Self-digestive solution of Lysobacter enzymogenes LE16 as a biofungicide to control plant powdery mildew

ABSTRACT The self-digestive solution (SDS) of the biocontrol bacterium Lysobacter enzymogenes LE16 shows strong antagonistic activities against multiple soil-borne phytopathogens but the positive evidence of this bacterium against plant foliar disease is still scanty. Thus, laboratory, greenhouse, and field experiments were carried out to estimate the efficacies of SDS, S-SDS (stored at room temperature for 12 months), and H-SDS (heated at 100°C for 30 min) against plant powdery mildew. This bacterium produced hydrolases (phosphatase, protease, lysozyme, chitinase, and β-1,3-glucanase) that degrade pathogen cell components and siderophores that compete for iron with phytopathogens. The top five antimicrobial metabolites identified in SDS were pyroglutamic acid, deoxycytidine, pyrrole-2-carboxylic acid, 13-oxo-9,11-tridecadienoic acid, and 3’-amino-3’-deoxythimidine. Among them, pyroglutamic acid may play a vital role in powdery mildew control. As a result, SDS, S-SDS, and H-SDS strongly inhibited the conidial germination of Erysiphe cichoracearum and Sphaerotheca fuliginea. The application of SDS significantly increased the activity of antioxidant enzymes in crop leaves and effectively controlled tobacco and cucumber powdery mildew in the greenhouse and the field. Therefore, L. enzymogenes LE16 can effectively control powdery mildew. The underlying mechanisms may be attributed to the induction of plant systemic resistance and the production of antifungal substances.


Introduction
Powdery mildew, a common foliar disease of plants worldwide, causes huge yield (14%-50%, depending on plant species) and economic losses in agriculture and horticulture each year (Singh 1999;Fondevilla and Rubiales 2012;Jin et al. 2013;Green et al. 2014). The conidia of powdery mildew repeatedly infect plants in environments with humidity 70%-90% and temperature 20°C-28°C, especially in the greenhouse (Carroll and Wilcox 2003;Guzman-Plazola et al. 2003;Keinath and Dubose 2012). Effective control of this foliar disease is important for crop cultivation in fields and greenhouses. In an agricultural or horticultural setting, resistant cultivars, non-chemical control, and chemical fungicide are often used to control crop powdery mildew (Moyer and Peres 2008;Gilardi et al. 2012). The cultivation of resistant cultivars is a fundamental strategy to control the disease. However, the long breeding cycles and variation of physiological races result in the loss of disease resistance, and many of them on the market provide only partial controls (Jiao et al. 2020). The efficacy of non-chemical products, such as milk, bicarbonates, soluble silicates, and oils, has provisions of the Budapest Treaty. L. enzymogenes LE16 grew on nutrient agar (NA) plates at 28°C for three days in the dark before collecting bacterial cells from the plates in sterile distilled water and diluting to 1 × 10 3 cells mL -1 as an inoculant.
For the detection of lytic enzymes and siderophores produced by L. enzymogenes LE16, 10 µL of the L. enzymogenes LE16 inoculant was spot-inoculated in the center of modified Pikovskaya's agar medium (Pikovskaya 1948), skimmed milk medium (Odhiambo et al. 2017), lysozyme test medium (Sugahara et al. 1982), chitinase test medium (Zhang and Yuen 2000), glucanase test medium (Palumbo et al. 2005), and Chrome Azurol S (CAS) medium (Schwyn and Jb 1987) for the detection of phosphatase, protease, lysozyme, chitinase, β-1,3-glucanase, and siderophores, respectively. Plates were then incubated at 28°C for three to six days depending upon the hydrolases, and the clear or colorful halos produced around bacterial colonies were considered positive.

Preparation of SDS from L. enzymogenes LE16
Aliquots (50 mL) of nutrient solution (agar-free NA, NB) were transferred into 100-mL Erlenmeyer flasks. After steam-sterilized, each flask was inoculated with 1 mL inoculant. Flasks were then incubated at 28°C with constant shaking at 150 rpm. L. enzymogenes LE16 cell density increased with time and reached a maximum at approximately two or three days after inoculation (Supplementary Figure S1A). Then, autolysis occurred, and cell density gradually decreased. The microscope examination also demonstrated no solid debris remaining in the broth culture after five bacterial incubation days at 28°C (Supplementary Figure S1B). Then, the SDS was prepared. Hereafter, SDS was subjected to storage at room temperature (18°C-32°C) for 12 months (S-SDS) and heat at 100°C for 30 min (H-SDS), respectively, and then used in later experiments.

Detection of soluble metabolites in SDS
SDS was passed through a 0.22 μM film, dried at −20°C into powder in a vacuum, and dissolved in an aqueous solution containing 5% methanol for the detection of metabolites by liquid chromatographmass spectrometry (LC-MS; Thermo Electron, San Jose, CA, USA) in the Morjorbio Company, Shanghai, China. The liquid chromatography was performed using a C 18 analytical column ACQUITY UPLC HSS T3 (100 mm × 2.1 mm; Waters, Milford, USA) with 95% water + 4.9% acetonitrile + 0.1% formic acid and 47.5% acetonitrile + 47.4% isopropanol + 5% water + 0.1% formic acid as mobile phase, respectively, at a speed of 0.4 mL min −1 . The column temperature was 40°C and 20 µL of samples were injected. For the mass spectrum (MS) analysis, an electron impact (EI) ionization source was used with an ion spray voltage of 5,500 V, a capillary temperature of 550°C, and a scan range of 35-500 m/z at a scan-interval of 0.30 s in the scan acquisition mode. N 2 was used as the nebulizer, heater, curtain, and collision gas. Reifycs Abf (Analysis Base File) Converter, MS-DIAL (Tsugawa et al. 2015), Shimadzu offline software, and the Fiehn library (Kind et al. 2009) were used for peak data processing (including raw extraction, identification, area integration to remove metabolites less than 3-fold blank, and so on), baseline filtering and calibration, and deconvolution analysis. Parameters were set as follows: 0.5 sigma window value, 5000 EI spectra cut-off, 0.5 min retention time tolerance, 0.5 Da m/z tolerance, 70% EI similarity cut-off, and 70% identification score cut-off. The alignment settings were: 0.075 minutes for retention time tolerance and 0.5 for retention time factor (Sadiq et al. 2020). The matched substances were retained based on a total similarity filter ≥ 800, fill ≥ 0.5%, fragment presence ≥ 800, and unmatched metabolites were removed. The metabolome analysis was performed using four replicate fermentation broths.

Conidial germination test
Erysiphe cichoracearum conidia were collected from naturally mildewed flue-cured tobacco leaves (hereafter, it is referred to as tobacco) and Sphaerotheca fuliginea from the cucumber. Source leaves were placed in a shaker and shaken 24 h before conidia harvest to dislodge the old and ensure high viability of inoculum. The fungi were identified based on conidial characteristics (Boesewinkel 1980). SDS, S-SDS, and H-SDS were dried at 40°C in a vacuum rotary evaporator. The solids obtained from the three solutions were individually weighted to PDA (pre-cooling to 45°C-50°C), making the final concentration of the crude extracts arrive at 0 (CK), 0.1%, 0.2%, and 0.4%, respectively. After solidification, the conidia with high viability from the shaken source leaves were plated onto the agar surfaces by sterile brushes and incubated at 25°C for 16 h with five replicate plates per treatment. An agar piece (1 cm × 1 cm) was randomly cut from each dish for the observation of conidial germination, and 100 conidia were checked by a microscope (Guzman-Plazola et al. 2003). Conidia germination rate (%) = (number of germinated conidia/total observed conidia) × 100%.

Greenhouse experiment
The experiment was carried out in the greenhouse of the College of Resources and Environment of Southwest University, Chongqing, China. A typical Udorthent (loam texture, pH 6.5) in pots (2.5 kg soil per pot) was fertilized with adequate N, P, and K (0.2 g (NH 2 ) 2 CO and 0.2 g KH 2 PO 4 kg −1 soil). Seedlings were grown from seeds sown in pots (one seedling per pot) for 45 days (tobacco) and 20 days (cucumber), arranged at random on the benches in the greenhouse under natural temperature (20°C-28°C) and humidity (78%-92%) during the growing season (from late April to early May). Seedlings were watered as needed every two or three days during the growing periods.
The conidia collected in the section 'Conidial germination test' were individually added to autoclaved aqueous solutions of 0.1% water agar and 0.0025% Tween-20 (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China). Conidia in the suspension can be held for up to 4 h with no apparent effect on viability. The conidial inoculants (3 × 10 7 conidia mL −1 ) were sprayed onto the whole plant until runoff on each leaf. The inoculated plants were then grown in the greenhouse until powdery mildew colonies became visible on the leaves. Spray treatments, including deionized water (CK), SDS, S-SDS, H-SDS, and 1000-time-diluted triadimefon solution (CH), were applied on whole plants, like the conidial inoculation. A drop of Tween 20 surfactant was added to 500 mL of all spray solutions before being used. The experiment was performed twice using a complete randomized design with four replicates for each treatment (15 pots per replicate). The colonies on each plant were counted before and on the 1st, 3rd, 5th, and 7th days after spray treatment. The colony counts of each plant were summed for data analysis. Disease severity of flue-cured tobacco was rated on a 0 to 4 scale, where 0 = no infection, 0.5 = less than 1%, 1 = 1% to 5%, 2 = 6% to 20%, 3 = 21% to 35%, 4 = more than 35% leaf area infected with powdery mildew (tobacco pest index and survey standard, 2008); cucumber was rated on a 0 to 5 scale, where 0 = no infection, 1 = less than 1%, 2 = 1% to 5%, 3 = 6% to 20%, 4 = 21% to 40%, 5 = more than 40% leaf area infected with powdery mildew (Yan et al. 2006). The experiment was stopped when the disease index exceeded 5 in the control treatment (water + PI). Disease index = sum of individual scores/total plants (or leaves) observed × maximum score × 100 (Wheeler 1969). And disease control efficacy (%) = ([disease index of controldisease index of treatment]/disease index of control) × 100% ) (the same calculations used in later experiments).
After the sprays of SDS and CH for seven days, the first two fully-expanded leaves were collected to determine the activities of antioxidant enzymes, levels of chlorophyll and malondialdehyde (MDA), and membrane permeability of leaf cells.
Leaf chlorophyll content was determined by the protocol described by Xu et al. (2018). Briefly, 100 mg of fresh leaves were homogenized in liquid nitrogen, and 5 mL of 95% ethanol was added. After being incubated in the dark for 1 h, the reaction mixture was centrifuged at 7,265 × g for 5 min. The supernatant was used to measure the absorbance spectrophotometrically at 663 and 645 nm against 95% ethanol as blank. The chlorophyll content (μg mL −1 ) = 20.2 (A 645 ) + 8.02 (A 663 ). MDA detection was performed using the method of Simaei et al. (2011). Five hundred mg of fresh leaves were homogenized in 10 mL of 0.1% (w/v) trichloroacetic acid (TCA). Furthermore, 1 mL of the homogenate was added to 4 mL of 0.5% (w/v) thiobarbituric acid + 20% (w/v) TCA, and the mixture was heated at 95°C for 30 min, chilled in an ice bath, and then centrifuged at 7,265 × g for 5 min. The absorbance of the supernatant was measured at 532 nm, and the value for non-specific absorption at 600 nm was subtracted. The concentration of MDA was calculated using the extinction coefficient of 155 mM −1 cm −1 and expressed as μmol g −1 . The measurement of leaf cell membrane permeability was described by Tariq et al. (2011). Then, 2 g fresh leaf with discs of 5 mm diameter was placed in 10 mL of distilled water for 24 h. The electric conductivity of solutions with and without leaf discs (EC 1 and EC 2 , respectively) was recorded by an electrical conductivity meter. The membrane permeability of leaf cell (or membrane damage) (%) = (EC 1 /EC 2 ) × 100%.

Field experiment
Mildewed tobacco (infected by E. cichoracearum) was selected for field investigation on powdery mildew control. There were three replicates per treatment, each with 50 plants per replicate. Foliar sprays of water (control), SDS, and CH were applied, respectively, to runoff onto the adaxial mildewed leaves in a tobacco field in Fenggang County, Zunyi city, Southwest China. The colonies on each plant were counted the same way described in the 'Greenhouse experiment'. Thereafter, the colony counts of each plant were summed for data analysis.

Data treatment
Results obtained from each experiment were evaluated by analysis of variance (ANOVA) using the SPSS 21.0 statistical software package (IBM, USA). Significant differences between means were tested by Fisher's protected least significant difference (LSD) (P < 0.05).

Inhibition of SDS against the germination of powdery mildew conidia in vitro
Compared with the CK, SDS, S-SDS, and H-SDS treatments significantly decreased the germination rates of E. cichoracearum (infected flue-cured tobacco) and S. fuliginea (infected cucumber) on the agar mediums (Table 2). For example, the germination rates of E. cichoracearum were decreased by 91.5%−99.6%, 91.0%−99.6%, and 90.3%−99.4% in the SDS, S-SDS, and H-SDS treatments compared with the blank control, respectively. The germination rate also decreased with increased SDS, S-SDS, and H-SDS concentrations. However, there was no significant difference in the germination rates among the three SDSs at equal concentrations in the agar medium.

Biocontrol of plant powdery mildew in the greenhouse
Foliar applications of SDS, S-SDS, H-SDS, and CH led to the disappearance of powdery mildew colonies on the leaves ( Figure 2) and a significant decrease in disease indexes (Table 3). The control efficacies of SDS, S-SDS, and H-SDS ranged from 70% to 85% (tobacco) and from 72% to 81% (cucumber), which were higher than the CH treatment (ranged from 65% to 68%). However, there was no significant difference between S-SDS and CH in tobacco and between H-SDS and CH in cucumber.

Influence of SDS on selected biochemical indicators of plants
Compared with the control (water spray on mildewed plants), foliar spray of SDS, S-SDS, H-SDS, and CH on mildewed plants significantly increased the activities of antioxidant enzymes (POD and SOD) in the leaves except for CAT, and CAT maintained almost unchanged in both tobacco and cucumber leaves (Figures 3a-3c). Compared with the control, for instance, the foliar spray treatments increased the activity of SOD by 23.2%-33.4% and POD by 14.6%-30.1% in mildewed cucumber leaves.  Similarly, the foliar spray treatments also decreased leaf MDA level and membrane permeability of tobacco and cucumber seedlings with conidial inoculation (Figures 3d and 3f). In contrast, the foliar sprays of SDS, S-SDS, H-SDS, and CH increased chlorophyll contents by 30.4%-40.2% for tobacco and 21.9%-38.5% for cucumber (Figure 3e).

Control of tobacco powdery mildew in the field
The foliar spray of SDS showed 79%-84% of the control efficacies against tobacco powdery mildew, which were significantly higher than CH (66%-70%) during the observation period (1-7 days after the foliar sprays, Figure 4). In addition, the control efficacies of both SDS and CH changed little during this period.

Discussion
The underlying mechanisms employed by biocontrol bacteria against plant diseases have been attributed to the production of hydrolases, the release of anti-pathogen siderophores, the synthesis of antimicrobial substances, and the induction of plant systemic resistance (Bargabus et al. 2004;Liu et al. 2013;Parnell et al. 2016;Miccoli et al. 2020). Autolytic cell destruction is initiated by releasing digestive enzymes from lysosomes into the cytoplasm and outside the medium. L. enzymogenes LE16 can produce extracellular phosphatase, protease, lysozyme, chitinase, and β-1,3-glucanase, which may contribute to its own and pathogen autolysis through the degradation of cell walls, membranes, and organelles, leaving no solid debris remaining in the SDS and contributing to the efficient control of powdery mildew. L. enzymogenes LE16 generated siderophores, which was consistent with some other biocontrol bacteria under iron-limiting conditions (Ghosh et al. 2015). Siderophores, a large group of ironchelating compounds with a molecular weight of 200-2000 Da, are beneficial to the producers to absorb iron and reduce available iron in the medium, resulting in iron deficiency in pathogens. If the L. enzymogenes LE16 contact with powdery mildew pathogens, the siderophores produced may make the pathogens deficient in iron and inhibit their growth and reproduction.
LC-MS analysis revealed that SDS was rich in antimicrobial metabolites, consistent with other L. enzymogenes strains (Tang et al. 2019). Among the top 5 antifungal substances, pyroglutamic acid has antifungal activities against Phytophthora infestans and Candida spp. (Gang et al. 2018;Wod et al. 2021). Deoxycytidine and 3'-amino-3'-deoxythimidine can greatly inhibit protein biosynthesis in fungal pathogen cells (Garavito et al. 2015). Pyrrole-2-carboxylic acid can slow the cell division and growth of the plant pathogen Aspergillus Niger (Gálvez-Iriqui et al. 2019). And 13-oxo-9,11-tridecadienoic acid damaged fungal cell membranes, leading to the leakage of cellular inclusions (Bito 2017). Thus, these antifungal substances varied in their targets in or on fungal pathogen cells. The presence of multiple antifungal substances in SDS may be favorable for the effective control of powdery mildew. Importantly, pyroglutamic acid is thermally stable, and the melting point exceeds 180°C, while the hydrolases and other antimicrobial substances in the SDS may be denatured by heat treatment. We noticed that the other mutant strain of L. enzymogenes (HYP18) in our lab lost the ability to synthesize pyroglutamic acid. At the same time, the control efficacy of its fermentation liquid against plant powdery mildew significantly decreased (Peng et al. unpublished data). The results suggest that pyroglutamic acid in H-SDS contributes greatly against the conidial germination of powdery mildew. Foliar spray of SDS significantly increased the activity of antioxidant enzymes SOD and POD in the leaves of mildewed plants. These results were consistent with the reports of Ardebili et al. (2011), Ji et al. (2020, and Rais et al. (2017), who found the biocontrol bacteria Pseudomonas fluorescens CHA0, Bacillus licheniformis W10, and Bacillus spp, respectively. KFP-5, KFP-7, and KFP-17 increased the activity of antioxidant enzymes in diseased plants. Pathogen infection induces the excessive production of reactive oxygen species (ROS) in plants, including superoxide radicals (O 2 • ), hydroxyl radicals ( • OH), and hydrogen peroxides (H 2 O 2 ) (Wu et al. 2011). The production of ROS in live cells damages cellular components, such as proteins, membrane lipids, and nucleic acids, which is one of the mechanisms of pathogens to injure host cells (Vellosillo et al. 2010;Herrera-Tellez et al. 2019;Lilai et al. 2022). Therefore, the application of SDS, S-SDS, and H-SDS can increase the activity of antioxidant enzymes, which is beneficial to eliminate ROS accumulated in mildewed plants and alleviate the damage to plant cells. MDA, one of the products of membrane lipid peroxidation, and leaf cell membrane permeability can be greatly reduced while chlorophyll levels increase in the leaves of mildewed plants.
In addition to the production of antifungal substances, P-salicylic acid was also detected in SDS (Supplementary Table S1). Salicylic acid is an important hormone in plants. Although this phytohormone effectively regulates plant growth and development, the most well-known role is related to plant defense responses (Zhang and Liu 2001). The application of exogenous salicylic acid confers resistance to many monocotyledonous and dicotyledonous plants against various pathogens, such as fungi (Sillero et al. 2012), bacteria (Alemu et al. 2019) and viruses (Luo et al. 2011). The underlying mechanisms are attributed to the deposition of callose plugs for the reinforcements of plant cell walls, the elimination of H2O2 that damage live cells, and the production of disease resistancerelated proteins upon contact with pathogens (Kumar 2014). The presence of salicylic acid in SDS and the increase in the activity of antioxidant enzymes support the theory that biocontrol microbes suppress plant diseases by the induction of systemic acquired resistance (Li et al. 2015).
Temperature, humidity, and sunlight often wave greatly in the fields, making the biocontrol efficacies of live microbes against crop diseases, particularly powdery mildew, relatively low and unpredicted (Keinath and Dubose 2012). Unlike other microbial agents that contain live cells, the foliar spray of SDS contains no live bacterial cells. It can effectively suppress powdery mildew in both greenhouse and fields, with higher control efficacies than the chemical fungicide triadimefon that has long been used in the local area to control plant powdery mildew. The lower control efficacy can be explained by the induction of pathogen resistance to this chemical fungicide. Thus, L. enzymogenes LE16 provides a safe, sustainable, and effective alternative for the control of powdery mildew.

Conclusions
The SDS of L. enzymogenes LE16 can effectively control crop powdery mildew. Its underlying mechanisms may be attributed to the induction of plant resistance to pathogens and the production of extracellular hydrolases, siderophores, and antifungal metabolites. It is necessary to conduct further studies on more crops and identify active ingredients produced by this bacterium.

Disclosure statement
No potential conflict of interest was reported by the author(s).