Viroids are the smallest plant pathogens with single-stranded, circular RNA molecules (246–401 nt). These pathogens do not code for proteins and belong to two families, Pospiviroidae and Avsunviroidae (Di Serio et al. 2014). The genus Pospiviroid in the family Pospiviroidae comprises ten species, one of which is columnea latentviroid (CLVd) (ICTV, 2024).
CLVd was originally isolated from asymptomatic Columnea erythrophae leaves in the USA (Owens et al. 1978). The natural hosts of CLVd include members of the Generiaceae and Solanaceae families, such as Brunfelsia undulata, Gloxinia spp., Nematanthus wettsteinii, Solanum stramonifolium, and tomatoes (S. lycopersicum) (Nielsen and Nicolaisen 2010; Singh et al. 1992; Spieker 1996; Tangkanchanapas et al. 2013; Verhoeven et al. 2004). In addition, chili pepper (Capsicum annuum), cucumber (Cucumis sativus), dahlia (Dahlia sp.), eggplant (S. melongena), marigold (Tagetes spp.), petunia (Petunia × hybrida), pot marigold (Calendula officinalis) and potato (Solanum tuberosum) have been reported as experimental host of CLVd (Hammond et al. 1989; Matsushita and Tsuda 2015, 2016; Verhoeven et al. 2004). CLVd is asymptomatic in some species of Solanaceae. However, symptoms such as leaf chlorosis, leaf epinasty, fruit distortion, and stem and leaf necrosis are apparent in tomatoes and potatoes, similar to those caused by potato spindle tuber viroid (PSTVd) (Owens et al. 1978). CLVd is mostly transmitted mechanically, as well as through seeds and pollen (Bhuvitarkorn and Reanwarakorn 2019; Tangkanchanapas et al. 2021). Since CLVd is known to be transmitted through seeds in tomato, with transmission rates of 0, 5.3, 50, and 100% depending on the cultivar (Matsushita and Tsuda 2016), representing one of the most significant transmission pathways for the viroid’s expansion. Additionally, Constable et al. (2019) detected CLVd in commercial tomatoes and pepper seeds. Thus, there is a possibility that the viroid has spread to other regions and countries through these contaminated seeds. CLVd has been reported in Canada, Costa Rica, Europe (Italy, France, and the Netherlands), Mali, and the USA (CABI 2023; EPPO 2023). Additionally, in Asia, CLVd was found several times in tomatoes and S. stramonifolium in Thailand (Tangkanchanapas et al. 2005, 2013, 2021) and in tomatoes and peppers in Vietnam (Choi et al. 2020). In the latest report that was used CLVd isolates in Thailand, CLVd isolated before 2014 in Thailand have been divided into severely pathogenic and mildly pathogenic groups (Tangkanchanapas et al. 2021). The mildly pathogenic group was the first CLVd isolate reported in Thailand that was eradicated. Since then, it has been unclear whether CLVd continues to occur in Thailand, as no outbreaks have been reported since 2014.
In this study, we report the detection of CLVd in tomato seeds produced in Thailand in 2021 during the 2022 plant quarantine for export inspection in Japan. The phylogenetic characteristics were analyzed based on a phylogenetic tree that was compared among our detected CLVd variants and other CLVd variants, including previously detected variants in Thailand, and the viability and pathogenicity of CLVd detected in the seeds were investigated through inoculation tests.
Among export inspections in Japan, 3,000 tomato seeds produced in Thailand were tested to confirm the absence of pospiviroid contamination. Total RNA was extracted from the seed samples as described by Yanagisawa et al. (2012). Reverse transcription-quantitative real-time PCR (RT-qPCR), which enables the detection of eight pospiviroids (Citrus exocortis viroid (CEVd), chrysanthemum stunt viroid (CSVd), PSTVd, CLVd, tomato apical stunt viroid (TASVd), tomato chlorotic dwarf viroid (TCDVd) and tomato planta macho viroid (TPMVd)) reported by Yanagisawa et al. (2017), was used to detect the viroids from each sample; each detection was performed in duplicate. PCR amplification was confirmed for each sample using SYBR Green with a CLVd-specific primer set (CLV-F1/R3). The cycle threshold (Ct) values were 36.0 and 36.7, and also the melting temperature (Tm) values were 85.0 and 85.1 ℃. At the time, the Tm values of positive control artificially synthesized based on the sequence of CLVd (GenBank/EMBL/DDBJ accession no. AY372392, Verhoeven et al. 2004) were both 84.7. Based on these results, the sample was determined to be CLVd positive. In contrast, the other pospiviroids were not detected using SYBR Green with a universal primer set that enabled the detection of six pospiviroids (CEVd, CSVd, PSTVd, TASVd, TCDVd, and TPMVd) and a PCFVd-specific primer set.
After the detection, to determine its complete sequence, direct sequencing analysis was carried out in triplicate using PCR products obtained from the primer set (Vid-FW/RE) (Verhoeven et al. 2004) and the new designed primer set (CLV-F22:5’-TTCGGGTTTCCTTCCTCTGC-3’ (nucleotide positions 200- 219 with reference to CLVd, No. JF446929) / CLV-R22:5’- CGCTCTTCCTCCCGTTC -3’ (nucleotide positions 63- 79)). RT-PCR was performed using the OneStep RT-PCR Kit (Qiagen, Hilden, Germany). Both PCR programs of these two primer sets included 30 min at 50 °C 15 min at 95 °C; 40 cycles of (30 s at 94 °C; 30 s at 55 °C; 1 min at 72 °C); 10 min at 72 °C. Amplified PCR products were separated by electrophoresis on 3% (w/v) agarose gels and stained with GelRed (Cosmo Bio, Tokyo, Japan). The length of each PCR product was then determined. A BLAST search revealed that the complete 370 nt sequence of the CLVd isolate (CLVd-TH, No. LC832423) obtained from the PCR products presented the highest identity (98.9%) with a CLVd variant (No. JF446929, Tangkanchanapas et al. 2021) that was isolated from tomatoes in Thailand in 2009.
Subsequently, to investigate the phylogenetic relationship between CLVd-TH and the previously detected CLVd variants in Thailand, a phylogenetic tree was constructed based on the complete genome sequences using the Maximum Likelihood (ML) method with 1,000 bootstrapping replicates performed in the MEGA version X program (Kumar et al. 2018) (Fig. 1). The results showed that CLVd-TH clustered with CLVd variants in Thailand (group B). Meanwhile, the CLVd variant (No. JF446929), which showed the highest similarity to CLVd-TH in the BLAST search, was grouped in a separate cluster (group A). Additionally, the cluster that CLVd-TH belongs to includes CLVd variants with severe pathogenicity (Tangkanchanapas et al. 2021).
Next, to confirm whether the detected CLVd maintained viability on tomato seeds, the inoculation was performed to tomato (cv. Rutgers) plants using extracted nucleic acids that were used for RT-qPCR detection of pospiviroids as the inocula. CLVd was mechanically inoculated into the young leaves of these seedlings as described by Yanagisawa and Matsushita (2017), and seven seedlings from each plant were inoculated. After inoculation, these plants were grown in a greenhouse at 23–25 ℃ with natural light. At the same time, the CLVd variant (No. AY373446, Verhoeven et al. 2004), isolated in the Netherlands, was also inoculated into tomatoes to compare its pathogenicity. One month after the inoculation, viroid infection was assessed in the uppermost leaves of each inoculated plant. Using the method described above, we confirmed the viroid infection in each inoculated plant and the viroid sequence after inoculation. RT-PCR and RT-qPCR analyses showed that five of the seven inoculated tomato plants tested positive. The tomato plants exhibited viroid-like symptoms such as severe stunting, leaf chlorosis, and leaf deformation, followed by necrosis of the stems and leaf veins (Fig. 2A, D). In addition, tomato plants inoculated with CLVd-TH were more virulent than those inoculated with the CLVd variant (No. AY373446) (Fig. 2A, B). According to Tangkanchanapas et al. (2021), the CLVd variant (no. AY373446) belongs to a group of strains that exhibit mild-to-intermediate symptoms. This demonstrates that CLVd-TH is highly virulent, similar to severe pathogenic variants. However, the CLVd sequence remained unchanged after inoculation. These results confirmed that CLVd maintained its viability on tomato seeds.
Despite CLVd being reported a few times in Thailand until 2014 (Tangkanchanapas et al. 2005, 2013, 2021), the fact that CLVd was detected in tomato seeds in 2021 suggests that CLVd will continue to persist in Thailand. Moreover, there were no CLVd variants in the NCBI database that completely matched CLVd-TH based on the BLAST results. Additionally, CLVd-TH belongs to a cluster that includes CLVd variants isolated from Thailand, including known variants with high pathogenicity (Fig. 1). Therefore, CLVd-TH is a new variant related to CLVd variants occurring in Thailand, and there may be a wide diversity of CLVd in Thailand. Furthermore, because the pathogenicity of CLVd-TH on the tomato plants was indicated as severe (Fig. 2), the potential damage of CLVd-TH spread would be enormous.
The Pospiviroidae genome is organized into five structural and functional domains: terminal left (TL), pathogenicity (P), central conserved (C), variable (V), and terminal right (TR) (Keese and Symons 1985). Tangkanchanapas et al. (2021) classified CLVd variants into five groups, each characterized by pathogenicity based on the V domain. The CLVd variant (No. JF446929) was classified into the mild group detected in Thailand. Comparison of the full genome of CLVd-TH with that of the CLVd variant (No. JF446929) showed that CLVd-TH had two (G142 → T, deletion of A144) mutations in the V domain and two (A170 → C, A173 → T) mutations in the TR domain. Additionally, a comparison of CLVd-TH with the CLVd variant AM698094, which belongs to the same group as the CLVd variant JF446929, showed that two mutations occurred (G142 → T, deletion of A144); however, the TR domain was unchanged. Mutations within a few nucleotides determine viroid virulence. For example, in PSTVd, nucleotides in the P domain are related to symptom expression (Owens et al. 1995), and it has been reported that a single base pair in the TR domain of TPMVd is associated with virulence (Li et al. 2017). These results suggested that the V domain has a different function in pathogenicity.
Because the CT values were relatively large when CLVd-TH was detected by RT-qPCR, it was expected that the number of viroids on the seeds would not be large. Nevertheless, we confirmed the viability of CLVd-TH in the seeds. Verhoeven et al. (2021) reported that viability could not be confirmed for viroids grown on tomato seeds. These differences might be due to the storage environment of the seed and viroid infection in the seed. Viroids can maintain viability by infecting the ovule inside the seed (Matsushita et al. 2018). In addition, Matsushita and Tsuda (2016) reported seed transmission of CLVd in tomato plants, suggesting the potential for seed transmission in this case. Moreover, Constable et al. (2019) detected pospiviroids, including CLVd, in commercial seeds, and found that the highest positive rate of tomato and pepper seed lots imported into Australia was recorded in 2013, followed by a decline. However, as in the present case, pospiviroids continue to be found. Therefore, continued testing of these viroids is necessary to prevent their spread to new areas.