BIOTROPIA Vol. 28 No. 3,2021: 239 - 252 DOI: 10.1 1598/btb.2021.28.3.1362 ANTAGONISTIC EFFECT OF YEAST, ACETIC ACID BACTERIA AND MANGOSTEEN RIND EXTRACT O N AFLATOXIGENIC Aspergifis f laws I N UNFERMENTED COCOA BEANS NIJMA NURFADILA'", SUTRISNO~, USMAN AH MAD^ AND SAMSUDIN~ 'SEAMEO BIOTROP, Jalan Raya Tajzlr Km. 6, Bogor 16134, Indonesia 2 S t ~ 3 Pmgram of Postharvest Technology, Faczllty ofAgricultural techno lo^, Instit~lt Pertdnian Bogor, Bogor 16680, Indonesia 3Balai Penelitan Tanaman Industri dan Penyegar (BALITTTU), Pamngkuda, St&abumi, Indonesia 43357 Received 22 May 2020 /Accepted 6 January 2023 ABSTRACT Yeasts and bacteria are two of common biocontrol agents to control mycotoxigenic fungi. Meanwhile, the mangosteen rind extract contains xanthone and gartanin compounds for antioxidant, antiproliferation, antiinflammation, antimicrobial, and anticancer. The objectives of this research were to test the effects of yeasts, acetic acid bacteria (AAB), and mangosteen rind extract on the aflatoxigenic A.ybegillz1sj7av.w growth and aflatoxin production in unfermented cocoa beans. Four yeast isolates, i.e., I~satcbenkia orientalis (lo) BIO 211291, 286 and 288, and Endomyces fbulger (E! BIO 132219; one bacteria isolate of Acetobacter aceti (Aa) FNCC0016; and mangosteen rind extract (MRE) were tested for their capabilities in inhibiting an aflatoxigenic A.javus (Afj BIO 3361/747 growth using the well method (in vitro). Two types of yeast (10 BIO 211291 and 288) were combined with A a and MRE in cocoa beans (in vivo). Aflatoxin production was analyzed using Thin h y e r Cbromathograph_v (TLC). The results showed that interaction of l o BIO 211291 and 288, and EfBIO 132219 on aflatoxigenic Af were interaction with inhibition zone > 2 mm (type D), while the interaction type of l o BIO 211286 on Afwere mutual intermingling growth, where both fungi grew into each other without any macroscopic sign of interaction (type A). The best treatment in agar media (in vitro) was Io BIO 211288 + A a on Potato Dextrose Agar + 12 g/L MRE. The highest l o population was 5.88 log cfu/g on cocoa beans inoculated by I o BIO 211291 + MRE in 1 day after inoculation, while the highest A. aceti population was 4.74 log cfu/g on cocoa beans with l o BIO 21 1291 + BIO 211288 + A a in 3 days after inoculation. Two best treatments were I o BIO 211288 + A a + MRE and I o BIO 211291 + BIO 211288 + A a + MRE, because there were no A.j7avuspopulation since 3 until 11 days after inoculation. Aflatoxin in all samples treatment was lower than limit detection B1 (< 2.20 ppb), B2 (< 3.50 ppb), (GI < 0.54 ppb), dan (G2 < 1.00 ppb). Keywords: aflatoxin, antagonistic, Aspergillusj7avus, mangosteen rind, yeasts INTRODUCTION Indonesia is the yd cocoa beans exporter countries after Ivory Coast and Ghana (Dickson 201 8). Agricultural Department (201 0) reported as much as 93% of cocoa beans in Indonesia was processed without fermentation (only sun- dried) conducted by farmers, while the remaining 7% of cocoa beans was processed by fermentation. Thompson et aL (2013) explained that cocoa beans fermentation process involves *Corresponding author, email: n.nurfadila@yahoo.co.id; nijma@biotrop.org some microorganisms, i.e., yeasts, acetic acid bacteria (AAB), lactic acid bacteria (LAB), BaciIIz~s and several other bacteria, as well as filamentous fungi. According to Nurhansyah (201 1) some importer countries such as Malaysia and United States of America reduced the amount of cocoa beans import from Indonesia, because the physicochemical quality of Indonesian cocoa beans were lower than those from Africa. The decreasing of cocoa beans quality was caused by unappropriate handhg during the harvesting, fermentation, drying, storing, and packaging processes that facilitates fungal contamination. BIOTROPIA Vol. 28 No. 3,2021 According to Asrul (2009) unfermented cocoa beans and mycotoxigenic fungi contamination, especially aflatoxigenic A.pe%iIIus javus, were problems in Indonesia. Some f m g were isolated from cocoa beans in Central Sulawesi, i.e., AspergiIIus javzls, A. niger, A. fumigatzls, PeniciIlium s p ., Fusarium s p ., Trichodema sp., T. virile, Rhi~0pu.r sp., Mucor sp., Ven'iciIlium sp., and Geotricbm sp. A.pe@IIus fIavu~ was isolated in cocoa beans after being harvested by farmers (7.2 x 108 cfu/mL), collector (4.5 x lo5 cfu/mL), and exporter levels (4.1 x lo3 cfu/mL). Aflatoxin B1 content in cocoa beans after being harvested by farmers (104.80 ppb), collectors (61.31 ppb), and exporters (47.74 ppb). Copetti et al. (2011) reported the occurrence of aflatoxigenic fungi and the presence of aflatoxin in 226 cocoa samples collected in Brazilian farms. The aflatoxigenic fungi isolated in cocoa beans were A . javus, A. parasiticus and A. nomius. A considerable increase in numbers of these species was observed during drying and storage processes. Pires et al. (2019) explained that the total aflatoxins in two cocoa bean samples from their study, from Bahia, from Par6 and from Rond6nia were 13.2, 16.3, 11.7, and 30 pg/kg, respectively. According to Basappa (2009) aflatoxin is a kind of toxins produced by A. flavus and A. parasiticus that causes liver cancer in human and animals. The types of aflatoxins found in foodstuffs and processed products are BI, B2, GI, and Gg but the most dangerous for human health is aflatoxin B1 (AFB1). One of the problems in unfermented cocoa beans is aflatoxigenic fung contamination. Therefore, formulating and concocting a combination treatment to inhibit aflatoxigenic fungi is necessary. Combinations of yeast, acetic acid bacteria, and mangosteen rind extract were used for testing the effectiveness in decreasing aflatoxigenic A. javm population and aflatoxin production. Dharmaputra et al. (201 8a) reported that Imctchenkia orientalis could inhibit 100°/o of ochratoxigenic A. ochraceus. It means that I. orientalis can be used as biocontrol agent, eventhough there is no one research explains that I. orientah can inhibit aflatoxigenic A . javus. Sabahannur and Ralle (2018) reported not only yeast, but Acetobacter aceti also supposed to maintain shelf life of food stuff. Meanwhile Yatman (2012) explained that mangosteen can be used as medicine, because it contains xanthone for antioxidant, antiproliferation, antiimflammatory, antitumor, and anticancer. Research conducted by Akao et al. (2008) showed that xanthone cc-mangostin compound of mangosteen rind exract could inhbit 50% of colon cancer cells growth. The inhibitive nature of the compound is similar with commercially available anticancer drugs, i.e., 5-FU, actinomycin D, and campotechn. Aisha et al. (2012) reported that xanthone extract, cc- mangostin, and y-mangostin inhibited 50% of cancer cell in 6.5 rf: 1.0 mg/mL, 5.1 rf: 0.2 pg/mL, and 7.2 ? 0.4 pg/mL. Therefore, combinations between yeast, acetic acid bacteria (AAB) and mangosteen rind extract (MRE) is expected to produce biocontrol agent on aflatoxigenic A. fIavus growth and decrease aflatoxin production in unfermented cocoa beans to be safe for next processing of chocolate products. The objectives of this research were to test the effects of yeasts, acetic acid bacteria (AAB), and mangosteen rind extract on the aflatoxigenic A. flavzls growth and aflatoxin production in unfermented cocoa beans. It is expected that the research result would show a potential combination of yeast, acetic acid bacteria (AAB), and mangosteen rind extract (MRE) as biocontrol agent to inhbit aflatoxigenic A.flavus to improve food safety in unfermented cocoa beans. MATERIALS AND METHODS Yeast Isolates, Acetic Acid Bacteria, Matoxigenic Fungus, Unfermented Cocoa Beans, and Mangosteen Rind Extract As many as four yeast isolates were used in interaction types and antagonistic test, i.e., 1 isolate of Endomycesjbuhger (Efj BIO 132219 and three isolates of Issatcbenkia orientalis (lo) BIO 211286, BIO 211288, and BIO 211291; only two yeast isolate were used in cocoa beans (in vivo) that showed the highest percentage of inhibition on A. fIavus in vitro, i.e., I. orientalis BIO 21 1291 and BIO 211288. Aflatoxigenic AspetgiIIusjavus BIO 3361/747 were obtained from Phytopathology Laboratory Culture Collections, Antagonistic effect of yeast, acetic acid bacteria, and mangosteen rind extraction - Nurfadila e t aL SEAMEO BIOTROP; and 1 acetic acid bacteria isolate (Acetobacter aceti FNCC0016) was obtained from Food and Nutrition Study Centre, Universitas Gadjah Mada, Yogyakarta. As much as 62 kg of unfermented cocoa beans were obtained from Sumedang Regency, West Java Province. As much as 2 kg of mangosteen rind extracts was obtained from RJ Herbal, Surabaya, East Java Province. Interaction Types Test Between Yeast and Atlatoxigenic A. flavus Four yeast isolates were tested on aflatoxigenic A. jlavzls BIO 3361/747 using direct opposition method (Dharmaputra et al 2018b) (Fig. 1). This method was used to determine the interaction types between the yeasts and aflatoxigenic A. jlavzls. Aflatoxigenic Aspevgillas jlavas BIO 3361 /747 was inoculated after 4 days of each yeast inoculation in the middle of Potato Dextrose Agar (PDA) media in observation on the interaction types was conducted macroscopically between the yeast and aflatoxigenic A. jlavas (Wheeler and Hochng 1993). The interaction types were shown in Table 1. The mathematical equation for calculating the percentage of inhibition between the yeast and aflatoxigenic A. fLavas is as follows: Notes: % I = percentage of inhibition, J1 = diameter of A. jlavus near to petridish, Jz = diameter of A. javas near to yeast. petri dishes (a dameter of 9 cm) with a &stance of 3 cm between each other. The petri dishes Figure 1 Scheme of antagonisms test between yeast isolate and toxigenic A. jlavus; A= toxigenic A. with each fungal and yeast were then incubated flauus, B = yeast isolate, J1 = diameter of A. at room temperature (27 & 2 "C) for 7 days. Five flauus near to petridish, Jz = diameter of A.jlauas replicates were used for each isolates. The near to yeast Table 1 Interaction types between two fungal colonies (Wheeler & Hocking 1993) Tvpe of interaction Description of classification A Mutual intermingling growth, where both fungi grew into each other without any macroscopic signs of interaction Mutual inhibition on contact or space between colonies small (< 2mm) Inhibition of one species on contact, the inhibited species continued to grow at a significantly reduced rate, while the inhibitor species grew at a slightly reduced rate or unchanged Mutual inhibition at a distance (> 2 mm) E Inhibition of one species on contact, the inhibitor species continuing to grow at a reduced rate through the inhibited colony F Inhibition of one species on contact or at a distance, the inhibitor species then continuing to grow at an unchanged rate through or over the inhibited colony a Notes: a: inhibitor fungi, b: inhibited species. BIOTROPIA Vol. 28 No. 3,2021 Obtaining of Unfermented Cocoa Beans and Antagonistic Test of Yeast, Acetobacter Mangosteen Rind Extract acetl; and Mangosteen Rind Extract f ipe lindak cocoa (bulk cocoa) fruits with Combinations on Aflatoxigenic A. ffavus in yellow color were harvested using sterile scissors Vitro from the trees. The cocoa beans were were then cut open using a knife and separated into parts of pods and cocoa beans with pulp. The next process were washing of cocoa beans from pulp and drying using sun-drying for 1 day (9 hours) to become unfermented cocoa beans. Obtaining of mangosteen rind extract is shown in Figure 2. As much as 12 g/L of mangosteen rind extract (MRE) was used in in vitro stage, while 12 g for 500 g of cocoa beans used in in vivo stage. According to Kusumaputri (2011), Dr. Berna Eliya as a phytochemist in Universitas Indonesia, explained that many people consume a glass of boiled water containing mangosteen rind extract. They usually use 60 g of fresh mangosteen rind equal to 12 g of mangosteen rind extract for 1 day. The doses of mangosteen rind extract (MRE) was obtained based on consumer's doses in MRE capsule product, as follows: Antagonistic test between yeast, Acetobacter aceti and mangosteen rind extract on aflatoxigenic A. flavas BIO 3361/747 was conducted using the well test method (Dharmaputra et aL 2016 with modification). Acetobacter aceti FNCC0016, Issatchenkia orientah BIO 211291, BIO 211288, BIO 211286, and Endomycesjbaliger BIO 13221 9) were cultured on Malt Extract Agar (MEA) media and incubated for 7 days, while the Acetobacter aceti FNCCOOl6 was cultured on Peptone Glucose Yeast Extract Agar (PGYA) media and incubated for 3 days at room temperature (27 k 2 OC). Aflatoxigenic Aspe~illusflavzzs BIO 3361/747 was cultured on Potato Dextrose Agar (PDA) media and incubated for 7 days at room temperature (27 k 2 OC). Five pieces (in 5 mm diameter) of pure culture of each yeast were placed into 25 mL of Nutrient Yeast Dextrose Broth (NYDB) media in an erlenmeyer flask (100 mL volume), whle the Acetobacter aceti isolate was placed into 25 mL of Nutrient Broth (NB) media. They were then incubated at 27 k 2 6~ for 7 days, and were shaken for 1 hour every 24 hours for 5 days. The conida cells of aflatoxigenic A.flavzls (5 x 106 cells/mL) were obtained by adding 15 mL of distilled water. then the A. flivas was scratched Notes: DE = doses of mangosteen rind extract (g), JI< = J on the upper surface using sterile inoculation number of capsules in 1 package of mangosteen rind extract product, BE = weight of mangosteen needle. The conidia cells were filtered by sterile rind extract in 1 capsule (mg), and BI< = number gauze on the funnel of erlenmeyer flask (100 mL of capsules to be cdnsumed &I 1 day. - volume). \I/ Grinding of mangosteen rind to become Sun-drying for 3 days mangosteen rind extract (MRE) (7 hours/day) Harvesting of ripe mangosteen fruits I Packaging of 2 kg mangosteen rind extract Sending of 2 kg MRE Separating of fruits and pericarp of mangosteen Figure 2 Stages of obtaining of mangosteen rind extract (MRE) Antagonistic effect of yeast, acetic acid bacteria, and mangosteen rind extraction - Nurfadila et al. The yeast, Acetobacter aceti, and A. flavtls cells were precipitated by centrifugation using a centrifuge at 7,000 rpm fixed angle rotor for 15 minutes and rinsed using twice sterile disulled water, and they were then resuspended in sterile distilledwater until the concentration reached 5 x 10' cells/mL (yeast cells) and 5 x lo6 cells/mL (Acetobacter aceti and aflatoxigenic A. flavm). Yeasts, Acetobacter aceti and A s p e ~ ~ ~ t l s f l a v a s cells were counted using a hemocytometer. A well (5 mm dameter) was prepared using a cork borer in the center of Potato Dextrose Agar (PDA) meda containing 15% cocoa beans juice and PDA media containing 15% cocoa beans juice and 12 g/L of mangosteen rind extract depending on the treatments into a petri dish (9 cm diameter). As much as 20 yL of 5x10' cells/mL yeast cells suspension were placed into the well. The petri dishes were left for 30 minutes to allow penetration of cells suspension into the well. Next, as much as 20 yL of 5 x 106 cells/rnl Acetobacter aceti was inoculated, then A. flavas was also inoculated into the well after 1 hour. Each yeast control was only inoculated with yeast cells suspension. Each treatment and controls were made in 3 replicates (= 5 petri dishes), which were then incubated at room temperature (27 k 2 "C). The growth of the aflatoxigenic A.flavzls, yeasts, and Acetobacter aceti in each petri dish was observed after 7 days of incubation. Total unit experiment for in vitro process was (4 yeast isolates x 2 within or without Acetobacter aceti x 2 kmds of media x 3 replicates) + ((4 yeast control + 1 A. flavtls controlj x 3 replicates) = 63. The radius of aflatoxigenic A. flavzas colony in each petri dish was measured before the colony reached to petri dish. Mathematical equation for the percentage of fermentor inhibition to aflatoxigenic A.flavzls is: Notes: O/O I = percentage of inhibition, Dl = diameter of A. Jlauzls control (mm), Dz = diameter of A. flavzrs in each treatment (mm). Testing of Combination Treatments to Inhibit Aflatoxigenic AspergiLIus ffavus in Vivo As many as two yeast isolates with the hghest percentage of inhibition in vitro stage, Acetobacter aceti FNCC00 1 6, and AspergiIIas flavas BIO 3361/747 were used in vivo stage. The steps in the testing of the combination treatments to inhbit aflatoxigenic A. jlavtls in vivo were similar to in vitro. Each treatments included positive and negative controls using 500 g of unfermented cocoa beans for each replicates. As much as 10 mL of 10' cells/mL yeast cells suspension were placed into 500 g of unfermented cocoa beans depending on the treatments. Each samples was then inoculated or not inoculated by 10 mL of lo6 cells/mL A. aceti (depend on the treatments) after 30 rnin, then as much as 10 mL of lo6 cells/mL aflatoxigenic A.flavas was inoculated into the samples depending on the treatments after 1 hour. Each cells inoculation was conducted sequentially to allow the cells penetration. Total unit experiment for in vivo process was 12 x 4 x 2 = 96 (12 = types of combination including positive and negative controls; 4 = days after inoculation (1,3, 6 and 11); 2 = replication). Determination of Issatchenkia orientafis, Acetobacter aceti, and Aflatoxigenic Aspergillus ffavus Populations, and Aflatoxin Production Yeast, Acetobacter aceti, and A. jlavtls were isolated using serial dilution method (10.' up to followed by pour plate method on Potato Dextrose Agar (PDA) and incubated for 7 days incubation at 27 k 2 u (INS 2008). Aflatoxin contents were determined using Thin Layer Chromatoghraphy (TLC) (Baiton et al. 2006). RESULTS AND DISCUSSION Interaction Types Between Yeast and Aflatoxigenic A. ffavus As many as four yeasts (one isolate of Endomycesjbtllger BIO 13221 9 and three isolates of Issatchenkia orientalis BIO 21 1286, BIO 211288, and BIO 211291) were used in testing the mechanisms of antagonism on aflatoxigenic A. flavtls BIO 3361 /747 using drect opposition method. The interaction type of antagonism mechanism between yeast and I. orientalis BIO 211291 (49.58%) and BIO 211288 (35.79%), and E. jbaiger 132219 (49.29%) with aflatoxigenic A . flavtls was D (Fig. 3a, 3c, 3d; Table 1). This interaction type showed that there BIOTROPIA Vol. 28 No. 3,2021 was a mutual inhibition with the inhibition zone of > 2 mm similar with Wheeler and Hocking (1 993). The interaction type of I. orientalis BIO 21 1286 with A.javvzls BIO 3361 /747 was A (Fig. 3b) where the percentage of each inhibition was 15.630/07 respectively (Table 1). This interaction showed mutual intermingling growth, where both fungi grew into each other without any macroscopic signs of interaction (Fig. 3; Table 1). Based on the percentage of the inhibition of A.3avu-r growth using direct opposition method, three isolates (I. orientalis BIO 21 1291, E. jbulger 13219, and I. orientalis BIO 211288) were higher than I. orientah BIO 211286. It means that I. orientah BIO 211291 and BIO 21 1288, and E. jbulger BIO 13219 were prefer as potential component inhibitor for aflatoxigenic A. j2avu-r in cocoa beans. Nevertheless, the three yeast isolates should be combined with additional materials to increase the ability of inhibition on aflatoxigenic A. javu-r in unfermented cocoa beans. However, this result was strengthened in the next step using the well in vitrO method. Figure 3 Mechanism of interaction between four yeast isolates (Issatchenka orientalis): (a) BIO 211291; (b) BIO 211286; (c) E~domycesjbahger BIO 133219; and (d) I. orientalis BIO 211288); aflatoxigenic A.pe~IIasJavt/s on Potato Dextrose Agar (PDA) media after 7 days of incubation at room temperature (28 + 2 OC) Table 1 Interaction types and the percentage of inhibition between yeast on aflatoxigenic A.perg'IIusJaut/s based on direct opposition method Mean of radius (rnrn) Isolate % Inhibition Interaction type Interaction figure 11 12 Issatchenkia orientah BIO 211291 vs A.jlauas OD I. orientalis BIO 21 1288 vs A. jlauas OD I. orientalis BIO 21 1286 vs 24 20.25 15.63 c A. jlaul~s Endomycesjbahger 28 14.2 49.29 a D BIO 13221 9 vs A.jlauas OD Antagonistic effect of yeast, acetic acid bacteria, and mangosteen rind extraction - Nurfadila et al. Antagonistic Test in Combination of Yeast, Acetobacter aceti and Mangosteen Rind Extract on Matoxigenic Aspergirrus flavus In Vitro According to Richard and Prusky (2002) yeast has some unique charactheristics such as fast growth, the ability to colonize surface of fruit, and the ability to join in nutrition competition with a pathogen, therefore it can be a biocontrol agent. Yeast also has an important role in fermentation, because it can convert glucose and maltose through anaerobic respiration. The well method is a method for determining on the percentage of inhibition between yeast and pathogenic fungi. All treatments except yeast controls were not inoculated by aflatoxigenic A. flavzts. Widiyanto et al. (2013) reported that unfermented cocoa beans has no any glucose, because the pulp has been removed. It means that additional material is needed to maintain beneficial microorganisms survival in unfermented cocoa beans. According to Maligan et al. (2018) mangosteen rind extract (MRE) contents in 100 g are 82.50% carbohydrate, 6.45% fats, 3.02% proteins, 5.87% water, and 2.10% total glucose. The nutritional content of MRE may increase the yeast survival. MRE also has xanthone for anticancer, antihyperglicemic, and antioxidant for human health. Other research, Acetobacter is one of bacteria that can oxidize glucose to gluconic and other organic acid in the same time to maintain the shelf life of foodstuff (Simanjuntak et al. 201 6). Therefore, in this research, each of four yeast isolates were combined with acetic acid bacteria and MRE in vitro stage to test the effectiveness of those treatment in different media (PDA + 15% cocoa beans juice with and without 12 g/L MRE). The highest percentage of inhibition on aflatoxigenic A. flavzls BIO 3361/747 was 100%, in treatment with Issatchenkia orientalis BIO 211288 + Acetobacter aceti FNCCOOl6 on Potato Dextrose Agar (PDA) + 15% cocoa beans juice + 12 g/L mangosteen rind extract (MRE) media (Fig. 4a; Table 2). It means those treatment more effective than other treatments to against the aflatoxigenic A.flauus in vitro. The 2nd highest percentage of inhibition was 51.98%, in treatment I. orientalis BIO 211291 on PDA + 15% cocoa beans juice + 12/g MRE (Fig. 4b; Table 2). Hafsari (2011) explained that the differences in fungal diameter indicated that yeast growth is faster than that of fungi and it also obtained more nutrition than pathogenic fungi. The stunting growth of fungi was shown from the diameter of colony that was lower than the control (+). According to Janisiewicz and Korsen (2002) the mechanism of space and nutrition competitions could happened if the yeast's effort was higher than pathogenic fungi to get nutrition and space. The similar research from Golubev (2006) explained that the capability of yeast antagonism would increase on other microorganism from different habitat, because the fungi are a new competitor that should be defeated to become a dominant in the available space and nutrition. Af BlO 3361 did not grow lo BIO 211291 Figure 4 (a) Issatchenkia orientalis BIO 211288 and (b) I. orientalis BIO 211291 vs toxigenic A.Jauzis BIO 3361/747 with Acetobacter aceti on Potato Dextrose Agar + 15% cocoa beans juice + 12 g MRE at room temperature (27 + 2 OC) after 7 days of incubation BIOTROPIA Vol. 28 No. 3,2021 Table 2 Effect of different combination and media treatments on the percentage of inhibition and growth of yeast, acetic acid bacteria, and aflatoxigenic A.ji'auzrs Treatment Diameter (mm) Inhbition Combination Meha Yeast Acetobacter aceti AspepxiIIus flavus YO) Issatcbenkia orientah 291 + A.flavus 15.45 37.20 20.00 I. orientalis 288 + A.flavus PDA + 15% cocoa 27.35 30.15 25.56 I. orientalis 286 + A.flavus beans juice ND 40.00 9.03 Endomycesjbul&er 21 9 + A.flauus ND 42.10 9.46 I. orientah 291 + A.flavus PDA + 15% cocoa 25.50 34.17 26.52 I. orientalis 288 + A.flavus beans juice + 12 g/L 34.47 25.23 45.74 I. orientalis 286 + A.flavus mangosteen rind ND 39.79 14.43 E. fibulzker 21 9 + A. flavus extract ND 38.13 18.00 I. orientalis 291 +Acetobacter aceti + 11.21 ND 30.60 34.1 9 Aspergilasflavus PDA + 15% cocoa I. orientalis 288 +A. aceti + A.fIauus beans juice 30.01 ND 24.57 47.16 I. orientah 286 +A. aceti + A. flavus ND ND 37.92 18.45 E.Jibuliger 21 9 + A. aceti + ~ . ~ a v u s ND ND 39.44 15.18 I. orientah 291 +A. aceti + A.flavus PDA + 15% cocoa 29.98 5.26 22.33 51.98 I. orientalis 288 +A. aceti + A.flauus beans juice + 12 g/L 34.45 18.62 ND 100.00 I. orientah 286 +A. aceti + A . flauus mangosteen rind ND ND 40.87 12.11 E.-fibuker 21 9 + A. aceti + A. flauus extract ND ND 39.77 14.47 I. orientalis 291 (control) 35 0 I. orientah 288 (control) 40 PDA + 15% cocoa 0 I. orientalis 286 (control) 15 beans juice 0 E$buhger 21 9 (control) 11 0 Alflavus (control+) 46.50 0 Notes: (-) = not inoculated; ND = not detected; 0 = no inhibition (control). Based on different yeast isolates, I. orientalis BIO 21 1291 and I. orientalis BIO 211288 were potential as a component inhibitor for A . j a v u s BIO 3361/747 in vitro. Based on different agar media, PDA + 15% cocoa beans juice + 12 g/L MRE media was more effective than PDA + 15% cocoa beans juice meda to inhibit aflatoxigenic A. flavus in vitro. It means that MRE influenced on decreasing of aflatoxigenic A. javus growth in vitro. Yatman (20 12) reported that xanthone of mangosteen rind extract can be used as antimicrobial, antioxidant, antifungi, and anticancer. Populations of Issatchenkia orientalis, Acetobacter acetl; and Aspergigus ffa vus In Vivo Two yeast isolates with the highest percentage of inhibition on aflatoxigenic A. javus in vitro were Ismtcbenkia orientalis BIO 21 1291 and BIO 211288. The yeast isolates were used in vivo stage. The highest population of yeast I. orientalis of cocoa beans in 1 day after inoculation was 5.88 log cfu/g, found in samples inoculated by with I. orientalir BIO 211291 + mangosteen rind extract (MRE) + A.javus. The highest I. orientalis population in 3 and 6 days after inoculation were 4.67 and 3.75 log cfu/g, found in samples inoculated by I. orientalis BIO 211291 + BIO 211288 + A. aceti + MRE + A. javus. The highest I. orientalir population in 11 days after inoculation was 2.82 log cfu/g, found in samples inoculated by I. orientalis BIO 211291 or BIO 211288 + MRE + A. flavus Fable 3). The negative and positive controls, sample with Acetobacter aceti + mangosteen rind extract + A. flavus were not inoculated by I. orientah. Based on dfferent days after inoculation, commonly I. orientalir population in 3 day after inoculation was higher than 1, 6 and 11 days after inoculation. Yeast population in all samples with MRE were not sipficant dfferent with samples without MRE. Jamili et al. (2016) reported the dominant yeasts were found in cocoa beans during fermentation, i.e., 1 isolate Candida kmsei, 3 isolates C. tropicalis, 1 isolate Saccbaromycopsis jbuhgera, 1 isolate Kloeckera sp., and 1 isolate Saccbaromyces cerevisiae. According to Ren et al. (2020) many microorganisms including bacteria, non-toxigenic fungi, and yeast strains have been investigated as potential biocontrol agents against to aflatoxigenic fungi. Lee et aL(2008) reported that peroxisomal 3-ketoagl-CoA tbiolase BIOTROPIA Vol. 28 No. 3,2021 influence the ability of A. j a v m growth during could be used to determine the ability of fungal storage are temperature, relative humidity, and growth. Not only temperature, relative humidity moisture content. Norlia et al. (2019) also and water activity, but also CO2 levels influence reported that relative humidity and water activity the fungal growth (Giorni e t aL 2018). (a,) in foods are interrelated to each other and Table 3 Issatchenkia orientah, Acetobacter aceti and A.pe@IIzls Yauzls populations in unfermented cocoa beans with combination treatments since 1 until 11 days after inoculation Issatchenkia orientah population Acetobacter aceti population Aspe@Izlsjlauzls population Treatment 0% cfu/g) 0% cfu/g) 0% cfu/g) Day after inoculation Day after inoculation Day after inoculation Negative control (without - 3.97 inoculation of microorganisms) I. orientalis 291 + I. orientalis 288 + 2.52 3.80 2.90 2.73 - - 2.52 2.48 0.48 0.48 MRE +A.flautls I. orientalis 288 + 4.38 4.30 3.20 2.82 - MRE + A. flauus I. orientalis 291 + 5.88 3.75 2.82 2.82 - MRE + A . flautls *I . orientalis 291 + I. orientalis 288 + Acetobacter aceti + 3.12 4.67 3.75 2.52 4.22 3.30 2.75 3.12 3.75 ND ND ND MRE +A.-flautls I. orientah 291 + I. orientalis 288 + Acetobacter aceti + 3.43 4.12 2.78 1.94 3.52 4.75 3.30 2.14 0.48 1.48 1.23 ND A. jYauus I. orientalir 288 + Acetobacter aceti + 2.00 4.12 2.43 ND 2.43 2.48 2.88 2.64 0.48 0.48 3.08 2.22 A. fl.uus I. orientalis 291 + Acetobacter aceti + 3.07 4.64 2.12 ND 3.26 4.73 2.37 2.37 2.74 ND 1.00 1.52 MRE +A.flauus *I . orientah 288 + Acetobacter aceti + 3.22 4.12 2.12 ND 3.75 3.56 2.12 2.00 0.48 ND ND ND MRE +A. flauus I. orientalis 291 + Acetobacter aceti + 3.56 3.60 2.67 ND 2.12 2.48 2.87 2.80 2.75 2.00 2.94 3.48 A. flauzls A. aceti + MRE + A. flnutls Positive control (A. flauzls) > " , Notes: MRE = mangosteen rind extract; ND = not detected; (-) = not inoculated. Antagonistic effect of yeast, acetic acid bacteria, and mangosteen rind extraction - Nurfadila e t al. 0 0 0 9 9 9 e - 3 v v v 0 0 0 9 9 9 3 3 3 v v v 0 0 0 9 9 9 3 3 3 v v v 0 0 0 9 9 9 3 3 3 v v v * * o 2 2 r q i v v s f 4 a, 0 * * o O 2 2 0 3 v v s l ; 0 * * m a 2 2 0 9 v v s h " a, a, d- 2 2 2 2 v v m I I 6 0 0 0 .- L? m X X " a, v v v g Lo 0 o o m L? L ? : 1 1 m m v V V 8 . .a 0 0 0 g X X X O v v v : I I 0 0 0 X 2 x 6 U v V V G 6- 0 o w 0 2 9 & v V X G 0 v v 0 0 0 , j h! " v 2 2 2 v r - i i - U cd Li U 0 0 - 3 X 2 : a -g g - 5 4 2 % q & G 9 7 3 s + g g p : Y 1 1 " -2 2 8 2 4 5 + " 2 SP: ;, g .. .% + $ .* $ :g g BIOTROPIA Vol. 28 No. 3,2021 Range of temperature and relative humidity I. orientalis BIO 21 1288 + A. aceti + mangosteen in the storage room after cocoa beans rind extract, because A . j a v a ~ was not grown in inoculation with various treatments were 27 - the samples for 3 until 11 days after inoculation. 28 OC dan 78 - 80%. Water activity and CO2 Based on aflatoxins content, all combination levels were not determined in this research. Suttajit (2014) reported that the optimum temperature for fungal growth was 25 - 40 OC. Aflatoxins content in positive control were aflatoxin B1 (74.01 ppb), G1 (54.05 ppb), and total aflatoxin (128.06 ppb), while all sample treatments included negative control contained aflatoxin B1 were lower than the limit detection, i.e., BI (< 2.20 ppb), Bz (< 3.50 ppb), GI (< 0.54 ppb), and G2 (< 1.00 ppb) (Table 4). There were no significant differences between combination treatments on aflatoxin production. It means that all unfermented cocoa beans with combination treatments within or without mangosteen rind extract could be processed into chocolate products, because the aflatoxins were relatively safe. According to Mazumder and Sasmal (2001) the maximum tolerable limit for aflatoxin in cocoa beans, cocoa butter, and cocoa powder in Bulgaria are 5 ppb, while in Uruguay and Malaysia the limits are 10 ppb. According to Scott and Pryzbylski (2020) the range of aflatoxins in raw cocoa beans from Trinidad and Ghana was 8 - 35 pg/kg. Maciel et al. (2018) reported as much as 38% cocoa beans in southern region of Bahia, Brazil were contaminated by aflatoxin in the range < LOD- 17.795 pg/kg, 25% and 18% of total samples were contaminated by AFBl and ochratoxin A in the range of