Conjugation of the Mycotoxins Alternariol and Alternariol Monomethyl Ether in Tobacco Suspension Cells

Article abstract of DOI:10.1021/acs.jafc.5b00806

The mycotoxins alternariol (AOH) and alternariol-9-O-methyl ether (AME) carry three and two phenolic hydroxyl groups, respectively, which makes them candidates for the formation of conjugated metabolites in plants. Such conjugates may escape routine methods of analysis and have therefore been termed masked or, more recently, modified mycotoxins. We report now that AOH and AME are extensively conjugated in suspension cultures of tobacco BY-2 cells. Five conjugates of AOH were identified by MS and NMR spectroscopy as β-D-glucopyranosides (attached in AOH 3- or 9-position) as well as their 6′-malonyl derivatives, and as a gentiobiose conjugate. For AME, conjugation resulted in the D-glucopyranoside (mostly attached in the AME 3-position) and its 6′- and 4′-malonyl derivatives. Pronounced differences were noted for the quantitative pattern of AOH and AME conjugates as well as for their phytotoxicity. Our in vitro study demonstrates for the first time that masked mycotoxins of AOH and AME can be formed in plant cells.

Full text of DOI:10.1021/acs.jafc.5b00806

Article
pubs.acs.org/JAFC
Conjugation of the Mycotoxins Alternariol and Alternariol
Monomethyl Ether in Tobacco Suspension Cells
Andreas A. Hildebrand, Beate N. Kohn, Erika Pfeiffer, Daniel Wefers, Manfred Metzler,
and Mirko Bunzel*
Institute of Applied Biosciences, Department of Food Chemistry and Phytochemistry, Karlsruhe Institute of Technology (KIT),
Adenauerring 20a, 76131 Karlsruhe, Germany
*
S Supporting Information
ABSTRACT: The mycotoxins alternariol (AOH) and alternariol-9-O-methyl ether (AME) carry three and two phenolic
hydroxyl groups, respectively, which makes them candidates for the formation of conjugated metabolites in plants. Such
conjugates may escape routine methods of analysis and have therefore been termed masked or, more recently, modified
mycotoxins. We report now that AOH and AME are extensively conjugated in suspension cultures of tobacco BY-2 cells. Five
conjugates of AOH were identified by MS and NMR spectroscopy as β-D-glucopyranosides (attached in AOH 3- or 9-position)
as well as their 6′-malonyl derivatives, and as a gentiobiose conjugate. For AME, conjugation resulted in the D-glucopyranoside
(
mostly attached in the AME 3-position) and its 6′- and 4′-malonyl derivatives. Pronounced differences were noted for the
quantitative pattern of AOH and AME conjugates as well as for their phytotoxicity. Our in vitro study demonstrates for the first
time that masked mycotoxins of AOH and AME can be formed in plant cells.
KEYWORDS: masked mycotoxins, modified mycotoxins, Alternaria toxins, malonyl glucosides, plant cell culture
INTRODUCTION
and AME were determined; nine were found to be active for
AOH and eight for AME.
■
The cosmopolitan genus Alternaria contains many species that
are important plant pathogens, infesting all the aerial parts of
plants and causing, e.g., early blight diseases of vegetables,
Over the past years, researchers and regulators have become
aware that conjugation of mycotoxins (and other compounds
such as pesticides) may also occur in plants. Such conjugation
reactions are believed to be part of the detoxification system of
1
brown spot of tangerines, or postharvest black rot of fruit. In
addition to economic losses due to preharvest and postharvest
decay, Alternaria species produce several groups of toxins,
which contaminate food and feed and may pose a health
10
the plant. Xenobiotic and endogenous compounds carrying
hydroxyl groups are frequently converted to glucosides and
further processed by addition of a malonyl, hexose, or pentose
2
problem. For example, Alternaria toxins have been associated
10
moiety to facilitate compartmentation and storage. Con-
jugated compounds escape routine methods of analysis if the
method aims exclusively at the detection of the parental
compounds. However, upon ingestion of the conjugate by
humans or animals, the parental form may be released in the
digestive tract and absorbed, thereby increasing the total
exposure to the compound. For the first time, this has been
unambiguously demonstrated for a glucoside of the mycotoxin
zearalenone, and the term “masked mycotoxin” has been
with an increased incidence of esophageal cancer in certain
areas of China. In 2011, the European Food Safety Authority
3
(
EFSA) published a Scientific Opinion on the risks for animal
and public health related to the presence of Alternaria toxins in
4
feed and food. The analysis of 11 730 occurrence data showed
that the most frequently detected Alternaria toxins were the
dibenzo-α-pyrones alternariol (AOH) and alternariol-9-O-
monomethyl ether (AME) (Figure 1A,B), which were common
contaminants of cereals, sunflower seeds, oilseed rape, olives,
11
4
coined. Today, several masked forms of zearalenone and
other mycotoxins, in particular of trichothecenes like
various fruits, and other food items. AOH and AME exhibit
genotoxicity in vitro by inducing gene mutations, DNA strand
1
2,13
5
−8
deoxynivalenol and nivalenol are known.
More recently,
breaks, and inhibition of topoisomerase I and IIα.
the term “modified mycotoxins” has been proposed for a
comprehensive definition including, e.g., derivatives formed by
thermal reactions or by plant metabolism (so-called “masked”
Both AOH and AME have free hydroxyl groups available for
metabolic conjugation (Figure 1A,B). In the mammalian
organism, formation of glucuronides is a major pathway of
detoxification and excretion. In vitro studies have shown that
AOH and AME are readily converted to glucuronides upon
incubation with hepatic and intestinal microsomes from
14
mycotoxins).
Because it is yet unknown whether AOH and AME are
converted to conjugated metabolites in plants, the aim of this
9
humans, rats, and pigs in the presence of UDPGA. Whereas
AOH gave rise to comparable amounts of the 3-O-glucuronide
and 9-O-glucuronide, AME was predominantly converted to
the 3-O-glucuronide. In the same study, the activities of 10
recombinant human UDP-glucuronosyltransferases for AOH
Received: February 11, 2015
Revised: April 23, 2015
Accepted: April 25, 2015
Published: April 25, 2015
©
2015 American Chemical Society
4728
DOI: 10.1021/acs.jafc.5b00806
J. Agric. Food Chem. 2015, 63, 4728−4736

Journal of Agricultural and Food Chemistry
Article
Figure 1. Chemical structures of alternariol (AOH, A), alternariol-9-O-monomethyl ether (AME, B), and AOH and AME metabolites formed in a
tobacco suspension cell culture system. 9-O-β-D-glucopyranosyl-AOH (AOH-2, C); 3-O-β-D-glucopyranosyl-AOH (AOH-3, D); 9-O-{β-D-
glucopyranosyl(1 → 6)-β-D-glucopyranosyl}AOH (AOH-1, E); 9-O-(6-O-malonyl-β-D-glucopyranosyl)AOH (AOH-4, F); 3-O-(6-O-malonyl-β-D-
glucopyranosyl)AOH (AOH-5, G); 7-O-β-D-glucopyranosyl-AME (AME-2, H); 3-O-β-D-glucopyranosyl-AME (AME-5, I); 3-O-(6-O-malonyl-β-D-
glucopyranosyl)AME (AME-6, J); 3-O-(4-O-malonyl-β-D-glucopyranosyl)AME (AME-7, K).
study was to analyze AOH and AME conjugation in a tobacco
BY-2 suspension cell culture as a model system.
acetonitrile and methanol were purchased from VWR International
(
(
Bruchsal, Germany). NMR solvents were from Deutero GmbH
Kastellaun, Germany) and VWR International.
Tobacco Suspension Cell Culture. Tobacco BY-2 cells
MATERIALS AND METHODS
■
(Nicotiana tabacum L. cv. Bright Yellow 2) were kindly provided by
Dr. Jan Maisch (Institute of Botany, KIT). The cells were cultured in
the dark at 25 °C on an orbital shaker at 150 rpm in 100 mL
Erlenmeyer flasks containing 30 mL of Murashige and Skoog medium
Chemicals and Reagents. AOH and AME were isolated from a
culture of Alternaria alternata strain TA7 grown on rice flour
containing media for 24 days at 25 °C as described earlier for the
15
isolation of other Alternaria toxins. The crude extract was analyzed
by LC-MS, and the fractions containing AOH and AME were isolated
by preparative HPLC. Identity was confirmed by NMR (see below),
and the purity was >98% according to HPLC analysis with UV
detection at 254 nm.
supplemented with 30 g/L sucrose, 200 mg/L KH PO , 100 mg/L
2 4
inositol, 1 mg/L thiamine, and 0.2 mg/L 2,4-dichlorophenoxyacetic
acid. The medium was adjusted to pH 5.8. The cells were subcultured
weekly, transferring 1.5 mL (about 0.5 g cell wet weight) of the
suspension into 30 mL of fresh medium. The exponential growth
phase of the cells started after 3 days and was finished on day seven
(about 10 g of cell wet weight or 0.4−0.5 g of cell dry weight in the
stationary phase).
Chemicals for the tobacco cell culture and other chemicals and
reagents were obtained from Sigma-Aldrich/Fluka (Taufenkirchen,
Germany) and Duchefa (Haarlem, The Netherlands) and were of the
highest quality available (96−100% depending on the chemical, with
the exception of Evan’s Blue (>75% purity)). HPLC-MS grade
Incubation of Tobacco Cells with AOH and AME. Stock
solutions (10 mM) of AOH or AME in DMSO were stable at ambient
4
729
DOI: 10.1021/acs.jafc.5b00806
J. Agric. Food Chem. 2015, 63, 4728−4736

Journal of Agricultural and Food Chemistry
Article
Figure 2. HPLC profiles of metabolites of alternariol (AOH) (A) and alternariol-9-O-monomethyl ether (AME) (B) in extracts of cultured tobacco
BY-2 cells after 48 h incubation (UV detection at 254 nm). The abbreviations used for the metabolites (AOH-X, AME-X) indicate which metabolite
is represented by which peak in the chromatogram (with X describing the peak number).
temperature in the dark for up to two months. The concentration of
the mycotoxin in the incubation medium was 50 or 100 μM. The
DMSO concentration did not exceed 1%. Control incubations were
carried out with DMSO but without mycotoxin. After 24, 48, or 96 h
of incubation, cells and medium were separated by filtration and
freeze-dried. The dried medium was dissolved in MeOH for
chromatographic analysis. The dried cells (10−12 g wet cells yield
about 0.6 g dry cells) were ground using mortar and pestle. To extract
AOH, AME, and their metabolites, an optimized extraction procedure
was used: 0.1 g of the ground material was extracted twice for 2 h at 25
rate was 8 mL/min. A linear gradient was started with 20% B held for
5 min, then ramped to 65% B in 20 min, ramped to 90% B in 1 min,
and held for 4 min. Fractions of the separated metabolites were
collected manually according to the chromatograms. After vacuum-
concentration to about half of their volumes, the fractions were freeze-
dried for HPLC-PDA-MS and NMR analysis. HPLC-PDA-MS was
conducted as described above and showed that the purity of the AOH
and AME metabolites used for NMR analysis was >97% (based on UV
chromatograms at 254 nm).
Carbohydrate Analysis. Carbohydrate analysis was performed by
high performance anion exchange chromatography (HPAEC) with
pulsed amperometric detection (PAD) after acidic hydrolysis with 2 M
trifluoroacetic acid for 30 min at 121 °C. After evaporation of the acid,
the sample was redissolved in water and analyzed on an ICS-5000
HPAEC system (Thermo Scientific Dionex, Sunnyvale, CA, USA)
using a CarboPac PA-20 column. The exact chromatographic
°
C with horizontal shaking. For the first extraction, 4.5 mL of MeOH/
CH Cl (2:1, v/v) was used, whereas 4.5 mL of MeOH/H O/acetic
2
2
2
acid (79/20/1, v/v/v) was used for the second extraction. The
combined extracts were evaporated to dryness using a vacuum
evaporator, and the residues were dissolved in 0.4 mL of MeOH and
analyzed by using HPLC-photo diode array (PDA)-MS.
1
7
Assay To Determine Cell Death. Cells that accumulated Evans
conditions are described in Wefers et al. (2014). To determine D-/
L-configuration of the glucose unit in 9-O-β-glucopyranosyl-AOH, the
glycosidic linkage was cleaved as described above. The residue after
evaporation was derivatized with 150 μL of (R)-2-octanol and 5 μL of
TFA at 130 °C overnight. After removal of the derivatization reagents,
the sample was silylated by using 80 μL of N,O-bis(trimethylsilyl)
trifluoroacetamide and 20 μL of pyridine. Sample analysis was
performed by GC-MS (GC-2010 Plus and GCMS-QP2010 Ultra,
Shimadzu, Kyoto, Japan) equipped with an Rxi-5Sil MS column (30 m
× 0.25 mm i.d., 0.25 μm film thickness, Restek, Bad Homburg,
Germany). Detailed chromatographic conditions are given in Wefers
16
blue were considered as dead cells. An aliquot of the cell suspension
200 μL) was incubated with 1.5 mL of 0.05% (w/v) Evans blue at
(
room temperature for 15 min, subsequently washed twice with 1 mL
of PBS, and resuspended in 500 μL of PBS. An aliquot of the
suspension (20 μL) was transferred on microscope slides, and a total
of 500 cells was scored under the light microscope, discriminating
between living (uncolored) and dead (blue stained) cells.
n
HPLC-PDA-MS Analysis. An LXQ Linear Ion Trap MS system
Thermo Fisher Scientific, Waltham, MA, USA) equipped with a
(
Finnigan Surveyor HPLC-PDA system was used. Separation was
carried out on a reversed phase column (Phenomenex Luna C8(2),
1
7
et al. (2014).
2
50 × 4.6 mm i.d.). Solvent A was H O with 0.1% formic acid, solvent
NMR Spectroscopy. NMR spectra were recorded on a Bruker
AscendM 500 spectrometer (Rheinstetten, Germany) equipped with a
Prodigy cryoprobe. Freeze-dried samples were dissolved in 500 μL of
2
B was acetonitrile with 0.1% formic acid, and the flow rate was 0.5
mL/min. The linear gradient started with 20% B, held for 5 min,
ramped to 50% B in 10 min, ramped to 70% B in 9 min, ramped to
DMSO-d . The structures of the isolated oligosaccharides were
6
1
1
00% B in 5 min, held for 2 min, decreased to 20% B in 1 min, and
identified by using H, H,H-correlated spectroscopy (COSY),
heteronuclear single quantum coherence (HSQC), and heteronuclear
multiple bond correlation (HMBC) experiments. Standard Bruker
pulse sequences were used; spectra were acquired at 298 K. Chemical
shifts (δ) were referenced to the central solvent signals (δH 2.50 ppm
equilibrated for 2 min before the next injection. The metabolites were
detected at 254 nm and characterized with MS operating in ESI
positive mode. Full scan mass spectra were recorded from m/z 100−
2
+
1000, MS of the [M + H] ions was conducted at CID 35 (35% of 5
1
8
V). Nitrogen was used as the sheath gas, auxiliary gas, and sweep gas
with flow rates of 30.0, 15.0, and 0.02 L/min, respectively. Spray
voltage was 4.47 kV, spray current 3.15 mA, capillary voltage 45.10 V,
capillary temperature 350 °C, and the tube lens voltage was 125.58 V.
Semiquantitative analyses were carried out by determining peak areas
at 254 nm, using AOH or AME as external standard compounds.
Isolation of the Metabolites of AOH and AME. About 12 g of
dry cells were collected from several incubations of each mycotoxin
and extracted by scaling up the procedure described above, yielding
about 40 mL of crude extracts dissolved in MeOH. These extracts
were fractionated in portions of 1.5 mL using a preparative HPLC-
UVD system (LC-8A pumps, SPD-20A UV detector, Shimadzu,
Kyoto, Japan) and a reversed phase column (Phenomenex, Luna
C18(2), 5 μm, 250 × 25.0 mm i.d.). Solvent A was water with 0.1%
formic acid, solvent B was MeOH with 0.1% formic acid, and the flow
and δC 39.5 ppm).
RESULTS AND DISCUSSION
■
Metabolite Patterns of AOH and AME Obtained from
Tobacco BY-2 Cells. AOH and AME were incubated in
tobacco BY-2 cell suspensions for 2 days during the exponential
growth phase. The incubation medium after cell separation did
not contain detectable amounts of free AOH and AME,
suggesting complete mycotoxin uptake into the cells. Analysis
of the cell extracts after disruption of the intact cells revealed
the presence of small amounts of parental AOH and AME. The
major part of the mycotoxins was, however, metabolized as
demonstrated by five and seven peaks with spectroscopic
4
730
DOI: 10.1021/acs.jafc.5b00806
J. Agric. Food Chem. 2015, 63, 4728−4736

Journal of Agricultural and Food Chemistry
Article
4
731
DOI: 10.1021/acs.jafc.5b00806
J. Agric. Food Chem. 2015, 63, 4728−4736

Journal of Agricultural and Food Chemistry
Article
characteristics suggesting AOH and AME conjugates, respec-
tively (Figure 2). None of these peaks was present in control
incubations without mycotoxins. Variation of the mycotoxin
concentration, incubation time, and cell mass did not change
the number of potential conjugates, although slight changes in
the size of the peaks were observed. All potential conjugates
eluted earlier than AOH or AME from the reversed-phase
column and had characteristic UV spectra similar to that of the
parental mycotoxins, i.e., with absorption bands at about 254,
2
88, 299, and 340 nm. The m/z values of the quasi-molecular
ions of the potential conjugates were higher than those of the
2
parental compounds, but their MS spectra showed the m/z
values of the corresponding mycotoxins as major fragments.
To obtain sufficient quantities of the metabolites for
structure elucidation, the combined cell extracts of several
incubations were fractionated using preparative HPLC. All five
metabolites of AOH and four dominant AME metabolites
represented by peaks 2, 5, 6, and 7 (Figure 2) were isolated in
quantities and purities that allowed for unambiguous structural
characterization by NMR spectroscopy. The minor AME
metabolites 1, 3, and 4 were isolated in small amounts only.
Tentative structures could, however, be proposed based on
their UV and mass spectrometric data as discussed later.
Structure Elucidation of the Major AOH and AME
Conjugates. AOH Metabolites. The molecular mass of the
AOH metabolite representing the dominant peak 2 (Figure 2A)
was determined as 420 (quasi-molecular ion with m/z 421 [M
+
+
H] ) consistent with a metabolite built of AOH and a hexose
sugar. The mass difference between the quasi-molecular ion and
+
the fragment m/z 259 [AOH + H] also suggests AOH being
attached to a hexose. HPAEC-PAD analysis after acidic
hydrolysis revealed the hexose to be glucose. The aromatic
1
region of the H NMR spectrum of AOH metabolite 2 showed
four AOH doublets with 2 Hz meta coupling constants. The
Figure 3. (A) HMBC spectrum (aglycone and anomeric region) of 9-
O-β-D-glucopyranosyl-alternariol. (B) Overlay of the carbohydrate
regions of the HSQC spectra of 9-O-β-D-glucopyranosyl-alternariol
carbons shifts of AOH were assigned by using the HSQC and
1
13
HMBC spectra. Comparison of the H and C chemical shifts
Table 1) with those of parental AOH showed larger
(blue correlation peaks) and 9-O-(6-O-malonyl-β-D-glucopyranosyl)-
(
alternariol (red correlation peaks).
differences for C9, H8, and H10 suggesting a modification at
the phenolic hydroxyl group attached to C9. Because DMSO
was used as a solvent, intermolecular proton exchanges were
slowed down resulting in several discrete OH signals in most
samples. By using the HMBC experiment the phenolic OH
protons can be assigned to specific carbons. Here, HMBC
correlation peaks showed coupling of the phenolic proton in
position 7, demonstrating that this phenolic group was not
conjugated. The anomeric proton signal of the glucose moiety
showed a coupling constant of 7.3 Hz implicating the β-anomer
of the glucose unit. The appearance of one anomeric proton
only and comparison of the H and C NMR data with those
of glucose suggested the glucose to be bound via its lactol
group to AOH. An HMBC correlation peak (Figure 3A)
between the anomeric proton and C9 of AOH confirmed the
glycosidic linkage in the suggested position and proved this
compound to be 9-O-β-D-glucopyranosyl-AOH (Figure 1C).
Structure elucidation of the compound representing peak 3
spectrum of peak 1 showed quasi-molecular ions with m/z 583
+
+
([M + H] ) and m/z 605 ([M + Na] ), indicating a molecular
mass of 582. Two mass losses of 162 Da resulting in m/z 421
and m/z 259 suggested that AOH is attached to a hexose
disaccharide. NMR data confirmed this assumption and
demonstrated an (1 → 6)-linkage between the two
glucopyranose units (diagnostic HMBC signals, downfield
shift of about 8 ppm for C6 of one glucose unit). Again, the
coupling constants of the anomeric protons (7.2 Hz, 7.8 Hz)
demonstrated both glucose anomers to be in their β-position.
NMR data of the aglycone suggested its conjugation via the
phenolic OH group at C9, which was unambiguously verified
by interpretation of the HMBC spectrum. Thus, the compound
representing peak 1 was identified as 9-O-{β-D-glucopyranosyl-
(1 → 6)-β-D-glucopyranosyl}AOH (Figure 1E).
1
13
The quasi-molecular ions of the compounds representing
+
+
(Figure 2A) showed many similarities to 9-O-β-D-glucopyr-
peaks 4 and 5 (m/z 507, [M + H] ; m/z 529, [M + Na] )
anosyl-AOH. However, NMR data clearly demonstrate the
linkage between the β-anomer of the glucose unit and the
phenolic hydroxyl group in position C3 of AOH via a glycosidic
linkage, identifying this compound as 3-O-β-D-glucopyranosyl-
AOH (Figure 1D).
suggest a molecular mass of 506 that cannot be explained by
the formation of simple AOH glycosides. The MS spectra
2
showed a 248 Da mass loss resulting in a fragment with m/z
+
259 [AOH + H] . A mass loss of 248 Da indicates the
elimination of an anhydro malonylglucoside as was also
19
The remaining three potential AOH conjugates had
molecular masses different from AOH glucosides. The mass
observed for isoflavone malonylglucosides. Interpretation of
the NMR spectra confirmed the substitution of the glucose unit
4
732
DOI: 10.1021/acs.jafc.5b00806
J. Agric. Food Chem. 2015, 63, 4728−4736

Journal of Agricultural and Food Chemistry
Article
Table 2. H and 13C NMR Data of Alternariol-9-O-monomethyl Ether (AME) and Its Metabolites
1
a
AME (Figure 1B)
AME-2 (Figure 1H)
AME-5 (Figure 1I)
AME-6 (Figure 1J)
AME-7 (Figure 1K)
δH
J
δC
δH
J
δC
δH
J
δC
δH
J
δC
δH
J
δC
1
2
3
3
4
4
6
6
7
7
8
9
1
1
1
1
1
1
2
3
4
5
6
6
138.30
117.42
158.33
137.75
116.70
158.30
138.22
117.94
157.60
138.08
117.71
157.18
138.48
117.95
157.65
6.73
d (2.3)
6.68
d (2.4)
6.98
d (2.4)
6.96
d (1.8)
7.01
d (2.5)
-OH
a
10.39
6.65
s
10.33
6.56
s
d (2.3)
101.29
152.52
nd
d (2.4)
100.72
152.67
nd
6.99
d (2.4)
102.21
152.22
nd
6.98
d (1.8)
102.28
152.11
nd
7.03
d (2.5)
102.21
152.19
nd
a
98.47
164.00
nd
98.62
164.04
98.54
163.94
99.02
164.16
161.30
-OH
11.83
6.63
s
11.78
6.69
11.80
6.68
11.81
6.71
d (1.9)
98.94
6.97
7.37
d (2.1)
d (2.1)
101.33
164.17
104.21
137.75
108.70
24.70
55.44
101.68
73.14
76.04
69.64
77.33
60.60
60.60
d (1.9)
d (1.9)
99.56
165.92
103.88
138.22
110.79
24.77
55.69
99.74
72.91
76.30
69.39
76.92
60.39
60.38
d (1.1)
d (1.1)
99.54
165.71
103.89
138.08
110.79
24.78
41.23
99.29
72.78
76.03
69.37
73.46
63.87
63.87
41.23
166.68
167.30
d (2.1)
d (2.1)
99.60
166.07
103.94
138.48
111.33
24.80
55.72
99.31
72.84
74.04
71.43
73.46
59.79
59.74
41.47
166.34
168.22
165.89
103.23
138.30
108.60
24.81
0
0a
0b
1
2
′
′
′
′
′
7.23
d (1.9)
7.30
7.3
7.32
2.74
3.91
s
s
2.75
3.93
5.01
3.39
3.31
3.16
3.43
3.73
3.45
2.80
3.93
5.03
3.26
3.28
3.16
3.44
3.71
3.45
2.80
3.93
5.1
2.81
3.93
5.17
3.37
3.76
4.68
3.55
3.57
3.35
3.42
55.60
d (7.7)
d (7.4)
d (7.3)
d (7.8)
3.29
3.32
3.20
3.75
4.37
4.12
3.39
′
′
Mal-CH2
Mal-COOR
Mal-COOH
nd
a
AME-2:7-O-β-D-glucopyranosyl-AME; AME-5:3-O-β-D-glucopyranosyl-AME; AME-6:3-O-(6-O-malonyl-β-D-glucopyranosyl)AME; AME-7:3-O-
4-O-malonyl-β-D-glucopyranosyl)AME. δ /δ are given in ppm, J in Hz. nd = not determined, s = singlet, d = doublet.
(
H
c
with a malonyl group. Linkage of the malonyl group to the
the difference between AOH and AME. Complete interpreta-
tion of the data including the HMBC correlation peak between
the anomeric glucose proton and the AME C3 identified this
compound as 3-O-β-D-glucopyranosyl-AME (Figure 1I).
The mass spectrum of the compound representing peak 6
suggested a malonyl conjugated AME glucoside (quasi-
1
13
glucose 6 position was obvious from the H and C data if
compared to data of the non-malonyl conjugated AOH
glucoside (Figure 3B, Table 1) and was confirmed by HMBC
correlation peaks showing coupling of the malonyl carbonyl
carbon and glucose 6 protons. Again, the glucose unit was
proven to be in its β-configuration. NMR also revealed that the
compounds representing peaks 4 and 5 differ in the AOH
position to which the 6-O-malonyl-β-D-glucopyranosyl unit is
attached. HMBC correlation peaks show the attachment to
AOH position 9 (compound representing peak 4) and 3
+
molecular ions of m/z 521 [M + H] and m/z 543 [M +
+
2
+
Na] , MS : m/z 273 [AME + H] ). Independent interpretation
of the NMR data but also comparison to the data of 3-O-(6-O-
malonyl-β-D-glucopyranosyl)AOH suggested this compound to
be 3-O-(6-O-malonyl-β-D-glucopyranosyl)AME. The linkage
positions between AME, glucose, and the malonyl group were
unambiguously confirmed by HMBC signals, identifying this
compound as 3-O-(6-O-malonyl-β-D-glucopyranosyl)AME
(Figure 1J).
(
compound representing peak 5) identifying these AOH
metabolites as 9-O-(6-O-malonyl-β-D-glucopyranosyl)AOH
and 3-O-(6-O-malonyl-β-D-glucopyranosyl)AOH, respectively
(
Figure 1F,G).
AME Metabolites. Different from AOH, AME carries two
phenolic hydroxyl groups only. Because of the methoxyl group
Another malonyl glucoside of AME was obtained from peak
7. Again, both the quasi-molecular of m/z 521 ([M + H] ) and
+
in position 9, glycosidation in this position as shown for AOH
is not an option. The quasi-molecular ion of peak 2 (Figure 2B)
the loss of 248 Da (representing an anhydro malonylglycoside)
were indicative for such a structure. NMR data of the AME
moiety of this compound were closely comparable to those of
3-O-(6-O-malonyl-β-D-glucopyranosyl)AME; however, large
differences were found for the NMR data of the glucose
moiety. A downfield shift of the glucose C4 signal and upfield
shifts of both the C3 and C5 signals indicate the attachment of
the malonyl group to the glucose 4 position, which was also
confirmed by interpretation of the HMBC spectrum. Using the
described information, this compound was identified as 3-O-(4-
O-malonyl-β-D-glucopyranosyl)AME (Figure 1K).
+
(
m/z 457, [M + Na] ) and the fragment m/z 273 ([AME +
+
H] , mass loss of 162 Da) suggested AME to be conjugated
with a hexose. NMR data (Table 2) indicated glucose
conjugation of AME in position 7, which was confirmed by
an HMBC correlation peak between the anomeric proton (β-
configuration) and carbon 7 of AME identifying this compound
as 7-O-β-D-glucopyranosyl-AME (Figure 1H).
The compound representing peak 5 has a molecular weight
+
of 434 (quasi-molecular ion m/z 435, [M + H] ]), and its
NMR-data are closely comparable to those of 3-O-β-D-
Thus, all metabolites of AOH and AME contain β-
glucopyranosyl-AOH with the exception of those data showing
glucopyranose linked through a glycosidic bond to various
4
733
DOI: 10.1021/acs.jafc.5b00806
J. Agric. Food Chem. 2015, 63, 4728−4736

Journal of Agricultural and Food Chemistry
Article
hydroxyl groups of the mycotoxins. To prove that glucose was
in its D-configuration, 9-O-β-D-glucopyranosyl-AOH was hydro-
lyzed under acidic conditions. The liberated glucose was
derivatized with (R)-2-octanol, silylated, and analyzed by GC-
MS confirming that glucose was present in its D-form. From
these results it was assumed that all glucose units in the
identified AOH and AME conjugates were in their D-
configuration. Whereas both available phenolic hydroxyl groups
pattern of the metabolites of AOH and AME was noted. After
48 h incubation of AME (50 μM in medium), about 75% of the
applied amount was recovered in the form of extractable
conjugates. Conjugation with glucose took place at the phenolic
hydroxyl groups located at positions C3 (about 90%) and C7
(10%) (Figure 2B). About 10% of the recovered metabolites
remained as the monoglucoside 3-O-β-D-glucopyranosyl-AME,
whereas 80% were further metabolized by malonylation,
preferentially (70%) at the glucose 6 position but also at the
4 position (10%). The only notable change in the pattern of
AME metabolites due to varying incubation times concerned
the proportion of 3-O-β-D-glucopyranosyl-AME, which de-
creased from 25% after 24 h incubation to 5% after 96 h, with a
concomitant increase of 3-O-(6-O-malonyl-β-D -
glucopyranosyl)AME (data not shown).
(in 3- and 7-position) of AME were found conjugated with
glucose, a conjugate in which the phenolic hydroxyl group in
position 7 of AOH was substituted with glucose was not found
as a major AOH conjugate. Lack of conjugation may be due to
the hydroxyl group being engaged in hydrogen bonding with
the neighboring carbonyl group at C6 (Figure 1). In contrast,
7
-O-glucosylation occurred in the conjugation of AME to a
small extent; 3-O-glucosylation was, however, predominant (see
below). A similar difference in the conjugation of AOH and
AME was observed in an in vitro study on the glucuronidation
with human hepatic microsomes and recombinant human
UDP-glucuronosyl transferases (UGTs). AOH was glucuroni-
dated at the C3 and C9, but not C7 hydroxyl groups, and the
rate and ratio of C3 and C9 conjugation were dependent on the
After 48 h incubation of AOH (50 μM in medium), the total
recovery of the applied amount as extractable conjugates was
only about 50%, and the major proportion (80%) of the
recovered material consisted of the monoglucosides 9-O-β-D-
glucopyranosyl-AOH (60%) and 3-O-β-D-glucopyranosyl-AOH
(20%). Notably, the corresponding malonylated 9-O- and 3-O-
glucosides of AOH together accounted for only 13% of the
extractable conjugates. This pattern did not differ much at
incubation times of 24 and 96 h. The major difference was an
increase of 9-O-{β-D-glucopyranosyl(1 → 6)-β-D-
glucopyranosyl}AOH from 7% at 24 h to 25% at 96 h,
accompanied by a decrease of 9-O-β-D-glucopyranosyl-AOH
(data not shown). In addition, the total recovery of extractable
AOH metabolites decreased to about 35%.
9
individual UGT isoform. For AME, glucuronidation was
clearly preferred at C3 by most UGT isoforms, but a few
isoforms, e.g., UGT1A7, catalyzed the conjugation at C7.
Similar differences in the conjugation can be assumed for plant
cells, which are known to differ in their pattern of UDP-
12
glucosyltranferases. For example, Kovalsky Paris et al. (2014)
recently identified an UDP-glucosyltransferase from barley,
which did not catalyze the common glucosylation of
zearalenone at the 14-hydroxyl group only, which is equivalent
to the 9-hydroxyl of AOH, but also at the 16-hydroxyl group,
which is neighboring a carbonyl group similar to the 7-hydroxyl
Implications. Our study demonstrated that the Alternaria
toxins AOH and AME are efficiently conjugated in cultured
tobacco BY-2 cells, a widely used in vitro model for plant cell
metabolism. The first metabolic step for both mycotoxins is
conjugation with glucose. Metabolic conjugation during phase
II metabolism is comparable in mammals and plants and
considered an important first detoxification step for endoge-
20
groups of AOH and AME.
The monoglucosides are, in part, further conjugated at the
glucopyranoside moiety with another β-D-glucopyranose
22
(
forming a gentiobiose conjugate). A gentiobiose conjugate
nous and exogenous compounds. Major differences, however,
between mammals and plants exist in the fate of such
conjugated metabolites. Whereas animals and humans are
able to rapidly excrete phase II metabolites via urine and bile,
plants are basically unable to excrete phase II metabolites. Thus,
storage of the metabolites either as soluble conjugates in
vacuoles or as insoluble conjugates bound to cell wall
components becomes essential. Both options require further
modification of the primary conjugates. Whereas malonylation
of the glucoside is considered a signal for transport into
vacuoles, further glycosylation is supposed to enable incorpo-
was unambiguously identified for AOH only. However, the MS
spectra and HPLC retention times of AME peaks 1 and 3
(
Figure 2B) also suggest the existence of an AME-7-diglucoside
and an AME-3-diglucoside, respectively. Because NMR spectra
were not obtained from the minor quantities of these
metabolites, the position of conjugation was deduced from
the observation that metabolites with identical conjugate
moiety are eluted earlier if conjugated at C7 as compared to
those conjugated at C3. In addition, two malonyl glucosides of
AOH and two malonyl glucosides of AME were unambiguously
identified. On the basis of its mass spectrum and elution
behavior a fifth malonyl glucoside was tentatively identified as
22
ration into the cell wall. Both processes are, however, not well
understood on a molecular and cellular level. From our first
semiquantitative data it appears that malonyl conjugation of the
primarily formed glucosides is the preferred pathway for AME,
whereas AOH is glycosylated, but less readily malonylated.
Thus, it can be assumed that AME detoxification through
conjugation and storage in the vacuole is a more efficient
process for AME as compared to AOH. The efficient formation
of malonyl glucosides and their potential storage in the vacuole
may also explain the comparatively high recoveries for AME in
our experiments, whereas much less AOH was recovered. Thus,
it can be assumed that AOH is preferentially stored in the cell
wall, a hypothesis that, however, needs to be investigated in
future studies.
7
-O-(6-O-malonyl-β-D-glucopyranosyl)AME. Earlier studies
describe a malonyl migration for malonyl glucosides of various
phytochemicals such as isoflavones. Acyl migration, especially 4
21
→
6 migrations, are frequently described in the literature.
Exemplarily, 4 → 6 migration was observed for malonylgenis-
1
9
tin. Therefore, solutions of the isolated AME malonyl
glucosides in MeOH and DMSO were tested for their stability.
Both compounds 3-O-(6-O-malonyl-β-D-glucopyranosyl)AME
and 3-O-(4-O-malonyl-β-D-glucopyranosyl)AME were stable
within the test period of up to 2 weeks.
Quantitative Aspects of AOH and AME Conjugation in
Tobacco BY-2 Cells. Although it was not the aim of this study
to exactly quantitate the metabolites and investigate the kinetics
of their formation, a striking difference in the quantitative
Another difference observed between AOH and AME
incubations was related to cell growth. Growth of cells
4
734
DOI: 10.1021/acs.jafc.5b00806
J. Agric. Food Chem. 2015, 63, 4728−4736

Journal of Agricultural and Food Chemistry
Article
incubated with AME was comparable to control cells without
mycotoxin; a reduction of cell mass was, however, noted if the
cells were incubated with AOH. Therefore, cell cultures were
treated during the lag phase of growth with the mycotoxins (50
μM in medium) for 24, 48, and 72 h, and the wet weight of
cells was determined. An aliquot of the cells was stained with
Evans blue to indicate dead cells and scored by light
microscopy. Whereas the cell mass of AME-treated cells
increased up to 6 fold with a slight decrease of dead cells, no
significant increase in the cell mass was observed in AOH-
treated cells, but the percentage of dead cells increased up to
REFERENCES
■
(
1) Logrieco, A.; Moretti, A.; Solfrizzo, M. Alternaria toxins and plant
diseases: An overview of origin, occurrence and risks. World Mycotoxin
J. 2009, 2, 129−140.
(2) Ostry, V. Alternaria mycotoxins: An overview of chemical
characterization, producers, toxicity, analysis and occurrence in
foodstuffs. World Mycotoxin J. 2008, 1, 175−188.
(
3) Liu, G. T.; Qian, Y. Z.; Zhang, P.; Dong, W. H.; Qi, Y. M.; Guo,
H. T. Etiological role of Alternaria alternata in human esophageal
cancer. Chin. Med. J. (Engl) 1992, 105, 394−400.
4) EFSA. Scientific opinion on the risks for animal and public health
(
related to the presence of Alternaria toxins in feed and food. EFSA J.
011, 9 (2407), 97.
5) Lehmann, L.; Wagner, J.; Metzler, M. Estrogenic and clastogenic
potential of the mycotoxin alternariol in cultured mammalian cells.
Food Chem. Toxicol. 2006, 44, 398−408.
(6) Brugger, E. M.; Wagner, J.; Schumacher, D. M.; Koch, K.;
Podlech, J.; Metzler, M.; Lehmann, L. Mutagenicity of the mycotoxin
alternariol in cultured mammalian cells. Toxicol. Lett. 2006, 164, 221−
30.
7) Pfeiffer, E.; Eschbach, S.; Metzler, M. Alternaria toxins: DNA
strand-breaking activity in mammalian cells in vitro. Mycotoxin Res.
007, 23, 152−157.
8) Fehr, M.; Pahlke, G.; Fritz, J.; Christensen, M. O.; Boege, F.;
40% over time. Thus, the impact of Alternaria toxins and their
2
(
conjugates on phytotoxicity is another factor that needs to be
investigated in future studies.
While we unambiguously proved the formation of conjugates
of Alternaria toxins in cell suspensions, the formation of these
conjugates also needs to be demonstrated in field plants as a
first step to evaluate a potential impact of AOH and AME
conjugates on food and feed safety. A very first approach to
measure conjugated Alternaria toxins was performed by
Walravens and co-workers who developed an UPLC-MS/MS
method to analyze glucosylated AOH and AME using
2
(
2
(
23
synthesized standard compounds. Syntheses of 9-O-β-D-
glucopyranosyl-AOH, 3-O-β-D-glucopyranosyl-AOH, and 3-O-
Altemoller, M.; Podlech, J.; Marko, D. Alternariol acts as a
topoisomerase poison, preferentially affecting the IIalpha isoform.
Mol. Nutr. Food Res. 2009, 53, 441−451.
24
β-D-glucopyranosyl-AME were described by Mikula et al., but
only 3-O-β-D-glucopyranosyl-AOH and 3-O-β-D-glucopyrano-
syl-AME were used for the method development mentioned
above. Analyses of 24 cereal-based food products did not detect
any 3-O-β-D-glucopyranosyl-AOH (the less abundant glucose-
conjugated AOH metabolite in our study) or 3-O-β-D-
glucopyranosyl-AME in these samples, which, however, did
not contain AME, too, and only trace amounts of AOH.
Therefore, more studies that include the full array of
metabolites identified in our study using food products from
different plant families are required to estimate the impact of
AOH and AME conjugates on food and feed safety.
̈
(9) Pfeiffer, E.; Schmit, C.; Burkhardt, B.; Altemoller, M.; Podlech, J.;
Metzler, M. Glucuronidation of the mycotoxins alternariol and
alternariol-9-methyl ether in vitro: Chemical structures of glucuronides
and activities of human UDP-glucuronosyltransferase isoforms.
Mycotoxin Res. 2009, 25, 3−10.
(
10) Sandermann, H. Plant metabolism of xenobiotics. Trends
Biochem. Sci. 1992, 17, 82−84.
11) Gareis, M.; Bauer, J.; Thiem, J.; Plank, G.; Grabley, S.; Gedek, B.
Cleavage of zearalenone-glycoside, a masked mycotoxin, during
digestion in swine. J. Vet. Med. B 1990, 37, 236−240.
(12) Berthiller, F.; Crews, C.; Dall’Asta, C.; De Saeger, S.; Haesaert,
G.; Karlovsky, P.; Oswald, I. P.; Seefelder, W.; Speijers, G.; Stroka, J.
Masked mycotoxins: A review. Mol. Nutr. Food Res. 2013, 57, 165−
(
1
86.
ASSOCIATED CONTENT
■
(13) EFSA. Scientific opinion on the risks for human and animal
health related to the presence of modified forms of certain mycotoxins
in food and feed. EFSA J. 2014, 12 (3916), 107.
14) Rychlik, M.; Humpf, H. U.; Marko, D.; Danicke, S.; Mally, A.;
Berthiller, F.; Klaffke, H.; Lorenz, N. Proposal of a comprehensive
definition of modified and other forms of mycotoxins including
masked″ mycotoxins. Mycotoxin. Res. 2014, 30, 197−205.
15) Fleck, S. C.; Burkhardt, B.; Pfeiffer, E.; Metzler, M. Alternaria
*
S
Supporting Information
HSQC and HMBC spectra of all AOH and AME metabolites
identified in this study. The Supporting Information is available
free of charge on the ACS Publications website at DOI:
(
1
0.1021/acs.jafc.5b00806.
″
(
AUTHOR INFORMATION
toxins: Altertoxin II is a much stronger mutagen and DNA strand
breaking mycotoxin than alternariol and its methyl ether in cultured
mammalian cells. Toxicol. Lett. 2012, 214, 27−32.
■
*
Corresponding Author
Tel: +4972160842936. E-mail: mirko.bunzel@kit.edu.
(16) Turner, J. G.; Novacky, A. Quantitative relation between plant
and bacterial cells involved in hypersensitive reaction. Phytopathology
Notes
1
(
974, 64, 885−890.
The authors declare no competing financial interest.
17) Wefers, D.; Tyl, C. E.; Bunzel, M. Novel arabinan and galactan
oligosaccharides from dicotyledonous plants. Front. Chem. 2014, 2,
ACKNOWLEDGMENTS
100.
(
■
18) Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. NMR chemical shifts
of common laboratory solvents as trace impurities. J. Org. Chem. 1997,
2, 7512−7515.
19) Yerramsetty, V.; Mathias, K.; Bunzel, M.; Ismail, B. Detection
The authors are grateful to Dr. Jan Maisch (Institute of Botany,
KIT) for providing BY-2 cells.
6
(
and structural characterization of thermally generated isoflavone
ABBREVIATIONS
■
malonylglucoside derivatives. J. Agric. Food. Chem. 2011, 59, 174−83.
AME, alternariol-9-O-monomethyl ether; AOH, alternariol;
HPAEC, high performance anion exchange chromatography;
PAD, pulsed amperometric detection; PDA, photo diode array;
UGT, UDP-glucuronosyl transferase
(20) Kovalsky Paris, M. P.; Schweiger, W.; Hametner, C.; Stuckler,
R.; Muehlbauer, G. J.; Varga, E.; Krska, R.; Berthiller, F.; Adam, G.
Zearalenone-16-O-glucoside: a new masked mycotoxin. J. Agric. Food.
Chem. 2014, 62, 1181−1189.
4
735
DOI: 10.1021/acs.jafc.5b00806
J. Agric. Food Chem. 2015, 63, 4728−4736

Journal of Agricultural and Food Chemistry
Article
(
21) Yoshimoto, K.; Tsuda, Y. General path of O-acyl migration in D-
glucose derivatives: acyl migration of methyl mono-O-myristoyl-α- and
β-D-glucopyranosides and mono-O-myristoyl-D-glucopyranoses. Chem.
Pharm. Bull. 1983, 31, 4324−34.
(
22) Sandermann, H. Higher-plant metabolism of xenobiotics-the
green liver concept. Pharmacogenetics 1994, 4, 225−241.
(
23) Walravens, J.; Mikula, H.; Rychlik, M.; Asam, S.; Ediage, E. N.;
Di Mavungu, J. D.; Van Landschoot, A.; Vanhaecke, L.; De Saeger, S.
Development and validation of an ultra-high-performance liquid
chromatography tandem mass spectrometric method for the
simultaneous determination of free and conjugated Alternaria toxins
in cereal-based foodstuffs. J. Chromatogr. A 2014, 1372, 91−101.
(
24) Mikula, H.; Skrinjar, P.; Sohr, B.; Ellmer, D.; Hametner, C.;
Froehlich, J. Total synthesis of masked Alternaria mycotoxins-sulfates
and glucosides of alternariol (AOH) and alternariol-9-methyl ether
(
AME). Tetrahedron 2013, 69, 10322−10330.
4
736
DOI: 10.1021/acs.jafc.5b00806
J. Agric. Food Chem. 2015, 63, 4728−4736