Structure of the reovirus membrane-penetration protein, Mu1, in a complex with is protector protein, Sigma3

Abstract

Cell entry by nonenveloped animal viruses requires membrane penetration without membrane fusion. The reovirus penetration agent is the outer-capsid protein, Mu1. The structure of Mu1, complexed with its "protector" protein, Sigma3, and the fit of this Mu1(3)Sigma3(3) heterohexameric complex into the cryoEM image of an intact virion, reveal molecular events essential for viral penetration. Autolytic cleavage divides Mu1 into myristoylated Mu1N and Mu1C. A long hydrophobic pocket can receive the myristoyl group. Dissociation of Mu1N, linked to a major conformational change of the entire Mu1 trimer, must precede myristoyl-group insertion into the cellular membrane. A myristoyl switch, coupling exposure of the fatty acid chain, autolytic cleavage of Mu1N, and long-range molecular rearrangement of Mu1C, thus appears to be part of the penetration mechanism.

Figure 1. Structure of the μ1₃σ3₃ Heterohexamer

(A) Surface of a reovirus particle (Dryden et al., 1993), with an example of each of the four exposed proteins labeled. The yellow triangle outlines one μ1₃σ3₃ complex. (B) A μ1₃σ3₃ complex, viewed from the side. The bottom of the complex would contact σ2 proteins on the surface of the reovirus core. Three σ3′s (41 kDa; shown in green, cyan, and magenta) bind to the upper part of the underlying trimer of μ1 trimer (76 kDa; shown in red, blue, and yellow). The three μ1 chains are wound tightly around each other. Each σ3 contacts two μ1’s. (C) View from the top after a 90° rotation of the model in (B) around a horizontal axis within the plane of the paper. (D) View from the bottom. Insets: Molecular surfaces of the μ1₃σ3₃ complex showing μ1 and σ3 in the same color code.

Figure 2. The μ1 Protein

(A) The μ1 trimer shown without bound σ3. One μ1 subunit is colored by domain (domain I, light and dark blue [μ1N, μ1C]; domain II, light and dark green [μ1N, μ1C]; domain III, red; domain IV, yellow); the two other μ1 subunits are in gray. The β-octyl glucoside in the colored subunit is in yellow; the sulfate ion in green and red. The σ3 binding site is indicated by an arrow. (B) Ribbon diagram of an isolated μ1 subunit (coloring as in [A]). (C) Domain segmentation of the amino acid sequence as determined from the three-dimensional structure. Domain color code as in (A). The central domain II contains domains I and III as “inserts,” and domain III similarly contains domain IV. (D) Amino acid sequence alignment of reovirus μ1 proteins. One representative of each of the three major human serotypes is shown: μ1 T1L (Chandran et al., 1999), μ1 T2J (Wiener and Joklik, 1988), and μ1 T3D (Jayasuriya et al., 1988). Also shown is the sequence of VP5 of grass carp hemorrhagic virus (GenBank accession number AAG17823), an aquareovirus outer capsid protein with 29% sequence identity to μ1 T1L. Amino acid numbering as in the μ1 T1L sequence. Secondary structure elements, derived from the coordinates using the program DSSP (Kabsch and Sander, 1983), are presented as tubes for α helices and arrows for β strands. The segments of the four domains are shown in the same color code as in (A). The arrowhead just before p2 shows the position of the autolytic cleavage. Dots indicate amino acid identities; dashes show gaps.

Figure 3. Fold of the μ1 Monomer

Left: Folding diagram of the μ1 polypeptide chain. α helices are shown as rectangles, β strands as arrows. Right: Ribbon representations of the four individual domains. The view for domains II and III is the same as for the colored μ1 monomer in Figure 2A; the view for domains I and IV is the same as for the yellow μ1 monomer in Figure 1B. The color code for the four domains is the same as in Figure 2. The β-octyl glucoside molecule bound to domain II is in yellow; the sulfate ion bound to domain IV, in green and red. The arrow shows the position of autolytic cleavage in domain I.

Figure 4. Autolytic Cleavage Site in μ1

(A) μ1 trimer viewed from the bottom. The three μ1N peptides are in light blue, yellow, and red; all three μ1C are in gray. The three β-octyl glucoside molecules are in green. The first ordered residue at the N terminus, Thr10, is indicated with “N.” (B) Close up of the three μ1N peptides with the adjacent helices A of the three μ1C chains. For the blue μ1 molecule, μ1N is shown in light blue, μ1C in dark blue, side chains of Asn42 and Pro43 in green, and side chains of Thr104, Glu105, His106, and Lys113 in close vicinity to the cleavage site in yellow. (C) The cleaved μ1 polypeptide chain as seen from the side. The intermolecular salt bridge Glu105/Lys113 is indicated.

Figure 5. Characteristic Features of the μ1 Molecule

(A) Putative N-myristoyl binding site in domain II framed by a helix-turn-helix motif and a short antiparallel β sheet. The β-octyl glucoside molecule is shown in light gray; hydrophobic side chains contributing to the pocket, in yellow. (B) Sulfate binding site in domain IV. The sulfate, bound on the upper surface of the μ1 trimer, is coordinated by five main chain amide groups from a β-hairpin insertion into the jelly-roll fold. The sulfate ion is in green; the β-hairpin, in red; the jelly-roll β strands, in yellow.

Figure 6. Fit of Atomic Structures into a Cryo-EM Image of the T1L Virion

(A) View from outside the particle. The contours of the EM map of the virion are in blue, with contours for the core also shown in red. Six μ1₃σ3₃ complexes were placed in the virion map as shown by the colored models (see Experimental Procedures). The jelly-roll domains IV of μ1 subunits in neighboring trimers contact each other through loops between strands 13 and 14, 15 and 16, and 19 and 20; the central domains II make contact through helices K and through the loops between helices C and D, and E and F; the basal domains I make contact through helices B and through the loops between helix A′ and strand 5. The isolated blob (arrow) in the center of the ring of six μ1 trimers probably represents a clustering of the C-terminal arms from the six neighboring μ1 subunits. The loops of μ1 containing residues 51–62 and (probably) the disordered loop containing residues 72–96 mediate variable contacts with the 150 σ2 subunits on the underlying surface of the core. (B) Cross-sectional view. A pentameric λ2 turret is on the left; the σ2 monomers, bound on the surface of the λ1 shell, support the μ1 lattice. (C) A synthetic, quasi-atomic model of the reovirus particle. The outer shell of 200 μ1₃σ3₃ heterohexamers is represented on the upper left, with μ1 in green and σ3 in white; the core (Reinisch et al., 2000) is on the upper right, with λ1 in red, λ2 in blue, and σ2 in pink and yellow; the combination of the two models, which includes all icosahedrally ordered proteins in the virion (a total of 100 MDa), is at the bottom.

Figure 7. Schematic Model of μ1 Transformations in Reovirus Entry

Proteolytic removal of σ3 (left) prepares the μ1 trimer to undergo a conformational change in the endosome (right). Unfolding of μ1C is suggested by uncoiling of the schematic trimer; the actual rearrangement may be considerably more extensive. Release of μ1N allows the myristoyl group to insert into the membrane. The stage at which autolytic cleavage separates μ1N from μ1C is not yet known.