Anti-inflammatory amino acid derivatives from the ascidian Herdmania momus

Article abstract of DOI:10.1021/np200397g

Four new amino acid derivatives, herdmanines A-D (1-4), were isolated from the marine ascidian Herdmania momus. Herdmanines A-C contain the unusual d-form of arginine. Compounds 3 and 4 had a moderate suppressive effect on the production of NO, with IC₅₀ values of 96 and 9 μM, respectively. These compounds were found to inhibit the mRNA expression of iNOS. The inhibitory activities on the production and mRNA expression of pro-inflammatory cytokines such as TNF-α, IL-1β, and IL-6 were evaluated.

Full text of DOI:10.1021/np200397g

ARTICLE
pubs.acs.org/jnp
Anti-inflammatory Amino Acid Derivatives from the Ascidian
Herdmania momus
Jian Lin Li,† Sang Chul Han,‡ Eun Sook Yoo,‡ Sook Shin,§ Jongki Hong,^ Zheng Cui,|| Huayue Li,†
and Jee H. Jung*,†
†College of Pharmacy, Pusan National University, Busan 609-735, Korea
‡School of Medicine, Jeju National University, Jeju 690-756, Korea
§Department of Life Science, Sahmyook University, Seoul 139-742, Korea
^College of Pharmacy, Kyung Hee University, Seoul 130-701, Korea
Shenyang Pharmaceutical University, Shenyang 110016, People's Republic of China
S Supporting Information
b
ABSTRACT:
Four new amino acid derivatives, herdmanines AꢀD (1ꢀ4), were isolated from the marine ascidian Herdmania momus.
Herdmanines AꢀC contain the unusual D-form of arginine. Compounds 3 and 4 had a moderate suppressive effect on the
production of NO, with IC50 values of 96 and 9 μM, respectively. These compounds were found to inhibit the mRNA expression of
iNOS. The inhibitory activities on the production and mRNA expression of pro-inflammatory cytokines such as TNF-R, IL-1β, and
IL-6 were evaluated.
arine ascidians are a very rich source of chemically unique
and biologically active secondary metabolites that have
biological evaluation of the amino-acid-derived metabolites of
the ascidian H. momus.
M
attracted the interest of both chemists and pharmacologists.1
Ascidians are particularly rich in amino-acid-derived metabolites
such as cyclic peptides and alkaloids. The cyclic depsipeptide
didemnin B, isolated from the ascidian Trididemnum solidum, is
the first marine natural product to enter phase I and II clinical
trials as an anticancer agent, but discontinued for its high cardiac
and neuromuscular toxicity.2
’ RESULTS AND DISCUSSION
Combined MeOH and CH2Cl2 extracts of H. momus were
subjected to solvent partitioning between CH2Cl2 and H2O. The
H2O layer was partitioned with n-BuOH, and the n-BuOH-
soluble fraction was subjected to chromatographic separation
and purification to afford a series of amino-acid-derived meta-
bolites (1ꢀ4).
The solitary tunicate Herdmania momus (order Pleurogona,
family Pyuridae) is a frequently encountered species distributed
primarily in the Indo-Pacific Ocean. However, only one chemical
study of this organism has been reported.3 Over the past few
years, there has been an unusual bloom of this ascidian around
the Korean peninsula, and this has caused severe economic and
ecological problems. In an approach to utilize H. momus in a
productive manner, we performed a chemical investigation of the
bioactive metabolites of this ascidian, which we collected off the
coast of Jeju Island, Korea. By solvent partition and chromatog-
raphy, we isolated four new amino acid derivatives, herdmanines
AꢀD (compounds 1ꢀ4), and evaluated their anti-inflammatory
activity. In this paper, we describe the structure elucidation and
Herdmanine A (1) was obtained as a yellow gum. The (+)-
FABMS spectrum showed a pseudomolecular ion peak at m/z
438 [M + H]+, and the molecular formula of this compound was
established as C22H23N5O5 on the basis of the (+)-HRFABMS
and NMR data, indicating 14 degrees of unsaturation.
The 1H NMR spectrum of 1 (Table 1) revealed resonances of
an AA0XX0 spin system at δH 7.15 (H-10/14) and 7.81 (H-11/
13), attributable to a p-disubstituted phenyl moiety. This was
further corroborated by COSY data (Figure 1). Furthermore, an
Received: May 11, 2011
Published: July 19, 2011
r
Copyright
2011 American Chemical Society and
1792
dx.doi.org/10.1021/np200397g J. Nat. Prod. 2011, 74, 1792–1797
American Society of Pharmacognosy
|

Journal of Natural Products
ARTICLE
1
Table 1. H (500 MHz) and 13C (100 MHz) NMR Data for 1a
and 2b
herdmanine A (1)
δH (J in Hz)
12.17, brs
herdmanine B (2)
δC δH (J in Hz)
c
position
1-NH
δC
2
134.7 8.33, s
133.4 8.01, s
106.0
3
105.9
3a
126.5
120.9
4
121.0 8.02, dd (6.0, 1.0)
122.3 7.87, d (8.5)
5
122.4 7.24, ddd (7.5, 6.0, 1.5) 113.1 6.78, dd (8.5, 2.0)
123.4 7.26, ddd (7.5, 6.0, 1.0) 155.2
6
7
113.3 7.42, dd (6.0, 1.5)
137.2
98.4 6.88, d (2.0)
7a
139.4
8
163.8
164.8
9
153.1
155.7
10
122.2 7.15, d (8.5)
129.0 7.81, d (8.5)
132.3
123.3 7.33, dd (8.5, 2.0)
129.7 7.95, dd (8.5, 2.0)
132.7
11
12
13
129.0 7.81, d (8.5)
122.2 7.15, d (8.5)
165.1
129.7 7.95, dd (8.5, 2.0)
123.3 7.33, dd (8.5, 2.0)
168.5
14
15
16-NH
17
8.78, d (9.0)
55.1 4.19, m
29.9 1.87, m
1.81, m
indole spin system was suggested, which was further supported
by the UV absorption at 283 nm. In the HMBC spectrum
(Figure 1), the correlations from H-4 to C-3, C-6, and C-7a,
from H-5 to C-3a and C-7, from H-6 to C-4 and C-7a, and from
H-7 to C-5 and C-3a were assigned to an indole moiety.
Correlations from the indole proton H-2 to C-3, C-3a, and
C-7a were also observed in the HMBC spectrum. The H-4 and
H-2/C-2 signals were significantly shifted downfield owing to the
carbonyl substitution at C-3 of the indole group, and a correla-
tion from H-2 to the carbonyl carbon C-8 (δC 163.8) was
observed in the HMBC spectrum. The carbon resonance at δC
55.1 and the resonance due to the associated proton at δH 4.19
indicated the presence of an amino acid moiety, and the carbon
signal at δC 158.0 (C-22) was in good agreement with the
expected value for a terminal guanidyl group. The presence of the
guanidyl group was further corroborated by the formation of a
pyrimidine derivative in the reaction between 1 and acetylace-
tone (Figure 2). Peaks due to six methylene protons, a methine
proton, and two exchangeable protons appeared at δH 8.78 (16-
NH) and 9.29 (21-NH), and these protons were thought to be
associated with a 2,5-diaminopentanoic acid moiety. Successive
COSY connections from 16-NH to 21-NH suggested the pre-
sence of an arginine moiety. The amide proton (16-NH) and R-
proton (H-17) of arginine showed HMBC correlations to the
carbonyl carbon C-15, which in turn showed correlations with
the aromatic protons H-11/13, indicating that 1 contained a p-
oxybenzoyl group directly attached to an arginine moiety. On the
basis of the molecular formula of 1, it was assumed that the N-(p-
oxybenzoyl)arginine and 3-carboxyindole units were joined via
an ester linkage. This was further supported by the chemical
shifts of C-8 (δC 163.8) and C-9 (δC 153.1) and the results of
FAB-CID-MS/MS analysis (Figure 3). The arginine moiety was
suggested to be in the D-form by comparison of the specific
rotation of 1 ([R]23D ꢀ18) with that of the analogous D-arginine
55.8 4.50, dd (7.0, 5.0)
31.3 2.00, m
18a
18b
19
1.87, m
25.3 1.58, m
41.2 3.15, m
9.29, s
26.1 1.68, quint (7.5)
42.1 3.21, m
20
21-NH
22
158.0
157.6
23-NH
7.20, brs
24-NH2
25-COOHd 177.5
a Spectra recorded in d6-DMSO. b Spectra recorded in CD3OD.
c Chemical shifts obtained from gHSQC and gHMBC spectra. d Proton
signals were not observed.
7.20, brs
178.3
was confirmed by Marfey’s method;6 the pyrimidine derivative of
1 (1a) was hydrolyzed and allowed to react with Marfey’s reagent
to give 1c. TLC analysis showed that 1c was identical to the
compound derived from authentic D-arginine.
Herdmanine B (2) was isolated as a white powder. The
pseudomolecular ion [M + H]+ peak at m/z 454.1755 in the
(+)-HRFABMS spectrum matched well with the expected
formula C22H24N5O6. The spectroscopic properties of 2 were
1
similar to those of 1. The only difference between the H NMR
spectra of 1 and 2 was in the signals associated with the indole
moiety. COSY and H NMR analysis indicated that three of
1
the aromatic protons were part of an AMX system. Correla-
tions were observed from H-4 to C-3, C-6, and C-7a, from H-5
to C-3a and C-7, and from H-7 to C-5 and C-3a. The chemical
shift of C-6 (δC 155.2) was indicative of hydroxylation. The
negative specific rotation of 2 ([R]23D ꢀ33) suggested that it
contained D-arginine, and this was confirmed by the analogy of
the specific rotation with that observed for (ꢀ)-herdmanine
A ([R]23D ꢀ18).
(5: [R]20D ꢀ133) and L-arginine model compounds (6: [R]20
Herdmanine C (3) was isolated as a white powder, and the
(+)-HRFABMS data of this compound supported the molecular
D
+3.6; 7: [R]23 +14; 8: [R]23 +26; 9: [R]23 +19.5).4,5 This
D
D
D
1793
dx.doi.org/10.1021/np200397g |J. Nat. Prod. 2011, 74, 1792–1797

Journal of Natural Products
ARTICLE
Figure 1. Selected HMBC and COSY correlations of compound 1.
Figure 3. Prominent fragmentations of the [M + H]+ ions of com-
pounds 1ꢀ3 in FAB-CID tandem mass spectrometry.
that the molecule contained a 3,5,6-trisubstituted indole moiety and
a tyrosine moiety. Comparison of the spectroscopic data with those
of 6-bromo-5-hydroxyindole6 and the chemical shift of the qua-
ternary carbons of the indole ring, δC 150.0 (C-5) and 107.0 (C-6),
indicated that those positions were substituted by a hydroxy group
and a bromine atom, respectively. A correlation was observed be-
tween H-2 and C-8 (δC 163.2), indicating that the C-3 position
of indole is substituted by a carboxy group. The molecular formula
of 4 suggested that these two moieties must be linked between
C-8 and C-9 via an ester linkage. The negative specific rotation
of 4 ([R]23D ꢀ4.7) suggested that the tyrosine moiety had the
L-configuration. This was corroborated by the hydrolysis of 4 and
derivatization with Marfey’s reagent, followed by TLC analysis.
Herdmanine D contains a rare 6-bromo-5-hydroxyindole moiety.
This moiety has been found only in the metabolites of the muricid
gastropod Drupella fragum,7 the sponge Oceanapia bartschi,8 and
the ascidians Syncarpa oviformis9 and Eudistoma olivaceum.10
Arginine derivatives (5ꢀ9) isolated from the ascidian Lepto-
clinides sp.4,5 are analogous to compounds 1ꢀ3. However, it is
noteworthy that compounds 1ꢀ3 include the less common D-
arginine unit. The occurrence of D-amino acids in vivo has been
shown by several authors.11,12 Specifically, microorganisms trans-
form L-amino acids to D-isomers with amino acid oxidase,
transaminases, and epimerases (racemases).13,14 From the pre-
sence of the unusual D-arginine moiety, we can speculate that
compounds 1ꢀ3 of this tunicate are the metabolites of symbiotic
microorganisms with an arginine epimerase. On the other hand,
it has been reported that the total tissue concentration of the D-
amino acids in 18 species of marine invertebrates ranges from
0.04 to 0.44 mM.15 This may support another suggestion that the
uncommon D-arginine originates from seawater, which contains
significant amounts of D-amino acids.15
Figure 2. Derivatization of compound 1 for Marfey analysis: (i) H2O
(25 μL), EtOH (50 μL), triethylamine (25 μL), and acetylacetone
(50 μL), 110 ꢀC, 4 h; (ii) 2 N HCl, 110 ꢀC, 6 h; (iii) FDAA (1-fluoro-2,4-
dinitrophenyl-5-L-alanine amide), 40 ꢀC, 2 h.
1
formula C20H22N4O6. The H NMR and 13C NMR data of 3
showed two sets of AA0XX0 spin systems, suggesting the presence
of two p-disubstituted phenyl moieties: δH 6.92 (H-2/6) and
7.79 (H-3/5); δH 7.10 (H-9/13) and 7.98 (H-10/12). The
HMBC and COSY correlations for the remaining proton signals
suggested the presence of an arginine moiety. Comparison of the
NMR data of 3 and 1 suggested that the 3-carboxyindole ring was
replaced with a p-hydroxybenzoyl moiety in the former. The
planar structure of 3, derived from the observations mentioned
above, was further corroborated by FAB-CID-MS/MS analysis
(Figure 3). The negative specific rotation of 3 ([R]23 ꢀ22)
D
suggested that it contained D-arginine.
Herdmanine D (4) was obtained as a white powder. The (+)-
FABMS spectrum showed isotopic [M + H]+ ion peaks at m/z
421/419 (the peak intensities were in a 1:1 ratio), which is
characteristic of a monobrominated compound. The molecular
formula of 4 was deduced as C18H15BrN2O5 on the basis of the
(+)-HRFABMS and NMR data. Peaks due to all 18 carbons were
visible in the 13C NMR spectrum, and HSQC analysis allowed
the appropriate assignment of 10 carbon-bound protons (seven
aromatic, one methine, and two methylene protons). The five
remaining protons were exchangeable. 1H NMR analyses and the
correlations observed in the COSY and HMBC spectrum indicated
The effects of compounds 1, 3, and 4 on LPS-induced nitric
oxide (NO) production in RAW 264.7 cells were screened by
measuring the accumulated nitrite in the culture medium.
Compounds 3 and 4 had moderate suppressive effects on the
production of NO, with IC50 values of 96 and 9 μM, respectively,
and compound 4 had minimal cytotoxicity, with a IC50 value of
1794
dx.doi.org/10.1021/np200397g |J. Nat. Prod. 2011, 74, 1792–1797

Journal of Natural Products
ARTICLE
Figure 5. Inhibitory effect of compounds 3 and 4 on PGE2 production
(A) and mRNA expression of COX-2 (B) in the LPS-stimulated
RAW264.7 cells. (A) Cells were stimulated for 24 h with LPS (1 μg/
mL) only or with LPS containing the test samples at different concen-
trations (50 and 25 μM). PEG2 amounts in the culture supernatant were
determined by ELISA. The data represent the mean ( SD of triplicate
experiments. (B) mRNA expression was determined by RT-PCR. The
graph represents the changes in the mRNA levels of COX-2 normalized
by β-actin. The data represent the mean ( SD.
Figure 4. Inhibitory effect of compounds 1, 3, and 4 on NO production
(A) and mRNA expression of iNOS of compounds 3 and 4 (B) in the
LPS-stimulated RAW264.7 cells. (A) Cells were stimulated for 24 h with
pure LPS (1 μg/mL) or with LPS containing the test samples at different
concentrations (250, 50, 10, and 2 μM). NO production (columns) was
determined using the Griess reagent method. Cell viability (lines) was
determined using the WST method. The data represent the mean ( SD
of triplicate experiments. (B) Total RNA was isolated in an RNase-free
environment, and iNOS mRNA expression was determined by Western
blotting. β-Actin was used as an internal control.
fractions. However, the aforementioned three compounds did
not show significant activity up to a concentration of 30 μg/disk
(agar diffusion assay).
170 μM (Figure 4A). To examine whether this suppression of
NO production was due to the reduced iNOS expression, RT-
PCR analysis of LPS-stimulated cells was performed. Both com-
pounds inhibited the mRNA expression of iNOS (Figure 4B). The
results indicated that these compounds inhibit NO production via
modulation of iNOS gene expression. Compounds 3 and 4 were
also applied to RAW 264.7 cells to investigate the inhibitory effect
on the production of cyclooxygenase-2 (COX-2)-derived prosta-
glandin E2 (PGE2). As depicted in Figure 5A, compounds 3 and 4
showed moderate inhibition of PGE2 production. The correlation
of PGE2 production with the mRNA expression of COX-2 was
confirmed (Figure 5B).
In a subsequent investigation on the ability of the compounds
to inhibit the production of pro-inflammatory cytokines (TNF-
R, IL-1β, and IL-6), RAW 264.7 cells were cultured with 3 or 4 in
the presence of LPS, and the respective cytokines were measured
using ELISA. Compounds 3 and 4 showed moderate suppressive
effects on the production of IL-6 (Figure 6A). Compound 4 also
showed clear inhibition of the mRNA expression of IL-6
(Figure 6B).
’ EXPERIMENTAL SECTION
General Experimental Procedures. Optical rotations were
measured using a Jasco DIP-370 digital polarimeter. UV spectra were
recorded using a Shimadzu UV-1601 UV/visible spectrophotometer.
1D and 2D NMR spectra were recorded using Varian UNITY 400 and
Varian INOVA 500 spectrometers, respectively. Chemical shifts were
reported with reference to the respective residual solvent or deuterated
solvent peaks (δH 2.50 and δC 39.7 for DMSO). FABMS data were
obtained using a JEOL JMS SX-102A. HRFABMS data were obtained
using a JEOL JMS-700 Mstation mass spectrometer. FAB accurate mass
measurement was carried out at a resolution of 10 000. HPLC was
performed using a Gilson 370 pump with a YMC-packed J’sphere ODS-
H80 column (250 ꢁ 10 mm, 4 μm, 80 Å) and a Shodex packed NH-5E
column (250 ꢁ 10 mm, 5 μm) using Shodex RI-71 and Jasco UV-975
detectors.
Animal Material. Herdmania momus was collected by hand with
scuba (15 to 25 m depth) in 2008, off the coast of Jeju Island, Korea. The
fresh specimens were immediately frozen and maintained at ꢀ20 ꢀC
until the chemical investigation. A voucher specimen (J08J-1) was
taxonomically identified by one of the authors (S.S.) and deposited at
the Laboratory of Phylogenetic Systematics, Sahmyook University.
Compounds 1, 3, and 4 were also evaluated for antibacterial
activity against human pathogenic bacteria (Staphylococcus aureus,
Escherichia coli, Klebsiella aerogenes, Salmonella typhimurium, and
Enterobacter cloacae) because 3 and 4 were isolated from antibacterial
1795
dx.doi.org/10.1021/np200397g |J. Nat. Prod. 2011, 74, 1792–1797

Journal of Natural Products
ARTICLE
Hz, H-3, H-5), 7.26 (3H, brs, 22-NH, 23-NH2), 7.10 (2H, d, J = 9.0 Hz,
H-9, H-13), 6.92 (2H, d, J = 8.5 Hz, H-2, H-4), 4.20 (1H, m, H-16), 3.15
(2H, m, H-19), 1.89 (1H, m, H-17a), 1.81 (1H, m, H-17b), 1.57 (2H,
quintet, J = 6.5 Hz, H-18); 13C NMR (DMSO, 100 MHz) δ 177.6 (C-
24), 165.0 (C-14), 164.7 (C-7), 163.5 (C-1), 157.9 (C-21), 153.1 (C-8),
132.9 (C-3, C-5), 132.5 (C-11), 129.0 (C-10, C-12), 122.0 (C-9, C-13),
119.8 (C-4), 116.3 (C-2, C-6), 55.0 (C-16), 41.2 (C-19), 29.9 (C-15),
25.3 (C-18); (+)-FABMS m/z 415 [M + H]+; (+)-HRFABMS m/z
415.1595 (calcd for C20H23N4O6, 415.1618); FAB-CID-MS/MS m/z
417 (100), 398 (1.2), 356 (0.7), 294 (2.0), 241 (3.4), 121 (1.8).
Herdmanine D (4): white powder; [R]23D ꢀ4.7 (c 0.30, MeOH); 1H
NMR (DMSO, 500 MHz) δ 12.07 (1H, brs, 1-NH), 10.14 (1H, brs, 18-
NH2), 9.95 (1H, brs, 18-NH2), 7.80 (1H, s, H-7), 7.61 (1H, s, H-4), 7.33
(2H, d, J = 8.5 Hz, H-11, H-13), 7.16 (2H, d, J = 8.5 Hz, H-10, H-14),
3.45 (1H, dd, J = 8.5, 4.5 Hz H-16), 3.16 (1H, dd, J = 14.5, 4.5 Hz
H-15a), 2.91 (1H, dd, J = 14.5, 8.0 Hz H-15b); 13C NMR (DMSO, 100
MHz) δ 170.4 (C-17), 163.2 (C-8), 150.1 (C-9), 150.0 (C-5), 135.2 (C-
12), 135.1 (C-2), 131.7 (C-7a), 130.9 (C-11, C-13), 127.0 (C-3a), 122.5
(C-10, C-14), 116.8 (C-7), 107.0 (C-6), 106.3 (C-4), 105.3 (C-3), 56.1
(C-16), 37.0 (C-15). (+)-FABMS m/z 421/419 [M + H]+, 375/373
[M ꢀ COOH]+; (+)-HRFABMS m/z 421.0224/419.0223 (calcd for
C18H16BrN2O5, 421.0222/419.0242).
Preparation of Pyrimidine Derivative and Marfey Analysis
of 1. To a 1 mL vial containing pure 1 (0.5 mg) was added a mixture of
H2O (25 μL), EtOH (50 μL), triethylamine (25 μL), and acetylacetone
(50 μL), and the resulting solution was maintained at 110 ꢀC for 4 h in a
screw-cap vial. The reaction mixture was evaporated to dryness under a
stream of N2 to yield the dimethylpyrimidine derivative of 1 (1a).
Compound 1a was dissolved in 6 N HCl (1 mL) and hydrolyzed at
110 ꢀC for 24 h. The acid hydrolysate was dried under N2 to yield the
arginine dimethylpyrimidine derivative 1b. To the acid hydrolysate,
0.1% FDAA (1-fluoro-2,4-dinitrophenyl-5-L-alanine amide) solution in
acetone (100 μL) and 1 M NaHCO3 (20 μL) were added, and the
mixture was heated at 40 ꢀC for 1 h. After cooling to room temperature,
the reaction mixture containing 1c was neutralized with 2 M HCl
(10 μL) and diluted with 0.5 mL of MeOH for TLC analysis (Figure 2).
(1c and those derived from authentic D- and L-arginine showed Rf values
of 0.36 and 0.41, respectively, on a silica gel TLC plate; mobile phase, 3:1
CHCl3/MeOH.) 1H NMR data of 1b in CD3OD: δH 6.38 (1H, s), 4.05
(1H, t), 3.39 (2H, m), 2.21 (6H, s), 1.83 (2H, m), 1.65 (2H, m).
Acid Hydrolysis and Marfey Analysis of 4. Compound 4 (0.5
mg) was dissolved in 6 N HCl (1 mL) and hydrolyzed at 110 ꢀC for 24 h.
The acid hydrolysate was dried under N2. The residue was treated with
FDAA prior to TLC analysis, as described above. (D-Tyr-FDAA and
L-Tyr-FDAA showed Rf values of 0.46 and 0.54, respectively, on an RP-
18 F254 TLC plate; mobile phase, 1:1 MeOH/H2O.)
Nitrite Assay. The production of NO was measured, as previously
described by Ryu et al.,16 by using Griess reagent (Sigma). Briefly, RAW
264.7 cells were stimulated with LPS (1 μg/mL), and 100 μL of the
supernatant was mixed with 100 μL of the Griess reagent (0.1%
naphthalene diamine dihydrochloride, 1% sulfanilamide, and 2.5%
H3PO4). The mixture was incubated for 10 min at room temperature.
The absorbance at 540 nm was measured using an ELISA reader
(Amersham Pharmacia Biotech), and the results were compared against
a calibration curve using sodium nitrite as the standard.
Cytotoxicity Assay. Cell viability was determined by the WST-1
(4-[3-iodophenyl]-2-(4-nitrophenyl)-2H-5-tetrazolio)-1,3-benzene dis-
ulfonate) assay. Briefly, the cells were stimulated for 24 h with 1 μg/mL of
pure LPS or LPS containing different concentrations of compounds 1, 3,
and 4. Then, the cell proliferation reagent WST-1 was added, and the cells
were reincubated for 2 h. Thereafter, the formazan formed was quantified
by measuring the light absorbance at 450 nm using an ELISA plate reader.
Determination of PGE2. Raw 264.7 cells were seeded in wells and
incubated for 24 h. After incubation, the cells were incubated with
Figure 6. Inhibitory effect of compounds 3 and 4 on production (A)
and mRNA expression (B) of cytokines (IL-6, IL-1β, and TNF-R) in the
LPS-stimulated RAW264.7 cells. (A) The RAW264.7 cells (1.5 ꢁ 105
cells/mL) were stimulated for 24 h with pure LPS (1 μg/mL) or with
LPS containing test samples at different concentrations (50 and 25 μM).
The production of cytokines was determined using the Griess ELISA
method. The data represent the mean ( SD of triplicate experiments.
(B) Effects of compounds 3 and 4 on the mRNA expression of cytokines
in the LPS-stimulated RAW 264.7 cells. Total RNA was isolated in an
RNase-free environment, and the cytokine and β-actin mRNA levels
were measured by RT-PCR.
Extraction and Isolation. The frozen ascidians (0.4 kg) were
chopped into small pieces and extracted with MeOH and CH2Cl2 at
room temperature. The combined extract was partitioned between
CH2Cl2 and H2O. The H2O layer was further extracted by n-BuOH.
The n-BuOH fraction (6 g) was subjected to reversed-phase MPLC
column chromatography (YMC gel ODS-A, 60 Å, 230 mesh) with a step
gradient solvent system, 30% to 100% MeOH/H2O, to afford 21
fractions. Fraction 11 (37.3 mg), which gave interesting signals in the
1H NMR spectrum, was subjected to reversed-phase HPLC (eluent:
50% MeOH) to afford 1 (3.8 mg). Fraction 8 (136.9 mg), which gave 1H
NMR signals similar to those given by fraction 11, was first purified on a
Shodex packed NH-5E column using 85% MeOH as the eluent and
further purified on a YMC-packed J’sphere ODS-H80 column using 35%
MeOH as the eluent to afford 2 (1.0 mg). Fraction 9 (120.3 mg), one of
the antibacterial fractions, was subjected to reversed-phase HPLC
(eluent: 50% MeOH) to obtain 3 (2.6 mg) and 4 (5.8 mg).
Herdmanine A (1): yellow gum; [R]23D ꢀ18 (c 0.10, MeOH); 1H and
13C NMR (see Table 1); (+)-FABMS m/z 438 [M + H]+; (+)-
HRFABMS m/z 438.1774 (calcd for C22H24N5O5, 438.1771); FAB-
CID-MS/MS m/z 438 (100), 421 (0.4), 294 (2.5), 264 (1.8), 144 (3.4),
121 (0.7).
Herdmanine B (2): white powder; [R]23D ꢀ33 (c 0.09, MeOH); 1H
and 13C NMR (see Table 1); (+)-FABMS m/z 454 [M + H]+; (+)-
HRFABMS m/z 454.1755 (calcd for C22H24N5O6, 454.1727); FAB-
CID-MS/MS m/z 454 (100), 437 (0.5), 294 (1.6), 280 (1.0), 160 (2.6).
Herdmanine C (3): white powder; [R]23D ꢀ22 (c 0.10, MeOH); 1H
NMR (DMSO, 500 MHz) δ 9.17 (1H, brs, 20-NH), 8.85 (1H, d, J = 7.5
Hz, 15-NH), 7.89 (2H, d, J = 9.0 Hz, H-10, H-12), 7.79 (2H, d, J = 8.5
1796
dx.doi.org/10.1021/np200397g |J. Nat. Prod. 2011, 74, 1792–1797

Journal of Natural Products
ARTICLE
different concentrations of 3 and 4 in the presence of LPS (1 μg/mL) for
24 h. The PGE2 concentration in the culture medium was determined
using an ELISA kit.
(10) Elena, A. S.; Vladimir, A. D. Nat. Prod. Commun. 2008,
3, 1617–1620.
(11) Friedman, M. J. Agric. Food Chem. 1999, 47, 3457–3479.
(12) Fujii, N.; Saito, T. Chem. Rec. 2004, 4, 267–278.
(13) Yoshimura, T.; Goto, M. FEBS J. 2008, 275, 3527–3537.
(14) Yoshimura, T.; Esak, N. J. Biosci. Bioeng. 2003, 96, 103–109.
(15) Preston, R. L. Comp. Biochem. Physiol. 1987, 87B, 55–62.
(16) Ryu, S. Y.; Oak, M. H.; Yoon, S. K.; Cho, D. I.; Yoo, G. S. Planta
Med. 2000, 66, 358–360.
Measurement of Production of Pro-inflammatory Cyto-
kines (IL-6, IL-1β, and TNF-r). The inhibitory effects of the isolated
compounds on IL-6, IL-1β, and TNF-R production were determined by
the method previously described by Cho et al.17 The samples were
dissolved with EtOH and diluted with DMEM. The final concentration
of chemical solvents did not exceed 0.1% in the culture medium. Under
these conditions, none of the solubilized solvents altered IL-6, IL-1β, or
TNF-R production in RAW 264.7 cells. Before stimulation with LPS
(1 μg/mL) and testing samples, RAW 264.7 cells were incubated for
18 h in 24-well plates under the same conditions. Stimuli and the testing
samples were then added to the culture cells for 24 h. Supernatants were
then collected and assayed for IL-6, IL-1β, and TNF-R content using
mouse ELISA kits (R & D Systems Inc.).
(17) Cho, J. Y.; Baik, K. U.; Jung, J. H.; Park, M. H. Eur. J. Pharmacol.
2000, 398, 399–407.
RNA Preparation and RT-PCR. Total RNA was extracted from
cells in an RNase-free environment using the Tri-Reagent method
(MRC) in accordance with the manufacturer’s instructions. Reverse
transcription was performed using a First-Strand cDNA Synthesis kit
(Promega). Briefly, total RNA (1 μg) was incubated with oligo(dT)18
primer at 70 ꢀC for 5 min and cooled on ice for 5 min. After the addition
of M-MuLV reverse transcriptase (Promega), dNTP (0.5 μM), and 1 U
RNase inhibitor, the reactions were incubated at 25 ꢀC for 5 min and
37 ꢀC for 60 min. M-MuLV reverse transcriptase was inactivated by
heating at 70 ꢀC for 15 min. The PCR was performed using a DNA gene
cycler (BIO-RAD), and the amplification was performed in 30 cycles:
94 ꢀC for 45 s (denaturing), 50ꢀ58 ꢀC for 45 s (annealing), and 72 ꢀC
for 1 min (extension). The PCR products were separated on 1.2%
agarose gel.
’ ASSOCIATED CONTENT
Supporting Information. 1H and 13C NMR spectra of
S
b
compounds 1ꢀ4. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*Tel: 82-51-510-2803. Fax: 82-51-513-6754. E-mail: jhjung@
pusan.ac.kr.
’ ACKNOWLEDGMENT
This study was financially supported by grants (no. 20090400000
and no. 20090083538) from the National Research Foundation
of Korea.
’ REFERENCES
(1) Bradley, S. D. Chem. Rev. 1993, 93, 1771–1991.
(2) Tourneau, C. L.; Raymond, E.; Faivre, S. Curr. Pharm. Des. 2007,
13, 3427–3439.
(3) Cheng, J.; Li, R.; Lin, G. Chin. J. Org. Chem. 1995, 15, 445–448.
(4) García, A.; Vꢀazquez, M. J.; Qui~noꢀa, E.; Riguera, R.; Debitus, C.
J. Nat. Prod. 1996, 59, 782–785.
(5) Anthony, R. C.; Vicky, M. A. J. Nat. Prod. 2009, 72, 696–699.
(6) Marfey, P. Carlsberg Res. Commun. 1984, 49, 591–596.
(7) Ochi, M.; Kataoka, K.; Ariki, S.; Iwatsuki, C.; Kodama, M.;
Fukuyama, Y. J. Nat. Prod. 1998, 61, 1043–1045.
(8) Cafieri, C.; Fattorusso, E.; Mahajnah, Y.; Mangoni, A. Z.
Naturforsch. B 1993, 48, 1408–1410.
(9) Rinehart, K. L.; Kobayashi, J.; Harbour, G. C.; Hughes, R. G.;
Mizsak, S. A.; Scahill, T. A. J. Am. Chem. Soc. 1984, 106, 1524–1526.
1797
dx.doi.org/10.1021/np200397g |J. Nat. Prod. 2011, 74, 1792–1797