Seragamides A-F, new actin-targeting depsipeptides from the sponge Suberites japonicus Thiele

Article abstract of DOI:10.1016/j.tet.2006.01.099

Six new depsipeptides, seragamides A-F (1-6), and a known geodiamolide I (7) have been isolated as cytotoxic metabolites from the Okinawan sponge Suberites japonicus. Their structures were elucidated by means of spectroscopic analysis and chemical transfo

Full text of DOI:10.1016/j.tet.2006.01.099

Tetrahedron 62 (2006) 3536–3542
Seragamides A–F, new actin-targeting depsipeptides from
the sponge Suberites japonicus Thiele
a
Chiaki Tanaka, Junichi Tanaka,
a,
b
*
Robert F. Bolland, Gerard Marriott and Tatsuo Higa
c
a
a
Department of Chemistry, Biology, and Marine Science, University of the Ryukyus, Nishihara, Okinawa 903-0213, Japan
b
University College/Asian Division, University of Maryland, PSC 80, Box 14149, APO AP 96367, USA
Department of Physiology, University of Wisconsin-Madison, 1300 University Av., Madison, WI 53706, USA
c
Received 24 October 2005; revised 30 January 2006; accepted 31 January 2006
Available online 28 February 2006
Abstract—Six new depsipeptides, seragamides A–F (1–6), and a known geodiamolide I (7) have been isolated as cytotoxic metabolites from
the Okinawan sponge Suberites japonicus. Their structures were elucidated by means of spectroscopic analysis and chemical
transformations. Seragamide A (1) promotes the polymerization of G-actin and stabilizes F-actin filaments.
q 2006 Elsevier Ltd. All rights reserved.
4
jaspamide/jasplakinolide, which stabilizes F-actin, trisoxa-
,5
1
. Introduction
6
–8
9 10
swinholide A, hectochlorin, and
zole macrolides,
amphidinolide H.
1
1
The eukaryotic cytoskeleton is composed of three groups of
protein polymers: microtubules, intermediate filaments and
microfilaments. Among these, the microtubule apparatus is
critical for mitosis and is the target of anticancer agents that
include vinca alkaloids and paclitaxel. The microfilament
apparatus on the other hand is rarely considered as a target
for chemical therapy. Recent studies, however, suggest that
actin and the actin filament system of tumor cells, like
microtubules, may serve as a critical target for small
During a functional screen for cytotoxic drugs that inhibit
the actin-driven processes of cytokinesis, we found that a
lipophilic extract of the yellow sponge Suberites japonicus
killed cells at 2 mg/mL with the formation of characteristic
morphological changes similar to that found for trisoxa-
zole macrolides, swinholide A, and misakinolide A. At a
lower concentration range of 0.2–0.02 mg/mL, the extract
led to the formation of multiple nuclei, which suggests that
the active drug might target the actin cytoskeleton. We
isolated and chemically characterized the bioactive
products and determined their mode of action. The
seragamides A–F identified from these studies represent
new and novel molecular tools to study the regulation of
the actin cytoskeleton and may prove useful as anti-cancer
drugs.
1
molecule drugs.
Microfilaments are composed of F-actin and associated
proteins; these filamentous assemblies play essential roles in
cytokinesis, cell motility, and cell adhesion. Actin filament
2
dynamics in eukaryotic cells is regulated by a large number
of actin-binding proteins (ABP) such as cofilin, gelsolin,
thymosin b4 and profilin. Given the complexity of the cell
cytoskeleton, studies that aim to understand the regulation
of actin filament dynamics usually employ small molecule
drugs such as fluorescent phalloidin, a marker of F-actin,
and cytochalasin D, which inhibits actin polymerization.
Recently, several highly specific actin-targeting drugs have
been identified from marine natural products. The first, and
most frequently used marine-derived drugs are latrunculins
A and B, which block polymerization by plugging the ATP
2. Results and discussion
2
.1. Structures of 1–7
3
A small amount of the sponge S. japonicus Thiele was
initially collected at Seragaki, Okinawa. Its lipophilic
extract gave rise to two new cytotoxic depsipeptides
named seragamides A (1) and B (2). A larger amount of
the sponge collected off Manza yielded seragamides C–F
(3–6) in addition to 1, 2, and the known geodiamolide I (7).
site of G-actin. Other actin-targeting marine drugs include
Keywords: Geodiamolide; Actin polymerization; Cytokinesis; Jaspamide;
Jasplakinolide.
Corresponding author. Tel.: C81 98 895 8560; fax: C81 98 895 8565;
e-mail: jtanaka@sci.u-ryukyu.ac.jp
*
0040–4020/$ - see front matter q 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.01.099

C. Tanaka et al. / Tetrahedron 62 (2006) 3536–3542
3537
OH
downfield shifted b-methylene signals (d 2.93 and 3.20).
The presence of a 1,2,4-trisubstituted benzene ring was
shown by observing three aromatic protons at d 6.90 (d, JZ
2
4
O
Seragamide A (1): R = Me, X = I
Seragamide B (2): R = Me, X = Br
Seragamide C (3): R = Me, X = Cl
Seragamide D (4): R = H, X = I
HN
8
1
7
X
O
27
8.5 Hz), 7.09 (dd, JZ8.5, 2.1 Hz), and d 7.50 (d, JZ
20
12
O
N
O
2.1 Hz). The chemical shifts of the proton at d 6.90 and the
carbon at d 154.0 suggested the ring to be phenolic. The
b-methylene protons showed HMBC correlation with three
aromatic carbons at d 130.3, 138.2 and 130.6, whereas
aromatic protons at d 7.50 and 7.09 showed correlation to
the beta carbon at d 32.8 indicating the residue as a tyrosine
derivative. The aromatic signal (d 7.50) showed correlation
to a characteristic high-field carbon signal at d 85.5,
suggesting the presence of iodine atom on this carbon.
HO
Me
O
2
Seragamide E (5): R = CH OH, X = I
14
R
N
H
1
29
O
HN
Geodiamolide A (8): X = I
Geodiamolide B (9): X = Br
Geodiamolide C (10): X = Cl
X
O
O
O
O
N
HO
Me
1
13
N
H
The H and C NMR data for the iodo-tyrosine residue
were comparable with those of the same moiety of
geodiamolide A (8).
OH
1
2
Seragamide F (6): R =
Geodiamolide I (7): R =
OH
26
O
Cl
1
1
The last amino acid unit was assigned as Ala by observing
H NMR signals for a methine at d 4.77, a methyl at d 1.13,
1
HN
12
O
OH
1
8
8
R
O
O
and an amide proton at d 6.49. Both the methyl and the
methine protons showed HMBC correlation to a carbonyl
carbon at d 174.5, while the amide proton exhibited HMBC
cross peak to another carbonyl at d 175.3.
Br
N
Me
1
5
Jaspamide/
Jasplakinolide (11): R =
NH
1
33
NH
O
Br
The remaining portion of the molecule C H O was
1
2 20 2
OH
elucidated as a polyketide composed of four units of
propionates by the following NMR data. Four methyl
groups were separated into three aliphatic doublets (d 0.87,
1.15, 1.25) and one vinyl methyl (d 1.52). HMBC
correlations from these four methyl groups to the rest of
the residue allowed us to construct a fatty acid residue as
2,4,6-trimethyl-non-4-enoyl, the same moiety observed in
geodiamolides and jaspamides. The E configuration of the
4,5-double bond was determined by the NOE signals
between the vinyl methyl (d 1.52) and the methine proton
24
O
Me
N
O
1
2
8
OH
HO
2
C
N
1
4
N
H
1
H
O
1
2
2
0
I
HO C
2
O
Me
O
1
1
N
8
OH
HO
15
1
N
N
H
H
2
8
18
O22
HO
1
3
1
9
Br
N
H
(d 2.22) at C-6, and between the olefinic proton (d 4.91br d)
and one of the methylene protons (d 2.02) at C-3.
Seragamide A (1) obtained as an amorphous solid was
analyzed for C H N O I by HRFABMS. It showed IR
The cyclic nature of seragamide A (1) was evident from its
ten degrees of unsaturation, nine of which were accounted
for four carbonyls, one aromatic, and one double bond.
Connection of the amino acid residues and the alkanoate
unit was obtained by HMBC correlations: the amide proton
at d 6.71 and a carbonyl of I-N-Me-Tyr at d 170.0s,
N-methyl group of I-N-Me-Tyr at d 3.00 and a carbonyl of
Ala at d 174.5s, the amide proton at d 6.49 and a carbonyl of
the alkanoate at d 175.3s, and the oxymethine proton at d
4.87 and the ester carbonyl at d 169.4s. Therefore, the
structure of seragamide A was assembled as shown in 1.
2
9 42 3 7
K1
absorption bands due to hydroxyl (3320 cm ), ester
K1 K1
1732 cm ), and amide (1635 cm ) functionalities. The
(
peptidic nature of compound 1 was suggested by two amide
signals (d 6.49 and 6.71; d 170–176) in NMR in addition to
the IR absorption. The H NMR data exhibited the presence
1
of an N-methyl (d 3.00) and aromatic protons (d 6.90, 7.09,
and 7.50). NMR analysis indicated the presence of three
amino acid residues: Thr, iodo-N-methyl-tyrosine (I-N-Me-
Tyr), and Ala.
A downfield signal (d 6.71) for an amide proton showed
a COSY correlation with a methine proton signal at d 4.42
Absolute configurations of the amino acid residues were
determined by applying Marfey’s method on its acid
hydrolysate (see Section 3). The result revealed the amino
acids to be L-Ala, L-Thr, and D-N-Me-Tyr. All of the three
amide bonds are in trans geometry as shown by NOEs
between NH-Ala (d 6.49) and H-2 (d 2.36), NH-Thr (d 6.71)
and alpha proton (d 5.32) of I-N-Me-Tyr, and N-Me-Tyr (d
(
d 58.1), which in turn coupled to an oxymethine proton
resonance at d 4.29 (d 67.7). This signal further coupled to a
methyl signal at d 1.09 (d 18.7). Thus, this unit was deduced
to Thr. The amide proton signal showed HMBC correlation
to a carbonyl carbon signal (d 170.0), while a resonance
(d 4.42) for an alpha proton showed cross peaks with d 170.0
and another carbonyl at d 169.4.
1
13
3.00) and alpha proton (d 4.77) of Ala. When H and
C
NMR data of 1 were compared with those of 8 as shown in
Table 1, most of the coupling constants and chemical shifts
for the fatty acid portion were nearly identical suggesting
that the three chiral centers were the same as those of
geodiamolide A (8) and its congeners. For further
Second amino acid residue was assigned as I-N-Me-Tyr as
follows. An N-methyl resonance (d 3.00) showed an HMBC
cross peak to an alpha carbon signal at d 56.7. A signal
(
d 5.32) for the alpha proton showed COSY cross peaks with

3538
C. Tanaka et al. / Tetrahedron 62 (2006) 3536–3542
Table 1. NMR data for seragamide A (1) and for geodiamolide A (8)
Seragamide A (1)
12
8
1
3
1
13
1
C#
1
2
C
H (J in Hz)
HMBC
COSY
C
H
175.3s
42.0d
42.9t
170.9s
42.4d
43.3t
2.36ddq (4.0, 11.0, 7.0)
a2.02dd (4.0, 14.0)
b2.16m
H-3ab,29
H-2,3b,5
H-2,3a
2.32m
a2.04dd
b2.16d
3
C-4
C-4
4
5
6
7
133.2s
131.2d
29.0d
43.4t
129.6s
131.6d
29.0d
43.7t
4.91br d (9.0)
2.22m
a1.38ddd (5.8, 8.0, 13.7)
b1.63m
4.87ddq (6.0, 7.0, 6.5)
C-27
H-3a,6,28
H-5,7ab,27
H-6,7b,8
H-6,7a,8
4.93d
2.16m
a1.34m
b1.59m
4.91m
C-5,6,26,27
C-5,6,26,27
C-7,9
8
9
71.8d
169.4s
58.1d
170.0s
56.7d
174.5s
45.8d
19.7q
30.6q
32.8t
H-7ab,26
71.0d
168.8s
49.0d
175.1s
56.7d
174.4s
45.9d
18.8q
30.7q
32.7t
1
1
1
1
1
1
1
1
0
1
2
3
4
5
6
7
4.42dd (3.0, 8.5)
5.32dd (7.6, 9.2)
C-9,11
H-10NH
H-17ab
4.75dq
5.21dd
C-11,13,16,17
4.77dq (6.7, 6.7)
1.13d (6.7)
3.00s
a2.93dd (9.2, 14.6)
b3.20dd (7.6, 14.6)
C-13,15
C-13,14
C-12,13
C-11,12,19,23
C-12,18,19
H-14NH,15
H-14
4.46dq
1.34q
2.97s
a2.95dd
b3.15dd
H-12,17b
H-12,17a
1
1
2
2
2
2
2
2
2
2
2
2
1
1
8
9
0
1
2
3
4
5
6
7
8
9
0NH
4NH
130.3s
138.2d
85.5s
133.0s
132.2d
85.1s
154.5s
116.1d
129.4d
18.7q
7.50d (2.1)
C-17,18,20,21
H-23
7.29d
154.0s
115.1d
130.6d
67.7d
18.7q
20.6q
20.5q
17.8q
18.9q
6.90d (8.5)
7.09dd (2.1, 8.5)
4.29dq (3.0, 6.5)
1.09d (6.5)
1.25d (6.5)
0.87d (6.5)
1.52d (1.2)
1.15d (7.0)
6.71d (8.5)
6.49d (6.7)
5.55br s
C-18,20,21,23
C-17,19,21
H-23
H-19,22
H-25
H-24
H-8
H-6
H-5
H-2
H-10
H-14
6.87d
7.05dd
1.02d
C-10,24
C-7,8
C-5,6,7
C-3,4,5
C-1,3
C-11
20.6q
18.2q
17.7q
20.4q
1.24d
0.86d
1.49s
1.14d
6.52d
6.59d
6.27s
C-1
TyrOH
ThrOH
3.49s
K1
confirmation, ring-opened derivatives 12 and 13 were
prepared by saponification of 1 and 11, respectively, and
their NMR data were compared. Except for the H-2
chemical shifts (Dd 0.09), all other chemical shifts and
coupling constants for the polyketide portion were in good
agreement as shown in Table 2, indicating the same
stereochemistry for the portion in 1 and 11. Thus, we
concluded the stereochemistry of seragamide A (1) as
shown.
similar absorptions (3320, 1733, 1650 cm ) to those of 1.
Inspection of NMR spectra of 2 indicated that almost all
signals in 2 are the same with those of 1 except for the
aromatic portion of Tyr residue (Tables 1 and 3). Further
1
2
comparison of the data with those of geodiamolide B (9),
which retains Br-N-Me-Tyr residue, confirmed that 2
contained a bromine atom in the place of iodine in 1.
Marfey analysis of the acid hydrolysate of 2 determined the
residues as L-Ala and L-Thr. As in the discussion of
seragamide A (1), the remaining chiralities of 2 are regarded
to be the same.
The molecular formula of seragamide
B
(2),
C H N O Br, suggested that it had a bromine atom
2
9 42 3 7
Seragamide C (3) was found to have a molecular formula
C H N O Cl by HRFABMS. Except for aromatic
instead of an iodine atom in 1. The IR spectrum showed
2
9
42
3
7
1
Table 2. Comparison of the H NMR data for the polyketide portion of 12
and 13 (J in Hz)
portions of Tyr residue in 3, most NMR signals are similar
to those of 1 and 2. The aromatic protons (d 6.94, 7.03, 7.17)
are virtually identical to those (d 6.93, 7.01, 7.15) of
H#
12
13
1
3
geodiamolide C (10), which has Cl-N-Me-Tyr residue.
The structure of 3 was elucidated as shown.
2
2.47ddq (7.1, 8.1, 6.8)
1.04d (6.8)
1.96dd (8.1, 13.9)
2.26dd (7.1, 13.9)
1.57s
4.95d (9.3)
2.43dddq (5.9, 8.8, 9.3, 6.6)
0.89d (6.6)
1.34ddd (5.9, 13.4, 6.1)
1.43ddd (8.8, 13.4, 6.1)
3.69ddq (6.1, 6.1, 6.1)
1.12d (6.1)
2.38ddq (7.3, 8.0, 6.8)
1.02d (6.8)
1.94dd (8.0, 13.7)
2.25dd (7.3, 13.7)
1.56s
4.91d (9.8)
2.39dddq (5.9, 9.8, 9.3, 6.6)
0.87d (6.6)
1.31ddd (5.9, 13.4, 6.1)
1.41ddd (9.3, 13.4, 6.1)
3.68ddq (6.1, 6.1, 6.1)
1.12d (6.1)
2
3
3
4
5
6
6
7
7
8
8
-Me
a
b
-Me
The molecular formula of seragamide D (4), C H N O I,
2
8 40 3 7
suggested that it is a demethyl analogue of seragamide A
1). The methyl group (d 19.7) of Ala residue in 1
(
disappeared in 4, and two characteristic downfield shifted
methylene protons (d 3.80, 4.17) appeared instead. As most
other portions showed similar NMR data (Tables 3 and 4) as
1, seragamide D (4) was elucidated as a congener having
Gly in the place of Ala in seragamide A (1). The linkage of
-Me
a
b
-Me

C. Tanaka et al. / Tetrahedron 62 (2006) 3536–3542
3539
1
Table 3. H NMR data for seragamides B–E (2–5) (J in Hz)
B (2)
C (3)
D (4)
E (5)
2
3
3
5
6
7
7
8
1
1
1
2.36m
2.36m
2.47tq (7.9, 6.7)
2.12d (7.9, 2H)
2.42ddq (4.0, 11.0, 6.7)
2.08dd (4.0, 14.0)
2.16m
4.93d (9.2)
2.20m
1.39ddd (5.2, 7.9, 13.7)
1.60m
4.92m
4.41dd (2.7, 8.8)
5.31dd (7.6, 8.8)
4.83m
a
b
2.02dd (3.0, 14.0)
2.17dd (11.5, 14.0)
4.91d (10.0)
2.22m
1.39m
1.64m
4.87ddq (5.2, 7.2, 6.5)
4.42dd (3.1, 8.8)
5.33dd (7.3, 9.2)
4.77dq (6.5, 6.5)
2.02dd (3.0, 14.0)
2.17dd (11.3, 14.0)
4.91d (8.8)
4.98d (7.9)
2.23m
2.20m
a
b
1.39ddd (6.1, 7.9, 13.7)
1.63ddd (6.7, 7.1, 13.7)
4.87ddq (6.1, 7.1, 6.4)
4.42dd (3.0, 8.8)
5.34dd (7.6, 8.8)
4.77dq (6.7, 6.7)
1.44ddd (6.8, 8.2, 13.7)
1.70ddd (6.1, 6.1, 13.7)
4.80ddq (6.1, 6.8, 6.1)
4.46dd (2.4, 9.2)
5.18dd (6.4, 9.5)
a3.80dd (3.0, 17.7)
b4.17dd (4.3, 17.7)
0
2
4
1
1
1
1
1
2
2
2
2
2
2
2
2
1
1
1
2
2
5
6
7a
7b
9
2
3
4
5
6
7
8
9
0NH
4NH
5OH
1OH
4OH
1.12d (6.5)
3.00s
1.12d (6.7)
3.00s
3.59m (2H)
3.08s
2.96s
2.94dd (9.2, 14.6)
3.21dd (7.3, 14.6)
7.31d (2.1)
6.94d (8.3)
7.07dd (2.1, 8.3)
4.29m
1.09d (6.5)
1.25d (6.5)
0.88d (6.5)
1.52d (1.2)
1.15d (7.0)
6.72d (8.8)
6.48d (6.5)
2.93dd (8.8, 14.6)
3.21dd (7.6, 14.6)
7.17d (2.1)
6.94d (8.2)
7.03dd (2.1, 8.2)
4.30m
1.09d (6.4)
1.25d (6.4)
0.87d (6.7)
1.52d (1.2)
1.15d (7.0)
6.69d (8.8)
6.47d (6.7)
2.83dd (6.4, 13.7)
3.26dd (9.5, 13.7)
7.53d (1.8)
6.91d (8.5)
7.10dd (1.8, 8.5)
4.22m
0.99d (6.7)
1.27d (6.1)
0.90d (6.7)
1.56s
2.93dd (8.8, 14.6)
3.23dd (7.6, 14.6)
7.50d (2.1)
6.91d (8.5)
7.09dd (2.1, 8.5)
4.33m
1.08d (6.4)
1.25d (6.1)
0.89d (6.7)
1.52d (1.2)
1.18d (6.7)
6.66d (8.8)
6.70d (6.4)
3.55m
1.17d (6.7)
6.64d (9.2)
6.48br s
5.55br s
3.49br s
5.52br s
3.49br s
5.35s
3.49br s
5.47br s
3.50d (4.9)
the residues was confirmed by HMBC correlations (H-10/C-
1, NH-Thr/C-11, N-Me/C-12,13, H-14/C-1).
corresponding to Ala residue in 1, while it exhibited signals
for a primary alcohol (d 65.1; d 3.59), suggesting the
presence of Ser residue instead of Ala in 1. This disposition
was also supported by HMBC correlations (N-Me/C-12,13,
H-14/C-13). The remaining portions showed nearly the
same NMR data with other seragamides (Tables 3 and 4).
The absolute stereochemistry of Ser was determined to be L
by Marfey’s method.
1
Seragamide E (5), C H N O I, was analyzed to have an
2
9 42 3 8
additional oxygen atom on the formula of seragamide A (1).
In the NMR spectra of 5, it lacked methyl signals
13
Table 4. C NMR data for seragamides B–E (2–5)
B (2)
C (3)
D (4)
E (5)
Seragamide F (6), C H N O Cl, showed the presence of
3
4 44 3 7
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
175.2s
41.9d
42.8t
133.2s
131.1d
29.0d
43.4t
175.2s
41.9d
42.8t
133.3s
131.0d
29.1d
43.5t
71.8d
169.4s
58.0d
170.0s
56.7s
174.5s
45.8d
18.6q
30.6q
33.0q
129.5s
129.0d
119.8s
150.3s
116.4d
128.8d
67.7d
19.7q
20.5q
20.6q
17.9q
19.0q
175.8s
41.8d
42.9t
133.5s
131.0d
29.2d
43.4t
72.6d
169.2s
57.8d
169.2s
57.7d
169.9s
42.1t
177.4s
42.0d
42.9t
132.9s
131.6d
29.1d
43.8t
71.5d
169.5s
58.0d
169.8s
57.0d
171.8s
52.5d
65.1t
31.0s
32.8t
130.2s
138.2d
85.6s
154.0s
115.2d
130.6d
67.6d
19.8q
20.7q
20.5q
18.0q
18.9q
14
two aromatic rings as observed for geodiamolide I (7). Of
the two aromatic rings in 6, a p-disubstitued one is virtually
the same as in 7, while 1,2,4-trisubstituted one (d 7.13d,
6.90d, 6.99dd) is similar to the one found for seragamide C
(
3) (d 7.17d, 6.94d, 7.03dd). Therefore, the structural
difference between 6 and 7 lies at the halogen substitution
on the latter aromatic ring. Close similarities of NMR data
of 6 to 7 indicate that they have the same stereochemistry.
71.8d
169.4s
58.0d
169.9s
56.6d
174.5s
45.8d
19.7q
30.6q
32.9t
129.8s
132.0d
110.1s
151.2s
116.2d
129.6d
67.7d
18.6q
20.5q
20.5q
17.8q
18.9q
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
2.2. Biological activity
The IC50 values of seragamides A–E (1–5) were determined
as 0.064, 0.12, 0.10, 0.18 and 0.58 mM, respectively.
Seragamides A–E (1–5) caused multinuclei formation in
cells at 0.01, 0.02, 0.01, 0.01, and 0.04 mg/mL, respectively.
Seragamide F (6) was not tested.
29.8q
32.5t
130.4s
138.6d
85.7s
154.0s
115.2d
130.8d
68.1d
19.5q
20.6q
20.6q
18.2q
18.4q
15
Using a Prodan-actin polymerization assay we found that
seragamide A (1) in the range of 20–200 nM facilitated the
polymerization of 1 mM G-actin, as shown in Figure 1. In
the Prodan-actin depolymerization assay, 1 nM seragamide
A (1) inhibited the depolymerization of F-actin at 100 nM
(Fig. 2). Since these activities are similar to those observed
for jaspamide/jasplakinolide (11), we suggest that

3540
C. Tanaka et al. / Tetrahedron 62 (2006) 3536–3542
3
.2. Animal material
The yellow round sponge was identified as S. japonicus
Thiele (Order Hadromerida, Family Suberitidae) by Dr.
John N.A. Hooper, Queensland Museum, Brisbane. The
specimen is deposited at the museum (QMG317382).
3
.3. Extraction and isolation
The first collection (40 g, wet) of the sponge S. japonicus
was made in 2001 by R.F.B. at a depth of 60 m by hand
using SCUBA at Seragaki, Okinawa. The fresh sample was
soaked and kept in MeOH until extraction. The MeOH
solution was concentrated and the residual material was
Figure 1. The effect of seragamide A (1) on polymerization of G-actin.
Prodan-labeled G-actin (1 mM) is mixed with seragamide A (1) at the
concentration of 0 nM (a), 1 nM (b), 10 nM (c), 20 nM (d), 50 nM (e),
partitioned between EtOAc and H O. The residue of the
2
EtOAc extract (33 mg) was chromatographed on a column
packed with Cosmosil using MeOH–H O to give six
100 nM (f), and 200 nM (g).
2
fractions. These fractions were submitted for a cell-based
high content imaging assay; fractions 1 and 2 were found to
be active (fraction 1: 0.2 mg/mL and fraction 2: 2 mg/mL),
showing vacuoles, blebs, arboratization similar to that
1
6
observed for misakinolide A. Each fraction was separated
by reversed phase HPLC using MeOH–H O (60/40) to give
compounds 1 (4.1 mg) and 2 (1.6 mg).
2
The second specimen (0.5 kg, wet), collected at a depth of
50 m off Manza, was repeatedly extracted with MeOH and
then with acetone. After concentration the combined
extracts were partitioned between EtOAc and H O. The
2
residue (6.4 g) of the organic extract was subjected to
vacuum flash column chromatography (VFC) on silica gel
(
hexane, hexane–EtOAc, EtOAc, and EtOAc–MeOH) to
1
Figure 2. The effect of seragamide A (1) on depolymerization of F-actin.
Prodan-labeled F-actin (28 mM) is mixed to a final concentration at 100 nM
with seragamide A (1) at 0 nM (a), 0.05 nM (b), 0.1 nM (c), 1 nM (d), and
afford four fractions. H NMR spectrum of each fraction
was compared with those of 1 and 2. The spectrum of
fraction 3 clearly indicated the presence of 1. The fraction
was roughly separated by MPLC (Lobar column) using 60%
aqueous MeOH to give three subfractions. Each of these
fractions was purified on reversed phase HPLC using the
same solvent (60% aqueous MeOH) to afford compounds
1–7 in the amount of 145.8, 23.0, 2.0, 1.5, 1.8, 0.4, and
10 nM (e).
seragamides also function as cell permeable, F-actin
stabilizing drugs.
0
.9 mg, respectively.
2
7
3
C45.6 (c 0.215, CHCl ); IR (neat) n : 3320, 1732,
.3.1. Seragamide A (1). White amorphous solid, [a]D
3
. Experimental
3
max
K1
635 cm ; UV l
4
(MeOH): 207 (3 3.4!10 ) and
1
2
4
6
max
3
.1. General experimental procedures
3
C
86 nm (2.8!10 ); FABMS: m/z 672 ([MCH] ), 546,
23; HRFABMS: m/z 672.2133, calcd for C H N O I
72.2146; H and C NMR (CDCl ): Table 1.
1
13
29 43 3 7
H and C NMR spectra were recorded on a Jeol a-500
spectrometer. Mass spectra were measured on Jeol JMS-
D300. IR, UV spectra and the optical rotations were
measured using a Shimadzu FTIR-8300A instrument, a
Hitachi U-2001 spectrophotometer, and a Jasco DIP-1000
polarimeter, respectively. Fluorescence spectra were taken
on a Jasco FP6500 fluorometer. A microplate reader
TECAN GENios was used for cytotoxicity. High-perform-
ance liquid chromatography (HPLC) was performed on a
Hitachi L-6000 pump equipped with a Hitachi RI monitor
655A-30) and a Hitachi L-4000 UV detector, using
Cosmosil packed ODS HPLC column (5C18-AR-II, 10!
50 mm). Merck silica gel was used for vacuum flash
chromatography. Nacalai Cosmosil was used for reversed
phase column chromatography. All solvents used were
reagent grade.
1
13
3
2
7
3
C39 (c 0.090, CHCl ); IR (neat) n : 3320, 1733,
.3.2. Seragamide B (2). White amorphous solid, [a]D
3
max
K1
650 cm ; UV l
4
(MeOH): 205 (3 3.1!10 ) and
1
2
max
3
83 nm (2.3!10 ); FABMS: m/z 626, 624 ([MCH] );
C
7
9
HRFABMS: 624.2240, calcd for C H N O Br 624.2284;
29 43 3 7
1
13
H and C NMR (CDCl ): Tables 3 and 4.
3
23
(
3.3.3. Seragamide C (3). Colorless glass, [a] C53 (c
D
K1
0.10, CHCl ); IR (neat) n : 3330, 1738, 1651 cm ; UV
3
max
4
lmax (MeOH): 217 (3 1.2!10 ) and 281 nm (2.3!10 );
3
2
C
FABMS: m/z 580, 582 ([MCH] ); HRFABMS: m/z
3
5
580.2790, calcd for C H N O Cl 580.2787; H and
1
13
C
2
9 43 3 7
NMR (CDCl ): Tables 3 and 4.
3

C. Tanaka et al. / Tetrahedron 62 (2006) 3536–3542
3541
2
.3.4. Seragamide D (4). Colorless glass, [a]D C46 (c
3
3
residue was stored in a freezer until HPLC analysis. The
FDAA derivatives of L-Ala, D- and L-Thr, D- and L-allo-Thr,
and D- and L-N-Me-Tyr were prepared in the same manner.
K1
.075, CHCl ); IR (neat) n : 2290, 1734, 1665 cm ; UV
0
lmax (MeOH): 217 (3 1.2!10 ) and 285 nm (2.6!10 );
3
max
4
3
C
FABMS: m/z 658 ([MCH] ); HRFABMS: m/z 658.1989,
1 13
calcd for C H N O I 658.1975; H and C NMR
HPLC analyses of FDAA derivatives were carried out by
using a reverse phase column (RP-18, Mightysil 250!
2
8 41 3 7
(
CDCl ): Tables 3 and 4.
3
10 mm) eluting with buffer–MeOH–MeCN (65/15/20) (flow
2
4
3
.3.5. Seragamide E (5). Colorless glass, [a] C33 (c
rate: 0.8 mL/min) and detecting UV absorption at 340 nm.
The buffer was prepared by adjusting 20 mM potassium
hydroxide to pH 2–3 with 85 wt% phosphoric acid.
Individual amino acid was identified by comparing retention
times with those of FDAA derivatives of standard amino
acids. The retention times were: L-Thr (24.5 min), D-Thr
D
K1
0
lmax (MeOH): 218 (3 1.1!10 ) and 285 nm (2.5!10 );
.090, CHCl ); IR (neat) n : 3340, 1732, 1653 cm ; UV
3 max
4
3
C
FABMS: m/z 688 ([MCH] ); HRFABMS: m/z 688.2095,
1 13
calcd for C H N O I 688.2103; H and C NMR: Tables
2
9 43 3 8
3
and 4.
(
40.6 min), L-allo-Thr (25.6 min), D-allo-Thr (29.9 min),
2
4
.3.6. Seragamide F (6). Colorless glass, [a]D C10 (c 0.02,
3
L-Ala (39.2 min), D-Ala (42.8 min), L-N-Me-Tyr (43.0 min),
and D-N-Me-Tyr (44.8 min). Derivatives of seragamide E
(5) and standards were analyzed on a different column using
the same solvents as above to give retention of L-Ser
(14.6 min) and D-Ser (16.0 min).
K1
CHCl ); IR (neat) n : 3350, 1716, 1684 cm ; UV l
max
3
max
3
3
MeOH): 229 (3 4.7!10 ) and 279 nm (1.4!10 );
(
FABMS: m/z 642 ([MCH] ); HRFABMS: m/z 642.2946,
C
3
7
5
calcd for C H N O Cl 642.2906; H NMR (CDCl ): d
1
3
4
45
3
3
0
.85 (d, 3H, JZ6.7 Hz; H-32), 1.06 (d, 3H, JZ6.7 Hz;
H-16), 1.08 (d, 3H, JZ6.4 Hz; H-31), 1.17 (d, 3H, JZ
.0 Hz; H-34), 1.17 (m, 1H; H-7), 1.33 (m, 1H; H-7), 1.58
d, 3H, JZ4.0 Hz; H-33), 1.92 (br d, 1H, JZ15.6 Hz; H-3),
3.6. Saponification of 1 and 11
7
(
Seragamide A (1, 10.8 mg) was dissolved in 1 mL of MeOH
and 75 mL of 1 M KOH, and the mixture was kept stirring at
room temperature for 4 h. The resulting solution was
neutralized by adding 0.1 M HCl and partitioned between
EtOAc and water. The organic layer was purified by
preparative TLC (hexane/EtOAc, 1–6) on silica to give
3.0 mg (27%) of 12.
2
2
2
1
1
.27 (m, 1H; H-6), 2.39 (dd, 1H, JZ11.7, 15.6 Hz; H-3),
.55 (m, 1H; H-2), 2.61 (dd, 1H, JZ6.0, 15.3 Hz; H-10),
.69 (dd, 1H, JZ4.6, 15.3 Hz; H-10), 2.91 (dd, 1H, JZ10.6,
4.7 Hz; H-18), 2.92 (s, 3H; H-17), 3.25 (dd, 1H, JZ6.4,
4.7 Hz; H-18), 4.63 (m, 1H; H-8), 4.78 (m, 1H; H-5), 4.79
(
m, 1H; H-15), 5.26 (m, 1H; H-11), 5.45 (dd, 1H, JZ6.4,
1
2
6
0.6 Hz; H-13), 6.66 (d, 1H, JZ7.0 Hz; 15-NH), 6.76 (d,
H, JZ8.8 Hz; H-27, 29), 6.90 (d, 1H, JZ8.2 Hz; H-23),
.99 (dd, 1H, JZ1.8, 8.2 Hz; H-24), 7.06 (d, 2H, JZ8.2 Hz;
Pure jaspamide (11) was isolated from a lipophilic extract of
a yellow sponge Jaspis sp. collected off Makassar,
1
7
H-26, 30), 7.13 (d, 1H, JZ1.8 Hz; H-20), 7.43 (br d, 1H,
JZ9.2 Hz; 11-NH).
Indonesia.
Its authentic nature was confirmed by
4
,5
NMR. Jaspamide (11, 1.6 mg) was dissolved in 500 mL
of MeOH and 100 mL of 1 M KOH, and the mixture was
stirred at 40–50 8C for 5 h. Similar work-up as for 12
furnished 1.3 mg (79%) of 13.
2
4
3
.3.7. Geodiamolide I (7). Colorless glass, [a] C37 (c
D
14
0
1
.045, CHCl ; lit. C39.3); IR (neat) nmax: 3330, 1734,
3
K1
684 cm ; UV l
4
(MeOH): 227 (3 1.0!10 ) and
max
1
3
82 nm (2.5!10 ); H and C NMR (CDCl ) data were
13
1
3.6.1. Compound 12. H NMR (CDCl –CD OD, 9–1) d
2
identical with those reported.
3
3
3
1
4
1.00 (d, 3H, JZ6.8 Hz; H-15), 1.09 (d, 3H, JZ6.1 Hz; H-
5), 2.84 (dd, JZ11.0, 14.6 Hz; H-17a), 2.98 (s, 3H; H-16),
3.25 (dd, JZ5.6, 14.6 Hz; H-17b), 4.22 (m, 2H; H-10, 24),
.61 (q, JZ7.1 Hz; H-14), 5.33 (m; H-12), 6.76 (d, JZ
2
3
.4. Acid hydrolysis of 1, 2 and 5
4
To a sample (0.3 mg) of 1 in a small glass tube was added
M HCl (0.5 mL). After sealing, it was heated at 110 8C for
8.3 Hz; H-22), 7.00 (dd, JZ1.9, 8.3 Hz; H-23), 7.46 (br s;
H-19). For H-2 to 8-Me, see Table 2. ESIMS m/z 688
[MKH] , m/z 690 [MCH] , 712 [MCNa] .
6
K
C
C
over night in an oven. The reaction mixture was cooled to
ambient temperature and partitioned between EtOAc and
H O. Each layer was concentrated to dryness under N
1
3.6.2. Compound 13. H NMR (CDCl –CD OD, 9–1) d
2
2
3
3
stream. The H O layer was used for making FDAA
2
derivatives. The same procedure was applied for the
hydrolysis of 2 and 5.
0.78 (d, 3H, JZ7.1 Hz; H-16), 2.60 (d, 2H, JZ6.8 Hz; H-
10), 2.94 (s, 3H; H-17), 3.18 (dd, JZ10.7, 14.9 Hz; H-18a),
3.40 (m; H-18b), 4.59 (q, JZ7.1 Hz; H-15), 5.22 (br; H-13),
5.26 (t, JZ6.2 Hz; H-11), 6.60 (d, 2H, JZ8.5 Hz; H-29,
3.5. Derivatization of amino acid with Marfey’s reagent
and HPLC analysis
31), 6.99 (d, 2H, JZ8.5 Hz; H-28, 32), 7.04 (t, JZ7.8 Hz;
H-24), 7.11 (t, JZ7.8 Hz; H-23), 7.27 (d, JZ7.8 Hz; H-22),
7
.48 (d, JZ7.8 Hz; H-25). For H-2 to 8-Me, see Table 2.
K C
ESIMS m/z 725, 727 [MKH] , m/z 749, 751 [MCNa] ,
To a 2 mL reaction vial containing 0.3 mg standard D-Ala in
C
771, 773 [MC2NaKH] .
10 mL of water was added 50 mL of N,a-(2,4-dinitro-5-
fluorophenyl)-L-alaninamide (FDAA, Marfey’s reagent)
acetone solution (10 mg/mL) followed by 20 mL of 1 M
NaHCO . The mixture was heated at 40 8C for 1 h. After
cooling at room temperature it was neutralized by adding
3.7. Cell assay
3
NBT-T2 cells (BRC-1370) were purchased from Riken and
cultured under a standard protocol using DMEM. Aliquots
of test compounds in MeOH were added to culture wells
10 mL of 2 M HCl solution. The resulting solution was
filtered and concentrated by N stream, and the resulting
2

3542
C. Tanaka et al. / Tetrahedron 62 (2006) 3536–3542
plated cells 1 day before. After incubating the sample wells
for 1 or 2 days, the toxic effect of a drug was observed under
a microscope. The IC50 values were measured by MTT
colorimetric method. Cultured cells were inoculated into
each well (24-well plate) with 1 mL of the medium. After
Dr. Hideo Naoki, Okinawa Health Biotechnology Research
Development Center, for allowing us to use a fluorometer.
References and notes
preincubation (24 h, 37 8C, 5% CO ), the sample solution
2
was added to each well and it was incubated for 48 h. And
then, MTT solution (100 mL, 5 mg/mL in PBS) was added to
each well and incubated for 3 h. The medium was removed
by aspiration. The residual formazan was dissolved in 1 mL
of dimethylsulfoxide (DMSO). The resulting solution was
diluted with DMSO (four times) and the absorbance was
measured at 540 nm.
1. Giganti, A.; Friederich, E. Prog. Cell Cycle Res. 2003, 5,
511–525.
2. Pollard, T. D. Nature 2003, 422, 741–745.
3. Spector, I.; Shochet, N. R.; Kashman, Y.; Groweiss, A. Science
1983, 219, 493–495.
4. Zabriskie, T. M.; Klocke, J. A.; Ireland, C. M.; Marcus, A. H.;
Molinski, T. F.; Faulkner, D. J.; Xu, C.; Clardy, J. C. J. Am.
Chem. Soc. 1986, 108, 3123–3124.
3
.8. Actin polymerization/depolymerization assay
5. Crews, P.; Manes, L. V.; Boehler, M. Tetrahedron Lett. 1986,
2
7, 2797–2800.
Prodan-labeled actin was prepared by the method described
by G.M. Polymerization of G-actin was carried out by
6. Klenchin, V. A.; Allingham, J. S.; King, R.; Tanaka, J.;
Marriott, G.; Rayment, I. Nat. Struct. Biol. 2003, 10,
1058–1063.
1
5
making a Prodan-G-actin solution at 1 mM in F-buffer
100 mM KCl, 2 mM MgCl , 0.2 mM CaCl , 1 mM DTT,
(
7. Tanaka, J.; Yan, Y.; Choi, J.; Bai, J.; Klenchin, V. A.;
Rayment, I.; Marriott, G. Proc. Natl. Acad. Sci. USA 2003,
100, 13851–13856.
2
2
0
.1 mM ATP, 5 mM Tris, pH 8.0) in a cuvette. The time
course of F-actin polymerization was recorded for 200 s
using an excitation wavelength of 385 nm and emission at
8. Allingham, J. S.; Tanaka, J.; Marriott, G.; Rayment, I. Org.
Lett. 2004, 6, 597–599.
4
90 nm in the presence of 1, 10, 20, 50, 100, and 200 nM of
seragamide A (1), or without 1 as control. The depolymeri-
zation assay of Prodan-F-actin was performed by diluting a
preformed 28 mM solution of Prodan-F-actin to 100 nM in
G-buffer (0.2 mM CaCl , 1 mM DTT, 0.1 mM ATP, 5 mM
2
Tris, pH 8.0). Time course experiments were similarly
carried out as for polymerization in the presence of 0.05,
9. Klenchin, V. A.; King, R.; Tanaka, J.; Marriott, G.; Rayment,
I. Chem. Biol. 2005, 12, 287–291.
10. Marquez, B. L.; Watts, K. S.; Yokochi, A.; Roberts, M. A.;
Verdier-Pinard, P.; Jimenez, J. I.; Hamel, E.; Scheuer, P. J.;
Gerwick, W. H. J. Nat. Prod. 2002, 65, 866–871.
11. Usui, T.; Kazami, S.; Dohmae, N.; Mashimo, Y.; Kondo, H.;
Tsuda, M.; Terasaki, A. G.; Ohashi, K.; Kobayashi, J.; Osada,
H. Chem. Biol. 2004, 11, 1269–1277.
0.1, 1, and 10 nM of seragamide A (1) or without 1 as
control.
1
2. Chan, W. R.; Tinto, W. F.; Manchand, P. S.; Todaro, L. J.
J. Org. Chem. 1987, 52, 3091–3093.
Acknowledgements
13. De Silva, E. D.; Andersen, R. J.; Allen, T. M. Tetrahedron
Lett. 1990, 31, 489–492.
This research was supported by a grant from The Cabinet
Office of Japan through Research Institute for Subtropics in
Okinawa, a grant titled as 21st century COE program: the
Comprehensive Analyses on Biodiversity in Coral Reef and
Island Ecosystems in Asian and Pacific Regions, a grant
14. Tinto, W. F.; Lough, A. J.; McLean, S.; Reynolds, W. F.; Yu,
M.; Chan, W. R. Tetrahedron 1998, 54, 4451–4458.
15. Marriott, G.; Zechel, K.; Jovin, T. M. Biochemistry 1988, 27,
6214–6220.
16. Tanaka, J.; Higa, T.; Kobayashi, M.; Kitagawa, I. Chem.
Pharm. Bull. 1990, 38, 2967–2970.
(
No. 16550145) from MEXT of Japan, and a grant NIH R01
HL069970-01 to G.M. We thank Dr. T. Ichiba, Okinawa
Industrial Technology Center, for mass measurements, and
17. Roy, M. C.; Ohtani, I. I.; Ichiba, T.; Tanaka, J.; Satari, R.;
Higa, T. Tetrahedron 2000, 56, 9079–9092.