Antifungal activity of vanilla juice and vanillin against Alternaria alternata

ABSTRACT Vanilla juice has been shown empirically to have antifungal activity against some fungal strains; however, there are no activity reported against Alternaria genre. In this work, the chemical profile of vanilla juice was obtained and its antifungal activity against fungal strains from the family Pleosporaceae, isolated from sorghum- and barley-diseased plants, was tested. The strains were identified as Alternaria alternata by their molecular and morphological characteristics. The vanilla juice characterization from Vanilla planifolia pods showed the presence of vanillin, vanillic acid, p-hydroxybenzaldehyde, p-hydroxybenzoic acid, guaiacol, glucovanillin, vanillyl alcohol, and furfural. Vanilla juice showed a fungistatic effect against all A. alternata strains tested in this study and increased the lag time from 50 to 112 h, and no conidia were produced. This result indicates the possible application of vanilla juice as an alternative to control agricultural crops such as barley and sorghum in Mexico.


Fungal cultures
A. alternata fungal strains used in this study were isolated from diseased plants of sorghum (Sorghum sp.) and barley (Hordeum vulgare), collected at different times (Table 1). To isolate fungi, some infected sorghum fragments and barley plants were placed in a moist chamber and incubated in darkness at 25°C for 3 days. After that, each fungus was isolated and subcultured on potato dextrose agar, PDA (Sigma, St. Louis, MO), at room temperature for 7 days. Fungal strains were maintained in PDA medium at 4°C and the spore suspension at −87°C.

Molecular and morphological characterization
Strains isolated (JCP13, JCP25, JCP31, JCP32, JCP49, JCP56) were characterized molecularly and the sequences were deposited in the GenBank. Strain ITV3 was donated and characterized in the genetics laboratory from the Instituto Tecnológico de Veracruz (ITVer). Morphological descriptions are based on comparisons of A. alternata (Ariyawansa et al., 2015;Lawrence, Rodonto, & Gannibal, 2016) augmented by new observations, as noted. The standard medium used to assess morphology and growth rate was PDA. The morphology, colony color, and diffusing pigment in the agar were recorded. Twenty-five measurements of conidia and conidiophores were made. Measurements of the characters were taken from images using the software IOS 7, 8-megapixel iSight camera, iPhone 5S version Microsystems (Hollyhill, Cork, Republic of Ireland). Scanning electron microscope micrographs were taken with the Cryo Transfer System (Jeol, model IT300) at the Escuela Superior de Apan (ESAp-UAEH).

Vanilla juice extraction
Vanilla juice was obtained from the sweating stage during the curing process of vanilla in Gutierrez Zamora, stated of Veracruz, Mexico. Vanilla juice was collected in sterile bottles 2 days after initiating the sweating cycles from a batch containing 1000 kg of vanilla pods. Samples were kept on ice during their transport to the laboratory.

pH measurements
The pH measurements were performed in the vanilla juice using a potentiometer (pH 510 Series Benchtop Meter Oakton model 00702-93) with a magnetic agitator. The measurements were performed in triplicate.

Chromatographic profile
The chromatographic profile was performed according to the technique proposed by Pérez-Silva et al. (2006), using an HPLC (Varian ProStar model 240) equipped with a UV detector (Waters model 2487), 230 nm wavelength, and a C18 column (Microsorb TM-MV). The mobile phase was a mixture of methanol-acidified water that was prepared as follows: the HPLC grade water (J.T. Baker) was acidified with 0.1 mol L −1 phosphoric acid (HYCEL), filtered through a 0.45-μm membrane, stirred for 5 min, and then sonicated (West prime Systems) during 30 min. HPLC grade methanol (J.T. Baker) was subjected to the same treatment. For the whole test, a flow of 0.7 mL min −1 was used.
Vanilla juice samples were diluted with the mobile phase and filtered through a membrane (0.22-μm) before being injected into the HPLC. To determine glucovanillin, vanillyl alcohol, p-hydroxybenzoic acid, p-hydroxybenzaldehyde, and vanillic acid a column was used, with temperature at 35°C, a ratio phase (methanol-acidified water, 78:22, v/v), and a pressure of 5.269 × 10 6 Pa. Vanillin and guaiacol were identified using a column temperature at 30°C. A ramp to the mobile phase was applied also, starting with a proportion of 78:22 decreasing 3 mL every 2 min until reaching a proportion of 65:35. The pressure ranged from 7.295 × 10 6 Pa to 9.727 × 10 6 Pa. Furfural was determined using a column temperature at 30°C, an initial phase with a ratio of 90:10, and a decreasing ramp of 2 mL every 3 min to reach a ratio 78:22. Pressure ranged from 5.269 × 10 6 to 5.472 × 10 6 Pa. Calibration curves were performed with an external control using: vanillic acid (VA; Fluka), vanillyl alcohol (VAL; Fluka), guaiacol (GYL; Fluka), vanillin (V; Sigma), p-hydroxybenzoic acid (BHA; Fluka), p-hydroxybenzaldehyde (PHB; Fluka), and furfural (FUR; Sigma-Aldrich). Glucovanillin (GLU) was extracted with the method reported by Odoux (2000) and was donated by the CIRAD Institute (France).

In vitro antifungal assays
The antifungal effect of vanilla juice and commercial-grade vanillin was evaluated in vitro on PDA medium. A volume of 2 µL with 2.0 × 10 6 spores was deposited in the center of each Petri dish, containing PDA medium as a control, PDA with vanilla juice (adjusted to 250 mg L −1 of vanillin), and PDA with commercial-grade vanillin. Experiments were performed at a temperature of 25°C during 7 days. The pure vanilla (Sigma) was used at different concentrations (250,500,750,1000,1250, and 1500 mg/L). Radial growth (mm) was measured every 24 h during 8 days. The tests were performed in triplicate and the results were analyzed using the STATISTICA software (Hill & Lewicki, 2007).
In this study, we isolated seven fungal strains from diseased plants; six strains were isolated from barley (JCP13, JCP25, JCP31, JCP32, JCP49, and JCP56) and one strain from sorghum plants (ITV3). Colonies in PDA medium showed rapid growth, the fungi reached a diameter of 37-40 mm in 7 days, at 25°C (Figure 1(a)). The initial growth was hairy, with a gray color that later changed in the center of the colony acquiring a darker tone, with more or less intense tones but still with gray borders. The reverse of the colonies showed a black color (Figure 1(b)). Hyphae had a filamentous shape with simple conidiophores, septa, from their end muriform conidia were formed, of a grayish brown color, and transverse and vertical multisepta in an irregular arrangement (Figure 1(c)). The new conidia are formed by gemmation from the apical cell, giving rise to a long chain of more than 10 conidia (Figure 1(d,e)). The primary conidiophores can be curved or straight, from short to very large in size, simple or branched with one or more terminal conidiogenous loci.
The results observed in this study are within the range of observed values for other A. alternata strains reported by other authors (Ariyawansa et al., 2015;Lawrence et al., 2016) and lower than those reported by (Nagrale, Gaikwad, & Sharma, 2013;Ramjegathesh & Ebenezar, 2012) for the maximum length of conidia.

pH measurement
Vanilla juice had similar organoleptic characteristics to the vanilla extract, with a predominantly acidic pH, in a range between 4.4 and 5.6. The low pH was associated with the organic acids present, such as acetic, propionic, butyric and isobutyric, propanoic acids, and others (Pérez-Silva et al., 2006), the maturity degree of vanilla beans, the sweating cycles (8 to 13), and the harvest time of the juice (Rosado, 2006). This pH result corresponds to the pH reported by Pacheco (2009) for green vanilla extract and vanilla cured extract, Figure 1. Alternaria alternata colony on PDA medium (7 days, 25°C) front plate (a), morphology colony of A. alternata on PDA medium reverse plate (b), Optic microscopy of conidia (c), micrograph of hyphae and conidia (d-e). Figura 1. Colonia de Alternaria alternata en medio PDA (7 días, 25°C), anverso de la placa (a), morfología de la colonia de A. alternata en medio PDA, reverso de la placa (b), conidios vistos por microscopía óptica (c), micrografías de hifas y conidios (d-e).
5.12 ± 0.19 and 5.08 ± 0.24, respectively. Vanilla juice is a byproduct obtained from approximately a ton of vanilla beans; therefore, the juice is a concentrated product containing large amounts of the chemical compounds present in vanilla beans. The vanilla juice had an oily or viscous consistency and a brown color. The juice was collected during the first cycles of sweating (18-20 h) as inferred by the amount produced (Tapia et al., 2011). At this stage, the vanilla bean cells destruction is accelerated, which leads to the formation of vanillin and the generation of large amounts of chemical compounds (Purseglova et al., 1981;Romero, 2003;Walton, Mayer, & Narbad, 2003). During the sweating phase, temperatures reached up to 65°C, which may facilitate the release of simple aromatic compounds and other polymers such as waxes, resins, gums, and essential oils (Ranadive, 1994;Rosado, 2006).

Chemical compounds in vanilla juice
Chromatographic characterization showed the same profile as the cured vanilla bean extract obtained from a traditional curing process (Figure 3(a)). The chemical concentration of the compounds in vanilla juice was evident in the chromatograms and was higher than those reported in the cured vanilla (Figure 3(b)). Furfural, guaiacol, and vanillyl alcohol compounds have been reported as traces in the cured vanilla extract (Pérez-Silva et al., 2006), and their presence in the vanilla juice is due to the amount produced (5 L ton −1 ; personal communication), therefore, all compounds are concentrated. The juices from two batches (fourth and sixth) showed the highest vanillin concentrations whereas the juices from the other three batches (first, second, and sixth) had the greatest glucovanillin concentration. Moreover, the vanillic acid and p-hydroxybenzoic acid and p-hydroxybenzaldehyde showed no significant differences.
Chemical compounds identified in the juice are summarized in Table 2.
The presence of glucovanillin and vanillyl alcohol in the juice was very important as vanillin precursors. The results obtained show that conversion of vanillin does not reach 100% in cured vanilla beans.
The four main aromatic compounds in the juice are due to their formation during the first 24 h after the curing process had started (Havkin-Frenkel et al., 2004). Dignum, Kerler, and Verpoorte (2001b) confirmed the presence of vanillin and vanillic acid in the first sweating cycles, but the glucovanillin presence has been reported in the ninth cycle of sweating at concentrations of up to 0.08 g 100 mL −1 . Other authors reported concentrations of p-hydroxybenzoic acid between 0.019 g 100 mL −1 and 0.034 g 100 mL −1 (Gassenmeier, Riesen, & Magyar, 2008).
Vanillin concentrations ranged from 0.10 g 100 mL −1 to 0.61 g 100 mL −1 . Odoux (2000) reported that vanillin concentration increased with each sweating cycle at around 0.6 g kg −1 in dry matter (DM), in the green pod, and at a concentration of 0.2 g kg −1 DM in cured pods. Dignum et al. (2001b) reported the presence of guaiacol in the cured vanilla extract obtained during the killing stage under controlled conditions. Guaiacol showed higher concentration on the first day of the killing stage until reaching a maximum conversion on the fourth day, which explains the changes in concentrations found in this research. Farmers collected the juice during the first sweating cycles because in this step the vanilla beans lose up to 40% water (Havkin-Frenkel et al., 2004;Purseglova et al., 1981). Unlike the later sweating cycles, where there is less dehydration and the amount of juice produced is minimal (Rosado, 2006).
During the sweating stage, the temperature may reach 65°C or higher (Rosado, 2006) and can form degraded  products such as furfural (Rao & Ravishankar, 2000). The vanilla juice showed different concentrations in the five batches studied. In general, the chemical compounds identified in the juice had a higher concentration (except vanillin) than in the cured vanilla beans (Pérez-Silva et al., 2006;Rosado, 2006). During the sweating and drying stages, the development of aromatic chemical compounds increases and can be dragged with water (Dignum et al., 2001a;Rosado, 2006).

Antifungal action
The vanilla juice was adjusted to 250 mg/L of vanillin concentration and was compared to commercial-grade vanillin, results did not show antifungal effect at the concentration tested. This can be attributed to the vanillin concentration tested and the difference in the pH value at which the tests were conducted, since the pH of the juice was 4.4; whereas the PDA medium in which vanillin is contained has a pH of 5.0 (Delaquis et al., 2005). The decrease in radial growth suggests that the fungus was unable to maintain a positive tannin pressure in the mycelium to counteract the effect of the chemical compounds in the growth medium (Tijerina-Ram írez, Lira-Méndez, Moreno-Medina, González-Prieto, & Mayek-Pérez, 2014) by limiting the availability of free water for growth (Harris, 1981). A. alternata is a fast-growing fungus of wet environments (Pontón, Moragues, Gené, Guarro,    , 2002). The effect of vanillin depends on the fungal species, solute molecule, and the incubation temperature (Olaya & Abawi, 1996). Ramírez, Chulze, and Magan (2004) report that when they exposed a strain of Fusarium graminearum to limited water availability, the strain used more energy to balance the water potential of the cytoplasm with respect to the external environment, altering the physiology of the organism and reflecting the decreased metabolism and growth.
These results indicate the possible application of vanilla juice as an alternative for the control of agricultural crops, such as tomatoes, that are attacked by this phytopathogenic fungus (Sanchez, Bautista, & Castillo, 2007). Until now, there are no reports on commercial-grade vanillin or vanilla extract effect against A. alternata, but there are reports on some natural extracts that inhibit the growth of some strains of Alternaria spp. (Mironas, Pokrib, Matescu, Boca, & Jurcoane, 2004). Furthermore, other authors report the antimicrobial activity of natural extracts obtained from three kinds of mangoes against A. alternata, reporting a reduction of 89.78% at a concentration of 6.25 mg mL −1 (Vega et al., 2013).

Antifungal vanillin activity
The 250 mg/L vanillin concentration decreased the growth of the Alternaria strains tested to 0.2 mm h −1 (Figure 4); an increase in the lag time was observed from 50 to 112 h, and a diminished mycelial growth inhibition of up to 37.5%. Moreover, when a vanillin concentration of 500 mg/L at 30°C and pH 5.0 was used, the A. alternata strain showed higher sensitivity by increasing the lag time to 192 h (Figure 4). At concentrations of 750 mg/L of vanillin, a fungicidal effect against A. alternata was observed. The vanillin concentrations of 1000, 1250, and 1500 mg/L showed the same behavior described at 750 mg/L. This result was very interesting because it is known that A. alternata produces melanin, which is a pigment that gives some resistance to fungal enzyme mixtures, such as chitinases and glucanases (Bloomfield & Alexander, 1967).
The vanillin concentration used in this work was less than that reported by Lopez-Malo, Alzadora, and Argaiz (1997) Figura 4. Efecto de la concentración de vainillina de grado comercial sobre la tasa de crecimiento radial de la cepas de A. alternata en tres tiempos diferentes.
against A. ochraceus strain, this effect was achieved at pH 3 and ≤25°C or pH 4 and ≤15°C. Lopez-Malo, Alzamora, and Argaiz (1998) reported that the germination time and radial growth rate of Aspergillus flavus, A. niger, A. ochraceus, and A. parasiticus was affected significantly when vanillin concentrations of 500, 750, and 1000 mg/L were used at pH 3.0 and 4.0 with a w 0.98. In addition, Delaquis et al. (2005) reported that with the increase of pH, the vanillin effect goes from bacteriostatic to bactericidal. This is because the organic antimicrobial agents, including phenolic acids, are more active in a dissociated state, because the pH is close to the pKa, explaining the bactericidal effect of vanillin (pKa 7.4) against some species of Listeria at pH 7.0.
Vanillin concentrations used in this research were based on the research reported by Lopez-Malo et al. (1997), (1998). However, the results were not consistent with expectations. These results may be due to the fact that the vanillin  concentration was low for the Alternaria strains tested. Minimum inhibitory concentrations reported for certain yeasts have been 2000 and 3000 mg/L (Cerruti & Alzamora, 1996;Rivera-Carriles et al., 2004). Vanilla juice and commercial-grade vanillin effect on growth parameters are shown in Figures 5 and 6(a-c), revealing that the juice exerted a fungistatic effect against barley and sorghum A. alternata strains tested in this study and increased the lag time from 50 h to 112 h. Moreover, when A. alternata strains grown on PDA medium were observed with the scanning electron microscope, there was absence of conidia production ( Figure 6(d)) as compared to the control (Figure 6(e)). Likewise, somatic structures showed an atypical morphology ( Figure 6  (d-e)). This result corroborates the fungistatic effect of vanilla juice and commercial-grade vanillin observed in the A. alternata strains tested in this study. On the other hand, the juice showed the same chromatographic profile as that of the vanilla extract; therefore, it was inferred that chemical compounds like phenols and acids can contribute with an additive, synergistic or antagonistic effect. Some acids, such as acetic and butyric, were also detected in vanilla juice (data not shown). The vanilla juice effect showed better results than the commercial-grade vanillin against A. alternata (at the concentrations tested); it diminished the mycelial growth to 50%.

Conclusion
Vanilla juice shows a fungistatic effect against all A. alternata strains studied and increased the pathogen's lag time from 50 to 112 h. A synergistic, additive, or antagonistic effect by the other chemical compounds present in vanilla juice was observed as compared to commercial vanillin. This vanilla juice can be concentrated and could be used in programs to evaluate a large number of extracts as natural antifungal products. The information obtained in this research would be useful to obtain the four aromatic compounds reported in the cured vanilla bean (vanillin, vanillic acid, p-hydroxybezaldehyde, and p-hydroxybenzoic acid) and precursors of vanillin (glucovanillin and vanillic acid). The presence of these compounds in the vanilla juice decreases its yield concentration in the cured vanilla beans at the end of the processing. Vanilla juice could be used as an alternative option for the control of fungi responsible for diseases in plants of commercial interest in Mexico.

Disclosure statement
No potential conflict of interest was reported by the authors.