Direct measurement of distances and angles in biomolecules by NMR in a dilute liquid crystalline medium.

In isotropic solution, internuclear dipolar couplings average to zero as a result of rotational diffusion. By dissolving macromolecules in a dilute aqueous nematic discotic liquid-crystalline medium containing widely spaced magnetically oriented particles, a tunable degree of solute alignment with the magnetic field can be created while retaining the high resolution and sensitivity of the regular isotropic nuclear magnetic resonance (NMR) spectrum. Dipolar couplings between 1H-1H, 1H-13C, 1H-15N, and 13C-13C pairs in such an oriented macromolecule no longer average to zero, and are readily measured. Distances and angles derived from dipolar couplings in human ubiquitin are in excellent agreement with its crystal structure. The approach promises to improve the accuracy of structures determined by NMR, and extend the size limit.

Biophysical characterization of one-, two-, and three-tandem repeats of human mucin (muc-1) protein core.

Until recently mucin tandem repeat protein cores were believed to exist in random-coil conformations and to attain structure solely by the addition of carbohydrates to serine and threonine residues. Matsushima et al. (Proteins Struct. Funct. Genet., 7: 125-155, 1990) recently proposed a model of the secondary structure of proline rich tandem repeat proteins that has challenged this idea, especially for the case of the human polymorphic epithelial mucin encoded by the muc-1 gene. We report here results of structural analyses of the muc-1 protein core by using synthetic peptide analogues. Synthetic peptides were prepared to correspond to one-, two-, and three-tandem repeats of muc-1. Results of one- and two-dimensional 1H NMR correlation spectroscopy on these peptides confirm that the muc-1 protein core is not a random-coil secondary structure. Long-lived amide protons are protected in D2O, and increasing spectral complexity in the region of the beta-protons of Asp2 and His 15 reveals that structural changes are occurring as the number of repeats increases. The greatest changes occur when the number of repeats increases from one to two. These results are supported by the reactivity of a panel of monoclonal antibodies raised against tumor associated muc-1 with these synthetic peptides in enzyme-linked immunosorbent assay. The primary immunodominant mucin epitope, PDTRP, does not appear to attain a native conformation in the single repeat peptide (20 amino acids, starting with P), but is expressed on peptides with multiple repeats. Intrinsic viscosity measurements of the peptide containing three repeats indicate that an ordered structure present in solution is rod shaped. The circular dichroism spectrum of the same peptide is dominated by proline in the trans conformation. These results are all consistent with the prediction that the muc-1 tandem repeat polypeptide core forms a polyproline beta-turn helix.

High precision solution structure of the C-terminal KH domain of heterogeneous nuclear ribonucleoprotein K, a c-myc transcription factor.

Among it's many reported functions, heterogeneous nuclear ribonucleoprotein (hnRNP) K is a transcription factor for the c- myc gene, a proto-oncogene critical for the regulation of cell growth and differentiation. We have determined the solution structure of the Gly26-->Arg mutant of the C-terminal K-homology (KH) domain of hnRNP K by NMR spectroscopy. This is the first structure investigation of hnRNP K. Backbone residual dipolar couplings, which provide information that is fundamentally different from the standard NOE-derived distance restraints, were employed to improve structure quality. An independent assessment of structure quality was achieved by comparing the backbone15N T1/T2ratios to the calculated structures. The C-terminal KH module of hnRNP K (KH3) is revealed to be a three-stranded beta-sheet stacked against three alpha-helices, two of which are nearly parallel to the strands of the beta-sheet. The Gly26-->Arg mutation abolishes single-stranded DNA binding without altering the overall fold of the protein. This provides a clue to possible nucleotide binding sites of KH3. It appears unlikely that the solvent-exposed side of the beta-sheet will be the site of protein-nucleic acid complex formation. This is in contrast to the earlier theme for protein-RNA complexes incorporating proteins structurally similar to KH3. We propose that the surface of KH3 that interacts with nucleic acid is comparable to the region of DNA interaction for the double-stranded DNA-binding domain of bovine papillomavirus-1 E2 that has a three-dimensional fold similar to that of KH3.

Structural analysis of the N-terminal domain of the human T-cell leukemia virus capsid protein.

The N-terminal domain of the retroviral capsid (CA) protein is one of the least conserved regions encoded in the genome. Surprisingly, the three-dimensional structures of the CA from different genera exhibit alpha-helical structural features that are highly conserved. The N-terminal residues of the human immunodeficiency virus type 1 (HIV-1) and Rous sarcoma virus (RSV) capsid proteins form a beta-hairpin. To determine if this feature is conserved in the retroviral family, we cloned, expressed, purified, and solved the structure of a N-terminal 134 amino acid fragment (CA(134)) from the human T-cell leukemia virus type 1 (HTLV-I) using high resolution nuclear magnetic resonance (NMR) spectroscopy. The CA(134) fragment contains an N-terminal beta-hairpin and a central coiled-coil-like structure composed of six alpha-helices. The N-terminal Pro1 residue contacts Asp54 in the helical cluster through a salt bridge. Thus, the beta-hairpin is conserved and the helical cluster is structurally similar to other retroviral CA domains. However, although the same Asp residue defines the orientation of the hairpin in both the HTLV-1 and HIV-1 CA proteins, the HTLV-I hairpin is oriented away, rather than towards, the helical core. Significant differences were also detected in the spatial orientation and helical content of the long centrally located loop connecting the helices in the core. It has been proposed that the salt bridge allows the formation of a CA-CA interface that is important for the assembly of the conical cores that are characteristic of HIV-1. As HTLV-I forms spherical cores, the salt-bridge feature is apparently not conserved for this function although its role in determining the orientation of the beta-hairpin may be critical, along with the central loop. Comparison of three-dimensional structures is expected to elucidate the relationships between the retroviral capsid protein structure and its function.

Solution structure of human GAIP (Galpha interacting protein): a regulator of G protein signaling.

The solution structure of the human protein GAIP (Galpha interacting protein), a regulator of G protein signaling, has been determined by NMR techniques. Dipolar couplings of the oriented protein in two different liquid crystal media have been used in the structure calculation. The solution structure of GAIP is compared to the crystal structure of an homologous protein from rat (RGS4) complexed to the alpha-subunit of a G protein. Some of RGS4 residues involved in the Galpha-RGS binding interface have similar orientations in GAIP (free form), indicating that upon binding these residues do not suffer conformational rearrangements, and therefore, their role does not seem to be restricted to Galpha interaction but also to RGS folding and stability. We suggest that other structural differences between the two proteins may be related to the process of binding as well as to a distinct efficiency in their respective GTPase activating function.

Dynamical behavior of the HIV-1 nucleocapsid protein.

The HIV-1 nucleocapsid protein (NC) contains two CCHC-type zinc knuckle domains that are essential for genome recognition, packaging and infectivity. The solution structure of the protein has been determined independently by three groups. Although the structures of the individual zinc knuckle domains are similar, two of the studies indicated that the knuckles behave as independently folded, non-interacting domains connected by a flexible tether, whereas one study revealed the presence of interknuckle NOE cross-peaks, which were interpreted in terms of a more compact structure in which the knuckles are in close proximity. We have collected multidimensional NMR data for the recombinant, isotopically labeled HIV-1 NC protein, and confirmed the presence of weak interknuckle NOEs. However, the NOE data are not consistent with a single protein conformation. 15N NMR relaxation studies reveal that the two zinc knuckle domains possess different effective rotational correlation times, indicating that the knuckles are not tumbling as a single globular domain. In addition, the 1H NMR chemical shifts of isolated zinc knuckle peptides are very similar to those of the intact protein. The combined results indicate that the interknuckle interactions, which involve the close approach of the side-chains of Phe16 and Trp37, are transitory. The solution behavior of NC may be best considered as a rapid equilibrium between conformations with weakly interacting and non-interacting knuckle domains. This inherent conformational flexibility may be functionally important, enabling adaptive binding of NC to different recognition elements within the HIV-1 psi-RNA packaging signal.

The use of dipolar couplings for determining the solution structure of rat apo-S100B(betabeta).

The relative orientations of adjacent structural elements without many well-defined NOE contacts between them are typically poorly defined in NMR structures. For apo-S100B(betabeta) and the structurally homologous protein calcyclin, the solution structures determined by conventional NMR exhibited considerable differences and made it impossible to draw unambiguous conclusions regarding the Ca2+-induced conformational change required for target protein binding. The structure of rat apo-S100B(betabeta) was recalculated using a large number of constraints derived from dipolar couplings that were measured in a dilute liquid crystalline phase. The dipolar couplings orient bond vectors relative to a single-axis system, and thereby remove much of the uncertainty in NOE-based structures. The structure of apo-S100B(betabeta) indicates a minimal change in the first, pseudo-EF-hand Ca2+ binding site, but a large reorientation of helix 3 in the second, classical EF-hand upon Ca2+ binding.

Refined solution structure and backbone dynamics of HIV-1 Nef.

The tendency of HIV-1 Nef to form aggregates in solution, particularly at pH values below 8, together with its large fraction of highly mobile residues seriously complicated determination of its three-dimensional structure, both for heteronuclear solution NMR (Grzesiek et al., 1996a, Nat Struct Biol 3:340-345) and for X-ray crystallography (Lee et al., 1996, Cell 85:931-942). Methods used to determine the Nef structure by NMR at pH 8 and 0.6 mM concentration are presented, together with a detailed description of Nef's secondary and tertiary structure. The described techniques have general applicability for the NMR structure determination of proteins that are aggregating and/or have limited stability at low pH values. Extensive chemical shift assignments are reported for backbone and side chain 1H, 13C, and 15N resonances of the HIV-1 Nef deletion mutants NEF delta 2-39, NEF delta 2-39, delta 159-173, and of NEF delta 2-39, delta 159-173 in complex with the SH3 domain of the Hck tyrosine protein kinase. Besides a type II polyproline helix, Nef's structure consists of three alpha-helices, a 3(10) helix, and a five-stranded anti-parallel beta-sheet. The analysis of 15N relaxation parameters of the backbone amide sites reveals that all the secondary structure elements are non-mobile on the picosecond to nanosecond and on the millisecond time scale. A large number of slowly exchanging amide protons provides evidence for the stability of the Nef core even on the time scale of hours. Significant internal motions on the ps to ns time scale are detected for residues 60 to 71 and for residues 149 to 180, which form solvent-exposed loops. The residues of the HIV-1 protease cleavage site (W57/L58) do not exhibit large amplitude motions on the sub-nanosecond time scale, and their side chains insert themselves into a hydrophobic crevice formed between the C-terminus of helix 1 and the N-terminus of helix 2. A refined structure has been determined based on additional constraints for side-chain and backbone dihedral angles derived from a large number of three-bond J-coupling and ROE data.

Protein backbone (15)N relaxation rates as a tool for the diagnosis of structure quality.

In the work reported herein we define a structure validation factor that depends on protein backbone (15)N relaxation rates. This is an alternative method to the previously defined quality factors derived from anisotropic chemical shifts or residual dipolar couplings. We have used the structure dependence of (15)N relaxation rates of anisotropically tumbling proteins to calculate this structure diagnosis factor and have used it to demonstrate the improvement of protein structures refined with residual dipolar couplings.

Direct structure refinement against residual dipolar couplings in the presence of rhombicity of unknown magnitude.

Residual dipolar couplings arising from small degrees of alignment of molecules in a magnetic field provide unique long-range structural information. The potential of this approach for structure refinement has recently been demonstrated for a protein-DNA complex in which the magnetic susceptibility tensor was axially symmetric. For most macromolecules and macromolecular complexes, however, axial symmetry cannot be assumed. Moreover, the presence of significant rhombicity will clearly affect the accuracy of the resulting coordinates. In this Communication we present a simple calculational strategy that makes use of simulated annealing refinement against the residual dipolar couplings in combination with a grid search, to simultaneously refine the structures and ascertain the magnitude of the axial and rhombic components of the tensor.

Direct refinement against proton-proton dipolar couplings in NMR structure determination of macromolecules.

The computational tools necessary for making use of (1)H-(1)H dipolar couplings in macromolecular structure refinement are presented. Potentials are described for direct refinement against (1)H-(1)H dipolar couplings of known sign as well as of unknown sign. In addition, a multiple potential is developed for prochiral protons whose stereospecific assignments are unknown. The utility of direct (1)H-(1)H dipolar coupling refinement is illustrated using the small protein ubiquitin. It is shown that direct (1)H-(1)H dipolar coupling refinement leads to improvements in the precision, accuracy, and quality of the resulting structures.

Measurement of dipolar couplings for methylene and methyl sites in weakly oriented macromolecules and their use in structure determination.

A simple and effective method is described for simultaneously measuring dipolar couplings for methine, methylene, and methyl groups in weakly oriented macromolecules. The method is a J-modulated 3D version of the well-known [1H-13C] CT-HSQC experiment, from which the J and dipolar information are most accurately extracted by using time-domain fitting in the third, constant-time dimension. For CH2-sites, the method generally yields only the sum of the two individual 13C-1H couplings. Structure calculations are carried out by minimizing the deviation between the measured sum, and the sum predicted for each methylene on the basis of the structure. For rapidly spinning methyl groups the dipolar contribution to the splitting of the outer 13C quartet components can be used directly to constrain the orientation of the C-CH3 bond. Measured sidechain dipolar couplings are in good agreement with an ensemble of NMR structures calculated without use of these couplings.

Structure and self assembly of a retrovirus (FeLV) proline rich neutralization domain.

The 60 amino acid proline-rich neutralization domain of the external surface unit glycoprotein of feline leukemia virus was chemically synthesized in total and in fragments. We examined the ability of these retroviral peptides to form ordered conformations using 1H-NMR, circular dichroism spectroscopy, and intrinsic viscosity measurements. One dimensional nuclear magnetic resonance spectroscopy revealed that the 60 amino acid peptide could form a stable, folded structure that was long-lived, as shown by the ability to protect amide-protons in D20. Peptides corresponding to the N-terminal 42, N-terminal 20 amino acids, and middle 20 amino acid sections could also form stable structures. The C-terminal segment did not protect any protons in D20. Interestingly, self assembly of the N-terminal 42 and C-terminal 16 amino acid peptides into a structure very close to that of the 60 amino acid domain was observed. The circular dichroism results reveals a large negative cotton effect at 198 nm that is characteristic of the proline-rich beta-turn helixes which consist predominantly of trans-proline. The intrinsic viscosity results suggest a non-random coil structure that is rod shaped. Our conclusion is that PRN60 forms a beta-turn helix and that this region of FeLV-gp70 is a separate folding domain of the retroviral surface unit glycoprotein. The unique conformational properties of PRN60 and its critical role as the predominant target for neutralizing antibody responses suggest that this peptide is a reasonable candidate for producing a synthetic peptide vaccine for FeLV.

High-resolution heteronuclear NMR of human ubiquitin in an aqueous liquid crystalline medium.

A mixture of dihexanoyl phosphatidylcholine and dimyristoyl phosphatidylcholine in water forms disc-shaped particles, often referred to as bicelles [Sanders and Schwonek (1992) Biochemistry, 31, 8898-8905]. These adopt an ordered, liquid crystalline phase, which can be maintained at very low concentrations of the bicelles (down to 3% w/v). At this concentration the spacing between individual bicelles, on average, exceeds 300 A. The bicelles are shown to have a negligible effect on the rotational diffusion of ubiquitin as judged by the 15N T1p values of the backbone amides relative to those in isotropic aqueous solution. The protein exhibits a residual degree of alignment which is proportional to the bicelle concentration, and approximately collinear with ubiquitin's rotational diffusion tensor. The degree of alignment obtained offers unique opportunities for studying the protein's structure and dynamics.

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.

Slow motions in oriented phospholipid bilayers and effects of cholesterol or gramicidin. A 19F-NMR T1 rho study.

In an extension of our earlier work (Peng, Z.-y., V. Simplaceanu, I. J. Lowe, and C. Ho. 1988. Biophys. J. 54:81-95), the rotating-frame nuclear spin-lattice relaxation (T1 rho) technique has been used to investigate the slow molecular motions (10(-4) - 10(-6) s) in lipid bilayers prepared from protonated or perdeuterated 19F-labeled phospholipids in the absence and presence of cholesterol or gramicidin as membrane-interacting molecules. Complications caused by the 19F-1H cross-polarization observed previously can be removed by the substitution of 2H for 1H in the acyl chains. Only a weak dependence of the T-1(1 rho) on the locking field strength is found for a phospholipid molecule with perdeuterated acyl chains, indicating that there are no slow motions with a single, well-defined correlation time between 5 x 10(-6) and 4 x 10(-5) s. However, the orientation dependences of the T-1(1 rho) can be well fitted by motional models with either one slow motion having an unspecified geometry or with a superposition of two specific types of slow motions. Cholesterol and gramicidin show distinct effects in altering either the geometry or the weighting of slow motions in phospholipid bilayers, as reflected by changes in the orientation dependence. These two additives also exhibit quite different label-position specificities. A qualitative understanding of the induced effects of cholesterol and gramicidin on the dynamics of phospholipid bilayers will be discussed.

Multidimensional 1H and 15N NMR investigation of glutamine-binding protein of Escherichia coli.

Specific and uniform 15N labelings along with site-directed mutagenesis of glutamine-binding protein have been utilized to obtain assignments of the His156, Trp32 and Trp220 residues. These assignments have been made not only to further study the importance of these 3 amino acid residues in protein-ligand and protein-protein interactions associated with the active transport of L-glutamine across the cytoplasmic membrane of Escherichia coli, but also to serve as the starting points in the sequence-specific backbone assignment. The assignment of H epsilon 2 of His156 refines the earlier model where this particular proton forms an intermolecular hydrogen bond to the delta-carbonyl of L-glutamine, while assignments of both Trp32 and Trp220 show the variation in local structures which ensure the specificity in ligand binding and protein-protein interaction. Using 3D NOESY-HMQC NMR, amide connectivities can be traced along 8-9 amino acid residues at a time. This paper illustrates the usefulness of combining 15N isotopic labeling and multinuclear, multidimensional NMR techniques for a structural investigation of a protein with a molecular weight of 25,000.