Principles of Virology. Jane Flint

only VP1. (Right) An enlarged view of the density below a single VP1 penton, indicating the thickness of the radial layers of DNA and the spacing between them. A model of double-stranded DNA as it appears when wrapped on a human histone (blue) is superimposed. Adapted from Hurdiss DL et al. 2016. Structure 24:528–536, licensed under CC BY 4.0. Courtesy of N.A. Ranson, University of Leeds, United Kingdom.

Figure 4.21 Conserved organization of the RNA-packaging proteins of nonsegmented (−) strand RNA viruses. Ribbon diagrams of the N proteins indicated are shown at the top, colored from purple at the N terminus to red at the C terminus. Their electrostatic surfaces from negative (red) to positive (blue) are shown in the space-filling models below, with the molecules rotated as indicated to show the RNA-binding cleft (blue) most clearly. Although differing in structural details, these N proteins share a two-lobed structure (top) and an RNA-binding cleft between the two lobes. Adapted from Ruigrok RW et al. 2011. Curr Opin Microbiol 14:504–510, with permission. Courtesy of D. Kolakofsky, University of Geneva.

Viruses with Envelopes

In addition to the capsids described previously, many virus particles possess an envelope formed by a viral protein-containing membrane that is derived from the host cell. However, viral envelopes vary considerably in size, morphology, and complexity, and they differ in lipid composition, the number of proteins they contain, and their location. The membranes form the outermost layer of enveloped animal viruses, but in some bacteriophages and archaeal viruses the membrane lies beneath an icosahedral capsid (Box 4.8). Typical features of viral envelopes and their proteins are described in the next section, to set the stage for consideration of the structures of envelope proteins and the various ways in which they interact with internal components of the virion.

Viral Envelope Components

      The foundation of the envelopes of all animal viruses is a lipid membrane acquired from the host cell during assembly. The lipid composition is variable, because viral envelopes can be derived from different kinds of cellular membranes. Embedded in the membrane are viral proteins, the great majority of which are glycoproteins that carry covalently linked sugar chains, or oligosaccharides. Sugars are almost always added to the proteins posttranslationally, during transport to the cellular membrane at which progeny virus particles assemble. Intra- or interchain disulfide bonds, another common chemical feature of these proteins, are also acquired during transport to assembly sites. These covalent bonds stabilize the tertiary or quaternary structures of viral glycoproteins.

       Envelope Glycoproteins

Viral glycoproteins are integral membrane proteins firmly embedded in the lipid bilayer by a short membrane-spanning domain (Fig. 4.22). The membrane-spanning domains of viral proteins are hydrophobic α-helices of sufficient length to span the lipid bilayer. They generally separate large external domains that are decorated with oligosaccharides from smaller internal segments. The former contain binding sites for cell surface virus receptors, major antigenic determinants, and sequences that mediate fusion of viral with cellular membranes during entry. Internal domains, which make contact with other components of the virus particle, are often essential for virus assembly.

      With few if any exceptions, viral membrane glycoproteins form oligomers, which can comprise multiple copies of a single protein or may contain two or more protein chains. The subunits are held together by noncovalent interactions and disulfide bonds. On the exterior of particles, these oligomers can form surface projections, often called spikes. Because of their critical roles in initiating infection, the structures of many viral glycoproteins have been determined.

      The hemagglutinin (HA) protein of human influenza A virus is a trimer that contains a globular head with a top surface that is projected ~135 Å from the viral membrane by a long stem (Fig. 4.23A). The latter is formed and stabilized by the coiling of α-helices present in each monomer. The membrane-distal globular domain contains the binding site for the host cell receptor. This important functional region is located >100 Å away from the lipid membranes of influenza virus particles. Other viral glycoproteins that mediate cell attachment and entry, such as the E protein of the flavivirus tick-borne encephalitis virus, adopt a quite different orientation (and structure); the external domain of E protein is a flat, elongated dimer that lies on the surface of the viral membrane rather than projecting from it (Fig. 4.23B). Despite their lack of common structural features, the HA and the E proteins are both primed for dramatic conformational change to allow entry of internal virion components into a host cell (Chapter 5).

BOX 4.8

      DISCUSSION

       A viral membrane directly surrounding the genome

      The membranes present in particles of animal viruses are external structures separated from the genome by at least one protein layer. As we have seen, internal protein layers contribute to condensation and organization of the genome via interactions of the nucleic acid with specialized nucleic acid-binding proteins or the internal surfaces of capsids. However, this arrangement is not universal: the particles of some archaeal and bacterial viruses, as well as giant viruses that infect eukaryotes, contain an internal membrane derived from their host cells.

      This property is exemplified by Sulfolobus turreted icosahedral virus, which infects a hyperthermophilic archaeon. This virus has a double-stranded DNA genome, a major capsid protein containing two β-barrel jelly roll domains, and pentons built from dedicated viral proteins. The capsid encases a lipid membrane rather than an internal nucleoprotein core. As shown in panel A of the figure, a large space separates the capsid and the membrane, with contact between the capsid and the membrane limited to the fivefold axes of icosahedral symmetry, where the most internal domain of the penton base protein contacts a viral transmembrane protein. Particles purified from Sulfolobus turreted icosahedral virus-infected cells include forms that lack the capsid and ex-hibit the size and morphology of lipid cores alone. These observations suggest that the membrane, rather than the capsid, is the major determinant of particle stability.

      The internal membrane of the Sulfolobus turreted icosahedral virus is built from membrane-forming lipids synthesized specifically in thermophilic and hyperthermophilic archaea: they comprise long chains (e.g., C₄₀, compared to C₁₆ to C₁₈ typical of mammalian cells) that comprise cyclopentane rings and branched, isoprenoid-like units ether-linked at either end to various polar head groups. Because of the latter property, these lipids can form monolayer membranes, in contrast to the lipid bilayers formed in animal cells (panel B). The ether linkages, cyclopentane rings, and branched acyl chains considerably increase the stability of membranes formed from these specialized lipids, facilitating survival of the organism and protection of the Sulfolobus turreted icosahedral virus genome during transit through the harsh environments (e.g., pH 3 and temperature of 80°C) inhabited by its host.

      The internal membrane of an archaeal virus. (A) Cross section through a near-atomic-resolution reconstruction of Sulfolobus turreted icosahedral virus, showing the unique vertex structures (turrets) and the separation of the capsid shell from the membrane. The internal surface of the membrane (yellow) is in