the nuclear Overhauser enhancements and the chemical shifts of the resonances, we deduced the secondary structure of the protein. Additional structural information was obtained from amide proton exchange data and J-coupling constants. The protein consists of two a-helices which are connected by a hinge region around residue 21. From spin-label experiments, the location of the protein relative to the DodPCho micelles was determined. One, hydrophobic, helix spans the micelle, and another, amphipathic, helix, is located beneath the surface of the micelle. For gVTTTp in SDS micelles, we found a micellar structure which is distorted near the C-terminus of the protein; whereas for DodPCho micelles, both distorted and regular elliptical micelles occur. This distortion is probably due to the interaction of the positively charged lysine side chains with the negatively charged head group of the detergent molecules.Keywords: major coat protein; bacteriophage M I 3 ; NMR; dodecylphosphocholine; structural analysis.Our knowledge of biological processes involving membranebound proteins is limited in comparison with those involving cytoplasmic proteins. This is due, in large part, to the lack of structural information caused by difficulties in applying the methods of X-ray diffraction and NMR to membrane-bound proteins. Membrane-associated proteins are more difficult to crystallize than globular proteins. Multidimensional high-resolution NMR methods, which work well in the elucidation of the structures of small water-soluble proteins (Wiithrich, 1986) are less well-suited for studying large, slowly reorienting complexes, like membrane-bound proteins. To obtain structural information for these proteins with these techniques, model systems, mostly in the form of detergent micelles, are used as membrane mimetics. In high-resolution NMR, glucagon (Braun et al., 1981) and melittin (Brown et al., 1982) were the initial model systems Correspondence to F. J. M. Van de Ven,