A total of 160 bacterial strains were isolated from soil samples, of which ten showed antifungal activity against S. maydis and S. macrospora (Figures 1 and 2, respectively). The strains 13, 35, 55, 135, and 160 were Gram positive and the strains 11, 16, 19, 21, and 156 were Gram negative.

The identity of the isolated bacteria with antifungal activity was determined by amplification and sequencing of the 16S rRNA. The edited sequences were deposited in GeneBank, under accession numbers GU220577 (11), GU220584 (13), GU220578 (16), GU220579 (19), GU220576 (21), GU220581 (35), GU220580 (55), GU220582 (135), GU220575 (156) and GU220583 (160).

The phylogenetic analysis showed that the Gram positive strains 13, 35, 55, 135, and 160 were related with Bacillus subtilis (Figure 3). Most of the Gram negative strains were related with Pseudomonas genera, specifically strains 156, 21, and 11. The bacterial 16 was grouped with Pseudomonas fluorescens, whereas the 19 strain was grouped with Pantoea agglomerans (Figure 4).

After determining the growth curve of each isolated bacterial strain with antifungal activity, different extracellular filtrates were obtained, as described in Materials and Methods, in this case the sampling was at 8, 16 and 32 h for the Gram negative bacteria and for the Gram positive bacteria was at 12, 24 and 50 h, to correspond at logarithmic, stationary early and stationary late phases respectively. The filtrates obtained from Gram negative bacteria showed a low inhibition or did not inhibit fungal development. However, the samples obtained in all growth phases from the Pseudomonas spp. 11 inhibited up to 70% of both fungi and the filtrates obtained in logarithmic phase from the P. fluorescens 16 inhibited 54% the growth of S. maydis. In general, the filtrates obtained from all strains of B. subtilis inhibited considerably the growth of both fungi. Most of the samples with high activity were obtained in the stationary phase (Table 1).

The extracellular filtrates with the best antifungal activity were selected to measure their stability at different conditions of temperature and pH. From Pseudomonas spp. 11 and P. fluorescens 16, the filtrates were obtained from a culture of 26 and 16 h that corresponded to the stationary and logarithmic phase, respectively. In contrast, in the case of B. subtilis 55, the filtrates were obtained from early stationary phase (22 h), and the filtrates selected from B. subtilis 13, 35, 135, and 160 strains were those obtained at 36, 35, 38, and 48 h respectively (late stationary phase).

The filtrates obtained from Pseudomonas spp. 11 and P. fluorescens 16 did not show stability at the different temperatures and pH conditions assayed, in contrast with the filtrates from B. subtilis, which maintained their antifungal activity even after being incubated for 30 days at different temperatures and pH values (Figures 5 and 6). Furthermore, five filtrates obtained from the B. subtilis 13, 5, 55, 135, and 160 strains showed stability to sterilization (Figure 7). These thermostable extracts did not show cytotoxicity against brine shrimp. In general, all extracts showed higher antifungal activity on S. maydis than on S. macrospora.

When the same fltrates were inoculated in plates of YEPD with a conidia suspension of S. maydis and S. macrospora, no colony development were observed. However, when the plates were observed under the microscope they exhibited abnormal germination tube and no hyphae grew from the swollen cells, which are like chlamydospores and not were present in cultures without extracts (Figure 8).

The use of biological control in the management of agriculture pests and diseases is an effective alternative to the use of pesticides, which are often accumulated in plants and are lethal to beneficial organisms present in the soil [13]. Nowadays, the development of new environmentally friendly methodologies, are based on the search for new genetic, chemical, and biological sources that are effective for phytosanitary issues [14]. Although different microorganisms have been utilized as control agents, antibiotics produced by bacteria isolated from the soil contribute to the suppression or inhibition of the growth of phytopathogens [10]. In the present work, the isolation of bacteria from soil of corn crops was performed with the idea of using these strains as biological agents in the integrated management of ear rot caused by S. maydis and S. macrospora.

Of the 160 bacteria isolated, only 10 showed antifungal activity over the studied fungi through diffusible substances in agar, of which some Gram-negative bacteria showed a higher inhibitory effect than the Gram-positive strains. However, the extracellular filtrates obtained from these same microorganisms presented low antifungal activity. Numerous Gram-negative bacteria have been used in the control of fungal plant pathogens. These bacteria exert their antifungal activity through the production of extracellular lytic enzymes, siderophores, salicylic acid, antibiotics, and volatile metabolites, such as hydrogen cyanide [15–20]. It is possible that some Gram-negative bacteria isolated in this work produce antifungal volatile compounds or enzymatic activities that are lost in the process of obtaining the filtrates. On the other hand, it can be assumed that the presence of bacteria is necessary for the substances responsible of the inhibition, and could be associated with the bacterial envelopes.

Molecular identification showed that most of the Gram-negative bacteria are included in the Pseudomonas genus (11, 16, 21, and 156 strains), specifically the 16 strain was grouped in the cluster of P. fluorescens. Numerous species of this genus are used as biological control agents of soil pathogens for their ability to colonize plant roots. Some of them can increase plant growth due to a property that inhibits the development of plant pathogenic microorganisms [15,16,20–23]. At present, there are many products containing Pseudomonas with moderate to excellent efficacy that are used for plant disease control [24]. Recently, the genomic analysis of this bacterial group has enabled the development of tools for the effective management of biological control agents in the field. However, the inoculation of native strains is important to exploit the beneficial properties of these microorganisms in agriculture because these microorganism are best adapted to conditions in the place when the phytosanitary problem is present [25].

Strain 19 was identified as P. agglomerans. This microorganism is used as a biological control agent for the bacterial pathogens P. syringae pv. syringae and Erwinia amylovora causal agents of the basal kernel blight of barley and fire blight in orchards, respectively [26,27]; however, their main use is in post-harvest biological control of fungi in fruits and citrus, inhibiting the development of Penicillium expansum, Botritis cinereae, Rhizopus stolonifer, Penicillium digitatum, and Monilinia laxa, and therefore, could be an alternative to chemical fungicides [28–34].

The most active filtrates were obtained from the Gram-positive bacteria that were identified as B. subtilis. The samples that presented higher inhibition of the fungi under study were obtained in the stationary phase of growth, indicating that the compounds are secondary metabolites. Within the Gram-positive bacteria used as biological control agents, some species of the Bacillus genus (B. subtilis, B. licheniformis, B. pumilus, B. amyloliquefaciens, and B. cereus) inhibit the development of different fungal plant pathogens due to their capacity to produce a broad variety of antibiotics [35–37].

The filtrates obtained from the five strains of B. subtilis (13, 35, 55, 135, and 160) maintained their antifungal activity even after being autoclaved, besides being stable at different pH values. Some species of Bacillus produce antibiotics with a broad spectrum of action besides being stable to high temperatures and different pH conditions. These compounds are cyclic lipopeptides with an amphipathic character that are divided into three groups: surfactins, iturins, and fengicins, which are secondary metabolites [13,35]. It is possible that the antifungal activities found in the five strains of B. subtilis correspond or are similar to these compounds, and could potentially be used in the field of ear rot management caused by Stenocarpella spp. due to their stability at different temperatures and pH conditions.

In a B. amyloliquefaciens strain was reported that Iturin A is the principal compound with antifungal activity that inhibits different postharvest fungal pathogens. The inhibitory effect was observed in abnormal conidial germination and germ tube development when conidia were treated when different lipopeptides extracts [38]. The thermostable filtrates obtained in this work are not cytotoxic and have the ability to cause abnormal germination of conidia.

The method most used for the control of S. maydis and S. macrospora in the field is crop rotation, however this practice requires suspending corn production by several years with the objective of decreasing the spore inoculum in the soil. On the other hand, the utilization of commercial hybrids has rarely been effective and losses are very important during harvest and post-harvest. In this last aspect, farm animals rejected the stored grains infected by Stenocarpella and their nutritional value is diminished [8]. Even more, S. maydis produces the diplodiatoxin a mycotoxin associated with different ailments in domestic ruminants, ducks, chickens, and horses [4–6]. The biological control of ear rot has scarcely been studied, with one report of isolation and utilization of actinomycetes at the experimental level, focused on strains of the Streptomyces genus with the potential to be used in the field [12].

The strains used in this study were S. maydis 2 and S. macrospora 533-4 kindly provided by Dr. Dan Jeffers from the Centro Internacional de Mejoramiento de Maíz y Trigo (CIMMYT), Mexico.

The fungal strains were grown in Petri plates containing potato dextrose agar medium (PDA) and YEPD solid medium (1% yeast extract, 2% peptone, 2% dextrose and 2% agar). The plates were maintained at 28 °C for 8 days, and then kept at −4 °C.

The following media were used to grow the bacterial strains, trypticasein soy agar medium (TSA), PDA medium at 28 °C, Kings medium B and potato dextrose broth medium (PDB) on a rotary shaker at 200 rpm at 28 °C for 20 h, and each strain was maintained in 25% glycerol at −80 °C.

Bacteria were isolated from rhizospheric soil samples obtained from corn crops in the state of Hidalgo, located in the central part of Mexico. Suspensions with 10 g of the sampled soil and 90 mL of sterile distilled water were prepared; each suspension was diluted from 10⁻¹ until 10⁻⁸. From these dilutions, 0.1 mL were inoculated in plates with TSA and Kings medium B and incubated at 28 °C. The plates were observed every 12 h for 72 h. Colonies with different characteristics were selected and preserved in TSA.

All isolated bacteria were tested for antifungal activity as follows: a 1-cm² fungal plug (S. maydis or S. macrospora) was inoculated in the center of a plate with PDA, each isolated bacterium was sown with a sterile stick at a distance of 2.5 cm from the fungus. The plates were then incubated at 28 °C for 72 h and verified every 12 h. The strains with capacity to inhibit proliferation of the fungus were selected and preserved. For an initial classification of the bacterial strains, was necessary to perform a morphological characterization of the colony and Gram staining [39].

The bacterial strains were identified by molecular methods by the amplification and sequencing of the 16S rRNA using the following universal primers: 8F 5′-AGAGTTTGATCCTGGCTCAG-3′ and 1492R 5′-GTTACCTTGTTACGACTT-3′. The PCR conditions were initial denaturalization for 5 min at 95 °C, 30 cycles of 60 s at 95 °C, 60 s at 54°C, 90 s at 72 °C and the final extension for 10 min at 72 °C. The PCR products were cleaned with the QIAquick PCR purification kit (Qiagen, USA) following the manufacturer instructions. The PCR-products were sequenced in both directions in an Applied Biosystems model 3730XL automated DNA sequencing system. Both strands of DNA were compared and edited manually. Multiple alignments of sequences were carried out with Clustal X [40]. A maximum likelihood (ML) analysis was done in PAUP*(v.4.0b10) [41] using a heuristic search strategy with tree bisection-reconnection: ACCTRAN and 10 random-taxon-addition replicates, starting from a random tree. Confidence at each node was assessed using a bootstrap analysis based on 5000 pseudoreplicates with 10 random-taxon-addition replicates per pseudoreplicate.

To obtain extracellular filtrates with antifungal activity, samples were collected in logarithmic phase and at the beginning and end stationary phase of the growth of each isolated bacterium with antifungal activity, cultured in PDB medium, at 28 °C, at 180 rpm. The samples were centrifuged at 8000 rpm and the supernatants (extracellular fraction) were sterilized by filtration, using a 0.22-μm pore diameter membrane, and then 200 μL to the filtrate was added in small bottomless tubes placed on PDA plates at a distance of 2.5 cm from the center to where the fungi were inoculated. The plates were incubated at 28 °C for 120 h and verified every 12 h. The filtrates with antifungal activity were those that inhibited the development of the fungi.

To evaluate the effect of the pH and the temperature in the stability of the extracellular filtrates, they were incubated during 1, 7, 15, and 30 days at different pH conditions (3.6, 7, and 10 pH values) and different temperatures (−20, 4, 18, 37, 50 °C); in the first case, 500 μL of the filtrate was mixed with 500 μL of different buffers (0.5 M citrate buffer, pH 3.6; 0.5 M phosphate buffer, pH 7; 0.5 M Tris-HCl buffer, pH 10), as a control, 500 μL of PDB was mixed with the different buffers. One sample of these filtrates was sterilized in an autoclave. All filtrates were tested to determine their antifungal activity, as described previously. Results are reported as percentage antifungal activity. The brine shrimp lethality test was used to evaluated the cytotoxicity of those filtrates that were thermostable according at the procedures described by Solis et al. [42].

The conidia were obtained from pycnidia, induced by growing the fungi for 25 days at 28 °C with periods of 12 h light and 12 h darkness in 200 wood toothpicks were placed in beakers of 100 mL with 25 mL of PDB diluted 1:2 with distilled water. The pycnidia were collected in sterile water and were then vigorously shaken to liberated the conidia. This solution was diluted until 10⁻⁵ and 0.1 mL were mixed with 300 μL of the each thermostable extracellular filtrate and inoculated in YEPD plates, then were incubated for 96 h at 28 °C. Each plate was observed in an optical microscope.