Pentahydroxyscirpene, a novel trichothecene-type compound, was isolated from Fusarium-inoculated rice. The structure of pentahydroxyscirpene was elucidated by 1D and 2D NMR spectroscopy and X-ray single-crystal diffraction. The conformation in solution was determined by NOESY experiments supported by quantum chemical calculations. In vitro toxicity tests showed that pentahydroxyscirpene inhibits protein synthesis as do other trichothecenes.
Pentahydroxyscirpene, a novel trichothecene-type compound, was isolated from Fusarium-inoculated rice. The structure of pentahydroxyscirpene was elucidated by 1D and 2D NMR spectroscopy and X-ray single-crystal diffraction. The conformation in solution was determined by NOESY experiments supported by quantum chemical calculations. In vitro toxicity tests showed that pentahydroxyscirpene inhibits protein synthesis as do other trichothecenes.
Trichothecenes,
a major class
of mycotoxins, are among the most prominent fungal secondary metabolites
harmful to human and animal health, causing a range of acute and chronic
symptoms.[1] Toxicity studies showed that
trichothecenes inhibit eukaryotic protein synthesis.[2] When consumed through contaminated foods, trichothecenes
are neurotoxic, immunosuppressive, and nephrotoxic.[3] Trichothecene-producing genera include Fusarium, Myrothecium, Stachybotrys, Cephalosporium, Trichoderma, and Trichothecium.[4] Within the genus Fusarium, some species are important plant pathogens, causing
head blight in small grain cereals and ear rot in maize.[5] Surveillance studies showed that worldwide contamination
of cereal grains and other feeds with Fusarium mycotoxins
is of global concern.[6]Over 200 trichothecenes
have been reported so far, all containing
a tricyclic 12,13-epoxytrichothec-9-ene core structure.[7] They are divided into four groups (types A–D)
according to their chemical properties.[8] Type A trichothecenes include compounds containing a hydroxyl group
(e.g., neosolaniol), an ester functionality (e.g., T-2 toxin), or
no oxygen substituent at C8 (e.g., 4,15-diacetoxyscirpenol). Type
B trichothecenes, such as nivalenol (1) or deoxynivalenol
(2), possess a carbonyl functionality at C8 (Figure 1). Type C trichothecenes have a C7/C8 epoxide, and
compounds of type D have an additional macrocycle between C4 and C15.
Figure 1
Structures
of important type A and B trichothecenes.
Structures
of important type A and B trichothecenes.Here we report the isolation and structural elucidation of
pentahydroxyscirpene
(3) (Figure 2) and results of
an in vitro toxicity test. The conformation in solution
is described as obtained by NOESY experiments supported by quantum
chemical calculations. Additionally, a fast and simple procedure for
the preparation of 3 starting from nivalenol is presented,
and the proposed biosynthetic pathway of the novel mycotoxin is discussed.
Figure 2
Structure
of pentahydroxyscirpene (3).
Structure
of pentahydroxyscirpene (3).During isolation of nivalenol from rice inoculated with the
nivalenol
producing strain IFA189 following a procedure described by Krska et
al.,[9] we discovered a novel trichothecene.
The compound was purified as described in the Supporting Information. 3 was obtained as a crystalline,
optically active compound {[α]25D = −36.1
(c 0.2; H2O)}. The molecular formula,
determined by HRMS measurement of the deprotonated ion (m/z 313.1290), was calculated to be C15H22O7 ([M – H]− calcd
313.1293), indicating 5 degrees of unsaturation. Due to four rings
and one C=C double bond present in type A trichothecenes, we
first considered 3 to be of classic type A structure
having no carbonyl group at C-8. The 1H NMR spectrum (see Supporting Information) of 3 clearly
displayed two methyl groups (δH 1.08, 1.81), of which
one is attached to an sp2carbon (allylic position), leading
to a higher chemical shift of this signal. An olefinic proton at δH 5.55 indicated the presence of a trisubstituted double bond.
Furthermore, 10 clearly separated signals between δH 2.91 and 4.42 were ascribable to protons geminal to an
oxygen atom (OH or OR). The 13C NMR spectrum of 3 contained 15 carbon resonances, which were further classified by
APT spectra as three quarternary carbons (including one olefinic carbon
at δC 140.1), one oxygenated quarternary carbon (δC 65.7), one olefinic (δC 123.9) and six oxygenated
methines, two oxygenated methylenes (δC 62.0, 46.8),
and two methyl groups (δC 20.8, 9.2). These data
were consistent with the signals observed in the 1H NMR
spectrum. All proton signals could be unambiguously assigned to their
respective carbons through analysis of the HSQC spectrum (see Supporting Information). The structure of 3 was established based on the detailed analysis of 1H–1H COSY, HMBC, and NOESY measurements (Figure 3).
Figure 3
(A) Selected 1H–1H COSY,
HMBC, and
(B) NOESY correlations.
(A) Selected 1H–1H COSY,
HMBC, and
(B) NOESY correlations.The configuration of 3 in solution was determined,
supported by quantum chemical calculations [DFT, B3LYP, 6-31++G(d,p),
methanol (IEFPCM), Gaussian 09; for further details see Supporting Information] (Figure 4A). X-ray single-crystal diffraction of 3 with
Mo Kα radiation confirmed the proposed structure (Figure 4B; for further details see the Supporting Information). Furthermore UV and CD measurements
were performed for comparison of 1, 2, and 3 (Figure 5).
Figure 4
(A) Optimized geometry
of 3 in MeOH (B3LYP, 6-311++G(d,p))
on the basis of NOESY correlations. (B) X-ray crystal structure of 3 (monohydrate; relative configuration).
Figure 5
(A) UV and (B) CD spectra of nivalenol (1), deoxynivalenol
(2), and pentahydroxyscirpene (3).
(A) Optimized geometry
of 3 in MeOH (B3LYP, 6-311++G(d,p))
on the basis of NOESY correlations. (B) X-ray crystal structure of 3 (monohydrate; relative configuration).(A) UV and (B) CD spectra of nivalenol (1), deoxynivalenol
(2), and pentahydroxyscirpene (3).To elucidate the origin of 3 as well as to ensure
that this compound is not an artifact or degradation product, the
whole fermentation process using the same Fusarium strain was repeated on rice. Both nivalenol (1) and
pentahydroxyscirpene (3) were detected in the cultures.3 does not contain a chromophoric moiety (Figure 5A) and may therefore be missed by routine analysis
using LC-UV.Since 3 contains an epoxide functionality,
which plays
an important part in the toxicity of trichothecences,[10] this natural compound was predicted to be toxic. In vitro toxicity tests, chosen based on the inhibition
of protein synthesis by trichothecenes, were performed with 3 as well as with 1 and 2. Translation
in rabbit reticulocyte lysates[11] was suppressed
to 50% (IC50) at 5.0 μM 3, 2.0 μM 1, and 1.8 μM 2 (Figure 6).
Figure 6
Inhibition of luciferase translation by 1, 2, and 3 (rabbit reticulocyte lysate).
Inhibition of luciferase translation by 1, 2, and 3 (rabbit reticulocyte lysate).Furthermore, a fast and simple procedure for the
synthesis of 3 starting from commercially available 1 was
developed (Scheme 1A). The prepared 3 was identical (NMR, [α]25D, HRMS/MS)
with the isolated material, proving the absolute configuration as
shown in Scheme 2A. The stereoselectivity of
this Luche reduction[12] can be explained
by the Cram chelation model[13] and/or substrate-controlled
hydride addition (Scheme 1B).
Scheme 1
Stereoselective
Luche Reduction of 1 to 3
Scheme 2
Proposed Biosynthetic
Pathway for 3 in Context to Previously
Described Routes for Type A and B Trichothecenes[14]
Including already
identified
genes encoding an enzymatic step; FPP = farnesyl pyrophosphate, sp.
= spontaneous.
A biogenetic pathway of 3 is proposed
based on previously
described routes to type A and B trichothecences (Scheme 2).[14] The biosynthetic
precursor of 1 and 3 is most likely 7,8-dihydroxy-3,4,15-triacetoxyscirpenol
(<span class="Chemical">7,8-dihydroxy-TAS). An unknown oxidoreductase plays an important
role in the production of 1 and 3. This
enzyme is predicted to catalyze the oxidation of the C8-OH, to form
the acyloin moiety that is typically present in type B trichothecenes
produced by various Fusarium species. In case the
dehydrogenation of 7,8-dihydroxy-TAS at the C8-OH is limiting, deacetylation
leads to the formation of 3.
Proposed Biosynthetic
Pathway for 3 in Context to Previously
Described Routes for Type A and B Trichothecenes[14]
Including already
identified
genes encoding an enzymatic step; FPP = farnesyl pyrophosphate, sp.
= spontaneous.With pentahydroxyscirpene (3) available, it should
become possible to purify the enzyme performing the conversion of
the C8-OH into a keto group leading to the biosynthesis of mycotoxins
of the important class type B trichothecenes.
Experimental
Section
General Experimental Procedures
Thin-layer chromatography
(TLC) was performed over silica gel 60 F254 (Merck). The chromatograms
were visualized by heat staining with ceric ammonium molybdate in
ethanol–sulfuric acid. LC-ESIMS/MS was performed on an HCT
ion trap mass spectrometer (Bruker, Germany) in full scan mode. Chromatographic
separation was done on a 1200 series HPLC system (Agilent Technologies,
Germany) using a Luna RP-C18 column (3.0 × 150 mm,
3 μm particle size, Phenomenex, Germany), and application of
pure substances was achieved using a TLC-MS interface (Camag, Germany).
Preparative column chromatography was performed on silica gel 60 (40–63
μm, Merck, Germany) using a SepacoreTM Flash System (Büchi,
Switzerland). HRMS and HRMS/MS spectra were obtained on a 6550 iFunnel
QTOF (Agilent Technologies) using negative electrospray ionization
(ESI). The 1290 Infinity UHPLC system equipped with a Zorbax SB C18 rapid-resolution high-definition column (150 × 2.1
mm; 1.8 μm particle size) (Agilent Technologies) was used with
a methanol–water gradient containing 0.1% formic acid. 1H and 13C NMR spectra were recorded on an Avance
DRX-400 MHz spectrometer (Bruker, Germany). Data were recorded and
evaluated using TOPSPIN 1.3 (Bruker Biospin). All chemical shifts
are given in ppm relative to tetramethylsilane. The calibration was
done using residual solvent signals. UV spectra were recorded by using
a Shimadzu UV-1800 UV spectrophotometer. Each sample was first measured
within the range 190–800 nm. Since no absorption was observed
in the range 400–800 nm, all samples were measured twice in
the range 190–400 nm. CD spectra were recorded with a JASCO
J-815 CD spectrophotometer with a N2 monitor. The nitrogen
flow was maintained at 3 L/min. After background measurement triplicates
of each sample were recorded and mean values were used for visualization.
Nivalenol (1) and deoxynivalenol (2) were
obtained from Romer Laboratories (Tulln), and all other chemicals
were purchased from Sigma-Aldrich (Austria/Germany).
Fermentation
and Purification
Following a previously
described procedure,[9] the Fusarium strain IFA189 was cultivated in a liquid synthetic nutrient-poor
medium, based on SNA (synthetic nutrient-poor agar), but without the
addition of agar, for 48 h after inoculation with mycelium. Rice was
inoculated with the spores and mycelium obtained suspension and incubated
in the dark at 28 °C for 28 days. At the end of the incubation
period, 1 kg of culture material was freeze-dried over a period of
72 h and the residue obtained was extracted with 2 L of methanol.
After soaking overnight, the solution was filtered and the extraction
solvent was evaporated. The remaining oily residue was dissolved in
a minimum amount of methylene chloride for preparative silica gel
chromatography (150 g of silica gel, CHCl3–MeOH,
90:10). After application of the diluted raw extract, the substances
were eluted with the same solvent mixture as described above. The
eluate was collected automatically with a fraction collector, and
all fractions (20 mL) were analyzed by TLC. Appropriate fractions
were pooled and concentrated on a rotary evaporator to obtain a crude
mixture of 1 and 3. A 120 mg amount of this
fraction of 1 and 3 was dissolved in approximately
10 mL of CH2Cl2–MeOH (2:1). After addition
of 0.75 g of silica gel and a short shaking period the solvent was
removed and the remaining material was transferred to an 8 g chromatography
column, which was subsequently filled up with silica gel and connected
to another 8 g silica gel column. The toxins were eluted (CH2Cl2–MeOH, 90:10, 10 mL/min, 5 mL fraction size),
and appropriate fractions were identified by heat staining after TLC
analysis (CH2Cl2–MeOH = 90/10, R(3) = 0.17, R(1) = 0.25).
Inhibition of Protein Synthesis in Rabbit Reticulocyte Lysates
The effect of 1, 2, and 3 on mammalian ribosomes was tested in a rabbit reticulocyte lysate-coupled in vitro transcription/translation system (TnT T3 coupled
reticulocyte lysate system, Promega) using the newly translated luciferase
as a reporter for translational activity. The transcription/translation
reactions were performed in a total volume of 15 μL for 20 min
according to the manufacturer’s instructions in the presence
of either 1, 2, or 3, respectively
or water as a control. The translation reactions were stopped by the
addition of 1 μL of 1 mM cycloheximide. Efficiency of translation
was determined by measuring activity of the firefly luciferase reporter
(Steady-Glo luciferase assay system, Promega) using the 2300 EnSpireMultimode
plate reader (PerkinElmer). Two independent experiments were performed
using the separately produced dilutions of the mycotoxins in water.
X-ray Single-Crystal Diffraction (CCDC 933314)
A crystal
of the monohydrate of 3 was embedded in perfluorinated
oil and attached to a Kapton mount. Intensity data were collected
on a Bruker KAPPA APEX II diffractometer equipped with a CCD detector
in ω- and φ-scan modes with 1° frame width using
MoKα radiation. The full reciprocal sphere
up to 2θ = 60° was collected. The frames were reduced to
intensity values using SAINT-Plus, and absorption correction was applied
using the multiscan approach implemented in SADABS.[15] The structure was solved using charge flipping implemented
in SUPERFLIP[16] and refined against F with JANA2006.[17] The H atoms
were located in the difference Fourier maps. C–H atoms were
refined as riding on the parent C atoms, and the O–H distances
were restrained to 0.820(1) Å. The absolute configuration was
deduced by synthesis from a precursor with known chiral centers. The
monohydrate of the title compound crystallizes in space group P212121 with one crystallographically
unique molecule of 3 and water, respectively. As expected,
the molecular conformation and the packing are defined by strong O–H···O
hydrogen bonds (Table S1, Supporting Information): The molecular configuration is stabilized by one intramolecular
hydrogen bond. Each molecule connects to three other molecules. The
water molecule donates two and the acceptor one hydrogen bond. Thus
an intricate three-dimensional network is formed.
Preparation
of 3 by Stereoselective Luche Reduction
of 1
Nivalenol (10.0 mg, 0.032 mmol) and CeCl3·7H2O (23.9 mg, 0.064 mmol) were dissolved
in dry MeOH (3 mL). NaBH4 (2.4 mg, 0.064 mmol) was added,
and the reaction mixture was stirred for 60 min at room temperature.
The solvent was evaporated, and the remaining residue was dissolved
in 2 mL of H2O–MeOH (95:5). Purification using RP-HPLC
(H2O–MeOH gradient elution) afforded 7.2 mg (72%)
of 3 as a white solid. The obtained compound was identical
(NMR, [α]25D, HRMS/MS) with the isolated
material.
Authors: R Krska; R C Schothorst; H P van Egmond; R D Josephs; J Lepschy; H Pettersson; D Chan; F Berthiller; R Schuhmacher; W Kandler; A Parich; E Welzig Journal: Anal Bioanal Chem Date: 2005-07-09 Impact factor: 4.142
Authors: Elisabeth Varga; Gerlinde Wiesenberger; Philipp Fruhmann; Alexandra Malachová; Thomas Svoboda; Marc Lemmens; Gerhard Adam; Franz Berthiller Journal: Toxins (Basel) Date: 2016-10-14 Impact factor: 4.546
Authors: Philipp Fruhmann; Christian Hametner; Hannes Mikula; Gerhard Adam; Rudolf Krska; Johannes Fröhlich Journal: Toxins (Basel) Date: 2014-01-10 Impact factor: 4.546
Authors: Matias Pasquali; Marco Beyer; Antonio Logrieco; Kris Audenaert; Virgilio Balmas; Ryan Basler; Anne-Laure Boutigny; Jana Chrpová; Elżbieta Czembor; Tatiana Gagkaeva; María T González-Jaén; Ingerd S Hofgaard; Nagehan D Köycü; Lucien Hoffmann; Jelena Lević; Patricia Marin; Thomas Miedaner; Quirico Migheli; Antonio Moretti; Marina E H Müller; Françoise Munaut; Päivi Parikka; Marine Pallez-Barthel; Jonathan Piec; Jonathan Scauflaire; Barbara Scherm; Slavica Stanković; Ulf Thrane; Silvio Uhlig; Adriaan Vanheule; Tapani Yli-Mattila; Susanne Vogelgsang Journal: Front Microbiol Date: 2016-04-06 Impact factor: 5.640