Institute: ARIAD Pharmaceuticals, Inc. Cambridge, MA USA
Keywords:
Abstract:
A variety of high-affinity ligands are known for FKBP (FK506 binding protein). These ligands encompass natural products e.g. rapamycin and FK506 as well as lower molecular weight synthetic molecules. This high-affinity interaction has been used as the basis of a series of chemical inducers of dimerization which have found broad utility in biological and medical applications utilizing controlled protein-dimerization events, specifically in the area of regulated gene therapy. Using the three-dimensional structures of FKBP:ligand complexes, the ligand-protein interface has been redesigned to enhance selectivity over native complexes by incorporating FKBP "holes" and complementary ligand "bumps". One synthetic ligand binds a F36V mutant of FKBP with an affinity of ~100pM. The X-ray structure of this complex shows a hydrophobic "hole" has opened at the site of the mutation which is partially occupied by the ethyl "bump" of the ligand (Proc. Natl. Acad. Sci. USA 95, 10437-10442, 1998). A series of ligand analogs has been prepared which display "bumps" of differing sizes and with differing stereochemistries in an effort to optimize the binding of the ligand to the "holed" protein. In the present study, NMR spectroscopy has been used to determine the solution structure of several of these analogs to correlate co-structural features with binding affinity. Interestingly, in several of these complexes we observe a switched binding mode where the "bump" no longer engages the engineered protein "hole", rather occupies a adjacent natural cavity on the surface of the protein, and the tetrahydropyrone (THP)-mimicking group expected to occupy the cavity occupies the engineered "hole". As we show, the ligand binding mode is dictated by the stereochemistry of the "bump" substituent while relative ligand affinities result from the combination of preferences for the "bump" and THP-mimic sidechains by both "hole" and cavity in either binding mode. The present study illustrates the utility of NMR structure determination in studying a series of related compounds where differences in binding orientation make simple SAR analysis problematic.

032
Session Assigned: M&T P
Title: Solution structure of substrate-based ligands when bound to hepatitis C virus NS3 protease

Institute: Bio-Mega, Laval, Quebec, Canada
Keywords: Peptide Structure and Dynamics
Abstract:
The interactions of NS3 protease with inhibitors that are based on N-terminal cleavage products of peptide substrates were studied by transferred nuclear Ovehauser effect NMR methods. These experiments showed that inhibitors bind the protease in a well-defined extended conformation. Protease-induced line-broadening studies helped identify the segments of inhibitors which come into contact with the protease. A comparison of the free and protease-bound conformations suggests that these ligands undergo siginficant rigidification upon complexation. This work provides the first solution structure of an inhibitor when bound to NS3 protease and should be valuable for desigining more potent inhibitors.

033
Session Assigned: M&T P
Title: Conformational Study of Azapeptide as a Template to fix specific secondary Structure in Peptides.

Institute: Advanced Analysis Center, KIST, Seoul, Korea
Keywords: Azapeptide, beta-turn, 1H NMR
Abstract:
It is well known that beta-turn structure is critical for the bioactivities of peptides since beta-turn structure in peptides may be part of enzyme recognition site. Furthermore, the beta-turn are common conformations for many bioactive peptides.

We designed and synthesized azapeptides, which are Boc-Phe-AzLeu-Ala-OMe(1) and Ac-Aib-AzGly-NH2(2), as a template of the beta-turn mimics. Their conformational preference have been determined by 1H NMR. The temperature coefficient of amide protons in chemical shifts, coupling constants and NOE data revealed that the designed azapeptides (1) and (2) forms beta-turn structures.

A more detailed conformation of azapeptides and their unique conformational preference will be presented.

034
Session Assigned: M&T P
Title: Probing Structural Heterogeneity in Mutacin 1140, a Novel Lantibiotic

Institute: University of Florida
Keywords:
Abstract:
We are investigating possible structural variability of mutacin 1140, a peptide lantibiotic produced in Streptococcus mutans JH1140 [Hillman, J.D., Novák, J., Segura, E., Gutierrez, J. A., Brooks, A., Crowley, P.J., Hess, M., Azizi, A., Leung K.-P.,Cvitkovitch, D., and Bleiweis, A. S. Infect. Immun. 66 (6), 2743-2749 (1998).]. Lantibiotics are ribosomally produced antibiotics that contain a large number of post-translational modifications, such as dehydrated amino acids and thioether bridges. The thioether bridges form between cysteine residues and modified serine and threonine residues. NMR spectroscopy has been shown to be an important method to characterize the modified lantibiotic residues [e.g. van de Ven, F.J.M. and Jung, G. (1996) Antonie van Leeuwenhoek. 69(2), 99-107.].

The heterogeneous covalent structure of mutacin 1140 has been partially characterized by Edman sequencing, mass spectroscopy, and NMR spectroscopy. Edman sequencing of the ethanethiol derivatized mutacin 1140 has identified the sequence positions of the post-translationally modified amino acid residues. Chlorination of mutacin 1140 followed by Edman sequencing has revealed the possibility of heterogeneity. Mass spectroscopy and NMR further confirm the presence of heterogeneity. The covalent structure predicted by mass spectroscopy differs from the covalent structure predicted by NMR.

Based on these results, there appears to be a mixture of different covalent products in mutacin 1140. We predict that the structural variability is due to the cysteines forming alternative thioether bridging patterns among the different modified serine and threonine residues. We will make different site-directed mutations designed to force the modifications into different homogeneous forms. These mutants will be structurally characterized by NMR spectroscopy and assayed for biological function in vivo. In the future we will solve the three-dimensional structures of the wild-type and mutant mutacins in order to better understand the structural consequences of the heterogeneous covalent modifications.

035
Session Assigned: M&T P
Title: NMR and Molecular Modeling Studies on Beta-Estradiol, Genistein, and Daidzein

Institute: Chemistry, Long Island University, Brooklyn, NY
Keywords:
Abstract:
NMR and molecular modeling are used to study the 3-dimensional structures of beta-estradiol, genistein, and daidzein. Beta-estradiol is the drug of choice for the prevention of osteoporosis (OP) in postmenopausal women. The same number of women over 50 die from osteoporosis as die from breast cancer. Despite this, osteoporosis rarely gets the attention it deserves. A problem associated with the use of beta-estradiol is the increased risk of developing breast and uterine cancer. We are thus looking at genistein, an agonist of beta-estradiol, as a potential therapeutic. ¹H NMR spectra of 10mM genistein, daidzein, and beta-estradiol were acquired in DMSO. The 1D spectra were assigned using DQFCOSY and TOCSY 2D NMR data. A 1D ¹³C NMR spectrum of beta-estradiol was assigned and a ¹³C-{¹H} HETCOR was used to help make the ¹H assignments for this compound. The ¹H assignments are a prerequisite to obtaining 3D spatial information from the 2D ROESY data used in the molecular modeling. Preliminary results show structural similarities in conformation between beta-estradiol and its agonist, genistein. Comparisons are made to the control flavenoid, daidzein.

036
Session Assigned: M&T P
Title: STUDY OF INCLUSION COMPLEXES OF CYCLODEXTRINS WITH CHOLINESTERASE INHIBITORS BY T1, ROESY AND MOLECULAR MODELLING

Institute: Laboratoire de RMN-Biologique, ICSN, CNRS
Keywords: complexes, cyclodextrins, cholinesterase inhibitors, ROESY, MODELLING
Abstract:
Some derivatives of tetrahydrofuran A (Phenyl (OCH3)3 Tetrahydrofuran CH2OCO(CH2)n-Pyridinium+Br-), which have an activity against platelet aggregation (PAF antagonists) show also a consequential potential anticholinesterasic activity in the treatment of Alzheimer s disease. From molecular modelling investigations of potential cholinesterase inhibitors, previous studies in our laboratory have shown that some esters of type B have an anticholinesterasic activity comparable with Tacrin. We have also demonstrated that to reinforce the anticholinesterases of type B compounds (RCOO (CH2)6 Pyridinium+Cl-), acridin,which by itself has a strong anticholinesterasic activity, appears as a suitable R substituent. For an eventual pharmaceutical application of these compounds, the inconvenience is the very poor solubility of acridin in water (2.6 10^-4M). Inclusion complexation with cyclodextrins may eliminate such a problem. Cyclodextrins (Cds) are able to form inclusion complexes with aromatic molecules due to the influence of non covalent interactions. These complexes can improve the solubility and bioavailibility of molecules of pharmaceutical interest. To study the biosolubility of the anticholinesterasic and hydrophobic compounds, the inclusion of acridin within either an unsubstituded native cyclodextrin the betaCD, or a highly water-soluble methylated cylodextrin, the heptakis (2,6,di-O-methyl) cyclomaltoheptose (DMbetaCD) was studied.

NMR titration experiments (Bruker DMX 500) showed that the change in the chemical shift observed for specific protons of acridin upon the addition of varying concentration of cyclodextrin allowed an average complex stability constant K = 320 M for the system betaCD/acridin. The chemical shifts changes in the pure inclusion complex were also evaluated by fitting the experimental values to an adequate equation showing a 1:1 stoechiometry. T1 selective, non selective and null inversion were assessed to demonstrate the complexation. Geometrical features of the host-guest inclusion complexes were deduced from intermolecular dipolar interactions obtained by 2D adiabatic off-resonance ROESY experiments. Molecular diffusion DOSY experiments also showed the presence of a complex. A more detailed picture of the solution structure of the inclusion complexes was obtained by combining NMR structural information and molecular modelling (docking, dynamics under constraints and energy calculations with Tripos force field). The acridin molecule can be approximated by an ellipsoid which is partially inserted along its principal axis into the betaCD cavity. The results obtained for methylated cyclodextrin show that the inclusion of the acridin molecule is partially hampered by the formation of a hydrophobic plug on the OCH3 primary site and that a second molecule of acridin could interact with the DMbetaCD but remains outside the cavity.

037
Session Assigned: M&T P
Title: ³¹P NMR Studies of Infectious Parasites: Pyrophosphate and a Potentially New Route To Chemotherapy

Institute:
Keywords:
Abstract:
    We have used high field one- and two-dimensional ³¹P NMR spectroscopy to investigate perchloric acid extracts of Trypanosoma cruzi, Trypanosoma brucei, Leismania mexicana,and Toxoplasma gondii (the causative agents of Chagas Disease, African Sleeping Sickness, Leishmaniasis, and Toxoplasmosis), and found that they contain large stores of short chain polyphosphates, the majority of which are present as inorganic pyrophosphate, P₂O₇^4-. Specific enzymatic assays show that in the clinically relevant T. cruzi amastigote, inorganic pyrophosphate is approximately 10 times more abundant than ATP. NMR spectroscopy and electron microscopy of subcellular fractions indicate that pyrophosphate is mostly associated with acidocalcisomes, storage organelles that have been shown previously to contain high amounts of phosphorus, calcium, magnesium, sodium, and zinc.

Based on the hypothesis that this pyrophosphate is necessary for the survival of these organisms, a non-metabolizable pyrophosphate analog, 3-amino-1-hydroxypropane -1,1- diphosphonic acid (pamidronate, Aredia), was assayed by uracil incorporation and Giemsa staining for inhibitory activity on in vitro cultures of T. cruzi amastigotes and T.gondii tachyzoites, and found to have a 50% inhibitory concentration (without toxicity) of 65 µM and 45 µM, respectively. While the protozoal targets for this drug are as yet not all well defined, bis-phosphonates are FDA approved compounds used routinely in simliar dosage in the treatment of various bone disorders, and offer a potentially new route to chemotherapy of these infectious diseases.

038
Session Assigned: M&T P
Title: Conformational Changes in the H+ Translocating Subunit c of the F1FO ATPase

Institute: Albert Einstein College of Medicine, Bronx, NY USA
Keywords: membrane protein; structure
Abstract:
F₁F_O ATP synthases use a transmembrane H⁺ gradient to drive the synthesis of ATP from ADP and P_i. Subunit c of the membranous F_O portion of the enzyme is involved in both H⁺ translocation, and in conformational coupling. The Asp61 side chain of subunit c is responsible for transporting H⁺ across the membrane. Changes in the protonation state of this side chain lead initially to conformational changes in the cytoplasmic-exposed loop, which ultimately drive conformational changes at the distant catalytic sites of the enzyme resulting in release of ATP product. Purified subunit c folds normally in a mixed solvent system (4:4:1 CHCl₃:MeOH:H₂O). We have previously determined its structure at pH 5.0, where Asp61 is protonated. Here we present complete ¹H, ¹³C, and ¹⁵N assignments and initial structure calculations for the protein at pH 8.0, where Asp61 is ionized. A structural comparison of subunit c at pH 5.0 and 8.0 will show what conformational changes take place near Asp61, and how these changes are linked to structural changes in the cytoplasmic loop of the subunit.

039
Session Assigned: M&T P
Title: NMR and CD Spectroscopic Studies on Changes in Secondary Structures of the Sodium Channel Inactivation Gate Peptides as Caused by the Pentapeptide KIFMK

Institute: Kyoto University, Kyoto, Japan
Keywords: sodium channel; KIFMK; III-IV linker
Abstract:
The pentapeptide, KIFMK, which contains three clustered hydrophobic amino acid residues of isoleucine, phenylalanine, and methionine (IFM) in the sodium channel inactivation gate on the cytoplasmic linker between domains III and IV (III-IV linker), is known to restore the fast inactivation to the mutant sodium channels having a defective inactivation gate. In order to investigate the docking site of KIFMK and to clarify the mechanisms for restoring the fast inactivation to the sodium channel, we have studied the interactions between the KIFMK and the fragment peptide in the III-IV linker, GGQDIFMTEEQK (MP-1A; G1484-K1495 in rat brain IIA), by one- and two-dimensional NMR and CD spectroscopies. It was found that KIFMK increases the helical content of MP-1A in TFE solution. In contrast, KDIFMTK, which is ineffective to restore inactivation, decreased the helical content. Moreover, KIFMK increased the helical content of MP-1EQA (E1492Q), while decreased those of MP-1NA (D1487N) and MP-1QEA (E1493Q). Thus, KIFMK, by allowing each Lys residue to interact with D1487 and E1492, respectively, stabilized the helical structure of the III-IV linker at around the IFM residues. In conclusion, this helix stabilizing effect of KIFMK on the III-IV linker restored and/or promoted the fast inactivation to the sodium channels having a defective inactivation gate.

040
Session Assigned: M&T P
Title: Structural basis of precursor protein processing using NMR spectroscopy

Institute: University of Florida
Keywords:
Abstract:
FMRFamide-like peptides (FLPs) form a large family of neuropeptides which have been observed in all animals, and are obtained by post-translational modifications of precursor proteins. The mature peptides have potent and diverse biological activities. Some precursor proteins contain as many as 28 copies of closely related peptide sequences (and thus have not been well optimized for NMR studies!) that are separated by basic amino acids (K, R, KR, or RK). No three-dimensional structure of a FLP precursor protein has been solved. Our overall goal in this study is to learn how the three-dimensional structures of precursor proteins influence peptide processing. These precursor proteins are not particularly simple to study, since they contain multiple proteolytic sites and large amounts of highly repetitive amino acid sequences. Therefore, we are approaching the problem from two ends. First, we are "building up" synthetic peptide fragments with a small number of unprocessed neuropeptides and processing sites. Second, we are expressing full-length precursor proteins from two different genes with the ultimate aim of solving their structures and conducting in vitro processing studies.

The FLP precursor protein from the snail Lymnaea stagnalis (FLP-Ls) contains 7 RGD sequences, which also form cleavage sites. Peptides with RGD sequences bind to integrin receptors and have been the subjects of numerous NMR investigations. Those studies have established that RGD-containing sequences form reverse turns in solution. It has also been suggested (but less firmly established) that reverse turns are necessary for proper proteolytic processing of precursor proteins. Initial NMR studies on a portion of FLP-Ls with two identical halves separated by an RGD site (GDPFLRFGRGDPFLRF-NH2) reveals distinct chemical shifts for each half of the peptide and the presence of secondary structure by NOEs. We are investigating additional extensions of the repeating sequence GDPFLRFG to see if the pattern of distinct chemical shifts continues. These studies are providing important initial data on relatively simple samples.

We will also report our progress on the more complicated samples of complete precursor proteins with either 6 or 8 repeating peptide fragments. Both proteins (afp-1 from Ascaris and flp-1 from C. elegans) are about 160 amino acids long and have two main halves, a peptide-encoding half and a non-peptide-encoding half. The flp-1 protein encodes 8 FLPs, all of which share the common sequence PNFLRF. The afp-1 protein encodes 6 FLPs, all with the common sequence PGVLRF. Both proteins have been expressed in bacteria and are ready for NMR and processing studies.

041
Session Assigned: M&T P
Title: Comparision of Solution Structures of Two Analogues of Cecropin B

Institute: Dept.of Chemistry,Natl.Tsing Hua University,Taiwan
Keywords:
Abstract:
As a self-defense response against bacterial infections, insects produce several antibacterial proteins. Of these Cecropins are known to be highly potent. In the present study we have chemically synthesized two analogs of Cecropin B (from Bombyx mori ), namely Cecropin B1 and B3 (CB-1 and Cb-3). Both CB-1 and CB-3 are 34 amino acid long peptides, deviod of cysteine residues. CB-1 and CB-3 are known to exhibit significant difference(s) in their membrane lytic properties. To understand the structural basis for the disparity in the membrane lytic actions of these Cecropin analogs, we have determind the three-dimensional, solution structures of CB-1 and CB-3 using homonuclear, multidimensional NMR spectroscopy and dynamic simulated annealing techniques. Although the solution structures of both CB-1 and CB-3 reveal two distinct helical segments, there exist subtle but interesting differences in their backbone and side-chain conformations. The structure-function relationship of CB-1 and CB-3 would be discussed in greater detail.

042
Session Assigned: M&T P
Title: Conformational Studies of the MCR-4 Cyclic Peptide Agonist and Antagonist by NMR

Institute: Pfizer Inc., Groton, CT, USA
Keywords: NMR;simulated_annealing;MCR-4;cyclic_peptides
Abstract:
The melanocyte-stimulating hormone receptor (MCR-4), a G-coupled receptor, is a molecular target for obesity. Two cyclic peptides which bind to the MCR-4 receptor and act as an agonist or antagonist have been identified (1). The synthesized agonist and antagonist peptides are cyclized through the side chains of Asp and Lys to form a lactam bridge, and differ only in the identity of one amino acid. MTII - agonist: Ac-Nle-Asp-His-dPhe-Arg-Trp-Lys-NH2 SHU-9119 - antagonist: Ac-Nle-Asp-His-dNal-Arg-Trp-Lys-NH2 The structures of both peptides were determined by NMR to identify any structural differences for the agonist and antagonist with particular emphasis on the conformation of residues, -His-dPhe-Arg-Trp-, shown to be necessary for agonist activity. Any structural differences may be useful in the design of small molecule agonists and antagonists. NMR studies were carried out by 2D homonuclear techniques and structures were determined in aqueous solution as well as DMSO, and refined by simulated annealing methods. Complete analysis of structural information by NMR, including interproton distances and dihedral angles, indicates that these peptides are highly restrained in solution. The conformation of both peptides will be presented and the structural differences will be discussed.

1) Fan, W. et. al. Nature, Vol 385, 1997, 165-168

043
Session Assigned: M&T P
Title: Characterization of Interface Peptides Interactions by NMR and Molecular Dynamics

Institute: Portland State University, Portland, OR
Keywords: aggregation dimerization peptide protein
Abstract:
The NPs are homodimeric proteins that exhibit thermodynamic linkage between dimerization and their biological function, the binding of peptide hormones oxytocin (OT) and vasopressin (VP). The dimerization interface from the x-ray structure consists of residues at both of the NP domains. Here we demonstrate the use of interface peptides to study the dimerization of neurophysin (NP) by NMR spectroscopy. Multiple interface peptides derived from the NP monomer-monomer contact region have been synthesized. These peptides range from seven to more than twenty amino acids. Preliminary studies showed that the presence of interface peptides derived from N-terminal domain of NP interact with NP. Interface peptide interactions with NP can provide a monomeric NP model to study the dimerization and the structure of the larger dimeric protein. In order to understand to what extent the secondary structures of these peptides stabilize the peptide-bound NP monomer, the structures of these peptides have been studied using NMR and molecular dynamics calculations, both alone and in the presence of NP. The NMR signals of these peptides were assigned using TOCSY and NOESY experiments. NOE distance constraints were calculated and X-PLOR was employed to characterize the tertiary structures. The results obtained from these studies will identify the specific features of the amino acids which are necessary for dimerization of NP. Furthermore, the results may aid in understanding the protein aggregation problem.

044
Session Assigned: M&T P
Title: A Hybrid Approach to Conformational Flexibility in Short Peptides

Institute: University of Florida, Gainesville, FL USA
Keywords:
Abstract:
The numbers and functions of small biologically active peptides are increasing dramatically in all areas of biology. In particular, large numbers of neuropeptides ranging from 4 to about 15 amino acids are being discovered through molecular biology or genome sequencing projects. As an example, Caenorhabditis elegans, a small nematode with ~300 neurons, has well over 50 similar peptides which all end in the amino acid sequence "Arg-Phe-NH2". The functions of these peptides, when known, range from excitation or relaxation of muscle cells to modulation of feeding behavior. In mammals similar peptides modulate the action of morphine, the heart rate, and blood pressure.

We have recently collected NMR data on a series of very similar peptides whose receptor binding affinities have been measured. We have found a very convincing inverse linear relationship between the population of reverse turn (as determined by ROE intensities and shifts in pH titrations) and the receptor binding affinity. This structure/function relationship is intriguing, because multiple similar peptides are often released at the same site. We proposed a model in which the conformations of certain peptides in a mixture will lower their effective concentration in solution with respect to a particular receptor, thus changing the peptide "bouquet" that is presented to receptors.

The problem with the experimental data is that it is not enough to fully characterize the peptide structures and it is conformationally averaged. We are taking a hybrid approach to this problem. First, an interactive clustering algorithm is used to separate clusters of low energy structures obtained from multiple independent Monte Carlo simulated annealing (MCSA) conformational searches at the molecular mechanics level of approximation. Each MCSA conformational search starts from a random conformation. For our initial studies, the conformational searches are carried out in the absence of any experimental NMR data. Representative conformations from the lowest energy clusters are geometry optimized using Density Functional Theory (DFT). Using the DFT energies and final geometries, we will calculate Boltzmann averages, from which the NMR parameters will be calculated and compared to experimental measurements.

045
Session Assigned: M&T P
Title: Heteronuclear NMR studies of the A. fulgidus endonuclease III - DNA complex

Institute: University at Albany - SUNY, Albany, NY
Keywords: protein - DNA interactions
Abstract:
Endonuclease III is the prototype of [4Fe-4S] family of DNA repair enzymes with a broad specificity for damaged bases in DNA. It specifically recognizes oxidized pyrimidines and is involved in the DNA base excision repair pathway. The enzyme can excise damaged bases through glycosylase activity and cleave the resulting AP (apyrimidinic) sites by its AP-lyase activity. Biochemical studies of AP-DNA - Endo III interactions suggest that the AP-lyase mechanism proceeds through the formation of a covalent protein - AP-DNA intermediate involving the invariant residue, Lys120. Though the crystal structure of the free E. coli Endonuclease III was solved ten years ago, there are no co-crystal structures of the enzyme bound to its DNA substrate, leaving many questions open on the structural basis of molecular recognition and catalysis.

To structurally characterize the Endo III-DNA interaction by multidimensional NMR, we cloned, overexpressed, and purified Endonuclease III from the thermophile Archeoglobus fulgidus. Endonuclease III of A. fulgidus is an alpha helical protein with molecular weight of 24 kDa and the secondary structure is similar to E.coli Endo III as revealed by [¹H-¹⁵N] NOESY-HSQC data and chemical shift indices. Nearly complete assignments of free [U-¹⁵N, ¹³C] Endo III were achieved using a series of triple resonance experiments (HNCA, HN(CO)CA, HNCO, HNCACO, and HNCACB) combined with [¹H-¹⁵N] HSQC experiments on specifically labeled Endonuclease III. We chose to use a reduced AP-containing oligonucleotide (rAP-ds13mer) as a noncleavable substrate to study protein-DNA interaction. Titration of rAP-ds13mer into the protein solution revealed that Endonuclease III is in slow exchange with its substrate. At 30 ^oC, the protein-DNA complex is well-behaved and amenable to structural NMR studies. We monitored changes in chemical shifts in HNCO spectra of free and bound Endonuclease III to map the protein-DNA binding surface. Definition of the structural characteristics of the protein binding surface and description of the changes in Watson-Crick base pairing of rAP-13mer upon complex formation suggests a possible recognition mechanism of the damaged DNA.

046
Session Assigned: M&T P
Title: Structure, Interactions, and Dynamics of the Central Domain of Human DNA Repair Protein XPA

Institute: Nara Institute of Science and Technology, Japan
Keywords:
Abstract:
Ultraviolet light and some chemicals frequently cause damages on DNA which can potentially be mutagenic or lethal to cells. Nucleotide excision repair (NER) is the ubiquitous pathway by which various kinds of damages on DNA are removed from the genome. Defects in NER give rise to an inherited disease, xeroderma pigmentosum (XP), which is characterized by high frequency of skin cancer on sun-exposed areas and by neurological complications. About 30 protein factors are thought to be involved sequentially in the human NER system. XPA is one of these NER factors, and has been shown to be involved in the damage recognition step of NER. XPA binds to other repair factors, and acts as a key element in NER complex formation. The central domain of human XPA consisting of 122 amino acids from Met98 to Phe219 is responsible for the preferential binding to damaged DNA and to another NER factor, replication protein A (RPA).

We present the three-dimensional structure [1], interactions with DNA and RPA [1], and dynamic properties [2] of the central domain of human XPA. Almost all of the resonances of the backbone and sidechain ¹H, ¹⁵N and ¹³C nuclei were assigned. Structures were calculated by the simulated annealing procedure with 1,389 distance and 83 angle constraints. The structures showed that the domain consists of a zinc-containing subdomain with a zinc ion tetrahedrally coordinated by four Cys residues, a C-terminal subdomain with a positively charged cleft in a novel structure, and a linker sequence connecting these two subdomains. Chemical shift perturbations of ¹⁵N-labelled XPA with a domain of RPA, RPA70_181-422 and with a 24-mer synthesized DNA showed that the zinc-containing subdomain binds to RPA70 and the cleft in the C-terminal subdomain binds to DNA. ¹⁵N T₁, T₂, and NOE relaxation data were obtained at two static magnetic fields, and analyzed with the model-free formalism under the assumption of isotropic or anisotropic rotational diffusion. Exchange contributions were also estimated by analysis of the spectral density function at zero frequency. The results showed that the domain exhibits a rotational diffusion anisotropy of 1.38, and most of flexible regions exist on the DNA binding surface in the cleft in the C-terminal subdomain. This flexibility may be involved in the interactions of XPA with various kinds of damaged DNA.

[1] Ikegami, T. et al. (1998) Nature Struct. Biol. 5, 701-706.
[2] Ikegami, T. et al. (1999) J. Biochem., in press.

047
Session Assigned: M&T P
Title: NMR Structure Determination of a Novel Metalloenzyme

Institute: MIT, Cambridge, MA, USA
Keywords: biochemical oxidation
Abstract:
A novel metalloenzyme with two distinct isoforms has been identified in the methionine salvage pathway of Klebsiella pneumoniae. The enzyme, presently named E2, has been cloned and expressed in E. coli. Surprisingly, induction of the cloned gene results in overexpression of two distinct isoforms, E2 and E2^|, which are separable by chromatography. The two forms differ (at least in composition) only by their metal ion contents. Both enzymes catalyze the oxidative degradation of 1,2-dihydroxy-3-keto-5-methylthiopentene anion, but the product of the degradation depends upon the identity of the metal ion bound in the active site (Dai & Abeles, in preparation). Ni⁺² bound in the active site results in the formation of carbon monoxide and formate from the degradative process. If Fe⁺² is bound to the enzyme, only formate is produced. Multidimensional NMR methods are being used to characterize the structural differences between the two forms of the enzyme. Nearly complete ¹H, ¹³C and ¹⁵N resonance assignments have been made for the native Ni⁺² containing E2 enzyme, and a global fold for the enzyme has been determined. There is preliminary evidence that structural differences between the two isoforms may be due to the presence (in E2) or absence (in E2^|) of a disulfide bridge between the only two cysteines in the protein.

048
Session Assigned: M&T P
Title: Long Range Distance Constraints in Perdeuterated Back-Exchanged DNA-Binding Proteins by 15N-Edited NOESY

Institute: Duke University, Durham, North Carolina, U.S.A.
Keywords: global fold
Abstract:
Using 3D and 4D 15N-edited NOESY, we are studying the range of dipolar cross-relaxation that can be observed in perdeuterated back-exchanged proteins at high field. Such observations provide long range HN-HN distance restraints that are valuable in deducing the global main-chain folds of proteins. NOE build-up studies suggest that the low density of proton spins in proteins thus labeled allows direct dipolar cross-relaxation to be observed using mixing times of four to five hundred milliseconds. To correlate NOE signal intensities with the relatively long distances observed, we have employed an internal distance calibration method based on the sequence/distance relation in presumed helical regions. These methods have been applied to the human nucleotide excision repair protein XPA, in which a 15 kDa domain binds preferentially to damaged DNA. This domain was metabolically labeled with 15N and 2H and studied at 600 and 750 MHz. A set of HN-HN NOEs observed at 600 MHz allowed the global fold of this domain to be deduced. It was found to fold as two sub-domains, one of which contains a Zn-finger motif. Similar studies are being carried out on metJ, the repressor of the E. coli methionine operon.

049
Session Assigned: M&T P
Title: Structural Alteration in vnd/NK2 Homeodomain by an Embryonically Lethal Single Ala->Thr Mutation

Institute: National Institutes of Health, Bethesda, MD, USA
Keywords:
Abstract:
Homeodomain containing proteins are conserved throughout evolution, bind DNA, act as transcription regulators and are involved in various aspects of development. Mutations in the homeodomain are associated with abnormal development and genetic disease across all species including human. The high degree of structural homology seen thus far for the homeodomain both in the free and DNA bound states suggests the possibility of some common rules in the study of the structural basis of developmental abnormalities and genetic disease. In this study, the three-dimensional structure of an embryonically lethal single amino acid mutation from alanine to threonine in position 35 of the vnd/NK2 homeodomain from Drosophila Melanogaster is described. From the structure of the wild type homeodomain in the absence of DNA, it is known that alanine-35 is a surface exposed residue where replacement with threonine would be expected to result in minimal structural disruption. However, in the free state, the threonine-35 mutant vnd/NK2 homeodomain is unstructured, even at 0 ^oC, whereas the corresponding wild type homeodomain is structured below 25 ^oC. The mutant homeodomain binds an 18-base-pair vnd/NK2 target DNA sequence with an affinity that is 50 fold lower than that of the corresponding wild type homeodomain. The mutant homeodomain maintains the canonical helix-turn-helix motif although there are significant distortions in the relative orientations of helix I and helix II. It is possible that the magnitude of the hydrophobic effect associated with the threonine side-chain methyl group is sufficient to produce the observed structural alterations.

050
Session Assigned: M&T P
Title: The Secondary and Tertiary Structures of the D-Alanyl Carrier Protein Established by Multinuclear NMR

Institute: Northwestern University, Evanston, IL
Keywords:
Abstract:
The incorporation of D-alanine into membrane-associated D-alanyl-lipoteichoic acid (LTA) in gram-positive organisms requires the 56 kDa D-alanine--D-alanyl carrier protein ligase (Dcl) and the 8.9 kDa D-alanyl carrier protein (Dcp). It was proposed that Dcp is a homolog of the acyl carrier protein (ACP) involved in fatty acid biosynthesis (Debabov et al. J. Bacteriol. 178:3869-3876, 1996). However, because the sequence identity between Dcp and ACP is only 20%, it was necessary to compare the three dimensional structures of Dcp and ACP. Since ACPs have not been crystallized, NMR was the method of choice for establishing both the secondary and tertiary structures of Dcp. To accomplish this goal, samples of ¹⁵N and ¹⁵N,¹³C labeled apo-Dcp were prepared with the plasmid pDCP1 containing dltC in BL21(DE3) for multidimensional NMR. This paper will report the complete assignment of all spin systems, the determination of the secondary structure, and will compare the three dimensional structure with that of ACP. Helix 1 was assigned to D3 through L17, helix II to M40 through Q52, and helix III to N71 through Q79. The .alpha-helical content measured by NMR is similar to the content calculated from CD (53%) and is similar to that found for ACP (47%). In contrast to ACP, Dcp does not have two metal binding sites. The assignments of W67, F32, F53, and F62 provided a group of aromatic amino acid residues for monitoring the denaturation of apo-Dcp by NMR in the presence and absence of Ca²⁺. While the thermal transition of Dcp is higher than that of ACP, there is no stabilization of the Dcp structure by divalent metal cations as measured by NMR. Thus, while there are significant specificity differences in the D-alanine incorporation reaction between the two carrier proteins, the present structural studies support the conclusion that Dcp is homologous to the ACP involved in fatty acid biosynthesis.

051
Session Assigned: M&T P
Title: Secondary Structure and Allosteric Activation of cAMP Receptor Protein(CRP) by the Binding of cAMP

Institute: Seoul National University, Seoul, Korea
Keywords:
Abstract:
Cyclic AMP receptor Protein(CRP) from E. coli plays a key role in regulation of the expression of more than 20 genes of the bacterium. CRP binds in the presence of cAMP to a specific target site near the promoter of each gene under its regulation. CRP is a dimer (Mr~47,000) of two identical subunit. In order to elucidate the structure-function relationship of CRP, we have taken NMR study of this protein and its complex with cAMP, oligonucleotides and RNA polymerase .alpha subdomain.

To assign the resonances of protein, we used stable isotope labeling(¹⁵N, ¹³C, Deuterium) labeling techniques. 3D-HNCA, HN(CA)CB, HN(COCA)CB, HNCO, NOESY-HSQC experiments were acquired with deuterated sample. ¹H-¹⁵N HSQC spectra were recorded respectively for 12 amino-acid specific ¹⁵N-labeled samples. NMR signals could be assigned by using these techniques and secondary structural elements were calculated by CSI method. NMR signals monitored in this study indicate that the reorientation of two cAMP binding domains in CRP occurred by the binding of cAMP. Subsequently, the two DNA binding domains become closer automatically, and also the loop around .beta-strands 9 and 10 is exposed to the surface to make a direct protein-protein interaction with RNA polymerase.

052
Session Assigned: M&T P
Title: Backbone Chemical Shift Assignments and Non-target DNA Binding Study on T4 Enconuclease V by NMR

Institute: Seoul National University, Seoul, Korea
Keywords:
Abstract:
T4 Endonuclease V (M.W. 16 kDa) is a DNA repair enzyme from bacteriophage T4 which catalyzes the first reaction step of the pyrimidine-specific base excision repair pathway. As the first step to understand the solution structure and the mechenism of DNA scanning and recognition of this enzyme at atomic level, we have fulfilled NMR backbone peak assingments using 3D double and triple resonance and amino acid-specific ¹⁵N-labeling techniques. Total 14 amino-acid specific ¹⁵N-labeled samples were made and ¹H-¹⁵N HSQC spectra were recorded respectively. We could assign the ¹HN, ¹⁵N, ¹³C.alpha, ¹³CO backbone resonances and secondary structural elements were calculated by CSI(Chemical Shift Index) method. To study the DNA recognition of T4 Endonuclease V, we added normal -TT- DNA oligomer to the ¹⁵N-arg, ¹⁵N-tyr and ¹⁵N-thr labeled enzyme and recorded ¹H-¹⁵N HSQC spectra respectively. In the case of ¹⁵N-thr, the peak of thr-58 disappeared. Thr-58 would play an important role when T4 Endonuclease V recognizes -TT- DNA semi-specifically.

053
Session Assigned: M&T P
Title: A novel DNA recognition mode adopted by the three-stranded .beta-sheet in the GCC-box binding domain

Institute: AIST-NIBHT, Tsukuba, Japan
Keywords: Protein-DNA complex
Abstract:
The 3D solution structure of the GCC-box binding domain (GBD) of a transcription factor from Arabidopsis thaliana in the complex with its target DNA fragment has been determined by heteronuclear multidimensional NMR in combination with simulated annealing and restrained molecular dynamic calculation. The GBD domain consists of a three-stranded anti-parallel .beta-sheet and an .alpha-helix packed approximately parallel to the .beta-sheet. Arginine and tryptophan residues in the .beta-sheet are identified to contact eight of the nine consecutive basepairs in the major groove, and at the same time bind to the sugar-phosphate backbones. The target DNA bends slightly at the central CG step thereby allowing the DNA to follow the curvature of the .beta-sheet.

054
Session Assigned: M&T P
Title: NMR Studies of Muty, an Adenine Glycosylase DNA Repair Enzyme Active on G-A Mismatches

Institute: University of Texas Medical Branch, Galveston TX
Keywords:
Abstract:
MutY is an adenine glycosylase DNA repair enzyme which is active on G-A mismatches, and its N-terminal domain, whose structure has been solved by X-ray crystallography, has homology to endonuclease III. However, the structure of the C-terminal domain has proven difficult to solve by crystallographic methods. We have applied a variety of 3D NMR experiments, including the HNCACB, HNCO, HCCH-TOCSY and N-15,C-13 edited NOESY experiments, to elucidate most of the resonance assignments of the C-terminal domain of MutY. Well defined helices and beta-strands have been determined and we are currently determining the tertiary fold of MutY.

055
Session Assigned: M&T P
Title: NMR Structure of Human Fibroblast Collagenase Complexed with a Sulfonamide Derivative of a Hydroxamic Acid Compound

Institute: Wyeth-Ayerst Reseach, Pearl River, NY
Keywords: NMR Solution Structure; Matrix Metalloproteinase; Hydroxamic acid
Abstract:
The solution structure of the catalytic fragment of human fibroblast collagenase (MMP-1) complexed with a sulfonamide derivative of a hydroxamic acid compound (CGS-27023A) has been determined using two-dimensional and three-dimensional heteronuclear NMR spectroscopy. The solution structure of the complex was calculated by means of hybrid distance geometry-simulated annealing using a combination of experimental NMR restraints obtained from the previous refinement of the inhibitor-free MMP-1 (1) and recent restraints for the MMP-1:CGS-27023A complex. The hydroxamic-acid moiety of CGS-27023A was found to chelate to the "right" of the catalytic zinc where the O-methoxy-phenyl sits in the S1’ active site pocket, the isopropyl group is in contact with H83 and N80 and the pyridine ring is essentially solvent exposed. The sulfonyl oxygens are in hydrogen-bonding distance to the backbone NHs of L81 and A82. This is similar to the conformation determined by NMR of the inhibitor bound to stromelysin (2). A total of 48 distance restraints were observed between MMP-1 and CGS-27023A from 3D 13C-edited/12C-filtered NOESY and 3D 15N-edited NOESY experiments. An additional 16 intra-molecular restraints were observed for CGS-27023A from a 2D 12C-filtered NOESY experiment. A minimal set of NMR experiments were used to assign the MMP-1 1H, 13C and 15N resonances in the MMP-1:CGS-27023A complex based on the free MMP-1 assignments. The assignments of CGS-27023A in the complex were obtained from 2D 12C-filtered NOESY and 2D 12C-filtered TOCSY experiments.

(1). Moy, F. J., Chanda, P. K., Cosmi, S., Pisano, M. R., Urbano, C., Wilhelm, J., and Powers, R. (1998) Biochemistry 37, 1495-1504
(2). Gonnella, N. C., Li, Y.-C., Zhang, X., and Paris, C. G. (1997) Bioorg. Med. Chem. 5, 2193-2201

Institute: Rutgers University, Piscataway, NJ, USA
Keywords:
Abstract:
The Lipari-Szabo model-free formalism [J. Am. Chem. Soc. 104: 4546] and its extensions [J. Am. Chem. Soc. 112: 4989] have proven to be extremely popular for the analysis of NMR relaxation data. However, as a prerequisite to the model-free analysis of internal motions, one must first obtain a description of the overall tumbling of the macromolecule in solution. Methods currently used to estimate these global parameters involve assumptions which may not always be valid.This can produce unreliable estimates of the values and uncertainties of the global parameters. Since the local and global parameters often exhibit substantial statistical correlations, unreliable estimates of the global parameters can translate into unreliable estimates of the local dynamic parameters and their uncertainties.

We present a general data analysis formalism based on products of Bayesian marginal probability densities which can be used to efficiently combine the information content from multiple experiments, such as R₁, R₂, and NOE data collected at multiple magnetic field strengths, or data from cross-correlation or rotating frame relaxation dispersion experiments. We show how such additional data makes model selection unnecessary and avoids assumptions which are made in the commonly-used implementation of the model-free formalism. By expressing the estimates of the dynamical parameters as probability distributions in the parameter space we avoid the need for repeated nonlinear optimizations, resulting in substantial computational savings compared to an equivalent classical statistical approach. Furthermore, the resulting probability distributions allow for easy visualization of the information content of the data and detection of outliers due to systematic errors or model inadequacy. We demonstrate our approach using the simple and extended Lipari-Szabo formalisms in the presence of isotropic and anisotropic tumbling.

057
Session Assigned: M&T P
Title: Dynamic Fluctuations Elucidated Using Heteronuclear Spin Probes

Institute: University of Virginia
Keywords:
Abstract:
Magnetic Resonance Dispersion, MRD, is a benefical tool in the understanding of molecular dynamics of solutions, small molecules and proteins. Previous work from this lab has focused on studying solution and small molecule dynamics using a two-magnet dispersion instrument consisting of a variable electromagnet (0-1.4 T), and a superconducting solenoid magnet at 7.05 T, which provides superior signal and resolution compared to solenoid field switching instruments. The spin system is equilibrated in the superconducting magnet providing maximum signal and then moved adiabatically and rapidly to a lower field for varying times. The longnitudinal relaxation time, T₁, is then acquired for different field strengths resulting in a spectral density plot of the interacting system for the desired probe nuclei. Current experiments have studied translational, rotational, hyperfine coupling, and exchange correlation times, t_c, in a dynamic range of 10µs to 10 ps by studing nuclei such as ¹⁹F, ¹H, ³Li, ³¹P and ¹¹³Cd. The current focus of this work is on methods for studying dynamic protein intramolecular fluctuations to elucidate binding site dynamics and enzyme-substrate interactions.

058
Session Assigned: M&T P
Title: Influence of CSA-CSA cross-correlation on multiple-quantum relaxation rates

Institute: Leicester University
Keywords:
Abstract:
The relaxation of multiple-quantum coherence due to intramolecular dipole-dipole and chemical shift anisotropy (CSA) relaxation mechanisms is described. It is shown that cross-correlation between the CSAs of the active spins can affect the relaxation of multiple-quantum coherences. An experiment that enables this effect to be monitored for heteronuclear two spin coherences is discussed. Heteronuclear multiple-quantum data is presented for Human light chain 1 protein and protein G. The data can be analysed to yield either a value for the CSA of one of the spins that give rise to the CSA-CSA cross correlation term, or information on the relative orientations of the CSA tensors of the two spins concerned.

References

R. Konrat and H. Sterk, Chem. Phys. Lett. 203, 75 (1993).

Session Assigned: M&T P
Title: Concerted use of CSA and DD relaxation in attempts to separate geometry and dynamic parameters in liquids

Institute: L. Kossuth University, H-4010 Debrecen, HUNGARY
Keywords: CSA/DD relaxation; cross-correlation; dynamics
Abstract:
Gradient- and sensitivity-enhanced experiments (1D and 2D variants) are proposed in both laboratory and rotating frame for measuring longitudinal and transverse cross-correlation rates between chemical shift anisotropy (CSA) and dipolar (DD) mechanisms (1,2). The experiments have been tested for medium sized molecules at natural abundance of 13C/15N nuclei, and also for 15N-labeled proteins.

Simultaneous evaluation of the longitudinal and transverse CSA/DD cross-correlated relaxation rates with the conventional heteronuclear relaxation parameters (T1, T2 and NOE) using an extended Lipari-Szabo (3) analysis may improve the accuracy of the obtained dynamic parameters and potentially allows the separation of the CSA magnitude and angular terms.

A dynamics- and exchange contribution-free evaluation of CSA from the longitudinal relaxation rates will be also considered.

1) Werbelow, in "Encyclopedia of NMR" eds. Grant & Harris (Wiley, New York), Vol. 6 pp. 4072-4078 (1996)Farrar & Stringfellow, in "Encyclopedia of NMR" eds. Grant & Harris (Wiley, New York), Vol. 6 pp. 4101-4107 (1996), and other refs. cited there
2) Tjandra et al. (1996), J. Amer. Chem. Soc. 118, 6986-6991. Tessari et al. (1997), J. Magn. Reson. 127, 128-133. Brüschweiler & Ernst (1992), J. Chem. Phys. 96, 1758-1766. Kroenke et al. (1998), J. Amer. Chem. Soc. 120, 7905-7915. Batta et al. (1999), J. Magn. Reson. in press Fushmand & Cowburn (1998), J. Amer. Chem. Soc. 120, 7109-7110.
3) Lipari & Szabo (1982), J. Amer. Chem. Soc. 104, 4559-4570.

060
Session Assigned: M&T P
Title: Nuclear Spin Relaxation and the Molecular Diffusion Tensor: Effects on Dipolar, CSA and Dipolar-CSA Interference Rates

Institute: Yale University, New Haven, CT USA
Keywords:
Abstract:
The theoretical development of nuclear spin relaxation is presented in the context of an arbitrary rotational diffusion tensor in the solution state. The effect of dipolar, CSA and dipolar-CSA interference relaxation is discussed using a consistent set of relaxation formalisms from the review by Spiess¹. The respective orientations of all interactions relative to the diffusion frame are used to give a more accurate description of the relaxation processes, in particular the orientation of the CSA tensor is specifically calculated with respect to the diffusion frame. In this manner the coupled relaxation matrix for a molecule is constructed, from which transient NOE, T1 and T2 information can be extracted. A computer program, entitled Yet Another Relaxation Matrix (YARM), was developed to perform these calculations, and will be presented. The relaxation rates of a variety of samples, from a two-spin (1H-13C) pair to a large multispin molecule, such as a DNA, were simulated in order to investigate the importance of these processes on measurable NMR relaxation parameters. We hope to demonstrate the potential of YARM as a research tool for relaxation in macromolecules.

¹Spiess, K. W. In NMR: Basic Principles and Progress, Diehl, P., Fluck, E., Kosfeld, R. Ed.; Springer-Verlag: NY, NY, 1978.