Abbreviations : Report Help
Alexander V. Karasev, Indranil Dasgupta, Marc Fuchs, Toru Iwanami, Karel Petrzik, Hélène Sanfaçon, Jeremy R. Thompson, Ioannis E. Tzanetakis, René van der Vlugt, Thierry Wetzel and Nobuyuki Yoshikawa
Edited by Nick J. Knowles, Stuart G. Siddell and Peter Simmonds
A summary of this ICTV Report chapter has been published as an ICTV Virus Taxonomy Profile article in the Journal of General Virology, and should be cited when referencing this online chapter as follows:
Thompson, J.R., Dasgupta, I., Fuchs, M., Iwanami, T., Karasev, A.V., Petrzik, K., Sanfaçon, H., Tzanetakis, I. E., van der Vlugt, R., Wetzel, T., Yoshikawa, N., and ICTV Report Consortium. 2017, ICTV Virus Taxonomy Profile: Secoviridae, Journal of General Virology, 98, 529–531.
Members of the family Secoviridae are non-enveloped viruses with mono- or bipartite (RNA-1 and RNA-2) linear positive-sense ssRNA genomes of 9 to 13.7 kilobases in total (Table 1.Secoviridae). Secoviruses are related to picornaviruses and are classified in the order Picornavirales. The majority of known members infect dicotyledonous plants and many are important plant pathogens (e.g. grapevine fanleaf virus and rice tungro spherical virus).
Table 1.Secoviridae. Characteristics of members of the family Secoviridae
cowpea mosaic virus, (RNA-1: X00206; RNA-2: X00729), species Cowpea mosaic virus, genus Comovirus
Non-enveloped 25–30 nm in diameter with icosahedral symmetry
9 to 13.7kb of positive-sense, mono- or bipartite RNA
In association with intracellular membranes derived from the endoplasmic reticulum
Directly from genomic RNA as large polyproteins, which are cleaved by 3C-like proteinases
Plants (mainly dicots), transmitted mainly by insects or nematodes. Some seed transmission demonstrated
Realm Riboviria, order Picornavirales. The family includes 1 subfamily with 3 genera, 5 additional genera and 86 species
Comovirus - (subfamily Comovirinae). Bipartite genome. Comoviruses usually have a narrow host range. Mosaic and mottle symptoms are characteristic. Transmission in nature is exclusively by beetles, especially members of the family Chrysomelidae. Beetles retain their ability to transmit virus for days or weeks.
Fabavirus - (subfamily Comovirinae). Bipartite genome. Fabaviruses have a wide host range among dicotyledonous and some families of monocotyledonous plants. Symptoms are typically ringspots, mottling and wilting. In nature, they are transmitted by aphids in a non-persistent manner.
Nepovirus - (subfamily Comovirinae). Bipartite genome. The genus includes 40 species that are widely distributed in temperate regions. Ringspot symptoms are characteristic. Many nepoviruses are transmitted non-persistently by longidorid nematodes. Seed and/or pollen transmission are also common. In herbaceous plants, the symptoms induced are often transient with a so-called “recovery” phenomenon.
Cheravirus. Bipartite genome. Symptoms are usually mild or absent. Cherry rasp leaf virus is transmitted by nematodes in the field.
Sadwavirus. Includes a single bipartite species, Satsuma dwarf virus, members of which have a wide host range. The natural mode of transmission is unknown.
Torradovirus. Bipartite genome. The torradovirus genome contains an open reading frame (ORF) upstream and partially overlapping the large ORF2 in RNA-2. Some torradoviruses are transmitted by whiteflies in a semi-persistent manner; one torradovirus is transmitted by aphids.
Sequivirus. Monopartite genome. The natural host range of sequiviruses includes plants in several families. Transmission is by aphids in a semi-persistent manner, but is dependent on the presence of a helper virus in the genus Waikavirus.
Waikavirus. Monopartite genome. The natural host range of waikaviruses is usually restricted to plants within a few families. Field transmission is semi-persistent by aphids or leafhoppers. Some waikaviruses are helper viruses for the insect transmission of other viruses, for example, rice tungro spherical virus is the helper virus for rice tungro bacilliform virus (family Caulimoviridae).
Species unassigned to a genus. Bipartite genome. The species Strawberry mottle virus, Black raspberry virus and Chocolate lily virus A are unassigned in the family but members form a divergent monophyletic group with isolates of Satsuma dwarf virus in phylogenetic trees using the conserved Pro-Pol region (Figure 4.Secoviridae). The nature of their capsid protein(s) and their genomic organization are not known. The genome organization of strawberry latent ringspot virus, a member of the species Strawberrry latent ringspot virus, is more related to that of cheraviruses (with the exception of the number of capsid proteins) and it groups more closely with cheraviruses than with sadwaviruses in the phylogenetic trees using the Pro-Pol sequence.
Virions are non-enveloped 25–30 nm in diameter and exhibit icosahedral symmetry (T=1, pseudo T=3, Figure 1.Secoviridae). Many virus preparations contain empty particles. In the case of viruses with a bipartite genome, the two RNAs are encapsidated in separate virions.
Figure 1.Secoviridae. (Top left): Molecular rendering of the cowpea mosaic virus particle. (Top central): Diagrammatic representation of a T=1 lattice. A=Small capsid protein, B=C-terminal domain of the large capsid protein and C=N-terminal domain of the large capsid protein. (Top right): Molecular rendering of the red clover mottle virus particle. (Center): Diagram of the three types of comovirus particles with the B-particle containing one molecule of RNA-1, the M-particle containing one molecule of RNA-2 and the T-particle being empty. (Bottom): Negative-contrast electron micrograph of particles of cowpea mosaic virus. The bar represents 100 nm.
Different classes of virions are distinguished according to their buoyant densities (top, middle and bottom components, also termed T, M and B), (Figure 1.Secoviridae). The main virion components (M and B) contain RNA. Viruses belonging to the genera Sequivirus and Waikavirus, which have a large monopartite genome, sediment with S20W values of 150–190S. For viruses with a bipartite genome, virions containing RNA-1 (B component) sediment at 110–135S. Virions containing RNA-2 (M component) sediment at 84–128S and contain one or two molecules of RNA-2. In cases where the lengths of RNA-1 and RNA-2 are similar, the M and B components may be difficult to separate. Empty shells (T component) sediment with S20W values of 49–63S depending on the virus considered.
The genome consists of one or two molecules of linear positive-sense ssRNA with lengths that differ among genera (Table 2). The genomic RNA(s) contain a 3ʹ-terminal poly(A) tract of variable length. The only known exception is the genomic RNA of a sequivirus (parsnip yellow fleck virus), which is apparently not poly-adenylated. For several comoviruses and nepoviruses, and for strawberry latent ringspot virus, a protein, designated VPg (2–4 kDa) has been shown to be covalently bound at the 5ʹ-end. The presence of a 5ʹ-linked VPg has not been confirmed for other genera but has been suggested because in many cases, infectivity of the RNA(s) has been shown to be protease-sensitive.
Table 2.Secoviridae. Accession numbers and genome content (bases), of representative viruses in the family Secoviridae
cowpea mosaic virus
broad bean wilt virus 2-ME
grapevine fanleaf virus-F13 (Subgroup A)
beet ringspot virus-S (Subgroup B)
Tomato ringspot virus-Raspberry (Subgroup C)
cherry rasp leaf virus-USA
satsuma dwarf virus-S58
tomato torrado virus-PRI-0301
parsnip yellow fleck virus-P121
rice tungro spherical virus-Shen
Unassigned species in the family
strawberry latent ringspot virus-MEN 454
strawberry mottle virus-Thompson
black raspberry necrosis virus- 1
chocolate lily virus A-KP2
Dioscorea mosaic associated virus-goiana
Nepoviruses have a single capsid protein (CP) of 52–60 kDa. Comoviruses, fabaviruses, sadwaviruses and strawberry latent ringspot virus have two CPs of 40–45 kDa and 21–29 kDa. Cheraviruses, torradoviruses, sequiviruses and waikaviruses have three CPs of similar sizes (24–35 kDa, 20–26 kDa and 20–25 kDa). The size and number of the CP(s) of viruses from three unassigned species of the family (Strawberry mottle virus, Black raspberry necrosis virus and Chocolate lily virus A) have not been determined yet. Virions have 60 copies of each CP per particle. For three comoviruses (cowpea mosaic virus, bean pod mottle virus and red clover mottle virus) and three nepoviruses (tobacco ringspot virus, grapevine fanleaf virus and arabis mosaic virus), the atomic structure has been solved and found to be very similar (pseudo T=3) to that of viruses belonging to the family Picornaviridae (Chandrasekar and Johnson 1998, Schellenberger et al., 2011, Lai-Kee-Him et al., 2013, Lin et al., 2000, Chen et al., 1989). Each capsid subunit consists of three beta-barrels (jelly roll domains) that can be present in one large CP with three jelly roll domains (Nepovirus), two CPs (one large CP including two jelly roll domains and one smaller CP with a single jelly roll domain; Comovirus, Fabavirus and Sadwavirus and strawberry latent ringspot virus) or three CPs each containing a single jelly roll domain (Cheravirus, Torradovirus, Sequivirus, Waikavirus) (Figure 2.Secoviridae).
Figure 2.Secoviridae. Architecture of the capsid of members of the family Secoviridae. In sequiviruses, waikaviruses, cheraviruses and torradoviruses, each subunit is composed of three separate small capsid proteins (CPs) each containing a single beta-barrel domain (VP1-VP3, top). In comoviruses, fabaviruses and sadwaviruses, the three beta-barrels are present in two CPs (VPL with two barrels and VPS with a single barrel, middle). In nepoviruses, the single CP is folded in three barrels (bottom).
Unfractionated viral RNA is highly infective. In the case of viruses with a bipartite genome, neither RNA species alone can infect plants systemically. RNA-1 carries all the information required for replication and can replicate in individual cells in the absence of RNA-2 although no virus particles are produced (as demonstrated for comoviruses and nepoviruses).
Viral proteins are usually expressed as large polyproteins, which are cleaved by 3C-like proteinases. Each RNA usually encodes a single polyprotein (Figure 3.Secoviridae). A notable exception is the RNA-2 of torradoviruses, which contains two ORFs. Another exception is the RNA-2 of comoviruses. Although a single large ORF is present, internal initiation at a second AUG allows the formation of two distinct polyproteins. In some cases, extensive regions of sequence identity between RNA-1 and RNA-2 are found in the 5ʹ- and/or 3ʹ-untranslated regions (UTRs) (Figure 3.Secoviridae).
Figure 3.Secoviridae. Genome organization of representative members of the family Secoviridae. Each RNA is shown with the ORFs represented with the boxes. Circles depict VPg molecules covalently attached at the 5′-end of the RNAs. Black circles represent VPgs confirmed experimentally and open circles represent putative VPgs. Poly(A) tails are represented at the 3′-end of the RNAs when present [A(n)]. Red and blue arrows above the sequences represent regions of extensive sequence identity between RNAs 1 and 2. In the latter for torradoviruses this identity is also characterized by conserved indels. Protein domains with conserved motifs for the putative NTP-binding protein (NTB, shown in orange), VPg (purple), proteinase (Pro, yellow), RNA-directed RNA polymerase (Pol, red), movement protein (MP, green) and capsid protein(s) (CP, blue) are shown. The star represents a conserved motif found in the Co-Pro (see comovirus section) protein of comoviruses and in the equivalent protein of other viruses. Proteinase cleavage sites identified experimentally or deduced by sequence comparisons are shown by the solid or dotted vertical lines, respectively. Possible ORFs in the genome of waikaviruses are shown with the dotted squares and putative subgenomic RNAs are shown by dotted arrows below the waikavirus genome. The hatched square for unassigned viruses represents the presence of a variable additional sequence at the C-terminus of the polyprotein. The Nepovirus genus can be divided into subgroups (Sg A, B, C) based on sequence and genome organization.
Within the polyproteins, protein domains are organized in a manner common to that of other members of the order Picornavirales (Figure 3.Secoviridae). The replication block contains domains characteristic of NTP-binding proteins (NTB or putative helicase), 3C-like proteinase (Pro) and RNA-directed RNA polymerase (Pol). In viruses with a monopartite genome, the structural proteins are located upstream of the replication block in the single polyprotein. In viruses with a bipartite genome, structural proteins are contained in the RNA-2-encoded polyprotein. In comoviruses, cheraviruses and nepoviruses, the movement protein is located upstream of the CP(s), and enables viral movement to adjacent cells. Both movement protein and CP(s) are required for cell-to-cell movement of the virus. The movement protein of comoviruses and nepoviruses is a structural component of tubular structures that traverse the cell wall and contain virus-like particles (Laporte et al., 2003, Pouwels et al., 2004). Putative movement proteins have been suggested to be encoded upstream of the CP(s) coding regions for many other viruses in the family but their biological function has not been confirmed.
The RNA-1-encoded 3C proteinase cleaves both RNA-1 and RNA-2-encoded polyproteins. The cleavage site specificity of the proteinase differs with the specific genera (and in the case of nepoviruses it differs with the specific subgroup, Table 3) (Wellink and van Kammen 1988, Gorbalenya et al., 1989, Margis and Pinck 1992, Thole and Hull 1998, Carrier et al., 1999, Ferriol et al., 2016). An amino acid in the substrate-binding pocket of the proteinase interacts directly with the amino acid in the -1 position of the cleavage site and plays a key role in the specificity of the proteinase.
Table 3.Secoviridae. Cleavage site specificity of the 3C-like proteinase of viruses in the family Secoviridae
Proteinase substrate binding pocket#
Dipeptide at cleavage site†
Q/G, Q/M, Q/S, Q/T*, Q/A*
Q/S*, Q/A*, Q/G*
R/G, C/S, C/A, A/S, G/E*, G/V*, C/G*
K/S, K/A, R/A, R/S*, R/G*
Q/G, Q/S, D/S
R/G, T/S, T/N, A/N, A/S, A/A
Q/A, Q/S, Q/V
Q/S*, Q/M*, Q/V*, Q/A*
Unassigned species in the family
Strawberry latent ringspot virus
Strawberry mottle virus
Q/G*, E/G*, Q/S*
Black raspberry necrosis virus
Dioscorea mosaic associated virus
Q/G*, Q/A*, Q/S*, E/G*
Chocolate lily virus A
Formation of replication complexes has been studied for comoviruses and nepoviruses. Replication occurs in association with intracellular membranes derived from the endoplasmic reticulum. Two RNA-1-encoded proteins (the NTB protein and the protein immediately upstream of NTB) interact directly with ER membranes and have been implicated in the proliferation of membrane vesicles in the cytoplasm of infected cells and in the assembly of the replication complex (Sanfacon 2012). This has not been studied for other viruses in the family.
All members of the family infect plants. Host range and symptoms vary with the genera and viruses considered (Table 4). Many viruses in the family have a known biological vector, although some (sequiviruses) require a helper virus and others do not have a known vector. Most viruses are transmissible experimentally by mechanical inoculation. However, waikaviruses are not known to be sap-transmissible. Many viruses are readily transmissible by seed or pollen.
Table 4.Secoviridae. Biological properties of viruses in the family Secoviridae
Seed or pollen transmission
Nematode (most), mite (blackcurrant reversion virus) or unknown
Wide or narrow
Nematode (cherry rasp leaf virus) or unknown
Whitefly or aphid
Aphid (requires helper virus)
Aphid or leafhopper
Strawberry latent ringspot virus
Strawberry mottle virus
Few host species tested
Virus preparations are usually good immunogens and polyclonal antibodies prepared against purified virus particles recognize all CPs. Viruses of species belonging to the same genus can be serologically interrelated, but often distantly.
Seco: derived from the amalgamation of the previous families Sequiviridae and Comoviridae.
Como: from Cowpea mosaic virus, the type species of the Comovirus genus
Faba: derived from the Latin faba, bean; also Vicia faba, broad bean
Nepo: from nematode-transmitted, polyhedral particles
Chera: from Cherry rasp leaf virus, the type species of the Cheravirus genus
Sadwa: from Satsuma dwarf virus, the type species of the Sadwavirus genus
Torrado: derived from Tomato torrado virus, the type species of the Torradovirus genus. In Spanish, torrado means toasted to refer to the severe necrosis (burnt-like phenotype) observed in the disease induced by ToTV.
Sequi: from Latin sequi, to follow, accompany (in reference to the dependent aphid transmission of parsnip yellow fleck virus)
Waika: from Japanese, describing the symptoms induced in rice by infection with rice tungro spherical virus alone (i.e. in the absence of rice tungro bacilliform virus)
There is currently only one subfamily (Comovirinae) which groups together three genera (Comovirus, Fabavirus and Nepovirus). These three genera are closely related to each other in phylogenetic analyses (Figure 4.Secoviridae). Given the presence of only one subfamily, formal demarcation criteria have not been defined.
The criteria demarcating genera in the family are:
Not all criteria may need to be met simultaneously.
Useful criteria to demarcate species are:
Not all criteria need to be met simultaneously. In some cases, sequence information alone can be a good indicator of a distinct species (i.e., when the percentage of sequence identity in both the Pro-Pol and CP(s) regions is well below the proposed cut-off). However, analysing only one region of the genome is generally not sufficient and both the Pro-Pol and CP(s) regions should be considered. In cases where the percentage of sequence identity in one or both sequences is near the proposed cut-off (e.g., between 75 and 85% in the Pro-Pol region or between 70 and 80% in the CP(s) region), other criteria should be considered and information on biological properties of the virus (host range, vector specificity, possibility of reassortment between RNAs) is useful. For example, beet ringspot virus (BRSV) and tomato black ring virus (TBRV) (genus Nepovirus) are closely related in the Pro-Pol sequence (89% sequence identity) but are much more divergent in the CP sequence (62% sequence identity). They differ in their antigenic reactions and also in the specificity of nematode transmission (BRSV is transmitted more efficiently by Longidorus elongatus and TBRV is transmitted more efficiently by Longidorus attenuatus).
Members of the family Secoviridae were previously classified in two different families: Comoviridae (including the genera Comovirus, Fabavirus and Nepovirus) and Sequiviridae (including the genera Sequivirus and Waikavirus) and in two unassigned genera: Cheravirus and Sadwavirus. The families and genera were amalgamated to create the new family Secoviridae, which includes all plant viruses that are members of the order Picornavirales (Sanfacon et al., 2009).
The conserved Pro-Pol region on RNA-1, delineated by the “CG” motif of the 3C-like proteinase and the “GDD” motif of the polymerase, has been used to determine the relationship among members of the order Picornavirales. Comparison of the Pro-Pol sequence among members of the family Secoviridae allows the definition of branches that generally correspond to the distinct genera. Members of the sub-family Comovirinae (genera Comovirus, Fabavirus and Nepovirus) are more closely related to each other than to other genera within the family (Figure 4.Secoviridae). Within this sub-family, fabaviruses and comoviruses are more closely related to each other than to nepoviruses. Nepovirus subgroups (see nepovirus section) are not clearly separated in the Pro-Pol tree (with the exception of subgroup B which constitutes a separate branch) but are more clearly separated in phylogenetic trees using the CP sequence (not shown).
Figure 4.Secoviridae Maximum likelihood inferred phylogenetic tree of members of the family Secoviridae based on an alignment of amino acid sequences of the conserved domains between the “CG” motif of the 3C-proteinase and the “GDD” motif of the polymerase (Pro-Pol region) using T-Coffee (Di Tommaso et al., 2011) . The tree was generated with PhyML (Guindon et al., 2010) (1000 bootstrap replicates) in the TOPALi suite (Milne et al., 2009) using an RtRev (Dimmic et al., 2002) +I+G evolutionary model selected by Protest (Darriba et al., 2011). Results are presented as a midpoint rooted tree. The bar represents the genetic distance. The nepovirus subgroups A, B and C are indicated along with genus and subfamily assignments. Bootstrap values > 70% are shown. This phylogenetic tree and corresponding sequence alignment are available to download from the Resources page.
Members of the family Secoviridae are related to members of other families in the order Picornavirales (Thompson et al., 2014). They all share a common virion structure, organization of the replication block within the polyproteins and conserved properties of the replication proteins, including the 3C-like proteinase. Members of the Secoviridae are also related to members of the families Potyviridae and Caliciviridae in some aspects (common replication block, polyprotein strategy, VPg bound to the 5ʹ-end of the genome and poly(A) tail at the 3ʹ-end of the genome) but differ in other properties.
Support for preparation of the Online Report and Report Summaries has been provided by: