LlU Wen-wen, XlN Min, CAO Meng-ji, QlN Meng, LlU Hui, ZHAO Shou-qi, WANG Xi-feng

1 State Key Laboratory for Biology of Plant Diseases and Insect Pests, Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing 100193, P.R.China

2 National Citrus Engineering Research Center, Citrus Research Institute, Southwest University, Chongqing 400712, P.R.China

3 National Agricultural Technology Extension and Service Center, Beijing 100125, P.R.China

Abstract To identify the possible quarantine viruses in seven common sunflower varieties imported from the United States of America and the Netherlands, we tested total RNAs extracted from the leaf tissues using next-generation sequencing of small RNAs. After analysis of small RNA sequencing data, no any quarantine virus was found, but a double-stranded RNA(dsRNA) molecule showing typical genomic features of endornavirus was detected in two varieties, X3939 and SH1108.Full-length sequence and phylogenetic analysis showed that it is a novel endornavirus, temporarily named as Helianthus annuus alphaendornavirus (HaEV). Its full genome corresponds to a 14 662-bp dsRNA segment, including a 21-nt 5′untranslated region (UTR), 3′ UTR ending with the unique sequence CCCCCCCC and lacking a poly(A) tail. An open reading frame (ORF) that encodes a deduced 4 867 amino acids (aa) polyprotein with three domains: RdRP, Hel and UGT(UDP-glycosyltransferase). HaEV mainly distributed in the cytoplasm but less in the nucleus of leaf cells by fluorescence in situ hybridization (FISH) experiment. This virus has a high seed infection rate in the five varieties, X3907, X3939, A231,SH1108 and SR1320. To our knowledge, this is the first report about the virus of the family Endornaviridae in the common sunflower.

Keywords: common sunflower (Helianthus annuus), next-generation sequencing (NGS), double-stranded RNA,Endornaviridae

1. lntroduction

As one of the major oil crops in the world, the common sunflower (Helianthus annuus) is grown in many countries with a harvested area over 26.2 million hectares and seed production of about 47.3 million tons (FAOSTAT 2016).Although the common sunflower is highly adaptable, i.e.,a relatively high tolerance to drought and insensitivity to photoperiod, diseases and pests have caused high production losses (Gornik and Lahuta 2017). Naturally,many viruses can infect common sunflower including Sunflower mosaic virus, Sunflower leaf curl virus, Sunflower chlorotic mottle virus, Cucumber mosaic virus and Sunflower ring blotch virus (Dujovny et al. 1998; Gulya et al. 2002;Bejerman et al. 2010; Cabrera Mederos et al. 2017).However, no endornavirus has been found in the common sunflower until now.

Viruses in the family Endornaviridae usually infect plants(Sabanadzovic et al. 2016), fungi (Khalifa and Pearson 2014; Shang et al. 2015) and oomycetes (Hacker et al.2005). These viruses are different from “traditional” viruses because they don’t have the capsid (Moriyama et al. 1995).Endornaviruses are composed of a linear double-stranded RNA (dsRNA) genomes that ranges from 9.6 to 17.6 kb and encodes a large polyprotein containing several functional domains (Okada et al. 2013; Khalifa and Pearson 2014).Viral RNA-dependent RNA polymerase (RdRp) is the only universally conserved domain in all endornaviruses. Other domains like methyltransfersase (MTR), helicase (Hel) and glucosyltransferase (GT) vary in their presence or absence in different genera (Roossinck et al. 2011). Along with the characterization of more endornaviruses, viruses in the genus Endornavirus, family Endornaviridae have been changed to genus Alphaendornavirus and genus Betaendornavirus according to the 10th Report of International Committee on Taxonomy of Viruses (ICTV) in 2016, which recognizes 18 species in the genus Alphaendornavirus and 4 species in the genus Betaendornavirus. Among them, 13, 8, and 1 species infect plants, fungi and oomycete, respectively(Adams et al. 2017).

Currently, all plant-infecting endornaviruses have been assigned to the genus Alphaendornavirus by the 10th Report of ICTV. They infect several economically important crops such as avocado, common bean, spinach, bottle gourd,pepper and rice (Moriyama et al. 1999; Okada et al. 2011;Villanueva et al. 2012; Okada et al. 2013; Kwon et al. 2014;Okada et al. 2014; Lim et al. 2015). These dsRNA viruses occur at every developmental stages and tissues of host,and replicate in the host cells (Moriyama et al. 1995). Mostendornaviruses are generally persistent in the plants and do not alter phenotypes or traits of their hosts, but a few endornaviruses also have an impact on host plants (Grill and Garger 1981; Sela et al. 2012).

Industry of the common sunflower is developing rapidly in China. In order to satisfy the needs of seeds, many commercial varieties have been imported from other countries. In July 2016, seven sunflower varieties imported from the United States of America and the Netherlands were grown in the plant quarantine experimental bases of the National Agricultural Technology Extension and Service Center, Beijing, China. We used next-generation sequencing of small RNA (sRNA-seq) to detect and identify potential quarantine viruses. Here, we reported the molecular and biological properties of an endornavirus,named as Helianthus annuus alphaendornavirus (HaEV),identified from these imported varieties of the common sunflower.

2. Materials and methods

2.1. Sample collection of the common sunflower(H. annuus)

The leaves of seven common sunflower varieties were collected from the plant quarantine experimental bases of the National Agricultural Technology Extension and Service Center, Beijing, China on July 5, 2016. Varieties X3907, X3939 and FFH702 were imported from the Netherlands, but varieties A231, SH1220, SH1108 and SR1320 from the United States (Appendix A). The leaves of each variety were collected in their own sealed bag,respectively, then taken to our laboratory for total RNA purification.

2.2. Small RNA sequencing

Total RNA was extracted from leaf tissues of seven common sunflower varieties using the Total RNA Purification Kit (Invitrogen, Waltham, MA, USA) according to the manufacturer’s instructions. The total RNA was quantified by measuring absorbance at 260 nm using a Nanodrop 2000 Spectrophotometer (Infinigen Biotechnology, City of Industry, CA, USA). Small RNA was enriched by PEG8000 precipitation from 10 μg of the total RNA from each sample.Proprietary adapters were then ligated to the 5′ and 3′termini of these small RNAs, and then the ligated small RNAs were used as templates for cDNA synthesis. The cDNA was amplified to produce sRNA libraries using TruSeq Small RNA SamplePrep Kit (Illumina, San Diego, CA, USA)(Xin et al. 2017a). The libraries were quantified by ECO(Illumina), then sequenced with an Illumina HiSeq 2500 according to the manufacturer’s instructions.

2.3. Analysis of small RNA sequencing data

The image files generated by the sequencer were then processed to produce high quality data. After masking of adapter sequences and removal of contaminated reads,we got the clean reads of full-length small RNA sequences for further analysis. High quality sequences were clustered into unique reads, and small RNAs with a length of 18–30 nt were identified. The clean sRNA reads were de novo assembled with a k-mer of 17 using the Velvet Program (Wu et al. 2012). The assembled contigs were subsequently compared against the non-redundant (nr) database from the National Center for Biotechnology Information (NCBI)using Blastn and Blastx search.

2.4. Amplification and analysis of full genomes

After confirmation of the novel virus (HaEV) infection, RNA was extracted from the virus-infected plants. Then cDNAs were obtained with designed primers from HaEV contigs,which named as 1–2 R, 2–3 R, 3–4 R and 4–5 R (Appendix B).Based on these contigs, RT-PCRs (Okada et al. 2017)were performed to cover the gaps using MMoLV Reverse Transcriptase (Promega, Madison, WI, USA) and LA Taq DNA polymerase (TaKaRa, Dalian, China) with the primer pairs 1–2 F, 1–2 R, 2–3 F, 2–3 R, 3–4 F, 3–4 R and 4–5 F,4–5 R (Appendix B). In addition, 5′ and 3′ RACE (rapidamplification of cDNA ends) System for Rapid Amplification of cDNA Ends Kits (Thermo Fisher Scientific, Waltham,MA, USA) were used to amplify the 5′- and 3′-terminal sequences of the virus genome with specific primers(Appendix B). Following the manufacturer’s instructions,the PCR products were purified with Wizard SV Gel and PCR Clean-Up Kit (Promega) and cloned into pMD18T cloning vector (TaKaRa). Then the clones were sequenced by Beijing Genomics Institute (Beijing, China). These sequences were spliced into the whole genome by Vector NTI Software (Thermo Fisher Scientific). Open reading frames (ORFs) and conserved domains were discovered using the NCBI ORF Finder (https://www.ncbi.nlm.nih.gov/orffinder/) and Conserved Domain Database (CDD)(https://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml),respectively (Marchler-Bauer et al. 2015).

2.5. Extraction of dsRNA

The dsRNA was extracted from the infected leaves of common sunflower varieties SH1108 using the CF-11 cellulose chromatography method (Valverde et al. 1990).The dsRNA was treated with S1 nuclease and DNase I(TaKaRa) to eliminate the contamination of single-stranded RNA (ssRNA) and DNA, electrophoresed on a 1% agarose gel, and then visualized by staining with ethidium bromide.

2.6. Phylogenetic analysis

Phylogenetic relationships among HaEV and other members of the family Endornaviridae were analyzed using polyprotein, RdRp and Hel sequences from GenBank(Table 1). The phylogenetic trees were generated using the neighbor-joining algorithm by MEGA 7.0 (Kumar et al.2016) with bootstrap of 1 000 replicates for HaEV and other viruses. We handled gaps and missing data using the option “pairwise deletion”. All sequence alignments were adjusted post hoc by visual inspection to ensure that the alignments were biologically relevant. Symbols indicate viruses infecting the same group of host, plants, fungi and oomycetes.

2.7. Detection of seed infection rate

We planted the seeds of seven common sunflower varieties X3907, X3939, FFH702, A231, SH1220, SH1108 and SR1320 in the soil that had been sterilized at 180°C for 2 h.After 17 days, the leaves from each variety were collected separately for virus detection. Total RNA of leaves was extracted using Total RNA Purification Kit (Invitrogen).RT-PCR was performed for virus detection with specially designed primers E1-F, E1-R, E2-F and E2-R (Appendix B).The PCR program was 94°C, 5 min; 32 cycles of (94°C,45 s; 54.2°C, 45 s; 72°C, 1.5 min); 72°C, 10 min.

2.8. Fluorescence in situ hybridization (FlSH)

Pieces of virus-infected leaves (1 cm×1 cm) were prechilled in chemical fixatives FAA (75% ethanol, 5% acetic acid and 5% formalin) at 4°C for 24 h fixation (Appendix C).Then samples were paraffin-embedding followed by the microwave method reported previously (Inada and Wildermuth 2005). The paraffin sections were then washed 3 times with fresh 100% xylene for 5 min each time, then rehydrated in an ethanol series (5 min each: 100, 90 and 70%). After incubating in 3% H2O2at 37°C for 10 min, the samples were washed three times with PBS for 5 min each time, then soaked in 0.1 mol L–1HCl for 30 min.Subsequently, samples were soaked in 0.5% Triton X-100 for 30 min and PBS two times for 10 min to remove the HCl.After a proteinase K digestion at 37°C for 10 min, samples were fixed with 4% paraformaldehyde solution for 5 min and washed with PBS three times for 10 min. The probe(Appendices B and C) which was synthesized by Sangon Biotech (Shanghai) Co., Ltd., was diluted with hybridization buffer with 1:100 at 88°C for 5 min and 37°C for 2 min. Then the samples were combined with probe labeled with Cy5(red) and sealed with rubber cement in a moist chamber at 37°C overnight for hybridization. After incubating in 3%(w/v) BSA for 30 min, they were exposed to the HRP-labeled secondary antibody at 37°C for 1 h. After processing in Tyramide Signal Amplification System (Invitrogen), samples were stained with 2 μg mL–1DAPI in PBS for observing under the microscope (Nikon 80i, Japan).

3. Results

3.1. Small RNA analysis of common sunflower varieties

After the analysis of the small RNA data from leaf tissues of seven different common sunflower varieties, X3907, X3939, FFH702, A231,SH1220, SH1108 and SR1320, no any quarantine virus was found, but a dsRNA molecule showing typical genomic features of endornavirus was found in two varieties, X3939 and SH1108 which had typical viral symptoms such as dwarf and leaf crinkle in the field (Fig. 1-A and B). In the Blastn and Blastx search of the assembled viral contigs from these two varieties against the non-redundant (nr)database of the NCBI, seven and eight contigs from variety X3939 and SH1108 respectively shared significant nucleotide and amino acid (aa) similarities with the genes encoded by the viruses of family Endornaviridae. These results showed the potential presence of an endornavirus in the analyzed samples of the common sunflower varieties.

3.2. Amplification and analysis of viral genome

Based on the viral contigs from the leaves of infected plants, we spliced five long contigs (1 830, 3 800, 1 506,2 334 and 2 691 nt, respectively) by vector NTI. Then we obtained four gaps (164,229, 577 and 1 838 bp, respectively)between these contigs using RT-PCR,respectively. Finally, the full-length viral genome was amplified by 5′ and 3′ RACE experiments. The whole viral genome consists of one dsRNA segment 14 662 bp long, including a 21 nt and 3′ UTR ending with the unique sequence CCCCCCCC and lacking a poly(A)tail (Fig. 2-A). It was also observed similar length band on the agarose gel of dsRNA extracted from HaEV infected leaves of the sunflower variety SH1108 (Fig. 2-B). We named this virus as H. annuus alphaendornavirus(HaEV). Its genome includes a 14 604 nt open reading frame (ORF) coding for a deduced polyprotein (GenBank accession MF362666) of 4 867 aa with three domains, Hel (helicase), UGT(UDP-glycosyltransferase) and RdRp(RNA dependent RNA polymerase)(Fig. 2-A). The viral-derived small interfering RNAs(vsiRNAs) derived from both varieties X3939 and SH1108 cover the entire viral genome with a distribution profile exhibiting same hotspot (Fig. 2-C and D). Some hotspots were particularly prominent between 2 000–3 000 nt in both varieties, but some hotspots between 7 000–8 000 nt were found only in X3939 (Fig. 2-C).

Fig. 1 Plants of common sunflower showed dwarfing and leaf crinkle symptoms. A and B, variety X3939 (A) and SH1108 (B) in the field. C and D, variety X3939 (C) and SH1108 (D) in the greenhouse. Leaf from each boxed area is on the right.

Fig. 2 Analysis of the Helianthus annuus alphaendornavirus (HaEV) genome. A, diagrammatic representation of HaEV genome organization. The large box represents the open reading frame (ORF). The smaller boxes depict domains identified in the encoded polyprotein. B, agarose gel electrophoresis of the dsRNAs extracted from leaves of the sunflower variety SH1108. The dsRNAs were treated with both RNase-free DNase I and S1 nuclease. Lane M, DL15 000 bp DNA marker; lane 1, double-stranded RNA(dsRNA) extracted from HaEV-infected leaves; lane 2, dsRNA extracted from control leaves. C, distribution profile of the HaEV-vsiRNAs (viral-derived small interfering RNAs) in variety X3939. D, distribution profile of the HaEV-vsiRNAs in variety SH1108.Hel, viral helicase superfamily; UGT, UDP-glycosyltransferase; RdRp, viral RNA dependent RNA polymerase.

The profile of siRNAs derived from HaEV was characterized to gain insight into the host sRNA induced by the virus. A total of 528 169 reads were mapped to HaEV,accounting for 1.44% of the total reads in variety X3939.Meanwhile, 540 081 reads that accounted for 2.27% of the total reads were mapped to the virus in variety SH1108.Basic groups (U, G, C, A) were distributed uniformly among 18–26 nt and were almost the same in the two varieties(Fig. 3-A). The 21 and 22 nt size classes of the HaEV-derived sRNAs were the most abundant in both varieties.Nevertheless, the 21 nt class and the 22 nt class had almost the same abundance in variety X3939, while the abundance of 21 nt class was much higher than the 22 nt class in variety SH1108 (Fig. 3-B). In addition, the polarity analysis of HaEV-sRNAs showed that more sRNAs were derived from the sense genomic strand (53.66%) than the antisense genomic strand (46.34%) in variety X3939. The same result was also found for variety SH1108; more sRNAs were derived from the sense genomic strand (56.86%) than the antisense genomic strand (43.14%) (Fig. 3-C).

Fig. 3 Small RNA profiling of Helianthus annuus alphaendornavirus (HaEV) from infected leaves in sunflower varieties X3939 and SH1108. A, relative frequency of four different 5′-terminal nucleotides for 18–26 nt vsiRNAs (viral-derived small interfering RNAs) of HaEV. B, size distribution of vsiRNAs derived from HaEV. C, diagram of vsiRNAs derived from HaEV mapped to sense and antisense strand.

3.3. Polyprotein features and phylogenetic relationships of HaEV with other endornaviruses

The deduced polyprotein of HaEV shared higher amino acid identities with alphaendornaviruses, i.e., 20.9% identity with Phytophthora alphaendornavirus 1 (PEV1), 20.7%with Phaseolus vulgaris alphaendornavirus 2 (PvEV2) and 20.4% with Cucumis melo alphaendornavirus (CmEV)(Table 1). The conserved domain RdRp and Hel of HaEV had also higher identities with alphaendornaviruses. RdRp showed 45.7% identity with Erysiphe cichoracearum alphaendornavirus (EcEV) and Hel had 31.7%identity with hot pepper alphaendornavirus (HpEV)(Table 1).

Meanwhile, the phylogenetic relationships of polyprotein,RdRp and Hel of HaEV had been analyzed with other endornaviruses. Based on multiple alignment results, the phylogenetic tree of endornaviruses contained two clades:alphaendornaviruses in Clade I and betaendornaviruses in Clade II (Fig. 4). The polyprotein of HaEV and Helicobasidium mompa alphaendornavirus 1 (HmEV1) formed one branch,and then clustered together with other alphaendornaviruses(Fig. 4-A). Hel of HaEV, closest to PEV1 was also clustered with other alphaendornaviruses (Fig. 4-B). Although RdRp of HaEV was still in the alphaendornaviruses clade, it had evolved into an independent branch (Fig. 4-C). All of these results suggested that HaEV should be a member of the genus Alphaendornavirus.

Fig. 4 Phylogenetic tree of Helianthus annuus alphaendornavirus (HaEV) and other members of the family Endornaviridae based on the amino acid sequences of the polyprotein, RdRp and Hel. Trees were generated using the neighbor-joining algorithm in MEGA 7.0 with bootstrap of 1 000 replicates. A, tree based on the amino acid sequences of the polyprotein. B, the amino acid sequences of Hel. C, the amino acid sequences of RdRp. Symbols indicate sequences infecting the same group of host, plants,fungi, or oomycetes. Virus abbreviations are defined in Table 1.

3.4. Detection of HaEV seed infection rate in different varieties

In the infection rate test, we planted seven varieties withsterilized soil in the greenhouse. RNAs were extracted from the leaves of different seedlings for HaEV detection.The infection efficiency was 70, 65, 85, 71 and 77% in X3907, X3939, A231, SH1108 and SR1320, respectively(Table 2). No plants of varieties FFH702 and SH1220 were virus-infected (Table 2). In addition, HaEV-infected varieties X3939 and SH1108 usually accompanied with symptoms like dwarfed and leaf crinkle (Fig. 1-C and D), while no symptom was observed in other HaEV-infected varieties.

Table 2 Percentage infection of common sunflower varieties by Helianthus annuus alphaendornavirus (HaEV) detected by RT-PCR

3.5. Distribution of HaEV in leaf cells of common sunflower

FISH analysis was used to examine the cellular distribution in HaEV-infected leaves. It is obvious that many fluorescent signals of HaEV (red) were found in the infected leaves of varieties SH1108 (Fig. 5-A), but no fluorescent signals in the control leaves (Fig. 5-B). The viral fluorescent signals (red)were located around the cell nucleus (blue), and few were found in the nucleus of infected leaves (Fig. 5-A). This result suggested that HaEV can exist both in the cytoplasm and nucleus, although abundant only in the cytoplasm.

4. Discussion

Vicia faba alphaendornavirus (VfEV), as the first virus in the family Endornaviridae, was originally discovered in broad bean by extracting dsRNA from the leaf tissues (Grill and Garger 1981). Subsequently, other endornaviruses were also discovered by extracting dsRNA from their host plants,fungi and oomycetes (Moriyama et al. 1995; Hacker et al.2005; Osaki et al. 2006). The extraction of dsRNA was the main method for discovery the endornavirus for many years. Recently, combination of next-generation sequencing(NGS) technologies and bioinformatics has accelerated the discovery of novel viruses, because the small RNAs can generate during an antiviral defense response in the plants infected by viruses, and then these small RNA could be used to identify known and novel viruses through small RNA sequencing (Massart et al. 2014). Thus, NGS technologies have been widely used to detect viruses in the animals,insects and plants (Hadidi et al. 2016; Shi et al. 2016; Xin et al. 2017b, c; Zhang et al. 2017). Here, we identified a novel endornavirus from common sunflower samples using small RNA sequencing.

This virus, HaEV, has three domains, the RdRp domain,Hel domain and UDP-glycosyltransferase (UGT) domain,which are present in all known alphaendornaviruses. The main differences between the genomes of the two genera in the family Endornaviridae are the combination of domains.Alphaendornaviruses usually code for the RdRp domain, GT domain, Hel domain and lack an identifiable MTR domain(Ong et al. 2016; Sabanadzovic et al. 2016). Phylogenetic analysis showed that RdRp and Hel domains encoded by HaEV and other species of alphaendornaviruses are all belong to the same evolutionary lineage. Commonly,genomes of alphaendornaviruses (more than 13 kb) are also longer than those of betaendornaviruses (up to 12 kb)(Adams et al. 2017), and the whole viral genome of HaEV consists of one dsRNA segment more than 14 kb. Thus,the domain composition, phylogenetic analysis and the genome length suggest that HaEV belongs to the genus Alphaendornavirus, family Endornaviridae.

Fig. 5 Distribution and localization of Helianthus annuus alphaendornavirus (HaEV) in the leaf cells of common sunflower variety SH1108 were detected by fluorescence in situ hybridization. Nucleus was stained with DAPI (4,6-diamino-2-phenyl indole) (blue),and probe for HaEV was labeled with Cy5 (red). Green arrows indicate intense fluorescence signal for HaEV in the cytoplasm.A, HaEV-infected leaves. B, negative control leaves.

Compared with betaendornaviruses only infected fungi,the hosts of alphaendornaviruses were more abundant including 13 plants, four fungi, and one oomycetes.However, HaEV, identified from the common sunflower, has high homologous with the alphaendornavirus infected fungi and oomycete. The deduced polyprotein of HaEV has the highest homology with HmEV1 in the phylogenetic analysis and PEV1 in the identity analysis (Fig. 4-A; Table 1). RdRp and Hel-1 of HaEV also showed the highest homology with fungi-infection alphaendornaviruses (Fig. 4-B and C; Table 1). In some previous studies, the evolution of alphaendornaviruses had possibly involved horizontal transmission between host types, e.g., between fungi and plants (Gibbs et al. 2000; Roossinck et al. 2011; Khalifa and Pearson 2014), our result supports this hypothesis.

Viruses in the family Endornaviridae can persist in their hosts over multiple generations (Roossinck et al. 2011).In fungi, they transmit horizontally via hyphal anastomosis and vertically via spores (Osaki et al. 2006; Tuomivirta et al.2009). In plants, they only were transmitted vertically via seeds or pollen with high efficiency (Horiuchi et al. 2003;Du et al. 2016; Fukuhara et al. 2006). Our study showed that although HaEV was not detected from varieties FFH702 and SH1220, it was still had a high infection rate (＞60%)in other five varieties X3907, X3939, A231, SH1108 and SR1320. Because no virus was found in their field samples of varieties FFH702 and SH1220 by analysis of small RNA sequencing data. It is suggested that once the sunflower was infected with HaEV, the virus would be transmitted to the next generation at a high rate.

Endornaviruses which infected plants are usually localized in the cytoplasm of host cells, and they can be transmitted efficiently to progeny plants. For example, cytoplasmic inheritance of these viruses was reported for F1hybrids of cultivated and wild rice (Moriyama et al. 1999). FISH analysis of HaEV suggested this virus is highly abundant in the cytoplasm of leaf cells, but low in the nucleus. Although the biological significance of this phenomenon is unclear,we propose that this distribution may be responsible for the highly efficient vertical transmission of HaVE.

Endornaviruses are not usually accompanied by visible viral symptoms or altered phenotypes or traits of their hosts,nor does the growth and development of the host seem to be affected (Du et al. 2016; Sabanadzovic et al. 2016).However, there were still some endornaviruses that could have a degree of influence on the host, such as VfEV, which are associated with the cytoplasmic male sterility trait in Vicia faba (Grill and Garger 1981; Pfeiffer 1998) and a bell pepper alphaendornavirus that can activate RNA silencing in the host (Sela et al. 2012). In addition, HmEV 1 has been reported having a virulence role in the host fungus (Osaki et al. 2006). In our study, two varieties X3939 and SH1108 that infected with HaEV usually accompanied plant dwarfing and leaf crinkle symptoms, but no such symptoms in other varieties. Because the deduced polyprotein of HaEV has the highest homology with HmEV1 in the phylogenetic, this virus was likely to be harmful to host plants as well. In the nextstep, we will do the further study whether the occurrence of these symptoms is related to HaEV infection or other factors.

5. Conclusion

In this work, we detected the viruses in seven common sunflower varieties imported from the United States of America and the Netherlands using high-throughput sequencing of small RNAs. A novel endornavirus HaEV was found in the two varieties X3939 and SH1108. The genomic organization, phylogenetic relationship, and biological characterization showed that it belonged to the genus Alphaendornavirus, family Endornaviridae. HaEV usually accompanied by high rates of infection in some varieties and mainly distributed in the cytoplasm but less in the nucleus of infected leaf cells. Unlike most other plantinfecting endornaviruses, this virus showed more similarity with alphaendornavirus which infected fungi and oomycete.

Acknowledgements

This research was supported by the Inter-Governmental S&T Cooperation Proposal between China and Czech Republic (2016YFE0131000) and the Beijng Nova Program,China (Z171100001117036). We thank Dr. Beth E. Hazen of Willows End Scientific Editing and Writing for editing the English and scientific language.

Appendicesassociated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm