Cellular Microbiology (2011)

doi:10.1111/j.1462-5822.2011.01653.x

MicroReview

cmi_1653

1..9

Class II enveloped viruses

Marie-Christine Vaney1,2 and Felix A. Rey1,2* 1 Institut Pasteur, Dpartement de Virologie, Unit de Virologie Structurale, F-75724 Paris Cedex 15, France. 2 Centre National de la Recherche Scientique, Unit de Recherche Associe 3015, F-75724 Paris Cedex 15, France. Summary A number of viruses transport their genomic material from cell to cell enclosed within a lipid bilayer that is in turn encased within a symmetric protein shell. This review focuses in a group of RNA viruses that have this type of virions. This group includes several of important human pathogenic viruses, such as the hepatitis C virus, dengue virus, chikungunya virus, rubella virus and the bunyaviruses. The best studied are the aviviruses and the alphaviruses, which have a b-sheet rich class II viral fusion protein used for entry into susceptible cells. We extend here the class II concept to encompass symmetric viruses in which the envelope proteins are derived from a precursor polyprotein containing two transmembrane glycoproteins arranged in tandem. The rst glycoprotein acts as chaperone for the folding of the second one, which carries the membrane fusion function. Since the bunyaviruses, included here, are very similar to the class I arenaviruses in other respects, this analysis highlights the patchwork nature of the various viral functional modules acting at different stages of the virus cycle, which appear assembled from genes of different origins. Introduction Viruses perpetuate themselves in nature by using the cellular metabolic and translational machineries for their own replication. They have thus evolved elaborate systems to transport their genome from cell to cell, and from organism to organism. The vehicle used for their transmission is called the virion, which is an assembly of
Received 20 June, 2011; revised 8 July, 2011; accepted 13 July, 2011. *For correspondence. E-mail felix.rey@pasteur.fr; Tel. (+33) 145 688 563; Fax (+33) 145 688 993. 2011 Blackwell Publishing Ltd

proteins and in many cases also lipids that maintains the viral nucleic acid protected from the environment during the extracellular journey. This vehicle is at the same time capable of recognizing a susceptible cell often by binding to a specic receptor into which its nucleic acid cargo can be delivered. The virion is thus programmed to react to specic signals encountered when exposed to a susceptible cell, initiating the infection by translocating the genetic material of the virus across the lipid bilayer into the cytoplasm. The translocation machinery used in this process depends on the type of hosts that the virus has evolved to infect, whether they are bacteria, archaea, protozoa, fungi, animals or plants. Animal viruses The virions of animal viruses, which include viruses causing disease in humans, rst bind to a receptor at the cell surface, and this initial interaction triggers a set of cellular signalling pathways that end in viral uptake by the cell. Uptake can lead to translocation of the viral nucleic acid directly at the plasma membrane or, more commonly, within an internal compartment of the cell (Mercer et al., 2010). The mechanism of genome translocation depends on whether the virion contains a lipid bilayer or not. In the rst case, the interactions with the host trigger a membrane fusion reaction to merge viral and cellular lipid bilayers, resulting in the release of the viral nucleic acid within the cell cytoplasm (Harrison, 2008). For viruses in which the virion lacks a lipid bilayer, this process has to take place by other means, requiring a local disruption of the cellular membrane, in a process that is not well understood (Banerjee and Johnson, 2008). Enveloped animal viruses The term enveloped viruses designates those whose virions have a lipid bilayer, or envelope, in which are anchored one or several different virus-encoded glycoproteins. These proteins are exposed at the virion surface and are responsible for the interactions with target cells that lead to entry. They are also the main targets of neutralizing antibodies produced by the host as part of its defence. The type of organization of the envelope proteins at the virion surface provides a further

cellular microbiology

M.-C. Vaney and F. A. Rey

Fig. 1. Similar organization of the genes coding for the envelope proteins in class II viruses. Each line indicates a separate genus within each of the three viral families represented, with a box indicating the polyprotein precursor ORF. The individual viral proteins, generated by proteolytic cleavage of the precursor by signalase and other host or viral proteases, are indicated in different colours and labelled. Non-structural proteins (NSP) present in the same polyprotein are shown in grey, and small integral membrane proteins in black. The fusion protein is coloured light blue, and the companion chaperone protein pale green. Full and empty arrowheads denote the location of the maturation cleavage site of the chaperone and the fusion proteins respectively. The diagrams are to scale (scale bar at bottom right) and are lined up to match at the N-terminus of the membrane fusion protein (arrows). The double wiggle symbol at the 3 end of the ORF of members of the Flaviviridae family indicates that it was truncated to t the Figure.

distinction into regular and irregular enveloped viruses. In the former, the enveloped glycoproteins display specic lateral interactions to form a rigid shell, often with icosahedral symmetry. This shell completely encases the viral membrane. In contrast, in the virions with irregular envelopes HIV, inuenza and herpes viruses are examples the glycoproteins appear to be randomly distributed at the surface, and do no form a symmetric scaffold. The membrane fusion protein of these viruses was shown to fold into a trimeric alpha-helical coiled coil during a fusogenic conformational change triggered by interactions with the host to drive lipid merger (Harrison, 2008). This central coiled coil is the hallmark of the class I (Colman and Lawrence, 2003) and the more recently described class III (Backovic and Jardetzky, 2009) viral fusion proteins, which are present in the irregular enveloped viruses studied so far. Regular enveloped viruses There are two types of viruses for which structural studies are available, in which the virions display a lipid bilayer encased within a symmetric protein shell (reviewed by Huiskonen and Butcher, 2007). The rst type includes a subset of large double-stranded DNA viruses infecting bacteria, archaea and eukaryotes of which mimivirus (Klose et al., 2010), which infects amoeba, is one

example. These viruses have in common an external icosahedral capsid architecture, formed by a capsid protein folded as a double jelly-roll motif related to the adenovirus hexon protein. They have been proposed to form a specic clade, known as the adenovirus-PRD1 lineage (Abrescia et al., 2004; Cockburn et al., 2004). This lineage includes viruses that infect mammals, like the African swine fever virus. However, in the latter viruses, a second envelope surrounding the icosahedral shell that encloses the inner membrane is acquired by the extracellular particle upon budding at the plasma membrane (Hawes et al., 2008). Entry into a target cells occurs by fusion of this external envelope with endosomal membranes, driven by a membrane fusion protein which has not been characterized structurally present at the surface of the extracellular virion and not involved in regular interactions with its neighbours. Viruses having the second type of regular enveloped virions, which include many important human pathogens, are the focus of this review. The membrane fusion proteins of these viruses are involved in forming the external symmetric coat that covers the viral membrane. They are single-stranded RNA (ssRNA) viruses encoding a class II membrane fusion protein (Kielian, 2006). The alphaviruses, the aviviruses and the phleboviruses (Fig. 1) have been shown to form regular enveloped particles, with the membrane encased within a relatively
2011 Blackwell Publishing Ltd, Cellular Microbiology

Enveloped virions with symmetric coats 3

Fig. 2. Comparison of the virion organization in class II virus families. Central slices across the glycoprotein shell of virions belonging to the three viral families dened in Fig. 1. A. The two panels show an 80 thick slice through the centre of the particle, made from the atomic model inferred by docking the X-ray structures into the cryo-EM density. Segments that were absent from the crystal structures [trans-membrane segments and stem regions of glycoproteins M and E for aviviruses and E2 and E1 for alphaviruses, coloured lime green and blue, respectively) were modelled into the remaining unoccupied electron density in the cryo-EM reconstruction (Zhang et al., 2003; Mukhopadhyay et al., 2006). The lipids were not modelled, but the rough location of the lipid heads of the outer and inner leaets of the viral membrane is indicated by the red arc (top-right) and central circle, respectively, in the avivirus panel and by two black arcs in the alphavirus panel (top-right, at roughly 1 oclock), with the trans-membrane regions crossing the bilayer. The alphavirus core protein (C) has a structured N-terminal domain of about 200 aa folded as chymotrypsin (Choi et al., 1991), which makes the orange scaffold within the membrane, whereas its 100 N terminal amino acids are disordered and interact with the genomic RNA (not visible). The avivirus C protein is only about 100 amino acids long; it is as basic as the C-terminal tail of alphavirus C, and also makes a complex with the genomic RNA that is not visible in the reconstruction. Note the similar volume for the genomic RNA present in both particles. B. Central slice 4 thick of a 20 cryo-EM reconstruction of the Rift valley fever virus particle [adapted from Huiskonen et al. (2009), with permission]. No X-ray structures are available yet for docking into the reconstruction, so that the details of the organization of the virion are not known. The density is represented in a scale from white (zero or negative electron density) to black (high density, corresponding to protein and lipid heads). The two leaets of the lipid bilayer are clearly apparent, as are the glycoprotein complexes at the surface. Note that the volume available for the three bunyavirus genomic segments is about an order of magnitude bigger than that of the aviviruses and alphaviruses (the scale is the same for all three panels). The total number of nucleotides in the genome is comparable, however, being between 11 and 12 kilobases for the three viruses. In bunyaviruses as in all nsRNA viruses, the genome is tightly packed in a complex with several thousand copies of the viral nucleoprotein N. In addition, the particle contains several copies of the viral polymerase L.

rigid glycoprotein shell of icosahedral symmetry. These viruses form, respectively, individual genera within the Togaviridae, the Flaviviridae and the Bunyaviridae families of ssRNA viruses, listed in Fig. 1. The Togaviridae family includes viruses with a positive sense RNA genome (psRNA, i.e., the genome is itself an mRNA). The genomic mRNA has a single open reading frame (ORF) coding for a polyprotein from which derive all non-structural proteins. The ORF coding for the polyprotein precursor of the structural proteins, outlined in Fig. 1, is present in a sub-genomic mRNA. The viruses of the Flaviviridae family also have a psRNA genome, in which the genomic mRNA has a single ORF encoding the polyprotein precursor of all viral proteins, structural and non-structural (NSP in Fig. 1). Finally, the Bunyaviridae is a family in which the genome is divided in three segments of negative-polarity ssRNA (nsRNA, i.e., the corresponding mRNAs are complementary to the genomic ssRNA). The envelope proteins are encoded in
2011 Blackwell Publishing Ltd, Cellular Microbiology

the middle (M) segment, also giving rise to a polyprotein precursor to the envelope proteins, and in some genera also to non-structural proteins (Fig. 1). The class II fold X-ray crystallography studies have been carried out on the envelope glycoproteins of aviviruses (Rey et al., 1995; Modis et al., 2003; Nybakken et al., 2006; Li et al., 2008) and of alphaviruses (Lescar et al., 2001; Roussel et al., 2006; Li et al., 2010; Voss et al., 2010); (Figs 2 and 3). Flavivirus E is responsible for receptor binding and for inducing membrane fusion, whereas alphavirus E1 drives membrane fusion and E2 binds to the viral receptor. E and E1 are elongated, thin molecules displaying a complex fold rich in b-sheet, displayed coloured according to domains in Fig. 3B. They have a central beta-barrel (called domain I, red) featuring two insertions between consecutive beta-strands to form a long fusion

M.-C. Vaney and F. A. Rey

Fig. 3. Regular packing of envelope proteins at the surface of infectious class II virions. The avivirus (top row) and alphavirus (bottom row) are displayed in surface rendering (A and C) and as ribbons (B). A. Both panels (top and bottom) are at the same scale. In the avivirus particle, each E subunit is in three different environments, resulting in non-equivalent contacts with its neighbours. Of the 90 E-protein dimers present at the surface, 30 have the corresponding E subunits related by the icosahedral twofold axes of the particle (displayed in blue, except for the one at the centre, which is coloured as in B and C and outlined with a black contour, for reference). The remaining 60 E dimers are in general positions. The E subunits of the local dimers lie one by the vefold (mauve) and the other by the threefold (pale yellow). The alphavirus particle (bottom panel) has triangulation T = 4 and is made of 80 trimeric spikes made of E1/E2 heterodimeric subunits. E2 is coloured grey, and E1 is coloured according to domains, as in the other panels. B. Ribbon diagram of an E homodimer (top), and the E2/E1 heterodimer (bottom), with E1 and one E subunit coloured by domains (red, yellow and blue for domains I, II and III, respectively, with the fusion loop in orange). E2 is shown in grey, as is the second E subunit in the homodimer. The membrane is below the dimer, in the plane of the paper (top panel), whereas it is perpendicular to the paper in the bottom panel. The blue arrow points to a cartoon drawing of the stem region (absent from the crystal structure) with E1 in blue and E2 in grey, crossing the viral membrane. Three E1/E2 heterodimers associate about a vertical axis to form each of the 80 spikes that compose the virion (shown in A in same colours, but solid surfaces). The black squares mark the region highlighted in (C), with a black arrow indicating the view of avivirus E shown in (C). For E1/E2, the view in C is the same as that in (B). C. Interactions of the fusion loop within the glycoprotein shell. The unit in the foreground is displayed in ribbons, and the one in the background is shown as a grey surface. A tryptophan residue at the centre of the fusion loop is represented in sticks, as well as a phenylalanine that is essential for fusion. Disulde bonds are displayed as green tubes.

domain (domain II, yellow), with a fusion loop (orange) at the tip of the rst insertion at the distal end of the molecule. The third domain (domain III, blue) displays an immunoglobulin superfamily fold. Both avivirus E and alphavirus E1 display the same complex fold despite the lack of detectable sequence conservation. This nding strongly suggests that the corresponding genes have diverged from a common ancestor, as only mutations compatible with the 3D fold are tolerated during the mutation and selection process driving evolution. Furthermore, the structural studies revealed a similar way of maintaining the fusion loop buried on the infectious particle, at the E homodimer interface in the case of the avivirus and at the E1/E2 heterodimer interface in the alphaviruses (Fig. 3C), with a similar pH activation mechanism.

Icosahedral organization of avivirus and alphavirus particles Electron cryo-microscopy (cEM) 3D reconstructions of intact avivirus (Zhang et al., 2003) and alphavirus (Mukhopadhyay et al., 2006) infectious virions to about 9 resolution have shown the organization of the whole particles. In addition, an atomic model for the whole external assembly was inferred by docking the crystal structures into the cEM reconstructions for viruses in both genera. These models identied the residues that are involved in the icosahedral interactions at the particle surface, those that are exposed and available for interaction with cellular receptors, and those that form the main epitopes recognized by antibodies. Furthermore, the resolution of the cEM maps has allowed to assign alpha 2011 Blackwell Publishing Ltd, Cellular Microbiology

Enveloped virions with symmetric coats 5 helical regions of segments of the polypeptide chain that were missing from the crystal structures, like the transmembrane regions, and the membrane proximal stem region (Fig. 2A). In addition, as discussed below, viruses in the avivirus genus undergo a maturation step on the virion, which assembles in the ER of the infected cell as immature particles. The structural studies have provided very important insight into the complex process of particle maturation, which involves a dramatic rearrangement of the particle surface (Li et al., 2008; Yu et al., 2008). The structural biology of these viruses has been reviewed recently (Mukhopadhyay et al., 2005; Jose et al., 2009; Sanchez-San Martin et al., 2009) Organization of bunyavirus particles In the case of the Bunyaviridae, the particles have a larger diameter (Fig. 2) and are more fragile and difficult to study. For a long time, the virions in this family were thought to be pleomorphic, in spite of early reports indicating that the particles of Uukuniemi virus (UUKV, a member of the phlebovirus genus, see Fig. 1), had icosahedral symmetry with triangulation T = 12 (von Bonsdorff and Pettersson, 1975; Pettersson, 1975). Electron microscopy studies using negative-stained samples had also revealed specic surface features of bunyavirus particles, characteristic of each genus, some of which appeared compatible with icosahedral symmetry (Martin et al., 1985). Analysis of vitried hydrated samples of particles of La Crosse virus, which belongs to the orthobunyavirus genus, have shown spherical virions of different sizes similar to those of the phleboviruses but displaying longer spikes which could be compatible with icosahedral symmetry of various triangulations, although the symmetry of the particles was not identied (Talmon et al., 1987). Icosahedral organization of phlebovirus particles More recently, the advent of electron cryo-tomography allowed substantial progress in the structural characterization of viruses of this family. This technique allows low-resolution 3D reconstructions calculated from multiple projections of the same particle [reviewed in (Grunewald and Cyrklaff, 2006)] instead of averaging the projections of many randomly oriented particles as in the standard cEM single particle approach, which requires all particles to be identical. Averaging of individual tomograms obtained for individual particles led to initial reconstructions of UUKV particles (Overby et al., 2008) and for the human pathogenic Rift Valley fever virus (RVFV, also a phlebovirus) (Freiberg et al., 2008) to a resolution of about 60 , conrming the icosahedral symmetry with triangulation T = 12 for phleboviruses. These initial studies contributed to realizing that the phlebovirus
2011 Blackwell Publishing Ltd, Cellular Microbiology

virions, when correctly preserved upon purication, are identical, which in a real tour-de-force led to subsequent reconstructions by cEM and single particle averaging to about 20 resolution (Huiskonen et al., 2009; Sherman et al., 2009). The resulting structures demonstrate the presence of ve- and six-fold co-ordinated capsomers, which were interpreted as being multimers of Gn/Gc heterodimeric building blocks. As shown in Fig. 2, the lateral packing of capsomers is somewhat reminiscent of that of the alphavirus spikes, which are formed by E2/E1 heterodimers, but which are organized as trimers instead of hexamers and pentamers. The larger volume inside the virion, and the fact that there is no rigid nucleocapsid within the membrane underpinning the glycoprotein shell, explains the fragility of the particles and the difficulties in preserving the symmetry. Organization of hantavirus particles An interesting electron cryo-tomography study was carried out on Tula virus, a member of the hantavirus genus. The viral particles display considerable heterogeneity, but the use of sub-tomogram averaging resulted in a 3D reconstruction of the particles to about 30 resolution (Huiskonen et al., 2010). The particles had a median diameter of 1350 , with some elongated particles with a length of 3500 and a diameter of 800 . This study showed that the Tula virions display spikes with fourfold symmetry, arranged on a square lattice making various patches at the virion surface, conrming earlier observations by negative staining (Martin et al., 1985). These spikes were interpreted as being tetramers made of Gn/Gc heterodimeric subunits (Huiskonen et al., 2010). The lateral packing of spikes suggested that they induce curvature upon interactions with their neighbours to form a closed particle. Fourfold symmetry is, however, not compatible with icosahedral symmetry, opening the question of how these interactions lead to particle closure during budding, given that an important part of the driving force for budding appears to be derived from the lateral packing between spikes, as for the other viruses described above. Bunyaviruses and the class II fold To summarize, the postulate that the bunyavirus Gc proteins are class II fusion proteins stems from the following observations: (i) the secondary structure predicted from their amino acid sequences indicates a high b-sheet content, (ii) they are synthesized within a polyprotein precursor, downstream of a companion glycoprotein (Gn, Fig. 1), (iii) in all bunyaviruses for which studies have been made, Gn and Gc heterodimerize in the ER before folding is complete, and are then transported to the Golgi, where

M.-C. Vaney and F. A. Rey civiruses will also be the second glycoprotein in the polyprotein precursor and that it displays a class II fold. Rubivirus In the case of rubella virus, the only member of the rubivirus genus, it was shown that E2 and E1 heterodimerize in the ER, and that E1 folds much more slowly than E2 (Baron and Forsell, 1991; Hobman et al., 1993; 1995), before transport to the Golgi, where budding of new particles takes place (Risco et al., 2003). Glycoprotein E1 was shown to contain a fusion peptide, and hence that it is the membrane fusion protein (Qiu et al., 2000), as predicted by comparison with the alphaviruses. However, electron microscopy studies have not detected icosahedral symmetry on the virus particles, perhaps because the symmetry is difficult to preserve during manipulation of the virions, as in the bunyavirus case. Hepaciviruses and pestiviruses There are no experimental data showing that the second protein (E2 in this case, Fig. 1) is the viral membrane fusion protein for any virus in these two genera. Yet the organization of the precursor polyprotein is the same as in the others. The majority of the studies have been carried out with the hepatitis C virus (HCV, hepacivirus genus). It was shown that HCV E1 and E2 form a heterodimer in the ER of the infected cell. Particle morphogenesis takes place in the ER lumen in tight association with the very low density lipoprotein assembly pathway (Bartenschlager et al., 2011). The building blocks of the HCV particle are E1/E2 heterodimers (reviewed by Voisset and Dubuisson, 2004). The ectodomain of the E2 protein of both HCV and the pestiviruses is about 100 amino acids shorter than that of the aviviruses and alphaviruses, indicating that, if E2 is indeed a class II fusion protein, its ectodomain must be truncated. A controversy about the identity of the fusion protein arose when a study suggested that a segment in E1 had a putative fusion-loop-like sequence (Flint et al., 1999), but without providing experimental evidence for it. This prompted others to model E1 as just domain II of avivirus E, proposing a sort of truncated version of the class II molecules (Garry and Dash, 2003). Knowing the importance of hairpin formation to induce the membrane fusion reaction, which in class II fusion involves a relocation of domain III bringing the stem and trans-membrane segment towards the fusion loop (Kielian, 2006; SanchezSan Martin et al., 2009), it is difficult to see how a molecule having just domain II could be functional for membrane fusion. This same study, however, proposed a truncated model having both domains II and III within the same molecule, with a truncated domain I, which would make a more plausibly functional molecule (pestivirus E2 is even
2011 Blackwell Publishing Ltd, Cellular Microbiology

budding occurs [reviewed in (Schmaljohn and Nichol, 2007)] and (iv) the Gn/Gc heterodimers make an ordered scaffold at the virion surface, completely (or partially in the hantavirus case) enclosing the viral membrane. These four properties are shared by the glycoproteins of the aviviruses and the alphaviruses, for which the structure is known. Flaviviruses bud into the ER lumen as immature particles containing the prM/E heterodimer, and alphaviruses into the external medium as mature virions containing E2/E1 heterodimers, where E2 derives from furin cleavage of its p62 precursor. The issue of maturation has not been studied in full yet in bunyaviruses, although in the nairovirus genus Gn is also derived from a longer precursor (PreGn, Fig. 1). The overall implication of this analysis is that the bunyavirus Gc proteins are homologous to each other and to both avivirus E and alphavirus E1 (Fig. 1), and therefore share the same 3D fold (Fig. 3B). With respect to bunyavirus Gn, its counterparts in avivirus (prM) and in alphavirus (p62) are not structural homologues of each other although both proteins are rich in b-strands. It is not possible, therefore, to predict whether the bunyavirus Gn may display structural homology to either. Moreover, the Gn proteins from different bunyavirus genera may not display structural homology to each other either. The secondary structure predictions suggest that they are also likely to be rich in b-sheet. Only structural studies of the bunyavirus Gn/Gc heterodimers to high enough resolution to trace the polypeptide chains will be able to provide the answer. A corollary of the above analyses is that the bunyavirus Gc proteins are the membrane fusion proteins. This has been experimentally veried for La Crosse virus, an orthobunyavirus (Plassmeyer et al., 2005; 2007) and for the Andes hantavirus (Tischler et al., 2005). The latter study proposed a model for the hantavirus Gc glycoprotein obtained by threading the Gc sequence on the template provided by the crystal structure of the avivirus tick-borne encephalitis virus E protein (Rey et al., 1995). As mentioned above, there is no clear alignment between the proteins; however, a computational analysis had proposed a class II fold by noting that the amino acid sequence of Gc of certain viruses in the phlebovirus genus had signicant amino acid similarity with E1 from some members of the alphavirus group (Garry and Garry, 2004). The region of the fusion loop proposed by this analysis was later experimentally conrmed as involved in membrane fusion using La Crosse virus (Plassmeyer et al., 2007). The other genera in the Togaviridae and Flaviviridae family The similar organization of the envelope proteins in all the virus genera listed in Fig. 1 suggests that the fusion protein of the rubiviruses, the pestiviruses and the hepa-

Enveloped virions with symmetric coats 7 shorter than HCV E2, and it is likely that these very close genera have a similar membrane fusion mechanism). Recently, a model was proposed for the tertiary structure of HCV E2 based in the connectivity of the nine disulde bonds present in the molecule, as well as distant sites that are known to be part of the interaction site with one of the HCV receptors, CD81 (Krey et al., 2010). This model also allows for three domains, and points to a conserved region in the predicted domain II of E2, with properties of a fusion loop. With respect to the particle organization, no 3D data are available for pestivirus particles, but cEM studies on HCV virion-like particles have been reported (Yu et al., 2007), also suggesting an icosahedral organization of the particle, although the virion-like particles used in this study were highly heterogeneous in nature. A fairly featureless 30 resolution cEM 3D reconstruction was reported, showing a particle with a diameter of about 500 in which the structure of the avivirus E dimer was docked according to the herringbone organization of the mature avivirus particle (Fig. 3A). The low resolution of these studies and the lack of the crystal structure of the HCV E1/E2 heterodimer did not allow, however, a clear-cut understanding of the organization of the HCV particle. Furthermore, a more recent structural and biophysical analysis of infectious HCV particles (Gastaminza et al., 2010) suggested the presence of several morphologies, with enveloped particles displaying a diameter of about 600 . One important problem in deriving a structure of infectious HCV particles is that they incorporate cellular proteins, in particular apolipoproteins (reviewed by Bartenschlager et al., 2011), which may blur the original symmetry if there is one of the particles. However, the basic principle of particle assembly is likely to involve specic, regular lateral contacts between E1/E2 building blocks. Implications for virus evolution Although this analysis awaits conrmation of the class II status of the Gc proteins of the Bunyaviridae as well as the fusion proteins of HCV, pestiviruses and rubella virus by high-resolution structural studies, the implications are that the membrane fusogenic proteins of three different virus families were derived from a common ancestor, indicating divergent evolution of the corresponding gene. The genome of viruses in the rst two families listed in Fig. 1 the Flaviviridae and the Togaviridae is a single psRNA molecule, whereas in the Bunyaviridae family it is composed of three different nsRNA molecules (or three genomic segments). Their replication strategy is therefore totally unrelated. The Bunyaviridae are, however, very similar in their overall genome organization to the Arenaviridae, which are viruses with two segments of nsRNA. Both families encode for related RNA poly 2011 Blackwell Publishing Ltd, Cellular Microbiology

merases and share a similar replication strategy (Morin et al., 2010; Reguera et al., 2010), but the arenavirus envelope protein is clearly a class I fusion protein, with a predicted central alpha-helical coiled coil (Gallaher et al., 2001; Eschli et al., 2006), and the arenavirus virions are pleomorphic. They therefore constitute a striking example and not the only one of two related viral families in which the genes coding for the proteins that form the vehicle for transmission from cell to cell were derived from a source different to that of the genes responsible for genome replication. This is a clear illustration of the polyphyletic origin of viruses, which display a mosaic of genes of different origins that are pooled together to act as separate modules at different steps of the viral cycle, and which have been adapted throughout evolution to carry their functions in synergy. This point is further illustrated when considering the relation between retroviruses and retrotransposons with long terminal repeats. Retroviruses from vertebrates appear to have emerged from long terminal repeats-retrotransposons by the acquisition of a third ORF: the envelope (env) gene. The env gene typically encodes a prototypical class I envelope protein. However, the nematode endogenous retroviruses originated by acquiring an env gene from a phleboviral source (Malik et al., 2000; Frame et al., 2001), i.e., a class II envelope protein gene. The main conclusion from this analysis is that for many viruses, it is not possible to talk about an overall lineage, but about the lineages of specic modules of the virus cycle: of the viral genes used for genome replication, or for the vehicle for virus dissemination, or for other functions important for the right interactions with the host to ensure virus perpetuation in nature.

References
Abrescia, N.G., Cockburn, J.J., Grimes, J.M., Sutton, G.C., Diprose, J.M., and Butcher, S.J. et al. (2004) Insights into assembly from structural analysis of bacteriophage PRD1. Nature 432: 6874. Backovic, M., and Jardetzky, T.S. (2009) Class III viral membrane fusion proteins. Curr Opin Struct Biol 19: 189196. Banerjee, M., and Johnson, J.E. (2008) Activation, exposure and penetration of virally encoded, membrane-active polypeptides during non-enveloped virus entry. Curr Protein Pept Sci 9: 1627. Baron, M.D., and Forsell, K. (1991) Oligomerization of the structural proteins of rubella virus. Virology 185: 811819. Bartenschlager, R., Penin, F., Lohmann, V., and Andre, P. (2011) Assembly of infectious hepatitis C virus particles. Trends Microbiol 19: 95103. von Bonsdorff, C.H., and Pettersson, R. (1975) Surface structure of Uukuniemi virus. J Virol 16: 12961307. Choi, H.K., Tong, L., Minor, W., Dumas, P., Boege, U., Rossmann, M.G., et al. (1991) Structure of Sindbis virus core protein reveals a chymotrypsin-like serine proteinase and the organization of the virion. Nature 354: 3743.

M.-C. Vaney and F. A. Rey

(2009) Electron cryo-microscopy and single-particle averaging of Rift Valley fever virus: evidence for Gn-Gc glycoprotein heterodimers. J Virol 83: 37623769. Huiskonen, J.T., Hepojoki, J., Laurinmaki, P., Vaheri, A., Lankinen, H., Butcher, S.J., et al. (2010) Electron cryotomography of Tula hantavirus suggests a unique assembly paradigm for enveloped viruses. J Virol 84: 48894897. Jose, J., Snyder, J.E., and Kuhn, R.J. (2009) A structural and functional perspective of alphavirus replication and assembly. Future Microbiol 4: 837856. Kielian, M. (2006) Class II virus membrane fusion proteins. Virology 344: 3847. Klose, T., Kuznetsov, Y.G., Xiao, C., Sun, S., McPherson, A., and Rossmann, M.G. (2010) The three-dimensional structure of Mimivirus. Intervirology 53: 268273. Krey, T., dAlayer, J., Kikuti, C.M., Saulnier, A., Damier-Piolle, L., Petitpas, I., et al. (2010) The disulde bonds in glycoprotein E2 of hepatitis C virus reveal the tertiary organization of the molecule. PLoS Pathog 6: e1000762. Lescar, J., Roussel, A., Wien, M.W., Navaza, J., Fuller, S.D., Wengler, G., et al. (2001) The Fusion glycoprotein shell of Semliki Forest virus: an icosahedral assembly primed for fusogenic activation at endosomal pH. Cell 105: 137 148. Li, L., Lok, S.M., Yu, I.M., Zhang, Y., Kuhn, R.J., Chen, J., et al. (2008) The avivirus precursor membrane-envelope protein complex: structure and maturation. Science 319: 18301834. Li, L., Jose, J., Xiang, Y., Kuhn, R.J., and Rossmann, M.G. (2010) Structural changes of envelope proteins during alphavirus fusion. Nature 468: 705708. Malik, H.S., Henikoff, S., and Eickbush, T.H. (2000) Poised for contagion: evolutionary origins of the infectious abilities of invertebrate retroviruses. Genome Res 10: 1307 1318. Martin, M.L., Lindsey-Regnery, H., Sasso, D.R., McCormick, J.B., and Palmer, E. (1985) Distinction between Bunyaviridae genera by surface structure and comparison with Hantaan virus using negative stain electron microscopy. Arch Virol 86: 1728. Mercer, J., Schelhaas, M., and Helenius, A. (2010) Virus entry by endocytosis. Annu Rev Biochem 79: 803 833. Modis, Y., Ogata, S., Clements, D., and Harrison, S.C. (2003) A ligand-binding pocket in the dengue virus envelope glycoprotein. Proc Natl Acad Sci USA 100: 69866991. Morin, B., Coutard, B., Lelke, M., Ferron, F., Kerber, R., Jamal, S., et al. (2010) The N-terminal domain of the arenavirus L protein is an RNA endonuclease essential in mRNA transcription. PLoS Pathog 6: e1001038. Mukhopadhyay, S., Kuhn, R.J., and Rossmann, M.G. (2005) A structural perspective of the avivirus life cycle. Nat Rev Microbiol 3: 1322. Mukhopadhyay, S., Zhang, W., Gabler, S., Chipman, P.R., Strauss, E.G., Strauss, J.H., et al. (2006) Mapping the structure and function of the E1 and E2 glycoproteins in alphaviruses. Structure 14: 6373. Nybakken, G.E., Nelson, C.A., Chen, B.R., Diamond, M.S., and Fremont, D.H. (2006) Crystal structure of the West Nile virus envelope glycoprotein. J Virol 80: 11467 11474.
2011 Blackwell Publishing Ltd, Cellular Microbiology

Cockburn, J.J., Abrescia, N.G., Grimes, J.M., Sutton, G.C., Diprose, J.M., Benevides, J.M., et al. (2004) Membrane structure and interactions with protein and DNA in bacteriophage PRD1. Nature 432: 122125. Colman, P.M., and Lawrence, M.C. (2003) The structural biology of type I viral membrane fusion. Nat Rev Mol Cell Biol 4: 309319. Eschli, B., Quirin, K., Wepf, A., Weber, J., Zinkernagel, R., and Hengartner, H. (2006) Identication of an N-terminal trimeric coiled-coil core within arenavirus glycoprotein 2 permits assignment to class I viral fusion proteins. J Virol 80: 58975907. Flint, M., Thomas, J.M., Maidens, C.M., Shotton, C., Levy, S., Barclay, W.S., et al. (1999) Functional analysis of cell surface-expressed hepatitis C virus E2 glycoprotein. J Virol 73: 67826790. Frame, I.G., Cuteld, J.F., and Poulter, R.T. (2001) New BELlike LTR-retrotransposons in Fugu rubripes, Caenorhabditis elegans, and Drosophila melanogaster. Gene 263: 219230. Freiberg, A.N., Sherman, M.B., Morais, M.C., Holbrook, M.R., and Watowich, S.J. (2008) Three-dimensional organization of Rift Valley fever virus revealed by cryoelectron tomography. J Virol 82: 1034110348. Gallaher, W.R., DiSimone, C., and Buchmeier, M.J. (2001) The viral transmembrane superfamily: possible divergence of Arenavirus and Filovirus glycoproteins from a common RNA virus ancestor. BMC Microbiol 1: 1. Garry, C.E., and Garry, R.F. (2004) Proteomics computational analyses suggest that the carboxyl terminal glycoproteins of Bunyaviruses are class II viral fusion protein (beta-penetrenes). Theor Biol Med Model 1: 1026. Garry, R.F., and Dash, S. (2003) Proteomics computational analyses suggest that hepatitis C virus E1 and pestivirus E2 envelope glycoproteins are truncated class II fusion proteins. Virology 307: 255265. Gastaminza, P., Dryden, K.A., Boyd, B., Wood, M.R., Law, M., Yeager, M., et al. (2010) Ultrastructural and biophysical characterization of hepatitis C virus particles produced in cell culture. J Virol 84: 1099911009. Grunewald, K., and Cyrklaff, M. (2006) Structure of complex viruses and virus-infected cells by electron cryo tomography. Curr Opin Microbiol 9: 437442. Harrison, S.C. (2008) Viral membrane fusion. Nat Struct Mol Biol 15: 690698. Hawes, P.C., Netherton, C.L., Wileman, T.E., and Monaghan, P. (2008) The envelope of intracellular African swine fever virus is composed of a single lipid bilayer. J Virol 82: 79057912. Hobman, T.C., Woodward, L., and Farquhar, M.G. (1993) The rubella virus E2 and E1 spike glycoproteins are targeted to the Golgi complex. J Cell Biol 121: 269281. Hobman, T.C., Woodward, L., and Farquhar, M.G. (1995) Targeting of a heterodimeric membrane protein complex to the Golgi: rubella virus E2 glycoprotein contains a transmembrane Golgi retention signal. Mol Biol Cell 6: 720. Huiskonen, J.T., and Butcher, S.J. (2007) Membranecontaining viruses with icosahedrally symmetric capsids. Curr Opin Struct Biol 17: 229236. Huiskonen, J.T., Overby, A.K., Weber, F., and Grunewald, K.

Enveloped virions with symmetric coats 9

Overby, A.K., Pettersson, R.F., Grunewald, K., and Huiskonen, J.T. (2008) Insights into bunyavirus architecture from electron cryotomography of Uukuniemi virus. Proc Natl Acad Sci USA 105: 23752379. Pettersson, R.F. (1975) The structure of Uukuniemi virus, a proposed member of the bunyaviruses. Med Biol 53: 418 424. Plassmeyer, M.L., Soldan, S.S., Stachelek, K.M., MartinGarcia, J., and Gonzalez-Scarano, F. (2005) California serogroup Gc (G1) glycoprotein is the principal determinant of pH-dependent cell fusion and entry. Virology 338: 121132. Plassmeyer, M.L., Soldan, S.S., Stachelek, K.M., Roth, S.M., Martin-Garcia, J., and Gonzalez-Scarano, F. (2007) Mutagenesis of the La Crosse Virus glycoprotein supports a role for Gc (10661087) as the fusion peptide. Virology 358: 273282. Qiu, Z., Yao, J., Cao, H., and Gillam, S. (2000) Mutations in the E1 hydrophobic domain of rubella virus impair virus infectivity but not virus assembly. J Virol 74: 66376642. Reguera, J., Weber, F., and Cusack, S. (2010) Bunyaviridae RNA polymerases (L-protein) have an N-terminal, inuenza-like endonuclease domain, essential for viral cap-dependent transcription. PLoS Pathog 6: e1001101. Rey, F.A., Heinz, F.X., Mandl, C., Kunz, C., and Harrison, S.C. (1995) The envelope glycoprotein from tick-borne encephalitis virus at 2 A resolution. Nature 375: 291 298. Risco, C., Carrascosa, J.L., and Frey, T.K. (2003) Structural maturation of rubella virus in the Golgi complex. Virology 312: 261269. Roussel, A., Lescar, J., Vaney, M.C., Wengler, G., and Rey, F.A. (2006) Structure and interactions at the viral surface of the envelope protein E1 of Semliki Forest virus. Structure 14: 7586. Sanchez-San Martin, C., Liu, C.Y., and Kielian, M. (2009) Dealing with low pH: entry and exit of alphaviruses and aviviruses. Trends Microbiol 17: 514521. Schmaljohn, C.S., and Nichol, S.T. (2007) Bunyaviridae. In Fields Virology, 5th edn. Knipe, D.M., and Howley, P.M. (eds). Philadelphia, PA: Lippincott Williams & Wilkins, pp. 17411789. Sherman, M.B., Freiberg, A.N., Holbrook, M.R., and Watowich, S.J. (2009) Single-particle cryo-electron microscopy of Rift Valley fever virus. Virology 387: 1115. Talmon, Y., Prasad, B.V., Clerx, J.P., Wang, G.J., Chiu, W., and Hewlett, M.J. (1987) Electron microscopy of vitriedhydrated La Crosse virus. J Virol 61: 23192321. Tischler, N.D., Gonzalez, A., Perez-Acle, T., Rosemblatt, M., and Valenzuela, P.D. (2005) Hantavirus Gc glycoprotein: evidence for a class II fusion protein. J Gen Virol 86: 29372947. Voisset, C., and Dubuisson, J. (2004) Functional hepatitis C virus envelope glycoproteins. Biol Cell 96: 413420. Voss, J.E., Vaney, M.C., Duquerroy, S., Vonrhein, C., GirardBlanc, C., Crublet, E., et al. (2010) Glycoprotein organization of Chikungunya virus particles revealed by X-ray crystallography. Nature 468: 709712. Yu, I.M., Zhang, W., Holdaway, H.A., Li, L., Kostyuchenko, V.A., Chipman, P.R., et al. (2008) Structure of the immature dengue virus at low pH primes proteolytic maturation. Science 319: 18341837. Yu, X., Qiao, M., Atanasov, I., Hu, Z., Kato, T., Liang, T.J., et al. (2007) Cryo-electron microscopy and threedimensional reconstructions of hepatitis C virus particles. Virology 367: 126134. Zhang, W., Chipman, P.R., Corver, J., Johnson, P.R., Zhang, Y., Mukhopadhyay, S., et al. (2003) Visualization of membrane protein domains by cryo-electron microscopy of dengue virus. Nat Struct Biol 10: 907912.

2011 Blackwell Publishing Ltd, Cellular Microbiology