Diketopiperazine-Type Alkaloids from a Deep-Sea-Derived Aspergillus puniceus Fungus and Their Effects on Liver X Receptor α

Article abstract of DOI:10.1021/acs.jnatprod.9b00055

Eight new diketopiperazine-type alkaloids including four oxepin-containing diketopiperazine-type alkaloids, oxepinamides H-K (1-4), and four 4-quinazolinone alkaloids, puniceloids A-D (5-8), together with two known analogues (9 and 10), were isolated from the culture broth extracts of the deep-sea-derived fungus Aspergillus puniceus SCSIO z021. Their structures were elucidated by spectroscopic analyses, and their absolute configurations were determined by Marfey's method along with comparison of their specific rotations and ECD spectra. The absolute configurations of 4 and 5 were further confirmed by a single-crystal X-ray diffraction analysis. Compounds 1-8 showed significant transcriptional activation of liver X receptor α with EC₅₀ values of 1.7-50 μM, and 7 and 8 were the most potent agonists.

Full text of DOI:10.1021/acs.jnatprod.9b00055

Article
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
pubs.acs.org/jnp
Diketopiperazine-Type Alkaloids from a Deep-Sea-Derived
Aspergillus puniceus Fungus and Their Effects on Liver X Receptor α
†
,‡
§
§
§
†
,†
Xiao Liang, Xuelian Zhang, Xinhua Lu, Zhihui Zheng, Xuan Ma, and Shuhua Qi*
†CAS Key Laboratory of Tropical Marine Bio-resources and Ecology, Guangdong Key Laboratory of Marine Materia Medica,
Institution of South China Sea Ecology and Environmental Engineering, South China Sea Institute of Oceanology, Chinese
Academy of Sciences, 164 West Xingang Road, Guangzhou, Guangdong 510301, People’s Republic of China
‡
University of Chinese Academy of Sciences, Beijing 100049, People’s Republic of China
§
New Drug Research & Development Co., Ltd, North China Pharmaceutical Group Corporation, Shijiazhuang, Hebei 050015,
People’s Republic of China
*
S Supporting Information
ABSTRACT: Eight new diketopiperazine-type alkaloids including four oxepin-containing diketopiperazine-type alkaloids,
oxepinamides H−K (1−4), and four 4-quinazolinone alkaloids, puniceloids A−D (5−8), together with two known analogues (9
and 10), were isolated from the culture broth extracts of the deep-sea-derived fungus Aspergillus puniceus SCSIO z021. Their
structures were elucidated by spectroscopic analyses, and their absolute configurations were determined by Marfey’s method
along with comparison of their specific rotations and ECD spectra. The absolute configurations of 4 and 5 were further
confirmed by a single-crystal X-ray diffraction analysis. Compounds 1−8 showed significant transcriptional activation of liver X
receptor α with EC50 values of 1.7−50 μM, and 7 and 8 were the most potent agonists.
iver X receptors (LXRα and LXRβ) are critical
modulators of cholesterol and lipid metabolism, inflam-
K (1−4), and four 4-quinazolinone alkaloids, puniceloids A−D
(5−8), together with two known alkaloids, oxepinamides F
L
1
,2
4
matory responses, and innate immunity. LXRs are ligand-
activated transcription factors that belong to a family of
and D (9 and 10), were isolated from the culture broth
extracts of the fungal strain SCSIO z021. These compounds
were evaluated for their transactivation effects on LXRα as well
as their enzyme inhibitory activity against IDO1 (indoleamine
1
,2
hormone nuclear receptors. Their important regulating roles
have led to expanded interest in developing new small-
3
2
,3-dioxygenase 1), LDHA (lactate dehydrogenase A), and five
molecule modulators for these receptors. The vast majority of
different phosphatases, including SHP1, SHP2, PTP1B,
TCPTP, and MEG2, and antifungal activity. Herein, we
describe the isolation and structure elucidation of the new
compounds as well as their bioactivities.
research conducted to find LXR modulators of therapeutic
3
utility has been directed toward developing LXR agonists.
LXR agonists have been shown to have potential use in the
treatment of atherosclerosis, diabetes, anti-inflammation, and
Alzheimer’s disease. Currently, the focus of the field is to
develop selective liver X receptor modulators to avoid the
undesirable side effects caused by the first generation of LXR
RESULTS AND DISCUSSION
The molecular formula of oxepinamide H (1) was assigned as
■
3
1
C H N O according to HRESIMS and NMR data. The H
modulators.
20 19
3
4
NMR data (Table 1) displayed characteristic signals for a
In order to explore new bioactive alkaloids from marine-
derived fungi, we studied the secondary metabolites of the
fungus Aspergillus puniceus SCSIO z021 isolated from deep-sea
sediments. Eight new alkaloids including four oxepin-
containing diketopiperazine-type alkaloids, oxepinamides H−
monosubstituted benzene ring at δ 7.02 (d, J = 7.0 Hz, 2H),
H
Received: January 20, 2019
©
XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.9b00055
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 1. Key HMBC and COSY correlations of compounds 1, 2, 4,
6
, and 7.
7.22 (t, J = 7.0 Hz, 1H), and 7.27 (t, J = 7.0 Hz, 2H), four
olefinic protons at δ 5.83 (t, J = 5.5 Hz), 6.24 (dd, J = 11.0,
H
5
.5 Hz), 6.31 (d, J = 5.5 Hz), and 6.59 (d, J = 11.0 Hz), one
methoxy group at δ 3.02 (s), and one methyl group at δ 0.32
H
H
13
Figure 2. Key NOESY correlations of compounds 1 and 2.
(
2
d, J = 7.0 Hz). The C NMR spectral data (Table 1) showed
0 carbon signals including one methyl, one methoxy, one
methylene, one methine, nine olefinic or aromatic methines,
one oxygenated nonprotonated carbon, four nonprotonated
carbons, and two carbonyl carbons. These data showed great
and the only obvious differences between them were the
changes of chemical shifts of H -16 (2: δ 1.47; 1: δ 0.32), C-
3
H
H
4
similarity to those of 10. The only obvious difference between
1 (2: δ 168.5; 1: δ 166.3), and C-17 (2: δ 40.3; 1: δ 46.6).
C C C C
1
1
and 10 was the additional presence of one methoxy group in
. This was supported by the 2D NMR analysis (Figure 1).
The 2D NMR analysis (Figure 1) proved that 2 had the same
planar structure as 1. Comparison of the electronic circular
dichroism (ECD) spectra (Figure 3) and specific rotation
The HMBC spectrum showed a correlation of δH 3.02 (s,
OCH ) with C-3 (δ 89.2, C), suggesting that the additional
methoxy group was attached at C-3. The configuration of the
Ala subunit was assigned as D-form by Marfey’s method,
indicating the C-15 R-configuration. The NOESY spectrum
2
5
25
values of 2 and 1 ([α] −235 for 2 and [α] +10.7 for 1)
3
C
D
D
suggested they were a pair of stereoisomers. The D-Ala residue
was established by Marfey’s method, which determined the C-
15 R-configuration in 2. Correspondingly, the absolute
configuration of C-3 was determined as R based on the
NOESY spectrum (Figure 2) showing an NOE correlation
between H -16 and 3-OCH . Analysis of the 3D structures of 1
5
(Figure 2) showing NOE correlations of H-15 with 3-OCH3
and H -16 with H-2′/H-3′ suggested the methyl group of the
3
Ala unit and the benzyl group at C-3 were on the same side of
the ring, which indicated the S-configuration for C-3.
3
3
and 2 (Figure 2) showed that the methyl group (CH -16) in 1
3
Oxepinamide I (2) had the same molecular formula of
C H N O as 1 according to HRESIMS and NMR data. The
H and C NMR data for 2 were greatly similar to those for 1,
was located in the shielded region of the benzene ring,
explaining the 0.32 ppm chemical shift of H -16 in 1. However,
2
0
19
3
4
3
1
13
the methyl group in 2 was not in the shielded region.
Table 1. H (500 MHz) and 13C NMR (125 MHz) Data for 1−4
1
1
2
3
4
position
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
3
4
6
8
9
1
1
1
1
1
1
1
166.3, C
168.5, C
168.8, C
166.1, C
-NH
9.29, s
9.57, s
9.31, s
7.41, s
10.62, s
89.2, C
87.3, C
50.3, CH3 3.31, s
153.0, C
161.0, C
143.4, CH 6.30, d (5.5)
117.4, CH 5.83, t (5.5)
82.5, C
125.6, C
-OCH /−OH 50.4, CH3 3.02, s
3
154.7, C
162.0, C
143.7, CH 6.31, d (5.5)
117.3, CH 5.83, t (5.5)
155.4, C
161.5, C
143.4, CH 6.29, d (5.5)
117.4, CH 5.81, t (5.5)
150.0, C
162.1, C
143.3, CH 6.22, d (5.5)
117.4, CH 5.79, t (5.5)
0
1
2
3
5
6
7
128.7, CH 6.24, dd (11.0, 5.5) 128.9, CH 6.25, dd (11.0, 5.5) 128.5, CH 6.23, dd (11.0, 5.5) 127.7, CH 6.19, dd (11.0, 5.5)
125.0, CH 6.59, d (11.0)
110.4, C
159.5, C
51.6, CH
124.9, CH 6.59, d (11.0)
110.7, C
159.7, C
125.0, CH 6.60, d (11.0)
110.1, C
159.8, C
125.4, CH 6.65, d (11.0)
109.5, C
159.8, C
51.7, CH
18.1, CH3 1.48, d (7.0)
117.7, CH 7.14, s
4.55, q (7.0)
51.8, CH
4.55, q (7.0)
52.0, CH
4.66, q (7.0)
5.02, q (7.0)
17.3, CH3 0.32, d (7.0)
46.6, CH2 3.11, d (13.0)
18.4, CH3 1.47, d (7.0)
40.3, CH2 3.23, d (13.5)
3.61, d (13.5)
18.4, CH3 1.53, d (7.0)
43.0, CH2 3.16, d (13.5)
3.73, d (13.5)
3
.48, d (13.0)
1
2
3
4
′
133.8, C
134.2, C
135.0, C
133.1, C
′, 6′
′, 5′
′
130.6, CH 7.02, d (7.0)
128.6, CH 7.27, t (7.0)
127.5, CH 7.22, t (7.0)
130.7, CH 7.26, d (7.0)
127.9, CH 7.20, t (7.0)
126.8, CH 7.15, t (7.0)
130.8, CH 7.36, d (7.0)
127.8, CH 7.20, t (7.0)
126.6, CH 7.14, tt (7.0, 2.5)
129.7, CH 7.62, d (7.5)
128.7, CH 7.44, t (7.5)
128.7, CH 7.36, t (7.5)
B
DOI: 10.1021/acs.jnatprod.9b00055
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 3. Comparison of the experimental ECD spectra of compounds 1−8 and 10 in CH OH.
3
Oxepinamide J (3) had a molecular formula of C H N O
4
1
9
17
3
1
13
based on HRESIMS and NMR spectra. The H and C NMR
data for 3 were greatly similar to those for 2 with the absence
of a methoxy group (δ 50.3, δH 3.31) and the additional
C
presence of an active hydrogen (δ 7.41, s) in 3, suggesting a
H
hydroxy replaced a methoxy group attached at C-3 in 3. The
2
D NMR spectra confirmed the planar structure of 3 was the
same as that of 10. Comparison of the ECD spectra (Figure 3)
25
and specific rotation values of 3 and 10 ([α] −212 for 3 and
D
2
5
[
α] +30.2 for 10) indicated that 3 and 10 were a pair of
D
stereoisomers, which was further supported by the NOESY
spectrum showing an NOE correlation of H -16 and 3-OH in
3
3
. The existence of a D-Ala residue was determined by Marfey’s
method. So, the absolute configuration of 3 was determined to
be the same as that of 2.
The molecular formula of oxepinamide K (4) was
determined as C H N O by HRESIMS and NMR spectra.
19
15
3
3
1
13
The H and C NMR data for 4 were similar to those for 1−3,
suggesting that 4 had the same skeleton as 1−3. The most
obvious difference between 4 and 1 was the presence of a
Figure 4. X-ray crystallographic structure of compound 4.
trisubstituted double bond (δ 125.6 (C), 117.7 (CH)) and
C
configuration of 5 was established by the NOESY correlations
between H -15 and H-2′/H-3′, which indicated CH -15 and
the absence of signals for a methoxy, an oxygenated
nonprotonated carbon, and a methylene in 4. The HMBC
correlations from H-17 (δH 7.14, s) to C-4/C-2′/C-6′ and
from NH (δH 10.62, s) to C-3 (δC 125.6, C) located the
double bond between C-3 and C-17. The R-configuration was
assigned for C-15 by Marfey’s method, and the double bond
between C-3 and C-17 was determined to have the Z-
configuration by a single-crystal X-ray diffraction experiment
3
3
the benzyl group were located on the same side of the ring.
The absolute configuration of C-14 was determined as R by
Marfey’s method, and correspondingly, C-3 was assigned as the
S-configuration. These assignments were further confirmed by
a single-crystal X-ray diffraction analysis (Figure 5).
Puniceloid B (6) had the molecular formula C H N O on
1
9
17
3
4
1
13
the basis of HRESIMS and NMR spectra. The H and
C
(Figure 4).
Puniceloid A (5) exhibited the same molecular formula of
C H N O as 1 by analyzing the HRESIMS and NMR
2
0
19
3
4
1
13
spectra. The H and C NMR data for 5 were highly similar to
those for 1, but the UV absorptions of 5 and 1 were
significantly different, implying they possessed a different
1
13
conjugated ring systems. Comparison of the H and C NMR
data for 5 and 1 displayed that the main differences between
them were the additional presence of a phenolic hydroxy
proton (δ 9.99, br s) and a deshielded nonprotonated carbon
H
(
1
C-7: δ 153.1) in 5 instead of an olefinic methine (CH-8 in
). Besides that, the chemical shifts and the coupling constants
C
of H-8/H-9/H-10 revealed the presence of a 1,2,3-
trisubstituted benzene ring instead of an oxepine ring in 5.
The HMBC correlations (Figure 1) from H-10 to C-6/C-12,
from H-9 to C-7/C-11, and from H-8 to C-6 suggested the
phenolic hydroxy group was attached at C-7. The relative
Figure 5. X-ray crystallographic structure of compound 5.
C
DOI: 10.1021/acs.jnatprod.9b00055
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 2. H (500 MHz) and 13C NMR (125 MHz) Data for 5−8
1
5
6
7
8
position
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
3
4
6
7
7
8
9
1
1
1
1
1
1
166.8, C
165.6, C
169.8, C
166.6, C
-NH
9.24, s
2.98, s
9.22, s
9.19, s
10.52, s
89.5, C
84.3, C
83.0, C
126.0, C
-OCH /-OH
50.0, CH3
147.2, C
135.6, C
153.1, C
6.82, br s
7.22, br s
3
150.8, C
135.6, C
152.7, C
148.2, C
136.0, C
153.1, C
144.1, C
136.1, C
152.8, C
-OH
9.99, br s
9.76, br s
10.00, br s
9.75, s
119.0, CH 7.32, dd (8.0, 1.5) 118.4, CH 7.30, dd (8.0, 1.5) 118.8, CH 7.30, dd (8.0, 1.5) 118.4, CH 7.26, dd (8.0, 1.0)
128.3, CH 7.42, t (8.0) 128.0, CH 7.40, t (8.0) 128.1, CH 7.39, t (8.0) 128.1, CH 7.35, t (8.0)
115.8, CH 7.54, dd (8.0, 1.5) 115.8, CH 7.52, dd (8.0, 1.5) 115.8, CH 7.55, dd (8.0, 1.5) 116.0, CH 7.58, dd (8.0, 1.0)
120.8, C
159.2, C
50.9, CH
17.9, CH3
46.1, CH2
0
1
2
4
5
6
120.5, C
159.3, C
51.2, CH
17.8, CH3
47.4, CH2
120.9, C
159.6, C
51.8, CH
18.3, CH3
43.1, CH2
120.4, C
159.5, C
51.2, CH2
18.2, CH3
4.73, q (7.0)
0.35, d (7.0)
3.12, d (12.5)
4.61, q (7.0)
0.44, d (7.0)
3.12, d (12.5)
3.91, d (12.5)
4.83, q (7.0)
1.59, d (7.0)
3.17, d (13.5)
4.26, d (13.5)
5.22, q (7.0)
1.50, d (7.0)
117.2, CH 7.74, s
4
.06, d (12.5)
1
2
3
4
′
134.7, C
135.2, C
135.3, C
134.0, C
′, 6′
′, 5′
′
130.9, CH 7.06, d (7.0)
128.3, CH 7.18, t (7.0)
127.1, CH 7.15, m
130.8, CH 7.06, d (7.0)
128.2, CH 7.18, t (7.0)
127.0, CH 7.15, m
131.4, CH 7.53, d (7.5)
127.5, CH 7.12, t (7.5)
126.3, CH 7.07, tt (7.5, 2.5)
129.4, CH 7.65, d (7.5)
128.7, CH 7.47, t (7.5)
127.6, CH 7.37, m
NMR data for 6 were similar to those for 5 except for the
additional presence of a hydroxy proton and the loss of a
methoxy group in 6. Further analysis of the HMBC and COSY
spectra suggested that 6 had the same planar structure as 5
except for a hydroxy group instead of a methoxy group
acid, TFA) as the eluting solvent. However, after concentration
under reduced pressure, the four compounds could inter-
convert, and finally they could transform into 4, which was a
stable compound under the acidic conditions. In addition, 5−7
also exhibited a similar phenomenon, and they could slowly
transform into 8 in the acidic solution. As we know,
chloroform can decompose under light to generate hydrogen
attached at C-3. The NOE correlations between H -15 and H-
3
2
′/H-3′ in the NOESY spectrum of 6 indicated that CH -15
3
1
and the benzyl group were on the same side of the ring. The
configuration of C-14 in 6 was determined to be R by Marfey’s
method. Correspondingly, the configuration of C-3 in 6 was
inferred to be S.
chloride and create an acidic environment. When the H and
1
3
C NMR spectroscopic data for 5 were recorded in CDCl3,
about 11 h later, it was observed that 5 almost completely
converted into 8 in the NMR tube (Figure S72). It is
speculated that 4 and 8 might be artificial products derived
Puniceloid C (7) shared the same molecular formula of
1
C H N O as 6 on the basis of HRESIMS data. The H and
from 1−3, 10, and 5−7, respectively. As CH OH was used in
1
9
17
3
4
3
1
3
C NMR data for 6 and 7 showed great similarity. Analysis of
the isolation process and compounds 10 and 3 are hemi-
aminals, there is a strong possibility that 1 and 2 are isolation
artifacts, with 10 and 3 being the original natural products. A
similar situation exists for compound 5. So, 1, 2, and 5 are also
likely artifacts derived from 10, 3, and 6, respectively. The
oxepin-containing alkaloids 1−3, 9, and 10, quinazolinone
alkaloids 5−7, and their elimination products 4 and 8 exhibited
three distinctive UV spectra, respectively (Figures S73 and
S74). It is easy to differentiate the skeletons of these kinds of
compounds according to their UV characteristics.
the HMBC and COSY spectra suggested 6 and 7 had the same
planar structure. However, the NOESY spectrum of 7 showed
an NOE correlation of H -15 with 3-OH, which indicated that
3
CH -15 and 3-OH were cofacial. The absolute configuration of
3
C-14 in 7 was also determined to be R by Marfey’s method.
Correspondingly, the configuration of C-3 in 7 was inferred to
be R. Thus, 6 and 7 were a pair of epimers at C-3.
Puniceloid D (8) had the molecular formula C H N O
based on HRESIMS data. The H and C NMR data for 8
were similar to those for 5, with the additional presence of a
1
9
15
3
3
1
13
The known compounds 9 and 10 had been reported to show
6
trisubstituted double bond between C-3 (δ 126.0, C) and C-
moderate transcriptional activation of LXRα. Therefore, the
C
1
6 (δ 117.2, CH) and the absence of signals for a methoxy, an
analogues 1−10 were tested for their transactivation effects on
C
oxygenated nonprotonated carbon, and a methylene in 8. The
planar structure of 8 was confirmed as shown by analysis of the
HMBC and COSY spectra. The NOESY correlation of H-16
with 7-OH indicated that the double bond between C-3 and
C-16 was also of the Z-configuration as that in 4. The
configuration of C-14 was determined to be R by Marfey’s
method.
Compounds 1−3 and 10 were relatively stable in neutral
solvent, while they were unstable in acidic solvent. We had
attempted to isolate them by preparative HPLC with an ODS
column using CH OH/H O (containing 0.03% trifluoroacetic
LXRα by the same bioassay methods as reported in the
6
literature. The results (Table 3) showed that 1−3 and 5−8
had significant transcriptional activation on LXRα with EC
5
0
values of 1.7−16 μM, and 7 and 8 were the most potent
agonists, with EC values of 1.7 μM. It was obvious that the
5
0
transactivation effects of quinazolinone alkaloids 5−8 were
more potent than those of oxepine-containing alkaloids 1−4, 9,
and 10. The values of corresponding maximum fold induction
of 1−8 were from 3.0 to 4.2. Comparison of the structures and
transactivation effects on LXRα of 1−10 indicated that when
the benzene ring moiety was converted to an oxepin unit in
3
2
D
DOI: 10.1021/acs.jnatprod.9b00055
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
instrument. Optical rotations were measured with an MCP 500
polarimeter (Anton Paar). ECD and UV spectra were measured with
a Chirascan circular dichroism spectrometer (Applied Photophysics
Ltd.). IR spectra were recorded with an IR Affinity-1 Fourier
Table 3. Transcriptional Activation of LXRα of 1−10
LXRα
compound
EC50 (μM)
maximum fold induction
1
13
transform infrared spectrophotometer (Shimadzu). H, C, and 2D
NMR spectra were acquired with a Bruker AV-500 MHz NMR
spectrometer with tetramethylsilane as reference. ESIMS and
HRESIMS spectroscopic data were acquired with an amaZon SL
ion trap mass spectrometer and MaXis quadrupole-time-of-flight mass
spectrometer (Bruker), respectively. The crystallographic data were
collected on a Rigaku MicroMax 007 diffractometer equipped with Cu
Kα radiation and a graphite monochromator. Semipreparative
reversed-phase (SP-RP) HPLC was performed on a Shimadzu LC-
1
2
3
4
5
6
7
8
9
15
15
16
3.2
3.7
3.4
3.0
4.2
4.0
3.9
3.6
2.7
3.5
19.3
50
5.1
5.3
1.7
1.7
41
20A preparative liquid chromatography system with a YMC-Pack
ODS column, 250 × 20 mm, S-5 μm, 12 nm. RP-MPLC (reversed-
phase medium-pressure preparative liquid chromatography) was
carried out using the CHEETAH MP200 system (Agela Technolo-
gies) and Claricep Flash columns filled with ODS (40−63 μm, YMC).
Sephadex LH-20 (GE Healthcare) was used for column chromatog-
raphy (CC). Silica gel (200−300 mesh) for CC and GF254 for thin-
layer chromatography (TLC) were obtained from Yantai Jiangyou
Silica Gel Development Co., Ltd. Sea salts were obtained from
Guangzhou Hai Li Aquarium Technology Co., Ltd.
1
0
47
0.53
TO901317
this kind of alkaloid, their bioactivities would decrease to a
certain extent, implying the quinazolinone skeleton plays an
important role in the bioactivity. Previous studies reported that
quinazolinones LN6500 and IMB-170 were potent LXRα
2
,7
agonists. This study further confirms that the quinazolinone
skeleton can be a model for synthesizing LXR agonists.
Compounds 1−10 were also tested for their enzyme
inhibitory activity toward five different phosphatases including
TCPTP, SHP1, MEG2, SHP2, and PTP1B and two other
enzymes, IDO1 and LDHA. The results (Table S2) showed
that 8−10 selectively exhibited inhibition against the six
individual enzymes, with IC values of 14−87 μM, while 1−7
did not show inhibition activity, and no compounds showed
inhibition activity against LDHA. Comparison of the structures
and inhibitory activities between 1−4, 9, and 10 suggested that
the substituents on the oxepin-containing diketopiperazine
skeleton could significantly affect their enzyme inhibitory
activity.
Furthermore, antifungal activities of compounds 1−10 and
fractions A and B that mainly contained 1−8 and 10 were also
evaluated. Four phytopathogenic fungi including Curvularia
australiensis, Colletotrichum gloeosporioides, Fusarium oxyspo-
rum, and Pyricularia oryzae were used as indicator strains. A
preliminary antifungal assay of fractions A and B was
performed by a standard disc diffusion method. Fraction B
exhibited obvious antifungal activity against P. oryzae at 100
μg/disc, while fraction A did not show antifungal activity.
HPLC analysis profiles of fractions A and B, tested at the same
concentration (Figure S78), displayed that the contents of 5−8
in fraction B were higher than in fraction A. It was speculated
that 5−8 might play important roles in the antifungal activity
of fraction B. Thus, compounds 1−10 and fractions A and B
were further evaluated for their antifungal activities against the
above indicator strains by the microbroth dilution method in a
Fungal Materials. The fungus Aspergillus puniceus SCSIO z021
was isolated from deep-sea sediments collected from Okinawa Trough
(
27°34.01′ N and 126°55.59′ E, ∼1589 m depth) that was about 4.7
8
km away from active hydrothermal vents. The strain was identified
according to the ITS rDNA sequence (GenBank accession number
KX258801) and deposited in the RNAM Center, South China Sea
Institute of Oceanology, Chinese Academy of Science.
Fermentation and Extraction. The strain was cultured on a
potato dextrose agar (PDA) plate containing 3% sea salt at 28 °C for
days. Then the spores were collected and mixed with liquid medium
to make a spore suspension, which was transferred into 210 × 1 L
Erlenmeyer flasks, each containing 300 mL of culture medium (2%
glucose, 20% potato, 3% sea salt). Static fermentation was carried out
at 26 °C for 25 days; after that, the broth and mycelia were separated
by cheesecloth. The broth was extracted with XAD-16 resin,
50
6
successively eluting with H
collected and concentrated under vacuum to obtain a broth extract
11 g). The mycelia were extracted with acetone. The acetone fraction
O and EtOH. The EtOH fraction was
2
(
was evaporated under reduced pressure to afford an aqueous solution,
which was further extracted with EtOAc three times to yield a mycelia
extract (77 g). HPLC analysis (Figure S77) showed that the chemical
profiles of the broth and mycelia extracts were similar, but the main
components were obviously different. In order to enrich the trace
components, the extracts from the broth and mycelia were combined
for further processing.
Isolation and Purification. The extracts (88 g) were fractionated
on a normal-phase column using a stepped gradient elution with
CH Cl /CH OH (v/v, 100:0, 98:2, 95:5, 92:8, 90:10, 80:20, 70:30)
2
2
3
to obtain nine subfractions (Fr.1−Fr.9). Fr.5 (2.7 g) was filtered to
remove most of the insoluble white crystals, and the remaining
residue was subjected to an ODS column eluting with CH OH/H O/
3
2
TFA (v/v/v, from 27:73:0.03 to 100:0:0.03) to give 14 fractions
Fr.5-1−Fr.5-14). Fr.5-7 was separated by Sephadex LH-20 eluting
with CH Cl /CH OH (v/v, 1:1) to give four fractions, and then Fr.5-
(
9
6-well culture plate at a concentration of ∼0.6 mM and 200
2
2
3
μg/mL, respectively. However, these compounds showed low
7-1 was further purified by preparative HPLC eluting with CH OH/
3
percent inhibition (<50%, see Table S1) at a concentration of
H O/TFA (v/v/v, 64:36:0.03, 5 mL/min) to give 4 (11.2 mg, t =
2
R
∼
0.6 mM, so their MIC values were not evaluated.
Comparison of the HPLC profiles of the broth and mycelia
35.0 min) and 8 (6.9 mg, t
R
= 30.1 min). Fr.5-8 was separated by
preparative HPLC eluting with CH OH/H O/TFA (v/v/v,
3
2
6
6:34:0.03, 5 mL/min) to give 9 (40.2 mg, tR = 33.1 min). Fr.6
extracts of the strain (Figure S77) indicated that all of the
above alkaloids mainly existed in the fermentation broth, and
the main components of the mycelia remain to be determined
and are the subject of continuing studies.
(10 g) was subjected to a silica gel column using CH Cl /acetone (v/
2
2
v, 100:0, 95:5, 90:10, 80:20, 70:30, 50:50) to give nine fractions (Fr.6-
−Fr.6-9). Fr.6-6 was separated on an ODS column eluting with
CH OH/H O/TFA (v/v/v, from 18:82:0.03 to 100:0:0.03) to afford
1
3
2
1
9 fractions. Fr.6-6-10 was purified by preparative thin-layer
chromatography (prep-TLC) using CH Cl /CH OH (v/v, 50:3) as
the developing solvent to yield 2 (5.3 mg, R = 0.92), 1 (3.3 mg, R =
EXPERIMENTAL SECTION
■
2
2
3
General Experimental Procedures. Melting points were
measured with an SGWX-4 digital display micromelting point
f
f
0.67), 3 (4.2 mg, R = 0.53), and 10 (4.0 mg, R = 0.25). Fr.6-6-11 was
f
f
E
DOI: 10.1021/acs.jnatprod.9b00055
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
purified by the same method as Fr.6-6-10 to yield 5 (4.4 mg, R =
parameters. All hydrogen atoms were located on idealized positions
and refined as riding atoms with the relative isotropic parameters.
Crystallographic data for 4 and 5 in this paper have been deposited
with the Cambridge Crystallographic Data Centre as supplementary
publication numbers CCDC 1885363 and 1885365, respectively.
Crystallographic data for 4: C H N O ·CH OH, fw = 365.38;
f
0
.60), 7 (2.0 mg, R = 0.45), and 6 (2.0 mg, R = 0.29).
f
f
2
5
Oxepinamide H (1): pale yellow powder; [α]
+10.7 (c 0.10,
D
CH OH); UV (CH OH) λ (log ε) 258 (3.68), 348 (3.51) nm;
3
3
max
ECD (0.55 mM, CH OH) λ (Δε) 224 (−0.34), 265 (−6.66), 351
3
max
(
+1.28) nm; IR (film) νmax 3188, 3048, 1668, 1574, 1526, 1456, 1429,
19
15
3
3
3
3
1
394, 1317, 1229, 1200, 1115, 1084, 1020, 881, 775, 742, 725, 707
orange crystal from CH OH; crystal size = 0.2 × 0.2 × 0.1 mm ; T =
3
−
1
1
13
cm ; H and C NMR, Table 1; HRESIMS m/z 388.1265 [M +
99.9(2) K; monoclinic, space group P2 ; unit cell parameters: a =
1
+
Na] (calcd for C H N O Na, 388.1268).
9.6126(2) Å, b = 7.02930(10) Å, c = 12.7647(3) Å, α = 90°, β =
20
19
3
4
2
5
D
3
3
Oxepinamide I (2): pale yellow powder; [α]
−235 (c 0.10,
97.404(2)°, γ = 90°, V = 855.32(3) Å , Z = 2, D = 1.419 g/cm ,
calc
1
−
CH OH); UV (CH OH) λ (log ε) 257 (3.65), 348 (3.48) nm;
3
3
max
F(000) = 384.0, μ = 0.829 mm . A total of 8468 reflections were
collected, which yielded 3335 independent reflections. The final
refinement gave R = 0.0471, wR = 0.1252 [I ≥ 2σ(I)], S = 1.109,
ECD (0.55 mM, CH OH) λ (Δε) 211 (−1.99), 226 (+4.77), 257
3
max
(
+0.56), 267(+0.89), 343 (−3.45) nm; IR (film) νmax 3198, 2941,
1
2
1
1
3
668, 1578, 1541, 1456, 1429, 1394, 1315, 1227, 1182, 1107, 1051,
Flack parameter = −0.03(13).
−
1
1
13
022, 742, 702 cm ; H and C NMR, Table 1; HRESIMS m/z
Crystallographic data for 5: 2 × C H N O , fw = 730.76;
colorless crystal from CH OH; crystal size = 0.3 × 0.2 × 0.02 mm ; T
2
0
19
3
4
+
66.1453 [M + H] (calcd for C H N O , 366.1448).
3
2
0
20
3
4
3
2
5
Oxepinamide J (3): pale yellow powder; [α]
−212 (c 0.10,
D
= 100.0(4) K; triclinic, space group P1; unit cell parameters: a =
7.5258(2) Å, b = 8.4100(3) Å, c = 15.2735(4) Å, α = 84.233(3)°, β =
CH OH); UV (CH OH) λ (log ε) 258 (3.56), 338 (3.39) nm;
3
3
max
ECD (0.57 mM, CH OH) λ (Δε) 211 (−3.20), 224 (+3.35), 250
3
3
max
80.971(3)°, γ = 67.427(3)°, V = 880.71(5) Å , Z = 1, D = 1.378 g/
calc
−1.10), 272 (+0.34), 344 (−2.52) nm; IR (film) ν 3236, 1670,
3
−1
max
cm , F(000) = 384.0, μ = 0.805 mm . A total of 17 342 reflections
were collected, which yielded 6509 independent reflections. The final
refinement gave R = 0.0356, wR = 0.0922 [I ≥ 2σ(I)], S = 1.066,
582, 1541, 1456, 1394, 1313, 1225, 1103, 1047, 1024, 1005, 762,
−
1
1
13
44, 704 cm ; H and C NMR, Table 1; HRESIMS m/z 374.1103
+
1
2
19 17 3 4
Flack parameter = 0.14(7).
Oxepinamide K (4): orange crystal (CH OH); mp 255−261 °C;
3
Determination of the Absolute Configuration of the
Alanine Residues from 1−8. Compounds 1−8 (0.3 mg each)
were dissolved in 1 mL of 6 N HCl and heated at 115 °C for 21 h.
The hydrolysate was extracted with EtOAc to remove organic
substances and obtain an aqueous phase. Then it was dried and
25
[α]
+243 (c 0.10, CH OH); UV (CH OH) λ
(log ε) 239
D
3
3
max
(
(
(
3
1
3
3.84), 321 (3.79), 376 (3.86) nm; ECD (0.60 mM, CH OH) λmax
3
Δε) 208 (+6.68), 210 (+6.53), 222 (+9.86), 236 (+6.26), 249
+8.26), 280 (−1.80), 303 (−2.84), 375 (+2.25) nm; IR (film) ν
max
200, 2922, 2853, 1699, 1672, 1506, 1435, 1389, 1366, 1312, 1221,
dissolved in 100 μL of H O. To this solution were added 20 μL of 1
−1
1
13
2
186, 1100, 760 cm ; H and C NMR, Table 1; HRESIMS m/z
56.0998 [M + Na] (calcd for C H N O Na, 356.1006).
M NaHCO and 100 μL of a 2 mg/mL FDAA (Marfey’s reagent)
+
3
1
9
15
3
3
acetone solution, and the mixture was incubated at 40 °C for 1 h. The
reaction was quenched by the addition of 20 μL of 2 N HCl; then the
Puniceloid A (5): white powder (generally) or colorless crystal
2
5
(
(
CH OH); mp 272−277 °C; [α]
+10.8 (c 0.10, CH OH); UV
CH OH) λ (log ε) 242 (3.67), 298 (3.63), 323 (3.67), 337 (2.63)
3
D
3
dried mixture was dissolved in CH OH for HPLC analysis. The
3
3
max
standards D/L-Ala were also derivatized with FDAA in the same way.
All of the samples were analyzed with HPLC under the following
conditions: YMC-Pack ODS-A column, 250 × 4.6 mm, S-5 μm, 12
nm; ECD (0.55 mM, CH OH) λ (Δε) 209 (+22.60), 224 (−5.46),
3
max
2
47 (−0.65), 254 (−2.98), 257 (−2.84), 261 (−3.33), 292 (+2.87)
nm; IR (film) ν 3420, 1668, 1593, 1580, 1456, 1393, 1366, 1312,
max
−
1
1
13
nm; mobile phase: CH
gradient starting with 15% CH
3
CN/H
CN and finishing with 45% CH
3
2
O (0.03% TFA in H
2
O), linear
1
2
3
202, 1136, 1065, 1024, 760, 717, 700 cm ; H and C NMR, Table
; HRESIMS m/z 388.1272 [M + Na] (calcd for C H N O Na,
+
3
CN
2
0
19
3
4
over 45 min; flow rate was 1 mL/min, with UV detection at an
absorbance of 340 nm. Retention times for the standards were 19.56
min for L-Ala-DAA, 24.39 min for D-Ala-DAA, and 20.63 min for
FDAA. Through analysis of the DAA derivatives of these hydrolysates,
an obvious peak at 24.4 min was observed, suggesting all of the Ala
residues were the D-form in these compounds.
88.1268).
Puniceloid B (6): white powder; [α]25 +28 (c 0.05, CH OH); UV
D
3
(
CH OH) λ (log ε) 240 (4.04), 295 (3.52), 323 (3.59), 335 (2.55)
3 max
nm; ECD (0.28 mM, CH OH) λ (Δε) 209 (+17.63), 225 (−3.61),
3
max
2
45 (−0.64), 253 (−2.08), 256 (−1.91), 259 (−2.21), 289 (+2.43)
nm; IR (film) ν 3416, 3225, 2941, 1663, 1599, 1456, 1396, 1317,
max
−
1
1
Transactivation Effects of 1−10 on LXRα. Human hepatocyte
L02 cells were cultured in RPMI-1640 medium supplemented with
1
302, 1225, 1207, 1105, 1057, 1020, 872, 764, 717, 700 cm ; H and
1
3
+
C NMR, Table 2; HRESIMS m/z 374.1107 [M + Na] (calcd for
1
0% fetal bovine serum at 37 °C in a humidified 5% CO incubator
2
4
C H N O Na, 374.1111).
19
17
3
4
2
5
and seeded into 96-well plates at a concentration of 3 × 10 cells/well
Puniceloid C (7): white powder; [α] −125 (c 0.10, CH OH);
D
3
1
day before transfection. The transfection was performed as follows.
UV (CH OH) λ (log ε) 239 (4.02), 290 (3.47), 323 (3.54), 335
3
max
Test compounds, negative control (0.1% DMSO), and positive
control (TO901317) were added 6 h after transfection. Following
incubation for 24 h, cells were lysed, and the luciferase activity was
(
(
2.50) nm; ECD (0.57 mM, CH OH) λ (Δε) 207 (−17.14), 224
3
max
−7.57), 233 (−8.26), 259 (+1.54), 322 (−1.32) nm; IR (film) ν
max
3
1
335, 3217, 2941, 1682, 1653, 1599, 1456, 1393, 1306, 1223, 1178,
4
,6
−
1
1
13
detected using the Luciferase Assay System (Promega).
153, 1020, 872, 762, 704 cm ; H and C NMR, Table 2;
+
Phosphatase Enzyme Inhibition Assays. Recombinant human
phosphatase PTP1B, PTP1B, SHP1, SHP2, MEG2, and TCPTP were
expressed in Escherichia coli and purified. The enzyme inhibition
assays were measured using p-nitrophenyl phosphate (pNPP) as a
substrate in a 96-well plate with a final volume of 100 μL. Human
recombinant PTP1B, SHP1, SHP2, MEG2, or TCPTP (0.05 μg) in
HRESIMS m/z 352.1296 [M + H] (calcd for C H N O ,
1
9
18
3
4
3
52.1292).
Puniceloid D (8): yellow crystal (CHCl /CH OH); mp 238−245
3
3
C; [α]25 +153 (c 0.10, CH OH); UV (CH OH) λ (log ε) 241
°
(
(
(
D
3
3
max
3.91), 299 (3.72), 356 (3.81) nm; ECD (0.60 mM, CH OH) λmax
3
Δε) 201 (+12.10), 237 (−6.13), 269 (+2.12), 285 (+0.04), 320
5
0 μL of reaction buffer (pH 6.5) containing 50 mM HEPES, 100
mM NaCl, 1 mM EDTA, and 1 mM dithiothreitol and test
compounds was added to each well of a 96-well plate. Na VO was
+3.29), 353 (+1.13), 369 (+1.10) nm; IR (film) ν 3377, 2922,
max
2
852, 1668, 1593, 1580, 1440, 1389, 1312, 1215, 1103, 1057, 750
−
1
1
13
cm ; H and C NMR, Table 2; HRESIMS m/z 356.1003 [M +
Na] (calcd for C H N O Na, 356.1006).
3
4
+
used as the positive control and DMSO as the negative control to
evaluate the high-throughput screening system. After preincubation
for 15 min at room temperature, 50 μL of reaction buffer containing
50 mM pNPP was added and incubated at 37 °C for 60 min. The
phosphatase activity was determined by measuring the absorbance at
405 nm for the amount of produced p-nitrophenol. IC50 values were
determined by analyzing the data using Gen5 software (Synergy2-
19
15
3
3
X-ray Crystallographic Analyses of 4 and 5. The crystallo-
graphic data were collected on a Rigaku MicroMax 007 diffractometer
equipped with Cu Kα radiation and a graphite monochromator. The
structure was solved by direct methods with the SHELXTL software
package and refined by full-matrix least-squares techniques. All non-
hydrogen atoms were refined with anisotropic displacement
F
DOI: 10.1021/acs.jnatprod.9b00055
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Multi-Mode microplate reader, BioTek Instruments, Inc.). Each
compound was assayed in triplicate.
IDO Enzyme Inhibition Assays. Recombinant human IDO was
ACKNOWLEDGMENTS
■
The authors are grateful for the financial support provided by
the National Natural Science Foundation of China (81673326
and 41376160), Regional Innovation Demonstration Project of
Guangdong Province Marine Economic Development
(GD2012-D01-002), Strategic Leading Special Science and
Technology Program of Chinese Academy of Sciences
expressed in E. coli and purified. The enzyme inhibition assay was
9
performed as described by Dolusic et al. Details are described in the
Supporting Information.
Human LDHA Enzyme Inhibition Assays. The enzyme
10
inhibition assay was performed as described by Granchi et al.
Details are described in the Supporting Information.
(
XDA100304002), and National Marine Public Welfare
Antifungal Bioassays. Four strains of phytopathogenic fungi,
including Curvularia australiensis BY01, Colletotrichum gloeosporioides
HNM 1003, Fusarium oxysporum HNM 1003, and Pyricularia oryzae
HNM 1003, were used as tested fungus. They were isolated from
lesions on the leaves of pineapple (Ananas comosus), yam (Dioscorea
alata), banana (Musa nana), and rice (Oryza sativa) and deposited in
the RNAM Center, South China Sea Institute of Oceanology, Chinese
Academy of Science.
Research Project of China (201305017). The authors also
appreciate the analytical facility center (Z. Xiao, A. Sun) of the
South China Sea Institute of Oceanology, Chinese Academy of
Sciences, for acquiring NMR, HRESIMS, and experimental
CD data and the use of the HPCC for all numeric simulations.
REFERENCES
■
(
Antifungal activities of fractions A and B were screened by the disc
diffusion assay. Briefly, the tested fungi were cultured on the PDA
plate for 6−8 days at 30 °C. Then some of the mycelia were
transferred into PD medium, and the fungal suspensions were mixed.
The mixture was filtered by cotton to obtain the spore suspension,
(
1) Wojcicka, G.; Jamroz-Wisniewska, A.; Horoszewicz, K.;
Beltowski, J. Postepy Hig Med. Dosw (Online). 2007, 61, 736−759.
2) Li, N.; Wang, X.; Zhang, J.; Liu, C.; Li, Y.; Feng, T.; Xu, Y.; Si, S.
Biochem. Pharmacol. 2014, 92, 438−447.
3) El-Gendy, B. E. M.; Goher, S. S.; Hegazy, L. S.; Arief, M. M. H.;
Burris, T. P. J. Med. Chem. 2018, 61, 10935−10956.
4) Lu, X. H.; Shi, Q. W.; Zheng, Z. H.; Ke, A. B.; Zhang, H.; Huo,
(
6
which was further diluted to approximately 1 × 10 CFU with PD
medium. The diluted spore suspensions (100 μL) were inoculated
and spread on a PDA plate. The two samples were made up to 20
mg/mL in DMSO. Each paper disc loaded with 100 μg of sample was
placed on the inoculated plate. DMSO (5 μL/disc) and amphotericin
B (5 μg/disc) were used as negative and positive controls,
respectively. The inhibition zone was observed after the plates were
incubated at 30 °C for 2 or 3 days.
(
C. H.; Ma, Y.; Ren, X. A.; Li, Y. Y.; Lin, J.; Jiang, Q.; Gu, Y. C.; Kiyota,
H. Eur. J. Org. Chem. 2011, 2011, 802−807.
(5) Fujii, K.; Ikai, Y.; Mayumi, T.; Oka, H.; Suzuki, M.; Harada, K.
Anal. Chem. 1997, 69, 3346−3352.
6) Zheng, Z. H.; Lu, X. H.; Zhang, H.; Lv, G. P.; He, J. G.; Zhao, B.
H.; Si, S. Y. Anal. Bioanal. Chem. 2010, 396, 1721−1730.
7) Albers, M.; Blume, B.; Schlueter, T.; Wright, M. B.; Kober, I.;
Kremoser, C.; Deuschle, U.; Koegl, M. J. Biol. Chem. 2006, 281,
920−4930.
8) Zhang, J.; Sun, Q.-l.; Zeng, Z.-g.; Chen, S.; Sun, L. Microbiol. Res.
015, 177, 43−52.
9) Dolusic, E.; Larrieu, P.; Blanc, S.; Sapunaric, F.; Pouyez, J.;
(
The percent growth inhibition of the compounds was evaluated by
(
11,12
a microbroth dilution method in a 96-well culture plate.
Tested
samples (10 mg/mL) and positive controls (5 mg/mL) including
hymexazol, carbendazim, and amphotericin B were dissolved in
4
(
2
(
DMSO, respectively. PD medium (48 μL), 2 μL of tested sample, and
6
5
0 μL of spore suspensions diluted with PD medium (1 × 10 CFU)
were added to 96-well microplates successively. The negative control
was treated with 2 μL of DMSO. The optical density (OD) of each
well was measured with a microplate reader at 600 nm when the plate
was incubated at 28 °C for 0 and 48 h, respectively. The percent
growth inhibition was calculated by the following formula:
Moineaux, L.; Colette, D.; Stroobant, V.; Pilotte, L.; Colau, D.;
Ferain, T.; Fraser, G.; Galleni, M.; Frere, J. M.; Masereel, B.; Van den
Eynde, B.; Wouters, J.; Frederick, R. Eur. J. Med. Chem. 2011, 46,
3
(
058−65.
10) Granchi, C.; Roy, S.; Giacomelli, C.; Macchia, M.; Tuccinardi,
T.; Martinelli, A.; Lanza, M.; Betti, L.; Giannaccini, G.; Lucacchini, A.;
Funel, N.; Leon, L. G.; Giovannetti, E.; Peters, G. J.; Palchaudhuri, R.;
Calvaresi, E. C.; Hergenrother, P. J.; Minutolo, F. J. Med. Chem. 2011,
percent inhibition = (C − T)/C × 100%
where C and T represent the OD48h − OD of the negative control
0h
5
4, 1599−612.
11) Li, X. J.; Zhang, Q.; Zhang, A. L.; Gao, J. M. J. Agric. Food
Chem. 2012, 60, 3424−3431.
12) Wang, J. F.; He, W. J.; Huang, X. L.; Tian, X. P.; Liao, S. R.;
Yang, B.; Wang, F. Z.; Zhou, X. J.; Liu, Y. H. J. Agric. Food Chem.
016, 64, 2910−2916.
well and treated well, respectively. Each sample was independently
performed at least three times.
(
(
ASSOCIATED CONTENT
■
2
*
S
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.9b00055.
Human LDHA enzyme inhibition assay and IDO
enzyme inhibition assay; 1D/2D NMR and HRESIMS
spectra of 1−8 (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*Tel: (86) 020-89022112. E-mail: shuhuaqi@scsio.ac.cn.
ORCID
Shuhua Qi: 0000-0003-3802-3139
Notes
The authors declare no competing financial interest.
G
DOI: 10.1021/acs.jnatprod.9b00055
J. Nat. Prod. XXXX, XXX, XXX−XXX