eIF4E, the mRNA cap binding protein, is a master switch that controls eukaryotic translation. To be active, it must bind eIF4G and form the eIF4F complex, which also contains eIF4A. Translation is downregulated by association of eIF4E with 4E-BP, which occupies the eIF4G binding site. Signaling events acting on 4E-BP cause it to dissociate from eIF4E, and eIF4E is then free to bind eIF4G to form active eIF4F complex. We have solved the structure of the yeast eIF4E/m7Gpp complex in a CHAPS micelle. We determined the position of the first nucleotide of the RNA in a complex with m7GpppA, and identified the 4E-BP binding site. eIF4E has a curved eight-stranded antiparallel beta-sheet, decorated with three helices on the convex face and three small helices inserted in connecting loops. The m7G of the cap structure is intercalated into a stack of triptophans in the concave face. The 4E-BP binding site is located in a region encompassing one edge of the beta-sheet, the adjacent helix a2 and several regions of non-regular secondary structure. It is adjacent to, but does not overlap the cap-binding site.
The amino-terminal domain of the alpha subunit of E.coli RNA polymerase (aNTD) plays a key role in assembly of the core enzyme, which consists of alpha2-beta-betaprime. aNTD is composed of 239 amino acids (26kDa protein) and forms a dimer. This protein aggregates non-specifically at an NMR concentration resulting in a poor spectrum. We examined the solution condition using 1H-15N HSQC spectra, and found that detergents which have steroid ring structure like CHAPS, sodium deoxycholate and so on were effective for disaggregating protein. The solution condition of NMR sample was optimized by monitoring translational diffusion using a PFG NMR experiment. The optimized condition was 0.6mM aNTD 20mM Tris pH8.5 0.1mM EDTA 7mM deoxycholate. aNTD is large 52kDa protein as a dimer. The pH of solution is high (8.5) and the concentration of the protein is only 0.6mM. And we performed HNCA, HN(CO)CA, CT-HNCA, CT-HN(CO)CA, HN(CA)CB, HN(COCA)CB experiments with a deuterated sample. In spite of these disadvantages we achieved 80% of assignments of aNTD backbone nuclei. The rest of resonance was lost because amid proton exchange was very fast at a high pH condition. The secondary structure of aNTD was determined by chemical shift deviations of Ca, Cb and CO resonances from those of random coil. It was found that aNTD contains three alpha helices and two beta sheets. Crucial residues for subunit assembly identified by mutagenesis experiments are located on the three alpha helices, suggesting they comprise interface for the subunit assembly.
The 1H, 13C and 15N NMR chemical shift assignments of a cloned catalytic fragment (12kDa) of gamma delta resolvase were undertaken to investigate the structure and catalytic mechanism of this site-specific DNA recombinase. The 1H-15N HSQC spectra showed broad peaks due to apparent conformational heterogeneity at low salt conditions while under higher salt conditions, a more structured conformation was observed. High quality 3D 15N and 13C edited NOESY data, HCCH-TOCSY and 3D triple resonance data [HNCA, HN(CO)CA, HNCACB and HN(CO)CACB] have been recorded. The backbone connectivities were completed for all residues except 64 to 73. The results indicate the domain is well-folded consisting of alpha and beta secondary structure as well as regions of less well-ordered structure. One such loop region is found to participate in phosphate contacts as determined from a phosphate ion binding study. This region would appear to represent the catalytic active site.
Intestinal fatty acid-binding protein (I-FABP) is a cytosolic 15.1 kDa protein that appears to function in the intracellular transport and metabolic trafficking of fatty acids. It binds a single molecule of long-chain fatty acid in an enclosed cavity surrounded by two five-stranded antiparallel -sheets and a helix-turn-helix domain. To investigate the role of the helical domain, we engineered and characterized a variant of I-FABP by deleting 17 contiguous residues and inserting a Ser-Gly linker (Kim K, Cistola DP, Frieden C. 1996. Biochemistry 35: 7553-7558). This variant, termed d-17-SG, was remarkably stable, exhibited a high -sheet content and was able to bind fatty acids with some features characteristic of the wild-type protein. In the present study, we determined the structure of the d-17-SG/palmitate complex at atomic resolution using triple-resonance 3D NMR methods. Sequence-specific 1H, 13C, and 15N resonance assignments were established at pH 7.2 and 25C and used to define the consensus 1H/13C chemical shift-derived secondary structure. Subsequently, an iterative protocol was used to identify 2,544 NOE-derived interproton distance restraints and to calculate its tertiary structure using a unique distance geometry/simulated annealing algorithm. In spite of the sizable deletion, the d-17-SG structure exhibits a backbone conformation that is nearly superimposable with the -sheet domain of the wild-type protein. The selective deletion of the helical domain creates a very large opening that connects the interior ligand-binding cavity with exterior solvent. Unlike wild-type I-FABP, fatty acid dissociation from d-17-SG is structurally and kinetically unimpeded, and a protein conformational transition is not required. The d-17-SG variant of I-FABP is the only wild-type or engineered member of the intracellular lipid-binding protein family whose structure lacks helices. Thus, d-17-SG I-FABP consititues a unique model system for investigating the role of the helical domain in ligand protein recognition, protein stability and folding, lipid transfer mechanisms and cellular function.
The hematopoietic cellular kinase (Hck) is a Src-family protein tyrosine kinase that is involved in intracellular signal transduction. Recent X-ray structures of Hck and Src reveal that intramolecular interaction between the SH3 domain and a linker region that joins the SH2 domain to the kinase domain are important for the negative regulation of the kinase. We have developed a novel construct consisting of the Hck SH3 domain and the proline-rich region of the GTPase activating protein (GAP), referred to as [SH3Pro], in order to study the structural and dynamical properties of intramolecular association. Hck SH3 binds this region of GAP with low affinity in a bimolecular fashion but binds this region tightly as a unimolecular construct. We have shown using immunoblotting experiments that this construct has reduced affinity for exogenous substrates. NMR diffusion measurements reveal that the protein behaves as a monomer of similar molecular weight as the SH3 domain alone. We have prepared a 13C, 15N dually enriched sample of [SH3Pro] and have acquired a variety of triple-resonance NMR data sets including HNCO, HNCA, CBCACONH, CBCANH, HCCH COSY and HCCH-TOCSY. The resonance assignments of Hck SH3Pro are being compared to those recently reported for Hck SH3 (Horita et al., In Press). Evidence for intramolecular binding and the nature of the binding surface based on 3D and 4D, 15N- and 13C-edited NOE experiments will be presented.
Chemokines (chemotactic cytokines) are key elements to the communication process within the immune system. They belong to a superfamily of small dimeric proteins (8-10 kDa per subunit) with four cysteines in very similar positions. The superfamily has been divided into subfamilies based on the relative position of the first two cysteines. One subfamily, named CXC-chemokines, has an amino acid insertion between the two cysteines. They function as neutrophil activators and their genes are coded on chromosome 4. The other major subfamily (CC-chemokines) activates monocytes and lymphocytes and are coded on chromosome 17. Recently discovered stromal cell-derived factor-1 (SDF-1) is a CXC-chemokine by cysteine positioning, however, its complete sequence has about equal similarity between both subfamilies. Unlike other chemokines, the SDF-1 gene is located on chromosome 10, the protein is expressed in many different tissues, and it is active on monocytes and lymphocytes but not active on neutrophils. A receptor for SDF-1 has been identified as CXCR-4 and found to be a necessary coreceptor for viral entry of T cell line-tropic (T-tropic) HIV-1. As such, SDF-1 blocks the entry of T-tropic HIV-1 into the helper T lymphocytes. Thus far, all of the chemokine monomers have very similar secondary and tertiary structures with 20-40% sequence identity across the whole superfamily. However, the quaternary structure of the dimer is dramatically different between the CXC- and CC-subfamilies. We are presently determining the solution structure of SDF-1, by heteronuclear NMR, which may lend insight into its' function and clues for new drug therapy in treating HIV-1 infection.
NMR Studies of the Ethylene-Responsive DNA Binding Protein *Nagawa, Yoshinobu; Takagi, Masaru; Torigoe, Hidetaka; Nemoto, Tadashi; Nakanishi, Hiroshi; National Institute of Bioscience and Human-Technology (NIBH) and The Institute of Physical and Chemical Research (RIKEN) , Tsukuba, Japan The ethylene-responsive element binding proteins (EREBPs) interact with the GCC box which is an 11-bp sequence (TAAGAGCCGCC) conserved in the 5' upstream region of ethylene-inducible pathogenesis-related protein genes in plants. The DNA binding domain of EREBPs is identified within a region of 59 amino acid residues. The solution structure of EREBP-2(88-164), which include the DNA binding domain, has been studied by a variety of homonuclear and heteronuclear NMR experiments. The combination of 3D 15N-edited TOCSY, 15N-edited NOESY, HNCA, HN(CO)CA, HNCO, and HN(CA)CO spectra were employed to obtain sequential assignments for the amino acid residues of the protein. The secondary structural elements in EREBP-2(88-164) were identified based on NOE information and chemical shift index. EREBP-2(88-164) contains four short b-strands (extending from residues 91-94, 100-103, 109-116 and 124-129) and one relatively long a-helix (132-146). Detailed structure and a view of the interactions between the protein and DNA will be presented.
Putidaredoxin (Pdx) is a paramagnetic, 106-residue protein whose major, diamagnetic part has been analyzed previously by 1H NMR methods and molecular modeling1. However, the paramagnetic region of the oxidized and reduced forms of Pdx is not well characterized by 1H detected NMR techniques because rapid relaxation of the protons in this region causes their NMR signals to be extremely broad and overlapped, thus inhibiting the use of heteronuclear shift correlation for spectral assignment. Recently, we have demonstrated that the direct observation of 15N NMR spectra of Pdx and the measurement of 15N T1 values offer a useful probe of the paramagnetic zone of this protein2. This zone has now been studied in greater detail by directly observed 15N NMR and chemical shift/temperature dependence studies of oxidized and reduced Pdx preparations in which the six cysteine residues are selectively labeled with 15N. Four of these cysteines (C39, C45, C48, and C86) are coordinated to the two iron atoms, one (C73) is outside the paramagnetic zone, and the remaining one (C85) is adjacent to one of the coordinating cysteines. 15N T1 values have been measured for oxidized Cys6-15N-Pdx, because of the ease of describing the electronic and paramagnetic environment around the two Fe3+ ions. By use of an average spin number for the iron-sulfur cluster2 of 1.67 together with the 15N T1 values, we have calculated reduced distances of the cysteine backbone 15N nuclei from the center of the cluster. These distances are consistent with those calculated from X-ray crystal structure data for five other ferredoxins, and confirm the structural similarity of the Fe2S2 clusters in Pdx and these ferredoxins. 1. Thomas C. Pochapsky, Xiao Mei Ye, Gayathri Ratnaswamy, and Teresa A. Lyons, Biochemistry 33, 6424-6432(1994). 2. Bruce Coxon, Nese Sari, Marcia J. Holden, and Vincent L. Vilker, Magn. Reson. Chem. 35, 741-751 (1997).
Ral guanine nucleotide dissociation stimulator (RalGDS) represents a guanine exchange factor for Ral, a Ras-related GTPase, whose biological role is not clear. The C-terminal domain, comprising the residues 681 to 787, has been found to be the prototype member for a new Ras interacting domain family, called RA (for Ras associating). This domain type has been shown to interact directly with Ras proteins, often specifically with only a subset of Ras or Ras-related GTPases. Since RA-domains are often coupled with other Ras-interacting domains, like CDC25, the new domain seems to be involved in the downstream signalling of the Ras-cascade. We have investugated into the solution structure of the RA-domain of human RalGDS. The structure is similar to the Ras-binding domain (RBD) of the serine/threonine kinase c-Raf1, despite that the sequence homology is low and the domain length is different. The new Ras-binding domain exhibits an ubiquitin-like fold, comprising a central helix and a five-stranded mixed beta-sheet. In order to map the binding determinants quantitatively we have mutated a set of residues involved in the putative binding interface and measured the affinity alterations of the mutants by surface plasmon resonance spectroscopy.
NMR Structure Studies of Alzheimer's Plaque Forming Protein R.E. Botto, D.M. Gregory, T.L. Benzinger-Smith#, D.G. Lynn+ and S.C. Meredith# Chemistry Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, IL 60439, and Departments of Chemistry+ and Pathology#, University of Chicago, Chicago, IL 60637 Alzheimers Disease is characterized histopathologically by the deposition of amyloid plaques (or fibrils), especially in areas of neuron loss. The primary component of these plaques is a 40-43 amino acid peptide called b-amyloid (Ab), which spans parts of the transmembrane and extracellular domains. The amyloid peptides are found in solution in the cerebrospinal fluid. The degree of clinical dementia and the neuron loss at autopsy correlate well with the number of amyloid plaques. Our structural goals are directed ultimately towards solving the conformation of Ab and the inter-strand contacts in amyloid plaques. We have implemented the DRAWS method for measuring interatomic distances between 13C- 13C spin pairs in several, mono- and di-labeled Ab synthons. Molecular modelling calculations using distance information from DRAWS and x-ray data on peptides indicate that amyloid fibril is comprised of Ab peptides which adopt a b-strand structure, and upon fibrilization individual peptides are organized in anti-parallel fashion. The advantage of the approach we are taking is that inter-strand distance measurements will allow us to construct a comprehensive solid-state structure of the peptide in the plaque, piece by piece. We would then be in a position to determine the locus and ultimately, the shape of turns and other structural features of the peptide which might appear in the plaque. These studies are now in progress. _________________ *Work performed under the auspices of the Office of Basic Energy Sciences, Division of Chemical Sciences, U. S. Department of Energy, under Contract # W-31-109-ENG-38.
Members of the intracellular lipid-binding protein family exhibit differences in specificity for fatty acids, retinoids and sterols (bile salts). This is particularly true of the lipid-binding proteins from the small intestine. The global folds of these proteins are nearly identical, so the determinants of lipid specificity most likely reside in the composition of key side-chains lining the lipid-binding cavity and entry portal, and in the dynamic properties of the molecule. We are investigating the determinants of specificity in the bile salt binding protein, human ileal lipid-binding protein (I-LBP). Initially, the three-dimensional solution structure of unliganded (apo) form of I-LBP is being determined by NMR. Sequence-specific 1H, 13C, and 15N resonance assignments were established at pH 6.3 and 25 degrees using a series of gradient enhanced 3-D pulse sequences. The proton and carbon assignments were used to determine the consensus chemical shift derived secondary structure (1). This secondary structure exhibits ten beta-strands and suprisingly, two full alpha helices. The NMR structure and dynamic properties of a related protein, apo intestinal fatty acid binding protein (I-FABP), shows local disorder and increased flexibility in the second helix which is mechanistically tied to ligand binding (2). The secondary structure of apo I-LBP also lacks a loop observed in the structures of other related proteins. The chemical shift derived secondary structure along with 3-D and 4-D NOESY data are being used in the determination of the tertiary structure of human apo I-LBP in solution. The results will be compared with those of other related proteins (2,3). (This work was supported by NIH grants DK48046 and DK09758) 1. Wishart, D.S., & Sykes, B.D. (1994) J. Biomol. NMR 4, 171-180. 2. Hodsdon, M.E., & Cistola, D.P. (1997) Biochemistry 36, 1450-1460. 3. Lucke, C., Zhang, F., Ruterjans, H., Hamilton, J.A., & Sacchettini, J.C. (1996) Structure 4, 785-800.
To elucide the mode of the receptor binding domain (RBD) of human alpha-2-macroglobulin with its receptor, the low density lipoprotien receptor-related protein(LRP), we have used 3D multinuclear NMR spectroscopy to determine the secondary structure of RBD. Assignment of the bcakbone NMR resonances of RBD was made by using C-13/N-15 and N-15 enriched RBD expressed in E. Coli. The Secondary structure was determined using H -1 and C-13 chemical shift indices and intrachain nuclear Overhauser enhancements. The secondary structure consisted of 8 strands beta-sheet and a short helix. The two lysines identified as being critical for receptor binding are located in and immediately adjacent to the alpha helix. This region interacts specifically with repeats from LRP and the low density lipoprotein receptor.
Enzyme I is the first enzyme involved in the phosphotransferase system of bacteria and is used to transport a variety of carbon sources into bacteria. Structural studies on the 259 amino terminal domain of Enzyme I, EIN, complexed with the 85 amino acid HPr will be discussed. The biological function of Enzyme I is to transfer a phosphorous to HPr which also occurs with EIN. Backbone and sidechain assignments were carried out using several perdeuterated and non-deuterated samples for EIN and EIN:HPr. Assignments for the complex, EIN:HPr, were carried out separately for the EIN and HPr subunits by using unlabelled HPr during the assignment of EIN and using unlabelled EIN during the assignment of HPr. The determination of the structure of the complex EIN:HPr will be presented. Previous work has shown that the binding site of HPr on EIN is localized in the alpha domain and involves helices H1, H2, H4 and the hinge region between H2 and H2'. The present work will show the details of the interaction between EIN and HPr.
Sequential dNN NOEs, chemical shift indices, 3JNa coupling constants and amide NH temperature coefficients reveal that the 73-residue full-length human p53 transcriptional activation domain (TAD) in the unbound state has, in water, two amphiphatic alpha-helices. This result is in clear contrast to popular "unstructured" models of TADs and in fact represents the first direct observation in solution of the alpha-helical structures hypothesized for TADs [Giniger, E. and Ptashne, J. (1987) Nature 330, 670] . The first helix encompassing the residues 19F-26L is a stable, highly amphipathic helix while the second, formed by the residues 48D-55T, is a transient helix. Amphipathic helices observed for the unbound p53 TAD suggest that formation of the transcriptional preinitiation complexes between the p53 TAD and its target proteins does not require the proposed induced fit mechanism. Two amphipathic helices we have deciphered in the p53 TAD provide a reasonable explanation for the observed synergism in transcriptional activity and may serve as a general structural motif in other non-proline, non-glutamine type TADs.
A variety of chemically and structurally distinct DNA lesions are removed from the eukaryotic genome by the nucleotide excision repair (NER) pathway. XPA (31 kDa) is essential to NER and is known to interact with damaged DNA and other proteins (RPA, ERCC1, and TFIIH) to repair the damage. To understand the structural basis for XPAs roles in NER, the solution structure of the minimal DNA binding domain of human XPA (XPA-MBD: M98-F219) was determined using NMR spectroscopy. The results represent the most detailed structural insights to date of a class IV zinc dependent DNA binding protein not bound to its DNA substrate. An elongated structure with two distinct regions, a zinc associated core (D101-K137) and an elastic domain (L138-F219), was observed. In the zinc associated core, the four Cys sulfydryl groups, C105, C108, C126 and C128, are coordinated to zinc in an S configuration. Two of the Cys residues are near the loop of an antiparallel -sheet between Y102-C105 and K110-M113 while the other two Cys residues form part of an alpha helix from C126-K137. The -sheet and the helix are separated by a loop-like structure. The zinc binding core is succeeded immediately by the elastic domain which contains a triple strand anti-parallel -sheet (L138-T140, L182-M178, and K163 K168), three unordered regions (K151-L162, N169-D177, and Q208-F219) and three helices (K141-L150, K183-W194, and Q197-R207). Two helices (K141-L150 and K183-W194) fold over the -sheet. The terminal 26 residues of XPA-MBD, which are essential for binding to damaged DNA, form an helix (Q197-R207) followed by an unordered region (Q208-F219). On the basis of the solution structure determined for XPA-MBD, a model for XPA:DNA interactions is proposed involving two distinct structural moieties; a zinc associated core for general binding to double stranded DNA and a flexible domain that may preferentially bind to single-stranded DNA associated with DNA lesions.
The solution structure of the recombinant wild-type, doubly isotope labelled monomer hemoglobin from the erythrocytes of the marine annelid Glycera dibranchiata has been achieved. The unusual protein is a 16 kD heme protein that functions in oxygen transport and storage, and unlike mammalian heme-globins shows a distinct proclivity towards auto-reduction. The overall solution structure of the polypeptide chain is highly similar to that of the many characterized myoglobins, however, two unusual features are probably significant with respect to altered function. The first if these is that the distal histidine, which is normally found in mammalian heme-globins, is replaced by Leucine. This obviates any structural stabilization by H-bonding from this distal residue. The second of these is the unusual appearance of a Phe at primary sequence position 31 (B10), also in the heme ligand binding pocket. Also unusual in comparison to sperm whale myoglobin, are the main chain dynamics as documented by nitrogen relaxation times, and the H/D isotope exchange dynamics. Taken together these measurements define a molecule structurally related to myoglobin, but dynamically different.
Abstract: A major problem when studying medium sized and larger paramagnetic metallo-proteins by proton NMR is the lack of spectral dispersion caused by the high molecular mass and the comparatively larger resonance linewidths. For heme proteins like cytochrome c peroxidase (CcP; molecular weight 35 kD) and the Glycera dibranchiata monomer hemoglobin (GMH4; molecular weight 16 kD) the combination of high proton resonance density and linewidths of up to 800-1000 Hz at half height require special methods for obtaining information on the proteins' active sites. We have taken two approaches based on molecular biology that are simple and effective ways to simplify the study of these proteins by proton NMR spectroscopy. A main method is perdeuteration achieved by expressing the apo-proteins in a highly deuterated growth medium. A key feature of this technology is the possibility to selectively study fully protonated, or even isotopically labelled substrates of heme enzymes. This poster will present our preliminary NMR work evaluating perdeuteration of GMH4, CcP, and parallel proton NMR studies on active-site mutants of CcP.
The n-methyl-D-aspartate (NMDA) and acetylcholine receptors (AchR) are ion channel forming proteins that transiently form pores through membranes and are necessary for rapid chemo-electric transduction of cellular signals. Information from biochemical and mutagenesis experiments has converged to form detailed overviews of these receptor proteins. However, atomic resolution models have yet to be reported. We are using solution and solid-state NMR to determine the structures of functional membrane segments of the NMDA and AchR proteins in membrane environments. The sequences corresponding to the proposed pore lining segments (M2) of the receptors have been expressed in E. Coli as fusion proteins and purified using traditional chromatography techniques. 15N and 13C labeled peptides have been incorporated into micelles for solution NMR structure determination. Uniformly 15N labeled peptides have also been incorporated into lipid bilayers and uniaxially oriented between glass plates for solid-state NMR structure determination. Solution NMR structures of the M2 peptides will be shown. Through the use of high resolution, multidimensional solid-state NMR experiments a structure of the AchR M2 in planar lipid bilayers has been determined and is compared to the calculated solution NMR structure. The results are discussed in the context of recently proposed receptor protein structural models.
X-linked agammaglobulinemia (XLA) is an inheritable disease; the patients fail to mount antibody response and consequently suffer from recurrent, sometimes lethal, bacterial infections. Point mutation in the gene encoding a non-receptor tyrosine kinase, Bruton's tyrosine kinase (BTK), leading to deletion of 14 aa residues of BTK SH3 domain, was found in a patient family. BTK interacts with several proteins during signal transduction. Absence of functional BTK leads to failure of B cell differentiation. We therefore studied the binding with ligands of a 58-residue peptide (aa 216-273) corresponding to SH3 domain of BTK and a 44 residue peptide (216-259) lacking the C-terminal 14 aa residues of BTK SH3 domain (truncated SH3 domain). The latter peptide represents the direct factor that causes XLA. Based on CD and NMR studies, BTK SH3 domain was found to adopt a b-barrel conformation, whereas the truncated BTK SH3 domain fails to fold properly and exists as a random coil. BTK TH domain binds to SH3 domains of various proteins. We found that truncated SH3 domain binds to proline rich peptides of BTK TH domain less avidly than does normal SH3 domain. Truncation in our case and mutation in vivo thus weakens binding between BTK SH3 and TH domains. Furthermore, we found that binding between truncated SH3 domain and proline rich peptide of p120cbl was considerably weaker than with normal SH3 domain. Truncated SH3 domain exhibits weak binding, probably because it fails to present to the ligand crucial residues in the correct context. Weaker binding of BTK SH3-TH domains suggests that the mutation-deletion makes for greater availability of both domains even in the unactivated state-an event that would for normal BTK, occur upon arrival of the signal. This disruption may impair the signaling cascade, resulting, eventually, in XLA. Our results delineate the importance of C-terminal in and ligand binding of SH3 domains and indicate also that improper folding and the altered binding behavior of mutant BTK SH3 domain likely causes XLA.
Human Replication Protein A (RPA) binds with high affinity to single-stranded DNA and interacts with multiple proteins that are involved in DNA metabolism processes including DNA repair, replication and recombination. RPA forms a heterotrimeric complex composed of subunits of 70-, 32- and 14-kDa. The N-terminal 18 kDa domain of the 70 kDa subunit exists as a structurally distinct domain. This fragment consisting of 169 residues does not affect the single-stranded DNA-binding activity, but stimulates DNA polymerase alpha activity and interacts with the tumor suppressor p53, transcription factors and SV40 large T antigen. The solution structure reveals that 115 residues of the N-terminal domain are embedded in a structure. The remaining 54 residues appear to be unstructured and flexible in solution thus representing a linker sequence to the DNA-binding domain of RPA70. Nevertheless, there is evidence that these random-coiled residues become structured when connected to the single-stranded binding domain of the 70 kDa subunit of RPA. The backbone- (1H, 15N, 13CO, alpha-13C) and the beta-13C-resonances of the structural region were almost entirely assigned in a sequence specific manner using three- and four-dimensional (3D and 4D) triple resonance heteronuclear NMR methods. The sequential NOE connectivities, 3J(HN-Halpha) coupling constants, amide proton exchange measurements as well as alpha-13C- and beta-13C chemical shift data indicate the presence of a 5-stranded beta-sheet and three 7-9 residues long alpha-helices. The beta-sheet in which four strands are broken by bulges is surrounded by two alpha-helices. An antiparallel 3-stranded beta-sheet is connected parallel via an alpha-helix to an antiparallel 2-stranded beta-sheet.
Low-temperature solid-state NMR spectroscopic techniques are being developed (77K
Staphylococcal protein A is a Staphylococcus aureus cell surface pathogenicity factor which contains five extracellular, homologous IgG-binding domains designated E, D, A, B and C that mediate binding to both the Fc fragment of IgG and to many IgG Fab fragments. We have chosen to study the complex between the E-domain of protein A and the Fv portion of a humanized anti-HER2 antibody (4D5-8). The structures of both members of the complex are known (Starovasnik et al. 1996, Biochemistry 48 15558-69; Eigenbrot et al. 1992, J. Mol. Biol. 229 969-995). A detailed structural model of the Fv-protein A domain E interface is of considerable interest both from the point of view of molecular recognition and because bacteria that produce Ig-binding proteins are known to act as mitogens in lymphocytes. Mutagenesis data suggest that only the VH portion of the Fv fragment is involved in E-domain binding. The VL and VH chains of this fragment were therefore expressed in seperate E. coli fermentations; the VL chain was perdeuterated and the VH chain was either 15N / 13C-labeled or 15N / 13C / 2H-labeled (deuteration to a level of ~90%). Deuteration of the VH portion of the Fv fragment resulted in greatly reduced linewidths and increased sensitivity in backbone correlation experiments performed on the fragment and will be essential in obtaining complete resonance assignments of the 32kD E-domain / Fv complex. 1H, 15N and 13C resonance assignments of the VH portion of the free Fv fragment are presented along with evidence for dynamic processes in the core of the VH portion of the free Fv fragment.
Vascular endothelial growth factor (VEGF) is an endothelial cell-specific mitogen and is a potent angiogenic and vascular permeabilizing factor. It is also an important mediator of pathological angiogenesis associated with cancer, rheumatoid arthritis, and proliferative retinopathy. The binding of VEGF to its receptors is modulated by both cell-surface-associated heparin-like glycosaminoglycans and exogenous heparin. Heparin binding to VEGF165 has been localized to the C-terminal 55 residues. The structure of the 55-residue heparin-binding domain of VEGF165 has been solved using data from 2D and 3D heteronuclear NMR spectroscopy. The structure has two sub-domains, each containing two disulfide bridges. There is limited secondary structure, although each sub-domain has a short two-stranded antiparallel beta-sheet; the C-terminal sub-domain also contains a short alpha-helix. Hydrophobic interactions are limited to side chains packing against the disulfide bridges. The heparin-binding domain of VEGF165 has no sequence or structural similarity to any known heparin-binding protein and thus represents a novel heparin-binding structure. Most of the positively charged amino acid side chains are localized on one side of the C-terminal sub-domain or on an adjacent disordered loop in the N-terminal sub-domain. The observed distribution of surface charges suggests that these residues constitute a heparin-interaction site.
The solution structure of murine macrophage inflammatory protein-2 (MIP-2), a heparin-binding chemokine secreted by macrophages in response to inflammatory stimuli, has been determined using two-dimensional homonuclear and heteronuclear NMR spectroscopy. Structure calculations were carried out by means of torsion-angle molecular dynamics using the program X-PLOR. The structure is based on a total of 2390 experimental restraints, comprising 2246 NOE-derived distance restraints (976 intraresidue, 608 sequential, 612 medium- and long-range, and 50 intersubunit restraints), 44 distance restraints for 22 hydrogen bonds, and 100 torsion angle restraints. A total of 20 structures were calculated, with the backbone (N, Ca, C) and heavy atom atomic rms distribution about the mean coordinates for residues 9-69 of the dimer being 0.57 0.16 and 0.96 0.12 , respectively. The N- and C-terminal residues (1 to 8 and 70 to 73, respectively) are disordered. The overall structure of the MIP-2 dimer is similar to that reported previously for the NMR structures of MGSA and IL-8, and consists of a six-stranded antiparallel beta-sheet (residue 25-29, 39-44 and 48-52) packed against two C-terminal antiparallel alpha-helices. A best fit superposition of the NMR structure of MIP-2 on the NMR structures of MGSA and IL-8 yield backbone atomic rms differences of 1.18 and 1.27 , respectively for the monomers, and 1.21 and 1.55 , respectively for the dimers (IL-8 residues 7-14 and 16-68). At the tertiary level the main differences between the MIP-2 solution structure and the IL-8 and MGSA structures involve the N-terminal loop between residues 9 to 23 and the loops formed by residues 30 to 38 and residues 53 to 58. At the quaternary level the difference between MIP-2 and MGSA and IL-8 results from differing interhelical angles and separations.
Iron is a component of most respiratory proteins where it acts as a ligand carrier. Structural studies of these proteins have been reported using both diffraction and spectroscopic techniques. In this work, we use 57Fe as a probe of structure/function relationships, using quantum chemical calculations and experimental measurements of the nuclear magnetic resonance chemical shieldings and Mssbauer quadrupole splittings. We have calculated the 57Fe shielding and electric field gradient (efg) tensors and their orientations using density functional theory. The shielding tensors were evaluated with the gauge-including atomic orbital (giao) method, using a locally dense basis set scheme and the DFT/B3LYP hybrid functional method. Among the compounds investigated were a cytochrome c model, isopropylisocyanide and carbonmonoxy-myoglobin (MbCO) model systems. We found good agreement between theory and experiment for cyt c and Mb 57Fe chemical shifts and shielding tensors and very good (0.10 mm sec-1 rmsd) agreement for the Mssbauer quadrupole splittings, or efg tensors. The tensor orientations are generally close to obvious molecular symmetry axes, with the skew of the shielding tensor reversing sign on transition from strong to weak ligand fields. Poor agreement was obtained for the 57Fe chemical shifts when MbCO models having highly distorted (x-ray) geometries were employed, suggesting that the Fe-C-O is close to the porphyrin normal. Structure and bonding of phenylnitroso and oxy ligands have also been investigated, and the very different efg orientations in oxy versus CO hemoglobin seen experimentally are well reproduced by the calculations. The ability to now quantitatively investigate both 57Fe shifts, shift and efg tensors (via Mssbauer spectroscopy) represents a useful new approach for understanding the detailed electronic structure (and functions) of metalloproteins.
Nitrophorin 1 (NP1) is a beta-barrel protein with a heme inside the barrel, bound to His-59 of the protein. There is a large distal pocket where NO or histamine can readily be bound. This ferriheme protein from the saliva of the blood-sucking insect Rhodnius prolixus carries nitric oxide (NO) reversibly as a diamagnetic complex, and releases NO when injected into the tissues of the victim. NO can then diffuse into the capillaries to cause vessel dilation and allow more blood to reach the site of the bite. NP1 thereafter picks up the histamine secreted by the victim, which delays the victim's immune response to the bite, thereby providing the insect with a second way of insuring a sufficient blood meal. We are investigating the proton and C-13 NMR spectra of NP1 in the absence and presence of imidazole and histamine, in order to understand the intimite environment of the heme-binding site. We have found that despite the very open-appearing heme binding site, only one heme orientation is observed; denatured protein readily re-folds to produce first the random A,B heme orientations, but within 12 hours has re-equilibrated to one orientation. NOEs from heme methyls to other heme residues allow assignment of most heme resonances. Two heme methyl resonances are buried within the diamagnetic protein envelope, a finding that is consistent with the orientation of His-59, determined from the X-ray crystal structure. We will show that analysis of the contact and pseudocontact shifts of the resolved heme methyls (3-Me and 5-Me) allows prediction of the other two heme methyl shifts, which can then be detected by advanced NMR techniques. Heterogeneity within the protein, probably resulting from cis/trans isomerization about the two disulfide bonds, divides the heme resonances into closely-spaced pairs and complicates the assignment of the heme resonances. Nevertheless, a set of consistent assignments has been achieved, and these results will be presented and discussed.
Proteins containing winged helix DNA binding motif have been shown to regulate tissue specific gene expression. One of the difficulties in NMR study of this family of proteins is highly uneven signal intensity. By using 15N, 13C and 2H enriched Genesis, formerly HFH-2, we observed an improved spectra quality for both DNA free Genesis and a Genesis-DNA complex. Although Genesis is highly sequence homologous to another winged helix protein HNF-3g with the known crystal structure, Genesis and its DNA complex both contain an extra helix where a loop was observed in HNF-3g/DNA complex. Thus, our data support a previous observation from molecular biological study that the DNA binding specificity in winged helix proteins is regulated by a sequence outside the primary DNA contact sequence. We also observed structure perturbation after Genesis interacts with its cognate DNA binding site.
A functional TraM protein is essential for the conjugative transfer of the F-like resistance plasmid R1. Based on mutational analyses it was shown that thereby TraM has a dual role in conjugation. First, a functional TraM protein was found to be required for normal levels or transfer gene expression, the second function is probably a direct involvement of TraM in the structural and functional organisation of the nucleoprotein complex at the origin of the conjugative DNA transfer, oriT.
This poster deals with the structure of a TraM mutant containing 55 amino acids, as well as with its binding to DNA. Combining NMR data and molecular modelling calculations led to a structure which consists of an extended alpha-helical part at the N-terminal domain, followed by two parallel and one antiparallel beta-strand. The binding studies between the TraM mutant and a palindromic DNA are currently under investigation.
The 3D solution structure of the recombinant oncogenic human proteins p13MTCP1 and p8MTCP1, coded by the MTCP1 oncogene were determined by homonuclear and 15N edited proton NMR methods at 600 MHz. These two proteins result from a differential splicing of MTCP1, a gene associated with a rare subset of T-cell leukemias. The small 8 kDa (68 residues) mitochondrial protein p8 was found overexpressed in proliferative T-cells, but is also expressed at low level in non-malignant tissues and is presumably an ubiquitary protein of unknown function. P13 (13 kDa, 107 residues) was found to be only associated to an heterogeneous group of mature T-cell proliferations. The primary sequence of p13 is highly and only homologous to that one of p14TCL1 coded by TCL1, a gene which was also found associated with such diseases, and they are probably the first members identified from a new family of oncogenic proteins. The 3D structure of p8 and p13 were calculated from NMR restraints using the programs DYANA and AMBER. The solution structure of p8 is a bundle of three helices, where two helices are firmly packed together through disulfide bridges in an original alpha-helical motif. The solution structure of p13 mainly consists in an eight-stranded orthogonal beta-barrel with a new topology. Two beta-pleated loops emerge from this barrel and might constitute the interaction surface with a potential molecular partner.
The goal of our work is to solve the structure of the two-EGF domain fragment of thrombomodulin, TMEGF(4-5). We have produced the glycosylated TMEGF(4-5) in the P. pastoris system incorporating uniform 15N-labeling. It has been of interest in our lab for some time to produce these 15N-labeled proteins in large quantites, however the standard fermentation protocols use NH4OH as the sole nitrogen source, and the amount of 15N-labeled NH4OH required for a 1 L fermentation costs approx. $20,000. Recently, we have developed an alternative fermentation scheme that allows the substitution of (NH4)2SO4 for NH4OH. The difficulty that had to be overcome was the high concentrations of salt (in the form of K2SO4) that become present during the fermentation under these conditions. We have successfully produced 60 mg of purified 100% 15N-labeled TMEGF(4-5) at a cost of $2700 using this method. Media exchange methods have been incorporated that sllow for control of the total amount of salt build-up using the 15N-labeled (NH4)2SO4 as the sole nitrogen source, so that high yields can be obtained. This method results in yields that are 50 - 100 fold increased over shake-flask production methods after purification.
This work was supported by NIH R01-HL47463. MJW is the recipient of an NIH Graduate Traineeship in Heme and Blood Proteins.
Epidermal growth factor (EGF) has a (1-3,2-4,5-6) disulfide bonding pattern. This pattern is found in nearly all EGF-like domains, despite wide variation in sequences. Three different disulfide bonding patterns of the fifth EGF-like domain of thrombomodulin all bind to thrombin with similar affinities, suggesting that the different disulfide bonded isomers of this domain may have similar backbone folds. These results prompted a test of whether the EGF fold could accommodate alternate disulfide bonding patterns. The disulfide bonds in murine EGF were altered to six different patterns, which after energy minimization, had similar energies and superimposed with an RMSD of the backbones of 2.1 . Structure calculations incorporating two different NMR distance restraint sets (Khoda & Inagaki, 1992; Tejero et al., 1996) showed that the three molecules containing a disulfide bond between the fifth and sixth cysteines (1-2,3-4,5-6), (1-3,2-4,5-6), (1-4,2-3,5-6) satisfied all the restraints. These three disulfide bonded isomers superimposed with an RMSD of 1.3 . A similar calculation was carried out with the fifth EGF-like domain of thrombomodulin. This domain has a (1-2,3-4,5-6) disulfide bonding pattern and the backbone fold differs substantially from EGF. For this domain, a few restraints were violated in the alternate disulfide bonded isomers, and the error functions were higher, although the differences were still not significant enough to rule out these structures.
Notch is a Drosophila neurogenic protein with 36 EGF-like domains in the extracellular portion of the protein. EGF-like domains 11 and 12 have been implicated in the binding of Notch to another Drosophila protein, Delta. The interaction of cells expressing Notch with cells expressing Delta is calcium dependent. Each of the two domains has the consensus sequence for calcium binding EGF-like domains. The fragment of Notch containing only EGF domains 11 and 12 has been expressed in P. pastoris yeast. Titration of the protein with calcium has been carried-out by monitoring the chemical shifts of aromatic protons. Each calcium-binding consensus sequence contains an aromatic amino acid: tyrosine in domain 11, and phenylalanine in domain 12. Each domain also contains a second phenylalanine so that there are 3 phenylalanines and one tyrosine in the entire protein fragment. Chemical shift changes are observed for a resonance at 7.80 ppm, with a half maximal change occurring at approx. 0.3 mM calcium. A new signal appears at 7.15 ppm at 0.8 mM calcium. These changes are probably caused by two different calcium binding events, one to domain 11 and one to domain 12.
This work was supported by funds from NIH/NCI #T32 CA09523, NIH-HL47463; HAG was supported
We have investigated the question of how CO and the isoelectronic ligand i-propylisocyanide bind to model hemes and heme proteins with a combination of nuclear magnetic resonance(NMR), 57 Fe Moessbauer, infra-red spectroscopy and density functional theory(DFT). The13C isotropic shift and shift anisotropy for the A0 substate of Physeter catodon CO myoglobin are virtually the same as that of the A0 model compound Fe(5,10,15,20-tetraphenylporphinato)
(CO)(N-methylimidazole), as are the17O chemical shift, and the 17O nuclear quadrupole coupling constant(NQCC), and lead to most probable ligand tilt(t) and bend(b) angles of 0o, 1o. Results for the A1 substate are also consistent with close to linear and untilted Fe-CO geometries(t= 4o, b= 7o), with the small changes in ligand spectroscopic parameters being attributed to electrostatic field effects. This poster details how one can combine all of the above spectroscopic observables to obtain a unified structural picture by using parameter surfaces compared using quantum chemistry, together with the Bayesian probability approach. Extension of this approach to the isoelectronic isocyanide complexes of heme proteins will be outlined.
In order to function as a tumor suppressor, p53 must assemble into a tetramer and bind DNA sequence specifically. In our work, the structural effect of mutants in the p53 oligomeriztion domain are examined with solution NMR spectroscopy. We have determined the structure of one mutant human p53 domain that forms a dimer and are nearing the completion of the wild-type trout p53 oligomerization domain. A comparison of these structures with the wild-type human p53 tetramerization domain clearly shows the dramatic affect of hydrophobic side chain size on the tertiary and quaternary structure of p53. Understanding the differences observed in mutant and wild-type structures helps us to redesign the oligomer interface, for example, to form better dimers or to form novel tetramers. Designed tetramers, coupled with a wild-type DNA binding domain have been made and show good transactivation and tumor suppression in cells. Active chimeric p53 tumor suppressors that would not be susceptible to the formation of inactivating hetero-oligomeric complexes could be used to replace inactive mutant p53 in tumor cells.
We also demonstrate that tetramerization and dimerization domains can interact to form hetero-oligomers. Preliminary data on one complex is consistent with the formation of hetero-tetramers, which requires a large change in the tertiary structure of the otherwise stable mutant dimer.
Double Quantum Heteronuclear Local Field Spectroscopy (2Q-HLF) is a solid-state NMR technique that has been recently developed to determine individual torsional angles around 1H-13C-13C-1H structural units. Obtaining such subtle structural information is essential for unraveling the molecular details behind important biological processes. 2Q-HLF has a number of features which makes it highly suitable for studying the conformations of individual molecular bonds in biomolecules: (i) the experiment has high sensitivity and resolution, since it operates under magic angle spinning conditions; (ii) the torsional angle determination does not require assumptions as to the orientations of chemical shift anisotropy tensors, and (iii) the use of double-quantum coherence suppresses the signals from natural abundance 13C spins in the protein backbone and chromophore. We applied this technique to determine the molecular torsional angle around the C10=C11 double bond in the retinylidene chromophore of rhodopsin and an intermediate of its photosequence, metarhodopsin I. We determine the torsional angles to be 160 10 degrees for rhodopsin and 180 25 degrees for metarhodopsin I. This highly detailed structural information for the chromophore of this 40 kDa intrinsic membrane protein responsible for visual signal transduction of vertebrates, will help to give us a better understanding of the structural changes in the ligand, that drive the signal transduction. In the near future, the methodology will be extended to other torsional angles in the chromophore and other photointermediates of the visual signal transduction, in order to refine our knowledge about this intriguing photochemical reaction. More information is available at http://www.fos.su.se/~mhl
Rotational Resonance is a solid-state NMR dipolar recoupling technique. The effect of the dipolar interaction between two spin one-half nuclei, such as 13C, can be effectively reintroduced in the evolution of the NMR experiment by matching the MAS rotation frequency of the sample to the isotropic chemical shift difference between the two sites. Recently, we have found that, for doubly 13C-labeled retinals, in the absence of chemical shift anisotropy, there is a linear dependence between the observed splitting of the resonances at the rotational resonance condition and the dipolar interaction (Verdegem et al., J. Am. Chem. Soc. 119, 169-174 (1997)). We used this phenomenon to determine intraligand distances in the retinylidene chromophore of bovine rhodopsin, an intermediate of its photosequence, metarhodopsin I and a rhodopsin analog, 10-methylrhodopsin. It is shown that the one-dimensional approach to rotational resonance in these systems has several practical advantages to the more conventional two-dimensional approach involving Zeeman magnetization transfer between the two sites, and generally, the experimental times needed are relatively short. The analysis procedure is fast and the accuracy of the distances in the range up to 0.4 nm is good. The measured intraligand distances were used to construct a detailed structure of the isomerization region of the chromophore of rhodopsin using ab initio Car-Parinello Molecular Dynamics. The isomerization region (C10-C11=C12-C13-C20) of the chromophore of rhodopsin appears 45 degrees out-of-plane distorted, partly due to the steric interaction between C20 and H10. Concomitantly, where the H10 is replaced by a methyl moiety, this provokes a larger distortion, as shown by our NMR results. The chromophore of metarhodopsin I adapts an all-E structure. This detailed structural information for the chromophore of this 40 kDa intrinsic membrane protein responsible for visual signal transduction of vertebrates will help to give us a better understanding of the structural changes in the ligand, that drive the signal transduction. More information is available at http://www.fos.su.se/~verdegem
Simian Virus 40 ( SV40 ) large tumor antigen is a member of small DNA tumor viruses. The oncoproteins of these small DNA tumor viruses inactivate pRB as part of their transformation process, and have become useful tools to study both cellular transformation and the function of the RB family proteins. The N-terminus of SV40 large tumor antigen ( TAg ) containing the highly conserved LXCXE pRB family protein interaction domain is required for replication and appear to contribute to transformation. Recent studies show that this part of TAg induces the rapid turnover of p130 and reduces the level of phosphorylated p130 and p107. The N-terminus of TAg ( N-TAg ) has homology to the J-domain of DnaJ molecular chaperone proteins which can activate Hsp70 proteins through a HPD motif.
The work presented was carried out in two steps. First, we determined the structure of J-domain ( 1-82 ) polypeptide. It contains three well-defined helices. The helices II and III form an antiparallel helical coiled-coil with helix I approximately perpendicular to them. A strong correlation is seen between the extent of sequence-conservation of hydrophobic residues in the family of J-domain homologues, and the structural organization of the hydrophobic core in these proteins. The HPD motif is exposed to solvent in a well defined loop region between helices II and III.
Next, we investigated a longer fragment ( 1-108 ) to include a conserved Tryptophan-rich domain and the pRB interacting LXCXE motif. Surprisingly, the J-domain spectra changed dramatically as the tryptophan-rich domain forms a forth helix laying on top of helices II and III to bring the LXCXE motif very close to the HPD motif. Extending the fragment to a.a. 135 showed that the region after residue 108, which includes the nuclear localization domain, is not structured.
This structural study provides the basis for designed mutants to disect the multifunction of N-TAg so that its mechanism of cell transformation can be better understood.
Metallothioneins (MTs) are small (6-7 kDa), ubiquitous, cysteine-rich proteins thought to function as intracellular metal ion distributors and mediators. The mammalian isoforms bind seven divalent metal ions in three- and four-metal clusters located in two structurally independent domains. The structures of the extensively studied isoforms MT1 and MT2 lack regular secondary structural elements and the tertiary fold is dictated by the metal clusters. The recently discovered, brain specific isoform MT3 (or GIF, Growth Inhibitory Factor) is implicated in the pathogenesis of Alzheimer's Disease and inhibits growth and survival of rat neuronal cultures. Despite the same number of cysteines (20 out of 68 residues) at conserved positions, this isoform differs from MT1 and MT2 with a hexapeptide insertion at the C-terminus and a unique Cys-Pro-Cys-Pro tetrapeptide at the N-terminus. The recombinant expression and purification of human Cd7-MT3 have been optimized and unlike earlier reports of extensive line broadening, well-resolved one dimensional and two dimensional (TOCSY, NOESY) 1H NMR spectra were acquired. The status of the sequential assignments currently in progress will be reported and the implications of the sequence novelties on the secondary structure will be discussed.
Crucial to the tertiary structure elucidation of MTs is the establishment of specific metal to Cys connectivities. This was accomplished by 1H-113Cd HMQC (heteronuclear multiple quantum correlation) experiments that revealed a conserved cluster structure similar to MT1 and MT2. ( NIH DK18778 )
As a model for calcineurin, a protein phosphatase central to the immune response, the structural, metal binding, and enzymatic properties of bacteriophage lamba protein phosphatase are being investigated using 1H and 113Cd NMR methods. Expression and purification procedures which yield 50 mg of 99% pure recombinant apo-phosphatase from 1 L of bacterial culture will be presented. In addition, metal reconstitution will be probed using 113Cd NMR techniques to monitor the relative affinity of the dinuclear site for Cd as well as other metals. Finally, representative 1- and 2-dimensional 1H NMR spectra of both apo- and metal reconstituted enzyme will be presented and discussed.
(NIH GM49858)
Pacific Northwest National Laboratory's Environmental Molecular Sciences Laboratory is a new national user facility housing a variety of state-of-the-art NMR instruments. We have developed a 'Virtual NMR Facility' that provides full access to these resources via the Internet. Users of the Virtual Facility have secure remote acquisition control of the EMSL spectrometers, supported by a full suite of collaboration tools. These tools include remote spectrometer time scheduling, real-time videoconferencing, computer screen sharing for live consultation on setup and monitoring of remote experiments and for joint data analysis sessions, and a World Wide Web based Electronic Notebook. This poster details the capabilities available through the Virtual NMR Facility and describes how they are being used on the Facility's first project: an EMSL-LBNL collaboration on multidimensional NMR experiments for assignment and modeling of a 110 amino acid Heat Shock Factor protein molecule. Other than the initial sample preparation, done at LBNL, and the insertion of the sample into the
spectrometer at the EMSL, all aspects of the experiment have been done collaboratively over the Internet. Literature references and sample preparation procedures were shared via the electronic notebook prior to the NMR experiments. Setup of the experiments was done directly from LBNL via encrypted X Windows connections to the EMSL 750 MHz spectrometer. Several experiments have now been completed this way, and we have found it to be an effective means for running NMR experiments. We anticipate that a Virtual NMR Facility is likely to become a valuable way for more researchers to access expensive state-of-the-art high field instruments and share data more quickly and efficiently. All EMSL Collaborative software tools we use are cross-platform (Unix, PC, Mac) and are available for free download from http://www.emsl.pnl.gov:2080/docs/collab/.
The 2-Fe 2-S ferredoxin putidaredoxin (Pdx) is the redox partner of cytochrome P450cam. The structure of the protein further than 8 angstroms from the metal center has been determined by NMR methods. However, 1H lines are excessively broadened near the metal center in both the oxidized and reduced protein, rendering standard NMR structural methods useless for the metal binding site. However, resonances of 15N and 13C are less susceptible to broadening, and structural information is available from these resonances in the form of distance restraints from the metal center via relaxation measurements and hyperfine couplings from ENDOR experiments. In order to take advantage of these possibilities, we have prepared samples of Pdx which are selectively labeled with 15N and/or 13C at one or two amino acids in order to assign resonances in a sequence-specific manner. This permits the use of double-resonance or two-dimensional NMR methods to be used in making sequence-specific assignments of resonances in the metal binding site.
The high-resolution solution structure of the inhibitor-free catalytic fragment of human fibroblast collagenase (MMP-1), a protein of 18.7 kDa which is a member of the matrix metalloproteinase family, has been determined using three-dimensional heteronuclear NMR spectroscopy. A total of 30 structures were calculated by means of hybrid distance geometry-simulated annealing using a total of 3333 experimental NMR restraints, consisting of 2409 approximate interproton distance restraints, 84 distance restraints for 42 backbone hydrogen bonds, 426 torsion angle restraints, 125 3JH restraints 153 C-alpha restraints and 136 C-beta restraints. The atomic rms distribution about the mean coordinate positions for the 30 structures for residues 7-137,145-163 is 0.42 0.04 for the backbone atoms and 0.80 0.04 for all atoms. The overall structure of MMP-1 is composed of a beta-sheet consisting of five beta-strands in a mixed parallel and anti-parallel arrangement and three alpha-helices. A best fit superposition of the NMR structure of inhibitor-free MMP-1 with the 1.56 resolution X-ray structure by Spurlino et al. (1994) complexed with a hydroxamate inhibitor yields a backbone atomic rms difference of 1.22 . The majority of differences between the NMR and X-ray structure occur in the vicinity of the active site for MMP-1. This includes an increase in mobility for residues 138-144, a displacement for the Ca+2-loop (residues 74-80) and a slow conformational exchange for residues comprising the active site (helix B, zinc ligated histidines and the nearby loop region). In contrast to the X-ray structures, only the slow conformational exchange is lost in the presence of an inhibitor. Distinct differences were observed for side-chain torsion angles, in particular, the chi-1 for N80 is -60o in the NMR structure compared to 180o in the X-ray. This results in the side-chain of N80 occupying and partially blocking access to the active site of MMP-1.
Hyperfine-shifted resonances arising from the heme group of hemoproteins provide a wealth of information on the electronic and magnetic structure of the heme active site. Furthermore, these resonances also provide a spectroscopic handle to study inter- and intra-protein interactions. The complete assignment of these resonances, however, can be challenging due to asymmetric delocalization of unpaired electron density into the porphyrin macrocycle. In addition, some heme proteins consist of mixtures of heme isomeric forms where the population of the two heme isomers is approximately 1:1, such as outer mitochondrial membrane (OM) cytochrome b5. The spectroscopic consequence of heme isomerism is the virtual doubling of all resonances arising from the heme, thus increasing the complexity of the spectrum.
In an attempt to circumvent these problems, the heme of OM cytochrome b5 was labeled with 13C at different positions utilizing [3-13C]-, [5-13C]- and [1,2-13C]- aminolevulinic acid as a heme precursor (1). The resultant labeled protein samples were used to carry out the assignment of heme resonances in ferricytochrome b5. This was accomplished with a combination of HMQC, HMBC, NOESY, and 1H-13C-13C edited 1H NMR experiments (2). These experiments yielded the assignment of all 1H heme resonances and most 13C heme resonances, including those arising from quaternary heme carbons in both isomeric forms.
(1) Rivera, M. & Walker, F. A. Anal. Biochem. 230, 295-302 (1995).
(2) Qiu, F., Rivera, M., & Stark, R. E. J. Magn. Reson. 129, 000-000 (1997).
Packaging of the RNA genome into the maturing viral particle is necessary for the production of infectious HIV-1 virions. Deletion of the packaging signal (also called the Psi-site) in the RNA genome or the nucleocapsid (NC) protein results in noninfective virions that are defective in RNA packaging. We have determined the NMR structure of the HIV-1 nucleocapsid protein bound to the SL3 stem-loop Psi-site RNA using 15N,13C-labeled RNA and protein. The NC protein has two zinc knuckle motifs separated by a flexible linker region and flanked by two disordered tails. Upon binding to the RNA, the N-terminal tail assumes a 310-helix conformation that packs in the major groove of the RNA A-helical stem. The side chain of the highly conserved Asn5 in the 310-helix forms specific hydrogen bonds to bases in the RNA stem. The majority of protein-RNA interactions occur between the RNA -G6G7A8G9- tetraloop and the two NC zinc knuckles. The base of G7 projects inside a hydrophobic cleft formed by protein aromatic and hydrophobic side chains. There are base-specific H-bonds from the exocyclic amino and oxygen groups of G7 to backbone carbonyl and amide groups of the C-terminal zinc knuckle. G9 forms similar base-specific hydrophobic and hydrogen bonding interactions with the N-terminal zinc knuckle. A8 packs against the N-terminal knuckle and forms H-bond with conserved Arg32 in the NC linker region. The NC-SL3 structure reveals new protein-RNA recognition motifs, explains the role of the highly conserved NC residues, and provides an alternative rational drug design strategy aimed at disrupting RNA packaging.
Methane monooxygenase (MMO) is a nonheme iron-containing enzyme which consists of three protein components: a hydroxylase (MMOH), an NADH linked reductase (MMOR) and a small B component (MMOB) that appears to play a regulatory role. NMR and CD spectroscopy have been used to structurally characterize the 138 amino acid MMOB protein. CD data indicate that MMOB is comprised of 32% alpha-helix, 20% beta-sheet and 27% turn. At 1 mM MMOB, proton NMR spectra show broader resonances than expected for a monomeric protein of 15,000 Daltons. Pulse field gradient NMR diffusion measurements yield diffusion coefficients which indicate predominant formation of dimers at 1mM MMOB and monomers at below 0.2 mM. Heteronuclear multidimensional NMR spectroscopic techniques are being used to assign the proton, carbon and nitrogen resonances at 0.2 mM . The secondary structure of MMOB will be analyzed by NOE correlations, amide exchange, J- coupling constants and chemical shift index prior to elucidation of the tertiary fold of the protein.
This research has been supported by grants from the NIH.
Much progress has been made both towards increasing the size of proteins that can be studied by high-resolution NMR spectroscopy and towards simplifying the assignment and subsequent structure determination processes. Deuteration and 13C/15N isotopic labeling have been instrumental in these endeavors. Perdeuteration at non- exchangeable sites in a protein permits the observation of very long-range 1HN-1HN NOEs in data acquired utilizing the 4D 15N/15N-separated NOESY experiment. Direct NOEs have been observed between amide nuclei separated by over 9 angstroms in a 12 kDa core packing mutant of thioredoxin (L78K-TRX). The global folds of human carbonic anhydrase II (29 kDa) and L78K-TRX have been calculated using backbone and sidechain restraints obtained soley from 4D 15N/15N- separated NOESY data along with a conservative set of dihedral angle restraints obtained from the assignment of the carbon chemical shifts. We will present our most recent efforts towards increasing the size of proteins amenable to NMR structural studies.
Transcription factor 1, TF1, is a 22 kDa homodimer, type-II DNA-binding protein encoded by the Bacillus subtilis bacteriophage SPO1. Its NMR solution structure includes a dimeric core which is composed of two alpha helices from each monomer at the bottom and two three-stranded beta sheets at the top. From this core, two beta-ribbon arms extend upward and fold back as two kinked C-terminal alpha helices. A double mutant of TF1, p15/32 TF1, in which amino acid 15th was changed from glutamate to glycine (E15G) and residue 32nd was changed from threonine to isoleucine (T32I), was of special interest. Although the mutations occurred outside the DNA binding domain, p15/32 TF1 exhibited strikingly different properties: it bound DNA 40 times more tightly; its thermal stability was increased by 21 degrees C and the mutant homodimer displayed a tighter subunit association. Uniformly 13C/15N lablelled p15/32 TF1 was overexpressed and purified for NMR studies. Assignments for the 1H, 15N and 13C nuclei and the chemical shift-derived secondary structure for p15/32 TF1 wil be presented.
SPO1, a Bacillus subtilis bacteriophage, incorporates 5-(hydroxymethyl)-2'-deoxyuridine (HmU) into its DNA instead of thymine to facilitate phage multiplication. HmU is a pyrimidine base whose hydroxymethyl group replaces the methyl group of thymine. The presence of HmU in the phage genome is an absolute requirement for its DNA to bind to transcription factor 1, an SPO1-encoded, type-II DNA binding protein. We have synthesized a 17-mer that mimics the consensus sequence of the TF1 binding site: d(5'-ACCHACHCHHHGHAGGT-3'). 1H and 31P chemical shifts have been assigned and the solution structures of the single-stranded and double-stranded HmU-containing DNAs will be described.
The cystolic fatty acid-binding proteins (FABP) are thought to function in the binding and transport of fatty acids (FA) within cells. Whereas all other mammalian FABP's have been found to contain a single binding site for long-chain FA, previous studies suggest that LFABP binds 2-3 molecules of FA. It is also known that LFABP binds heme, squalene, eicosanoids, lysophopholipids, among others, in addition to FA's and their CoA esters. Our chemical shift index studies and NOE connections show that, both in apo- and holo-states, LFABP adopts a similar structural fold to other FABP's and the structure is predominantly beta-sheet. However, the NH-resonances of apo-LFABP is characterized to have narrow shift dispersion and undergo very rapid solvent exchange. Twenty out of 124 backbone NH's were found to have doubled crosspeaks and 15 others were not observable. Thus, apo-LFABP adopts the fold with substantial flexibility. The N15-HSQC spectra show that LFABP is locked into a well-defined structure after 2 eq. of palmitate or oleate are added. As more FA is added, the linewidth of some crosspeaks becomes progressively sharpened, but the spectrum remains essentially the same. As many as 8 eq. of oleate may be added to LFABP at pH7 and the solution remains clear. The 2 specifically bound FA molecules have been studied with CT-C13-HSQC, HCACO, and HCCHCOSY experiments. Two alpha-CH and two methyl resonances were clearly detected from the bound FA's. The CO resonances for the two FA molecules were not distinguishable, however. This result confirms a previous report (1) by Cistola et al. The X-ray study (2) of the holo-LFABP (apo-structure is not yet available) with oleate showed 2 specifically bound oleates in different part of the binding cavity and segments of aliphatic chain were noted near surface. It was suggested that the two oleates bind with LFABP in sequence. With H1, C13, N15 assignments for both apo- and holo-LFABP in hand in our laboratory, we are able to investigate the binding of FA in sequence, and to delineate the conformational changes associated with the addition of each FA molecule.
Human platelet factor-4 (PF4) demonstrates various physiologic effects. It accelerates thrombin - antithrombin complex formation, potentiates platelet aggregation and inhibits megakaryocytopoiesis, angiogenesis and solid tumor growth. Many of these biological activities of PF4 appear to be related to its known interaction with the poly-sulfated glycosaminoglycan heparin. Therefore, it becomes crucial to understand the structural details of this protein-carbohydrate complex. Most experiment performed with heparin have used commerically available heparin which due to its heterogeneity in size and sequence redundancy, allows only some general insight into structural aspects of the complex. In order to avoid these and other technical problems, the present NMR-based study used a homogeneous heparin-derived dodecasaccharide fragment and C-13/N-15 uniformly labelled PF4. Translational diffusion coefficients derived from pulsed filed gradient (PFG) NMR experiments, provide insight into the changing size of the complex at different heparin - PF4 ratios. NOESY data and chemical shift changes in PF4 and in the heparin fragment indicate minor conformational changes and suggest where heparin is binding on the protein. Experimental NMR data are supplemented with computer simulations which allow one to estimate the geometry of the molecular complex and suggest detailed information on interactions between functional groups on PF4 and heparin. Information derived from this study can then be used to help design better pharmaceutical agents.
This research has been supported by the American Heart Association.
Factor Xa is an important enzyme component of the coagulation process. This enzyme is a serine protease which catalyzes the activation of prothrombin to thrombin when in the prothrombinase complex. Several research studies have suggested the existence on factor Xa of subtle secondary recognition sites distinct from the active site, called exosites. Tick anticoagulant peptide (TAP) is an inhibitor of factor Xa which is uniquely specific for factor Xa. Tick anticoagulant peptide is also unique in that it appears to contain both active site and exosite interactions with factor Xa. A mutant of TAP, R3A-TAP, has the active site interactions with factor Xa negated, leaving only the exosite interactions with factor Xa remaining. These exosite interactions are expected to be instrumental in the process by which TAP recognizes its specific substrate. We believe structural elucidation of these subtle secondary binding sites on TAP should aid in the explanation for this inhibitor's specificity. Furthermore, elucidation of the binding sites of R3A-TAP combined with knowlege gained from the available crystal structure of human factor Xa will aid in our investigation of the secondary binding sites present in factor Xa.
Progress will be reported describing the solution structure of R3A-TAP obtained by Nuclear Magnetic Resonance spectroscopic techniques. Studies of R3A-TAP complexed with bovine factor Xa will also be presented.
map30 is an anti-HIV and anti-tumor protein from plant Momordica charantia. It is effective against various types of human cancer in mouse with EC50 in the range 0.21-0.86 nM. In addition, the protein inhibits infection of HIV-1 virus in lymphocytes and monocytes as well as the replication of the virus in already infected cells. Furthermore, it is not cytotoxic to normal cells because it recognizes and enters only viral infected or tumor-transformed cells, and unable to enter healthy cells. The protein contains a single chain of 263 amino acids (30 kDa). Its size and spin complexity pose significant challenges and difficulties to solution NMR spectroscopiests. We report here the 3D solution structure of map30 determined using solution NMR. The structure contains 10 alpha helices, 9 beta strands and a random coil region of ~15 residues of C-terminus. The overall determined structure of map30 resembles the structure of racin A chain, a toxin which is known to be lethal to normal human cells ( and life). The structural similarity between two proteins are also suggested by their ~30% sequence homology. However, racin A chain does not possesses many activities that map30 has. Moreover, racin A chain kills cells indiscriminately, in contrast to what map30 does to the cells. The determined map30 structure will allow us to examine structural bases for map30's unique anti-HIV and anti-tumor activities.
In addition to presenting the structural features of map30, we also would like to discuss the technical difficulties we have encountered in determining the structure, and some of new techniques we have developed to overcome those difficulties.
Prions cause neurodegenerative diseases in humans and animals [1]. The infectious prions are largely composed of a hostencoded glycoprotein denoted PrP-Sc. Limited digestion of PrP-Sc by proteinase K yields an N-terminally truncated polypeptide corresponding primarily to residues 90-231, which remains infectious when injected into normal healthy animal. A recombinant prion protein (rPrP) of 142 residues corresponding to residues 90-231 of Syrian hamster (SHa) was expressed in E. coli, purified and refolded into an alpha-helical form resembling the normal cellular isoform PrP-C [2]. The structure of SHa rPrP(90-231) has been studied using multi-dimensional NMR with 15N, 15N/13C-labeled samples at 600 and 750 MHz [3]. The backbone 1H, 13C and 15N resonances were assigned by establishing intra-residue and sequential connectivities of N, HN, Ca, Cb resonances using 3D triple resonance experiments HNCA, HNCACB, CBCA(CO)HN. The side-chain resonances were assigned with HBHA(C)HN, HCCH-TOCSY and 15N-HSQC-NOESY experiments. Distance constraints were obtained using 100 ms, 750 MHz, 15N-NOESY-HSQC and 13C-NOESY-HSQC spectra. The program Aria/Xplor [4] was used for the assignment of ambiguous restraints and the structure calculations, followed by Amber restrained MD and energy minimization.
[1] S. B. Prusiner, Science 252, 1515-22 (1991).
[2] I. Mehlhom, et al., Biochemistry 35, 5528-5537 (1996).
[3] T. J. James, et al., PNAS. USA Vol. 94, pp. 10086-10091, (1997)
[4] M. Nilges, et al., JMB 269, 408-422 (1997)
In the course of a protein design study, we generated a series of core mutants of the B1 domain of streptococcal protein G (A. Su and S. L. Mayo, Protein Science, 6, 1701-1707, 1997). Here we present the solution structures, obtained by standard homonuclear NMR and hybrid distance geometry / simulated annealing methods, of two of these: a three-fold and a six-fold mutant. We also present an analysis of the mutants' backbone dynamics based on 15N relaxation.
Thermal melts had previously shown the three-fold mutant to be slightly more stable than wild type GB1 (Tm = 91 oC, compared to Tm = 87 oC for GB1), while the six-fold mutant is somewhat less stable (Tm = 73 oC). Since a protein's hydrophobic core is a critical determinant of its structure and stability, we felt it was important to compare the structures of these mutants to that of GB1. Both mutants adopt the characteristic protein G fold, with only small deviations from wild type super- secondary structure parameter values. The presence of a large number of long-range NOEs and large amide hydrogen exchange protection factors indicate that many of the interactions which stabilize the GB1 structure are present in both mutants. Of particular interest, chi1 angles for core residues are well-defined, attesting to efficient packing in the redesigned cores, even with 6 out of 10 core residues mutated. Finally, a preliminary analysis of 15N relaxation data indicates that the motional behaviour of the mutants' backbones is similar to that of the wild type.
We have investigated the involvement of Glu-681 in the red blood cell anion exchange protein, band 3, in binding of Cl to the external-facing anion transport site, by using 35Cl NMR. The fact that protons and sulfate are apparently transported together across the permeability barrier suggests that the proton and sulfate binding sites might be near each other in a substrate "pocket" within band 3. Furthermore, the competition between sulfate and chloride suggests that the monovalent anion binding site may lie within the same pocket. To see whether or not the external Cl transport site is located near Glu-681, we reacted intact red blood cells with the band 3 inhibitor Woodward's Reagent K ( WRK) to selectively modify Glu-681, and then used single and double-quantum 35Cl NMR to monitor external Cl binding. Binding was measured in 50% hematocrit intact red blood cell suspensions by fitting the magnitude calculation of the external 35Cl free induction decay to a biexponential function (internal Cl signal is eliminated by acquisition delay) . As we have shown previously, the relaxation rate of the rapidly decaying component of the 35Cl FID (R2f), which arises from the satellite transitions in the spin 3/2 system, is very sensitive to Cl binding to the external transport site, while the transverse relaxation rate of the central transition is very weakly dependent (if at all) on binding to the transport site. Surprisingly, R2f of WRK-treated cells is considerably larger than that for control cells with DIDS (blocks Cl binding to the transport site), and addition of DIDS to WRK-treated cells reduces R2f to the same value as observed for control cells with DIDS. Similarly, the doulbe-quantum-filtered 35Cl signal which is due to the transport site is not eliminated by WRK treatment, but is reduced when DIDS is added. The results demonstrate that both the Cl and DIDS binding sites are not very near to Glu-681, supporting kinetic evidence that separate gating regions in band 3 may control proton and Cl transport.
(Supported by NIH Grant DK27495).
Poster Slot Number: 232
Structural Characterization of the Complex Formed by the Fv Portion of a Humanized anti-HER2 Antibody and the E-Domain of Protein A
Institute: Genentech, Inc., South San Francisco;
Poster Slot Number: 233
Solution structure of the heparin-binding domain of vascular endothelial growth factor
Institute: Genentech, Inc., South San Francisco;
Poster Slot Number: 234
Structure of Macrophage Inflammatory Protein-2 by NMR and Torsion Angle Molecular Dynamics
Institute: Walt Disney Memorial Cancer Institute, Orlando, FL;
Poster Slot Number: 235
An Experimental and Density Functional Theory Investigation of 57Fe Chemical Shifts and Mössbauer Quadrupole Splittings (Field Gradients) i
Institute: University of Illinois at Urbana-Champaign;
Poster Slot Number: 236
NMR Studies of the Paramagnetic Forms of Nitrophorin 1, the Reversible NO-Carrying Protein from the Saliva of R. prolixus
Institute: Department of Chemistry, University of Arizona, Tucson, Arizona 85721;
Poster Slot Number: 237
NMR Investigation of a perdeuterated winged helix DNA binding protein, Genesis, and its DNA complex
Institute: Dept of Biochem, Univeristy of Illinois at Chicago;
Poster Slot Number: 238
TraM Protein, its Structure and DNA Binding Properties
Institute: Institute of Organic Chemistry, Graz;
Poster Slot Number: 239
Solution structure of the recombinant human proteins p8 and p13 from the the MTCP1 oncogene.
Institute: Centre de Biochimie Structurale, Montpellier, France;
Poster Slot Number: 240
PRODUCTION OF LARGE QUANTITIES OF 15N-LABELED PROTEINS IN PICHIA PASTORIS BY FERMENTATION
Institute: Dept. Of Chem. & Biochem., University of California, San Diego, La Jolla;
Poster Slot Number: 241
DISULFIDE BOND PLASTICITY IN EPIDERMAL GROWTH FACTOR
Institute: Dept. of Chem. and Biochem., University of California, San Diego, La Jolla;
Poster Slot Number: 242
Calcium-binding to the Notch (11-12) EGF Domains
Institute: Dept. of Chem. & Biochem., University of California, San Diego, La Jolla;
Poster Slot Number: 243
The Z-surface Approach to Obtaining Structure: Applications to Heme Isocyanide and CO Complexes
Institute: University of Illinois at Urbana-Champaign;
Poster Slot Number: 244
Structure and interaction of mutant p53 oligomerization domains.
Institute: The Wistar Institute, Philadelphia PA;
Poster Slot Number: 245
Direct Determination of a Molecular Torsional Angle in Rhodopsin and Metarhodopsin I by Double-Quantum Solid-State NMR.
Institute: Division of Physical Chemistry, Stockholm University, Stockholm, Sweden;
Poster Slot Number: 246
Obtaining Intraligand Distances in Rhodopsin, Metarhodopsin I and a Rhodopsin Analog using One-Dimensional Rotational Resonance.
Institute: Leiden Institute of Chemistry, Leiden University, Leiden, The Netherlands;
Poster Slot Number: 247
Solution Structures of pRB interaction domain ( 1-108 ) of SV40 large T antigen
Institute: Harvard Medical School, Boston;
Poster Slot Number: 248
Structural Implications on the Growth Inhibitory Function of Metallothionein Isoform 3
Institute: Department of Biochemistry, University of Minnesota, Minneapolis, MN 55455;
Poster Slot Number: 249
NMR and Metal Binding Studies of Lambda-Phage Protein Phosphatase: A Model for the Catalytic Dinuclear Metal Site in Calcineurin
Institute: Department of Biochemistry, University of Minnesota, Minneapolis, MN 55455;
Poster Slot Number: 250
Remote Research Using the EMSL Virtual NMR Facility: 750 MHz Studies of a Heat Shock Factor Protein
Institute: Pacific Northwest National Laboratory, Richland;
Poster Slot Number: 251
Selective 15N and 13C labeling of paramagnetic proteins for structural analysis
Institute: Department of Chemistry, Brandeis University, Waltham, MA 02254-9110;
Poster Slot Number: 252
High-Resolution Solution Structure of the Inhibitor-Free Catalytic Fragment of Human Fibroblast Collagenase Determined by Multidimensional NMR
Institute: Wyeth-Ayerst Research, Pearl River, NY;
Poster Slot Number: 253
Assignment of hyperfine-shifted carbon resonances in both heme isomeric forms of outer mitochondrial membrane cytochrome b5
Institute: Oklahoma State University, Stillwater, OK; #CUNY College of Staten Island;
Poster Slot Number: 254
Structure of HIV-1 Nucleocapsid Protein Bound to Packaging Signal RNA
Institute: Howard Hughes Medical Institute-UMBC, Baltimore;
Poster Slot Number: 255
Structural Characterization of Component B from Methane Monooxygenase
Institute: Department of Biochemistry, University of Minnesota at Twin Cities;
Poster Slot Number: 256
Structure Determination of Perdeuterated Proteins
Institute: Duke University NMR Center Durham, NC 27710;
Poster Slot Number: 257
1H-, 15N- and 13C-NMR assignments and secondary structure of p15/32 TF1, a double mutant of the Bacillus subtilis SPO
Institute: University of California, San Diego; La Jolla;
Poster Slot Number: 258
1H- and 31P-NMR studies of transcription factor 1 binding site: a 17-mer duplex DNA containing 5-(hydroxymethyl)-2'-deoxyuridine
Institute: University of California, San Diego; La Jolla;
Poster Slot Number: 259
Studies of the Sequential Fatty Acid Binding by Rat Liver Fatty Acid-Binding Protein (LFABP) Using Uniformly C13-labeled Fatty Acids and N15-labeled L
Institute: CUNY-College of Staten Island, NY; #Rutgers University, New Brunswick, NJ;
Poster Slot Number: 260
NMR studies of heparin derived dodecasaccharide binding to platelet factor-4
Institute: Department of Biochemistry, University of Minnesota, Minneapolis MN;
Poster Slot Number: 261
Solution Structural Studies of the R3A Mutant of Tick Anticoagulant Peptide
Institute: Emory Univ., Atlanta, GA and Univ. of Vermont, Burlington, VT;
Poster Slot Number: 262
Determining the 3D structure of a 30 kDa singular anti-HIV and anti-cancer protein, map30, by solution NMR
Institute: Molecular Structural Biology Section, NIDR/NIH;
Poster Slot Number: 263
Determination of the Solution Structure of Recombinant Syrian Hamster Prion Protein rPrP(90-231)
Institute: Dept. Pharm. Chem., University of California, San Francisco, CA 94143;
Poster Slot Number: 264
Structures and Backbone Dynamics of Protein G Core Mutants
Institute: Howard Hughes Medical Institute and Division of Biology, Caltech, Pasadena;
Poster Slot Number: 265
Location of Band 3 binding bites for Cl and DIDS relative to Glu-681: 35Cl NMR in the presence of Woodward's Reagent K
Institute: University of Rochester, Rochester, NY;
Click here to return to the Poster Topics page.