Genus Rotavirus

Type Species	Rotavirus A	(RV-A)

Distinguishing Features

Virus particles have a ‘wheel-like’ appearance when viewed by negative contrast electron microscopy (Fig. 7), from which the genus derives its name. The triple layered capsid encloses a genome of 11 dsRNA segments and is formed in a unique morphogenic pathway involving the acquisition of a transient lipid envelope following budding of immature particles into the endoplasmic reticulum. Viruses only infect vertebrates and are transmitted by a fecal-oral route.

Virion Properties

Morphology

The data from Simian rotavirus A/SA11 (SiRV-A/SA11), represents a paradigm for the other viruses within the genus. The mature infectious virion has an overall diameter of approximately 100  nm and is made up of three concentric protein layers, with no lipid-containing envelope. The detailed topology of these layers and their protein components has been revealed using a combination of cryoelectron microscopy followed by image processing (Fig. 8) and analysis of the virus like particles formed using baculovirus recombinants expressing specific rotavirus structural proteins. The innermost layer composed of VP2, is approximately 51  nm in diameter and 3.5  nm thick. This layer appears to be directly comparable to the internal capsid layer of some other genera of the family Reoviridae (for example the VP3(T2) layer of Bluetongue virus, which is referred to as the orbivirus “subcore”). The rotavirus VP2(T2) layer surrounds the genomic dsRNAs and two structural proteins [VP1(Pol) and VP3(Cap)] that are organized as a series of up to 12 enzymatic complexes, attached to the inner surface of VP2(T2) at the 5-fold axes of symmetry. The assembled genomic RNAs, together with the enzymatic complexes, have a collective diameter of approximately 44  nm and have been referred to as the rotavirus “subcore”. However, this is not regarded as a distinct coherent particle and is not equivalent to the ‘subcores’ of some other genera (see Orbivirus).

The two outer layers of the intact rotavirus particle both have a surface lattice with T = 13 l (laevo) icosahedral symmetry and a uniquely characteristic set of 132 large channels that span both layers, linking the outer surface with the inner most VP2 protein layer. The middle capsid layer is composed of 780 copies of VP6(T13), arranged as 260 trimeric morphological units, positioned at all the local and strict 3-fold axes of the icosahedral lattice. This VP6(T13) layer forms the outer surface of the ‘double layered’ particle (approximately 70.5  nm in diameter) and is directly comparable to one of the capsid layers of viruses from some other genera within the family Reoviridae. For example the ‘core’ particle of the orbiviruses has an outer core layer composed of 780 copies of the orbivirus VP7(T13) protein, which is structurally similar to the VP6(T13) protein of the rotaviruses. The outermost layer of the rotavirus capsid, which is composed of two proteins (VP4 and VP7) and is required for infectivity, has a diameter of approximately 75  nm. The glycoprotein VP7 makes up the surface of the outermost shell and also appears to be arranged in trimers, that interact with the tips of the trimers of VP6(T13). The VP4 protein associates to form 60 dimeric spikes, approximately 20  nm in length which project approximately 12  nm from the surface of the outer shell, giving a final maximum diameter of approximately 100  nm. The VP4 spikes also extend into the particle, so that they must interact with VP7 and VP6 and possibly even have contact with VP2(T2).

Physicochemical and Physical Properties

The triple layered infectious (complete) rotavirus particle (Fig. 7 and 8) has a density of 1.36  g/cm³ in CsCl and sediments at 520-530S in sucrose. Virus infectivity is absolutely dependent upon the presence of the outermost protein layer, the integrity of which requires calcium. This outer layer can be removed by treatment of virions with calcium-chelating agents, such as EGTA or EDTA. Infectivity is stable within the pH range 3-9 and, when stabilized by the presence of 1.5  mM CaCl₂, is not significantly affected by storage for months at 4°C or even 20°C. Infectivity is also relatively thermostable at 50°C but can be destroyed by repeated cycles of freeze-thawing. Infectivity is also generally resistant to fluorocarbon extraction, treatment with solvents such as ether and chloroform, or non-ionic detergents such as deoxycholate, all of which reflect the absence of a lipid containing envelope on the mature particle. However, infectivity is lost by treatment with sodium dodecyl sulphate (0.1%), or a number of disinfectants such as betapropiolactone, chlorine, formalin and phenols, with 95% ethanol, which removes the outer shell, being perhaps the most effective disinfectant against these viruses. Haemaglutinating activity of the intact particle is lost rapidly at 45°C, as a result of freeze-thawing, or through the removal of the VP4 spikes by treatment at pH 10. Some variation has been observed in the physicochemical properties and stability of intact virions of different rotavirus strains. For example, human rotaviruses do not all exhibit haemaglutinating activity and lose the proteins of their outer layer more easily than other strains. In reassortment studies, some of this variation has been attributed to the parental origin of the VP4 present in the virus particle.

Double layered particles are non-infectious (Fig. 8), have a density of 1.38  g/cm³ in CsCl and sediment at 380-400S in sucrose. Single layered particles can be produced by treatment of double-shelled particles with either chaotropic agents such as sodium thiocyanate or high concentrations of CaCl₂. Single layered particles have a density of 1.44  g/cm³ in CsCl and sediment at 280S in sucrose.

Nucleic Acid

The genome of Rotavirus A (RV-A) is packaged within the innermost protein shell of the particle. In each particle the genome is composed of a single copy of 11 discrete segments of dsRNA, which range in size (RV-A isolates) from 3302  bp to 663  bp (Table 5), and have an average combined size of 18,550  bp. RV-A strains (not including the isolates from avian species), normally have a 4:2:3:2 pattern of segments following fractionation of genomic dsRNAs by PAGE. The genome segments are numbered 1-11 in order of increasing mobility during PAGE, although even within RV-A, the migration order of cognate segments, particularly in the segment 7-9 triplet, does vary. In other rotavirus species deviation from the 4:2:3:2 pattern is indicative of a genome segment concatemerisation, in which as much as an additional 1800  bp are packaged into viable virions, in the form of a partial duplication of a genomic segment. The genome RNA sequences are A+U rich (58-67%). The segments are completely base paired and the plus-sense strand contains a 5-terminal cap structure (m⁷GpppG^(m)GPy) but does not have a polyadenylation signal near its 3-end. In contrast to the genus Orthoreovirus, there is no evidence for the presence of single-strand oligonucleotides in the virion. Two levels of terminal sequence conservation are evident in RV-A isolates; firstly all genomic segments share short conserved 5- and 3-termini. The 5-terminal conservation is 10  nts and has a consensus sequence of 5-(GGC^A/_U^A/_UU^A/_UA^A/_U^A/_U . . . )-3 and the 3-terminus has an 8  nts conservation whose consensus is 5-( . . . ^A/_UU^G/_U^U/_G^G/_U^A/_GCC)-3. Immediately internal to these terminal regions at each end of the different segments, there is a second region of conservation of at least 30-40  nts which is genome segment specific. These two levels of conservation may be indicative of cis-acting signals that are important for controlling transcription, replication, and segment selection for packaging. The 5-NTRs vary in length but are all less than 50  nts and in all segments are followed by at least one long ORF after the first AUG. Some gene segments contain additional in-frame (genes 7, 9 and 10) or 2nd frame (gene 11) ORFs but only in the case of genes 9 and 11 are these used to give more than a single primary translational product (i.e., monocistronic) from each genome segment. The 3-NTRs vary in length from 17  nts (segment 1) to 182  nts (segment 10) which in the latter case constitutes ~25% of the total segment length.

Detailed genomic information for rotavirus species RV-B and RV-C is more sparse. The genome is also made up of 11 segments although in all cases the PAGE pattern differs most significantly from RV-A isolates in the absence of the tight triplet of migrating bands (segments 7-9). Sequence information is available for representatives of almost all RV-C (10 out of 11) genome segments but relatively few RV-B segments (4/11). In both cases, the segments have broadly similar properties in terms of length and presence of ORFs.

Proteins

Thirteen primary gene products have been defined in the case of RV-A with two viral genes (9 and 11) each encoding two primary translation products. In the case of gene 9, two initiation codons in the same reading frame are both used, giving largely overlapping forms of the protein product VP7. Gene 11 contains two long ORFs in different reading frames, translation of which gives two unrelated non-structural proteins NSP5 and NSP6 (Table 5).

Several nomenclature systems have been employed for rotavirus proteins but recently one in which proteins are numbered according to their migration rates on SDS-PAGE, starting with the slowest (i.e., highest molecular weight), with structural proteins being given the prefix VP and non-structural proteins the prefix NSP, has become accepted as logical and likely to minimise confusion. An abbreviation to indicate structure/function has been added in brackets to some protein names to facilitate comparisons between different viruses/genera (Table 5). Six structural proteins have been identified and their approximate localization within the mature virus particle defined. The viral core containing the dsRNA genome, has three proteins associated with it, two of which VP1(Pol) and VP3(Cap) are directly associated with the genome whilst the third [VP2(T2)] makes up the core shell, the innermost protein shell of the capsid. VP1(Pol), the largest viral protein at Mr 125 10³, is thought to carry the RNA-dependent RNA polymerase activity which can be assayed in purified virus preparations. VP3(Cap) is a Mr 88 10³ protein which has been shown to carry guanylyl-transferase activity when expressed by recombinant baculoviruses. It is therefore thought to be involved in adding the 5-cap structure present on viral mRNAs. The amounts of each of these proteins in the virion is known to be low (<25 molecules/particle) but has not been measured precisely. VP2(T2) (Mr 94 10³) is the most abundant protein of the viral core with an estimated 120 molecules per virion. It has nucleic acid binding activity although this does not appear to show any sequence specificity. From its deduced amino acid sequence, VP2(T2) contains two leucine zipper motifs, which are thought to be characteristically involved in the dimerisation of nucleic acid binding proteins. This suggests that VP2 may also form a functional dimer.

The middle protein shell of the virion is made up of 780 molecules of VP6(T13) (Mr 41 10³) arranged in 260 trimeric units. The final two proteins of the virion, VP4 (Mr 88 10³) and VP7 (Mr 38 10³) of which there are 120 and 780 molecules per virion respectively make up the outermost shell. The spike protein VP4 (776 aa in most animal strains and 775 aa in most human strains) contains a trypsin cleavage site approximately one third of the way along it. Cleavage of the protein by treatment with protease in-vitro produces two products VP5 (Mr 60 10³) and VP8 (Mr 28 10³) and enhances virus infectivity. VP7 which makes up the surface of the outer shell is formed by N-linked gylcosylation of the primary translation products (vpr7) from genome segment 9. As indicated earlier, segment 9 contains two in frame initiation codons, of which the first is in a weak Kozak consensus and two bands of VP7 are seen following SDS-PAGE analysis of purified virions. Together with biochemical studies, this suggests that these bands arise from post-translational processing of VP7 initiated from each of the two initiation codons. It also implies that the two forms of VP7, with largely overlapping primary sequence, are synthesized and incorporated into virions but at present there is no formal proof of this hypothesis.

Six non-structural proteins are also encoded by the viral genome and much less is known about the functions of these. The largest, NSP1 (Mr 53 10³), is the most variable of all the rotavirus proteins within a single rotavirus species, showing as much as 65% sequence diversity between strains of RV-A. Lower levels of variation are seen in the amino acid sequences of proteins within a single rotavirus species (e.g. <25% variation in VP2(T2) and <45% variation in VP4). However variation in these proteins can also be high between virus species (>87% for VP4 and >84% for VP2). Despite this variation, NSP1 does have a conserved cysteine-rich motif near its amino terminus, which suggests a ‘zinc finger’ metal binding domain. Such domains are present in some nucleic acid binding proteins and NSP1 has been shown to bind both zinc and ssRNA. NSP1 is a component of the pre-core replication intermediate (RI), found in infected cells, suggesting that it has a role in the early stages of virus assembly, which include the process of genome segment selection. However, the precise nature of this role has not yet been defined. NSP2(ViP) (Mr 35 10³) is also found in early RIs. Viruses which contain temperature sensitive mutations in genome segment 8 (encoding NSP2(ViP)), have an RNA-negative phenotype at the non-permissive temperature. These viruses fail to replicate the viral RNA, indicating that NSP2(ViP) has a direct role in the mechanism of virus replication, although this is as yet undefined. NSP2(ViP) also has non-specific RNA binding activity for both ssRNA and dsRNA, with no apparent sequence specificity. NSP3 (Mr 34 10³) is again found in early RIs suggesting a role in viral morphogenesis. It self assembles into multimers and has ssRNA binding activity, which in this case appears to show sequence specificity for a conserved sequence present at the 3-end of viral mRNAs. NSP3 also interacts with the human initiation factor eIF4GI. Current data indicate that by taking the place of poly(A) binding protein present in eIF4GI, NSP3 is responsible for the shut-off of cellular protein synthesis.

NSP4 (Mr 20 10³ and 28 10³) is post-translationally modified by N-linked glycosylation (like the structural protein VP7). However, in contrast to VP7 where the non-glycosylated form of the protein is only observed if inhibitors of N-linked glycosylation are used, the faster migrating non-glycosylated form of NSP4 is observed in normally infected cells. Not all of the protein can be chased into the glycosylated form, indicating that NSP4 exists in two forms that may have different functions. NSP4 has been shown to be involved in the later stages of virion maturation in the ER and recently has been shown to have function as an enterotoxin.

The final two non-structural proteins, NSP5 (Mr 26 10³) and NSP6 (Mr 12 10³), are encoded in two different reading frames of the same viral gene. NSP5 which is rich in serine and threonine is post-translationally modified by being both phosphorylated and O-link glycosylated. It also has RNA binding activity but its function is unknown. The ORF for NSP6 is conserved in most virus isolates examined, although its function is also undefined. Several lower molecular weight polypeptides that are serologically related to NSP5 are also phosphorylated. Recent studies have shown that NSP5 takes multiple forms, as differentially phosphorylated proteins, some of which are autophosphorylated. The kinase domain in NSP5 has not yet been mapped. The highly phosphorylated NSP5 may play a structural role in the organization of viroplasms. NSP5 forms dimers and can interact with NSP2(ViP) and NSP6. NSP5 can bind poly(U) and ssRNA. This binding appears to be enhanced by the presence of NSP2(ViP). The role of NSP5 may therefore be critical for RNA replication. NSP6 is also phosphorylated and localizes in the viroplasms.

Information on the proteins of species other than RV-A is very sparse and is primarily drawn from sequence analysis of viral genes. It is clear that these viruses have homologues of the proteins characterized in RV-A viruses but better systems for routine cultivation in tissue culture are required to facilitate detailed characterization of their proteins.

Lipids

None reported. Although the immature particles acquire a transient membrane during its passage through the ER.

Carbohydrates

Three viral proteins have been shown to be glycosylated. In two cases (VP7 and NSP4) linkage of the sugar is through an amino linkage to asparagine and in the third (NSP5) through an O-linkage to serine and/or threonine.

Genome Organization and Replication

The complete RNA-protein coding assignments have been determined for several RV-A isolates. The coding assignments, for the RV-A Simian SA11 virus strain, are given in Table 5. The replication cycle which is completed in 10-12 hours at 37°C has been studied primarily in continuous cell cultures derived from monkey kidneys. There is little conclusive information about the early steps in the replication cycle. There is increasing evidence that VP4 is the viral attachment protein but the cellular receptor has not been identified. Some rotavirus strains initially attach to the sialic acid residues on the cell surface. Current data suggests that rotaviruses may enter cells by two mechanisms. Viruses can enter cells through receptor mediated endocytosis, but there may also be an alternative mechanism of direct virus entry. In both cases the virus entry process removes the outer virus shell and releases the transcriptionally active double-shelled particle into the cytoplasm of the infected cell. Virion associated enzymes produce 5-capped, non-polyadenylated mRNAs, which are full length transcripts from the minus strand of each of the virion genome segments. Gene expression is regulated by the level of transcription from individual genomic segments, with differences evident in both the kinetics and level of production of different mRNAs. The viral mRNAs derived from each of the genome segments serve two functions, firstly they are translated to generate the viral proteins encoded by the segment and some further control of individual gene expression occurs during this process. For example there is ~250-fold difference in the level of expression between the most (NSP4) and least [VP1(Pol)] abundant protein. Secondly viral mRNAs are also the templates for genome replication. As with other members of the family Reoviridae, it remains unclear whether a given mRNA molecule has the potential to fulfill either role, or if there are two different forms of mRNA from individual genome segments, that can each fulfill only one of these roles. Genome segment assembly takes place by selection of the different viral mRNAs required to form the precore RI and may involve an association with the cytoskeleton. Assembly of the eleven mRNAs is followed by minus strand synthesis, which occurs in ‘core-RI’ and ‘VP6(T13)-RI’, that are present in the ‘viroplasms’ found within the cytoplasm of the infected cell. The next steps in the morphogenesis of progeny virions are unique to rotaviruses and involve the double layered particle budding into the ER in a process that involves NSP4. This results in the particle transiently acquiring an envelope that is lost during the final maturation step(s) when the outer virion shell of VP4 and VP7 is added.

Antigenic Properties

Three viral proteins (VP4, VP6(T13) and VP7) of RV-A have been subjected to detailed antigenic characterization. VP6(T13), which forms the intermediate capsid shell, is both a highly conserved and highly immunogenic protein, carrying both virus group and sub-group determinants. It does not elicit the production of neutralizing antibodies but may play a role in the induction of protective immunity. It is the major target of diagnostic assays for rotaviruses. The outer shell glycoprotein VP7 was the first to be recognized as eliciting a virus type specific neutralizing antibody response and hence has been subjected to extensive molecular and epidemiological analysis. Tissue culture based neutralization assays have allowed the recognition of 14 glycoprotein or G serotypes and where tested in animal cross protection studies, these serotype definitions have been confirmed. Sequencing of genes encoding VP7 from multiple isolates has shown that those falling into the same serotype have <10% sequence variation, whereas between serotypes variation falls in the 15-25% range. VP4 has also been shown to elicit neutralizing antibodies but in this case serotypic definition, is less advanced. Cross neutralization studies have identified 11 proteases or P serotypes, with subtypes being recognized in four of these. As with the gene encoding VP7, there has been extensive epidemiological analysis of the gene encoding VP4. This has allowed the recognition of 19 genotypes in which members falling within a genotype show less than approximately 10% sequence divergence and on the whole there has been correspondance between genotyping and definitive serotyping, but it is important that the latter takes precedence.

In the case of other rotaviruses, although homologues of VP4, VP6(T13) and VP7 have been identified at least in RV-B and RV-C, no information is available at present on the existence or extent of antigenic diversity.

Biological Properties

All rotaviruses have proved difficult to cultivate in-vitro with growth being restricted to a few epithelial cell lines, that have been derived mainly from monkey kidneys. The infection of these cells can be enhanced by pre-treatment of virus with trypsin. This restriction of virus growth in-vitro parallels the in-vivo situation where virus replication is normally restricted to the terminally differentiated enterocytes lining the tips of the microvilli in the small intestine.

There are several mechanisms of pathogenesis, including the destruction of enterocytes that leads to malabsorption and an osmotic diarrhea. Prior to the appearance of histologic changes, a watery diarrhea is often seen and this is thought to be secretory, possibly induced by the action of the rotavirus enterotoxin. Finally it has been proposed that the destruction of these enterocytes causes a loss of the permeability barrier between the gut lumen and the vasculature, resulting in the osmotic pull of fluid from the circulation into the gut and the ‘watery’ characteristic of rotavirus infection. These viruses infect a wide range of avian and mammalian species, with disease being restricted in the great majority of cases to the young. However, RV-B has caused large epidemics in human adults.

List of Species Demarcation Criteria in the Genus

In common with the other genera within the family Reoviridae, it has been agreed that the prime determinant for inclusion of virus isolates within a single virus species will be “ability to exchange (reassort) genome segments during co-infection, thereby exchanging genetic information and generating viable and novel progeny virus strains”. However, data providing direct evidence of segment reassortment between isolates is very limited and serological comparisons, together with comparisons of RNA or protein sequences, represent major factors used to examine the level of similarity that exists between isolates. These and other data can also be used to predict the “compatibility” of strains for reassortment.

The rotaviruses are currently divided into five species (RV-A to E) with two possible additional species (RV-F and RV-G). Viruses within different species are thought to be unlikely, or unable to reassort their genome segments under normal circumstances and each species may therefore represent a separate gene pool.

Members of a single rotavirus species may be identified by:

1.	The ability to exchange genetic material by genome segment reassortment during dual infections, thereby producing viable progeny virus strains.
2.	High levels of serological cross-reaction by ELISA, using either polyclonal sera, or monoclonal antibodies against VP6(T13), or its homologue in groups other than RV-A.
3.	High levels of RNA sequence conservation in terminal and near terminal regions of the genome. Two levels of sequence conservation, namely, short (~10  nts) conserved consensus sequences at the termini of all genomic segments and, internal to these, longer (~40-50  nts) segment specific consenus conservations have been found by terminal RNA fingerprinting. The actual conserved fingerprint patterns are different in each of the virus species and hence can be used as a diagnostic feature for species assignment.
4.	High levels of RNA sequence similarities in “conserved” genome segments. Viruses within the same species will normally show less then approximately 10% sequence variation in genome segment 6 (encoding the major inner structural protein, VP6(T13)). Viruses in different species will normally contain > 30% sequence variation in genome segment 6. These differences are also reflected in the amino acid sequences of the viral proteins, such as VP2(T2) (Fig. 10).
5.	Identification by virus serotype with a virus type already classified within a specific rotavirus species. None of the serotypes within different species will cross neutralize.
6.	Identification of host range. For example RV-E has to date only been found in pigs. RV-D and the putative RV-F and RV-G have only been isolated from avian species.

List of Species in the Genus

Official virus species names are in italics. Tentative virus species names, alternative names ( ), strains or serotypes are not italicized. Virus names, genome sequence accession numbers [ ], and assigned abbreviations ( ) are:

Species in the Genus

Rotavirus A		(RV-A)
Simian rotavirus A/SA11	Seg 1: [X16830], Seg 2: [X16831], Seg 3: [X16062], Seg 4: [X14204], Seg 5: [X14914], Seg 6: [X00421], Seg 7: [X00355], Seg 8: [J02353], Seg 9: [K02028], Seg 10: [KO1138], Seg 11: [X07831]	(SiRV-A/SA11)
Rotavirus B	Seg 3: [X16949], Seg 4: [M91434, V03556], Seg 5: [M55982], Seg 6: [M84456], Seg 9: [M33872, D00911], Seg 11: [M34380, D00912]	(RV-B)
Rotavirus C		(RV-C)
Porcine rotavirus C/Cowden	Seg 1: [M74216], Seg 2: [M74217], Seg 3: [M74218], Seg 4: [M74219], Seg 5: [M29287], Seg 6: [M69115], Seg 7: [X60546], Seg 8: [M61100], Seg 10: [M81488]	(PoRV-C/Co)
Rotavirus D		(RV-D)
Chicken rotavirus D/132		(AvRV-D/132)
Rotavirus E		(RV-E)
Porcine rotavirus E/DC-9		(PoRV-E/DC-9)

Tentative Species in the Genus

Rotavirus F	(RV-F)
Chicken rotavirus F/A4	(AvRV-F/A4)
Rotavirus G	(RV-G)
Chicken rotavirus G/555	(AvRV-G/555)

Note, the cognate genes do not necessarily correspond to the RNA segments with the same number (e.g., PoRV-C/Co segments 5-8 correspond to SiRV-A/SA11 segments 6, 7, 5, and 9, respectively).

Phylogenetic Relationships within the Genus