A Novel Mycovirus That Is Related to the Human Pathogen Hepatitis E Virus and Rubi-Like Viruses

Abstract

Previously, we reported that three double-stranded RNA (dsRNA) segments, designated L-, M-, and S-dsRNAs, were detected in Sclerotinia sclerotiorum strain Ep-1PN. Of these, the M-dsRNA segment was derived from the genomic RNA of a potexvirus-like positive-strand RNA virus, Sclerotinia sclerotiorum debilitation-associated RNA virus. Here, we present the complete nucleotide sequence of the L-dsRNA, which is 6,043 nucleotides in length, excluding the poly(A) tail. Sequence analysis revealed the presence of a single open reading frame (nucleotide positions 42 to 5936) that encodes a protein with significant similarity to the replicases of the “alphavirus-like” supergroup of positive-strand RNA viruses. A sequence comparison of the L-dsRNA-encoded putative replicase protein containing conserved methyltransferase, helicase, and RNA-dependent RNA polymerase motifs showed that it has significant sequence similarity to the replicase of Hepatitis E virus, a virus infecting humans. Furthermore, we present convincing evidence that the virus-like L-dsRNA could replicate independently with only a slight impact on growth and virulence of its host. Our results suggest that the L-dsRNA from strain Ep-1PN is derived from the genomic RNA of a positive-strand RNA virus, which we named Sclerotinia sclerotiorum RNA virus L (SsRV-L). As far as we know, this is the first report of a positive-strand RNA mycovirus that is related to a human virus. Phylogenetic and sequence analyses of the conserved motifs of the RNA replicase of SsRV-L showed that it clustered with the rubi-like viruses and that it is related to the plant clostero-, beny- and tobamoviruses and to the insect omegatetraviruses. Considering the fact that these related alphavirus-like positive-strand RNA viruses infect a wide variety of organisms, these findings suggest that the ancestral positive-strand RNA viruses might be of ancient origin and/or they might have radiated horizontally among vertebrates, insects, plants, and fungi.

Sclerotinia sclerotiorum is a destructive soilborne plant pathogenic fungus with a wide host range that includes more than 450 species and subspecies among 64 genera of plants (6). As an important and unique plant pathogenic fungus, the sequence of the whole genomic DNA of S. sclerotiorum has been determined (http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi?organism=fungi). Double-stranded RNA (dsRNA)-associated hypovirulence in S. sclerotiorum was first reported for strain 91 (5) and later reported for strain Ep-1PN (36). In strain Ep-1PN, three dsRNA segments, designated L-, M-, and S-dsRNAs, with estimated sizes of 7.4, 6.4, and 1.0 kbp, respectively, were associated with hypovirulence of S. sclerotiorum (36). Of these three dsRNA segments, only M-dsRNA was consistently detected in association with the hypovirulence phenotype (36). Evidence was recently presented that the M-dsRNA was derived from the genomic RNA of a positive-strand RNA virus, Sclerotinia sclerotiorum debilitation-associated RNA virus (SsDRV) (66). Furthermore, sequence analysis of the S-dsRNA segment showed that it is a defective RNA derived from SsDRV. The L-dsRNA segment may represent a novel mycovirus different from SsDRV, since it lacks sequence similarity to SsDRV as determined by Northern hybridization analysis (D. Jiang, unpublished data).

Discovery of novel mycoviruses may expand our knowledge of global ecology and evolution of viruses. Although mycoviruses typically have isometric particles and dsRNA genomes (e.g., members of the families Totiviridae, Chrysoviridae, and Partitiviridae) (14), viruses in these families also infect organisms other than fungi. Whereas some members of the family Totiviridae infect protozoa, a number of the viruses in the families Partitiviridae and Chrysoviridae infect plants. The mycoreoviruses from hypovirulent strains of Cryphonectria parasitica and Rosellinia necatrix represent a distinct group of dsRNA mycoviruses with reovirus-like particle morphology, and they are most closely related to the tick-borne animal pathogens belonging to the genus Coltivirus in the family Reoviridae (19, 52). Viruses with dsRNA genomes infect a broad range of hosts (vertebrates, invertebrates, fungi, plants, protozoa, and bacteria) and are grouped in six families of dsRNA viruses: Totiviridae, Birnaviridae, Partitiviridae, Cystoviridae, Chrysoviridae, and Reoviridae (10). A comparative analysis of the amino acid sequences of proteins encoded by dsRNA viruses revealed little similarity between viruses of different genera, even between those belonging to the same family, e.g., those belonging to the family Reoviridae. Even though the RNA-dependent RNA polymerase (RdRp) genes are the most highly conserved genes among RNA viruses, phylogenetic analysis of the RdRps suggests a polyphyletic origin for dsRNA viruses. The dsRNA viral RdRps tend to group with different supergroups of the positive-strand RNA viruses (29). The concept of multiple origins of dsRNA viruses from diverse lineages of positive-strand RNA viruses is presently well accepted (1, 29).

Recently, there has been an increasing number of reports of positive-strand RNA mycoviruses whose RdRp and helicase gene lineages are within the lineages of positive-strand RNA plant viruses, e.g., the potexvirus-like mycoviruses FgV-DK21 in Fusarium graminearum (33), Botrytis virus X (22), and Oyster mushroom spherical virus (64). Many of these positive-strand RNA mycoviruses do not encode coat proteins, and they occur in their hosts as dsRNA derivatives of their genomic positive-strand RNAs but are phylogenetically related to plant viruses. The mycoviruses that lack typical virions include members of the genus Hypovirus that infect C. parasitica, with lineage to plant potyviruses (44); SsDRV, an unassigned mycovirus from S. sclerotiorum (66), which is related to allexiviruses in the family Flexiviridae, and Diaporthe ambigua RNA virus (DaRV), with lineage to tombusviruses (47). Mitoviruses that infect C. parasitica (46), Ophiostoma novo-ulmi (20), and Botrytis cinerea (65) are phylogenetically related to positive-strand RNA bacteriophages in the family Leviviridae. Considering the fact that these related positive-strand RNA viruses infect a wide variety of organisms, the ancestral RNA virus might be of ancient origin and/or might have spread out horizontally among animals, plants, fungi, protozoa, and prokaryotes.

Some mycoviruses are associated with debilitation/hypovirulence of their hosts, and these mycoviruses are potential biocontrol agents to combat plant fungal diseases and to probe the pathogenicity of hosts on the molecular level (44). Of the debilitation/hypovirulence-associated mycoviruses, the hypovirus/C. parasitica system has been the most thoroughly studied. Significant insight has been gained into the molecular basis of hypovirulence in this system and its potential implementation for biological control of chestnut blight (24, 38, 44, 55). The depth of knowledge gained from studying the hypovirus/C. parasitica system should now pave the way for investigations of other similar fungal virus systems.

In a recent study, Li et al. (37) identified a small number of genes whose expression was downregulated in the virus-infected S. sclerotiorum strain Ep-1PN and discussed the probability that the predicted depleted levels of the corresponding proteins may contribute to the characteristic debilitation and hypovirulence of this strain. In the present study, molecular cloning and sequencing of the L-dsRNA segment from a debilitated fungal strain were carried out, and the sequences generated were assembled and subjected to sequence and phylogenetic analyses to determine whether the L-dsRNA is related to previously characterized mycoviruses and to examine its relationships to viruses infecting organisms other than fungi.

MATERIALS AND METHODS

Fungal strains.

S. sclerotiorum hypovirulent strain Ep-1PN, which contained L-dsRNA, M-dsRNA (SsDRV), and S-dsRNA, was originally isolated from a sclerotium collected from a diseased eggplant (35). Strain Ep-1PNA367, a virus-free strain, was a single-ascospore isolate derived from Ep-1PN. Strains Ep-1PNSA-8, Ep-1PNSA-23, and Ep-1PNSA-34 were isolated from individual sclerotia of strain Ep-1PN. All fungal strains were grown at 18 to 22°C on potato dextrose agar medium (PDA) and stored on PDA slants at 4 to 8°C.

Extraction of dsRNA.

The procedure for dsRNA extraction, previously described by De Paulo and Powell (7), was used with minor modifications. For the extraction of dsRNA from mycelia of strains Ep-1PNSA-8, Ep-1PNSA-23, Ep-1PNSA-34, and Ep-1PNA367, small agar mycelial plugs were placed on cellophane membranes placed on top of PDA (CM-PDA) in petri plates for 2 days, and the mycelia were then harvested from the cellophane membranes. To extract the dsRNA from strain Ep-1PN, the mycelium growing on CM-PDA for up to 1 week was harvested and then homogenized in a sterilized mortar with a pestle. The homogenate was spread on fresh CM-PDA plates for 2 days, and then the mycelium was harvested and stored at −80°C.

cDNA synthesis, molecular cloning, and sequencing.

To obtain sequence information for the L-dsRNA, dsRNA (1.0 μg) was mixed with 0.1 μg random hexamer primers and 3 μl 100% dimethyl sulfoxide, and diethyl pyrocarbonate-treated double-distilled H₂O was added to a final volume of 12 μl. The mixture was heated to 95 to 98°C for 14 min and chilled on ice for 3 min. First- and second-strand cDNAs were synthesized as described by Sambrook et al. (51). The resulting cDNA was purified by filtration through a Sephadex G-50 column and A-tailed with Taq DNA polymerase and deoxynucleoside triphosphate at 72°C for 30 min. The A-tailed ds cDNA was ligated into the pMD18-T vector according to the manufacturer's instructions (TaKaRa) and transformed into competent cells of Escherichia coli JM109. Sequence-specific primers were used for reverse transcription (RT)-PCR to amplify parts of the genome which were not cloned by the initial random cDNA synthesis. Denatured dsRNA was reverse transcribed using RevertAid Moloney murine leukemia virus (M-MuLV) reverse transcriptase (Fermentas) and a sequence-specific reverse primer and incubated for 60 min at 45°C. After RT, the mixture was treated with RNase H (1 U at 37°C for 30 min; TaKaRa), and 2% of the reaction volume was used for PCR amplification with the pertinent forward and reverse primers, GC buffer, and LA Taq DNA polymerase (TaKaRa). The resulting PCR product was fractionated by electrophoresis on 1% agarose gel and purified using a gel extraction kit (Axygen). The PCR product was cloned into the pMD18-T vector.

Clones for the terminal sequences of the dsRNA were generated by T4 RNA ligase oligonucleotide-mediated amplification as described by Lambden et al. (34). The 3′ terminus of each strand of dsRNA was ligated at 5 to 15°C for 16 to 18 h with the 5′-end phosphorylated oligonucleotide 5′-GCATTGCATCATGATCGATCGAATTCTTTAGTGAGGGTTAATTGCC-(NH₂)-3′ using T4 RNA ligase (Fermentas). The oligonucleotide-ligated dsRNA was denatured and used for the RT reaction with RevertAid M-MuLV reverse transcriptase and 10 pmol of a primer with a sequence complementary to that of the oligonucleotide used for the RNA ligation (oligoREV, 5′-GGCAATTAACCCTCACTAAAG-3′). The reaction product was treated with RNase H, as described above, and the cDNA was amplified with another primer complementary to the RNA ligation oligonucleotide (5′-TCACTAAAGAATTCGATCGATC-3′) and the sequence-specific primer corresponding to the 5′- and 3′-terminal sequences of the dsRNA, respectively. The expected PCR products were recovered and purified with a gel extraction kit (Axygen) and cloned into the pMD18-T vector (TaKaRa).

Sequencing was carried out by the dideoxynucleotide termination method using a Big Dye terminator sequencing kit (BigDye terminator v. 2.0; ABI) and an ABI Prism 377-96 automated sequencer (Beijing Sunbiotech). M13 universal primers or sequence-specific primers were used for sequencing, and each base was determined by sequencing at least two independent clones (usually three to five clones) from both orientations.

Sequence and phylogenetic analyses.

Viruses selected for phylogenetic analysis are given in Table 1. The DNAMAN version 5.2.9 (Lynnon Biosoft) software package was used for sequence annotations, including nucleotide statistics and open reading frame (ORF) searching. Sequence similarity searches of GenBank, Swiss-Prot, and EMBL databases were conducted using the BLAST program (3). Searches for amino acid signatures and protein motifs were conducted using the programs included in the ExPASy proteomics tools (http://www.expasy.org/tools/). Multiple alignments of amino acids were made with the program MUSCLE version 3.6 (9), and the resulting alignment was manually adjusted according to Koonin's alignments (28). Two independent methods for the generation of tentative phylogenetic trees were used, namely the neighbor-joining (NJ) algorithm and the maximum likelihood (ML) method. The NJ algorithm was performed using PAUP* 4.0b10 (Sinauer Associates, Sunderland, MA), assuming the BLOSUM 62 matrix (18). Bootstrap values were calculated from 1,000 bootstrap replicates. The ML method was performed using the program TREE-PUZZLE version 5.2 (54). Likelihoods were calculated using the variable time model of amino acid substitution (42) and the relevant parameter values estimated from the data (available upon request).

TABLE 1.

Viruses selected for phylogenetic analysis

Family and genus	Speciesa	Abbreviation	Accession no.	Positions for phylogenetic analysis (aa)
				RdRp	Helicase
Unassigned
Hepevirus	Hepatitis E virus Yam67	HEV-Yam67	AAM66329	1361-1616
	Swine hepatitis E virus	HEV-swine	ABB88699	1373-1628	969-1199
	Avian hepatitis E virus*	AHEV	AAS45830	1201-1457	800-1029
	Hepatitis E virus JRA1	HEV-JRA1	BAB93536		971-1201
Closteroviridae
Closterovirus	Beet yellows virus	BYV	AAC25115	2701-2963	2231-2519
	Mint virus 1*	MV1	YP_224091	90-354
			YP_224291		1575-1857
Ampelovirus	Plum bark necrosis and stem pitting-associated virus*	PBNSTaV	YP_001552324	170-434
			YP_001552323		2029-2312
	Little cherry virus 2	LChV-2	AAP87784	176-440
			AAP87783		1327-1609
Unassigned	Mint vein banding virus*	MVBV	AAS57939	171-435
Unassigned
Tobamovirus	Tobacco mosaic virus	TMV	AAD44327	1285-1548	815-1087
	Pepper mild mottle virus	PMMoV	CAC59955	1285-1548	815-1087
	Rehmannia mosaic virus*	ReMV	YP_001041889	1285-1548	815-1087
Tetraviridae
Omegatetravirus	Helicoverpa armigera stunt virus	HaSV	NP_049235	1030-1288	593-851
	Dendrolimus punctatus tetravirus*	DpTV	YP_025094	1021-1279	584-842
Betatetravirus	Nudaurelia capensis beta virus	NβV	NP_048059	981-1232	548-805
Togaviridae
Alphavirus	Sindbis virus	SINV	NP_740669	274-537
			NP_062888		708-969
	Semliki Forest virus	SFV	NP_740668	269-532
			NP_463457		705-963
	Aura virus	AURAV	NP_819013	268-531
Rubivirus	Rubella virus	RUBV	NP_062883	1775-2035	1333-1578
Unassigned
Endornavirus	Oryza sativa endornavirus	OsEV	YP_438200	4219-4482	793-1055
	Oryza rufipogon endornavirus	OrEV	YP_438202	4275-4537	1548-1811
	Vicia faba endornavirus	VfEV	YP_438201	5375-5637	1947-2212
	Phytophthora endornavirus 1*	PEV1	YP_241110	4259-4522
	Gremmeniella abietina type B RNA virus XL*	GaBRV-XL	YP_529670	3052-3308
Tymoviridae
Tymovirus	Turnip yellow mosaic virus	TYMV	NP_663297	1478-1729	958-1207
Marafivirus	Maize rayado fino virus	MRFV	NP_734076	218-468
Maculavirus	Grapevine fleck virus	GFkV	NP_542612	1588-1837	1050-1279
Flexiviridae
Potexvirus	Potato virus X	PVX	NP_040882	1140-1397	717-965
Allexivirus	Shallot virus X	ShVX	NP_620648	1319-1569	910-1144
Capillovirus	Apple stem grooving virus	ASGV	NP_044335	1262-1517
Carlavirus	Potato virus M	PVM	NP_705707	183-436
Foveavirus	Apple stem pitting virus	ASPV	NP_604464	1867-2117	1356-1635
Mandarivirus	Indian citrus ringspot virus	ICRSV	NP_203553	1341-1598
Trichovirus	Apple chlorotic leaf spot virus	ACLSV	NP_040551	1539-1789	1040-1307
Vitivirus	Grapevine virus A	GVA	NP_619662	1375-1625
Luteoviridae
Luteovirus	Soybean dwarf virus	SbDV	NP_150431	484-757
Unassigned
Umbravirus	Tobacco bushy top virus	TBTV	NP_733848	124-401
Unassigned
Benyvirus	Beet necrotic yellow vein virus F2, France	BNYVV-B-Eur	CAA28795	1737-1997	927-1182
	Beet soilborne mosaic virus	BSBMV	NP_612601	1745-2005	929-1184
Coronaviridae
Coronavirus	Infectious bronchitis virus	IBV	NP_066134		5132-5447
Unassigned
Mycovirus	Sclerotinia sclerotiorum debilitation-associated RNA virus	SsDRV	YP_325662	1273-1530
	Sclerotinia sclerotiorum RNA virus L	SsRV-L	ACE88957	1275-1530	640-878

^aTentative members of the designated genera are indicated by asterisks, and their names are not italicized.

Northern blot hybridization.

Northern hybridization analysis was performed as previously described (26). To verify the authenticity of the cDNA clones generated with the purified dsRNA, the cDNA clones were labeled with [³²P]dCTP using a radiolabeling kit (TaKaRa) and used to probe the RNA blot.

RT-PCR.

Total RNA from isolates derived from the debilitated strain Ep-1PN was isolated according to the method described by Sambrook et al. (51). First-strand cDNA was synthesized using RevertAid M-MuLV reverse transcriptase (Fermentas). To detect M-dsRNA, the reverse primer SsDRV-PCRpREV (5′-CAGTCCCTAGTTTCATCTCGTTCC-3′) was used, and the first-strand cDNA was then subjected to PCR using the SsDRV-PCRpREV primer and the forward primer SsDRV-PCRpFOR (5′-TGCAGGAAACAGTCATGGCAAC-3′), with a predicted size of 871 bp for the amplicon. To detect L-dsRNA, the reverse primer Ss-7.4RP (5′-GAAGCCACAGGGACAGCAAG-3′) was used, and the first-strand cDNA was then subjected to PCR using the Ss-7.4RP primer and the forward primer Ss7.4-FP (5′-CCACCGACGCAGGCAAATAC-3′), with a predicted size of 721 bp for the amplicon. The conditions for cDNA amplification included an initial denaturation step of 4 min at 94°C, followed by 30 cycles of 30 s at 94°C, 30 s at 61°C, and 1 min at 72°C, with a final elongation step of 10 min at 72°C. PCR products were fractionated by gel electrophoresis on 1% agarose gels and stained with ethidium bromide.

Mycelial growth and virulence test.

To evaluate the effect of Sclerotinia sclerotiorum RNA virus L (SsRV-L) on colony morphology and the virulence of S. sclerotiorum to rapeseed, SsRV-L-infected strains were compared with the original strain Ep-1PN and virus-free ascospore offspring of strain Ep-1PN using the procedures described by Li et al. (35).

Nucleotide sequence accession number.

The sequence for the full-length L-dsRNA cDNA was deposited in GenBank under accession no. EU779934.

RESULTS

Synthesis and sequencing of cDNA from L-dsRNA.

The L-dsRNA segment extracted from mycelia of strain Ep-1PN was electrophoretically separated on a 1% agarose gel, purified with a gel extraction kit, and then subjected to cDNA synthesis using random primers (hexamer). The ds cDNA fragments were cloned, and the cloned cDNAs were transformed into E. coli strain JM109. More than 40 cDNA clones with inserts of 200 to 800 bp were obtained and confirmed to be derived from the L-dsRNA by using reverse Northern dot blot hybridization analysis (data not shown). Among these, 10 randomly selected cDNA clones were sequenced in both directions (Fig. 1A, clones A to J). RT-PCR was conducted to fill the gaps between clones with specific primers designed based on these cDNA sequences, and an oligoligation strategy was used to obtain the sequences of the 5′ and 3′ termini. A total of 26 clones were obtained and sequenced (Fig. 1A). Computer-assisted sequence assembly showed that the full-length L-dsRNA cDNA is 6,043 bp in length, excluding the poly(A) tail. The cloning strategy for dsRNA is outlined in Fig. 1A. The sequence was deposited in GenBank.

FIG. 1.

Schematic representation of the strategy used in cDNA cloning of SsRV-L dsRNA and predicted genome organization. (A) cDNA clones (A to J; red lines) were synthesized using random hexamer primers and denatured dsRNA as a template. Sequences of the regions of the dsRNA that were not covered by these cDNA clones were obtained from cloned RT-PCR products using sequence-specific primers (designed based on the sequences of these cDNA clones), and clones corresponding to 5′ and 3′ termini were amplified using the method described by Lambden et al. (34) (clones 1 to 26; black lines). (B) Diagrammatic representation of the genomic organization of SsRV-L dsRNA showing the presence of a single ORF. The ORF encodes a putative protein containing a methyltransferase domain, a helicase domain typical of superfamily 1 of viral RNA helicases, and eight conserved motifs characteristic of RdRps of positive-strand RNA viruses. UTR, untranslated region.

Sequence analysis of the complete cDNA of the L-dsRNA.

Analysis of the complete cDNA sequence revealed the presence of a single large ORF (nucleotides 42 to 5936), potentially encoding a polypeptide of 1,964 amino acid residues with a predicted mass of 213.4 kDa (Fig. 1B). Motif Scan (http://myhits.isb-sib.ch/cgi-bin/motif_scan) searches showed that this protein contains conserved methyltransferase, viral RNA helicase, and RdRp domains characteristic of the replicases of many of the positive-strand RNA viruses in the alphavirus-like supergroup of viruses (Fig. 1B). The 5′ noncoding region consists of 41 nucleotides and could be folded into a hairpin structure, and the 3′ noncoding region consists of 107 nucleotides and could be folded into three independent hairpin structures (see Fig. S1 in the supplemental material). Such hairpin structures are also found in other viruses in the alphavirus-like supergroup (15, 17, 41, 43, 58). Homology searches of the methyltransferase, helicase, and RdRp conserved motifs of the L-dsRNA-encoded replicase indicated that they are related to several animal viruses belonging to the genus Hepevirus, including Hepatitis E virus (HEV), Swine hepatitis E virus (HEV-swine) and Avian hepatitis E virus (AHEV) (Fig. 2). BLASTP database searches of the L-dsRNA RdRp conserved domain revealed that it shares significant sequence similarity (E values of 3e⁻¹¹ or lower) with the RdRp encoded by the hepeviruses and by the closteroviruses. The RdRp domain of L-dsRNA shares 27% identity and 43% similarity with HEV and 28% identity and 44% similarity with HEV-swine (Table 2). These L-dsRNA RdRp conserved motifs are also related to those of plant viruses in the genus Closterovirus. The RdRp conserved motifs of SsRV-L share 23% identity and 40% similarity with Mint virus 1. Likewise, the identity and similarity scores between SsRV-L and Plum bark necrosis virus and Stem pitting-associated virus are 24% and 41%, respectively. The corresponding scores for Mint vein banding virus are 24% and 40% (Table 2). Thus, these results suggest that L-dsRNA probably represents the replicative form or a replicative intermediate of the genomic RNA of a positive-strand mycovirus coinfecting the debilitated S. sclerotiorum strain Ep-1PN along with SsDRV. The newly characterized virus coinfecting strain Ep-1PN was designated Sclerotinia sclerotiorum RNA virus L (SsRV-L).

FIG. 2.

Amino acid sequence alignment of the putative methyltransferase (A), helicase (B), and RdRp (C) motifs of SsRV-L and those of selected viruses in the genus Hepevirus. The positions of the conserved motifs in these motifs (shaded areas) correspond to those previously described (28, 31, 49) and are indicated with horizontal lines above the shaded areas. Asterisks indicate identical amino acid residues, and colons indicate similar residues. Numbers in brackets refer to the amino acid position in the ORF. See Table 1 for abbreviations of virus names and viral protein accession numbers.

TABLE 2.

Percent amino acid sequence identity and similarity between the methyltransferase, helicase, and RdRp motifs of SsRV-L and those of other selected alphavirus-like positive-strand RNA viruses

Motif	Genus	Virusa	Abbreviation	Accession no.	Identity (%)b	Similarity (%)b	Reference
Methyltransferase	Hepevirus	Avian hepatitis E virus*	AHEV	AAS45830	63/241 (26)	93/241 (38)	23
	Tobamovirus	Tomato mosaic virus	ToMV	CAD10425	57/240 (23)	97/240 (40)	Unpublished
		Tobacco mosaic virus	TMV	AAD44327	57/240 (23)	97/240 (40)	56
		Pepper mild mottle virus	PMMoV	CAC59955	56/237 (23)	98/237 (41)	62
		Cucumber green mottle mosaic virus	CGMMV	BAA87620	54/217 (24)	92/217 (42)	60
		Rehmannia mosaic virus*	ReMV	YP_001041889	56/240 (23)	95/240 (39)	67
	Betatetravirus	Nudaurelia capensis beta virus	NβV	NP_048059	63/256 (24)	99/256 (38)	16
RNA helicase	Hepevirus	Avian hepatitis E virus*	AHEV	AAS45830	67/228 (29)	94/228 (41)	23
		Hepatitis E virus JRA1	HEV-JRA1	BAB93536	66/229 (28)	96/229 (41)	59
		Swine hepatitis E virus	HEV- swine	ABB88699	82/284 (28)	122/284 (42)	Unpublished
	Omegatetravirus	Helicoverpa armigera stunt virus	HaSV	NP_049235	68/240 (28)	104/240 (43)	15
		Dendrolimus punctatus tetravirus*	DpTV	YP_025094	66/241 (27)	106/241 (43)	63
RdRP	Hepevirus	Hepatitis E virus Yam67	HEV-Yam67	AAM66329	68/250 (27)	108/250 (43)	25
		Swine hepatitis E virus	HEV- swine	ABB88699	70/245 (28)	108/245 (44)	Unpublished
	Closterovirus	Beet yellows virus	BYV	AAC25115	79/335 (23)	130/335 (38)	45
		Mint virus 1*	MV1	YP_224091	65/276 (23)	111/276 (40)	61
		Grapevine leafroll-associated virus 2	GLRAV-2	AAC40856	65/280 (23)	110/280 (39)	68
		Grapevine rootstock stem lesion-associated virus*	GRSLAV	NP_835337	66/280 (23)	112/280 (40)	Unpublished
		Citrus tristeza virus	CTV	AAO12717	70/284 (24)	112/284 (39)	53
	Unassigned Closteroviridae	Mint vein banding virus*	MVBV	AAS57939	92/384 (23)	152/384 (39)	61
	Ampelovirus	Plum bark necrosis and stem pitting-associated virus*	PBNSTaV	YP_001552324	84/346 (24)	144/346 (41)	2
		Little cherry virus 2	LChV-2	AAP87784	75/308 (24)	124/308 (40)	50

^aTentative members of the designated genera are indicated by asterisks, and their names are not italicized.

^bThe number of identical or similar amino acid residues/total number of residues included in the sequence comparison analysis is shown. Values in parentheses are the identity or similarity percentage scores as calculated by the BLAST program.

Sequence comparison of the viral RNA helicase motifs of SsRV-L showed that they share significant sequence similarity with viruses in the genera Hepevirus and Omegatetravirus (E values of 4e⁻⁶ or lower). The helicase motif identity and similarity scores for SsRV-L and AHEV or SsRV-L and HEV are 28% and 41% or 29% and 42%, respectively. Interestingly, the RNA helicase motifs of SsRV-L also share significant sequence similarities with the insect viruses Helicoverpa armigera stunt virus (HaSV) and Dendrolimus punctatus tetravirus (DpTV), and the percentages of identity and similarity of the helicase domain between SsRV-L and those two insect viruses are 27% and 41%, respectively (Table 2). Both HaSV and DpTV belong to the genus Omegatetravirus in the family Tetraviridae (15, 63).

Furthermore, the methyltransferase motifs of SsDRV-L and AHEV share sequence similarities with the tobamoviruses Tomato mosaic virus, Tobacco mosaic virus, Pepper mild mottle virus, Cucumber green mottle mosaic virus, and Rehmannia mosaic virus, and the insect betatetravirus Nudaurelia capensis beta virus (Table 2).

Phylogenetic analysis.

ML distance comparisons of amino acid sequences of the RdRp domain of SsRV-L and representative viruses of alphaviruses, endornaviruses, tymo-like viruses, rubi-like viruses, and tobamo-like viruses showed that SsRV-L is most closely related to HEV, which belongs to the genus Hepevirus (see Table S1 in the supplemental material). Phylogenetic trees based on multiple alignments of RdRp conserved motifs of SsRV-L and these viruses were independently generated with the NJ algorithm and ML. The resulting NJ and ML trees had similar topologies and showed that SsRV-L clusters with several rubi-like viruses, including benyviruses, hepeviruses, omegatetraviruses, and rubivirus. The ML tree is shown in Fig. 3A.

FIG. 3.

Phylogenetic analysis of the conserved motifs and flanking sequences of RdRp (A) and viral RNA helicase (B) derived from aligned deduced amino acid sequences of SsRV-L and selected viruses. The NJ algorithm and ML were used to generate tentative phylogenetic trees independently. The NJ algorithm was performed using PAUP* 4.0b10 (Sinauer Associates, Sunderland, MA), and ML was performed using the program TREE-PUZZLE version 5.2 (54). The resulting ML tree is shown. The first number indicated at the nodes represents the bootstrap values (%) calculated from the NJ tree inferred from 1,000 bootstrap replicates, and the second number represents the quartet puzzling support values (%) inferred from 10,000 puzzling steps; a minus sign (−) indicates that a node is absent in the corresponding NJ method. Only bootstrap or quartet puzzling support values of >50% are indicated. The scale relates branch lengths to the number of substitutions per site. The tree was outgroup rooted to the viruses in the carmo-like group for RdRp analysis and to the arteri-like viruses (Infectious bronchitis virus) for viral RNA helicase analysis. See Table 1 for abbreviations of virus names, viral protein accession numbers, and the amino acid positions in the replicase sequences of the viruses used for phylogenetic analysis. Yellow and red stars, positions of the two viruses coinfecting S. sclerotiorum.

A phylogram based on multiple alignments of viral RNA helicase conserved motifs of SsRV-L and representative alphaviruses, endornaviruses, tymo-like viruses, rubi-like viruses, and tobamo-like viruses was similar to the tree generated by multiple alignments of RdRp conserved motifs (Fig. 3B). Likewise, a similar tree was generated based on the multiple alignments of methyltransferase conserved motifs of SsRV-L and selected viruses (see Fig. S2 in the supplemental material).

Evidence for autonomous replication of SsRV-L.

The single-sclerotium isolates Ep-1PNSA-8, Ep-1PNSA23, and Ep-1PNSA34, which were derived from strain Ep-1PN, lacked the M-dsRNA and S-dsRNA segments and contained only the L-dsRNA segment. This observation was confirmed with RT-PCR detection (Fig. 4). Thus, SsRV-L could replicate independently in S. sclerotiorum.

FIG. 4.

Northern hybridization analysis of S. sclerotiorum strains singly or doubly infected with SsRV-L. Five strains of S. sclerotiorum were used: strain Ep-1PN is doubly infected with SsRV-L and SsDRV; strains Ep-1PNSA-8, Ep-1PNSA-23, and Ep-1PNSA-34 are single-sclerotium isolates derived from strain Ep-1PN and contain only SsRV-L; and strain Ep-1PNA367 is a single-ascospore isolate derived from strain Ep-1PN and is virus free. (A) Electrophoretic analysis of dsRNA samples on agarose gels. (B) dsRNA samples were hybridized with an α-³²P-labeled cDNA probe of SsRV-L. (C) Total RNA samples were reversely transcribed with RevertAid M-MuLV reverse transcriptase and a specific RT primer (5′-CAGTCCCTAGTTTCATCTCGTTCC-3′) designed based on the sequence of SsDRV and PCR amplified with SsDRV-specific primers (reverse primer 5′-CAGTCCCTAGTTTCATCTCGTTCC-3′ and forward primer 5′-TGCAGGAAACAGTCATGGCAAC-3′).

Compared to the virus-free strain Ep-1PNA367, an ascospore isolate derived from strain Ep-1PN, the hyphal growth and virulence of strains Ep-1PNSA-8, Ep-1PNSA23, and Ep-1PNSA34 on detached rapeseed leaves were slightly reduced, but their growth and virulence were significantly higher than those of the hypovirulent strain Ep-1PN (Fig. 5). The sclerotial growth was normal on the PDA, and there were little or no differences in colony morphology between these strains and that of the wild-type strain of S. sclerotiorum. Thus, SsRV-L contributes little if any to the debilitation of strain Ep-1PN.

FIG. 5.

Effect of SsRV-L on hyphal growth and virulence of its host. Five strains of S. sclerotiorum were used: strain Ep-1PN is doubly infected with SsRV-L and SsDRV; strains Ep-1PNSA-8, Ep-1PNSA-23, and Ep-1PNSA-34 are single-sclerotium isolates derived from strain Ep-1PN and contain only SsRV-L; and strain Ep-1PNA367 is a single-ascospore isolate derived from strain Ep-1PN and is virus-free. (A) Comparison among five selected strains of S. sclerotiorum for their mycelial growth rate on a PDA plate at 20°C. d, day. (B) Comparison among the five strains for their virulence on detached leaves of rapeseed (Brassica napus) as determined by induced lesion diameter (in centimeters) at 20°C for 48 h.

DISCUSSION

This study represents the first report of a molecular characterization of a positive-strand RNA mycovirus, SsRV-L, that is phylogenetically related to the human pathogen HEV and rubi-like viruses. Sequence analysis of the full-length cDNA clone containing the coding sequence of a putative viral replicase revealed the presence of conserved methyltransferase, viral RNA helicase, and RdRp-2 superfamily motifs characteristic of the viral RNA replicase proteins of positive-strand RNA viruses and with significant similarity to the human pathogen HEV, a member of the genus Hepevirus (30). Like HEV, the genome of SsRV-L also has a short 5′-untranslated leader sequence and 3′-untranslated region with a poly(A) tail. However, unlike the genomes of HEV and HEV-swine, which contain three ORFs, SsRV-L genomic RNA contains only a single ORF (ORF1), with significant sequence similarity to ORF1 of HEV and the ORF1s of HEV-swine and AHEV. Furthermore, ORF1 of SsRV-L lacks coding sequences for papain peptidase or papain peptidase-like proteins. Interestingly, the viral RNA helicase and methyltransferase motifs of SsRV-L and HEV (including HEV-swine and AHEV) share significant sequence similarity with the insect tetraviruses and the plant tobamo-like viruses (Table 2).

The positive-strand RNA viruses were classified into three superfamilies: alpha-like, picorna-like, and flavi-like. The superfamily of alpha-like viruses comprises three lineages: rubi-like, tobamo-like, and tymo-like viruses (31). Whereas the tobamo- and tymo-like viruses infect plants, the rubi-like viruses infect plants, vertebrates, and insects. Recently, evidence was presented that the benyvirus (rubi-like lineage) Beet necrotic yellow vein virus (BNYVV) replicates in its plasmodiophorid vector Polymyxa betae, suggesting that BNYVV may also be considered a virus of plasmodiophorids (39). Our results from phylogenetic analysis support the conclusion that SsRV-L could be classified with the rubi-like viruses; thus, the host range of rubi-like viruses is more diverse than once thought. The reason for the broad host range of rubi-like viruses is not clearly known.

Although SsRV-L shares sequence similarities with HEV and the insect viruses HaSV and DpTV, it is more likely that these positive-strand RNA viruses were derived independently from plant viruses, even those plant viruses related to tobamoviruses. Since vertebrates and insects are not hosts of S. sclerotiorum, it is unlikely that S. sclerotiorum acquired the ancestral SsRV-L from these organisms. S. sclerotiorum, a plant pathogenic fungus, might have acquired the ancestral SsRV-L-like virus from a virus-infected plant, since it shares many hosts with viruses in the genus Tobamovirus. As a matter of fact, the hypovirulent strain Ep-1PN was originally isolated from a diseased eggplant plant (Solanum melongena) (35). Humans might have obtained HEV from herbivores; swine were one of the reservoirs of HEV (11), and wild animals can be infected by HEV or HEV-like viruses (57). Insects, such as H. armigera and D. punctatus, could have obtained HaSV and DpTV from plants, and it is feasible that birds obtained AHEV from virus-infected insects. Although these inferences are consistent with the hypothesis that the ancestral alpha-like positive-strand RNA virus might have spread horizontally among plants, fungi, vertebrates, and insects, they do not rule out the possibility of the ancient origin of the progenitor virus being a single cell type prior to the separation of fungi, plants, and animals. Recently, Koonin et al. (32) presented evidence that picorna-like virus evolution antedates the radiation of eukaryotic supergroups.

The genomic organization of SsRV-L is similar to that of SsDRV; both lack unnecessary genes, including genes for coat protein and movement protein. Hypoviruses and DaRV are also examples of mycoviruses that lack coat and movement proteins. These viruses, like SsRV-L and SsDRV, are phylogenetically related to plant viruses. Botrytis virus F (BVF) represents an example of a mycovirus that is phylogenetically related to plant viruses (potexviruses) but which codes for a coat protein in addition to the replicase (21). Unlike potexviruses, however, BVF does not code for a movement protein. It is not known how viruses delete unnecessary genes from their genomes; SsDRV, SsRV-L, DaRV, hypoviruses, and similar viruses may represent examples of regressive evolution by viruses in fungi.

The occurrence of SsRV-L and SsDRV in the hypovirulent strain Ep-1PN of S. sclerotiorum represents a novel type of coinfection involving two positive-strand RNA mycoviruses. Mixed infections with two or more dsRNA viruses appear to be a common occurrence among mycoviruses (14). Examples of mixed infections with dsRNA viruses belonging to distinct virus families include the totivirus Hv190SV and the chrysovirus Hv145SV that coinfect Cochliobolus victoriae (12, 13). There are many examples of mixed infections by distinct members of the same virus family, including two totiviruses, SsRV-1 and SsRV-2, that coinfect Sphaeropsis sapinea (48), two totiviruses, ScV-L-A and ScV-L-BC, that infect Saccharomyces cerevisiae, and the partitiviruses Penicillium stoloniferum viruses S and F that infect Penicillium stoloniferum.

As previously reported, it is not clear whether SsRV-L contributes to the hypovirulence phenotype of strain Ep-1PN (36), since S. sclerotiorum isolates that carry SsRV-L alone exhibit a normal phenotype. Thus, we assume that S. sclerotiorum might have acquired SsRV-L earlier than SsDRV, and SsDRV could have been acquired through hyphal anastomosis with other SsDRV-infected fungal strains, since mycoviruses could be transmitted between vegetatively incompatible strains, though at lower frequencies.

Unlike C. parasitica, which can produce both ascospores and conidial spores, S. sclerotiorum can only produce ascospores that form in apothecia germinating from dormant sclerotia. Thus, SsRV-L and SsDRV cannot be dispersed via conidial spores. Furthermore, our previous work showed that neither SsRV-L nor SsDRV could be transmitted through ascospores of strain Ep-1PN (27, 66), and hyphae do not present valid dormant material for dispersal of S. sclerotiorum. Therefore, SsRV-L and SsDRV could be transmitted and dispersed only by sclerotia, and the distribution of SsRV-L and SsDRV would then be confined to limited areas. Moreover, the survival ability of fungal strains doubly infected with SsRV-L and SsDRV is predicted to be very low, since they grow slowly, virtually lose their virulence, and produce only a few sclerotia (35). Thus, we reasoned that horizontal transfer of the ancestral viruses of SsRV-L and SsDRV from other fungi or plants to S. sclerotiorum might have occurred relatively recently. Because S. sclerotiorum has a broad host range of plants known to be susceptible to many closteroviruses and potexviruses, the possibility that closterovirus-like (SsRV-L) and potexvirus-like (SsDRV) viruses might have transferred from plants to S. sclerotiorum seems feasible (8, 40).

SsRV-L appears to be more stable than SsDRV in S. sclerotiorum, since subcultures that contain SsRV-L but lack SsDRV are relatively easy to obtain by hyphal tipping and single sclerotium isolation. Like SsDRV, SsRV-L is also eliminated through sexual reproduction of S. sclerotiorum; this phenomenon is common among mycoviruses of higher ascomycetous hosts, e.g., the hypovirus/C. parasitica system (4), but the underlying mechanism is not known.

Currently, it is not understood whether there is an interaction between SsRV-L and SsDRV in doubly infected S. sclerotiorum and whether this interaction has any bearing on the debilitation phenotype, since the two viruses can replicate independently. S. sclerotiorum strains singly infected with SsRV-L show little or no adverse effects, whereas strains that are singly infected with SsDRV exhibit a debilitated phenotype. In a recent study, Li et al. (37) identified several genes whose expression was downregulated in a doubly infected S. sclerotiorum strain. The hypovirulence/debilitation system of S. sclerotiorum and its associated mycoviruses presents an attractive system to explore the molecular basis of pathogenicity in this devastating plant pathogen (37). Future construction of infectious full-length cDNA clones for these two viruses and the development of RNA transfection systems would be useful in deciphering the interaction between SsRV-L and SsDRV and their effects on their common host.