1760 Journal of Food Protection, Vol. 77, No. 10, 2014, Pages 1760-1767 doi: 10.4315/0362-028X.JFP-13-360 Copyright © , International Association for Food Protection

Removal of Aflatoxin B1 and Inhibition of Aspergillus flavus Growth by the Use of Lactobacillus plantarum on Olives FATEN KACHOURI,* HAMIDA KSONT1NI,

and

MOKTAR HAMDI

Laboratory of Microbial Ecology and Technology (LETMl), National Institute of Applied Sciences and Technology (1NSAT), BP: 676.1080, Tunis, Tunisia; and Superior School of Food Industry at Tunis (ES1AT), 58 Street Alain Savary, 1003, Tunis, Tunisia MS 13-360: Received 2 September 2013/Accepted 12 January 2014

ABSTRACT Olives can be contaminated with a wide variety of molds (Aspergillus and/or Penicillium) that can be occurring naturally on fresh and processed olives and could support mycotoxin production. The aim of this work was to investigate aflatoxin B1 (AFB,) production by fungi and its bioaccumulation in olives during storage and to study the impact of the application of Lactobacillus plantarum on the inhibition of mold development and production of AFB,. Two different treatments were applied: (i) olives with natural microflora and (ii) olives inoculated with Aspergillus flavus after elimination of natural microflora. AFBi has been extracted from olives and quantitated by high-performance liquid chromatography using a fluorescence detector. Results showed the absence of this metabolite in the olives for the season 2008 to 2009. In 2009 to 2010, AFB! was detected at the level of 11 pg/ kg. The application of L. plantarum during the storage of olives favors the reduction of the level of AFB, to 5.9 pg/kg correlated with a decrease in the amount of molds (86.3%). The images obtained by environmental scanning electron microscopy showed that L. plantarum was able to adhere to the olive surface and probably produce a biofilm that inhibits the multiplication of yeast and fungi by oxygen competition. Results showed an increase of antioxidant activity and amount of total phenolic compounds of olives, respectively, by 24 and 8.6%. In many olives contaminated with A. flavus, AFB, was present at an initial level of 5.15 pg/ kg and increased to 6.55 pg/kg after 8 days of storage. The biological detoxification of AFB] in olives by L. plantarum is confirmed by the reduction of the level of AFB] to 2.12 pg/kg on day 0 and its absence after 4 days of storage.

Fungal growth occurs under favorable environmental conditions, and it is associated with the production of secondary metabolites, many of which can be hazardous to vertebrates (animals and humans) (28). The conditions that promote mycotoxin production are usually more restricted than those for mold growth. However, the presence of mycotoxins in food products and the amounts produced depend entirely on the ecological and processing parameters of the foodstuff (16). Aflatoxins, secondary metabolites of some mold strains of Aspergillus flavus, Aspergillus parasiticus, and Asper­ gillus nomius, can occur as natural contaminants of foods and feeds (7). Aflatoxin B1 (AFB,), the most toxic aflatoxin, is of particular interest because it is a frequent contaminant of many food products and one of the most potent naturally occurring mutagens and carcinogens known (23). Aflatoxins have been found as contaminants of numerous crops, including peanuts, cereals, and figs (3, 5, 47); however, few studies have reported that olives (15,37), olive products (8, 54), and even olive oil (10) and some vegetable oils (35, 43) could be contaminated with aflatoxins, to a lesser extent.

* Author for correspondence. Tel: + 216 71770959; Fax: + 2 1 6 71770339; E-mail: [email protected].

Olives can be contaminated with a wide variety of naturally occuning molds on fresh and processed olives. Reports by several authors showed that olives could support mycotoxin production because they are often stored for weeks in conditions that promote the growth of molds. Therefore, toxinogenesis is possible and leads to the accumulation of mycotoxin in the olives and their possible transfer into olive oil. In effect, Aspergillus and/or Penicillium are able to develop on olives and produce ochratoxin A and/or citrinin and/or AFB after harvest, during the drying and storage of olives (11). AFB, was detected at low concentrations in olives (55), and the range of AFB! production was from 0.15 to 0.65 ng/g in olives with natural microflora and from 0.15 to 1.13 ng/g in olives inoculated with A. parasiticus after elimination of natural microflora (17). Olive and olive derivatives are very important products, and attention should be given to their aflatoxin content because of the contradictory reports. Several strategies, including physical, chemical, and biological methods have been investigated to prevent the growth of aflatoxinproducing fungi and to eliminate or reduce the levels of aflatoxins or to degrade or detoxify aflatoxins in foods and feeds. One of the most effective ways to control the problems caused by aflatoxins is to prevent the growth of fungi in the substrate. For example, the use of chemical inhibitors suppresses the spore germination of the fungi, as

J. Food Prot., Vol. 77, No. 10

well as the development of the fungal mycelium, in the substrate susceptible to contamination by these toxins. Extracts and powders of various spices, herbs, and essential oils have been reported to have antimicrobial activity against aflatoxin-producing fungi, and some of them also inhibit aflatoxin formation (29, 45, 46, 56). Biological detoxification of mycotoxins in food, raw material, and concentrated feed as well as in human and animal organisms is a new and very promising method. Particular attention is paid to lactic acid bacteria (LAB) Lactobacillus sp., and yeast Saccharomyces sp., because they can contribute to the inhibition of molds development and production of mycotoxins (12-14, 31, 34, 39, 51, 53, 57). LAB have several potential applications and are widely used for the production of fermented foods and are also part of intestinal microflora. Research reports indicate that LAB have beneficial health effects in humans and have a long history of use in foods. They produce some antagonistic compounds able to control pathogenic bacteria and undesir­ able spoilage microflora, in particular. Using LAB to control mold growth could be an interesting alternative to physical and chemical methods because these bacteria have been reported to have strong antimicrobial properties. However, the antifungal activity of lactic strains remains unclear. A limited number of reports have shown that a good selection of LAB could allow the control of mold growth and improve the shelf life of many fermented products and, therefore, reduce health risks due to exposure to mycotoxins (18). Moreover, several lactobacilli have antioxidative activity, and are able to decrease the accumulation risk of reactive oxygen species during the ingestion of food (27), and could be involved in many uses (30). Therefore, the purpose of the current study was to detect the presence of AFB ( in olives during storage for two consecutive seasons, to investigate the ability of Lactoba­ cillus plantarum, frequently encountered in the fermentation of plant materials where phenolic compounds are abundant, to remove AFB] production, and to explain the mechanism of APBi removal by L. plantarum. MATERIALS AND METHODS Olive samples. AFB, production was examined in olives during 16 days of storage at ambient temperature. The samples used were (i) olives without any treatment, holding their natural microflora, (ii) olives after the treatment presented below, to eliminate their natural microflora, and inoculated with A. flavus produces AFB1( and (iii) olives without (i) and with (ii) treatment and inoculated with L. plantarum. We obtained pitted olives of the variety “ Chetoui,” the main olive groves in the north of Tunis (Tunisia). Experiments were performed in triplicate for two consecutive seasons 2008 to 2009 and 2009 to 2010 (12 samples for each season). Olive treatment. In the present study, treatment was applied only to olives before their inoculation with A. flavus. Before their treatment, olives were washed with water and cut into pieces. To eliminate their natural microflora, they were dipped in 70% ethanol and shaken for 1 min; subsequently, ethanol was rejected. Olives were washed with 1.25% NaOCl for 3 min and rinsed with sterile distilled water twice (50).

OLIVE PRESERVATION

1761

Preparation of spore inoculum. The aflatoxigenic strain A. flavus used throughout this study was isolated from spices and obtained from the Regional Laboratory of Public Health, Tunisia. Spore inoculum was prepared by growing A. flavus on Sabouraud broth with chloramphenicol (500 pg/ml) for 72 h at 30°C, and spores were harvested aseptically using 10 ml of sterile 0.01% (vol/vol) Tween 80 solution (33). The density of suspension was adjusted to an optical density of 600 nm equal to 1.8 and was used to inoculate olives with 2 x 106 spores per g. Preparation of L. plantarum inoculum. L. plantarum was isolated from different tests of fermented olives (26). Cells of L. plantarum cultivated on de Man Rogosa Sharpe (MRS) agar were harvested by centrifugation for 15 min at 6,000 x g after 18 h of incubation at 37°C. The cell pellets were then quickly washed with deionized water and resuspended in sterile saline water (0.9%). The preparation was used to inoculate olives with 2 x 106 cells per g. Enumeration and identification of viable cells. The number of bacterial cells that adhered to the olives was determined after 0, 4, 8, 12, and 16 days of storage. Ten grams of olives was taken from each sample. The olive samples were covered in 90 ml of the dilution medium, which was 1% peptone water, to remove the planktonic cells, followed by the removal of the adhered cells using previously sterilized standardized swabs. The swabs were transferred to test tubes containing 10 ml of dilution medium and stirred vigorously in a vortex for 1 min. For each measuring period, two samples were used as replicates. For LAB counts, the samples were plated on MRS agar and incubated at 37°C for 48 h. Viable yeasts and fungi were enumerated on Sabouraud agar medium with chloramphenicol (500 pg/ml) and incubated for 48 to 72 h at 30°C. Yeast identification was conducted by morphological tests; biochemical and carbohydrate assimilation tests determined using specific API C AUX (API system, Biomerieux, Marcy l’Etoile, France) (22). Fungi were identified according to Samson et al. (49). Environmental scanning electron microscopy. After 1 and 4 days of storage, olives were gently rinsed three consecutive times in sterile distilled water (19) to remove nonadherent cells and observed using environmental scanning electron microscopy (Quanta 200) equipped with a tungsten filament (FEI, Hillsboro, OR). The signal was collected using a gaseous secondary electron detector. AFB, analysis in olive samples. The high-performance liquid chromatography (HPLC) equipment (Agilent Technologies, Santa Clara, CA) consisted of an Agilent 1100 series quaternary pump with vacuum degasser, auto-injector, and fluorescence detector. Detection of the hemiacetal derivative AFB2a was carried out at /.cx = 365 nm and Xcm = 425 nm. Data collection and subsequent processing were performed using an HP ChemStation. The analytical column was a Zorbac C l8 (250 by 4.6 mm inside diameter, particle size 4 pm) and operated at room temperature. The mobile phase was a mixture of water-acetonitrile-methanol (74/15/11, vol/vol/vol). The chromatographic program run was isocratic, and the flow rate was set at 1 ml/min, and the retention time of AFB2a was 12 min. The determination of AFB, in olives has been described in detail by Daradimos et al. (10). Olive samples (20 g) and 40 ml of hexane were poured into a 250-ml separating funnel and extracted with three 40-ml portions of methanol-water extraction solvent (60/ 40, vol/vol) containing 4% NaCl. The mixture was extracted with two 20-ml portions of aqueous 4% NaCl solution for 2 min each time. Following this, the aqueous and aqueous-methanolic extracts

1762

KACHOURI ET AL.

were combined in a separating funnel and washed twice with 20 ml of hexane each time. Then, AFB, was extracted from the aqueous phase with two 13-ml portions of chloroform and shaken for 1 min each time. The chloroform fractions were combined in a 125-ml flask and evaporated to dryness using a rotatory evaporator (42). Silica-based C18 cartridges, Mega Bond Elut C18 (Agilent Technologies), were prewashed with 3 ml of hexane and 3 ml of dichloromethane. The suitability of the cartridges was checked by adding AFB] standard (3 ml of dichloromethane containing 0.1 pg of AFBi) to the cartridges, with a satisfactory recovery of 97% (analysis was in triplicate). The sample residue was dissolved in 3 ml of dichloromethane and added to the cartridges with a flow rate of 6 ml/min. The sample residue container was rinsed with two 1-ml portions of dichloromethane, and rinses were added to the cartridges. The cartridges were washed with 3 ml of hexane, 3 ml of anhydrous ethyl ether, and 3 ml of dichloromethane (at a flow rate of 6 ml/min). The vacuum was turned off. The extraction system cover was removed, and the vial placed under the cartridge. AFB i was eluted under normal gratvity with 4 ml of chloroformacetone (90/10, vol/vol). The eluate was evaporated to dryness in a water bath (45°C) under a stream of nitrogen before derivatization following the method of Park et al. (42). The derivative of AFB! (AFB2a, hemiacetal of AFB,) was prepared by adding 200 pi of hexane and 200 pi of trifluoroacetic acid to the evaporated solution of AFB!, heating at40°C in a water bath for 10 min, evaporating to dryness under nitrogen, and redissolving in 200 pi of water-acetonitrile (9 + 1); the derivative was then analyzed by HPLC. The AFB2a shows enhanced fluorescence compared with AFBi (52). Extraction for measurement of phenols and total antiox­ idant activity. Olive paste (0.5 g) was extracted using 3 x 5 ml of methanol. The extracts were evaporated under N2 and taken up with 5 ml of methanol (4). The total phenolic contents of olive oil were measured using a modified Folin-Ciocaltau colorimetric method (4). Methanolic extract (0.5 ml) was dissolved in 0.5 ml of Folin-Ciocalteu reagent (1/10 dilution) and 1 ml of distillated water; 1.5 ml of 20% sodium carbonate was added after 1 min, and the final mixture was shaken thoroughly. Absorbance was measured at 760 nm versus a blank after 2 h in the dark at room temperature. The results were expressed in milligrams of gallic acid per liter of olive oil. The total antioxidative activity was assessed by using the linolenic acid test (LA-test), evaluates the ability of the samples to inhibit LA (Sigma-Aldrich, St. Louis, MO) oxidation (41). The standard of LA in 96% ethanol (1:100) was diluted in isotonic saline (1:125). To 0.4 ml of LA preparation was added 0.01% sodium dodecyl sulphate (Sigma) and the sample (0.045 ml). Incubation was stalled by adding 0.1 mM FeS04 (Sigma) and was monitored at 37°C for 60 min. After the interruption of the reaction by adding 0.25% butylated hydroxytoluene (Sigma), the mixture was treated with 0.5 ml of acetate buffer (pH 3.5; Sigma) and heated with freshly prepared 1% thiobarbituric acid solution (Sigma) at 80°C for 40 min. After cooling, the mixture was acidified by adding 0.5 ml of cold 5 M HC1, extracted with 1.7 ml of cold 1-butanol (Sigma), and centrifuged 10 min at 3,000 x g. Absorbance of the butanol fraction was then measured. The total antioxidative activity of the sample corresponds to its inhibition effect on LA-standard peroxidation as follows: [1 —A534 (sample)/ A534 (LA as control)] x 100. The higher numerical value (percentage) of the total antioxidative activity is an indication of the higher antioxidant phenomenon in the sample. Peroxidation of LA-standard in the isotonic saline (without samples) served as control.

J. Food Prot., Vol. 77, No. 10

Statistical analysis. Statistical analyses were performed using the analysis of variance test: DATASET1.ISD by GraphPad in stats demo version 3.0 software to determine differences between the means. Statistical significance was determined at a 5% probability level.

RESULTS AND DISCUSSION AFBi production in olives with natural microflora during storage. The in-house characterization of the method for determinination of AFB, in olives has been reported in detail by Daradimos et al. (10). The working standard solutions used for the establishment of the calibration curve were at a concentration of 0.5, 1.0, 2.5, and 5.0 pg/kg. Each concentration was injected in triplicate. The chromatogram corresponding to the standard solution of AFBi at 2.5 pg/kg, show that AFB! is eluted at a retention time of 12 min. Using this calibration method, the limit of detection (signal-to-noise ratio = 3) was 0.15 pg/ kg, with a linear range covering the whole calibration curve with a correlation coefficient of regression r2 = 0.9996 and a mean recovery of 97%, with a mean standard deviation (SD) of 6.3%. These results demonstrated the suitability of the analytical method for the detection of AFBi in olives. This study started by a screening survey, which was undertaken to determine the presence and levels of AFB 1 in local olive products for two consecutive seasons (2008 to 2009 and 2009 to 2010). Thus, samples were analyzed to quantify AFB! levels in these products. The chromatographic analysis of the olive samples, harvested in season 2008 to 2009, showed no AFBi content throughout the storage period. The absence of AFBi in the studied samples could be explained by the presence of antimicrobial constituents, such as caffeic acid, coumarins, flavones, catechins, and phenolic compounds (44), and could also be due to the absence of mycoflora producing AFBi ° r to the simultaneous presence of microorganisms that generate competition for substrate. However, the presence of other microorganisms may restrict fungal growth and possibly mycotoxin production. It is known that a natural fermentation in olives takes place for which a complex microflora is needed. More recently, it was reported that degradation of ochratoxin A is possible by some microorganisms, such as Lactobacillus sp., Strepto­ coccus sp., and Aspergillus niger sp. (59). Also, an atoxigenic Aspergillus strain was found to decompose ochratoxin A (58), although it was shown that the presence of penicillia decreases significantly the level of AFBi. The aflatoxin production by A.flavus may be inhibited when A. niger is present in the environment. The identification of yeast and fungal flora of the olive samples studied showed the presence of five species of yeast (Candida famata [32%], Candida ciferrii [10%], Rhodotorula mucilaginosa [21%], Crytococcus laurentii [16%], and Pichia guilliermondii [16%]) and six species of fungi (A. niger [14%], Aspergillus fumigatus [14%], Geotrichum penicillatum [10%], Geotrichum candidum [10%], Geotri­ chum captitum [10%], and Penicillium sp. [42%]). These results indicated the absence of fungi-producing aflatoxins such as A.flavus, A. parasiticus, and A. nomius. Bircan (2)

J. Food Prot., Vol. 77, No. 10

1763

OLIVE PRESERVATION

TABLE 1. AFBj determination in the olive samples (harvested in season 2009 to 2010)“

Control olive samples Olive samples inoculated by L. plantarum

Day 0

Day 4

Day 8

Day 12

Day 16

11 5.9

0 0

0 0

0 0

0 0

“ Values are in micrograms per kilogram; each value is the mean for three replicates.

showed that Penicillium, Geotrichum, and Rhizopus are the major species present in olives during storage, and Penicillium is the most abundant. Tantaoui-Elaraki and Le Tutour (54) indicated that Penicillium species are commonly found in olives and olive products. For the season 2009 to 2010, the chromatographic analysis showed the presence of AFB, in one of olive samples at a mean level of 11 pg/kg with an SD of 6.3% (n = 3; Table 1). The presence of AFB! in the sample could be due to the existence of a toxigenic species that has not been identified because foods such as olives are not homogeneously contaminated by aflatoxins. Indeed, Leontopoulos et al. (33) reported that contamination of olives by molds may not be uniform. During the storage period (from day 4 to day 16), no AFB! was observed (Table 1). This result could be explained partly by the degradation of these metabolites that were present at the beginning of storage (day 0) (33). Indeed, Nguyen (39) claims that certain strains of fungi can degrade aflatoxins. On the other hand, it could be due to the presence of phenolic compounds that are known for their antimicrobial and antifungal activity and prevention against the production of aflatoxins (3, 44). Nowadays, control by naturally produced agents is becoming increasingly important. Natural plant compounds have been used traditionally to preserve foods. Extracts and powders of various spices, herbs, and essential oils are reported to have antimicrobial activity against aflatoxinproducing fungi, and some of them also inhibit aflatoxin formation (29, 45, 46, 56). Moreover, the use of LAB and their metabolites to control mold development and mycotoxin accumulation appears to be a promising biocontrol strategy in perishable foods or feed frequently contaminated by toxigenic fungal strains, including vegetable products, particularly cereals, fruits, and by-products (9). The application of LAB has been mainly limited to fermented products until now, using bacterial strains that have been recognized as safe and that do not have an impact on the taste and appearance of the supplemented product. Among the identified applications is the use of LAB as starter cultures in the brewing industry. Two additional strategies are currently being evaluated; incorporation of LAB into biofilms and encapsulation processes. AFBi removal by L. plantarum in olives with natural microflora during storage. The olives inoculated with L. plantarum have shown a mean level of AFB] of 5.9 pg/kg, with a SD of 5.6% (w = 3) on day 0, less than the amount of AFBi of 11 pg/kg present in the control olive sample. Then, no AFB) was detected during storage (Table 1).

This is in agreement with other studies on aflatoxin removal by LAB (13), in which Lactobacillus rhamnosus strains GG and Lc 705 bound approximately 80% AFB j within 0 h. Turbic et al. (57) showed that 77 to 99% AFBi were removed by L. rhamnosus strains in high and moderate amounts. Moreover, El Khoury et al. (12) found also that Lactobacillus bulgaricus and Streptococcus thermophilus strains, used in Lebanese dairy industries, were shown effective in reducing the extent of free AFMi content in liquid culture medium and during yogurt processing. Microflora interactions should have an impact on the food deterioration, especially metabolite production. Therefore, LAB seem to play a crucial role in AFMi removal and could be used as a biological agent for AFMi reduction. Even though the mechanism of aflatoxin removal by LAB is still unknown, it has been suggested that aflatoxin molecules are bound to bacterial cell wall components rather than metabolically degraded (13, 31). Haskard et al. (20) suggested that AFBi is bound to bacteria through weak noncovalent interactions, such as associations with hydrophobic pockets on the bacterial surface. Another mechanism by which probiotic bacteria induced its protective effect was suggested by Kodali and Sen (28), who reported that these bacteria synthesize extracellular polysaccharides with signif­ icant physiological and therapeutic activities and have significant antioxidant and free radical scavenging activities. Hathout et al. (21) evaluated the protective role of Lactobacillus casei and Lactobacillus reuteri against aflatoxin-induced oxidative stress in rats and reported that the protective role of these strains may be due to their aflatoxins binding activity as well as their antioxidant properties. Several LAB have been able to bind AFBi in vitro and in vivo. This model suggests the attachment of AFBi molecules to the surface of the organism and takes two processes into consideration: binding and release of aflatoxin to and from the binding site on the surface of the microorganisms (6, 32). The binding property displayed by some selected LAB, resulting in a decrease of mycotoxin bioavailability, could therefore be used in novel approaches to decontaminate food and feed. However, little is known at this time about the stability and toxicity of LAB-mycotoxin complexes. In this study, the removal of AFB, by L. plantarum, could be explained by different mechanisms. The first mechanism could be attributed to the ability of L. plantarum to reduce the amount of molds during the storage, which could foster the reduction of their level of AFB], and is in agreement with the study of Yassa (60) that reported that LAB inhibit fungal growth and AFBi production. Accord­ ing to Magnusson et al. (36), three mechanisms may explain

1764

KACHOURI ET AL.

Storage (days) FIGURE 1. Evolution o f L. plantarum cellular concentration (A) and percent reduction of molds ( ■ ) of olive samples inoculated with L. plantarum during storage (haiyest in season 2009 to 2010). the antim icrobial efficiency o f LAB: the yield o f organic acid, com petition for nutrients, and production o f antago­ nistic com pounds. Several species or subspecies, such as Lactococcus lactis subsp. lactis, Lactococcus lactis subsp. cremoris, Lactococcus lactis subsp. diacetylactis, Lactoba­ cillus acidophilus, L. plantarum, and Lactobacillus curvatus, are able to synthesize peptides or antim icrobial proteins know n as bacteriocins, w hose activity is only directed against closely taxonom ically related bacteria. The num eration o f viable cells after 0, 4, 8, 12, and 16 days o f storage show ed that the application o f L. plantarum favors the reduction o f the am ount o f m olds (86.3% ; Fig. l).T h e significant reduction o f yeast and fungi can be due to the ability o f L. plantarum to adhere to the olive surface and produce biofilm and probably affected the final population o f the biofilm on the surface, resulting from nutrient and oxygen com petition or from synthesis o f antagonistic com pounds. N ow adays, protective biofilm form ation o f food industry surfaces can also be beneficial because its presence m ay effectively m odify the physico­ chem ical properties o f substrates and, as such, reduce adhesion o f the undesirable planktonic m icroorganism . In recent years, biofilm s o f LAB have received considerable attention for their potential use in the settlem ent o f a com petitive flora and changes to cell surface physicochem ­ ical properties for increased adhesion o f protective biofilms. The assessm ent o f adhesion o f L. plantarum to olives at different times o f contact has show n a higher increase in the total num ber o f L. plantarum adhering, w hich increased exponentially over 1.3 x 108 cells per g (Fig. 1). M oreover, the im ages obtained by environm ental scanning electron m icroscopy m icroscopy (Fig. 2) show ed that the olive surface was covered by a uniform and com pact biofilm. A lso, observations revealed that the biofilm form ed from 1 to 4 days had the sam e m orphological characteristics and the b iofilm ’s volum e increased during storage. T he adhesion m ay be due to the correlation found betw een the cell surface hydrophilicity o f L. plantarum and its ability to adhere to the hydrophilic surface o f olive. M oreover, results show ed that the application o f L. plantarum during the storage o f olive fruits favors the increase o f antioxidant activity and the am ount o f total phenolic com pounds, respectively, with 24 and 8.6%

J. Food Prot., Vol. 77, No. 10

(Fig. 3). This m ay be due to the ability o f L. plantarum to use the oxygen present in the solution that was responsible for the auto-oxidation o f phenolic com pounds. In fact, previous w ork carried out in our laboratory show ed that viable cells o f L. plantarum had an antioxidant activity and also show ed that the antioxidant activity increased w ith the concentration o f L. plantarum. From this result, w e suggest that L. plantarum adhesion could probably induce an insufficiency o f oxygen supply to the olive surface, resulting in inhibit fungal grow th and A F B | production. Indeed, aflatoxins are highly oxygenated m olecules, and the decrease in oxygen concentration using antioxidant enzym es, such as xanthine oxidase, superoxide dism utase, and glutathione peroxidase, decreases the production o f A FB, by decreasing lipid peroxidation, w hich affects thenbiosynthesis (2). In fact, m any LAB possess enzym atic and nonenzym atic antioxidative m echanism s and m inim ize generation o f reactive oxygen species to levels that are not harm ful to the cells. LAB lack m any o f the com ponents o f the respiratory chain, which facilitate the utilization o f 0 2 as a term inal electron acceptor. However, m any LAB synthesize the coupled N A D H oxidase/NA D H peroxidase system that balances the N A D + /N A D H ratio, catalyzes the reduction o f 0 2 to H20 2, and decom poses H20 2 to H20 for the purpose of protection. Jansch et al. (24) show ed that the defense o f Lactobacillus sanfranciscensis to oxygen toxicity is the involvem ent o f oxygen in the m etabolism by N A DH -oxidase activity. Thus, lactobacilli differ in their response to oxygen and their ability to use 0 2 in central carbon flux. The second m echanism is also related to the ability o f L. plantarum to capture oxygen and consequently preserve phenolic com pounds, present in the olive samples, against oxidation by the oxygen present in the m edium . The increase o f the am ount o f phenolic com pounds in inoculated olives could have an inhibitory effect on AFB \ level. In fact, previous studies have show n that the addition o f phenolic com pounds extracted from olives to a m edium containing a culture o f A .fla vu s inhibits up to 90% o f AFB! production w ithout affecting fungal grow th (44) and that the action m ode o f phenolic com pounds on A FB i is related to their am ounts. From another side, phenolic com pounds, espe­ cially caffeic acid and catechins, show ed antibacterial and antifungal activities. A ziz et al. (1) reported that com pounds extracted from olive fruits, such as oleuropein inhibited aflatoxin production. M oreover, grow th and A F B , produc­ tion by A. flavus and A. parasiticus were com pletely inhibited by vanillic acid, caffeic acid, p-coum aric acid, phydroxybenzoic acid, and quercetin. Indeed, previous work carried out in our laboratory show ed that the application o f L. plantarum during the olive oil process increases the concentration o f phenolic com pounds, particularly orthodi­ phenols (26). M oreover, K achouri and H am di (25) studied the transform ation o f phenolic com pounds contained in olive mill w astew ater (OM W ) into valuable products using L. plantarum to increase their transportation from O M W to olive oil. Incubation o f olive oil samples with ferm ented O M W by L. plantarum caused polyphenols to decrease in O M W and increase in oil. Ferm entation with L. plantarum induced reductive depolym erization o f O M W , w hich is

OLIVE PRESERVATION

J. Food Prot., Vol. 77, No. 10

1765

FIGURE 2. Electromicrographs of L. plantarum adhered onto olives during storage, visualized by environmental scan­ ning electron microscopy: Olive adhered by L. plantarum after (a) 1 day and (b) 4 days of contact.

more soluble in olive oil. The analysis of the phenolic compounds found in olive oil after storage showed that the application of L. plantarum favors the increase of all phenolic compounds in olive oil, especially by depolymer-

0

4

8

12

16

0

4

8

12

16

Storage (days) FIGURE 3. Effect of L. plantarum on phenolic content (a) and total antioxidant activity (b) of methanolic extract of olives ([W] control and [M] inoculated by L. plantarum,) during storage (harvest in season 2009 to 2010).

ization and by reductive conversion of phenolic compounds of olive and oxygen fixation. In fact, the capacity of the strain to convert phenolic compounds participates positively in the antifungal activity and, consequently, in the prevention against the production of aflatoxins. To the best of our knowledge, only one article has dealt with the involvement of a phenolic compound produced by LAB in the antifungal activity (38). This phenolic compound, which remains to be identified, was produced by Pediococcus acidilactici LAB 5 and showed varying degrees of antifungal activity against a number of foods, feedbome molds, and plant pathogenic fungi. AFBj removal by L. plantarum in olives contami­ nated with A. flavus during storage. The effect of L. plantarum on the production of AFB1; by artificial contamination of olives with A. flavus, was performed. This strain was used because, according to Mahjoub and Bullerman (37), it is a stable and strong aflatoxigenic producer. Before inoculation, olives were treated to eliminate their natural microflora. The numeration of viable cells showed that the treatment favors the reduction of the amount of yeasts and fungi. AFB, was not detected in all samples of olives before and after treatment. The chromatographic analysis showed that the mean level of AFB i on day 0 of storage in the olives inoculated with A. flavus was 5.15 pg/kg, with an SD of 6.6% (n = 3). This level increased to a maximum of 6.55 pg/kg, with an SD of 4.2% (n = 3), after 8 days of storage, and then it decreased to 5.7 pg/kg, with an SD of 7.1% (n = 3), after 12 days of storage and to 2.35 pg/kg, with an SD of 5.9% (n = 3), after 16 days of storage (Table 2). The decrease in the

1766

KACHOURI ET AL.

J. Food Prot., Voi. 77, No. 10

T A B L E 2. A F B / determ ination in the artificially contam inated olive sam plesa

O live sam ples inoculated w ith A . fla v u s O live sam ples inoculated w ith A . fla v u s and L. plantarum

Day 0

Day 4

Day 8

Day 12

Day 16

5.15 2.12

5.3 0

6.55 0

5.7 0

2.35 0

" V alues are in m icrogram s p e r kilogram ; each value is the m ean for three replicates.

AFB! level could be attributed to its degradation and to the action of phenolic compounds (2). For the olives samples inoculated with A. flavus and L. plantarum, the level of AFB! was 2.12 pg/kg, with an SD of 7.4% (n = 3) on the day of inoculation, and then no AFB! was detected after 4, 8, 12, and 16 days of storage (Table 2). The level of AFB] was found to be lower in the case of the olives inoculated with both A. flavus and L. plantarum (2.12 pg/kg) than in the olives inoculated with only A. flavus (5.15 pg/kg). These finding demonstrated clearly the inhibitory effect of L. plantarum on the production of AFBi in olives during storage by a factor of 58.8%. The removal of AFBi by L. plantarum was accompanied with a reduction of the amount of molds (70.1%; Fig. 4). The percentage of reduction increased to a maximum (92.4%), after 8 days of storage. Recently, Salminen et al. (48) reported that several bacterial strains have the ability to reduce exposure to dietary mycotoxins, and these strains were identified as mycotoxin binders. The greatest aflatoxin removal capacity strains described to date are L. rhamnosus, and they are currently used in food products and present a potential cost-effective and commercially viable approach to detoxifying aflatoxincontaminated products. Turbic et al. (57) indicated that specific strains of LAB have proved highly effective in removing AFB] in model systems. According to Oatley et al. (40), the Bifidobacteria strain binds 25 to 60% of the AFB|. Despite the low incidence of AFBi in Tunisian olive products compared with other regions worldwide, the current results in this study imply that more emphasis should be given to the determination of AFB[ levels in Tunisian olive and derived products. Olives could support mycotoxin production, as often, they are stored for weeks in conditions that promote the growth of molds. Therefore, toxinogenesis is possible and leads to the accumulation of mycotoxin in the olives and possible transfer into olive oil.

Also, LAB, L. plantarum strain, frequently encountered in the fermentation of plant materials in which phenolic compounds are abundant, were shown effective in reducing the extent of free AFBi content in olives during storage. Therefore, microbial ecology and chemical composition of the foodstuff could play a crucial role in the presence of mycotoxins in food products. REFERENCES 1.

2.

3.

4.

5.

6.

7. 8. 9.

10.

11.

8,5 12.

8 7,5 7

00

£ u.

13.

6,5 o fi 6 'B .2 5,5

14.

5 4,5 15. Storage (days) FIG U R E 4. E volution o f L . plantarum cellular concentration (A ) a n d p e rc e n t reduction o f A . flavus ( ■ ) o f olive sam ples inoculated with A. flavus a n d L. plantarum during storage (harvest in season 2009 to 2010).

16. 17.

Aziz, N. H., S. E. Farag, L. A. A. Moussa, and M. A. Abo-Zaid. 1998. Comparative antibacterial and antifungal effects of some phenolic compounds. Microbios 93:43-54. Bircan, C. 2005. Determination of aflatoxin contamination in olives by immunoaffinity column using high-performance liquid chroma­ tography. J. Food Quality 29:126-138. Bircan, C. S., A. Barringer, U. Ulken, and R. Pehlivan. 2008. Aflatoxin levels in dried figs, nuts and paprika for export from Turkey. Int. J. Food Sci. Technol. 43:1492-1498. Boskou, G., F. N. Salta, S. Chrysostomou, A. Mylona, A. Chiou, and N. K. Andrikopoulos. 2006. Antioxidant capacity and phenolic profile of table olives from the Greek market. Food Chem. 94:558-564. Boyacioglu D., and M. Gonul. 1990. Sutvey of aflatoxin contami­ nation of dried figs grown in Turkey in 1986. FoodAddit. Contam. 7: 235-237. Bueno, D. J., C. H. Casale, R. P. Pizzolitto, M. A. Salano, and G. Olivier. 2006. Physical adsorption of Aflatoxin B 1 by lactic acid bacteria and Saccharomyces cerevisiae: a theoretical model. J. Food Prot. 70:2148-2154. Bullerman, L. B. 1986. Mycotoxins and food safety. Food Technol. 40:59-66. Costabeler, I., R. Angulo, and M. Jodral. 1996. Aflatoxin in olive oil cake. Microbiol. Aliments Nutr. 14:271-274. Dalie, D. K. D., A. M. Deschamps, and F. Richard-Forget. 2010. Lactic acid bacteria-potential for control of mould growth and mycotoxins: a review. Food Control 21:370-380. Daradimos, E., P. Marcaki, and M. Koupparis. 2000. Evaluation and validation of two fluorometric HPLC methods for the determination of aflatoxin B1 in olive oil. Food Addit. Contam. 12:445^t50. El Adlouni, C., M. Tozlovanu, F. Naman, M. Faid, and A. PfohlLeszkowicz. 2006. Preliminary data on the presence of mycotoxins (ochratoxin A, citrinin and aflatoxin B l) in black table olives “ Greek style” of Moroccan origin. Mol. Nutr. Food Res. 50:507-512. El Khoury, A., A. Atoui, and J. Yaghi. 2011. Analysis of aflatoxin M l in milk and yogurt and AFM1 reduction by lactic acid bacteria used in Lebanese industry. Food Control 22:1695-1699. el-Nezami, H., P. Kankaanpaa, S. Salminen, and J. T. Ahokas. 1998. Physicochemical alterations enhance the ability of dairy strains of lactic acid bacteria to remove aflatoxin from contaminated media. J. Food Prot. 61:466-468. el-Nezami, H., H. Mykkanen, P. Kankaanpaa, S. Salminen, and J. Ahokas. 2000. Ability of Lactobacillus and Propionibacterium strains to remove aflatoxin Bl from the chicken duodenum. J. Food Prot. 63:549-552. Eltem, R. 1996. Growth and aflatoxin Bl production on olives and olive paste by moulds isolated from “ Turkish-style” natural black olives in brine. Int. J. Food Microbiol. 32:217-223. Filtenborg, O., J. C. Frisvad, and U. Thrane. 1996. Moulds in food spoilage. Int. J. Food Microbiol. 33:85—102. Ghitakou, S, K. Koutras, E. Kanellou, and P. Markaki. 2006. Study of aflatoxin B] and ochratoixin A production by natural microflora and

OLIVE PRESERVATION

J. Food Prot., Vol. 77, No. 10

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

Aspergillus parasiticus in black and green olives of Greek origin. Food Microbiol. 23:612-621. Gourama, H., and L. B. Bullerman. 1995. Inhibition of growth and aflatoxin production of Aspergillus flavus by Lactobacillus species. / . Food Prot. 58:1249-1256. Hamadi, F., H. Latrache, M. Mabrrouki, A. El Ghmari, A. Outzourhit, M. Ellouali, and A. Chtaini. 2005. Effect of pH on distribution and adhesion of Staphylococcus aureus to glass. J. Adhes. Sci. Technol. 19:73-85. Haskard, C. A., H. S. El-Nezami, P. E. Kankaanpaa, S. Salminen, and J. T. Ahokas. 2001. Surface binding of aflatoxin B (1) by lactic acid bacteria. Appl. Environ. Microbiol. 67:3086-3091. Hathout, A. S., S. R. Mohamed, A. A. El-Nekeety, N. S. Hassan, S. E. Aly, and M. A. Abdel-Wahhab. 2011. Ability of Lactobacillus casei and Lactobacillus reuteri to protect against oxidative stress in rats fed aflatoxins-contaminated diet. Toxicon 58:179-186. Hernandez, A., A. Martin, E. Aranda, F. Perez-Nevado, and M. G. Cordoba. 2007. Identification and characterization of yeast isolated from the elaboration of seasoned green table olives. Food Microbiol. 24:346-351. International Agency for Research on Cancer. 1993. Some naturally occurring substances: food items and constituents, heterocyclic aromatic amines and mycotoxins, vol. 56, p. 19-23. In IARC monographs on the evaluation of carcinogenic risks to humans. Jansch, A., S. Freiding, J. Behr, and R. F. Vogel. 2011. Contribution of the NADH-oxidase (Nox) to the aerobic life of Lactobacillus sanfranciscensis DSM20451T. Food Microbiol. 28:29—37. Kachouri, F„ and M. Hamdi. 2004. Enhancement of polyphenols in olive oil by contact with fermented olive mill wastewater by Lactobacillus plantarum. Process Biochem. 39:841-845. Kachouri, F., and M. Hamdi. 2006. Use Lactobacillus plantarum in olive oil process and improvement of phenolic compounds content. J. Food Eng. 77:746-752. Kaizu, M., M. Sasaki, H. Nakajima, and Y. Suzuki. 1993. Effect of antioxidative lactic acid bacteria on rats fed a diet deficient in vitamin E. J. Dairy Sci. 76:2493—2499. Kokkonen, K., M. Jestoi, and A. Rizzo. 2005. The effect of substrate on mycotoxin production of selected Penicillium strains. Int. J. Food Microbiol. 99:207-214. Krishnamurthy, Y. L., and J. Shashikala. 2006. Inhibition of aflatoxin B 1 production of Aspergillus flavus isolated from soybean seeds by certain natural plants products. Lett. Appl. Microbiol. 43:469^174. Kullisaar, T., M. M. Zilmer, T. Mikelsaar, H. Vihalemm, C. Annuk, and A. Kilk. 2002. Two antioxidative lactobacilli strains as promising probiotics. Int. J. Food Microbiol. 72:215-224. Lahtinen, S. J., C. A. Haskard, A. C. Ouwehand, S. J. Salminen, and J. T. Ahokas. 2004. Binding of aflatoxin B1 to cell wall components of Lactobacillus rhamnosus strain GG. Food. Addit. Contam. 21:158—164. Lee, Y. K., H- El-Nezami, C. A. Haskard, S. Gratz, K. Y. Puong, S. Salminen, and H. Mykkanen. 2003. Kinetics of adsorption and desoiption of aflatoxin B 1 by viable and nonviable bacteria. J. Food Prot. 66:426^130. Leontopoulos, D., A. Siafaka, and P. Markaki. 2003. Black olives as substrate for Aspergillus parasiticus growth and aflatoxin B1 production. Food Microbiol. 20:119—26. Madrigal-Santilla, E., E. Madrigal-Bujaider, and R. Maquez-Maquez. 2006. Antigenotoxic effect of Saccharomyces cerevisiae on the damage produced in mice fed with aflatoxin B 1 contaminated com. Food Chem. Toxicol. 44:2058-2063. Magnarini, C., T. Mezzetti, L. Cossignani, V. Cecchetti, C. Cubbiotti, F. Santinelli, G. Burini, and P. Damiani. 1990. RP HPLC determination of aflatoxins in edible oils. Rass Chimie 42:273-276. Magnusson, J., K. Strom, S. Roos, J. Sjogren, and J. Schniirer. 2003. Broad and complex antifungal activity among environmental isolates of lactic acid bacteria. FEMS Microbiol. Lett. 219:129-135. Mahjoub, A, and L. B. Bullerman. 1987. Mold growth and aflatoxin production on whole olives and olives pastes. Sci. Aliments 7:629636.

38.

39.

40. 41.

42.

43. 44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55. 56.

57.

58.

59. 60.

1767

Mandal, V., S. K. Sen, and N. C. Mandal. 2007. Detection, isolation and partial characterization of antifungal compound(s) produced by Pediococcus acidilactici LAB 5. Nat. Prod. Commun. 2:671-674. Niderkom, V., H. Boudra, and D. Morgavi. 2006. Binding of Fusarium mycotoxins by fermentative bacteria in vitro. / . Appl. Microbiol. 101:849-856. Oatley, J. T., M. D. Rarick, G. E. Ji, and J. E. Linz. 2000. Binding of aflatoxin B1 to Bifidobacteria in vitro. J. Food Prot. 63:1133-1136. Pahkla, R., M. Zilmer, T. Kullisaar, and L. Rago. 1998. Comparaison of the antioxidant activity of melatonin and pinoline in vitro. / . Pineal Res. 24:96-101. Park, D. L., M. W. Trucksess, S. Nesheim, M. Stack, and R. F. Newell. 1994. Solvent-efficient thin layer chromatographic method for the determination of aflatoxins B l, B2, G1 and G2 in com and peanut products: collaborative study. J. AO AC Int. 77:637-646. Parker, W. A., and D. Melnick. 1996. Absence of aflatoxins from refined vegetable oils. J. Am. Oil Chem. Soc. 43:635-638. Paster, N., B. J. Juven, and H. Harshemesh. 1988. Antimicrobial activity and inhibition of aflatoxin B 1 formation by olive plant tissue constituents. J. Appl. Bacteriol. 64:293—297. Razzaghi-Abyaneh, M., M. Shams-Ghahfarokhi, T. Yoshinari, M. B. Rezaee, K. Jaimand, H. Nagasawa, and S. Sakuda. 2008, Inhibitory effects of Satureja hortensis L. essential oil on growth and aflatoxin production by Aspergillus parasiticus. Int. J. Food Microbiol. 123:228—233. Reddy, K. R. N., S. B. Nurdijati, and B. Salleh. 2010. An overview of plant-derived products on control of mycotoxigenic fungi and mycotoxins. Asian J. Plant Sci. 9:126—133. Saleemullah, S., A. Iqbal, I. A. Khalil, and H. Shah. 2006. Aflatoxin contents of stored and artificially inoculated cereals and nuts. Food Chem. 98:699-703. Salminen, S., S. Nybom, J. Meriluoto, M. C. Collado, S. Vesterlund, and H. El-Nezami. 2010. Interaction of probiotics and pathogensbenefits to human health. Curr. Opin. Biotechnol. 21:157-167. Samson, R. A., E. S. Hoekstra, F. Lund, O. Filtenborg, and J. C. Frisvad. 2000. Methods for the detection, isolation and characteriza­ tion of food-bome fungi, p. 283-297. In R. A. Samson, E. S. Hoekstra, J. C. Frisvad, and O. Filtenborg (ed.), Introduction to foodand airborne fungi, 6th ed. Centralbureau voor Schimmelcultures, Utrecht, The Netherlands. Scott, G. E., and N. Zummo. 1995. Size of maize sample needed to determine percent kernel infection by Aspergillus flavus. Plant Dis. 79:861-864. Shetty, P., and L. Jespersen. 2006. Saccharomyces cerevisiae and lactic acid bacteria as potential mycotoxin decontaminating agents. Trends Food Sci. Technol. 17:48-55. Stubblefield R. D. 1987. Optimum conditions for formation of aflatoxin Ml-trifluoroacetic acid derivative. J. Assoc. Off. A nal Chem. 70:1047-1049. Styriak, I., E. Conkova, V. Kmet, J. Bom, and E. Razzazi. 2001. The use of yeast for microbial degradation of some selected mycotoxins. Mycotoxin Res. 17:24—27. Tantaoui-Elaraki, A., and B. Le Tutour. 1985. Possible mycotoxin contamination of olives and olive products: latest development. Oleagineux 40:451^154. Tantaoui-Elaraki, A., and A. Marmioui. 1996. Antifungal treatment trial of “ Greek style” black olives. Microbiol. Aliments Nutr. 14:5-14. Thanaboripat, D., N. Mongkontanawut, Y. Suvathi and V. Ruangrattametee. 2004. Inhibition of aflatoxin production and growth of Aspergillus flavus by citronella oil. KMITL Sci. J. 4:1-8. Turbic, A., J. T. Ahokas, and C. A. Haskard. 2002. Selective in vitro binding of dietary mutagens, individually or in combination, by lactic acid bacteria. Food Addit. Contam. 19:144-152. Varga, J., Z. Peteri, K. Tabori, J. Teren, and C. Vagvolgyi. 2005. Degradation of ochratoxin A and other mycotoxins by Rhizopus isolates. Int. J. Food Microbiol. 99:321-328. Varga, J., K. Rigo, and J. Teren. 2000. Degradation of ochratoxin A by Aspergillus species. Int. J. Food Microbiol. 59:1-7. Yassa, I. A. 1995. Some factors affecting mold growth and aflatoxin production in olives. Ann. Agric. Sci. 40:59-65.

Copyright of Journal of Food Protection is the property of Allen Press Publishing Services Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Removal of aflatoxin B1 and inhibition of Aspergillus flavus growth by the use of Lactobacillus plantarum on olives.

Olives can be contaminated with a wide variety of molds (Aspergillus and/or Penicillium) that can be occurring naturally on fresh and processed olives...
6MB Sizes 1 Downloads 7 Views