113CD-1H spin-spin couplings in homonuclear 1H correlated spectroscopy of metallothionein. Identification of the cysteine 1H spin systems

Zinc finger structure determination by NMR: Why zinc fingers can be a handful

Zinc fingers can be loosely defined as protein domains containing one or more tetrahedrally-co-ordinated zinc ions whose role is to stabilise the structure rather than to be involved in enzymatic chemistry; such zinc ions are often referred to as "structural zincs". Although structural zincs can occur in proteins of any size, they assume particular significance for very small protein domains, where they are often essential for maintaining a folded state. Such small structures, that sometimes have only marginal stability, can present particular difficulties in terms of sample preparation, handling and structure determination, and early on they gained a reputation for being resistant to crystallisation. As a result, NMR has played a more prominent role in structural studies of zinc finger proteins than it has for many other types of proteins. This review will present an overview of the particular issues that arise for structure determination of zinc fingers by NMR, and ways in which these may be addressed.

Brownian motion, spin diffusion and protein structure determination in solution

This paper presents my recollections on the development of protein structure determination by NMR in solution from 1968 to 1992. The key to success was to identify NMR-accessible parameters that unambiguously determine the spatial arrangement of polypeptide chains. Inspired by work with cyclopeptides, model considerations showed that enforcing short non-bonding interatomic distances imposes «ring closure conditions» on polypeptide chains. Given that distances are scalar parameters, this indicated an avenue for studies of proteins in solution, i.e., under the regime of stochastic rotational and translational motions at frequencies in the nanosecond range (Brownian motion), where sharp pictures could not be obtained by photography-related methods. Later-on, we used distance geometry calculations with sets of inter-atomic distances derived from protein crystal structures to confirm that measurements of short proton-proton distances could provide atomic-resolution structures of globular proteins. During the years 1976-1984 the following four lines of research then led to protein structure determination by NMR in solution. First, the development of NMR experiments enabling the use of the nuclear Overhauser effect (NOE) for measurements of interatomic distances between pairs of hydrogen atoms in proteins. Second, obtaining sequence-specific resonance assignment solved the "phase problem" for protein structure determination by NMR. Third, generating and programming novel distance geometry algorithms enabled the calculation of atomic-resolution protein structures from limited sets of distance constraints measured by NMR. Fourth, the introduction of two-dimensional NMR provided greatly improved spectral resolution of the complex spectra of proteins as well as efficient delineation of scalar and dipole-dipole ¹H-¹H connectivities, thus making protein structure determination in solution viable and attractive.

Lessons on the critical interplay between zinc binding and protein structure and dynamics

Zinc is one of the most important micronutrients for virtually all living organisms, and hence, it is important to understand the molecular mechanisms for its homeostasis. Besides proteins involved in transmembrane transport, both extra- and intracellular zinc-binding proteins play important roles in the respective metabolic networks. Important examples for extracellular zinc transporters are mammalian serum albumins, and for intracellular zinc handling, certain metallothioneins are of relevance. The availability of protein structures including relevant metal binding sites is a fundamental prerequisite to decipher the mechanisms that govern zinc binding dynamics in these proteins, but their determination can prove to be surprisingly challenging. Due to the spectroscopic silence of Zn(2+), combinations of biophysical techniques including electrospray ionisation mass spectrometry (ESI-MS) and multinuclear NMR, isothermal titration calorimetry (ITC) and extended X-ray absorption fine structure (EXAFS) spectroscopy, coupled with site-directed mutagenesis and molecular modelling have proven to be valuable approaches to understand not only the zinc-binding properties of metallothioneins and albumins, but also the influence of other physiologically relevant competing agents. These studies have demonstrated why the bacterial metallothionein SmtA contains a site inert towards exchange with Cd(2+), why the plant metallothionein EC from wheat is partially unfolded in the presence of Cd(2+), and how fatty acids impact on the zinc-binding ability of mammalian serum albumins.

Use of (113)Cd NMR to probe the native metal binding sites in metalloproteins: an overview

Our laboratories have actively published in this area for several years and the objective of this chapter is to present as comprehensive an overview as possible. Following a brief review of the basic principles associated with (113)Cd NMR methods, we will present the results from a thorough literature search for (113)Cd chemical shifts from metalloproteins. The updated (113)Cd chemical shift figure in this chapter will further illustrate the excellent correlation of the (113)Cd chemical shift with the nature of the coordinating ligands (N, O, S) and coordination number/geometry, reaffirming how this method can be used not only to identify the nature of the protein ligands in uncharacterized cases but also the dynamics at the metal binding site. Specific examples will be drawn from studies on alkaline phosphatase, Ca(2+) binding proteins, and metallothioneins.In the case of Escherichia coli alkaline phosphatase, a dimeric zinc metalloenzyme where a total of six metal ions (three per monomer) are involved directly or indirectly in providing the enzyme with maximal catalytic activity and structural stability, (113)Cd NMR, in conjunction with (13)C and (31)P NMR methods, were instrumental in separating out the function of each class of metal binding sites. Perhaps most importantly, these studies revealed the chemical basis for negative cooperativity that had been reported for this enzyme under metal deficient conditions. Also noteworthy was the fact that these NMR studies preceded the availability of the X-ray crystal structure.In the case of the calcium binding proteins, we will focus on two proteins: calbindin D(9k) and calmodulin. For calbindin D(9k) and its mutants, (113)Cd NMR has been useful both to follow actual changes in the metal binding sites and the cooperativity in the metal binding. Ligand binding to calmodulin has been studied extensively with (113)Cd NMR showing that the metal binding sites are not directly involved in the ligand binding. The (113)Cd chemical shifts are, however, exquisitely sensitive to minute changes in the metal ion environment.In the case of metallothionein, we will reflect upon how (113)Cd substitution and the establishment of specific Cd to Cys residue connectivity by proton-detected heteronuclear (1)H-(113)Cd multiple-quantum coherence methods (HMQC) was essential for the initial establishment of the 3D structure of metallothioneins, a protein family deficient in the regular secondary structural elements of α-helix and β-sheet and the first native protein identified with bound Cd. The (113)Cd NMR studies also enabled the characterization of the affinity of the individual sites for (113)Cd and, in competition experiments, for other divalent metal ions: Zn, Cu, and Hg.

Identification of zinc-ligated cysteine residues based on 13Calpha and 13Cbeta chemical shift data

Although a significant number of proteins include bound metals as part of their structure, the identification of amino acid residues coordinated to non-paramagnetic metals by NMR remains a challenge. Metal ligands can stabilize the native structure and/or play critical catalytic roles in the underlying biochemistry. An atom's chemical shift is exquisitely sensitive to its electronic environment. Chemical shift data can provide valuable insights into structural features, including metal ligation. In this study, we demonstrate that overlapped 13Cbeta chemical shift distributions of Zn-ligated and non-metal-ligated cysteine residues are largely resolved by the inclusion of the corresponding 13Calpha chemical shift information, together with secondary structural information. We demonstrate this with a bivariate distribution plot, and statistically with a multivariate analysis of variance (MANOVA) and hierarchical logistic regression analysis. Using 287 13Calpha/13Cbeta shift pairs from 79 proteins with known three-dimensional structures, including 86 13Calpha and 13Cbeta shifts for 43 Zn-ligated cysteine residues, along with corresponding oxidation state and secondary structure information, we have built a logistic regression model that distinguishes between oxidized cystines, reduced (non-metal ligated) cysteines, and Zn-ligated cysteines. Classifying cysteines/cystines with a statistical model incorporating all three phenomena resulted in a predictor of Zn ligation with a recall, precision and F-measure of 83.7%, and an accuracy of 95.1%. This model was applied in the analysis of Bacillus subtilis IscU, a protein involved in iron-sulfur cluster assembly. The model predicts that all three cysteines of IscU are metal ligands. We confirmed these results by (i) examining the effect of metal chelation on the NMR spectrum of IscU, and (ii) inductively coupled plasma mass spectrometry analysis. To gain further insight into the frequency of occurrence of non-cysteine Zn ligands, we analyzed the Protein Data Bank and found that 78% of the Zn ligands are histidine and cysteine (with nearly identical frequencies), and 18% are acidic residues aspartate and glutamate.

205Tl NMR Methods for the Characterization of Monovalent Cation Binding to Nucleic Acids

Monovalent cations play an important role in many biological functions. The guanine rich sequence, d(G4T4G4), requires monovalent cations for formation of the G-quadruplex, d(G4T4G4)2. This requirement can be satisfied by thallium (Tl+), a potassium (K+) surrogate. To verify that the structure of d(G4T4G4)2 in the presence of Tl+ is similar to the K+-form of the G-quadruplex, the solution structure of the Tl+-form of d(G4T4G4)2 was determined. The 10 lowest energy structures have an all atom RMSD of 0.76 +/- 0.16 A. Comparison of this structure to the identical G-quadruplex formed in the presence of K+ validates the isomorphous nature of Tl+ and K+. Using a 1H-205Tl spin-echo difference experiment we show that, in the Tl+-form of d(G4T4G4)2, small scalar couplings (<1 Hz) exist between 205Tl and protons in the G-quadruplex. These data comprise the first 1H-205Tl scalar couplings observed in a biological system and have the potential to provide important constraints for structure determination. These experiments can be applied to any system in which the substituted Tl+ cations are in slow exchange with the bulk ions in solution.

Sensitivity-enhanced double-TROSY experiment for simultaneous measurement of one-bond 15N-1H, 15N-13C' and two-bond 1H-13C' couplings

A recently published experiment for the measurement of 1JHN, 1JNC', and 2JHC' coupling constants [J. Am. Chem. Soc. 125 (2003) 11504] was modified to yield a double-TROSY experiment which selects 1 of the 16 multiplet components from a 15N-HSQC spectrum recorded of a uniformly 15N/13C-labelled protein. Subspectra containing any 1 of the 16 multiplet components can be generated allowing accurate coupling constant measurements. The experiment is sensitivity enhanced, turning all magnetization components precessing during the evolution time into observable magnetization during the detection time. The experiment is discussed with regard to the previously published alpha/beta-filtered HN(alpha/beta-NC'-J) experiment [J. Magn. Reson. 140 (1999), 32] which measures the same coupling constants.

Solution 1H, 15N NMR spectroscopic characterization of substrate-bound, cyanide-inhibited human heme oxygenase: water occupation of the distal cavity

A solution NMR spectroscopic study of the cyanide-inhibited, substrate-bound complex of uniformly (15)N-labeled human heme oxygenase, hHO, has led to characterization of the active site with respect to the nature and identity of strong hydrogen bonds and the occupation of ordered water molecules within both the hydrogen bonding network and an aromatic cluster on the distal side. [(1)H-(15)N]-HSQC spectra confirm the functionalities of several key donors in particularly robust H-bonds, and [(1)H-(15)N]HSQC-NOESY spectra lead to the identification of three additional robust H-bonds, as well as the detection of two more relatively strong H-bonds whose identities could not be established. The 3D NMR experiments provided only a modest, but important, extension of assignments because of the loss of key TOCSY cross-peaks due to the line broadening from a dynamic heterogeneity in the active site. Steady-state NOEs upon saturating the water signal locate nine ordered water molecules in the immediate vicinity of the H-bond donors, six of which are readily identified in the crystal structure. The additional three are positioned in available spaces to account for the observed NOEs. (15)N-filtered steady-state NOEs upon saturating the water resonances and (15)N-filtered NOESY spectra demonstrate significant negative NOEs between water molecules and the protons of five aromatic rings. Many of the NOEs can be rationalized by water molecules located in the crystal structure, but strong water NOEs, particularly to the rings of Phe47 and Trp96, demand the presence of at least an additional two immobilized water molecules near these rings. The H-bond network appears to function to order water molecules to provide stabilization for the hydroperoxy intermediate and to serve as a conduit to the active site for the nine protons required per HO turnover.

Interactions of bismuth complexes with metallothionein(II)

Bismuth complexes are widely used as anti-ulcer drugs and can significantly reduce the side effects of platinum anti-cancer drugs. Bismuth is known to induce the synthesis of metallothionein (MT) in the kidney, but there are few chemical studies on the interactions of bismuth complexes with metallothionein. Here we show that Bi(3+) binds strongly to metallothionein with a stoichiometry bismuth:MT = 7:1 (Bi(7)MT) and can readily displace Zn(2+) and Cd(2+). Bismuth is still bound to the protein even in strongly acidic solutions (pH 1). Reactions of bismuth citrate with MT are faster than those of [Bi(EDTA)](-), and both exhibit biphasic kinetics. (1)H NMR data show that Zn(2+) is displaced faster than Cd(2+), and that both Zn(2+) and Cd(2+) in the beta-domain (three metal cluster) of MT are displaced by Bi(3+) much faster than from the alpha-domain (four metal cluster). The extended x-ray absorption fine structure spectrum of Bi(7)MT is very similar to that for the glutathione and N-acetyl-L-cysteine complexes [Bi(GS)(3)] and [Bi(NAC)(3)] with an inner coordination sphere of three sulfur atoms and average Bi-S distances of 2.55 A. Some sites appear to contain additional short Bi-O bonds of 2.2 A and longer Bi-S bonds of 3.1 A. The Bi(3+) sites in Bi(7)MT are therefore highly distorted in comparison with those of Zn(2+) and Cd(2+).

NMR structure of the sea urchin (Strongylocentrotus purpuratus) metallothionein MTA

The three-dimensional structure of [(113)Cd7]-metallothionein-A (MTA) of the sea urchin Strongylocentrotus purpuratus was determined by homonuclear(1)H NMR experiments and heteronuclear [(1)H, (113)Cd]-correlation spectroscopy. MTA is composed of two globular domains, an N-terminal four-metal domain of the amino acid residues 1 to 36 and a Cd4Cys11cluster, and a C-terminal three-metal domain including the amino acid residues 37 to 65 and a Cd3Cys9cluster. The structure resembles the known mammalian and crustacean metallothioneins, but has a significantly different connectivity pattern of the Cys-metal co-ordination bonds and concomitantly contains novel local folds of some polypeptide backbone segments. These differences can be related to variations of the Cys sequence positions and thus emphasize the special role of the cysteine residues in defining the structure of metallothioneins, both on the level of the domain architecture and the topology of the metal-thiolate clusters.

Biphasic kinetics of Zn2+ removal from Zn metallothionein by nitrilotriacetate are associated with differential reactivity of the two metal clusters

We investigated the kinetics of nitrilotriacetate (NTA) extraction of Zn2+ from Zn7-metallothionein (MT) and a metal-hybrid derivative, Zn4Ag6MT, in which the Zn2+ and Ag+ ions occupy sites in the C-terminal alpha and N-terminal beta domains of the protein, respectively. Biphasic kinetics were observed for Zn7MT under pseudo-first-order conditions. Rate constants were (5.2 +/- 0.6) x 10(-3) and (1.0 +/- 0.3) x 10(-4)s-1 in 20 mM phosphate, 100 mM KF, pH 7.5 at 23 degrees C. In contrast, Zn4Ag6MT showed a single kinetic step with a rate constant of (2.9 +/- 0.4) x 10(-3)s-1. These results indicate that the biphasic reactivity of Zn7MT stems from differential susceptibility of the metal in the two metal-thiolate clusters to removal by competing ligands, with Zn2+ in the more stable alpha-domain cluster reacting faster than that in the less stable beta-domain cluster. Such behavior suggests that the structures of the two domains of mammalian MT may have evolved to assure that Cu binding does not compromise the structural characteristics that allow Zn to be rapidly transferred from MT to essential cellular ligands.

Calculations of one-, two- and three-bond nuclear spin-spin couplings in a model peptide and correlations with experimental data

We present ab initio calculations of the Fermi contact term and experimental correlations of six coupling constants, 3JHNH alpha, 1JC alpha H alpha, 2JC'H alpha, 1JC alpha N, 2JC alpha N and 1JC'N, in a peptide as functions of the backbone dihedral angles, phi and psi. Given estimates of experimental uncertainties, we find semiquantitative experimental correlations for 3JHNH alpha, 1JC alpha N and 2JC alpha N, qualitative correlations for 1JC alpha H alpha and 2JC'H alpha, but no experimental correlations of practical utility for 1JC'N, owing to its complex dependence on at least four dihedral angles. Errors in the estimation of dihedral angles from X-ray crystallographic data for proteins, which result from uncertainties in atom-to-atom distances, place substantial limitations on the quantitative reliability of coupling constant calculations fitted to such data. In the accompanying paper [Edison, A.S. et al., J. Biomol. NMR, 4, 543-551] we apply the results of the coupling constant calculations presented here to the estimation of phi and psi angles in staphylococcal nuclease from experimental coupling constants.

Estimates of phi and psi torsion angles in proteins from one-, two- and three-bond nuclear spin-spin couplings: application to staphylococcal nuclease

Calculated coupling constants (3JHNH alpha, 1JC alpha H alpha, 2JC'H alpha, 1JC alpha N and 2JC alpha N) from our accompanying paper [Edison, A.S. et al. (1994) J. Biomol. NMR, 4, 519-542] have been used to generate error surfaces that can provide estimates of the phi and psi angles in proteins. We have used experimental coupling data [3JHNH alpha: Kay, L.E. et al. (1989) J. Am. Chem. Soc., 111, 5488-5490; 1JC alpha H alpha: Vuister, G. W. et al. (1993) J. Biomol. NMR, 3, 67-80; 2JC'H alpha: Vuister, G.W. and Bax, A. (1992) J. Biomol. NMR, 2, 401-405; 1JC alpha N and 2JC alpha N: Delaglio, F. et al. (1991) J. Biomol. NMR, 1, 439-446] to create error surfaces for selected residues of the protein staphylococcal nuclease. The residues were chosen to include all those with five experimental couplings, as well as some with four experimental couplings, to demonstrate the relative importance of 3JHNH alpha and 1JC alpha H alpha. For most of the cases, we obtained good agreement between the X-ray structure [Loll, P.J. and Lattman, E.E. (1989) Protein Struct. Funct. Genet., 5, 183-201] and the NMR data.

Comparison of the solution conformations of human [Zn7]-metallothionein-2 and [Cd7]-metallothionein-2 using nuclear magnetic resonance spectroscopy

The solution structure of native human [Zn7]-metallothionein-2 has been compared with the previously determined structure of human [Cd7]-metallothionein-2. The comparison was based on complete sequence-specific 1H nuclear magnetic resonance assignments for human [Zn7]-metallothionein-2 obtained using the sequential assignment method. The secondary structure was found to be very similar in the [Zn7]- and [Cd7]- forms of the protein. Only seven amide protons in [Zn7]- metallothionein-2 were found to have exchange rates lower than approximately 0.2 min-1 at pH 7.0 and 10 degrees C, which corresponds closely to the results of amide proton exchange studies with the [Cd7]- form of the protein. Finally, the 1H-1H distance constraints determined from nuclear Overhauser enhancement spectroscopy for human [Zn7]-metallothionein-2 were checked for compatibility with the [Cd7]-metallothionein-2 structure. Overall, although no direct method is available for identifying the metal-polypeptide co-ordinative bonds in the Zn(2+)-containing protein, these measurements provided several independent lines of evidence showing that the [Zn7]- and [Cd7]- forms of human metallothionein-2 have the same molecular architecture.

Three-dimensional structure of human [113Cd7]metallothionein-2 in solution determined by nuclear magnetic resonance spectroscopy

The three-dimensional structure of human [113Cd7]metallothionein-2 was determined by nuclear magnetic resonance spectroscopy in solution. Sequence-specific 1H resonance assignments were obtained using the sequential assignment method. The input for the structure calculations consisted of the metal-cysteine co-ordinative bonds identified with heteronuclear correlation spectroscopy, 1H-1H distance constraints from nuclear Overhauser enhancement spectroscopy, and spin-spin coupling constants 3JHN alpha and 3J alpha beta. The molecule consists of two domains, the beta-domain including amino acid residues 1 to 30 and three metal ions, and the alpha-domain including residues 31 to 61 and four metal ions. The nuclear magnetic resonance data present no evidence for a preferred relative orientation of the two domains. The polypeptide-to-metal co-ordinative bonds in human metallothionein-2 are identical to those in the previously determined solution structures of rat metallothionein-2 and rabbit metallothionein-2a, and the polypeptide conformations in the three proteins are also closely similar.

GAL4 transcription factor is not a "zinc finger" but forms a Zn(II)2Cys6 binuclear cluster

The DNA-binding domain of the transcription factor GAL4, consisting of the 62 N-terminal residues and denoted GAL4(62*), contains a Cys-Xaa2-Cys-Xaa6-Cys-Xaa6-Cys-Xaa2-Cys-Xaa6+ ++-Cys motif, which has been shown previously to bind two Zn(II) or Cd(II) ions. Binding of Zn(II) or Cd(II) is essential for the recognition by GAL4 of the specific palindromic DNA sequence to which it binds upstream of genes for galactose-metabolizing enzymes, the UASG sequence. On the basis of the 113Cd NMR chemical shifts of the two bound 113Cd(II) ions, we propose a binuclear cluster model for this Zn(II)-binding subdomain. 1H-113Cd heteronuclear multiple-quantum NMR spectroscopy and phase-sensitive double-quantum filtered 1H correlation spectroscopy of the 112Cd(II)- and 113Cd(II)-substituted GAL4(62*) derivatives provide direct evidence that the two bound 113Cd(II) ions are coordinated only by the six cysteine residues, two of which form bridging ligands between the two 113Cd(II) ions. The latter can be identified from the pattern of 1H-113Cd J coupling. Thus a binuclear metal ion cluster rather than a "zinc finger" is formed by the six cysteine residues of the GAL4 DNA-binding domain. This model can be directly applied to eight other fungal transcription factors which have been shown to contain similarly spaced Cys6 clusters. 1H NMR spectra of apo-GAL4(62*) suggest conformational fluctuation of the metal-binding subdomain upon removal of Zn(II) or Cd(II). Both Cd(II)2- and Zn(II)2-containing species of GAL4 can be formed, and the similar 1H NMR spectra suggest similar conformations.

Heteronuclear filters in two-dimensional [1H,1H]-NMR spectroscopy: combined use with isotope labelling for studies of macromolecular conformation and intermolecular interactions

The use of heteronuclear filters enables the editing of complex 1H nuclear magnetic resonance (NMR) spectra into simplified subspectra containing a lesser number of resonance lines, which are then more easily amenable to detailed spectral analysis. This editing is based on the creation of heteronuclear two-spin or multiple-spin coherence and discrimination between protons that do or do not participate in these heteronuclear coherences. In principle, heteronuclear editing can be used in conjunction with one-dimensional or multidimensional 1H-NMR experiments for studies of a wide variety of low-molecular-weight compounds or macromolecular systems, and is implicitely applied in a wide range of heteronuclear NMR experiments with proton detection (e.g. Bax et al. 1983; Griffey & Redfield, 1987). In the present article we shall focus on the use of heteronuclear filters in two-dimensional (2D) [1H, 1H]-NMR experiments. The selection of the material covered was primarily motivated by its impact on the practice of protein structure determination in solution, and on NMR studies of intermolecular interactions with biological macromolecules. Section 2 surveys potential applications of heteronuclear filters in this area. The remainder of the article is devoted to an introduction of the theoretical principles used in heteronuclear filters, and to a detailed description of the experimental implementation of these measurements. In writing the review we tried to minimize redundancy with the recent article in Quarterly Review of Biophysics by Griffey & Redfield (1987) and to concentrate on experiments that were introduced during the period 1986–9.

Complete structure of the polysaccharide from Streptococcus sanguis J22

The cell wall polysaccharides of certain oral streptococci such as Streptococcus sanguis strains 34 and J22, although immunologically distinct, act as receptors for the fimbrial lectins of Actinomyces viscosus T14V. We report the complete covalent structure of the polysaccharide from S. sanguis J22 which is composed of a heptasaccharide subunit linked by phosphodiester bonds. The repeating subunit, which contains alpha-GalNAc, alpha-rhamnose, beta-rhamnose, beta-glucose, and beta-galactose all in the pyranoside form and beta-galactofuranose, is compared with the previously published structure of the polysaccharide from strain 34. The structure has been determined almost exclusively by high-resolution nuclear magnetic resonance methods. The 1H and 13C NMR spectra of the polysaccharides from both strains 34 and J22 have been completely assigned. The stereochemistry of pyranosides was assigned from JH-H values determined from phase-sensitive COSY spectra, and acetamido sugars were assigned by correlation of the resonances of the amide 1H with the sugar ring protons. The 13C spectra were assigned by 1H-detected multiple-quantum correlation (HMQC) spectra, and the assignments were confirmed by 1H-detected multiple-bond correlation (HMBC) spectra. The positions of the glycosidic linkages were assigned by detection of three-bond 1H-13C correlation across the glycosidic linkage in the HMBC spectra. The positions of the phosphodiester linkages were determined by splittings observed in the 13C resonances due to 31P coupling and also by 1H-detected 31P correlation spectroscopy.

Conformation of [Cd7]-metallothionein-2 from rat liver in aqueous solution determined by nuclear magnetic resonance spectroscopy

The three-dimensional structure of [Cd7]-metallothionein-2 from rat liver was determined in aqueous solution, using nuclear magnetic resonance spectrometry and distance geometry calculations. The experimental data provided proton-proton distance constraints from measurements of nuclear Overhauser effects, constraints on the geometry of the metal-cysteine clusters determined by heteronuclear correlation spectroscopy, and dihedral angle constraints derived from both coupling constants and nuclear Overhauser effects. The structure calculations were performed with the program DISMAN. As in previous studies with rabbit liver metallothionein-2a, the structure calculations were performed separately for the alpha and beta-domains containing the 4 and 3-metal clusters, respectively, since no interdomain constraints were found. For both domains, the global polypeptide fold, the location of polypeptide secondary structure elements, the architecture of the metal-sulfur cluster and the local chirality of the metal co-ordination are very similar to the solution structure of rabbit metallothionein-2a, but show considerable difference relative to the crystal structure of rat metallothionein-2.

Three-dimensional structure of rabbit liver [Cd7]metallothionein-2a in aqueous solution determined by nuclear magnetic resonance

In previous work the metal-polypeptide co-ordinative bonds in the major protein species of a reconstituted [113Cd7]metallothionein-2 preparation from rabbit liver in aqueous solution were determined, the secondary polypeptide structure was found to contain several half-turns and 3(10)-helical segments, and a preliminary characterization of the overall polypeptide backbone fold in the beta-domain containing the three-metal cluster, and the alpha-domain containing the four-metal cluster, was obtained. Using a new, more extensive set of nuclear magnetic resonance data these earlier structures were improved by new structure calculations. The new experimental data consist of distance constraints from measurements of nuclear Overhauser effects, and dihedral angle constraints derived from both coupling constants and nuclear Overhauser effects. The structure calculations were performed with the program DISMAN. Since no information on the orientation of the two domains relative to each other could be obtained, the structure calculations were performed separately for the alpha-domain and the beta-domain. The average of the pairwise root-mean-square distances among the 20 structures with the least residual violations of input constraints was 2.9 A for the beta-domain and 1.4 A for the alpha-domain (1 A = 0.1 nm). The overall chirality of the polypeptide fold is right-handed for the beta-domain and left-handed for the alpha-domain. For each of the seven metal ions the local chirality of the co-ordination of the four cysteinyl Sy atoms is clearly defined. The improved structures of both domains show the previously noted differences relative to the recently published crystal structure of metallothionein-2a from rat liver.

Sequence-specific 1H-NMR assignments in rat-liver metallothionein-2

As a basis for the determination of the metal-coordination topology and the three-dimensional fold of the polypeptide chain, sequence-specific assignments were obtained for the 1H nuclear magnetic resonance (NMR) spectrum of metallothionein-2 from rat liver. The 1H spin systems of the 20 metal-bound cysteines were identified from comparison of two metal-homogeneous protein preparations obtained by reconstitution of the apometallothionein-2 with 113Cd2+ and 112Cd2+, respectively. The identification of the spin systems for the remaining amino acid residues and sequential assignments were then obtained with two-dimensional 1H-NMR experiments at 500 MHz. The assignments are complete except for two backbone amide protons, which were not observed, and the side-chain hydrogen atoms beyond beta CH2 for all eight lysines. The chemical shifts are presented at pH 7.0 and 25 degrees C.

Chemistry and biochemistry of metallothionein

A wealth of chemical, spectroscopic and structural data attest to the uniqueness of the metallothioneins as a group of novel bioinorganic structures. Their earmarking feature is the arrangement of "soft" metal ions in complexes with cysteine side chains to form discrete metal-thiolate clusters. In this review an account is given of the chemical characteristics of the 52 metallothioneins whose primary structures are now known completely or in part. Also included is an up-to-date summary of the spectroscopic properties and of the spatial structure models derived from X-ray diffraction crystallographic analysis and from two-dimensional nuclear magnetic resonance spectroscopy.

Sequence-specific 1H-NMR assignments in rabbit-liver metallothionein-2

The complete sequence-specific assignment of the 1H nuclear magnetic resonance spectrum of a major subform of rabbit liver metallothionein-2 is presented. The sequential assignment procedures revealed a number of differences with regard to results obtained by earlier partial chemical sequencing of a preparation now known to be microheterogeneous. In particular, the present data indicate a polypeptide chain length of 62 amino acid residues as compared to the occurrence of 61 amino acids in all other known mammalian metallothioneins. In the new sequence, which was also fully confirmed by chemical means, the additional amino acid residue was identified as Ala8' inserted between Ala8 and Ala9 of the standard amino acid numeration. In addition to the predominant protein species all preparations contained a minor component, for which the two-dimensional 1H-nuclear magnetic resonance features are compatible with a chemically different, homologous metallothionein.

The reaction of metallothionein with mercuribenzoate. A dialysis and 113Cd-n.m.r. study

Reaction of rat liver cadmium-metallothionein-II(Cd-MT-II) with p-hydroxymercuribenzoate(pHOHgBzO-) causes displacement of bound Cd. When pHOHgBzO- -induced displacement of 109Cd is observed after dialysis of the reaction mixture, the stoichiometry is consistent with stepwise displacement of tetraco-ordinate Cd atoms by non-random entry of reagent into the polynuclear clusters. 113Cd n.m.r. allows direct observation of the effects on bound Cd of stepwise titration of 113Cd-MT-II with pHOHgBzO-. The first equivalent reduces all resonances approximately equally. Subsequently differential reactivity of the protein thiolates towards the reagent gives rise to differential decreases in the 113Cd signal intensities. Resonances previously attributed to a three-metal cluster are lost before those arising from the four-metal cluster. These results are interpreted in terms of current models of the MT structure. They are distinct from the results of reaction of MT with 5,5'-dithiobis-(2-nitrobenzoic acid), which distinguishes between only two classes of thiolates, terminal and bridging. Such different patterns of reactivity of the protein thiolates may underlie a biological activity of this protein.

Crystal structure of Cd,Zn metallothionein

The anomalous scattering data from five Cd in the native protein were used to determine the crystal structure of cadmium, zinc (Cd,Zn) metallothionein isoform II from rat liver. The structure of a 4-Cd cluster was solved by direct methods. A 2.3 A resolution electron density map was calculated by iterative single-wavelength anomalous scattering. The structure is folded into two domains. The amino terminal domain (beta) of residues 1 to 29 enfolds a three-metal cluster of one Cd and two Zn atoms coordinated by six terminal cysteine thiolate ligands and three bridging cysteine thiolates. The carboxyl terminal domain (alpha) of residues 30 to 61 enfolds a 4-Cd cluster coordinated by six terminal and five bridging cysteine thiolates. All seven metal sites have tetrahedral coordination geometry. The domains are roughly spherical, and the diameter is 15 to 20 A; there is limited contact between domains. The folding of alpha and beta is topologically similar but with opposite chirality. Redundant, short cysteine-containing sequences have similar roles in cluster formation in both alpha and beta.

Polypeptide fold in the two metal clusters of metallothionein-2 by nuclear magnetic resonance in solution

The solution conformation of rabbit liver Cd27+-metallothionein-2 was determined by nuclear magnetic resonance (n.m.r.) and distance geometry. The n.m.r. data are based on complete sequence-specific resonance assignments for the polypeptide chain. This letter describes the global arrangement of the polypeptide chain, which forms two distinct domains containing metal clusters of three and four Cd ions, respectively.

Systematic application of high-resolution, phase-sensitive two-dimensional 1H-NMR techniques for the identification of the amino-acid-proton spin systems in proteins. Rabbit metallothionein-2

Novel strategies for elucidation and classification of amino acid 1H-NMR spin systems in proteins were developed exploiting recently introduced two-dimensional NMR techniques such as phase-sensitive double-quantum-filtered correlated spectroscopy, relayed coherence transfer spectroscopy, double quantum spectroscopy and nuclear Overhauser spectroscopy. Due to the improved resolution in phase-sensitive spectra, the fine structure of cross peaks could be exploited as a powerful source of information for establishing 1H-1H connectivities. Principles for the interpretation of multiplet structures of absorption mode cross peaks are discussed. With these methods the 1H spin systems of rabbit liver metallothionein-2 were elucidated and classified according to amino acid types. Despite the intrinsically difficult situation arising from the unusual amino acid composition of this protein, a more complete characterization of the 1H spin systems prior to the step of sequential resonance assignments was achieved with the presently introduced methodology than was possible in earlier studies of proteins of similar size.