Isotopic Labeling Studies Reveal the Patulin Detoxification Pathway by the Biocontrol Yeast Rhodotorula kratochvilovae LS11

Article abstract of DOI:10.1021/acs.jnatprod.8b00539

Patulin (1) is a mycotoxin contaminant in fruit and vegetable products worldwide. Biocontrol agents, such as the yeast Rhodotorula kratochvilovae strain LS11, can reduce patulin (1) contamination in food. R. kratochvilovae LS11 converts patulin (1) into desoxypatulinic acid (DPA) (5), which is less cytotoxic than the mycotoxin (1) to in vitro human lymphocytes. In the present study, we report our investigations into the pathway of degradation of patulin (1) to DPA (5) by R. kratochvilovae. Isotopic labeling experiments revealed that 5 derives from patulin (1) through the hydrolysis of the ?-lactone ring and subsequent enzymatic modifications. The ability of patulin (1) and DPA (5) to cause genetic damage was also investigated by the cytokinesis-block micronucleus cytome assay on in vitro human lymphocytes. Patulin (1) was demonstrated to cause much higher chromosomal damage than DPA (5).

Full text of DOI:10.1021/acs.jnatprod.8b00539

Article
pubs.acs.org/jnp
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
Isotopic Labeling Studies Reveal the Patulin Detoxification Pathway
by the Biocontrol Yeast Rhodotorula kratochvilovae LS11
Cristina Pinedo,† Sandra A. I. Wright,‡ Isidro G. Collado,† Rebecca J. M. Goss,§ Raffaello Castoria,⊥
∥
,#
,†
́
́
Patrizia Hrelia, Francesca Maffei,* and Rosa Duran-Patron*
†
́
́
Departamento de Química Organica, Facultad de Ciencias, Universidad de Cadiz, Campus Universitario Río San Pedro s/n, Torre
sur, 4a planta, 11510, Puerto Real, Cadiz, Spain
́
‡
̈
Section of Biology, Faculties of Health and Occupational Studies & Engineering and Sustainable Development, University of Gavle,
̈
801 76 Gavle, Sweden
§School of Chemistry, Biomedical Sciences Research Complex, University of St Andrews, Fife, Scotland KY169ST, United Kingdom
⊥
̀
Dipartimento Agricoltura, Ambiente, Alimenti, Universita degli Studi del Molise, Via F. De Sanctis snc, 86100 Campobasso, Italy
∥
̀
Dipartimento di Farmacia e Biotecnologie, Alma Mater Studiorum-Universita di Bologna, Via Irnerio, 48, 40126 Bologna, Italy
#
̀
̀
Dipartimento di Scienze per la Qualita della Vita, Alma Mater Studiorum-Universita di Bologna, Campus Rimini, Corso D’Augusto
237, 47921 Rimini, Italy
S
* Supporting Information
ABSTRACT: Patulin (1) is a mycotoxin contaminant in fruit
and vegetable products worldwide. Biocontrol agents, such as
the yeast Rhodotorula kratochvilovae strain LS11, can reduce
patulin (1) contamination in food. R. kratochvilovae LS11
converts patulin (1) into desoxypatulinic acid (DPA) (5),
which is less cytotoxic than the mycotoxin (1) to in vitro
human lymphocytes. In the present study, we report our
investigations into the pathway of degradation of patulin (1)
to DPA (5) by R. kratochvilovae. Isotopic labeling experiments
revealed that 5 derives from patulin (1) through the
hydrolysis of the γ-lactone ring and subsequent enzymatic
modifications. The ability of patulin (1) and DPA (5) to cause
genetic damage was also investigated by the cytokinesis-block
micronucleus cytome assay on in vitro human lymphocytes. Patulin (1) was demonstrated to cause much higher chromosomal
damage than DPA (5).
atulin (1) is a well-known mycotoxin, biosynthesized by
contamination of foodstuffs, including the use of chemical
Pspecies belonging to the genera Penicillium, Aspergillus, and physical methods.3,21−24 In recent years, biocontrol agents
have emerged as a promising method to prevent infection of
stored fruits by P. expansum and, as a consequence, eliminate
patulin (1) contamination in fruit derivatives.25,26
Byssochlamys (anamorph Paecilomyces), Gymnoascus, Paecilo-
myces, and Scopulariopsis.1,2 It commonly occurs in fruits and in
some other foods that have been attacked by these fungi.
Patulin’s (1) most noteworthy producer is P. expansum Link.
(Trichocomaceae family), which causes blue mold of apples
during postharvest storage.3
Patulin (1) is toxic to humans and animals, as well as to
bacteria, yeasts, and plants.3−18 In mammals, exposure to 1 can
induce genotoxicity, embryotoxicity, immunotoxicity, and
neurotoxicity.3,7−18 Thus, human exposure to low levels of 1
is considered a major concern for public health. The Joint
FAO/WHO Expert Committee on Food Additives has
recommended a maximum tolerable daily intake of 0.4 μg/kg
body weight/day.19 In 2006 the European Commission set a
maximum permitted level of 50 μg/L for fruit juice and 10 μg/
kg in products for children.20
Various microbial species, including biocontrol agents,
exhibited the ability to degrade patulin (1). Saccharomyces
cerevisiae degraded 1 to (E)- and (Z)-ascladiol (2, 3) during
alcoholic fermentation of apple juice for cider production.27
More recently, it was reported that the bacteria Gluconobacter
oxydans28 and Lactobacillus plantarum,29 the marine yeast
Kodameae ohmeri,30 and the terrestrial yeasts Candida
guillermondi31 and Pichia caribbica32 also transformed 1 into
2 and 3. An additional metabolite, hydroascladiol (4), was
formed in a cell-free supernatant of a culture of L. plantarum
supplemented with 1.29 (Z)-Ascladiol (3) has been reported to
be the product of nonenzymatic isomerization of (E)-ascladiol
Several postharvest control strategies have been imple-
mented to prevent, reduce, or eliminate patulin (1)
Received: July 3, 2018
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.8b00539
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Chart 1
(2) catalyzed by sulfhydryl compounds, such as homocysteine,
cysteine, and glutathione or dithiothreitol.33
resting cell cultures of P. expansum FS-7. A time-course study
was carried out to determine the optimal time for feeding the
isotopically labeled acetate to enable maximum production of
highly labeled metabolite. As a result of this time-course study
for the production of patulin (1), four-day-old resting cell
cultures of P. expansum were fed with the above-mentioned
13C-labeled sodium acetates. Extraction and purification of the
fermentation broths led to the isolation of two 13C-labeled
In previous studies, we observed that the biocontrol yeast
Rhodotorula kratochvilovae LS11 (basionym Rhodosporidium
kratochvilovae, Sporidiobolaceae family) metabolized 1, gen-
erating a new degradation product, desoxypatulinic acid (DPA,
5).34 Studies carried out with another basidiomycete yeast
allowed us to demonstrate that Sporobolomyces sp. IAM 13481
was also able to convert 1 to 5, in addition to 2 and 3.35 In a
further report, a phytogenetically related yeast species,
Rhodosporidium paludigenum (syn. Rhodotorula paludigena),
was demonstrated to produce a biodegradation product of 1,
which was putatively identified as DPA (5).36
Cytotoxicities of patulin degradation products have been
investigated. (E)- and (Z)-ascladiol (2, 3), unlike 1, were
devoid of cytotoxic effects on human cell lines of liver, kidney,
intestine, and immune system.37 In addition, 2 did not produce
cytotoxicity in an ex vivo porcine intestinal model tissue.38 On
the other hand, DPA (5) was also much less cytotoxic than 1
to in vitro human lymphocytes34 and hepatic cell lines.36 These
conversions by microbial species can therefore be considered
detoxification processes of the mycotoxin (1).
analogues of patulin (1), which were analyzed by means of 13
C
NMR spectroscopy.
When P. expansum was fed with sodium [1-13C]-acetate,
significant enrichments with 13C were observed at C-2, C-3a,
and C-7 of patulin (1), while the feeding experiment with
sodium [1,2-13C2]-acetate led to enhanced signals for C-2, C-3,
and C-6 and enhanced and spin−spin coupled signals between
C-3a and C-4 and between C-7 and C-7a (Figure 1, Table 1),
Previous isotopic labeling experiments with 13C-labeled
patulin (1) have allowed us to demonstrate that DPA (5) is
derived from the metabolism of 1 by R. kratochvilovae strain
LS11.34 However, the biodegradation mechanism by which
this conversion occurs is still unknown. In order to elucidate
this detoxification pathway, we followed herein the fate of
several isotopic labels during the biodegradation process of 1
into 5 by the strain LS11. For this purpose, we prepared two
13C-labeled analogues of 1 through administering sodium
[1-13C]- and [1,2-13C2]-acetates, separately, to resting cell
cultures of P. expansum FS-7. We incubated the strain LS11
with these isotopically labeled analogues of 1 or with unlabeled
1 in the presence of either H2O, 2H2O, or H218O to follow the
degradation of the mycotoxin (1) by the yeast into DPA (5).
Furthermore, to gain a better understanding of the toxicity of
both patulin (1) and DPA (5), we evaluated and compared the
chromosomal damage and the cytotoxic effects induced by the
two compounds in cultured human lymphocytes by using the
cytokinesis-block micronucleus cytome (CBMN cyt) assay.39
Figure 1. Labeling pattern resulting in patulin (1) after feeding
Penicillium expansum FS-7 with sodium [1-13C]- and [1,2-13C2]-
acetates.
indicating that these positions resulted from the intact
incorporation of two acetates. These findings are in agreement
with the 13C incorporation detected in patulin (1) by HREIMS
after administration of sodium [1-13C]-acetate to P. expansum
that we previously reported, where up to three 13C labels were
incorporated in 0.3% of the molecules.34
The 13C-labeled analogues of patulin (1) obtained in the
above-mentioned experiments were subsequently administered
to R. kratochvilovae LS11 cultures in order to follow the fate of
the isotopic label during the metabolism of the compounds
into DPA (5). The resulting labeled DPAs (5) were extracted
after 3 days at acidic pH, purified, and analyzed by 13C NMR
spectroscopy. Feeding experiment with the 13C-labeled
RESULTS AND DISCUSSION
■
Patulin (1) was labeled with 13C by pulse feeding sodium
[1-13C]- and [1,2-13C2]-acetates, separately, to two series of
B
DOI: 10.1021/acs.jnatprod.8b00539
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 1. Isotopic Enrichment of Carbons and Coupling
Constants in Patulin (1) after Feeding Penicillium expansum
FS-7 with Sodium [1-13C]- or [1,2-13C2]-Acetates
Table 2. Isotopic Enrichment of Carbons and Coupling
Constants in DPA (5) after Feeding Rhodotorula
kratochvilovae LS11 with Molecules of Patulin (1) Deriving
from Cultures of Penicillium expansum FS-7 Fed with
Sodium [1-13C]- or [1,2-13C2]-Acetate
a
atom % 13C excess
J (Hz)
carbon atom (N) δC (CD3OD)
[1-13C]-acetate
[1,2-13C2]-acetate
a
atom % 13C excess
J (Hz)
2
3
3a
4
170.8
111.0
153.5
89.9
3.23
−0.22
3.36
enriched
enriched
56.8
carbon atom (N) δC (CD3OD)
[1-13C]-acetate
[1,2-13C2]-acetate
2
3
4
5
6
7
8
69.5
36.9
193.8
114.4
164.4
32.1
0
enriched
40.0
40.0
71.4
71.4
0.22
56.8
1.51
−0.61
5.12
−0.68
−0.49
4.46
6
7
60.3
109.6
0.00
3.36
enriched
84.9
7a
147.9
−0.07
84.9
a
Atom % 13C excess = {(RN/RN(UL) × 1.1} − 1.1, where RN is the
enriched
enriched
176.2
ratio of the peak intensity at N-position in labeled compound
calculated on the basis of the peak intensity at the C-6-position.
Similarly, RN(UL) is the ratio of the peak intensity at N-position in
unlabeled compound. Value 1.1 is the theoretical 13C natural
abundance (atom %).
a
Atom % 13C excess = {(RN/RN(UL) × 1.1} − 1.1, where RN is the
ratio of the peak intensity at N-position in labeled compound
calculated on the basis of the peak intensity at the C-2-position.
Similarly, RN(UL) is the ratio of the peak intensity at N-position in
unlabeled compound. Value 1.1 is the theoretical 13C natural
abundance (atom %).
analogue of patulin (1) from sodium [1-13C]-acetate led to
DPA (5) with enhanced signals at carbons C-3, C-5, and C-8.
When strain LS11 was fed with labeled patulin (1) deriving
from sodium [1,2-13C2]-acetate, the 13C NMR spectrum of
DPA (5) showed the incorporation of two intact C2 units for
the carbon pairs C-3/C-4 and C-5/C-6. In addition, 13C
enrichments at positions C-2, C-7, and C-8 were observed
(Figure 2, Table 2). This pattern of observed couplings and
from labeled patulin (1) derived from [1-13C]-acetate feeding,
and 18O from H218O present in the culture medium. The 13C
NMR spectrum of the resultant DPA (5) showed an
isotopically shifted resonance 0.024 ppm for C-8, which is in
the expected order of magnitude for 13C attached by a simple
bond to 18O (Figure 3).41 This result clearly confirms the
hydrolysis of the patulin (1) lactone ring by the water within
the medium, leading to enol 6 outlined in Scheme 1, which is
then rapidly isomerized into the more stable keto form 7.
The hydrolysis of patulin (1) could be catalyzed under
slightly acidic or basic conditions. In order to investigate the
role of LS11 in the hydrolysis of patulin (1), the pH of the
yeast culture medium was measured at the beginning (pH 4.5)
and at the end (pH 7.4) of the biodegradation experiment.
Since at the beginning of the biodegradation process the
culture medium had an acidic pH, patulin (1) was dissolved in
water at pH 2 and stirred for 1 week at room temperature.
After this time period, there was no evidence of any
transformation product as detected by TLC analysis. In
addition, patulin (1) was dissolved in water at pH 8 and also
stirred for 1 week at room temperature, without degradation
products being observed by TLC. These results, which are in
agreement with those previously published on the stability of
patulin (1) at acidic pH,42,43 suggest that the hydroxy group
bound to C-8 in DPA (5) derives from medium water by the
action of a hitherto unidentified hydrolase (Scheme 1).
The subsequent experimental step was aimed at confirming
the reduction and subsequent dehydration of the hydrolysis
product (7) of patulin (1) to yield DPA (5). For this purpose,
an experiment of patulin (1) biodegradation by R.
kratochvilovae LS11 was carried out in a culture medium
containing 25% (v/v) of 2H2O. The 2H NMR spectrum of the
resultant DPA (5) showed only one signal at δ 3.00 ppm,
corresponding to D-7, as a consequence of the reduction of the
double bond C-5/C-7 by the conjugate addition of a hydride
to the α,β-unsaturated acid 7 and subsequent dehydration of
the resultant intermediate 8, as shown in Scheme 1. Recent
studies have shown that 1 produces a marked increase,
compared to the control, of the short-chain dehydrogenase−
reductase protein, as well as the level of expression of the
Figure 2. Labeling pattern resulting in DPA (5) after feeding
Rhodotorula kratochvilovae LS11 with molecules of patulin (1)
deriving from cultures of Penicillium expansum FS-7 fed with sodium
[1-13C]- or [1,2-13C2]-acetate.
enrichments suggests that DPA (5) derives from the
metabolization of patulin (1) by the yeast through the
biodegradation pathway shown in Scheme 1.
In order to explore the hydrolysis of the patulin (1) lactone
ring, an experiment was conducted in which the 13C-labeled
mycotoxin (1) from sodium [1-13C]-acetate was biodegraded
by R. kratochvilovae LS11 in a culture medium containing 25%
(v/v) of H218O. 18O does not possess a magnetic moment and
therefore is not detectable by NMR. However, it can be
indirectly detected by 13C NMR spectroscopy through the
isotope-induced shift in the carbon atom directly attached to
18O.40 The low levels of incorporation of isotopically labeled
water achievable in biosynthetic studies can be difficult to
observe. To enhance this, it is standard practice to carry out
the labeling study on material that is 13C enriched in the
position to be oxygenated in order to make the indirect
detection of 18O more sensitive. In this experiment, labeling of
DPA (5) was achieved by simultaneously incorporating 13C
C
DOI: 10.1021/acs.jnatprod.8b00539
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
a
Scheme 1. Degradation Pathway of Patulin (1) to DPA (5) by Rhodotorula kratochvilovae LS11
a
Labeling pattern resulting from experiments of LS11 incubation with H218O, H2O, and 13C-labeled patulin (1) deriving from sodium [1-13C]-
2
acetate.
the resultant intermediate 7 by the action of an unidentified
dehydratase lead to DPA (5).
Previously, we have reported that DPA (5) was much less
cytotoxic than patulin (1) to in vitro human lymphocytes,
probably as a consequence of the loss of the hemiacetal group
and the hydrolysis of the lactone ring.34 Patulin (1), unlike
DPA (5), reacts with the thiol group present in the antioxidant
tripeptide glutathione (GSH) and in enzymes, thus blocking
their biological activity.34,45 Since patulin (1) is toxic to yeasts,
they need to detoxify it in order to thrive. It has been observed
that in Saccharomyces cerevisiae 12786 and Sporobolomyces sp.
IAM 13481, the patulin (1) biodegradation process occurred
in two steps, where the first involved an inhibition of growth
and conditioning of cells to the toxic environment, and the
second involved resumed microbial growth and successive
biotransformation of patulin (1).46,47 As observed with the
above-mentioned yeast, R. kratochvilovae needs to detoxify
patulin (1). Therefore, it can be stated that the pathway
depicted in Scheme 1 shows the process of detoxification of
patulin (1) to DPA (5) by R. kratochvilovae.
In order to have a better understanding of the toxic effects of
patulin (1) and DPA (5), the CBMN cyt assay was performed
on in vitro human lymphocytes. This assay is a comprehensive
system for measuring DNA damage, cytostasis, and cytotox-
icity. DNA damage events were scored specifically in once-
divided binucleated (BN) cells and include (a) MNi, a
biomarker of chromosome breakage and/or whole chromo-
some loss, (b) NPBs, a biomarker of DNA misrepair and/or
telomere end-fusions, and (c) nuclear buds (NBUDs), a
biomarker of the elimination of amplified DNA and/or DNA
repair complexes. Cytostatic effects were measured via the
proportion of mono-, bi-, tri-, and tetranucleated cells, and
cytotoxicity effects via necrotic and/or apoptotic cell ratios.39
Patulin (1) and DPA (5) were obtained as described in the
Experimental Section and dissolved in sterile double-distilled
water for treatment of human lymphocyte cultures. Colcemid
(COL) and ethyl methanesulfonate (EMS) served as the
standard positive controls. All lymphocyte cultures were
treated for 24 h with these compounds.
Figure 3. Region of the 13C NMR spectrum of DPA (5) showing the
18O isotope-induced shift in the signal for C-8 after incubation of
Rhodotorula kratochvilovae LS11 in a culture medium containing
H218O (25% v/v) and 13C-labeled patulin (1), deriving from sodium
[1-13C]-acetate.
corresponding gene in R. mucilaginosa.44 This finding
reinforces the reduction step proposed in the detoxification
pathway of patulin (1).
Taken together, all these results are consistent with the
patulin (1) biodegradation pathway proposed in Scheme 1,
according to which the α,β,γ,δ-unsaturated γ-lactone ring of
the mycotoxin 1 is hydrolyzed by the action of a hitherto
unidentified hydrolase to afford the intermediate 6, which is
rapidly isomerized to 7, probably without the need for any
enzyme. Subsequent enzymatic reduction of the double bond
C-5/C-7, through Michael-type reaction, and dehydration of
As shown in Table 3, patulin (1) at the highest
concentration tested (10 μM) induced a high percentage of
necrotic cells (more than 20%), while at the dose of 5 μM
significantly decreased the nuclear division index (NDI)
D
DOI: 10.1021/acs.jnatprod.8b00539
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 3. Mean ( SD) Frequencies of MNi, NPBs, and BUDs in BN Cells and NDI in Human Lymphocytes Treated in Vitro
with Patulin (1) or DPA (5)
treatment (μM)
MNi/1000 BN cells
5.3 1.00
NPBs/1000 BN cells
BUDs/1000 BN cells
1.0 1.00
NDI
patulin (1)
0.1
0.5
1.0
5.0
10.0
0.1
0.5
1.0
5.0
10.0
1.3 0.58
1.3 0.58
1.7 0.58
toxic
1.7 0.03
1.7 0.04
1.6 0.03
toxic
a b
,
a
12.3 2.08
14.7 2.08
toxic
4.7 1.53
6.7 2.08
toxic
a b
,
ab
,
toxic
toxic
toxic
toxic
DPA (5)
5.0 1.00
6.3 1.53
7.3 1.15
1.0 1.00
1.0 1.00
1.0 1.00
1.3 0.58
1.7 0.58
1.3 0.58
8.7 1.52
1.7 0.58
1.3 0.58
2.0 1.00
2.3 1.15
4.7 1.52
6.3 1.52
1.3 0.58
7.3 2.52
2.3 1.15
1.8 0.03
1.7 0.03
1.6 0.03
1.6 0.03
1.6 0.04
1.8 0.08
1.6 0.04
1.7 0.03
a
a
a
9.0 1.00
a
9.7 1.15
solvent control (double-distilled water)
EMS (240 μg/mL)
COL (0.02 μg/mL)
5.3 1.55
13.3 1.53
20.3 2.08
a
a
a
b
Statistically significant as compared to the solvent control (P < 0.05). Statistically significant as compared to DPA treatment (P < 0.05).
(−35%) with respect to the solvent control (sterile double-
distilled water). These cytotoxic effects prevented a sufficient
recovery of BN cells to allow chromosomal damage analysis.
Patulin (1) also induced a significant increase of MNi
frequency; the maximum increase observed at 1 μM was
about 3-fold higher than the corresponding control value. The
increase of MNi was concentration dependent. No significant
increase in the frequency of NPBs with respect to the solvent
control was recorded in the range of 0.10−1 μM concen-
trations of patulin (1), whereas the frequencies of BUDs were
significantly higher at doses of 0.5 and 1.0 μM. These results
were consistent with previous observations made in both
human and mammalian cells9−12 and confirm the ability of
patulin (1) to cause chromosomal damage and to interfere
with DNA. As expected, we recorded a significant increase in
MNi frequency with respect to the solvent control in the
treatment with COL. Moreover, the treatment with EMS
induced a significant increase in the frequencies of all three
DNA damage events as compared to those observed in the
solvent control.
Interestingly, DPA (5) did not induce any cytotoxic effect in
the concentration range of 0.1−10 μM; thus analysis of the
three DNA damage events could be performed in the
lymphocyte cultures. The treatment with DPA (5) exhibited
significant differences as compared to the treatment with
patulin (1) for MNi and BUDs. No elevated level of either
MNi and BUDs was observed in lymphocytes treated with
0.5−1.0 μM 5, and the frequencies recorded were significantly
lower than those induced by patulin (1) in the same
concentration range. Only the highest concentrations of 5 (5
and 10 μM) induced significant differences as compared to the
solvent control for MNi and BUDs. On the other hand,
treatment with DPA (5) at any concentration did not increase
the frequency of NPBs.
nucleophilic centers of DNA bases, both in the absence and in
the presence of GSH, to yield several mono- and disubstituted
DNA base adducts as well as mixed patulin−GSH−nucleobase
adducts. The initial step for obtaining these diadducts is a
Michael-like addition of the thiol group of GSH or the amine
group of the DNA bases to the δ position of the unsaturated
lactone moiety of patulin (1) to yield a reactive enol-lactone,
which can add nucleophiles of DNA bases to the carbonyl
carbon of the mycotoxin (1). Formation of patulin−
nucleobase diadducts could contribute to the increase in
MNi and BUD observed in cells treated with patulin (1). On
the other hand, formation of patulin−GSH and patulin−
GSH−nucleobase di- or triadducts could cause indirect DNA
damage by depletion of the intracellular pool of GSH and
antioxidant enzymes (superoxide dismutase and catalase)51
and subsequent increase of ROS16,52 and of lipid peroxidation
products.53
At a molecular level, DPA (5) is a carboxylic acid whose
structure contains an α,β-unsaturated ketone, and the addition
of nucleophiles to C-6 of 5 through a Michael-type addition
reaction could be expected. However, it was shown that DPA
(5) does not react with GSH.34 In addition, it has also been
reported that the adducts resulting from the reactions of
patulin (1) with 4-bromothiophenol, N-acetylcysteine, or
GSH, whose structures are similar to 5, did not undergo
further addition of sulfur-containing nucleophiles.45,54 The lack
of reactivity of DPA (5) could be due to the mesomeric effect
+ M produced by ring oxygen on the C-6 of 5, which renders
this carbon, in contrast to C-7 of patulin (1), insufficiently
electrophilic for Michael-type additions by nucleobases or
cellular thiols. The electrophilic structure of DPA (5) could
therefore explain the lower genotoxicity of this patulin (1)
degradation and detoxification product.
In conclusion, the isotopic labeling experiments reported
here have enabled us to propose a detoxification pathway of
patulin (1) to DPA (5) by the yeast R. kratochvilovae LS11,
which involves the hydrolysis of the γ-lactone ring of 1 and
subsequent reduction of the double bond C-5/C-7 and
dehydration of the resultant intermediate 7. It has also been
shown that DPA (5) causes much less chromosomal damage
than patulin (1) on human lymphocytes in vitro, possibly due,
as indicated above, to its lack of reactivity with cellular
nucleophilic amines and thiols as a consequence of the
deactivation of its carbon C-6 by the oxygen ring.
The α,β-unsaturated carbonyl group is commonly recog-
nized as an alert for mutagenicity due to its reactivity with
cellular nucleophilic amines and thiols.48 Patulin (1), a α,β,γ,δ-
unsaturated γ-lactone, could display its cytotoxic and
chromosome-damaging effects by forming covalent adducts
with these types of compounds, since small lactones are known
to act as alkylating agents.49 It has been suggested that 1 can
interact irreversibility with nucleophilic groups of DNA,
leading to chromosomal damage and cell cycle disturbance.14
Recently, Pfenning et al.50 reported that patulin (1) reacts with
E
DOI: 10.1021/acs.jnatprod.8b00539
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
medium) supplemented with 150 ppm of patulin (1). The yeast was
incubated at 25 °C and 150 rpm. The biodegradation was monitored
on a daily basis by TLC, using toluene−ethyl acetate−formic acid
5:4:1 (v/v/v) as the eluent system. After complete biodegradation of
patulin (1), the culture medium and the yeast cells were separated by
centrifugation for 20 min at 6000 rpm. The broth was acidified to pH
2 with an aqueous solution of 2 M HCl, saturated with NaCl, and
extracted with ethyl acetate (×3). The EtOAc extract was washed
with H2O (×3) and dried over anhydrous Na2SO4, and the solvent
evaporated under reduced pressure. The residue was purified by
chromatography on a Si gel column, using toluene−ethyl acetate−
formic acid, 5:4:1 (v/v/v), as the solvent system. Fractions containing
DPA (5) were detected by TLC, collected, and analyzed by 1H NMR.
The spectrum achieved was identical to that shown by a real sample.
Biodegradation of Labeled Patulin (1) Deriving from
Sodium [1-13C]-Acetate. R. kratochvilovae LS11 was grown for 4
days in the presence of labeled patulin (1) from the feeding of P.
expansum with sodium [1-13C]-acetate in accordance with the above
procedure. Extraction of the broth yielded a crude extract (26.6 mg),
which was purified as described above to afford 18.0 mg of labeled
DPA (5). The relative incorporations are shown in Table 2.
Biodegradation of Labeled Patulin (1) Deriving from
Sodium [1,2-13C2]-Acetate. The yeast was grown for 3 days in
the presence of labeled patulin (1) from the feeding of P. expansum
with sodium [1,2-13C2]-acetate (see above). Extraction of the broth
yielded a crude extract (26.1 mg), which was purified as above to
obtain 10.2 mg of labeled DPA (5). The coupling constants are
reported in Table 2.
The biodegradation pathway elucidated sets the foundation
for identifying the microbial genes involved in the various
enzymatic steps. As a result, new strategies for monitoring or
reducing the levels of patulin (1) in food can be developed.
EXPERIMENTAL SECTION
■
1
General Experimental Procedures. H and 13C NMR spectra
were recorded on Agilent 400 and 600 MHz spectrometers using
CD3OH as solvent (Merck, Darmstadt, Germany). TLC was
performed on Merck Kiesegel 60 F254, 0.25 mm thick. Silica gel
60PF254 (60−100 mesh, Merck) was used for column chromatog-
raphy.
Reagents and Labeled Precursors. All reagents were purchased
from Sigma-Aldrich (Milan, Italy). Sodium [1-13C]-acetate (isotopic
purity 99 atom % 13C), sodium [1,2-13C2]-acetate (99 atom % 13C),
and H218O (97 atom % 18O) were obtained from Euriso-Top (Saint
Aubin, France). D2O (99.9 atom % D) was obtained from Merck
(Darmstadt, Germany). Both isotope-labeled sodium acetates were
dissolved in water and sterilized through 0.22 μm Millex GP
(Millipore) filters before being added to the flasks containing the
culture medium for microbial growth. H218O and D2O were filter-
sterilized prior to use.
Microorganisms. All microbial strains used in this study were
obtained from the culture collection of the Dipartimento Agricoltura,
̀
Ambiente, Alimenti, Universita del Molise, Italy. Conidial stock
suspensions of Penicillium expansum FS-7 were maintained viable in
80% glycerol at −40 °C. Cells of the yeast Rhodotorula kratochvilovae
LS11, originally isolated from an olive tree, were preserved on
modified PDA medium (200 g of potatos, 20 g of D-glucose, and 20 g
of agar per liter of distilled water) at 4 °C.
Biodegradation of Patulin (1) in H218O. A 50 mL Erlenmeyer
flask, containing 20 mL of Lilly-Barnett medium in 25% of H218O and
supplemented with 150 ppm of labeled patulin (1) obtained from the
feeding of P. expansum with sodium [1-13C]-acetate, was inoculated
with a suspension of R. kratochvilovae LS11 cells (1.0 × 105 CFU/
mL). The yeast was incubated as described above for 6 days.
Extraction of the broth yielded a crude extract (3.4 mg), which was
purified as above to afford 1.2 mg of labeled DPA (5).
Preparation of Labeled Patulin. P. expansum strain FS-7 was
cultivated at 25 °C and 150 rpm in Erlenmeyer flasks (500 mL)
containing 300 mL of PDB medium (200 g of potatos and 20 g of D-
glucose per liter of distilled water) diluted with H2O to a final ratio of
1:10. Each flask was inoculated with fresh conidia to a final
concentration of 4 × 106 conidia/mL. After 4 days of incubation,
the mycelia were transferred into the same number of Erlenmeyer
flasks (500 mL), each containing 300 mL of PDB medium diluted as
above and a filter-sterilized aqueous solution of the labeled precursor.
Three days after administration of the precursor, the culture medium
and mycelia were separated by filtration. The broth was saturated with
NaCl and extracted with ethyl acetate (×3). The organic extract was
washed with H2O (×3) and dried over anhydrous Na2SO4, and the
solvent evaporated under reduced pressure.34 The residue was
purified by column chromatography on Si gel with an increasing
gradient of ethyl acetate in n-hexane as eluent. Fractions containing
Biodegradation of Patulin (1) in 2H2O. A suspension of R.
kratochvilovae LS11 cells (1.0 × 105 CFU/ml) was added to an
Erlenmeyer flask (100 mL) containing 40 mL of Lilly-Barnett medium
in 25% of D2O supplemented with 150 ppm of unlabeled patulin (1).
The yeast was incubated as described above for 5 days. Extraction of
the broth yielded a crude extract (6.1 mg), which was purified as
above to afford 3.2 mg of labeled DPA (5).
Isolation of Unlabeled Patulin (1) and DPA (5). P. expansum
strain FS-7 was cultivated at 25 °C and 150 rpm in six Erlenmeyer
flasks (500 mL), each containing 300 mL of PDB medium diluted
with H2O to a final ratio of 1:10. Each flask was inoculated with fresh
conidia to a final concentration of 4 × 106 conidia/mL. After 4 days of
incubation, the culture medium and mycelia were separated by
filtration. The broth was extracted and purified using the methodology
described above for the isolation of labeled patulin (1), which
provided 128.0 mg of 1.
Four Erlenmeyer flasks (500 mL), each containing 200 mL of Lilly-
Barnett medium supplemented with 150 ppm of 1, were inoculated
with a suspension of R. kratochvilovae LS11 (1.0 × 105 CFU/mL)
cells from a 2−3-day-old culture on PDA. The yeast was incubated for
4 days at 25 °C and 150 rpm. At this time point, the culture medium
and the yeast cells were separated by centrifugation. The medium was
extracted and purified in accordance with the methodology described
above for the isolation of labeled DPA (5), which yielded 34.0 mg of
5.
Block Micronucleus Cytome Assay. The assay was set up using
blood samples from three healthy, nonsmoking adult males who were
less than 40 years old, as provided by AVIS (Italian Association of
Voluntary Blood Donors). Lymphocytes were separated from whole
blood using a density gradient (Histopaque 1077), washed twice in
RPMI 1640 medium, and then cultured at a concentration of 2 × 106
cells in 5 mL of RPMI 1640 medium (4 × 105/mL) supplemented
with 15% fetal calf serum (FCS) (v/v), 0.5% phytohemagglutinin (v/
v), 1% penicillin−streptomycin solution (v/v), and 1 mM L-glutamine
1
patulin (1) were detected by TLC, collected, and analyzed by H
NMR. The signals observed in the 1H NMR spectrum were
coincident with those shown by the commercial sample.
Labeled Patuline (1) from Sodium [1-13C]-Acetate. In
accordance with the above procedure, four flasks inoculated with P.
expansum FS-7 were fed with a filter-sterilized solution of sodium
[1-13C]-acetate in H2O to a final concentration of 300 ppm.
Extraction of the broth yielded a crude extract (120.2 mg), which
was purified as described in the general method to afford 55.0 mg of
labeled patulin (1). The isotope incorporations are shown in Table 1.
Labeled Patulin (1) from Sodium [1,2-13C2]-Acetate. A filter-
sterilized aqueous solution of sodium [1,2-13C2]-acetate was added to
one flask inoculated with P. expansum to a final concentration of 233
ppm. Extraction of the broth yielded a crude extract (35.9 mg), which
was purified as described above to yield 20.7 mg of labeled patulin
(1). The coupling constants are shown in Table 1.
Biodegradation of Patulin (1) by R. kratochvilovae LS11. A
suspension of R. kratochvilovae LS11 cells (1.0 × 105 CFU/mL)
obtained from a 2−3-day-old culture on PDA was added to a 500 mL
Erlenmeyer flask containing 200 mL of Lilly-Barnett medium (10.0 g
of D-glucose, 2.0 g of L-asparagine, 1.0 g of KH2PO4, 0.5 g of MgSO4·
7H2O, 0.01 mg of FeSO4·7H2O, 8.7 mg of ZnSO4·7H2O, 3.0 mg of
MnSO4·7H2O, 0.1 mg of biotin, and 0.1 mg of thiamine per liter of
F
DOI: 10.1021/acs.jnatprod.8b00539
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
solution. The cultures were incubated at 37 °C in a humid
atmosphere in the presence of 5% CO2.55
ACKNOWLEDGMENTS
■
This research was supported by grants from MINECO
(AGL2015-65684-C2-1-R and HI2007-0026) and the Italian
Ministry of University and Research (Project PRIN no.
2006072204). Use of NMR facilities at the Servicio Centra-
lizado de Ciencia y Tecnologia (SCCYT) of the University of
Cadiz is acknowledged.
After 44 h of incubation, all cultures were supplemented with
cytochalasin-B (Cyt B) at a final concentration of 6 μg/mL. After 48
h, lymphocyte cultures were treated with an aqueous solution of
patulin (1) or DPA (5) in sterile double-distilled water to obtain final
concentrations of 0.10, 0.50, 1, 5, and 10 μM. Sterile double-distilled
water was used as the solvent control, whereas positive controls were
EMS (240 μg/mL), an alkylating agent, and COL (0.02 μg/mL), a
microtubule-disrupting tubulin-binding agent.
́
́
REFERENCES
■
At the end of the 72 h incubation period, the lymphocytes were
collected and treated with a mild hypotonic treatment (RPMI 1640
medium−H2O, 1:2, supplemented with 2% FCS) for 2 min and then
fixed in ice-cold methanol−acetic acid (1:1). The slides were prepared
by cytocentrifugation, air-dried, and stained with May Grunwald-
Giemsa.56 All were coded prior to scoring. Scoring was done by using
a Zeiss Axioplan light microscope at 1000× magnification under oil
immersion. MNi, NPBs, and BUDs were scored according to the
criteria described by Fenech et al.57 by examining at least 2000 BN
cells for each concentration of the compounds under assessment
(1000 for each culture). The frequencies of MNi, NPBs, or BUDs in
1000 BN cells were then calculated for each treatment. Cell-cycle
parameters were evaluated by classifying 1000 cells according to the
number of nuclei. The nuclear division index was calculated according
to eq 1, where M1 through M4 indicate the number of cells with 1−4
nuclei and N indicates the total number of scored cells.56 Necrotic
and apoptotic cells were scored according to the guidelines reported
by Fenech et al.57 So, 1000 cells were counted and scored as either
mononucleated, binucleated, trinucleated, tetranucleated, apoptotic,
or necrotic, and percentages of the different types of cells were
calculated. Assays with high % of necrotic or apoptotic or few BN cells
were classified as “toxic”.
(1) Benkerroum, N. Int. Dairy J. 2016, 62, 63−75.
(2) Pitt, J. J.; Hocking, A. D. Fungi and Food Spoilage; Blackie
Academic & Professional: London, 1997.
(3) Wright, S. A. I. Curr. Opin. Food Sci. 2015, 5, 105−109.
(4) Madhyastha, M. S.; Marquardta, R. R.; Masi, A.; Borsa, J.;
Frohllch, A. A. J. Food Prot. 1994, 57, 48−53.
(5) Iwahashi, Y.; Hosoda, H.; Park, J. H.; Lee, J. H.; Suzuki, Y.;
Kitagawa, E.; Murata, S. M.; Jwa, N.-S.; Gu, M.-B.; Iwahashi, H. J.
Agric. Food Chem. 2006, 54, 1936−1942.
(6) Bottalico, A.; Lerario, P.; Visconti, A. Phytopathol. Mediterr.
1980, 19 (2−3), 196−198.
(7) Puel, O.; Galtier, P.; Oswald, I. P. Toxins 2010, 2, 613−631.
(8) De Melo, F. T.; De Oliveira, I. M.; Greggio, S.; Dacosta, J. C.;
Gueceva, T. N.; Saffi, J.; Henriques, J. A.; Rosa, R. M. Food Chem.
Toxicol. 2012, 50, 3548−3555.
(9) Zhou, A. M.; Jiang, L. P.; Geng, C. Y.; Cao, J.; Zhoumg, L. F.
Toxicon 2009, 53, 584−586.
(10) Ayed-Boussema, I.; Abassi, H.; Bouaziz, C.; Hlima, W. B.; Ayed,
Y.; Bacha, H. Environ. Toxicol. 2013, 6, 299−306.
(11) Alves, I.; Oliveira, N. G.; Laires, A.; Rodrigues, A. S.; Rueff, J.
Mutagenesis 2000, 15, 229−234.
(12) Liu, B. H.; Yu, T. S.; Li, S. Y.; Su, M. C.; Wang, M. C.; Shih, S.
M. Toxicol. Appl. Pharmacol. 2003, 191, 255−263.
(13) Donmez-Altuntas, H.; Gokalp-Yildiz, P.; Bitgen, N.; Hamurcu,
Z. Mycotoxin Res. 2013, 29, 63−70.
(14) Glaser, N.; Stopper, H. Food Chem. Toxicol. 2012, 50, 1796−
1801.
M1 + 2M2 + 3M3 + 4M4
NDI =
(1)
N
The experiments were performed in triplicate, and each
concentration was tested in two parallel cultures. Data from the
three experiments were pooled.
Statistical Analysis. All values from toxicity experiments are
means SD. Student’s t-test was used for comparing the mean of a
specific treatment with a control mean. P ≤ 0.05 was considered as
statistically significant.
(15) Zhou, S. M.; Jiang, L. P.; Geng, C. Y.; Cao, J.; Zhong, L. F.
Toxicon 2010, 55, 390−395.
(16) Ferrer, E.; Garcia, J.; Font, G.; Ruiz, M. J. Toxicol. In Vitro
2009, 23, 1504−1509.
(17) Korte, A. Mutat. Res., Genet. Toxicol. Test. 1980, 78, 41−49.
(18) Song, E.; Xia, X.; Su, C.; Dong, W.; Xian, Y.; Wang, W.; Song,
Y. Food Chem. Toxicol. 2014, 7, 122−127.
ASSOCIATED CONTENT
(19) World Health Organization. Evaluation of certain food additives
and contaminants. Forty-fourth report of the joint FAO/WHO expert
committee on food additives. WHO Technical Report Series 859,
Geneva, 1995; pp 36−38.
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.8b00539.
(20) European-Commission. Regulation (EC) No 1881/2006 of 19
December 2006 setting maximum levels for certain contaminants in
foodstuffs, L 364/5, 2006.
1
1H and 13C NMR spectra for patulin (1) and H, 13C,
(21) Moake, M. M.; Padilla-Zakour, O. I.; Worobo, R. W. Compr.
Rev. Food Sci. Food Saf. 2005, 1, 8−21.
2
and H NMR spectra for DPA (5) (PDF)
(22) Kabak, B.; Dobson, A. D. W.; Var, I. Crit. Rev. Food Sci. Nutr.
2006, 46, 593−619.
AUTHOR INFORMATION
(23) Assatarakul, K.; Churey, J. J.; Manns, D. C.; Worobo, R. W. J.
Food Prot. 2012, 75, 717−724.
■
Corresponding Authors
(24) Sant’Ana, A. S.; Rosenthal, A.; Massaguer, P. R. Food Res. Int.
2008, 41, 441−453.
*E-mail: francesca.maffei@unibo.it. Tel: +39 0541 434656.
Fax: +39 0541 4343600.
*E-mail: rosa.duran@uca.es. Tel: +34 956 012718.
́
́
́
(25) Halasz, A.; Lasztity, R.; Abonyi, T.; Bata, A. Food Rev. Int. 2009,
25, 284−298.
(26) Mahunu, G. K.; Zhang, H.; Yang, Q.; Li, C.; Zheng, X. Crit. Rev.
Microbiol. 2016, 42 (4), 643−655.
ORCID
Isidro G. Collado: 0000-0002-8612-0593
Rebecca J. M. Goss: 0000-0001-9869-8205
(27) Moss, M. O.; Long, M. T. Food Addit. Contam. 2002, 19 (4),
387−399.
(28) Ricelli, A.; Baruzzi, F.; Solfrizzo, M.; Morea, M.; Fanizzi, F. P.
Appl. Environ. Microb. 2007, 73, 785−792.
́
́
Rosa Duran-Patron: 0000-0002-5307-4164
Notes
(29) Hawar, S.; Vevers, W.; Karieb, S.; Ali, B. K.; Billington, R.; Be, J.
Food Control 2013, 34, 502−508.
The authors declare no competing financial interest.
G
DOI: 10.1021/acs.jnatprod.8b00539
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(30) Dong, X.; Jianga, W.; Lia, C.; Maa, N.; Xua, Y.; Menga, X. Food
Addit. Contam. A 2015, 32 (3), 352−360.
(31) Chen, Y.; Peng, H.-M.; Wang, X.; Li, B.-Q.; Long, M.-Y.; Tian,
S.-P. Toxins 2017, 9, 2−13.
(32) Zheng, X.; Li, Y.; Zhang, H.; Apaliya, M. T.; Zhang, X.; Zhao,
L.; Jiang, Z.; Yang, Q.; Gu, X. Biol. Control 2018, 123, 127−136.
(33) Sekiguchi, J.; Shimamoto, T.; Yamada, Y.; Gaucher, G. M. Appl.
Environ. Microl. 1983, 45, 1939−1942.
(34) Tannous, J.; Snini, S. P.; El Khoury, R.; Canlet, C.; Pinton, P.;
Lippi, Y.; Alassane-Kpembi, I.; Gauthier, T.; El Khoury, A.; Atoui, A.;
Zhou, T.; Lteif, R.; Oswald, I. P.; Puel, O. Arch. Toxicol. 2017, 91,
2455−2467.
(35) Maidana, L.; Gerez, J. R.; El Khoury, R.; Pinho, F.; Puel, O.;
Oswald, I. P.; Bracarense, A. P. F. R. L. Food Chem. Toxicol. 2016, 98,
189−194.
́
́
(36) Castoria, R.; Mannina, L.; Duran-Patron, R.; Maffei, F.;
Sobolev, A. P.; De Felice, D. V.; Pinedo-Rivilla, C.; Ritieni, A.;
Ferracane, R.; Wright, S. A. I. J. Agric. Food Chem. 2011, 59, 11571−
11578.
́
́
(37) Ianiri, G.; Idnurm, A.; Wright, S. A.; Duran-Patron, R.;
Mannina, L.; Ferracane, R.; Ritieni, A.; Castoria, R. Appl. Environ.
Microbiol. 2013, 79, 3101−3115.
(38) Zhu, R.; Feussner, K.; Wua, T.; Yan, F.; Karlovsky, P.; Zheng,
X. Food Chem. 2015, 179, 1−5.
(39) Fenech, M. Nat. Protoc. 2007, 2, 1084−1104.
(40) Vederas, J. C. Nat. Prod. Rep. 1987, 4, 277−337.
(41) Hansen, P. E. Annu. Rep. NMR Spectrosc. 1984, 15, 105−234.
(42) Lovett, J.; Peeler, J. T. J. Food Sci. 1973, 38, 1094−1095.
(43) Drusch, S.; Kopka, S.; Kaeding, J. Food Chem. 2007, 100, 192−
197.
(44) Zheng, X.; Yang, Q.; Zhao, L.; Apaliya, M. T.; Zhang, X.;
Zhang, H. Sci. Rep. 2017, 7, DOI: 10.1038/s41598-017-14078-5.
(45) Fliege, R.; Metzler, M. Chem. Res. Toxicol. 2000, 13, 373−381.
(46) Sumbu, Z. L.; Thonart, P.; Bechet, J. Appl. Environ. Microbiol.
1983, 45, 110−115.
(47) Ianiri, G.; Panfili, G.; Fratianni, A.; Pinedo, C.; Castoria, R.
Toxins 2017, 9, 61.
(48) Ashby, J.; Tennant, R. W.; Zeiger, E.; Stasiewicz, S. Mutat. Res.,
Genet. Toxicol. Test. 1989, 223, 73−103.
̈
(49) Brase, S.; Encinas, A.; Keck, J.; Nising, C. F. Chem. Rev. 2009,
109, 3903−3990.
(50) Pfenning, C.; Esch, H. L.; Fliege, R.; Lehmann, L. Arch. Toxicol.
2016, 90, 433−448.
(51) Zhang, B.; Peng, X.; Li, G.; Xu, Y.; Xia, X.; Wang, Q. Toxicon
2015, 94, 1−7.
(52) Janzowski, C.; Glaab, V.; Mueller, C.; Straesser, U.; Kamp, H.
G.; Eisenbrand, G. Mutagenesis 2003, 18, 465−470.
́
(53) Gueraud, F.; Atalay, M.; Bresgen, N.; Cipak, A.; Eckl, P. M.;
Huc, L.; Jouanin, I.; Siems, W.; Uchida, K. Free Radical Res. 2010, 44,
1098−1124.
(54) Fliege, R.; Metzler, M. Chem. Res. Toxicol. 2000, 13, 363−372.
(55) Maffei, F.; Carbone, F.; Cantelli Forti, G.; Buschini, A. M.; Poli,
P.; Rossi, C.; Marabini, L.; Radice, S.; Chiesara, E.; Hrelia, P. Environ.
Int. 2009, 35, 1053−1061.
(56) Eastmond, D. A.; Tucker, J. D. Environ. Mol. Mutagen. 1989, 13,
34−43.
(57) Fenech, M.; Crott, J.; Turner, J.; Brown, S. Mutagenesis 1999,
14, 605−612.
H
DOI: 10.1021/acs.jnatprod.8b00539
J. Nat. Prod. XXXX, XXX, XXX−XXX