Plant tissue recollection, isolation and identification of the microorganism

The collected plant tissue came from fruits, bark and branches of mango (Mangifera indica L.), coconut (Cocos nucifera L.), passion fruit (Passiflora edulis L.), papaya (Carica papaya L.), guaraná (Paullinia cupana L.) and cocoa (Theobroma cacao L.). The samples were collected at the Cotaxtla and Huimanguillo experimental fields from the National Institute of Agricultural and Livestock Forestry Research (INIFAP) at the states of Veracruz and Tabasco, Mexico. The Cotaxtla field is located at 34.5 km of the federal Veracruz-Córdoba highway, Medellín de Bravo, Veracruz; the Huimanguillo experimental field is located at 1 km of Huimanguillo-Cárdenas road, Huimanguillo, Tabasco. The isolation was carried out at the phytopathology laboratory of the parasitology department of the Universidad Autónoma Agraria Antonio Narro. At this stage, we used plant tissues with infection symptoms such as pycnidia and black rot. These were cut into small pieces and disinfested with 1.5 % sodium hypochlorite in water (v:v) for 2 min. Then washed 1 min with sterile distilled water by triplicate, dried on absorbent paper in a laminar flow hood. Four tissue pieces from each sample were placed on Petri dishes with PDA medium (^{Contreras, 2016}). The samples were placed located at cardinal points within the same dishs. The plates were incubated 7 d at 28±1 °C, re-sowing was performed by monosporic culture. The strains were identified by macroscopic and microscopic morphological comparison based on color, colony form, mycelial growth, sporulation and shape of conidia (^{Alves et al., 2008}).

Jasmonic acid production: liquid fermentation

To assess the JA production by liquid fermentation we followed the methodology proposed by ^{Michelena (2001)}. 250 mL Erlenmeyer flasks with 100 mL of modified Miersch medium were autoclaved for 15 min. To inoculate the medium, we placed three 5 mm diameter fragments from the precultured mycelium (obtained from plantings in Petri dishes) on PDA at 28 °C for 3 d. The flasks were then incubated 15 d in the dark at a constant 28 °C temperature without aeration or agitation.

Growing medium

The composition of the modified Miersch medium (g L^-1) was: 50 g sucrose, KNO₃ 3 g; MgSO₄7 H₂O 0.2g; KCl 0.1g; FeSO₄. 7H₂O 0.01g; ZnSO₄. 7H₂O 0.01g; MnSO₄ 0.001g; Na₂MoO₄. 2H₂O 0.001g; CuSO₄. 5H₂O 0.001 g and 0.1 g yeast extract (^{Lorenzo et al., 2007}).

Jasmonic acid determination by HPLC

At the end of the incubation, we removed the fermentation broth from the mycelium by vacuum filtration using Whatman No. 4 filter paper. 5 mL aliquots of the filtered culture were adjusted to pH 3.0 with HCl (4 M), and from those, three extractions with ethyl acetate (1:1) in a separatory funnel were made. We then dehydrated the JA fractions with anhydrous sodium sulfate, dried by rotoevaporation at 50 °C (triplicate analysis), resuspended in 1 mL of ethyl acetate into amber flasks and kept them refrigerated at -4 °C until its assessment (^{Dathe et al., 1981}).

For the JA determination and quantification, we followed the technique described by ^{Kramell (1999)} with a Hewlett-Packard 1050 high-performance liquid chromatography equipment, in which an HP79854A ultraviolet detector (Palo Alto, California, USA) was used. The control was JA laboratory grade from SIGMA commercial house. The mobile phase was composed of methanol-water (60:40) with 1 % acetic acid (v:v). The flow was of 0.85 mL min^-1 with a Hypresil ODS column (Thermo scientific, Waltham, Massachusetts, USA) (25 cmx4.6 mmx5 µm id). Detection was set at a 295 nm wavelength.

Kinetics of JA production and analytical procedures

Once the best yield strain was selected, its production capacity was evaluated about time, after 5, 10 and 15 d. Production conditions were the same as in the previous stage. At the end of the fermentation, a rapid filtration was performed and the supernatant recovered for subsequent substrate consumption analyzes (^{Dubois et al., 1956}), pH (^{Willard et al., 1974}), biomass (^{Arnáiz et al., 2000}) and JA. The experimental design was completely random, the assessed variable was JA production capacity for each strain, the data were analyzed with ANOVA, and the means were compared by the Tukey test (p≤0.05) for which the SAS 9.2 statistical software was used.

Production of jasmonic acid by liquid fermentation

At the end of the morphological characterization process the JA production was evaluated. The resulting fermentation samples were prepared and HPLC analyzed. Among the 20 isolates eight showed JA produced in the liquid fermentation systems. The production of JA ranged from 6.5 to 552 mg L^-1. The 3B isolate had the lowest yield, while the 3C isolate showed the highest concentration, 552 mg L^-1 (Table 2). The ability of fungal microorganisms to synthesize JA has high variability among strains, even from the same species; thus, there are strains that produce JA in concentrations higher than 1000 mg L^-1, while others produce few mg L^-1 (^{Farbood et al., 2001}). In addition, Gibberella fujikuroi (^{Miersch et al., 1993}), Fusarium oxysporum (^{Cole et al., 2014}) and Pseudomonas syringae (^{Gimenez-Ibanez et al., 2016}) produce JA in small concentrations. Still, since the first report of JA microbial production (^{Broadbent et al., 1968}), so far B. theobromae has shown that has a greater production potential.

Table 2 Isolates with jasmonic acid production capacity. 

* Values with different letters are statistically different (p≤0.05).

The produced JA concentration was similar or above the average range in other experiments in which they produced JA using the same liquid fermentation system, in culture media containing high concentrations of carbohydrates (^{Dhandhukia and Thakkar, 2007}; ^{Eng et al., 2008}).

Statistical analysis showed that strains 3C, 11E and 2A were different (p≤0.05) to the other strains and showed the highest JA yield. Therefore, we used strain 3C for the following experiments related to production kinetics.

Jasmonic acid production kinetics

The variables concerning time were: substrate consumption, pH, biomass production and JA production. By the end of day five, the microorganism consumed about 60 % of the added carbohydrates. The pH continuously decreased from 7 to 4.1 the first 5 d of fermentation. This change was due to metabolites accumulation resulting from the substrate degradation process (^{Marero et al., 1997}). The results of the biomass production show a classical microbial growth curve: during the first 5 d of the bioprocess presented an exponential growth phase and from day ten on a stationary phase. In this study the microbial death stage was not assessed because the experiment did not reach the necessary time for it. The results of the JA evaluation show that from day 10, metabolite production starts at 35.45±1.2 ppm, and 5 d latter dramatically increases to 564.99±2.1 ppm. The observed behavior was similar to that reported by ^{Pinakin (2007)}, who performed a kinetic study of several strains of B. theobromae and obtained a 50-100 mg L^-1 production. This shows that the microorganism can produce JA as a secondary metabolite in a liquid fermentation bioprocess. Also, ^{Castillo (2014)} quantified and identified various hormones in a fermented broth from three B. theobromae strains which production levels were close to 100 ppm (Figure 2).

Figure 2 Kinetics of jasmonic acid production by liquid fermentation. Evaluated variables: pH, substrate consumption, biomass production and production of jasmonic acid in modified Miersch medium.