Evaluation of Hydrangea Cultivars for Tolerance Against Root Rot Caused by Fusarium oxysporum

The floricultural industry, including ornamental landscape plants, is growing continuously. The demand for the production and selection of new cultivars with superior traits such as novel color, shapes, forms, and floral longevity is also increasing (Naing et al. 2021). Ornamental plants grown as cut flowers, pot plants, or landscape plants are often challenged by several pathogens with effects lasting long after the purchase (Leus 2018). Hydrangea L. is an important ornamental genus comprising more than 75 species, with sales of $47.43 million as pot plants, $1.3 million as landscape plants, and $1.2 million as cut flowers in 2019 (Dirr 2004; NASS 2019). Hydrangea macrophylla (Thunb. Ex J.A. Murr.) Ser., is the most widely grown member of Hydrangea genus, and now popularity and the consumer interest is increasing in panicle (H. paniculata Siebold), smooth (H. arborescens L.), and oakleaf (H. quercifolia Bartram) hydrangea (Dirr 2004).

H. macrophylla is a popular summer-flowering shrub valued for its large round inflorescence that’s comprised of showy sterile flowers and a few fertile flowers present inside brightly colored corymbs. Plants can grow from 1 to 2 or as high as 3 m and are listed to grow in USDA hardiness zones 6 to 9 (Dirr 2004; Reed 2002). H. quercifolia is a large ornamental shrub native to the southeastern United States and is grown in USDA hardiness zone 5 to 9 with about 40 described cultivars which have become increasingly popular among consumers and landscape designers in recent years (Dirr 2009; Reed and Alexander 2015; van Gelderen and van Gelderen 2004). Most oakleaf hydrangea cultivars primarily differ in plant size and floral characteristics. The inflorescences are exceptionally showy, with longer panicles (up to 30 cm length) with creamy white sepals, and their leaves are large and coarse, turning deep shades of maroon, burgundy, and red in the fall (Cochran et al. 2014). H. paniculata is native to Japan and China. It is cold hardy and grown in USDA hardiness zone 4, can grow 3 to 6 m high, and produces 15- to 20-cm-long panicles with white to pale pink flowers in midsummer (Reed 2004). H. arborescens is a smooth hydrangea that can grow to 1 to 3 m tall, is quite compact, and has a largely unbranched stem (Dirr 2004).

Several foliar diseases caused by different fungi, bacteria, and viruses as well as root diseases caused by Phytophthora, Pythium, and Armillaria are limiting the healthy production of Hydrangea spp. (Baysal-Gurel et al. 2016a, b). Fusarium oxysporum, a soilborne fungi with nonpathogenic and pathogenic strains is an Ascomycota fungus under class Sordariomycetes and family Nectriaceae of order Hypocreales with more than 120 formae speciales. The fungus has a wide host range, including 30 ornamental species (Burgess et al. 1981; Lecomte et al. 2016; Michielse and Rep 2009). Oakleaf hydrangea ‘Pee Wee’ and ‘Queen of Hearts’ with root rot symptoms were brought to Baysal-Gurel laboratory at Otis L. Floyd Nursery Research Center, Tennessee State University, McMinnville, TN, in May 2018. Disease incidence levels recorded from a nursery in Tennessee were 40 and 60% out of 500 ‘Pee Wee’ and ‘Queen of Hearts’ plants, respectively. The causal organism was identified as F. oxysporum using morphological and molecular diagnosis process (Neupane et al. 2023).

During infection of plants by Fusarium, a well-known nonspecific phytotoxin fusaric acid (FA) is produced that plays a direct role in pathogenesis by altering membrane permeability, ATP synthesis inhibition, or chelation of metal ions and is potentially toxic to animals (Bouizgarne et al. 2006; DʼAlton and Etherton 1984; Hidaka et al. 1969; Ruiz et al. 2015). The compound is reported accelerating disease development in plants by changing plant physiology (Dong et al. 2012; López-Díaz et al. 2018; Telles-Pupulin et al. 1996; Wu et al. 2008). It was first discovered from laboratory culture of F. heterosporum Nees. FA and was the first metabolite used in the pathogenesis of F. oxysporum f. sp. lycopersici Schlecht. emend. Snyd and Hans. causing wilt in tomato (Gaumann 1957; Yabuta et al. 1937). Toxicity may vary from low to high depending on the pathogen species—low: <100 μg/g; moderate: 100 to 500 μg/g; and high: >500 μg/g based on the amount of FA extracted from the culture of different species of pathogen (Bacon et al. 1996). Unlike other secondary metabolites such as deoxynivalenol, moniliformin, or zearalenone, which are produced by limited species of Fusarium, FA is produced by at least 78 different strains of nine different species (Bacon et al. 1996). Testing the cultivars for the sensitivity to metabolites and secretions from fungi is a low-cost method in breeding for resistance (Barna et al. 2011). Toxin is required by the pathogen for causing disease, so cultivars/varieties which are tolerant to toxin produced by pathogens are tolerant to disease (Švábová and Lebeda 2005). The effect of FA has been previously used for in vitro selection of Fusarium-resistant variants such as in banana plants (Matsumoto et al. 1995), barley (Barna et al. 2011), pineapple (Borras et al. 2001), gladiolus (Remotti et al. 1997), and tomato (Shahin and Spivey 1986).

Stress conditions which include biotic (pathogen infection) and abiotic (drought, heat, cold, and salinity) stress can cause impact on growth and yield of plants (Suzuki et al. 2014). Plants and fungi are found producing mannitol which may serve as a metabolic store or osmolyte as well as a powerful quencher of reactive oxygen species (ROS) (Meena et al. 2015). Mannitol, a six-carbon sugar alcohol, is widely distributed in nature and accumulates in response to abiotic stress as a compatible solute; it is found naturally in more than 100 higher plants and is also present in fungi, algae, and lichens (Patel and Williamson 2016). In plants, it is produced during the stressed condition as osmo-protectant, and it is reported with roles in metabolism and pathogenesis by fungi (Cheng et al. 2009). It can be extracted from both infected leaves (apoplast) and fungal spores (Voegele et al. 2005). Mannitol function in response to diverse biotic stress appears to be protection during biotrophic pathogens attack by quenching the pathogen-induced ROS (Stoop et al. 1996). Many phytopathogenic fungi are found to synthesize mannitol and use it to quench plant defense mediated by ROS (Jennings et al. 1998). Mannitol, which is a sugar storage carbohaydrate, is found in all parts of pathogens such as mycelia, fruiting body, and spores. Under the starvation, mannitol helps in spore germination (Meena et al. 2016). It is also shown that pathogen attack can induce mannitol dehydrogenase (MTD) expression in the nonmannitol-containing host. During pathogen invasion, mannitol might help to counter fungal suppressive mechanism by catabolizing mannitol derived from fungal origins (Jennings et al. 2002).

Diseases caused by Fusarium spp. are managed primarily using chemical soil fumigation and resistant cultivars. Methyl bromide was heavily used as a soil fumigant but has been phased out due to its harmful effect on the environment. Use of resistant cultivars is one of the most cost effective and environmentally safe methods in plant disease management (Fravel et al. 2003; Matthiessen and Kirkegaard 2006). Efforts for developing superior cultivars are mostly based on traits such as growth habit and pest resistance. Efficient screening techniques to select suitable cultivars are important before conducting any breeding program (Shakoor et al. 2017). These screening and bioassays with pathogens should be repeatable, controlled, and reflect plants’ performance regarding resistance to pathogen diversity (Leus 2018). This study was carried out to evaluate the response of 15 different cultivars of hydrangea from four different species against root rot caused by F. oxysporum by assessing root rot severity, pathogen recovery, and recovery of FA from the infected roots. Also, mannitol concentration was analyzed to see the role of mannitol in pathogenesis by F. oxysporum to the infected plants.

Materials and Methods

Preparation of rooted cuttings of hydrangea cultivars

Softwood cuttings from 2020 and 2021 spring flush were obtained from USDA plant germplasm collection present at the Tennessee State University Otis L. Floyd Nursery Research Center in McMinnville, TN. Cultivar details are provided in Table 1. The cuttings were sprinkled with water and placed in a cool chamber for 1 h before preparing cuttings for rooting. Rooting dip was prepared by adding 1,500 ppm of indole-3-butyric acid (IBA) with 750 ppm of naphthaleneacetic acid (NAA). Cuttings were prepared by keeping at least three internodes and two leaves at growing tips that were dipped in the rooting hormone for 5 s and placed in potting mix pine bark (Sims Bark Co., Inc., Tuscumbia, AL, U.S.A.) + Osmocote Pro (Bes-Tex Supply, LLC, San Angelo, TX, U.S.A.) inside a moist chamber for rooting at 98% relative humidity and an average mean temperature of 27°C.

Table 1. Description of all cultivars used in the experiment

Cultivar	Scientific name	Length and width	Characteristics	Citation
Picnic Hill	Hydrangea quercifolia Bartram	Generally, 2 m in height and width higher than length	Medium sized flowers, white colored and turns pink	Dirr 2004
Alice	H. quercifolia	Can grow up to 4 m tall, exhibit apical dominance	Sepals turn pink or brown after aging	Dirr 2004
John Wayne	H. quercifolia	Can grow 1.8 m high and 2 m wide	Midsummer flowering, creamy white panicles and sterile flowers turn pink	https://www.backyardgardener.com/plantname/hydrangea-quercifolia-john-wayne-hydrangea/
Queen of Hearts	H. quercifolia	Can grow 1.9 to 3.5 m tall	Early summer flowering, inflorescence 12 cm wide and 30 cm long, sepals are open white turning pink and then to vibrant rose color	Reed and Alexander 2015
Munchkin	H. quercifolia	Compact, can grow 0.9 m high and 1.4 m wide	Summer flowering, inflorescence 12 cm wide and 17 cm long, open white to medium pink flowers	Reed 2010
Back Porch	H. quercifolia		Vigorous, large white flowers that turn pink, early flowering	Dirr 2004
Pee Wee (quercifolia)	H. quercifolia	Compact, can grow 0.6 to 1 m in height and width, dwarf selection	Small inflorescence, sterile flowers sepals do not cover inflorescence	Dirr 2004
Harmony	H. quercifolia		Fully sterile flower	Dirr 2004
Snow Flake	H. quercifolia	Can grow 2 to 2.5 m tall	Panicles 30 to 40 cm long, double flower inflorescence; white flowers	Dirr 2004
Snow Giant	H. quercifolia	Can grow 2 to 2.5 m tall	Large, lightly fragrant, and sterile snow-white flowers	Dirr 2004
Snow Queen	H. quercifolia	Compact, can grow as high as 1.8 m	Inflorescences are 15 to 20 cm long, white flowers which turn pink in maturity	Dirr 2004
Frosty	H. arborescens subsp. discolor	Can grow 0.9 to 1.2 m tall	Sterile flowers with single sepals and forms mophead inflorescence, flowers are white in color	Dirr 2004
Pee Wee (paniculata)	H. paniculata	Can grow 3 to 6 m tall, most cold hardy	Inflorescences are 15 to 20 cm long, white to pale pink flowers	Reed 2004 and Dirr 2004
Hayes Starburst	H. arborescens	Compact, can grow 0.9 to 1.2 m tall	Double white sepals, outer shooting florets and turns green, white to parchment after aging	Dirr 2004
Nikko Blue	H. macrophylla	Can grow 1.5 to 2 m tall	Free flowering, entire sepals, cream flowers with blue margin	Dirr 2004

View as image HTML

Preparation of fungal inoculum

F. oxysporum culture (Isolate FBG2018_490 GenBank accession number ON227492) isolated from a hydrangea plant in 2018 from a nursery in McMinnville, TN, was obtained from Baysal-Gurel laboratory. The cultures were maintained on Fusarium selective medium at room temperature (25°C) for 10 days before inoculum preparation. A modified Czapeck-Dox medium containing antimicrobial agents was used to prepare Fusarium selective medium (Tio et al. 1977). Next, 20 g of dextrose (Sigma-Aldrich, St. Louis, MO, U.S.A.), 0.5 g of KH₂PO₄ (Sigma-Aldrich), 2 g of NaNO₃ (Thermo Fisher Scientific, Branchburg, NJ, U.S.A.), 0.01 g of FeSO₄.7H₂O (VWR Life Sciences, Solon, OH, U.S.A.), 0.5 g of MgSO₄.7H₂O (VWR Life Sciences), 1 g of yeast extract (Thermo Fisher Scientific), and 20 g of agar (Thermo Fisher Scientific) were autoclaved with 995 ml of distilled water at 121°C at 15 psi for 30 min. After cooling at 55°C, 0.75 g of PCNB (pentachloronitrobenzene [Sigma-Aldrich]), 0.1 g of streptomycin sulfate (Thermo Fisher Scientific), and 0.01 g of neomycin sulfate (Sigma-Aldrich) were added (Tio et al. 1977). For inoculum, conidial suspensions were prepared by scraping the fungal mycelium and dissolving in distilled water at the rate of four plates/liter of water. The final solution was adjusted to 1 × 10⁶ conidia/ml with the number of conidia counted using a hemocytometer (Hausser Scientific, Horsham, PA, U.S.A.).

Experimental design and conditions

The experiment was set up as a completely randomized design with six replications of each treatment (inoculated and noninoculated) in a greenhouse facility of the Tennessee State University, Otis L. Floyd Nursery Research Center in McMinnville, TN. Two trials were conducted: trial 1 between 15 February and 29 June 2021 and trial 2 between 30 September 2021 and 15 February 2022. Six-month-old rooted cuttings, about 15 to 20 cm in height, were grown in 16.0- by 20.0-cm-diameter containers (Nursery Supplies Inc., Chambersburg, PA, U.S.A.) using potting mix (Morton’s Nursery Mix: Canadian sphagnum peat [55 to 65%]) (Morton’s Horticultural Products, McMinnville, TN, U.S.A.). Two weeks after transplanting, 12 plants per cultivar were divided for the experiment; six remained noninoculated (controls) and received 150 ml of distilled water (without F. oxysporum), while six were drench inoculated with 150 ml of conidial suspension of the isolate (F. oxysporum maintaining concentration of 1 × 10⁶ conidia/ml). Plants were watered twice a day for 2 min using an overhead irrigation system with 150 ml of water per event. Initial plant height and width were measured on 16 February 2021 (trial 1) and 1 October 2021 (trial 2). Four months after treatment, final height, width, total plant fresh weight, and total root fresh weight were recorded on 29 June 2021 (trial 1) and 14 February 2022 (trial 2). Mannitol and FA were extracted from the root portion of inoculated and noninoculated plants for further analysis. For the first trial period, maximum, minimum, and average temperatures were measured as 28.37, 16.93, and 21.9°C, respectively. The maximum, minimum, and average relative humidity were 100, 88.2, and 98.12%, respectively. For the second trial period, maximum, minimum, and average temperatures were 27.21, 16.2, and 21.1°C, respectively, and maximum, minimum, and average relative humidity were 100, 89.6, and 99.6%, respectively.

Extraction of FA

FA was extracted as described by Dong et al. (2016). Roots from both inoculated and noninoculated hydrangea cultivars were weighed (50 mg) and crushed in a bullet blender (Storm 24 Quasar Instruments, Colorado Springs, CO, U.S.A.) using equal volumes of 1% KH₂PO₄ (Sigma Aldrich) and MeOH (Thermo Fisher Scientific). The suspension was centrifuged for 15 min at 10,000 rpm and extracted. The pH of the supernatant was adjusted to 2.5 using 2M hydrochloric acid (Thermo Fisher Scientific), which was then added to 25 ml of methylene chloride for extraction using a rotary evaporator at 45°C. The residue was extracted with 3 ml of methanol (Thermo Fisher Scientific) and stored at −20°C until analysis using high-performance liquid chromatography (HPLC manufactured by Shimadzu Corporation, Kyoto, Japan).

FA analysis using HPLC

FA analysis using HPLC was carried out by the process as described by Ding et al. (2015) using an Agilent HC-C18 column (Luna 5 μm, manufactured by Phenomenex, Torrance, CA, U.S.A.). Elution was done using a mobile phase (20% methanol + 48% filtered HPLC grade water + 32% H₃PO₄ (0.43%) for 15 min with a UV detector at 228 nm, and the flow rate was 1 ml/min using an autosampler (Thermo Fisher Scientific). The samples were filtrated through 0.45-μm filters (Whatman Uniflo Syringe Filter, Sigma-Aldrich) before being injected through the autosampler. Pure FA was used to prepare five standard solutions: 10, 20, 30, 40, and 50 μg/ml, which was then used to determine absorption peak and retention time. The area of the peak of standards was used to plot a curve and a linear regression equation X = (Y + 80,681)/33,462 (where X is the concentration [μg/g] in a given sample, and Y was the height of peak recorded at 2.9 min) was created from the standard curve to find the concentration of samples tested.

Extraction and analysis of mannitol

Mannitol extraction and analysis from roots of inoculated and noninoculated hydrangea plants were done as described by Dong et al. (2016). The roots were dried at 60°C for 48 h and then ground using Cyclone Sample Mill (UDY Corporation, Fort Collins, CO, U.S.A.). Dried root powder (0.5 g) was mixed with 10 ml of distilled water; boiled for 2 h at 100°C; and extracted, and the extraction process was repeat after adding 10 ml of water. The filtrate collected was adjusted to a 25-ml volume, and 1 ml of solution was mixed with 1 ml of 0.015 mol/liter sodium periodate (Thermo Fisher Scientific) and left for 10 min. The solution was mixed with 2 ml of 0.1% rhamnose (Sigma-Aldrich) and 4 ml of Nash reagent. Nash reagent was prepared by fresh mixing an equal volume of 2 mol/liter ammonium acetate (Fisher Chemical, Fair Lawn, NJ, U.S.A.); acetic acid (Fisher Chemical); and acetyl acetone (TCI America, Portland, OR, U.S.A.). The final mixture was placed in a water bath for 15 min at 53°C. Pure mannitol (Thermo Fisher Scientific) was used to prepare five standard solutions: 10, 20, 30, 40, and 50 μg/ml. The absorbance of all samples and standards were measured on a spectrophotometer (BioTek Instruments, Winooski, VT, U.S.A.) at 412 nm. A curve was prepared plotting the standard absorbance and a liner regression equation X = (Y – 0.1001)/0.0052 (where X is the concentration [μg/g] in a given sample, and Y is the spectrophotometer absorption reading) was created from the standard curve to find the concentration of samples tested.

Data collection and statistical analysis

Plant height and width were measured at the beginning and end of each experiment; total plant fresh weight (root and shoot) and total root fresh weight (roots were cut from the plant at the base of the root collar) were recorded at the end of each experiment for analyses. Total height and width increase was measured by subtracting the height and width of a plant recorded at the beginning from the height and width of the plant recorded at the end of the experiment. Roots were washed with water to remove debris. Root disease severity was assessed visually using a scale of 0 to 100% of total root affected at the end of each trial, where 0% is no disease severity, and 100% is the death of plants. Ten randomly selected root samples (1 cm long) from the root area of each hydrangea plant (both inoculated and noninoculated) were plated individually on Fusarium selective medium to determine the percent recovery of F. oxysporum from root samples. Recovered F. oxysporum out of total roots plated were presented in percentage value by recording the number of roots with pathogen growth to the total number of roots plated and incubated for 3 days. The mean was calculated from six replications. Pathogen identity was confirmed by DNA extraction and amplification using primers ITS1/ITS4 (White et al. 1990), T1F/T222 (Stefańczyk et al. 2016), and EF1/EF-2 (O’Donnell et al. 1998) from the recovered pathogens.

Data analysis was performed using SAS software, Version 9.4 (SAS Institute Inc., Cary, NC, U.S.A.). Mean and standard error (SE) were calculated for plant height and width increase, total plant fresh weight, and root fresh weight. Analysis of variance (ANOVA) was used to partition the variance in disease severity, FA concentration, mannitol concentration, and pathogen recovery into sources attributable to cultivar and inoculation level (inoculated or noninoculated). When all values for noninoculated plants were zero (as observed for disease severity, FA concentration, and pathogen recovery), one-way ANOVA models were specified; a two-way ANOVA was performed to see the effects of inoculation in mannitol concentration. PROC GLIMMIX with Beta distribution was used for the percentage-valued plant parameters disease severity and pathogen recovery, and PROC MIXED was used to analyze continuous variables (FA and mannitol concentrations). Multiple comparisons were done using Sidak’s posthoc test at P < 0.05 when the interactions were significant. Pearson’s correlation coefficient value was calculated to estimate the relation among mannitol with disease severity and FA with disease severity of infected plants in both trials.

Results

Hydrangea cultivar screening for Fusarium root rot

Hydrangea roots were analyzed for root rot severity 4 months after inoculation of F. oxysporum in 15 cultivars, and all inoculated cultivars showed disease severity in both trials, while some cultivars were tolerant to disease. A significant difference in root rot severity was observed among the different species used for screening. Both trial results showed that the cultivars from species such as H. macrophylla ‘Nikko Blue’, H. paniculata ‘Pee Wee’, and H. arborescens ‘Frosty’ and ‘Hayes Starburst’ showed significantly lower root rot severity compared to cultivars from H. quercifolia (Fig. 1; P < 0.0001 [trial 1]; P < 0.0001 [trial 2]). There was no significant difference in root rot severity between the two cultivars from H. arborescens in both trials. Further, root rot severity was significantly different among 11 cultivars screened from H. quercifolia species in both trials (Fig. 1; P < 0.0001 (trial 1); P < 0.0001 (trial 2). In trial 1, among the cultivars from H. quercifolia, the highest root rot severity was observed in ‘Harmony’ (59.9%) and was lowest in cultivars such as ‘Alice’, ‘John Wayne’, ‘Picnic Hill’, and ‘Snow Flake’. In trial 2, root rot severity was highest in ‘Back Porch’ (60%) and lowest in ‘John Wayne’ (46.6%) among H. quercifolia. ‘Nikko Blue’ was the cultivar among all 15 evaluated cultivars with lowest root rot severity in both trials (17.5% in trial 1 and 22.5% in trial 2). Root rot severity was not recorded from noninoculated hydrangea from all cultivars evaluated in the experiment (data not shown).

Download as PowerPoint

Recovery percentage of F. oxysporum from hydrangea roots was significantly different among different species used for screening. Both trial results showed that the cultivars from species such as H. macrophylla ‘Nikko Blue’, H. paniculata ‘Pee Wee’, and H. arborescens ‘Frosty’ and ‘Hayes Starburst’ were showing significantly lower F. oxysporum recovery compared to cultivars from H. quercifolia (Fig. 2; P < 0.0001 [trial 1]; P < 0.0001 [trial 2]). There was no significant difference in F. oxysporum recovery between the two cultivars from H. arborescens in both trials. Further, F. oxysporum recovery comparison among 11 cultivars from H. quercifolia was significantly different in trial 1 (Fig. 1; P < 0.0001); ‘Back Porch’ and ‘Harmony’ showed the highest recovery of F. oxysporum (63.3% each). The lowest F. oxysporum among the H. quercifolia was from ‘John Wayne’ (51.6%). In trial 2, there were no significant differences in F. oxysporum recovery among the cultivars from H. quercifolia. In both trials, among all cultivars from four different Hydrangea spp., the lowest recovery of F. oxysporum was from ‘Nikko Blue’ (23.2% in trial 1 and 22.5% in trial 2). F. oxysporum was not recovered from any noninoculated hydrangea plants from all cultivars of four species of hydrangea used in the experiment in both trials (data not shown).

Download as PowerPoint

All cultivars were evaluated for height increase, width increase, total plant fresh weight, and total root fresh weight at the end of both trials (data not shown).

FA analysis

FA was analyzed after extraction from the inoculated and noninoculated hydrangea plants from 15 different cultivars. No FA peak was observed from any noninoculated plants in the HPLC chromatogram. However, there was a significant difference among the hydrangea species evaluated for FA recovery in both trials (Fig. 3; [P < 0.0001, trial 1] and [P < 0.0001, trial 2]). Results from trial 1 showed that recovery of FA from inoculated roots of H. macrophylla and H. paniculata was lowest compared to four different species. There was a significant difference between two cultivars from H. arborescens, ‘Hayes Starburst’ and ‘Frosty’, for FA recovery in trial 1. Significantly higher FA concentration was recovered from ‘Hayes Starburst’ (31.7 μg/g) compared to ‘Frosty’ (11.9 μg/g; Fig. 3). Among 11 cultivars from H. quercifolia species, there was a significant difference in FA recovery concentration. The highest FA recovery concentration was observed from ‘Harmony’ (34 μg/g), and the lowest was from ‘Snow Flake’ (9.1 μg/g). Significant (P < 0.0001) positive Pearson’s correlation coefficient (0.434) was found between root rot severity and FA concentration in trial 1.

Download as PowerPoint

Results from trial 2 showed that recovery of FA from inoculated roots of H. macrophylla, H. arborescens, and H. paniculata was lowest compared to four different species except for one cultivar from H. quercifolia, ‘Snow Flake’. There was no significant difference between two cultivars from H. arborescens, ‘Hayes Starburst’ and ‘Frosty’, in FA recovery in trial 2. Among 11 cultivars from H. quercifolia species, there was a significant difference in terms of FA recovery concentration. The highest FA recovery concentration was observed from ‘Harmony’ (37.7 μg/g), and the lowest was from ‘Snow Flake’ (10.8 μg/g). Significant (P < 0.0001) positive Pearson’s correlation coefficient (0.439) was found between root rot severity and FA concentration in trial 2.

Mannitol analysis

There was no significant effect of treatments (inoculated/noninoculated) on mannitol concentration in the cultivars from H. macrophylla, H. arborescens, and H. paniculata in trial 1. However, in H. quercifolia cultivars, all noninoculated hydrangea plants have significantly lower mannitol concentration compared to the inoculated plants (Fig. 4; P < 0.0001). Higher mannitol concentration was observed from cultivars such as ‘Back Porch’, ‘Snow Flake’, ‘Snow Giant’, and ‘Snow Queen’ in trial 1. Significant (P < 0.0001) positive Pearson’s correlation coefficient (0.49) was found between root rot severity and mannitol concentration in trial 1.

Download as PowerPoint

The concentration of mannitol was not significantly different between inoculated and noninoculated hydrangea cultivars from H. macrophylla, H. arborescens, and H. paniculata in trial 2. Among 11 cultivars from H. quercifolia species, mannitol concentration was significantly different on inoculated and noninoculated plants. Mannitol concentration was significantly higher in the inoculated plants than noninoculated plants from the same cultivar from H. quercifolia. Mannitol concentration in inoculated roots of cultivars such as ‘Black Porch’, ‘Harmony’, ‘Snowflake’, ‘Snow Queen’, and ‘Snow Giant’ was higher compared to other cultivars from the same species (Fig. 5). Significant (P < 0.05) positive Pearson’s correlation coefficient (0.30) was found between root rot severity and mannitol concentration in trial 2.

Download as PowerPoint

Discussion

This study reports the first experiment conducted in selection of tolerant hydrangea cultivars against root rot caused by F. oxysporum based on disease severity and pathogen, FA, and mannitol recovery from the infected roots. Some hydrangea cultivars were previously evaluated for different diseases such as powdery mildew caused by Erysiphe polygoni (Vaňha) Weltzien (Li et al. 2009) and leaf spot disease caused by six pathogens Corynespora cassiicola (Berk. and M.A. Curtis) C.T. Wei, Cercospora spp., Myrothecium roridum Tode, Glomerella paniculata (Stoneman) Spauld. and H. Schrenk, Phoma exigua Sacc., and Botrytis cinerea Pers. (Mmbaga et al. 2012). Results from the current experiment show that none of the hydrangea cultivars tested were tolerant to F. oxysporum, and cultivars from H. quercifolia were more susceptible to root rot compared to cultivars from other Hydrangea species. H. macrophylla ‘Nikko Blue’ was more tolerant to root rot compared to other evaluated cultivars in this study based on consistently low disease severity, pathogen recovery, and FA concentration. ‘Nikko Blue’ was also found resistant to leaf spot disease caused by different pathogens (Mmbaga et al. 2012), although it was highly susceptible to powdery mildew caused by E. polygoni (Li et al. 2009). Among the cultivars from H. quercifolia, ‘Snow Flake’ and ‘John Wayne’ were more tolerant to F. oxysporum.

FA recovery from the hydrangea roots was used as one of the parameters for evaluating different cultivars against root rot caused by F. oxysporum. FA was recovered from the roots of all hydrangea plants inoculated with F. oxysporum and may play an important role in disease severity development. Previous studies have reported FA as a major mycotoxin responsible for causing disease by Fusarium spp., facilitating rapid symptom development, foliar desiccation, and necrosis in tomato wilt caused by F. oxysporum f. sp. lycopersici (Singh and Upadhyay 2014) and dry rot of Solanum tuberosum L. (Venter and Steyn 1998). Cultivars such as ‘Snow Giant’, ‘Harmony’, ‘Snow Queen’, and ‘Munchkin’ showed relatively higher FA concentration, disease severity, and F. oxysporum recovery from infected roots, showing that these cultivars are more susceptible to root rot disease compared with the other cultivars evaluated. Using FA culture filtrate as a tool for screening genotypes for conferring resistance in several plants has been previously reported, but pathogen-produced toxin during disease development is comparatively lower than in the pure culture due to favorable growth condition and nutritional composition of culture media (Matsumoto et al. 2010). So, plants inoculated with pathogen and recovery from the infected area is more accurate. The plants grown in experimental settings closely resemble the actual field settings with actual plant-pathogen interaction (Predieri 2001). Results from both trials of this experiment showed a significant positive correlation between FA concentration and root rot severity, and this was consistent with some previous findings in some other plants. Positive correlation was found between FA (recovered from all the parts of diseased banana plants) and disease incidence symptoms in banana plants infected with F. oxysporum f. sp. cubense (Dong et al. 2012); FA and virulence of F. oxysporum f. sp. carthami strains in safflower (Chakrabarti and Basu Chaudhary 1980); FA production and disease incidence from different Fusarium spp. in potato (El-Hassan et al. 2007); and virulence of 12 F. oxysporum isloates and FA causing dry rot of potato (Venter and Steyn 1998). In contrast, in an experiment with rice plants infected with F. proliferatum (Matsush.) Nirenberg ex Gerlach and Nirenberg, F. fujikuroi Sawada) Wollenw., and F. verticillioides (Sacc.) Nirenberg, the Fusarium species were able to cause bakanae disease, but no significant correlation was established between disease severity and FA content (Bashyal et al. 2016).

In this experiment, mannitol from roots of inoculated and noninoculated plants were analyzed to see the role in pathogenesis. Results showed that mannitol concentration significantly increased after pathogen infection, and the increase was significant (between inoculated and noninoculated) in the susceptible cultivars from H. quercifolia species. The increase was not significant in cultivars from H. arborescens, H. macrophylla, and H. paniculata. This rise indicated the role of mannitol in causing disease, and its role in fungal pathogenicity was previously explained (Patel and Williamson 2016). According to previously published reports, a dramatic rise of mannitol was observed after the infection of Uromyces fabae (Pers.) J. Schröt. in Vinca faba L. (Voegele et al. 2005) and after infection of Alternaria brassicola (Schwein.) Wiltshire in Brassica oleracea leaves (Calmes et al. 2013), and increases of mannitol and proline were observed after infection of tomato by A. alternata (Fr.) Keissl (Meena et al. 2016). Plants produce ROS in apoplast in response to pathogen attack, and this ROS serves as a signal to start downstream resistance which includes systematic acquired resistance and hypersensitive resistance (Meena et al. 2015). In response to mannitol production due to fungal infection, both mannitol and nonmannitol-producing plants produces mannitol catabolic enzyme MTD (Patel and Williamson 2016). This NAD-dependent MTD (mannitol dehydrogenase-nonspecific resistance gene in plants) was positively involved in plant response after the pathogen attack (Stoop et al. 1996; Williamson et al. 1995). Pathogen attack induced MTD expression in tobacco (nonmannitol-containing host), which showed that plants can counter fungal pathogenicity and catabolize mannitol produced by the tobacco pathogen A. alternata (Jennings et al. 1998). A transgenic eggplant (Solanum melongena L.) containing mannitol-1-phosphate dehydrogenase (mtlD) exhibited increased resistance against fungal wilts caused by F. oxysporum, Rhizoctonia solani J.G. Kuhn, and Verticilllium dahlia Kleb. under both in vitro and in vivo conditions, as well as resistance against abiotic stress (Prabhavathi et al. 2002; Prabhavathi and Rajam 2007). However, if MTD production is low or too late, mannitol accumulation forms a protective shell around the plant membrane, which will further protect the fungi and contribute to pathogenesis (Patel and Williamson 2016). Though a positive correlation between mannitol production and disease severity was observed in this study, production of MTD in nonmannitol-producing hydrangea was not sufficient to overcome mannitol production by pathogen, so fungal mannitol played a role in pathogenesis. Mannitol extracted from noninoculated plants may be the production to overcome stress from transplanting shock and change in environmental conditions from cuttings preparation to transplanting.

In conclusion, results from our experiments showed that F. oxysporum could be a serious issue for hydrangea, and no cultivars are resistant to disease. Cultivars from H. arborescens, H. macrophylla, and H. paniculata are more tolerant compared to cultivars from H. quercifolia species. Among the cultivars from H. quercifolia species, ‘Snow Flake’, ‘John Wayne’, and ‘Alice’ are more tolerant compared to others and can be further utilized for breeding or selection of hydrangea cultivars to grow in Fusarium-infected areas.