Bioactivities of Ketones Terpenes: Antifungal Effect on F. verticillioides and Repellents to Control Insect Fungal Vector, S. zeamais

Maize is one the most important staple foods in the world. However, numerous pests, such as fungal pathogens, e.g., Fusarium verticillioides, and insects, such as Sitophlilus zeamais, attack maize grains during storage. Many F. verticillioides strains produce fumonisins, one of the most important mycotoxin that causes toxic effects on human and animal health. This situation is aggravated by the insect fungal vector, Sitophlilus zeamais, which contributes to the dispersal of fungal spores, and through feeding damage, provide entry points for fungal infections. The aim of this study was to evaluate in vitro bioassays, the antifungal activity on F. verticillioides M3125 and repellent effects against S. zeamais of ketone terpenes. In addition, we performed Quantitative structure–activity relationship (Q-SAR) studies between physico-chemical properties of ketone terpenes and the antifungal effect. Thymoquinone was the most active compound against F. verticillioides (Minimum Inhibitory Concentration, MIC: 0.87) affecting the lag phase and the growth rate showing a total inhibition of growth at concentration higher than 2 mM (p < 0.05). The Q-SAR model revealed that the antifungal activity of ketone compounds is related to the electronic descriptor, Pi energy. Thymoquinone showed a strong repellent effect (−77.8 ± 8.5, p < 0.001) against S. zeamais. These findings make an important contribution to the search for new compounds to control two stored pests of maize.


Introduction
Maize is one the most important staple foods in the world with an annual production over 700 million metric tons/year [1]. Argentina is the second largest exporter of maize in the world (27 million tons) [2]. Numerous pests, such as fungal pathogens, e.g., Fusarium verticillioides, and insects, such as Sitophlilus zeamais, attack maize grains during storage (9.6% of maize grain production is lost) [3], causing substantial damage to cereals, manifested as general spoilage. Many F. verticillioides strains produce fumonisins, one of the most important mycotoxins that cause toxics effects on human and animal health [4][5][6][7]. Consequently, the occurrence of F. verticillioides and fumonisins cause great economic losses due to grain quality deterioration. Mycotoxin formation can occur in pre and post-harvest stages. However, a high production of fumonisin takes place during grain storage, when the temperature and the humidity favor fungal growth and secondary metabolite production [8]. This situation is aggravated by the insect fungal vector, Sitophlilus zeamais Motschulky (Coleoptera:Curculionidae), an important pest of the stored maize [9], which contributes to the dispersal of fungal spores and through feeding damage provides entry points for fungal infections, favoring mycotoxin production by F. verticillioides [10,11].
Synthetic compounds are used as fumigants to protect stored-products [11,12]. However, the extensive use of synthetic pesticides is associated with problems in human health and other negative effects on the environment [13]. In addition, the use of these synthetic fumigants is related to the development of resistant populations of fungi and insects. Modern strategies of biological control of pests are being investigated as alternatives to synthetic pesticides; monoterpenes found mainly as components of essential oils of plants are an alternative. Previous studies reported bioactivities (antifungal, insecticidal and repellence) of terpenes with different structural groups on stored grain pests, with ketones being the most bioactive [14][15][16][17]. However, little is known about the bioactivity of natural ketone compounds against biotrophic pathogens. The aim of this study was to evaluate the antifungal activity on F. verticillioides and repellent effects against insect fungal vector, S. zeamais of ketone terpenes using in vitro bioassays. In addition, we performed Q-SAR studies between physico-chemical properties of ketone terpenes and the antifungal effect.

Medium and Culture Conditions
The culture conditions were carried out according to Pizzolitto et al. [19]. Stock solutions of each ketone were prepared in dimethyl sulfoxide (DMSO), and the appropriate amount to obtain the concentrations of 0.25; 0.5; 0.75; 1.0; 1.5; 2.0 and 3 mM was added to the culture medium. In the control treatments the equivalent amount of DMSO was added to the culture medium.
For the evaluation of antifungal activities, experiments were performed using Czapek-Dox agar in Petri dishes (90 mm). The culture medium was autoclaved at 120 °C for 15 min before cooling at 45 °C an appropriated volume of ketones was added to the media to reach the intended concentrations mentioned above. Inoculum was prepared by growing F. verticillioides M 3125 on Czapek agar for 7 days at 25 °C to obtained heavily sporulating cultures. A conidial suspension was placed in aqueous solution, after homogenizing, the suspension was counted using a Neubauer chamber and adjusted to 10 6 conidia· mL −1 . Czapek-Dox plates were inoculated centrally with 10 µL of the spore suspension. All the experiments were repeated twice.

Growth Assessment
The antifungal activity was measured as a growth assessment. Two perpendicular diameters of the growing colonies were measured daily (mm) until the colony reached the edge of the plate (approx. 7 days). The radios of the colonies were plotted against time, and a linear regression applied in order to obtain the growth rate as the slope of the line. The effect of each ketone compound on Lag phase for growth, was defined as the time (days) in which each colony reaches 5 mm of diameter [15].
The inhibition percentage was plotted against concentration to each compound. Minimum inhibitory concentration (MIC) was defined as the lowest concentration of the ketone compounds at which no fungal growth was observed and lethal doses values (LD25) were calculated by interpolation in the above mentioned lineal regression. The p values and R 2 parameters were used to check the validity of the regression analysis.

Molecular Modeling and Calculation of Molecular Parameters
The hydrophobic, electronic, polar, steric, geometric and topological descriptors were obtained from Herrera et al. [20].

Insect Growth
S. zeamais adults were obtained from Córdoba, Argentina. The colony was maintained in the laboratory for two year without exposure to insecticides. Insect rearing was carried under controlled temperature and humidity (28 °C and 60%-70%) and light/dark 12:12 h [21] on sterilized whole maize grain in sealed containers Adult weevils of mixed sex and age were used for the experiments. Experiments were conducted under complete darkness; temperature (28 °C) and relative humidity (60%-70%) remained constant in brood chamber.

Repellency Bioassays
Behavioral responses of S. zeamais adults to individual ketones were evaluated using a two-choice olfactometer, as described by Herrera et al. [17]. Briefly, a glass tube (30 cm × 1 cm of diameter) was connected to two Erlenmeyers (250 mL). In the center (15 cm from the ends) of a glass tube a small hole was made (1 cm × 1 cm). The connections were sealed with rubber plugs and they were covered with film to prevent gas leakage. The compounds were put on a filter paper (2 cm diameter) within each flask. Twenty insects deprived of food (24 h) were introduced into olfactometer (through the hole of the glass tube). After 2 h, we recorded the location of the insects in the flaks. Insects were given a choice between a specific dose of the test compound and the solvent (n-hexane) used as a control. The experiments being carried out between 10:00 and 16:00 h. Five replicates were performed for each assay. The position of the flasks was interchanged between consecutive tests to avoid spatial asymmetries. Insects was used once and afterwards discarded. Propionic acid was used as positive control [22].
A response index (RI) was calculated by using the equation: Whereas T is number of insects responding to the treatment. C is number of insects responding to the control and Tot is the total number of insects released [23]. Negative values of RI indicate repellence to the treatment while positive ones indicate attraction.

Statistical Analysis
In the antifungal assay, the data were analyzed by one-way analysis of variance (ANOVA). The normality of the data was tested using the Shapiro-Wilk test. Comparisons between the control and treatment data sets were carried out using by the Di Rienzo, Guzmá n and Casanoves Test (DGC test) [24] with results giving p values < 0.05 being considered significant.
Multiple linear regression analyses (MLR) were calculated in order to examine the quantitative relationships between linear combinations of the dependent variable (log 1/LD25) and the predictor variables (structure and molecular properties). Molar concentrations of the LD25 values were used for the MLR analyses. In the MLR equations, N is the number of data points, r is the correlation coefficient between observed values of the dependent variable and the values calculated from the equation, and R 2 is the square of the correlation coefficient and represents the goodness of fit. The quantitative structureactivity relationship (QSAR) model was validated with the root mean square prediction error (RMSPE) obtained by the cross validation leave-one out procedure. Results with p values < 0.05 were considered significant.
The significance of the mean RI in each treatment of Repellency bioassays was evaluated by the Student's t-test for paired comparisons [25]. The most significant positive or negative mean values of RI were first analyzed by an analysis of variance and subsequently ranked by using the Duncan multiple range test (p ≤ 0.05).
All analyses were performed using the InfoStat software Professional 2010 p [26].

Antifungal Activity of Ketones
The antifungal effect of ketone compounds against F. verticillioides M 3125 was evaluated in a range of concentration between 0 and 3 mM. Thymoquinone was the most active compound followed by S-carvone and R-carvone (MIC values: 0.87 mM; 3.84 mM and 4.56 mM, respectively) and the inhibition was widely dependent upon the compound concentration. On the other hand, dihydrocarvone, camphor, menthone and α-thujone had no significant effect on F. verticillioides growth at the evaluated concentrations ( Table 1).
The effect of ketones on growth rate is shown in Figure 2A. The results revealed that this growth parameter was affected depending on the compounds and the concentration in the culture medium. The most active ketone was thymoquinone showing a total inhibition of growth at concentration higher than 2 mM (p < 0.05). Moreover, the lag phase increased significantly with the amount of thymoquinone, pulegone and R-carvone ( Figure 2B).

Molecular Modeling and Calculation of Molecular Parameters
The analysis was carried out using LD25 of thymoquinone, carvone, pulegone, dihydrocarvone, verbenone and camphor, in order to determine which physicochemical and topological characteristics (Tables 2 and 3) were better at inhibiting F. verticillioides growth. The LD25 values were calculated by interpolation in the above-mentioned lineal regression. The LD25 values of dihydrocarvone and camphor were outside of the concentration evaluated range.
Equation (1)   A second QSAR analysis was performed in order to include the ketones menthone and α-thujone, which did not show antifungal activity at the tested concentrations (>3 mM). Subsequently, we arbitrarily assumed that the LD25 of the inactive compounds was 11.89 mM (outside the concentration range tested), and these compounds were included in the new MLR analysis, Equation  The descriptors were obtained by [20]. The descriptors were obtained by [20]; * the chirality was not taken into account.
The Equation (2) is shown in Figure 4. The obtained model showed a value of prediction error (RMSPE = 18.3%) higher than in Equation (1). Also, an increase in the value of Pi energy representing a greater antifungal activity is shown above.
On the other hand, dihydrocarvone, menthone, α-thujone and thymoquinone showed direct relation between concentration and repellent insect (p < 0.05), but they did not show significant repellent effects at low concentration.

Discussion
Considering the impact of synthetic pesticides on human health and ecosystems, the use of eco-friendly compounds becomes necessary to protect stored-products. In this study, antifungal and repellent activities of nine ketones were evaluated against two stored-pests. Few studies report the antifungal activity of pure ketones and essential oils rich in these compounds against F. verticillioides [27,28]. Thymoquinone was the most active ketone against F. verticillioides having significant effect on growth parameters. Similar results were obtained by phenolic compounds [14,15,29,30].
The model obtained in Equations (1) and (2) reveal that a rise in Pi energy values increase the antifungal activity of ketone compounds against F. verticillioides M3125. This electronic descriptor was calculated by the Huckel method and is related to the localization energies for electrophilic and nucleophilic attraction in a molecule [31]. The number of double bonds present in ketones increases the values of Pi energy.
In this study, α,β-unsaturated ketones (thymoquinone, R-S carvone, pulegone and verbenone) were the most inhibitory compounds. These chemicals presented an extra double bond to the carbonyl group, therefore the Pi energy values are greater than saturated ketones. The α,β-unsaturated ketones could present keto-enol tautomerism, referring to the interconversion of the two forms involves the movement of a α-hydrogen and the shifting of bonding electrons [32][33][34][35][36]. So, these compounds might react by mechanism such as Michael reaction, etc. On the other hand, the toxicity of α,β-unsaturated ketones is dependent on the substituents at the β-carbon [31][32][33][34]. Verbenone and pulegone present alkyl substituents at β-carbon. Hence, this could explain their lowest antifungal effect than thymoquinone and R-S carvone [33].
Furthermore, the behavioral responses of S. zeamais adults to individual ketones evaluated using a two-choice olfactometer revealed that thymoquinone and thujone were the best repellents at higher concentrations, while camphor was repellent at low concentrations. In contrast, pulegone and verbenone acted as attractants at low concentrations but repellents at higher concentrations. In previous studies, some volatile compounds from fungi and grain showed attractant effects at low concentrations and repellent effects at higher concentrations against S. granarius and S. zeamais, thus, showing the ability of the species to detect and respond differentially to variations in concentration [17]. In addition, previous studies reported from our group showed high insecticidal activity of terpene ketones against S. zeamais [16,35].
Here, the antifungal effect of eight natural ketone compounds on F. verticillioides, and the repellent effect against their insect vector, S. zeamais was reported. However, the antifungal effect of the natural compounds is not necessarily related tothe anti-fumonisin effect. The mycotoxins are secondary metabolites, and their production is dependent on the stress condition. So, when the antifungal compounds are applied at sub-lethal concentrations, the fungus could reach a stress state, and mycotoxin production could be stimulated. The stimulation of fumonisin production by natural products has been previously reported [37]. Further studies will be carried out in order to determine and predict the effect of the evaluated compounds on fumonisin production.

Conclusions
The results obtained in the present work showed the capacity of thymoquinone as a potential tool for controlling F. verticillioides M3125, due to its antifungal activity, and repellent effect against the insect fungal vector, S. zeamais. In addition, the QSAR model revealed that the antifungal activity of ketone compounds could be predicted by the electronic descriptor, Pi energy, revealing the importance of α,β-unsaturation in the antifungal activity of these ketones. This study, contributes to the search of novel active compounds to control two storage pests of maize.