Introduction

Melon (Cucumis melo L.) is a cucurbit crop of worldwide economic importance whose production is favored by high temperature and low humidity conditions, which results in high-quality fruits (Pandey et al. 2016; Li et al. 2017). Global melon production in 2019 totaled approximately 27.3 million tons, 46.5% of which was produced in China, the main melon-producing country (FAO 2021). In Brazil, the Northeast region stands out with 95.8% (563,378 tons) of the national melon production (IBGE 2021). However, the expansion of farming areas along with intense and successive cultivation increases the incidence and severity of plant diseases, causing yield losses worldwide (Porto et al. 2019; Sales Júnior et al. 2019). In this scenario, powdery mildew is highlighted as the main melon shoot disease on a global scale (Li et al. 2017) and is usually caused by the fungi Podosphaera xanthii (Castag.) U. Braun & N. Shish. (previously Sphaerotheca fuliginea (Schlecht.Ex.Ft.) Pollacci) and Golovinomyces cichoracearum (S. Blumer) (Rur et al. 2018).

The white colonies produced by these fungi on the leaf surface reduce the photosynthetic activity of plants (Dallagnol et al. 2015). Subsequently, as the disease progresses, the infected tissues become chlorotic and show premature senescence, thus reducing fruit yield and quality (Wang et al. 2016; He et al. 2017). The colonies contain very light, wind-dispersed spores (Wang et al. 2016).

P. xanthii is the predominant melon pathogen (Wang et al. 2021). This species is adapted to high temperatures and dry climates because it does not require moisture on the leaf surface to infect (Ramos et al. 2013). However, although fungicides are known to reduce disease severity (Wang et al. 2016), their use might result in the emergence of pesticide-resistant P. xanthii and there are restrictions imposed by melon importers pertaining to the maximum residue limit (MRL) of agrochemicals and government agencies.

Consequently, there is interest in sustainable, effective alternative products to fungicides for managing P. xanthii on melon, including biological control, fertilizers, and raw milk. Biological control entails using microorganisms such as the fungus Trichoderma, notably effective in pathogen suppression by employing mechanisms such as antibiosis, competition, parasitism, and predation through one or more antagonistic mechanisms (Martínez et al. 2016; Sarhan et al. 2020).

Another interesting approach is to supply silicon (potassium silicate) via foliar application. Silicon acts directly on the fungal pathogen through the induction of enzymes involved in plant defense, producing phenolic compounds and phytoalexins at the site of infection (Dallagnol et al. 2012). This element also acts indirectly by forming a physical barrier resulting from the increased pH and osmotic potential after water evaporation (Dallagnol et al. 2015).

Raw milk is another option that provides calcium and phosphate salts, iron, magnesium, proteins, vitamins, amino acids, and microorganisms. These components increase after milk application and remain stable on the leaves, providing protection against powdery mildew (Hamann and Krömker 1997). Raw milk has a satisfactory cost–benefit ratio, accessibility by producers, and significant environmental acceptance (Medeiros et al. 2012). Applying raw milk has been shown effective against powdery mildew in several cucurbit crops: cucumber, pumpkin, and melon (Faria et al. 2011).

Plant defense can also be induced by some commercial fertilizers, including Agro Mos®, Copper Crop®, Soil Set®, and Fertisilício®, whose composition includes nutrients and enzyme-activating amino acids. Reduction in severity of powdery mildew is associated with increase in the activity of chitinase (QUI), β-1,3-glucanase (GLU), peroxidase (PO), and polyphenol oxidase (PPO), among other enzymes related to pathogenicity (Soliman and El-Mohamedy 2017). These compounds act by inhibiting conidial germination and mycelial growth, resulting in a smaller leaf area affected by the pathogen (Sarhan et al. 2020).

This study aimed to determine the efficacy of alternative products to conventional fungicides for managing powdery mildew caused by P. xanthii in melon.

Material and methods

Location of the study

The study was conducted in a plant nursery at the Universidade Federal Rural do Semi-Árido (UFERSA) in Mossoró, RN, Brazil. According to the Köppen climate classification, the region has a BSwh’ climate, i.e., hot, dry, and typically semi-arid. The mean annual temperature is 27 °C, with irregular annual rainfall (annual mean of 673.9 mm) and a mean air relative humidity of 68.9% (Alvares et al. 2013).

Collection, storage, and inoculation of P. xanthii

The P. xanthii isolate was collected from leaves of the melon hybrid Goldex at the Agrícola Salutaris Farm in the municipality of Afonso Bezerra, RN. Since the species is an obligate parasite, the material was sent to the plant nursery in paper bags to preserve the isolate under in vivo conditions.

Seedlings of the melon hybrid Goldex were provided by the TopPlant® company and transplanted 10 days after sowing (DAS) to pots containing 14.3 dm3 of soil from a producing area in Baraúna, RN. Inoculation was performed 18 DAS when the plants had three true leaves. The plants were inoculated artificially by depositing a small amount of conidia two spots (at each side of and equidistant from, the main leaf vein) on the second true leaf of each plant (Yuste-Lisbona et al. 2010).

A drip irrigation system was used to provide high water-use efficiency (Batista et al. 2009), and macro- and micronutrients were supplied according to the nutrient requirements of the crop (Cury et al. 2014).

Experimental design

A randomized block design (RBD) was used with eight treatments and eight replications, with one plant per replicate. Treatments were as follows: 1 = control, 2 = Amistar Top®, 3 = Suppress-L™, 4 = Agro Mos®, 5 = Copper Crop®, 6 = Soil Set®, 7 = Fertisilício®, and 8 = raw milk.

Preparation and application of treatments

Suppress-L™ was used as a biocontrol. This product was multiplied in rice grains at the concentration of 109 conidia mL−1 and applied at 5 kg/ha, as recommended by the manufacturer. Amistar Top® (300 mL/ha), Agro Mos® (900 mL/ha), Copper Crop® (300 mL/ha), Soil Set® (300 mL/ha), and Fertisilício® (600 mL/ha) were diluted in water and administered as recommended by the manufacturers. Amistar Top® (azoxystrobin and difenoconazole) was chosen because it is a fungicide commonly used by farmers in the region for this disease, and is registered in the Ministry of Agriculture, Livestock and Food Supply for this crop and pathogen. Raw milk was diluted in water and applied at a concentration of 20% (Zatarim et al. 2005). Whereas Suppress-L™ and Fertisilício® were applied every 5 days, the remaining treatments were applied weekly, as recommended by the manufacturers. The applications began 1 day after inoculation with powdery mildew. All applications were performed using a sprayer commercial hand held, and the volume applied was increased according to the crop development stage to achieve full shoot coverage.

Epidemiological components and plant evaluation

Plant evaluation was performed weekly, beginning after the first signs of P. xanthii (6 days after inoculation). The following epidemiological components were evaluated: incubation period (IP), disease incidence (INC), disease severity (SEV), and area under the disease progress curve (AUDPC).

The IP was defined as the number of days between inoculation and the appearance of the first disease symptoms. The INC was estimated as the percentage of symptomatic plants in each treatment. The SEV was determined using the scale proposed by He et al. (2017) with adaptations, where 0 = no visible symptom; 1 = up to 5% of the leaves covered with powdery mildew; 3 = 6–10% of the leaves covered with powdery mildew; 5 = 11–20% of the leaves covered with powdery mildew; 7 = 21–40% of the leaves covered with powdery mildew; 9 = 41–60% of the leaves covered with powdery mildew; 11 = 61–80% of the leaves covered with powdery mildew; and 13 = 81–100% of the leaves covered with powdery mildew. The SEV data was used to calculate AUDPC using the equation AUDPC = Σ [((y1 + y2)/2)*(t2 -t1))], where y1 and y2 are two consecutive evaluations performed at times t1 and t2, respectively.

Melon growth analysis

The variables of plant height (PH), stem diameter (SD), and number of leaves (NL) were determined weekly. At the end of the experiment, leaf area (LA) was obtained by the leaf disk method (Lins et al. 2018).

Post-harvest analyses

The post-harvest variables examined were pulp firmness (PF), fruit weight (FW), longitudinal diameter (LD), transverse diameter (TD), and °Brix. Fruit color was used as the defining parameter for harvest (Câmara et al. 2007).

Biochemical analyses

Fully expanded melon leaves were collected from all plants of each treatment 12 h after the last application of products. These samples were stored in coolers containing ice and transported to the Laboratory of Plant Biotechnology at UFERSA, where they were weighed, processed, and stored in a freezer (− 43 °C) for later analysis.

The leaf samples (0.5 g) were mechanically homogenized in 4 mL of 100 mM sodium acetate buffer (pH 5.0) using a mortar and pestle. The homogenate was kept for 25 min at 4 °C in a centrifuge at 20,000 g, and the resulting supernatant was considered as an enzyme extract to determine the activity of chitinase (QUI) and β-1,3-glucanase (GLU) (Bezerra Neto and Barreto 2011).

Chitinase activity was evaluated by the release of soluble “CM-chitin-RBV” fragments from carboxymethyl chitin marked with Remazol Brilliant Violet (CM-chitin-RBV 4 mg mL−1, Loewe Biochemica GmbH). For that purpose, 200µL of the protein extract was mixed with 600 µL of the same extraction buffer and 200 µL of “CM-chitin-RBV” (2.0 mg mL−1). After incubation for 20 min at 40 °C, the reaction was stopped by adding 200 µL of 1 M HCl, followed by ice cooling and centrifugation at 10,000 g for 5 min. The absorbance of the supernatant was determined at 550 nm with extraction buffer in the reference cuvette. The results were expressed as enzyme unit mg−1 of protein after discounting the absorbance values of the control (800 µL of extraction buffer + 200 µL of “CM-chitin-RBV”) (Stangarlin et al. 2000).

The spectrophotometric determination of the activity of β-1,3-glucanase in the extracts was performed by transferring 25 µL of the enzyme extract to each test tube. Subsequently, 200 µL of the acetate buffer solution (0.1 M, pH 5.0) and 200 µL of laminarin were added to the tubes (15 mg mL−1). The mixture was incubated for 30 min at 37 °C, after which the glucose content produced by the hydrolysis of laminarin was determined (Bezerra Neto and Barreto 2011). Then, the reducing sugars formed after incubation were quantified and the absorbances were read at 410 nm in a spectrophotometer. Finally, the protein concentration was analyzed by the Bradford method (Bradford 1976), and the results were expressed as enzyme unit min−1 mL−1.

The polyphenol oxidase extract was prepared by homogenizing the frozen leaves at the maximum temperature of 4 °C in 10 mL of 0.05 M phosphate buffer (pH 7.0) containing 1 mg of polyvinylpyrrolidone-10. The homogenized material was then centrifuged at 4000 g for 20 min, and the precipitate was discarded. The supernatant was preserved in ice and used in the analyses of this enzyme (Campos et al. 2004). Polyphenol oxidase activity was determined according to the technique described by Hyodo and Yang (1971) with the following modifications: a cold tube received 3.6 mL of 0.05 M phosphate buffer at pH 6.0, 1 mL of enzyme extract, and 0.1 mL of 0.1 M catechol. Next, the tube with the mixture was agitated in a vortex for 15 s, then incubated at 30 °C for 30 min, and transferred to an ice bath. Then, the mixture received 0.2 mL of 1.4% perchloric acid and, after agitation in a vortex, the tube rested for 10 min. Absorbance was read in a spectrophotometer at 395 nm. In the control, the enzyme extract was replaced by water. Enzyme activity was expressed as enzyme unit (UE) min−1 of sample. One enzyme unit was defined as the amount of enzyme that caused an increase of 0.001 absorbance units per minute (Campos et al. 2004).

In the procedure to prepare the peroxidase extract, the leaves were weighed (1 g of sample) and ground in a mortar with 50 mg of PVP and approximately 10 mL of liquid nitrogen. During this process, 3 mL of the sodium acetate buffer solution (50 mM, pH 5.0) and 1 mM of EDTA were added. Subsequently, the homogenized material was centrifuged at 10,000 mg for 10 min at 4 °C (Bezerra Neto and Barreto 2011). Then, the supernatant was transferred to Eppendorf tubes for analysis.

Peroxidase activity was determined according to the techniques described by Bezerra Neto and Barreto (2011). First, a cold tube received 1 mL of phosphate buffer (0.2 M) at pH 6.0 and 100 µL of the enzyme extract. Then, the tubes were placed in a thermostatic bath at 25 °C until the temperature stabilized. Subsequently, 100 µL of guaiacol (0.5%) and 100 µL of hydrogen peroxide (0.08%) were added to the tubes and the mixture was agitated in a vortex, after which seven spectrophotometric readings at 470 nm were made every 30 s. Enzyme activity was estimated based on the difference of absorbance per minute and weight of fresh sample. In the control, the enzyme extract was replaced by distilled water. All analyses were performed in duplicate.

Statistical analyses

The experiment was conducted twice. A preliminary ANOVA was used to determine whether there were significant differences between the two experiments and whether the data could be combined. Upon meeting the assumptions, the analysis of variance was performed, and the Scott-Knott test was applied at 5% probability. The statistical software R was used in all statistical analyses (R Core Team 20l7) (Wei et al. 2017).

Results

According to the F test, the evaluated treatments significantly affected the IP, SEV, and AUDPC variables (p > 0.001) (Table 1).

Table 1 F test for the incubation period (PI), severity (SEV), and area under the disease progress curve (AUDPC) of melon plants after 57-day inoculation with powdery mildew
Full size table

The incubation period was longer in the plants that received Amistar Top® (mean of 16 days), delaying the onset of symptoms by 9.25 days relative to the control treatment (6.75 days) (Fig. 1A). Since all plants showed symptoms, there were no differences among treatments for the INC according to the Scott-Knott test at 5% probability (100%) (Fig. 1B).

Fig. 1
figure 1

Incubation period (A), incidence (B), severity (C), and area under the disease progress curve (AUDPC) (D) in melon plants after 57-day inoculation with powdery mildew. The error bars represent the standard deviation of the means. Means with different letters differ by the Scott-Knott test at 5% probability

Full size image

The lowest SEV was observed in the plants treated with Amistar Top® (4.5), although this fungicide did not differ significantly from raw milk (6.0). These two treatments resulted in reductions in SEV of 62 and 49%, respectively, relative to the control (11.75). Copper Crop® (8.25) and Agro Mos® (9.25) reduced the SEV by 30 and 21%, respectively, relative to the control. These treatments also resulted in lower SEV values than Soil Set® (12.00), Fertisilício® (12.50), and Suppress-L™ (13.00) which all were not significantly different from the control (Fig. 1C).

Most treatments reduced the AUDPC, except Suppress-L™ (Fig. 1D). Amistar Top® was the most effective treatment and resulted in an AUDPC of 756. This fungicide promoted a reduction of 81.60% compared to the control (3,115). Raw milk, Copper Crop®, and Agro Mos® showed AUDPC values of 1561, 2009, and 2072 and reductions of 59, 36, and 33%, respectively (Fig. 1D).

With regard to the plant growth variables (PH, SD, and LA), only LA was significantly affected by the treatments (P < 0.05) (Table 2). Among fruit quality parameters, only the °Brix showed a significant difference between treatments (P < 0.05).

Table 2 F test for plant height (PH), stem diameter (SD), leaf area (LA), and °Brix in melon plants after 57-day inoculation with powdery mildew
Full size table

According to the Scott-Knott test, raw milk and Amistar Top® differed from the other treatments at 5% probability, providing the best LA results, with mean values of 1865.50 and 1606.01 cm2, corresponding to increases of 48.43 and 27.78% relative to the control (Fig. 2A). The mean soluble solids in the fruits were higher than 9°Brix in all treatments, and raw milk stood out with a mean value of 11.8°Brix (Fig. 2B).

Fig. 2
figure 2

Melon leaf area (A) and fruit °Brix (B) after inoculation with powdery mildew. The error bars represent the standard deviation of the means. Means with different letters differ by the Scott-Knott test at 5% probability

Full size image

Regarding the activity of defense enzymes (PPO, POX, QUI, and GLU), there was a significant difference between treatments (P < 0.05) (Table 3).

Table 3 F test for the activity of polyphenol oxidase (PPO), peroxidase (POX), chitinase (QUI), and β-1,3-glucanase in melon plants 57 days after inoculation with P. xanthii
Full size table

All treatments increased polyphenol oxidase (PPO) activity relative to the control (Fig. 3), differing by the Scott-Knott test at 1% probability. Moreover, raw milk (8.61 E.U.−1 min−1 mg of sample) resulted in higher enzyme activity compared to the other treatments, showing a 46% increase compared to the control (Fig. 3A).

Fig. 3
figure 3

Activity of polyphenol oxidase (PPO) (A), peroxidase (POX) (B), chitinase (QUI) (C), and β-1,3-glucanase (GLU) (D) after the application of treatments and powdery mildew inoculation in melon plants. The error bars represent the standard deviation of the means. Means with different letters differ by the Scott-Knott test at 5% probability

Full size image

With regard to peroxidase (POX) activity, most treatments resulted in higher enzyme activity compared to the control, except Suppress-L™ (Fig. 3B). The highest POX activity occurred with raw milk (8.31 E.U.−1 min−1 mg of sample), followed by Copper Crop® (7.25 E.U.−1 min−1 mg of sample), Agro Mos® (4.92 E.U.−1 min−1 mg of sample), and Fertisilício® (4.54 E.U.−1 min−1 mg of sample), increasing by 4055, 3525, 2360, and 2170%, respectively (Fig. 3B).

The treatments with Fertisilício® (1.27 E.U.−1 mg of protein), Copper Crop® (1.39 E.U.−1 mg of protein), and raw milk (1.03 E.U.−1 mg of protein) increased chitinase activity (QUI) by 160, 107, and 54%, respectively, relative to the control (0.67 E.U.−1 mg of protein) (Fig. 3C).

The maximum activity of β-1,3-glucanase (GLU) was obtained with raw milk (5.14 E.U.−1 min−1 mL). However, Suppress-L™ (4.21 E.U.−1 min−1 mL), Copper Crop® (4.00 E.U.−1 min−1 mL), and Fertisilício® (3.95 E.U.−1 min−1 mL) were also effective in promoting the activity of β-1,3-glucanase in relation to the control. These treatments increased enzyme activity by 46, 19, 13, and 12%, respectively (Fig. 3D).

Discussion

The present study highlights the potential of raw milk as an alternative to fungicides to control powdery mildew, the main melon leaf disease. Raw milk induced plant defense mechanisms by promoting higher activities in three of the four enzymes studied (PPO, POX, and GLU). In addition to raw milk, the results documented that the fertilizer Copper Crop® was another alternative product that provided high protection.

Raw milk reduced disease severity and AUDPC through various action mechanisms, with its constituents acting directly against the fungus, especially phosphate and calcium, which degrade the cell wall of hyphae. Additionally, its constituents provide a slightly basic pH and cause disturbances in fungal development at the onset of the disease, in addition to stimulating the growth of antagonistic microorganisms on the leaf surface (Kamel et al. 2017). These results agree with Kamel and Afifi (2020), who observed a reduction in the AUDPC (499.3) after milk application in cucumber. On the other hand, indirect action occurs through the expression of enzymes that strengthen the plant defense system due to the biosynthesis of phytoalexins and phenols to form lignin, which acts as a barrier against pathogen penetration (Bettiol 1999; Zatarim et al. 2005; Sudisha et al. 2011; Li et al. 2015). Moreover, Bettiol and Astiarraga (1998) and Medeiros et al. (2012) found evidence of suppression of fungal mycelia, bacteria, and yeast strains on the leaf surface after milk application, which also explains the efficacy of this material.

Control of powdery mildew by applying raw milk was also documented by Kamel and Afifi (2020) in an experiment conducted with cucumber plants grown under greenhouse conditions. In a study with pumpkin cultivars, Barickman et al. (2017) observed lower powdery mildew severity after copper application. However, the same study stressed that plants treated with azoxystrobin showed a significant reduction in severity (13%) compared to copper, corroborating the results of the present study.

Due to the increase in plant resistance and the limited potential for environmental contamination, milk is widely used (Sudisha et al. 2011). Moreover, in agricultural crops such as common bean (Vigna unguiculata L. Walp), resistance usually depends on plant age and reaches its maximum at the beginning of the crop cycle (Ebrahim et al. 2011). However, under the conditions of the present study, protective values were also observed at the end of the melon crop cycle. Substances like raw milk cannot be legally used by farmers for disease control in some countries, including the USA, until reviewed and registered as a fungicide. However, milk is widely used for disease control in several countries, e.g., China, Brazil, and India (Sudisha et al. 2011). In Brazil, it is possible to use this tool, mainly in organic production.

Copper Crop® was effective due to the action of the cupric ion in the cell membrane, resulting in the denaturation of structural proteins and promoting enzyme reactions that block the respiratory activity, consequently inhibiting spore germination (Torre et al. 2018). Moreover, heavy metals such as copper can induce, by stress, enzymes related to plant defense (Chmielowska et al. 2010). Copper Crop® is a suitable alternative to a copper-based fungicide because it has complexed amino acids that facilitate absorption by the plant, and is an important alternative, as it increases the fresh and dry mass of the shoot, contributing to nutrition and, consequently, increasing plant vigor and disease resistance. In Brazil, this copper-based product is already being used by farmers in the agricultural market.

Amistar Top®, a conventional fungicide, was the only treatment that delayed the onset of symptoms. However, disease incidence was not affected by this fungicide or any other treatment due to the rapid spread of this fungus. The fungicide delays the onset of the disease due to its protective and curative properties, inhibiting the action of enzymes involved in fungal mitochondrial respiration and paralyzing sterol biosynthesis in the membranes through the activity of C-14 demethylase (Krämar and Schirmer 2007; Souders et al. 2019). Efficacy of azoxystrobin, one of the two active ingredients in Amistar Top®, was also observed by Barickman et al. (2017) in pumpkin cultivars affected by powdery mildew.

Therefore, the present study highlighted several beneficial effects, with raw milk resulting in increased PPO and GLU activity due to the high concentration of amino acids, which promote GLU activity, the presence of tyrosine (a PPO constituent), and glycine activity (Martins Filho 1987; Sudisha et al. 2011; Teixeira et al. 2017). In addition to raw milk, Fertisilício® also favored the phenolic compounds, increasing the content of protective enzymes such as PPO (Hasan et al. 2020).

The higher POX activity obtained with raw milk is attributed to the composition of this product since lactoperoxidase, an enzyme of the peroxidase family, is naturally found in the mammary glands of bovines. Moreover, the phosphate ion found in milk and applied to the leaves also significantly increases POX activity (Kussendrager and Van Hooijdonk 2000; Walters et al. 2005). High values of this enzyme are also important to ensure the balance of reactive oxygen species (Berger et al. 2016). From this perspective, another study reported a similar finding with millet after infection with downy mildew (Sclerospora graminicola) (Sudisha et al. 2011), with amino acids and raw milk promoting a fivefold increase in POX activity. The eliciting capacity of amino acids and raw milk also effectively reduced the severity of powdery mildew in Vitis vinifera L. and Cucurbita pepo L. (Bettiol 1999; Crisp et al. 2006).

Although the literature relates the application of silicon (Si) to the increase in enzyme activity only when applied to the roots (Dallagnol et al. 2015), the present study proved that foliar application also effectively increased QUI activity. The levels of this enzyme are usually low in plants, as observed in the control treatment. However, the application of inducers such as Fertisilício® favors enzyme activity through signal transduction, a biochemical process that induces plant defense and increases the activity of PR proteins such as QUI. QUI activity was also increased after the foliar application of Si under stress caused by Sporisorium scitamineum (Deng et al. 2020).

Larger leaf area on plants treated with raw milk likely is due mainly to phosphate and other salts (Ca, Fe, and Mg), components that induce cell division and growth (Epstein and Bloom 2006), and control of powdery mildew may have also contributed. Amistar Top® also increased leaf area, probably due to control of powdery mildew and perhaps also growth enhancement which has been documented for this and other strobilurin fungicides in some plants. Furthermore, raw milk positively affected fruit °Brix. As a potassium-rich material, raw milk plays a key role in solute transport through the phloem, favoring sugar accumulation in fruits (Rangel et al. 2018).

Raw milk was the best alternative method to control powdery mildew in melon, showing equivalent efficacy to chemical control (Amistar Top®), reducing disease severity by 49% compared to the control, and increasing the °Brix of melon by 19.5%. Copper Crop® also controlled powdery mildew in melon, reducing severity by 30%. Suppress-L™ and Fertisilício® were not effective for managing powdery mildew under the environmental conditions of this study. The present study highlighted that the activities of polyphenol oxidase, peroxidase, chitinase, and β-1,3-glucanase are involved with the plant defense mechanism of melon plants since the treatments with the highest enzyme activity values had the lowest powdery mildew severity. Among the treatments tested, raw milk promoted the highest activities of the studied enzymes.