Structure and function of the C-terminal PABC domain of human poly(A)-binding protein.

We have determined the solution structure of the C-terminal quarter of human poly(A)-binding protein (hPABP). The protein fragment contains a protein domain, PABC [for poly(A)-binding protein C-terminal domain], which is also found associated with the HECT family of ubiquitin ligases. By using peptides derived from PABP interacting protein (Paip) 1, Paip2, and eRF3, we show that PABC functions as a peptide binding domain. We use chemical shift perturbation analysis to identify the peptide binding site in PABC and the major elements involved in peptide recognition. From comparative sequence analysis of PABC-binding peptides, we formulate a preliminary PABC consensus sequence and identify human ataxin-2, the protein responsible for type 2 spinocerebellar ataxia (SCA2), as a potential PABC ligand.

Structural proteomics of an archaeon.

A set of 424 nonmembrane proteins from Methanobacterium thermoautotrophicum were cloned, expressed and purified for structural studies. Of these, approximately 20% were found to be suitable candidates for X-ray crystallographic or NMR spectroscopic analysis without further optimization of conditions, providing an estimate of the number of the most accessible structural targets in the proteome. A retrospective analysis of the experimental behavior of these proteins suggested some simple relations between sequence and solubility, implying that data bases of protein properties will be useful in optimizing high throughput strategies. Of the first 10 structures determined, several provided clues to biochemical functions that were not detectable from sequence analysis, and in many cases these putative functions could be readily confirmed by biochemical methods. This demonstrates that structural proteomics is feasible and can play a central role in functional genomics.

Design and solution structure of functional peptide mimetics of nerve growth factor.

The C-D loop in nerve growth factor (NGF) is involved in binding to the NGF receptor, TrkA. It is flexible and adopts several different types conformations in different NGF crystal forms. We have previously shown that a small cyclic peptide derived from the C-D loop of NGF binds to the TrkA receptor by mimicking the structure of this loop. To understand structure-function relationships in NGF C-D loop mimetics, we have produced a series of peptides predicted to form different types of beta-turns. The peptides were tested for their ability to promote cell survival in serum-free medium and to induce TrkA tyrosine phosphorylation. NMR structural studies were used to determined the backbone conformation and the spatial orientation of side chains involved in binding to the TrkA receptor. Peptides that form type I or type gammaL-alphaR beta-turns were the most active. The variety of active loop conformations suggests that the mimetics (and NGF) accommodate the binding site on TrkA by an 'induced fit' mechanism. In agreement with this hypothesis, NMR relaxation measurements detected both fast and slow motion in the peptides. We also characterized a retro-inverso peptide derived from the NGF C-D loop. This D-amino acid cyclic peptide did not adopt a conformation homologous to the NGF C-D loop and was inactive. This may be representative of difficulties in producing structural and functional mimetics by retro-inverso schemes.

Rapid fold and structure determination of the archaeal translation elongation factor 1beta from Methanobacterium thermoautotrophicum.

The tertiary fold of the elongation factor, aEF-1beta, from Methanobacterium thermoautotrophicum was determined in a high-throughput fashion using a minimal set of NMR experiments. NMR secondary structure prediction, deuterium exchange experiments and the analysis of chemical shift perturbations were combined to identify the protein fold as an alpha-beta sandwich typical of many RNA binding proteins including EF-G. Following resolution of the tertiary fold, a high resolution structure of aEF-1beta was determined using heteronuclear and homonuclear NMR experiments and a semi-automated NOESY assignment strategy. Analysis of the aEF-1beta structure revealed close similarity to its human analogue, eEF-1beta. In agreement with studies on EF-Ts and human EF-1beta, a functional mechanism for nucleotide exchange is proposed wherein Phe46 on an exposed loop acts as a lever to eject GDP from the associated elongation factor G-protein, aEF-1alpha. aEF-1beta was also found to bind calcium in the groove between helix alpha2 and strand beta4. This novel feature was not observed previously and may serve a structural function related to protein stability or may play a functional role in archaeal protein translation.

Solution structure of the PDZ2 domain from human phosphatase hPTP1E and its interactions with C-terminal peptides from the Fas receptor.

The solution structure of the second PDZ domain (PDZ2) from human phosphatase hPTP1E has been determined using 2D and 3D heteronuclear NMR experiments. The binding of peptides derived from the C-terminus of the Fas receptor to PDZ2 was studied via changes in backbone peptide and protein resonances. The structure is based on a total of 1387 nonredundant experimental NMR restraints including 1261 interproton distance restraints, 45 backbone hydrogen bonds, and 81 torsion angle restraints. Analysis of 30 lowest-energy structures resulted in rmsd values of 0.41 +/- 0.09 A for backbone atoms (N, Calpha, C') and 1.08 +/- 0.10 A for all heavy atoms, excluding the disordered N- and C-termini. The hPTP1E PDZ2 structure is similar to known PDZ domain structures but contains two unique structural features. In the peptide binding domain, the first glycine of the GLGF motif is replaced by a serine. This serine appears to replace a bound water observed in PDZ crystal structures that hydrogen bonds to the bound peptide's C-terminus. The hPTP1E PDZ2 structure also contains an unusually large loop following strand beta2 and proximal to the peptide binding site. This well-ordered loop folds back against the PDZ domain and contains several residues that undergo large amide chemical shift changes upon peptide binding. Direct observation of peptide resonances demonstrates that as many as six Fas peptide residues interact with the PDZ2 domain.

Fusion proteins could generate false positives in peptide phage display.

Fusion proteins are frequently used in the functional characterization of newly discovered proteins and to identify interacting partners. In our study of hPTP1E, a cytosolic protein tyrosine phosphatase, we used glutathione S-transferase (GST)-fusion protein of the second PDZ domain to identify interacting peptide motifs by peptide phage display. A consensus motif G X X V W L G was identified and found to be specific for binding to GST-PDZ2 as determined by ELISA, peptide displacement and by protein overlay. However, using nuclear magnetic resonance (NMR), no interaction of the peptide was observed with PDZ2 alone. In co-precipitation experiments using the consensus peptide cross-linked to Affi-Gel, only GST-PDZ2 (but not PDZ2 or GST alone) could be precipitated. These data suggest that there is a potential for identification of artifacts when using fusion proteins in peptide phage display, and one should exercise caution in interpreting these results. It is critical that the interaction be verified using a second, independent system.

Two stable unfolding intermediates of the disease-causing L68Q variant of human cystatin C.

In hereditary cystatin C amyloid angiopathy (HCCAA), presence of the Leu68 --> Gln substitution in cystatin C is coupled to a decreased concentration of this major cysteine proteinase inhibitor in cerebrospinal fluid and leads to its amyloid deposition in the brain. We established a high-yield expression system for L68Q cystatin C in Escherichia coli resulting in inclusion body accumulation at a level of 40% of the total cellular protein. Refolding of protein from purified inclusion bodies yielded a pure, almost completely monomeric and active inhibitor. CD and NMR spectroscopy demonstrated that so produced L68Q cystatin C is folded, conformationally homogeneous, and structurally very similar to wild-type cystatin C. Incubation at pH 7.0-5.5 caused the cystatin C variant to dimerize rapidly. The molecular form present at pH 6.0 displayed a slightly increased amount of hydrophobic parts on the surface as measured by 1-anilinonaphthalene-8-sulfonic acid (ANS) binding. NMR results showed that the dimer has a structure similar to that of the wild-type cystatin C dimer formed as a result of slight denaturation. Under more acidic conditions, at pH 4.5, another stable unfolding intermediate of L68Q cystatin C was identified. This molecular form exists in a monomeric state, is characterized by changes in secondary structure according to far UV CD spectroscopy, and shows an altered ANS binding resembling that of a molten globule state. The acidic pH also caused an almost complete monomerization of preformed dimers. The state of denaturation of L68Q cystatin C in vivo is thus a critical factor for the concentration of active cysteine proteinase inhibitor in cerebrospinal fluid and likely also for the development of amyloidosis, in HCCAA patients.

H(C)CH-COSY and (H)CCH-COSY experiments for 13C-labeled proteins in H2O solution.

We present three experiments which serve to identify carbon and proton sidechain resonances in 13C-labeled proteins. The first is an improvement on the previously published H(C)CH-COSY experiment and comprises the application of gradients for coherence selection and a reduction in the phase cycle. The second experiment is a new (H)CCH-COSY with two carbon dimensions. The (H)CCH-COSY presents several advantages over the H(C)CH-COSY experiment in terms of better sensitivity, improved resolution and easier identification of amino acid spins systems. The third experiment is a 2D proton-edited (H)C(C)H-COSY that allows suppression of methylene resonances. All three HCCH-COSY experiments show good sensitivity and excellent solvent suppression. The 2D version can be acquired in as little as 45 minutes and the 3D versions acquired overnight. The experiments are demonstrated on a 13C-labeled sample of the second PDZ domain from human phosphatase PTP1E in H2O solution.

Solution structure and internal motion of a bioactive peptide derived from nerve growth factor.

The conformation and internal dynamics of a bioactive cyclic peptide, N-acetyl-YCTDEKQCY, derived from the C-D loop of beta-nerve growth factor (beta-NGF) were analyzed by solution NMR spectroscopy. NMR experimental data were used to calculate an ensemble of peptide structures. All of the structures had a beta-turn at residues Asp4-Gln7 but could be divided into two families according the presence or absence of a hydrogen bond at Gln7. Comparison of the calculated structures with the corresponding C-D loops from the x-ray structures of the NGF revealed striking similarity. The orientation of Glu5, Lys6, and Gln7 side chains in the NGF mimetic was very similar to the C-D loop of NGF. These residues are known to participate in interactions with the TrkA receptor. Relaxation measurements of the peptidomimetic alpha-carbons at 13C natural abundance and calculated dynamic parameters suggest that the loop region of peptide is well structured but that residues Thr3, Asp4, Glu5, and Lys6 undergo slow conformational exchange. These results suggest that conformational similarity and possibly peptide dynamics are responsible for the bioactivity of the peptide.

Structural basis for different inhibitory specificities of human cystatins C and D.

Human cystatins C and D share almost identical primary structures of two out of the three segments proposed to be of importance for enzyme interactions but have markedly different profiles for inhibition of the target cysteine peptidases, cathepsins B, H, L, and S. To investigate if the N-terminal binding regions of the inhibitors are responsible for the different inhibition profiles, and thereby confer biological selectivity, two hybrid cystatins were produced in Escherichia coli expression systems. In one hybrid, the N-terminal segment of cystatin C was placed on the framework of cystatin D, and the second was engineered with the N-terminal segment of cystatin D on the cystatin C scaffold. Truncated cystatin C and D variants, devoid of their N-terminal segments, were obtained by incubation with glycyl endopeptidase and isolated, in a second approach to assess the importance of the N-terminal binding regions for cystatin function and specificity. The affinities of the four cystatin variants for cathepsins B, H, L, and S were measured. By comparison with corresponding results for wild-type cystatins C and D, it was concluded (1) that both the N-terminal and framework part of the molecules significantly contribute to the observed differences in inhibitory activities of cystatins C and D and (2) that the N-terminal segment of cystatin C increases the inhibitory activity of cystatin D against cathepsin S and cathepsin L but results in decreased activity against cathepsin H. These differences in specificity were explained by the residues interacting with the S2 subsite of peptidases (Val- and Ala-10 in cystatin C and D, respectively). Also, removal of the N-terminal segment results in total loss of enzyme affinity for cystatin D but not for cystatin C. Therefore, structural differences in the framework parts, as well as in the N-terminal segments, are critical for both inhibitory specificity and potency. Homology modeling was used to identify residues likely responsible for the generally reduced inhibitory potency of cystatin D.

Effect of peptide binding on amide proton exchange rates in the PDZ2 domain from human phosphatase hPTP1E.

Amide hydrogen-deuterium exchange rates were measured in the PDZ2 domain from human phosphatase hPTPIE by 1H-15N heteronuclear NMR spectroscopy. Protection factors were calculated for the slowly exchanging hydrogens in both the free PDZ2 domain and its complex with an octapeptide peptide, R-N-E-I-Q-S-L-V, derived from the C-terminus of the Fas receptor. Aside from a short alpha-helical region alpha1 (amino acids A-45 to D-49), the pattern of highly protected amides correlated well with the presence of hydrogen bonds in elements of the secondary structure. Hydrogen-bonded amides showed relatively fast exchange rates with half-lives of less than 9 h at pD 7.6 and 8 degrees C. Protection factors, calculated as the ratio of theoretical (denatured) and observed exchange rates, showed less dispersion in maximal values than did the actual exchange rates. This behavior and the large pH dependence of the exchange rates suggest that amide exchange is close to the EX2 limit. In this limit, exchange of the most protected amides occurs through a global unfolding mechanism. The free energy of the unfolding calculated from the largest protection factors is 4.8 +/- 0.4 kcal/mol (1 cal = 4.184 J). This deltaG(o) closely matches the value measured by experiments with guanidine hydrochloride and fluorescence emission spectroscopy. Peptide binding to PDZ2 resulted in mostly global effects and stabilized the folded domain by 1.4 kcal/mol.

Affinity purification and elimination of methionine oxidation in recombinant human cystatin C.

Recombinant human cystatin C (cC), a cysteine protease inhibitor, contained methionine sulfoxide [Met(O)] residues when expressed in Escherichia coli under aerobic conditions or upon allowing osmotic shock solutions from anaerobically grown cultures to warm to room temperature. Oxidation occurred in the periplasmic space or intracellularly during aerobic expression. Both Met14 and Met41 were subject to oxidation, as determined by NMR spectroscopy and mass spectrometry. Oxidation of Met110 was not observed. Growth under anaerobic conditions and modified purification procedures prevented oxidation. Through the use of a new form of affinity purification, cC was purified to > 99% in one step on E-64-papain-Sepharose (E-64 is 1-[N-[(L-3-trans-carboxyoxirane-2-carbonyl)-L-leucyl]amino]-4-g uanidinobutane), with elution with sodium trichloroacetate. The dissociation equilibrium constants (Kd) for the interaction of unoxidized cC, (Met(O)14)cC, and (Met(O)41)cC with S-(N-ethylsuccinimidyl)papain were experimentally identical: 1.8 (+/-0.2) x 10(-7), 1.6 (+/-0.2) x 10(-7), and 1.4 (+/-0.5) x 10(-7) M, respectively. This implies that the structure of the protease-binding region of mono-oxidized cC's was unchanged. The NMR observation of small, localized conformational changes was consistent with this. (Met(O)14)cC and (Met(O)14,Met(O)41)cC eluted earlier upon analytical affinity chromatography.

NMR structural studies of human cystatin C dimers and monomers.

Human cystatin C undergoes dimerization before unfolding. Dimerization leads to a complete loss of its activity as a cysteine proteinase inhibitor. A similar process of dimerization has been observed in cells, and may be related to the amyloid formation seen for the L68Q variant of the protein. Dimerization is barrier controlled, and no dimer/monomer interconversion can be observed at physiological conditions. As a consequence, very stable, "trapped" dimers can be easily separated from monomers. A study of the structural aspects of cystatin C dimer formation was undertaken using NMR spectroscopy. The monomer/dimer model was verified by (pulse field gradient NMR) self-diffusion molecular mass measurements. Complete backbone resonance assignments and secondary structure determination were obtained for the monomer using data from triple resonance experiments performed on 13C/15N doubly labeled protein. A marked similarity of the cystatin C secondary structure to that of chicken cystatin was observed. Using uniformly and amino-acid-specific 15N-enriched protein, backbone NH signals were assigned for cystatin C in its dimeric state. Comparison of 1H -15N correlation NMR spectra of the monomer and dimer shows that the three-dimensional structure remains unchanged in the dimer and that only local perturbations occur. These are localized to the amino acid residues comprising the cysteine proteinase binding site. Such a mode of dimerization readily explains the complete loss of the inhibitory activity in the dimer. The NMR results also demonstrate that the dimer is symmetric.

Structural studies of the major glycolipid from Saccharopolyspora genus.

A major glycolipid was isolated from the well characterized Saccharopolyspora species, S. hirsuta, S. rectivirgula, S. erythraea and one not completely identified strain (Saccharopolyspora sp.). On the basis of sugar and methylation analysis, specific enzymatic and chemical degradations of the carbohydrate moiety, its FAB mass spectrometry and NMR spectroscopy characterizations, the carbohydrate part was shown to be the glycerol linked dimannoside alpha-D-Manp-(1-->3)-alpha-D-Manp-(1-->1/3)Gro. The internal mannose residue is esterified at C-6 by one fatty acid residue, whereas another fatty acyl chain substitutes the primary methylene position of glycerol. The main fatty acyl residues are anteiso-branched heptadecanoic acid and the iso-branched fatty acids iso-17:0, iso-16:0, and iso-18:0, with the former species being predominant. The major glycolipid has potential value for taxonomic and diagnostic purposes, especially in the specific diagnosis of farmer's lung disease.

Folding-related dimerization of human cystatin C.

With the aim to improve our understanding of the structural basis for protein self-association and aggregation, in particular in relationship to protein refolding and amyloid formation, folding-related processes for human cystatin C have been studied. Using NMR spectroscopy together with chromatographic and electrophoretic methods, a self-association process resulting in dimer formation for protein samples treated with denaturing agents as well as for samples subjected to low pH or high temperature conditions could be studied with amino acid resolution. In all three cases, the dimerization involves properly folded molecules and proceeds via the reactive site of the inhibitor, which leads to complete loss of its biological activity. This dimerization process has potential relevance for amyloid formation by the brain hemorrhage-causing Leu58-Gln variant of cystatin C. The results also indicate that cystatin C dimerization and inactivation may occur in acidified compartments in vivo, which could be relevant for the physiological regulation of cysteine proteinase activity.

Tetraether lipids of Methanospirillum hungatei with head groups consisting of phospho-N,N-dimethylaminopentanetetrol, phospho-N,N,N-trimethylaminopentanetetrol, and carbohydrates.

Acyclic, standard tetraether and diether lipids each account for about 50% of the total ether lipids found in Methanospirillum hungatei. Sixteen ether lipids were purified and defined according to relative weight percentage and staining reactions on thin-layer plates. Structures were elucidated for six previously uncharacterized tetraether lipids. Four of these lipids had as one head group either alpha-glcp-(1-2)-beta-gal(f)-, or beta-gal(f)-(1-6)-beta-gal(f)-, in glycosidic linkage to the first glycerol of the lipid backbone, and either a N,N-dimethyl-aminopentanetetrol or a N,N,N-trimethylaminopentanetetrol moiety in phosphodiester linkage to the second glycerol of the backbone. A fifth lipid was a tetraether structure novel in having carbohydrate moieties at both head group positions; namely alpha-glcp-(1-2)-gal(f)- and beta-gal(f)-. Two other lipids, a diether and a tetraether, had a single head group consisting of alpha-glcp-(1-2)-beta-gal(f)- modified by O-acetylation of the gal(f) residue at C-6. In addition to the seven new lipids described above, diether and tetraether analogs of phosphatidylglycerol were found.

Metabolic Pathways in Methanococcus jannaschii and Other Methanogenic Bacteria.

Eleven strains of methanogenic bacteria were divided into two groups on the basis of the directionality (oxidative or reductive) of their citric acid pathways. These pathways were readily identified for most methanogens from the patterns of carbon atom labeling in glutamate, following growth in the presence of [2-C]acetate. All used noncyclic pathways, but members of the family Methanosarcinaceae were the only methanogens found to use the oxidative direction. Methanococcus jannaschii failed to incorporate carbon from acetate despite transmembrane equilibration comparable to other weak acids. This organism was devoid of detectable activities of the acetate-incorporating enzymes acetyl coenzyme A synthetase, acetate kinase, and phosphotransacetylase. However, incorporation of [1-C]-, [2-C]-, or [3-C]pyruvate during the growth of M. jannaschii was possible and resulted in labeling patterns indicative of a noncyclic citric acid pathway operating in the reductive direction to synthesize amino acids. Carbohydrates were labeled consistent with glucogenesis from pyruvate. Leucine, isoleucine, phenylalanine, lysine, formate, glycerol, and mevalonate were incorporated when supplied to the growth medium. Lysine was preferentially incorporated into the lipid fraction, suggesting a role as a phytanyl chain precursor.

Production of Specifically Labeled Compounds by Methanobacterium espanolae Grown on H(2)-CO(2) plus [C]Acetate.

Methanobacterium espanolae, an acidiphilic methanogen, required acetate for maximal growth on H(2)-CO(2). In the presence of 5 to 15 mM acetate, at a growth pH of 5.5, the mu(max) was 0.05 h. M. espanolae consumed 12.3 mM acetate during 96 h of incubation at 35 degrees C with shaking at 100 rpm. At initial acetate levels of 2.5 to 10.0 mM, the amount of biomass produced was dependent on the amount of acetate in the medium. C nuclear magnetic resonance spectra of protein hydrolysates obtained from cultures grown on [1-C]- or [2-C]acetate indicated that an incomplete tricarboxylic acid pathway, operating in the reductive direction, was functional in this methanogen. The amino acids were labeled with a very high degree of specificity and at greater than 90% enrichment levels. Less than 2% label randomization occurred between positions primarily labeled from either the carboxyl or methyl group of acetate, and very little label was transferred to positions primarily labeled from CO(2). The labeling pattern of carbohydrates was typical for glucogenesis from pyruvate. This methanogen, by virtue of the properties described above and its ability to incorporate all of the available acetate (10 mM or lower) from the growth medium, has advantages over other microorganisms for use in the production of specifically labeled compounds.

Hydroxydiether Lipid Structures in Methanosarcina spp. and Methanococcus voltae.

Hydroxylated diether lipids are the most abundant lipids in Methanosarcina acetivorans, Methanosarcina thermophila, and Methanosarcina barkeri MS and Fusaro, regardless of the substrate used for growth. Structural analysis of the lipid moiety freed of polar head groups revealed that the hydroxydiether lipids of all the Methanosarcina strains were hydroxylated at position 3 of sn-2 phytanyl chains. The finding that Methanosarcina strains synthesize the same hydroxydiether structure suggests that this is a taxonomic characteristic of the genus. Methanococcus voltae produced minor amounts of the 3-hydroxydiether characteristic of Methanosarcina spp. and also the 3'-hydroxydiether described previously for Methanosaeta concilii.