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.

PDBe: Protein Data Bank in Europe.

The Protein Data Bank in Europe (PDBe; pdbe.org) is a partner in the Worldwide PDB organization (wwPDB; wwpdb.org) and as such actively involved in managing the single global archive of biomacromolecular structure data, the PDB. In addition, PDBe develops tools, services and resources to make structure-related data more accessible to the biomedical community. Here we describe recently developed, extended or improved services, including an animated structure-presentation widget (PDBportfolio), a widget to graphically display the coverage of any UniProt sequence in the PDB (UniPDB), chemistry- and taxonomy-based PDB-archive browsers (PDBeXplore), and a tool for interactive visualization of NMR structures, corresponding experimental data as well as validation and analysis results (Vivaldi).

On the structures of filamentous bacteriophage Ff (fd, f1, M13).

The filamentous bacteriophage (Inovirus) strain Ff (fd, f1, M13) is widely used in molecular biophysics as a simple model system. A low resolution molecular model of the fd protein coat has been reported, derived from iterative helical real space reconstruction of cryo-electron micrographs (cryoEM). This model is significantly different from the model previously derived from X-ray fibre diffraction and solid-state NMR. We show that the cryoEM model agrees neither with solid-state NMR data nor with X-ray fibre diffraction data of fd, and has some puzzling structural features, for instance nanometre holes through the protein coat. We refine the cryoEM model against the X-ray data, and find that the model after refinement closely approximates the model derived directly from X-ray fibre diffraction and solid-state NMR data. We suggest possible reasons for the differences between the models derived from cryoEM and X-ray diffraction.

Molecular structure of fd (f1, M13) filamentous bacteriophage refined with respect to X-ray fibre diffraction and solid-state NMR data supports specific models of phage assembly at the bacterial membrane.

Filamentous bacteriophage (Inovirus) is a simple and well-characterized model system. The phage particle, or virion, is about 60 angstroms in diameter and several thousand angstrom units long. The virions are assembled at the bacterial membrane as they extrude out of the host without killing it, an example of specific transport of nucleoprotein assemblages across membranes. The Ff group (fd, f1 and M13) has been especially widely studied. Models of virion assembly have been proposed based on a molecular model of the fd virion derived by X-ray fibre diffraction. A somewhat different model of the fd virion using solid-state NMR data has been proposed, not consistent with these models of assembly nor with the X-ray diffraction data. Here we show that reinterpreted NMR data are also consistent with the model derived from X-ray fibre diffraction studies, and discuss models of virion assembly.

A duplicated fold is the structural basis for polynucleotide phosphorylase catalytic activity, processivity, and regulation.

BACKGROUND: Polynucleotide phosphorylase (PNPase) is a polyribonucleotide nucleotidyl transferase (E.C.2.7.7.8) that degrades mRNA in prokaryotes. Streptomyces antibioticus PNPase also assays as a guanosine 3'-diphosphate 5'-triphosphate (pppGpp) synthetase (E.C.2.7.6.5). It may function to coordinate changes in mRNA lifetimes with pppGpp levels during the Streptomyces lifecycle. RESULTS: The structure of S. antibioticus PNPase without bound RNA but with the phosphate analog tungstate bound at the PNPase catalytic sites was determined by X-ray crystallography and shows a trimeric multidomain protein with a central channel. The structural core has a novel duplicated architecture formed by association of two homologous domains. The tungstate derivative structure reveals the PNPase active site in the second of these core domains. Structure-based sequence analysis suggests that the pppGpp synthetase active site is located in the first core domain. CONCLUSIONS: This is the first structure of a PNPase and shows the structural basis for the trimer assembly, the arrangement of accessory RNA binding domains, and the likely catalytic residues of the PNPase active site. A possible function of the trimer channel is as a contribution to both the processivity of degradation and the regulation of PNPase action by RNA structural elements.

The molecular structure and structural transition of the alpha-helical capsid in filamentous bacteriophage Pf1.

The major coat protein in the capsid of Pf1 filamentous bacteriophage (Inovirus) forms a helical assembly of about 7000 identical protein subunits, each of which contains 46 amino-acid residues and can be closely approximated by a single gently curved alpha-helix. Since the viral DNA occupies the core of the tubular capsid and appears to make no significant specific interactions with the capsid proteins, the capsid is a simple model system for the study of the static and dynamic properties of alpha-helix assembly. The capsid undergoes a reversible temperature-induced structural transition at about 283 K between two slightly different helix forms. The two forms can coexist without an intermediate state, consistent with a first-order structural phase transition. The molecular model of the higher temperature form was refined using improved X-ray fibre diffraction data and new refinement and validation methods. The refinement indicates that the two forms are related by a change in the orientation of the capsid subunits within the virion, without a significant change in local conformation of the subunits. On the higher temperature diffraction pattern there is a region of observed intensity that is not consistent with a simple helix of identical subunits; it is proposed that the structure involves groups of three subunits which each have a slightly different orientation within the group. The grouping of subunits suggests that a change in subunit libration frequency could be the basis of the Pf1 structural transition; calculations from the model are used to explore this idea.

Structure of the capsid of Pf3 filamentous phage determined from X-ray fibre diffraction data at 3.1 A resolution.

We have recorded X-ray diffraction patterns at 3.1 A resolution from magnetically aligned fibres of the Pf3 strain of filamentous bacteriophage (Inovirus). The patterns are similar to patterns from the higher-temperature form of the Pf1 strain, indicating that the Pf3 and Pf1 virions have the same helix symmetry and similar protein subunit shape. This is of particular interest, given that the primary structures of the two protein subunits are quite different; and the nucleotide/protein subunit ratio in the Pf3 virion is more than twice that in Pf1, indicating important differences in DNA packaging. We have built a molecular model of the Pf3 protein capsid based on the model of Pf1, and refined it against the diffraction data using simulated annealing. The refinement confirms that the two structures are similar, which may reflect a fundamental motif of alpha-helix packing. However, there are some differences between the structures: the Pf3 subunit appears to be completely alpha-helical, beginning at the N terminus, whereas the first few residues of the Pf1 subunit are not helical; and the structure of the C-terminal region of the Pf3 subunit at the inner surface of the tubular capsid indicates that DNA/protein interactions in this virion may involve both aromatic side-chains and positively charged side-chains, whereas those in the Pf1 virion involve predominantly only the latter. In the course of this work, we have developed new approaches to refinement and validation of helical structures with respect to continuous transform fibre diffraction data.

X-ray diffraction and far-UV CD studies of filaments formed by a leucine-rich repeat peptide: structural similarity to the amyloid fibrils of prions and Alzheimer's disease beta-protein.

The development of neuro-degenerative diseases often involves amyloidosis, that is the formation of polymeric fibrillar structures from normal cellular proteins or peptides. For example, in Alzheimer's disease, a 42 amino acid peptide processed from the amyloid precursor protein forms filaments with a beta-sheet structure. Because of this, the structure and dynamics of polymeric peptide filaments is of considerable interest. We showed previously that a 23 amino acid peptide constituting a single leucine-rich repeat (LRRN) polymerises spontaneously in solution to form long filaments of a beta-sheet structure, a property similar to that of Alzheimer's beta-amyloid and prion peptides. Here we report that a variant of LRRN in which a highly conserved asparagine residue is replaced by aspartic acid does not form either filaments or beta structure. By contrast, a variant which replaces this asparagine residue with glutamine forms filaments ultrastructurally indistinguishable from those of LRRN. Electron micrographs of LRRN filaments show that many consist of two interleaved strands which appear to have a ribbon-like morphology. X-ray diffraction patterns from oriented LRRN fibres reveal that they are composed of long beta-sheet arrays, with the interstrand hydrogen bonding parallel to the filament axis. This 'cross-beta' structure is similar to that adopted by beta-amyloid and prion derived fibres. Taken together, these results indicate that the LRR filaments are stabilised by inter- or intra-strand hydrogen bonded interactions comparable to the asparagine ladders of beta-helix proteins or the 'glutamine zippers' of poly-glutamine peptides. We propose that similar stabilising interactions may underlie a number of characterised predispositions to neuro-degenerative diseases that are caused by mutations to amide residues. Our finding that amyloid-like filaments can form from a peptide motif not at present correlated with degenerative disease suggests that a propensity for beta-filament formation is a common feature of protein sub-domains.

Dynamic properties of nucleated microtubules: GTP utilisation in the subcritical concentration regime.

Microtubule assembly kinetics have been studied quantitatively under solution conditions supporting microtubule dynamic instability. Purified GTP-tubulin (Tu-GTP) and covalently cross-linked short microtubule seeds (EGS-seeds; Koshland et al. (1988) Nature 331, 499) were used with and without biotinylation. Under sub-critical concentration conditions ([Tu-GTP] < 5.3 microM), significant microtubule growth of limited length was observed on a proportion of the EGS-seeds by immuno-electron microscopy. A sensitive fluorescence assay for microtubule GDP production was developed for parallel assessment of GTP utilisation. This revealed a correlation between the detected microtubule growth and the production of tubulin-GDP, deriving from the shortening phase of the dynamic microtubules. This correlation was confirmed by the action of nocodazole, a specific inhibitor of microtubule assembly, that was found to abolish the GDP release. The variation of the GDP release with tubulin concentration (Jh(c) plot) was determined below the critical concentration (Cc). The GDP production observed was consistent with the elongation of the observed seeded microtubules with an apparent rate constant of 1.5 x 10(6) M-1 second-1 above a threshold of approximately 1 microM tubulin. The form of this Jh(c) plot for elongation below Cc is reproduced by the Lateral Cap model for microtubule dynamic instability adapted for seeded assembly. The behaviour of the system is contrasted with that previously studied in the absence of detectable microtubule elongation (Caplow and Shanks (1990) J. Biol. Chem. 265, 8935-8941). The approach provides a means of monitoring microtubule dynamics at concentrations inaccessible to optical microscopy, and shows that essentially the same dynamic mechanisms apply at all concentrations. Numerical simulation of the subcritical concentration regime shows dynamic growth features applicable to the initiation of microtubule growth in vivo.

Role of capsid structure and membrane protein processing in determining the size and copy number of peptides displayed on the major coat protein of filamentous bacteriophage.

Filamentous bacteriophage virions can be engineered to display small foreign peptides in the N-terminal regions of all 2700 copies of the major coat protein (pVIII), but larger peptides can be accommodated only in hybrid virions, in which modified and wild-type coat protein subunits are interspersed. The copy number of peptides accepted in hybrid virions is generally believed to be related to peptide size: the larger the insert, the lower the number of modified coat protein subunits in the assembled virion. However, we show here that some large peptides can be displayed at a much higher copy number than smaller ones and that some relatively small peptides are poorly displayed, if at all, in hybrid virions. X-ray diffraction studies of a recombinant virion together with model building experiments with peptide and protein epitopes of known structure demonstrated that it is feasible to accommodate much larger structures, without perturbation of the capsid protein packing, than it has proved possible to generate in vivo. We show further that the insertion of certain peptides greatly slowed or even prevented the processing of the pVIII pro-coat by leader peptidase at the inner membrane of the Escherichia coli cell. A good correlation was found between the effect of the insert on the rate of the processing of the pro-coat, an essential step in virus assembly, and the number of the mature but modified proteins in the subsequently assembled hybrid virion. These results have important implications for the design of peptide display systems based on filamentous bacteriophage.

In vitro assembly of microtubule protein with GTP and 2'dGTP: kinetic evidence for a preassembly conformational change.

The assembly of chick brain microtubule protein in a NaCl-supplemented buffer has been examined with respect to nucleation and the subsequent elongation as a function of the nucleotide (GTP vs 2'dGTP), and the protein and nucleotide concentrations. The kinetics suggest that unassembled tubulin can exist in two conformational states (termed Tu1,GTP and Tu2,GIP when GTP is bound to the exchangeable site), with Tu1,GTP contributing to nucleation and Tu2,GTP participating in elongation. The extent of self-nucleation is proposed to be determined, in part, by the rate constant governing this conformational change. This analysis contrasts with that of earlier studies, which concluded that the number of subunits interacting to form an effective nucleus could be estimated from the dependency of self-nucleation on the initial concentration of unassembled tubulin.

Matching electrostatic charge between DNA and coat protein in filamentous bacteriophage. Fibre diffraction of charge-deletion mutants.

The virion of Ff (fd, f1, M13) filamentous bacteriophage consists of a long tube of coat protein subunits in a shingled, helical array, surrounding a genome of circular single-stranded DNA. Modified fd virions have been generated by a mutation (K48A) that removes one positive charge from each coat protein subunit in the C-terminal region of the polypeptide chain facing the DNA. The number of nucleotides in the mutant DNA is unchanged, but the K48A virions are 35% longer than wild-type. We have measured the X-ray diffraction attributable to single virions in hydrated gels of wild-type and K48A bacteriophages. Most of the diffraction pattern shows no significant difference between wild-type and K48A. Since the DNA is only about 12% by weight of the wild-type virion, the diffraction pattern is dominated by the protein contribution, and the absence of significant differences indicates that there are no significant changes in the symmetry or structure of the protein coat. But there is a change in the diffraction pattern in a region where the DNA and protein contributions are comparable. The diffraction pattern of the K48A mutant shows an increase in intensity of one of the weaker equatorial peaks, relative to wild-type, in a region where the protein contribution has negative sign but the DNA contribution has positive sign. This is consistent with a decrease in the ratio of DNA:protein per unit length of the K48A mutant. The results support the view that the protein forms a sheath lined with positive charges interacting electrostatically and non-specifically with a negatively charged DNA core of matching charge density. The lower positive charge density lining the capsid in the K48A mutant means that correspondingly fewer nucleotides can be packaged per coat protein subunit, which in turn requires an elongation of the DNA inside the virion. A longer virion is thus required to package the same amount of DNA. Within the error of measurement, the number of positive charges on the protein interacting with the DNA is the same in K48A as in the wild-type, despite the fact that the mutant is 35% longer than the wild-type.

Assembly of chick brain MAP2-tubulin microtubule protein. Analysis of tubulin subunit flux rates by immunofluorescence microscopy.

A filter-based immunofluorescence-microscopy method for obtaining microtubule lengths has been developed and evaluated. Kinetic constants and mean lengths obtained show close agreement with those obtained by complementary methods applied to chick brain MAP2-tubulin microtubule protein in NaCl-supplemented buffer. The filter-based method has been used to estimate tubulin subunit flux (Jon) resulting from isothermal dilution of microtubule populations to various free tubulin concentrations, (c). This experimental Jon(c) plot is significantly different from that predicted by a variety of theoretical models, but is consistent with a 'lateral cap' model of dynamic instability [Bayley, Schilstra & Martin (1990) J. Cell. Sci. 95, 33-48] adapted to accommodate the observed vectorial GTP hydrolysis.