First report of causing bulb rot on in China.

Lilium lancifolium Thunb., commonly known as Juandan lily and tiger lily, is widely cultivated in China for its edible bulbs and medicinal properties, with a commercial value worth of ~RMB 6 billion Yuan per year. Bulb rot is an increasingly common disease on L. lancifolium, significantly impacting both the quantity and quality of the main product, the scaled bulbs. Typically, the causal pathogens invade the plant through wounds in the root or the ends of the bulb, causing the roots and bulb to brown and rot, which can eventually lead to stem wilt and death of the whole plants. During pathogenesis, the infected bulbs typically turn from white to brown, with sunken lesions and later the scales flaking off from the base of the bulb (Figure 1A and 1B). Plants growing from infected bulbs are generally short, with discolored leaves, wilting, and death at an early stage. Bulb rot is commonly observed in fields with excess water and a history of continuous Juandan lily cultivation. For this study, wilted L. lancifolium plants with rotted bulbs were collected from Longshan in Hunan, Enshi in Hubei, Yixing in Jiangsu, and Lu'an in Anhui in 2018 and 2019. Infected bulbs were surface sterilized with 75% ethanol for 30 seconds, followed by disinfection with 2% sodium hypochlorite for 5 minutes, and then rinsing with sterile water three times. The surface-sterilized tissue was divided into small pieces of 0.5 × 0.5 cm in size, placed on potato dextrose agar (PDA) medium containing 50 mg/l streptomycin sulfate, and incubated at 25℃. Mycelia growing from diseased tissues were sub-cultured onto fresh PDA medium to obtain pure culture, which formed dense white hyphae after a few days (Figure 1C and 1D). Colonies on PDA produced abundant condia about 15 days after subculturing. Microconidia were abundant, solitary, thin walled, hyaline, ovoid, 0 to 1 septate, with an average size of 6.1 × 2.6 μm (n=50) (Figure 1E). Macroconidia had a curved apical cell and foot-like basal cell with 3 to 5 septa, with an average size of 35.4 × 4.3 μm (n=30) (Figure 1E). No chlamydospore was observed. These morphological characteristics of the causal pathogen were similar to those of Fusarium spp. (Leslie et al., 2006). To identify the Fusarium isolates to species level, DNA fragments of the internal transcribed spacer (ITS) regions of the ribosomal RNA gene cluster, translation elongation factor subunit 1-alpha (TEF1-α), and RNA polymerase II subunit 2 (RPB2) genes were amplified using primers ITS1/ITS4, EF1/EF2, and 7cF/11aR respectively and sequenced (Choi et al. 2018; Jiang et al. 2018; Choi et al. 2017). BLAST analyses showed that the ITS (GenBank Accession No. MT549849), TEF1-α (GenBank Accession No. MT553348), and RPB2 (Accession No. MW201686) sequences of our isolates shared the highest sequence identities (98-100%) with those of F. fujikuroi reference strains in GenBank. A phylogenetic tree showing the relationship between one of our strains, S106, and those of the closely related species within the F. fujikuroi species complex was constructed by the maximum likelihood method using MEGA X (Kumar et al. 2018) (Figure 2). Based on the morphological characteristics and DNA sequences, the strains were identified as F. fujikuroi sensu stricto. We used two methods, an ex vivo assay using Juandan lily bulb scales and an in vivo assay using potted Juandan lily plants, to confirm pathogenicity for one representative F. fujikuroi strain from each of the four geographic regions to fulfill Koch's postulates (Bian et al. 2016; Zeng et al. 2019). In the ex vivo assay, actively growing mycelia on PDA plates were cut into 5mm diameter fungal blocks as inocula. To prepare healthy Juandan lily bulb scales as test tissues, healthy fresh scales were first surface sterilized using 75% alcohol for 30 seconds, followed by treatment of 2% sodium hypochlorite for 5 minutes, and then rinsed with sterile water 3 times. The scales were punctured with sterilized dissecting needles, the 5mm mycelial blocks containing the PDA medium were then inoculated on the punctured wound of the scales. Sterile PDA culture medium without mycelia was inoculated on the punctured wound as a negative control. After inoculation, Juandan lily scales were placed in sterile culture dishes with two layers of sterilized filter paper and 5ml of sterile water in each dish. Six Juandan lily scales were placed in each dish, with different treatments placed in different dishes, and the dishes were placed in an incubator in the dark at 25℃. After 10 days of incubation, we found that the F. fujikuroi-inoculated Juandan lily bulb scales showed disease symptoms (brownish lesion) similar to those in the original field collected infected bulb samples (Figure 1F). However, such symptoms were not observed in the negative control group. The pathogenicity test was performed 3 times for each isolate, each with six repeats. In the in vivo pathogenicity test using potted lily plants, we prepared actively growing cultures of our F. fujikuroi strains by incubating them in a liquid medium, the potato dextrose broth, for 3 days in a shaker-incubator at 25℃ and 180rpm. The asexual spores conidia from the fungal cultures were harvested by filtration through eight layers of sterile cheese clothes and with spore concentrations adjusted to 1×107 conidia per ml. Healthy Juandan lily bulbs were selected and one bulb was planted in each pot containing sterilized soil. Each pot was inoculated with 1ml conidia suspension, at the base soil where the bulbs were planted. The pots were placed in a growth chamber at 25℃ with a 12 h light and 12 h dark cycle. Symptoms similar to those observed in diseased bulbs in the field were observed, with symptoms at 30 days after inoculations shown in Figure 3. Specifically, most of the roots, bulb plate and scale tissues of Juandan lily plants inoculated with F. fujikuroi conidia were rotten and turned black, with few new roots. In addition, the infected plants showed stunted growth (Figure 3). In contrast, the uninoculated plants grew normally, with dense new roots and healthy-looking bulbs, and no rot symptom (Figure 3). The fungi were re-isolated from the infected Juandan lily tissues from both pathogenicity assays, following the procedures described above for isolating and identifying the fungal cultures from infected field samples. These re-isolated fungi were shown to have colony morphology and DNA sequences at the three loci identical to those of our inoculated F. fujikuroi strains. Several Fusarium species have been reported as pathogens of lily plants in China, including F. oxysporum, F. solani and F. tricinctum (Li, et al., 1995; Li, et al., 2013). In addition, F. redolens has been reported previously in ornamental lily in Ukraine (Zerova, 1940). Indeed, Fusarium moniliforme, one of the disused synonyms of F. fujikuroi (Seifert et al. 2003), has been reported as a causal agent for diseases in lily. However, it's now known that the originally defined F. fujikuroi sensu lato is in fact a large species complex consisting of over 60 recognized species, including F. fujikuroi sensu stricto (Moussa et al. 2017; Choi et al. 2018). In addition, there are over 100 species in the genus Lilium as well as many other species with their common names including the word "lily" but are not in the Lilium genus. To our knowledge, this is the first confirmed report of bulb rot of Juandan lily L. lancifolium caused by F. fujikuroi sensu stricto in China. Our result should help with future monitoring and control of Juandan lily diseases.