In Korea, daffodil (Narcissus pseudonarcissus) cultivation largely depends on bulbs imported from abroad. Narcissus bulbs are mainly imported from major producing countries like the Netherlands and the UK (Hanks and Chastagner, 2017; Valouzi et al., 2022). According to Korea Agro-Trade Information (KATI), bulb imports increased over fivefold in the past decade, from 200 million KRW in 2014 to 1.1 billion KRW in 2024 (KATI; https://www.kati.net/statistics/periodPerformance.do). Vegetative propagation of Narcissus via bulbs allows vertical virus transmission (Raj et al., 2021; Shin et al., 2002). As a result of infections through infected bulbs and vectors, numerous viral diseases have been reported in Narcissus, posing a constraint to the production of high-quality bulbs (Probowati et al., 2022). To date, 26 viruses have been reported in Narcissus worldwide (Kim et al., 2024; Probowati et al., 2022). In the Republic of Korea, five viruses have been reported in Narcissus: cyrtanthus elatus virus A (CyEVA), narcissus mosaic virus (NMV), narcissus yellow stripe virus (NYSV), snowdrop virus Y (SVY), and tobacco rattle virus (TRV) (Kim and Jeong, 2024; Kim et al., 2024, 2025; The Korean Society of Plant Pathology, 2022). Furthermore, six viruses-arabis mosaic virus, raspberry ringspot virus, strawberry latent ringspot virus, tobacco ringspot virus, tomato black ring virus, tomato ringspot virus-are designated as quarantine pathogens for imported Narcissus bulbs by the Animal and Plant Quarantine Agency (APQA, 2022). Despite 26 virus species being documented worldwide, there is a growing concern regarding the potential introduction of various viruses into Korea via imported bulbs. To investigate viral pathogens in Korean Narcissus cultivations, symptomatic leaf samples from experimental and commercial fields were analyzed using metatranscriptome analysis and RT-PCR.
In May 2021, 75 Narcissus samples exhibiting virus-like symptoms were collected from experimental and commercial fields. These included 10 samples of five cultivars (‘Love Call’, ‘Replete’, ‘Tête-à-Tête’, ‘Tahiti’, and ‘Keitabrew’) from the Floriculture Research Institute in Yesan, and 65 samples from commercial farms in Wonju, Geoje, and Shinan, representing five cultivars (‘Replete’, ‘Goblet’, ‘Tahiti’, ‘Salou’, and ‘Erlicheer’) along with nine samples of unknown cultivars.
Metatranscriptome analysis was performed on an Illumina HiSeq 2000 platform (Macrogen, Seoul, Korea). 10 mg of leaf tissue from each of the 75 samples was pooled and homogenized in liquid nitrogen. Total RNA was extracted using the Maxwell® 16 LEV Plant RNA Kit (Promega, Madison, WI, USA) and subjected to construct a cDNA library with the TruSeq Stranded Total RNA LT Sample Prep Kit for Plants (Illumina, San Diego, CA, USA). De novo assembly was performed using Trinity (version r20140717), and viral contigs were identified based on BLASTn and BLASTx analyses (E-value cutoff 1.0E−5).
Metatranscriptome sequencing (60 Gbp) and assembly generated 96,215 contigs. Subsequent BLAST analyses identified 53 contigs related to plant viruses. These contigs were presumed to be derived from ten plant viruses: CyEVA, narcissus late season yellows virus (NLSYV), narcissus latent virus (NLV), NMV, NYSV, nerine latent virus (NeLV), SVY, TRV, turnip yellows virus (TuYV), and Japanese iris necrotic ring virus (JINRV, Betacarmovirus iridis). Among these, a single nearly complete JINRV-related contig (4,082 bp; 501,565 reads) was identified. The contig shared 66.04% nucleotide identity with the type isolate (accession no. D86123) and 84.11% with the Australian isolate (JQ807998). While JINRV has previously been reported only in Iris species (Wylie et al., 2012; Yasukawa et al., 1982, 1991), to the best of our knowledge, this study is the first to identify Narcissus as a natural host for JINRV.
RT-PCR was performed on all 75 Narcissus samples to confirm JINRV. Total RNA was extracted using the Easy-spin™ (DNA free) Total RNA Extraction Kit (iNtRON Biotechnology, Daejeon, Korea). Based on previously reported JINRV isolates (accession no. D86123 and JQ807998) and the contig obtained from the metatranscriptome, primers targeting a partial region of the coat protein (CP) gene were designed (Supplementary Table 1).
JINRV was detected in 22 of 75 samples (29.3%), all of which were collected from commercial fields in Shinan, Jeollanam-do, with a local infection rate of 55.0% (22/40) (Table 1). No JINRV infection was detected in samples collected from regions other than Shinan. Among the four cultivars sampled in Shinan (‘Goblet’, ‘Replete’, ‘Tahiti’, and ‘Salou’), JINRV was detected in ‘Goblet’, ‘Replete’, and ‘Tahiti’, but not in ‘Salou’. RT-PCR for the other nine viruses was conducted using specific primers (Supplementary Table 1). All JINRV-positive samples were co-infected with JINRV and two to five additional viruses, including NLSYV, NLV, NYSV, NeLV, SVY, TuYV and TRV (Supplementary Table 2). The JINRV-infected plants showed poor growth, chlorotic mosaic, leaf distortion, and yellow striping (Fig. 1); however, these symptoms may be associated with mixed infections including JINRV rather than being attributable to JINRV alone.
To determine the complete SA36 genome sequence, terminal regions were amplified via RACE-PCR and the internal region via RT-PCR, as previously described (Bak et al., 2024). Overlapping amplicons from RACE-PCR and RT-PCR (Supplementary Table 3) were cloned, sequenced, and assembled into the genome using DNAMAN 7.0. To analyze genetic variation among JINRV isolates, three isolates (Replete-SA36, Goblet-SA45, and Tahiti-SA46) were selected from infected cultivars, and the complete CP sequence (1,056 bp) was amplified using two overlapping primer sets (Supplementary Table 1).
Phylogenetic analyses were conducted using MEGA11 v11.0.13 (Kumar et al., 2018), applying the neighbor-joining method for the relatively conserved RNA-dependent RNA polymerase (RdRp) region and the maximum-likelihood method for the variable CP region to address complex substitutions (1,000 bootstrap replicates), respectively, comparing SA36 to 17 viruses in the family Tombusviridae (Supplementary Table 4). Furthermore, nucleotide and amino acid sequence identities among the three Narcissus- and Iris-derived isolates were analyzed using the Sequence Demarcation Tool v1.2 (Muhire et al., 2014).
The complete genome of JINRV isolate SA36 was 4,060 nucleotides long, with five putative open reading frames (ORFs) and untranslated regions (UTRs) at both termini (Fig. 2A). The coding region spans nucleotides 64-3,772, with 5′ and 3′ UTRs of 63 nt and 288 nt, respectively. ORF1 encodes a 27 kDa protein ending with an amber stop codon, while ORF2 is translated via readthrough to produce an 85-kDa protein. ORF2 contained a GDD motif, a conserved amino acid sequence commonly found in RdRps (Charon et al., 2022). ORF3 and ORF4 encode 8-kDa and 12-kDa proteins, respectively, putative movement proteins (MPs). The ORF5 at the 3′ end encodes a 37-kDa CP. The complete genome sequence of JINRV isolate SA36 was deposited in GenBank (accession no. LC878470). The genome organization of isolate SA36 was compared with that of previously reported isolates (Fig. 2B, 2C). JINRV isolate SA36 has a single amber stop codon, similar to the Australian isolate Marijiniup10, whereas the Japanese isolate has two. Accordingly, SA36 and Marijiniup10 each encode one readthrough-associated protein, whereas the Japanese isolate encodes two.
Sequence analysis of four protein regions, RdRp (p85), MP (p8), MP (p12), and CP (p37), revealed that JINRV isolate SA36 shared nucleotide and amino acid identities of up to 88.5% and 92.7%, respectively, with the closely related isolate Marijiniup10 (Table 2). Sequence analysis showed that isolate SA36 shared relatively high similarity with the Australian isolate Marijiniup10, whereas it exhibited lower similarity to the Japanese type isolate. Phylogenetic analysis based on the amino acid sequences of the RdRp and CP regions clustered the isolate SA36 within the same clade as previously reported JINRV isolates, clearly separated from other members of the family Tombusviridae (Fig. 3). Based on the genetic distance (branch length), the isolate SA36 was genetically closer to Marijiniup10 than to the type isolate.
The CP genes of the three Korean JINRV isolates determined in this study shared 98.8-100% nucleotide identity and 98.0-100% amino acid identity with each other (Fig. 4, Supplementary Table 5). Compared to Iris-derived isolates, the Narcissus-derived isolates exhibited 65.9-85.5% nucleotide identity and 71.9-91.5% amino acid identity. The two Iris-derived isolates shared 65.9% nucleotide identity and 71.6% amino acid identity in their CP sequences. The CP sequences of the three JINRV isolates obtained in this study have been deposited in the NCBI GenBank database under the following accession numbers: Replete-SA36 (LC878470), Goblet-SA45 (LC899180), and Tahiti-SA46 (LC899181).
Sap inoculation was performed using JINRV-infected leaves onto 16 indicator plants (Capsicum annuum var. grossum, Citrullus lanatus, Chenopodium amaranticolor, C. quinoa, Cucurbita pepo, Cucumis sativus, Lycopersicon esculentum, Nicotiana benthamiana, N. clevelandii, N. glutinosa, N. occidentalis, N. rustica, N. tabacum cv. KY 57, N. tabacum cv. Turkish, Phaseolus vulgaris, and Vigna unguiculata) (n = 3 per species), alongside mock-inoculated controls. Since no Narcissus sample singly infected with JINRV was identified, the inoculum (JINRV isolate SA42, selected due to sufficient availability of symptomatic leaf tissue) was naturally co-infected with JINRV, NLSYV, NLV, NeLV, SVY and TuYV. At 18 dpi, mild mosaic symptoms were observed only in the systemic leaves of C. pepo (Fig. 5). RT-PCR using specific primers for all six viruses present in the inoculum (Supplementary Table 3, Set 6) and Sanger sequencing confirmed that C. pepo was infected exclusively with JINRV, and the amplicon sequence shared 84.12% nucleotide identity with a previously reported isolate (accession no. JQ807998). No other co-infecting viruses were detected. The JINRV-infected leaf tissues were deposited in the Korean Agricultural Culture Collection (KACC) under the accession number CV251209-30.
JINRV isolates SA36 and Marijiniup10 exhibit a typical Betacarmovirus genome organization but differ from the Japanese type isolate in genome structure and sequence identity. According to the International Committee on Taxonomy of Viruses (ICTV) species demarcation criteria, members of the genus Betacarmovirus are considered distinct species when both RdRp and CP amino acid identities are below 75%. In this study, the CP amino acid identity between SA36/Marijiniup10 and the type isolate was below 75%, whereas the RdRp identity exceeded 75%, indicating that current evidence is insufficient to classify them as distinct species. At present, the three isolates are likely to belong to the same species, while the genome organization observed in SA36/Marijiniup10 may represent a common genome organization of JINRV. A similar case was reported for carnation mottle virus (CarMV, genus Alphacarmovirus), which was initially described as having two amber stop codons, but later isolates were reported to contain only one as additional isolates became available (Harbison et al., 1985; Jo et al., 2015). Whether the genome organization observed in SA36/Marijiniup10 is representative of JINRV will likely be clarified as more JINRV isolates are identified in the future. The relatively low amino acid sequence identity between the Narcissus-derived isolates and the two Iris-derived isolates may suggest host-associated divergence; likewise, this will require further validation as more JINRV isolates become available.
While HTS was utilized for the preliminary identification of viral pathogens, the presence of the viruses was subsequently confirmed through RT-PCR on individual samples. Because all JINRV-positive samples were co-infected, the observed symptoms cannot be attributed to JINRV alone. These symptoms resemble those previously reported for virus-infected Narcissus globally (Berniak et al., 2013; Kim et al., 2025; Mowat et al., 1988; Ward et al., 2009), confounding the identification of JINRV-specific symptoms. Given such frequent co-infections, interactions among co-infecting viruses should be considered. In general, the accumulation of multiple viruses in vegetatively propagated crops causes crop degeneration, reducing yield and quality (Asjes, 1996; Cho et al., 2025; Sutton et al., 1986). Indeed, in lily, another bulbous ornamental crop, more severe symptoms have been reported in co-infected samples compared to single-infected ones (Kim et al., 2019). In this context, considering the infection patterns of samples from the Shinan region, where three to six viruses were simultaneously detected, co-infections including JINRV may contribute to the poor growth and symptoms observed in Narcissus from this region. Although this study primarily focused on JINRV, these findings suggest the need for further studies to clarify the effects of co-infections on symptom development and the overall viral infection status of Narcissus in Korea.
The detection of JINRV in 22 of the 40 samples collected from Shinan is noteworthy, and the virus may have been introduced through imported bulbs. Because current Korean quarantine measures target only six viruses, these imports serve as a primary route for the introduction of novel viruses such as JINRV. Considering the continuous increase in the volume of Narcissus bulb imports to Korea, expanding the scope of viral monitoring for imported bulbs (Lee et al., 2024) is necessary to prevent the inadvertent introduction of unlisted pathogens. While this is admittedly the most probable introduction pathway, the possibility of natural transmission cannot be excluded. Since JINRV is a member of the family Tombusviridae, the potential for soil-borne transmission (Andika et al., 2016), or transmission via an unidentified vector, should also be considered. Without prior records of JINRV in Korea, it remains unclear whether the virus was recently introduced or has already distributed in Shinan. Therefore, further investigation of imported bulbs and additional field surveys are needed to clarify the distribution and introduction pathway of JINRV in Korea.
In conclusion, it is presumed that current Narcissus cultivation in Korea relies on bulbs co-infected with various viruses. Given this situation, although the immediate supply of virus-free bulbs may be difficult, it is crucial for future Narcissus cultivation to prevent the introduction and accumulation of novel viruses such as JINRV, thereby securing bulbs with minimal viral infection or those free from highly pathogenic viruses. These findings provide useful baseline information for viral monitoring of Narcissus bulbs and for future quarantine surveillance in Korea.
PDF Links
PubReader
ePub Link
Full text via DOI
Full text via PMC
Download Citation
Supplement1
Print
