Micelle-induced curvature in a water-insoluble HIV-1 Env peptide revealed by NMR dipolar coupling measurement in stretched polyacrylamide gel

The structure of a water-insoluble fragment encompassing residues 282-304 of the HIV envelope protein gp41 is studied when solubilized by dihexanoyl phosphatidylcholine (DHPC) and by small bicelles, consisting of a 4:1 molar ratio of DHPC and dimyristoyl phosphatidylcholine (DMPC). Weak alignment with the magnetic field was accomplished in a stretched polyacrylamide gel, permitting measurement of one-bond (1)H-(15)N, (13)Ca-(13)C', and (13)C'-(15)N dipolar couplings, which formed the basis for determining the peptide structure. In both detergent systems, the peptide adopts an alpha-helical conformation from residue 4 through 18. In the presence of the DHPC micelles the helix is strongly curved towards the hydrophobic surface, whereas in the presence of bicelles a much weaker curvature in the opposite direction is observed.

Morphology of three lyotropic liquid crystalline biological NMR media studied by translational diffusion anisotropy

The morphologies of three dilute liquid crystalline phases, which are widely used for biological NMR spectroscopy, are investigated by the study of tracer self-diffusion. The aqueous liquid crystalline media investigated include the common phospholipid bicelle medium, a phase consisting of a mixture of pentaethyleneglycol mono dodecyl ether and hexanol, and a medium containing cetylpyridinium bromide and hexanol. Threonine and water were used as tracer molecules for probing the aqueous environment, and tetramethylsilane (TMS) was for probing the lipophilic environment. Pulsed field gradient NMR was used to measure tracer self-diffusion rates in three orthogonal directions. Although results for the water-soluble tracers in bicelle media do not contradict the widely accepted disk-shaped bicelle model, the high TMS diffusion rate observed in the bilayer plane requires extensive transient edge-to-edge contacts of such disks. This morphology is essentially that of a heavily perforated lamellar bilayer phase and explains why this medium remains liquid crystalline well below the Onsager limit for disk-shaped nematogens. Below 25 degrees C, a bicelle mixture consisting of dimyristoyl phosphatidyl choline and dihexanoyl phosphatidyl choline remains isotropic, but tracer diffusion obstruction indicates that the particles are significantly oblate. The diffusion anisotropy in the penta(ethyleneglycol) mono dodecyl ether liquid crystals confirms the previously proposed alpha-lamellar phase. However, weak inhibition of aqueous-phase self-diffusion in the z direction points to the presence of bridge- or caplike obstructions, and the bilayers appear slightly permeable to water. If the previously proposed concentric cylinder superstructure of bilayers applies, the diffusion data indicate that the most outer cylinder must have a diameter greater than 50 microm. The tracer self-diffusion data for the cetylpyridinium bromide/hexanol medium is only compatible with a planar alpha-lamellar phase, with its local director orthogonal to the magnetic field, and a very large domain size over which the director remains parallel.

A simple apparatus for generating stretched polyacrylamide gels, yielding uniform alignment of proteins and detergent micelles

Compressed and stretched polyacrylamide hydrogels previously have been shown to offer a robust method for aligning proteins. A simple, funnel-like apparatus is described for generating uniformly stretched hydrogels. For prolate-shaped proteins, gels stretched in the direction of the magnetic field yield two-fold larger alignment than gels compressed to the same aspect ratio in this direction. Empirically, protein alignment is found to be proportional to (c-2.3)2 [(d(o/dN)3-1], where do and dN are the diameters of the cylindrical gels before and after stretching, respectively, and c is the polyacrylamide weight fraction in percent. Low gel densities, in the 4-7% range, are found to have minimal effects on macromolecular rotational correlation times, tauc, and no effect of the compression ratio on tauc could be discerned over the range studied (do/dN < or = 1.4). Application is demonstrated for a sample containing the first Ig-binding domain of protein G, and for a detergent-solubilized peptide.

Solution structure of Ca(2+)-calmodulin reveals flexible hand-like properties of its domains

The solution structure of Ca(2+)-ligated calmodulin is determined from residual dipolar couplings measured in a liquid crystalline medium and from a large number of heteronuclear J couplings for defining side chains. Although the C-terminal domain solution structure is similar to the X-ray crystal structure, the EF hands of the N-terminal domain are considerably less open. The substantial differences in interhelical angles correspond to negligible changes in short interproton distances and, therefore, cannot be identified by comparison of NOEs and X-ray data. NOE analysis, however, excludes a two-state equilibrium in which the closed apo conformation is partially populated in the Ca(2+)-ligated state. The difference between the crystal and solution structures of Ca(2+)-calmodulin indicates considerable backbone plasticity within the domains of calmodulin, which is key to their ability to bind a wide range of targets. In contrast, the vast majority of side chains making up the target binding surface are locked into the same chi(1) rotameric states as in complexes with target peptide.

Measurement of homonuclear proton couplings based on cross-peak nulling in CT-COSY

A method in which 1H-1H scalar and dipolar couplings are obtained from the cross-peak nulling condition in a series of constant-time (CT) COSY spectra, as a function of the duration of the CT period, is described. The method is best suited for measurement of 1H-1H couplings in the range 5-20 Hz. It is shown, however, that results can be sensitive to cross-correlated relaxation effects. Also, artifactual resonances, resulting from strong coupling, can be quite pronounced in CT-COSY spectra, even when /J(AB)/(deltaA-deltaB)/<0.1. The experiments are demonstrated for the DNA dodecamer d(CGCGAATTCGCG)2, both in isotropic solution and in a liquid crystalline phase.

Characterization of molecular alignment in aqueous suspensions of Pf1 bacteriophage

The phase diagram of Pf1 solutions has been studied indirectly by observation of 2H quadrupole splittings of the solvent signal and measurement of dipolar couplings in solute macromolecules. At low volume fractions of Pf1 and at high ionic strength, alignment of both the phage and the solute depends strongly on the strength of the magnetic field. Both the theoretical and experimentally determined phase diagram of Pf1 show that at low concentrations and high ionic strengths the solution becomes isotropic. However, just below the nematic phase boundary the behavior of the system is paranematic, with cooperative alignment which depends on the strength of the applied magnetic field. Above 16 mg/ml Pf1 is fully nematic up to 600 mM NaCl. Alignment of proteins with a significant electric dipole moment, which tends to be strong in Pf1, can be reduced by either high ionic strength or low phage concentration. Because ionic strength modulates both the orientation and magnitude of the alignment tensor in Pf1 medium, measurement at two ionic strengths can yield linearly independent alignment tensors.

Multiplet component separation for measurement of methyl 13C-1H dipolar couplings in weakly aligned proteins

A simple spectral editing procedure is described that generates separate subspectra for the methyl 13C-[1H3] multiplet components of 1H-13C HSQC spectra. The editing procedure relies on co-addition of in-phase and antiphase spectra and yields 1H-coupled constant-time HSQC subspectra for the methyl region that have the simplicity of the regular decoupled CT-HSQC spectrum. Resulting spectra permit rapid and reliable measurement of 1H-13C J and dipolar couplings. The editing procedure is illustrated for a Ca2+-calmodulin sample in isotropic and liquid crystalline phases.

H...N hydrogen bond lengths in double stranded DNA from internucleotide dipolar couplings

The ratio of the internucleotide dipolar coupling and the corresponding one-bond imino 15N-1H dipolar coupling provides a measure for the N...H/H-N distance ratio. Measurements were carried out for a dodecamer, d(CGCGAATTCGCG)2, in which a C-G and an A-T basepair were uniformly enriched in 15N. When assuming H-bonds to be perfectly linear, dipolar data indicate time-averaged hydrogen bond lengths of 1.80 +/- 0.03 A for A-T and 1.86 +/- 0.02 A for C-G. When using H-bond orientations from high resolution X-ray data, H-bond lengths are about 0.1 A shorter.

Measurement of 1H3'-31P dipolar couplings in a DNA oligonucleotide by constant-time NOESY difference spectroscopy

The ratios of cross peak intensities in a selective constant-time NOESY experiment, recorded with and without 31P decoupling, yield values for the sum of the H3'-P scalar and dipolar couplings. The selective refocusing of H3' resonances in this experiment results in excellent resolution and sensitivity, even in the liquid crystalline phase where the 1H spectrum is broadened by unresolved homonuclear dipolar couplings. The vicinal H3'-P scalar and dipolar couplings in the DNA oligomer d(CGCGAATTCGCG)2 were measured in both isotropic solution, and in a liquid crystalline phase. Isotropic values are in good agreement with values reported previously. Dipolar couplings are in excellent agreement with the NMR structure for this dodecamer, and to a somewhat lesser extent with the X-ray structures.

Measurement of homonuclear proton couplings from regular 2D COSY spectra

An interactive computer procedure is described which determines (1)H--(1)H couplings from fitting the cross-peak multiplets in regular phase-sensitive COSY spectra. The robustness and simplicity of the method rely on the fact that a given cross-peak intensity is not an independent variable in the fitting procedure, making it possible to measure couplings accurately even from individual cross peaks with unresolved multiplet structure.

A model for the abrogation of the SOS response by an SOS protein: a negatively charged helix in DinI mimics DNA in its interaction with RecA

DinI is a recently described negative regulator of the SOS response in Escherichia coli. Here we show that it physically interacts with RecA and prevents the binding of single-stranded DNA to RecA, which is required for the activation of the latter. DinI also displaces ssDNA from a stable RecA-DNA cofilament, thus eliminating the SOS signal. In addition, DinI inhibits RecA-mediated homologous DNA pairing, but has no effect on actively proceeding strand exchange. Biochemical data, together with the molecular structure, define the C-terminal alpha-helix in DinI as the active site of the protein. In an unusual example of molecular mimicry, a negatively charged surface on this alpha-helix, by imitating single-stranded DNA, interacts with the loop L2 homologous pairing region of RecA and interferes with the activation of RecA.

Solution structure of DinI provides insight into its mode of RecA inactivation

The Escherichia coli RecA protein triggers both DNA repair and mutagenesis in a process known as the SOS response. The 81-residue E. coli protein DinI inhibits activity of RecA in vivo. The solution structure of DinI has been determined by multidimensional triple resonance NMR spectroscopy, using restraints derived from two sets of residual dipolar couplings, obtained in bicelle and phage media, supplemented with J couplings and a moderate number of NOE restraints. DinI has an alpha/beta fold comprised of a three-stranded beta-sheet and two alpha-helices. The beta-sheet topology is unusual: the central strand is flanked by a parallel and an antiparallel strand and the sheet is remarkably flat. The structure of DinI shows that six negatively charged Glu and Asp residues on DinI's kinked C-terminal alpha-helix form an extended, negatively charged ridge. We propose that this ridge mimics the electrostatic character of the DNA phospodiester backbone, thereby enabling DinI to compete with single-stranded DNA for RecA binding. Biochemical data confirm that DinI is able to displace ssDNA from RecA.

Study of conformational rearrangement and refinement of structural homology models by the use of heteronuclear dipolar couplings

For an increasing fraction of proteins whose structures are being studied, sequence homology to known structures permits building of low resolution structural models. It is demonstrated that dipolar couplings, measured in a liquid crystalline medium, not only can validate such structural models, but also refine them. Here, experimental 1H-15N, 1Halpha-13Calpha, and 13C'-13Calpha dipolar couplings are shown to decrease the backbone rmsd between various homology models of calmodulin (CaM) and its crystal structure. Starting from a model of the Ca2+-saturated C-terminal domain of CaM, built from the structure of Ca2+-free recoverin on the basis of remote sequence homology, dipolar couplings are used to decrease the rmsd between the model and the crystal structure from 5.0 to 1.25 A. A better starting model, built from the crystal structure of Ca2+-saturated parvalbumin, decreases in rmsd from 1.25 to 0.93 A. Similarly, starting from the structure of the Ca2+-ligated CaM N-terminal domain, experimental dipolar couplings measured for the Ca2+-free form decrease the backbone rmsd relative to the refined solution structure of apo-CaM from 4.2 to 1.0 A.

Measurement of one-bond 15N-13C' dipolar couplings in medium sized proteins

A simple and accurate method is described for measurement of 1J(C'N) splittings in isotopically enriched proteins. The method is of the quantitative J correlation type, and the 1J(C'N) splitting is derived from the relative intensity in two 3D TROSY-HNCO spectra with 1J(C'N) dephasing intervals of approximately 1/(2 1J(C'N)) (reference intensity) and approximately 1/1J(C'N) (residual intensity). If the two spectra are recorded under identical conditions and with the same number of scans, the random error in the 1J(C'N) value extracted in this manner is inversely related to the signal-to-noise (S/N) in the reference spectrum. A S/N of 30:1 in the reference spectrum yields random errors of less than 0.2 Hz in the extracted 1J(C'N) value. Dipolar couplings obtained from the difference in 1J(C'N) splitting in the isotropic and liquid crystalline phase for the C-terminal domain of calmodulin are in excellent agreement with its 1.68-A crystal structure, but agree considerably less with the 2.2-A structure.