Identification of Fusarium oxysporum f. sp. melonis Race 2 in Central Taiwan and Its Potential Biocontrol Agent
Article information
Abstract
In 2022, a wilt disease was found in the melon in central Taiwan, resulting in severe yield losses. To identify the causal agent, both morphological and molecular identification were conducted. Together with the results of host range tests, the pathogen was identified as Fusarium oxysporum f. sp. melonis race 2 (FOM). To develop a sustainable and eco-friendly method for disease management, the culture broth and culture filtrate of two potential biocontrol agents, Bacillus mycoides strains BM02 and BM103, were evaluated against FOM in a greenhouse. The results revealed that B. mycoides strain BM02 consistently and significantly (P < 0.05) reduced the disease when applied via foliar spraying or soil-drenching, compared to the water control. To our knowledge, this is the first report confirming FOM race 2 in Taiwan and the biocontrol results suggested that BM02 could be promising for managing melon Fusarium wilt.
Melon (Cucumis melo) is affected by several diseases, with melon Fusarium wilt (MFW) being one of the most damaging diseases in Taiwan. The pathogen, Fusarium oxysporum f. sp. melonis (FOM), is a well-known species affecting melon crops worldwide (Schreuder et al., 2000; Zuniga et al., 1997). The annual global loss caused by MFW was estimated around 10–30% and sometimes up to 100% (Sadeghpour et al., 2023). FOM can produce three types of asexual spores: microconidia, macroconidia, and chlamydospores, primarily attacking melon plants from flowering to the fruiting stages. Chlamydospores of the pathogen can survive in the soil, germinate under favorable environmental conditions, and penetrate the roots via the root tips, natural openings, or wounds. The pathogen then colonizes the xylem vessels, producing polysaccharides and inducing tyloses, clogging the vascular system and leading to chlorosis and wilting symptoms. Field transmission of the disease mainly occurs through irrigation water or contaminated equipment, making the disease difficult to control (Luongo et al., 2012; Zhao et al., 2011).
Currently, four races (0, 1, 2, and 1.2) have been defined based on the host resistance genes overcome by pathogen variants. Two dominantly inherited resistance genes, Fom-1 and Fom-2, provide long-lasting resistance to races 0, 2, and 0, 1, respectively, but not to race 1.2. However, germplasms lacking Fom genes are susceptible to all races. Race 1.2, the most virulent and yield-limiting race, was further divided into the race 1.2 wilting strain and the race 1.2 yellowing strain based on the symptoms they cause. The resistance to race 1.2 is complex and appears to be controlled by multiple recessive genes (Chikh-Rouhou et al., 2021; Edel-Hermann and Lecomte, 2019; Perchepied and Pitrat, 2004; Risser et al., 1976; Sebastiani et al., 2017). Although Fusarium wilt of melon has been documented in Europe, North America, Israel, Japan, and South Africa, the geographical distribution of different physiological races seems not uniform. For instance, races 0, 1, 2, and 1.2 were reported in Europe, the USA, Japan, Türkiye, and Israel, whereas only three races were documented in South Africa (races 0, 1, and 2) (Jacobson and Gordon, 1990; Kurt et al., 2002; Luongo et al., 2015; Namiki et al., 2000; Schreuder et al., 2000; Zuniga et al., 1997). Currently, the physiological races of FOM present in Taiwan remain unknown.
Since FOM persists in the soil by producing chlamydospores as a survival structure and colonizing plant debris or the roots of non-susceptible rotation crops (Sebastiani et al., 2017), the application of fungicides for managing FOM is therefore less effective. Additionally, inappropriate use of chemical fungicides could result in pollution of the environment, or lead to the emergence of fungicide-resistant pathogens (Fang et al., 2022; Park et al., 2024). Although disease-resistant cultivars would be the best option for managing MFW, there is no resistant cultivar currently available in Taiwan. Hence, management of MFW relies primarily on integrating various cultural practices, such as grafting rootstocks, rotation with non-hosts, application of soil amendment, or relying on the application of nonpathogenic Fusarium strains or antagonistic microorganisms (Freeman et al., 2002).
Biological control is defined as the use of beneficial organisms to reduce the negative effects of plant pathogens and promote positive responses of the plants (Pandit et al., 2022). The most common approach in biological control is the selection of antagonistic microorganisms, investigation of the mode of action, and development the application methods for disease management. A broad variety of antagonistic microorganisms, such as Pseudomonas spp., Trichoderma spp., Streptomyces spp., and Bacillus spp., have been identified from the rhizosphere or the root endophytes as effective biocontrol agents (BCAs) against Fusarium wilt (Belgrove et al., 2011; Gava and Pinto, 2016; Kavino and Manoranjitham, 2018; Le et al., 2021). For instance, Bacillus strain Y-IVI had strong antagonistic activity in vitro against Fusarium oxysporum and was used to control muskmelon Fusarium wilt (Zhao et al., 2016). Jayanti and Joko (2020) reported that several endophytic bacteria isolated from healthy melon plants were evaluated for their efficacy in inhibiting the growth of F. oxysporum f. sp. melonis, among which Bacillus sp. and Burkholderia sp. exhibited the best results (Jayanti and Joko, 2020).
B. mycoides has previously been reported as a BCA and plant growth-promoting bacterium in agriculture. It was found effective in controlling the water-soaking lesions on cucumber leaves and reducing Pythium damping-off in greenhouse trials (Chen, 2017; Lin et al., 2018; Peng et al., 2017). Similar disease reduction effects had also been observed on cabbage seedlings inoculated with Pythium aphanidermatum (Huang et al., 2018). Ali et al. (2022) reported that B. mycoides PM35 is capable of promoting plant growth and increasing the productivity of maize. Moreover, B. mycoides is known for producing plant growth regulators. Two volatile antifungal compounds, phenylacetic acid and methyl phenylacetate, produced by B. mycoides BM02 were found to suppress plant diseases caused by Fusarium oxysporum f. sp. lycopersici and other pathogenic microorganisms (Wu et al., 2020). In addition, B. mycoides BM103 was found able to support plant growth, induce resistance-related gene expression, and decrease the disease severity of Fusarium wilt of strawberries (Chen, 2017; Lin et al., 2018). Based on these studies, we hypothesized that the B. mycoides strains, BM02 and BM103, may be effective in managing MFW. Hence, the efficacy of the two biocontrol strains in protecting melon plants from FOM infection was further explored in planta. The aims of this study were (1) to determine the physiological race of MFW pathogen found in central Taiwan, and (2) to evaluate the efficacy of two B. mycoides strains in protecting melon seedlings from MFW.
Materials and Methods
Plant materials and fungal isolates
The fungal pathogens used in this study were isolated from the melon plants showing Fusarium wilt symptoms in a field in Wufeng District, Taichung City (Fig. 1A). Briefly, diseased melon tissues were surface disinfested using 0.1% hypochlorite (HClO) and 75% ethanol, followed by rinsing three times in distilled water. Surface disinfested tissues were placed on a Petri dish with 2% of water agar and incubated for 48 h to encourage hyphae growth and sporulation. Then, single spores of Fusarium spp. were transferred onto potato dextrose agar (PDA; Difco, Franklin Lakes, NJ, USA) plates. The purified single spore isolates were incubated at 25°C for 14 days and used for further study. The initial identification of three representative fungal isolates, FOM02, FOM03, and FOM05, was conducted based on taxonomic keys described by Booth (1971) and Lombard et al. (2019). The mycelial, colony, and spore morphology grown on a modified water agar (1.5% (w/v) agar, 0.5% (w/v) glucose, 1 L H2O) were documented. The fungal morphology was observed under a Zeiss Axio Imager 2 microscope and photographed using an Axiocam 105 color camera (Carl Zeiss Microscopy GmbH, Germany).

The Fusarium wilt symptoms from melon in the field (A). Wilting and stunting symptoms observed on melon plants inoculated with (left side, B) and without (right side, B) FOM05 at 8 days after inoculation. The longitudinal section of melon vine exhibited vascular bundle discoloration (left side, C) after inoculated with FOM05, whereas uninoculated plants remained healthy (right side, C).
Melon (Cucumis melo var. makuwa cv. ‘Silver Light’) seeds were surface-disinfected with 1% chlorox for 15 min, followed by rinsing in sterile water for 3 times. Then, the seeds were transferred onto a filter paper and immersed in sterile water at 25°C to support further imbibition. After 48 h of imbibition, germinated melon seeds were sown in TS 30/70+perlite high tray substrate (Gramoflor, Vechta, Germany) and grown at 27°C in a greenhouse. The 14-day-old seedlings were used for further experiments.
Pathogenicity and host range
For the pathogenicity test, three fungal isolates, FOM02, FOM03, and FOM05, were grown in PDA medium for 2 weeks and then three pieces of approximately 0.5 × 0.5 cm2 mycelial plugs of each isolate were transferred to 150 mL of half-strength potato dextrose broth. The inoculum was incubated on a rotary shaker (120 rpm) at 28°C for 2 days and the cultural broth was filtered with four layers of cheesecloths. The root tips of 14-day-old melon seedlings were trimmed, leaving 6 cm of root from the base of hypocotyl, and then dipped in spore suspension (1 mL/plant, 106 spores/mL) for 15 min (Gordon et al, 2019). Each isolate had 10 seedlings, and the seedlings in negative control were treated with sterilized water. After treatment, seedlings were planted in the TS 30/70+perlite high tray substrate (Gramoflor) in a 3.5-inch pot and incubated at 28°C/26°C (day/night temperature) with 14 h light period in a greenhouse for 8 days. The experiments were conducted three times.
To investigate the host range of the three Fusarium isolates, 15 plant species in 6 families were inoculated (Table 1). Host range test was conducted using the aforementioned root trimming method with modification in the spore suspension concentration (105 spores/mL) and the number of inoculated plants (5 plants per isolate). The experiments were conducted three times.
Extraction of genomic DNA and PCR amplification of specific loci
The Fusarium isolates FOM02, FOM03, and FOM05 were cultured in PDA at 25°C for 14 days. Then, DNA extraction was conducted following the method of Lin et al. (2010b). In addition to the three FOM isolates, a total of 23 isolates of Fusarium spp., Fusarium oxysporum, and non-Fusarium isolates (Table 2) were included in the PCR amplification as positive and negative controls. For specific amplification of F. oxysporum, primer set of FnSc1 (5′-TACCACTTGTTGCCTCGGCGGATCAG-3′) and FnSc2 (5′-TTGAGGAACGCGAATTAACGCGAGTC-3′) was used (Lin et al., 2010b). The PCR reaction mixture consisted of 1 μL (10 ng) of DNA template, 10 μL of 2× PCR master Mix Kit (QIAGEN, Hilden, Germany), 1 μL (0.5 μM) of each forward primer and reverse primer, respectively, and added Milli-Q water to a final volume of 20 μL. The PCR amplification started with a denaturation step at 94°C for 90 s, followed by 30 cycles of denaturation at 94°C for 30 s, annealing at 62°C for 30 s, and extension at 72°C for 60 s, and a final extension at 72°C for 10 min.
To further identify the physiological race, F. oxysporum f. sp. melonis race 2 specific primer pair Fa15F (5′-TAGGGATGATAGCGGTCTGG-3′) and Fa15R (5′-GCTAGTTCGAGGCAAT TGGA-3′) was used for race-specific detection in tested isolates (Luongo et al., 2012). The composition of the PCR reaction mixture was mentioned in the previous paragraph. The race-specific PCR amplification started with a denaturation step at 94°C for 2 min, followed by 30 cycles of denaturation at 94°C for 30 s, annealing at 60°C for 30 s, extension at 72°C for 60 s, and a final extension at 72°C for 1 min.
Multilocus sequence analysis and construction of phylogenetic tree
Partial sequences of internal transcribed spacer (ITS) region, transcription elongation factor 1 alpha (tef1α), and the DNA-directed RNA polymerase II second largest subunit (rpb2) genes were amplified using primer pairs ITS1-F (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) (White et al., 1990), EF-1 (5′-ATGGGTAAGG ARGACAAGAC-3′) and EF-2 (5′-GGARGTACCAGTSATCATGTT-3′) (O’Donnell et al., 1998), and RPB2-5F2 (5′-GGGGWGAYCAGAAGAAGGC-3′) and fRPB2-7cR (5′-CCCATRGCTTGYTTRCCCAT-3′) (Liu et al., 1999; Sung et al., 2007), respectively. The PCR reaction mixture was prepared as aforementioned, and performed under the following conditions. Initial denaturation at 95°C for 2 min., then 35 cycles of 95°C for 30 s, annealing at respective temperatures (58°C for ITS1/ITS4, 53°C for RPB2-5F2/fRPB2-7cR primers, and 52°C for EF-1/EF-2 primers) for 45 s, and then 72°C for 1 min with a final extension at 72°C for 4 min (Liu et al., 1999; O’Donnell et al., 1998; Sun et al., 2022; Sung et al., 2007; White et al., 1990). The PCR products were submitted for Sanger sequencing with an ABI 3730 XL analyzer (Thermo Fisher Scientific Inc., Foster City, CA, USA).
The quality of the obtained sequences was checked using Geneious Prime (Geneious Prime 2024.0.7, https://www.geneious.com). The newly generated nucleotide sequences of ITS, tef1α, and rpb2 were deposited in National Center for Biotechnology Information GenBank and aligned with reference sequences of 15 Fusarium isolates, respectively (Supplementary Table 1). Then, sequences of the three loci were concatenated for phylogenetic analysis. A phylogenetic tree was constructed using the Maximum Likelihood (ML) method in MEGA X and the Bayesian inference (BI) method in Geneious Prime. For creating the ML tree, the Kimura 2-parameter model (Kimura, 1980) with a discrete Gamma distribution (+G, parameter = 0.62) was applied with 1,000 bootstrap replicates to estimate the branch support in MEGA X (Kumar et al., 2018; Nei and Kumar, 2000). For constructing the BI tree, a generalized-time-reversible substitution model was selected, and a Markov Chain Monte Carlo was conducted with one million generations and sampled every 5,000 generations. The first 20% of obtained trees were discarded as burned-in, and the remaining trees were used to estimate the posterior probability of the branches in the MrBayes v3.2.6 plugin (Huelsenbeck and Ronquist, 2001; Ronquist et al., 2012) in Geneious Prime.
Biocontrol activity of BM02 and BM103
Two strains of Bacillus mycoides Flügge, BM02 and BM103, respectively isolated from the tomato and strawberry rhizosphere in central Taiwan have been previously characterized (Chen, 2017; Lin et al., 2018; Peng et al., 2017). To further investigate the biocontrol activity of B. mycoides against MFW, BM02 and BM103 suspension (OD600 = 1.15, concentration ≈ 3.7 × 108 and 3.5 × 108 CFU/mL, respectively) were added to SYM broth (soybean meal-molasses broth, 3% (w/v) soybean meal, 0.5% (w/v) molasses, 1,000 mL H2O, autoclaved at 121°C for 20 min) at a 1:100 (v/v) ratio. After the cultural broth was incubated at 200 rpm at 30°C for 3 days (concentration of BM02 and BM103 cultural broth ≈ 2.3 × 108 and 2.0 × 108 CFU/mL, respectively), the cultural broth and cultural filtrate (filtered with 0.2 μM of Acrodisc Syringe Filters with Supor membrane, Pall Corporation, Port Washington, NY, USA) of both strains were further evaluated.
To evaluate the biocontrol efficacy of the cultural broth and the cultural filtrate against MFW, 200- and 400-times-diluted solutions were applied either by foliar spraying or by soil drenching on melon seedlings. For foliar spraying, the melon seedlings were sprayed at 7-day-old (approximately 1- to 2-true-leaf stage, 3 mL/seedling) and 14-day-old (approximately 2- to 3-true-leaf stage, 6 mL/seedling), respectively, before inoculation. For soil drenching, 10 mL of either the broth or the filtrate was drenched at the same seedling stages in foliar spraying experiments before inoculation. The control plants were treated with sterilized water.
For inoculation, the melon seedlings were uprooted one day after the second treatment. The root of each seedling was washed gently with tap water, trimmed off 4 cm from the root tip, and dipped in 1 mL (105 spores/mL) of FOM05 spore suspension for 15 min. The inoculated plants were incubated at a 28°C greenhouse for 21–42 days. Each treatment had 18 seedlings and the experiments were conducted three times.
Disease assessment
The disease severity of MFW was assessed based on the symptoms of the plant and the proportion of discolored vascular tissues in the stem 6 cm above the hypocotyl 21 days after inoculation. Plants were rated based on the 0 to 4 scale of Akhter et al. (2015) with modifications, in which 0 = healthy plants; 1 = stunted plant with less than half of yellowing leaves; 2 = plant with yellowing stem and more than half of the leaves were wilted; 3 = less than 10% discoloration in the stem vascular tissue; and 4 = more than 10% of discoloration in the stem vascular tissue. The mean disease severity in each treatment was calculated using the following formula:
, where Si was the disease scale; ni was the number of plants in each scale, and N meant the total number of plants surveyed.
Statistical analysis
The disease severity data in the biocontrol experiments were subjected to a non-parametric Kruskal-Wallis test followed by a Dunn test for pairwise comparison at P < 0.05 level. All statistical analyses were performed using tidyr and dplyr packages in R platform version 4.3.3 (R Core Team, 2024).
Results
Pathogenicity and host specificity
Pathogenicity test was conducted for the fungal isolates FOM02, FOM03, and FOM05 as per Koch’s postulates on 14-day-old melon seedlings. All FOM02, FOM03, and FOM05-inoculated melon plants were wilted 8 days after inoculation, displaying reddish brown lesions at the lower stem. Vascular bundle discoloration was observed when the lower stem was split longitudinally, whereas control plants remained healthy. The pathogens were re-isolated from the vascular bundle tissue of all inoculated plants, and the re-isolated pathogens were similar to the original culture, thus fulfilling Koch’s postulates (Fig. 1B and C).
To evaluate the host range of the pathogen, the three isolates were inoculated on oriental melon and 15 different plant species in 6 families, respectively. Among the tested plant species, only oriental melon was susceptible to the three fungal isolates. Specifically, FOM02-inoculated oriental melons exhibited severe wilting symptoms whereas FOM03- and FOM05-inoculated oriental melons exhibited necrotic lesions on leaves and wilting symptoms of the plants (Table 1, Fig. 2). The rest plant species remained healthy at the end of each experiment, suggesting that the three isolates caused wilt disease specifically on melon.

Results of host range tests after the plants were inoculated with Fusarium oxysporum isolates FOM02, FOM03, and FOM05, respectively, for 8 days. Only oriental melon (Cucumis melo var. makuwa) (A) exhibited yellowing and drying symptoms whereas the other 14 plant species (B–O) in 6 families were symptomless after inoculation. (B) Lettuce (Lactuca sativa var. crispa). (C) Cabbage (Brassica oleracea var. capitata). (D) Pai-Tsai (Brassica rapa subsp. chinensis). (E) Radish (Raphanus sativus). (F) Cantaloupe (Cucumis melo var. reticulatus). (G) Cucumber (Cucumis sativus). (H) Squash (Cucurbita moschata). (I) Watermelon (Citrullus lanatus). (J) Dwarf bean (Vigna unguiculata). (K) Sweet pea (Pisum sativum). (L) Strawberry (Fragaria ananassa). (M) Chili pepper (Capsicum annuum). (N) Sweet pepper (Capsicum annuum). (O) Tomato (Solanum lycopersicum).
Identification of the pathogen and determination of physiological race
The pathogen was identified based on morphological characteristics, amplification of specific loci, and multilocus sequence analysis. In terms of colony morphology, the color of FOM02, FOM03, and FOM05 colonies changed from white to pale purple after incubating on PDA for 14 days (Fig. 3A and B). All isolates produced macroconidia, microconidia, and chlamydospores. Macroconidia were sickle-shaped, hyaline, and multi-celled with two to three septa. The size of macroconidia ranged from 24.6 μm to 43.8 μm in length and 3.6 μm to 5.0 μm in width (n = 50) (Fig. 3D and E). Microconidia were oval-shaped, hyaline, and single or bi-celled, with smaller size than macroconidia, ranging from 6.6 μm to 12.5 μm in length and 2.5 μm to 4.6 μm in width (n = 50) (Fig. 3F). Chlamydospores were globose to sub-globose, formed intercalarily or terminally, 5 to 8 μm in diameter (n=50) (Fig. 3C). Based on the morphological characteristics, the three fungal isolates were identified as Fusarium.

Morphological characteristics of Fusarium oxysporum f. sp. melonis isolate FOM05. Front (A) and reverse (B) surface of the colony after incubated on potato dextrose agar with 12 h photoperiod at 25°C for 14 days. Microscopic characteristics of FOM05: terminal chlamydospores from conidiophores (C), aerial conidiophores producing macroconidia (D, E), and aerial conidiophore producing abundant microconidia that cohere in a false head (F) (scale bars = 10 μm).
To identify the species epithet and physiological race of the three isolates, F. oxysporum-specific primers (FnSc1/FnSc2) and F. oxysporum f. sp. melonis race 2-specific primers (Fa15F/Fa15R) were used. Both primer pairs were able to amplify specific fragments from the three Fusarium isolates, with amplicon sizes of 327 bp for FnSc1/FnSc2 and 301 bp for Fa15F/Fa15R, suggesting that the three isolates belonged to F. oxysporum f. sp. melonis race 2 (Fig. 4). Contrarily, neither FnSc1/FnSc2 nor Fa15F/ Fa15R primer pairs amplified any fragment from non-target fungal isolates, suggesting good specificity of the two primer pairs (Fig. 4).

PCR amplification results using Fusarium oxysporum-specific primers FnSc1/FnSc2 (A) and F. oxysporum f. sp. melonis race 2-specific primers Fa15F/Fa15R (B) from different Fusarium and non-Fusarium isolates. M, 100 bp ladder (500 bp, white triangle; 300 bp, red triangle). The labels for each lane are the same in both panels.
The species identification based on specific primers was well supported by the multilocus sequence analyses of partial sequences of ITS, tef1α, and rpb2. The ITS region (455–474 bp), tef1α (616–668 bp), and rpb2 (504 bp) of the three isolates exhibited 100%, 98.87–98.95%, and 99% sequence identities with F. oxysporum isolates JW 11005 sequences (GenBank accession nos. MZ890536 for ITS, MZ921881 for tef1α, and MZ921750 for rpb2), respectively. The concatenated sequences of the three isolates were clustered with Fusarium oxysporum in both ML and BI trees, hence a representative ML tree with the highest log likelihood was shown (Fig. 5). Based on the morphological characteristics, host specificity, specific primers, and multilocus sequence analysis results, the three isolates causing MFW in central Taiwan were identified as F. oxysporum f. sp. melonis race 2.

Phylogenetic tree of three Fusarium oxysporum isolates, FOM02, FOM03, FOM05, from this study and 15 reference sequences from National Center for Biotechnology Information GenBank (Supplementary Table 1) based on the concatenated partial sequences of ITS, tef1α, and rpb2 genes. The phylogenetic tree was constructed using Maximum Likelihood method with Kimura 2-parameter model and Bayesian Inference method. The bootstrap support (>90%) for Maximum Likelihood method and posterior probabilities (>0.99) for Bayesian Inference method are shown next to the branches. Lasiodiplodia mexicanensis strain IXBLT15 was used as an outgroup.
Biocontrol activity of Bacillus mycoides BM02 and BM103
To evaluate the biocontrol activity of two B. mycoides strains against MFW, the cultural broth and cultural filtrate were applied at 200- and 400-times dilution, respectively, by foliar spraying and soil drenching methods. In the foliar spraying experiment, 200-times diluted BM103 cultural broth, 400-times diluted BM02 cultural broth, and 400-times diluted BM103 cultural filtrate significantly reduced disease severity of MFW compared to the untreated control (Fig. 6A). In the soil drenching experiment, BM02 cultural broth at both dilution levels and its cultural filtrate at 200-times dilution significantly reduced the disease severity compared to the untreated control. However, the disease severity in BM103-treated plants was only numerically less severe than in untreated control (Fig. 6B). Microscopic examination of the lower stems of untreated melon seedlings also revealed characteristic vascular browning, whereas melon seedlings treated with BM02 and BM103 exhibited no symptoms or mild symptoms in vascular bundle tissues (Fig. 7).

Effect of Bacillus mycoides strains BM02 and BM103 on managing melon Fusarium wilt. Two hundred-times and 400-times diluted cultural broth and cultural filtrate were sprayed on leaves (A) or drenched in the soil (B) when the melon seedlings were 7-day and 14-day-old. One day after the second treatment, melon seedlings were inoculated with F. oxysporum f. sp. melonis FOM05 spore suspension, incubated at a 28°C greenhouse, and the disease severity was assessed at 21 days after inoculation. Means (n = 18) in the same dilution fold of each application method followed by the same letters are not significantly different (P > 0.05) according to Dunn’s post hoc test. Bars indicate standard error.

Vascular browning in melon stems caused by Fusarium oxysporum f. sp. melonis FOM05 after treated with and without Bacillus mycoides BM02 and BM103. Vascular browning and discoloration were observed in FOM05-inoculated plants treated with water (A, C). Contrarily, no obvious browning in BM02-treated (B) and mild browning in BM103-treated (D) melon plants 21 days after inoculation.
Discussion
Fusarium wilt in cucurbits is among the most damaging diseases affecting this plant family. Currently, eight cucurbit-infecting formae speciales have been described, including F. oxysporum f. sp. benincasae, f. sp. cucumerinum, f. sp. radicis-cucumerinum, f. sp. lagenariae, f. sp. luffae, f. sp. melonis, f. sp. momordicae, and f. sp. niveum (Edel-Hermann and Lecomte, 2019; Lombard et al., 2019). Among which F. oxysporum f. sp. melonis had been documented in Taiwan for many years (Huang, 2010), however, information on the existing physiological races was not available. Considering different melon cultivars were planted in Taiwan, elucidating the present physiological races should shed light on cultivar selection before planting.
In this study, FOM02, FOM03, and FOM05 causing wilt and discoloration in vascular bundle of melon plants were identified as F. oxysporum based on morphological characteristics and species-specific primers. The result was further supported by multilocus sequence analysis. The ITS locus alone is insufficient to resolve Fusarium species (O’Donnell et al., 2013), therefore, protein-coding loci such as tef1α and rpb2 are recommended as primary markers for identifying Fusarium isolates at the species level (Carbone and Kohn, 1999; Geiser et al., 2004; O’Donnell et al., 2000, 2007). The result of species-specific primer amplification was supported by multilocus sequence analysis of the ITS, tef1α, and rpb2 loci, suggesting that the species-specific primers are a useful tool for quick identification of the pathogen to the species epithet.
FOM02, FOM03, and FOM05 were found to induce wilt disease only in oriental melon after inoculating 15 plant species in Asteraceae, Brassicaceae, Cucurbitaceae, Fabaceae, Rosaceae, and Solanaceae families. No visible symptoms on tested plants other than oriental melon suggested that the isolates could be classified as F. oxysporum f. sp. melonis. To further determine the physiological races in F. oxysporum f. sp. melonis, the conventional method included inoculating differential cultivars, assessing the disease responses of the cultivars to each isolate, and then categorizing the pathogen into various races. However, the process is not only laborious and time-consuming but also sometimes difficult due to limited access to the germplasm of differential hosts. Hence, using race-specific markers as an alternative for identifying the physiological race was performed in this study. Based on the PCR result, race 2-specific marker Fa15F/Fa15R was able to amplify a 301 bp fragment specifically from the three isolates, but not from other formae speciales tested in this study. The result suggested that the three melon isolates belonged to F. oxysporum f. sp. melonis race 2. Although we did not collect F. oxysporum f. sp. melonis isolates from different areas, nor did we investigate the prevalence of different races in Taiwan, this is the first study suggesting the presence of FOM race 2 in Taiwan. Further studies focusing on the spatiotemporal population structure of the pathogen should be warranted in the future.
For integrated disease management, knowing the pathogen population was the first step which helps in determining the cultivars that could be planted in certain areas. Integrating different approaches for disease management was the next step, in which utilization of microbial control agents has provided an effective, safe, and sustainable alternative for the management of Fusarium diseases (Khan et al., 2021). Bacillus species, one of the most studied BCAs, was reported to suppress plant diseases by various mechanisms, such as competition for nutrients and space, facilitating the uptake of certain nutrients from the environment, biosynthesis of plant hormones, production of antibiotics, hydrolytic enzymes, siderophores, and induction of systemic resistance (Aly et al., 2022; Park et al., 2024; Shin et al., 2023; Woo et al., 2024).
To date, many studies have reported the successful control of soilborne diseases by inoculating plant roots with induced systemic resistance (ISR)-inducing BCAs (Molinari and Leonetti, 2019; Samaras et al. 2021; Tjamos et al., 2005). Although foliar application of ISR-inducing BCAs has less frequently been explored than root application in controlling soilborne diseases, control of soilborne pathogens by foliar application of BCA has been documented. For instance, a tomato endophyte Bacillus sp. G4L1 has been shown to significantly suppress tomato bacterial wilt under high pathogen pressure (Fu et al., 2020), suggesting a novel approach for managing soilborne pathogens. Foliar application of ISR-inducing Bacillus species is a simpler, less labor-intensive, and cost-effective approach compared to soil drenching for managing soilborne diseases if the strains are compatible with tank-mix application. In this study, the efficacy of Bacillus mycoides BM02 and BM103 in reducing MFW was assessed. After foliar spraying or soil drenching of the BCAs, BM02 was observed to possess a more consistent suppressive effect against MFW compared to strain BM103. Interestingly, the disease severity was reduced after spraying the cultural broth or cultural filtrate of either strain on leaves, suggesting that a certain level of resistant response was induced by these treatments in melon seedlings.
B. mycoides possesses a diverse range of metabolites that can be categorized based on their functional groups, including alcohols, amines, aromatic compounds, ketones like 4-methyl-2-oxopentanoic acid, acetyl compounds like N-acetyl-L-cysteine, and phosphates like ethanolamine (Baggio et al., 2021; Chen, 2017; Lin et al., 2018; Peng et al., 2017; Wu et al., 2020). Many metabolites produced by B. mycoides have also been reported effectively against various plant diseases. For example, B. mycoides strain A3 (BmA3) can produce phytohormones, including indole-3-acetic acid and gibberellic acid, and exhibits phosphate solubilizing and radical scavenging capabilities. BmA3 was also able to activate jasmonic acid and salicylic acid signaling pathways, inducing plant growth and tolerance against abiotic stresses in A. thaliana seedlings (Kurniawan and Chuang, 2021). The dimethyl-disulfide and ammonia produced by B. mycoides strains CHT2401 and CHT2402 were toxic to many plant pathogens, which may lead to direct inhibition of the pathogens (Huang et al., 2018). Hence, the reduction of MFW observed in this study might be attributed to the production of metabolites, the enhancement of nutrient absorption from soils, or due to the induction of disease resistance.
Root colonization by BCAs is considered an essential prerequisite for the successful biocontrol of soilborne diseases (Sachdev and Singh, 2018). When the introduced BCA populations decline due to the influence of abiotic and biotic factors, such as drought, flooding, soil pH, or competition among different microorganisms, the biocontrol efficacy in the field could be affected and short-persisted. Therefore, repeated drenching of BCAs at regular intervals was usually recommended to ensure consistent biocontrol efficacy and colonization of BCAs (Yendyo et al., 2017). However, the “booster” drenching of a sufficient volume of BCAs to protect the roots is technically difficult, labor-intensive, and costly. On the contrary, the booster application of BCAs by foliar spraying may be easier, which could also be incorporated with compatible pesticides in the tank mix, or be applied via a drone. Therefore, developing B. mycoides BM02 into a foliar spray formula for managing soilborne diseases could be a practical and innovative direction in the future.
Notes
Conflicts of Interest
No potential conflict of interest relevant to this article was reported.
Electronic Supplementary Material
Supplementary materials are available at The Plant Pathology Journal website (http://www.ppjonline.org/).