Comparison of forces acting on maxillary incisors during tracheal intubation with different laryngoscopy techniques: a blinded manikin study.

Dental trauma is a common complication of tracheal intubation. As existing evidence is insufficient to validly assess the impact of different laryngoscopy techniques on the incidence of dental trauma, the force exerted onto dental structures during tracheal intubation was investigated. An intubation manikin was equipped with hidden force sensors in all maxillary incisors. Dental force was measured while 104 anaesthetists performed a series of tracheal intubations using direct laryngoscopy with a Macintosh blade, and videolaryngoscopy with a C-MAC® , or the hyperangulated GlideScope® or KingVision® laryngoscopes in both normal and difficult airway conditions. A total of 624 tracheal intubations were analysed. The median (IQR [range]) peak force of direct laryngoscopy in normal airways was 21.1 (14.0-32.8 [2.3-127.6]) N and 29.3 (17.7-44.8 [3.3-97.2]) N in difficult airways. In normal airways, these were lower with the GlideScope and KingVision hyperangulated laryngoscopes, with a reduction of 4.6 N (p = 0.006) and 10.9 N (p < 0.001) compared with direct laryngoscopy, respectively. In difficult airways, these were lower with the GlideScope and KingVision hyperangulated laryngoscopes, with a reduction of 9.8 N (p < 0.001) and 17.6 N (p < 0.001) compared with direct laryngoscopy, respectively. The use of the C-MAC did not have an impact on the median peak force. Although sex of anaesthetists did not affect peak force, more experienced anaesthetists generated a higher peak force than less experienced providers. We conclude that hyperangulated videolaryngoscopy was associated with a significantly decreased force exerted on maxillary incisors and might reduce the risk for dental injury in clinical settings.

Treatment of multifocal breast cancer by systemic delivery of dual-targeted adeno-associated viral vectors.

Adeno-associated viral (AAV) vectors yield high potential for clinical gene therapy but, like for other vectors systems, they frequently do not sufficiently transduce the target tissue and their unspecific tropism prevents their application for multifocal diseases such as disseminated cancer. Targeted AAV vectors have been obtained from random AAV display peptide libraries but so far, all vector variants selected from AAV libraries upon systemic administration in vivo retained some collateral tropism, frequently the heart. Here we explored, if this impediment can be overcome by microRNA-regulated transgene cassettes as the combination of library-derived capsid targeting and micro-RNA control has not been evaluated so far. We used a tumor-targeted AAV capsid variant (ESGLSQS) selected from random AAV-display peptide libraries in vivo with remaining off-target tropism toward the heart and regulated targeted transgene expression in vivo by complementary target elements for heart-specific microRNA (miRT-1d). Although this vector still maintained its strong transduction capacity for tumor target tissue after intravenous injection, transgene expression in the heart was almost completely abrogated. This strong and completely tumor-specific transgene expression was used for therapeutic gene transfer in an aggressive multifocal, transgenic, polyoma middle T-induced, murine breast cancer model. A therapeutic suicide gene, delivered systemically by this dual-targeted AAV vector to multifocal breast cancer, significantly inhibited tumor growth after one single vector administration while avoiding side effects compared with untargeted vectors.

Novel random peptide libraries displayed on AAV serotype 9 for selection of endothelial cell-directed gene transfer vectors.

We have demonstrated the potential of random peptide libraries displayed on adeno-associated virus (AAV)2 to select for AAV2 vectors with improved efficiency for cell type-directed gene transfer. AAV9, however, may have advantages over AAV2 because of a lower prevalence of neutralizing antibodies in humans and more efficient gene transfer in vivo. Here we provide evidence that random peptide libraries can be displayed on AAV9 and can be utilized to select for AAV9 capsids redirected to the cell type of interest. We generated an AAV9 peptide display library, which ensures that the displayed peptides correspond to the packaged genomes and performed four consecutive selection rounds on human coronary artery endothelial cells in vitro. This screening yielded AAV9 library capsids with distinct peptide motifs enabling up to 40-fold improved transduction efficiencies compared with wild-type (wt) AAV9 vectors. Incorporating sequences selected from AAV9 libraries into AAV2 capsids could not increase transduction as efficiently as in the AAV9 context. To analyze the potential on endothelial cells in the intact natural vascular context, human umbilical veins were incubated with the selected AAV in situ and endothelial cells were isolated. Fluorescence-activated cell sorting analysis revealed a 200-fold improved transduction efficiency compared with wt AAV9 vectors. Furthermore, AAV9 vectors with targeting sequences selected from AAV9 libraries revealed an increased transduction efficiency in the presence of human intravenous immunoglobulins, suggesting a reduced immunogenicity. We conclude that our novel AAV9 peptide library is functional and can be used to select for vectors for future preclinical and clinical gene transfer applications.

The assembly-activating protein promotes capsid assembly of different adeno-associated virus serotypes.

Adeno-associated virus type 2 (AAV2) capsid assembly requires the expression of a virally encoded assembly-activating protein (AAP). By providing AAP together with the capsid protein VP3, capsids are formed that are composed of VP3 only. Electron cryomicroscopy analysis of assembled VP3-only capsids revealed all characteristics of the wild-type AAV2 capsids. However, in contrast to capsids assembled from VP1, VP2, and VP3, the pores of VP3-only capsids were more restricted at the inside of the 5-fold symmetry axes, and globules could not be detected below the 2-fold symmetry axes. By comparing the capsid assembly of several AAV serotypes with AAP protein from AAV2 (AAP-2), we show that AAP-2 is able to efficiently stimulate capsid formation of VP3 derived from several serotypes, as demonstrated for AAV1, AAV2, AAV8, and AAV9. Capsid formation, by coexpressing AAV1-, AAV2-, or AAV5-VP3 with AAP-1, AAP-2, or AAP-5 revealed the ability of AAP-1 and AAP-2 to complement each other in AAV1 and AAV2 assembly, whereas for AAV5 assembly more specific conditions are required. Sequence alignment of predicted AAP proteins from the known AAV serotypes indicates a high degree of homology of all serotypes to AAP-2 with some divergence for AAP-4, AAP-5, AAP-11, and AAP-12. Immunolocalization of assembled capsids from different serotypes confirmed the preferred nucleolar localization of capsids, as observed for AAV2; however, AAV8 and AAV9 capsids could also be detected throughout the nucleus. Taken together, the data show that AAV capsid assembly of different AAV serotypes also requires the assistance of AAP proteins.

Amphipols from A to Z.

Amphipols (APols) are short amphipathic polymers that can substitute for detergents to keep integral membrane proteins (MPs) water soluble. In this review, we discuss their structure and solution behavior; the way they associate with MPs; and the structure, dynamics, and solution properties of the resulting complexes. All MPs tested to date form water-soluble complexes with APols, and their biochemical stability is in general greatly improved compared with MPs in detergent solutions. The functionality and ligand-binding properties of APol-trapped MPs are reviewed, and the mechanisms by which APols stabilize MPs are discussed. Applications of APols include MP folding and cell-free synthesis, structural studies by NMR, electron microscopy and X-ray diffraction, APol-mediated immobilization of MPs onto solid supports, proteomics, delivery of MPs to preexisting membranes, and vaccine formulation.

Novel adenovirus-based helper system to support production of recombinant parvovirus.

Preclinical studies using various cell culture and animal systems highlight the potential of recombinant rodent parvoviruses (recPVs) for cancer therapy. Production of these viruses is, however, not efficient and this hampers the clinical applications of these agents. In this study, we show that the adenovirus genes E2a, E4(orf6) and VA RNA increase the production of recPVs by more than 10-fold and reduce the time of production from 3 to 2 days in HEK293T cells. The helper effects of these genes can be observed with different recPVs, regardless of the nature and size of the inserted transgene. Furthermore, we generated a recombinant Adenovirus 5 carrying the parvovirus VP transcription unit. This helper, named Ad-VP, allows recPVs to be efficiently produced through a protocol based only on cell infection, making possible to use cell lines, such as NB324K, which are good producers of parvoviruses but are hardly transfectable. Hence, we could further improve viral titers and reduce time and costs of production. This Ad-VP helper-based protocol could be scaled up to a bioreactor format for the generation of the large amounts of recPVs needed for future clinical applications.

Heart-targeted adeno-associated viral vectors selected by in vivo biopanning of a random viral display peptide library.

Selection of targeted vectors from virus display peptide libraries is a versatile and efficient approach to improve vector specificity and efficiency. This strategy has been used to target various cell types in vitro. Here, we report the screening of an adeno-associated virus type 2 (AAV2) display peptide library in vivo to select vectors specifically homing to heart tissue after systemic application in mice. Selected library clones indicated superior specificity of gene transfer compared with wild-type AAV2, AAV9 and a heparin binding-deficient AAV2 mutant. Such targeted vectors were able to reconstitute expression of delta-sarcoglycan in the heart of adult delta-sarcoglycan knockout mice after systemic gene transfer in vivo, attesting to the therapeutic potential of this approach.

Augmentation of AAV-mediated cardiac gene transfer after systemic administration in adult rats.

Adeno-associated virus (AAV)-6 or -9-pseudotyped vectors are suitable for efficient cardiac gene transfer after intravenous injection in mice. However, a systemic application in larger animals or humans would require very high doses of viral particles. Therefore, the aim of our study was to test if ultrasound-targeted microbubble destruction could augment cardiac transduction of AAV vectors after intravenous administration in rats. To analyze efficiency and specificity of gene transfer, microbubbles loaded with AAV-6 or -9 harboring a luciferase or enhanced green fluorescent protein (EGFP) reporter gene were infused into the jugular vein of adult Sprague-Dawley rats. During the infusion, high mechanical index ultrasound was administered to the heart. Control rats received the same amount of virus without microbubbles, but with ultrasound. After 4 weeks, organs were harvested and analyzed for reporter gene expression. In contrast to low cardiac expression after systemic transfer of the vector solution without microbubbles, ultrasound-targeted destruction of microbubbles significantly increased cardiac reporter activities between 6- and 20-fold. Analysis of spatial distribution of transgene expression using an AAV-9 vector encoding for EGFP revealed transmural expression predominantly in the left ventricular anterior wall. In conclusion, ultrasound targeted microbubble destruction augments cardiac transduction of AAV vectors in rats. This approach may be suitable for efficient, specific and noninvasive AAV-mediated gene transfer in larger animals or humans.

Protein-lipid interactions with Fusobacterium nucleatum major outer membrane protein FomA: spin-label EPR and polarized infrared spectroscopy.

FomA, the major outer membrane protein of Fusobacterium nucleatum, was expressed and purified in Escherichia coli and reconstituted from detergent in bilayer membranes of phosphatidylcholines with chain lengths from C(12:0) to C(17:0). The conformation and orientation of membrane-incorporated FomA were determined from polarized, attenuated total reflection, infrared (IR) spectroscopy, and lipid-protein interactions with FomA were characterized by using electron paramagnetic resonance (EPR) spectroscopy of spin-labeled lipids. Approximately 190 residues of membranous FomA are estimated to be in a beta-sheet configuration from IR band fitting, which is consistent with a 14-strand transmembrane beta-barrel structure. IR dichroism of FomA indicates that the beta-strands are tilted by approximately 45 degrees relative to the sheet/barrel axis and that the order parameter of the latter displays a discontinuity corresponding to hydrophobic matching with fluid C(13:0) lipid chains. The stoichiometry ( N b = 23 lipids/monomer) of lipid-protein interaction from EPR demonstrates that FomA is not trimeric in membranes of diC(14:0) phosphatidylcholine and is consistent with a monomeric beta-barrel of 14-16 strands. The pronounced selectivity of interaction found with anionic spin-labeled lipids places basic residues of the protein in the vicinity of the polar-apolar membrane interfaces, consistent with current topology models. Comparison with similar data from the 8- to 22-stranded E. coli outer membrane proteins, OmpA, OmpG, and FhuA, supports the above conclusions.

Incorporation of outer membrane protein OmpG in lipid membranes: protein-lipid interactions and beta-barrel orientation.

OmpG is an intermediate size, monomeric, outer membrane protein from Escherichia coli, with n beta = 14 beta-strands. It has a large pore that is amenable to modification by protein engineering. The stoichiometry ( N b = 20) and selectivity ( K r = 0.7-1.2) of lipid-protein interaction with OmpG incorporated in dimyristoyl phosphatidylcholine bilayer membranes was determined with various 14-position spin-labeled lipids by using EPR spectroscopy. The limited selectivity for different lipid species is consistent with the disposition of charged residues in the protein. The conformation and orientation (beta-strand tilt and beta-barrel order parameters) of OmpG in disaturated phosphatidylcholines of odd and even chain lengths from C(12:0) to C(17:0) was determined from polarized infrared spectroscopy of the amide I and amide II bands. A discontinuity in the protein orientation (deduced from the beta-barrel order parameters) is observed at the point of hydrophobic matching of the protein with lipid chain length. Compared with smaller (OmpA; n beta = 8) and larger (FhuA; n beta = 22) monomeric E. coli outer membrane proteins, the stoichiometry of motionally restricted lipids increases linearly with the number of beta-strands, the tilt (beta approximately 44 degrees ) of the beta-strands is comparable for the three proteins, and the order parameter of the beta-barrel increases regularly with n beta. These systematic features of the integration of monomeric beta-barrel proteins in lipid membranes could be useful for characterizing outer membrane proteins of unknown structure.

Augmented transgene expression in transformed cells using a parvoviral hybrid vector.

Autonomous parvoviruses possess an intrinsic oncotropism based on viral genetic elements controlling gene expression and genome replication. We constructed a hybrid vector consisting of the H1 parvovirus-derived expression cassette comprising the p4 promoter, the ns1 gene and the p38 promoter flanked by the adeno-associated viruses 2 (AAV2) inverted terminal repeats and packaged into AAV2 capsids. Gene transduction using this vector could be stimulated by coinfection with adenovirus, by irradiation or treatment with genotoxic agents, similar to standard AAV2 vectors. However, the latter were in most cases less efficient in gene transduction than the hybrid vector. With the new vector, tumor cell-selective increase in transgene expression was observed in pairs of transformed and non-transformed cells, leading to selective killing of the transformed cells after expression of a prodrug-converting enzyme. Preferential gene expression in tumor versus normal liver tissue was also observed in vivo in a syngeneic rat model. Comparative transduction of a panel of different tumor cell lines with the H1 and the H1/AAV hybrid vector showed a preference of each vector for distinct cell types, probably reflecting the dependence of the viral tropism on capsid determinants.

Cardio-specific long-term gene expression in a porcine model after selective pressure-regulated retroinfusion of adeno-associated viral (AAV) vectors.

Cornerstone for an efficient cardiac gene therapy is the need for a vector system, which enables selective and long-term expression of the gene of interest. In rodent animal models adeno-associated viral (AAV) vectors like AAV-6 have been shown to efficiently transduce cardiomyocytes. However, since significant species-dependent differences in transduction characteristics exist, large animal models are of imminent need for preclinical evaluations. We compared gene transfer efficiencies of AAV-6 and heparin binding site-deleted AAV-2 vectors in a porcine model. Application of the AAVs was performed by pressure-regulated retroinfusion of the anterior interventricular cardiac vein, which has been previously shown to efficiently deliver genes to the myocardium (3.5 x 10(10) viral genomes per animal; n=5 animals per group). All vectors harbored a luciferase reporter gene under control of a cytomegalovirus (CMV)-enhanced 1.5 kb rat myosin light chain promoter (CMV-MLC2v). Expression levels were evaluated 4 weeks after gene transfer by determining luciferase activities. To rule out a systemic spillover peripheral tissue was analyzed by PCR for the presence of vector genomes. Selective retroinfusion of AAV serotype 6 vectors into the anterior cardiac vein substantially increased reporter gene expression in the targeted distal left anterior descending (LAD) territory (65 943+/-31 122 vs control territory 294+/-69, P<0.05). Retroinfusion of AAV-2 vectors showed lower transgene expression, which could be increased with coadministration of recombinant human vascular endothelial growth factor (1365+/-707 no vascular endothelial growth factor (VEGF) vs 38 760+/-2448 with VEGF, P<0.05). Significant transgene expression was not detected in other organs than the heart, although vector genomes were detected also in the lung and liver. Thus, selective retroinfusion of AAV-6 into the coronary vein led to efficient long-term myocardial reporter gene expression in the targeted LAD area of the porcine heart. Coapplication of VEGF significantly increased transduction efficiency of AAV-2.

Genetic modifications of the adeno-associated virus type 2 capsid reduce the affinity and the neutralizing effects of human serum antibodies.

The high prevalence of human serum antibodies against adeno-associated virus type 2 (AAV) vectors represents a potential limitation for in vivo applications. Consequently, the development of AAV vectors able to escape antibody binding and neutralization is of importance. To identify capsid domains which contain major immunogenic epitopes, six AAV capsid mutants carrying peptide insertions in surface exposed loop regions (I-261, I-381, I-447, I-534, I-573, I-587) were analyzed. Two of these mutants, I-534 and I-573, showed an up to 70% reduced affinity for AAV antibodies as compared to wild-type AAV in the majority of serum samples. In addition, AAV mutant I-587 but not wild-type AAV efficiently transduced cells despite the presence of neutralizing antisera. Taken together, the results show that major neutralizing effects of human AAV antisera might be overcome by the use of AAV capsid mutants.

Identification of a heparin-binding motif on adeno-associated virus type 2 capsids.

Infection of cells with adeno-associated virus (AAV) type 2 (AAV-2) is mediated by binding to heparan sulfate proteoglycan and can be competed by heparin. Mutational analysis of AAV-2 capsid proteins showed that a group of basic amino acids (arginines 484, 487, 585, and 588 and lysine 532) contribute to heparin and HeLa cell binding. These amino acids are positioned in three clusters at the threefold spike region of the AAV-2 capsid. According to the recently resolved atomic structure for AAV-2, arginines 484 and 487 and lysine 532 on one site and arginines 585 and 588 on the other site belong to different capsid protein subunits. These data suggest that the formation of the heparin-binding motifs depends on the correct assembly of VP trimers or even of capsids. In contrast, arginine 475, which also strongly reduces heparin binding as well as viral infectivity upon mutation to alanine, is located inside the capsid structure at the border of adjacent VP subunits and most likely influences heparin binding indirectly by disturbing correct subunit assembly. Computer simulation of heparin docking to the AAV-2 capsid suggests that heparin associates with the three basic clusters along a channel-like cavity flanked by the basic amino acids. With few exceptions, mutant infectivities correlated with their heparin- and cell-binding properties. The tissue distribution in mice of recombinant AAV-2 mutated in R484 and R585 indicated markedly reduced infection of the liver, compared to infection with wild-type recombinant AAV, but continued infection of the heart. These results suggest that although heparin binding influences the infectivity of AAV-2, it seems not to be necessary.

Membrane protein folding on the example of outer membrane protein A of Escherichia coli.

The biophysical principles and mechanisms by which membrane proteins insert and fold into a biomembrane have mostly been studied with bacteriorhodopsin and outer membrane protein A (OmpA). This review de-scribes the assembly process of the monomeric outer membrane proteins of Gram-negative bacteria, for which OmpA has served as an example. OmpA is a two-domain outer membrane protein composed of a 171-residue eight-stranded beta-barrel transmembrane domain and a 154-residue periplasmic domain. OmpA is translocated in an unstructured form across the cytoplasmic membrane into the periplasm. In the periplasm, unfolded OmpA is kept in solution in complex with the molecular chaperone Skp. After binding of periplasmic lipopolysaccharide, OmpA insertion and folding occur spontaneously upon interaction of the complex with the phospholipid bilayer. Insertion and folding of the beta-barrel transmembrane domain into the lipid bilayer are highly synchronized, i.e. the formation of large amounts of beta-sheet secondary structure and beta-barrel tertiary structure take place in parallel with the same rate constants, while OmpA inserts into the hydrophobic core of the membrane. In vitro, OmpA can successfully fold into a range of model membranes of very different phospholipid compositions, i. e. into bilayers of lipids of different headgroup structures and hydrophobic chain lengths. Three membrane-bound folding intermediates of OmpA were discovered in folding studies with dioleoylphosphatidylcholine bilayers. Their formation was monitored by time-resolved distance determinations by fluorescence quenching, and they were structurally distinguished by the relative positions of the five tryptophan residues of OmpA in projection to the membrane normal. Recent studies indicate a chaperone-assisted, highly synchronized mechanism of secondary and tertiary structure formation upon membrane insertion of beta-barrel membrane proteins such as OmpA that involves at least three structurally distinct folding intermediates.

Radon progeny activity on skin and hair after speleotherapeutic radon exposure.

The objective of present investigation was to measure radon progeny activity on hair and skin (forehead, paraumbilical, paravertebral) and its decrease after speleotherapeutic radon exposure in the gallery of the Gasteiner Heilstollen. Radon progeny activity was measured by means of a halogen-quenched Geiger-Mueller tube with a mica window (density 1.5-2.0 mg/cm2) and an effective diameter of 45 mm; beta efficiency 32% (210Bi), alpha efficiency 18% (241Am). Results are in counts per minute (cpm). All 17 patients were being treated for rheumatic disease in the galleries of the Gasteiner Heilstollen Hospital. The following activity (mean +/- standard error of the mean) was measured 25 (+/- 5) min after leaving the treatment area: on hair of the head, 1235+/-141 cpm; forehead, 503+/-78 cpm; paraumbilical 460+/-85 cpm; paravertebral, 270+/-39 cpm. Taking a shower did not significantly reduce radon progeny activity. Speleotherapeutic radon exposure causes a considerable increase in radon progeny activity on skin. The large surface of hair causes much greater activity on hair than on skin. Owing to their high adhesive properties, radon progenies are not reduced by taking a shower.

[Genetic differentiation of pedunculate oak Quercus robur L. in the European part of Russia based on RAPD markers]. / Geneticheskaia differentsiatsiia duba chereshchatogo (Quercus robur L.) v Evropeiskoi chasti Rossii na osnove RAPD-markerov.

Using randomly amplified polymorphic DNA markers (RAPD), genetic variation and differentiation in four populations of pedunculate oak Quercus robur L. were examined. The populations occupy a large part of the Quercus robur range in the European Russia (Voronezh and Novgorod oblasts; Republics of Mordovia and Bashkortostan). With each of six random primers (A02, A09, A17, B01, B08, B11), 96 DNA samples were analyzed by PCR. In all, 48 putative polymorphic RAPD loci were detected. We failed to reveal population-specific DNA fragments for any primer although the frequencies of 14 fragments were significantly different among populations. The oak populations studied exhibited high variability: 73-90% of genes were polymorphic and the effective allele number was about 1.4. The total genetic variation varied from 0.202 (Vor) to 0.245 (Nov), which corresponded to the estimates for populations of this species from Central and Western Europe. The populations examined showed low among-population differentiation (GST = 0.098); gene flow Nem was 4.61. The proportion of among-population variation of the RAPD loci studied accounted for 7% of the total variability; more than 93% of the total variability was explained by individual and within--population variation.

Interaction of membrane-spanning proteins with peripheral and lipid-anchored membrane proteins: perspectives from protein-lipid interactions (Review).

Studies of lipid-protein interactions in double-reconstituted systems involving both integral and peripheral or lipid-anchored proteins are reviewed. Membranes of dimyristoyl phosphatidylglycerol containing either myelin proteolipid protein or cytochrome c oxidase were studied. The partner peripheral proteins bound to these membranes were myelin basic protein or cytochrome c, respectively. In addition, the interactions between the myelin proteolipid protein and avidin that was membrane-anchored by binding to N-biotinyl phosphatidylethanolamine were studied in dimyristoyl phosphatidylcholine membranes. Steric exclusion plays a significant role when sizes of the peripheral protein and transmembrane domain of the integral protein are comparable. Even so, the effects on avidin-linked lipids are different from those induced by myelin basic protein on freely diffusible lipids, both interacting with the myelin proteolipid protein. Both the former and the cytochrome c/cytochrome oxidase couple evidence a propagation of lipid perturbation out from the intramembrane protein interface that could be a basis for formation of microdomains.

Electron cryo-microscopy and image reconstruction of adeno-associated virus type 2 empty capsids.

Adeno-associated virus type 2 empty capsids are composed of three proteins, VP1, VP2 and VP3, which have relative molecular masses of 87, 72 and 62 kDa, respectively, and differ in their N-terminal amino acid sequences. They have a likely molar ratio of 1:1:8 and occupy symmetrical equivalent positions in an icosahedrally arranged protein shell. We have investigated empty capsids of adeno-associated virus type 2 by electron cryo-microscopy and icosahedral image reconstruction. The three-dimensional map at 1.05 nm resolution showed sets of three elongated spikes surrounding the three-fold symmetry axes and narrow empty channels at the five-fold axes. The inside of the capsid superimposed with the previously determined structure of the canine parvovirus (Q. Xie and M.S. Chapman, 1996, J. Mol. Biol., 264, 497-520), whereas the outer surface showed clear discrepancies. Globular structures at the inner surface of the capsid at the two-fold symmetry axes were identified as possible positions for the N-terminal extensions of VP1 and VP2.