J. Microbiol. Biotechnol. (2012), 22(12), 1740–1748http://dx.doi.org/10.4014/jmb.1207.07048First published online October 5, 2012pISSN 1017-7825 eISSN 1738-8872

Isolation and Identification of Fungi from a Meju Contaminated with Aflatoxins

Jung, Yu Jung1, Soo Hyun Chung

2*, Hyo Ku Lee

1, Hyang Sook Chun

3, and Seung Beom Hong

4

1Department of Food Science and Technology, Kongju National University, Yesan 340-702, Korea2Department of Food and Nutrition, College of Health Science, Korea University, Seoul 136-703, Korea3Food Safety Research Division, Korea Food Research Institute, Sungnam 463-746, Korea4Korean Agricultural Culture Collection, National Academy of Agricultural Science, RDA, Suwon 441-707, Korea

Received: July 24, 2012 / Revised: September 17, 2012 / Accepted: September 28, 2012

A home-made meju sample contaminated naturally with

aflatoxins was used for isolation of fungal strains. Overall,

230 fungal isolates were obtained on dichloran rosebengal

chloramphenicol (DRBC) and dichloran 18% glycerol

(DG18) agar plates. Morphological characteristics and

molecular analysis of a partial β-tubulin gene and the

internal transcribed spacer (ITS) of rDNA were used for

the identification of the isolates. The fungal isolates were

divided into 7 genera: Aspergillus, Eurotium, Penicillium,

Eupenicillium, Mucor, Lichtheimia, and Curvularia. Three

strains from 56 isolates of the A. oryzae/flavus group were

found to be aflatoxigenic A. flavus, by the presence of the

aflatoxin biosynthesis genes and confirmatory aflatoxin

production by high-performance liquid chromatography

(HPLC). The predominant isolate from DRBC plates was

A. oryzae (42 strains, 36.2%), whereas that from DG18

was A. candidus (61 strains, 53.5%). Out of the 230 isolates,

the most common species was A. candidus (34.3%) followed

by A. oryzae (22.2%), Mucor circinelloides (13.0%), P.

polonicum (10.0%), A. tubingensis (4.8%), and L. ramosa

(3.5%). A. flavus and E. chevalieri presented occurrence

levels of 2.2%, respectively. The remaining isolates of A.

unguis, P. oxalicum, Eupenicillium cinnamopurpureum, A.

acidus, E. rubrum, P. chrysogenum, M. racemosus, and C.

inaequalis had lower occurrence levels of < 2.0%.

Keywords: Meju, fungi, aflatoxigenicity, fungal frequency

In Korea, the term meju is used to describe dried fermented

soybeans that have been formed into a block. The meju

is an important starter material for Korean fermented

soybean products and strongly determines the quality of

the products including soybean paste, soy sauce, and

kochujang [20, 23]. The quality of traditional meju is

influenced by the metabolism of microorganisms during

the fermentation process. It is known that most of the fungi

grow predominantly on the dried and air-contacted surface

of the meju during its fermentation, whereas bacteria are

usually present inside the meju where the oxygen level

is low [5, 39]. The fungi presented in the meju were

recognized as effective microorganisms for fermentation

and composed mostly of Aspergillus oryzae, Mucor spp.,

and Penicillium spp. [21, 22].

Currently, large quantities of fermented soybean products

are manufactured commercially, and inoculation with A.

oryzae is used for mass fermentation [25]. In the case of

home-made meju, manufactured by traditional methods,

the quality of the meju depends on natural fermentation

[23], and differences in the quality of the product can occur

because of microbial diversity in place and time of

production [19]. The fungi that grow in meju or soybean

products are generally regarded as safe. However, it is

possible for home-made meju to be contaminated with

mycotoxin-producing fungi, which can produce mycotoxins

under the natural fermentation environment.

Mycotoxin contamination on agricultural commodities

has attracted worldwide attention because of its adverse

effects on human health, poultry, and livestock [4, 13].

Some mycotoxins are carcinogenic, mutagenic, teratogenic,

nephrotoxic, and immunosuppressive agents [8]. In particular,

it is possible for meju to be contaminated with aflatoxins

produced by A. flavus or A. parasiticus-those have the

morphological and biochemical similarities with A. oryzae

or A. sojae [20, 25, 29].

The method used for the detection and identification of

fungi has been dependent on the morphological and

cultural characteristics of the fungi. To date, molecular

techniques have greatly improved our understanding of

fungal ecology and have revolutionized the tools available

for exploring environmental fungal communities [12]. In

*Corresponding authorPhone: +82-2-940-2854; Fax: +82-2-941-7825;E-mail: chungs59@korea.ac.kr

1741 Jung et al.

particular, the specific amplification and the sequence of

the internal transcribed spacer (ITS) region of fungal DNA

(rDNA) are widely used for the phylogenetic study of

fungi [10, 14, 38]. The β-tubulin gene also has high

differentiation, and its regional sequence is relatively easy

to obtain [9, 15].

Currently, there is an increasing demand for the safety

verification of fungi that grow in traditional foods. To date,

there are only limited reports on the aflatoxigenic fungi

and fungal flora in meju. Park et al. [25] collected meju

samples from the southern area of Korea and performed

direct competitive enzyme-linked immunosorbent assays

to show that some Aspergillus isolates had produced

aflatoxins. However, the immunochemical methods used

to detect aflatoxigenic fungi can sometimes give false-

positive results [36]. Molecular approaches to detect

aflatoxigenic fungi, including polymerized chain reaction

(PCR) and gene sequencing, have been used recently with

improved simplicity and sensitivity. Previously, we used

multiplex PCR analysis for detecting aflatoxigenic fungi

from meju samples in Korea and found that 4 of the 65

isolates of Aspergillus section Flavi were potentially

aflatoxigenic strains [16].

In this study, 230 fungal isolates were selected from a

home-made meju sample manufactured using traditional

methods, in which aflatoxins were contaminated naturally.

The fungal isolates were identified by morphological and

molecular characteristics, and the phylogenetic positions

of the isolates were obtained from the fungal β-tubulin and

ITS sequence. The fungal composition of the meju sample

was presented, and we used PCR assay and HPLC to

determine which isolates were potential aflatoxigenic strains.

MATERIALS AND METHODS

Source of Sample and Fungal Isolation

A home-made meju sample naturally contaminated with aflatoxins

(aflatoxin B1 and B2: 210 ppb) was manufactured by the traditional

method in Chungnam Province and used for the isolation of fungal

strains. The meju was finely ground using a laboratory blender, and

20 g of the sample was added to 180 ml of peptone water 0.1% (w/v)

and maintained at room temperature for approximately 30 min. This

mixture was then shaken, and serial dilutions were obtained. One

hundred µl of each dilution was spread onto the surface of 2 types

of solid media: DRBC agar [peptone 5 g, glucose 10 g, KH2PO4

0.1 g, MgSO4·7H2O 0.05 g, dichloran (0.2% in ethanol) 1.0 ml, rose

bengal 0.025 g, chloramphenicol 0.1 g, agar 15 g, and distilled water

1 L] and DG18 agar [peptone 5 g, glucose 10 g, KH2PO4 0.1 g,

MgSO4·7H2O 0.05 g, dichloran (0.2% in ethanol) 1.0 ml, glycerol

220 g, chloramphenicol 0.1 g, agar 15 g, and distilled water 1 L]

[28]. The plates were incubated for 1 week at 25oC. On the last day

of incubation, the fungal colonies were transferred to Czapek agar

(CA) slants, and were incubated for 7 days at 25oC for further study.

Each species isolated from the sample was considered as an isolate.

Morphological Classification

To observe macro- and microscopic characteristics of the colonies,

the fungi were grown on CA, malt extract agar (MEA), and potato

dextrose agar (PDA) plates for 5 days at 25oC. Next, the conidial

heads, conidiophores, vesicles, conidia shapes, and roughness of the

conidial walls were observed under a microscope. Each strain was

identified in genus level according the standard methods provided

by Pitt and Hocking [28] and Samson et al. [31, 32]. Experiments

were conducted twice with 3 replicate plates.

Molecular Identification

The fungal isolates were grown on potato dextrose broth for 7 days

at 25oC. The mycelia were harvested from the plates and the total

genomic DNA was extracted using the DNeasy Plant Mini-Kit

(Qiagen, Valencia, CA, USA). A partial sequence of the β-tubulin

gene was amplified using 2 primers: Bt2a (5'-GGTAACCAAATC

GGTGCTGCTTTC-3') and Bt2b (5'-ACCCTCAGTGTAGTGACC

CTTGGC-3') [9]. The ITS region (including ITS1-5.8S rRNA-ITS2

region) of the rDNA were amplified using primers ITS1 (5'-

TCCGTAGGTGAACCTGCGG-3') and ITS4 (5'-TCCTCCGCTTAT

TGATATGC-3') [38]. Genomic DNA of the β-tubulin and the ITS

were added to the PCR reactions containing forward and reverse

primers, and PCR was performed using the thermal cycler PC708

(ASTEC, Japan). Each PCR reaction was performed in a total

volume of 50 µl, containing 5 µl of 10× PCR buffer, 3 µl of

deoxyribonucleotide triphosphate (dNTP; 2.5 mM), 0.4 µl of each

primer (100 pmol), 0.3 µl of Taq DNA polymerase (Solgent, Korea),

39.9 µl of sterile deionized water, and 1 µl of DNA template. The

PCR conditions were as follows: 4 min at 95oC; denaturation for

1 min at 95oC; annealing for 1 min at 65oC (β-tubulin) or 1 min at

54oC (ITS region); extension for 2 min at 72oC (35 cycles); and a

final extension for 7 min at 72oC. Amplicons were separated using

the Mupid-2 Plus submarine electrophoresis system (Advance, Japan)

on 1.5% (w/v) agarose gels. The gels were stained with ethidium

bromide and bands were visualized with a UV transilluminator. The

above-mentioned amplicons were also used for sequencing analysis

and were purified with the EzWay PCR Clean-up kit (KOMA

biotech, Korea). Finally, the purified PCR products were sequenced

by Solgent Co. Ltd (Daejeon, Korea).

DNA Sequence and Phylogenetic Analysis

DNA sequences were edited using the DNASTAR computer package,

and sequence alignment was performed using the CLUSTAL W

program [34]. These sequences were used with the BLAST program

(http://www.ncbi.nlm.nih.gov/BLAST) to identify the fungi. The

neighbor-joining (NJ) method was used for phylogenetic analysis, in

which the data were first analyzed using the Tamura–Nei parameter

distance calculation model with g-distributed substitution rates, and

then an NJ tree was constructed using MEGA version 4.0 [33]. A

bootstrap analysis was performed with 1,000 replications as confirmation

of each clade.

Multiplex PCR Analysis for Screening Aflatoxigenicity

In order to screen for aflatoxigenic isolates from the A. oryzae/flavus

group, 4 pairs of primers were used. One regulatory gene (aflR) and

3 structural genes (omtA, omtB, and ver-1) were amplified using the

appropriate primers, and 2 different sets of 2 primers, primer set I

(aflR/omtA) and primer set II (omtB/ver-1), were combined for

FUNGAL FLORA IN A MEJU CONTAMINATED WITH AFLATOXIN 1742

performing multiplex PCR by considering a consistent amplification

pattern for every expected amplicon. The PCR mixture reactions

and condition were performed as described in a previous report [16].

PCR products were electrophoresed on a 1.5% (w/v) agarose gel

with a 1 kb plus DNA size marker (Solgent Co. Ltd, Korea). A.

flavus KACC41403 from the Korea Agricultural Culture Collection

(Suwon, Korea) was used as a standard strain for aflatoxin production.

HPLC Analysis of Aflatoxigenicity

The production of aflatoxins by 56 isolates of the A. oryzae/flavus

group was determined using HPLC. Individual fungal spores were

removed from the mycelium after 7 days of growth at 25oC on a

PDA slant medium and suspended in sterile 0.05% (v/v) Tween 80.

A 0.1 ml aliquot of the spore suspension was used to inoculate into

50 ml of sterile Czapek yeast-extract (CY; NaNO3 3 g, KH2PO4 1 g,

KCl 0.5 g, MgSO4·7H2O 0.5 g, FeSO4·H2O 0.01 g, yeast extract 5 g,

sucrose 30 g, and distilled water 1 L), which was used for the

production of aflatoxins by the fungi. The inoculated flasks were

incubated for 14 days at 25oC. The fungal culture broth was filtered

through Whatman No.1 filter paper, and 10 ml of the filtrate was

then diluted with phosphate-buffered saline (pH 7.5) to 100 ml. The

mixture was passed through a Whatman GF/A glass filter and 50 ml

of the filtrate was loaded onto an immunoaffinity column (AflaTest

WB, Vicam Co., USA) at a flow rate of approximately 1 drop per

second for clean-up. After washing the column with 10 ml of water

at the same flow rate, aflatoxin was eluted with 2 ml of methanol.

The eluate was evaporated at 40oC under a stream of N2 until dry.

The dry residue was derivatized by adding 200 µl of trifluoroacetic

acid, and the mixture was left to stand for 30 min before it was

diluted with 800 µl of acetonitrile-water [10:90 (v/v)]. This derivatized

sample was filtered through a 0.22 µm membrane filter, and the

filtrate was used for HPLC analysis. Separation of aflatoxins B1, B2,

G1, and G2 from the injected 50 µl of samples was carried out using

a Nova-Pack C18 column (150 mm, 3.9 mm i.d., 4 µm; Waters,

USA). The mobile phase was acetonitrile-methanol-water [17:17:66

(v/v/v)], pumped at a constant flow rate of 0.5 ml/min. The quantitative

determination of each aflatoxin was carried out using a fluorescence

detector (excitation: 360 nm; emission: 440 nm).

RESULTS AND DISCUSSION

Morphological Classification of Fungal Isolates from

Meju

A total of 230 fungal isolates were obtained from the meju

sample that was naturally contaminated with aflatoxin. Of

all the isolated strains, 116 were from the DRBC plates

and 114 were from the DG18 plates. The cultural and

microscopic characteristics of the isolates were analyzed to

classify the fungal genus using adequate keys [28, 31, 32].

Most isolates had typical morphological features, which

classified them into 1 of 7 fungal genera: Aspergillus,

Eurotium, Penicillium, Eupenicillium, Lichtheimia, Mucor,

and Curvularia. The isolates of Eurotium spp. from the

DG18 plates showed reduced growth when transferred onto

PDA, and had similar anamorph and cleistothecia in

colony morphology and microscopic features. Most of the

other isolates showed faster growth on PDA, MEA, and

CA plates than Eurotium spp. at 25oC. Each fungal isolate

classified at the genus level was used in the next molecular

analysis for further identification.

Molecular Analysis for Identification of Fungal Isolates

DNA sequences with concordance analysis can provide

information that enables the identification of a fungal

species [27]. To identify the species of the 230 isolates,

molecular analysis was conducted using the β-tubulin gene

for Aspergillus, Eurotium, Penicillium, and Eupenicillium

spp. and rDNA-ITS gene for Mucor, Lichtheimia, and

Table 1. Molecular identification of fungal isolates from the meju sample using gene sequencing of β-tubulin and the ITS region.

Fungal species Gene regionsProducts

(bp)No. of base differences

(Identity, %)GenBank accession no.

Aspergillus acidus

β-Tubulin

510 0 (100) JF450869

Aspergillus candidus 517 0 (100) EU014092

Aspergillus oryzae/flavus 497 0-1 (99-100) EF661486, EF661483

Aspergillus tubingensis 506 0 (100) HE577808

Aspergillus unguis 402 0 (100) EF652333

Eurotium chevalieri 412 0 (100) EF651913

Eurotium rubrum 403 1 (99) EF651922

Eupenicillium cinnamopurpureum 432 0 (100) EF506216

Penicillium chrysogenum 431 0 (100) EF198568

Penicillium oxalicum 492 5 (99) JF521520

Penicillium polonicum 436 0 (100) EU128570

Lichtheimia ramosa

ITS1–5.8S–ITS2

816 2 (99) HQ285692

Mucor circinelloides 599 0 (100) DQ118989

Mucor racemosus 597 0 (100) HQ285603

Curvularia inaequalis 560 2 (99) AF313409

1743 Jung et al.

Curvularia spp., respectively. The identity of the isolates

and PCR product sizes are shown in Table 1, and the

phylogenetic trees produced using the NJ method are

presented in Fig. 1.

Ten species were identified by the sequencing of the β-

tubulin gene, with the exception of the A. oryzae/flavus

group, and 4 species were identified by the sequencing of

the rDNA-ITS region. The sizes of each fungal PCR

amplicon indicated its identity according to the reference

strains from the GenBank database. Five clusters of

Aspergillus were found in the meju sample after β-tubulin

gene analysis, including A. acidus, A. candidus, A.

tubingensis, and A. unguis, and their PCR amplicon size

and homology were identical (100%) to the reference

strains from the GenBank database. In the case of the A.

oryzae/flavus group, their cluster had a same-sized PCR

Fig. 1. Phylogenetic tree of the fungal genera isolates from the meju sample. (A) Aspergillus, Eurotium, Penicillium, and Eupenicillium: β-tubulin gene. (B) Mucor, Lichtheimia, and Curvularia: rDNA-ITS region. The tree was

constructed using neighbor-joining analyses of the β-tubulin and rDNA-ITS region gene sequences. The numbers above the nodes represent bootstrap

values (out of 1,000 bootstrap replications).

FUNGAL FLORA IN A MEJU CONTAMINATED WITH AFLATOXIN 1744

amplicon (497 bp) and high homology (99-100%) with

either reference strains of A. oryzae or A. flavus. E. rubrum

and E. chevalieri were identified with 99-100% homology

within each cluster. Three clusters of Penicillium and 1

cluster of Eupenicillium—P. chrysogenum, P. oxalicum, P.

polonicum, and Eupenicillium cinnamopurpureum—were

obtained from the meju sample, and each cluster showed a

highly consistent sequence (99-100%) when identified

from their relevant species. The phylogenetic tree constructed

from rDNA-ITS sequencing showed that 3 clusters of

Zygomycetes (M. circinelloides, M. racemosus, and L. ramosa)

and C. inaequalis were present in the meju sample. M.

circinelloides and M. racemosus had same-sized PCR

products and 100% sequence consistency with the reference

strains, and L. ramosa and C. inaequalis had 99% homology

with reference strains from the GenBank database.

Both A. flavus and A. oryzae belong to the Aspergillus

section Flavi, and A. flavus could be classified as a

separate species; however, it is almost genetically identical

to A. oryzae. Chang and Ehrlich [2] suggested that A.

oryzae may be a variant morphotype of typical A. flavus;

and Rank et al. [29] reported that the gene homology of the

two species was ca. 99.5%. Although the isolates of the A.

oryzae/flavus group from the meju sample showed similar

morphological characteristics (conidial heads in the shades

of yellow-green to brown, usually metulated vesicles and

dark clerotia) and β-tubulin gene homology (99-100%),

they could be divided into A. flavus and A. oryzae

according to the difference in aflatoxigenicity. It is known

that A. oryzae does not produce aflatoxins despite its close

relationship with A. flavus [1, 18, 37].

Aflatoxigenicity of Aspergillus oryzae/flavus Group

The aflatoxigenicity of 56 isolates belonging to the A.

oryzae/flavus group from the meju sample was examined

using multiplex PCR assay and HPLC confirmation. It is

known that genes involved in the aflatoxigenic biosynthesis

pathway could be used to differentiate aflatoxigenic

Aspergillus fungi from non-aflatoxigenic Aspergillus spp.

[3, 6, 35]. Molecular approaches such as PCR are reported

to be efficient for the detection of aflatoxigenic fungi,

because PCR can be conducted in vitro and is specific and

sensitive [24]. Previously, we had developed a multiplex

PCR assay for the detection of aflatoxigenic Aspergillus

isolates from meju samples [16]. The result showed that

multiplex PCR may be used for detecting the presence of

aflatoxigenic Aspergillus species in meju samples. We

used this assay to screen the aflatoxigenic strains from the

56 isolates of the A. oryzae/flavus group. Two sets of 2

primers were used to amplify the genes, including aflR,

omtA, omtB, and ver-1. The genes of aflR and omtA were

used to successfully detect aflatoxigenic strains by other

researchers because aflR is known to be an aflatoxin

biosynthesis regulatory gene and omtA is known to be an

aflatoxin biosynthesis structural gene.

Among the 56 isolates of the A. oryzae/flavus group, 5

isolates (KUFNM018, 027, 044, 084, and 098) showed

complete amplification for both primer sets used in the

multiplex PCR assay, and the remaining 51 isolates (91%)

showed only 3 bands after multiplex PCR, which represented

the amplified omtA, omtB, and ver-1 genes, respectively;

the regulatory gene aflR was deleted (Fig. 2 and Table 2).

In Fig. 2, the lanes with 4 amplicon bands were obtained

from A. flavus KACC 41403 (used as a standard strain for

aflatoxin production) and 5 isolates (KUFNM018, 027,

044, 084, and 098) of A. oryzae/flavus; the lanes with 3

bands were from the remaining 51 isolates (showing 5

representative isolates: KUFNM119–123).

The 56 isolates were then tested for aflatoxin production

by using culture filtrates, and the results are shown in

Table 3. The 5 isolates that showed 4 amplicon bands

including aflR, omtA, omtB, and ver-1 were divided in 2

Fig. 2. Agarose gel electrophoretic pattern of PCR products obtained from genomic DNA of the Aspergillus oryzae/flavus group. (A) Primer set I. (B) primer set II. M (bp): 1 kb plus DNA ladder.

1745 Jung et al.

subgroups: aflatoxin producers (KUFNM027, 044, and

084) and aflatoxin non-producers (KUFNM018 and 098),

which were then designated as aflatoxigenic A. flavus and

non-aflatoxigenic A. flavus, respectively. This result was

consistent with results published by other researchers.

Criseo et al. [6] reported that 36.5% of 134 non-aflatoxigenic

A. flavus contained 4 genes (aflR, omtA, ver-1, and nor-1)

as like aflatoxigenic strains; Degola et al. [7] showed that

A. flavus isolates containing 5 genes (aflD, aflO, aflaQ,

aflR, and aflS) were separated into aflatoxin producers and

non-producers. However, these studies did not clearly

discriminate between aflatoxigenic and non-aflatoxigenic

A. flavus, because only selected genes of the aflatoxin

synthetic pathway were analyzed. In addition, an A. flavus

strain that contains all the genes of the aflatoxin synthetic

pathway still requires other various molecular considerations

to be made, such as post-transcriptional level and/or

protein levels. Therefore, for the accurate detection of

aflatoxigenicity, immunological or cultural methods should

be applied. In this study, quantitative HPLC results showed

that the production of aflatoxin B1 (AFB1) and aflatoxin B2

(AFB2) by aflatoxigenic A. flavus strains was 114 µg/ml and

21µg/ml (KUFNM027), 55µg/ml and 3µg/ml (KUFNM044),

and 25 µg/ml and 6 µg/ml (KUFNM084), respectively.

The chromatograms for the aflatoxin production by the 3

aflatoxigenic A. flavus are shown in Fig. 3. In the case of

the 51 isolates with three bands of PCR amplicons (omtA,

omtB, and ver-1), their culture contained no detectable

amount of aflatoxins; therefore, they were confirmed to be

non-producers of aflatoxin. Previously, the aflR gene has

been used to discriminate between aflatoxin-producing and

aflatoxin-non-producing fungi [6]. Our result showed that

the 51 aflR-lacking isolates were non-aflatoxigenic fungi,

and they were classified as A. oryzae for further study.

In addition, it is assumed that the 3 isolates of aflatoxigenic

A. flavus might produce considerable amounts of aflatoxins

during meju fermentation; the contents of AFB1 and AFB2

in the meju sample were 206.2 ng/g and 3.8 ng/g, respectively.

However, the aflatoxin content in soybean paste and soy

sauce fermented for 6 months using meju blocks that had

the same origin (same household and same production

time) as the meju sample in this study was 7.1 ng/g and

Table 2. Genetic patterns of 56 Aspergillus oryzae/flavus group isolates from the meju sample.

Genetic pattern Number of isolates

Gene presence detected by multiplex PCR

Primer set I Primer set II

aflR omtA omtB ver-1

Four bands 5 +a + + +

Three bands 51 –b

+ + +

a +: Amplification in PCR.

b –: No amplification in PCR.

Table 3. Aflatoxin production by 56 Aspergillus oryzae/flavusgroup isolats from the meju sample.

Isolates of Aspergillus oryzae/flavus group

Aflatoxin (µg/ml)

B1 B2 G1 G2

5 isolates with 4 amplicons

KUFNM018 – – – –

KUFNM027 144 21 – –

KUFNM044 55 3 – –

KUFNM084 25 6 – –

KUFNM098 – – – –

51 isolates with 3 amplicons – – – –

–: Not detected

Fig. 3. HPLC chromatograms showing aflatoxin production: (A)A. flavus KUFNM027; (B) A. flavus KUFNM044; (C) A. flavusKUFNM084; and (D) Aflatoxin standard mixture. Czapek yeast-extract (CY; NaNO3 3 g, KH2PO4 1 g, KCl 0.5 g,

MgSO4·7H2O 0.5 g, FeSO4·H2O 0.01 g, yeast extract 5 g, sucrose 30 g,

and distilled water 1 L) was used for the production of aflatoxins by each

fungal isolate. The inoculated flasks were incubated for 14 days at 25oC,

and culture filtrates were used for the aflatoxin analysis.

FUNGAL FLORA IN A MEJU CONTAMINATED WITH AFLATOXIN 1746

0.4 ng/g of AFB1 (with no detectable amount of AFB2),

respectively (data not shown). It is well known that aflatoxins

are degraded during the fermentation and ripening process

of traditional soybean paste and soy sauce production [26].

Frequency of Fugal Isolates from the Meju Sample

For reliable mycological analysis of food or environmental

samples, the choice of culture media is important. In this

study, 2 types of media (DRBC and DG18) were used to

allow more complete recovery of fungal species present in

the meju sample of which the surface was dried and the

inside was somewhat wet. A total of 16 fungal species were

found in the meju sample after molecular analysis and

aflatoxigenecity confirmation (Table 4). The use of DRBC

medium made it possible to isolate 12 species with the

three major occurrences of A. oryzae (18.3%), M. circinelloides

(10.0%), and A. candidus (7.8%), whereas DG18 medium

presented 11 species isolated with A. candidus (26.5%), A.

oryzae (3.9%), and P. polonicum (3.9%) being the 3 most

frequent species. A. acidus, P. chrysogenum, P. oxalicum,

M. racemosus, and C. inaequalis only grew on the DRBC

agar, and A. unguis, E. chevalieri, E. rubrum, and E.

cinnamopurpureum only grew on the DG18 agar. Of all

the 230 isolates, the predominant fungal species was A.

candidus (34.3%), which is known to be a xerophile,

followed by fungi that grow at a relatively high aw, such as

A. oryzae (22.2%), M. circinelloides (13.0%), P. polonicum

(10.0%), A. tubingensis (4.8%), and L. ramosa (3.5%). A.

flavus had an occurrence level of 2.2%, and this species

was isolated on both DRBC and DG18 media. The

isolates of E. chevalieri, A. unguis, P. oxalicum, and E.

cinnamopurpureum had occurrence levels of 2.2%, 1.7%,

1.3%, and 1.3%, respectively. The remaining isolated

species—A. acidus, E. rubrum, P. chrysogenum, and M.

racemosus—had occurrence levels of <1.0%. The colonies

and microscopic morphological characters of the 16

representative isolates are shown in Fig. 4.

To date, researches performed in Korea on the mycobiota

that colonize meju have shown that the predominant isolates

were of A. oryzae/flavus group, Mucor, and Penicillium

spp. [5, 21, 22]. Those data were in accordance with the

mycological results found in this study when analysis was

conducted with DRBC. However, A. candidus, the most

common species in this study, was not reported to be found

in meju samples before, because the authors of previous

studies used standard media, which allowed the fungi to

grow at a relatively high aw, such as PDA, MEA, and CY.

The water activity (aw: 0.95) of the DG18 medium enables

the isolation and enumeration of fungal flora from dried

and semidried foods [11]; thus, xerophilic fungi, such as A.

candidus and Eurotium spp., were present on this medium.

To the best of our knowledge, this is the first report on

the fungal frequency in meju, which is a starter material in

Korean fermented soybean products. The molecular assays

of fungal β-tubulin and the ITS sequence were conducted

for the identification of 230 isolates from the meju sample,

using 2 types of media. The DRBC and DG18 allowed

more complete recovery of fungal species present in the

Table 4. Number and frequency of fungal species isolated from the meju sample on DRBC and DG18 agar plates.

Fungal species Number Frequency (%)

DRBCa DG18b Total DRBC DG18 Total

Aspergillus acidus 1 0 1 0.4 0 0.4

Aspergillus candidus 18 61 79 7.8 26.5 34.3

Aspergillus flavus 3 2 5 1.3 0.9 2.2

Aspergillus oryzae 42 9 51 18.3 3.9 22.2

Aspergillus tubingensis 3 8 11 1.3 3.5 4.8

Aspergillus unguis 0 4 4 0 1.7 1.7

Eurotium chevalieri 0 5 5 0 2.2 2.2

Eurotium rubrum 0 2 2 0 0.9 0.9

Penicillium chrysogenum 1 0 1 0.4 0 0.4

Penicillium oxalicum 3 0 3 1.3 0 1.3

Penicillium polonicum 14 9 23 6.1 3.9 10.0

Eupenicillium cinnamopurpureum 0 3 3 0 1.3 1.3

Mucor circinelloides 23 7 30 10.0 3.0 13.0

Mucor racemosus 2 0 2 0.9 0 0.9

Lichtheimia ramosa 4 4 8 1.7 1.7 3.5

Curvularia inaequalis 2 0 2 0.9 0 0.9

116 114 230 50.4 49.6 100

aDichloran glycerol 18% agar.

bDichloran rose bengal and chloramphenicol agar.

1747 Jung et al.

meju sample, which consisted of 16 species, including

xerophilic fungi. Three strains of aflatoxigenic A. flavus

and 2 non-aflatoxigenic A. flavus were successfully

differentiated from 56 isolates of the A. oryzae/flavus

group, using cultural experiments after multiplex PCR

assay of aflatoxigenic genes (aflR, omtA, omtB, and ver-1).

Cultural and quantitative HPLC analyses ascertained the

potential production of aflatoxins by the 3 strains of A.

flavus isolated from the sample.

Acknowledgments

This research was supported by a grant from Korea

University and a grant (11162KFDA995) from the Korea

Food and Drug Administration.

REFERENCES

1. Barbesgaard, P., H. P. Heldt-Hansen, and B. Diderichsen. 1992.

On the safety of Aspergillus oryzae: A review. Appl. Microbiol.

Biotechnol. 36: 569-572.

2. Chang, P. K. and K. C. Ehrlich. 2010. What does genetic

diversity of Aspergillus flavus tell us about Aspergillus oryzae?

Int. J. Food Microbiol. 15: 189-199.

3. Chang, P. K., B. W. Horn, and J. W. Dorner. 2005. Sequence

break points in the aflatoxin biosynthesis gene cluster and

flanking regions in nonaflatoxigenic Aspergillus flavus isolates.

Fungal Genet. Biol. 42: 914-923.

4. Chelkowski, J. 1991. Mycological quality of mixed feeds and

ingredients, pp. 217-227. In J. Chelkowski (ed.). Cereal Grain,

Mycotoxins, Fungi and Quality in Drying and Storage. Elsevier,

Amsterdam, London, New York.

5. Cho, D. H. and W. J. Lee. 1970. Microbiological studies of

Korean native soy-sauce fermentation; A study on the microflora

of fermented Korean maeju loaves. J. Kor. Agric. Chem.

Soc. 13.

6. Criseo, G., C. Racco, and O. Romeo. 2008. High genetic

variability in non-aflatoxigenic A. flavus strains by using

quadruplex PCR-based assay. Int. J. Food Microbiol. 125:

341-343.

7. Degola, F., E. Berni, C. Dall’Asta, E. Spotti, R. Marchelli,

I. Ferrero, and F. M. Restivo. 2006. A multiplex RT-PCR

approach to detect aflatoxigenic strains of Aspergillus flavus. J.

Appl. Microbiol. 103: 409-417.

8. Fraga, M. E., F. Curvello, M. J. Gatti, L. R. Cavaglieri, A. M.

Dalcero, and C. A. da Rocha Rosa. 2007. Potential aflatoxin

and ochratoxin A production by Aspergillus species in poultry

feed processing. Vet. Res. Commun. 31: 343-353.

9. He, C. H., Y. H. Fan, G. F. Liu, and H. B. Zhang. 2008.

Isolation and identification of a strain of Aspergillus tubingensis

with deoxynivalenol biotransformation capability. Int. J. Mol.

Sci. 9: 2366-2375.

10. Hillis, D. M. and M. T. Dixon. 1991. Ribosomal DNA:

Molecular evolution and phylogenetic inference. Q. Rev. Biol.

66: 411-453.

11. Hocking, A. D. and J. I. Pitt. 1980. Dichloran-glycerol medium

for enumeration of xerophilic fungi from low moisture foods.

Appl. Environ. Microbiol. 39: 488-492.

Fig. 4. Colony surface and conidial head of fungal speciesisolated from the meju sample on potato-dextrose agar. A, a: Aspergillus acidus; B, b: A. candidus; C, c: A. oryzae; D, d: A. flavus;

E, e: A. tubingensis; F, f: A. unguis; G, g: Eurotium rubrum; H, h: E.

chevalieri; I, i: Penicillium chrysogenum; J, j: P. oxalicum; K, k: P.

polonicum; L, l: Eupenicillium cinnamopurpureum; M, m: Mucor circinelloides;

N, n: M. racemosus; O, o: Lichtheimia ramosa; P, p: Curvularia inaequalis.

FUNGAL FLORA IN A MEJU CONTAMINATED WITH AFLATOXIN 1748

12. Horton, T. R. and T. D. Bruns. 2001. The molecular revolution

in ectomycorrhizal ecology: Peeking into the black box. Mol.

Ecol. 10: 1855-1871.

13. Hussein, S. H. and J. M. Brasel. 2001. Toxicity, metabolism,

and impact of mycotoxins on humans and animals. Toxicology

167: 101-134.

14. Jumpponen, A. 2009. Analysis of rhizosphere fungal communities

using rRNA and rDNA, pp. 29-40. In A. Varma and A. C.

Kharkwal (eds.). Symbiotic Fungi, Soil Biology, 18th. Ed.

Springer-Verlag, Berlin.

15. Kim, D. H., S. H. Kim, Y. K. Kim, S. O. Kim, S. J. Kim, and

S. B. Hong. 2009. Reidentification of Aspergillus spp. isolated

from clinical specimens of patients suspected as pulmonary

aspergillosis in Korea. Kor. J. Med. Mycol. 14: 133-144.

16. Kim, D. M., S. H. Chung, and H. S. Chun. 2011. Multiplex

PCR assay for the detection of aflatoxigenic and non-

aflatoxigenic fungi in meju, a Korean fermented soybean food

starter. Food Microbiol. 28: 1402-1408.

17. Kim, J. Y., S. H. Yeo, S. Y. Baek, and H. S. Choi. 2011

Molecular and morphological identification of fungal species

isolated from bealmijang meju. J. Microbiol. Biotechnol. 21:

1270-1279.

18. Kiyota, T., R. Hamada, K. Sakamoto, K. Iwashita, O. Yamada,

and S. Mikami. 2011. Aflatoxin non-productivity of Aspergillus

oryzae caused by loss of function in the aflJ gene product. J.

Biosci. Bioeng. 111: 512-517.

19. Kwon, D. J. 2002. Comparison of characteristics of koji

manufactured with Bacillus subtilis B-4 and Aspergillus oryzae

F-5. Kor. J. Food Sci. Technol. 34: 873-878.

20. Lee, C. H. and S. S. Lee. 2002. Cereal fermentation by fungi.

Appl. Mycol. Biotechnol. 2: 151-170.

21. Lee, S. S., K. H. Park, K. J. Choi, and S. A. Won. 1993.

Identification and isolation of Zygomycetous fungi found on

maeju, a raw material of Korean traditional soysauces. Kor. J.

Mycol. 21: 172-187.

22. Lee, S. S., K. H. Park, K. J. Choi, and S. A. Won. 1993. A

study on Hyphomycetous fungi found on maejus, a raw

material of Korean traditional soysauce. Kor. J. Mycol. 21:

242-272.

23. Lee, S. W., S. K. Park, and H. C. Kim. 2001. Characteristics of

red mold isolated from traditional meju. Kor. J. Post-harvest

Sci. Technol. 8: 199-205.

24. Niessen, L. 2008. PCR-based diagnosis and quantification of

mycotoxin-producing fungi. Adv. Food Nutr. Res. 54: 81-138.

25. Park, J. H., S. J. Kang, S. S. Oh, and D. H. Chung. 2001. The

screening of aflatoxin producing fungi from commercial meju

and soybean paste in western Gyeongnam by immunoassay. J.

Food Hyg. Safety 16: 274-279.

26. Park, K.-Y., K.-B. Lee, and L. B. Bullerman. 1988. Aflatoxin

production by Aspergillus parasiticus and its stability during the

manufacture of Korean soy paste (doenjang) and soy sauce

(kanjang) by traditional method. J. Food Prot. 51: 938-945.

27. Peterson, S. W. 2008. Phylogenetic analysis of Aspergillus

species using DNA sequences from four loci. Mycologia 100:

205-226.

28. Pitt, J. I. and A. D. Hocking. 2009. Fungi and Food Spoilage,

3rd Ed. Springer, New York.

29. Rank, C., M. L. Klejnstrup, L. M. Petersen, S. Kildgaard, J. C.

Frisvad, C. H. Gotfredsen, and T. O. Larsen. 2012. Comparative

chemistry of Aspergillus oryzae (RIB40) and A. flavus (NRRL

3357). Metabolites 2: 39-56.

30. Reddy, B. N. and C. R. Raghavender. 2007. Outbreaks of

aflatoxicoses in India. Afr. J. Food Agric. Nutr. Devel. 7: 1-15.

31. Samson, R. A., E. S. Hoekstra, and J. C. Frisvad. 2004.

Introduction to Food and Airborne Fungi, 7th Ed. Centraalbureau

voor Schimmelcultures, Utrecht, The Netherlands.

32. Samson, R. A., K. A. Seifert, A. F. A. Kuijpers, J. A. M. P.

Houbraken, and J. C. Frisvad. 2004. Phylogenetic analysis of

Penicillium subgenus Penicillium using partial β-tubulin

sequences. Stud. Mycol. 49: 175-200.

33. Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007. MEGA4:

Molecular evolutionary genetics analysis (MEGA) software

version 4.0. Mol. Biol. Evol. 24: 1596-1599.

34. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and

D. G. Higgins. 1997. The CLUSTAL X Windows interface:

Flexible strategies for multiple sequence alignment aided by

quality analysis tools. Nucleic Acids Res. 25: 4876-4882.

35. Tominaga, M., Y. H. Lee, R. Hayashi, O. Suzuki, K. Tamada,

K. Skamoto, K. Gotoh, and O. Akita. 2006. Molecular analysis

of an inactive aflatoxin biosynthesis gene cluster in Aspergillus

oryzae RIB strains. Appl. Environ. Microbiol. 72: 484-490.

36. Tsai, G. J. and S. C. Yu. 1997. An enzyme-linked immunosorbent

assay for the detection of Aspergillus parasiticus and Aspergillus

flavus. J. Food Prot. 60: 978-984.

37. Wei, D. L. and S. C. Jong. 1986. Production of aflatoxins by

strains of the Aspergillus flavus group maintained in ATCC.

Mycopathologia 93: 19-24.

38. White, T. J., T. Bruns, S. Lee, and J. W. Taylor. 1990.

Amplification and direct sequencing of fungal ribosomal RNA

genes for phylogenetics, pp. 315-322. In M. A. Innis, D. H.

Gelfand, J. J. Sninsky, and T. J. White (eds.). PCR Protocols: A

Guide to Methods and Applications. Academic Press Inc., NY.

39. Yoo, J. Y. and H. G. Kim. 1998. Characteristics of traditional

mejus of nation-wide collection. J. Kor. Soc. Food Sci. Nutr. 27:

259-267.