Structure determination of biological macromolecules in solution using nuclear magnetic resonance spectroscopy

A detailed understanding of the function of a biological macromolecule requires knowledge of its three-dimensional structure. Most atomic-resolution structures of biological macromolecules have been solved either by X-ray diffraction in single crystals or by nuclear magnetic resonance (NMR) in solution. This review surveys the method of NMR structure determination. First, a brief introduction to NMR and its basic concepts is presented. The main part of the article deals with the individual steps necessary for an NMR structure determination. At the end, the discussion turns to considerations on the influence of the molecular size of the macromolecules on the structure determination by NMR. New techniques are discussed that greatly enhance the possibilities of applying NMR to large molecular systems.     

High-resolution nuclear magnetic resonance studies of proteins

The combination of advanced high-resolution nuclear magnetic resonance (NMR) techniques with high-pressure capability represents a powerful experimental tool in studies of protein folding. This review is organized as follows: after a general introduction of high-pressure, high-resolution NMR spectroscopy of proteins, the experimental part deals with instrumentation. The main section of the review is devoted to NMR studies of reversible pressure unfolding of proteins with special emphasis on pressure-assisted cold denaturation and the detection of folding intermediates. Recent studies investigating local perturbations in proteins and the experiments following the effects of point mutations on pressure stability of proteins are also discussed. Ribonuclease A, lysozyme, ubiquitin, apomyoglobin, alpha-lactalbumin and troponin C were the model proteins investigated.     

Structural studies of MS2 bacteriophage virus particle disassembly by nuclear magnetic resonance relaxation measurements

In this article we studied, by nuclear magnetic resonance relaxation measurements, the disassembly of a virus particle-the MS2 bacteriophage. MS2 is one of the single-stranded RNA bacteriophages that infect Escherichia coli. At pH 4.5, the phage turns to a metastable state, as is indicated by an increase in the observed nuclear magnetic resonance signal intensity upon decreasing the pH from 7.0 to 4.5. Steady-state fluorescence and circular dichroism spectra at pH 4.5 show that the difference in conformation and secondary structure is not pronounced if compared with the phage at pH 7.0. At pH 4.5, two-dimensional (15)N-(1)H heteronuclear multiple quantum coherence (HMQC) spectrum shows approximately 40 crosspeaks, corresponding to the most mobile residues of MS2 coat protein at pH 4.5. The (15)N linewidth is approximately 30 Hz, which is consistent with an intermediate with a rotational relaxation time of 100 ns. The average spin lattice relaxation time (T(1)) of the mobile residues was measured at different temperatures, clearly distinguishing between the dimer and the equilibrium intermediate. The results show, for the first time, the presence of intermediates in the process of dissociation of the MS2 bacteriophage.     

Pulsed nuclear magnetic resonance studies in sickled erythrocytes

Pulsed proton magnetic resonance studies were made to determine T₁ (spin-lattice) and T₂ (spin-spin) relaxation times for normal and sickled blood cells in the oxygenated and deoxygenated states as a function of temperature between 0°C and 37°C. It was found that the sickling phenomenon affects T₂ values but not T₁ values in these cells. These results are ascribed to a proton-exchange mechanism. The functional relationship found between T₂ and temperature in the sickled cells is similar to that between the formation of hydrophobic bonds and temperature proposed by Nalbandian and Scheraga.     

High resolution nuclear magnetic resonance studies of nigeran

Polymer motion in solution can be studied by ¹³CNMR relaxation methods, which provide information about the correlation time for C-H vectors. ¹³C-Relaxation and Nuclear Overhauser Enhancement (NOE) data may frequently be combined to determine the dipole-dipole relaxation contribution. An alternative method is proposed based on a comparison of the proton spin-lattice relaxation rates of the centre proton resonances of an unlabelled molecule with the relaxation rates of the ¹³C satellites (from ¹³C labelled molecules).Selectively labelled nigeran which is an alternating $\text{1 → 3 and 1 → 4 α-}\text{d}\text{-glucan}$ has been investigated. The discussion in terms of the occurrence of different motions for each of the two units of the polymer requires an unambiguous assignment of the two anomeric carbons. For this reason a detailed assignment of the ¹H and ¹³C Nuclear Magnetic Resonance (NMR) spectra of nigeran in dimethylsulphoxide-d₆ is described, based on T₁ and NOE measurements in addition to selective homonuclear and heteronuclear spin decoupling experiments. These values are correlated with a conformation estimated by HSEA hard-spheres calculation. The measurements of the relaxation parameters for labelled and unlabelled compounds which provide an alternative determination of the ¹³C-¹H dipole-dipole relaxation contribution in a macromolecule agree well with ¹³C-{¹H} NOE experiments.     

Broad-line nuclear magnetic resonance studies of chloroperoxidase.

Chloroperoxidase, a heme glycoprotein isolated from the mold Caldariomyces fumago, was studied by NMR relaxation techniques. Interaction of the chloride ion substrate with the enzyme may be analyzed as consisting of at least three contributions: a weak interaction with the iron atom, nonspecific anion-protein interactions, and a specific interaction generated at low pH. The data indicate that a specific interaction, which develops in parallel with enzyme activity at low pH, does not occur at the iron atom first coordination sphere site. The results are summarized in terms of an enzymatic mechanism not involving chloride ion coordination to the iron atom.     

Carbon-13 nuclear magnetic resonance studies of cobalamins

The carbon-13 nuclear magnetic resonance spectra of aquocobalamin, adenosylcobalamin, methylcobalamin, and (carboxymethyl)cobalamin have been interpreted. The assignments were made by a comparison of the spectra with that of cyanocobalamin, by a study of the pH dependence of the chemical shifts, by an analysis of the effect of the axial ligands on the carbon atoms of the corrin ring, and by a study of the specific line broadening effect of the paramagnetic ions Mn2+ and Gd3+. The chemical shift changes that accompany the "base-on"----"base-off" conversion of the organocobalamins demonstrate that the conformation of the "western" half of the corrin ring and the conformations of the a, b, c, d, f, and g side chains are relatively constant. In contrast, the conformations of the "eastern" half of the corrin ring and the e propionamide side chain are highly variable.     

Interstitial sodium nuclear magnetic resonance relaxation times in perfused hearts.

Separation of intracellular and extracellular sodium nuclear magnetic resonance (NMR) signals would enable nondestructive monitoring of intracellular sodium. It has been proposed that differences between the relaxation times of intracellular and extracellular sodium be used either directly or indirectly to separate the signal from each compartment. However, whereas intracellular sodium relaxation times have been characterized for some systems, these times were unknown for interstitial sodium. In this study, the interstitial sodium NMR relaxation times have been measured in perfused frog and rat hearts under control conditions. This was achieved by eliminating the NMR signal from the extracardiac (perfusate) sodium, and then quantifying the remaining cardiac signal. The intracellular signal was measured to be 8% (frog) or 22% (rat) of the cardiac signal and its subtraction was found to have a negligible effect on the cardiac relaxation times. Therefore this cardiac signal is considered to provide a good estimate of interstitial relaxation behavior. For perfused frog (rat) hearts under control conditions, this signal was found to have a T1 of 31.6 +/- 3.0 ms (27.3 +/- 1.6 ms) and a biexponential T2 of 1.9 +/- 1.0 ms (2.1 +/- 0.3 ms) and 25.2 +/- 1.3 ms (26.3 +/- 3.2 ms). Due to the methods used to separate cardiac signal from perfusate signal, it is possible that this characterized only a part of the signal from the interstitium. The short T2 component attributable to the interstitial signal indicates that separation of the NMR signals from each compartment on the basis of relaxation times alone may be difficult.     

Conformational studies of N-Tyr-MIF-1 in aqueous solution by 1H nuclear magnetic resonance spectroscopy.

N-Tyr-MIF-1 (Tyr-Pro-Leu-Gly-NH2) is an endogenous brain peptide with multiple effects on animal behavior. However, there have been no studies on the conformation of this tetrapeptide. In this report, we studied the conformation of N-Tyr-MIF-1 in aqueous solution by conventional one-dimensional and two-dimensional (COSY and NOESY) 1H nuclear magnetic resonance spectroscopy at 300 MHz. A complete set of assignments for the resolved resonances and approximate assignments for the overlapping resonances were made. The results demonstrate that N-Tyr-MIF-1 is in slow exchange between two conformers, most likely determined by the cis and trans states of the proline residue. The minor conformation represents 30 +/- 3% of the population over the temperature range from 3 degrees to 73 degrees. In the major conformation, the tyrosine aromatic ring appears to be close enough to interact directly with the proline pyrrolidine ring, as indicated by a strong temperature dependence of the proline C beta H, C delta H and C delta H' chemical shifts. In contrast, this interaction of the tyrosine and proline rings is not present in the minor conformation.     

Conformation of alamethicin in oriented phospholipid bilayers determined by (15)N solid-state nuclear magnetic resonance

The conformation of the 20-residue antibiotic ionophore alamethicin in macroscopically oriented phospholipid bilayers has been studied using (15)N solid-state nuclear magnetic resonance (NMR) spectroscopy in combination with molecular modeling and molecular dynamics simulations. Differently (15)N-labeled variants of alamethicin and an analog with three of the alpha-amino-isobutyric acid residues replaced by alanines have been investigated to establish experimental structural constraints and determine the orientation of alamethicin in hydrated phospholipid (dimyristoylphosphatidylcholine) bilayers and to investigate the potential for a major kink in the region of the central Pro(14) residue. From the anisotropic (15)N chemical shifts and (1)H-(15)N dipolar couplings determined for alamethicin with (15)N-labeling on the Ala(6), Val(9), and Val(15) residues and incorporated into phospholipid bilayer with a peptide:lipid molar ratio of 1:8, we deduce that alamethicin has a largely linear alpha-helical structure spanning the membrane with the molecular axis tilted by 10-20 degrees relative to the bilayer normal. In particular, we find compatibility with a straight alpha-helix tilted by 17 degrees and a slightly kinked molecular dynamics structure tilted by 11 degrees relative to the bilayer normal. In contrast, the structural constraints derived by solid-state NMR appear not to be compatible with any of several model structures crossing the membrane with vanishing tilt angle or the earlier reported x-ray diffraction structure (Fox and Richards, Nature. 300:325-330, 1982). The solid-state NMR-compatible structures may support the formation of a left-handed and parallel multimeric ion channel.     

Surface plasmon resonance and nuclear magnetic resonance studies of ABAD-Abeta interaction

Abeta binding alcohol dehydrogenase (ABAD) is an NAD-dependent mitochondrial dehydrogenase. The binding between ABAD and Abeta is likely a direct link between Abeta and mitochondrial toxicity in Alzheimer's disease. In this study, surface plasmon resonance (SPR) was employed to determine the temperature dependence of the affinity of the ABAD-Abeta interaction. A van't Hoff analysis revealed that the ABAD-Abeta association is driven by a favorable entropic change (DeltaS = 300 +/- 30 J mol-1 K-1) which overcomes an unfavorable enthalpy change (DeltaH = 49 +/- 7 kJ/mol). Therefore, hydrophobic interactions and changes in protein dynamics are the dominant driving forces of the ABAD-Abeta interaction. This is the first dissection of the entropic and enthalpic contribution to the energetics of a protein-protein interaction involving Abeta. SPR confirmed the conformational changes in the ABAD-Abeta complex after Abeta binding, consistent with differences seen in the crystal structures of free ABAD and the ABAD-Abeta complex. Saturation transfer difference (STD) NMR experiments directly and unambiguously demonstrated the inhibitory effect of Abeta on the ABAD-NAD interaction. Conversely, NAD inhibits the Abeta-ABAD interaction. Binding of Abeta and binding of NAD to ABAD are likely mutually exclusive. Thus, Abeta binding induces conformational and subsequently functional changes in ABAD, which may have a role in the mechanism of Abeta toxicity in Alzheimer's disease.     

Solid-state nuclear magnetic resonance relaxation studies of the interaction mechanism of antimicrobial peptides with phospholipid bilayer membranes

An 18-residue peptide, KWGAKIKIGAKIKIGAKI-NH(2) was designed to form amphiphilic beta-sheet structures when bound to lipid bilayers. The peptide possesses high antimicrobial activity when compared to naturally occurring linear antimicrobial peptides, most of which adopt an amphipathic alpha-helical conformation upon binding to the lipids. The perturbation of the bilayer by the peptide was studied by static (31)P and (2)H solid-state NMR spectroscopy using POPC and POPG/POPC (3/1) bilayer membranes with sn-1 chain perdeuterated POPC and POPG as the isotopic labels. (31)P NMR powder spectra exhibited two components for POPG/POPC bilayers upon addition of the peptide but only a slight change in the line shape for POPC bilayers, indicating that the peptide selectively disrupted the membrane structure consisting of POPG lipids. (2)H NMR powder spectra indicated a reduction in the lipid chain order for POPC bilayers and no significant change in the ordering for POPG/POPC bilayers upon association of the peptide with the bilayers, suggesting that the peptide acts as a surface peptide in POPG/POPC bilayers. Relaxation rates are more sensitive to the motions of the membranes over a large range of time scales. Longer (31)P longitudinal relaxation times for both POPG and POPC in the presence of the peptide indicated a direct interaction between the peptide and the POPG/POPC bilayer membranes. (31)P longitudinal relaxation studies also suggested that the peptide prefers to interact with the POPG phospholipids. However, inversion-recovery (2)H NMR spectroscopic experiments demonstrated a change in the relaxation rate of the lipid acyl chains for both the POPC membranes and the POPG/POPC membranes upon interaction with the peptide. Transverse relaxation studies indicated an increase in the spectral density of the collective membrane motion caused by the interaction between the peptide and the POPG/POPC membrane. The experimental results demonstrate significant dynamic changes in the membrane in the presence of the antimicrobial peptide and support a carpet mechanism for the disruption of the membranes by the antimicrobial peptide.     

Quantitative in vivo nuclear magnetic resonance studies of hybridoma metabolism

Carbon-13 nuclear magnetic resonance (NMR) spectroscopy was used to study the metabolism of a murine hybridoma cell line at two feed glutamine concentrations, 4.0 and 1.7 mM. Carbon-13 labeling patterns were used in conjunction with nutrient uptake rates to calculate the metabolic fluxes through the glycolytic pathway, the pentose shunt, the malate shunt, lipid biosynthesis, and the tricarboxylic acid (TCA) cycle. Decreasing the feed glutamine concentration significantly decreased glutamine uptake but had little effect on glucose metabolism. A significant incrase in antibody productivity occurred upon decreasing the feed glutamine level. The increased antibody productivity in concert with decreased glutamine uptake and no apparent change in glucolytic metabolism suggests that antibody production was not energy limited. Metabolic flux calculations indicate that (1) approximately 92% of the glucose consumed proceeds directly through glycolysis with 8% channeled through the pentose shunt; (2) lipid biosynthesis appears to be greater than malate shunt activity; and (3) considerable exchange occurs between TCA cycle intermediates and amino acid metabolic pools, leading to substantial loss of (13)C label from the TCA cycle. These results illustrate that (13)NMR spectroscopy is a powerfulf tool in the calculation of metabolic fluxes, particularly for exchange pathways where no net flux occurs. (c) 1994 John Wiley & Sons, Inc.     

High-resolution proton nuclear magnetic resonance studies of human gastrin

High-resolution 1H nuclear magnetic resonance (NMR) spectroscopy at 300 MHz has been used to study the behavior of human gastrin in aqueous solution. A large number of resonances have been assigned by analysis of one- and two-dimensional NMR spectra and the effects of pH and by comparison with the spectrum of des-less than Glu1-gastrin. In gastrin, the ratio of cis to trans conformations around the Gly-2 to Pro-3 peptide bond is 3:7. This is reflected in splitting of the resonances of several neighboring residues and of a residue distant in the sequence, Tyr-12. The pKa of Tyr-12 is 10.7. Sulfation of this residue perturbs the resonances of Tyr-12 and Gly-13 but has very little effect on the rest of the spectrum. A study of the temperature dependence shows that several perturbed resonances move toward their expected positions as the temperature is raised but with a linear dependence on temperature, consistent with a redistribution of populations among accessible local conformations rather than a cooperative conformational change. Addition of Na+ or Ca2+ causes only minor changes in the spectrum. The paramagnetic metal ion Co2+ produces a number of spectral changes, reflecting strong binding to at least one site involving the Glu residues and weaker binding to Asp-16.     

Proton nuclear magnetic resonance studies of mast cell histamine

The state of histamine in mast cells was studied by 1H NMR spectroscopy. Spectra were measured for histamine in situ in intact mast cells, for histamine in suspensions of mast cell granule matrices that had been stripped of their membranes, and for histamine in solutions of heparin. The 1H NMR spectrum of intact mast cells is relatively simple, consisting predominantly of resonances for intracellular histamine superimposed on a weaker background of resonances from heparin and proteins of the cells. All of the intracellular histamine contributes to the NMR signals, indicating it must be relatively mobile and not rigidly associated with the negatively charged granule matrix. Spectra for intracellular histamine and for histamine in granule matrices are similar, indicating the latter to be a reasonable model for the in situ situation. The dynamics of binding of histamine by granule matrices and by heparin are considerably different; exchange of histamine between the bulk water and the granule matrices is slow on the 1H NMR time scale, whereas exchange between the free and bound forms in heparin solution is fast. The chemical shifts of resonances for histamine in mast cells are pH dependent, decreasing as the intragranule pH increases without splitting or broadening. The results are interpreted to indicate that histamine in mast cells is relatively labile, with rapid exchange between bound histamine and pools of free histamine in water compartments confined in the granule matrix.     

1H nuclear magnetic resonance double resonance study of oxytocin in aqueous solution.

Peptide NH resonances in the 250 MHZ 1H nuclear magnetic resonance (NMR) spectrum of oxytocin in H2O were assigned to specific amino acid residues by the "underwater decoupling" technique (i.e., decoupling from corresponding CalphaH resonances, which are buried beneath the intense water peak). These experiments confirm previous assignments of A. I. Brewster an V. J. Hruby ((1973), Proc. Natl. Acad. Sci. U.S.A. 70, 3806) and A. F. Bradbury et al. ((1974), FEBS Lett. 42, 179). Three methods of assigning NH resonances of peptides--solvent titration, underwater decoupling, and isotopic labeling--are compared. As the solvet composition is gradually changed from dimethyl sulfoxide to H2O, oxytocin undergoes a conformational change at 70-90 mol % of H2O. Exposure to solvent of specific hydrogens of oxytocin in H2O was studied by monitoring intensity changes of solute resonances when the solvent peak was saturated. Positive nuclear Overhauser effects (NOE's) of 14 +/- 5 were observed for the Tyr ortho CH and meta CH resonances, respectively. Comparative studies with deamino-oxytocin indicate that these effects result predominantly from intermolecular dipoledipole interaction between aromatic side chain CH protons and protons of the solvent. The NOE's therefore indicate intimate contact between water and the aromatic CH hydrogens of the Tyr side chain. The extent of saturation transferred by proton exchange between water and NH group varies with Ph in a manner which appears to reflect the acid-base catalysis of the protolysis reaction. There is no indication that any NH protons are substantially shiedled from the solvent.