Interactive Inhibition of Aflatoxigenic Aspergillus flavus and Ochratoxigenic Aspergillus carbonarius by Aspergillus oryzae under Fluctuating Temperatures

This study aimed to evaluate the effectiveness of A. oryzae in inhibiting aflatoxin B1 (AFB1) and ochratoxin A (OTA) production by A. flavus and A. carbonarius, respectively, under shifting temperatures. A. oryzae was tested on different agar, namely coconut cream agar (CCA) and chilibased agar to figure out the variation in the effectiveness of A. oryzae on the most appropriate medium for A. flavus and A. carbonarius to produce mycotoxin and under natural condition where they are predominantly found. On CCA, the temperatures applied were 20, 30, 35, 40, 20/30, 20/35, and 20/40 ◦C, while on chili-based agar, the temperatures imposed were 20, 40, and 20/40 ◦C, at varied water activity of 0.92 and 0.97aw. The findings indicated that A. oryzae was much more effective in inhibiting the growth of A. flavus rather than A. carbonarius, yet it was able to inhibit higher OTA concentration than AFB1 at fluctuating temperatures on CCA as the most appropriate medium for A. flavus and A. carbonarius. A. oryzae effectively inhibited AFB1 and OTA at static temperature of 20 ◦C and water activity of 0.97aw on chili-based agar. Under fluctuating temperatures (20/40 ◦C), A. oryzae was also able to control mycotoxin, particularly OTA at high water activity (0.97aw).


Introduction
Mycotoxins are low-molecular-weight natural products produced as secondary metabolites by filamentous fungi, and they have toxic effects on humans and animals [1]. Aflatoxin B 1 (AFB 1 ) and ochratoxin A (OTA) are primary mycotoxins studied as they are found mainly in crops. Aflatoxins mostly contaminate cereals, oilseeds, nuts, dried fruits, spices, and beans, while OTA is found in grapes and coffee beans [2,3]. Aflatoxins are difuranocoumarin derivatives produced by a polyketide pathway by many strains of Aspergillus flavus and A. parasiticus. Meanwhile, OTA is predominantly produced by A. ochraceus, A. niger, A. carbonarius, Penicillium verrucosum, and P. nordicum [4][5][6][7][8].
Chili is one of the ubiquitous crops known for its mycotoxin contamination. Previous studies have demonstrated the occurrence of aflatoxins and OTA in chili from several countries, such as Indonesia, Malaysia, Pakistan, Bangladesh, Qatar, Saudi Arabia, and Thailand. In addition, mostly isolated fungi from chili include A. parasiticus, A. flavus, A. nomius, and A. niger [9][10][11][12][13][14][15].
Among fungi that have been isolated from chili, A. oryzae is also found as a nontoxigenic fungi [16,17]. The strains of this species have various mutations within aflatoxin biosynthetic gene clusters, resulting in their inability to produce aflatoxins; in particular, the aflR gene is absent or significantly different in some A. oryzae compared to A. flavus.
Both the species are typically found in cereal, soil, or koji fermentations used for miso, sake, and other Japanese and Korean fermented products [18,19]. A. oryzae was used as a biocontrol agent for controlling mycotoxin producing fungi. In addition, it is known as a Generally Recognized as Safe (GRAS) microorganism [20,21]. Previous studies showed that A. oryzae reduced aflatoxin [22][23][24]. However, there are few studies on the optimum environmental conditions for A. oryzae to reduce aflatoxin and OTA production.
Detoxifications of mycotoxins can be performed by physical, chemical, and biological methods. However, the biological method remains preferable due to safety issues, possible losses in nutritional contents, limited efficacy, and cost implications in physical and chemical controls [25]. Several fungi have succeeded in controlling AFB 1 and OTA contamination, such as atoxigenic A. flavus and A. parasiticus, Trichoderma sp., Phoma sp., Rhizopus sp., Sporotrichum sp., Alternaria sp., and A. pullulans [26][27][28]. The environmental factors, including temperature and water activity, also affect their effectiveness. For instance, the maximum activity of protease produced by Trichoderma harzianum to control toxigenic fungi was obtained at a temperature of 40 • C [29]. Debaryomyces hansenii could stimulate aflatoxins production by A. parasiticus at 0.99a w , whereas a significant reduction of aflatoxins was observed at 0.92a w [30].
Although fungal competition have been intensively studied, most of them were carried out under static temperatures [31][32][33]. Meanwhile, fluctuating temperatures may occur in the environment due to climatic changes, and it is able to increase mycotoxin formations, such as deoxynivalenol and zearalenone by Fusarium graminearum [34], aflatoxin by A. parasiticus [35,36], and OTA by A. ochraceus [37]. In this study, we aimed to understand interaction of fungi isolated from chili and study the effectiveness of A. oryzae to inhibit AFB 1 and OTA under fluctuating temperatures.

Fungal Strains
There were three fungal strains used in this study, Aspergillus flavus strain CH141, A. carbonarius strain CH112, and A. oryzae strain CH084. They were isolated from chili in Thailand and stored as culture collections of Kasetsart University Research and Development Institute (Bangkok, Thailand). All the isolates of A. flavus and A. carbonarius have been proven to produce AFB 1 and OTA, respectively, and A. oryzae was proven to produce neither of them. The analysis was conducted using thin-layer chromatography (TLC) coupled with densitometry following extraction with the protocol described by Chuaysrinule, Maneeboon, Roopkham and Mahakarnchanakul [15].

Mycological Media
Coconut cream agar (CCA) was used in this study to analyse mycotoxin production at different temperatures merely through UV light (366 nm). Additionally for A. flavus, it was also a suitable medium for A. carbonarius in producing OTA, similar to other prominent synthetic media such as potato dextrose agar (data not shown). CCA was prepared by mixing coconut cream and water with a ratio of 50:50 prior to the addition of agar 1.5% and sterilization using autoclave at 121 • C for 15 min [38].
Meanwhile, chili powder agar was used to provide the fungi with similar living conditions as in chili. Furthermore, it was preferred rather than real chili due to its efficiency in adjusting the moisture content and controlling the desired conditions. Chili powder agar 2% was made with 20 g of chili powder, 20 g of agar and 1 L of distilled water, as detailed by Marín, et al. [39]. The medium was modified with 260.95 and 134.58 g of glycerol per liter of medium to obtain 0.92 and 0.97 a w, respectively. The values 0.92 and 0.97 a w were within the range of optimum condition for AFB 1 and OTA production; moreover, 0.97 a w was the highest water activity that could be achieved during trials in obtaining desired a w levels. All autoclaved media were poured in 9-cm petri dishes, and final a w levels were measured after solidification. The water activity of all media was determined using a water activity meter (AquaLab 4TE, Decagon Devices, Pullman, Washington, DC, USA).

Inoculum Preparation
Initially, each strain was inoculated on potato dextrose agar (PDA) and incubated at 25 • C for 7 days. A spore suspension was made by adding 0.1% (v/v) tween 20 in distilled water. The spore was harvested by addition of 5 mL of suspension previously prepared suspension. Subsequently, the spore suspension was counted using a hemocytometer and diluted to achieve a spore concentration of 10 5 spores/mL. In mixed culture experiments, a spore suspension of A. flavus or A. carbonarius was mixed with an equal volume of spore suspension of A. oryzae. In single cultures, spore suspension of each A. flavus, A. carbonarius, and A. oryzae was mixed with an equal volume of sterilized water.

Inoculation and Incubation
On CCA, 20-µL of spore suspension of single and mixed cultures, prepared as previously reported, were inoculated on the center of agar to achieve the final concentration of 10 3 spores/mL. Cultures inoculated on CCA were incubated at the static temperatures of 20, 30, 35, and 40 • C for 6 days as control, and at three levels of fluctuating temperatures, 20/30, 20/35, and 20/40 • C for 6 days in total. Under fluctuating temperatures, incubation times were 3 days at each initial and final temperature to give the same condition on the mycotoxin production. Six days were chosen because this incubation was generally sufficient for A. flavus to achieve significant AFB 1 accumulation [40]. The fluctuating temperatures were made manually by changing the temperature from low to high temperature after 3 days incubation at a low temperature.
On chili-based agar, a dual culture technique was performed. The inoculation procedure on CCA and chili-based agar was differentiated because the mycelium color of the fungi widely varied on CCA, allowing the influence of one fungus on mycelium growth to be clearly observed. On the other hand, fungi on chili-based agar have similar mycelium color that might interfere in identifying the fungus effect on mycelium growth if mixed cultures technique was applied. Dual culture was carried out by inoculating 10 µL of spore suspension of each fungal species on the medium on the opposite site at a distance of 3 cm between inoculums. The single culture was performed by inoculating only spore suspension of one isolate in a similar manner. Petri plates with chili-based agar on same water activity were stored together in polyethylene plastics and incubated at (1) static temperature of 20 and 40 • C for 7 days at each temperature as controls and at (2) fluctuating temperatures from 20 • C to 40 • C with the interval times of 15 h and 9 h at each temperature, respectively, to simulate day and night time. The cycling temperature was imposed for 4 days in total, but the fluctuating temperatures were performed after pre-incubation at a static temperature of 20 • C for 3 days. The temperatures of 20, 30, 35, and 40 • C were chosen as they are within the temperature range in Thailand [41].

Growth Assessment
The growth of fungi on CCA was recorded by measuring the diameter of mycelia in two directions at right angles to each other after 6 days of incubation. The diameter of a colony (cm) was divided by total incubation time (days) to obtain fungal growth rates (cm/day). The same approach to measure the growth was also conducted for cultures on chili-based agar, but the assessment was performed daily.

AFB 1 and OTA Analysis
The fungal colony grown on CCA was extracted and quantified for AFB 1 and OTA concentrations by thin-layer chromatography (TLC) coupled with a densitometer [42]. Five agar plugs of CCA were extracted by using 1 mL of methanol and sonicated for 60 min. Following sonication, 10 µL of supernatant in samples and 5 concentrations of AFB 1 and OTA standards (5, 10, 15, 20, 30 ng/mL) were spotted in thin-layer chromatography silica gel 60 plates (Merck KGaA, Darmstadt, Germany). Those standards were prepared to construct the calibration curve. The plates were developed in the mobile phase consisting of chloroform/acetone (9:1, v/v) for AFB 1 analysis and toluene/ethyl acetate/formic acid (6:3:1, v/v/v) for OTA analysis with a migration distance of 8 cm. They were then left to air-dry for 5 min and scanned using a densitometer TLC scanner 3 (Camag, Muttenz, Switzerland) at the wavelength of 366 nm and 333 nm for AFB 1 and OTA detection, respectively. Limit of detection (LOD) and limit of quantification (LOQ) were 0.61 and 0.91 ng/spot for AFB 1 , 3.14 and 3.39 ng/spot for OTA, respectively.
Mycotoxins from cultures on chili-based agar was assessed with different methods because the mycotoxin on chili-based agar was produced in low concentrations. Cultures cut from chili-based agar were weighted and then extracted by adding 10 mL of methanol and sonication for 60 min. The extract was filtered with a 0.45-µm syringe filter and stored in a 2-mL amber vial. According to method 991.31 in AOAC [43], reverse-phase HPLC was applied and combined with a fluorescent detector. Sample eluent was injected into HPLC (2690/95, Water Corporation, Milford, MA, USA) and a symmetry ® C18 column (5 µm, 3.9 × 150 mm) (Waters Corporation) was employed. For AFB 1 detection, water/acetonitrile/methanol (60:15:25, v/v/v) was used as mobile phase at a flow rate of 1 mL/min. Column temperature was set at 40 • C. The fluorescence detection was operated using a wavelength of 365 nm and 445 nm for excitation and emission wavelengths, respectively. The post-column derivatization was conducted with a photochemical reactor for enhanced aflatoxin detection (PHRED) (Aura Industries, San Diego, CA, USA). For OTA analysis, the mobile phase was 6% (v/v) acetic acid/acetonitrile/methanol (45:35:20, v/v/v) at a flow rate of 1 mL/min and column temperature was 35 • C. The excitation and emission wavelengths were 350 nm and 470 nm, respectively. LOD and LOQ obtained were 0.18 and 0.31 ng/g for AFB 1 , 0.23 and 0.36 ng/g for OTA, respectively. The recovery ranges for AFB 1 and OTA were 81.74-99.66% and 82.35-109.25%, respectively.

Statistical Analysis
One-way analysis of variance was performed to assess significant differences between mean values, then T-Tests and Duncan Multiple Range Test with p-value < 0.05 were applied to determine the mean differences. The analysis was carried out using Statistical Package for the Social Science (SPSS) ver. 20 program (IBM Corporation, Armonk, NY, USA). All mycotoxin production data were converted by y = ln (x) before ANOVA to homogenize the variance [44]. Experiments were conducted two times with three replications.

Competitive Interaction between Fungi
The study aimed to determine the effect of competition among fungi isolated from chili. Furthermore, the most appropriate temperature for A. oryzae to reduce AFB 1 and OTA contaminations was examined. Aspergillus oryzae was used as it was isolated along with A. flavus and A. carbonarius from chili collected from a market in Thailand, as reported in Chuaysrinule, et al. [15].
Growth rates of single and mixed cultures are shown in Figure 1a. The single culture of A. flavus had the highest growth rate at 35 • C at which fungal growth rate decreased by 40.66% in mixed cultures with A. oryzae. Besides at 35 • C, the reduction of growth rates in mixed cultures of A. flavus and A. oryzae compared to single culture of A. flavus was also observed at 20 and 40 • C by 10.17% and 14.29%, respectively. At 30 • C, the growth rate of mixed cultures of A. flavus and A. oryzae was found to increase by 50.94% compared to the single culture of A. flavus. In A. carbonarius, the optimum growth rate was observed at 20 • C at which the growth rate of mixed cultures was observed to be lower by 36.70% than the single culture of A. carbonarius (Figure 1b). The growth inhibition by A. oryzae found in the current study was in agreement with the study of Alshannaq, et al. [22], which reported that cell-free culture of A. oryzae was able to inhibit conidial germination of A. flavus. They hypothesized that specific genes were involved in unknown antifungal products that could inhibit fungal growth. single culture of A. flavus. In A. carbonarius, the optimum growth rate was observed at 20 °C at which the growth rate of mixed cultures was observed to be lower by 36.70% than the single culture of A. carbonarius (Figure 1b). The growth inhibition by A. oryzae found in the current study was in agreement with the study of Alshannaq, et al. [22], which reported that cell-free culture of A. oryzae was able to inhibit conidial germination of A. flavus. They hypothesized that specific genes were involved in unknown antifungal products that could inhibit fungal growth. Moreover, the increasing growth rates of mixed cultures compared to A. carbonarius were found at static temperatures of 30, 35, and 40 °C by 18.03%, 96.77%, and 100%, respectively. The increasing growth rate in mixed cultures is possibly due to the temperature imposed, such as 30 °C, which was not an optimum condition for A. flavus   Moreover, the increasing growth rates of mixed cultures compared to A. carbonarius were found at static temperatures of 30, 35, and 40 • C by 18.03%, 96.77%, and 100%, respectively. The increasing growth rate in mixed cultures is possibly due to the temperature imposed, such as 30 • C, which was not an optimum condition for A. flavus and A. carbonarius, yet was optimum for A. oryzae instead, contributing to the increasing growth rates in mixed cultures. The current study observed that A. oryzae was optimum at the temperature of 30-35 • C. It is supported by the study of Gomi (2014), reporting that A. oryzae had optimum growth temperature at 32-36 • C (±1 • C), and it was unable to grow above 44 • C.
Under fluctuating temperatures, the optimum growth temperature of A. flavus was 20/35 • C at which mixed cultures with A. oryzae resulted in significantly decreasing rates by 30.38% (Figure 1a). In the single culture of A. carbonarius, the optimum growth rate was observed at 20/30 • C at which no significant differences in growth rates were observed between the single culture of A. carbonarius and mixed cultures of A. carbonarius and A. oryzae. The mixed cultures of A. flavus and A. carbonarius resulted in a lower growth rate compared to that of the single culture of A. flavus and A. carbonarius, except at temperature of 20/40 • C. The reduction of growth rates in the mixed cultures with A. oryzae indicated that under fluctuating temperatures, A. oryzae would be more effective to inhibit the growth of A. flavus instead of A. carbonarius. It has been widely known that fungi generally were not the effective microorganisms to inhibit A. carbonarius. Kogkaki, et al. [45] reported that grape-associated fungal strains resulted in slight inhibition for the growth of A. carbonarius, particularly A. ibericus. However, de Almeida, et al. [46] showed that fungi had a high ability to inhibit the growth of A. carbonarius. The fungi used in their study were isolated from coffee beans, and they had growth inhibition potentials exceeding 70%, such as Rhizopus oryzae, A. westerdijkiae, A. niger, A. tamari, A. fumigatus, and Aspergillus sp. The possible inhibition growth mechanism is the nutrient and space competition. Control agents would utilize the nutrient before being used by the pathogen, and they have fast-growth covering a large area Besides, A. oryzae was possibly secreting certain substances, thereby inhibiting the growth of antagonist fungi. Skouri-Gargouri, et al. [47] reported that another nontoxigenic Aspergillus species, A. clavatus, could inhibit the growth of several plant pathogens such as F. oxysporum and A. niger through secretion of an antifungal peptide.
Under static temperatures, A. flavus had optimum AFB 1 production at 30 • C at which the highest inhibition of AFB 1 production by A. oryzae was obtained by 94% (Table 1). Meanwhile, A. carbonarius had an optimum temperature at 20 • C at which lower OTA production was obtained by 45% in mixed cultures with A. oryzae, and the highest OTA inhibition by A. oryzae was found at 35 • C accounted for 95%. It indicated that A. oryzae would be effective to inhibit AFB 1 and OTA at the static temperature of 30 and 35 • C, respectively. At 40 • C, induction of AFB 1 and OTA were observed, indicating that A. oryzae somehow could induce mycotoxin production under suboptimal conditions of A. flavus and A. carbonarius to produce mycotoxin. Similar trend was observed with mixed cultures of A. flavus and A. carbonarius with reduction of AFB 1 at all temperatures except at 40 • C at which AFB 1 production increased, and the concentration was higher than that in the mixed cultures of A. flavus and A. oryzae. It concluded that A. oryzae was significantly able to reduce AFB1 in wider range of temperature than A. carbonarius. Furthermore, the competition of nutrition and space with A. carbonarius was not sufficient to reduce AFB 1 production from A. flavus. Another mechanism was required to eliminate AFB 1 .  In the current study, A. oryzae was able to reduce AFB 1 and OTA contamination. This result is in agreement with the study by Lee et al. [23], who reported that A. oryzae isolated from Meju, a traditional Korean fermented soybean starter, has the potential to control AFB 1 in foods and feeds. Furthermore, Alshannaq, et al. [22] showed similar results, in which A. oryzae strains and A. oryzae cell-free PDB fermentate could control A. flavus growth and aflatoxin contamination in peanuts. The ability of A. oryzae to reduce AFB 1 was possibly due to several reasons, including (1) superior abilities of A. oryzae to sequester nutrients, (2) production of metabolites that might inhibit AFB 1 production, (3) non-protein substances that naturally occurred in fermentate, and (4) specific genes involved in unknown antifungal products inhibiting growth and mycotoxin production in A. flavus. In OTA, the reduction by A. oryzae could be caused by several factors, including (1) growth limitation of A. carbonarius by co-culture with A. oryzae leading to reduction of OTA production, (2) consumption by A. oryzae, the nutrients required to synthesize OTA, and (3) OTA degradation by A. oryzae. Several fungi could degrade OTA due to carboxypeptidase A activity such as A. niger, R. oryzae, Alternaria, Cladosporium, and Trichoderma [48]. At 40 • C, 20/35 • C, and 20/40 • C, the production of AFB 1 was higher in mixed cultures than in the single culture of A. flavus. In OTA, the rise in production was observed at 40 • C. Despite the non-significant increase in concentrations, it was still worth noting to investigate. The increase of mycotoxin production was presumably due to competition with A. oryzae and temperature stress. A. oryzae caused A. flavus and A. carbonarius to produce AFB 1 and OTA as their defense mechanisms against the threat. Fox, et al. [49], Magan, et al. [50] suggested that mycotoxins be able to act as chemical signals in communication, a competitive weapon to defend the habitat, or to inhibit the growth of competitors in the same niche. Furthermore, the presence of competitors is essential for mycotoxin production. It is supported by Horn, et al. [51], who reported that serial transfers on culture media within laboratory conditions, in which they lacked interaction among fungi and natural stressed conditions, resulted in the loss of aflatoxins production in A. flavus. Temperature stress also contributed to the increased production of AFB 1 and OTA because it can activate the mycotoxin biosynthesis genes [52,53].
Overall, A. oryzae seems to be appropriate fungi to inhibit mycotoxigenic fungi and eliminate AFB 1 and OTA. Moreover, it was much more effective to inhibit the growth of A. flavus instead of A. carbonarius both at static and fluctuating temperatures on CCA. In terms of mycotoxin inhibition at static temperatures, the effectiveness of A. oryzae in inhibiting AFB 1 and OTA production was based on temperature. The temperatures of 30 and 35 • C were optimum conditions for AFB 1 and OTA inhibition by A. oryzae, respectively. Under fluctuating temperatures, A. oryzae was much more effective in inhibiting OTA instead of AFB 1 as reduction of OTA was observed under all fluctuating temperature programs. In AFB 1 , the production was only reduced at 20/30 • C.

Interaction between Toxigenic and Non-Toxigenic Species on Chili-Based Agar
Since all the fungi used were isolated from chili, this part was conducted to determine the effectiveness of A. oryzae in inhibiting AFB 1 and OTA on chili-based agar. In the current part of the study, dual cultures were carried out by inoculating spore suspension of two fungi in two different areas on the same agar medium; thus, toxigenic fungi (A. flavus or A. carbonarius) and non-toxigenic fungi (A. oryzae) each have growth diameter. The fluctuating temperatures were performed by manually changing temperatures from 20 • C to 40 • C. Each temperature was imposed for 15 h and 9 h, respectively, for 4 days after pre-incubation at 20 • C for 3 days. The media was used with 2 a w levels, 0.92 and 0.97 which were selected because water activity of more than 0.90 resulted in high production of AFB 1 and OTA [15,54,55].
The growth of A. flavus at the end of incubation time was observed to be higher at 20 • C than at 40 • C (Figure 2a). By the presence of A. oryzae, the growth of A. flavus was not significantly different at 20 • C and 40 • C (Figure 2b). It indicated that low water activity (0.92 a w ) resulted in the inability of A. oryzae to inhibit the growth of A. flavus at both static temperatures of 20 • C and 40 • C. Meanwhile, growth of A. flavus at 20 and 40 • C was observed lower after dual cultures with A. oryzae at 0.97 a w (Figure 3). It showed that under static temperature, particularly temperatures of 20 and 40 • C, A. oryzae can inhibit the growth of A. flavus at water activity more than 0.92 a w , specifically 0.97 a w . On the contrary, previous study reported that A. oryzae can grow on media with water activity above 0.8 [56]. According to that, A. oryzae is supposed to grow and inhibit the growth of A. flavus at 0.92 a w . This condition is presumably due to dual cultures with A. flavus, which hinder the growth phase of A. oryzae. On the contrary, previous study reported that A. oryzae can grow on media with water activity above 0.8 [56]. According to that, A. oryzae is supposed to grow and inhibit the growth of A. flavus at 0.92 aw. This condition is presumably due to dual cultures with A. flavus, which hinder the growth phase of A. oryzae.    In A. carbonarius, no growth was observed at static temperatures of 40 °C, at 0.97 aw and 0.92 aw. At 20 °C and 0.92 aw, there were no significant differences observed between the growth of A. carbonarius in single and dual cultures with A. oryzae (Figure 4a). By increasing the water activity to 0.97 aw, A. carbonarius in single and dual cultures was also increasing at the same temperature. Under such condition, A. carbonarius in single culture was observed to be higher than in dual culture (Figure 4b). It indicated that A. oryzae would be adequate to inhibit the growth of A. carbonarius at a high water activity (0.97 aw). In A. carbonarius, no growth was observed at static temperatures of 40 • C, at 0.97 a w and 0.92 a w. At 20 • C and 0.92 a w , there were no significant differences observed between the growth of A. carbonarius in single and dual cultures with A. oryzae (Figure 4a). By increasing the water activity to 0.97 a w , A. carbonarius in single and dual cultures was also increasing at the same temperature. Under such condition, A. carbonarius in single culture was observed to be higher than in dual culture (Figure 4b). It indicated that A. oryzae would be adequate to inhibit the growth of A. carbonarius at a high water activity (0.97 a w ).  No toxin was produced under static temperatures on the fourth day of incubation as no AFB1 and OTA were detected. After incubation for 7 days, AFB1 was detected at 20 °C and 40 °C (Table 3). At 20 °C, AFB1 was detected only in the single culture of A. flavus at water activity of 0.97 aw, while no AFB1 was observed in dual cultures. It resulted in inhibition of AFB1 by A. oryzae (100%). At 40 °C, AFB1 was detected at 0.92 aw and 0.97 aw, at which AFB1 concentrations were observed to be lower by 9.37% and 14.43%, respectively, in dual cultures compared to those in the single culture. Meanwhile, OTA was only detected at a static temperature of 20 °C, at 0.92 and 0.97 aw. Under those conditions, the concentrations of OTA were observed to be lower in dual cultures with A. oryzae in comparison to single culture of A. carbonarius. At 20 °C with the water activity of 0.92 and 0.97 aw, the reduction of OTA in dual cultures was observed by 43.33% and 68.42%, respectively. Lower concentrations of AFB1 and OTA after dual cultures with A. oryzae indicated that A. oryzae could reduce AFB1 and OTA, either by inhibition of mycotoxin production or degradation of mycotoxin. A. oryzae would be optimal to inhibit AFB1 and OTA at low temperature and high water activity (20 °C, 0.97 aw).  No toxin was produced under static temperatures on the fourth day of incubation as no AFB 1 and OTA were detected. After incubation for 7 days, AFB 1 was detected at 20 • C and 40 • C (Table 3). At 20 • C, AFB 1 was detected only in the single culture of A. flavus at water activity of 0.97 a w , while no AFB 1 was observed in dual cultures. It resulted in inhibition of AFB 1 by A. oryzae (100%). At 40 • C, AFB 1 was detected at 0.92 a w and 0.97 a w , at which AFB 1 concentrations were observed to be lower by 9.37% and 14.43%, respectively, in dual cultures compared to those in the single culture. Meanwhile, OTA was only detected at a static temperature of 20 • C, at 0.92 and 0.97 a w . Under those conditions, the concentrations of OTA were observed to be lower in dual cultures with A. oryzae in comparison to single culture of A. carbonarius. At 20 • C with the water activity of 0.92 and 0.97 a w , the reduction of OTA in dual cultures was observed by 43.33% and 68.42%, respectively. Lower concentrations of AFB 1 and OTA after dual cultures with A. oryzae indicated that A. oryzae could reduce AFB 1 and OTA, either by inhibition of mycotoxin production or degradation of mycotoxin. A. oryzae would be optimal to inhibit AFB 1 and OTA at low temperature and high water activity (20 • C, 0.97 a w ). Under fluctuating temperatures of 20/40 • C with 0.92 a w , there were no significant differences between the growth of A. flavus in single and dual cultures during incubation (Figure 5a). It was also observed at water activity of 0.97 (Figure 5b), indicating that fluctuating temperatures would inhibit A. oryzae from controlling the growth of A. flavus even after the water activity had been raised to 0.97 a w . However, higher growth of A. flavus in dual cultures was observed at 0.97 a w than at 0.92 a w at the end of incubationday, which indicated that high water activity was favorable for the growth of A. flavus under fluctuating temperatures.
In A. carbonarius, no growth was observed in fluctuating temperatures of 20/40 • C and 0.92 a w. At 20/40 • C and 0.97 a w , the growth of A. carbonarius in dual cultures was observed to be higher than that in single culture (Figure 6). At the end of incubation, the growth of A. carbonarius in single and dual cultures under fluctuating temperatures was observed to be lower than their growth under static temperature (20 • C and 0.97 a w , Figure 5b). It indicated that fluctuating temperatures could cause stress conditions in A. carbonarius. Hence, the growth was resisted. However, the presence of A. oryzae could help A. carbonarius grow despite the stress-inducing situation, considering higher growth of A. carbonarius that was observed in dual cultures than in single culture under fluctuating temperatures. Furthermore, colonies of A. carbonarius started to be visible after the second day of incubation at static temperature, while their colonies were visible on the fourth day of incubation under fluctuating temperatures. It indicated that fluctuating temperatures caused A. carbonarius in both single and dual cultures to grow slower compared to the growth under static temperature of 20 • C. differences between the growth of A. flavus in single and dual cultures during incuba (Figure 5a). It was also observed at water activity of 0.97 (Figure 5b), indicating fluctuating temperatures would inhibit A. oryzae from controlling the growth of A. fl even after the water activity had been raised to 0.97 aw. However, higher growth o flavus in dual cultures was observed at 0.97 aw than at 0.92 aw at the end of incubationwhich indicated that high water activity was favorable for the growth of A. flavus un fluctuating temperatures. In A. carbonarius, no growth was observed in fluctuating temperatures of 20/4 and 0.92 aw. At 20/40 °C and 0.97 aw, the growth of A. carbonarius in dual cultures observed to be higher than that in single culture (Figure 6). At the end of incubation growth of A. carbonarius in single and dual cultures under fluctuating temperatures observed to be lower than their growth under static temperature (20 °C and 0.97 aw, Fig  5b). It indicated that fluctuating temperatures could cause stress conditions in carbonarius. Hence, the growth was resisted. However, the presence of A. oryzae could A. carbonarius grow despite the stress-inducing situation, considering higher growth o carbonarius that was observed in dual cultures than in single culture under fluctua temperatures. Furthermore, colonies of A. carbonarius started to be visible after the sec day of incubation at static temperature, while their colonies were visible on the fourth of incubation under fluctuating temperatures. It indicated that fluctuating temperat caused A. carbonarius in both single and dual cultures to grow slower compared to growth under static temperature of 20 °C. Under fluctuating temperatures, production of AFB1 at 0.92 aw was induced with the presence of A. oryzae (Table 4). However, the production at 0.97 aw was inhibited instead. It indicated that under fluctuating temperatures, the ability of A. oryzae to inhibit mycotoxin would be affected by water activity. With the current fluctuating temperature scenario, A. oryzae could not inhibit AFB1 at water activity of 0.92 aw or less than 0.97 aw. It is in agreement with the study by Peromingo, et al. [30], which demonstrated that Debaryomyces hansenii could stimulate aflatoxins production by A. parasiticus at 0.99 aw, whereas a significant reduction of aflatoxins was observed at 0.92 aw. In a study by La Penna, et al. [57], Kluyveromyces spp. were able to completely inhibit aflatoxin production of A. flavus and A. parasiticus at high aw (0.994 and 0.982), but they could increase aflatoxin levels with certain Aspergillus at low aw (0.955 and 0.937). In OTA, fluctuating temperatures caused A. oryzae to inhibit the production at 0.97 aw by 42.23%.  Under fluctuating temperatures, production of AFB 1 at 0.92 a w was induced with the presence of A. oryzae (Table 4). However, the production at 0.97 a w was inhibited instead. It indicated that under fluctuating temperatures, the ability of A. oryzae to inhibit mycotoxin would be affected by water activity. With the current fluctuating temperature scenario, A. oryzae could not inhibit AFB 1 at water activity of 0.92 a w or less than 0.97 a w . It is in agreement with the study by Peromingo, et al. [30], which demonstrated that Debaryomyces hansenii could stimulate aflatoxins production by A. parasiticus at 0.99 a w , whereas a significant reduction of aflatoxins was observed at 0.92 a w . In a study by La Penna, et al. [57], Kluyveromyces spp. were able to completely inhibit aflatoxin production of A. flavus and A. parasiticus at high a w (0.994 and 0.982), but they could increase aflatoxin levels with certain Aspergillus at low a w (0.955 and 0.937). In OTA, fluctuating temperatures caused A. oryzae to inhibit the production at 0.97 a w by 42.23%. a,b Different superscript lowercase letters denote significant differences between strains within the same temperature and a w (p < 0.05). NA, not analyzed since no growth was observed. ND, not detected. * Plus (+) or minus (−) indicates either induction or inhibition of mycotoxin, respectively, after dual cultures with A. oryzae on 7-day incubation.

Conclusions
A. oryzae was optimum to reduce AFB 1 and OTA contamination in crops at high water activity and low temperature (0.97 a w, 20 • C). In some cases in which the fluctuating temperatures could not be avoided during storage or shipment of crops, A. oryzae was still able to inhibit AFB 1 and OTA production despite decreasing inhibition rate in comparison to static temperatures. However, the current study suggested that it was used to inhibit OTA instead of AFB 1 . Further studies could address a greater variety of fungi and mycotoxin that A. oryzae could degrade under various abiotic stress that possibly occurred during postharvest to have a comprehensive study for the degradation of mycotoxins.