Condensation of Macrocyclic Polyketides Produced by Penicillium sp. DRF2 with Mercaptopyruvate Represents a New Fungal Detoxification Pathway

Article abstract of DOI:10.1021/acs.jnatprod.6b00295

Application of a refined procedure of experimental design and chemometric analysis to improve the production of curvularin-related polyketides by a marine-derived Penicillium sp. DRF2 resulted in the isolation and identification of cyclothiocurvularins 6-8 and cyclosulfoxicurvularins 10 and 11, novel curvularins condensed with a mercaptolactate residue. Two additional new curvularins, 3 and 4, are also reported. The structures of the sulfur-bearing curvularins were unambiguously established by analysis of spectroscopic data and by X-ray diffraction analysis. Analysis of stable isotope feeding experiments with [U-¹³C₃¹⁵N]-l-cysteine confirmed the presence of the 2-hydroxy-3-mercaptopropanoic acid residue in 6-8 and the oxidized sulfoxide in 10 and 11. Cyclothiocurvularins A (6) and B (7) are formed by spontaneous reaction between 10,11-dehydrocurvularin (2) and mercaptopyruvate (12) obtained by transamination of cysteine. High ratios of [U-¹³C₃¹⁵N]-l-cysteine incorporation into cyclothiocurvularin B (7), the isolation of two diastereomers of cyclothiocurvularins, the lack of cytotoxicity of cyclothiocurvularin B (7) and its methyl ester (8), and the spontaneous formation of cyclothiocurvularins from 10,11-dehydrocurvularin and mercaptopyruvate provide evidence that the formation of cyclothiocurvularins may well correspond to a 10,11-dehydrocurvularin detoxification process by Penicillium sp. DRF2.

Full text of DOI:10.1021/acs.jnatprod.6b00295

Article
pubs.acs.org/jnp
Condensation of Macrocyclic Polyketides Produced by Penicillium sp.
DRF2 with Mercaptopyruvate Represents a New Fungal
Detoxification Pathway
Marcos V. de Castro,†,# Laura P. Ioca,†,# David E. Williams,‡ Bruna Z. Costa,§ Carolina M. Mizuno,†,⊥,¶
́
†
†
†
⊥
∥
́
Mario F. C. Santos, Karen de Jesus, Everton L. F. Ferreira, Mirna H. R. Seleghim, Lara D. Sette,
Edenir R. Pereira Filho,∇ Antonio G. Ferreira,∇ Natalia S. Goncalves,○ Raquel A. Santos,○
̧
́
Brian O. Patrick,¤ Raymond J. Andersen,‡ and Roberto G. S. Berlinck*,†
†Instituto de Quimica de Sao Carlos, Universidade de Sao Paulo, CP 780, CEP 13560-970, Sao Carlos, SP, Brazil
̃
̃
̃
‡Departments of Chemistry and Earth, Ocean & Atmospheric Sciences, University of British Columbia, Vancouver, BC V6T 1Z1,
Canada
§Instituto de Quimica, Universidade Estadual de Campinas, Caixa Postal 6154, CEP 13083-970, Campinas, SP, Brazil
⊥Departamento de Ecologia e Biologia Evolutiva, Universidade Federal de Sao Carlos, Sao Carlos, SP, Brazil
∥Departamento de Bioquímica e Microbiologia, Instituto de Biocien
Campus Rio Claro, Avenida 24-A, 1515, Rio Claro, SP, Brazil
∇Departamento de Química, Universidade Federal de Sao Carlos, CEP 13565-905, Sao Carlos, SP, Brazil
̃
̃
̂
́
cias, Universidade Estadual Paulista “Julio de Mesquita Filho”,
̃
̃
○Laborator
Franca, SP, Brazil
́ ́ ́
io de Genetica e Biologia Molecular, Universidade de Franca, Avenida Dr. Armando Salles Oliveira, 201. Pq. Universitario,
¤Department of Chemistry, University of British Columbia, Vancouver, BC V6T 1Z1, Canada
S
* Supporting Information
ABSTRACT: Application of a refined procedure of experimental design and chemometric analysis to improve the production of
curvularin-related polyketides by a marine-derived Penicillium sp. DRF2 resulted in the isolation and identification of
cyclothiocurvularins 6−8 and cyclosulfoxicurvularins 10 and 11, novel curvularins condensed with a mercaptolactate residue.
Two additional new curvularins, 3 and 4, are also reported. The structures of the sulfur-bearing curvularins were unambiguously
established by analysis of spectroscopic data and by X-ray diffraction analysis. Analysis of stable isotope feeding experiments with
[U−13C315N]-L-cysteine confirmed the presence of the 2-hydroxy-3-mercaptopropanoic acid residue in 6−8 and the oxidized
sulfoxide in 10 and 11. Cyclothiocurvularins A (6) and B (7) are formed by spontaneous reaction between 10,11-
dehydrocurvularin (2) and mercaptopyruvate (12) obtained by transamination of cysteine. High ratios of [U−13C315N]-L-
cysteine incorporation into cyclothiocurvularin B (7), the isolation of two diastereomers of cyclothiocurvularins, the lack of
cytotoxicity of cyclothiocurvularin B (7) and its methyl ester (8), and the spontaneous formation of cyclothiocurvularins from
10,11-dehydrocurvularin and mercaptopyruvate provide evidence that the formation of cyclothiocurvularins may well correspond
to a 10,11-dehydrocurvularin detoxification process by Penicillium sp. DRF2.
dihydroxyphenylacetic acid lactones (DALs).5 The 12-mem-
bered macrocyclic dihydroxyphenylacetic acid lactone curvular-
in (1) was first isolated from growth media of Curvularia fungi.6
Curvularin (1) induces the formation of disordered micro-
acrocyclic polyketides constitute a remarkable group of
Msecondary metabolites presenting complex structural
features, potent biological activities, and essential adaptative
functions for the corresponding producing organisms.1 Fungi
are notable producers of structurally diverse polyketides with a
variety of structural motifs,2−4 including 10- to 18-membered
macrolides known as resorcinylic acid lactones (RALs) and
Received: April 8, 2016
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
tubule-organizing centers in centrosomes and of barrel-shaped
spindles in nuclei of sea urchin eggs.7 Several curvularin
derivatives have been isolated from fungal strains belonging to
the genera Alternaria,8 Aspergillus,9 Brickellia,10 Cochliobolus,11
Corynespora,12 Curvularia,13 Eupenicillium,14 and Penicillium.15
Curvularins exhibit an array of biological activities,16 including
phytotoxic,8a cytotoxic,7,9,13b,15i nematicidal,9 antibacterial,13a
kinase inhibitory,12 antifungal and antialgal,13c antitrypanoso-
mal, and anti-inflammatory activities15i as well as inhibiting
nitric-oxide synthase expression.17 Recent investigations on the
biosynthesis of curvularin-related metabolites highlighted the
details of the biochemical machinery involved in the formation
of this unique class of macrocyclic polyketides.18
Our current investigation of secondary metabolites from
marine-derived fungi19−21 led to the isolation and selection of a
Penicillium sp. DRF2 strain obtained from the sponge
Dragmacidon reticulatum.22 Penicillium DRF2 produced a
cytotoxic extract from which the known 10,11-dehydrocurvu-
larin (2) and the new 12-keto-10,11-dehydrocurvularin (3)
were isolated as the major metabolites when the fungus was
grown under standard conditions. Improved production of
minor metabolites detected by HPLC-UV-MS was accom-
plished by applying a refined version of our previously
described experimental design and chemometrics analysis
procedure.19,21 We have been able to promote the fungus
production, as well as to isolate and identify the new curvularin
4, as well as the novel sulfur-bearing curvularin derivatives 6−8,
10, and 11, which, along with 3 and 4, are the subject of the
present report.
RESULTS AND DISCUSSION
■
The extract obtained from the culture medium of Penicillium
DRF2 displayed cytotoxic activity and HPLC peaks with UV
absorptions at λmax 222−230, 270−280, and 300−310 nm and
[M + H]+ ions at m/z 300−320, along with minor compounds
with m/z > 400. A larger growth of the fungus in 3% malt
medium under standard conditions (see SI), followed by
purification of the medium extract by a series of chromato-
graphic separations, led to the isolation of 10,11-dehydrocur-
vularin (2) and 12-keto-10,11-dehydrocurvularin (3). Applica-
tion of a refined procedure of our previously described
experimental design and chemometric analysis19,21 (see SI),
followed by solid-phase extraction of the culture medium,
Sephadex LH20 column chromatography, and reversed-phase
HPLC purifications, enabled us to establish growth conditions,
optimize the production, and isolate minor curvularin
derivatives. These include the new curvularin 4, along with
the tetrahydrothiophene-bearing metabolites 6−8, 10, and 11.
The (+)-HRESIMS of 12-keto-10,11-dehydrocurvularin (3)
gave a [M + H]+ quasi-molecular ion at m/z 305.1050, which
corresponded to the molecular formula C16H16O6. This formula
required nine degrees of unsaturation, one more than that for
10,11-dehydrocurvularin (2), with one additional oxygen and
two hydrogens less than 2. The NMR data of 3 were very
similar to those of 2, the only difference being that the
resonances assigned to the CH2-12 methylene (δH 2.38 (m)/δC
33.0) in 213b had been replaced by an additional α,β-
unsaturated ketone resonating at δC 201.5 (C-12). The
assignment of a ketone at C-12 was confirmed by the
observation of HMBC correlations between both CH2-13 (δH
2.61/δC 37.0) and CH2-14 (δH 1.81 and 2.10/δC 32.0) and C-
12 (δ 201.5) and between the olefinic protons assigned to the
CH-10/CH-11 E double bond (J = 16.4 Hz) (δH 7.44/δC
Figure 1. Structures of curvularin (1), 15(S)-10,11-dehydrocurvularin
(2), 15(S)-12-keto-10,11-dehydrocurvularin (3), 15(S)-cis-10,11-epox-
ycurvularin (4), trans-10,11-epoxycurvularin (5), 10(R),11(R),15-
(S),18(S)-cyclothiocurvularin A (6), 10(S),11(S),15(S),18(R)-cyclo-
thiocurvularin B (7), 10(S),11(S),15(S),18(R)-cyclothiocurvularin B
methyl ester (8), 10(S),11(S),15(S),18(R)-dimethoxymethyl-
cyclothiocurvularin B methyl ester (9), 15(S),18(S)-cyclosulfoxicurvu-
larin (10), and 15(S),18(S)-cyclosulfoxicurvularin methyl ester (11).
139.6, δH 6.51/δC 135.0) and C-12 (δ 201.5). Since the specific
rotation of 2 isolated in the current investigation ([α]25 −8.2
D
(c 10.0, MeOH)) compared favorably to that reported in the
literature,13 the absolute configuration at C-15 is assumed to be
the same as that previously published. Similarly for 3, with a
specific rotation recorded as [α]25 −7.6 (c 0.05, MeOH), the
D
absolute configuration at C-15 is also assumed to be S.
Although C-12 oxygenation is unexpected since curvularins are
regular polyketides,23 12-oxocurvularin,24 curvulone A,13c and
8-β-hydroxy-7-oxocurvularin25 also possess oxygenation at this
position.
HRESIMS analysis of cis-10,11-epoxycurvularin (4) gave a
[M + H]+ ion at m/z 307.11883 and a [M + Na]+ ion at m/z
329.10063, appropriate for the molecular formula C16H18O6
requiring eight degrees of unsaturation, the same as 10,11-
dehydrocurvularin (2) with the addition of H2O. Although
1
analysis of the H, 13C, HSQC, COSY, and HMBC spectra
1
indicated a close similarity between 4 and 3 (Table 1), the H
and 13C resonances of the C-10/C-11 double bond were no
longer observed. Differently from 2 and 3, compound 4
B
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 1. NMR Data (1H 400 MHz, 13C 100 MHz, MeOH-d4)
for Curvularins 3 and 4
spectrum H-10 correlated to C-9 (δC 208.3), to the quaternary
C-18 (δC 87.6), and to the C-19 carboxylic carbon (δC 176.3).
H-11 also correlated with C-9 along with C-10, C-12, and C-13,
while H-17a/H17b correlated with C-10, C-11, and C-18 and
weakly with C-19. The aforementioned HMBC correlations
confirmed the presence of a 3-hydroxytetrahydrothiophene-3-
carboxylic acid structural moiety in 6 (Figure 2). The trans
a
12-keto-10,11-
dehydrocurvularin (3)
cis-10,11-epoxycurvularin (4)
position
δC, type
170.7, C
δH (J in Hz)
δC, type
δH (J in Hz)
1
2
173.2, C
41.5, CH2 3.71 (d, 17.2);
3.68 (d, 17.2)
39.4, CH2 3.57 (d, 18.2);
3.08 (d, 18.4)
3
138.5, C
135.8, C
4
113.5, CH
163.5, C
6.36 (d, 2.4)
6.26 (d, 2.6)
111.2, CH
162.5, C
102.9, CH
159.7, C
120.2, C
204.8, C
62.0, CH
63.8, CH
6.32 (d, 1.9)
6.19 (d, 2.1)
5
6
101.9, CH
164.9, C
7
8
113.9, C
9
197.2, C
10
11
139.6, CH
135.0, CH
7.44 (d, 16.4)
6.51 (d, 16.4)
3.83 (d, 2.2)
Figure 2. Key HMBC correlations observed for cyclothiocurvularin A
(6).
3.13 (ddd, 2.2,
3.0, 10.2)
12
13
14
201.5, C
33.7, CH2 1.92 (m); 1.02
(m)
37.0, CH2 2.61 (dd, 5.7
and 6.6)
22.2, CH2 1.77 (m); 1.02
(m)
cyclic ring junction between the tetrahydrothiophene ring and
the macrocyclic ring was suggested by a vicinal coupling of 8.5
Hz between H-10 (δH 4.50) and H-11 (δH 3.68), indicating an
anti-relationship. Crystallization of 6 from MeOH gave
colorless block crystals that were suitable for single-crystal X-
ray diffraction analysis. The absolute configuration of cyclo-
thiocurvularin A (6) was thus established as 10(R), 11(R),
15(S), 18(S) on the basis of the refined Flack parameter,
0.05(5) (Figure 3).
32.0, CH2 1.81 (m); 2.10
(m)
35.1, CH2 1.72 (m); 1.44
(m)
15
16
71.1, CH
4.85 (m)
75.4, CH
4.54 (m)
19.4, CH3 1.17 (d, 6.8)
20.7, CH3 1.11 (d, 6.2)
a
The signal of the 7-OH group (δ 12.1, s) was observed in DMSO-d6.
presented two oxymethine signals at δH 3.83/δC 62.0 (CH-10)
and δH 3.13/δC 63.8 (CH-11), assigned to an epoxide group at
C-10 and C-11. The small coupling constant (2.2 Hz) observed
between H-10 and H-11,15i,26 as well as the molecular formula,
The (+)-HRESIMS of cyclothiocurvularin B (7) gave a [M +
H]+ ion at m/z 411.11299 and a [M + Na]+ ion at m/z
433.09405, appropriate for the same molecular formula as that
of 6 (C19H22O8S). The NMR data of 7 (Table 2) were very
similar to those of 6. Only minor differences were observed in
1
provided further support for this assignment. A H NOEDiff
experiment irradiating at δ 3.83 (H-10) showed a clear NOE on
H-11 at δ 3.13 (Figure S11, SI), indicating a cis relative
stereochemistry at C-10/C-11. Compound 4 corresponds to
the cis- stereoisomer of the trans-stereoisomer 5, which has
been previously named 10,11-epoxycurvularin.15i trans-10,11-
Epoxycurvularin (5) was erroneously drawn as 4.15i The
structure of 5 is herein correctly drawn as the trans-
stereoisomer of 10,11-epoxycurvularin. The NMR data
reported for trans-10,11-epoxycurvularin (5)15i are clearly
different from the data we obtained for 4. The stereochemistry
at C-15 was suggested as 15(S) based on the configuration of 2
and 3, produced in the same pool of polyketides by Penicillium
sp. DRF2, but the relative configuration between the epoxide
group and C-15 could not be established. Compound 4 was
named 15(S)-cis-10,11-epoxycurvularin.
1
the H and 13C NMR chemical shifts assigned to 6 and 7,
suggesting that the two compounds were diastereomers (Table
2). Treatment of 7 with diazomethane generated colorless plate
crystals of the dimethoxymethyl ester of 7 (9) that were
suitable for single-crystal X-ray diffraction analysis. The result of
the analysis points to an absolute configuration for 7 of 10(S),
11(S), 15(S), 18(R) (Figure 3), which as suspected is
diastereomeric with 6.
Cyclothiocurvularin B methyl ester (8) gave a [M + H]+ ion
at m/z 425.12865 and a [M + Na]+ ion at m/z 447.10943
(calcd 447.10841) that were appropriate for the molecular
formula C20H24O8S, which differs from that of 6 and 7 by the
addition of CH2. Other than the presence of a methyl ester at
δH 3.62/δC 53.3 (CH3-20), the NMR data obtained for 8
(Table 3) were strikingly similar to those of 7 (Table 2).
Therefore, the structure of 8 was assigned as the methyl ester of
7.
HRESIMS analysis of cyclothiocurvularin A (6) gave a [M −
H]− ion at m/z 409.0948, appropriate for the molecular
formula C19H22O8S, requiring nine sites of unsaturation and
differing from the molecular formula of 2 by the addition of
Sulfoxides 10 and 11 were also isolated from the growth
medium of Penicillium sp. DRF2 obtained under the same
improved conditions. The HRESIMS of compound 10 gave a
[M + H]+ ion at m/z 427.1049 and a [M + Na]+ ion at m/z
449.0855, appropriate for the molecular formula C19H22O9S,
which differs from that of 6 and 7 by the addition of an oxygen
atom. The only significant differences observed in the 13C
spectrum of 9 when compared with 6−8 was that in 10 C-10
(δC 58.4) was more shielded than in 6−8 (64.0 < δC < 64.9),
and C-11 in 10 was deshielded (δC 59.8) compared to 6−8
(49.8 < δC < 50.4), as was the C-17 methylene resonance in 10
(δC 65.5) relative to 6−8 (40.9 < δC < 41.2). Such changes,
C3H4O3S. Analysis of the H, 13C, HSQC, COSY, and HMBC
1
spectrum of 6 in MeOH-d4 indicated close similarity to
compound 2. The resonances assigned to the C-10/C-11 olefin
seen in 2 were also not observed in 6. Instead, C-10 and C-11
were assigned to saturated methines at δH 4.50 (d, 8.5 Hz)/δC
64.5 and δH 3.68 (bt, 8.5 Hz)/δC 49.8, respectively. Methines
CH-10 and CH-11 coupled to one another in the COSY
experiment. H-11 also vicinally correlated to the methylene
CH2-12 at δH 1.96 and 1.50/δC 35.8. The additional C3H4O3S
fragment was assigned to a 2-hydroxy-3-mercaptopropanoic
acid moiety connected to CH-10 and CH-11. In the HMBC
C
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 3. ORTEP diagram generated from the X-ray diffraction analysis of cyclothiocurvularin A (6) and of the dimethoxymethyl ether of
cyclothiocurvularin B (9), respectively.
Table 2. NMR Data (1H 400 MHz, 13C 100 MHz) for Cyclothiocurvularins A (6) and B (7)
a
b
a
cyclothiocurvularin A (6)
cyclothiocurvularin B (7)
cyclothiocurvularin B (7)
position
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
1
171.7, C
40.7, CH2
138.4, C
112.5, CH
163.3, C
103.4, CH
161.4, C
120.9, C
208.3, C
64.5, CH
49.8, CH
35.8, CH2
23.4, CH2
30.3, CH2
73.5, CH
17.7, CH3
41.2, CH2
87.6, C
171.4, C
38.4, CH2
136.2, C
112.0, CH
161.6, C
103.3, CH
159.0, C
121.8, C
209.3, C
64.2, CH
49.3, CH
34.6, CH2
27.3, CH2
31.2, CH2
75.6, CH
21.7, CH3
40.9, CH2
87.2, C
172.6, C
38.5, CH2
135.9, C
111.6, CH
162.7, C
103.4, CH
160.2, C
121.7, C
209.7, C
64.4, CH
50.1, CH
34.9, CH2
27.7, CH2
31.6, CH2
76.4, CH
21.7, CH3
40.9, CH2
88.0, C
2
4.33 (d, 10.6); 3.27 (d, 10.5)
6.19 (d, 1.5)
4.07 (d, 12.1); 3.16 (d, 12.2)
6.48 (bs)
4.13 (d, 11.9); 3.12 (d, 11.9)
6.45 (d, 1.9)
3
4
5
6
6.25 (d, 1.5)
6.25 (bs)
6.22 (d, 1.9)
7
8
9
10
11
12
13
14
15
16
17
18
19
4.50 (d, 8.5)
4.10 (d, 11.2)
4.27 (d, 11.0)
3.68 (bt, 8.5)
3.88 (dt, 2.2, 11.3)
1.56 (m); 1.37 (m)
1.53 (m); 1.43 (m)
1.67 (m); 1.45 (m)
4.64 (dq, 6.0, 10.0)
1.18 (d, 6.2)
3.93 (t, 11.5)
1.96 (m); 1.50 (m)
1.60 (m); 1.43 (m)
1.67 (m); 1.34 (m)
4.98 (m)
1.66 (m); 1.36 (m)
1.50 (m)
1.73 (m); 1.42 (m)
4.64 (m)
1.03 (d, 4.3)
1.21 (d, 6.3)
3.44 (d, 7.3); 2.91 (d, 7.5)
3.43 (d, 11.4); 2.91 (d, 11.4)
3.45 (d, 11.4); 2.92 (d, 11.4)
176.3, C
174.8, C
176.6, C
a
b
MeOH-d4. MeCN-d3.
along with the additional oxygen atom and the lack of evidence
for any additional oxygen-substituted carbons, suggested the
presence of a sulfoxide in the structure of 10. In all other
respects cyclosulfoxicurvularin (10) was the same as cyclo-
thiocurvularin B (7), including the trans-junction between the
tetrahydrothiophene moiety and the macrolactone ring,
indicated by the coupling constant (J = 8.2 Hz) between H-
10 and H-11 in 10. The C-16 chemical shift suggests the same
configuration as 7 and 8 at C-15. The HRESIMS of compound
11 gave a [M − H]− ion at m/z 439.1056, appropriate for the
molecular formula C20H24O9S, which differs from that of 10 by
the addition of CH2. As with 8 the presence of a methyl ester
(δH 3.75 (s)/δC 53.6) was readily established, and 11 was
assigned as the methyl ester of sulfoxide 10. Although the
coupling constant of 8.1 Hz between H-10 and H-11 in 10
indicates the same trans-relationship as observed for 6−8, the
absolute configurations of 10 and 11 at C-10, C-11, and C-18
and of the sulfoxide group were not assigned because we have
been unable to obtain suitable crystals for X-ray diffraction
analyses.
The occurrence of sulfur-bearing macrolides produced by
fungi is rare. The few examples reported in the literature
include pandangolide 2, bearing a 2-mercaptoacetic acid,27 and
pandangolide 4, which presents a sulfide bridge connecting two
macrocyclic polyketides.28 Pandangolide 3 contains an
uncyclized 2-hydroxy-3-mercaptopropanoic acid (mercaptolac-
tate) moiety,28 the same sulfur-bearing residue observed in
sumalarins A−C, recently isolated from culture media produced
D
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 3. NMR Data (1H 400 MHz, 13C 100 MHz) for Cyclothiocurvularin B (8) and Cyclosulfoxicurvularins 10 and 11
a
b
b
cyclothiocurvularin B methyl ester (8)
cyclosulfoxicurvularin (10)
cyclosulfoxicurvularin methyl ester (11)
position
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
1
172.7, C
38.7, CH2
136.4, C
112.2, CH
163.1, C
103.2, CH
160.5, C
121.7, C
209.8, C
64.9, CH
50.4, CH
35.6, CH2
28.1, CH2
31.8, CH2
76.5, CH
22.0, CH3
41.0, CH2
87.6, C
170.7, C
37.7, CH2
135.2, C
110.9, CH
161.4, C
102.8, CH
159.5, C
119.7, C
206.0, C
58.4, CH
59.8, CH
22.4, CH2
24.6, CH2
30.6, CH2
74.7, CH
21.6, CH3
65.5, CH2
84.0, C
172.7, C
38.9, CH2
137.2, C
112.6, CH
161.1, C
103.3, CH
163.7, C
120.5, C
207.5, C
59.8, CH
62.3, CH
24.0, CH2
26.1, CH2
32.0, CH2
76.6, CH
21.9, CH2
66.2, CH2
85.0, C
2
4.07 (d, 12.1); 3.01 (d, 12.0)
6.35 (d, 2.4)
4.10 (d, 7.9); 3.15 (d, 8.0)
6.34 (d, 1.4)
4.36 (d, 7.9); 3.12 (d, 7.9)
6.44 (d, 1.3)
3
4
5
6
6.13 (d, 2.2)
6.23 (d, 1.4)
6.25 (d, 1.3)
7
8
9
10
11
12
13
14
15
16
17
18
19
20
4.04 (d, 11.2)
4.55 (d, 8.2)
n.o.
3.84 (dt, 2.2, 11.3)
1.55 (m); 1.28 (m)
1.44 (m)
3.50 (t, 8.2)
3.72 (dd, 1.2, 8.1)
1.88 (m); 1.47 (m)
1.68 (m); 1.55 (m)
1.88 (m); 1.59 (m)
4.74 (m)
1.70 (m); 1.26 (m)
1.49 (m); 1.43 (m)
1.75 (m); 1.52 (m)
4.62 (m)
1.65 (m); 1.37 (m)
4.56 (m)
1.11 (d, 6.3)
1.11 (d, 6.3)
1.24 (d, 4.1)
3.48 (d, 11.4); 2.80 (d, 11.5)
3.90 (d, 9.1); 2.68 (d, 8.9)
4.02 (d, 9.4); 3.13 (d, 9.4)
174.4, C
53.3, CH3
173.0, C
173.1, C
53.6, CH3
3.62 (s)
3.75 (s)
a
b
MeCN-d3. MeOH-d4.
by Penicillium sumatrense.29 To the best of our knowledge, only
a single compound has been previously reported with a
mercaptolactate moiety incorporated into a tetrahydrothio-
phene ring: pharbitic acid, a giberellin derivative isolated from
the Japanese morning-glory Pharbitis nil.30 The isolation of
cyclothiocurvularins and cyclosulfoxicurvularin provides addi-
tional examples of this unique class of secondary metabolites.
By feeding cultures of Penicillium sp. DRF2 with
[U−13C315N]-L-cysteine the biogenetic origin of the mercapto-
lactate residue in both 6 and 7 has been unambiguously
established. The 13C NMR spectrum of cyclothiocurvularin B
(7) isolated from this stable isotope feeding experiment showed
>99% 13C incorporation at C-17, C-18, and C-19 (Table 4 and
Figure S24, SI). This incorporation ratio was calculated by
integration of the 13C-enriched signals of C-17, C-18, and C-19
and division of the integral values by the integral values of the
13C-nonenriched signals of the same carbons. These results
demonstrate that L-cysteine is the amino acid precursor of the
mercaptolactate moiety in cyclothiocurvularins. The very high
levels of [U−13C315N]-L-cysteine incorporation into cyclo-
thiocurvularins suggest that, when added to Penicillium sp.
DRF2 growth medium, labeled cysteine is completely
converted into 3-mercaptopyruvate, which is subsequently
condensed to the 10,11-dehydrocurvularin scaffold. Earlier
studies on cysteine metabolism provided only indirect evidence
that mercaptolactate is derived from cysteine through 3-
mercaptopyruvate (12, Scheme 1).31 It has also been
demonstrated that Aspergillus nidulans can synthesize cysteine
and methionine from inorganic sulfate.32 Since the growth
medium of Penicillium sp. DRF2 contains Na2SO4 in 2 mM
concentration (SI), the availability of sulfate may stimulate the
production of cyclothiocurvularins and cyclosulfoxicurvularins
under the optimal growth conditions used for the production of
these metabolites.
Table 4. Incorporation Ratios (%) of [U−13C315N]-L-
Cysteine into Cyclothiocurvularin B (7)
a
position
δ
13C
[U−13C315N1]-L-cysteine
1
172.5
38.5
1.80
2.03
2.13
2.28
1.85
2.37
1.84
2.19
1.09
2.42
2
3
136.3
111.8
162.8
103.6
160.5
121.7
209.7
64.0
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
b
50.0
n.d.
34.9
2.30
2.45
2.11
2.18
2.30
99.0
99.9
99.9
27.7
31.5
76.3
21.7
40.9
88.0
176.6
a
b
100 MHz, 12 h acquisition time. n.d.: overlapped by the solvent
signal.
7, raised the hypothesis that both 6 and 7 were products of a
spontaneous reaction between 10,11-dehydrocurvularin (2)
and mercaptopyruvate (12). We hypothesized that when the
fungus increases the production of 10,11-dehydrocurvularin, it
starts to produce 3-mercaptopyruvate, which would sponta-
neously react with 2 in order to diminish its concentration in
the growth medium. This hypothesis was tested by performing
a reaction between 2 and 12 in a buffer medium. Four
transaminases were evaluated for the catalytic conversion of L-
cysteine into 3-mercaptopyruvate (12) (SI). Conversions
superior to 80% were observed. The ω-transaminase ATA-
The isolation of two cyclothiocurvularin stereoisomers at C-
10 and C-11, as well as the very high levels of [U−13C315N]-L-
cysteine incorporation into the mercaptolactate moiety of 6 and
E
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Scheme 1. Spontaneous Formation of Cyclothiocurvularin B (7) from 10,11-Dehydrocurvularin (2) and Mercaptopyruvate (12)
Figure 4. Spontaneous conversion of 10,11-dehydrocurvularin (2) to cyclothiocurvularins A (6) and B (7) in the presence of mercaptopyruvate
(12). Black line: Cyclothiocurvularin B (7) standard (tR 1.99 min). Mauve line: Cyclothiocurvularin A (6) standard (tR 1.87 min). Green line: 10,11-
Dehydrocurvularin (2) + L-cysteine + ω-ATA 113 transaminase reaction replicate 1. Red line: 10,11-Dehydrocurvularin (2) + L-cysteine + ω-ATA
113 transaminase reaction replicate 2. 10,11-Dehydrocurvularin (2) is observed at tR 2.17 min. uPLC-qTOF analysis conditions: column: Acquity
UPLC BEH C18 (1.7 μm, 2.1 × 50 mm); eluent: gradient of MeCN (+ 0.01% HCO2H) in H2O (+ 0.01% HCO2H), starting at 10% to 100% MeCN
(+ 0.01% HCO2H) in 4 min. Detection: MSE continuum during 5 min, m/z 250−450 molecular weight range; detection mode: ESI(+); scan time:
0.2 s; collision energy: 3 V; ramp collision energy: 20−30 V.
113 gave the best conversion of L-cysteine into mercaptopyr-
uvate (94% in 24 h). The subsequent reaction of 10,11-
dehydrocurvularin (2) in the presence of L-cysteine and ω-
transaminase ATA-113 successfully yielded cyclothiocurvularins
A (6) and B (7) in >90% conversion and in a ratio of 1:3, by
comparison with authentic standards (Figure 4).
for the spontaneous formation of 6 and 7 from 2 and 12
(Figure 4). The isolation ratio of 6/7 from the experiments
using [U−13C315N]-L-cysteine is 2:7. The ratio for the
spontaneous formation of 6 and 7 from 2 and 12 is explained
considering the most stable conformation of 15(S)-2 (Scheme
1), which shows an upper more sterically hindered face, which
does not favor the reaction between 15(S)-2 and 12. The
spontaneous reaction between 15(S)-2 and 12 leading to 7 is
likely thermodynamically driven at the bottom face of 15(S)-2,
which is sterically less hindered. Formation of 7 involves a syn-
addition to the double bond si−si face of 2. The resulting
product from the addition at the bottom face is 10(S),11-
(S),15(S),18(S)-cyclothiocurvularin B (7). 10(R),11(R)-Cyclo-
thiocurvularin A (6) is the minor compound formed by syn-
addition of 12 at the upper face of the 10,11-dehydrocurvularin
(2) enone double bond. The above results clearly demonstrate
a chemical rather than a biochemical conversion of 2 and 12
into 6 and 7.
Our results unambiguously showed that, while L-cysteine
requires enzymatic conversion to mercaptopyruvate (12),
formation of cyclothiocurvularins A (6) and B (7) result
from the spontaneous condensation of 12 with 10,11-
dehydrocurvularin (2), via a Michael-type addition to the
conjugated double bond followed by cyclization at the α-
carbonyl group of mercaptopyruvate (Scheme 1). It is well
known that 2 suffers Michael-type additions at the enone
group.8,19 The high levels of [U−13C315N]cysteine incorpo-
c
ration into cyclothiocurvularin B (7) is thus explained by its
efficient transamination to mercaptopyruvate, followed by
condensation with 2. Cyclothiocurvularin B (7) is the major
cyclothiocurvularin isolated from the growth medium of
Penicillium sp. DRF2, either in the presence of [U−13C315N]-
cysteine or not. The overall isolation ratio of 1:12 between 6
and 7 is related to the 1:3 ratio of 6/7 obtained by reaction of 2
with L-cysteine in the presence of the transaminase, observed
The spontaneous formation of cyclothiocurvularins and
cyclosulfoxicurvularin is related to the recent report that
when the enone-bearing polyketide Sch-642305, produced by
the endophyte Phomopsis sp. CMU-LMA, is added to the
culture of Aspergillus niger ATCC 16404, the product of Sch-
F
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
642305 condensation with mercaptolactate was isolated. The
authors suggest that the resulting product is a protected form of
Sch-642305, from its reaction with mercaptolactate, itself
derived from cysteine.31a An analogous pathway is proposed
for the formation of paraphaeosphaeride A by the endophytic
fungus Paraphaeosphaeria neglecta FT462, although the authors
do not indicate a protective role for the producing strain in
providing paraphaeosphaeride A from paraphaeosphaeride C
and either mercaptolactate or 3-mercaptopyruvate.33 A similar
Michael-type addition reaction of mercaptopyruvate (12) with
an enone, followed by cyclization, is proposed in the pathway
for the biogenesis of the mercaptolactate-bearing plant
metabolite pharbitic acid.30 The fact that 10,11-dehydrocurvu-
larin (2) and cyclothiocurvularins A (6) and B (7) have the
same absolute configuration at C-15 is related to the recent
report that the thioesterase domain of curvularins promotes a
stereospecific cyclization at C-15, leading exclusively to the
15(S) stereoisomer.34
Isolation of cyclothiocurvularins and cyclosulfoxicurvularin
and the high level of cysteine incorporation have implications
for the understanding of curvularin production by Penicillium
sp. DRF2. The nonenzymatic condensation of 3-mercaptopyr-
uvate (12) with 10,11-dehydrocurvularin (2) leading to
cyclothiocurvularins 6 and 7, possibly followed by further
oxidation to cyclosulfoxicurvularin 10, probably results from the
enhanced production of 2 by the fungus under the optimal
conditions herein established. High concentrations of 2 are very
likely toxic to the fungus due to its intrinsic reactivity as a
Michael acceptor. We propose that in order to minimize high
concentrations of 2, the fungus produces 3-mercaptopyruvate
(12) from cysteine, to react with 2. Thus, high concentrations
of 10,11-dehydrocurvularin may well stimulate the biocatalytic
formation of cyclothiocurvularins as detoxification products.
Compounds 3, 4, 7, and 8 were tested against the A549
(lung), HeLa (cervix), and MCF-7 (breast) tumor cell lines, as
well as against the nontumoral breast cell line MCF-10A.
Compounds 3, 4, 7, and 8 were essentially inactive in these
assays. The uncyclized sumalarin A displayed a much more
potent cytotoxicity to cancer cells29 than compounds 6 and 7.
Therefore, the formation of cyclothiocurvularins and cyclo-
sulfoxicurvularin clearly abolishes the cytotoxic activity of
10,11-dehydrocurvularin.
enone group. Thus, cyclothiocurvularins 6 and 7, as well as
cyclosulfoxicurvularin 9, are very likely detoxification products
produced under conditions of fungal stress. Our findings
provide significant insights into the metabolic regulation of
fungal secondary metabolite production in culture conditions.
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
measured in MeOH on a PerkinElmer 341 MC polarimeter at 20
°C or on a Jasco P-2000 polarimeter at 25 °C. NMR spectra were
recorded on a Bruker AvanceIII 9.4 T instrument, operating at 400.35
1
MHz for H and 100.10 MHz for 13C channels, respectively, or in a
1
Bruker AvanceIII 14.1 T instrument, operating at 600.23 MHz for H
and 150.94 MHz for 13C, respectively, with a 5 mm cryoprobe. All
NMR spectra were obtained at 25 °C using tetramethylsilane as an
internal reference. Solvents used for extraction and column
chromatography were glass distilled prior to use. HPLC-grade solvents
were used without further purification in HPLC separations. TLC
analyses were performed with precoated TLC sheets of Si gel on
polyester, eluting with different mixtures of MeOH in CH2Cl2. Plates
were observed under a UV lamp (λmax 254 and 365 nm).
Semipreparative HPLC separations were performed with a Waters
quaternary pump 600 and double beam UV detector 2487 monitored
by a Waters Millenium 32. HPLC-LRESIMS analyses were performed
using a Waters Alliance 2695 coupled online with a Waters 2996
photodiode array detector, followed by a Micromass ZQ2000 MS
detector with an electrospray interface. The photodiode array scanned
the samples at λmax 200−400 nm. The MS detector was optimized to
the following conditions: capillary voltage, 3.00 kV; sample cone
voltage, 30.0 V; source block temperature, 100 °C; desolvation
temperature, 350 °C, operating in electrospray positive and negative
mode, detection range, 200−800 Da with total ion count extracting
acquisition. The cone and desolvation gas flow were 50 and 350 L/h,
respectively, with a Nitrogen Peak Scientific N110DR nitrogen source.
Data acquisition and processing were performed using Empower 2.0.
HPLC-HRMS analyses were recorded on a Micromass Q-tof Micro, in
ES+ mode, using the following experimental conditions: capillary
voltage, 3.0 kV; sample cone, 42.0 V; sample was infused at 10 μL/min
in 1:1 MeOH/H2O.
Isolation and Identification of Penicillium sp. DRF2.
Penicillium sp. DRF2 was isolated from the sponge Dragmacidon
reticulatum, collected in Sao Sebastiao, SP, Brazil, in September 2005,
̃
̃
by scuba diving, at depths between 5 and 10 m. Exact GPS positioning
collecting points are 23°45′32″ south; 45°15′8″ west and 23°47′759″
south; 45°09′380″ west. The strain isolation procedure is described in
detail in the SI. After 1 week of growth, fungal strains were
exhaustively purified before preservation using Castellani’s meth-
od.35,36 Penicillium sp. DRF2 was identified by molecular analyses, as
described in the SI.
CONCLUSION
■
HPLC-UV-MS screening and refinement of our previously
reported experimental design and chemometric analysis to
improve the production of minor metabolites of fungal strains21
enabled us to obtain good production of cyclothiocurvularins A
(6) and B (7) and cyclosulfoxicurvularin (9) and, in so doing,
allowed us to fully characterize and investigate the biogenesis
and biological activities of these unusual metabolites. To the
best of our knowledge, cyclothiocurvularins A (6) and B (7)
and cyclosulfoxicurvularin (9) are the first members of such
macrolides bearing the very rare mercaptolactate residue
embedded in a tetrahydrothiophene or in a tetrahydrothio-
phene-1-oxide cyclic moiety. The nonenzymatic formation of
compounds 6, 7, and 9 from mercaptopyruvate and 10,11-
dehydrocurvularin (2) probably results from the overproduc-
tion of 10,11-dehydrocurvularin (2). In high concentrations,
compound 2 may well be toxic to the producer strain due to its
intrinsic cytotoxicity as a Michael addition acceptor, leading to
an efficient conversion of cysteine to mercaptopyruvate before
its spontaneous condensation with the 10,11-dehydrocurvularin
Fungal Growth, Extraction, and Isolation of Compounds 3,
4, and 6−8, 10 and 11. A small-scale growth and metabolic
production experiment under standard conditions was performed
starting with inocula of previously grown Penicillium sp. DRF2 during
7 days. Inocula were transferred into eight 500 mL Schott flasks
containing 250 mL of growth medium each (malt 2% in 1 L of
sterilized ASW), for growth in still mode at 25 °C during 30 days. At
the end of growth, 250 mL of EtOAc was added to each Schott flask.
Each flask containing the mixture EtOAc + growth medium was
blended and shaken overnight. The blended mixture was filtered
through a Celite pad, transferred to a separatory funnel, and subjected
to liquid−liquid partitioning. The organic phase was evaporated to
dryness, to give 354 mg of the EtOAc crude extract.
The EtOAc extract was partitioned between hexane and 95:5
MeOH/H2O. The polar fraction (244 mg) was triturated with 50:50
MeOH/H2O. The soluble portion (165 mg) was subjected to a solid-
phase extraction on a C18 reversed-phase cartridge (2 g) using H2O/
MeOH 75:25 (50 mL), 50:50 (50 mL), 25:75 (50 mL) and 100%
MeOH (50 mL), to give fractions DRF2-F1 (36.4 mg), DRF2-F2
(94.7 mg), DRF2-F3 (25.4 mg), and DRF2-F4 (2.3 mg). Fractions
G
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
DRF2-F2 and DRF2-F3 were pooled after TLC analysis (CH2Cl2/
MeOH, 9:1, sprayed with phosphomolybdic acid reagent heated at 100
°C during 2 min). Fraction DRF2-F2/F3 (120.1 mg) was separated by
HPLC, using a reversed-phase C18 CSC-Inertsil ODS2 column (250 ×
94 mm, 150 Å, 5 μm), to give 8.5 mg of 10,11-dehydrocurvularin (2)11
and 1.5 mg of 12-keto-10,11-dehydrocurvularin (3).
427.1049 [M + H]+ (calcd for C19H23O9S 427.10573) and m/z
449.0855 [M + Na]+ (calcd for C19H22O9SNa, 449.08767).
Cyclosulfoxicurvularin methyl ester (11): colorless, glassy solid;
[α]D −52.6 (c 0.015, MeOH); UV (MeOH) λmax (log ε) 209 (4.10),
277 (3.68), 303 (3.60) nm; 1H NMR and 13C NMR (Table 3);
(−)-HRESIMS m/z 439.1056 [M − H]− (calcd for C20H23O9S,
439.1063).
A larger scale growth and metabolic production by Penicillium sp.
DRF2 was performed under improved conditions defined by a refined
version of our experimental design and chemometrics analysis21
(described in detail in the SI). Two optimized growth conditions have
been established. Growth condition #11 (GC#11): 10% of salts
concentration of the artificial seawater (ASW) recipe,21 80% of
nutrients concentration of the standard M3 medium preparation,36
temperature of growth at 15 °C, under shaking during 35 days.
Growth condition #17 (GC#17): 10% of salts concentration of the
artificial seawater (ASW) recipe, 80% of nutrients concentration of the
standard M3 medium preparation, temperature of growth at 15 °C, in
still mode during 35 days. For each of the two optimized growth
conditions, 2 L of M3 culture medium was prepared and inoculated
with Penicillium sp. DRF2. At the end of the growth period, the
medium of each growth experiment was separately processed. After
filtration through a Celite pad, the growth media of both growth
experiments were separately subjected to a solid-phase extraction using
a C18 reversed-phase silica gel column (10 g) eluted with a gradient of
MeOH in H2O. Fractions F1 to F4 (F1 = 75%:25% H2O/MeOH,
discarded; F2 = 50:50 H2O/MeOH; F3 = 25:75 H2O/MeOH; F4 =
100% MeOH, discarded) were dried in vacuo followed by HPLC-UV-
MS analysis. The growth experiment using GC#11 yielded 987 mg of
F2 and 66 mg of F3, while the growth experiment using GC#17
yielded 1.1 g of F2 and 224 mg of F3. Analysis by HPLC-UV-MS of
fractions F2 and F3 obtained from both growth experiments showed
good similarity and were pooled to give 2.377 g of a fraction named
E1117-23. This fraction was subjected to a complex separation
procedure (described in detail in the SI), to give 0.6 mg of 3, 2.6 mg of
4, 4.4 of 6, 60 mg of 7, 3.4 mg of 8, 2.9 mg of 10, and 0.3 mg of 11. In
order to obtain crystals of 6 suitable for X-ray diffraction analysis, a
portion of 6 was further purified by C18 reversed-phase HPLC using an
InertSustain, 5 μm, 25 × 1.0 cm column, with 7:3 (0.05% TFA/H2O)/
MeCN as eluent, and 6 then recrystallized by slow evaporation from
MeOH.
X-ray crystal data for 6: C39H55O22S2, fw 942.66, orthorhombic,
crystal size 0.10 × 0.12 × 0.35 mm3, space group P212121, a =
10.7952(7) Å, b = 17.2819(11) Å, c = 23.5916(13) Å, V = 4401.3(5)
Å3, Z = 4, T = 100(2) K, Dcalcd = 1.483 g/cm3, μ = 2.06 cm−1, F(000) =
1999.0, reflections collected 30 225 (Mo Kα radiation, 2.92° ≤ 2θ ≤
50.81°), independent reflections 8055 (Rint = 0.050, Rsigma = 0.068),
final R indexes for I ≥ 2σ(I), R1 = 0.041, wR2 = 0.094, final R indexes
for all data R1 = 0.056, wR2 = 0.106, the goodness-of-fit on F2 is 1.02,
the Flack parameter37 x is 0.05(5). Crystallographic data for 6 have
been deposited in the Cambridge Crystallographic Data Center
(CCDC) as deposit no. CCDC 1472488. The data can be obtained
free of charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/getstructures.
X-ray crystal data for 7: C22H28O8S, fw 452.50, tetragonal, crystal
size 0.03 × 0.12 × 0.14 mm3, space group P43, a = 10.019(3) Å, b =
10.019(3) Å, c = 42.875(12) Å, V = 4304(3) Å3, Z = 4, T = 90(2) K,
Dcalcd = 1.397 g/cm3, μ = 1.97 cm−1, F(000) = 1920.0, reflections
collected 26 253 (Mo Kα radiation, 1.90° ≤ 2θ ≤ 48.6°), independent
reflections 6912 (Rint = 0.098, Rsigma = 0.115), final R indexes for I ≥
2σ(I), R1 = 0.117, wR2 = 0.304, final R indexes for all data R1 = 0.112,
wR2 = 0.300, the goodness-of-fit on F2 is 1.05, the Flack parameter37 x
is 0.04(7). Crystallographic data for 7 have been deposited in the
Cambridge Crystallographic Data Center (CCDC) as deposit no.
CCDC 1472487. The data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
getstructures.
Preparation of Dimethoxycyclothiocurvularin B Methyl
Ester (9). Cyclothiocurvularin B (7) (2 mg) was treated with
diazomethane that was generated in situ by the addition of 0.7 mL of
2.0 M trimethylsilyldiazomethane in hexanes to 0.7 mL of anhydrous
MeOH in 1.0 mL of C6H6. The reaction mixture was left stirring for 16
h at rt. After evaporation of the reagents the sample was purified by
HPLC using a C18 reversed-phase InertSustain column (5 μm, 25 ×
0.46 cm), with 3:2 MeCN/H2O as eluent, to yield 1.3 mg of
dimethoxycyclothiocurvularin B methyl ester (9).
Dimethoxycyclothiocurvularin B methyl ester (9): colorless plate
crystals; mp 176−180 °C; UV (MeOH) λmax 219, 244, 284, 309 nm;
1H NMR (600 MHz, DMSO-d6) 8.53 (s); 5.99 (s); 5.23 (s); 5.14 (s);
4.99 (bt, 5.3 Hz); 4.02 (s); 3.31 (s); 2.68 (m); 2.13−2.03 (m), 1.90
(bd); 1.87 (bd); 1.76 (m); 1.50 (s); 1.30 (m); 1.11 (m); 1.00 (s); 0.95
(s); 0.70 (d, 7.0 Hz); 13C NMR (150 MHz, DMSO-d6) 198.2, 143.6,
140.6, 136.1, 132.8, 121.6, 115.1, 64.1, 62.4, 41.9, 40.5, 39.8, 37.6, 37.5,
32.6, 32.1, 30.9, 23.3, 21.4, 16.3, 16.2, 13.9; (+)-HRESIMS m/z
475.1400 [M + Na]+ (calcd for C22H28O8SNa, 475.1403).
Administration of Unlabeled and of Stable-Isotope-Labeled
Substrates to Cultures of Penicillium sp. DRF2. In order to verify
the highest concentration of cysteine tolerated by the fungus
Penicillium sp. DRF2 for the [U−13C315N]-L-cysteine administration
experiments, as well as a possible alteration in the cyclothiocurvularin
production in the presence of L-Cys, a series of duplicate growth
experiments were performed with unlabeled L-Cys. These growth
experiments were performed in a total volume of 100 mL of culture
media each, distributed in two Schott flasks inoculated with 5 × 106
spores of Penicillium sp. DRF2, followed by incubation in the GC#17
medium. Unlabeled L-Cys was added in six growth experiments at final
concentrations of 0.5, 1.0, 2.0, 3.0, 4.0, and 5.0 mg/mL, at the sixth day
of microbial growth. After 20 days of incubation, the cultures were
harvested and filtered through Celite. The liquid media of each growth
experiment was subjected to a solid-phase extraction using a C18
reversed-phase silica gel column (2 g) eluted with a gradient of MeOH
in H2O. Fractions F1 to F4 (F1 = 100% H2O, discarded; F2 = 75:25
H2O/MeOH, discarded; F3 = 25:75 H2O/MeOH; F4 = 100% MeOH,
discarded) were dried in vacuo followed by HPLC-UV-MS analysis
12-Keto-10,11-dehydrocurvularin (3): colorless, glassy solid;
[α]25 −7.6 (c 0.05, MeOH); UV (MeOH) λmax (log ε) 212 (4.8),
D
1
298 (3.8), 340 (3.0) nm; H NMR and 13C NMR (Table 1); (+)-
HRESIMS m/z 305.1050 [M + H]+ (calcd for C16H17O6, 305.10196).
cis-10,11-Epoxycurvularin (4): colorless, glassy solid; [α]25D +110.3
(c 0.13, MeOH); UV (MeOH) λmax (log ε) 220 (4.0), 237 (3.7), 273
(3.5), 300 (3.0) nm; 1H NMR and 13C NMR (Table 1);
(+)-HRESIMS m/z 307.11883 [M + H]+ (calcd for C16H19O6,
307.11761) and m/z 329.10063 [M + Na]+ (calcd for C16H18O6Na,
329.09956).
Cyclothiocurvularin A (6): colorless block crystals; mp 193−196
°C; [α]25D −18.2 (c 0.05, MeOH); UV (MeOH) λmax (log ε) λmax 205
(4.16), 275 (3.71), 304 (3.60) nm; 1H NMR and 13C NMR (Table 2);
(−)-HRESIMS m/z 409.0948 [M − H]− (calcd for C19H21O8S,
409.0957).
Cyclothiocurvularin B (7): colorless, glassy solid; [α]25 −14.0 (c
D
0.5, MeOH); UV (MeOH) λmax (log ε) 203 (4.20), 276 (3.70), 306
1
(3.60) nm; H NMR and 13C NMR (Table 2); (+)-HRESIMS m/z
411.11299 [M + H]+ (calcd for C19H23O8S, 411.11081) and m/z
433.09405 [M + Na]+ (calcd for C19H22O8Na, 433.09276).
Cyclothiocurvularin methyl ester (8): colorless, glassy solid; [α]25
D
−22.8 (c 0.17, MeOH); UV (MeOH) λmax (log ε) 204 (4.20), 276
(3.70), 306 (3.60) nm; 1H NMR and 13C NMR (Table 3);
(+)-HRESIMS m/z 425.12865 [M + H]+ (calcd for C20H25O8S,
425.12646) and m/z 447.10943 [M + Na]+ (calcd for C20H24O8SNa,
447.10841).
Cyclosulfoxicurvularin (10): colorless, glassy solid; [α]25D −52.0 (c
0.105, MeOH); UV (MeOH) λmax (log ε) 210 (4.10), 277 (3.68), 302
1
(3.62) nm; H NMR and 13C NMR (Table 3); (+)-HRESIMS m/z
H
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
using a reversed-phase C18 column (X-Terra, 250 × 4.6 mm, 5 μm)
eluted with a MeOH/H2O 0.1% formic acid gradient.
Cytotoxicity Assays. Breast cell lines MCF-10A and MCF-7 were
purchased from National Cancer Institute Cell Bank, Rio de Janeiro,
Brazil. The cervical cancer derived HeLa and human alveolar basal
epithelial cell line A549 were a kind donation from Laboratory of
Small-scale [U−13C315N]-L-cysteine feeding experiments were
performed in a total volume of 150 mL of culture media distributed
in three Schott flasks with 50 mL each. These Schott flasks were
inoculated with 5 × 106 spores of Penicillium sp. DRF2, followed by
incubation in GC#17 medium (SI). [U−13C315N]-L-Cysteine was
added at a final concentration of 2 mM, on the sixth day of fungal
growth. Cultures were harvested after 20 days of growth and filtered
through Celite. The growth media was then subjected to a solid-phase
extraction using a C18 reversed-phase silica gel column (10 g) eluted
with a gradient of MeOH in H2O. Fractions F1 to F4 (F1 = 75%:25%
H2O/MeOH, discarded; F2 = 50:50 H2O/MeOH; F3 = 25:75 H2O/
MeOH; F4 = 100% MeOH, discarded) were dried in vacuo followed
by HPLC-UV-MS analysis using a reversed-phase C18 analytical
column (X-Terra, 250 × 4.6 mm, 5 μm) eluted with a MeOH/H2O
0.1% formic acid gradient.
Cytogenetics and Mutagenesis, Medical School, Ribeirao Preto, SP,
̃
Brazil. All cell lines were expanded and cultured in complete medium
composed of DMEM + HAM F10 (1:1, v/v) (Sigma-Aldrich)
supplemented with 10% fetal calf serum (Sigma, St Louis, MO, USA)
and 1% penicillin/streptomycin stabilized solution (Sigma-Aldrich) in
cell culture flasks of 25 mm2 (TPP, Swizerland). Cells were incubated
at 37 °C and 5% CO2 in a humidified atmosphere. Experiments were
performed between the third and eighth passage. The IC50 was
determined with XTT metabolic using the Cell Proliferation Kit II
(Roche) following the manufacturer’s instructions. Briefly, cells were
trypsinized and cultured at 1 × 104 cells/well in a 96-well plate. After
24 h cells were treated with 3, 4, 7, and 8 (7.8, 15.6, 31.2, 62.5, 125,
250, 500, and 1000 μM) for an additional 24 h, after which the culture
medium was removed and cells were incubated with DMEM medium
without phenol red plus XTT reagent for 4 h. Absorbance was
measured at 492 nm with a reference at 620 nm, using a microplate
Large-scale [U−13C315N]-L-cysteine feeding experiments were
performed using a total volume of 0.5 L of culture media distributed
in five Schott flasks with 100 mL each, inoculated with 107 spores of
Penicillium sp. DRF2, followed by incubation in GC#17 medium using
the same concentration of [U−13C315N]-L-cysteine. Cultures were
harvested after 20 days of growth and filtered through Celite. At the
end of incubation, the liquid media were subjected to a solid-phase
extraction using a C18 reversed-phase silica gel column (10 g) eluted
with a gradient of MeOH in H2O. Fractions F1 to F4 (F1 = 75%:25%
H2O/MeOH, discarded; F2 = 50:50 H2O/MeOH; F3 = 25:75 H2O/
MeOH; F4 = 100% MeOH, discarded) were dried in vacuo followed
by HPLC-UV-MS analysis. A total amount of 265.9 mg of the crude
F2 fraction was obtained. The F2 fraction was purified by HPLC-UV
using a reversed-phase C8 preparative column (Inertsil C8-4, 250 × 14
mm, 5 μm) eluted with 66% H2O and 34% MeCN. This separation
yielded four new fractions, of which F2-D corresponded to compound
2 (1.0 mg). Fraction F2-A was purified by HPLC using a reversed-
phase C18 semipreparative column (Inertsil ODS2, 250 × 9.4 mm, 5
μm) eluted with 50% H2O, 33% MeOH, and 17% MeCN.
Compounds 6 (2.7 mg) and 7 (9.3 mg) were obtained from the
[U−13C315N]-L-cysteine incorporation experiment. Incorporation
ratios have been calculated as previously described.38
reader (Sunrise, Tecan, Manndorf, Switzerland). The IC50 values were
̈
calculated using GraphPad Prism 5.0. Doxorrubicin was used as
positive control for A549 (lung cancer, 0.19 μM), HeLa (cervical
cancer, 0.14 μM), MCF-7 (breast cancer, 0.12 μM), and MCF10A
(normal breast cells, 0.4 μM).
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.6b00295.
Procedures of chemical and biological screening,
experimental design, and chemometric analysis, con-
1
version of L-cysteine into mercaptopyruvate, H and 13C
NMR data of compounds 3, 4, 6−11, and 13C-labeled 7.
(PDF)
Experimental details for the X-ray diffraction analysis for
compounds 6 and 9. (PDF)
Crystallographic data (CIF)
[13C-17,13C-18,13C-19 ]-Cyclothiocurvularin B (7): colorless, glassy
solid; 13C NMR (100 MHz, MeOH-d4) δ 176.0 (d, 60.4, C-19), 87.4
(dd, 60.6 and 38.0), 40.8 (dd, 38.1 and 2.0, C-17); (+)-HRESIMS m/z
414.1113 [M + H]+ (calcd for C1613C3H23O8S 411.3989). The
measured HRMS value results from the 99.9% incorporation ratio of
[U−13C,15N]-L-cysteine into cyclothiocurvularin B.
Crystallographic data (CIF)
AUTHOR INFORMATION
Corresponding Author
*Tel: +55-16-33739954. Fax: +55-16-33739952. E-mail:
rgsberlinck@iqsc.usp.br.
■
Conversion of 10,11-Dehydrocurvularin (2) into Cyclo-
thiocurvularins A (6) and B (7). A mixture of triethanolamine
hydrochloride (0.1 g, 100 mmol/L), acetone (75 μL, 100 mmol/L),
and pyridoxalphosphate (1.3 mg, 1 mmol/L) was added to H2O (4
mL). The pH of the mixture was adjusted to 7.5 with NaOH (5 mol/
L), and the volume was adjusted to 5 mL with H2O. Then, 450 μL of
this mixture was transferred to a 2 mL Eppendorf flask containing 2
mg of transaminase ATA-113 [from Codex ATA screening kit
(ATASK 201000P, lot no. N13012) produced by Codexis, Inc.
(Redwood City, CA, USA)]. L-Cysteine hydrochloride (50 μL, 70
mmol/L in H2O) and 10,11-dehydrocurvularin (2, 7 mmol/L in
DMSO, 50 μL) were separately added to the reaction mixture, which
was kept at 30 °C and 200 rpm for 48 h. Reaction controls in the
absence of enzyme were also developed in order to evaluate the
substrate stability in the reaction mixture. The enzymatic reaction was
quenched with MeOH (500 μL) and centrifuged (× rpm, × min). The
reaction supernatant was diluted in MeOH (100×) and analyzed by
UPLC-qTOF. UPLC-qTOF analysis conditions: column, Acquity
UPLC BEH C18 (1.7 μm, 2.1 × 50 mm); eluent, gradient of MeCN (+
0.01% HCO2H) in H2O (+ 0.01% HCO2H), starting at 10% to 100%
MeCN (+ 0.01% HCO2H) in 4 min. Detection: MSE continuum
during 5 min, m/z 250−450 molecular weight range; detection mode,
ESI(+); scan time, 0.2 s.; collision energy, 3 V; ramp collision energy,
20−30 V.
Present Address
¶Unit of Molecular Biology of the Gene in Extremophiles,
Department of Microbiology, Institut Pasteur, 75015 Paris,
France.
Author Contributions
#M. V. de Castro and L. P. Ioc
́
a contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Prof. A. J. Marsaioli (Universidade Estadual
de Campinas) for assistance and providing reagents for
biocatalytic reactions, as well as for fruitful discussions. The
authors also thank Prof. J. C. Vederas (University of Alberta)
and Drs. S. Romminger, S. Newminster, and D. Hansen (Life
Sciences Institute, University of Michigan) for fruitful
discussions. Profs. D. R. Cardoso and A. J. dos Santos Neto
(IQSC, USP) are acknowledged for assistance with the HRMS
measurements. Financial support was provided by FAPESP
I
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Debbab, A.; Clements, C.; Edrada-Ebel, R.; Orlikova, B.; Diederich,
M.; Wray, V.; Lin, W. H.; Proksch, P. Bioorg. Med. Chem. 2011, 19,
414−421.
BIOTA/BIOprospecTA grants 2005/60175-2, 2010/50190-2
and 2013/50228-8 and the FAPESP/NSF ICC grant 2012/
50026-3 to R.G.S.B., as well as an FAPESP scholarship to
C.M.M. and an NSERC grant to R.J.A. The authors also thank
the Brazilian federal funding agencies CAPES and CNPq for
the Ph.D. scholarships awarded to M.V.C., K.J., L.P.I., E.F.F.F.,
and M.F.C.S.
(16) Shen, W.; Mao, H.; Huang, Q.; Dong, J. Eur. J. Med. Chem.
2015, 97, 747−777.
(17) Yao, Y.; Hausding, M.; Erkel, G.; Anke, T.; Forstermann, U.;
Kleinert, H. Mol. Pharm. 2003, 63, 383−391.
(18) (a) Xu, Y.; Espinosa-Artiles, P.; Schubert, V.; Xu, Y.-M.; Zhang,
́
W.; Lin, M.; Gunatilaka, A. A. L.; Sussmuth, R.; Molnar, I. Appl.
̈
Environm. Microbiol. 2013, 79, 2038−2047. (b) Xua, Y.; Zhou, T.;
REFERENCES
(1) Ousborn, A. Trends Genet. 2010, 26, 449−457.
(2) Cox, R. J.; Simpson, T. J. In Comprehensive Natural Products
Chemistry; Meth-Cohn, O., Barton, S. D., Nakanishi, K., Eds.; Elsevier:
Amsterdam, 1999; Vol. I, pp 347−383.
(3) Hanson, J. R. The Chemistry of Fungi; RSC Publishing:
Cambridge, UK, 2008; pp 47−72.
(4) Cox, R. J. Org. Biomol. Chem. 2007, 5, 2010−2026.
(5) Xu, J.; Jiang, C.-S.; Zhang, Z.-L.; Ma, W.-Q.; Guo, Y.-W. Acta
Pharmacol. Sin. 2014, 35, 316−330.
■
Zhou, Z.; Sua, S.; Roberts, S. A.; Montfort, W. R.; Zeng, J.; Chen, M.;
Zhang, W.; Lin, M.; Zhanb, J.; Molnar
2013, 110, 5398−5403.
(19) Ferreira, E. L. F.; Williams, D. E.; Ioc
́
, I. Proc. Natl. Acad. Sci. U. S. A.
́
a, L. P.; Morais-Urano, R.
P.; Santos, M. F. C.; Patrick, B. O.; Elias, L. M.; Lira, S. P.; Ferreira, A.
G.; Passarini, M. Z. R.; Sette, L. D.; Andersen, R. J.; Berlinck, R. G. S.
Org. Lett. 2015, 17, 5152−5155.
(20) Mercado-Marin, E.; Garcia-Reynaga, P.; Romminger, S.;
Pimenta, E. F.; Romney, D.; Lodewyk, M.; Williams, D. E.;
Andersen, R. J.; Miller, S.; Tantillo, D.; Berlinck, R. G. S.; Sarpong,
R. Nature 2014, 509, 318−324.
(21) Pimenta, E. F.; Vita-Marques, A. M.; Tininis, A.; Seleghim, M.
H. R.; Sette, L. D.; Veloso, K.; Ferreira, A. G.; Williams, D. E.; Patrick,
B. O.; Dalisay, D. S.; Andersen, R. J.; Berlinck, R. G. S. J. Nat. Prod.
2010, 73, 1821−1832.
(22) Passarini, M. R. Z.; Santos, C.; Lima, N.; Berlinck, R. G. S.;
Sette, L. D. Arch. Microbiol. 2013, 195, 99−111.
(23) (a) Birch, A. J.; Musgrave, O. C.; Rickards, R. W.; Smith, H. J.
Chem. Soc. 1959, 3146−3152. (b) Arai, K.; Rawlings, B. J.; Yoshizawa,
Y.; Vederas, J. C. J. Am. Chem. Soc. 1989, 111, 3391−3399. (c) Liu, Y.
Q.; Li, Z.; Vederas, J. C. Tetrahedron 1998, 54, 15937−15958.
(24) Lai, S.; Shizuri, Y.; Yamamura, S.; Kawai, K.; Niwa, M.;
Furukawa, H. Heterocycles 1991, 32, 307−310.
(25) Lai, S.; Shizuri, Y.; Yamamura, S.; Kawai, K.; Terada, Y.;
Furukawa, H. Chem. Lett. 1990, 589−592.
(26) Pavia, D. L.; Lampman, G. M.; Kriz, G. S.; Vyvyan, J. R.
Introduction to Spectroscopy, 4th ed.; Brooks/Cole, Cengage Learning:
Belmont, CA, p 243.
(6) (a) Musgrave, O. C. J. Chem. Soc. 1956, 4301−4305.
(b) Musgrave, O. C. J. Chem. Soc. 1957, 1104−1108. (c) Birch, A.
J.; Rickards, R. W.; Moore, B. J. Chem. Soc. 1962, 220−222.
(d) Munro, H. D.; Musgrave, O. C.; Templeton, R. J. Chem. Soc. C
1967, 0, 947−948. (e) Musgrave, O. C.; Templeton, R.; Munro, H. D.
J. Chem. Soc. C 1968, 250−255. (f) Munro, H. D.; Musgrave, O. C.;
Templeton, R. J. Chem. Soc. C 1971, 95−98. (g) Ghisalberti, E. L.;
Hockless, D. C. R.; Rowland, C. Y.; White, A. H. Aust. J. Chem. 1993,
46, 571−575.
(7) Kobayashi, A.; Hino, T.; Yata, S.; Itoh, T. J.; Sato, H.; Kawazu, K.
Agric. Biol. Chem. 1988, 52, 3119−3123.
(8) (a) Robeson, D. J.; Strobel, G. A. Z. Naturforsch. C 1981, 1081−
1083. (b) Vurro, M.; Evidente, A.; Andolfi, A.; Zonno, M. C.;
Giordano, F.; Motta, A. Plant Sci. 1998, 138, 67−79. (c) Jeon, Y. T.;
Ryu, K. H.; Kang, M. K.; Park, S. H.; Yun, H.; Pham, Q.; Kim, S. U.
Han'guk Eungyong Sangmyong Hwahakhoeji 2010, 53, 39−42.
(9) Kusano, M.; Nakagami, K.; Fujioka, S.; Kawano, T.; Shimada, A.;
Kimura, Y. Biosci., Biotechnol., Biochem. 2003, 67, 1413−1416.
(10) He, J.; Wijeratne, E. M. K.; Bashyal, B. P.; Zhan, J. X.; Seliga, C.
J.; Liu, M. P. X.; Pierson, E. E.; Pierson, L. S.; VanEtten, H. D.;
Gunatilaka, A. A. L. J. Nat. Prod. 2004, 67, 1985−1991.
(11) (a) Ghisalberti, E. L.; Rowland, C. Y. J. Nat. Prod. 1993, 56,
2175−2177. (b) Almassi, F.; Ghisalberti, E. L.; Skelton, B. W.; White,
A. H. Aust. J. Chem. 1994, 47, 1193−1197.
(27) Smith, C. J.; Abbanat, D.; Bernan, V. S.; Maiese, W. M.;
Greenstein, M.; Jompa, J.; Tahir, A.; Ireland, C. M. J. Nat. Prod. 2000,
63, 142−145.
(28) Jadulco, R.; Proksch, P.; Wray, V.; Sudarsono; Berg, A.; Grafe,
̈
U. J. Nat. Prod. 2001, 64, 527−530.
(29) Meng, L.-H.; Li, X.-M.; Lv, C. T.; Li, C. S.; Xu, G. M.; Huang, C.
G.; Wang, B. G. J. Nat. Prod. 2013, 76, 2145−2149.
(30) Yokota, T.; Yamazaki, S.; Takahashi, N.; Iitaka, Y. Tetrahedron
Lett. 1974, 15, 2957−2960.
(12) Ebrahim, W.; Aly, A. H.; Man
G.; Wray, V.; Lin, W. H.; Dai, H.; Proksch, P.; Kurtan
Eur. J. Org. Chem. 2012, 2012, 3476−3484.
(13) (a) Bicalho, B.; Goncalves, R. A. C.; Zibordi, A. P. M.; Manfio,
G. P.; Marsaioli, A. J. Z. Naturforsch., C: J. Biosci. 2003, 58, 746−751.
(b) Greve, H.; Schupp, P. J.; Eguereva, E.; Kehraus, S.; Kelter, G.;
́
di, A.; Totzke, F.; Kubbutat, M. H.
́
, T.; Debbab, A.
̧
(31) (a) Adelin, E.; Martin, M.-T.; Bricot, M.-F.; Cortial, S.;
Retailleau, P.; Ouazzani, J. Phytochemistry 2012, 84, 135−140.
(b) Castillo-Lozano, M. L.; Delile, A.; Spinnler, H. E.; Bonnarme,
P.; Landaud, S. Appl. Microbiol. Biotechnol. 2007, 75, 1447−1454.
(c) Cooper, A. J. L. Annu. Rev. Biochem. 1983, 52, 187−222.
(32) (a) Paszewski, A. Acta Biochim. Polonica 1993, 40, 445−449.
(b) Grynberg, M.; Topczewski, J.; Godzik, A.; Paszewski, A.
Microbiology 2000, 146, 2695−2703.
Maier, A.; Fiebig, H. H.; Konig, G. M. Eur. J. Org. Chem. 2008, 2008,
̈
5085−5092. (c) Dai, J.; Krohn, K.; Florke, U.; Pescitelli, G.; Kerti, G.;
̈
Papp, T.; Kover
T. Eur. J. Org. Chem. 2010, 2010, 6928−6937.
́
, K. E.; Ben
́ ́
yei, A. C.; Draeger, S.; Schulz, B.; Kurtan,
̈
(14) Xie, L. W.; Ouyang, Y. C.; Zou, K.; Wang, G. H.; Chen, M. J.;
Sun, H. M.; Dai, S. K.; Li, X. Appl. Biochem. Biotechnol. 2009, 159,
284−29.
(33) Li, C.-S.; Ding, Y.; Yang, B.-J.; Miklossy, G.; Yin, H.-Q.; Walker,
L. A.; Turkson, J.; Cao, S. Org. Lett. 2015, 17, 3556−3559.
(15) (a) Fennell, D.; Raper, K. B.; Stodola, F. H. Chem. Ind. 1959,
1382−1383. (b) Raistrick, H.; Rice, F. A. H. J. Chem. Soc. C 1971,
3069−3070. (c) Vesonder, R. F.; Ciegler, A.; Fennell, D.; Tjarks, L.
W.; Jensen, A. H. J. Environ. Sci. Health, Part B 1976, 11, 289−297.
(d) Kobayashi, A.; Hino, T.; Yata, S.; Itoh, T. J.; Sato, H.; Kawazu, K.
Agric. Biol. Chem. 1988, 52, 3119−3123. (e) Lai, S.; Shizuri, Y.;
Yamamura, S.; Kawai, K.; Terada, Y.; Furukawa, H. Tetrahedron Lett.
1989, 30, 2241−2244. (f) Lai, S.; Shizuri, Y.; Yamamura, S.; Kawai, K.;
Furukawa, H. Bull. Chem. Soc. Jpn. 1991, 64, 1048−1050.
(g) Malmstrom, J.; Christophersen, C.; Frisvad, J. C. Phytochemistry
2000, 54, 301−309. (h) Zhan, J.; Wijeratne, E. M. K.; Seliga, C. J.;
Zhang, J.; Pierson, E. E.; Pierson, L. S., III; Vanetten, H. D.;
Gunatilaka, A. A. L. J. Antibiot. 2004, 57, 341−344. (i) Aly, A. H.;
́
(34) Xu, Y.; Zhou, T.; Zhang, S.; Xuan, L.-J.; Zhan, J.; Molnar, I. J.
Am. Chem. Soc. 2013, 135, 10783−10791.
(35) Capriles, C. H.; Mata, S.; Middelveen, M. Mycopathologia 1989,
106, 73−79.
(36) Kossuga, M. H.; Romminger, S.; Xavier, C.; Milanetto, M. C.;
Valle, M. Z.; Pimenta, E. F.; Morais, R. P.; Carvalho, E.; Mizuno, C.
M.; Coradello, L. F. C.; Barroso, V. M.; Vacondio, B.; Seleghim, M. H.
R.; Cavalcanti, B. C.; Pessoa, C.; Moraes, M. O.; Lima, B. A.;
Gonca̧ lves, R.; Bonugli-Santos, R. C.; Sette, L. D.; Berlinck, R. G. S.
Rev. Bras. Farmacogn. 2012, 22, 257−267.
(37) Parsons, S.; Flack, H. D.; Wagner, T. Acta Crystallogr., Sect. B:
Struct. Sci., Cryst. Eng. Mater. 2013, B69, 249−259.
J
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(38) Kubanek, J.; Andersen, R. J. J. Nat. Prod. 1999, 62, 777−779.
K
DOI: 10.1021/acs.jnatprod.6b00295
J. Nat. Prod. XXXX, XXX, XXX−XXX