Refined structure of the gene 5 DNA binding protein from bacteriophage fd

Structure and assembly of filamentous bacteriophages

Filamentous bacteriophages are interesting paradigms in structural molecular biology, in part because of the unusual mechanism of filamentous phage assembly. During assembly, several thousand copies of an intracellular DNA-binding protein bind to each copy of the replicating phage DNA, and are then displaced by membrane-spanning phage coat proteins as the nascent phage is extruded through the bacterial plasma membrane. This complicated process takes place without killing the host bacterium. The bacteriophage is a semi-flexible worm-like nucleoprotein filament. The virion comprises a tube of several thousand identical major coat protein subunits around a core of single-stranded circular DNA. Each protein subunit is a polymer of about 50 amino-acid residues, largely arranged in an α-helix. The subunits assemble into a helical sheath, with each subunit oriented at a small angle to the virion axis and interdigitated with neighbouring subunits. A few copies of "minor" phage proteins necessary for infection and/or extrusion of the virion are located at each end of the completed virion. Here we review both the structure of the virion and aspects of its function, such as the way the virion enters the host, multiplies, and exits to prey on further hosts. In particular we focus on our understanding of the way the components of the virion come together during assembly at the membrane. We try to follow a basic rule of empirical science, that one should chose the simplest theoretical explanation for experiments, but be prepared to modify or even abandon this explanation as new experiments add more detail.

Sequence-structure mapping errors in the PDB: OB-fold domains

The Protein Data Bank (PDB) is the single most important repository of structural data for proteins and other biologically relevant molecules. Therefore, it is critically important to keep the PDB data, as much as possible, error-free. In this study, we have analyzed PDB crystal structures possessing oligonucleotide/oligosaccharide binding (OB)-fold, one of the highly populated folds, for the presence of sequence-structure mapping errors. Using energy-based structure quality assessment coupled with sequence analyses, we have found that there are at least five OB-structures in the PDB that have regions where sequences have been incorrectly mapped onto the structure. We have demonstrated that the combination of these computation techniques is effective not only in detecting sequence-structure mapping errors, but also in providing guidance to correct them. Namely, we have used results of computational analysis to direct a revision of X-ray data for one of the PDB entries containing a fairly inconspicuous sequence-structure mapping error. The revised structure has been deposited with the PDB. We suggest use of computational energy assessment and sequence analysis techniques to facilitate structure determination when homologs having known structure are available to use as a reference. Such computational analysis may be useful in either guiding the sequence-structure assignment process or verifying the sequence mapping within poorly defined regions.

Improving macromolecular atomic models at moderate resolution by automated iterative model building, statistical density modification and refinement

An iterative process for improving the completeness and quality of atomic models automatically built at moderate resolution (up to about 2.8 A) is described. The process consists of cycles of model building interspersed with cycles of refinement and combining phase information from the model with experimental phase information (if any) using statistical density modification. The process can lead to substantial improvements in both the accuracy and completeness of the model compared with a single cycle of model building. For eight test cases solved by MAD or SAD at resolutions ranging from 2.0 to 2.8 A, the fraction of models built and assigned to sequence was 46-91% (mean of 65%) after the first cycle of building and refinement, and 78-95% (mean of 87%) after 20 cycles. In an additional test case, an incorrect model of gene 5 protein (PDB code 2gn5; r.m.s.d. of main-chain atoms from the more recent refined structure 1vqb at 1.56 A) was rebuilt using only structure-factor amplitude information at varying resolutions from 2.0 to 3.0 A. Rebuilding was effective at resolutions up to about 2.5 A. The resulting models had 60-80% of the residues built and an r.m.s.d. of main-chain atoms from the refined structure of 0.20 to 0.62 A. The algorithm is useful for building preliminary models of macromolecules suitable for an experienced crystallographer to extend, correct and fully refine.

Aggregation and self-assembly of hydrophobins from Trichoderma reesei: low-resolution structural models

Hydrophobins are secreted fungal proteins, which have diverse roles in fungal growth and development. They lower the surface tension of water, work as adhesive agents and coatings, and function through self-assembly. One of the characteristic properties of hydrophobins is their tendency to form fibrillar or rod-like aggregates at interfaces. Their structure is still poorly known. In a step to elucidate the structure/function relation of hydrophobin self-assembly, we present the low-resolution structure of self-assembled fibrils of the class II hydrophobin HFBII from Trichoderma reesei based on small and wide-angle x-ray scattering. We first studied the solution state (10 mg/mL) of both HFBI and HFBII and showed that they formed assemblages in aqueous solution, which have a radius of gyration of ~24 A and maximum dimension of ~65 A, corresponding to the size of a tetramer. This result was supported by size-exclusion chromatography. Undried samples of HFBII fibrils had a monoclinic crystalline structure, which changed to hexagonal when the material was dried. A low-resolution structure for the HFBII fibrils is suggested. There are data in the literature based on staining properties suggesting that hydrophobins of class I form assemblies with an amyloid structure. Comparison of the HFBII data (x-ray results, staining with thioflavin T) to published data showed that the HFBII assemblages are not amyloid.

Identification of the primer binding domain in human immunodeficiency virus reverse transcriptase

We have labeled the primer binding domain of HIV1-RT with 5'-32P-labeled (dT)15 primer using ultraviolet light energy. The specificity of the primer cross-linking to HIV1-RT was demonstrated by competition experiments. Both synthetic and natural primers, e.g., p(dA)15, p(dC)15, and tRNA(Lys), inhibit p(dT)15 binding and cross-linking to the enzyme. The observed binding and cross-linking of the primer to the enzyme were further shown to be functionally significant by the observation that tRNA(Lys) inhibits the polymerase activity on poly(rA).(dT)15 template-primer as well as the cross-linking of p(dT)15 to the enzyme to a similar extent. At an enzyme to p(dT)15 ratio of 1:3, about 15% of the enzyme can be cross-linked to the primer. To identify the domain cross-linked to (dT)15, tryptic peptides were generated and purified by a combination of HPLC on a C-18 reverse-phase column and DEAE-Sephadex chromatography. A single peptide cross-linked to p(dT)15 was identified. This peptide corresponded to amino acid residues 288-307 in the primary sequence of HIV1-RT as judged by amino acid composition and sequence analyses. Further, Leu(289)-Thr(290) and Leu(295)-Thr(296) of HIV1-RT appear to be the probable sites of cross-linking to the primer p(dT)15.

Map-likelihood phasing

The recently developed technique of maximum-likelihood density modification [Terwilliger (2000), Acta Cryst. D56, 965-972] allows a calculation of phase probabilities based on the likelihood of the electron-density map to be carried out separately from the calculation of any prior phase probabilities. Here, it is shown that phase-probability distributions calculated from the map-likelihood function alone can be highly accurate and that they show minimal bias towards the phases used to initiate the calculation. Map-likelihood phase probabilities depend upon expected characteristics of the electron-density map, such as a defined solvent region and expected electron-density distributions within the solvent region and the region occupied by a macromolecule. In the simplest case, map-likelihood phase-probability distributions are largely based on the flatness of the solvent region. Though map-likelihood phases can be calculated without prior phase information, they are greatly enhanced by high-quality starting phases. This leads to the technique of prime-and-switch phasing for removing model bias. In prime-and-switch phasing, biased phases such as those from a model are used to prime or initiate map-likelihood phasing, then final phases are obtained from map-likelihood phasing alone. Map-likelihood phasing can be applied in cases with solvent content as low as 30%. Potential applications of map-likelihood phasing include unbiased phase calculation from molecular-replacement models, iterative model building, unbiased electron-density maps for cases where 2F(o) - F(c) or sigma(A)-weighted maps would currently be used, structure validation and ab initio phase determination from solvent masks, non-crystallographic symmetry or other knowledge about expected electron density.

Identification and functional characterization of a member of the PUR-alpha family from Schistosoma mansoni

Schistosoma mansoni p14 gene encodes an eggshell precursor that is expressed only in vitelline cells of mature female worms in response to a male stimulus. The upstream region of the p14 gene contains several potential cis-acting regulatory sequences. We used the upstream region of the p14 gene as bait in a yeast-one-hybrid screen of a S. mansoni cDNA library to identify interacting proteins. We report the identification and characterization of a cDNA (S. mansoni PUR-alpha (SmPUR-alpha)) encoding a protein homologous to single-stranded DNA transcription activator PUR-alpha, that binds to the p14 upstream region and activates transcription of the HIS3 reporter gene in yeast. SmPUR-alpha has a predicted molecular mass of 30 kDa and shares an overall homology of 63% with mammalian PUR-alpha. The DNA binding domain of SmPUR-alpha is highly conserved. We show by gel shift assays that GST-SmPUR-alpha binds to oligonucleotides comprising the p14 upstream region. SmPUR-alpha binds preferentially to single-stranded DNA and also binds RNA. Unlike the mammalian homologue, SmPUR-alpha exhibits little specificity for the PUR element GGn, but shows strong preference for a sequence containing alternating pyrimidines. Our data support that SmPUR-alpha is a single-copy gene and through reverse transcriptase-polymerase chain reaction and in situ hybridization, we show that SmPUR-alpha is constitutively transcribed in many cell types and thus likely plays a role as a general transcription activator in schistosomes.

Knowledge-based interaction potentials for proteins

We discuss the derivation of atomic-level potentials of mean force from the known protein structures and their applicability for structural evaluation applications. In the derivation process, rigorous density estimation methodology is used to estimate the probability density functions (PDFs) for the distributions of interatomic distances in the protein structures. Potentials of mean force are then derived from these density functions using simple Boltzmann's relation. We also test the potentials against pairs of current and superseded protein structures in the Protein Data Bank. Using PDF potentials to evaluate each structure pair, we are able to identify, with high accuracy, which of the two structures is of higher resolution or better quality. This result shows that the PDF potentials are sensitive to details in protein structures as the current and superseded atomic coordinates generally do not differ by more than 1 A in root-mean-square deviation, and that the PDF potentials could potentially be used for X-ray structure refinement and protein structure prediction.

The unr gene: evolutionary considerations and nucleic acid-binding properties of its long isoform product

The unr transcription unit is located just upstream of the N-ras gene in the genome of mammals, in which unr, like N-ras, is ubiquitously expressed. To determine at what point in evolution the unr/N-ras linkage was created, analysis of nucleic acids by Southern and Northern blotting was performed, allowing us to track the presence of the unr gene to the start of vertebrate evolution and the unr/N-ras linkage to the time at which the reptilian and bird lines diverged. We have investigated, with specific anti-unr antibodies, a potential relation between unr protein levels and cellular processes in which N-ras is implicated. A positive correlation in the proliferation of 3T3 cells, but not differentiation of PC12 cells induced by nerve growth factor (NGF), was found. To study the nucleic acid-binding properties of unr, a protein with multiple repeats of a nucleic acid-binding motif, we expressed the long splicing isoform in a eukaryotic cell line and purified it in native form. The results obtained-a high affinity of unr for single-stranded DNA and RNA and lower affinity for double-stranded DNA without regard to nucleic acid sequence, and its intracellular localization in both the nuclear and non-nuclear compartments, together with its ubiquious expression in mammalian tissues-provide molecular information about the function of one of the closest gene tandems in mammalian cells (unr-N-ras).

Energy strain in three-dimensional protein structures

BACKGROUND

Steric strain in protein three-dimensional structures is related to unfavorable inter-atomic interactions. The steric strain may be a result of packing or functional requirements, or may indicate an error in the coordinates of a structure. Detailed energy functions are, however, usually considered too noisy for error detection.

RESULTS

After a short energy refinement, a full-atom, detailed energy function becomes a sensitive indicator of errors. The statistics of the energy distribution of amino acid residues in high-resolution crystal structures, represented by models with idealized covalent geometry, were calculated. The interaction energy of each residue with the whole protein structure and with the solvent was considered. Normalized deviations of amino acid residue energies from their average values were used for detecting energy-strained and, therefore, potentially incorrect fragments of a polypeptide chain. Protein three-dimensional structures of different origin (X-ray crystallography, NMR spectroscopy, theoretical models and deliberately misfolded decoys) were compared. Examples of the applications to loop and homology modeling are provided.

CONCLUSIONS

Elevated levels of energy strain may point at a problematic fragment in a protein three-dimensional structure of either experimental or theoretical origin. The approach may be useful in model building and refinement, modeling by homology, protein design, folding calculations, and protein structure analysis.

DNA-binding studies of XSPTSPSZ, derivatives of the intercalating heptad repeat of RNA polymerase II

The synthesis, solution conformation, and interaction with DNA of three 8-residue peptides structurally related to the heptad repeat unit found at the C-terminus of RNA polymerase II are reported. Peptides QQ, XQ, and PQ are derived from the parent sequence YSPTSPSY (peptide YY), which was reported to bind to DNA by bisintercalation [M. Suzuki (1990) Nature, Vol. 344, pp. 562-565], and contain either a 2-quinolyl (Q), 2-quinoxolyl (X), or 5-phenanthrolyl (P) group in place of the aromatic side chains of the N- and C-terminal tyrosine residues present in the parent sequence. The combined results of linear dichroism and induced CD measurements of peptides QQ, XQ, and PQ with calf thymus DNA are consistent with weak binding of the peptides to DNA in a preferred orientation in which the chromophores are intercalated. Small increases in the melting temperatures of poly[d(A-T)2] are also consistent with the peptides interacting with DNA. While enzymatic footprinting with DNase I showed no protection from cleavage by the enzyme, chemical footprinting with fotemustine showed that the peptides modify the reactivity of the major groove, presumably via minor groove binding. Peptide QQ inhibited fotemustine alkylation significantly more than either XQ or PQ, and slightly more than YY. In aqueous solution, nmr experiments on QQ, XQ, and PQ show a significant population of a conformation in which Ser2-Pro3-Thr4-Ser5 form both type I and type II beta-turn conformations in equilibrium with open chain conformations. Nuclear magnetic resonance titration experiments of PQ with (GCGTACGC)2 showed small changes in chemical shifts, consistent with the formation of a weak nonspecific complex. Analogous experiments, using peptides QQ and XQ with (GCGTACGC)2, and peptide YY with (CGTACG)2, showed no evidence for the interaction of the peptides with these oligonucleotides. These results show that peptides of general structure XSPTSPSZ are weak nonspecific DNA binders that differ significantly from previously characterized S(T)PXX DNA-binding motifs that are generally AT-selective minor groove binders.

Analysis of protein-protein interaction sites using surface patches

Protein-protein interaction sites in complexes of known structure are characterised using a series of parameters to evaluate what differentiates them from other sites on the protein surface. Surface patches are defined in protomers from a data set of 28 homo-dimers, 20 different hetero-complexes (segregated into large and small protomers), and antigens from six antibody-antigen complexes. Six parameters (solvation potential, residue interface propensity, hydrophobicity, planarity, protrusion and accessible surface area) are calculated for the observed interface patch and all other surface patches defined on each protein. A ranking of the observed interface, relative to all other possible patches, is calculated. With this approach it becomes possible to analyse the distribution of the rankings of all the observed patches, relative to all other surface patches, for each data set. For each type of complex, none of the parameters were definitive, but the majority showed trends for the observed interface to be distinguished from other surface patches.

Raman spectroscopy of the Ff gene V protein and complexes with poly(dA): nonspecific DNA recognition and binding

Raman spectra of crystals and solutions of the single-stranded DNA binding protein of bacteriophage Ff (gene V protein, gVp) and of solution complexes of gVp with single-stranded poly-(deoxyadenylic acid) [poly(dA)] reveal the following: (i) The gVp secondary and tertiary structures are similar in solution and in the crystal and are dominated by beta-sheet domains, in agreement with NMR and X-ray findings. (ii) Subunit conformation and side chain environments of gVp are virtually unchanged over a wide range of salt concentration (0 < [NaCl] < 100 mM); however, the solution conformation of poly(dA) exhibits sensitivity to added salt. The perturbed Raman markers indicate subtle changes in helix backbone geometry with accompanying small differences in base stacking as the concentration of NaCl is changed. (iii) In complexes with poly(dA), neither the conformation of gVp nor its side chain environments are altered significantly in comparison to the free protein. This is the case at both high salt (nucleotide-to-subunit binding stoichiometry n = 4) and low salt (n = 3). (iv) The Raman signature of poly(dA) undergoes small perturbations upon gVp binding, indicative of small changes in base stacking and phosphodiester backbone conformation. The present results show that the different stoichiometric binding modes of gVp to poly(dA) are accomplished without significant changes in gVp subunit structure and with only modest changes in the single-stranded poly(dA) ligand. This contrasts sharply with sequence-specific double-stranded DNA binding proteins, such as the phage lambda and D108 repressors, which undergo substantial structural changes upon DNA binding, and which also alter more dramatically the Raman fingerprints of their DNA target sites. Thus, nonspecific and specific nucleic acid recognition modes are distinguishable by Raman spectroscopy. The Raman signature of gVp also allows examination of hydrogen bonding interactions of unique side chains within the hydrophobic core (cysteine 33) and at the binding interface (tyrosine 41). These are discussed in relation to the recently published gVp crystal structure.

Direct measurement of oligonucleotide binding stoichiometry of gene V protein by mass spectrometry

The binding stoichiometry of gene V protein from bacteriophage f1 to several oligonucleotides was studied using electrospray ionization-mass spectrometry (ESI-MS). Using mild mass spectrometer interface conditions that preserve noncovalent associations in solution, gene V protein was observed as dimer ions from a 10 mM NH4OAc solution. Addition of oligonucleotides resulted in formation of protein-oligonucleotide complexes with stoichiometry of approximately four nucleotides (nt) per protein monomer. A 16-mer oligonucleotide gave predominantly a 4:1 (protein monomer: oligonucleotide) complex while oligonucleotides shorter than 15 nt showed stoichiometries of 2:1. Stoichiometries and relative binding constants for a mixture of oligonucleotides were readily measured using mass spectrometry. The binding stoichiometry of the protein with the 16-mer oligonucleotide was measured independently using size-exclusion chromatography and the results were consistent with the mass spectrometric data. These results demonstrate, for the first time, the observation and stoichiometric measurement of protein-oligonucleotide complexes using ESI-MS. The sensitivity and high resolution of ESI-MS should make it a useful too] in the study of protein-DNA interactions.

DNA-binding surface of RecA protein photochemical cross-linking of the first DNA binding site on RecA filament

The first DNA-binding site (site I) of RecA protein on the filament has been mapped. RecA protein was covalently cross-linked with a 55-base synthetic single-stranded DNA which was a good substrate for the RecA-mediated strand exchange reaction. The cross-linking sites of protein were determined in the regions spanning RecA residues 64-68, 89-106, 178-183, 199-216 and 257-280. The cross-linking in the residues 64-68, 89-106, 199-216 and 257-280 would be due to the cross-linking of Tyr65, Tyr103, disordered loop 2, and Tyr264, respectively. These regions form a DNA-binding surface centered around the beta-sheet spanning residues 243-257. In the P6(1) crystal filament, the DNA-binding surface is near the RecA-RecA interface but are not in the filament axis. The data implicate a mechanism whereby the DNA binding surface would be led into the filament axis by a conformational change from inactive filament as the P6(1) structure to active filament as the RecA-DNA-ATP complex.

Feed-forward neural networks for secondary structure prediction

A feed-forward neural network has been employed for protein secondary structure prediction. Attempts were made to improve on previous prediction accuracies using a hierarchical mixture of experts (HME). In this method input data are clustered and used to train a series of different networks. Application of an HME to the prediction of protein secondary structure is shown to provide no advantages over a single network. We have also tried various new input representations, chosen to incorporate the effect of residues a long distance away in the one-dimensional amino acid chain. Prediction accuracy using these methods is comparable to that achieved by other neural networks.

Site-directed mutagenesis of the M13 gene 5 protein: the role of Arg-21, Tyr-26 and Tyr-41

The gene 5 protein of bacteriophage M13 is a single stranded DNA binding protein essential for phage replication. We have generated the mutations R21A, Y26F and Y41A in the gene 5 protein and purified the mutant proteins for functional characterisation in vitro. The complex of Y26F with single-stranded DNA is disrupted at 0.8 M NaCl, the same salt concentration as that required to dissociate the native complex. However, the mutant proteins R21A and Y41A are considerably less stable and dissociate from single-stranded DNA at at 0.4 M NaCl. The fluorescence of the mutant proteins and the DNA-protein complexes they form has been compared with the wild-type protein to allow an assessment of the contribution from individual residues. We conclude that the fluorescence of Tyr-26 is 50% quenched in the complex with DNA, whereas that of Tyr-41 is fully quenched. Fluorescence titrations of the mutant proteins with poly(dT) show that all three mutant proteins can bind DNA but, in the case of Y41A, with a change of stoichiometry suggesting a loss of cooperativity. Gel retardation analysis of Y41A also shows anomalous behaviour in binding to oligonucleotides, consistent with the proposed involvement of Tyr-41 in dimer-dimer contacts in the nucleoprotein complex.

Single-stranded DNA binding protein encoded by the filamentous bacteriophage M13: structural and functional characteristics

The single-stranded DNA binding protein, or gene V protein (gVp), encoded by gene V of the filamentous bacteriophage M13 is a multifunctional protein that not only regulates viral DNA replication but also gene expression at the level of mRNA translation. It furthermore is implicated as a scaffolding and/or chaperone protein during the phage assembly process at the hostcell membrane. The protein is 87 amino acids long and its biological functional entity is a homodimer. In this manuscript a short description of the life cycle of filamentous phages is presented and our current knowledge of the major functional and structural properties and characteristics of gene V protein are reviewed. In addition models of the superhelical complexes gVp forms with ssDNA are described and their (possible) biological meaning in the infection process are discussed. Finally it is described that the 'DNA binding loop' of gVp is a recurring motif in many ssDNA binding proteins and that the fold of gVp is shared by a large family of evolutionarily conserved gene regulatory proteins.

Analysis of the DNA binding site of Escherichia coli RecA protein

To investigate the DNA binding site of RecA protein, we constructed 15 recA mutants having alterations in the regions homologous to the other ssDNA binding proteins. The in vivo analyses showed that the mutational change at Arg243, Lys248, Tyr264, or simultaneously at Lys6 and Lys19, or Lys6 and Lys23 caused severe defects in the recA functions, while other mutational changes did not. Purified RecA-K6A-K23A (Lys6 and Lys23 changed to Ala and Ala, respectively) protein was indistinguishable from the wild-type RecA protein in its binding to DNA. However, the RecA-R243A (Arg243 changed to Ala) and RecA-Y264A (Tyr264 changed to Ala) proteins were defective in binding to both ss- and ds-DNA. In self-oligomerization property, RecA-R243A was proficient but RecA-Y264A was deficient, suggesting that the RecA-R243A protein had a defect in DNA binding site and the RecA-Y264A protein was defective in its interaction with the adjacent RecA molecule. The region of residues 243-257 including the Arg243 is highly homologous to the DNA binding motif in the ssDNA binding proteins, while the eukaryotic RecA homologues have a similar structure at the amino-terminal side proximal to the nucleotide binding core. The region of residues 243-257 would be a part of the DNA binding site. The other parts of this site would be the Tyr103 and the region of residues 178-183, which were cross-linked to ssDNA. These three regions lie in a line in the crystal structure.

Conformational stability of dimeric proteins: quantitative studies by equilibrium denaturation

The conformational stability of dimeric globular proteins can be measured by equilibrium denaturation studies in solvents such as guanidine hydrochloride or urea. Many dimeric proteins denature with a 2-state equilibrium transition, whereas others have stable intermediates in the process. For those proteins showing a single transition of native dimer to denatured monomer, the conformational stabilities, delta Gu (H2O), range from 10 to 27 kcal/mol, which is significantly greater than the conformational stability found for monomeric proteins. The relative contribution of quaternary interactions to the overall stability of the dimer can be estimated by comparing delta Gu (H2O) from equilibrium denaturation studies to the free energy associated with simple dissociation in the absence of denaturant. In many cases the large stabilization energy of dimers is primarily due to the intersubunit interactions and thus gives a rationale for the formation of oligomers. The magnitude of the conformational stability is related to the size of the polypeptide in the subunit and depends upon the type of structure in the subunit interface. The practical use, interpretation, and utility of estimation of conformational stability of dimers by equilibrium denaturation methods are discussed.

Crystal structure of inorganic pyrophosphatase from Thermus thermophilus

The 3-dimensional structure of inorganic pyrophosphatase from Thermus thermophilus (T-PPase) has been determined by X-ray diffraction at 2.0 A resolution and refined to R = 15.3%. The structure consists of an antiparallel closed beta-sheet and 2 alpha-helices and resembles that of the yeast enzyme in spite of the large difference in size (174 and 286 residues, respectively), little sequence similarity beyond the active center (about 20%), and different oligomeric organization (hexameric and dimeric, respectively). The similarity of the polypeptide folding in the 2 PPases provides a very strong argument in favor of an evolutionary relationship between the yeast and bacterial enzymes. The same Greek-key topology of the 5-stranded beta-barrel was found in the OB-fold proteins, the bacteriophage gene-5 DNA-binding protein, toxic-shock syndrome toxin-1, and the major cold-shock protein of Bacillus subtilis. Moreover, all known nucleotide-binding sites in these proteins are located on the same side of the beta-barrel as the active center in T-PPase. Analysis of the active center of T-PPase revealed 17 residues of potential functional importance, 16 of which are strictly conserved in all sequences of soluble PPases. Their possible role in the catalytic mechanism is discussed on the basis of the present crystal structure and with respect to site-directed mutagenesis studies on the Escherichia coli enzyme. The observed oligomeric organization of T-PPase allows us to suggest a possible mechanism for the allosteric regulation of hexameric PPases.

Side-chain entropy and packing in proteins

What role does side-chain packing play in protein stability and structure? To address this question, we compare a lattice model with side chains (SCM) to a linear lattice model without side chains (LCM). Self-avoiding configurations are enumerated in 2 and 3 dimensions exhaustively for short chains and by Monte Carlo sampling for chains up to 50 main-chain monomers long. This comparison shows that (1) side-chain degrees of freedom increase the entropy of open conformations, but side-chain steric exclusion decreases the entropy of compact conformations, thus producing a substantial entropy that opposes folding; (2) there is side-chain "freezing" or ordering, i.e., a sharp decrease in entropy, near maximum compactness; and (3) the different types of contacts among side chains (s) and main-chain elements (m) have different frequencies, and the frequencies have different dependencies on compactness. mm contacts contribute significantly only at high densities, suggesting that main-chain hydrogen bonding in proteins may be promoted by compactness. The distributions of mm, ms, and ss contacts in compact SCM configurations are similar to the distributions in protein structures in the Brookhaven Protein Data Bank. We propose that packing in proteins is more like the packing of nuts and bolts in a jar than like the pairwise matching of jigsaw puzzle pieces.

Crystal structure of CspA, the major cold shock protein of Escherichia coli

The major cold shock protein of Escherichia coli, CspA, produced upon a rapid downshift in growth temperature, is involved in the transcriptional regulation of at least two genes. The protein shares high homology with the nucleic acid-binding domain of the Y-box factors, a family of eukaryotic proteins involved in transcriptional and translational regulation. The crystal structure of CspA has been determined at 2-A resolution and refined to R = 0.187. CspA is composed of five antiparallel beta-strands forming a closed five-stranded beta-barrel. The three-dimensional structure of CspA is similar to that of the major cold shock protein of Bacillus subtilis, CspB, which has recently been determined at 2.45-A resolution. However, in contrast to CspB, no dimer is formed in the crystal. The surface of CspA is characteristic for a protein interacting with single-stranded nucleic acids. Due to the high homology of the bacterial cold shock proteins with the Y-box factors, E. coli CspA and B. subtilis CspB define a structural framework for the common cold shock domain.

Structural and functional properties of the evolutionarily ancient Y-box family of nucleic acid binding proteins

The Y-box proteins are the most evolutionarily conserved nucleic acid binding proteins yet defined in bacteria, plants and animals. The central nucleic acid binding domain of the vertebrate proteins is 43% identical to a 70-amino-acid-long protein (CS7.4) from E. coli. The structure of this domain consists of an antiparallel five-stranded beta-barrel that recognizes both DNA and RNA. The diverse biological roles of these Y-box proteins range from the control of the E. coli cold-shock stress response to the translational masking of messenger RNA in vertebrate gametes. This review discusses the organization of the prokaryotic and eukaryotic Y-box proteins, how they interact with nucleic acids, and their biological roles, both proven and potential.

Recognition of errors in three-dimensional structures of proteins

A major problem in the determination of the three-dimensional structure of proteins concerns the quality of the structural models obtained from the interpretation of experimental data. New developments in X-ray crystallography and nuclear magnetic resonance spectroscopy have accelerated the process of structure determination and the biological community is confronted with a steadily increasing number of experimentally determined protein folds. However, in the recent past several experimentally determined protein structures have been proven to contain major errors, indicating that in some cases the interpretation of experimental data is difficult and may yield incorrect models. Such problems can be avoided when computational methods are employed which complement experimental structure determinations. A prerequisite of such computational tools is that they are independent of the parameters obtained from a particular experiment. In addition such techniques are able to support and accelerate experimental structure determinations. Here we present techniques based on knowledge based mean fields which can be used to judge the quality of protein folds. The methods can be used to identify misfolded structures as well as faulty parts of structural models. The techniques are even applicable in cases where only the C alpha trace of a protein conformation is available. The capabilities of the technique are demonstrated using correct and incorrect protein folds.

Characterization of the Pf3 single-strand DNA binding protein by circular dichroism spectroscopy

We have used circular dichroism (CD) spectroscopy and gel electrophoresis to characterize the single-strand DNA binding protein (ssDBP) of the bacteriophage Pf3 and its complexes with Pf3 DNA and various DNA and RNA homopolymers. The secondary structure of Pf3 ssDBP had < 1% alpha-helix and therefore was probably a beta-sheet structure like the fd gene 5 protein (g5p). From CD titrations, the binding stoichiometry of Pf3 ssDBP was two nucleotides per protein monomer (n = 2) for complexes formed with all of the nucleic acids except poly[r(U)], for which n = 3 (in a buffer of 10 mM Tris-HCl and 70 mM NaCl, pH 8.2). Evidence of an additional binding mode of n = 4 for complexes formed with Pf3 DNA was found by gel electrophoresis experiments. Pf3 ssDBP showed a marked sequence dependence in binding affinities similar to that known for the fd g5p.

Super-secondary structures involving triple-strand beta-sheets

Triple-strand beta-sheets having up- and -down topology are widespread in proteins and occur in two forms denoted here as S-like and Z-like beta-sheets. In many cases they are included in super-secondary structures of higher order. A number of such structures is described in this paper. An important feature of these super-secondary structures is that they have a unique handedness. Another feature is that some of them only involve S-like beta-sheets and others only Z-like beta-sheets.

A calculation strategy for the structure determination of symmetric dimers by 1H NMR

The structure determination of symmetric dimers by NMR is impeded by the ambiguity of inter- and intramonomer NOE crosspeaks. In this paper, a calculation strategy is presented that allows the calculation of dimer structures without resolving the ambiguity by additional experiments (like asymmetric labeling). The strategy employs a molecular dynamics-based simulated annealing approach to minimize a target function. The experimental part of the target function contains distance restraints that correctly describe the ambiguity of the NOE peaks, and a novel term that restrains the symmetry of the dimer without requiring the knowledge of the symmetry axis. The use of the method is illustrated by three examples, using experimentally obtained data and model data derived from a known structure. For the purpose of testing the method, it is assumed that every NOE crosspeak is ambiguous in all three cases. It is shown that the method is useful both in situations where the structure of a homologous protein is known and in ab initio structure determination. The method can be extended to higher order symmetric multimers.

Exploring the DNA binding domain of gene V protein encoded by bacteriophage M13 with the aid of spin-labeled oligonucleotides in combination with 1H-NMR

The DNA binding domain of the single-stranded DNA binding protein gene V protein encoded by the bacteriophage M13 was studied by means of 1H nuclear magnetic resonance, through use of a spin-labeled deoxytrinucleotide. The paramagnetic relaxation effects observed in the 1H-NMR spectrum of M13 GVP upon binding of the spin-labeled ligand were made manifest by means of 2D difference spectroscopy. In this way, a vast data reduction was accomplished which enabled us to check and extend the analysis of the 2D spectra carried out previously as well as to probe the DNA binding domain and its surroundings. The DNA binding domain is principally situated on two beta-loops. The major loop of the two is the so-called DNA binding loop (residues 16-28) of the protein where the residues which constitute one side of the beta-ladder (in particular, residues Ser20, Tyr26, and Leu28) are closest to the DNA spin-label. The other loop is part of the so-called dyad domain of the protein (residues 68-78), and mainly its residues at the tip are affected by the spin-label (in particular, Phe73). In addition, a part of the so-called complex domain of the protein (residues 44-51) which runs contiguous to the DNA binding loop is in close vicinity to the DNA. The NMR data imply that the DNA binding domain is divided over two monomeric units of the GVP dimer in which the DNA binding loop and the tip of the dyad loop are part of opposite monomers. The view of the GVP-ssDNA binding interaction which emerges from our data differs from previous molecular modeling proposals which were based on the GVP crystal structure (Brayer & McPherson, 1984; Hutchinson et al., 1990). These models implicate the involvement of one or two tyrosines (Tyr34, Tyr41) of the complex loop of the protein to participate in complex formation with ssDNA. In the NMR studies with the spin-labeled oligonucleotides, no indication of such interactions has been found. Other differences between the models and our NMR data are related to the structural differences found when solution and crystal structures are compared.

Exploration of the single-stranded DNA-binding domains of the gene V proteins encoded by the filamentous bacteriophages IKe and M13 by means of spin-labeled oligonucleotide and lanthanide-chelate complexes

Scrutiny of NOE data available for the protein encoded by gene V of the filamentous phage IKe (IKe GVP), resulted in the elucidation of a beta-sheet structure which is partly five stranded. The DNA-binding domain of IKe GVP was investigated using a spin-labeled deoxytrinucleotide. The paramagnetic-relaxation effects observed in the 1H-NMR spectrum of IKe GVP, upon binding of this DNA fragment, could be visualized using two-dimensional difference spectroscopy. In this way, the residues present in the DNA-binding domain of IKe GVP can be located in the structure of the protein. They exhibit a high degree of identity with residues in the gene V protein encoded by the distantly related phage M13 (M13 GVP), for which similar spectral perturbations are induced by such a spin-labeled oligonucleotide. Binding studies with negatively charged lanthanide-1,4,7,10-tetraazacyclodecanetrayl-1,4,7-10- tetrakis(methylene)tetrakisphosphonic acid (DOTP) complexes, showed that these complexes bind to IKe and M13 GVP at two spatially remote sites whose affinities have different pH dependencies. Above pH 7, there is one high-affinity binding site for Gd(DOTP)5-/M13 GVP monomer, which coincides with the single-stranded DNA-binding domain as mapped with the aid of spin-labeled oligonucleotide fragments. The results show that single-stranded DNA binds to conserved (phosphate binding) electropositive clusters at the surface of M13 and IKe GVP. These positive patches are interspersed with conserved or conservatively replaced hydrophobic residues. At pH 5, a second Gd(DOTP)(5-)-binding site becomes apparent. The corresponding pattern of spectral perturbations indicates the accommodation of patches of conserved, or conservatively replaced, hydrophobic residues in the cores of the M13 and IKe dimers.

Genetic fusion of subunits of a dimeric protein substantially enhances its stability and rate of folding

The gene V protein of bacteriophage f1 is a single-stranded DNA and RNA-binding protein composed of two identical subunits. We have constructed single-chain variants of the protein using short peptide linkers of five or six amino acids to connect the carboxyl terminus of one monomer to the amino terminus of the second monomer. The resulting subunit-fusion gene V proteins were found to bind single-stranded DNA nearly as tightly as the wild-type protein. Denaturation measurements show that the subunit-fusion gene V proteins are 5 kcal/mol (1 kcal = 4.18 kJ) more stable than the wild-type protein at a protein concentration of 10 microM. The rate of unfolding of the protein is essentially unaffected by the fusion of monomeric subunits, whereas the rate of folding is greatly enhanced. Our results suggest a simple way of obtaining a substantial thermodynamic stabilization for some oligomeric proteins.

Universal nucleic acid-binding domain revealed by crystal structure of the B. subtilis major cold-shock protein

The cold-shock response in both Escherichia coli and Bacillus subtilis is induced by an abrupt downshift in growth temperature. It leads to the increased production of the major cold-shock proteins, CS7.4 and CspB, respectively. CS7.4 is a transcriptional activator of two genes. CS7.4 and CspB share 43 per cent sequence identity with the nucleic acid-binding domain of the eukaryotic gene-regulatory Y-box factors. This cold-shock domain is conserved from bacteria to man and contains the RNA-binding RNP1 sequence motif. As a prototype of the cold-shock domain, the structure of CspB has been determined here from two crystal forms. In both, CspB is present as an antiparallel five-stranded beta-barrel. Three consecutive beta-strands, the central one containing the RNP1 motif, create a surface rich in aromatic and basic residues that are presumably involved in nucleic acid binding. Preferential binding of CspB to single-stranded DNA is observed in gel retardation experiments.

Structure in solution of the major cold-shock protein from Bacillus subtilis

The cold-shock domain (CSD) is found in many eukaryotic transcriptional factors and is responsible for the specific binding to DNA of a cis-element called the Y-box. The same domain exists in the sequence of the Xenopus RNA-binding proteins FRG Y1 and FRG Y2 (refs 1, 3). The major cold-shock proteins of Escherichia coli (CS7.4) and B. subtilis (CspB) have sequences that are more than 40 per cent identical to the cold-shock domain. We present here the three-dimensional structure of CspB determined by nuclear magnetic resonance spectroscopy. The 67-residue protein consists of an antiparallel five-stranded beta-barrel with strands connected by turns and loops. The structure resembles that of staphylococcal nuclease and the gene-5 single-stranded-DNA-binding protein. A three-stranded beta-sheet, which contains the conserved RNA-binding motif RNP1 as well as a motif similar to RNP2 in two neighbouring antiparallel beta-strands, has basic and aromatic residues at its surface which could serve as a binding site for single-stranded DNA. CspB binds to single-stranded DNA in gel retardation experiments.

Cloning, expression and in vitro characterisation of the M13 gene 5 protein

The gene 5 protein encoded in the genome of bacteriophage M13 is a single stranded DNA binding protein essential for phage replication. We have cloned a fragment of the M13 genome containing gene 5, and investigated the effect of upstream elements on expression of the gene by means of Bal 31 deletion analysis. The gene was also expressed from the lac promoter of the phagemid vector pUC119, and the recombinant protein purified and characterised for DNA binding. The affinity of the recombinant protein for single-stranded DNA was shown to be essentially identical to that of wild type gene 5 protein. Wild type gene 5 protein has a glutamic acid residue at position 30 which, on the basis of the crystal structure, was believed to play a role in maintaining the tertiary structure of the protein through the formation of a salt bridge with arginine-80. We show that substitution of glutamic acid at position 30 by lysine does not impair DNA binding, suggesting that a salt bridge between glutamate-30 and arginine-80 is not essential for the structural integrity of the gene 5 protein as previously proposed.

Sequence of cDNA comprising the human pur gene and sequence-specific single-stranded-DNA-binding properties of the encoded protein

The human Pur factor binds strongly to a sequence element repeated within zones of initiation of DNA replication in several eukaryotic cells. The protein binds preferentially to the purine-rich single strand of this element, PUR. We report here the cloning and sequencing of a cDNA encoding a protein with strong affinity for the PUR element. Analysis with a series of mutated oligonucleotides defines a minimal single-stranded DNA Pur-binding element. The expressed Pur open reading frame encodes a protein of 322 amino acids. This protein, Pur alpha, contains three repeats of a consensus motif of 23 amino acids and two repeats of a second consensus motif of 26 amino acids. Near its carboxy terminus, the protein possesses an amphipathic alpha-helix and a glutamine-rich domain. The repeat region of Pur cDNA is homologous to multiple mRNA species in each of several human cell lines and tissues. The HeLa cDNA library also includes a clone encoding a related gene, Pur beta, containing a version of the 23-amino-acid consensus motif similar, but not identical, to those in Pur alpha. Results indicate a novel type of modular protein with capacity to bind repeated elements in single-stranded DNA.

Sequence of cDNA comprising the human pur gene and sequence-specific single-stranded-DNA-binding properties of the encoded protein

The human Pur factor binds strongly to a sequence element repeated within zones of initiation of DNA replication in several eukaryotic cells. The protein binds preferentially to the purine-rich single strand of this element, PUR. We report here the cloning and sequencing of a cDNA encoding a protein with strong affinity for the PUR element. Analysis with a series of mutated oligonucleotides defines a minimal single-stranded DNA Pur-binding element. The expressed Pur open reading frame encodes a protein of 322 amino acids. This protein, Pur alpha, contains three repeats of a consensus motif of 23 amino acids and two repeats of a second consensus motif of 26 amino acids. Near its carboxy terminus, the protein possesses an amphipathic alpha-helix and a glutamine-rich domain. The repeat region of Pur cDNA is homologous to multiple mRNA species in each of several human cell lines and tissues. The HeLa cDNA library also includes a clone encoding a related gene, Pur beta, containing a version of the 23-amino-acid consensus motif similar, but not identical, to those in Pur alpha. Results indicate a novel type of modular protein with capacity to bind repeated elements in single-stranded DNA.

Characterization of the nucleic acid-binding activities of the isolated amino-terminal head domain of the intermediate filament protein vimentin reveals its close relationship to the DNA-binding regions of some prokaryotic single-stranded DNA-binding proteins

In order to demonstrate that the nucleic acid-binding activities of vimentin are dictated by its Arg-rich N-terminal head domain, this was cut off at position Lys96 with lysine-specific endoproteinase and analysed for its capacity to associate with a variety of synthetic and naturally occurring nucleic acids. The isolated polypeptide (vim NT) showed a preference for single-stranded (ss) polynucleotides, particularly for ssDNAs of high G-content. A comparison of the sequence and predicted secondary structure of vim NT with that of two prokaryotic ssDNA-binding proteins, G5P and G32P of bacteriophages fd and T4, respectively, revealed that the nucleic acid-binding region of all three polypeptides is almost entirely in the beta-conformation and characterized by a very similar distribution of aromatic amino acid residues. A partial sequence of vim NT can be folded into the same beta-loop structure as the DNA-binding wing of G5P of bacteriophage fd and related viruses. As in the case of G5P, nitration of the Tyr residues with tetranitromethane was blocked by single-stranded nucleic acids. This and spectroscopic data indicate intercalation of the Tyr aromatic ring systems between the bases of the nucleic acids and thus the contribution of a stacking component to the binding reaction. The binding was accompanied by significant changes in the ultraviolet absorption spectra of both vim NT and single-stranded nucleic acids. Upon mixing of vim NT with nucleic acids, massive precipitation of the reactants occurred, followed by the quick rearrangement of the aggregates with the formation of specific and soluble association products. Even at very high ionic strengths, at which no electrostatic reaction should be expected, a distinct fraction of vim NT incorporated naturally occurring ssRNAs and ssDNAs into fast sedimenting complexes, suggesting co-operative interaction of the polypeptide with the nucleic acids. In electron microscopy, the complexes obtained from 28 S rRNA appeared as networks of extended nucleic acid strands densely covered with vim NT, in contrast to the compact random coils of uncomplexed RNA. The networks produced from fd DNA were heterogeneous in appearance and their nucleoprotein strands in rare cases were very similar to the rod-like structures of G5P-fd DNA complexes.

Fluorescence studies of the binding of bacteriophage M13 gene V mutant proteins to polynucleotides

This investigation describes how the binding characteristics of the single-stranded DNA-binding protein encoded by gene V of bacteriophage M13, are affected by single-site amino acid substitutions. The series of mutant proteins tested includes mutations in the purported monomer-monomer interaction region as well as mutations in the DNA-binding domain at positions which are thought to be functionally involved in monomer-monomer interaction or single-stranded DNA binding. The characteristics of the binding of the mutant proteins to the homopolynucleotides poly(dA), poly(dU) and poly(dT), were studied by means of fluorescence-titration experiments. The binding stoichiometry and fluorescence quenching of the mutant proteins are equal to, or lower than, the wild-type gene V protein values. In addition, all proteins measured bind a more-or-less co-operative manner to single-stranded DNA. The binding affinities for poly(dA) decrease in the following order: Y61H greater than wild-type greater than F68L and R16H greater than Y41F and Y41H greater than F73L greater than R21C greater than Y34H greater than G18D/Y56H. Possible explanations for the observed differences are discussed. The conservation of binding affinity, also for mutations in the single-stranded DNA-binding domain, suggests that the binding to homopolynucleotides is largely non-specific.