Flavivirus E protein

The atomic structures of the ectodomain (residues 1-395) of TBEV, DENV-2 and DENV-3 E proteins have been solved using X-ray crystallographic methods and are structurally very similar (Fig. 1) (Rey et al 1995, Modis et al 2003, 2005). The E protein ectodomain, minus the C-terminal —100 residues consisting of the stem-anchor regions, displays three domains—domain I, the central domain that contains the N-terminus of the protein; domain II, the extended, finger-like 'dimerization' domain that makes most of the intra-dimeric contacts; and domain III, the immunoglobulin-like domain that is thought to be involved in receptor binding. Domains I and II are connected by four polypeptide chains, whereas

Domain m fusion peptide

Domain 1 Domain II

FIG. 1. Structure of the DENV-2 E protein ectodomain, residues 1—395 as determined using X-ray crystallography. The figure shows a monomer of the E protein in the neutral pH conformation. Domains I, II and III are as denoted. The fusion peptide at the distal tip of domain II is also labelled.

Domain m

Protein Structure Dengue

FIG. 1. Structure of the DENV-2 E protein ectodomain, residues 1—395 as determined using X-ray crystallography. The figure shows a monomer of the E protein in the neutral pH conformation. Domains I, II and III are as denoted. The fusion peptide at the distal tip of domain II is also labelled.

domains I and III are connected by a single polypeptide chain. At the distal tip of domain II is a small hydrophobic sequence, the so called fusion peptide, that initiates the fusion process (Allison et al 2001). The E protein structure was found to be remarkably similar to the structure of the E1 protein of Semliki Forest virus (SFV), an alphavirus belonging to the Togaviridae family (Lescar et al 2001). Unlike flaviviruses, the alphaviruses separate the functions of receptor binding and membrane fusion into two transmembrane glycoproteins termed E2 and E1, respectively. This distinctive three-domain arrangement of the E protein, shared by fusion proteins from these two viral families, is conserved because of a fundamental structural requirement of the Class II membrane fusion process (Kuhn et al 2002, Lescar et al 2001).

Recently, the structures of native DENV-2 and WNV virus were solved using cryo-electron microscopy (cryo-EM) and image reconstruction methods (Kuhn et al 2002, Mukhopadhyay et al 2003). The atomic structure of the E protein determined by X-ray crystallography can be computationally docked into the cryo-EM density to obtain a 'pseudo-atomic' structure of the virus. Fitting the atomic structure of the E protein into the DENV EM density revealed that there were 90 E dimers lying flat on the viral surface (Fig. 2). Surprisingly, they were not organized into the predicted conventional T = 3 icosahedral arrangement. Instead the E proteins can be grouped in sets of three, nearly parallel dimers with these sets arranged to form a 'herringbone' pattern on the viral surface (Kuhn et al 2002). There is a 21° angular difference between domains I and II in the crystal structure of the E protein compared with its position in the fitted structure of the virus particle

Virus Denv
FIG. 2. (A) Surface-shaded view of the cryo-EM reconstruction of mature DENV-2 at 14 Ä resolution. (B) The fit of the E protein structure into the cryo-EM density of the mature virus. One raft, consisting of three parallel dimers, is highlighted. The E protein is shaded as in Fig. 1.

(Mukhopadhyay et al 2005, Zhang et al 2005). It has been suggested that this arrangement of E dimers on the viral surface positions the protein in a metastable state that can be released and spring into a fusogenic trimer during entry (Stiasny et al 2001). During the low pH-induced fusion between the viral and cell membranes in the endosome, the E protein undergoes extensive conformational changes and the E homodimers dissociate into monomers and then re-associate to form fusion-competent homotrimers such that their fusion peptides are exposed at the tip of the trimer. The structure of the trimer of E proteins in the putative postfusion state was solved using X-ray crystallography and shows domain III folded back by about 30 A toward domain II and rotated —20°. The domain II of each of the E proteins in the trimer lies parallel to each other with their fusion peptides exposed at the tip of the trimer (Modis et al 2004). The fusion peptides presumably penetrate the endosomal membrane and initiate the fusion process.

The structures of the immature forms of DENV-2 and YFV have also been solved using cryo-EM methods (Zhang et al 2003). The immature particles are dramatically different than the mature native virus. These immature virions are bigger (— 60 nm) and present spike-like projections on the surface that are formed by trimers of prM-E heterodimers. Similar to the mature virion, these immature particles lack classical T = 3 symmetry. Fitting of the E protein atomic structure into the cryo-EM density revealed that the hinge angle between domains I and II is 6° different from the crystal structure and 27° different than the E fitted in the mature virus (Fig. 3) (Mukhopadhyay et al 2005, Zhang et al 2005). The three prM proteins, which were identified by subtracting the density of E proteins from the

Protein Structure Dengue

FIG. 3. (A) Surface-shaded view of the cryo-EM reconstruction of immature DENV-2 at 16 A resolution. (B) Arrangement of the E protein structure in the immature virus particle obtained by fitting the E protein structure into the EM density for the immature particle. The E protein is shaded as in Fig. 1.

FIG. 3. (A) Surface-shaded view of the cryo-EM reconstruction of immature DENV-2 at 16 A resolution. (B) Arrangement of the E protein structure in the immature virus particle obtained by fitting the E protein structure into the EM density for the immature particle. The E protein is shaded as in Fig. 1.

total immature density, cover the fusion peptide of the three E proteins in the spike in order to prevent premature fusion. During late maturation steps in the Golgi, prM is cleaved by furin leading to a rearrangement of the E proteins from prM-E heterodimers that are projecting from the viral surface to E homodim-ers that lie flat on the viral surface. Thus, the E protein undergoes dramatic con-formational changes during assembly of the virus particle and during fusion. During this conversion the E protein undergoes a —30° shift in the hinge region between domains I and II.

The DENV-2 E protein structure solved at Harvard University by Modis and Harrison was determined both with and without the detergent, n-octyl-P-D-glu-coside (BOG), bound to the E protein (Modis et al 2003). This molecule was found in a hydrophobic pocket that lies in the region between domains I and II that acts as a 'hinge' to promote domain flexing. The E protein has been shown to undergo significant conformational changes in this region during membrane fusion as well as in maturation of the virus particle. Mutations in various flavivi-ruses that affect virulence and alter the pH threshold required for fusion were mapped to this region between domains I and II (Rey et al 1995). The molecular mechanisms that lead to membrane fusion are not yet completely understood, but the initial events must involve pH-mediated conformational changes resulting in dimer dissociation. The BOG binding may have stabilized the E protein dimer in a certain conformation making the dimer amenable for crystal formation. The DENV-2 E protein structure solved independently at Purdue University by Rossmann and colleagues was similar to the structure solved by Modis and Harrison except that there was a hinge motion between domain I and II of —10° and lacked the presence of BOG even though it was present in the crystallization buffer (Zhang et al 2005). These data indicate that the flexible hinge region is involved in the numerous conformational changes necessary for membrane fusion and during transition from the immature to the mature viral form. This hinge region thus presents an appealing binding site for compounds that could exhibit anti-viral activity by interfering with the various transitions that the E protein undergoes. Furthermore, sequence analysis of the E proteins from flaviviruses reveals —40% amino acid identity and these proteins are predicted to be very similar in their overall fold and domain arrangement including the hinge region (Burke & Monath 2001). Thus, there is the possibility for the development of a compound that acts as an inhibitor against a broad spectrum of pathogenic flaviviruses.

Was this article helpful?

+2 0
Swine Influenza

Swine Influenza

SWINE INFLUENZA frightening you? CONCERNED about the health implications? Coughs and Sneezes Spread Diseases! Stop The Swine Flu from Spreading. Follow the advice to keep your family and friends safe from this virus and not become another victim. These simple cost free guidelines will help you to protect yourself from the swine flu.

Get My Free Ebook


Responses

  • sophie
    What is flavivirus eprotein?
    4 months ago

Post a comment