Fig 1.
Thermostability and general organization of the FAdV-C4 capsid.
(a) Effect of heating on HAdV-C5 and FAdV-C4 virions analysed by extrinsic fluorescence. Circles represent the normalized fluorescence ± STD for HAdV-C5 and FAdV-C4. Continuous lines correspond to the Boltzmann sigmoid fitting to estimate the T0.5 for each condition tested. (b) Radially coloured surface rendering of the 3.3 Å resolution FAdV-C4 map viewed down the 2-fold icosahedral axis. One facet is identified by a white triangle. Icosahedral symmetry axes are indicated with white symbols: 5-fold (pentagon), 3-fold (triangle) and 2-fold (oval). (c) Molecular models of the major and minor coat proteins traced in the icosahedral FAdV-C4 facet. Top: view from outside the capsid. Bottom: view from inside the capsid. Protein colours as indicated by the legend at the left-hand side. The four hexon trimers in the AU are numbered H1-H4.
Fig 2.
Hexon and penton structures in FAdV-C4.
(a) Superposition of the HAdV-C5 (PDB ID 6b1t, in white) and FAdV-C4 (coloured by RMSD) hexon monomers. P1 and P2 indicate the two jelly rolls. The horizontal dashed line indicates the equatorial region. (b) Structural superposition of available hexon structures (PDB IDs: 6cgv: HAdV-C5, 5tx1: HAdV-D26, 6yba: HAdV-F41, 2obe: SAdV-25, 3zif: BAdV-3, and 6qi5: LAdV-2). Conserved residues (with < 5Å RMSD compared to the rest of structures) are shown in dark blue. Residues with RMSD higher than 5 Å are represented in white for all hexons, except for the > 5 Å RMSD FAdV-C4 residues, which are represented in red, to be distinguished from the rest. Conserved residues are mainly forming the jelly rolls in the base. Variable regions are mostly located at the top of the hexons, conforming the specific HVRs. Note the unique big insertion of FAdV-C4 (zoom at the right), not present in any of the other hexon structures. (c) Superposition of the HAdV-C5 (PDB ID 6b1t, white) and FAdV-C4 (coloured by RMSD) penton base monomers. The hypervariable and variable loops (HVL, VL and VL’) along with other diff regions are indicated. (d) Top view of the penton base pentamers of HAdV-C5 and FAdV-C4, shown with one monomer rainbow-colored and with the HVL, VL and VL’ indicated, and the rest of monomers in grey tones.
Fig 3.
Protein VIII divergent structure in FAdV-C4.
(a) Sequence alignment and secondary structure elements of protein VIII in HAdV-C5 (top) and FAdV-C4 (bottom). Squiggles indicate α-helices and arrows indicate β-strands. Sequence numbering corresponds to HAdV-C5. Residues are coloured according to similarity: black (low similarity), red with blue frame (high similarity) and white on red background (strict identity). N- and C-terminal fragments are delimited by the AVP-processed peptides shadowed in grey. An arrowhead below the FAdV-C4 sequence indicates where the C-terminal fragment tracing starts, seven amino acids downstream of the AVP site. Note the divergence in α2/α2’ and that FAdV-C4 α2” and α2”’ are displaced with respect to β3 and β4 in HAdV-C5. (b) Superposition of protein VIII in HAdV-C5 (white, PDB ID: 6b1t) and FAdV-C4 (coloured by RMSD). (c) Rainbow coloured structures of protein VIII in HAdV-C5 (left) and FAdV-C4 (right). The body, neck (dashed box) and head domains are indicated. Dashed gaps corresponding to the AVP cleaved segment (not traced) are indicated with scissors. Note that α2 in FAdV-C4 is shorter than α2 in HAdV-C5, and along with α2’ follows a different path compared to α2 and subsequent coil structure in HAdV-C5 (curved red arrows). β3 and β4 in HAdV-C5 differ from α2” and α2”’ in FAdV-C4 (dashed frame). Helix α2’ in FAdV-C4 is bent by 115 º with respect to α2. (d) Detail of the HAdV-C5 (white) and FAdV-C4 (orange) C-terminal fragment of protein VIII.
Fig 4.
FAdV-C4 minor coat protein IIIa.
(a) Superposition of protein IIIa in HAdV-C5 (PDB ID: 6b1t, white) and FAdV-C4 (coloured by RMSD), presented in their native position in the capsid. The view is from inside the capsid, with the 5-fold symmetry axis at the top of the page. The GOS-glue, connecting helix, VIII-binding and core proximal domains are indicated. (b) As in (a), but comparing with LAdV-2 protein IIIa in grey (PDB ID: 6qi5). The GOS-glue domains and two thirds of the connecting helices mostly overlap in the three structures (FAdV-C4, HAdV-C5 and LAdV-2). The N-terminal region in FAdV-C4 presents fewer ordered amino acids (filled arrowheads), lacking a ~ 12 residue stretch ordered in HAdV-C5 and LAdV-2 (hollow arrowheads). The VIII-binding domain is significantly rotated with respect to its position in HAdV-C5 and slightly rotated with respect to LAdV-2. (c) Centre: superposition of the three protein IIIa structures: FAdV-C4 (orange), HAdV-C5 (white) and LAdV-2 (grey), represented in their native position in the capsid. Left: comparison of the three connecting helices. Horizontal black lines indicate the region where the three helices overlap; dashed lines indicate their bending away from each other (diff 2). The dashed black rectangles show the VIII-binding domains in the three viruses, with the small handle formed by diff3 labelled as a landmark. A cylinder indicates the first helix of the VIII-binding domain in FAdV-C4. (d) Focus on the connecting helix and the first helix in the VIII-binding domain showing how their relative orientations vary between the three viruses. For convenience a cartesian coordinate system is represented with the connecting helix of FAdV-C4 along the Y axis and the first helix of the VIII-binding domain of HAdV-C5 along the X axis. The origin of coordinates is located at the last amino acid of the FAdV-C4 connecting helix.
Fig 5.
Additional internal components of the FAdV-C4 capsid.
(a) FAdV-C4 map, as viewed from inside of the capsid. Left: only the density corresponding to the traced proteins is shown. The two-fold (oval), three-fold (triangle) and five-fold (pentagon) symmetry axes, the four hexons in one AU (H1-H4), and the facet (dashed triangle) are indicated for reference. Right: remnant densities (RD) are coloured in red. (b) Remnant map components colour-coded as RD1-4. Right: a zoom showing only the colour-coded RDs. RDs considered too small to be analysed are left in grey. Densities correspond to the unsharpened map contoured at 1.5 σ. Hexagon contours indicate hexon positions.
Fig 6.
Differences in interactions established by protein IIIa in HAdV-C5 and FAdV-C4.
(a) Protein-protein interfaces between IIIa (yellow surface) and its neighbouring copy IIIa’ (symmetry copy coloured in white) for HAdV-C5 (left) and FAdV-C4 (right). (b) Interfaces between IIIa (yellow surface) and VIII (orange surface). (c) Protein-protein interfaces between IIIa (yellow surface) and neighbouring hexons (blue surface). In each panel, the interacting molecules are shown at the top as seen from inside the vertex region (indicated with a black pentagon), and one of them is separated and rotated as indicated at the bottom, to reveal the interfacing residues (coloured in red). Dashed arrows indicate that the molecule orientation is maintained between the top and bottom panels. The number of contacting residues is indicated in red inside circles.
Fig 7.
Interfaces between protein VIII and hexons in HAdV-C5 and FAdV-C4.
(a) Protein- protein interfaces between VIII (orange surface) and hexons (blue surface), with contacting residues highlighted in red, and their number indicated as in Fig 6. The N-termini of HAdV-C5 hexons are shorter than their FAdV-C4 counterparts (ovals). (b) Zoom in on protein VIII. The traced N-termini of the HAdV-C5 hexons (blue) are not interlaced with protein VIII (orange) (left, ovals) whereas the N-termini of the FAdV-C4 hexons (most notably H2) are clasping protein VIII (right, ovals). (c) Ribbon representation of the molecules represented as surfaces in (a) and (b), with the non-traced first residues of the hexons indicated by dashed lines. Blurry curved arrows represent the flexibility of the hexon N-termini. Scissors indicate the AVP sites. Dashed frames are drawn to facilitate comparisons. Hexon chain IDs (according to S4 Table) are indicated in parenthesis.
Fig 8.
Organization of the core varies among adenovirus genera.
(a) Multiple protein sequence alignment of core proteins pVII (top) and pµ (bottom), showing their respective length (in amino acids) encircled. NCBI GenBank accession numbers for the genomes chosen as representative members of each genus are as follows: MastAdV, HAdV-C5 (AC_000008); AviAdV, FAdV-A1 (AC_000014); SiAdV, FrogAdV-1 (NC_002501); AtAdV, OAdV-7 (NC_004037); IchtAdV, WSAdV (MK_101347). Predicted AVP cleavage sites are indicated by dashed vertical lines following the consensus patterns for proteins pVII and pµ. Amino acids are coloured by polarity, and the top histograms indicate the mean pairwise identity over all pairs in the column. Notice that the number of cleavage sites also varies between genera, but the sites delimiting the conserved domains of protein VII and µ are also conserved. Blue rectangles highlight high similarity regions corresponding to the pVII N-terminal peptide (pVIIn1) and C-terminal region of pµ. (b) Central section of a HAdV (as an example we use HAdV-F41, EMDB code:10768, left) and FAdV-C4 (right) cryo-EM maps. Arrows point to more pronounced concentric shells in FAdV-C4. (c) Radial average profiles of FAdV-C4 and HAdV-F41 maps. The arrows indicate the 25 Å spacing between shells.