Naturally occurring singleton residues in AAV capsid impact vector performance and illustrate structural constraints.

Vectors based on the adeno-associated virus (AAV) are attractive and versatile vehicles for in vivo gene transfer. The virus capsid is the primary interface with the cell that defines many pharmacological, immunological and molecular properties. Determinants of these interactions are often restricted to a limited number of capsid amino acids. In this study, a portfolio of novel AAV vectors was developed after a structure-function analysis of naturally occurring AAV capsid isolates. Singletons, which are particular residues on the AAV capsid that were variable in otherwise conserved amino acid positions, were found to impact on vector's ability to be manufactured or to transduce. Data for those residues that mapped to monomer-monomer interface regions on the particle structure suggested a role in particle assembly. The change of singleton residues to the conserved amino acid resulted in the rescue of many isolates that were defective on initial isolation. This led to the development of an AAV vector portfolio that encompasses six different clades and 3 other distinct AAV niches. Evaluation of the in vivo gene transfer efficiency of this portfolio after intravenous and intramuscular administration highlighted a clade-specific tropism. These studies further the design and selection of AAV capsids for gene therapy applications.

Identification of aleutian mink disease parvovirus capsid sequences mediating antibody-dependent enhancement of infection, virus neutralization, and immune complex formation.

Aleutian mink disease parvovirus (ADV) causes a persistent infection associated with circulating immune complexes, immune complex disease, hypergammaglobulinemia, and high levels of antiviral antibody. Although antibody can neutralize ADV infectivity in Crandell feline kidney cells in vitro, virus is not cleared in vivo, and capsid-based vaccines have proven uniformly ineffective. Antiviral antibody also enables ADV to infect macrophages, the target cells for persistent infection, by Fc-receptor-mediated antibody-dependent enhancement (ADE). The antibodies involved in these unique aspects of ADV pathogenesis may have specific targets on the ADV capsid. Prominent differences exist between the structure of ADV and other, more-typical parvoviruses, which can be accounted for by short peptide sequences in the flexible loop regions of the capsid proteins. In order to determine whether these short sequences are targets for antibodies involved in ADV pathogenesis, we studied heterologous antibodies against several peptides present in the major capsid protein, VP2. Of these antibodies, a polyclonal rabbit antibody to peptide VP2:428-446 was the most interesting. The anti-VP2:428-446 antibody aggregated virus particles into immune complexes, mediated ADE, and neutralized virus infectivity in vitro. Thus, antibody against this short peptide can be implicated in key facets of ADV pathogenesis. Structural modeling suggested that surface-exposed residues of VP2:428-446 are readily accessible for antibody binding. The observation that antibodies against a single target peptide in the ADV capsid can mediate both neutralization and ADE may explain the failure of capsid-based vaccines.

Structural and kinetic analysis of the chemical rescue of the proton transfer function of carbonic anhydrase II.

Histidine 64 in human carbonic anhydrase II (HCA II) functions in the catalytic pathway of CO(2) hydration as a shuttle to transfer protons between the zinc-bound water and bulk water. Catalysis of the exchange of (18)O between CO(2) and water, measured by mass spectrometry, is dependent on this proton transfer and was decreased more than 10-fold for H64A HCA II compared with wild-type HCA II. The loss of catalytic activity of H64A HCA II could be rescued by 4-methylimidazole (4-MI), an exogenous proton donor, in a saturable process with a maximum activity of 40% of wild-type HCA II. The crystal structure of the rescued complex at 1.6 A resolution shows 4-MI bound in the active-site cavity of H64A HCA II, through pi stacking interactions with Trp 5 and H-bonding interactions with water molecules. In this location, 4-MI is about 12 A from the zinc and approximates the observed "out" position of His 64 in the structure of the wild-type enzyme. 4-MI appears to compensate for the absence of His 64 and rescues the catalytic activity of the H64A HCA II mutant. This result strongly suggests that the out conformation of His 64 is effective in the transfer of protons between the zinc-bound solvent molecule and solution.

Structure of the Maize streak virus geminate particle.

The Geminiviridae is an extensive family of plant viruses responsible for economically devastating diseases in crops worldwide. Geminiviruses package circular, single-stranded DNA (ssDNA) genomes. The characteristic twinned or "geminate" particles, which consist of two joined, incomplete T = 1 icosahedra, are unique among viruses. We have determined the first structure of a geminivirus particle, the Nigerian strain of Maize streak virus (MSV-N), using cryo-electron microscopy and three-dimensional image reconstruction methods. The particle, of dimensions 220 x 380 A, has an overall 52-point-group symmetry, in which each half particle "head" consists of the coat protein (CP) arranged with quasi-icosahedral symmetry. We have modeled the MSV-N CP as an eight-stranded, antiparallel beta-barrel motif (a structural motif common to all known ssDNA viruses) with an N-terminal alpha-helix. This has produced a model of the geminate particle in which 110 copies of the CP nicely fit into the reconstructed density map. The reconstructed density map and MSV-N pseudo-atomic model demonstrate that the geminate particle has a stable, defined structure.

Mutational analysis of the adeno-associated virus type 2 (AAV2) capsid gene and construction of AAV2 vectors with altered tropism.

Adeno-associated virus type 2 (AAV2) has proven to be a valuable vector for gene therapy. Characterization of the functional domains of the AAV capsid proteins can facilitate our understanding of viral tissue tropism, immunoreactivity, viral entry, and DNA packaging, all of which are important issues for generating improved vectors. To obtain a comprehensive genetic map of the AAV capsid gene, we have constructed 93 mutants at 59 different positions in the AAV capsid gene by site-directed mutagenesis. Several types of mutants were studied, including epitope tag or ligand insertion mutants, alanine scanning mutants, and epitope substitution mutants. Analysis of these mutants revealed eight separate phenotypes. Infectious titers of the mutants revealed four classes. Class 1 mutants were viable, class 2 mutants were partially defective, class 3 mutants were temperature sensitive, and class 4 mutants were noninfectious. Further analysis revealed some of the defects in the class 2, 3, and 4 mutants. Among the class 4 mutants, a subset completely abolished capsid formation. These mutants were located predominantly, but not exclusively, in what are likely to be beta-barrel structures in the capsid protein VP3. Two of these mutants were insertions at the N and C termini of VP3, suggesting that both ends of VP3 play a role that is important for capsid assembly or stability. Several class 2 and 3 mutants produced capsids that were unstable during purification of viral particles. One mutant, R432A, made only empty capsids, presumably due to a defect in packaging viral DNA. Additionally, five mutants were defective in heparan binding, a step that is believed to be essential for viral entry. These were distributed into two amino acid clusters in what is likely to be a cell surface loop in the capsid protein VP3. The first cluster spanned amino acids 509 to 522; the second was between amino acids 561 and 591. In addition to the heparan binding clusters, hemagglutinin epitope tag insertions identified several other regions that were on the surface of the capsid. These included insertions at amino acids 1, 34, 138, 266, 447, 591, and 664. Positions 1 and 138 were the N termini of VP1 and VP2, respectively; position 34 was exclusively in VP1; the remaining surface positions were located in putative loop regions of VP3. The remaining mutants, most of them partially defective, were presumably defective in steps of viral entry that were not tested in the preliminary screening, including intracellular trafficking, viral uncoating, or coreceptor binding. Finally, in vitro experiments showed that insertion of the serpin receptor ligand in the N-terminal regions of VP1 or VP2 can change the tropism of AAV. Our results provide information on AAV capsid functional domains and are useful for future design of AAV vectors for targeting of specific tissues.

A beta-stranded motif drives capsid protein oligomers of the parvovirus minute virus of mice into the nucleus for viral assembly.

The determinants of nuclear import in the VP-1 and VP-2 capsid proteins of the parvovirus minute virus of mice strain i (MVMi) synthesized in human fibroblasts were sought by genetic analysis in an infectious plasmid. Immunofluorescence of transfected cells revealed that the two proteins were involved in cooperative cytoplasmic interactions for nuclear cotransport. However, while VP-1 translocated regardless of extension of deletions and did not form capsid epitopes by itself, VP-2 seemed to require cytoplasmic folding and the overall conformation for nuclear transport. The sequence (528)KGKLTMRAKLR(538) was found necessary for nuclear uptake of VP-2, even though it was not sufficient to confer a nuclear localization capacity on a heterologous protein. In the icosahaedral MVMi capsid, this sequence forms the carboxy end of the amphipathic beta-strand I (betaI), and all its basic residues are contiguously positioned at the face that in the unassembled subunit would be exposed to solvent. Mutations in singly expressed VP-2 that either decrease the net basic charge of the exposed face (K530N-R534T), perturb the hydrophobicity of the opposite face (L531E), or distort the betaI conformation (G529P) produced cytoplasmic subviral oligomers. Particle formation by betaI mutants indicated that the basic residues clustered at one face of betaI drive VP oligomers into the nucleus preceding and uncoupled to assembly and that the nuclear environment is required for MVMi capsid formation in the infected cell. The degree of VP-1/VP-2 transport cooperativity suggests that VP trimers are the morphogenetic intermediates translocating through the nuclear pore. The results support a model in which nuclear transport signaling preserves the VP-1/VP-2 stoichiometry necessary for efficient intranuclear assembly and in which the beta-stranded VP-2 nuclear localization motif contributes to the quality control of viral morphogenesis.

Biochemical and physical characterization of parvovirus minute virus of mice virus-like particles.

The VP-2 major capsid protein of the prototype strain of the parvovirus minute virus of mice (MVMp) was expressed, using a baculovirus vector, in Sf9 insect cells. Immunogold electron microscopy of infected Sf9 cells showed VP-2 localized in the nucleus and cytoplasm as is observed in mammalian cells during natural infections. The VP-2 subunits self-assembled into empty parvovirus-like particles (VLPs), which appeared morphologically similar to and immunogenically indistinguishable from native empty MVMp particles, which also contain the minor capsid protein, VP1. Incubations under different pH and temperature conditions showed that the MVMp VLPs and native empty MVMp capsids share comparable stability. Once heated the particles can be similarly and specifically cleaved by trypsin at the VP-2 N-terminal domain. This process mimics the further maturation of the "rat-like" parvovirus virions, following viral DNA encapsidation, indicating that biologically relevant features of the MVMp capsid are maintained in the VLPs. Crystals have been obtained for the MVMp VLPs which were isomorphous to those used for the high-resolution structure determination of virions and native empty particles of the immunosuppressive strain of MVM (MVMi). The VLP crystals diffracted X rays to beyond 3-A resolution and are in space group C2 (a = 448.7, b = 416.6, c = 306.1 A, and beta = 95.9 degrees ). This is the first report of crystals from parvoviral particles produced in a heterologous system diffracting X rays to high resolution, indicating that VP-2 of some parvovirus capsids can self-assemble into ordered T = 1 icosahedral capsids in the absence of other viral and host cell functions.

Three-dimensional structure of Aleutian mink disease parvovirus: implications for disease pathogenicity.

The three-dimensional structure of expressed VP2 capsids of Aleutian mink disease parvovirus strain G (ADVG-VP2) has been determined to 22 A resolution by cryo-electron microscopy and image reconstruction techniques. A structure-based sequence alignment of the VP2 capsid protein of canine parvovirus (CPV) provided a means to construct an atomic model of the ADVG-VP2 capsid. The ADVG-VP2 reconstruction reveals a capsid structure with a mean external radius of 128 A and several surface features similar to those found in human parvovirus B19 (B19), CPV, feline panleukopenia virus (FPV), and minute virus of mice (MVM). Dimple-like depressions occur at the icosahedral twofold axes, canyon-like regions encircle the fivefold axes, and spike-like protrusions decorate the threefold axes. These spikes are not present in B19, and they are more prominent in ADV compared to the other parvoviruses owing to the presence of loop insertions which create mounds near the threefold axes. Cylindrical channels along the fivefold axes of CPV, FPV, and MVM, which are surrounded by five symmetry-related beta-ribbons, are closed in ADVG-VP2 and B19. Immunoreactive peptides made from segments of the ADVG-VP2 capsid protein map to residues in the mound structures. In vitro tissue tropism and in vivo pathogenic properties of ADV map to residues at the threefold axes and to the wall of the dimples.

Functional implications of the structure of the murine parvovirus, minute virus of mice.

BACKGROUND: Minute virus of mice (MVM) is a single-stranded (ss) DNA-containing, murine parvovirus with a capsid built up of 60 icosahedrally related polypeptide chains, each of which consists of the C-terminal region common to two structural proteins, VP1 and VP2. In infectious virions, most VP2 molecules are cleaved to VP3 by the removal of about 20 amino acids from the N terminus. Of the 587 amino acids in VP2, approximately half are identical to those in the analogous capsid protein of the antigenically distinct canine parvovirus (CPV), the crystal structure of which has previously been determined. The three-dimensional structure determination of MVMi (the immunosuppressive strain of MVM) was previously reported to 3.5 A resolution. RESULTS: We report here an analysis of the MVMi virus structure and provide insights into tissue tropism, antigenicity and DNA packaging. Amino acids determining MVM tissue tropism were found to cluster on, or near, the viral surface. A conserved, glycine-rich, N-terminal peptide was seen to run through a cylindrical channel along each fivefold axis and may have implications for antigenicity. Density within the virion was interpreted as 29 ssDNA nucleotides per icosahedral asymmetric unit, and accounts for over one-third of the viral genome. CONCLUSIONS: The presence of the glycine-rich sequence in the fivefold channels of MVMi provides a possible mechanism to explain how the unique N-terminal region of VP1 becomes externalized in infectious parvovirions. Residues that determine tropism may form an attachment recognition site for a secondary host-cell factor that modulates tissue specificity. The ordering of nucleotides in a similar region of the interior surface in the CPV and MVMi capsids suggests the existence of a genomic DNA-recognition site within the parvoviral capsid.

Structural analysis of the Spiroplasma virus, SpV4: implications for evolutionary variation to obtain host diversity among the Microviridae.

BACKGROUND: Spiroplasma virus, SpV4, is a small, non-enveloped virus that infects the helical mollicute Spiroplasma melliferum. SpV4 exhibits several similarities to the Chlamydia phage, Chp1, and the Coliphages alpha 3, phi K, G4 and phi X174. All of these viruses are members of the Microviridae. These viruses have isometric capsids with T = 1 icosahedral symmetry, cause lytic infections and are the only icosahedral phages that contain single-stranded circular DNA genomes. The aim of this comparative study on these phages was to understand the role of their capsid proteins during host receptor recognition. RESULTS: The three-dimensional structure of SpV4 was determined to 27 A resolution from images of frozen-hydrated particles. Cryo-electron microscopy (cryo-EM) revealed 20, approximately 54 A long, 'mushroom-like' protrusions on the surface of the capsid. Each protrusion comprises a trimeric structure that extends radially along the threefold icosahedral axes of the capsid. A 71 amino acid portion of VP1 (the SpV4 capsid protein) was shown, by structural alignment with the atomic structure of the F capsid protein of phi X174, to represent an insertion sequence between the E and F strands of the eight-stranded antiparallel beta-barrel. Secondary structure prediction of this insertion sequence provided the basis for a probable structural motif, consisting of a six-stranded antiparallel beta sheet connected by small turns. Three such motifs form the rigid stable trimeric structures (mushroom-like protrusions) at the threefold axes, with hydrophobic depressions at their distal surface. CONCLUSIONS: Sequence alignment and structural analysis indicate that distinct genera of the Microviridae might have evolved from a common primordial ancestor, with capsid surface variations, such as the SpV4 protrusions, resulting from gene fusion events that have enabled diverse host ranges. The hydrophobic nature of the cavity at the distal surface of the SpV4 protrusions suggests that this region may function as the receptor-recognition site during host infection.

Expression of Aleutian mink disease parvovirus capsid proteins in defined segments: localization of immunoreactive sites and neutralizing epitopes to specific regions.

The capsid proteins of the ADV-G isolate of Aleutian mink disease parvovirus (ADV) were expressed in 10 nonoverlapping segments as fusions with maltose-binding protein in pMAL-C2 (pVP1, pVP2a through pVP2i). The constructs were designed to capture the VP1 unique sequence and the portions analogous to the four variable surface loops of canine parvovirus (CPV) in individual fragments (pVP2b, pVP2d, pVP2e, and pVP2g, respectively). The panel of fusion proteins was immunoblotted with sera from mink infected with ADV. Seropositive mink infected with either ADV-TR, ADV-Utah, or ADV-Pullman reacted preferentially against certain segments, regardless of mink genotype or virus inoculum. The most consistently immunoreactive regions were pVP2g, pVP2e, and pVP2f, the segments that encompassed the analogs of CPV surface loops 3 and 4. The VP1 unique region was also consistently immunoreactive. These findings indicated that infected mink recognize linear epitopes that localized to certain regions of the capsid protein sequence. The segment containing the hypervariable region (pVP2d), corresponding to CPV loop 2, was also expressed from ADV-Utah. An anti-ADV-G monoclonal antibody and a rabbit anti-ADV-G capsid antibody reacted exclusively with the ADV-G pVP2d segment but not with the corresponding segment from ADV-Utah. Mink infected with ADV-TR or ADV-Utah also preferentially reacted with the pVP2d sequence characteristic of that virus. These results suggested that the loop 2 region may contain a type-specific linear epitope and that the epitope may also be specifically recognized by infected mink. Heterologous antisera were prepared against the VP1 unique region and the four segments capturing the variable surface loops of CPV. The antisera against the proteins containing loop 3 or loop 4, as well as the anticapsid antibody, neutralized ADV-G infectivity in vitro and bound to capsids in immune electron microscopy. These results suggested that regions of the ADV capsid proteins corresponding to surface loops 3 and 4 of CPV contain linear epitopes that are located on the external surface of the ADV capsid. Furthermore, these linear epitopes contain neutralizing determinants. Computer comparisons with the CPV crystal structure suggest that these sequences may be adjacent to the threefold axis of symmetry of the viral particle.

Structure determination of minute virus of mice.

The three-dimensional crystal structure of the single-stranded DNA-containing ('full') parvovirus, minute virus of mice (MVM), has been determined to 3.5 A resolution. Both full and empty particles of MVM were crystallized in the monoclinic space group C2 with cell dimensions of a = 448.7, b = 416.7, c = 305.3 A and beta = 95.8 degrees. Diffraction data were collected at the Cornell High Energy Synchrotron Source using an oscillation camera. The crystals have a pseudo higher R32 space group in which the particles are situated at two special positions with 32 point symmetry, separated by (1/2)c in the hexagonal setting. The self-rotation function showed that the particles are rotated with respect to each other by 60 degrees around the pseudo threefold axis. Subsequently, a more detailed analysis of the structure amplitudes demonstrated that the correct space-group symmetry is C2 as given above. Only one of the three twofold axes perpendicular to the threefold axis in the pseudo R32 space group is a 'true' crystallographic twofold axis; the other two are only 'local' non-crystallographic symmetry axes. The known canine parvovirus (CPV) structure was used as a phasing model to initiate real-space electron-density averaging phase improvement. The electron density was easily interpretable and clearly showed the amino-acid differences between MVM and CPV, although the final overall correlation coefficient was only 0.63. The structure of MVM has a large amount of icosahedrally ordered DNA, amounting to 22% of the viral genome, which is significantly more than that found in CPV.

Structural analysis of a mutation in canine parvovirus which controls antigenicity and host range.

A single mutation in canine parvovirus (CPV) of VP2 residue 300 from alanine to aspartic acid causes a loss of canine host range and alters the antigenic properties of the virus. The three-dimensional structure of this mutant has been solved to 3.25 A resolution. Crystals of full particles were triclinic, with cell dimensions of a = 267.6, b = 268.5, c = 274.3 A. alpha = 61.9, beta = 62.6, and gamma = 60.2 degrees. The native structure of CPV was used as an initial model. Phases were improved by real-space electron density averaging. In spite of the relative low percentage of observed reflections (32.5% of the data between 15.0 and 3.25 A resolution), the presence of 60-fold noncrystallographic redundancy allowed the averaging procedure to converge smoothly. The mutant aspartic acid at residue 300 forms a salt bridge with Arg81 in an icosahedrally threefold-related subunit, inducing local changes within the antigenic site B on the CPV surface. In addition, the loop between residues 359 and 374 adopts a conformation similar to that displayed by feline panleukopenia virus. The ability of the Ala300-->Asp mutant to evade antibody binding can be associated with the change of charge distribution and structure in the antigenic binding site. The variation in host range behavior may be due to the increased stability as a result of formation of the salt bridge between adjacent subunits.

Cryo-electron microscopy studies of empty capsids of human parvovirus B19 complexed with its cellular receptor.

The three-dimensional structures of human parvovirus B19 VP2 capsids, alone and complexed with its cellular receptor, globoside, have been determined to 26 resolution. The B19 capsid structure, reconstructed from cryo-electron micrographs of vitrified specimens, has depressions on the icosahedral 2-fold and 3-fold axes, as well as a canyon-like region around the 5-fold axes. Similar results had previously been found in an 8 angstrom resolution map derived from x-ray diffraction data. Other parvoviral structures have a cylindrical channel along the 5-fold icosahedral axes, whereas density covers the 5-fold axes in B19. The glycolipid receptor molecules bind into the depressions on the 3-fold axes of the B19:globoside complex. A model of the tetrasaccharide component of globoside, organized as a trimeric fiber, fits well into the difference density representing the globoside receptor. Escape mutations to neutralizing antibodies map onto th capsid surface at regions immediately surrounding the globoside attachment sites. The proximity of the antigenic epitopes to the receptor site suggests that neutralization of virus infectivity is caused by preventing attachment of viruses to cells.