A total of 10 Bletilla striata seedlings with severe disease were collected from two B. striata nurseries of the Chengdu Academy of Agriculture and Forestry Sciences (103°51′26.3412″ E, 30°42′11.7396″ N) in Chengdu, Sichuan Province, China. The root surface was washed with sterile distilled water three times (to remove surface dirt) and dried at room temperature. Pure cultures were obtained from single conidia on PDA plates based on the Chomnunti method (Chomnunti et al., 2014). They were then preserved until identification. Plant and pathogen specimens were deposited in the Herbarium of Sichuan Agricultural University (SICAU) and the Culture Collection of Sichuan Agricultural University (SICAUCC), China, respectively.

Cultures were grown on PDA for 7 days, at 25°C, under 12 h light/12 h dark for recording growth rates, shape, texture and colour of the colonies. For morphological identification, the microscopic characteristics of the pathogen were observed. We observed microscopic characteristics, such as conidia pile, conidiophores, conidiogenous cells, conidia and number of septa, using an Olympus BX43 (Olympus in the Japanese). No fewer than 50 measurements were made for each feature using the Image Frame Work (IFW 0.9.0.7).

The New Plant Genomic DNA Kit (Beijing Aidlab Biotechnologies Co., Ltd., Beijing, China) was used to extract genomic DNA from the fresh fungal mycelium of two representative isolates (SICAUCC 23-0086 and SICAUCC 23-0087). In Fusarium systematics, through NCBI comparison and morphological characteristics description, fungi in this study belongs to the Fusarium solani species complex (Xie et al., 2022). Subsequently, polymerase chain reaction (PCR) amplification was performed for the ribosomal RNA internal transcribed spacer (rDNA-ITS), translation elongation factor 1-alpha region (tef1-α), and RNA polymerase II subunit B (rpb2) (Gräfenhan et al., 2011; Wanasinghe et al., 2021). The primer pairs are listed in Table 1. Polymerase chain reaction (PCR) was performed in 25 μL reaction mixture containing 22 μL Master Mix (Beijing LABLEADBiotech Co., Ltd., Beijing, China), 1 μL DNA template and 1 μL each of forward and reverse (10 μM) primers. The amplification reactions were performed as described by White et al. (1990), O’Donnell et al. (1998), and Liu et al. (1999). PCR products were sequenced at Hangzhou Youkang Biotech Co., Ltd., Chengdu, China. Based on BLAST searches in GenBank and recent publications (O’Donnell et al., 2012; Lombard et al., 2015; Xu et al., 2020), using the ITS, rpb2, and tef1-α and sequence data, reference sequences were downloaded and separate phylogenetic analyses, based on single gene datasets were carried out to initially determine the placement of the species. Information on the taxa used and GenBank accession numbers of our novel species are listed in Table 2. Alignments for the individual locus matrices were generated with the online version of MAFFT version 7.429 (Katoh et al., 2019) and ambiguous regions were excluded using BioEdit version 7.0.5.3 (Hall, 2004). Phylogenetic analyses of Fusarium solani were performed using ITS, rpb2, and tef1-α dataset (Vaidya et al., 2011), and rooted by Fusarium illudens NRRL 22090. A maximum likelihood (ML) phylogenetic tree was constructed as described by Xu et al. (2020). The tree was constructed and edited using FigTree v1.4.2.

Based on Koch’s hypothesis, pathogenicity tests were performed on the pathogen using the method of Yang Z. L. et al. (2022). The test strain was the representative strain SICAUCC 22-0086, and the inoculated materials comprised Bletilla striata seedlings of 12 plants that were 2 years old, healthy, and of consistent size (provided by the Chengdu Academy of Agriculture and Forestry Sciences).

The wheat grains were soaked in water for 12 h, weighed and bagged in about 300 g. The grains were sterilized in an autoclave at 121°C for 30 min. After cooling under aseptic conditions about 10 mycelia discs of 9 mm in diameter were inoculated and mixed with wheat grains, while the control was inoculated with mycelia discs without mycelium. Incubate in an artificial incubator at 25°C for 15 days. When the mycelium was full or wrapped with wheat grains, it was ready to be used. Wheat grains with mycelia were evenly mixed with soil at a ratio of 1:40, and about 1 kg of the soil mixture was weighed and used to plant nine potted Bletilla striata seedlings. As a control, wheat grains without mycelia were mixed with soil in the same manner and used to plant three potted B. striata seedlings. Artificial climate incubator culture conditions were set to 25°C, 12/12 h light/dark, and a humidity of about 52%. Disease symptoms and severity were observed and recorded every 5 days. Three seedlings per treatment and the experiment was replicated three times (n = 9). After symptoms appeared, diseased root samples were taken for re-isolation and identification of the pathogen.

To explore the effects of ten diverse culture media on the mycelial growth and sporulation of representative strain SICAUCC 22-0086, a 9 mm activated colony was placed in the center of ten media. This fungus inoculated, respectively, on potato dextrose agar (PDA), potato saccharose agar (PSA), corn meal agar (CMA), czapek-dox medium (Czapek), chalmers agar modified (modified Richard), peberdy, modified frey medium (modified Fries), extract agar (MEA), oatmeal agar (OA) czapek doxsolution agar (CDA), synthetic low nutrient agar (SNA) and incubated under sterile conditions for 7 days at 25°C and 12/12 h light/dark conditions. The colony diameter was measured to assess mycelial growth, and the colony characteristics [including shape, color, edge features, texture (gloss), and elevation] were observed and recorded. Additionally, sterile water was dropped into each plate to form a spore suspension, and sporulation was then assessed using a hemocytometer plate. There were three tests per treatment and three replicates (n = 9) per test.

To explore the effects of various carbon and nitrogen sources on mycelial growth and sporulation (Song et al., 2020; Li et al., 2021), five carbon sources (glucose, sucrose, maltose, starch, and lactose, plus no carbon source as a control) and eight nitrogen sources (potassium nitrate, sodium nitrate, ammonium sulfate, ammonium nitrate, urea, peptone, beef paste, and yeast extract, plus no nitrogen source as a control) were tested. A 9 mm activated colony was placed in the center of the medium (using modified Richard medium as the basic medium) and incubated under sterile conditions for 7 days at 25°C and 12/12 h light/dark conditions. There were three tests per treatment and three replicates (n = 9) per test. Mycelial growth and sporulation were measured as described in section 2.4.

To explore the effect of temperature on mycelial growth and sporulation, a 9-mm activated colony was placed in the center of a PDA plate and incubated for 7 days at temperatures of 15°C, 20°C, 25°C, 30°C, and 35°C under aseptic and 12/12 h light/dark conditions, respectively. To explore the effect of the photoperiod on mycelial growth and sporulation, a 9 mm activated colony was placed in the center of a PDA plate and incubated under sterile conditions for 7 days at 25°C under full light (24 h light), full darkness (24 h dark), or alternating light and dark (12/12 h light/dark). To explore the effect of pH on mycelial growth and sporulation, a 9 mm activated colony was placed in the center of a PDA plate and incubated for 7 days at aseptic, 25°C, and 12/12 h light/dark conditions at pH 3, 4, 5, 6, 7, 8, and 9. There were three tests per treatment and three replicates (n = 9) per test. Mycelial growth and sporulation were measured as described in section 2.4.

The inhibition rate of seven fungicides (which are detailed in Table 3) was determined by the mycelial growth rate method (Xin et al., 2020). First, 1 × 10⁴ mg/L active ingredient solution was prepared and diluted to 10, 50, 100, 200, 400 and 800 μg/mL with sterile water and mixed with PDA medium (1:9 ratio). PDA medium without a fungicide was used as the blank control. A 9 mm activated colony of representative strain SICAUCC 22-0086 was placed in the center of each fungicide-containing/blank control PDA plate and cultured at 25°C for 7 days. There were five replicates per fungicide concentration. Mycelial growth and sporulation were measured as described in section 2.4.

The following formula was used to determine the antifungal rate of each fungicide at each concentration (where D is the control colony diameter and d is the treatment colony diameter):

Microsoft Excel 2016 was used to calculate a toxicity regression equation (where x is the logarithm of the fungicide concentration (μg/mL) and y is the inhibition rate) and correlation coefficient for each fungicide. DPS v2.0 was used to calculate the half maximal effective concentration (EC₅₀) value of each fungicide.

The approximate ranges of antimicrobial rate of different agent concentrations were initially obtained through a one-way test, and the three fungicides with the highest inhibition rates were selected as the three factors in the response surface methodology (RSM) analysis using the Box–Behnken design, with antimicrobial rate (based on colony diameter) as the response value. From there, the factors and level values required for the orthogonal test were determined, and then the orthogonal test design was used to further optimize the optimal fungicide concentration scheme.

The data were analyzed using SPSS v24.0, and the ANOVA was done after conforming to the normal distribution test, means were analyzed for significance of differences using Duncan’s new complex polar method, and labeled using sequential letter labeling. Toxicity regression equations, correlation coefficients, and EC₅₀ values were generated using Microsoft Excel 2016 and DPS v2.0. Graphs were constructed using GraphPad Prism v8. RSM was performed using Design-Expert v12.

Bletilla striata plants with typical white root rot symptoms (Figure 1) were observed in plant nurseries of the Chengdu Academy of Agriculture and Forestry Sciences in Chengdu, Sichuan Province, China. All symptoms were observed in both plant nurseries surveyed. Field observation demonstrated that the underground roots had severe decay and necrosis. Their roots are covered with a mass of white filamentous to arachnoid mycelia. A white radial mycelia or rhizomorph develop on the old/main root, displaying a relatively loose and soft structure. The rhizomorph could expand into the soil, become thinner, and sometimes fill the gaps in the soil. The mycelia passed through the cortex and cambium, entering deep into the xylem, resulting in full root decay, leaf narrowing, gradual leaf yellowing, early leaf shedding and, ultimately, death of the entire plant. The incidence of B. striata white root rot was about 5%.

Ten Bletilla striata samples with typical symptoms were collected from the two plant nurseries. We successfully isolated 10 Fusarium spp. (identified based on morphology) from the 10 diseased B. striata plants assessed.

Indoor culture characteristics: Colonies on PDA in the dark at 25°C reached a diameter of 70–75 mm after 7 days. Colonies regularly circular; reverse side white; filiform to arachnoid white mycelia (Figure 2). Sporulation from conidiophores formed on aerial mycelia. Conidiophores simple or branched one to several times, with each branch bearing a single terminal monophialide. Conidiogenous cells monophialidic, cylindrical, and could produce microconidia and macroconidia. Microconidia formed on aerial conidiophores, clustered in false heads at monophialide tips, and hyaline, oval or obovoid, symmetrical or gently bent dorsoventrally, smooth, and thin-walled, 0(−1) septate. 0-septate conidia: 9.07–18.22 × 3.83–8.26 μm (x̅ = 15.6 × 6.3 μm, n = 50); 1-septate conidia: 23.14–37.52 × 7.83–10.26 μm (x̅ = 29.6 × 9.3 μm, n = 50). Macroconidia hyaline, clavate to falcate, slight to moderate dorsal curvature, elongate and slender, 2–4 septate, predominantly 3 septate. 2-septate conidia: 48.26–53.74 × 8.59–9.52 μm (x̅ = 49.5 × 9.2 μm, n = 50); 3-septate conidia: 50.14–58.36 × 8.75–11.33 μm (x̅ = 52.6 × 10.2 μm, n = 50); 4-septate conidia: 55.78–62.04 × 9.05–11.5 μm (x̅ = 57.2 × 9.3 μm, n = 50); contained oil drops, apical cells blunt or papillate, sporadically hooked, basal cells indistinct. Chlamydospores not observed.

Based on these morphological characteristics, the isolates matched the description of the genus Fusarium (Li et al., 2017; Crous et al., 2021; Han et al., 2023). Detailed morphological observation indicated that the 10 strains had completely consistent morphological characteristics with each other. To further validate this result, we selected two representative strains (SICAUCC 23-0086 and SICAUCC 23-0087) for phylogenetic analysis.

Molecular identification of two representative isolates (SICAUCC 23-0086 and SICAUCC 23-0087) was required to verify the accuracy of the morphological identification. Alignment against the NCBI GenBank database using BLASTn indicated that the ITS sequence from both strains shared 99.49% identity with that of Fusarium solani CBS 140079 with 100% query coverage, indicating that these strains could be a member of the F. solani species complex. Next, the rpb2 and tef1-α gene fragments were sequenced. The identity of each gene sequence (ITS, rpb2 and tef1-α) of the two isolates was 100%.

To identify the two isolates more accurately, a phylogenetic tree based on the concatenated sequences of ITS, rpb2, and tef1-a was constructed (Figure 3). Fusarium illudens NRRL 22090 was used as an outgroup and sequences from 47 taxa were included. The alignment contained 3,951 base pairs (ITS = 1,198, rpb2 = 2009, tef1-α = 744), including gaps. The phylogenetic tree showed that the two isolates clustered with F. solani NRRL43474 and F. solani NRRL43468. Bootstrap support values at the nodes were 100%. Therefore, the two isolates were identified as F. solani based on both morphological and molecular evidence.

Mycelial discs of representative strain SICAUCC 22-0086 cultured for 7 days were used to inoculate wheat grains, which were then placed in the dark for 15 days. The mixture was added to soil, which was used to plant potted Bletilla striata seedlings (Figure 4). On day 5, the inoculated plants exhibited disease symptoms such as brown lesions, narrow leaves, and wilted tips (Figures 4A1,B1), similar to the symptoms observed in the field. There were a few white mycelia on the surface of the roots (Figures 4C1,D1). On day 10, the brown necrotic leaf lesions had gradually expanded, some leaves had withered or even shed (Figures 4A2,B2), the roots appeared rotten and were covered with white mycelia (Figures 4C2,D2), and the plant growth was significantly weakened. On day 15, all inoculated plants had wilting leaves (Figures 4A3,B3), the root epidermis and cortex were decaying, internal tissues were dark brown, and many white mycelia covered the rotting tissues (Figures 4C3,D3), similar to the symptoms observed in the field. In contrast, all non-inoculated plants remained healthy on day 15 (Figures 4A4–D4).

Next, diseased Bletilla striata seedling leaves and roots were randomly selected for pathogen re-isolation and identification. The re-isolated pathogens were consistent with the inoculated pathogens, fulfilling Koch’s postulates. Thus, we identified Fusarium solani as the pathogen causing white root rot in B. striata.

The pathogen was able to grow and sporulate on all 10 media, but with significant differences in 7-day colony diameter and sporulation (Figure 5A). PDA was ranked first for growth (colony diameter: 74.5 mm, p < 0.05 vs. other treatments). Peberdy (colony diameter: 71 mm) and CMA (colony diameter: 69 mm) both ranked second (p > 0.05) for growth. MEA was ranked last for growth (colony diameter, 56 mm, p < 0.05 vs. other treatments).

PDA was ranked first for sporulation (2.74 × 10⁹ conidia/mm², p < 0.05 vs. other treatments). OA was ranked second for sporulation (2.25 × 10⁹ conidia/mm², p < 0.05 vs. other treatments). MEA (0.63 × 10⁹ conidia/mm²) and PSA (0.6 × 10⁹ conidia/mm²) were both ranked last (p > 0.05) for sporulation.

Based on both the colony diameter and sporulation, the optimum medium for the pathogen was PDA.

The pathogen grew and sporulated on the five carbon sources plus the control (Figure 5B). Sucrose was ranked first for both growth (colony diameter, 73 mm, p < 0.05 vs. other treatments) and sporulation (2.36 × 10⁹ conidia/mm², p < 0.05 vs. other treatments). Glucose was good for growth (colony diameter, 71 mm, p < 0.05 vs. other treatments), but its effect on sporulation was middling (1.31 × 10⁹ conidia/mm², p < 0.05 vs. other treatments). Control (no carbon source) was ranked last (colony diameter, 24 mm, p < 0.05 vs. other treatments; sporulation, 0.17 × 10⁹ conidia/mm², p < 0.05 vs. other treatments). Overall, sucrose was the optimum carbon source.

The pathogen grew and sporulated on the eight nitrogen sources plus the control (Figure 5C). Potassium nitrate (colony diameter, 73 mm), yeast extract (colony diameter, 72 mm), and sodium nitrate (colony diameter, 69 mm) were all ranked first (p > 0.05) for growth. Sodium nitrate (2.1 × 10⁹ conidia/mm²) and yeast extract (1.95 × 10⁹ conidia/mm²) were ranked first (p > 0.05) for sporulation. Urea (colony diameter, 62 mm) and peptone (colony diameter, 65 mm) were both ranked second (p > 0.05) for growth. Control (no nitrogen source) was ranked last (colony diameter, 16 mm, p < 0.05 vs. other treatments; 0.21 × 10⁹ conidia/mm², p < 0.05 vs. other treatments). Overall, sodium nitrate and yeast extract were the optimum nitrogen sources.

The pathogen grew and sporulated at all temperatures tested (15–35°C), but with significant differences in colony diameter and sporulation (Figure 6A). At 25°C, growth was optimum (colony diameter, 72 mm, p < 0.05 vs. other treatments), as was sporulation (2.6 × 10⁹ conidia/mm², p < 0.05 vs. other treatments). At 15°C (colony diameter, 31 mm; 0.2 × 10⁹ conidia/mm²) and 35°C (colony diameter, 14 mm; 0.09 × 10⁹ conidia/mm²), growth and sporulation were low. Therefore, 25°C was the optimum growth temperature, and lower and higher temperatures were unsuitable.

The pathogen grew and sporulated under all photoperiods tested (Figure 6B). Both 24 h light (colony diameter, 72 mm; 2.55 × 10⁹ conidia/mm²) and 24 h dark (colony diameter, 74 mm; 2.85 × 10⁹ conidia/mm²) were ranked first (p > 0.05) for both growth and sporulation. Overall, light vs. dark did not significantly affect growth or sporulation.

The pathogen grew and sporulated at all pH values tested (Figure 6C). At pH 7 and 8, growth was optimum (colony diameter, 73 and 71 mm, p < 0.05 vs. other treatments) and at pH 7 sporulation was optimum (2.37 × 10⁹ conidia/mm², p < 0.05 vs. other treatments). Excessive acid and alkali were not conducive to growth or sporulation. Overall, the optimum pH was 7.

Different fungicide types and concentrations led to different inhibition and sporulation rates of representative strain SICAUCC 22-0086 after culture on fungicide-containing PDA for 7 days (Figures 7, 8). At 200 μg/mL, difenoconazole had the highest inhibition rate (79.25%, p < 0.05 vs. other fungicides; sporulation, 0.23 × 10⁹ conidia/mm²). At 400 μg/mL, difenoconazole had the highest antifungal rate (80.11%; sporulation, 0.21 × 10⁹ conidia/mm²), followed by pyraclostrobin (78.18%; sporulation, 0.27 × 10⁹ conidia/mm²) and thiophanate-methyl (78.05%; sporulation, 0.24 × 10⁹ conidia/mm²) (p < 0.05 between these three vs. other fungicides). Chlorothalonil had the lowest inhibition rate (41.55%, p < 0.05 vs. other fungicides; sporulation, 1.33 × 10⁹ conidia/mm²). At 800 μg/mL, difenoconazole had the highest antifungal rate (81.74%; sporulation, 0.15 × 10⁹ conidia/mm²; p < 0.05 vs. other fungicides), followed by pyraclostrobin (80.82%; sporulation, 0.25 × 10⁹ conidia/mm²), thiophanate-methyl (80.03%; sporulation, 0.21 × 10⁹ conidia/mm²), and iprodione (79.45%; sporulation, 0.23 × 10⁹ conidia/mm²), p < 0.05 between these four vs. other fungicides. Chlorothalone had the lowest inhibition rate (48.86%, p < 0.05 vs. other fungicides; sporulation, 0.87 × 10⁹ conidia/mm²). In conclusion, among the seven fungicides tested, difenoconazole had the best inhibition rate, followed by pyraclostrobin and thiophanate-methyl, while chlorothalonil had the worst inhibition rate.

For each fungicide used in the fungicide assays, the original fungicide concentration and formulation type, toxicity regression equation (where x is the logarithm of the fungicide concentration (μg/mL) and y is the inhibition rate), correlation coefficient, and EC₅₀ value are shown in Table 4. The best fungicide was 10% difenoconazole WG (EC₅₀, 142.773 μg/mL), followed by 70% thiophanate-methyl WP (EC₅₀, 183.924 μg/mL) and 25% pyraclostrobin EC (EC₅₀, 235.593 μg/mL). 75% chlorothalonil WP was the worst (EC₅₀, 693.091 μg/mL). In summary, difenoconazole, thiophanate-methyl, and pyraclostrobin effectively suppressed the colony diameter, with difenoconazole exhibiting the most profound impact.

In the RSM analysis, the concentration factor levels of three agents, difenoconazole, pyrrologlucoside and thiophenate-methyl, were set based on the results of a one-way experiment. The inhibition rate was used as the response value to optimize these factor levels. Design-Expert v8.0.6 was used for RSM (including statistical analysis of variance, regression coefficients, and regression equations).

RSM yielded the following quadratic multinomial regression equation (where A is difenoconazole, B is pyraclostrobin, and C is thiophanate-methyl):

The high F-value of the model (75.38) was highly significant (p < 0.001) (Table 5). C, A, and C² had highly significant effects on inhibition rate (p ≤ 0.001), while B and B² also significantly affected the inhibition rate (p < 0.05). Based on these results, the order of factors affecting the inhibition rate was as follows: thiophanate-methyl (C) > difenoconazole (A) > pyraclostrobin (B).

The coefficient of determination (R²) was 0.9898 and the adjusted coefficient of determination (adjusted R²) was 0.9767 (Table 6), indicating that the model was sufficient to represent the relationship between the independent and response variables. Therefore, the model was statistically reasonable.

The optimum fungicide concentration scheme for inhibition rate was determined by RSM (Figure 9). The optimum concentrations of difenoconazole, pyraclostrobin, and thiophanate-methyl were 395.42 μg/mL (Figure 8A), 781.03 μg/mL (Figure 8B), and 561.11 μg/mL (Figure 8C), respectively. Under this scheme, the inhibition rate was predicted to be 91.78 ± 0.18%. In practice, under this scheme, the inhibition rate was 92.24 ± 0.34% (based on three replicates), which was consistent with the prediction, thereby confirming the reliability of RSM.