Molecular Diagnosis and Vegetative Compatibility Group Analysis of Fusarium Wilt of Banana in Nepal

Fusarium wilt of banana (FWB), caused by Fusarium oxysporum f. sp. cubense (Foc), is the most important constraint of the banana industry globally. In Nepal, epidemics resembling FWB have been increasingly observed on the Malbhog cultivar in the past several years. However, the disease has not been officially reported yet, and consequently, little is known about the pathogen present across the country. In this study, we characterized 13 fungal strains isolated from banana plants of the Malbhog cultivar (Silk, AAB) showing symptoms similar to FWB in banana plantations in Nepal. All of the strains were typed as belonging to the F. oxysporum and caused FWB symptoms when inoculated in the Malbhog and Cachaco (Bluggoe, ABB) cultivars. No symptoms were observed in the Williams cultivar (Cavendish, AAA). Vegetative compatibility group (VCG) analysis classified the strains as VCG 0124 or VCG 0125. PCR analyses conducted with primers specific for Foc race 1 (Foc R1) or Foc tropical race 4 (TR4) revealed that all the strains reacted positively for Foc R1 and none for TR4. Altogether, our results demonstrated that the pathogen populations causing FWB of the Malbhog cultivar in Nepal were Foc R1. This work reported, for the first time, the occurrence of FWB in Nepal. Further studies with larger Foc populations are needed to better understand disease epidemiology to design sustainable disease management strategies.


Introduction
Banana is a major fruit crop and a high-value agricultural product in Nepal. It is cultivated on 18,329 ha areas across the country, with a total production of 278,890 tons and a productivity of 16.79 t/ha spanning 68 out of the 77 districts of Nepal [1,2]. Among them, Chitwan district is considered the major hub for banana production, followed by the Saptari, Jhapa, Morang, and Rupandehi districts. In Chitwan, banana farming is mainly concentrated in Bharatpur, Ratnanagar, Kalika, Khaireni, Jagatpur, and Thimura [3]. Nearly 15% of the area of banana-growing regions in Nepal is covered by a variety called Malbhog (Silk, AAB), which accounts for a 2493 ha area with 41,833 tons of production [4].

Sample Collection
In December 2017, an unusual phenomenon of banana plants (Malbhog, AAB) showing severe yellowing wilt symptoms before collapse was first reported in the Chitwan district of Nepal. Plant sampling from Malbhog bananas was conducted during two field surveys in December 2019 in the Chitwan and Nawalparasi districts. Visited locations were at an altitude of 170 m to 210 m. Diseased plants with typical external symptoms were found, and some xylem vessels appeared reddish-brown when the pseudostems were cut longitudinally. Altogether, thirteen samples from symptomatic tissue were collected from seven farmers' fields of the Chitwan district and four farmers' fields of the Nawalparasi district (Table 1).

Isolation and Identification
All collected samples were dried at room temperature before the isolation of the pathogen. Every sample was cut into small squares with a width of 0.5 cm using scissors. The scissors and tweezers were repeatedly heat-sterilized for every sample isolation. To prevent cross-contamination, gloves were changed when handling each sample. The cut pieces were immersed in 70% alcohol for 30 s for disinfection, followed by immersion in 0.1% mercury chloride for 30 s in the same way, and then rinsed three times with sterile water and placed on the plates with PDA medium with 50 µg/mL ampicillin. Then, the plates were kept in the dark at 30 • C for 5-7 days. The morphologies of the developed colonies were observed under a stereomicroscope, and the typical structures of Fusarium were observed using a compound microscope.

Pathogenicity Test
Tissue-cultured plantlets of Malbhog were planted in a steam-sterilized potting medium consisting of soil, sand, and farmyard manure in the ratio of 3:1:1 at the National Plant Pathology Research Centre (NPPRC), Khumaltar, Nepal. The pathogenicity test was performed when the plants reached [10][11][12][13][14][15] cm in height at the 3-4 leaf stage to ensure that the vascular system was fully developed. Six plants were used per treatment. These plants were inoculated with NP11 or NP12 single-spore isolates from the Chitwan and Nawalparasi districts, respectively. The conidial suspension adjusted was counted using a hemocytometer and was 6.3 × 10 6 conidia/mL. The roots of plants were slightly wounded by a sterile knife, and 100 mL of spore suspension of the F. oxysporum isolate was then added to the plantlets by drenching. Plants drenched with water served as the control. All the plants were placed in the greenhouse. Inoculated plants were monitored regularly for the appearance of disease symptoms. When external symptoms appeared, the internal symptoms were also investigated by dissecting the pseudostem and corm. The pathogen was re-isolated from the inoculated plants and further characterized by PCR.
Pathogenicity tests were also carried out at the Institute of Plant Protection Research, Guangxi Key Laboratory of Biology for Crop Diseases and Insect Pests, Guangxi Academy of Agricultural Sciences, China. Banana plantlets from the Cachaco (Bluggoe, ABB) and Williams (Cavendish, AAA) cultivars were grown and inoculated as afore mentioned. Nine plants were used per treatment, and each treatment was repeated three times. The wounded roots of the inoculated plantlets were soaked in a 10 6 conidia/mL spore suspension for 30 min and then planted in the pots. Those plantlets that soaked in sterile water for 30 min served as the healthy control. Twenty-four days after inoculation, the disease index was investigated by dissecting the corm. The classification criteria were as follows: scale 0, no symptoms in the corm; scale 1, discoloration area in the central longitudinal section of the corm was less than 25%; scale 3, discoloration area in the central longitudinal section of the corm was 26-50%; scale 5, discoloration area in the central longitudinal section of the corm was 51-75% area; scale 7, discoloration area in the central longitudinal section of the corm was more than 75%. The disease index was calculated according to the following equation [25]: Disease index equation = [Σ (number of infected plants at a given disease scale × the disease scale)/(total tested plants × 7)] × 100 Statistical analysis was performed using R Studio version 4.0.3 (Agricolae package) for the disease index data. The data was transformed into log 10 (1 + data) for normalization, and 1 in each data set was added to remove potential infinity values from the logtransformed data. The analysis was measured using two-way ANOVA with the LSD test.

PCR Analyses
Five-to seven-day-old cultures were used for polymerase chain reaction (PCR) amplification. Approximately 100 mg of mycelium was taken with a sterile toothpick and placed into a sterilized PCR tube with 50 µL of lysis buffer for microorganisms (Takara Bio, https://www.takarabio.com, accessed on 13 January 2020). After 15 min of thermal denaturation at 80 • C in the PCR machine, it was centrifuged briefly, and 1.5 µL of the supernatant was used as a template for the PCR reaction. The translation elongation factor 1α gene (TEF-1α) was amplified with primers EF-1 and EF-2 as internal positive controls using the following program: 95 • C for 2 min and 35 cycles of 95 • C for 30 s; 50 • C for 30 s; and 72 • C for 1 min, followed by an additional extension time for 10 min at 72 • C [21]. The primer sets W1805F/W1805R and W2987F/W2987R from Li et al. (2012) [20] were used to identify Foc R1 (354 bp) and Foc R4 (593 bp), respectively. Another pair of W2987F/W2987R from Li et al. (2013) [28] was used to identify TR4 (452 bp). [Note: the name of primer set W2987F/W2987R from Li et al. (2013) [28] is the same as that from Li et al. (2012) [20], so in this research, we labeled it as TR4-W2987F/TR4-W2987R.] The sequence information of all the primer sets is shown in Table 2. The PCR was performed using the Gold Mix (Code: TSE 101, tsingke, Beijing, China) according to the manufacturer's protocol.
PCR products were electrophoresed in 1% agarose gel in an electrophoretic tank (Takara Bio) containing 1X TAE buffer for 20 min at 135 V. BIO-RAD ChemiDocTMXRS+ was used to take an image after electrophoresis. Table 2. The primer sets used to identify the Foc pathogen.

Phylogenetic Analysis
The TEF-1α sequences of isolates were amplified by PCR. These PCR products were confirmed with aimed bands by agarose gel electrophoresis and purified using the DNA Gel Extraction kit TSP601 (Tsingke, Beijing, China). PCR products were then cloned and sequenced in the pMD18-T vector. The TEF-1α sequences were showed in Supplementary  Table S1. Additionally, TEF-1α sequences of other F. oxysporum isolates were obtained from the publicly available genomes (https://www.ncbi.nlm.nih.gov/, accessed on 30 July 2021). The analysis involved 31 isolate sequences that were aligned with MUSCLE [29], and the phylogenetic analyses were performed using maximum likelihood (ML). The phylogenetic tree was constructed using the MEGA 6.0 software with the Kimura 2-parameter model [30]. The TEF-1α sequence from Ramlispora sorghi was used as the outgroup taxon (KM087651.1). Branches were tested for the inferred tree by bootstrap analysis on 1000 random trees.

Disease Diagnosis with Field Symptoms
The problem of wilt disease has existed in banana orchards in Nepal since 2017. The disease was observed in some new and old orchards growing the Malbhog variety. Symptoms appeared as yellowing of the oldest leaves and splitting at the base of different types of infected plants. Some diseased, old yellow leaves split at the base. Vascular discoloration varied from pale yellow in the early stage to dark red or almost black in the later stage when viewed by cutting a cross-section or splitting the pseudostem base. Brownish streaks of the vascular tissue were also observed during the field survey ( Figure 1). However, wilt disease was not observed in the Grand Naine banana variety.

Phylogenetic Analysis
The TEF-1α sequences of isolates were amplified by PCR. These PCR products were confirmed with aimed bands by agarose gel electrophoresis and purified using the DNA Gel Extraction kit TSP601 (Tsingke, Beijing, China). PCR products were then cloned and sequenced in the pMD18-T vector. The TEF-1α sequences were showed in Supplementary  Table S1. Additionally, TEF-1α sequences of other F. oxysporum isolates were obtained from the publicly available genomes (https://www.ncbi.nlm.nih.gov/, accessed on 30 July 2021). The analysis involved 31 isolate sequences that were aligned with MUSCLE [29], and the phylogenetic analyses were performed using maximum likelihood (ML). The phylogenetic tree was constructed using the MEGA 6.0 software with the Kimura 2-parameter model [30]. The TEF-1α sequence from Ramlispora sorghi was used as the outgroup taxon (KM087651.1). Branches were tested for the inferred tree by bootstrap analysis on 1000 random trees.

Disease Diagnosis with Field Symptoms
The problem of wilt disease has existed in banana orchards in Nepal since 2017. The disease was observed in some new and old orchards growing the Malbhog variety. Symptoms appeared as yellowing of the oldest leaves and splitting at the base of different types of infected plants. Some diseased, old yellow leaves split at the base. Vascular discoloration varied from pale yellow in the early stage to dark red or almost black in the later stage when viewed by cutting a cross-section or splitting the pseudostem base. Brownish streaks of the vascular tissue were also observed during the field survey ( Figure 1). However, wilt disease was not observed in the Grand Naine banana variety.

Morphology of the Pathogen Isolate
Thirteen suspected pathogen isolates were extracted from thirteen corresponding samples collected from banana fields in the Chitwan and Nawalparasi districts (Table 1), and their morphological characteristics were studied. The isolates produced dense white and purple colonies, and mycelia were evenly spread all over the PDA medium (Figure 2A,B).
The microscopic examination of fungal isolates showed the presence of three types of asexual spores: macroconidia, microconidia, and chlamydospores. The macroconidia were thin-walled with a definite foot cell and a pointed apical cell ( Figure 2C). Microconidia were oval-or kidney-shaped and were produced on micro-conidiophores in aerial mycelia ( Figure 2D). Chlamydospores were thick-walled ( Figure 2E), and these were usually produced singly in macroconidia, were intercalary, or were in terminal hyphae [31].

Morphology of the Pathogen Isolate
Thirteen suspected pathogen isolates were extracted from thirteen corresponding samples collected from banana fields in the Chitwan and Nawalparasi districts (Table 1), and their morphological characteristics were studied. The isolates produced dense white and purple colonies, and mycelia were evenly spread all over the PDA medium ( Figure  2A,B).
The microscopic examination of fungal isolates showed the presence of three types of asexual spores: macroconidia, microconidia, and chlamydospores. The macroconidia were thin-walled with a definite foot cell and a pointed apical cell ( Figure 2C). Microconidia were oval-or kidney-shaped and were produced on micro-conidiophores in aerial mycelia ( Figure 2D). Chlamydospores were thick-walled ( Figure 2E), and these were usually produced singly in macroconidia, were intercalary, or were in terminal hyphae [31].

Pathogenicity Test
The pathogenicity test was performed on three banana cultivars: Malbhog (AAB), Cachaco (ABB), and Williams (AAA). Yellowing and wilting symptoms were observed in inoculated Malbhog plantlets after 21 days of pathogen inoculation ( Figure 3A: left two plants). The non-inoculated plants (controls) showed no symptoms, and the leaves remained green ( Figure 3A: right two plants). Dark spots and brownish discoloration of the vascular tissue in the pseudostem were also observed ( Figure 3B). In severe cases, inoculated plantlets showed complete wilting or even death.
Similar symptoms were observed in inoculated Cachaco plantlets ( Figure 4A,C). Wilting symptoms in plantlets inoculated with the isolates NP1-1, NP2-1, NP4-1, NP5-1, NP6-1, NP7-1, NP8-1, and NP10-1 were noticed after 24 days of pathogen inoculation. There was a highly significant response of the disease index on the varieties and interactions of isolates with varieties, while the response was non-significant in the case of isolates without varieties ( Table 3). The pathogenicity of the F. oxysporum isolates to Cachaco and Williams was extremely significantly different (p < 2×10 -16 ). All of the inoculated Cachaco plantlets showed different degrees of symptoms, and the disease index was greater than 26.19, while the inoculated Williams plantlets nearly had no disease symptoms. The disease index of Cachaco inoculated with the isolate NP4-1 was 66.67, which was significantly higher than that caused by NP2-1, NP5-1, NP6-1, and NP8-1 (p < 0.0465).

Pathogenicity Test
The pathogenicity test was performed on three banana cultivars: Malbhog (AAB), Cachaco (ABB), and Williams (AAA). Yellowing and wilting symptoms were observed in inoculated Malbhog plantlets after 21 days of pathogen inoculation ( Figure 3A: left two plants). The non-inoculated plants (controls) showed no symptoms, and the leaves remained green ( Figure 3A: right two plants). Dark spots and brownish discoloration of the vascular tissue in the pseudostem were also observed ( Figure 3B). In severe cases, inoculated plantlets showed complete wilting or even death.
Similar symptoms were observed in inoculated Cachaco plantlets ( Figure 4A,C). Wilting symptoms in plantlets inoculated with the isolates NP1-1, NP2-1, NP4-1, NP5-1, NP6-1, NP7-1, NP8-1, and NP10-1 were noticed after 24 days of pathogen inoculation. There was a highly significant response of the disease index on the varieties and interactions of isolates with varieties, while the response was non-significant in the case of isolates without varieties ( Table 3). The pathogenicity of the F. oxysporum isolates to Cachaco and Williams was extremely significantly different (p < 2 × 10 −16 ). All of the inoculated Cachaco plantlets showed different degrees of symptoms, and the disease index was greater than 26.19, while the inoculated Williams plantlets nearly had no disease symptoms. The disease index of Cachaco inoculated with the isolate NP4-1 was 66.67, which was significantly higher than that caused by NP2-1, NP5-1, NP6-1, and NP8-1 (p < 0.0465). The pathogen was re-isolated from the inoculated, symptomatic plants and confirmed with the original one using microscopic observation and PCR. The pathogen was re-isolated from the inoculated, symptomatic plants and confirmed with the original one using microscopic observation and PCR.   The pathogen was re-isolated from the inoculated, symptomatic plants and confirmed with the original one using microscopic observation and PCR.    Values are the means of three replications. Data were transformed using log 10 (1 + data), the presented means are original, and the values inside the parentheses indicate transformed data; ANOVA was conducted on the transformed data. The mean values in the same column with the same letter are not statistically significantly different (p < 0.05).

PCR Analysis
PCR analyses were performed using four primer sets ( Figure 5). The W1805F/W1805R primer set previously associated with Foc R1 populations [20] produced the predicted 354 bp amplicon in all of the 13 tested isolates and from the Foc R1 strain used as the reference control. None of the tested isolates reacted to the Foc R4 or TR4 primer sets. The TR4 strain used as the positive control produced the predicted 593 and 452 bp amplicons, respectively. Altogether, our PCR results suggested that all Foc strains isolated from Malbhog bananas in Nepal were associated with Foc R1 populations.

Discussion
FWB causes severe yield losses globally, especially the TR4 strain, which is considered the biggest threat to global banana production [33]. Since 2017, typical symptoms of FWB have been increasingly observed in banana plantations in Nepal. However, identification and characterization of the causal agent is still lacking. In this research, we isolated and characterized the pathogen responsible for causing wilt disease on Malbhog banana plantations in the Chitwan and Nawalparasi districts, Nepal. Isolates, which exhibited morphological characteristics similar to the Fusarium oxysporum [34], were able to cause typical FWB symptoms in the Malbhog and Cachaco cultivars but not in the Cavendish cultivar, Williams. Field observations also revealed that the Cavendish cultivar, Grand Naine, which is resistant to Foc R1, had successfully replaced the susceptible Malbhog in the affected areas. Our PCR and VCG analyses grouped all the strains isolated as Foc R1 and revealed that six of these strains belonged to VCGs 0124 and 0125.
The phylogenetic tree based on TEF-1α divided the thirteen Nepal isolates into two clusters. The first cluster included most isolates grouped with Foc NRRL 25607 and CAV 2612, both of which belonged to VCG 0124 [18,32]. NP5-1, which belonged to VCG 0125, was also grouped with them. NP4-1 and NP10-1, both of which belonged to VCG 0124, were grouped into a separate branch. Results of genetic diversity derived from VCG and the phylogenetic tree were not fully consistent. It was probably due to the complex genetic diversity of Foc for convergent evolution or the horizontal gene transfer of F. oxysporum [10,21,22]. Moreover, TEF-1α was shown in some cases to be insufficient for distinguishing different formae speciales of Fusarium [35]. Nonetheless, results of the VCG revealed the diversity of Foc R1 in Nepal, although a larger number of isolates collected from other banana-producing regions may be suitable to further understand the Foc population biology in Nepal.
Banana is a high-value agricultural product and a major fruit in Nepal, in terms of production and domestic consumption (https://www.icimod.org/expanding-commercialbanana-production-in-nepal/, accessed on 8 June 2021). The current epidemic and spread of Foc R1 have severely affected the production of the traditional banana cultivar, Malbhog, which is one of the preferred local cultivars in Nepal. Our surveys revealed that FWB was prevalent in almost 90% of the Malbhog-growing area of the Chitwan and Nawalparasi districts. Surveys and farmer interviews during field visits indicated that such a rapid spread of FWB could be associated with the informal exchange of planting material from infested to healthy areas. In the absence of effective management practices, many growers chose to plant the Foc R1-resistant cultivar, Grand Naine. However, farmers and extension officers need to consider that Grand Naine is highly susceptible to Foc TR4, a more aggressive Foc strain already present in Bihar and Uttar Pradesh, India [23,24]. Therefore, prevention and exclusion measures against Foc TR4 using anticipatory approaches, as well as continuous Foc population monitoring, are strongly encouraged.

Conclusions
In conclusion, 13 fungal strains were isolated from banana plants of the Malbhog cultivar (Silk, AAB) showing symptoms similar to FWB in banana orchards in Nepal. Molecular diagnosis and vegetative compatibility group analysis were conducted for the evaluation of the FWB. All the strains caused FWB symptoms when inoculated in the Malbhog and Cachaco (Bluggoe, ABB) cultivars. No symptoms were observed in the Williams cultivar (Cavendish, AAA). PCR analyses with primers specific for Foc race 1 (R1) or Foc tropical race 4 (TR4) revealed that all the strains reacted positively for Foc R1 and none for TR4 in this study. Vegetative compatibility group (VCG) analysis further confirmed most strains as VCG 0124 or VCG 0125. The results demonstrated that the pathogen in this study causing FWB in Nepal are associated with Foc R1 in these limited samples. Larger Foc populations are necessary to better understand disease epidemiology for making sustainable disease management strategies in the next step.