Varioxiranols A-G and 19-O-Methyl-22-methoxypre-shamixanthone, PKS and Hybrid PKS-Derived Metabolites from a Sponge-Associated Emericella variecolor Fungus

Article abstract of DOI:10.1021/acs.jnatprod.5b00578

Chemical examination of a sponge (Cinachyrella sp.)-associated Emericella variecolor fungus resulted in the isolation of seven new polyketide derivatives, namely, varioxiranols A-G (1-7), and a new hybrid PKS-isoprenoid metabolite, 19-O-methyl-22-methoxypre-shamixanthone (8), together with nine known analogues. Their structures were elucidated on the basis of extensive spectroscopic analyses, including ECD effects, Mosher's method, X-ray diffraction, and chemical conversion for the determination of absolute configurations. Varioxiranols F and G were found for the first time to link a xanthone moiety with a benzyl alcohol via an ether bond, while the dioxolanone group of 5 is unusual in nature. A cell-based lipid-lowering assay revealed that pre-shamixanthone (12) exerted significant inhibition against lipid accumulation in HepG2 cells without cytotoxic effects, accompanying the potent reduction of total cholesterol and triglycerides. Real-time quantitative PCR indicated that pre-shamixanthone (12) mediated the reduction of lipid accumulation related to the down-regulation of the expression of the key lipogenic transcriptional factor SREBP-1c and its downstream genes encoding FAS and ACC.

Full text of DOI:10.1021/acs.jnatprod.5b00578

Article
pubs.acs.org/jnp
Varioxiranols A−G and 19‑O‑Methyl-22-methoxypre-shamixanthone,
PKS and Hybrid PKS-Derived Metabolites from a Sponge-Associated
Emericella variecolor Fungus
Qi Wu,† Chongming Wu,‡ Hailin Long,† Ran Chen,‡ Dong Liu,† Peter Proksch,§ Peng Guo,*,‡
and Wenhan Lin*,†
†State Key Laboratory of Natural and Biomimetic Drugs, Peking University, Beijing 100191, People’s Republic of China
‡Pharmacology and Toxicology Research Center, Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences,
Peking Union Medical College, Beijing 100193, People’s Republic of China
§Institute of Pharmaceutical Biology and Biotechnology, Heinrich-Heine University, 40225 Duesseldorf, Germany
S
* Supporting Information
ABSTRACT: Chemical examination of a sponge (Cinachyr-
ella sp.)-associated Emericella variecolor fungus resulted in the
isolation of seven new polyketide derivatives, namely,
varioxiranols A−G (1−7), and a new hybrid PKS-isoprenoid
metabolite, 19-O-methyl-22-methoxypre-shamixanthone (8),
together with nine known analogues. Their structures were
elucidated on the basis of extensive spectroscopic analyses,
including ECD effects, Mosher’s method, X-ray diffraction, and
chemical conversion for the determination of absolute
configurations. Varioxiranols F and G were found for the
first time to link a xanthone moiety with a benzyl alcohol via
an ether bond, while the dioxolanone group of 5 is unusual in nature. A cell-based lipid-lowering assay revealed that pre-
shamixanthone (12) exerted significant inhibition against lipid accumulation in HepG2 cells without cytotoxic effects,
accompanying the potent reduction of total cholesterol and triglycerides. Real-time quantitative PCR indicated that pre-
shamixanthone (12) mediated the reduction of lipid accumulation related to the down-regulation of the expression of the key
lipogenic transcriptional factor SREBP-1c and its downstream genes encoding FAS and ACC.
clusters and that ecological impacts may induce structural
modifications of the secondary metabolites.15,16
he fungus Emericella variecolor, a teleomorph of Aspergillus
Tvariecolor (synonym of A. stellatus and A. stellifer), is
widely distributed in marine and terrestrial environments.1
Previous chemical investigation of E. variecolor uncovered a
variety of natural products with diverse and unusual scaffolds. A
strain of E. variecolor derived from a Caribbean sponge
produced benzyl alcohols (varitriol, varioxirane), prenylxan-
thones (varixanthone, shamixanthone, tajixanthone hydrate),
and cyclopentanones,2 while the same species associated with
the sponge Haliclona valliculata generated anthraquinones,
prenylxanthones (isoemericellin, shamixanthone), and strome-
mycin.3 In addition, a strain from marine sediments afforded
sesterterpenes with various scaffolds (ophiobolins,4 astellatol,5
andilesin A,6 variecolin7) and the unique polyketide shima-
lactone A.8 A UV-induced mutant strain of A. variecolor
produced meroterpenoid metabolites such as andibenin B.9
Alternatively, a strain of A. variecolor from terrestrial locations
generated the secondary metabolites, which differed from those
produced in marine-derived strains, as exemplified by the
presence of isoechinulin-type alkaloids,10 quinine-related
derivatives,11 xanthones,12 and others.13,14 These findings
indicate that E. variecolor possesses diverse biosynthesis gene
In our program aimed at discovery of bioactive natural
products from sponge-derived microorganisms, the fungus E.
variecolor XSA-07-2, isolated from a Cinachyrella sp. sponge in
the South China Sea, showed a remarkable diversity of
secondary metabolites in a preliminary HPLC and NMR
screening and was therefore subjected to detailed chemical
analysis.
RESULTS AND DISCUSSION
■
Chromatographic separation of the EtOAc extract of E.
variecolor XSA-07-2 cultured on solid rice medium, including
semipreparative HPLC purification, resulted in the isolation of
17 derivatives, of which eight are new compounds.
Varioxiranol A (1) had a molecular formula of C15H22O4 as
deduced by HRESIMS and NMR data. The IR absorption
suggested the presence of a hydroxy group (3251 cm−1). The
1H NMR spectrum exhibited aromatic signals for an ABC spin
system at δH 6.87 (d, J = 8.0 Hz, H-3), 7.22 (t, J = 8.0 Hz, H-4),
Received: June 30, 2015
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
and 7.10 (d, J = 8.0 Hz, H-5), indicating the presence of a
trisubstituted aromatic ring. The substitutions of a hydroxy-
methylene (δH 4.54 d, J = 5.2 Hz, H2-7) and a methoxy group
(δC 56.1, δH 3.78, s) at C-1 (δC 127.0) and C-2 (δC 158.0),
respectively, were assigned by the HMBC interactions of the
aromatic carbons with the protons of the substituted groups. In
addition, COSY correlations established a linear side chain from
C-8 to C-14, in which a double bond with E geometry (JH‑8/H‑9
= 16 Hz) resided at C-8 (δC 127.5) and C-9 (δC 134.1).
Additional COSY correlations of the oxymethine H-10 (δH
3.95, ddd, J = 6.0, 5.2, 4.0 Hz) and H-11 (δH 3.39, ddt, J = 5.6,
5.2, 4.0 Hz) with the D2O exchangeable protons (δH 4.84 and
4.42), respectively, confirmed C-10 (δC 75.6) and C-11 (δC
74.0) to be hydroxylated. Thus, the side chain was identified as
an (E)-hept-1-ene-3,4-diol segment, which was linked to C-6
(δC 138.9) of the aromatic ring, based on the HMBC
interaction of H-8 (δH 6.94, d, J = 16 Hz) with C-1, C-5 (δC
118.5) and C-6 (δC 138.9). The relative configuration of C-10
and C-11 was determined by the 3JH,H value in association with
the NOE interactions.17,18 The JH‑10/H‑11 value (4.0 Hz) was
3
indicative of a gauche relationship of H-10 and H-11. The NOE
Figure 1. Key NOE interactions of compounds 1 and 2.
B
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 2. ΔδRS (δR − δS) values (in ppm) for the MPA esters of 1a.
ab
,
1
Table 1. H and 13C NMR Data of 1−5
1
2
3
4
5
no.
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
1
2
3
4
5
6
7
8
127.0, C
127.0, C
127.0, C
127.0, C
127.3, C
158.0, C
158.0, C
158.0, C
158.0, C
157.9, C
110.1, CH
128.8, CH
118.5, CH
138.9, C
6.87, d (8.0)
7.22, t (8.0)
7.10, d (8.0)
110.1, CH
128.8, CH
118.5, CH
138.9, C
6.92, d (8.0)
7.23, t (8.0)
7.19, d (8.0)
110.1, CH
128.8, CH
118.4, CH
138.9, C
6.87, d (8.0)
7.23, t (8.0)
7.10, d (8.0)
110.2, CH
128.9, CH
118.5, CH
138.7, C
6.88, d (8.0)
7.23, t (8.0)
7.10, d (8.0)
110.6, CH
129.0, CH
118.4, CH
138.1, C
6.90, d (8.0)
7.23, t (8.0)
7.06, d (8.0)
54.0, CH2 4.54, d (5.2)
53.9, CH2 4.55, d (5.3)
53.9, CH2 4.55, s
53.9, CH2 4.56, d (5.1)
53.8, CH2 4.55, s
127.5, CH
134.1, CH
75.6, CH
74.0, CH
6.94, d (16.0)
127.9, CH
133.0, CH
73.4, CH
76.1, CH
75.3, CH
67.8, CH
6.95, d (16.0)
128.0, CH
132.9, CH
73.5, CH
74.9, CH
72.1, CH
77.3, CH
6.95, d
(16.0)
128.4, CH
132.3, CH
73.4, CH
75.2, CH
75.2, CH
61.3, CH
6.95, d (16.0)
129.4, CH
129.7, CH
70.7, CH
78.3, CH
79.6, CH
65.8, CH
7.15, d
(16.0)
9
6.27, dd (16.0,
6.0)
6.27, dd (16.0,
7.2)
6.26, dd
(16.0, 7.2)
6.24, dd (16.0,
7.6)
6.11, dd
(16.0, 4.8)
10
11
12
13
3.95, ddd (6.0,
5.2, 4.0)
4.34, ddd (7.2,
5.0, 4.0)
4.34, dd
(7.2, 4.0)
4.34, ddd (7.6,
5.1, 4.0)
4.54, dd
(4.8, 3.0)
3.39, ddt (5.6,
5.2, 4.0)
3.40, ddd (8.8,
5.0, 4.0)
3.36, dd
(8.8, 4.0)
3.58, ddd (8.4,
5.9, 4.0)
4.67, dd
(3.6, 3.0)
35.3, CH2 1.30, m;
1.48, m
3.30, ddd (8.8,
5.0, 4.0)
3.59, dd
(8.8, 3.2)
3.41, ddd (8.4,
5.9, 3.6)
4.44, dd
(3.6, 3.0)
19.0, CH2 1.30, m;
1.48, m
3.80, ddq (6.3,
5.0, 4.0)
3.52, qd
(6.3, 3.2)
4.48, dq (6.8,
3.6)
3.84, dq
(6.5, 3.6)
14
15
14.6, CH3 0.88, t (6.9)
17.9, CH3 1.04, d (6.3)
13.0, CH3 1.04, d (6.3)
17.9, CH3 1.39, d (6.8)
18.3, CH3 0.92, d (6.5)
155.3, C
4.55, br
5.89, br
3.50, br
7-OH
4.64, t (5.2)
4.84, d (5.2)
4.42, d (5.6)
4.63, t (5.3)
4.94, d (5.0)
4.65, t (5.1)
5.00, d (5.1)
4.90, d (5.9)
5.23, d (5.9)
10-OH
11-OH
12-OH
13-OH
2-OMe
4.73, d (5.0)
4.54, d (5.0)
4.42, d (,5.0)
5.31, brs
56.1, CH3 3.78, s
56.1, CH3 3.78, s
56.1, CH3 3.77, s
55.8, CH3 3.20, s
56.1, CH3 3.78, s
56.1, CH3 3.77 s
13-
OMe
a
b
Chemical shifts are in ppm; J values in Hz are in parentheses. Measured at 400 MHz in DMSO-d6.
interactions between H-9/H-12, H-10/H-12, and H-9/OH-11
(Figure 1) in association with the absence of an NOE
interaction between H-10/OH-11 and OH-10/H-12 suggested
the relative configuration at the side chain of 1 to be 10R* and
11S*, which were the same as those of the known analogue
(1E,5E)-1-(2-(hydroxymethyl)-3-methoxyphenyl)hepta-1,5-
diene-3,4-diol (9).19 The absolute configuration of 1 was
determined on the basis of the modified Mosher’s method used
for the determination of the configuration of the acyclic 1,2-
diol.20 The hydroxy group OH-7 was first protected by TBSCl
to yield the analogue 1a. Treatment of 1a with (R)- and (S)-
MPA, respectively, afforded (R)- and (S)-MPA diesters (1b and
1c). Diagnostic ΔδRS (δ(R)‑MPA ester − δ(S)‑MPA ester) data (Figure
2) resulted in positive ΔδRS values for H-9 and H-10, whereas
H-11 and H-12 showed negative ΔδRS values. These data were
in accordance with 10R and 11S configurations, which were the
same as those of 9, whose enantiomer was previously obtained
using total synthesis.19
same backbone. The significant differences were attributed to
the side chain, where two additional oxymethines were present.
The COSY and HMQC cross-peaks assigned the oxymethines
to C-12 (δC 75.3) and C-13 (δC 67.8), while the COSY
relationships between H-12 (δH 3.30)/OH-12 (δH 4.54 d, J =
5.0 Hz) and H-13 (δH 3.80)/OH-13 (δH 4.42 d, J = 5.0 Hz)
allowed the substitution of OH groups at C-12 and C-13,
respectively. The HMBC interaction from H3-14 (δH 1.04, d, J
= 6.3 Hz) to C-12 and C-13 further supported the assignments.
The JH‑8/H‑9 value (16.0 Hz) was indicative of the E geometry of
the double bond. The coupling constants of 3JH‑10/H‑11 (4.0 Hz),
3JH‑11/H‑12 (8.8 Hz), and 3JH‑12/H‑13 (4.0 Hz) reflected the gauche
relationships between H-10/H-11 and H-12/H-13, as well as
the anti-orientation for H-11/H-12. These data suggested the
side chain of 2 adopted a single dominant conformer.18 The
NOE interactions in association with the 3JH,H values helped to
assign the relative configuration of the side chain. In the 1D
NOE experiment, irradiation of H-10 induced the NOE
enhancement of H-8, H-11, H-12, and OH-12 (Figure 1b);
irradiation of H-11 resulted in the NOE enhancement of OH-
12, H-13, CH3-14, and OH-12 (Figure 1c); in addition the
Varioxiranol B (2) had a molecular formula of C15H22O6
according to its HRESIMS and NMR data. The NMR data of 2
(Table 1) were comparable to those of 1, while 2D NMR
spectroscopic analyses established both 1 and 2 contained the
C
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
NOE enhancements of H-10, H-13, and H3-14 (Figure 1d)
were observed when H-12 was irradiated in the 1D NOE
experiment. These findings in association with the J values
assigned the relative configuration to be 10R*, 11R*, 12S*, and
13S*. Acidic hydrolysis21 of varioxirane (11) yielded a product
(Figure 3) whose NMR data and specific rotation were
Varioxiranol F (6) had a molecular formula of C40H48O10,
which was established by the HRESIMS data. Analyses of 1D
and 2D NMR data (Table 2) revealed the structure of 6 to
consist of two moieties. Moiety A was identical to compound 1
based on the comparison of their NMR spectroscopic data. In
regard to moiety B, its NMR data including 12 aromatic
carbons and a carbonyl carbon at δC 184.5 (C-13) were
characteristic of a xanthone nucleus, structurally related to
tajixanthone hydrate,12 also isolated from the same fraction.
The COSY correlations of H-20 (δH 2.76, ddd, J = 2.8,2.9,3.0
Hz) with H2-21 (δH 4.44, 4.36) and H-19 (δH 5.43, brd, J = 2.8
Hz), in association with the HMBC interactions from H-19 to
the aromatic carbons C-7 (δC 149.5), C-8 (δC 121.1), and C-12
(δC 116.9) and between H2-21 and C-7, showed the fusion of a
tetrahydropyran ring to the aromatic ring at C-7 and C-8. An
isopropene unit linked to C-20 (δC 44.9) and a hydroxy group
bonded to C-19 (δC 63.2) were defined by the HMBC
correlations of olefinic methylene H2-23 (δH 4.60, 4.82) and the
methyl protons H3-24 (δH 1.87, s) to C-21 and C-20, in
addition to the COSY correlation between H-19 and OH-19.
Moreover, a 15,16-dioxygenated isopentane unit was recog-
nized by the COSY correlation between H2-14 (δH 3.11, 2.72)
and H-15 (δH 3.92, dd, J = 10.0, 2.0 Hz), as well as the HMBC
interactions from the methyl protons H3-17 (δH 1.43, s) and
H3-18 (δH 1.50, s) to C-16 (δC 78.0) and C-15 (δC 76.6). This
unit was positioned at the aromatic carbon C-4 (δC 116.9)
according to the HMBC interactions from H2-14 to C-3 (δC
138.2), C-4, and C-10 (δC 153.0) and between H-3 (δH 7.60, d,
J = 8.4 Hz) and C-14 (δC 31.1). The remaining OH (δH 12.64,
s) and a methyl unit (δH 2.36, s) were located at C-1 (δC 160.2)
and C-6 (δC 138.3), respectively, on the basis of their HMBC
interactions (Figure 5). The NOE interaction between H3-24
and H-19 indicated the same orientation of H-19 and the
isopropene unit. Accordingly, moiety B was identical to
tajixanthone hydrate.12 The HMBC interaction between H2-
7′ (δH 4.58, 4.74) and C-16 allowed the linkage of both
moieties A and B through an ether bond across C-16 and C-7′.
Acidic hydrolysis24 of 6 yielded two products, with one product
being identical to compound 1 based on the same NMR data
and specific rotation (Figure 6). The 2D NMR data including
NOESY data indicated the gross structure of the second
product to be the same as tajixanthone hydrate.12 Mosher’s
method in association with the NOE relationships supported
the absolute configuration of the second product to have 15S,
19R, and 20S configurations, which are the same as those of
tajixanthone hydrate.
Figure 3. Acidic hydrolysis of 11 to 2.
identical to those of 2. As the absolute configuration of
varioxirane was determined by total synthesis,19 the stereogenic
centers in 2 were confirmed as 10R, 11R, 12S, and 13S.
The HRESIMS and NMR data determined the molecular
formula of varioxiranol C (3) (C16H24O6) to be one CH2 unit
more than that of 2. Apart from an additional methoxy group
(δH 3.20, s) observed in 3, the NMR spectroscopic data of both
2 and 3 were very similar. The 2D NMR (COSY, HMQC, and
HMBC) data assigned 3 to be a 13-O-methyl analogue of 2, as
evident from the methoxy protons correlated to C-13 (δC 77.3)
in the HMBC spectrum. The similar NOE interaction and J
values of the side chain in both 2 and 3 (Table 1) indicated 3
shared the same relative configuration as 2.
Varioxiranol D (4) was found to possess a molecular formula
of C15H21ClO5, as determined by the HRESIMS data, with a Cl
atom to replace an OH group of 2. The NMR data for the
aromatic nucleus in 4 were identical to those of 2, while the
NMR data for the side chain of both 4 and 2 were mostly
similar. The difference was attributed to the shielded C-13 (δC
61.3) and the deshielded H-13 (δH 4.48, dq, J = 6.8, 3.6 Hz) in
comparison with those of 2. These findings in association with
the absence of OH at C-13 allowed assignment of a chlorinated
3
C-13. On the basis of JH,H values between adjacent methines
from H-8 to H-13 (Table 2) and NOE interactions similar to
those of 2, the configurations of the stereogenic centers in 4
were assumed to be the same as those of 2.
The molecular formula of varioxiranol E (5) was determined
as C16H20O7 on the basis of its HRESIMS and NMR data,
requiring seven degrees of unsaturation. The IR absorption at
1784 cm−1 suggested the presence of a dioxolanone
functionality.22,23 Inspection of the 13C NMR spectrum
revealed six aromatic carbons for a phenyl ring and two
olefinic carbons for a double bond, which was similar to the
benzyl alcohol skeleton such as 2. The 2D NMR data
established the molecular backbone of 5 to be the same as
that of 2, while the side chain carbons from C-10 (δC 70.7) to
C-13 (δC 65.8) were oxygenated. The COSY correlations
between H-10 (δH 4.54)/OH (δH 5.89) and H-13 (δH 3.84)/
OH (δH 5.31) showed C-10 and C-13 to be hydroxylated. In
addition, the HMBC interactions of the carbonyl carbon C-15
(δC 155.3) with both H-11 (δH 4.67) and H-12 (δH 4.44)
established a 1,3-dioxolan-2-one ring, which was located
between C-11 (δC 78.3) and C-12 (δC 79.6). The relative
configuration of 5 was established by NOE and coupling
constants of the side chain, while the stereogenic centers from
C-10 to C-13 were all determined to be of the R configuration
by the X-ray single-crystal diffraction analysis using the Flack
parameter (Figure 4).
The HRESIMS and NMR data provided the molecular
formula of varioxiranol G (7) as C40H46O10, with 2 amu less
than that of 6. The NMR data of 7 were closely similar to those
of 6 (Table 2), while the 2D NMR data indicated the presence
of two moieties. Comparison of the NMR data revealed that 7
possessed the same xanthone moiety (moiety B) as that of 6,
while the remaining moiety was identical to compound 9 by the
evidence that two additional olefin protons (δH 5.54 and 5.79)
1
were observed in the H NMR spectrum. The location of the
double bond at C-12′ and C-13′ was confirmed by the COSY
correlation. The connection of C-7′ and C-16 through an ether
bond was assembled by the HMBC interaction from H2-7′ (δH
4.59, 4.71) to C-16 (δC 78.0). Acidic hydrolysis of 7 to afford
compound 9 and tajixanthone hydrate supported the structural
assignment, while the similar ECD data of both 6 and 7 (Figure
7) reinforced the same absolute configuration of both
compounds.
D
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
ab
,
1
Table 2. H and 13C NMR Data of 6−8
6
7
8
no.
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
δC, type
δH (J in Hz)
1
160.2, C
109.9, CH
138.2, CH
116.9, C
119.2, CH
138.3, C
149.5, C
121.1, C
109.2, C
153.0, C
152.1, C
116.9, C
184.5, C
31.1, CH2
160.2, C
109.8, CH
138.2, CH
117.0, C
119.2, CH
138.2, C
149.5, C
121.1, C
109.2, C
153.0, C
152.1, C
117.0, C
184.5, C
31.2, CH2
162.9, C
109.0, CH
137.8, CH
123.6, C
120.5, CH
133.3, C
147.0, C
119.5, C
112.8, C
157.2, C
155.3, C
122.5, C
199.2, C
27.6, CH2
2
6.78, d (8.4)
6.83, d (8.3)
7.61, d (8.4)
6.42, d (8.4)
3
7.60, d (8.4)
7.21, s
7.25, d (8.4)
6.82, s
4
5
7.21, s
6
7
8
9
10
11
12
13
14
3.11, dd (14.3, 2.0)
2.72, dd (14.3, 10.0)
3.92, dd (10.0, 2.0)
3.11, dd (14.2, 2.3)
2.72, dd (14.2, 10.0)
3.93, dd (10.0, 2.3)
3.26, d (6.9)
5.28, t (6.9)
15
16
17
18
19
20
21
76.6, CH
78.0, C
76.6, CH
78.0, C
121.8, CH
133.5, C
19.4, CH3
22.1, CH3
63.2, CH
44.9, CH
64.6, CH2
1.43, s
19.7, CH3
22.1, CH3
63.2, CH
44.9, CH
64.6, CH2
1.42, s
17.8, CH3
25.8, CH3
70.4, CH
45.0, CH
63.0, CH2
1.76, s
1.50, s
1.50, s
1.71, s
5.43, brd (2.8)
2.76, ddd (2.8, 2.9, 3.0)
4.44, dd (10.8, 2.9)
4.36, dd (10.8, 3.0)
5.43, brd (2.6)
2.75, m
4.57, brs
2.19, dd (9.8, 3.6)
4.26, dd (10.5, 3.6)
4.30, dd (10.5, 9.8)
4.36, dd (10.7, 2.7)
4.44, dd (10.7, 3.2)
22
23
24
25
1′
2′
3′
4′
5′
6′
7′
142.6, C
142.6, C
75.0, C
112.3, CH2
22.6, CH3
17.4, CH3
123.8, C
4.60, s; 4.82, s
1.87, s
112.3, CH2
22.6, CH3
17.4, CH3
123.9, C
4.60, s; 4.83, s
1.87, s
23.9, CH3
22.6, CH3
16.8, CH3
1.20, s
1.05, s
2.22, s
2.36, s
2.36, s
157.7, C
157.7, C
110.1, CH
129.1, CH
119.4, CH
138.7, C
6.82, d (8.0)
7.26, t (8.0)
7.09, d (8.0)
110.1, CH
129.1, CH
119.3, CH
138.3, C
6.79, d (8.0)
7.25, t (8.0)
7.07, d (8.0)
54.4, CH2
4.58, d (9.4)
4.74, d (9.4)
6.99, d (15.8)
6.20, dd (15.8, 6.7)
4.26, dd (6.7, 3.7)
3.78, m
54.4, CH2
4.59, d (9.5)
4.71, d (9.5)
8′
9′
10′
11′
12′
130.8, CH
130.2, CH
75.7, CH
73.9, CH
34.5, CH2
19.1, CH2
14.1, CH3
130.6, CH
130.4, CH
75.5, CH
75.5, CH
129.0, CH
130.0, CH
17.9, CH3
7.00, d (15.8)
6.16, dd (15.8, 5.8)
4.31, dd (5.8, 4.0)
4.18, dd (7.1, 4.0)
5.54, ddd(15.5,7.1, 1.5)
5.79, dq (15.5, 6.1)
1.71, d (6.1)
1.43, m
13′
14′
1.54, m; 1.38, m
0.92, t (7.3)
12.63, s
1-OH
2′-OMe
10-OH
11-OH
19-OMe
22-OMe
12.64, s
13.92, s
55.8, CH3
3.82, s
55.9, CH3
3.82, s
10.87, s
10.64, s
3.17, s
3.15, s
57.6, CH3
48.7, CH3
a
b
Chemical shifts are in ppm; J values in Hz are in parentheses. Measured at 400 MHz in CDCl3.
Varioxiranols F and G (6, 7) appear to be products derived
Compound 8 had a molecular formula of C27H34O7, as
deduced from HRESIMS data. The 13C NMR spectrum
displayed a total of 27 resonances, including 12 aromatic
carbons for two phenyl rings (rings A and B). In regard to
aromatic ring A, an AB spin system was observed by the COSY
correlation between H-2 (δH 6.42, d, J = 8.4 Hz) and H-3 (δH
7.25, d, J = 8.4 Hz). The presence of an isoprenyl unit was
recognized by the COSY correlation between methylene H2-14
(δH 3.26, d, J = 6.9 Hz) and olefinic proton H-15 (δH 5.28, t, J =
from the intermolecular condensation of tajixanthone (13) with
varioxiranol A (1) and (1E,5E)-1-(2-(hydroxymethyl)-3-
methoxyphenyl)hepta-1,5-diene-3,4-diol (9), respectively. The
highly regioselective nucleophilic attack of the hydroxy group at
C-7 of 1 or 9 to C-16 of the oxirane in 13, in addition to the
observation of 6 and 7 in the HPLC chromatogram of the
EtOAc extract, led to the hypothesis that an enzymatic step is
necessary for the condensation.
E
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 7. ECD spectra of 6 and 7.
Figure 4. Structure of 5 determined by X-ray single-crystal diffraction
analysis.
6.9 Hz), along with the HMBC relationships from methyl
protons H3-17 (δH 1.76, s) and H3-18 (δH 1.71, s) to C-16 (δC
133.5) and C-15 (δC 121.8). The linkage of the isoprenyl unit
to C-4 (δC 123.6) was confirmed by the HMBC interactions
from H2-14 to C-3 (δC 137.8), C-4, and C-10 (δC 157.2).
Additional HMBC relationships from H-3 to C-1 (δC 162.9)
and C-10 and from H-2 to C-9 (δC 112.8) and C-4 assigned all
carbons in aromatic ring A, while the deshielded chemical shifts
of C-1 and C-10 were indicative of their oxygen substitution.
Aromatic ring B presented a lone proton H-5 (δH 6.82, s) for a
pentasubstituted aromatic ring, while the formation of a
chroman ring was based on the COSY relationships from H-
20 (δH 2.19, brdd, J = 9.8, 3.6 Hz) to H2-21 (δH 4.26, 4.30) and
H-19 (δH 4.57, brs), in association with the HMBC interactions
from H2-21 to C-7 (δC 147.0) and from H-19 to C-7, C-8 (δC
119.5), and C-12 (δC 122.5). In addition, a methoxyisopropane
unit linked to C-20 (δC 55.0) was evident from the HMBC
correlation between MeO (δH 3.15, s) and C-22 (δC 75.0) and
from both methyl singlets δH 1.20 (3H, s, H3-23) and 1.05 (3H,
s, H3-24) to C-22 and C-20. An additional HMBC interaction
between MeO (δH 3.17, s) and C-19 (δC 70.4) confirmed C-19
to be methoxylated. Moreover, the methylation and hydrox-
Figure 5. Key HMBC and COSY correlations of 6.
Figure 6. Acidic hydrolysis of 6 and 7.
F
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
ylation at C-6 (δC 133.3) and C-11 (δC 155.3), respectively,
were defined by the HMBC correlations from methyl protons
(δH 2.22, s, H3-25) to C-5, C-6, and C-7 in addition to OH-11
(δH 10.87, s) correlating to C-5, C-11, and C-12. The linkage of
both aromatic rings A and B to form a benzophenone across C-
13 (δC 199.2) was confirmed by the observation of long-range
HMBC interactions from the carbonyl C-13 to H-2 and H-5.
Thus, the structure of 8 was determined as 19-O-methyl-22-
methoxypre-shamixanthone. The NOE interaction between H-
19 and H3-23/H3-24 determined a trans orientation of H-19
and H-20, while similar ECD effects of both 8 and pre-
shamixanthone (12) (Figure 8) revealed both compounds
shared the same configurations at C-19 and C-20.
Figure 9. Effects of compounds on oleic acid-elicited intracellular lipid
accumulation. Positive control: simvastatin; blank: DMEM; OA: oleic
acid. Compounds were determined by spectrophotometry at 358 nm
after Oil Red O staining. Bars depict the means SEM in triplicate.
**p < 0.01, OA vs blank; ##p < 0.01, test group vs OA group.
Figure 8. ECD spectra of 8 and pre-shamixanthone.
On the basis of the MS and NMR spectroscopic analyses and
the comparison of the specific rotation, compound 9 was
identified as the known fungal metabolite (1E,5E)-1-(2-
(hydroxymethyl)-3-methoxyphenyl)hepta-1,5-diene-3,4-diol25
and as the enantiomer of the synthesized product.16 Natural 9
was initially reported with configuration information,25 which
has now been confirmed by Mosher’s method (Figures S59 and
S60, Supporting Information), indicating 11R and 12S
configurations. Compounds 10 and 11 were identical to
varitriol2,26 and varioxirane,2 respectively, according to the
spectroscopic comparison. Analyses of 1D and 2D NMR
spectroscopic data identified 12 as pre-shamixanthone, a
product from A. nidulans when its cryptic PKS genes were
activated.19 In addition, five xanthone analogues were identical
to tajixanthone (13),24 tajixanthone methanoate (14),12
tajixanthone hydrate (15),12 15-acetyltajixanthone hydrate,20
and shamixanthone.12
Compounds 1−12 were tested for lipid-lowering effects
against oleic acid (OA)-elicited lipid accumulation in HepG2
liver cells.27,28 Three compounds, 1, 9, and 12, exerted
inhibitory effects against lipid accumulation at a dose of 10
μM as measured by Oil Red O staining (Figure 9) and showed
no toxicity up to 100 μM toward HepG2 cells in the MTT
assay (Figure S62, Supporting Information). Among the active
compounds, compound 12 was the most active, with an
inhibitory effect comparable to the positive control simvastatin.
Further bioassay revealed that compound 12 exerted significant
inhibition against intracellular triglyceride (TG) levels (Figure
10A) and dramatically reduced total cholesterol (TC) (Figure
10B) at a dose of 10 μM. It is noted that compound 1 mainly
reduced TG levels with weak effects on TC, whereas compound
9 induced the reduction of TC with weak effect on TG. These
findings suggested distinct mechanisms of 1 and 9 for the
antihyperlipidemic effects, which is surprising considering that
9 differs from 1 only by having an extra double bond at C-12/
Figure 10. Inhibitory effects of compounds toward (A) total
cholesteroland and (B) triglycerides. Positive control: simvastatin;
blank: DMEM; OA: oleic acid. Intracellular levels of triglycerides and
total cholesterol were measured by kits according to the
manufacturer’s instructions. Bars depict the means SEM in triplicate.
**p < 0.001, *p < 0.05, OA vs blank; #p < 0.05, ##p < 0.01, compound
vs OA.
C-13. Real-time quantitative PCR was performed29 to
determine how compound 12 regulated the expression of
lipid metabolic genes. As shown in Figure 11, compound 12
dramatically down-regulated sterol regulatory element-binding
transcription factor 1 (SREBP-1c) and the mRNA levels of
Figure 11. Effect of 12 on the transcription of lipogenic genes. Gene
expression was quantified by quantitative real-time PCR analysis. The
gene expression levels were normalized to β-actin mRNA levels. Values
represent mean SD. Results are representative of three independent
experiments. *p < 0.05, **p < 0.01, ***p < 0.001 vs control group.
G
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
transferred to fermentation medium and further incubated for 30 days
at 28 °C statically.
genes downstream of SREBP including genes encoding acetyl-
CoA carboxylase (ACC) and fatty acid synthase (FAS).
SREBP-1c is a key transcriptional factor promoting lipid
production,30 while ACC and FAS are two sequential enzymes
in the fatty acid biosynthetic pathway whose expression is
regulated by SREBP-1c.31 The experimental results indicated
that 12 may decrease lipid accumulation through down-
regulation of the SREBP-1 pathway.
The present work revealed additional chemical diversity from
the marine sponge-associated fungal strain E. variecolor XSA-07-
2. Varioxiranols F and G (6, 7), featuring the linkage of a benzyl
alcohol with a xanthone, were found for the first time from
marine-derived fungi, while the 1,3-dioxolan-2-one unit in
compound 5 is an unusual moiety, which was reported
previously in diterpenes and anthrodioxolanone from
plants,32,33 as well as in cyclohexenes from fungi.34 Cytosporin
E is the only one bearing a dioxolanone unit from a marine-
derived fungus. It was isolated from a strain of Eutypella
scoparia associated with a marine pulmonate mollusc.35 The
potent inhibition of compound 12 against lipid accumulation
by regulation of the SREBP pathway, with low cytotoxicity,
suggested 12 to be a potential lead compound for the
development of an antihyperlipidemic agent.
Extraction and Isolation. The rice cultures were extracted with
EtOAc (4 times). The EtOAc solution was concentrated in vacuo (32
°C) to yield an extract (40 g), which was subsequently subjected to
silica gel column chromatography eluting with petroleum ether/
acetone (4:1, v/v) to afford four fractions, F1−F4. F1 (3.6 g) was
further fractionated by silica gel column chromatography and eluted
with a petroleum ether/acetone gradient (from 60:1 to 10:1) to obtain
three fractions (F1a−F1c). F1b (2.5 g) was subjected to an ODS
column (10 μm) eluting with constant MeOH/H2O (75%) to yield 8
(7.1 mg), 12 (28.7 mg), 6 (14.6 mg), and 7 (8.0 mg). F1c (1.1 g) was
separated through semipreparative HPLC (C18) with MeCN/H2O
(55%) as a mobile phase to obtain 9 (28.3 mg), 10 (36.1 mg), and 4
(11.0 mg). A gel filtration over Sephadex LH-20 of F1c (30.2 mg)
eluting with MeOH was employed to afford compound 2 (10.5 mg).
Fraction F2 (200 mg) was separated by semipreparative HPLC with
MeCN/H2O (55%) as a mobile phase to yield 1 (12.0 mg), 11 (22.4
mg), 3 (9.3 mg), and 5 (15.5 mg). Fraction F4 (7.0 g) was subjected
to a silica gel (300−400 mesh) column separation eluting with
CH2Cl2/MeOH (10:1) to obtain tajixanthone (13, 2.0 g), tajixanthone
methanoate (14, 3.0 g), and tajixanthone hydrate (15, 60.0 mg). The
remaining collection (85 mg) was further separated on semi-
preparative HPLC with acetonitrile/H2O (65%) as a mobile phase
to yield 15-acetyltajixanthone hydrate (9.3 mg) and shamixanthone
(22.4 mg).
Varioxiranol A (1): white, amorphous powder; [α]20D +5.28 (c 0.26,
MeOH); UV (MeOH) λmax (log ε) 220 (2.8), 251 (2.3), 294 (1.7)
EXPERIMENTAL SECTION
■
1
nm; IR (KBr) νmax 3251, 2957, 2906, 2868, 1575 cm−1; H and 13C
General Experimental Procedures. Melting points were
measured on an X-5 micro melting point apparatus (Kexian Co.,
China). Optical rotations were measured using an Autopol III
automatic polarimeter (Rudolph Research Co.). UV spectra were
recorded by a 3300-ELSD UV detector (Alltech Co.). IR spectra were
measured on a Thermo Nicolet Nexus 470 FT-IR spectrometer. NMR
spectra were recorded on a Bruker DRX-400 NMR (or on a Bruker
NMR data, Table 1; HRESIMS m/z 289.1410 [M + Na]+ (calcd for
C15H22O4Na, 289.1407).
Varioxiranol B (2): white, amorphous powder; [α]20D +15.0 (c 0.24,
MeOH); UV (MeOH) λmax (log ε) 220 (2.8), 251 (2.3), 294 (1.6)
nm; IR (KBr) νmax 3316, 2954, 2931, 2876, 1673, 1258 cm−1; 1H and
13C NMR data, Table 1; HRESIMS m/z 321.1327 [M + Na]+ (calcd
for C15H22O6Na, 321.1322).
1
DRX-500 NMR) spectrometer in CDCl3 or DMSO-d6 with H and
13C nuclei observed at 400 and 100 MHz (500 and 125 MHz),
respectively, using TMS as an internal standard. HRESIMS spectra
were obtained on a Bruker APEX IV 70e FT-MS spectrometer and on
a Thermo DFS spectrometer using a matrix of 3-nitrobenzyl alcohol.
Column chromatography was carried out with silica gel (200−300
mesh), and HF254 silica gel for TLC was obtained from Qingdao
Marine Chemistry Co. Ltd. ODS gel (50 μm) and Sephadex LH-20
(18−110 μm) were obtained from YMC (Japan) and Amersham
Pharmacia Biotech AB, Uppsala, Sweden. HPLC was performed with
an Alltech 426 pump employing a UV detector, and the Chromasil C18
column (semipreparative, 10 μm) was purchased from Pharmacia. All
chemicals used herein were of analytical grade. HepG2 cells (ATCC
CRL-10741) were obtained from the American Type Culture
Collection. DMEM medium, fetal bovine serum, and penicillin/
streptomycin were supplied by Gibco, while oleic acid and lovastatin
were obtained from Sigma-Aldrich.
Fungal Material. Fungal strain Emericella variecolor XSA-07-2 was
isolated from a marine Cinachyrella sp. sponge, which was collected
from Yongxin Island in the South China Sea, in April 2013. The fungal
clone germinated from the cut of sponge tissue was repurified under
sterile conditions using standard methods. Morphological scrutiny of
hyphae and spores combined with the 18S rDNA ITS sequence
(GenBank number KP202154) led to the identification of its species.
This strain (XSA-07-2) was deposited at the State Key Laboratory of
Natural and Biomimetic Drugs, Peking University, China.
Varioxiranol C (3): white, amorphous powder; [α]20D +14.3 (c 0.34,
MeOH); UV (MeOH) λmax (log ε) 220 (2.5), 252 (1.7) nm; IR (KBr)
νmax 3439, 2978, 2897, 1681, 1575 cm−1; 1H and 13C NMR data, Table
1; HRESIMS m/z 335.1471 [M + Na]+ (calcd for C16H24O6Na,
335.1475).
Varioxiranol D (4): colorless gum; [α]20 +23.3(c 0.30, MeOH);
D
UV (MeOH) λmax (log ε) 220 (2.63), 252 (1.27) nm; IR (KBr) νmax
3368, 2934, 1752, 1250 cm−1; 1H and 13C NMR data, Table 1;
HRESIMS m/z 339.0975 [M + Na]+ (calcd for C15H21O5ClNa,
339.0982).
Varioxiranol E (5): colorless crystal; mp 76.5 °C; [α]20 +168 (c
D
0.42, MeOH); UV (MeOH) λmax (log ε) 221 (2.63), 252 (1.99) nm;
1
IR (KBr) νmax 3315, 2957, 2875, 1784, 1677 cm−1; H and 13C NMR
data, Table 1; HRESIMS m/z 347.1107 [M + Na]+ (calcd for
C16H20O7Na, 347.1096).
Varioxiranol F (6): yellow gum; [α]20D +5.54 (c 0.26, MeOH); UV
(MeOH) λmax (log ε) 203 (3.27), 241 (3.10), 254 (3.07), 268 (3.09),
294 (2.42) nm; ECD (c 3.8 × 10−4 M, MeOH) λmax (Δε) 220 (−2.5),
242 (−0.1), 276 (−3.0) nm; IR (KBr) νmax 3383, 2925, 2854, 1726,
1655, 1598, 1677, 1243 cm−1; 1H and 13C NMR data, Table 2;
HRESIMS m/z 711.3145 [M + Na]+ (calcd for C40H48O10Na,
711.3136).
Varioxiranol G (7): yellow gum; [α]20D +8.57 (c 0.14, MeOH); UV
(MeOH) λmax (log ε) 203 (3.61), 248 (3.44), 253 (3.40), 273 (3.42),
294 (2.89) nm; ECD (c 2.0 × 10−4 M, MeOH) λmax (Δε) 227 (−2.5),
248 (−2.1), 300 (−2.8) nm; IR (KBr) νmax 3410, 2934, 2893, 1643,
Fermentation of the Fungus. Fermentation of the strain was
initiated in 40 500 mL sized Erlenmeyer flasks, each preloaded with 80
g of rice and 100 mL of sterilized artificial seawater (NaCl 26.726 g,
MgCl2 2.26 g, MgSO4 3.248 g, CaCl2 1.153 g, NaHCO3 0.198 g, KCl
0.721 g, NaBr 0.058 g, H3BO3 0.058 g, Na2SiO3 0.0024 g, Na2Si4O9
0.0015 g, H3PO4 0.002 g, Al2Cl6 0.013 g, NH3 0.002 g, LiNO3 0.0013
g, H2O 1 L). The seed was prepared by inoculating activated fungal
cakes from an agar Petri dish into 200 mL of potato dextrose broth
medium. Approximately 20 mL aliquots of the inoculum were then
1
1359, 1245 cm−1; H and 13C NMR data, Table 2; HRESIMS m/z
685.3013 [M − H]− (calcd for C40H45O10, 685.3026).
19-O-Methyl-22-methoxypre-shamixanthone (8): yellow gum;
[α]20 −8.10 (c 0.42, MeOH); UV (MeOH) λmax (log ε) 202
D
(3.16), 232 (2.73), 289 (2.61), 368 (2.14) nm; ECD (c 8.9 × 10−4 M,
MeOH) λmax (Δε) 238 (−0.2), 270 (−0.1), 311 (−1.1) nm; IR (KBr)
νmax 3352, 2972, 2924, 1617, 1424, 1349, 1223 cm−1; H and 13C NMR
H
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 3. Primers Used in Real-Time Quantitative PCR Analysis
name
forward (5′−3′)
reverse (5′−3′)
SREBP-1c
FAS
CCATGGATGCACTTTCGAA
CGGTACGCGACGGCTGCCTG
TGATGTCAATCTCCCCGCAGC
CCAGCATAGGGTGGGTCAA
GCTGCTCCACGAACTCAAACACCG
TTGCTTCTTCTCTGTTTTCTCCCC
ACC
data, Table 2; HRESIMS m/z 469.2226 [M − H]− (calcd for
C27H33O7, 469.2215).
methods using SHELXS-97 and refined using full-matrix least-squares
difference Fourier techniques. Crystal data of 5: C16H20O7, M =
324.32, crystal size 0.40 × 0.20 × 0.16 mm; orthorhombic, a =
9.1496(5) Å, b = 12.6638(4) Å, c = 13.4684(4) Å, α = 90.00°, β =
90.00°, γ = 90.00°, U = 1560.57(11) Å3, T = 99.2 K, space group
P212121 (no.19), Z = 4, μ(Cu Kα) = 0.917 mm−1, F(000) = 688; 5368
reflections measured, 2888 unique [(Rint = 0.0218(inf-0.9 Å)] were
used in all calculations; the calculated density was 1.380 mg/mm3; the
final R indexes (all data) were R1 = 0.0357 and wR2 = 0.0852; the
Flack parameter was −0.03(17). The crystallographic data for the
structure of 5 have been deposited in the Cambridge Crystallographic
Data Centre (deposition number: CCDC 1019445). Copies of the
data can be obtained free of charge, on application to the director,
CCDC, 12 Union Road, Cambridge CB21EZ, UK (http://www.ccdc.
cam.ac.uk; fax: +44-(0)1223-336033, or e-mail: deposit@ccdc.cam.ac.
uk).
Cell-Based Lipid Accumulation Assay. HepG2 cells were
maintained in DMEM medium supplemented with 10% fetal bovine
serum and penicillin/streptomycin (100 μg/mL). The cells with 70−
80% confluence were incubated in DMEM + oleic acid (100 μM) for
12 h and then were treated with the compounds (each, 10 μM) and
the positive control lovastatin in DMEM/100 μM oleic acid with
DMEM/100 μM oleic acid as a blank for an additional 6 h.
Subsequently, the cells were subjected to Oil Red O staining or TC
and TG determination as described previously.27,28 Each experiment
(n = 8 for Oil Red O staining or n = 3 for TC and TG determination)
was repeated in triplicate.
Quantitative Real-Time PCR. The mRNA levels of lipid
metabolism-related genes were determined by real-time quantitative
PCR. Total RNA extraction, cDNA synthesis, and quantitative PCR
assays were performed as described previously.28 Samples were cycled
40 times using a Fast ABI-7500 sequence detector (Applied
Biosystems). ABI-7500 cycle conditions were as follows: 5 min at 95
°C followed by 40 cycles of 15 s at 95 °C, 30 s at 60 °C, and 30 s at 72
°C. Cycle threshold (CT) was calculated under default settings for
real-time sequence detection software (Applied Biosystems). At least
three independent biological replicates were performed to check the
reproducibility of the data. The gene-specific primers used for
quantitative PCR are listed in Table 3.
(1E,5E)-1-(2-(Hydroxymethyl)-3-methoxyphenyl)hepta-1,5-diene-
3,4-diol (9): [α]20 +18.5 (c 0.22, EtOH) (lit.25 [α]27 +20.2 (c 0.98,
D
D
EtOH) and [α]24.5 +42.7 (c 0.22, CHCl3)).
D
Varioxirane (11): [α]20D −23.6 (c 0.18, CHCl3) (lit.21 [α]30D −24.6
(c 0.13, CHCl3)).
MPA Esterification of 9. Compound 9 (0.047 mmol) was
dissolved in DMF (0.5 mL), and TBSCl (0.052 mmol) and imidazole
(0.142 mmol) were added sequentially at 0 °C. Then the mixture was
stirred for 5 h at room temperature (rt). The reacted mixture was
washed with H2O and then extracted with EtOAc and dried in in
vacuo. The residue was purified on semipreparative HPLC with
MeCN/H2O (90%) as a mobile phase to give a colorless oil, 9a. Both
(R)- and (S)-MPA esters of 9a were obtained by the treatment of 9a
(2.4 mg, respectively) with (R)- and (S)-MPA (3.16 mg) and
dicyclohexylcarbodiimide (3.93 mg) in dry CDCl3 (0.6 mL) catalyzed
with dimethylaminopyridine (2.32 mg) and stirring at rt overnight.
The MPA esters, 9b (3.0 mg) and 9c (2.1 mg), were purified by
semipreparative HPLC using MeCN (100%) as a mobile phase.
MPA Esterification of 1. Following the same protocol as for 9, the
TBS protection and Mosher’s reaction of 1 were performed to yield
the (R)-MPA ester of 1a (2.9 mg) and the (S)-MPA esters of 1a (3.1
mg).
(R)-MPA ester of 1a: 1H NMR (CDCl3, 400 MHz) δH 6.83 (1H, d,
J = 8.1 Hz, H-3), 7.23 (1H, t, J = 8.1 Hz, H-4), 7.03 (1H, d, J = 7.6 Hz,
H-5), 4.79 (2H, d, J = 3.80 Hz, H-7), 3.84 (3H, s, MeO-2), 7.15 (1H,
d, J = 15.8 Hz, H-8), 6.06 (1H, dd, J = 15.9, 7.5 Hz, H-9), 5.69 (1H,
ddd, J = 7.1, 4.2, 1.0 Hz, H-10), 4.95 (1H, td, J = 8.4, 4.4 Hz, H-11),
1.18 (2H, m, H-12), 0.74 (2H, m, H-13), 0.52 (3H, dd, J = 7.1 Hz, H3-
14); ESIMS m/z 697.83 [M + H]+.
(S)-MPA ester of 1a: 1H NMR (CDCl3, 400 MHz) δH 6.77 (1H, d,
J = 7.6 Hz, H-3), 7.20 (1H, t, J = 8.0 Hz, H-4), 6.81 (1H, d, J = 7.7 Hz,
H-5), 4.67 (2H, d, J = 11.8 Hz, H-7), 3.83 (3H, s, MeO-2), 6.82 (1H,
d, J = 16.0 Hz, H-8), 5.65 (1H, dd, J = 16.3, 8.1 Hz, H-9), 5.39 (1H,
dd, J = 8.1, 3.0 Hz, H-10), 5.33 (1H, td, J = 8.3, 3.0 Hz, H-11), 1.52
(2H, m, H-12), 1.32 (2H, m, H-13), 0.89 (3H, t, J = 7.2 Hz, H3-14);
ESIMS m/z 697.83 [M + H]+.
Chemical Conversion of 11 to 2. To a solution of 11 (1.5 mg) in
MeCN (3 mL) was added dropwise aqueous citric acid (20%, 2 mL).
The mixture was stirred at rt until most of the starting material was
transformed. A saturated aqueous NaHCO3 solution (1 mL) was
added to quench the reaction. The mixture was extracted by EtOAc
(10 mL). The organic layers were combined and washed with brine,
dried over anhydrous Na2SO4, and concentrated. The crude product
was subjected to flash chromatography on silica gel using EtOAc/
petroleum ether (1:1.5) to give compound 2 (0.8 mg).
Acidic Hydrolysis of 6 and 7. To a solution of 6 (1.5 mg) in
CH2Cl2 (1 mL) at rt was added aqueous H3PO4 (85 wt %, 0.12 mL)
dropwise. The mixture was stirred for 14 h, and TLC assay showed the
reaction was complete. Then 3 mL of H2O was added. The mixture
was extracted with EtOAc (4 mL). The combined EtOAc phase was
dried over magnesium sulfate and concentrated in vacuo to give an
extract, which was purified by semipreparative HPLC using 40%
MeOH/H2O as a mobile phase to yield compound 1 (0.4 mg) and
tajixanthone hydrate (0.65 mg). Following the same hydrolysis
protocol as for 6, compound 7 (1.5 mg) was converted to 9 (0.5
mg) and tajixanthone hydrate (0.7 mg).
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.5b00578.
NMR spectroscopic data for the new compounds (1−8)
including 1H, 13C, and 2D NMR spectra, IR and ESIMS/
MS data, and X-ray data for 5 (PDF)
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: pguo@implad.ac.cn.
*E-mail: whlin@bjmu.edu.cn. Tel: ++86-10-82806188. Fax: +
+86-10-82806188.
X-ray Crystallographic Analysis of 5. A white crystal of 5 was
obtained in MeOH. The crystal data were recorded on an Oxford
Diffraction Gemini E diffractometer with graphite-monochromated Cu
Kα (λ = 1.5418 Å) radiation. The structures were solved by direct
Notes
The authors declare no competing financial interest.
I
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(29) Eberle,
Biochimie 2004, 86, 839−848.
́
D.; Hegarty, B.; Bossard, P.; Ferre,
́
P.; Foufelle, F.
ACKNOWLEDGMENTS
■
This work was supported by grants from the 973 Project
(2015CB755906), NSFC (No. 30672607), NSFC-Shangdong
Join Fund for Marine Science (U1406402), the National Hi-
Tech 863-Projects (2011AA090701, 2013AA092902),
COMRA (DY125-15-T-01), and Sino-German Project GZ816.
(30) Shimomura, I.; Bashmakov, Y.; Ikemoto, S.; Horton, J. D.;
Brown, M. S.; Goldstein, J. L. Proc. Natl. Acad. Sci. U. S. A. 1999, 96,
13656−13661.
(31) Luo, J.; Li, Y.; Wang, J.; Lu, J.; Wang, X.; Luo, J.; Kong, L. Chem.
Pharm. Bull. 2012, 60, 195−204.
(32) Zhang, F.; Zhang, C.; Tao, X.; Wang, J.; Chen, W.; Yue, J.
Bioorg. Med. Chem. Lett. 2014, 24, 3791−3796.
(33) Chen, J.; Lin, W.; Liao, C.; Shieh, P. J. Nat. Prod. 2007, 70, 989−
992.
REFERENCES
■
(1) Zalar, P.; Frisvad, J. C.; Gunde-Cimerman, N.; Varga, J.; Samson,
R. A. Mycologia 2008, 100, 779−795.
(2) Malmstrom, J.; Christophersen, C.; Barrero, A. F.; Oltra, J. E.;
Justicia, J.; Rosales, A. J. Nat. Prod. 2002, 65, 364−367.
(3) Bringmann, G.; Lang, G.; Steffens, S.; Gunther, E.; Schaumann,
K. Phytochemistry 2003, 63, 437−443.
(4) Wei, H.; Itoh, T.; Kinoshita, M.; Nakai, Y.; Kurotaki, M.;
Kobayashi, M. Tetrahedron 2004, 60, 6015−6019.
(5) Sadler, I. H.; Simpson, T. J. J. Chem. Soc., Chem. Commun. 1989,
6012−6014.
(34) Vatcharin, R.; Nachamon, R.; Saranyoo, K.; Souwalak, P.;
Kawitsara, B.; Jariya, S. J. Nat. Prod. 2014, 77, 2375−2382.
(35) Ciavatta, M. L.; Lopez-Gresa, M. P.; Gavagnin, M.; Nicoletti, R.;
Manzo, E.; Mollo, E.; Guo, Y.; Cimino, G. Tetrahedron 2008, 64,
5365−5369.
(6) Mclntyre, C. R.; Scott, F. E.; Simpson, T. J.; Trimble, L. A.;
Vederas, J. C. J. Chem. Soc., Chem. Commun. 1986, 501−503.
(7) Hensens, O. D.; Zink, D.; Williamson, J. M.; Lotti, V. J.; Chang,
R. S. L.; Goetz, M. A. J. Org. Chem. 1991, 56, 3399−3403.
(8) Wei, H.; Itoh, T.; Kinoshita, M.; Kotoku, N.; Aoki, S.; Kobayashi,
M. Tetrahedron 2005, 61, 8054−8058.
(9) Simpson, T. J.; Ahmed, S. A.; Mclntyre, C. R.; Scott, F. E.; Sadler,
I. H. Tetrahedron 1997, 53, 4013−4034.
(10) Wang, W.; Lu, Z.; Tao, H.; Zhu, T.; Fang, Y.; Gu, Q.; Zhu, W. J.
Nat. Prod. 2007, 70, 1558−1564.
(11) Wang, W.; Zhu, T.; Tao, H.; Lu, Z.; Fang, Y.; Gu, Q.; Zhu, W. J.
Antibiot. 2007, 60, 603−607.
(12) Pornpakakul, S.; Liangsakul, J.; Ngamrojanavanich, N.;
Roengsumran, S.; Sihanonth, P.; Piapukiew, J.; Sangvichien, E.;
Puthong, S.; Petsom, A. Arch. Pharmacal Res. 2006, 29, 140−144.
(13) Wang, W.; Zhu, T.; Tao, H.; Lu, Z.; Fang, Y.; Gu, Q.; Zhu, W.
Chem. Biodiversity 2007, 4, 2913−2919.
(14) Wang, W.; Liu, P.; Zhang, Y.; Li, J.; Tao, H.; Gu, Q.; Zhu, W.
Arch. Pharmacal Res. 2009, 32, 1211−1214.
(15) Sarkar, A.; Funk, A. N.; Scherlach, K.; Horn, F.; Schroeckh, V.;
Chankhamjon, P.; Westermann, M.; Roth, M.; Brakhage, A. A.;
Hertweck, C.; Horn, U. J. Biotechnol. 2012, 160, 64−71.
(16) Scherlach, K.; Hertweck, C. Org. Biomol. Chem. 2009, 7, 1753−
1760.
(17) Matsumori, N.; Kaneno, D.; Murata, M.; Nakamura, H.;
Tachibana, K. J. Org. Chem. 1999, 64, 866−876.
(18) Iwai, T.; Kubota, T.; Kobayashi, J. J. Nat. Prod. 2014, 77, 1541−
1544.
(19) Sudhakar, G.; Raghavaiah, J. J. Org. Chem. 2013, 78, 8840−8846.
(20) Seco, J. M.; Martino, M.; Quinoa, E.; Riguera, R. Org. Lett. 2000,
2, 3261−3264.
(21) Deng, X.; Su, J.; Zhao, Y.; Peng, L.; Li, Y.; Yao, Z.; Zhao, Q. Eur.
J. Med. Chem. 2011, 46, 4238−4244.
(22) Zhang, C. R.; Fan, C. Q.; Zhang, L.; Yang, S. P.; Wu, Y.; Lu, Y.;
Yue, J. M. Org. Lett. 2008, 10, 3183−3186.
(23) Chen, J. J.; Lin, W. J.; Liao, C. H.; Shieh, P. C. J. Nat. Prod.
2007, 70, 989−992.
(24) Li, B.; Berliner, M.; Buzon, R.; Chiu, C. K. F.; Colgan, S. T.;
Kaneko, T.; Keene, N.; Kissel, W.; Le, T.; Leeman, K. R.; Marquez, B.;
Morris, R.; Newell, L.; Wunderwald, S.; Witt, M.; Weaver, J.; Zhang,
Z.; Zhang, Z. J. Org. Chem. 2006, 71, 9045−9050.
(25) Dunn, A. W.; Robert, A. W.; Johnstone, R. A. W. J. Chem. Soc.,
Perkin Trans. 1 1979, 2122−2123.
(26) Ghosh, S.; Pradhan, T. K. J. Org. Chem. 2010, 75, 2107−2110.
(27) Zhang, X.; Wu, C.; Wu, H.; Sheng, L.; Su, Y.; Zhang, X.; Luan,
H.; Sun, G.; Sun, X.; Tian, Y.; Ji, Y.; Guo, P.; Xu, X. PLoS One 2013, 8,
e61922.
(28) Wu, C.; Feng, J.; Wang, R.; Liu, H.; Yang, H.; Rodriguez, P. L.;
Qin, H.; Liu, X.; Wang, D. PLoS One 2012, 7, e35764.
J
DOI: 10.1021/acs.jnatprod.5b00578
J. Nat. Prod. XXXX, XXX, XXX−XXX