Fusarium oxysporum f.sp. niveum
(Fusarium wilt of watermelon)

F. oxysporum f.sp. niveum is not a quarantine plant pathogen, because it occurs worldwide.

F. oxysporum f.sp. niveum only infects watermelon (Citrullus lanatus).

Transmission

F. oxysporum f.sp. niveum can be transmitted through seeds, animal manure, fertilizer, compost, tools, irrigation, soil, wind and rain. In China, transmission was reported through animal manure. The chlamydospores of the pathogen can survive in plant debris and infect watermelon roots after passing through the intestines of livestock or poultry and through compost processing. Another route of infection reported in China is through seed, which can transport the pathogen between fields and introduce it to new watermelon-growing areas (Chen et al., 1993). Injuries to the roots allow entry of the pathogen (LüGuiYun et al., 2014). In the USA, disease incidence (proportion of plants displaying symptoms) did not differ on cultivar Fascination (resistant to race 1, susceptible to race 2) in field plots infested with F. oxysporum f.sp. niveum race 2 alone or F. oxysporum f.sp. niveum together with southern root-knot nematode (Meloidogyne incognita) eggs or larvae (Keinath et al., 2019b). Contrary to these results, Hua et al. (2019) reported that co-inoculation of Fascination, Calhoun Gray and Plant Introduction (PI) 296341-FR with M. incognita and F. oxysporum f.sp. niveum race 1 or race 2 (PI 296341-FR only) resulted in earlier symptoms and increased severity.

Life-Cycle

The hyphae of the pathogen penetrate watermelon roots via the root tip meristematic zone and the epidermis of the zones of root elongation and maturation. Hyphae can also penetrate via ruptures caused by new lateral roots (Holliday, 1970; LüGuiYun et al., 2014). As the watermelon seedlings grow, hyphae of the pathogen enter the vascular system of the plant and, by growing and extending in or between the parenchymatous cells of the host, can infect the whole plant (Zhang et al., 2015). Watermelon fruits are infected through the vascular system or through wounds. Seed can be infected through the pistil (Petkar and Ji, 2017). Colonization of roots of the resistant cultivar Plant Introduction 296341-FR was reduced compared to colonization of the susceptible cultivar Black Diamond (LüGuiYun et al., 2014).

A latent period of 17 days, after soil inoculation, and 30 days, after inoculation of upper parts of the plant, at an optimal temperature of 25-27°C, is needed for infection. At late stages of colonization, conidia may be produced on the surfaces of diseased plants, especially the stems. The conidia can be transferred from one infected plant to another through irrigation and tools.

Factors affecting the survival of F. oxysporum f. sp. niveum have been extensively studied (Hopkins and Elmstrom, 1977; Huang and Sun, 1978, 1982; Bora et al., 1982).The pathogen may become saprophytic in the form of hyphae or dormant as chlamydospores in plant debris and soil, which become primary infection sources in the following season (Nishimura, 1971; Wang et al., 1993). Long-term persistence of the pathogen in infested soil has been reported by many authors (Smith, 1899; Mondal et al., 1994; Martyn, 2014).

Epidemiology

Epidemics of the disease are normally determined by the primary population of the pathogen, which is accumulated annually in the soil. The severity of the disease is influenced by soil type, irrigation, fertilizer application, crop management practices and seedling preparation. In the USA, the effect of inoculum density on the severity of the disease has been investigated (Summer, 1972; Martyn and McLaughlin, 1983). Watermelon cultivars that were slightly resistant or susceptible to F. oxysporum f.sp. niveum were severely wilted at 28 days, when grown in soil containing 10²-10³ propagules/g air-dry soil. Seedling emergence was reduced at inoculum densities above 10³ propagules/g. Similarly, when roots of watermelon cultivars were dipped in suspensions of 10³ to 10⁶ microconidia/ml, more cultivars were rated susceptible, slightly resistant, or moderately resistant at lower concentrations than at higher concentrations.

Monoculture can increase disease severity. In the USA, a year-to-year increase in the disease was reported in 10 cultivars (Hopkins and Elmstrom, 1984), but in the fourth year of monoculture, the lowest level of wilt and the highest yields were recorded in all cultivars. This was explained as the induction of suppressive soil by monoculture (see section on Prevention and Control). In other soils, however, continuous cropping of susceptible cultivars or cultivars resistant to race 1 can increase the proportion of race 2 isolates in the local population (Hopkins et al., 1992).

The optimum soil temperature for the disease in seedlings is 27°C, but under heavy soil infestation, severe wilt can occur over 20-30°C. Infection declines rapidly at temperatures above 30°C and does not occur above 33°C (Walker, 1941). In the southern USA, the soil temperature at 5- to 10-cm depth averaged over the 4-week period after transplanting was negatively correlated with disease incidence (Keinath et al., 2019a).

Physiological Races

In the USA, Barnes (1972) discovered differential pathogenicity of F. oxysporum f.sp. niveum to certain wilt-resistant watermelon cultivars. Martyn and Bruton (1989) reported 42 isolates of F. oxysporum f.sp. niveum from eight states in the USA and Israel that were race-typed on the basis of disease reaction of three watermelon differential cultivars. Race 1 was identified from each of the eight states and accounted for 45% of the isolates. Race 2 was isolated from three states (Texas, Oklahoma and Florida) and accounted for 33% of the isolates. Race 0 was found in the same states as Race 2 but only accounted for 9% of the isolates.

Pathogenic differences within F. oxysporum f.sp. niveum (Zhang and Wang, 1991; Zhang and Rhodes, 1993) have also been reported in China. Zhou and Kang (1996) identified eight isolates of F. oxysporum f.sp. niveum, collected from watermelon growing areas of Beijing, using a root-dip inoculation with purified spore suspensions at 100,000/ml, on watermelon varieties, Sugar Baby, Charleston Gray, Calhoun Gray, Jingxin 1 and Sumi 1 as differential hosts. Results showed that all eight isolates belonged to physiological race 1. Gu et al. (1994) reported that the existence of physiological races of 10 isolates collected from the suburbs of Shanghai may differ from those reported in the USA and elsewhere. In three provinces in the Aegean region of Turkey,  races 0, 1, and 2 accounted for 5%, 92%, and 3% of the isolates collected (Filiz and Turhan, 1992).

Race 2 is now common within the southern USA states and other watermelon-producing regions of the world. In 2000 in Maryland and Delaware, USA, 21, 57 and 22% of the isolates recovered from symptomatic watermelon plants in commercial fields were race 0, 1 and 2, respectively, while 76% of the isolates from a research field in Maryland were race 2 (Zhou and Everts, 2003). In South Carolina, USA, 2, 26 and 72% of pathogenic isolates collected between 2005 and 2013 were races 0, 1 and 2, respectively (Keinath et al., 2020). Race 3, originally identified in Maryland, USA, is also present in Florida and Georgia (Zhou et al., 2010; Amaradasa et al., 2018; Petkar et al., 2019). In Georgia, USA, 5, 39 and 56% of isolates collected in 2012-2013 were identified as race 0, 2 and 3, respectively; no race 1 isolates were found (Petkar et al., 2019).

The standard set of differential cultivars to identify races of F. oxysporum f.sp. niveum includes Sugar Baby (or Black Diamond), susceptible to all races; Charleston Gray, resistant to race 0 and susceptible to races 1, 2 and 3; Calhoun Gray, highly resistant to race 1 and susceptible to races 2 and 3; and PI 296341-FR, resistant to races 0, 1 and 2, and susceptible to race 3. As seed of Calhoun Gray is not available commercially, the cultivars Dixielee or Allsweet may be substituted (Larkin et al., 1990; Zhou and Everts, 2003).

Bacillus subtilis is a bacterial antagonist which can colonize the rhizospheres of watermelon and produce substances that are antagonistic to F. oxysporum f.sp. niveum (Lin et al., 1990). Trichoderma viride and T. harzianum, fungal antagonists that can live in or colonize the rhizospheres of watermelon, inhibit F. oxysporum f.sp. niveum through the production of antagonistic substances, nutrient competition and/or hyper-parasitic action (Sivan and Chet, 1986; Zhao et al., 1998).

Incidence

F. oxysporum f.sp. niveum has been recovered from watermelon seed lots in Egypt, Tunisia and the USA (McLaughlin and Martyn, 1982; Michail et al., 1989; Boughalleb and El Mahjoub, 2006). Tests on watermelon cultivars Xinhongbao, Xingqing No. 1, Zhemi No. 1 and Zhengza No. 7, bought from a local seed company in Zhejiang Province, China, indicated that 0.25, 0.49, 1.79 and 0.11% of the seed, respectively, contained F. oxysporum f.sp. niveum (Chen et al., 1993). When watermelon fruit pericarps were inoculated, 4.5% of the seed produced in inoculated fruits contained the pathogen (Petkar and Ji, 2017).

Effect on Seed Quality

There are no reports of any detrimental effects of the pathogen on seed appearance, germination and vigour.

Pathogen Transmission

An important route of transmission in China is through seed, which can transport the pathogen between fields and introduce it into new watermelon-growing areas (Chen et al., 1993). The fungus can enter watermelon fruits and the mature seed through the vessels. The population of mycelia and conidia in the hypocotyledonary axis and cotyledon node were found to be high after germination (Wang et al., 1993).

Seed Treatments

Geng et al. (2019) reported that fludioxonil as a seed treatment reduced the number of diseased seedlings produced by an infested seed lot by 67%. El-Shami et al. (1985) showed that garlic extract could inhibit spore germination and mycelial growth of the fungus in a similar manner to five fungicides. Soaking watermelon seeds in the extract gave better control of seedling wilt than seed treatment with benomyl, carboxin, carboxin/captan or carboxin/thiram.

Seed Health Tests

Seed tests for F. oxysporum f.sp. niveum could be adapted from seed-washing tests, seed-incubation methods and seedling symptoms because both the inside and outside of watermelon seeds can be contaminated with the conidia and mycelia of the pathogen. Isolates of the fungus can be obtained for assessing pathogenicity as follows:

Seedling inoculation in pots by the standard root-dip method (after Kleczewski and Egel, 2011; Jo et al., 2015):

Watermelon seedlings were germinated in vermiculite, sand, or other potting media under the conditions described for the continuous-dip inoculation technique. Seedlings with not more than two true leaves are removed and roots are washed with running water, then dipped in the conidial suspensions of Fusarium isolates for 30 seconds. Seedlings were transplanted into peat pots (10 x 10 x 15 cm). Disease incidence was observed after 10-21 days and expressed as the percentage of seedling mortality compared with that of uninoculated controls. Experiments consisted of four replicates with 10 seedling per treatment arranged in a completely random design. The minimum number of seedlings that should be inoculated per isolate is 10. Tests should be repeated to ensure reproducibility.

One-sided wilting of watermelon plants that appears as vines begin to lengthen combined with a reddish-brown discoloration of the xylem in the main stem or at the base of main vines are diagnostic symptoms for rapid identification of Fusarium wilt in the field that can be used by scouts and agricultural advisors to make a preliminary diagnosis.

Fusarium oxysporum f.sp. niveum has been identified using polymerase chain reaction (PCR) assays, which can be used for rapid and reasonably accurate diagnosis and to distinguish pathogenic from non-pathogenic isolates of F. oxysporum (Lin et al., 2010). These primers correctly identified non-pathogenic isolates 96% of the time but correctly identified pathogenic isolates representing races 0, 1 and 2 from South Carolina, USA, only 71% of the time (Keinath et al., 2020). Other primers developed by Zhang et al. (2005) yielded non-specific reactions with several formae speciales in tests in another laboratory and are not accurate enough for diagnosis or identification of F. oxysporum f.sp. niveum ((W Patrick Wechter, unpublished data, USDA, Charleston, Souith Carolina, USA). A small sample of F. oxysporum f.sp. niveum race 0 and race 1 isolates was reported to contain copies of secreted-in-xylem (SIX) elicitor number 6, but this protein was absent in race 2 isolates (Niu et al., 2016). Primers designed to amplify this protein correctly identified non-pathogenic isolates and isolates of race 0, 1 and 2 from South Carolina, USA, 100, 33, 54 and 72% of the time, respectively (Keinath et al., 2020).

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Host-Plant Resistance

One of the best control measures against Fusarium wilt of watermelon is the use of resistant cultivars (Hopkins and Elmstrom, 1984). The cultivar Calhoun Gray is highly resistant to Fusarium wilt caused by race 1 of the pathogen in many countries and regions, including Queensland, Australia (Inch et al., 1972), Bangladesh (Mondal and Rashid, 1990), Italy (Cirulli, 1974), Montenegro (Mijuskovic and Vucinic, 1977), California, USA (Paulus et al., 1976), Florida, USA (Hopkins and Elmstrom, 1975, 1981) and South Carolina, USA (Barnes, 1972). However, after race 2 appeared in Israel and Texas, USA, Calhoun Gray became susceptible (Nepa et al., 1985; Netzer and Martyn, 1989).

In certain locations, cultivars in addition to Calhoun Gray also performed well in Fusarium-infested soil or had improved horticultural characters. In the USA, Elmstrom and Crall (1979) reported that the content of soluble solids and resistance of Dixielee watermelon to F. oxysporum f.sp. niveum was higher than in Crimson Sweet or Charleston Gray. Elmstrom and Hopkins (1981) reported that Smokylee and Summit were highly resistant, with <20% seedling wilt, and produced adequate yields even under heavy infestation. Norton et al. (1985) reported that AU-Jubilant, an inbred line from the cross Jubilee X PI271778, and AU-Producer, an inbred line from the cross Crimson Sweet X PI189225, were high-yielding and had resistance to race 1 of the pathogen, and were adapted for the southeastern part of USA. In Florida, Crall and Elmstrom (1986) developed the 'icebox' (small fruited) cultivars Minilee and Mickylee, which are resistant to the pathogen; both had a long shelf-life and appeared to be suitable for year-round production in Florida.

In China, the cultivar Calhoun was found to have resistance to Fusarium wilt (Zhang et al., 1995; Zhou and Kang, 1996a). The cultivar Jingkang 2 is resistant to Fusarium wilt and yields up to 60-75 ton/ha (Zhou, 1995). In order to quicken the procedure to find resistant breeding lines to Fusarium wilt, Huang et al. (1981) and Yu and Wang (1990) developed a simple, rapid and effective method using toxic metabolites extracted from the pathogen to screen resistant cultivars. Results showed that resistance to the toxin was correlated with resistance of the cultivars to the pathogen. A method for identifying resistance in watermelon to Fusarium wilt was recommended; this procedure combined root dip inoculation and spore culturing at the seedling stage with evaluation in the field at a later stage (Wang and Zhang, 1988; Zhang and Wang, 1991). The results of Yu et al. (1995) showed that the inheritance of resistance conformed to the additive-dominance model. The additive effect was major and the susceptibility was partially dominant.

The ability to map resistance genes coupled with marker-assisted selection is accelerating breeding watermelon cultivars resistant to race 2. Several research groups in China and the USA have identified quantitative-trait loci (QTL) located on different chromosomes and linked to resistance to race 1 or race 2. Resistance to race 1 in most modern hybrid cultivars, including Calhoun Gray, has been mapped to chromosome 1 (Yi et al., 2015; Fall et al., 2018). Branham et al. (2019) reported another race 1 resistance gene located on chromosome 9. QTLs for race 2 resistance have been mapped to chromosomes 9, 10, and 11 (Yi et al., 2015; Meru and McGregor, 2016; Branham et al., 2017).

Resistant Rootstocks

In Japan, the new bottle gourd cultivar Renshi was bred for use as a rootstock for watermelon, which prevented acute wilt of watermelon grafted on bottle gourd. It is tolerant of both very dry and wet soil. Graft compatibility with watermelon was good and the growth, quality and cropping characteristics of watermelons grafted onto Renshi were similar to those on other rootstocks (Matsuo et al., 1985).

In Korea, 19 cultivars, including Cucurbita moschata cv. Choseun, C. maxima cv. HA Sintojwa and C. pepo cv. Vegetable Spaghetti were selected as resistant to F. oxysporum f.sp. cucumerinum, niveum and melonis. A number of cultivars were selected as promising breeding lines for rootstocks, including Taeyang, Kangryeog, Strong Ilhwi and Vegetable Spaghetti. These cultivars grew at low temperatures and were resistant to Fusarium wilt (Kim et al., 1997).

In China, watermelon rootstock ChaoFeng F1 was found to be immune to Fusarium wilt, and watermelons grafted onto this rootstock matured early. Yields were 15-17.6% higher than those of watermelons grafted on a common gourd rootstock. The grafted watermelons had thinner skins and more deeply-coloured flesh (Zheng, 1995).

In Italy, D’Amore et al. (1996) reported that grafting watermelon onto rootstocks of the genera Cucurbita or Lagenaria gave protection from F. oxysporum f.sp. niveum and improved the absorption of water and nutritive elements. Yields were increased and there was an improvement in the quality of fruits.

In Bangladesh, watermelon cv. Top Yield was grafted onto nine different cucurbit rootstock seedlings: C. moschata cultivars Mammoth King, Round and Oblong; three Lagenaria leucantha (L. siceraria) cultivars; Benincasa hispida; and wild watermelon (C. maxima). Comparison was made with ungrafted watermelons. Fruit yields ranged from 13.6 kg/plant on B. hispida to 29.6 kg/plant on L. leucantha cv. Summerking, compared with 15.4 kg/plant in the ungrafted controls. Fusarium wilt was a problem only in plants on wild watermelon rootstocks (10% were affected) and in ungrafted plants (46% affected) (Mondal et al., 1994).

In the absence of watermelon cultivars resistant to Fusarium wilt caused by race 2, grafting susceptible cultivars onto interspecific hybrid squash rootstocks (C. maxima × C. moschata) or bottle gourd (L. siceraria) rootstocks protects grafted scions from Fusarium wilt (Miguel et al., 2004; Davis et al., 2008; Keinath and Hassell, 2014b). Interspecific hybrid squash and bottle gourd possess nonhost resistance to F. oxysporum f. sp. niveum races 1 and 2 (Yetıșır et al., 2003; Keinath and Hassell, 2014a). Thus, grafting is effective regardless of which race is present or predominates in a field. In Turkey, grafting ‘Crimson Tide,’ a diploid watermelon cultivar resistant to race 1, onto bottle gourd increased yields in soil infested with race 2 of F. oxysporum f.sp. niveum (Yetıșır et al., 2003). In Spain, in soil infested with unidentified races of F. oxysporum f.sp. niveum, grafting triploid watermelon onto interspecific hybrid squash ‘Shintoza’ increased yields over threefold (Miguel et al., 2004). In Mexico, grafting triploid watermelon susceptible to Fusarium wilt onto interspecific hybrid squash consistently increased total weight of fruit produced in Fusarium-infested soil (Álvarez-Hernández et al., 2015).

Biological Control

In experiments conducted in Florida, USA, soil suppressiveness of F. oxysporum f.sp. niveum occurred through more than five successive greenhouse plantings of the watermelon cultivar Florida Giant (susceptible to F. oxysporum f.sp. niveum). The authors suggested that cultivar differences were responsible for the promotion of differences in rhizosphere microflora populations associated with soil suppressiveness (Hopkins et al., 1987; Larkin et al., 1993b). Specific isolates of non-pathogenic F. oxysporum from suppressive soil were the only organisms consistently effective in reducing the disease (35-75% reduction) The mode of action of these saprophytic isolates of F. oxysporum was induced systemic resistance with biological control potential (Larkin et al., 1996).

In Taiwan, Huang et al. (1989) reported that five saprophytic isolates of fungi and 10 isolates of bacteria were obtained from watermelon roots planted in eight soils from Taiwan. F. oxysporum, F. solani and Trichoderma sp. suppressed watermelon wilt caused by F. oxysporum f.sp. niveum.

In Egypt, Michail et al. (1989) reported that Fusarium wilt of watermelon could be controlled by cross protection. Prior inoculation of plants with F. oxysporum f.sp. cucumerinum, which causes cucumber wilt disease, followed by the pathogen 5 days later resulted in no apparent wilt symptoms on watermelon. Yu and Wang (1989) also reported cross protection using a weakly virulent isolate of F. oxysporum f.sp. niveum or an isolate of F. solani to inoculate plants 5 to 15 days before challenging them with the pathogen.

Soil Minerals

Calcium, phosphate and potassium deficiencies induce a higher incidence of Fusarium wilt. Calcium compounds and phosphate salts such as Ca(OH)2, Ca(NO3)2.4H2O, CaCO3, CaSO4, K2HPO4 and NaH2PO4.2H2O were strongly inhibitory to chlamydospore germination and promoted lysis of germ tubes. Mycelial growth of F. oxysporum f.sp. niveum in conducive soil was inhibited by Ca(OH)2, K2HPO4 and NaH2PO4.2H2O. Raising soil pH in Florida to 7.2-7.5 with hydrated lime reduced Fusarium wilt and increased yields of watermelon (Jones et al., 1975). To avoid decreasing the pH, nitrogen fertilizer must be applied as nitrate, not as ammonium. However, Tsao and Zentmyer (1979) reported that the population of F. oxysporum f.sp. niveum was reduced from 96.45 to 66.0 and 71.4%, respectively, by the application of 1% urea plus Ca-superphosphate and potassium nitrate plus Ca-superphosphate, in sandy loam soil. Results from these studies taken together suggest that calcium could play an important role in suppressing F. oxysporum f.sp. niveum in soil.

However, this conclusion is contradicted by results of Lin et al. (1996), which indicate that CaCl₂ improved the germination and germ tube growth of chlamydospores of the pathogen in vitro. In addition, Hopkins and Elstrom (1976) found no significant differences in Florida in wilt incidence or yields between treatments of high soil pH (7-7.3) and all nitrate nitrogen, and lower soil pH (5.2-6) and 25% ammonia nitrogen.

Chemical Control

In China, homodemycine, a copper complex of several different amino acids which is non-residual and has low toxicity, was more effective at controlling F. oxysporum f.sp. niveum on watermelons in field trials than carbendazim and thiophanate-methyl. It was also shown to stimulate growth of uninfected plants and could be used to increase production (Li and Liu, 1990).

Applications of prothioconazole at transplanting as a drench or applied after transplanting through drip irrigation or a foliar spray reduced Fusarium wilt in multiple locations. Prothioconazole combined with thiophanate-methyl was slightly more effective than prothioconazole in one experiment (Everts et al., 2014; Miller et al., 2020). A new fungicide, pydiflumetofen was more effective than prothioconazole in North Carolina, USA, and also increased weight and number of marketable fruit when applied twice, once as a drench at transplanting and 14 days later as a foliar spray (Miller et al., 2020).

Soil Solarization

In Korea, Kye and Kim (1985) reported that Fusarium wilt of watermelon may be effectively controlled by soil solarization in a closed plastic house during the hot summer season.

In the USA, Martyn and Hartz (1986) also reported that soil solarization for either 30 or 60 days delayed the onset of wilt symptoms and reduced total disease incidence in a F. oxysporum f.sp. niveum-susceptible cultivar Sugar Baby, but did not provide complete control of the disease. The effects lasted over two growing seasons, control being best during the first year.

Freeman and Katan (1988) reported that sublethal heating of conidia and chlamydospores of F. oxysporum f.sp. niveum at 38-42°C caused up to 33% reduction in propagule viability and weakened the surviving propagules. This weakening effect was expressed as a delay in germination, reduction in the growth of conidial and chlamydospore germ tubes and an enhanced decline of the population density of viable conidia in soil. Disease incidence in watermelon seedlings inoculated with heat-treated conidia of the pathogen was reduced by 35-82%.

In field trials conducted in Cyprus during 1984-86 on soil naturally infested with F. oxysporum f.sp. niveum, solarization raised the soil temperature by 7-10°C and reduced soil inoculum density by ca 90% (Ioannou and Poullis, 1990).

30/03/20 Reviewed by:

Anthony Keinath, Professor of Plant Pathology, Clemson University, Coastal Research and Education Center, Charleston, SC, USA