Orthidines A-E, tubastrine, 3,4-dimethoxyphenethyl-β-guanidine, and 1,14-sperminedihomovanillamide: potential anti-inflammatory alkaloids isolated from the New Zealand ascidian Aplidium orthium that act as inhibitors of neutrophil respiratory burst

Article abstract of DOI:10.1016/j.tet.2008.04.012

In addition to the known dihydroxystyrylguanidine alkaloid tubastrine (1), five new dimers, orthidines A-E (2-6) and the biosynthetically unrelated 1,14-sperminedihomovanillamide (orthidine F, 7) were isolated from the New Zealand ascidian Aplidium orthium. The structures of the new compounds, elucidated by interpretation of spectroscopic data, encompass benzodioxane neolignan-type scaffolds (2-5) and a 1,2,3,4-tetrasubstituted cyclobutane (6), the latter likely having arisen via [_π2_s+_π2_s] dimerization of tubastrine. The subunit head-to-tail orientation of dimer 6 was established unambiguously by interpretation of data from a ²J,³J-HMBC NMR experiment. The structure of 7 was also confirmed by facile synthesis. Compounds 1-4, 6, and 7 inhibited the in vitro production of superoxide by PMA-stimulated human neutrophils in a dose-dependent manner with IC₅₀s of 10-36 μM and this was associated with inhibition of superoxide production by neutrophils in vivo in a murine model of gouty inflammation.

Full text of DOI:10.1016/j.tet.2008.04.012

Tetrahedron 64 (2008) 5748–5755
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Orthidines A–E, tubastrine, 3,4-dimethoxyphenethyl-b-guanidine, and
1,14-sperminedihomovanillamide: potential anti-inflammatory alkaloids
isolated from the New Zealand ascidian Aplidium orthium that act as
inhibitors of neutrophil respiratory burst
A. Norrie Pearce a, Elizabeth W. Chia b, Michael V. Berridge b, Elizabeth W. Maas c,
,
Michael J. Page d, Jacquie L. Harper b, Victoria L. Webb c, Brent R. Copp a
*
a Terramarine Pharmaceuticals, Department of Chemistry, The University of Auckland, Private Bag 92019, Auckland, New Zealand
b Malaghan Institute of Medical Research, PO Box 7060, Wellington South, New Zealand
c National Institute for Water and Atmospheric Research (NIWA) Ltd, Private Bag 14-901, Kilbirnie, Wellington, New Zealand
d National Institute for Water and Atmospheric Research (NIWA) Ltd, PO Box 893, Nelson, New Zealand
a r t i c l e i n f o
a b s t r a c t
Article history:
In addition to the known dihydroxystyrylguanidine alkaloid tubastrine (1), five new dimers, orthidines
A–E (2–6) and the biosynthetically unrelated 1,14-sperminedihomovanillamide (orthidine F, 7) were
isolated from the New Zealand ascidian Aplidium orthium. The structures of the new compounds, elu-
cidated by interpretation of spectroscopic data, encompass benzodioxane neolignan-type scaffolds (2–5)
and a 1,2,3,4-tetrasubstituted cyclobutane (6), the latter likely having arisen via [p2sþp2s] dimerization of
tubastrine. The subunit head-to-tail orientation of dimer 6 was established unambiguously by in-
terpretation of data from a 2J,3J-HMBC NMR experiment. The structure of 7 was also confirmed by facile
synthesis. Compounds 1–4, 6, and 7 inhibited the in vitro production of superoxide by PMA-stimulated
Received 29 January 2008
Received in revised form 17 March 2008
Accepted 3 April 2008
Available online 8 April 2008
Keywords:
Marine natural products
Aplidium orthium
Tubastrine
human neutrophils in a dose-dependent manner with IC50s of 10–36 mM and this was associated with
inhibition of superoxide production by neutrophils in vivo in a murine model of gouty inflammation.
Cyclobutane
Ó 2008 Elsevier Ltd. All rights reserved.
Guanidine
Anti-inflammatory
1. Introduction
spectroscopic methods with confirmation of 7 also being achieved
by synthesis. In vitro and in vivo anti-inflammatory activities are
As part of our ongoing screening campaign of New Zealand biota
to discover new anti-inflammatory natural products1–4 we found
that an extract of the marine organism Aplidium orthium (Ascidia-
cea), collected from Northeast Island of the Three Kings Islands,
strongly inhibited superoxide production by stimulated human
neutrophils.5 Bioassay-guided fractionation of this extract using
combinations of C8 and C18 reversed phase and Sephadex LH-20
flash column chromatography led to the isolation of the known
guanidinostyrene marine natural product tubastrine (1),6,7 four
new benzodioxanes (orthidines A–D, 2–5), a cyclobutane dimer of
tubastrine (orthidine E, 6), and a biosynthetically unrelated diho-
movanillamide derivative of spermine (orthidine F, 7). Investigation
of the chemical components of non-biologically active chroma-
tography fractions yielded guanidine 8, previously known as
a synthetic compound.8 The structures of 1–8 were determined by
reported for the natural products 1–4 and 6–8.
2. Results and discussion
2.1. Chemistry
The biologically active crude MeOH extract of the organism was
subjected to reversed phase C18 column chromatography with
a steep H2O (0.05% TFA) to MeOH gradient. Biological activity was
detected in the H2O (TFA) and 10–20% MeOH/H2O (0.05% TFA)
fractions, further purification of which using reversed phase C8 and
Sephadex LH-20 gel permeation chromatography afforded the
previously reported alkaloid tubastrine (1) and five new com-
pounds (2, 3, 4, 6, 7). A second collection of the organism from
the same location yielded the same series of compounds as well
as the additional stereoisomer 5. Investigation of the biologically
inactive 30% MeOH/H2O (0.05% TFA) C18 column fraction afforded
* Corresponding author. Tel.: þ64 9 373 7599x88284; fax: þ64 9 373 7422.
E-mail address: b.copp@auckland.ac.nz (B.R. Copp).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.04.012

A.N. Pearce et al. / Tetrahedron 64 (2008) 5748–5755
5749
5'
OH
OH
OH
OR
OR
5
4a
8a
O
O
3
O
O
OH
2'
2
Δ
1"
NH
NH
NH2
8
2b
HN
H2N
2"
HN
H2N
NH2
HN
H2N
H2N
3"
NH2
H2N
NH2
NH2
1 R = H
8 R = Me, Δ saturated
2
3
OH
5
H2N
NH2
NH
H2N
O
NH2
NH
H2N
NH2
NH
HO
5
8
O
O
3
2
6
7
O
8
HN
H2N
HN
OH
HN
H2N
OH
OH
OH
OH
NH2
10
H2N
NH2
NH2
OH
4
5
6
16
OMe
17
HO
3
O
Relative stereochemistry only
10
14
N
N
H
2
H
7
7
5
3,4-dimethoxyphenethyl-
b
-guanidine (8), which is reported for the
spectra, accounting for the remaining composition. HMBC corre-
lations observed between H-8 ( 6.97,
7.00, d, J¼2.0 Hz) and H-6 (
dd, J¼8.4, 2.0 Hz) to 143.4 (C-4a) and from H-5 ( 6.92, d, J¼8.4 Hz)
to 142.7 (C-8a) unequivocally assigned carbons C-4a and C-8a,
respectively. Interfragment connectivity was established by ob-
serving HMBC correlations between H-2 ( 5.63) and C-8a and
between H-3 ( 4.91) and C-4a leading to the construction of a 2,3-
first time as a natural product.8
d
d
The first eluting bioactive compound was identified as tubas-
trine (1), previously isolated from the hard coral Tubastrea aurea6
and the ascidian Dendrodoa grossularia,7 through combined use of
HRESIMS and 1H and 13C, and two dimensional NMR spectroscopy.
Its identity was confirmed by direct comparison of 1H and 13C NMR
data with that previously reported.7
Orthidine A (2) was assigned a molecular formula of C18H20N6O4
on the basis of HRESIMS and NMR data. In addition to the pseudo-
molecular ion at m/z 385, a doubly ionized ion at m/z 193 attrib-
utable to [Mþ2H]2þ was also observed. The UV–vis spectrum of
2 exhibited absorption maxima at 207, 226, and 284 nm with
a shoulder at 311 nm indicative of an extended aromatic chromo-
phore. Interpretation of the 1H (Table 1) and 13C (Table 2) NMR
spectra including HSQC and HMBC data established the presence of
a tubastrine-like substructure accounting for C9H9N3O2 of the
required formula. In addition, a partial fragment consisting of a 3,4-
dihydroxyphenyl group, two oxymethine carbons (dC 77.5, dH 4.91,
d, J¼5.6 Hz; dC 79.8, dH 5.63, d, J¼5.6 Hz), and a guanidine group (dC
158.7) was established by interpretation of HSQC and HMBC NMR
d
d
d
d
d
dihydrobenzodioxane framework for orthidine A (2). The relative
stereochemistry and conformation of 2 was investigated by analysis
3
3
of NOESY NMR correlations, and JHH and JCH coupling constants.
NOESY correlations were observed between H-2 and H-3/H-20/H-60
and between H-3 and H-2/H-20/H-60. While the magnitude of the
vicinal coupling between H-2 and H-3 (5.6 Hz) suggested a trans
stereochemical relationship, an H-2 axial, H-3 axial arrangement
was discounted due to the observation of strong NOESY correla-
tions between H-2 and H-3. This inferred an H-2 equatorial, H-3
equatorial arrangement for 2 (Fig. 1). Further support for this
relative stereochemistry and conformation was obtained from
analysis of a J-HMBC NMR experiment,9 which furnished coupling
constants of 8.0 Hz for both 3JC-8a–H-2 and 3JC-4a–H-3. Such a coupling
constant is consistent with C-8a/H-2 and C-4a/H-3 being in
Table 1
1H NMR (600 MHz, CD3OD) data for 2–5
Position
2
3
4
5
dH (mult., J in Hz)
dH (mult., J in Hz)
dH (mult., J in Hz)
dH (mult., J in Hz)
2
3
5
6
5.63 (d, 5.6)
4.91 (d, 5.6)
6.92 (d, 8.4)
6.97 (dd, 8.4, 2.0)
7.00 (d, 2.0)
6.86 (d, 1.8)
6.78 (m)
6.76 (m)
6.22 (d, 14.0)
7.08 (d, 14.0)
5.87 (d, 1.9)
5.23 (d, 1.9)
7.03 (m)
7.03 (m)
7.05 (m)
6.96 (d, 1.4)
6.84 (m)
6.85 (m)
6.23 (d, 14.0)
7.12 (d, 14.0)
4.91 (d, 5.7)
5.63 (d, 5.7)
6.88 (d, 8.4)
6.96 (dd, 8.4, 2.0)
7.05 (d, 2.0)
6.90 (d, 1.7)
6.79 (m)
6.80 (m)
6.23 (d, 13.9)
7.12 (d, 13.9)
5.23 (d, 2.0)
5.86 (d, 2.0)
6.94 (d, 8.4)
7.03 (m)
7.13 (d, 2.0)
6.96 (m)
6.84 (m)
6.86 (m)
6.24 (d, 14.0)
7.12 (d, 14.0)
8
20
50
60
100
200

5750
A.N. Pearce et al. / Tetrahedron 64 (2008) 5748–5755
Table 2
13C NMRa and HMBCb data for 2–5 (CD3OD)
Position
2
3
4
5
dC
dC
dC
dC
2
3
4a
5
79.8 (3)
78.8 (3)
77.4 (20, 60)
79.8
141.9 (3, 5, 6, 8)
118.7
76.6 (20, 60)
78.8
140.6 (3, 5, 6, 8)
119.3
77.5 (2, 20, 60)
143.4 (3, 5, 6, 8)
118.6 (6)
76.6 (20, 60)
143.7 (8)
119.1
6
7
8
121.2 (8, 100)
131.5 (5, 200)
115.5 (6, 100)
142.7 (2, 5, 8)
128.0 (2, 3, 20, 50)
115.4 (3, 60)
146.8 (20, 50)
147.5 (20, 60)
116.4
120.3 (3, 20)
116.8 (6, 200)
122.0 (100)
156.2 (200)
158.7 (2)
121.4 (8, 100)
132.2 (5, 200)
116.1 (6, 100)
141.5 (2, 5, 8)
126.9 (3, 50)
114.2 (3, 60)
146.9 (50)
121.2 (8, 100)
131.5 (5, 200)
115.3 (6, 100)
144.2 (2, 5)
128.0 (3, 50)
115.5 (2, 60)
146.7 (20, 50)
147.4 (20, 60)
116.5
121.6 (8, 100)
131.9 (5, 200)
115.9 (6, 100)
144.6 (2, 5, 8)
126.9 (2, 20, 50)
114.3 (2, 60)
146.9 (50)
8a
10
20
30
40
50
60
100
200
400
2b
147.1 (20, 60)
116.6
147.1 (20, 60)
116.5
118.5 (3, 20)
116.7 (6, 8, 200)
122.3 (100)
156.2 (200)
159.1 (2)
120.4 (2, 20)
116.9 (6)
118.6 (2, 20)
116.6 (6, 8)
122.2 (100)
156.2 (200)
159.1 (3)
121.9
156.0 (200)
158.6 (3)
a
Obtained at 150 MHz.
Numbers in parentheses are protons to which the carbon correlated to in 1H–13C HMBC NMR experiments.
b
antiperiplanar relationships (Fig. 1a). Thus it was concluded that
orthidine A (2) bears the 2-guanidyl and 3-(3,4-dihydroxyphenyl)
groups in a 2,3-trans diaxial (H-2 equatorial, H-3 equatorial) con-
figuration. The lack of specific rotation and absence of CD maxima
implied that 2 was a racemate.
1,4-Benzodioxane derivatives resulting from dimerization have
been reported from a number of sources including neolignans
americanol A and isoamericanol A isolated from the seeds of the
plant Phytolacca americana,10 ascidian metamorphosis-inducing
sulfated 3,4-dihydroxystyrene dimers isolated from the marine
sponge Jaspis sp.,11 and N-acetyldopamine dimers isolated from the
sclerotization of insect cuticles.12–14 Facile racemization of the latter
dimers was noted13 and more recently such dimers have been
reported to exhibit antioxidant and anti-inflammatory biological
activities.14 Tubastrine and related compounds15,16 have been
reported from a number of divergent phyla (Chordata, Porifera, and
Cnidaria)dthis is the first report of tubastrine dimers.
orthidine B (3) from H-2 to H-3, H-20, and H-60 and from H-3 to H-2,
H-20, and H-60. A J-HMBC NMR experiment determined the
3JC-8a–H-2 coupling constant to be 10.3 Hz while the lack of an ob-
3
served correlation attributed a coupling constant for JC-4a–H-3 of
less than 1 Hz. These results are consistent with H-2 being anti-
periplanar to C-8a (i.e., guanidyl axial) and H-3 being orthogonal
to C-4a (i.e., dihydroxylphenyl equatorial) (Fig. 1b). A lack of de-
tectable specific rotation or CD maxima indicated that 3 was also
isolated as a racemate.
Spectroscopic and spectrometric data observed for orthidine C
(4) were very similar to those observed for orthidine A (2). In-
spection of 1H and 13C NMR data (Tables 1 and 2) and analysis of
HSQC and HMBC spectra indicated the presence of the same sub-
structural fragments, i.e., a 3,4-dioxa-phenyl-(E)-ethenyl guanidine
fragment similar to tubastrine and a second fragment comprised of
a 3,4-dihydroxyphenyl, two oxymethines (dC 77.4, dH 4.91, d,
J¼5.7 Hz; dC 79.8, dH 5.63, d, J¼5.7 Hz), and a guanidine group (
158.6). Interfragment HMBC correlations between H-2 and H-5 to
C-8a ( 144.2) and from H-3/H-6/H-8 to C-4a ( 141.9) indicated that
d
Orthidine B (3) had the same molecular formula as 2 and
exhibited very similar 1H and 13C NMR resonances (Tables 1 and 2).
Major differences were observed, however, for resonances associ-
ated with the oxymethines C-2 and C-3 (dC 78.8, dH 5.87; dC 76.6, dH
5.23). Analysis of HSQC and HMBC NMR data led to the same planar
structure as orthidine A (2) with an HMBC correlation observed
d
d
the 3,4-dihydroxyphenyl and guanidine groups were substituted
on C-2 and C-3, respectively; the reversed attachment order to that
observed for orthidine A (2). The magnitude of the 3JH-2–H-3 vicinal
coupling constant (5.7 Hz) supported a trans relative configuration
and as with 2, J-HMBC NMR data indicated that both 3JC-8a–H-2 and
3JC-4a–H-3 coupling constants were 7.6 Hz, inferring that both H-2
and H-3 were equatorial. Lack of chiroptical properties indicated 4
was also a racemate.
between H-2 and C-8a (d 141.5) being important in establishing
3
regiochemistry. The observation of a JH-2–H-3 vicinal coupling
constant of 1.9 Hz indicated a cis relative configuration between
H-2 and H-3. As with 2, NOESY correlations were also observed for
3
3
Figure 1. Relative stereochemistry of (a) orthidine A (2) and (b) orthidine B (3) as determined by interpretation of NOESY, JHH and JCH NMR data.

A.N. Pearce et al. / Tetrahedron 64 (2008) 5748–5755
5751
Table 3
H-7 to C-7 and from H-8 to C-8 suggested that 6 was a 1,2,3,4-
tetrasubstituted cyclobutane related to tubastrine by a [p2sþp2s]
cycloaddition reaction. Such photochemically allowed reaction
products can be described as deriving from head-to-head or head-
to-tail cycloaddition. Examples of natural products resulting from
head-to-head, e.g., sceptrin,17 sagerinic acid,18 monochaetin,19
moslolignans,20 and mooniines21 or head-to-tail, e.g., p-coumaric
and ferulic acid dimers,22 4,40-dihydroxytruxillic acid,23 achyro-
dimers A–C,24 and piperarborenines A and B25 dimerization have
been reported.
1H (600 MHz), 13C (150 MHz), and HMBCa data for 6 (CD3OD)
Position
dH (mult., J in Hz)
dC
1H–13C HMBC
NOESY
1, 10
146.2
146.7
116.9
128.0
121.0
116.6
50.1
3/30, 5/50
6/60
5/50, 7/70
6/60, 7/70, 8/80
3/30, 7/70
2, 20
3, 30
6.84 (d, 1.9)
7/70, 8/80
7/70, 8/80
4, 40
5, 50
6.75 (dd, 8.1, 1.9)
6.80 (d, 8.1)
3.88 (dd, 8.0, 6.7)
4.66 (dd, 8.0, 6.7)
7.94 (br d)b
6, 60
7, 70
3/30, 5/50, 7/70, 8/80
7/70, 8/80
8/80, 3/30, 5/50
7/70, 3/30, 5/50
8, 80
53.0
NH-9, 90
10, 100
To discern between the possible regioisomeric structures of 6
resulting from head-to-head (A) or head-to-tail (B) cycloaddition
(Fig. 2a), a 2J,3J-HMBC long-range NMR experiment was used.26 This
experiment, which introduces a STAR operator into proton-detec-
ted long-range heteronuclear correlation pulse sequences, allows
differentiation between two-bond and three-bond long-range
correlations to protonated carbons. By use of a Jscale parameter, an
158.2
8/80
a
Protons to which the carbon correlated to in 1H–13C HMBC NMR experiments.
Obtained in DMSO-d6 (300 MHz).
b
Compound 6 (orthidine E) was isolated as an optically inactive
pale yellow gum and exhibited a pseudo-molecular ion at m/z 387
([MþH]þ C18H23N6O4) under electrospray ionization. As seen for
orthidines A–C, a doubly ionized ion, this time at m/z 194 attrib-
utable to [Mþ2H]2þ, was also detected. Analysis of 1H, 13C, HSQC-
DEPT, and HMBC NMR data (Table 3) established the presence of
the now familiar 1,2,4-trisubstituted dihydroxylated phenyl ring,
two coupled methines (H-7 (dH 3.88, dd, J¼8.0, 6.7 Hz, dC 50.1); H-8
2
F1 dimension skew is apparent in JCH correlations while no mod-
3
ulation of JCH correlations occurs. As shown in Figure 2b and c,
correlations observed in the 2J,3J-HMBC experiment of 6 carried out
in both CD3OD and D2O solvents clearly show that H-7/70 correlates
3
to C-7/70 with a JCH-type signal, as does H-8/80 with C-8/80, while
2
H-7/70 exhibits a staggered F1 skewed JCH correlation to C-8/80 as
also observed for H-8/80 to C-7/70. Thus orthidine E (6) represents a
relatively rare example of a naturally occurring head-to-tail cyclo-
addition product.
(
dH 4.66, dd, J¼8.0, 6.7 Hz, dC 53.0)), and guanidine (
d 158.2) frag-
ments. These fragments accounted for exactly half of the required
molecular formula, indicating that 6 was a dimer. Long-range
HMBC correlations were observed from H-7 (
116.9) and C-5 ( 121.0) of the catechol ring and C-8 (
from H-8 ( 4.66) to C-7 ( 50.1) and to the guanidine carbon, C-10
158.2), consistent with the presence of a dihydrotubastrine
substructure. Inter-substructure HMBC correlations observed from
d
3.88) to C-3 (
d
Data obtained in an NOESY NMR experiment was used to de-
termine the relative configuration of 6. There are three relative
stereochemical geometries possible for 6, being cis–trans–cis (C),
cis–cis–cis (D) or trans–trans–trans (E) as shown in Figure 3.27–29
NOESY correlations observed between the cyclobutane protons
d
d
53.0) and
d
d
(d
Figure 2. (a) Alternative head-to-head (A) or head-to-tail (B) cycloaddition structures of orthidine E (6) and relevant regions of 2J,3J-HMBC NMR spectra (Jscale 16) of 6 obtained in
(b) CD3OD and (c) D2O highlighting ‘normal’ shaped correlations between H-7/70/C-7/70 and H-8/80/C-8/80 and ‘F1 skew’ shaped correlations between H-7/70/C-8/80 and H-8/80/C-7/
3
70 establishing the JCH bonding arrangement between H-7/70 and C-7/70 and H-8/80 and C-8/80 and hence defining orthidine E as being head-to-tail (i.e., B).

5752
A.N. Pearce et al. / Tetrahedron 64 (2008) 5748–5755
Figure 3. Alternative relative configurations (C–E) of orthidine E (6) with observed NOESY correlations indicated on C.
H-7/70 and H-8/80 are not consistent with the all-trans configura-
tion E. NOESY correlations were also observed between H-7/70 and
H-8/80 to H-3/30 and H-5/50, which is possible for either of config-
urations C and D. While all-cis configurations such as D have been
discounted on the grounds of steric congestion,28 previous studies
implying dimerization through C-15. This was supported by the
observation of HMBC correlations from CH2-15 (dH 1.65) to C-15 (dC
22.4) leading to the conclusion that orthidine F was the 1,14-
dihomovanillamide derivative of spermine (7). Reaction of sper-
mine with homovanillic acid in the presence of coupling agent
PyBOP yielded 7 (39%) that was identical to the natural product in
all respects. Structurally, orthidine F (7) is closely related to the
plant derived dihydrocaffeamide spermine alkaloid kukoamine A,30
which exhibits hypotensive activity and also inhibits the trypano-
somatid parasite enzyme trypanothione reductase.31
indicating cyclobutane 1H–1H-coupling constants of 3Jcis 11 Hz and
27,29
3Jtrans 6 Hz
provide stronger evidence that the coupling con-
stants observed for 6 (H-7/H-8, dd, J¼8.0, 6.7 Hz) arise from a cis–
trans–cis (i.e., C), rather than the all-cis (D), relative configuration.
Orthidine F (7) had a molecular formula of C28H42N4O6 as
determined by HRFABMS. Inspection of the 1H and 13C NMR spectra
(DMSO-d6) (Table 4) accounted for 14 carbons and 21 protons, in-
cluding 3 exchangeables, indicating that 7 was another symmetri-
cal dimer. The presence of a homovanilloyl fragment was suggested
Inspection of non-biologically active column fractions by
HPLC-DAD and NMR for related metabolites that could aid in
structure–activity relationship studies yielded compound 8. In-
spection of 1H, 13C, COSY, HSQC, and HMBC NMR data established 8
by inspection of 1H NMR [
J¼1.8 Hz, H-3; 6.70, 1H, d, J¼8.0 Hz, H-6; 6.64, 1H, dd, J¼8.0, 1.8 Hz,
H-5; 3.74, 3H, s, OCH3; 3.29, 2H, s, H2-7] and 13C NMR [
147.2 (s),
d
8.83, 1H, br s, OH-17; 6.85, 1H, d,
as being 3,4-dimethoxyphenethyl-
b-guanidine, previously repor-
ted as a synthetic product.8
d
Investigation of a second collection of A. orthium yielded 5
(orthidine D), which proved very difficult to purify and so was
characterized as a 2:1 mixture with 4. Whilst signal overlap ham-
pered analysis of the 1H NMR spectrum, adequate 13C resonance
dispersion allowed for complete assignment of the structure of the
major component of the mixture (Tables 1 and 2). The planar
structure of 5 was determined to be the same as orthidine C (4) (i.e.,
2-(3,4-dihydroxyphenyl) and 3-guanidyl substituted) by in-
145.0 (s), 127.0 (s), 121.1 (d), 115.2 (d), 113.2 (d), 55.5 (q), 41.9 (t)]
data. Intra-fragment HSQC and long-range 1H–13C HMBC NMR
correlations confirmed this. An isolated 1,5-diazaheptanyl 12-pro-
ton spin system, -NHCH2CH2CH2NHCH2CH2-, was also identified by
interpretation of COSY, selective TOCSY, HSQC, and HMBC NMR
data. Connection to the homovanilloyl fragment via an amide bond
was established by observation of HMBC correlations from NH-9
3
(d
8.23) and CH2-10 (dH 3.11) to C-8 (dC 170.9). This combined
terpretation of HSQC and HMBC NMR data. A JH-2–H-3 vicinal
substructure accounted for all observed 1H and 13C NMR reso-
nances and amounted to exactly half of the molecular formula,
coupling constant of 2.0 Hz indicated a C-2/C-3 cis relative config-
uration, which was supported by NOESY NMR correlations (data
3
not shown). A J-HMBC NMR experiment determined the JC-4a–H-3
coupling constant to be 10.2 Hz while the lack of an observed
correlation attributed a coupling constant for 3JC-8a–H-2 of less than
1 Hz. These results are consistent with H-3 being antiperiplanar to
C-4a (i.e., guanidyl axial) and H-2 being orthogonal to C-8a (i.e.,
dihydroxylphenyl equatorial). The mixture of 4 and 5 was optically
inactive suggesting that 5 was a racemate.
Table 4
1H (400 MHz), 13C (100 MHz), and HMBCa data for 7 (DMSO-d6)
Position
dH (mult., J in Hz)
dC
1H–13C HMBC
1, 10
145.0
147.2
113.2
127.0
121.1
115.2
41.9
OH-17/170, 3/30, 5/50
OH-17/170, 6/60, 16/160
5/50, 7/70
2, 20
3, 30
6.85 (d, 1.8)
The biogenesis of orthidines A–D is likely to be similar to that
proposed for neolignans or lignoids, which proceeds via phenol
4, 40
6/60, 7/70
5, 50
6.64 (dd, 8.0, 1.8)
6.70 (d, 8.0)
3.29 (s)
3/30, 7/70
6, 60
OH-17/170
oxidation, followed by O-b radical coupling and ring closure of re-
7, 70
3/30, 5/50
sultant quinone methides (Fig. 4).32 If not enzymatically controlled,
the coupling and ring closure steps would be expected to yield
regio- and stereoisomers as observed in the current study.
8, 80
170.9
7/70, NH-9/90, 10/100
NH-9, 90
10, 100
11, 110
12, 120
NH-13, 130
14, 140
15, 150
16, 160
OH-17, 170
8.23 (br t)
3.11 (m)
35.7
25.7
44.3
11/110
1.77 (m)
2.82 (m)
9.05 (br m)
2.83 (m)
1.65 (m)
3.74 (s)
12/120
10/100, 11/110
2.2. Biological evaluation
45.8
22.4
55.5
14/140, 15/150
Tubastrine (1) and orthidines A, B, C, E, and F (2, 3, 4, 6, 7)
inhibited in vitro superoxide production by PMA-stimulated hu-
man neutrophils in a dose-dependent manner (Table 5).5 The
compounds exhibited good selectivity, with low or non-detectable
8.83 (br s)
a
Protons to which the carbon correlated to in 1H–13C HMBC NMR experiments.

A.N. Pearce et al. / Tetrahedron 64 (2008) 5748–5755
5753
O
OH
O
OH
OH
O
OH
OH
O
O
OH
O
NH
NH2
HN
H2N
NH
NH2
HN
H2N
HN
H2N
H2N
NH2
HN
H2N
H2N
NH2
NH2
NH2
Figure 4. Proposed biogenesis of orthidines A and B from tubastrine.32
Table 5
Summary of biological activities observed for compounds 1–4, 6–8a
in determining the head-to-tail dimerization substructure of the
cyclobutane analogue orthidine E. While structure elucidation of
the numerous examples of naturally occurring cyclobutane dimers
to date has relied upon mass spectrometric fragmentation patterns,
X-ray crystallography or chemical shift comparisons with pre-
viously reported compounds, the ability to differentiate between
b
c
Compound
AI50
AP50
X/XOd
1
2
3
4
6
7
8
SOD
27.21ꢂ0.03 (n¼2)
15.06ꢂ0.23 (n¼2)
11.80ꢂ1.34 (n¼2)
14.13ꢂ1.13 (n¼2)
10.67ꢂ1.57 (n¼2)
36.25ꢂ6.18 (n¼2)
262.8ꢂ48.77 (n¼2)
0.65ꢂ0.06
>258 (n¼2)
118.0 (n¼1)
156.0ꢂ6.38 (n¼2)
133.4 (n¼1)
>1000 (n¼2)
>165 (n¼2)
118.4ꢂ2.64 (n¼2)
67.47ꢂ2.78 (n¼2)
67.96ꢂ6.91 (n¼2)
215.8ꢂ13.46 (n¼2)
>893 (n¼2)
3
2JCH and JCH HMBC correlations provides a rapid and robust alter-
>129 (n¼2)
native method. A sixth new compound, the dihomovanillamide
derivative of spermine was characterized by spectroscopic tech-
niques and confirmed by synthesis. Several of the compounds were
found to inhibit the in vitro production of superoxide by PMA-
stimulated human neutrophils and were also active in an in vivo
acute model of gouty arthritis providing further evidence that in-
tervention of neutrophil-mediated processes can led to the dis-
covery of potential anti-inflammatory agents. These results further
highlight the tremendous diversity of natural product architecture
biosynthesized by ascidians of the genus Aplidium.37
101.2ꢂ13.69 (n¼2)
>893 (n¼2)
Methotrexate
Allopurinol
6.92 nMꢂ0.84 nM
42.24ꢂ5.86
a
Values are IC50 values (units of mM unless stated otherwise) representing the
mean of at least two determinationsꢂstandard error.
b
AI50: anti-inflammatory activity. Concentration of compound required to inhibit
in vitro superoxide production by phorbol-12-myristate 13-acetate (PMA)-stimu-
lated human neutrophils by 50%.
c
AP50: antiproliferative activity. Concentration of compound required to inhibit
HL60 cell proliferation by 50%.
d
X/XO: xanthine/xanthine oxidase inhibitory activity. Concentration of com-
pound required to inhibit superoxide production in xanthine/xanthine oxidase assay
by 50%.
4. Experimental section
4.1. General methods (chemistry)
antiproliferative activity (AP50) against human leukemic HL60 cells.
Significantly weaker suppression of superoxide production was
observed for 3,4-dimethoxyphenethyl-b-guanidine (8), presumably
Flash column chromatography was performed using reversed
phase Merck Lichroprep RP-8 or RP-18 40–63 mm, and size exclu-
due to the lack of phenolic functionality. In an effort to determine
the influence of non-specific antioxidant effects, all compounds
were tested in a xanthine/xanthine oxidase enzyme-based super-
oxide scavenging assay.4 As also shown in Table 5, all of the com-
pounds were considerably less effective as superoxide scavengers
than as suppressors of superoxide production by neutrophils. The
compounds were also tested for inhibition of neutrophil infiltration
and superoxide production in an acute model of gouty arthritis as
sion chromatography on Pharmacia Biotech SephadexÒ LH-20. UV–
vis spectra were run as MeOH solutions on a UV-2102 PC Shimadzu
UV–vis scanning spectrophotometer. IR spectra were acquired as
either dry films or Nujol mulls on a Spectrum One FTIR spectrom-
eter with the 1603 cmꢀ1 absorption band of polystyrene being used
as reference. NMR spectra were recorded on either a Bruker Avance
DRX-600 spectrometer at 600 MHz for 1H and 150 MHz for 13C, or
a Bruker Avance DRX-400 spectrometer at 400 MHz for 1H and
100 MHz for 13C. Proto-deutero solvent signals were used as ref-
erence (DMSO-d6: dH, 2.50; dC, 39.4. CD3OD: dH, 3.30; dC, 49.0.
CDCl3: dH, 7.25; dC, 77.0 ppm), or in the case of 1H and 13C NMR data
for 6 acquired in D2O, external reference to DSS (dH 0.00; dC
0.0 ppm) was used. MS were recorded on either a Bruker Daltonics
MicrOTOF Spectrophotometer or a VG-7070 mass spectrometer.
Compound purity was determined by reversed phase HPLC (Waters
previously described.4 At a dose of 25
mmol/kg, administered orally,
all of the compounds inhibited superoxide production by neutro-
phils in vivo (30–70%) with no indication of acute liver toxicity
compared to controls (data not shown). Interestingly, tubastrine (1)
and orthidine E (6) also inhibited neutrophil infiltration in vivo. In
contrast, compound 8 failed to inhibit in vivo neutrophil in-
filtration, providing further evidence that free phenolic groups play
a key role in the bioactivity of the orthidine alkaloids.
600 HPLC photodiode array system, Alltech Econosil C18, 3 mm,
33ꢁ7 mm, H2O (0.05% TFA) to MeCN over 13.5 min at 2.0 mL/min
3. Conclusion
and monitoring at 280 nm).
Our search for new classes of anti-inflammatory agents has led
to the isolation of a series of guanidine-containing alkaloids33–36
from the New Zealand ascidian A. orthium. The known dihydroxy-
styrylguanidine tubastrine (1) was isolated in addition to five new
alkaloids that contained benzodioxane (orthidines A–D) and
cyclobutane (orthidine E) skeletons arising from the dimerization
of tubastrine. In the case of orthidines A–D, relative stereochem-
istry assignment relied upon the acquisition and interpretation of
3JCH coupling constants that can be more readily obtained via re-
cently reported indirect methods than the more classical carbon-
detection methods. The use of a 2J,3J-HMBC experiment was pivotal
4.2. Biological material
The ascidian A. orthium38 was collected (1263 g) by SCUBA at
a depth of ꢀ30 m from Northeast Island, Three Kings Island group,
New Zealand in November 2002 and a second collection (1022 g)
was made in the same location in December 2003. Both collections
were kept frozen until used. Specimens were identified by one of us
(M.P.) and a voucher specimen is held at the National Institute for
Water and Atmospheric Research, Private Bag 14-901, Kilbirnie,
Wellington, New Zealand as MNP7030.

5754
A.N. Pearce et al. / Tetrahedron 64 (2008) 5748–5755
4.3. Isolation and purification
[Mþ2H]2þ 193.0848, C9H11N3O2 requires 193.0846; purity 99%
tR¼4.84 min.
The frozen specimens from the first collection were freeze-dried
(82.4 g) and extracted with MeOH (6ꢁ100 mL) followed by CH2Cl2
(2ꢁ100 mL). The combined extracts were filtered and dried in
vacuo to produce a crude extract (26.2 g) that was subjected to
reversed phase C18 flash column chromatography with a steep
gradient from H2O (0.05% trifluoroacetic acid (TFA)) to MeOH
(0.05% TFA). The anti-inflammatory assay showed the activity to be
concentrated in the water and 10–20% MeOH/H2O (0.05% TFA)
fractions. Repeated C18 and C8 flash column chromatography (H2O
(0.05% TFA) to MeOH (0.05%TFA)) followed by size exclusion col-
umn chromatography on Sephadex LH-20 yielded 1 (12.5 mg, 0.02%
dry wt), 2 (8.9 mg, 0.01% dry wt), 3 (15.3 mg, 0.02% dry wt), 4
(20.2 mg, 0.02% dry wt), 6 (16.1 mg, 0.02% dry wt), 7 (4.3 mg,
0.005% dry wt), and 8 (17.2 mg, 0.02% dry wt). A second collection
of the same species (51.1 g dry wt, 15.4 g crude extract) was pro-
cessed in the same manner yielding 1 (5.6 mg, 0.01% dry wt) and 5
in a 3:2 ratio with 3 (6.6 mg, 0.01% dry wt). Compounds 2, 4, 6, 7,
and 8 were detected by analytical HPLC but not isolated.
4.3.6. Orthidine E (6)
Pale yellow gum; 1H, 13C, and HMBC NMR data are reported in
Table 3. 1H NMR (600 MHz, D2O)
d
6.96 (2H, d, J¼8.2 Hz, H-6/60),
6.90 (2H, d, J¼2.2 Hz, H-3/30), 6.85 (2H, dd, J¼8.2, 2.2 Hz, H-5/50),
4.63 (2H, dd, J¼8.2, 6.5 Hz, H-8/80), 3.93 (2H, dd, J¼8.2, 6.5 Hz, H-
7/70); 13C NMR (100 MHz, D2O)
d
158.8 (C-10/100), 146.7 (C-2/20),
146.2 (C-1/10), 130.1 (C-4/40), 123.3 (C-5/50), 118.9 (C-3/30, 6/60), 53.7
(C-8/80), 50.5 (C-7/70); UV (MeOH) lmax (log
) 206 (4.45), 230
3
(3.83), 285 (3.45) nm; IR nmax (smear) 3335, 3173, 1671, 1610, 1521,
1444, 1369, 1287, 1201, 1119, 1022, 990 cmꢀ1; ESIMS m/z (rel int %)
387 (30), 194 (100); HRESIMS m/z [MþH]þ 387.1768, C18H23N6O4
requires 387.1775; [Mþ2H]2þ 194.0918, C9H12N3O2 requires
194.0924; purity 99% tR¼1.14 min.
4.3.7. Orthidine F (7)
Pale yellow gum; 1H, 13C, and HMBC NMR data reported in Table
4. UV (MeOH) lmax (log 3) 206 (4.10). 230 (3.46), 285 (3.10) nm; IR
nmax (smear) 3328, 3188, 1677, 1514, 1433, 1270, 1202, 1132,
4.3.1. Tubastrine (1)6,7
800 cmꢀ1
;
FABMS m/z [MþH]þ 531, HRFABMS 531.3188,
Pale yellow gum; UV (MeOH) lmax (log
3
) 205 (3.79), 221 (3.79),
C28H43N4O6 requires 531.3183; purity 97% tR¼5.66 min.
287 (3.79), 306 sh (3.68) nm; IR nmax (smear) 3326, 3176, 1677, 1607,
1522, 1445, 1280, 1195, 1113, 930 cmꢀ1; ESIMS m/z [MþH]þ 194,
HRESIMS m/z 194.0909, C9H12N3O2 requires 194.0924; 1H NMR
4.3.8. Synthesis of orthidine F (7)
Homovanillic acid (98 mg, 0.54 mmol), spermine (54 mg,
0.27 mmol), and PyBOP (281 mg, 0.54 mmol) were stirred together
(DMSO-d6, 400 MHz)
d
10.06 (1H, d, J¼10.0 Hz, NH-9), 9.05 (1H, s,
OH-1), 8.80 (1H, s, OH-2), 7.60 (4H, br s, NH2ꢁ2), 7.10 (1H, dd,
J¼13.9, 10.2 Hz, H-8), 6.81 (1H, d, J¼1.4 Hz, H-3), 6.68 (2H, m,
H-5,6), 6.02 (1H, d, J¼13.9 Hz, H-7); 13C NMR (DMSO-d6, 100 MHz)
in DMF (1 mL). Triethylamine (225 mL, 1.62 mmol) was added and
the reaction mixture was stirred at ambient temperature under N2
for 22 h. The crude product was subjected to reversed phase C8
flash column chromatography followed by reversed phase phenyl-
bonded column chromatography (H2O to MeOH), with the product
eluting in the 100% MeOH fraction yielding 7 (55.6 mg 39%) as
a pale yellow gum. FABMS m/z [MþH]þ 531, HRFABMS 531.3184,
d
153.9 (C-10), 145.1 (C-2), 144.6 (C-1), 126.9 (C-4), 119.8 (C-8),
117.3 (C-5), 115.6 (C-6), 114.7 (C-7), 112.7 (C-3); purity 99%
tR¼0.90 min.
4.3.2. Orthidine A (2)
C
28H43N4O6 requires 531.3183; UV, IR, and 1H and 13C NMR data
Pale yellow gum; 1H, 13C, and HMBC NMR data are reported in
were identical to those observed for the isolated natural product.
Purity 98% tR¼5.64 min, and co-eluted with isolated natural
product.
Tables 1 and 2. UV (MeOH) lmax (log 3) 207 (4.56), 226 (4.17), 284
(4.02), 311 sh (3.73) nm; IR nmax (smear) 3333, 3189, 1676, 1606,
1507,1438,1267,1201, 1139, 810 cmꢀ1; ESIMS m/z (rel int %) 385 (7),
326 (23), 193 (100); HRESIMS m/z [MþH]þ 385.1606, C18H21N6O4
requires 385.1619; [Mþ2H]2þ 193.0858, C9H11N3O2 requires
193.0846; purity 99% tR¼5.30 min.
4.3.9. 3,4-Dimethoxyphenethyl-
b
-guanidine (8)8
) 205 (4.07), 229 (3.52),
Pale yellow gum; UV (MeOH) lmax (log
3
281 (3.11) nm; IR nmax (smear) 3350, 3184, 1671, 1517, 1466, 1262,
1235, 1202, 1139, 1024, 801 cmꢀ1; FABMS m/z [MþH]þ 224,
4.3.3. Orthidine B (3)
HRFABMS 224.1394,
(CD3OD, 400 MHz)
C
11H18N3O2 requires 224.1399; 1H NMR
6.89 (1H, d, J¼8.2 Hz, H-6), 6.85 (1H, d,
Pale yellow gum; 1H, 13C, and HMBC NMR data are reported in
d
Tables 1 and 2. UV (MeOH) lmax (log
3
) 205 (4.55), 226 (4.16), 284
J¼2.0 Hz, H-3), 6.78 (1H, dd, J¼8.2, 2.0 Hz, H-5), 3.82 (3H, s, H3-14),
3.79 (3H, s, H3-13), 3.42 (2H, t, J¼7.1 Hz, H2-8), 2.81 (2H, t, J¼7.1 Hz,
(4.08), 308 sh (3.83) nm; IR nmax (smear) 3344, 3187, 1677, 1606,
1507, 1439, 1267, 1202, 1141, 932 cmꢀ1; ESIMS m/z (rel int %) 385
(35), 326 (10), 193 (100); HRESIMS m/z [MþH]þ 385.1612,
H2-7); 13C NMR (CD3OD, 100 MHz)
d 158.7 (C-10), 150.6 (C-2), 149.4
(C-1), 132.2 (C-4), 122.3 (C-5), 113.8 (C-3), 113.3 (C-6), 56.6 (C-14),
56.5 (C-13), 43.9 (C-8), 35.6 (C-7); purity 99% tR¼4.89 min.
C
18H21N6O4 requires 385.1619; [Mþ2H]2þ 193.0861, C9H11N3O2 re-
quires 193.0846; purity 95% tR¼4.65 min.
4.3.4. Orthidine C (4)
4.4. Biological assays
Pale yellow gum; 1H, 13C, and HMBC NMR data are reported in
Tables 1 and 2. UV (MeOH) lmax (log
3
) 205 (4.59), 224 (4.32), 284
Details of general procedures for the in vitro and in vivo bi-
ological assays undertaken in the present study have been reported
previously.4
(4.23), 311 sh (3.97) nm; IR nmax (smear) 3326, 3173, 1677, 1606,
1507, 1442, 1263, 1197, 1118, 933 cmꢀ1; ESIMS m/z (rel int %) 385 (3),
326 (16), 193 (100); HRESIMS m/z [MþH]þ 385.1634, C18H21N6O4
requires 385.1619; [Mþ2H]2þ 193.0899, C9H11N3O2 requires
193.0846; purity 98% tR¼5.14 min.
Acknowledgements
4.3.5. Orthidine D (5)
This work was supported by the New Zealand Foundation for
Research Science and Technology, contract CO1X0205. We wish
to thank M. Walker and M. Schmitz for assistance with NMR
and R. Imatdieva and P. Greed (Waikato University) for MS
acquisition.
Pale yellow gum; 1H, 13C, and HMBC NMR data are reported in
Tables 1 and 2. UV (MeOH) lmax (log 3) 206 (4.59), 227 (4.23), 287
(4.25), 310 sh (4.01) nm; ESIMS m/z (rel int %) 385 (7), 193 (100);
HRESIMS m/z [MþH]þ 385.1622, C18H21N6O4 requires 385.1619;

A.N. Pearce et al. / Tetrahedron 64 (2008) 5748–5755
5755
18. Lu, Y.; Foo, L. Y. Phytochemistry 1999, 51, 91–94.
19. Isaza, J. H.; Ito, H.; Yoshida, T. Phytochemistry 2001, 58, 321–327.
References and notes
20. Wang, Q.; Terreaux, C.; Marston, A.; Tan, R.-X.; Hostettmann, K. Phytochemistry
2000, 54, 909–912.
21. Rahman, A.-U.; Khattak, K. F.; Nighat, F.; Shabbir, M.; Hemalal, K. D.; Till-
ekeratne, L. M. Phytochemistry 1998, 48, 377–383.
22. Ford, C. W.; Hartley, R. D. J. Sci. Food Agric. 1990, 50, 29–43.
23. Tachibana, S.; Ohkubo, K.; Towers, G. H. N. Phytochemistry 1992, 31, 81–83.
24. Sagawa, T.; Takaishi, Y.; Fujimoto, Y.; Duque, C.; Osorio, C.; Ramos, F.; Garzon, C.;
Sato, M.; Okamoto, M.; Oshikawa, T.; Ahmed, S. U. J. Nat. Prod. 2005, 68, 502–
505.
25. Lee, F.-P.; Chen, Y.-C.; Chen, J.-J.; Tsai, I.-L.; Chen, I.-S. Helv. Chim. Acta 2004, 87,
463–468.
26. Krishnamurthy, V. V.; Russell, D. J.; Hadden, C. E.; Martin, G. E. J. Magn. Reson.
2000, 146, 232–239.
27. Ulrich, H.; Rao, D. V.; Stuber, F. A.; Sayigh, A. A. R. J. Org. Chem. 1970, 35, 1121–
1125.
28. Wei, K.; Li, W.; Koike, K.; Chen, Y.; Nikaido, T. J. Org. Chem. 2005, 70, 1164–1176.
29. Montaudo, G.; Caccamese, S. J. Org. Chem. 1973, 38, 710–716.
30. Funayama, S.; Yoshida, K.; Konno, C.; Hikino, H. Tetrahedron Lett. 1980, 21, 1355–
1356.
1. Pearce, A. N.; Berridge, M. V.; Harper, J. L.; Maas, E. W.; Perry, N. B.; Webb, V. A.;
Copp, B. R. Chem. N.Z. 2007, 71, 47–49.
2. McNamara, C. E.; Larsen, L.; Perry, N. B.; Harper, J. L.; Berridge, M. V.; Chia, E. W.;
Kelly, M.; Webb, V. L. J. Nat. Prod. 2005, 68, 1431–1433.
3. Pearce, A. N.; Chia, E. W.; Berridge, M. V.; Maas, E. W.; Page, M. J.; Webb, V. L.;
Harper, J. L.; Copp, B. R. J. Nat. Prod. 2007, 70, 111–113.
4. Pearce, A. N.; Chia, E. W.; Berridge, M. V.; Clark, G. R.; Harper, J. L.; Larsen, L.;
Maas, E. W.; Page, M. J.; Perry, N. B.; Webb, V. L.; Copp, B. R. J. Nat. Prod. 2007, 70,
936–940.
5. Tan, A. S.; Berridge, M. V. J. Immunol. Methods 2000, 238, 59–68.
6. Sakai, R.; Higa, T. Chem. Lett. 1987, 127–128.
7. Barenbrock, J. S.; Ko¨ck, M. J. Biotechnol. 2005, 117, 225–232.
8. Short, J. H.; Biermacher, U.; Dunnigan, D. A.; Leth, T. D. J. Med. Chem. 1963, 6,
275–283.
9. Meissner, A.; Sørensen, O. W. Magn. Reson. Chem. 2001, 39, 49–52.
10. Fukuyama, Y.; Hasegawa, T.; Toda, M.; Kodama, M.; Okazaki, H. Chem. Pharm.
Bull. 1992, 40, 252–254.
11. Tsukamoto, S.; Kato, H.; Hirota, H.; Fusetani, N. Tetrahedron 1994, 50, 13583–
13592.
12. Andersen, S. O.; Jacobsen, J. P.; Roepstorff, P. Tetrahedron 1980, 36, 3249–3252.
13. Noda, N.; Kubota, S.; Miyata, Y.; Miyahara, K. Chem. Pharm. Bull. 2000, 48, 1749–
1752.
14. Xu, M.-Z.; Lee, W. S.; Han, J.-M.; Oh, H.-W.; Park, D.-S.; Tian, G.-R.; Jeong, T.-S.;
Park, H.-Y. Bioorg. Med. Chem. 2006, 14, 7826–7834.
31. Ponasik, J. A.; Strickland, C.; Faerman, C.; Savvides, S.; Karplus, P. A.; Ganem, B.
Biochem. J. 1995, 311, 371–375.
32. Arnoldi, A.; Merlini, L. J. Chem. Soc., Perkin Trans. 1 1985, 2555–2557.
33. Berlinck, R. G. S.; Kossuga, M. H. Nat. Prod. Rep. 2005, 22, 516–550.
34. Berlinck, R. G. S. Nat. Prod. Rep. 2002, 19, 617–649.
35. Berlinck, R. G. S. Nat. Prod. Rep. 1999, 16, 339–365.
36. Berlinck, R. G. S. Nat. Prod. Rep. 1996, 13, 377–409.
37. Zubı´a, E.; Ortega, M. J.; Salva´, J. Mini-Rev. Org. Chem. 2005, 2, 389–399.
38. Millar, R. H. N.Z. Oceanogr. Inst. Mem. 1982, 85, 5–117.
15. Urban, S.; Capon, R. J.; Hooper, J. N. A. Aust. J. Chem. 1994, 47, 2279–2282.
16. Sperry, S.; Crews, P. J. Nat. Prod. 1998, 61, 859–861.
17. Walker, R. P.; Faulkner, D. J.; Van Engen, D.; Clardy, J. J. Am. Chem. Soc. 1981, 103,
6772–6773.