HT-2 toxin 4-glucuronide as new T-2 toxin metabolite: Enzymatic synthesis, analysis, and species specific formation of T-2 and HT-2 toxin glucuronides by rat, mouse, pig, and human liver microsomes

Article abstract of DOI:10.1021/jf302571y

Glucuronides of the mycotoxin T-2 toxin and its phase I metabolite HT-2 toxin are important phase II metabolites under in vivo and in vitro conditions. Since standard substances are essential for the direct quantitation of these glucuronides, a method for the enzymatic synthesis of T-2 and HT-2 toxin glucuronides employing liver microsomes was optimized. Structure elucidation by nuclear magnetic resonance spectroscopy (NMR) and mass spectrometry revealed that besides T-2 toxin glucuronide and HT-2 toxin 3-glucuronide also the newly identified isomer HT-2 toxin 4-glucuronide was formed. Glucuronidation of T-2 and HT-2 toxin in liver microsomes of rat, mouse, pig, and human was compared and metabolites were analyzed directly by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). A distinct, species specific pattern of glucuronidation of T-2 and HT-2 toxin was observed with interesting interindividual differences. Until recently, glucuronides have frequently been analyzed indirectly by quantitation of the aglycone after enzymatic cleavage of the glucuronides by β-glucuronidase. Therefore, the hydrolysis efficiencies of T-2 and HT-2 toxin glucuronides using β-glucuronidases from Helix pomatia, bovine liver, and Escherichia coli were compared.

Full text of DOI:10.1021/jf302571y

Article
pubs.acs.org/JAFC
HT‑2 Toxin 4‑Glucuronide as New T‑2 Toxin Metabolite: Enzymatic
Synthesis, Analysis, and Species Specific Formation of T‑2 and HT‑2
Toxin Glucuronides by Rat, Mouse, Pig, and Human Liver
Microsomes
Tanja Welsch and Hans-Ulrich Humpf*
Institute of Food Chemistry, Westfalische Wilhelms-Universitat Munster, Corrensstraße 45, D-48149 Munster, Germany
̈
̈
̈
̈
ABSTRACT: Glucuronides of the mycotoxin T-2 toxin and its phase I metabolite HT-2 toxin are important phase II
metabolites under in vivo and in vitro conditions. Since standard substances are essential for the direct quantitation of these
glucuronides, a method for the enzymatic synthesis of T-2 and HT-2 toxin glucuronides employing liver microsomes was
optimized. Structure elucidation by nuclear magnetic resonance spectroscopy (NMR) and mass spectrometry revealed that
besides T-2 toxin glucuronide and HT-2 toxin 3-glucuronide also the newly identified isomer HT-2 toxin 4-glucuronide was
formed. Glucuronidation of T-2 and HT-2 toxin in liver microsomes of rat, mouse, pig, and human was compared and
metabolites were analyzed directly by liquid chromatography coupled with tandem mass spectrometry (LC−MS/MS). A distinct,
species specific pattern of glucuronidation of T-2 and HT-2 toxin was observed with interesting interindividual differences. Until
recently, glucuronides have frequently been analyzed indirectly by quantitation of the aglycone after enzymatic cleavage of the
glucuronides by β-glucuronidase. Therefore, the hydrolysis efficiencies of T-2 and HT-2 toxin glucuronides using β-
glucuronidases from Helix pomatia, bovine liver, and Escherichia coli were compared.
KEYWORDS: mycotoxin, Fusarium, trichothecene, T-2 toxin, HT-2 toxin, liver microsomes, glucuronidation, β-glucuronidase,
enzymatic synthesis, glucuronide, LC−MS/MS, NMR, rat, mouse, pig, human
tolerable daily intake of 100 ng/kg b.w. was recently established
for the sum of T-2 and HT-2 toxin.2,3
INTRODUCTION
■
The mycotoxins T-2 toxin and HT-2 toxin are toxic secondary
fungal metabolites, produced by several Fusarium species like
Fusarium sporotrichioides, Fusarium poae, or Fusarium langse-
thiae. These fungi infect crops in the field or during storage.
Therefore, T-2 toxin and HT-2 toxin are common contam-
inants in cereals like wheat, barley, maize and especially oat and
products thereof. T-2 and HT-2 toxin are prominent members
of the group of trichothecene mycotoxins, which are
characterized by a tetracyclic sesquiterpenoid 12,13-epoxy-
trichothec-9-ene structure (Figure 1).1,2 Among the toxic
effects observed for T-2 toxin are inhibition of protein and
DNA synthesis, induction of apoptosis, and immuno- and
hematotoxicity. HT-2 toxin is not only a natural contaminant in
cereals, but also the main metabolite of T-2 toxin. Toxic effects
can partly be ascribed to HT-2 toxin, and for this reason, a
T-2 toxin is rapidly metabolized and depending on the test
system and the animal species a variety of metabolites is
formed. Main reactions involved in T-2 metabolism are
hydrolysis, hydroxylation, de-epoxidation, acetylation and
conjugation to polar moieties. Deacetylation of T-2 toxin at
position 4 forming HT-2 toxin is one of the most important
phase I metabolic reactions, shown for example in micro-
somes,4 human cells in primary culture5,6 or animals.3,7,8
Additionally, phase II reactions, especially glucuronidation of
T-2 toxin, HT-2 toxin and further phase I metabolites, are
essential for the metabolism and excretion. The incubation of
T-2 toxin with rat liver microsomes in the presence of uridine
5′-diphosphoglucuronic acid (UDPGA) leads to the formation
of HT-2 toxin 3-glucuronide.9
After administration of tritium labeled T-2 toxin and analysis
of the metabolites by thin layer chromatography (TLC)
combined with radioactivity detection it was observed in
several in vitro and in vivo studies that a considerable percentage
of the radioactivity remained at the initial spotting sites of the
TLC plates. These polar substances were supposed to be
glucuronides. When these samples were subsequently treated
with β-glucuronidase, mainly HT-2 toxin was liberated. After
perfusion of rat liver for 1 h with tritium labeled T-2 toxin, for
Received: June 13, 2012
Revised: September 11, 2012
Accepted: September 12, 2012
Published: September 12, 2012
Figure 1. Chemical structure of T-2 toxin and HT-2 toxin.
© 2012 American Chemical Society
10170
dx.doi.org/10.1021/jf302571y | J. Agric. Food Chem. 2012, 60, 10170−10178

Journal of Agricultural and Food Chemistry
Article
type B1 from bovine liver, type H5 from Helix pomatia and type IX-A
from Escherichia coli were bought from Sigma-Aldrich GmbH (Seelze,
Germany). Livers of female pigs (Angler/Sattel x Pietrain) were
obtained from Gut Wewel (Senden, Germany). In vitro CYP M human
example, analysis of the bile showed that polar conjugates
remaining at the initial spotting site of the TLC plate accounted
for 88% of the radioactivity in bile. After hydrolysis of these
polar compounds with β-glucuronidase HT-2 toxin accounted
for 80% of the radioactivity, 3′-hydroxy HT-2 toxin for 11% and
T-2 tetraol for 1%.10 In another study, about half of the 34% of
radioactivity that appeared at the initial spotting site of TLC
plates after 24 h of incubation of rat hepatocytes with T-2 toxin
was associated with HT-2 toxin.11
liver microsomes (50 donors pooled), female Gottingen minipig liver
̈
microsomes, female ICR/CD mouse liver microsomes and female
Wistar rat liver microsomes were purchased from Celsis In Vitro
Technologies (Brussels, Belgium). T-2 and HT-2 toxin were isolated
from cultures of F. sporotrichioides in our laboratory17 and d3-T-2 toxin
was synthesized in our laboratory according to Beyer et al.17
Preparation of Microsomes. Pig livers were removed shortly
after the death of the animals, frozen in liquid nitrogen and stored at
−80 °C until further usage. Microsomes were prepared according to
Lake.18 Protein content was determined by Bradford-assay19 using
bovine serum albumin as reference material.
When rats were administered tritium labeled T-2 toxin
intravascularly or intraduodenally, roughly half of the radio-
activity in bile was linked to 3′-hydroxy HT-2 toxin and 25% to
polar compounds remaining at the initial spotting site of the
TLC plate. Enzymatic treatment of these polar fractions
resulted in 59% HT-2 toxin and only 17% 3′-hydroxy HT-2
toxin, although this was the main unconjugated metabolite in
bile.12
Mainly T-2 toxin, HT-2 toxin, 3′-hydroxy T-2 toxin and 3′-
hydroxy HT-2 toxin were detected in bile and urine of pigs 4 h
after intravascular dosing with T-2 toxin. Enzymatic treatment
revealed that only about half of the 3′-hydroxy HT-2 toxin was
present in the form of its glucuronide, whereas T-2 and HT-2
toxin were almost completely conjugated to glucuronic acid,
thus emphasizing the importance of HT-2 toxin glucuronide as
a metabolite of T-2 toxin. Overall, glucuronides represented
about 77% of the radioactivity excreted in the bile and 63% of
the radioactivity excreted in the urine of these pigs.13
Incubation with Liver Microsomes. For preparative isolation of
glucuronides, pig liver microsomes were first mixed with 10 μg of
alamethicin and kept on ice for 10 min before the other substances
were added. Incubations were carried out in a total volume of 250 μL
in 25 mM TRIS buffer pH 7.4 containing 5 mM MgCl2, 3.8 mM
UDPGA, 0.8 mM T-2 toxin and 1 mg of microsomal protein. After
incubation for 2 h at 37 °C, the reaction was stopped by addition of
500 μL of acetonitrile. The precipitate was removed by centrifugation
(16 000g for 10 min) and several assays were combined for further
purification. For comparison of glucuronidation with different liver
microsomes, incubation was slightly modified; 50 mM TRIS buffer,
containing 10 mM MgCl2, 0.34 mM UDPGA, 0.086 mM T-2 toxin
and 500 μg of microsomal protein was used. No alamethicin was
added. Incubations were carried out in triplicate.
Isolation and Purification of Glucuronides. For the isolation of
glucuronides, acetonitrile in the combined reaction mixtures was
evaporated using a BA-VC-300H vacuum concentrator (H. Saur,
Reutlingen, Germany). The aqueous residue was then extracted three
times with equal volumes of chloroform. The aqueous phase was
acidified with formic acid to pH = 3 and loaded on Bond Elut Plexa
SPE columns (200 mg) (Agilent, Waldbronn, Germany) which had
been conditioned with 5 mL of methanol and equilibrated with 5 mL
of water. Columns were washed ten times with 3 mL of water and five
times with 3 mL of 5% (v/v) methanol. Substances were eluted with
60% (v/v) methanol. After concentration in a vacuum concentrator,
glucuronides were purified by preparative HPLC on a Varian Pro Star
system (Varian, Waldbronn, Germany) using a 250 × 9.4 mm i.d.; 5
μm, Eclipse XDB C18 column (Agilent, Waldbronn, Germany).
Mobile phase was water (A) and a mixture of acetonitrile and water in
a ratio of 99.9%:0.1% (v/v) (B), both containing 5 mM ammonium
acetate. The gradient was as follows: 22% B for 22 min, then a change
to 30% B in 1 min and an isocratic step at 30% B for 9 min. The
column was washed for 1 min with 90% B and equilibrated for 8 min
at starting conditions. Flow rate was set to 4.5 mL/min. Fractions (30
s) were collected, checked for glucuronides using LC−MS/MS,
combined and concentrated in a vacuum concentrator. To remove
salts, a final purification was performed on a Bond Elut Plexa SPE
cartridge as described above.
Purity Determination by HPLC-ELSD. Purity of isolated
glucuronides was determined on a Shimadzu LC 20AT HPLC system
(Shimadzu, Kyoto, Japan) equipped with a UV (220 nm, 254 nm) and
an evaporative light scattering detector (ELSD) ELSD LT (Shimadzu,
Kyoto, Japan). ELSD parameters were: nebulizer temperature 40 °C,
air pressure 2.5 bar. Separation was performed on a 250 mm × 4.6 mm
i.d., 5 μm, Gemini C18 column (Phenomenex, Aschaffenburg,
Germany), using the mobile phase as described above for preparative
HPLC at a flow rate of 1 mL/min. The gradient started at 25% B,
which was held for 3 min, and afterward increased linearly in 27 min to
100% B, which was held for 2 min. For calculating the purity, the peak
area of the peak corresponding to the glucuronide was compared to
the sum of all peak areas.
In the studies cited above, analysis of glucuronides was done
indirectly by hydrolysis with β-glucuronidase. This widely
employed method for the quantitation of glucuronides
compares the amount of a metabolite before and after
enzymatic cleavage of its glucuronides, assuming the difference
to be the glucuronide content of the sample. This method,
however, suffers from several drawbacks, for instance loss of
structural information, time-consuming sample preparation or
risk of underestimation of glucuronide content due to
incomplete cleavage by β-glucuronidase treatment. Conse-
quently, during recent years, efforts have been made to directly
quantitate glucuronides.14−16
For direct quantitative analysis, glucuronide standards are
necessary, but they are not commercially available. Therefore,
one aim of this work was to synthesize glucuronides of T-2
toxin and HT-2 toxin enzymatically using liver microsomes.
There are only limited data on glucuronidation of T-2 and
HT-2 toxin by humans. Only one report exists about the
occurrence of T-2 toxin glucuronide and HT-2 toxin 3-
glucuronide in a human colon carcinoma derived cell line (HT-
29) as well as in human primary renal proximal tubule epithelial
cells (RPTEC).6 Additionally, information on species specific
differences in glucuronidation of T-2 and HT-2 toxin is scarce.
Thus, glucuronidation of T-2 and HT-2 toxin by liver
microsomes of rat, mouse, pig and human was compared and
glucuronides were analyzed directly by LC−MS/MS. More-
over, the degree of hydrolysis of T-2 and HT-2 toxin
glucuronides by different β-glucuronidases was examined.
MATERIALS AND METHODS
■
Chemicals and Materials. All chemicals were purchased from
Sigma-Aldrich GmbH (Seelze, Germany), Carl Roth GmbH + Co. KG
(Karlsruhe, Germany), VWR International GmbH (Darmstadt,
Germany) or AppliChem GmbH (Darmstadt, Germany). Water was
purified by a Milli-Q Gradient A10 system (Millipore, Schwalbach,
Germany). Uridine 5′-diphosphoglucuronic acid, β-glucuronidases
Exact Mass Measurements and Fragmentation. Measure-
ments of the exact mass and the fragmentation patterns of
glucuronides were done by Fourier transformation mass spectrometry
(FTMS) on a LTQ Orbitrap XL mass spectrometer (Thermo Fisher
10171
dx.doi.org/10.1021/jf302571y | J. Agric. Food Chem. 2012, 60, 10170−10178

Journal of Agricultural and Food Chemistry
Article
Scientific, Bremen, Germany) using heated electrospray ionization
(HESI). Data were processed using Xcalibur software version 2.0.7
SP1 (Thermo Fisher Scientific). Solutions of the purified glucuronides
were introduced by a syringe pump (5 μL/min). The mass
spectrometer was set to positive mode and a resolution of 100 000.
Parameters were as follows: capillary temperature 225 °C, vaporizer
temperature 50 °C, sheath gas flow 8, aux gas flow 5, source voltage 3
kV, capillary voltage 12 V, tube lens voltage 102 V. Product ion spectra
were acquired using collision induced dissociation (CID) at 20%.
Exact masses measured by FTMS were as follows: HT-2 toxin 3-
glucuronide, found m/z 618.2749, calculated for [C28H40O14 + NH4]+
618.2756; HT-2 toxin 4-glucuronide, found m/z 618.2754, calculated
for [C28H40O14 + NH4]+ 618.2756; T-2 toxin glucuronide, found m/z
660.2853, calculated for [C30H42O15 + NH4]+ 660.2862.
using LC−MS/MS. After incubation and centrifugation, supernatants
were diluted 1:10 with water, and 25 ng of d3-T-2 toxin was added as
internal standard. To calculate calibration curves for each analyte, the
ratios of the peak area of the analyte to the peak area of d3-T-2 toxin
were plotted against analyte concentration. Analysis was done on an
Agilent 1100 series HPLC (Agilent, Waldbronn, Germany) linked to
an API 4000 QTrap mass spectrometer (Applied Biosystems,
Darmstadt, Germany), using Analyst 1.4.2 software (Applied
Biosystems) for data acquisition.
HPLC Parameters. Separation was performed on a 50 mm × 2.0
mm i.d., 2.4 μm, Pursuit UPS C18 column (Agilent, Waldbronn,
Germany) equipped with a 4 × 2.0 mm i.d. C18 guard column
(Phenomenex, Aschaffenburg, Germany) using the mobile phase as
described above for preparative HPLC. The gradient started at 20% B
and increased to 100% B in 8 min. The column was washed with 100%
B for 1 min and equilibrated at starting conditions for 6 min. Flow rate
was 300 μL/min, column oven temperature was set to 40 °C and
injection volume to 50 μL.
MS/MS Parameters. The mass spectrometer was operated in the
positive multiple reaction monitoring (MRM) mode, quadrupoles
were set at unit resolution. For electrospray ionization, the ion voltage
was set to 5500 V, zero grade air served as nebulizer gas (35 psi) and
heated to 350 °C as turbo gas (45 psi). Nitrogen was employed as
curtain gas (20 psi) and as collision gas for fragmentation of the [M +
NH4]+ adducts (3.5 × 10−5 Torr). Parameters for each mass transition
were optimized by direct infusion of the analytes. Values for
declustering potential (DP), collision energy (CE) and cell exit
potential (CXP) are given in brackets. T-2 toxin, 484−305 (DP 65 V,
CE 19 V, CXP 19 V), 484−215 (DP 60 V, CE 24 V, CXP 14 V); d3-T-
2 toxin, 487−308 (DP 65 V, CE 19 V, CXP 19 V); HT-2 toxin, 442−
263 (DP 54 V, CE 22 V, CXP 18 V), 442−215 (DP 54 V, CE 18 V,
CXP 16 V); HT-2 toxin glucuronide, 618−215 (DP 61 V, CE 28 V,
CXP 17 V), 618−245 (DP 61 V, CE 24 V, CXP 19 V), 618−263 (DP
61 V, CE 26 V, CXP 17 V); T-2 toxin glucuronide, 660−305 (DP 65
V, CE 25 V, CXP 15 V), 660−215 (DP 65 V, CE 28 V, CXP 21 V).
Entrance potential (EP) was 10 V. Each mass transition was monitored
for 150 ms. The following transitions served as quantifiers: 484−305
(T-2 toxin), 442−215 (HT-2 toxin), 618−215 (HT-2 toxin 4-
glucuronide) and 618−263 (HT-2 toxin 3-glucuronide).
NMR Measurements. NMR spectra were recorded on a Bruker
DPX-400 (Bruker BioSpin, Rheinstetten, Germany) or on a Varian
600 unity plus (Varian, Palo Alto, CA) NMR spectrometer. Samples
were dissolved in d4-methanol and signals are reported referenced to
methanol. For structure elucidation and assignment of NMR signals,
1H, 13C, COSY, HMBC, HSQC, TOCSY and NOE measurements
were conducted.
1
HT-2 toxin 3-β-D-glucuronide: H NMR (400 MHz, d4-MeOD) δ:
5.75 (d, 1 H, J = 5.8 Hz, H-10), 5.31 (d, 1 H, J = 5.4 Hz, H-8), 4.69 (d,
1 H, J = 7.8 Hz, H-1′), 4.60 (d, 1 H, J = 3.0 Hz, H-4), 4.38 (dd, 1 H, J
= 5.0 Hz, 3.0 Hz, H-3), 4.31 (d, 1 H, J = 12.5 Hz, H-15), 4.23 (d, 1 H,
J = 5.9 Hz, H-11), 3.99 (d, 1 H, J = 12.5 Hz, H-15), 3.69 (d, 1 H, J =
9.5 Hz, H-5′), 3.65 (d, 1 H, J = 5.0 Hz, H-2), 3.49 (1 H, H-4′), 3.43 (t,
1 H, J = 8.9 Hz, H-3′), 3.33 (H-2′), 2.97 (d, 1 H, J = 4.0 Hz, H-13),
2.78 (d, 1 H, J = 4.0 Hz, H-13), 2.37 (dd, 1 H, J = 15.2 Hz, 5.8 Hz, H-
7), 2.16 (m, 2 H, H-18), 2.11−2.06 (m, H-19), 2.05 (s, 3 H, acetyl-
CH3), 1.97 (d, 1 H, J = 15.2 Hz, H-7), 1.73 (s, 3 H, H-16), 0.97 (d, 3
H, J = 3.2 Hz, H-20/21), 0.96 (d, 3 H, J = 3.2 Hz, H-20/21), 0.80 (s, 3
H, H-14).
13C NMR (100 MHz, d4-MeOD) δ: 174.1 (C-17), 172.2 (acetyl-
CO), 137.0 (C-9), 125.5 (C-10), 103.4 (C-1′), 86.1 (C-3), 80.3 (C-
2), 80.0 (C-4), 78.1 (C-3′), 74.4 (C-2′), 73.5 (C-4′), 69.6 (C-8), 68.6
(C-11), 65.9 (C-15), 65.3 (C-12), 50.0 (C-5), 47.3 (C-13), 44.6 (C-
18), 44.0 (C-6), 28.7 (C-7), 26.8 (C-19), 22.8 (C-20/21), 22.7 (C-20/
21), 21.1 (acetyl-CH3), 20.5 (C-16), 7.4 (C-14).
1
Animal Study. The animal study with pigs was performed at the
Institute of Animal Nutrition, Friedrich-Loeffler-Institute (FLI),
Braunschweig, Germany. Experiments and procedures were conducted
according to the European Community regulations concerning the
protection of experimental animals and the guidelines of the Regional
Council of Braunschweig, Lower Saxony, Germany (File Number
33.9.42502-04-054/09). The experimental details of this study will be
HT-2 toxin 4-β-D-glucuronide: H NMR (400 MHz, d4-MeOD) δ:
5.74 (d, 1 H, J = 5.9 Hz, H-10), 5.32 (d, 1 H, J = 5.5 Hz, H-8), 4.74 (d,
1 H, J = 2.6 Hz, H-4), 4.65 (d, 1 H, J = 7.8 Hz, H-1′), 4.32 (d, 1 H, J =
12.5 Hz, H-15), 4.29−4.25 (m, 2 H, H-3, H-11), 4.03 (d, 1 H, J = 12.5
Hz, H-15), 3.69 (d, 1 H, J = 9.0 Hz, H-5′), 3.51 (d, 1 H, J = 4.9 Hz, H-
2), 3.48 (1 H, H-4′), 3.42 (t, 1 H, J = 8.9 Hz, H-3′), 3.29 (H-2′), 3.04
(d, 1 H, J = 3.9 Hz, H-13), 2.87 (d, 1 H, J = 3.9 Hz, H-13), 2.38 (dd, 1
H, J = 15.2 Hz, 5.9 Hz, H-7), 2.17 (m, 2 H, H-18), 2.13−2.08 (m, H-
19), 2.06 (s, 3 H, acetyl-CH3), 1.90 (d, 1 H, J = 15.2 Hz, H-7), 1.73 (s,
3 H, H-16), 0.97 (d, 3 H, J = 3.3 Hz, H-20/21), 0.96 (d, 3 H, J = 3.3
Hz, H-20/21), 0.84 (s, 3 H, H-14).
published elsewhere by Danicke et al. and will only be briefly
̈
summarized here. Female pigs were dosed with a single dose of T-2
toxin either iv with 80 μg/kg b.w. (n = 2) or orally with 4 mg/animal
(n = 4) resulting in approximately 89−100 μg/kg b.w. Urine was
collected after 3, 6, 12, 24, and 48 h and selected samples were
analyzed after syringe filtration (0.45 μm, regenerated cellulose)
employing LC−MS/MS as described above.
13C NMR (100 MHz, d4-MeOD) δ: 174.2 (C-17), 172.3 (acetyl-
CO), 137.2 (C-9), 125.3 (C-10), 102.5 (C-1′), 86.4 (C-4), 80.8 (C-
2), 79.8 (C-3), 77.6 (C-3′), 74.0 (C-2′), 73.6 (C-4′), 69.5 (C-8), 68.5
(C-11), 66.0 (C-15), 65.9 (C-12), 50.1 (C-5), 47.7 (C-13), 44.5 (C-
18), 44.3 (C-6), 29.1 (C-7), 26.9 (C-19), 22.8 (C-20/21), 22.7 (C-20/
21), 21.5 (acetyl-CH3), 20.4 (C-16), 7.7 (C-14).
Incubation with β-Glucuronidase. Hydrolysis of T-2 and HT-2
toxin glucuronides was carried out in a total volume of 250 μL. To
cleave the glucuronides, 200 μL of a glucuronide solution in water or
in spiked pig urine was mixed with 50 μL of β-glucuronidase
suspended in 750 mM buffer and incubated for definite times at 37 °C.
For incubation with β-glucuronidase type B1 from bovine liver and
type H5 from H. pomatia, sodium acetate buffer (pH 4.5) was used in
contrast to incubation with β-glucuronidase type IX-A from E. coli
where KH2PO4/Na2HPO4 buffer (pH 6.5) was employed. Incubation
was terminated by addition of 250 μL of acetonitrile and the
precipitate was removed by centrifugation (16 000g for 5 min). After
1:10 dilution of the samples and addition of 25 ng of d3-T-2 toxin,
analysis was carried out as described above. To check for their stability,
T-2 and HT-2 toxin were incubated under the same conditions and
glucuronides were incubated without addition of β-glucuronidase to
the buffer. T-2 and HT-2 toxin as well as HT-2 toxin glucuronides
were used in a concentration of 10 mM in the 250 μL incubation
1
T-2 toxin 3-β-D-glucuronide: H NMR (600 MHz, d4-MeOD) δ:
6.03 (d, 1 H, J = 3.0 Hz, H-4), 5.78 (d, 1 H, J = 5.9 Hz, H-10), 5.32 (d,
1 H, J = 5.8 Hz, H-8), 4.66 (dd, 1 H, J = 5.0 Hz, 3.0 Hz, H-3), 4.47 (d,
1 H, J = 7.7 Hz, H-1′), 4.40 (d, 1 H, J = 5.9 Hz, H-11), 4.38 (d, 1 H, J
= 12.5 Hz, H-15), 4.09 (d, 1 H, J = 12.5 Hz, H-15), 3.79 (d, 1 H, J =
5.0 Hz, H-2), 3.60−3.30 (sugar), 3.03 (d, 1 H, J = 4.0 Hz, H-13), 2.83
(d, 1 H, J = 4.0 Hz, H-13), 2.39 (dd, 1 H, J = 15.1 Hz, 5.9 Hz, H-7),
2.16 (m, 2 H, H-18), 2.09 (s, 3 H, acetyl-CH3), 2.08 (s, 3 H, acetyl-
CH3), 1.91 (d, 1 H, J = 17.3 Hz, H-7), 1.73 (s, 3 H, H-16), 0.97 (d, 3
H, J = 4.5 Hz, H-20/21), 0.96 (d, 3 H, J = 4.4, H-20/21), 0.73 (s, 3 H,
H-14). H-19 is overlapping with the signals of acetyl-CH3.
LC−MS/MS Measurement. For the comparison of the metabo-
lism of T-2 toxin in liver microsomes of rat, mouse, pig and human,
T-2 toxin, HT-2 toxin and HT-2 toxin glucuronides were quantified
10172
dx.doi.org/10.1021/jf302571y | J. Agric. Food Chem. 2012, 60, 10170−10178

Journal of Agricultural and Food Chemistry
Article
Figure 2. Overview of the formed metabolites when incubating T-2 toxin with pig liver microsomes in the presence of the cofactor UDPGA.
mixture. The initial amount of T-2 toxin glucuronide could not be
quantified exactly; hydrolysis efficiencies were examined by comparing
the amount of T-2 toxin glucuronide before and after incubation. All
incubations were done in triplicate
RESULTS AND DISCUSSION
■
Glucuronide Synthesis and Isolation. Although glucur-
onides of T-2 toxin and its phase I metabolite HT-2 toxin are
key metabolites in in vitro and in vivo studies, these
glucuronides have until now usually been analyzed indirectly
after hydrolysis with β-glucuronidase.10−13 For direct analysis of
Figure 3. LC−MS/MS chromatogram of a mixture obtained after 2 h
glucuronides using sensitive and selective LC−MS/MS
techniques, standard substances are essential. Since they are
not commercially available, a method for the enzymatic
synthesis and purification of T-2 and HT-2 toxin glucuronides
using pig liver microsomes was developed and glucuronides
were characterized by MS and NMR.
To synthesize T-2 and HT-2 toxin glucuronides, T-2 toxin
was incubated in a buffered system with pig liver microsomes in
the presence of the cofactor uridine 5′-diphosphoglucuronic
acid (UDPGA). Liver microsomes served as a source for the
uridine 5′-diphospho-glucuronosyltransferases, which transfer
the glucuronic acid to a substrate like T-2 or HT-2 toxin
incubation of T-2 toxin with pig liver microsomes and UDPGA. The
following mass transitions are shown: T-2 toxin 484−305; HT-2 toxin
442−215; T-2 toxin glucuronide 660−305; HT-2 toxin glucuronides
618−215 (red), 618 − 263 (black).
well as HT-2 toxin were glucuronidated, since one peak with
mass transitions expected for the T-2 toxin glucuronide and
surprisingly two peaks with mass transitions expected for the
HT-2 toxin glucuronide were observed. No glucuronide peaks
were present when microsomes, T-2 toxin, or UDPGA was
omitted from the reaction mixture. Different incubation
conditions were tested to optimize the glucuronide formation.
A ratio of about 4 to 1 of molar amounts of UDPGA to T-2
toxin was chosen as a compromise between glucuronide yield
and high cost for UDPGA. Addition of alamethicin, a protein
forming pores in the microsomal vesicles, increased the
glucuronide yield, as described in literature for estradiol
glucuronidation.25
(Figure 2). Besides the Konigs-Knorr synthesis, this enzymatic
̈
approach is widely described in literature to obtain glucur-
onides.20−22 For instance, Roush et al.9 were able to isolate
HT-2 toxin 3-glucuronide after incubation of T-2 toxin with rat
liver microsomes and UDPGA. Figure 3 depicts a typical LC−
MS/MS chromatogram of an incubation mixture of T-2 toxin
with pig liver microsomes after 2 h of incubation. In accordance
to literature data,4,23 T-2 toxin was hydrolyzed quickly by
certain carboxylesterases24 present in pig liver microsomes
forming HT-2 toxin. Moreover, it was shown that T-2 toxin as
After incubation, the supernatant was extracted with
chloroform thereby removing most of the T-2 and HT-2
toxin, while the glucuronides remained in the aqueous phase.
During the following purification step by solid phase extraction,
10173
dx.doi.org/10.1021/jf302571y | J. Agric. Food Chem. 2012, 60, 10170−10178

Journal of Agricultural and Food Chemistry
Article
the amount of buffer salts was reduced. Separation of the
glucuronide isomers was then achieved by preparative HPLC.
The yield was about 5−6% for each HT-2 toxin glucuronide
isomer and below 1% for the minor component T-2 toxin
glucuronide. Similar yields were obtained by Roush et al.9 who
isolated about 10% of HT-2 toxin 3-glucuronide when
incubating T-2 toxin with rat liver microsomes. However, the
relatively low yield could be increased in our study by recovery
of T-2 and HT-2 toxin from the incubation mixture by solvent
extraction as described above and reuse in further incubation
experiments. Purity of the HT-2 toxin glucuronides, measured
with an evaporative light scattering detector (ELSD), was about
97%. HPLC-UV, LC−MS/MS and NMR measurements only
showed traces of impurities. For T-2 toxin glucuronide ELSD
measurements resulted in a purity of about 40%. Impurities in
the case of the T-2 toxin glucuronide are supposed to be caused
by salts as one larger injection peak occurred in the ELSD
chromatogram and NMR measurements of T-2 toxin showed
only some minor impurities.
newly described metabolite HT-2 toxin 4-glucuronide was
confirmed by HMBC data in which a correlation between the
anomeric H-1′ of the glucuronic acid moiety and C-4 of HT-2
toxin as well as a correlation between H-4 and C-1′ was
detected. The proton signal of the anomeric H-1′ of the
glucuronic acid residue appeared as a doublet. From the vicinal
coupling constant J = 7.8 Hz between H-1′ and H-2′ the
presence of two axial protons, and therefore, β-configuration of
the glucuronic acid moiety was inferred. 1H NMR signals could
be assigned for T-2 toxin glucuronide by comparing the signals
for T-2 toxin glucuronide to NMR data of T-2 toxin, although
only small amounts of T-2 toxin glucuronide were obtained and
1
the H NMR spectrum showed some impurities.
The exact masses measured for the ammonium adducts of
T-2 and the HT-2 toxin glucuronides were in accordance to the
calculated ones. In the product ion spectra of the ammonium
adducts of the glucuronides, fragment ions that are also typical
for unconjugated T-2 or HT-2 toxin were observed like m/z
215, 245, 305, or 365 for T-2 toxin glucuronide and m/z 215,
245, or 263 for the HT-2 toxin glucuronides (Figure 4).
Structure Elucidation by NMR and MS. Detailed
1
structure elucidation of the glucuronides was done using H,
13C, COSY, HSQC, HMBC and TOCSY NMR data. The HT-2
toxin glucuronide isomer eluting first in LC−MS/MS measure-
ments (Figure 3) was identified as HT-2 toxin 4-glucuronide, a
new metabolite which has not been described so far. The
second of the HT-2 toxin glucuronide peaks (Figure 3) was
1
found to be HT-2 toxin 3-glucuronide. H NMR data were in
accordance to literature data9 although some signals were
assigned differently based on the 2D-NMR experiments. First
hints for the identification of the glucuronidation sites of the
two HT-2 toxin glucuronide isomers were derived from
comparison of the 1H NMR data of the HT-2 toxin
glucuronides and those of HT-2 toxin (Table 1). Compared
1
Table 1. Comparison of H and 13C NMR Signals of Atoms
Near the Glucuronidation Site for HT-2 Toxin, HT-2 Toxin
3-Glucuronide, and HT-2 Toxin 4-Glucuronide
1H δ (ppm)
H-
atom HT-2 toxin HT-2 toxin 3-glucuronide HT-2 toxin 4-glucuronide
4
3
2
4.38
4.09
3.44
4.60
4.38
3.65
4.74
4.29−4.25
3.51
13C δ (ppm)
C-
atom HT-2 toxin HT-2 toxin 3-glucuronide HT-2 toxin 4-glucuronide
4
3
2
81.8
80.8
80.6
80.0
86.1
80.3
86.4
79.8
80.8
Figure 4. Product ion mass spectra of the [M + NH4]+ ions measured
by FTMS at collision induced dissociation (CID) of 20%. (A) HT-2
toxin 3-glucuronide, (B) HT-2 toxin 4-glucuronide, (C) T-2 toxin
glucuronide.
to HT-2 toxin, the resonance of H-4 is shifted downfield in
HT-2 toxin 3-glucuronide and even further downfield in HT-2
toxin 4-glucuronide. H-2 and H-3 on the other hand are more
deshielded in HT-2 toxin 3-glucuronide than in HT-2 toxin 4-
glucuronide and HT-2 toxin. Similar effects were observed in
the 13C NMR spectra, where the signals for C-3 or C-4 were
shifted downfield for HT-2 toxin 3-glucuronide or HT-2 toxin
4-glucuronide respectively (Table 1). The downfield shifts of
the 1H and 13C NMR signals of the atoms at the
glucuronidation site in relation to the signals of the parent
compound were comparable to those of other substances.22,26
The linkage of the glucuronic acid moiety to position 4 in the
Although these mass spectra of HT-2 toxin 3-glucuronide and
HT-2 toxin 4-glucuronide show similar fragment ions there are
distinct differences in their intensity. In the mass spectrum of
HT-2 toxin 4-glucuronide, signals resulting from the loss of the
isovaleric acid side chain (m/z 499) or the loss of the isovaleric
acid and acetic acid side chains (m/z 439) have a higher
intensity than the signal for HT-2 toxin (m/z 425) resulting
from the loss of the glucuronic acid moiety. For HT-2 toxin 3-
10174
dx.doi.org/10.1021/jf302571y | J. Agric. Food Chem. 2012, 60, 10170−10178

Journal of Agricultural and Food Chemistry
Article
Figure 5. Metabolites formed after 2 h of incubation of T-2 toxin with liver microsomes of rat, mouse, pig and human. (Left) HT-2 toxin, HT-2
toxin glucuronides, T-2 toxin shown as percent of these quantified metabolites. (Right) Details of the LC−MS/MS chromatograms depicting HT-2
toxin glucuronides (black) and T-2 toxin glucuronide (green) and their peak intensities.
glucuronide, however, the signal for m/z 425 has a higher
intensity than the signals for m/z 499 and 439. When
fragmenting HT-2 toxin 3-glucuronide the signal for m/z 263
is higher than the signal for m/z 215, in contrast to HT-2 toxin
4-glucuronide (Figure 4). Consequently, monitoring the ratio
of the mass transitions 618−215 and 618−263 is a further
confirmation of the structure of the respective HT-2 toxin
glucuronide besides the retention time in routine measure-
ments (Figure 3 and Figure 6). In the mass spectra of the
glucuronides in the negative mode the fragment ions m/z 175
and m/z 113 which are attributable to the glucuronic acid
moiety were observed. These fragments were also detected for
other glucuronides in the negative mode, further proving the
structure of the synthesized standard substances.21
Species Specific Glucuronide Formation by Liver
Microsomes of Rat, Mouse, Pig, and Human. Using the
synthesized glucuronide reference material, the formation of
T-2 and HT-2 toxin 3-glucuronide in a human colon carcinoma
cell line (HT-29) and primary human renal proximal tubule
epithelial cells (RPTEC) could be measured by LC-FTMS.6
Further information about T-2 toxin glucuronidation in
humans is lacking, especially data on glucuronide formation
in human liver. Microsomal glucuronidation of T-2 toxin has
only been examined using rat liver.9 To date, the in vivo
metabolism of T-2 toxin including glucuronidation has only
been studied in rats12 and pigs;13 however, glucuronides were
measured only indirectly by enzymatic cleavage. Therefore, to
investigate interspecies differences in glucuronidation, T-2 and
HT-2 toxin were incubated with microsomes of rat, mouse, pig,
and human livers. A LC−MS/MS method was developed
which allowed the direct analysis of T-2 toxin, HT-2 toxin and
their respective glucuronides.
less pronounced. Hydrolysis of T-2 toxin during incubation
with liver microsomes forming mainly HT-2 toxin besides other
minor hydrolysis products has frequently been reported in
literature.4,23,27 From experiments with different enzyme
inhibitors, it was concluded that certain carboxylesterases
were responsible for this hydrolysis.24 Ohta et al.27 reported a
higher capacity for hydrolysis of T-2 toxin in human liver
microsomes compared to mouse liver microsomes. Using rat
liver microsomes, even less HT-2 toxin was formed than with
mouse liver microsomes. When Kobayashi et al.23 or Knupp et
al.4 however incubated T-2 toxin with liver microsomes in the
presence of a NADPH generating system in order to study
hydroxylation, they noticed that T-2 toxin was much better
hydrolyzed by rat liver microsomes compared to mouse liver
microsomes, which is also in accordance to our results.
Liver microsomes of rat, mouse, pig, and human each
exhibited a distinct pattern of glucuronidation (Figure 5).
When incubating T-2 toxin with mouse liver microsomes, much
more T-2 toxin glucuronide was formed than by incubation
with microsomes of the other species. This might be due to the
slower hydrolysis of T-2 toxin by mouse liver microsomes. Only
trace amounts of T-2 toxin glucuronide were detected in
incubation mixtures containing rat liver microsomes (see
chromatograms in Figure 5). HT-2 toxin 3-glucuronide
accounted for about half of the quantified metabolites in
human liver microsomes. Liver microsomes of the other species
had a much lower glucuronidation rate (Figure 5). When
studying rat liver microsomes, HT-2 toxin 3-glucuronide
accounted for about 17% of the metabolites quantified.
Roush et al.9 found 49% of HT-2 toxin 3-glucuronide after 2
h of incubation of small amounts of tritium-labeled T-2 toxin
with β-naphthoflavone-induced hepatic microsomes from rats.
Differences in the extent of glucuronidation should be
considered when evaluating the toxicity of T-2 and HT-2
toxin toward several species. Interestingly, only pig liver
microsomes were able to form substantial amounts of HT-2
toxin 4-glucuronide, a new isomer, which has not been
identified before (Figure 5). In incubation mixtures with rat
liver microsomes, only trace amounts of HT-2 toxin 4-
glucuronide were found besides larger amounts of HT-2
A typical chromatogram of an incubation of T-2 toxin with
pig liver microsomes is displayed in Figure 3. Figure 5 depicts
the formation of metabolites after 120 min of incubation of T-2
toxin with different microsomes. T-2 toxin glucuronide could
not be quantified but differences between species could be
monitored by comparison of peak areas (see chromatograms in
Figure 5). As expected, T-2 toxin was hydrolyzed quickly by all
liver microsomes except for mouse liver, where hydrolysis was
10175
dx.doi.org/10.1021/jf302571y | J. Agric. Food Chem. 2012, 60, 10170−10178

Journal of Agricultural and Food Chemistry
Article
toxin 3-glucuronide, the only glucuronide identified by Roush
et al.9 When HT-2 toxin was employed (instead of T-2 toxin)
as starting material for incubations with microsomes, HT-2
toxin glucuronide formation was comparable. Stereo- and
regioselective formation of glucuronides as well as significant
differences in the rate of glucuronidation depending on the
animal species used as a source for the liver microsomes has
been observed for a variety of other substances.22,26,28,29 This
was supposed to be connected to differences concerning the
uridine 5′-diphospho-glucuronosyltransferase isoforms present
in each species.
Hydrolysis of Glucuronides with Different β-Glucur-
onidases. The enzymatic hydrolysis using β-glucuronidases is
routinely employed for indirect quantitation of glucuronide
metabolites. Various hydrolysis protocols exist and differences
in deconjugation efficiencies have been observed for several
substances depending e.g. on type and amount of β-
glucuronidase, matrix, incubation time or temperature.30,31
For T-2 and HT-2 toxin glucuronides this kind of information
is scarce. Therefore, to examine enzymatic hydrolysis of T-2
and HT-2 toxin glucuronides, buffered aqueous solutions or
spiked pig urine samples were treated with different amounts of
three commonly used types of β-glucuronidases (type H5 from
H. pomatia, type B1 from bovine liver, type IX-A from E. coli).
The degree of deconjugation over time was monitored by
comparing amounts of glucuronides before and after hydrolysis
as well as by quantitating formed T-2 toxin and HT-2 toxin. T-
2 and HT-2 toxin glucuronides were stable for 24 h at 37 °C in
buffered aqueous solutions or pig urine in the absence of β-
glucuronidase. The stability of the hydrolysis products T-2
toxin and HT-2 toxin in the incubation mixture was also tested.
While HT-2 toxin was stable for 24 h in the presence of any of
the three β-glucuronidases, T-2 toxin was only stable in
mixtures containing β-glucuronidase from bovine liver or E. coli.
When treated with β-glucuronidase from H. pomatia, T-2 toxin
was degraded and HT-2 toxin was formed. After 30 or 60 min
of incubation, only 76% or 37% of the originally applied T-2
toxin concentration remained. Similarly, after treatment of T-2
toxin glucuronide with β-glucuronidase from H. pomatia, not
only T-2 toxin was formed, but also HT-2 toxin 3-glucuronide
and HT-2 toxin. Consequently, β-glucuronidase from H.
pomatia is not suitable for the analysis of samples containing
T-2 toxin glucuronide or T-2 toxin.
To verify that HT-2 toxin 4-glucuronide was not only formed
in vitro by pig liver microsomes, urine of female pigs that were
given T-2 toxin orally or iv(obtained from S. Danicke, Institute
̈
of Animal Nutrition, Friedrich-Loeffler-Institute (FLI),
Braunschweig, Germany) was analyzed by LC−MS/MS. Figure
6 illustrates that HT-2 toxin 4-glucuronide, as well as HT-2
Figure 6. LC−MS/MS chromatogram of urine (collected after 6 h) of
a pig administered T-2 toxin (4 mg) orally. The chromatogram shows
2 mass transitions for HT-2 toxin glucuronides, 618−215 (red) and
618−263 (black).
Table 2 summarizes the hydrolysis efficiencies for 10 mM
solutions of HT-2 toxin glucuronides under various incubation
conditions. Results are given as percentage of HT-2 toxin
formed by enzymatic cleavage compared to the amount of
HT-2 toxin that would be formed after complete hydrolysis of
the applied glucuronide. In general it was observed, that longer
incubation times and higher amounts of enzyme improved
hydrolysis to a certain extent, which was also proven for other
glucuronides.30,32 Moreover, when glucuronides were dissolved
in urine instead of water, harsher reaction conditions were
toxin 3-glucuronide, was found in pig urine. Consequently,
both HT-2 toxin glucuronide isomers are also formed in vivo
and should be taken into account when studying T-2 toxin
metabolism. This information about HT-2 toxin glucuronide
isomers was lost when Corley et al.13 examined the metabolism
of T-2 toxin in pigs, since glucuronides in physiological samples
were analyzed indirectly after hydrolysis with β-glucuronidase.
Table 2. Hydrolysis of HT-2 Toxin 3-Glucuronide and HT-2 Toxin 4-Glucuronide with β-Glucuronidases under Different
Incubation Conditions
a
substance incubated
incubation time [h]
type (amount) of β-glucuronidase
matrix
% HT-2 toxin formed SD
b
HT-2 toxin 4-glucuronide
2
2
2
4
4
BL (250 U)
buffer
buffer
buffer
urine
urine
39.8 3.8
89.2 6.0
98.1 1.8
73.3 3.8
87.8 1.3
c
HP (250 U)
d
EC (250 U)
EC (500 U)
EC (1000 U)
HT-2 toxin 3-glucuronide
4
24
4
BL (500 U)
BL (1000 U)
HP (500 U)
HP (1000 U)
EC (250 U)
EC (2000 U)
EC (4000 U)
buffer
buffer
buffer
buffer
buffer
urine
urine
2.3 0.1
10.4 0.3
49.4 3.5
92.3 5.7
104.7 7.1
70.6 3.3
79.0 3.4
24
2
24
24
a
The amount of HT-2 toxin formed after hydrolysis standard deviation is given in percent of the amount of HT-2 toxin expected after complete
b
c
d
hydrolysis of the glucuronides. Bovine liver. H. pomatia. E. coli.
10176
dx.doi.org/10.1021/jf302571y | J. Agric. Food Chem. 2012, 60, 10170−10178

Journal of Agricultural and Food Chemistry
Article
(2) EFSA Panel on Contaminants in the Food Chain (CONTAM).
Scientific Opinion on the risks for animal and public health related to
the presence of T-2 and HT-2 toxin in food and feed. EFSA J. 2011, 9,
2481.
(3) Li, Y.; Wang, Z.; Beier, R. C.; Shen, J.; Smet, D. D.; De Saeger, S.;
Zhang, S. T-2 Toxin, a trichothecene mycotoxin: review of toxicity,
metabolism, and analytical methods. J. Agric. Food Chem. 2011, 59,
3441−3453.
necessary to achieve deconjugation probably due to interfer-
ences with the matrix. HT-2 toxin 4-glucuronide was
hydrolyzed more easily, whereas the hydrolysis of the more
prevalent HT-2 toxin 3-glucuronide required higher amounts of
enzyme and longer incubation times. Achieving the hydrolysis
of T-2 toxin glucuronide was even more difficult. When
incubated with 4000 U of E. coli β-glucuronidase for 24 h, about
half of the glucuronide was still not hydrolyzed. Distinct
differences were observed between the deconjugation efficien-
cies of the three enzymes. Bovine liver β-glucuronidase poorly
hydrolyzed HT-2 toxin 3-glucuronide. Similar problems were
observed by Pace10 when hydrolyzing the polar substances at
the initial spotting site that were left after analysis of
metabolites of T-2 toxin in bile by thin layer radio-
chromatography. The polar compounds proved to be poor
substrates for bovine liver β-glucuronidase, whereas after
hydrolysis with β-glucuronidase from limpets, about 92% of
the radiolabel was associated with free metabolites, mainly
HT-2 toxin. With enzymes from H. pomatia, a better hydrolysis
for HT-2 toxin glucuronides was obtained, but as discussed
above, this enzyme is not suitable for samples containing T-2
toxin. Applying E. coli β-glucuronidase, the highest hydrolysis
efficiency was achieved. Nevertheless, for hydrolysis of HT-2
toxin 3-glucuronide in urine, high amounts of enzyme and 24 h
of incubation were necessary. Although these were the only
conditions tested under which no more HT-2 toxin 3-
glucuronide could be detected in urine, only about 80% of
the expected amount of HT-2 toxin was recovered.
These examples demonstrate that not every deconjugation
protocol and enzyme is suitable for the hydrolysis of certain
glucuronides. Incomplete hydrolysis of glucuronides or side
reactions might lead to errors in the quantitation of these
metabolites. Moreover, sample preparation is time-consuming
and structural information is lost. These limitations emphasize
the need for glucuronide reference material for the direct
analysis of T-2 and HT-2 toxin glucuronides in physiological
samples. Direct analysis of glucuronides revealed species
specific differences in the formation of T-2 and HT-2 toxin
glucuronides by liver microsomes of rat, mouse, pig, and
human. The new metabolite HT-2 toxin 4-glucuronide was
detectable by LC−MS/MS not only in pig liver microsomes,
but also in vivo in pig urine.
(4) Knupp, C. A.; Swanson, S. P.; Buck, W. B. Comparative in vitro
metabolism of T-2 toxin by hepatic microsomes prepared from
phenobarbital-induced or control rats, mice, rabbits and chicken. Food
Chem. Toxicol. 1987, 25, 859−865.
(5) Konigs, M.; Mulac, D.; Schwerdt, G.; Gekle, M.; Humpf, H.-U.
̈
Metabolism and cytotoxic effects of T-2 toxin and its metabolites on
human cells in primary culture. Toxicology 2009, 258, 106−115.
(6) Weidner, M.; Welsch, T.; Hubner, F.; Schwerdt, G.; Gekle, M.;
̈
Humpf, H.-U. Identification and apoptotic potential of T-2 toxin
metabolites in human cells. J. Agric. Food Chem. 2012, 60, 5676−5684.
(7) Wu, Q.; Dohnal, V.; Huang, L.; Kuca, K.; Yuan, Z. Metabolic
pathways of trichothecenes. Drug Metab. Rev. 2010, 42, 250−267.
(8) Dohnal, V.; Jezkova, A.; Jun, D.; Kuca, K. Metabolic pathways of
T-2 toxin. Curr. Drug Metab. 2008, 9, 77−82.
(9) Roush, W. R.; Marletta, M. A.; Russo-Rodriguez, S.; Recchia, J.
Trichothecene metabolism studies. 2. Structure of 3-alpha-(1″beta-D-
glucopyranosiduronyl)-8-alpha-isovaleryloxy-scirpen-3,4-beta,15-triol
15-acetate produced from T-2 toxin in vitro. Tetrahedron Lett. 1985,
26, 5231−5234.
(10) Pace, J. G. Metabolism and clearance of T-2 mycotoxin in
perfused rat livers. Fundam. Appl. Toxicol. 1986, 7, 424−433.
(11) Thompson, W. L.; Pace, J. G.; O’Brien, J. C. In vitro metabolism
of T-2 mycotoxin. In Vitro Toxicol. 1989, 2, 139−150.
(12) Coddington, K. A.; Swanson, S. P.; Hassan, A. S.; Buck, W. B.
Enterohepatic circulation of T-2 toxin metabolites in the rat. Drug
Metab. Dispos. 1989, 17, 600−605.
(13) Corley, R. A.; Swanson, S. P.; Buck, W. B. Glucuronide
conjugates of T-2 toxin and metabolites in swine bile and urine. J.
Agric. Food Chem. 1985, 33, 1085−1089.
(14) Ketola, R. A.; Hakala, K. S. Direct analysis of glucuronides with
liquid chromatography-mass spectrometric techniques and methods.
Curr. Drug Metab. 2010, 11, 561−582.
(15) Kaushik, R.; Levine, B.; LaCourse, W. A brief review: HPLC
methods to directly detect drug glucuronides in biological matrices
(Part I). Anal. Chim. Acta 2006, 556, 255−266.
(16) Warth, B.; Sulyok, M.; Berthiller, F.; Schuhmacher, R.;
Fruhmann, P. Direct quantification of deoxynivalenol glucuronide in
human urine as biomarker of exposure to the Fusarium mycotoxin
deoxynivalenol. Anal. Bioanal. Chem. 2011, 401, 195−200.
(17) Beyer, M.; Ferse, I.; Humpf, H. U. Large-scale production of
selected type A trichothecenes: The use of HT-2 toxin and T-2 triol as
precursors for the synthesis of d3-T-2 and d3-HT-2 toxin. Mycotoxin
Res. 2009, 25, 41−52.
AUTHOR INFORMATION
Corresponding Author
*Tel: +49 251 8333391. Fax: +49 251 8333396. E-mail:
humpf@uni-muenster.de.
■
(18) Lake, B. G. Preparation and characterisation of microsomal
fractions for studies on xenobiotic metabolism. Biochem. Toxicol. 1987,
183−215.
Notes
The authors declare no competing financial interest.
(19) Bradford, M. M. A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the principle
of protein-dye binding. Anal. Biochem. 1976, 72, 248−254.
(20) Wu, X. N.; Murphy, P.; Cunnick, J.; Hendrich, S. Synthesis and
characterization of deoxynivalenol glucuronide: Its comparative
immunotoxicity with deoxynivalenol. Food Chem. Toxicol. 2007, 45,
1846−1855.
(21) Stevenson, D.; Hansen, R.; Loader, J.; Jensen, D.; Cooney, J.;
Wilkins, A. L. Preparative enzymatic synthesis of glucuronides of
zearalenone and five of its metabolites. J. Agric. Food Chem. 2008, 56,
4032−4038.
ACKNOWLEDGMENTS
We thank the working group of Sven Danicke (Institute of
Animal Nutrition, Friedrich-Loeffler-Institute (FLI), Braunsch-
weig, Germany) for providing the pig urine samples, Erika
Pfeiffer (KIT, Karlsruhe) for helpful hints concerning the
preparation of microsomes and the Kurzen family (Gut Wewel,
Senden) for supplying pig liver.
■
̈
REFERENCES
■
(22) Alonen, A.; Gartman, M.; Aitio, O.; Finel, M.; Yli-Kauhaluoma,
J.; Kostiainen, R. Synthesis, structure characterization, and enzyme
screening of clenbuterol glucuronides. Eur. J. Pharm. Sci. 2009, 37,
581−587.
(1) van der Fels-Klerx, H. J.; Stratakou, I. T-2 toxin and HT-2 toxin
in grain and grain-based commodities in Europe: Occurrence, factors
affecting occurrence, co-occurrence and toxicological effects. World
Mycotoxin J. 2010, 3, 349−367.
10177
dx.doi.org/10.1021/jf302571y | J. Agric. Food Chem. 2012, 60, 10170−10178

Journal of Agricultural and Food Chemistry
Article
(23) Kobayashi, J.; Horikoshi, T.; Ryu, J.-C.; Tashiro, F.; Ishii, K.;
Ueno, Y. The cytochrome P-450-dependent hydroxylation of T-2
toxin in various animal species. Food Chem. Toxicol. 1987, 25, 539−
544.
(24) Johnsen, H.; Odden, E.; Lie, O.; Johnsen, B. A.; Fonnum, F.
Metabolism of T-2 toxin by rat liver carboxylesterase. Biochem.
Pharmacol. 1986, 35, 1469−1473.
(25) Soars, M. G.; Ring, B. J.; Wrighton, S. A. The effect of
incubation conditions on the enzyme kinetics of UDP-glucuronosyl-
transferases. Drug Metab. Dispos. 2003, 31, 762−767.
(26) Martin, I. J.; Lewis, R. J.; Bernstein, M. A.; Beattie, I. G.; Martin,
C. A.; Riley, R. J.; Springthorpe, B. Which hydroxy? Evidence for
species differences in the regioselectivity of glucuronidation in rat, dog,
and human in vitro systems and dog in vivo. Drug Metab. Dispos. 2006,
34, 1502−1507.
(27) Ohta, M.; Ishii, K.; Ueno, Y. Metabolism of trichothecene
mycotoxins. 1. Microsomal deacetylation of T-2 toxin in animal tissues.
J. Biochem. 1977, 82, 1591−1598.
(28) Maier-Salamon, A.; Bohmdorfer, M.; Thalhammer, T.; Szekeres,
̈
T.; Jaeger, W. Hepatic glucuronidation of resveratrol: Interspecies
comparison of enzyme kinetic profiles in human, mouse, rat, and dog.
Drug Metab. Pharmacokinet. 2011, 26, 364−373.
(29) Pfeiffer, E.; Hildebrand, A.; Mikula, H.; Metzler, M.
Glucuronidation of zearalenone, zeranol and four metabolites in
vitro: Formation of glucuronides by various microsomes and human
UDP-glucuronosyltransferase isoforms. Mol. Nutr. Food Res. 2010, 54,
1468−1476.
(30) Taylor, J.; Grace, P.; Bingham, S. Optimization of conditions for
the enzymatic hydrolysis of phytoestrogen conjugates in urine and
plasma. Anal. Biochem. 2005, 341, 220−229.
(31) Ferchaud, V.; Courcoux, P.; Le Bizec, B.; Monteau, F.; Andre, F.
Enzymatic hydrolysis of conjugated steroid metabolites: Search for
optimum conditions using response surface methodology. Analyst
2000, 125, 2255−2259.
(32) Shibasaki, H.; Tanabe, C.; Furuta, T.; Kasuya, Y. Hydrolysis of
conjugated steroids by the combined use of beta-glucuronidase
preparations from Helix pomatia and Ampullaria: Determination of
urinary cortisol and its metabolites. Steroids 2001, 66, 795−801.
10178
dx.doi.org/10.1021/jf302571y | J. Agric. Food Chem. 2012, 60, 10170−10178