Biological Control Properties of Two Strains of Priestia megaterium Isolated from Tar Spots in Maize Leaves

Abstract

Priestia megaterium is a maize endophyte that may help the plant defend itself against bacterial and fungal pathogens. This study aimed to identify antimicrobials produced by two P. megaterium endophytes (FS10 and FS11) from maize and determine if seed coating with either strain could increase resistance to pathogens. Volatiles emitted by both isolates reduced the hyphal growth of fungi by 17–76%. Gas chromatography analysis found that each strain emitted isovaleric acid (IVA) and 3-methyl-1-butanol (3MB). Volatiles produced by each isolate inhibited bacterial growth, especially Clavibacter michiganensis ssp. michiganensis (Cmm). IVA killed all Cmm cells at 208 µL L⁻¹, while 3MB inhibited Cmm growth by 51% at 208 µL L⁻¹. Diluted cell-free extracts from FS10 and FS11 cultures stopped growth of Cmm, Erwinia amylovora and Ustilago maydis but did not arrest growth of Fusarium verticillioides. The treatment of corn seeds with FS10 or FS11 reduced leaf damage by 38–84% in young plants caused by Bipolaris maydis, Colletotrichum graminicola (Ces.) G.W. Wilson 1914, Exserohilum turcicum and Pythium sylvaticum. FS10 and FS11 isolates exuded volatile and soluble compounds that were more effective in slowing growth of bacteria than fungi. It is likely that corn seed treatment with FS10 and FS11 triggers induced systemic resistance, which mitigates leaf damage caused by maize pathogens.

1. Introduction

Bacteria can be utilized in agriculture to help supplement nutrients or augment plant defense against insects and fungal pathogens. There are 63 Bacillus bacteria strains registered as biopesticide active ingredients by the United States Environmental Protection Agency [1]. Forty-three percent of these Bacillus bacteria are classified as Bacillus thuringiensis Berliner 1915, which produces bioinsecticides. The remaining Bacillus biopesticides are primarily strains of Bacillus subtilis (Ehrenberg 1835) Cohn 1872 or Bacillus amyloliquefaciens (ex Fukomoto 1943) Priest et al. 1987, which are typically used for the control of fungal pathogens. Bacillus organisms synthesize many different kinds of small peptides that can inhibit the growth of fungi, but the three main classes of antimicrobial peptides are the fengycins, iturins and surfactins [2]. There is evidence that Bacillus bacteria only colonize the rhizosphere of plants and produce surfactins exclusively [3], which likely helps prevent pathogenic soilborne fungi from colonizing the roots. There is other evidence that suggests Bacillus bacteria can live inside the plant tissues, which is called endophytism, and they can provide protection from invading fungi throughout the plant [4,5]. Endophytes can be beneficial, neutral or pathogenic to plants [6].

One of the most common bacteria found in sweet corn stems is Bacillus megaterium (de Bary 1884) [7]. Bacillus megaterium was renamed Priestia megaterium (de Bary 1884) Gupta et al. 2020 in 2020 [8]. Despite its prevalence in maize stems, any role that P. megaterium serves in maize has not been established [9]. Studies with a GFP-expressing P. megaterium found that this bacterium strain could enter the roots, move up the stems and colonize the leaves [10]. It is possible that P. megaterium synthesizes compounds in maize that combat pathogenic microbes. Antimicrobial compounds, such as megacine, produced by P. megaterium, have been studied since the 1950s [11]. Since that time, other antimicrobial compounds, such as bacimethrin, fungitoxin, cytidines, and oxetanocin produced by strains of P. megaterium have been characterized [12,13,14,15,16,17]. Bacimethrin is an antibacterial biochemical and also affects yeasts when supplied at levels above 100 µg mL⁻¹ [18]. Some P. megaterium strains can synthesize the well-known antifungal compounds fengycin, surfactin, iturin A or bacillomycin D [19,20,21].

Bacteria can produce volatile organic compounds (VOCs) that can serve as biopesticides [22]. VOCs are low-molecular-weight compounds (<300 Da) that can move through soil and the atmosphere for long distances. VOCs are released from bacteria via catabolic reactions that include glycolysis, lipolysis, proteolysis, and fermentation, as well as through fatty acid biosynthesis and sulfur metabolism [23]. The VOCs made by bacteria include hydrocarbons, aldehydes, ketones, esters, organic acids, alcohols, terpenes, and sulfur- or nitrogen-containing molecules [22]. VOCs produced by Bacillus megaterium KU143 inhibited the in vitro growth of Aspergillus flavus (Link 1809), Penicillium fellutanum (Biourge 1923) and Penicillium islandicum (Sopp 1912) [24,25]. Gas chromatography coupled with mass spectroscopy identified 12 volatile compounds produced by B. megaterium KU143, which included 5-methyl-2-phenyl-1H-indole [25]; derivatives and related compounds of this volatile contained both antibacterial and antifungal activity [26,27]. Another strain of P. megaterium produced the antifungal volatiles hexadecenoic acid methyl ester (palmitic acid) and tetracosane [28].

The addition of bacteria to maize roots or to seeds can cause the leaves of the plants to be more resistant to fungal or bacterial pathogens [29,30]. This phenomenon, termed induced systemic resistance (ISR), is well documented in many plants [31,32]. P. megaterium has demonstrated the ability to cause ISR in Arabidopsis thaliana (L.) Heynh. 1842, oilseed rape (Brassica napus (L. 1753)), rice (Oryza sativa (L. 1753)), cucumber (Cucumis sativus (L. 1753)), Camellia sinensis (Kuntze 1881) and Capsicum chinense (Jacq. 1777, Habanero-type pepper) [33,34,35,36,37,38,39]. Hallmarks of ISR include an accumulation of salicylic acid and elevated expression of pathogenesis-related genes [33], which ultimately result in a reduction in disease. The examples of P. megaterium triggering ISR in a variety of plants indicate that ISR will likely be a component of plant resistance in any plant treated with P. megaterium.

Our laboratory recently isolated putative biological control agents from maize leaf tar spots, the reproductive structures of Phyllachora maydis (Maubl. 1904) [40]. Two of these organisms were identified as P. megaterium strains. Additional experiments with the two P. megaterium strains found that each was able to reduce the incidence of tar spot disease in field maize when applied individually to seed prior to sowing [41]; the greater leaf resistance in the P. megaterium-treated plants could be due to ISR. The mechanism(s) of the increased disease resistance in maize via each P. megaterium strain has not been characterized. Since there is evidence that P. megaterium can be transported from the roots to the leaves and could be in close contact with the tar spot pathogen, we examined whether volatiles or secreted molecules produced by the two P. megaterium strains have antifungal activity. In addition, we wanted to determine if maize seed treatment with each P. megaterium strain could reduce levels of leaf disease triggered by Exserohilum turcicum (Pass.) K.J. Leonard & Suggs 2018 (causes northern corn leaf blight), Bipolaris maydis (Y. Nisik & Miyake) Shoemaker 1959 (causes southern corn leaf blight), Colletotrichum graminicola (Ces.) G.W. Wilson 1914 (causes anthracnose disease) and Pythium sylvaticum (W.A. Campb. & F.F. Hendrix 1967), an oomycete that can cause seedling disease. We hypothesized that treatment of corn seed with FS10 and FS11 would result in a lower incidence of leaf disease. We also hypothesized that P. megaterium produces volatile or water-soluble antifungal compounds. The results of these comparative studies, which are described below, can indicate which mechanisms are likely to be the most important to the different pathogens, and thereby provide guidance for the most effective field application strategies.

2. Materials and Methods

2.1. Organisms

Priestia megaterium strains 10 and 11 (abbreviated from hereafter as FS10 and FS11) were isolated from tar spot stromata as previously described [40]. Clavibacter michiganensis ssp. michiganensis corrig. (Smith 1910) Davis et al. 1984 (abbreviated as Cmm) strains B-65397 and B-65398, Erwinia amylovora (Burrill 1882) Winslow et al. 1920 B-65467 and B-65468, Fusarium proliferatum (Matsush.) Nirenberg ex Gerlach and Nirenberg 1982 13569, Bipolaris maydis 3797, Colletotrichum graminicola 62057, Pythium sylvaticum 66995 and Ustilago maydis (DC) Cda. YB-255 and YB-380 were obtained from the USDA ARS Culture Collection (NRRL). Fusarium verticillioides (Sacc.) Nirenberg 1976 strain F4 was isolated from maize kernels as previously described [42]. Exserohilum turcicum strain 19StM09 was collected in Illinois [43]. Bacteria cultures were grown using liquid Luria–Bertani (LB) medium (Bacto™ tryptone (Becton, Dickinson and Company, Sparks, MD, USA), 10 g L⁻¹; sodium chloride, 10 g L⁻¹; and Bacto™ yeast extract (Thermo Fisher Scientific, Miami, FL), 5 g L⁻¹) or on LB Petri dishes (100 × 15 mm, Falcon brand, Corning Incorporated, Durham, NC, USA) with 15 g L⁻¹ of Difco™ agar (Becton, Dickinson and Company). For some experiments, Difco™ premixed LB broth (Becton, Dickinson and Company) was substituted. F. proliferatum and F. verticillioides were cultivated on V8 medium at room temperature or at 30 °C; the recipes for these media and the method for collecting spores from these fungi were previously described [42]. B. maydis and C. graminicola fungi were grown at room temperature on Difco™ potato dextrose broth (Becton, Dickinson and Company) with 20 g L⁻¹ of Difco™ agar in Petri dishes (100 × 15 mm). E. turcicum was grown on a V8 juice medium [44] at room temperature in the laboratory with a supplemental LED light set for 12 h of light per day. P. sylvaticum was grown on pea medium (NRRL Medium No. 48 [45]) in the dark at room temperature. U. maydis was grown in YM medium (per liter of medium: yeast extract, 3 g; malt extract, 3 g; proteose peptone, 5 g; glucose, 10 g; agar, 15 g) at 225 RPM at 30 °C. Maize inbred B37 seed was provided by the USDA Agricultural Research Service North Central Regional Plant Introduction Center and self-fertilized for several generations to increase the laboratory supply of seed for experiments.

2.2. Assays Using Cell-Free Extracts

One mL of an LB overnight culture of FS10 or FS11 was added to 20 mL of medium optimal for lipopeptide (MOLP) production [46] in a sterilized 250 or 300 mL Erlenmeyer flask; peptone from casein (Fluka Sigma Aldrich, St. Louis, MO, USA) was used for this medium. These cultures were incubated for four days at 30 °C and 200 RPM and then refrigerated. Aliquots of the culture were centrifuged at 21,130× g at 4 °C, and the supernatant was sterilized using surfactant-free cellulose acetate syringe filters with 0.2 µm pores (Corning Incorporated, Oneonta, NY, USA). FS10 and FS11 cell-free extracts (CFEs) were tested for their ability to inhibit growth of F. verticillioides and F. proliferatum spores, U. maydis YB-255 basidiospores and bacterial cells of Cmm strain B-65397 and E. amylovora B-65467. The assays were performed in sterile 24-well tissue culture plastic plates (Falcon 353047, Corning Incorporated). Each well contained 750 µL of CFE (FS10 or FS11), growth medium, and cells (bacteria, spores or basidiospores) up to a final volume of 1 mL. The growth medium for the bacteria was LB, YM for U. maydis, and PDB for F. verticillioides and F. proliferatum. MOLP medium substituted for the CFE in the control wells. The plate containing the cells was placed in an incubator with a temperature of 30 °C and shaken at 100 RPM. The experiments were stopped after 7 d, but some were stopped earlier if visible growth of the organisms was noted in the wells treated with FS10 and FS11 CFE.

2.3. pH Bioassay

After completing the CFE bioassay, the pH of the contents of each well (that did not contain fungi or bacteria) was measured. In some instances, the contents of some wells evaporated during the bioassay, and in these cases the contents of two wells were mixed, and the pH was measured. New bioassays in 24-well tissue culture plastic plates were initiated to determine if cells could grow in control solutions (75% MOLP, 25% LB or YM) altered to pH 5.5 and 6.9. These two pH values were chosen because they covered the range of the measured pH values in all the experiments. Overnight cultures of Cmm (B-65397), U. maydis (YB-255) or E. amylovora (B-65467) were diluted 1:100 using Ringer’s solution (NaCl, 7.2 g; CaCl₂, 0.17 g; and KCl, 0.37 g per L of water, pH 7.0–7.4). Four wells of a sterile 24-well plate received 500–1000 µL of medium of a specific pH (described above) and 2 µL of diluted cells or basidiospores. Two wells of medium without cells served as negative controls. The wells with these solutions did not have any visible turbidity after the experiment was set up. The plate was placed at 30 °C and 100 RPM and lasted 30–89 h. This method was similar to a study using unbuffered LB with an initial pH of 5.0 and 7.0 [47]. However, in this study culture absorbance was not monitored during cell growth.

2.4. Young Plant Bioassay

In this bioassay, maize inbred B37 was used because its leaves are susceptible at wound sites to fungi [48]. Seeds of maize inbred B37 were incubated at room temperature for 1 h in an LB solution containing 9000 FS10 colony-forming units (CFUs) µL⁻¹ or 47,000 FS11 CFUs µL⁻¹; LB was the dilutant and the control solution. Then the seeds were dried in a laminar flow safety cabinet. The next day, one seed was placed in one 4-inch pot that contained pre-wetted (washed with deionized water) propagation mix soil (Sungro Horticulture, Agawam, MA, USA). The plants were grown in a large climate-controlled room as previously described [49]. After 10–14 days each pot received a plug of Osmocote fertilizer (15-8-11, Everris, the Netherlands) to ensure the plants were healthy at the start of the bioassay. A portion of the second leaf from a five-leaf plant was placed in a 50 × 9 mm Petri dish with a tight-fitting lid (Falcon 351006, Corning Incorporated) containing about 5 mL of water agar medium (agar at 3% v/v). Because the sporulation of the target fungi was poor, media plugs containing actively growing mycelia (B. maydis, C. graminicola, E. turcicum and P. sylvaticum) were removed and used similarly to methodology previously described [50]. Approximately 1 mm² media pieces with actively growing mycelia of each fungus or oomycete were removed with a sterile scalpel and placed on 3 mm slits cut in leaf pieces with the mycelial side contacting the slit. Eight total slits were cut in each leaf piece, control 1 mm² media pieces were placed on the left four slits of the leaf piece, and fungal colonized media were placed on the right four slits of the leaf piece. The dish was closed tightly, and the dishes were placed in a Tupperware container that was wrapped in aluminum foil; all the containers were placed in an incubator set for 27 ± 1 °C. Zones of necrosis or chlorosis were recorded to the nearest mm daily for up to four days.

2.5. In Vitro Bioassay to Determine the Antifungal Activity of Volatiles Produced by FS10 and FS11

FS10 and FS11 cultures grew in LB medium with shaking (200–250 revolutions per minute) overnight at 30 °C. The cultures were diluted to 1:100 with Ringer’s solution. A hundred microliters of this dilution were added to one Petri dish (100 × 15 mm) containing BBL™ trypticase ™ soy broth medium with Difco™ agar (15 g L⁻¹, abbreviated as TSA). This plate served as the source of volatiles produced by the P. megaterium bacteria and contained an estimated 40,000–190,000 colony-forming units (CFUs). The concentration of CFUs in the original P. megaterium culture used for each experiment was calculated by making two additional 1:100 serial dilutions and spreading a volume of the last dilution onto one TSA Petri dish surface and counting the CFUs on the Petri dish after incubating overnight at 30 °C. The mean number of CFUs was calculated from triplicate dishes of the most dilute suspension of bacteria. The amount of CFUs in each experiment is noted in Supplementary Table S1.

Several fungal spores (the amounts for each experiment are listed in Supplementary Table S1) were placed in the center of V8 (for Fusarium fungi) or YM (for Ustilago fungi) medium contained in Petri dishes. The bottom part of the dish with the FS10 or FS11 bacteria was sealed to the bottom part of the V8 or YM dish containing the fungal spores using double layers of Parafilm. A single layer of Scotch^® Magic™ tape (3M, Saint Paul, MN, USA) was added on top of the Parafilm seal, and the dishes, prepared as triplicates, were incubated at 30 °C in the dark with the FS10/FS11 dish or control TSA dish on the bottom (Supplementary Figure S1). The size (the largest distance from left edge to right edge possible) of the fungal mass was measured by a ruler after several days of incubation (the duration of each experiment is noted in Table 1).

2.6. In Vitro Bioassay to Determine the Antibacterial Activity of Volatiles Produced by FS10 and FS11

A TSA plate containing an estimated 50,000–170,000 CFUs of FS10 or FS11 was prepared as described above (estimated numbers of P. megaterium cells can be found listed in Supplementary Table S2). Cultures of the bacteria receiving the volatiles were serially diluted to have approximately 50–100 CFUs on one LB Petri dish. The top lids of the FS10/FS11 dish and the dish containing the bacteria to receive volatiles were removed inside a laminar flow chamber, and then the bottom parts of each dish (containing the medium and bacteria) were sealed with a double layer of Parafilm. A single layer of Scotch^® Magic™ tape was added over the double Parafilm layers. The sealed dishes, with the FS10/FS11 or blank TSA dish on the bottom, were prepared as triplicates and incubated at 30 °C in the dark. The experiment was stopped after a few days (the duration of each experiment is noted in Table 2).

2.7. Measurement of Sizes of Bacterial CFUs

Images of Petri dishes with CFUs were digitized with the Chemidoc XRS+ Imager (Bio-Rad, Hercules, CA, USA) and then exported as 600 dpi jpg files. Images were opened in ImageJ 1.8.0 [51], and the threshold was adjusted so that the red area covered the bacterial CFUs. The “freehand tool” was then used to select an area with individual colonies to prevent the measurement of CFUs where multiple CFUs were touching each other. The “analyze particle function” was then used to calculate CFU pixel area with the minimum size set to 50. The pixel area values were exported to Microsoft Excel. The data points that fell below Q1 − 1.5 IQR or above Q3 + 1.5 IQR were outliers and removed from the dataset; IQR is the range between the first and the third quartiles, specifically Q1 and Q3, where IQR = Q3 − Q1. Means, standard errors, and statistical comparisons were computed using Microsoft Excel. The images and calculations are available in Supplementary Materials (Folders S1–S6).

2.8. Identification of Volatiles

The FS10 and FS11 bacteria were separately grown in LB broth overnight at 30 °C. Gas chromatography (GC) glass vials were autoclaved, and the vial caps were treated with 70% ethanol for several minutes before use; each GC vial contained 4–5 mL of sterile TSA medium. Two loopfuls (using a sterile inoculating loop) of bacteria from the overnight liquid culture were inoculated into a single GC vial. The caps were tightly secured on the vials and incubated for two d at 30 °C. Each bacterial strain was tested in triplicate, and there were two empty vials with TSA control medium that were analyzed as well. Headspace volatiles were analyzed on an ISQ gas-chromatograph mass spectrometer (GC-MS) (ThermoFisher Scientific, Waltham, MA, USA) equipped with a split/splitless injector, a Combi-Pal (CTC Analytics, Lake Elmo, MN, USA) autosampler with a solid-phase micro-extraction (SPME) head and a sample incubator, and Chromeleon v 7.3.2 software. The injector was equipped with an SPME insert and held at 250 °C. The GC column was a Phenomenex (Torrance, CA, USA) ZB-Waxplus™ (30 m length, 0.25 mm inner diameter, 0.25 μm phase thickness). The carrier gas was helium at constant flow (1.5 mL min⁻¹). The MS was operated in electron ionization mode at 70 eV and 310 °C and scanning mass/charge (m/z) 35–350. Samples were equilibrated at 30 °C for 30 min, then extracted with a SPME fiber for 30 min at the same temperature using a 50/30 μm divinylbenzene/Carboxen on polydimethylsiloxane on a StableFlex fiber (MilliporeSigma, Burlington, MA, USA). The fiber was immediately desorbed in the GC injection port for 5 min. Injection port was splitless until complete desorption followed by a 25 mL min⁻¹ split. The oven temperature program was 40 °C for 1 min, followed by a ramp of 10 °C min⁻¹ to 240 °C where it was held for 2 min. Peaks were identified using the National Institute of Standards and Technology (NIST-14) MS library as well as comparing retention time and mass spectra to commercial standards when available.

2.9. Bioassays Using the Volatiles Isovaleric Acid and 3-Methyl-1-butanol

An overnight culture of C. michiganensis ssp. michiganensis strain B-65397 was serially diluted with Ringer’s solution to have approximately 50–100 CFUs on one LB Petri dish. Varying amounts of isovaleric acid (IVA, Aldrich Chemical) and 3-methyl-1-butanol (3MB, Sigma Aldrich Chemical) were added to plastic caps that were placed in empty Petri dish bottoms. The plastic cap was secured to the Petri dish bottom with Scotch^® Magic™ tape. The LB Petri dish bottom containing the target bacteria was placed over the Petri dish bottom with the IVA or 3MB and sealed with two strips of Parafilm. A single layer of Scotch^® Magic™ tape was added on top of the double Parafilm seal (in the same manner described above), and the dishes, prepared as triplicates, were incubated at 30 °C in the dark. After a 3 or 4-day incubation, CFUs were counted, or images of the plates were digitized, and the sizes of the bacterial CFUs were estimated as described above.

2.10. Statistics

Student’s t-test in Microsoft Excel was used for most comparisons between mean values. Because of unequal plant numbers in treatments, analysis of variance (ANOVA) using Proc GLM in SAS (Version 9.4, SAS, Cary, NC, USA) was used to compare the means of the length of pathogen damage in the young plant assay treatments and controls. JMP Pro (V. 17.0, JMP Statistical Discovery, Cary, NC, USA) was used to compare volatile peak mean area counts by ANOVA. The Kruskal–Wallis test was used for benzaldehyde data because it was not normally distributed. Means were compared using Student’s t-test using an α-level of 0.05.

2.11. Data Collection

The data in the study were collected from October 2023 through September 2025.

3. Results

3.1. Antifungal Activity of Volatiles Produced by the P. megaterium Strains

In vitro bioassays using sealed Petri dishes indicated that both FS10 and FS11 produced volatiles that inhibited some of the fungi tested (

Table 1. Volatiles emitted by Priestia megaterium (de Bary 1884) Gupta et al. 2020 strains FS10 and FS11 reduce the average size of a fungal colony in most cases.

Fungus	Control	P. megaterium FS10 Volatiles	P. megaterium FS11 Volatiles	Length of Assay
Fusarium verticillioides (Sacc.) Nirenberg 1976	A 59 ± 2 a (3) 6% B 45 ± 0.3 a (3) 1%	47 ± 0.5 b (2) [20%] 2% 34 ± 0.3 b (3) [24%] 2%	46 ± 0.6 b (3) [22%] 2% 34 ± 0.3 b (3) [24%] 2%	119 h 96 h
Fusarium proliferatum (Matsush.) Nirenberg ex Gerlach and Nirenberg 1982	A 53 ± 0.6 a (3) 2% B 40 ± 0.8 a (3) 3%	37 ± 3 b (3) [30%] 12% 33 ± 0.7 b (3) [17%] 4%	40 ± 0 b (2) [24%] 0% 33 ± 1 b (3) [17%] 3%	96 h 96 h
Ustilago maydis (DC) Cda. YB-255	18 ± 0.6 a (3) 6%	4.3 ± 0.2 b (3) [76%] 7%	4.3 ± 0.2 b (3) [76%] 7%	72 h
U. maydis YB-380	17 ± 1 a (3) 12%	7 ± 0.3 b (3) [59%] 9%	7 ± 0.4 b (3) [59%] 11%	73 h

Independent biological replicates are indicated by A and B; means in mm, ± standard error; means in the same row followed by different letters are statistically significant at p < 0.05 using Student’s t-test; (technical replicates); [percent growth inhibition]; coefficient of variation.

, Figure 1). The growth of Fusarium fungi exposed to the bacterial volatiles was inhibited from 17 to 30% compared to the controls. Basidiospores of the fungus Ustilago maydis were the most sensitive to the volatiles produced by FS10 and FS11, inhibiting the growth of YB-255 76% and YB-380 59% (Table 1, Figure 1).

3.2. Identification of Volatiles Produced by the P. megaterium Strains

GC analysis of headspace in a glass vial was completed after FS10 and FS11 bacteria had grown for 2 d at 30 °C (Figure 2). The chromatograms of the two control vials had peaks at 7.57 and 10.3 min, which corresponded to 2,5-dimethylpyrazine and benzaldehyde, respectively (Figure 2 and Supplementary Table S4). These two compounds can be extracted from whole soybeans by water [52]. Since soybeans were part of the growth medium in the vials, it is likely that these compounds were volatilized from the growth medium in the control vials during the 2 d.

The 2,5-dimethylpyrazine peak was found in the chromatograms of the FS10 and FS11 vials at approximately the same levels (Figure 3); however, the benzaldehyde peaks were much smaller in 4 of the 6 FS10 or FS11 vials than in the control vials. This suggests that the bacteria were able to metabolize the benzaldehyde but not the 2,5-dimethylpyrazine. FS10 and FS11 bacteria produced 3-methyl-1-butanol (3MB, retention time of 5.96 min) and isovaleric acid (IVA, retention time of 12.01 min), with the relative area of 3MB being larger than the relative area of IVA (Figure 3). Both 3MB and IVA have demonstrated antifungal and antibacterial activity [53,54,55,56].

3.3. Inhibition of Bacterial Growth by the P. megaterium Volatiles

Since 3MB and IVA have antibacterial activity, bioassays like those performed with fungi in the present study were performed to test the effects of volatiles produced by FS10 and FS11 on bacteria CFU growth (

Table 2. Volatiles emitted by Priestia megaterium (de Bary 1884) Gupta et al. 2020 strains FS10 and FS11 reduced the average size of a bacterium CFU.

Bacterium	Control	P. megaterium FS10 Volatiles	P. megaterium FS11 Volatiles	Length of Assay
P. megaterium FS10	3759 ± 56 a (61) 12%	3084 ± 46 b (53) [18%] 11%	2705 ± 60 c (55) [28%] 16%	23 h
P. megaterium FS11	2787 ± 44 a (75) 14%	2659 ± 72 ab (80) [5%] 24%	2536 ± 71 b (73) [9%] 24%	23 h
Clavibacter michiganensis ssp. michiganensis corrig. (Smith 1910) Davis et al. 1984 (Cmm) B-65397	2701 ± 84 a (45) 21%	40 ± 1 b (73) [98%] 28%	34 ± 1 c (57) [99%] 20%	144 h
Cmm B-65398	1454 ± 33 a (61) 18%	33 ± 1 b (132) [98%] 48%	59 ± 3 c (121) [96%] 52%	89 h
Erwinia amylovora (Burrill 1882) Winslow et al. 1920 B-65467	1970 ± 48 a (17) 10%	363 ± 46 b (41) [82%] 81%	305 ± 38 b (38) [84%] 78%	49 h
E. amylovora B-65468	2290 ± 24 a (107) 11%	457 ± 8 b (172) [80%] 23%	341 ± 5 c (159) [85%] 19%	50 h

Means in pixels, ± standard error; means with different letters in the same row are significantly different using Student’s t-test in Microsoft Excel; (total number of CFUs, technical replicates); [percentage of growth inhibition]; coefficient of variation.

). The volatiles produced by strains FS10 and FS11 slightly inhibited (5–28% compared to the control) the growth of themselves (Table 2). There were some bacterial strains whose growth was significantly inhibited (96–99%) by volatiles produced by both FS10 and FS11, including Cmm lines B-65397 and B-65398, a pathogen of tomato.

The mean size of E. amylovora CFUs was reduced 80–85% by the volatiles produced by FS10 and FS11 (Figure 4). In three of six independent tests with bacteria, the average size of CFUs exposed to FS11 volatiles was smaller than the average size of CFUs exposed to FS10 volatiles; in two of six independent tests, there were no statistically significant differences between the average size of bacteria exposed to FS10 or FS11 volatiles; and in one of six independent tests, the average size of CFUs exposed to FS11 volatiles was greater than the size of CFUs exposed to FS10. All these experiments indicated that volatile compounds produced by both FS10 and FS11 suppressed bacterial growth more than fungal growth.

Since the tested bacteria were more sensitive to FS10 and FS11-produced volatiles than the tested fungi, in vitro growth experiments with Cmm B-65397 bacteria were performed using varying concentrations of 3MB and IVA. Cmm growth was significantly inhibited by IVA at 83 µL L⁻¹, and IVA completely inhibited growth at 208 µL L⁻¹ in two independent experiments (

Table 3. Isovaleric acid (IVA) inhibited the growth of Clavibacter michiganensis ssp. michiganensis corrig. (Smith 1910) Davis et al. 1984 B-65397 bacteria.

Treatment	Experiment 1, CFUs per Plate After 4 d	Experiment 2, CFUs per Plate After 4 d
Control	25 ± 3 (3) a [22%]	103 ± 7 (3) a [11%]
IVA at 83 µL L−1	5 ± 3 (3) b [108%]	35 ± 9 (3) b [43%]
IVA at 208 µL L−1	0 (3)	0 (3)
IVA at 417 µL L−1	0 (3)	0 (3)

Mean ± standard error in pixels; (technical replicates); means in the same column with different letters are statistically significant at p < 0.05 using Student’s t-test; [coefficient of variation].

and Table S3). The IVA treatment was removed, and the plates with Cmm bacteria were allowed to grow for an additional five days. The number of Cmm colonies increased on the plates that had received the 83 µL L⁻¹ IVA treatment: Experiment 1, 20 ± 5 (SE) CFUs, and Experiment 2, 91 ± 9 (SE) CFUs. These data indicate that some Cmm bacteria could recover from the 83 µL L⁻¹ IVA treatment. However, no Cmm bacteria grew on plates (after removal of IVA) that had been treated with 208 and 417 µL L⁻¹ IVA.

The same bacteria were used to test the effects of 3MB. In these experiments, the size of the CFU was negatively correlated to the amount of 3MB utilized (

Table 4. 3-methyl-1-butanol (3MB) inhibited the growth of Clavibacter michiganensis ssp. michiganensis corrig. (Smith 1910) Davis et al. 1984 B-65397 bacteria.

Treatment	Experiment 1, Mean Size of CFU After 3 d	Experiment 2, Mean Size of CFU After 3 d
Control	463 ± 12 a (74) [22%]	570 ± 27 a (89) [44%]
3MB at 208 µL L−1	226 ± 6 b (72) [22%]	282 ± 14 b (73) [42%]
3MB at 417 µL L−1	45 ± 2 c (53) [38%]	108 ± 7 c (46) [42%]
3MB at 625 µL L−1	0	38 ± 3 d (28) [37%]

Means in the same column with different letters are statistically significant at p < 0.05 using Student’s t-test; (technical replicates); [coefficient of variation].

and Supplementary Folders S7 and S8). It should be noted that no CFUs could be seen for the 625 µL L⁻¹ 3MB treatment of Cmm in Experiment 1 after 3 d. The 625 µL L⁻¹ 3MB treatment was removed, and the plates were placed at 30 °C; small Cmm CFUs could be seen on the plates after 113 h. The mean size of Cmm CFUs with 417 µL L⁻¹ 3MB treatment was 10% and 19% of the mean size of the control CFUs in experiments 1 and 2, respectively. Interestingly, the 208 µL L⁻¹ IVA treatment killed all the Cmm, but the 208 µL L⁻¹ 3MB treatment reduced the mean Cmm CFU size to half the CFU size of the control bacteria.

3.4. Inhibition of Fungal and Bacterial Growth Using P. megaterium Cell-Free Extracts

When the FS10 and FS11 CFEs were diluted 1:1.333, no growth was seen for U. maydis YB-255 basidiospores, Cmm B-65397 cells, or E. amylovora B-65467 cells (

Table 5. Priestia megaterium (de Bary 1884) Gupta et al. 2020 FS10 and FS11 CFEs arrested growth of fungi and bacteria.

Organism	Mean Number of Cells, Spores or Basidiospores per Well	Control Growth	Growth with FS10 CFE Prep 1	Growth with FS11 CFE Prep 1	Growth at Tested pH, Length of Assay
Fusarium verticillioides (Sacc.) Nirenberg 1976	85	Yes	Yes	Yes	ND
Ustilago maydis (DC) Cda. YB-255	100	Yes	No, pH 6.7 ± 0.05 (4) [1%]	No, pH 5.6 ± 0.03 (4) [0.9%]	Good growth at 5.5 and 6.9, 48 h
Clavibacter michiganensis ssp. michiganensis corrig. (Smith 1910) Davis et al. 1984 B-65397 Gr+	95	Yes	No, pH 6.0 ± 0.05 (4) [2%]	No, pH 5.6 ± 0.03 (3) [1%]	Weak growth at 5.5, good growth at 6.9, 89 h
Organism	Mean number of cells, spores or basidiospores per well	Control growth	Growth with FS10 CFE Prep 2	Growth with FS11 CFE Prep 2	Growth at tested pH
Fusarium proliferatum (Matsush.) Nirenberg ex Gerlach And Nirenberg 1982	26	Yes	No, pH 6.9 ± 0.12 (3) [3%]	Yes	ND
U. maydis YB-255	120	Yes	No, pH 6.2 ± 0.03 (4) [0.8%]	No, pH 5.7 ± 0.03 (4) [1%]	See above
E. amylovora (Burrill 1882) Winslow et al. 1920 B-65467 Gr−	38	Yes	No, pH 5.7 ± 0.03 (3) [1%]	No, 5.7 ± 0 (2) [0]	Good growth at 5.5 and 6.9, 30 h

Gr, Gram-stain + or −; ND, not determined; mean pH ± standard error (technical replicates) [coefficient of variation].

). However, F. verticillioides spores were able to grow under the same conditions. Fusarium proliferatum spores grew in diluted FS11 CFE but not in diluted FS10 CFE. Since the pH of the diluted CFE and growth medium might affect growth of fungi or bacteria during the bioassay, pH values were recorded for wells that had no visible growth (Table 5). New bioassays were conducted using 75% MOLP and 25% YM or LB that was adjusted to a specific pH value. Most of the cells, or spores, or basidiospores grew well at the tested pH values (in the far-right column of Table 5). One exception was Cmm B-65397 with a pH 5.5 medium, in which a small ball of material was visible in each well; the contents of the four wells were individually transferred to a sterile LB plate and incubated at 30 °C for 2 days; strong growth of the yellow bacteria was observed on each of the four plates, indicating that pH 5.5 was inhibiting growth of the bacteria. No pH test was performed for F. proliferatum since the pH of unaltered 75% MOLP and 25% PDB was 6.9, which matched the mean pH of the three replicates of the diluted FS10 CFEs.

3.5. Reduction in Maize Pathogen Leaf Damage with P. megaterium Seed Treatment

B. maydis and P. sylvaticum caused significant leaf necrosis 2 and 3 days after inoculation (

Table 6. Treatment of maize seed with Priestia megaterium (de Bary 1884) Gupta et al. 2020 FS10 or FS11 bacteria reduces seedling leaf damage caused by fungal and oomycete plant pathogens.

Treatment	Control	FS10	FS11
Bipolaris maydis (Y. Nisik. and C. Miyake) Shoemaker 1959, 2–day necrosis	9.1 ± 0.4 a (24) 21%	5.0 ± 0.5 b (24)] [45%] 54%	4.1 ± 0.4 b (32) [55%] 54%
Colletotrichum graminicola (Ces.) G.W. Wilson 1914, 4–day necrosis, non–zero ratings only	3.4 ± 0.3 a (12) 49%	1.9 ± 0.1 b (10) [44%] 39%	2.1 ± 0.3 b (11) [38%] 42%
Exserohilum turcicum (Pass.) K.J. Leonard and Suggs 2018, 2–day chlorosis	1.9 ± 0.4 a (24) 105%	0.3 ± 0.1 b (24) [84%] 164%	0.6 ± 0.2 b (32) [68%] 186%
Pythium sylvaticum W.A. Campb. and F.F. Hendrix 1967, 3–day necrosis	1.7 ± 0.2 a (24) 60%	0.8 ± 0.2 b (24) [53%] 113%	0.4 ± 0.1 b (32) [76%] 188%

The means ± standard error are listed first; means in rows followed by a different letter are significantly different by analysis of variance; (number of technical replicates); [percentage of inhibition]; coefficient of variation.

). Seed treatment with FS10 or FS11 reduced the necrosis caused by these pathogens by 45–76%. FS10 bacteria seed treatment reduced chlorosis caused by E. turcicum by 84% (Figure 5, the greatest reduction observed in these bioassays), but the FS11 treatment (68% reduction) was not as effective as the FS10 seed treatment (Table 6). Necrotic zones (including for controls) were not as numerous in the bioassays using C. graminicola, possibly because of less vigorous pathogen growth. Nevertheless, when using non-zero values, both FS10 and FS11 seed treatments significantly reduced the amount of necrosis caused by C. graminicola after 4 d.

4. Discussion

Bacteria produce many volatile compounds, some of which are also made by fungi. 3MB was the most commonly found volatile across 221 bacterial and fungal genera [57] and was more prevalent in bacterial genera (54.9%) than in fungal (45.1%) genera. 3MB was an effective antifungal molecule in several studies [55,58,59,60,61]. Only two reports of antibacterial activity of 3MB could be found in the literature: one on cyanobacteria [62] and one on Agrobacterium tumefaciens (Smith and Townsend 1907) Conn 1942 [56]. In one of the studies, twenty-five µmoles of 3MB inhibited the growth of A. tumefaciens in two-compartment Petri dishes (one compartment contained the bacteria while the other compartment contained the volatile), while 100 µmoles stopped all growth of the bacteria [56]. The lowest 3MB dose in the present study was 227 µmoles (Table 4), and Cmm B-65397 bacteria were able to grow at this dose. These conflicting results may be due to the different arrangements of the assay where the volatile was next to A. tumefaciens in the Sidorova et al. study [56], whereas in the current study the volatile was directly below the bacteria. In addition, the growth medium in the current study could have absorbed a large amount of 3MB that was then not available to affect the bacteria. Nevertheless, the results of the current study paralleled those of Sidorova et al. (2021) [56], which showed that increasing the dose of 3MB inhibited the growth of bacteria.

Cmm is a pathogen of tomato. Fumigation of tomato seeds with IVA or 3MB could be a possible treatment to reduce the number of Cmm that infect mature plants. Sidorova et al. (2021) tested the effects of 3MB on seedling growth and seed germination of Arabidopsis thaliana and demonstrated that the volatile inhibited both developmental stages [56]. Similar effects were found by another research group testing 3MB on wheat seeds [63]. 3MB was phytotoxic on orange fruits at concentrations of 1 and 0.66 µL mL⁻¹ of air space [55]. However, when the 3MB concentration was reduced to 0.33 µL mL⁻¹ of air space, the fruit was not damaged, and the treatment was able to reduce the number of lesions caused by the fungus Phyllosticta citricarpa (McAlpine) Aa 1973 [55]. This result indicates that effective fungal or bacterial pathogen control using 3MB on agricultural commodities, including seeds, might be possible with extensive testing to determine the amount of 3MB that has no toxicity to plant tissue but prevents pathogen growth.

IVA was found in 34 of 221 tested genera [57] and was more common in bacterial genera (79.4%) than in fungal genera (20.6%). IVA produced by Pseudoalteromonas haloplanktis (ZoBell and Upham 1944) Gauthier et al. 1995 was partially purified, and it contributed to the killing of Vibrio ordalii (Schiewe et al. 1982) [64]. Pure IVA inhibited the growth of several bacteria, although 2130 µM, or higher concentrations, was needed to inhibit 80% of growth of the bacteria, which included three Streptococcus species, Porphyromonas gingivalis (Coykendall et al. 1980) Shah and Collins 1988 emend. Hahnke et al. 2016, Aggregatibacter actinomycetemcomitans (Klinger 1912) Norskov-Lauritsen and Kilian 2006, and Fusobacterium nucleatum Knorr 1922 [54]. IVA (at 80 µL L⁻¹) was able to completely inhibit the growth of a fungus, Botrytis cinerea, inoculated onto table grapes after 4 d of fumigation [53]. Damage to the grapes with IVA levels from 20 to 640 µL L⁻¹ was not reported. The authors hypothesized that IVA induces fungal resistance in the grapes, which may explain the remarkable antifungal activity of IVA. Additional research found that IVA is a broad-spectrum phytotoxin that killed plants (Echinochloa crus-galli (L.) P. Beauv. 1812, Mazus japonicus (Thunb.) Kuntze, Taraxacum mongolicum Hand.-Mazz. 1907, and Brassica campestris L.) with a dosage of 196-588 mM seven days after treatment [65]. The authors did not indicate how much of the IVA solution was added to each plant. If IVA is to be used as a plant protection product in the future, then a concentration must be identified that kills the pathogen (bacterium or fungus) but does not harm the plant commodity.

It is difficult to know the concentration of volatiles produced by FS10 and FS11 within plants as they encounter other beneficial and pathogenic organisms. Based on the experiments in this study, the volatiles produced by FS10 and FS11 had only a marginal effect on the growth of fungi and would likely not be a major contributor to reducing the fungi attacking a plant. On the other hand, this study determined that 3MB and IVA were effective in reducing the growth of Cmm, a pathogenic plant bacterium. Both 3MB and IVA are used in the fragrance industry and are generally regarded as safe at allowable levels [66,67]. Therefore, experimentation with 3MB and IVA as fumigants may be beneficial for a variety of products in the agriculture industry.

MOLP medium is effective at inducing antifungal lipopeptide production in Bacillus strains at the stationary phase of growth [46]. MOLP medium induced fengycin and surfactin production in two Bacillus megaterium strains; lipopeptide extracts from the MOLP medium of these two strains inhibited mycelial growth of Fusarium sambucinum (Fuckel 1863) and Verticillium dahliae (Klebahn 1913) [21]. Secreted biochemical(s) from FS10 and FS11 grown in MOLP medium were effective in limiting the growth of Gram-positive and negative bacteria (Table 5). However, the antifungal activity of the FS10 and FS11 CFEs varied in potency with the tested fungi. U. maydis basidiospores were sensitive to both FS10 and FS11 CFEs, while FS10 CFE exhibited more potent antifungal activity than FS11 CFE when tested with F. proliferatum. This phenomenon is not unusual as CFE from one P. megaterium strain was able to inhibit fungal mycelial growth better than CFE from another P. megaterium strain in bioassays using V. dahliae [21]. Interestingly, CFE from three P. megaterium lines did not inhibit two tested fungal species but had variable effects on the growth of nine bacteria, although the authors did not use MOLP medium for growing their bacteria [16]. All these results suggest that the components of laboratory-derived media could play a big role in determining which antibiotics are synthesized by bacteria and secreted into the medium. Additionally, the pH bioassays in this study indicated that an acidic pH did not stop growth of the tested fungi or bacteria. This likely means that in nearly all the bioassays that authentic antifungal or antibacterial compounds are likely present in the CFEs. Detailed identification of the bioactive components of the FS10 and FS11 CFEs will be completed in the future. The soluble antibacterial compounds from FS10 and FS11 could be especially important for agricultural, biomedical or livestock applications with the global trend of increasing bacterial resistance to antibiotics [68,69].

Overnight cultures of FS10 and FS11 served as inoculum for seed coating or volatile plate assays in this study. The cultures were initiated by removing cells from a colony on a plate using a sterile plastic loop and shaking it in sterile LB medium. Rather than waiting to determine the cell concentration in the overnight cultures, the FS10 and FS11 cultures were diluted on the day after inoculation and equivalent dilute cell volumes spread onto plates in the same manner or used for seed inoculation. There was generally a large lawn of bacteria that grew on the plate the next day. For unknown reasons, the number of CFUs per mL in the overnight culture was generally greater for FS11 compared to FS10 cultures. However, in only 3 of 16 bioassays did FS11 volatiles (or ISR factors) outperform FS10 in slowing the growth of bacteria or fungi or in reducing the amount of plant damage. The three cases all involved reducing the size of bacterial CFUs. This finding indicates that having more FS11 cells on a plate or on a seed was not advantageous for greater reductions in growth or minimizing pathogen damage. On the other hand, the lowest amount of FS10 cells on a plate, 40,000, was able to reduce the size of U. maydis growth by 76% compared to the controls. In bioassays with F. proliferatum, 60,000 FS10 cells reduced growth of the fungal colony by only 17% compared to the controls. These data indicate that more in vitro experiments need to be performed to determine the minimum number of bacterial cells that are needed to cause a reduction in growth. This minimum number of bacteria might be different for causing significant inhibition in various target organisms.

In the present study, when the P. megaterium strains were applied to seeds, they reduced the damage to leaves compared to controls when three fungal pathogens and one oomycete maize pathogen were applied (Table 6). The results of this study agree with many other published reports showing that P. megaterium was an effective agent for reducing pathogen damage in plants other than maize [33,34,35,36,37,38,39]. Bacteria (Pseudomonas putida (Trevisan 1889) Migula 1895, Bacillus subtilis and Bacillus pumilus (Meyer and Gottheil 1901) inoculated onto maize roots or seeds induced systemic resistance to bacterial or fungal disease [29,30]. All of this data indicate that there is good potential for seed coating with FS10 or FS11 to lower disease levels in field-grown maize leaves or seedlings caused by B. maydis, C. graminicola, E. turcicum or P. sylvaticum since seed coating with FS10 or FS11 reduced the incidence and severity of tar spot pathogen in field-grown maize leaves [41].

5. Conclusions

It was surprising to discover that FS10 and FS11 volatile and secreted compounds were not very effective against fungi in our laboratory tests, which indicated our hypothesis was weakly met; these results might be due to the inherent resistance of the organisms tested. To our knowledge, this is the first report that IVA can be very effective at killing Cmm, a problem pathogen for tomato growers. This study also determined that FS10 and FS11 are producing antibacterial compounds at their stationary growth phase that can kill Gram-stain positive and negative bacteria. Sequencing the genomes of these bacteria in the future may help guide the isolation of the soluble antibacterial compounds. The maize young plant tests expanded the utility of FS10 and FS11 as inducers of plant resistance to one fungus (Phyllachora maydis) to four fungi and one oomycete. These results indicated that our hypothesis about treatment of corn seeds inducing plant resistance to pathogens was fulfilled. Additional research under commercial field conditions is needed to further evaluate the promising potential reported in the present study for P. megaterium FS10 and FS11. These strains of P. megaterium have the potential to reduce disease damage in maize and other crops using different application methods, thereby enhancing maize production to benefit growers, end users, and consumers.