Structure and bioactivity of erebusinone, a pigment from the Antarctic sponge Isodictya erinacea

Article abstract of DOI:10.1016/S0040-4020(00)00760-2

We have investigated the Antarctic sponge Isodictya erinacea as part of our ongoing study of Antarctic chemical ecology. I. erinacea was found to produce a tryptophan catabolite as its yellow pigment. The pigment, erebusinone, causes significantly reduced

Full text of DOI:10.1016/S0040-4020(00)00760-2

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 9057±9062
Structure and Bioactivity of Erebusinone, a Pigment from the
Antarctic Sponge Isodictya erinacea
Byoungho Moon,a Young Chul Park,a James B. McClintockb and Bill J. Bakera,
*
aDepartment of Chemistry, Florida Institute of Technology, Melbourne, FL 32901, USA
bDepartment of Biology, University of Alabama at Birmingham, Birmingham, AL 35294, USA
Received 5 June 2000; accepted 31 July 2000
AbstractÐWe have investigated the Antarctic sponge Isodictya erinacea as part of our ongoing study of Antarctic chemical ecology.
I. erinacea was found to produce a tryptophan catabolite as its yellow pigment. The pigment, erebusinone, causes signi®cantly reduced
molting and proportionally increased mortality at ecologically relevant concentrations when fed to sympatric predatory amphipods. This
appears to be the ®rst example of molt inhibition as a mechanism of chemical defense. q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
is a conspicuous member of the community due to its bright
yellow coloration, which contrasts sharply with the more
common colorless species. I. erinacea is an attractive target
for chemical investigation because it is free of predation
despite its lack of structural protection elements such as
spicules.6 We have previously reported that I. erinacea
derives protection against a major Antarctic sponge
predator, the sea star Perknaster fuscus, from its secondary
metabolite repertoire.9 The yellow pigment of this sponge
proved intractable in our earlier studies due to its instability
at ambient temperature. Recollection of the sponge with
attention paid to handling has enabled the isolation of this
previously unreported pigment, erebusinone (1), which
appears to be a tryptophan catabolite.
The marine biota of Antarctica has recently been the subject
of a number of chemical investigations.1 Interest in this
biota stems in part to the continuing effort to identify
novel chemical diversity for exploitation in drug discovery
programs2±4 and partly from fundamental questions
concerning forces that drive the evolution of secondary
metabolism.5 Both of these objectives derive their impetus
from the unique ecology of the Antarctic benthos. The
physical environment on the Antarctic benthos has been
stable for more than 20 million years, a period suf®cient
for the establishment of a `biologically accommodated'
ecosystem whereby predation and competition are the domi-
nant forces determining species distributions and propor-
tions.6,7 The continent has been isolated from its lower
latitude neighbors even longer. Factors such as physical
stability and isolation are important criteria for genetic
divergence and this has resulted in high levels of endemism
in Antarctica.8 While too little data are currently available to
ascertain the extent to which these factors will lead to high
levels of novel and/or diverse secondary metabolic path-
ways, it is clear that unique, chemically mediated biological
interactions have evolved.5 We describe herein the isolation
and characterization of a secondary metabolite from the
Antarctic sponge Isodictya erinacea which we believe to
be responsible for an unusual chemical defense mechanism.
Results
I. erinacea was collected by scuba from several locations in
Erebus Bay on the western coast of Ross Island, Antarctica,
at 220 to 230 m under the seasonal ice sheet in the austral
summer of 1996. A 105 g (wet) specimen of I. erinacea,
frozen in liquid nitrogen immediately after collection to
prevent degradation of the pigment, was blended and
exhaustively extracted in methanol. The methanol extract
was concentrated to a largely aqueous slurry and applied to
a bed of reversed phase silica gel. The extract was eluted
sequentially with water, methanol and ethyl acetate. The
deep yellow methanol fraction was concentrated, then
applied to LH-20 in methanol, and fractions were collected
for each column volume of eluent. Erebusinone (1) eluted
from LH-20 in the fourth column volume of methanol and
was puri®ed by repeated HPLC, yielding 2 mg (2£1023%)
of a yellow solid. The HREIMS spectrum provided a
molecular formula for erebusinone of C11H14N2O3 (Dmmu
0.8). The 1H NMR spectrum displayed six signals, including
The Antarctic benthos in the vicinity of McMurdo Sound
supports a thriving community dominated by sponges.6 The
sponge I. erinacea (Topsent, 1916) (Family Esperiopsidae)
Keywords: marine natural products; Antarctic chemical ecology; sponge;
alkaloid; pigment; chemical defense.
* Corresponding author. Tel.: 11-321-674-7376; fax: 11-321-674-8951;
e-mail: bbaker@®t.edu
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00760-2

9058
B. Moon et al. / Tetrahedron 56 (2000) 9057±9062
Table 1. 13C NMR shift comparison of erebusinone 1, hydroxyanthranilate
3 and 4 (CD3OD, 360 MHz)
Carbon #
1
3
4
C-1
C-2
C-3
C-4
C-5
C-6
C-7
C-8
120.2
145.5
141.7
127.6
115.3
128.9
201.9
39.7
111.8
146.2
142.2
118.1
116.2
122.7
170.3
120.1
151.7
138.4
120.3
120.1
121.8
207.0
38.9
Figure 1. Key HMBC correlations establishing the molecular structure of
erebusinone (1).
C-9
Methyl
Amide CvO
36.5
22.7
173.2
36.0
22.6
173.6
51.9
a three spin aromatic system (H-6, d 7.68, dd; H-4, d 7.53,
dd; H-5, d 6.52, t), two mutually coupled mid-®eld methyl-
ene triplets (H2-9, d 3.45 and H2-8, d 3.12), and an acetyl
1
singlet (d 1.87), accounting for 10 protons. The H NMR
down®eld apparently due to the electron withdrawing
effects of the carbonyl substituent. Erebusinone therefore
has the same aromatic substitution pattern as 3-hydroxy-
kynurenine, suggesting a biogenesis from the tryptophan
catabolic pathway.
spectrum in DMSO-d6 displayed the remaining four protons
as broad signals, and the H2-9 signal appeared in DMSO-d6
as a quartet due to a new coupling to an amide proton. An
HMQC spectrum allowed assignment of all protonated
carbons observed in the 13C NMR spectrum. The aromatic
ring and the mutually coupled methylene signals could be
assembled around the ketone carbonyl at d 201.9 based on
HMBC correlations observed from H-6, H2-8 and H2-9 to
C-7 (Fig. 1). The HMBC spectrum also revealed correla-
tions from H2-9 to the amide carbonyl, which also displayed
a correlation to the acetyl methyl group. Two valances on
the aromatic ring (X and Y, Fig. 1) could not be accounted
for by 2D NMR spectroscopy and an amine and phenolic
function remained to be assigned (Fig. 2).
Due to structural similarities between erebusinone (1) and
3-hydroxykynurenine (2) (see Discussion, below), erebusi-
none was evaluated for bioactivity against the predatory
amphipod Orchomene plebs. A freshly collected colony of
I. erinacea was extracted and the extract chromatographed
to yield a bright yellow LH-20 fraction bearing erebusinone;
on return to our home institution, this fraction was demon-
strated to be chromatographically homogeneous (see
Experimental). This erebusinone-bearing fraction was
reduced in volume, then divided into twelve aliquots.
Over the course of the experiment, agar blocks were
prepared from individual aliquots of the erebusinone sample
using agar enriched with krill (feeding stimulant) and sand
(to weight the agar down). Amphipods were fed ad libitum
either erebusinone-enriched disks removed from the agar
blocks or control disks prepared the same way as the experi-
mental disks except with an equal volume of methanol
rather than the erebusinone sample. Feeding was vigorous
in both groups of amphipods; over the course of the ®rst four
weeks, erebusinone-enriched diet was consumed at a signi®-
cantly higher rate than controls. Molt and mortality were
monitored twice daily (Fig. 3) for 33 days. At the conclusion
of the experiment, nearly twice as many amphipods
Based on biosynthetic grounds, it was compelling to place
the amine group of erebusinone (1) in the position ortho
(C-2) to the acyl group and the phenol in the respective
meta position (C-3), yielding a derivative of 3-hydroxy-
kynurenine (2), a known marine invertebrate-derived
product.10 However, the down®eld shift of the ortho posi-
tion (C-2, d 145.5) relative to meta (C-3, d 141.7) suggested
C-2 may be substituted by the more electronegative
hydroxyl group. Comparison of chemical shifts of erebusi-
none to isomeric analogues was thus undertaken; methyl
3-hydroxyanthranilate11 (3) and 412 were prepared and
their shifts (see Table 1) supported an assignment where
the carbon bearing the less electronegative atom resonates
Figure 2.

B. Moon et al. / Tetrahedron 56 (2000) 9057±9062
9059
Figure 3. (A) Molt events, as measured by shed O. plebs carapaces fed control (diamonds) and erebusinone-enriched (circles) diets, as a percentage of total
number of amphipods. (B) Mortality in O. plebs, as measured by deceased amphipods, fed experimental (circles) and control (diamonds) diets as a percentage
of total number of amphipods. Both variables were monitored twice daily.
receiving control diet had molted (Fig. 3A) whereas nearly
twice as many amphipods receiving experimental diet had
suffered mortality (Fig. 3B). Both results show statistical
difference (ANOVA using arcsine transformed percentage
data, P , 0:05† between experimental and control data sets,
suggesting that erebusinone interferes with molting in O.
plebs leading to premature mortality.
predators, often leading to premature death of the
predator.13
In the marine realm, crustacean arthropods molt by a
mechanism similar to that in insects.10 Ecdysone is metabo-
lized to 20-hydroxyecdysone, a crustacean molt-inducing
hormone. The molt regulatory pathway involves the cata-
bolism of tryptophan, via 5-hydroxytryptophan and
3-hydroxykynurenine (2), to xanthurenic acid, which
inhibits the cytochrome P450 oxidase responsible for
hydroxylation of ecdysone at C-20. Xanthurenic acid, there-
fore, is a naturally occurring molt inhibitor in crustaceans.
Discussion
Primitive plants evolved a defense against arthropod preda-
tion which takes advantage of the requirement of insects to
molt as part of their development.13 Molting, which is regu-
lated by a complex and still poorly understood series of
events, can be induced physiologically by ecdysone-derived
hormones. As a chemical defense mechanism, a number of
plant species produce derivatives of ecdysone, termed
phytoecdysones, some of which are more potent molting
inducers in insect predators than the natural hormones.
Phytoecdysones produce major abnormalities in insect
The structural similarity between erebusinone (1) and that of
3-hydroxykynurenine (2) led us to study the potential role of
erebusinone in mediating molting in an Antarctic predatory
crustacean. The Antarctic amphipod O. plebs is a general
benthic scavenger on the Antarctic benthos and can be a
voracious sponge predator; we have found the amphipod
boring into or otherwise associated with many McMurdo
Sound sponges. O. plebs feeding on a diet enriched at the

9060
B. Moon et al. / Tetrahedron 56 (2000) 9057±9062
concentration found naturally in I. erinacea suffered signi®-
cant molt inhibition (Fig. 3A), relative to controls, leading
to a proportional increase in mortality (Fig. 3B). Toxic
effects of erebusinone can be discounted as causing the
reduction in molting since such effects would be expected
to randomly kill amphipods, not just amphipods preparing
to molt. Lack of predation by O. plebs on I. erinacea may
well result from the detrimental consequences of doing so,
in which case this would appear to be the ®rst example in the
marine realm of a chemical defense mechanism involving
mediation of molting and an unusual example of molt
inhibition, rather than induction, as a means of chemical
defense.
Isolation of erebusinone
Freshly collected sponge (105.5 g wet) was extracted by
blending with methanol (3£400 mL). The extract was
®ltered and the solvent removed under reduced pressure.
The residual dark brown oil was subjected to reverse
phase vacuum chromatography (18 cm£4 cm) with metha-
nol, water, methanol, then ethyl acetate. The combined
methanol fractions (22.5 g after concentration) were chro-
matographed on Sephadexw LH-20 with methanol as eluent.
Four fractions were collected, two column volumes of
eluent per fraction. The fourth LH-20 fraction (2.7 g after
concentration) was subjected to puri®cation by reversed
phase HPLC with 0.01 M ammonium formate/methanol
(80:20), tR 13 min. Further puri®cation by reversed phase
HPLC with 0.05% TFA/methanol (80:20) provided
erebusinone (2 mg, 0.002% of wet weight, tR 25 min).
Because erebusinone (1) is a structural analog of a precursor
to the natural inhibitor rather than an analog to the inhibitor
itself, the mechanism of action of molt inhibition is of
interest. Erebusinone may function by being metabolized
into a xanthurenic acid analog which acts to swamp or
irreversibly bind the cytochrome P450 responsible for
molt hormone biosynthesis. Alternately, erebusinone may
act elsewhere, such as direct inhibition of the cytochrome
P450 oxidase or one of the many other enzymes involved in
molt regulation.10 Further evaluation of the biological
activities of erebusinone is underway.
Erebusinone (1). yellow solid; IR (KBr) nmax 3419, 1641,
1221, 1061, 998 cm21; UV (MeOH) lmax (e) 372 (176), 312
(372), 238 (1,238) nm; HREIMS found 222.0996 (M1
1
Dmmu 0.8 for C11H14N2O3); H NMR (CD3OD): d 7.68
(dd, Jˆ6.8, 1.4 Hz, H-6), 7.53 (dd, Jˆ6.5, 1.4 Hz, H-4),
6.52 (bt, Jˆ7.9 Hz, H-5), 3.45 (t, Jˆ6.5 Hz, H2-9), 3.12 (t,
Jˆ6.5 Hz, H2-8), 1.87 (s, NCOCH3); 13C NMR (CD3OD):
See Table 1.
Experimental
General procedures
Preparation of methyl 3hydroxyanthranilate (3).
Powdered K2CO3 (200 mg, 1.44 mmol, 1.1 equiv.) was
added to DMSO (3 mL) then, after stirring for 5 min,
3-hydroxyanthranilic acid (200 mg, 1.3 mmol, 1 equiv.)
was added, followed immediately by methyl iodide
(222 mg, 1.57 mmol, 1.2 equiv.). Stirring was continued
for 30 min after which the mixture was poured into water
(20 mL) and then extracted with methylene chloride
(3£20 mL). The combined organic extracts were washed
with water and brine, dried with anhydrous MgSO4 and
concentrated. The crude product was puri®ed by column
chromatography (5% EtOAc in n-hexane) to give 99 mg
(45%) of methyl 3-hydroxyanthranilate as a yellow solid:
IR (KBr) nmax 3414, 3328, 1709, 1301, 1281 cm21; EIMS
m/z 167 (M1); 1H NMR (CD3OD): d 7.28 (d, 1H), 6.76 (d,
1H), 6.41 (t, 1H), 3.78 (s, 3H); 13C NMR: See Table 1.
NMR spectral analyses were performed on an 8.46-T NMR
instrument operating at 360 MHz for H and 90 MHz for
1
1
13C. One bond heteronuclear H±13C connectivities were
determined by HMQC; two and three bond 1H±13C connec-
tivities were determined by HMBC optimized for 7 Hz
couplings; chemical shifts are reported in ppm with the
chemical shift of residual solvent nuclides used as internal
standards. The IR spectra were recorded on a Nicolet
Magna-IR 550 spectrometer. The UV spectra were recorded
on a Hewlett±Packard 8452A diode-array spectrometer.
Electron impact mass spectra were recorded on a VG
70SE at the University of Florida. A Waters 510 HPLC
pump system with a Waters 490E UV detector and YMC
ODS-Aq (10£250 mm, 5 mm) column were used for
HPLC puri®cation. All chromatographic solvents were
distilled from glass. Unless otherwise noted, chemicals
were commercially available and used without further
puri®cation.
Preparation of 4
A. Lithium diisopropylamide (LDA) (10 mL, 2.0 M in
hexane) was added dropwise to a stirred solution of
2-hydroxy-3-nitroacetophenone (1.63 g, 9 mmol) in
anhydrous THF (100 mL) under N2 at 2788C (dry-ice/
acetone). The reaction mixture was stirred at 2788C for
30 min. Formaldehyde vapor, obtained by heating para-
formaldehyde (7.00 g) at 1808C, was introduced into the
reaction vessel with vigorous stirring under nitrogen at
2788C. The reaction mixture was then poured into saturated
aqueous NH4Cl (50 mL) and acidi®ed to pH 3 with concen-
trated HCl. After extraction with ethyl ether (3£100 mL),
and EtOAc (100 mL), the organic layers were combined and
dried over anhydrous MgSO4 prior to concentration in
vacuo. Puri®cation of the residue by column chromato-
graphy on silica gel (EtOAc/hexane, 1:1) yielded 855 mg
(45%) of 20-hydroxy-30-nitro-3-hydroxypropiophenone as a
pale yellow solid: mp 708C; IR (KBr) nmax 3427, 3098,
Animal material
I. erinacea was collected at a depth of 20±30 m from
several locations on the coast of Ross Island in McMurdo
Sound, Antarctica in the Austral summer of 1996. A
voucher specimen (SC 1620) is on hand at Florida Tech.
O. plebs was collected in McMurdo Sound by placing
tethered ®sh tissue on the benthos overnight; upon retrieval,
the amphipods were removed from the ®sh tissue and trans-
ferred to the Crary Laboratory aquarium in bottles (2 L) of
ice cold sea water. The animals were maintained in these
bottles, kept at ambient temperature (218C) on a diet of 4%
krill in agar.

B. Moon et al. / Tetrahedron 56 (2000) 9057±9062
9061
1670, 1584, 1354,1288 cm21; UV (CH2Cl2) lmax (e) 344
(6701); HREIMS found 211.0476 (M1 Dmmu 0.5 for
C9H9NO5); H NMR (CDCl3): d 12.72 (s, 1H, ±OH), 8.35
(dd, Jˆ6.6, 1.6 Hz, 1H), 8.11 (dd, Jˆ6.2, 1.6 Hz, 1H), 7.08
(t, Jˆ8.0 Hz, 1H), 4.08 (t, Jˆ5.3 Hz, 2H), 3.38 (t, Jˆ5.8 Hz,
2H); 13C NMR (CDCl3): d 204.18 (s), 156.01 (s), 137.56 (s),
136.54 (d), 131.58 (d), 125.55 (s), 118.94 (d), 57.75 (q),
42.99 (q).
To the system was added thiolacetic acid (97 mL,
1.35 mmol). The reaction mixture was stirred at room
temperature for 4 h and concentrated in vacuo. The resulting
oil was subjected to ¯ash column chromatography on silica
gel (EtOAc/hexane, 1:1) to obtain 109 mg (90%) of the 20-
benzyloxy-30-nitro-3-acetamidopropiophenone as a pale
yellow solid: mp 758C; IR (KBr) nmax: 3295, 3078, 2933,
1709, 1657, 1532; UV (CH2Cl2) lmax (e): 302 (1662);
HREIMS found 342.1208 (M1 Dmmu 0.8 for
C18H18N2O5); 1H NMR (CDCl3): d 7.95 (dd, Jˆ6.5,
1.6 Hz, 1H), 7.73 (dd, Jˆ6.1, 1.7 Hz, 1H), 7.36 (s, 5H),
7.31 (t, Jˆ7.9 Hz, 1H), 6.08 (bs, 1Hex), 3.47 (q, Jˆ5.9 Hz,
2H), 3.09 (t, Jˆ5.8 Hz, 2H), 1.88 (s, 3H); 13C NMR
(CDCl3): d 201.08 (s), 170.19 (s), 150.37 (s), 144.90 (s),
136.64 (s), 135.06 (s), 133.60(d), 129.17 (d), 128.99 (d, 2
C), 128.93 (d, 2 C), 128.48 (d), 79.51 (t), 43.01 (t), 34.50 (t),
23.29 (q).
1
B. A mixture of 20-hydroxy-30-nitro-3-hydroxypropiophe-
none (438 mg, 2.08 mmol), anhydrous K2CO3 (286 mg,
2.08 mmol) and benzylbromide (246 mL, 2.08 mmol) in
dimethoxyethane (10 mL) and DMF (3 mL) was stirred
for 24 h at room temperature. The reaction mixture was
®ltered and then concentrated in vacuo. The residue was
diluted with water (50 mL), then extracted with CH2Cl2
(3£100 mL). The organic layer was dried over anhydrous
MgSO4, and concentrated under reduced pressure. Column
chromatography on silica gel (EtOAc/hexane, 4:6) yielded
380 mg (61%) of 20-benzyloxy-30-nitro-3-hydroxypropio-
E. A mixture of 20-benzyloxy-30-nitro-3-acetamidopropio-
phenone (68 mg, 0.19 mmol), and 10% Pd/C (30 mg) in
ethanol (15 mL) was shaken with hydrogen administered
via balloon for 1 h at room temperature (21 mL of hydrogen
was consumed). The reaction mixture was ®ltered to remove
the catalyst, and the ®ltrate was concentrated. Puri®cation of
the residue by column chromatography on silica gel
(methanol/CH2Cl2, 5:95) yielded 34 mg (81%) of 4 as a
yellow solid: yellow solid: IR (KBr) nmax: 3427, 3322,
2940, 2479, 1663, 1637, 1545, 1453, 1249; UV (MeOH)
lmax (e) 372 (219), 278 (679), 238 (1323); HREIMS
found 222.1052 (M1 Dmmu 4.8 for C11H14N2O3),
phenone as a pale brown oil: mp 578C; IR (KBr) nmax:
3600±3300 (b), 3085, 2953, 1703, 1611, 1538, 1354,
1242; UV (CH2Cl2) lmax (e): 302 (3,076); HRFABMS
1
found 302.1036 ((M1H)1 Dmmu 0.8 for C16H16NO5; H
NMR (CDCl3): d 7.94 (dd, Jˆ6.6, 1.6 Hz, 1H), 7.74 (dd,
Jˆ6.1, 1.6 Hz, 1H), 7.37 (m, 5H), 7.29 (t, Jˆ8.0 Hz, 1H),
3.83 (t, Jˆ5.4, 3H), 3.10 (t, Jˆ5.4 Hz, 3H); 13C NMR
(CDCl3): d 201.85 (s), 150.25 (s), 144.78 (s), 136.74 (s),
134.98 (s), 133.71 (d), 129.26 (d), 129.14 (d, 2C), 128.99 (d,
2 C), 128.49 (d), 125.08 (d), 79.70 (t), 58.14 (t), 45.54 (t).
1
163.0637 (M12CH3CONH2, Dmmu 0.4 for C9H9NO2); H
C. To a stirred solution of 20-benzyloxy-30-nitro-3-hydro-
xypropiophenone (350 mg, 1.16 mmol) in 5 mL of
anhydrous CH2Cl2 at 2788C under N2 was added methane-
sulfonyl chloride (135 mL, 1.74 mmol), followed by
triethylamine (243 mL, 1.74 mmol). After the mixture was
stirred for 1 h at 2788C, the solution was poured into water
(50 mL), separated, and the aqueous phase was extracted
with CH2Cl2 (2£50 mL). The combined organic layers
were dried over anhydrous Na2SO4, ®ltered, and concen-
trated to yield the mesylate (427 mg, 97%) that was used
without further puri®cation. The mesylate (427 mg,
1.13 mmol) in anhydrous DMF (5 mL) was treated with
NaN3 (73 mg, 1.13 mmol) with stirring under N2 for 3 h at
room temperature. The reaction mixture was then diluted
with ethyl ether (50 mL). A white precipitate was formed
which was removed by ®ltration. The ®ltrate was evapo-
rated under reduced pressure to leave an oil. Puri®cation
of the oil by column chromatography on silica gel (hexane/
EtOAc/CH2Cl2, 7:1.5:1.5) yielded 147 mg (40%) of 20-
benzyloxy-30-nitro-3-azidopropiophenone as pale yellow
oil: IR (NaCl) nmax: 3078, 2927, 2117 (±N3), 1709, 1611,
1249; UV (CH2Cl2) lmax (e): 302 (4909); 1H NMR (CDCl3):
d 7.99 (dd, Jˆ6.4, 1.7 Hz, 1H), 7.77 (dd, Jˆ6.1, 1.6 Hz,
1H), 7.39 (s, 5H), 7.32 (t, Jˆ7.9 Hz, 1H), 5.05 (s, 2H),
3.57 (t, Jˆ6.2 Hz, 2H), 3.13 (t, Jˆ6.3 Hz, 2H); 13C NMR
(CDCl3): d 199.07 (s), 150.56 (s), 144.80 (s), 136.53 (s),
135.15 (s), 133.98 (d), 129.27 (d), 129.06 (d, 2 C), 128.98
(d, 2 C), 128.67 (d), 79.70 (t), 46.09 (t), 42.26 (t).
NMR (CD3OD): d 7.19 (dd, Jˆ6.8, 1.3 Hz, 1H), 6.93 (dd,
Jˆ6.3, 1.4 Hz, 1H), 6.70 (t, Jˆ7.9 Hz, 1H), 3.51 (t,
Jˆ6.4 Hz, 2H), 3.21 (t, Jˆ6.4 Hz, 2H), 1.87 (s, 3H); 13C
NMR (CD3OD): See Table 1.
Bioassay. Agar blocks for amphipod bioassay were
prepared from freshly collected I. erinacea (160 g dry/
650 mL), which was extracted and chromatographed by
reversed phase and LH-20 methods, as described above.
The bright yellow LH-20 band bearing erebusinone was
reduced in volume then divided into twelve 2.0 mL aliquots.
Agar blocks were prepared as needed from individual
aliquots of the erebusinone sample by adding the 2.0 mL
solution (in methanol) to 52 mL (ˆ1/12 the volume of the
original sponge such that the concentration of erebusinone
in agar is equivalent to that of the sponge on a volume basis)
of 3% agar enriched with 4% krill (feeding stimulant) and
1.5% sand (to weight the agar down). Control blocks were
similarly prepared using 2.0 mL of pure methanol. HPLC
analysis of one aliquot of the erebusinone sample was
carried out on return to our home institution; erebusinone
(1), identi®ed by 1H NMR spectroscopy, was the sole
absorption in the HPLC chromatogram at 254 nm.
O. plebs were collected using ®sh-baited traps in McMurdo
Sound, Antarctica. Amphipods selected for experiments
were of similar size (mean lengthˆ9 mm). Six 250 mL
Erlenmeyer ¯asks containing 30 amphipods each were
maintained at ambient temperature (218C) in the Crary
Laboratory (McMurdo Station, Antarctica) with daily
water changes. Experimental animals in three of the ¯asks
were fed an ad libitum diet of disks removed from agar
D. Under a nitrogen atmosphere in a 1 mL round-bottom
¯ask equipped with a rubber septum was placed 20-benzyl-
oxy-30-nitro-3-azidopropiophenone (110 mg, 0.34 mmol).

9062
B. Moon et al. / Tetrahedron 56 (2000) 9057±9062
blocks of 4% ®nely ground dry krill in 3% agar containing
1.5% sand (to weight the food to the bottom of the ¯asks)
and a concentration of the erebusinone ecologically equiva-
lent to that found in the sponge. Control animals in the other
three ¯asks received the same diet ad libitum but containing
an equivalent amount of solvent (methanol) used to
solubilize erebusinone in the preparation of the experimen-
tal diet. All agar disks were replaced daily with fresh disks.
Both experimental and control amphipod ¯asks were
observed twice daily for molt (discarded carapace) and
mortality for a 33-day-period. The percent incidence of
molting and mortality over the entire experimental period
was compared between experimental and control treatments
(following arcsine transformation of percentage data)
employing an analysis of variance test.
References
1. Amsler, C. D.; Iken, K. B.; McClintock, J. B.; Baker, B. J.
Marine Chemical Ecology; McClintock, J. B., Baker, B. J., Eds.;
CRC Press: Boca Raton, FL, in press.
2. (a) Blunt J. W.; Munro, M. H. G.; Battershill, C. N.; Copp, B. R.;
McCombs, J. D.; Perry, N. B.; Prinsep, M.; Thompson, A. M. New
Journal of Chemistry 1990, 40, 761±775. (b) Perry, N. B.; Ettouati,
L.; Litaudon, M.; Blunt, J. W.; Munro, M. H. G. Tetrahedron 1994,
50, 3987±3992.
3. Jayatilake, G. S.; Baker, B. J.; McClintock, J. B. Tetrahedron
Letters 1997, 38, 7507±7510.
4. Fontana, A.; Ciavatta, M. L.; Amodeo, P.; Cimino, G.
Tetrahedron 1999, 55, 1143±1152.
5. (a) McClintock, J. B.; Baker, B. J. American Scientist 1998, 86,
254±263. (b) McClintock, J. B.; Baker, B. J. American Zoologist
1997, 37, 329±342.
Acknowledgements
6. Dayton, P. K.; Robilliard, G. A.; Paine, R. T.; Dayton, L. B.
Ecological Monographs 1974, 44, 105±128.
Field and laboratory assistance from Jim Mastro, Chuck
Amsler, Thomas Klinger, Thomas Barlow, Jill Baker and
Gamini Jayatilake are greatly appreciated. This research
was supported by the National Science Foundation (OPP-
9530735 to JBM and OPP-9526610 and -9901076 to BJB).
We gratefully acknowledge the services of Dr David
Powell, Director of Spectroscopic Services at the University
of Florida for the mass spectral data. Purchase of the NMR
spectrometer was assisted by a grant from NSF. The
Antarctic Support Associates Inc., the Antarctic Support
Services of the National Science Foundation, and the US
Naval Antarctic Support Force provided invaluable logisti-
cal support for this project. This manuscript is submitted in
honor of Paul J. Scheuer, whose inspiration and guidance
has contributed in many ways to make this work possible.
7. Dayton, P. K.; Mordida, B. J.; Bacon, F. American Zoologist
1994, 34, 90±99.
8. Arntz, W. E.; Brey, T.; Gallardo, V. A. Oceanography and
Marine Biology an Annual Review 1994, 32, 241.
9. Moon, B. H.; Baker, B. J.; McClintock, J. B. Journal of Natural
Products 1998, 61, 116±118.
10. Nakanishi, K. In Biomedical Importance of Marine Orga-
nisms; Fauntin, D. G., Ed.; California Academy of Science Press:
San Francisco, 1988; pp 59±67.
11. Kelly, T. R.; Martinez, C.; Mears, R. J. Heterocycles 1997, 45,
87±93.
12. Full details of the synthesis of 1 and 4 will be reported
elsewhere.
13. Williams, C. M. In Chemical Ecology; Sondheimer, E.,
Simeone, J. B., Eds.; Academic Press: New York, 1972; pp 103±132.