Ed Rybicki, October 2000
The world of RNA plant and animal viruses is a wide and complex one: they may have
and have single or multiple genome components; in simple naked, or complex enveloped virions. It is quite feasible that RNA genomes of viruses are the only extant lineal descendants of the primaeval RNA world, given that they still replicate as the progenitor genome is supposed to have - that is, by use of a template-specific RNA-dependent RNA polymerase (RdRp), which is also known as an RNA replicase. See here for a brief description of types of RNA viruses and their genomes.
This section will cover the essentials of the life cycles of a range of RNA plant and animal viruses in a comparative manner, from entry into, to release from the host cell, highlighting similarities among and differences between the main groups of viruses mentioned above, with specific examples where relevant.
Plant and animal virus taxa to be covered, and their broad properties, are shown in Table 1. It can readily be seen that although there are a number of genera and families that are unique to either type of host, there are also a number of viruses in the same families that have different hosts.
These viruses seem to be connected in that if they do not also infect arthropods, then they are related to viruses that do.
Plants, fungi and arthropods were the first complex organisms to colonise the terrestrial environment; a close association therefore developed between them before any chordates emerged from the oceans, to which arthropods then subsequently adapted as new hosts - meaning that similar arthropod-derived viruses could have adapted to very dissimilar alternative hosts. It is interesting that most plant viruses are ssRNA(+) and non-enveloped, while most animal viruses are enveloped: this could mean that early terrestrial plants contained mainly ssRNA(+) viruses, which then developed in isolation until terrestrial arthropods emerged. It may also mean that envelopes confer no survival advantage on plant viruses.
Other material on virus evolution may be found here.
All RNA viruses have linear genomes, without significant terminal repeat sequences, and all employ RdRps. These are template-specific, but do not have proofreading ability, and do not make use of RNA primers for replication, as do all DNA polymerases. The RdRps also all specifically recognise different origins of replication at the 3'-termini of both (+) and (-) sense RNAs, whatever the type of genome.
There is a fundamental difference in mechanisms employed to enter host cells between viruses infecting animal cells and viruses infecting plants. This is because animal cells are separated by barriers far less formidable than the thick, rigid and impermeable cell walls consisting of cellulose and pectin that separate plant cells from one another. This is covered in a general way here.
Because plant cell walls are so thick compared to the sizes of the viruses infecting them (>10 m m compared to largely <1 m m), plant viruses have not evolved mechanisms similar to those of bacteriophages for entering their host cells. The only ways that viruses can enter plant cells to cause a primary infection are via:
1) a purely mechanical injury that breaches the cell wall and transiently breaches the plasma membrane of underlying cells;
2) similar gross injury due to the mouthparts of a herbivorous arthropod, such as a beetle;
3) injection directly into cells through the piercing mouthparts of sap-sucking insects or nematodes;
4) carriage into plant tissue on or in association with cells of a fungal parasite;
5) vertical transmission through infected seed or by vegetative propagation;
6) transmission via pollen; and
7) grafting of infected tissue onto healthy tissue.
For example, the ssRNA(-) viruses Tomato spotted wilt virus (TSWV) and Crimean-Congo haemorrhagic fever virus (CCHFV) - both in family Bunyaviridae (see Table 1) - share a common particle morphology, and infect the cells of their respective arthropod hosts in similar ways: that is, by a specific attachment and a fusion or phagosomal uptake mechanism. CCHFV also infects the the cells of its mammalian hosts similarly (see below). However, TSWV infects plant cells by injection directly into cells via the piercing mouthparts of its insect vector, the Western flower thrips, and not via membrane interactions.
Once virions are in the cytoplasm, they are generally uncoated to some extent by a variety of processes, including simple dissociation and/or enzyme-mediated partial degradation of the particles, to release the viral genome as a naked RNA or as a nucleoprotein complex.
The initial phase of cell entry starts when attachment proteins on the virion surface attach to specific receptors on the cell surface. Both attachment proteins and receptors are normally glycoproteins; the cellular proteins can be things like transplantation markers (MHC proteins), adhesins, or simply sialyloligosaccharides (sugars attached to glycoproteins) in the case of ortho- and paramyxoviruses. The attachment is normally temperature- and pH-dependent, and is due to the same sorts of molecular structural complementarities - "lock and key" fit - as occur with enzyme-substrate and antibody-antigen binding.
There are essentially two different paths that are followed for entry into the cell: these are
The first is perhaps the primary means of viral cell entry, and is simply a subversion of a normal cellular process. Virus particles become attached at multiple sites to cellular receptors, as these consolidate within the plasma membrane. If these complexes migrate to coated pits, they are internalised as clathrin-coated vesicles as part of normal endocytosis. These vesicles quickly fuse with endosomes and then lysosomes, which renders their internal environment considerably more acidic and introduces a host of degradative proteases, lipases, and other enzymes. The pH shift generally triggers conformational changes in the attachment protein complexes, which in the case of enveloped virions, may expose lipophilic "fusion domains" that allow fusion of the viral envelope and the vesicle membrane. This has been show to occur with orthomyxoviruses, for example. In the case of non-enveloped virions, pH-induced conformational changes in the capsid may cause increased hydrophobicity / lipophilicity, which will allow interactions with the vesicle membrane that can cause pore formation. This is known to occur with picornaviruses. In either case, the result is the entry of an RNA-protein complex (=nucleoprotein or nucleocapsid) or of naked RNA into the cytoplasm, which is the most important part of the uncoating process.
Direct membrane fusion as a mode of entering cells is possible only with enveloped viruses, and is common among paramyxoviruses. The viruses require a fusion-promoting protein on their virion surfaces, which, in the presence of consolidated receptor-attachment protein binding, promotes fusion of cell and virion membranes and the release of the nucleoprotein into the cell cytoplasm. This is a pH-independent process, and may occur at the cell surface, or within an endosomal vesicle.
The replication of these viruses is intimately involved with the expression of their genomes: all of the viruses must produce all or most of the components of an RdRp, and often other proteins as well, in order to transcribe full-length complementary RNA molecules from RNA templates (see Figure 1). Whereas modes of entry of the viruses split largely along host lines, the exact type of genome of the virus determines the mode(s) of expression and replication. For example,
Additionally, all plant-infecting viruses possess one or more movement-related protein (MP) genes: these are very varied, although there are distinct groups of them, and they appear to derive from host plant genes for chaperonins and plasmodesmata-associated proteins (Melcher, 2000).
All of these viruses have wholly or partially translatable genomes, and as a result are usually infectious as naked RNA. Apart from this common feature, there are few other evolutionary similarities. After partial or complete uncoating upon entry into the cytoplasm, the genomic RNA is recognised by the translation initiation factors and ribosomal subunits and translation of the open reading frame (ORF) nearest the 5' end of the RNA(s) is initiated. If there are still proteins bound to the RNA, which in the case of plant viruses is most likely, this process efficiently strips them off. While all ssRNA(+) genomes have at least one ORF accessible for translation, and those with multicomponent genomes will have more than one, expression of any with more than one ORF per genome segment will suffer from the limitation that the eukaryotic translation machinery is heavily biased to expressing only the 5'-proximal ORF. Thus, these viruses have evolved two main strategies for expressing their whole genomes. These are:
The first strategy is typical of a group of viruses including the Picornaviridae and Potyviridae: these viruses make use of co- and post-translational cleavages by virus-coded endoproteinases, as well as of sequential and sometimes alternative cleavages, in order to make a large number of proteins as well as to regulate their own replication (see Figure 2) (Rueckert, 1996). The strategy results in near-equimolar amounts of the different proteins being made; because this is undesirable in the case of RdRp, the replicase complex for these viruses is "single-use", in that a freshly-synthesised RdRp polyprotein is needed to initiate replication on each new template. Another means of removing excess components is employed by potyviruses, which export their so-called "nuclear inclusion body" (NIa and NIb) or replicase protein subunits to the nucleus to be sequestered as insoluble aggregates. Comoviruses and nodaviruses resemble picornaviruses that have been cut into two segments; the bymoviruses in family Potyviridae similarly resemble a cleaved potyvirus genome. All of the viruses in a picornavirus-like supergroup (picorna-, poty-, como-, calici- and other viruses) use an RdRp that makes use of a protein as a primer for both (+) and (-) sense RNA production: this is part of the precursor RdRp and is cleaved off as elongation of the initial complex occurs, to become a 5'-genome-linked protein, usually known as Vpg. Viruses in this supergroup tend to have 3'-polyadenylated (polyA) genome segments, with the polyA sequences being part of the genome and copied into polyU in (-) sense RNA. The processed polyprotein strategy is shared by flaviviruses, whose genomes also contain a single large ORF; however, these virus genomes are not polyadenylated, and do not have a Vpg.
The second strategy is typical of two supergroups of viruses usually termed the alpha-like and carmo-like viruses in terms of sequence and genome organisation affinities (Strauss et al., 1996), and is exemplified by the generic tobamoviruses. The genome of Tobacco mosaic virus (TMV; see Figure 3) is initially expressed by means of translation of two proteins from its single 5'-proximal ORF: a 126 kDa protein is expressed ten times more abundantly than a 183 kDa protein, which is the product of read-through of a stop codon near the 3'-end of the large ORF. Both proteins are replicase components, along with at least one host protein (Lewandowski and Dawson, 2000). Further expression only occurs after synthesis of full-length RNA(-) from the RNA(+) template, and transcription of subgenomic mRNAs encompassing one or more of the 3'-proximal ORFs from this by internal initiation of transcription at RNA "promoter" sequences. These "promoters" differ from origins of replication in that they are not copied into mRNA transcripts, which are therefore not replicatable although they contain the 3'-origin. This allows temporal separation of early (=regulatory) and late (=structural) genes. The virus genomes usually have a cell mRNA-like 5'-cap structure (7-methyl guanosine triphosphate, m7Gppp), and many have a complex 3'-terminal structure, often resembling a tRNA (and aminoacylatable), and otherwise a series of pseudoknots. This serves both to protect the 3'-end from exonucleases and as a specific RdRp recognition site.
Some viruses utilise a mix of both strategies: for example, togaviruses (eg: Sindbis virus, genus Alphavirus) translate a proteolytically processed polyprotein that includes a RdRp; subsequent replication results in subgenomic mRNA production from the 3'-half of the genome, and production of another processable polyprotein. Viruses with segmented genomes may produce single proteins from single segments, or, in the case of bromoviruses or tobraviruses, for example, may have both monocistronic and multicistronic segments. In any case, all subgenomic mRNAs will be 3'-coterminal, as there appear to be no mechanisms for transcription termination. Coronavirus mRNAs appear to all have the same 5'-terminal leader sequence of 50-80 bases, indicating a more complicated form of transcription than recognition of internal promoter sequences in RNA(-) molecules.
Replication in all cases involves an initial transcription of full-length RNA(-) from an infecting RNA(+) template, and transcription from this of RNA(+), and perhaps also subgenomic mRNA. Replication complexes are usually closely associated with membrane complexes derived from the ER or perhaps nuclear membranes, and free RNA(-) is not found. In some cases it has been shown that coat protein (CP) helps regulate the expression of RNA(+), in that cp- mutants accumulate approximately equal amounts of both senses of RNA, while normal viruses accumulate much more RNA(+), especially as the CP concentration increases late in infection. dsRNA forms of viral genomes and of subgenomic RNAs can be isolated from infected cells for many ssRNA(+) viruses, including most plant viruses, some picorna-like insect viruses, and coronaviruses: this may be how some dsRNA viruses originated (see below).
These viruses seem to be an evolutionarily recent development, as they infect only higher eukaryotes, like arthropods, vertebrates, and higher plants. The viruses infecting plants probably do so as a result of close association of insects and host plants in recent evolutionary times; most of these still also infect an insect vector / alternative host. The group includes the only taxonomic order among RNA(-) viruses: this is the order Mononegavirales, including the families Orthomyxoviridae, Paramyxoviridae, and Filoviridae, all of which have single-component genomes and share a basic genome arrangement and significant sequence similarities (Figure 4). Nearly all of the RNA(-) viruses in Table 1 share a similar major RdRp subunit (L-type protein gene); there are also similarities in their nucleoproteins (N or NP genes).
Replication of all the viruses commences with the transcription by virion-associated RdRp of usually monocistronic mRNAs from genomic RNA(s) in the newly uncoated nucleoprotein complexes (see Figure 1). For the segmented genomes of bunya- and orthomyxoviruses, this usually means a single mRNA per segment; for the non-segmented mononegaviruses, this means multiple transcription initiation and termination events on a full-length RNA(-), at intergenic repeated sequences, with transcription apparently usually initiating at the genomic 3'-end with synthesis of a 50-base leader. Transcripts are capped and polyA tailed; the RdRp complex (consisting of L, N / NP and other proteins) adds caps, while tails are apparently added by RdRp stuttering at short polyU repeats at the end of genes. Independent transcription events in paramyxoviruses allow control of level of expression: these viruses transcribe far more mRNAs for structural protein genes at the 3'-end of the genome than for regulatory genes at the 5'-end, possibly due to the progressive failure of the RdRp complex at reinitiating multiple times down the length of the RNA(-) (Lamb and Kolakofsky, 1996).
Production of full-length RNA(+) rather than of mRNAs is triggered by binding mainly of newly-synthesised viral N (nucleoprotein) but also of P (RdRp minor subunit) proteins to the 5'-leader sequence, somehow causing the RdRp to ignore all termination and polyadenylation signals. The RNA(+) is then used as template for RNA(-) transcription: this also has a 5'-leader, which is also recognised as an assembly origin by N protein. Thus, concomitant genome or anti-genome synthesis and nucleoprotein assembly occur, with a bias for (-) strand synthesis, possibly due to preferential recognition of the (+) strand 3'-origin.
The trisegmented bunyaviruses (see Figure 5) have an interesting "cap stealing" strategy: virion-associated L or replicase protein cleaves cellular mRNAs 12-18 nt from their 5'-ends, and uses the capped leaders to prime transcription of non-polyadenylated mRNA on the three virion L, M and S (-) RNAs. These mRNAs are shorter than the genome segments, as transcription is apparently terminated by hairpin loops. It is not certain how bunyaviruses switch from mRNA to full-length RNA(+) transcription; however, the N protein may act similarly to the way it does in mononegaviruses. Two genera of the Bunyaviridae also have at least one ambisense RNA: phleboviruses and tospoviruses transcribe a mRNA from the 3'-end of the S segment RNA(+); the plant-infecting tospoviruses in addition have an ambisense M RNA, with the extra gene (5'-end of M RNA(-)) being involved in movement functions in plants (Schmaljohn, 1996).
The 8 component orthomyxoviruses are unusual in a number of respects, including having the most segmented genome among ssRNA(-) viruses, and the fact that both transcription and replication occur in the nucleus. Transcription occurs by the same cap-stealing mechanism as for bunyaviruses, but with termination and polyadenylation occurring as for mononegaviruses, with RdRp stuttering at short polyU repeats at the end of genes. The RdRp is also different to those of the other viruses, with 3 viral subunits (PB1-2 and PA). The switch from primer-dependent mRNA synthesis to RNA(+) and RNA(-) synthesis occurs after protein synthesis, possibly due to free NP binding. There is evidence of temporal regulation of expression, with regulatory proteins being made in greatest amounts at early times, and structural proteins later: this is due to selective replication of specific template RNA(-) into mRNA (Lamb and Krug, 1996).
Arenaviruses have a 2-component genome, each segment of which also has a (+) sense ORF at the 5'-end of RNA(-), which is transcribed from RNA(-) as an mRNA. Arenavirus transcription also makes use of some kind of priming, possibly by short capped oligoribonucleotides, and mRNAs are subgenomic and not polyadenylated: transcription is apparently terminated by intergenic stem-loop structures. Transcription occurs from both full-length RNA(-) and RNA(+) templates, with "early" products including the L and N proteins and "late" products including a membrane GP protein and a Z protein. There is a clear switch between transcription and replication, but although the N protein may be involved, this is not proven (Southern, 1996).
Commonalities in expression and replication of ssRNA(-) viruses appear to include distinct transcription and replication functions for the RdRp, probably triggered by binding of the virion nucleoprotein (N or NP) subunits. Thus, both RNA(-) and RNA(+) may be found complexed with N proteins in replication complexes. As for ssRNA(+) viruses, glycoproteins (GPs) are generally expressed as are cellular trans- or outer membrane proteins; that is, they have signal sequences that result in translocation into the rough ER during translation, and are subsequently glycosylated according to signals perceived by the cellular machinery.
While it is tempting to speculate that these viruses are the monophyletic survivors of a pre-DNA dsRNA genome era, the truth is that, although there is very wide diversity among dsRNA viruses, at least some of them may descend from ssRNA(+) viruses. Two distinct groups of dsRNA viruses have polymerase affinities with alpha-like and poty-like viruses respectively (Smart et al., 1999; Gibbs et al., 2000). Thus, the viruses are certainly polyphyletic in origin, and there is almost certainly a wide variety of mechanisms used for expression and replication. However, many of the viruses have not been well studied, so details are lacking.
Reoviruses are the best-studied dsRNA viruses. Representatives of the family infect plants, animals, and insects, and many infect an insect vector as well as an animal or plant alternate host. The viruses all have a double capsid structure, the outer layer of which is stripped off, partly due to proteolysis, during endocytotic entry. Naked core particles in the cytoplasm are able to transcribe capped and non-polyadenylated genome-segment-length monocistronic mRNAs, via an RdRp activity associated with the insides of the hollow spike structures at 5-fold rotational axes of symmetry. These are extruded into the cytoplasm as they are synthesised, and are translated. Viral products accumulate as viroplasms: associations of viral structural and polymerase proteins and mRNAs result in assembly of immature particles, inside which mRNAs are transcribed to give RNA(-) molecules with which they become base-paired. This is the best-characterised example of conservative replication for any organism. New "core" particles also produce mRNAs, but these appear to be largely uncapped (Nibert et al., 1996).
Partitiviruses appear to follow much the same genome expression strategy as reoviruses, in that monocistronic mRNAs are transcribed, which can act as templates for RNA(-) transcription (Strauss et al., 2000).
Birnaviruses have 2-component monocistronic genomes, with 5'-Vpgs, and transcribe genome-length capped mRNAs in virions in the cytoplasm, which then serve as template for the newly synthesised RdRp. Unlike viruses discussed above, one segment (A) encodes a polyprotein, which is cleaved to give virion proteins VP1, VP2 and VP3, while the other segment (B) produces a polymerase with a capping function (Roner, 1999).
Trichomonas vaginalis viruses are unusual among dsRNA viruses in having single-component genomes with multiple ORFs (Bessarab et al., 2000). Details are sketchy, but there are similarities with the larger totiviruses. These virus genomes have two large overlapping ORFs, and express a protein from the 5'-proximal ORF, and a larger fusion protein from both ORFs by means of a ribosomal frameshift. Partitiviruses may derive from totiviruses, as their polymerase sequences show some similarity (Ghabrial, 1998).
A newly characterised group, tentatively named the endornaviruses, were formerly regarded as dsRNA plasmids of plants. They resemble the potyvirus-resembling hypoviruses in lacking particles, but may be transmitted by seed or by grafting, and may have their origin within the alpha-like virus cluster: their 10kb dsRNAs have a single ORF with recognisable helicase and polymerase motif similarities (Gibbs et al., 2000). Presumably these exist as RdRp-associated replicative intermediates and multiply semi-conservatively, like ssRNA(+) viruses.
The processes of assembly of the virions are as varied as their structures; however, there is a logical divide between those with membranes, and those without. The former tend to be considerably more complex than the latter, which may be as simple as a nucleoprotein composed of a single type of protein. There is a commonality between all of the viruses in that their core nucleoproteins assemble either as helices (usually) or as isometric particles (Harrison et al., 1996). This assembly is usually a simple process, but often very specific, and is driven by increasing concentrations of genomic or pre-genomic RNA and of structural protein. Assembly takes place in the cytoplasm for all except the orthomyxoviruses, which assemble nucleoproteins containing N, PB1, PB2, and PA proteins in the nucleus, from where they are exported to the cytoplasm after association of the complexes with the M1 or matrix protein. The interaction of protein and RNA may be promoted by their sequestration in inclusion bodies or viroplasms, which are often associated with elaborations of internal ER-derived membranes.
For some of the simple naked isometric viruses, specific nucleation of assembly at low CP concentration is followed by complete nucleocapsid assembly as CP concentration increases. For picornaviruses, however, there is a complex assembly process. One model of the process involves assembly of a complete RNA-free provirion. This then undergoes autolytic protein cleavage due to its associating with genomic RNA, which is then encapsidated due to a complicated structural reorganisation (see here for an animation). In another model, viral RNA is complexed with smaller protein aggregates, which are then further processed. Reoviruses also have a complex assembly process, starting with the mRNA-protein complex, which becomes a RNAse-sensitive double capsid that does not contain nonstructural (NS) proteins. This synthesises RNA(-) strands, and then undergoes some structural changes to become RNAse-insensitive and to have NS proteins associated with it. Virions may collect in amorphous or paracrystalline arrays inside infected cells: plant viruses especially may accumulate at very high concentrations. Release of such virions may be induced by virus-induced cell lysis, such as is the case with some picornaviruses. However, in most cases release is by cell death followed by membrane degradation.
For most of the more complex virions, such as those of the mononegaviruses and coronaviruses, nucleocapsid or nucleoprotein assembly is followed by association of these with matrix (M) proteins. In the case of rhabdoviruses, soluble M protein appears to condense the loosely helical nucleoprotein aggregate into a more compact form resembling the virion interior (Wagner and Rose, 1996). With this virus and others, membrane-bound M protein binds specifically with nucleoproteins, first to localise the complexes to membrane sites that include the plasmalemma, and internal compartments such as the Golgi apparatus, ER, and elaborations of these. Increasing the number of M protein-nucleoprotein interactions causes recruitment of M proteins, which causes the membrane to fold around them in the start of the act of "budding". Membrane glycoproteins (GP) are an essential part of all enveloped viruses; the cytoplasmic stubs of these also interact with and are recruited by the M proteins (where present) to provide a virus-specific exterior to the budding virion. Virions can bud without glycoproteins in some cases (eg: coronaviruses); however, these are non-infectious. Localised patches of membrane in infected cells may have M and GPs associated with each other, which are then specifically bound by free nucleoprotein complexes.
Bunyaviruses, arenaviruses, and togaviruses do not have M proteins: instead, these virus nucleocapsids have a direct interaction with the cytoplasmic portion of transmembrane GPs. In the case of bunyaviruses the GPs are embedded in intracellular vesicles: cytoplasmic NPs then bud into the vesicles by association with the cytoplasmic portions, to produce enveloped virions within the vesicles.
Release of enveloped virions is a simple consequence of the final act of assembly. When a membrane containing M and/or GPs has completely folded around a nucleoprotein, it produces a vesicle: if this is external to the cell, then it has budded; if it is inside another vesicle, such as a post-Golgi vesicle, then the fusion of this with the plasmalemma in the normal course of cellular vesicle trafficking will result in extracellular budding.
Virions of ortho- and paramyxoviruses have haemagglutinin glycoproteins (HAs) that bind sialyloligosaccharides: as these HAs may contain the same sugars, both viruses also have virion-associated neuraminidases (NAs), which enzymatically destroy the receptors to negate the possibility of virions binding to one another during or after budding (Lamb and Krug, 1996; Lamb and Kolakofsky, 1996).
Plant virus genomes may also move from cell to cell via plasmodesmata, or complex membrane-lined channels that penetrate the cell wall: this is a complex process involving virus-coded MP(s) which specifically bind viral RNA, and in many cases involves transport of a nucleoprotein complex which is not an assembled capsid. It is possible that genomes of plant reoviruses and other dsRNA plant viruses may move as ssRNA(+) nucleoprotein complexes rather than as dsRNA.
Bessarab, I.N., Liu, H.W., Ip, C.F., and Tai, J.H. (2000). The complete cDNA sequence of a type II Trichomonas vaginalis virus. Virology 267, 350-359.
Ghabrial, S.A. (1998). Origin, adaptation and evolutionary pathways of fungal viruses. Virus Genes 16, 119-131.
Gibbs, M.J., Koga, R., Moriyama, H., Pfeiffer, P., and Fukuhara, T. (2000). Phylogenetic analysis of some large double-stranded RNA replicons from plants suggests they evolved from a defective single-stranded RNA virus. J. Gen. Virol. 81, 227-233.
Harrison, H., Wiley, D.C., and Skehel, J.J. (1996). Virus Structure. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 59-100. Lippincott-Raven, New York.
Lamb, R.A. and Kolakofsky, D. (1996). Paramyxoviridae: the Viruses and Their Replication. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 1177-1204. Lippincott-Raven, New York.
Lamb, R.A. and Krug, R.M. (1996). Orthomyxoviridae: the Viruses and Their Replication. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 1353-1396. Lippincott-Raven, New York.
Lewandowski, D.J. and Dawson, W.O. (2000). Functions of the 126- and 183-kDa proteins of tobacco mosaic virus. Virology 271, 90-98.
Melcher, U. (2000). The '30K' superfamily of viral movement proteins. J. Gen. Virol. 81, 257-266.
Murphy, F.A. (1996). Virus Taxonomy. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 15-57. Lippincott-Raven, New York.
Nibert, M.L., Schiff, L.A., and Fields, B.N. (1996). Reoviruses and Their Replication. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 1557-1596. Lippincott-Raven, New York.
Pringle, C.R. (1999). Virus Taxonomy - 1999. Arch.Virol. 144, 421-429.
Roner, M.R. (1999). Rescue systems for dsRNA viruses of higher organisms In: Advances in Virus Research (Maramorosch, K., Murphy, F.A., and Shatkin, A.J., Eds.) 355-367. Academic Press, San Diego.
Rueckert, R.R. (1996). Picornaviridae: the Viruses and Their Replication. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 609-654. Lippincott-Raven, New York.
Schmaljohn, C.S. (1996). Bunyaviridae: the Viruses and Their Replication. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 1447-1472. Lippincott-Raven, New York.
Smart, C.D., Yuan, W., Foglia, R., Nuss, D.L., Fulbright, D.W., and Hillman, B.I. (1999). Cryphonectria hypovirus 3, a virus species in the family hypoviridae with a single open reading frame. Virology 265, 66-73.
Southern, P.J. (1996). Arenaviridae: the Viruses and Their Replication. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 1505-1520. Lippincott-Raven, New York.
Strauss, E.E., Lakshman, D.K., and Tavantzis, S.M. (2000). Molecular characterization of the genome of a partitivirus from the basidiomycete Rhizoctonia solani. J. Gen. Virol. 81, 549-555.
Strauss, E.G., Strauss, J.H., and Levine, A.J. (1996). Virus Evolution. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 153-172. Lippincott-Raven, New York.
Wagner, R.R. and Rose, J.K. (1996). Rhabdoviridae: the Viruses and Their Replication. In: Fields Virology (Fields, B.N., Knipe, D.M., and Howley, P.M., Eds.) 2nd edition. 1121-1136. Lippincott-Raven, New York.
Agol, V.I., Paul, A.V., and Wimmer, E. (1999). Paradoxes of the replication of picornaviral genomes. Virus Res. 62, 129-147.
Gubareva, L.V., Kaiser, L., and Hayden, F.G. (2000). Influenza virus neuraminidase inhibitors. Lancet. 355, 827-835.
Jaspars, E.M. (1999). Genome activation in alfamo- and ilarviruses. Arch.Virol. 144, 843-863.
Neumann, G. and Kawaoka, Y. (1999). Genetic engineering of influenza and other negative-strand RNA viruses containing segmented genomes. Adv.Virus Res. 53:265-300, 265-300.
Portela, A., Zurcher, T., Nieto, A., and Ortin, J. (1999). Replication of orthomyxoviruses. Adv.Virus Res. 54:319-48, 319-348.
Rijnbrand, R.C. and Lemon, S.M. (2000). Internal ribosome entry site-mediated translation in hepatitis C virus replication. Curr.Top.Microbiol.Immunol 242:85-116.
Roberts, A. and Rose, J.K. (1999). Redesign and genetic dissection of the rhabdoviruses. Adv.Virus Res. 53:301-19, 301-319.
Suzuki, R., Suzuki, T., Ishii, K., Matsuura, Y., and Miyamura, T. (1999). Processing and functions of Hepatitis C virus proteins. Intervirology 42, 145-152.
capsid: protein shell made from virus-specified protein, usually contains viral genome
virion: particle containing all or part of the virus genome, always composed of a nucleocapsid, may also have extra shells of protein and/or a host-derived membrane envelope containing viral glycoproteins
nucleoprotein: protein-RNA complex (usually unstructured)
nucleocapsid: capsid protein-RNA complex (usually with a regular structure)
ssRNA: single stranded RNA
dsRNA: double-stranded RNA
ssRNA(+): single-stranded RNA of messenger polarity (can be translated directly)
ssRNA(-): single-stranded RNA of anti-messenger polarity(has to be transcribed to express protein)
ambisense RNA: RNA with both messenger and anti-messenger sense polarity
RdRp: RNA-dependent RNA polymerase (also called replicase)
Legends to Figures:
Pathways of information flow for RNA viruses. Double-stranded (dsRNA) viruses replicate conservatively via a full-length RNA(+) which is transcribed from dsRNA by a virion-associated virus-specific RNA-dependent RNA-polymerase (RdRp). The RNA(+) then acts as mRNA, for synthesis of viral proteins, then as a template for RNA(-) synthesis, to which it base-pairs. Single-stranded (ssRNA) (+)-sense virus genomes initially act as mRNAs, and translate a RdRp component. The RNA(+) then replicates via a full-length RNA(-), which is caught up in replicative complexes and is never free. This is used as template for mRNA transcription if this occurs, by the viral RdRp. Viruses with ssRNA(-) genomes replicate by means of a virion-associated RdRp taken into the cell. This initially acts as a transcriptase, to make subgenomic mRNAs. The genome replicates via transcription of full-length RNA(+). Translation is a cell-specific process; all transcription and replication is done by virus-specific RdRps.
A scheme describing polyprotein translation and processing for picornaviruses (modelled on poliovirus; see Rueckert, 1996). Vpg = 5'-genome-linked protein; polyA = polyadenylate sequence at 3'-end.
Depiction of the expression strategy of the Tobacco mosaic virus (TMV) genome. Red arrows indicate host-dependent translation; blue arrows indicate transcription from an RNA(-) template. Solid boxes are open reading frames (ORFs); these are shown in different colours. Hatched boxes indicate proteins. The 5'-ends of the viral RNA and the mRNAs have a 7-methyl guanosine triphosphate (m7Gppp) cap structure; the 3'-tRNA-like sequence of all RNAs is shown as a cloverleaf structure.
The ORF nearest the 5'-end of genomic RNA is translated into a 126 kDa protein. A "leaky" stop codon allows infrequent translational readthrough (1/10 times) to give a 183 kDa protein product. These two proteins together with (a) host protein(s) constitute the RdRp and replicase. This transcribes a full-length RNA(-) from genomic RNA(+). The RdRp can also transcribe the RNA(-) into RNA(+), and, by recognition of two or more "RNA promoter" sequences within the RNA(-) sequence, into at least two nested subgenomic mRNAs: all products of transcription from RNA(-) share the same 3'-terminus. Only the 5'-proximal ORF of each mRNA is translated.
Depiction of gene order and function and designations in viruses of the order Mononegavirales. The genomic 3'-end is a free -OH group; the 5'-end is phosphorylated but uncapped. The leader sequence is about 50 bases long and conserved in viruses in the same genus. Intergenic sequences are conserved within a virus, and include polyU sequences of 4-7 bases. The different open reading frames (ORFs) are shown in different colours. The gene order is conserved among viruses in the order, as is the function. N/NP are nucleoproteins which bind viral (-) and (+)-sense RNA; NS proteins are non-structural and involved in aspects of regulation; M or matrix protein is bound by assembled nucleocapsids and binds the cytoplasmic portion of the G/GP membrane glycoproteins, which span the cell-derived envelope in the assembled virions; the L protein is the main component of the RdRp, and is incorporated into virions.
Depiction of genome components and expression strategy of bunyaviruses. All viruses have three genomic ssRNA(-) components: these are L, M and S, coding for polymerase (L), glycoproteins (G1, G2) and non-structural (NSm), and nucleoprotein (N) and non-structural (NSs) proteins, respectively. Different open reading frames (ORFs) are shown in diferent colours. Genus-specific ORFs are indicated: only bunyaviruses have an extra NSs (small non-sructural protein) ORF internal to the N ORF; tospo- and phleboviruses have an ambisense (both (+) and (-) sense ORFs) S component, with an mRNA being transcribed off the RNA(+) form of the genome. Tospoviruses in addition have an ambisense M component, with the extra 5'-ORF coding for a host-derived movement protein (MP).