MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 354: 169?179, 2008
doi: 10.3354/meps07200
Published February 7
INTRODUCTION
The production of defensive metabolites is often
assumed to divert energy from basic metabolic re-
quirements including growth, reproduction, mainte-
nance, and resource acquisition. A variety of models
have been proposed to account for patterns of resource
allocation to defenses and these other fundamental
processes (Cronin 2001). The environmental stress the-
ory (EST) was proposed to explain increased palat-
ability of physiologically-stressed terrestrial plants; it
suggests that decreases in defensive compounds or
increases in nutritional value of the plants could
account for observed increases in herbivory (White
1984, Rhoades 1985). Changes in these plant biochem-
ical constituents might occur if a stressed individual
exhibits either (1) reduced efficiency in acquiring req-
uisite resources, or (2) increased maintenance costs
associated with physiological repair relative to an
unstressed individual. While the role of abiotic factors
in structuring marine communities has been well doc-
umented, the impacts of abiotic stressors on marine
algal and invertebrate defenses have rarely been
studied (Renaud et al. 1990, Cronin & Hay 1996a,
? Inter-Research 2008 ? www.int-res.com*Email: slattery@olemiss.edu
Indirect effects of bleaching on predator
deterrence in the tropical Pacific soft coral
Sinularia maxima
Marc Slattery1,*, Valerie J. Paul2
1Department of Pharmacognosy and the National Center for Natural Products Research, School of Pharmacy, 
The University of Mississippi, University, Mississippi 38677-1848, USA
2Smithsonian Marine Station at Fort Pierce, 701 Seaway Drive, Fort Pierce, Florida 34949, USA
ABSTRACT: The environmental stress theory (EST) suggests that stress-induced biochemical
changes will make an organism more susceptible to predation relative to unstressed individuals.
Bleaching represents a stress response in marine invertebrate host?zooxanthellae symbiont associa-
tions, including those species that rely on chemical defenses to reduce predation pressure. We exam-
ined the EST in the context of a natural soft coral bleaching event, and then by use of a 3 mo trans-
plant/shading field experiment that also resulted in bleaching. Feeding experiments using an
omnivorous pufferfish indicated that extracts of bleached soft corals were more palatable to preda-
tors than were those from unbleached individuals. Two biochemical constituents were significantly
reduced in naturally- and experimentally-bleached soft corals: lipid and defensive metabolite
(pukalide) concentration. While a sunscreen (palythine) was significantly lower immediately after a
natural bleaching event, it recovered and even exceeded pre-bleached levels in experimentally-
bleached soft corals. Feeding assays with pukalide and lipid at concentrations representative of
unbleached and bleached soft corals indicate that pufferfish are not deterred by bleached levels of
the defensive metabolite, but that pufferfish are less attracted to food with bleached levels of lipid.
Nonetheless, field observations of up to 4 times higher predation on bleached soft corals suggest that
defensive metabolite concentration may have a greater influence on predation than food quality. This
study indicates that a stress response (bleaching) has an indirect effect on predator deterrence in the
soft coral Sinularia maxima due to changes in biochemical constituents that affect food quality. 
KEY WORDS:  Environmental stress theory ? Chemical defenses ? Bleaching ? Sunscreens ?
Transplants
Resale or republication not permitted without written consent of the publisher
Mar Ecol Prog Ser 354: 169?179, 2008
Michalek-Wagner & Bowden 2000). However, the
importance of interactions between prey nutritional
quality and chemical defenses in predation has re-
ceived more attention in marine communities (Duffy &
Paul 1992, Hay et al. 1994, Cruz-Rivera & Hay 2003);
although this is untested, some of the prey caloric
value is likely due to lipid (Stimson 1987). 
Coral-bleaching is one of the most apparent environ-
mental stress responses in the marine environment
(Glynn 1993). This phenomenon, which can occur in
any invertebrate?zooxanthellae association, has been
reported with alarming and increasing frequency
worldwide (Brown 1997). The loss of zooxanthellae
symbionts by the invertebrate host tissue (i.e. ?bleach-
ing?) may prevent host cellular damage through oxida-
tive stress induced by elevated sea temperatures
(Lesser 1997), or radiation (UV-B, UV-A, and PAR: ? =
280?319 nm, 320?399 nm, 400?700 nm, respectively,
Baruch et al. 2005), or a combination of these factors
(Douglas 2003). Loss of zooxanthellae can have a sig-
nificant impact on the metabolic health and budget of
the host organism since these symbionts are important
in nitrogen and phosphorous cycling within the associ-
ation, and they contribute up to 95% of their photosyn-
thetic products to the host (Muller-Parker & D?Elia
1997). As photosynthesis declines with the reduction in
chl a levels during bleaching (Warner et al. 2002),
many corals are forced to utilize stored carbon reserves
in protein, carbohydrates, and lipid (Anthony et al.
2002, Grottoli et al. 2004). Protein and carbohydrates
typically represent a short-term solution to a bleached
coral?s metabolic requirements (Anthony et al. 2002),
while the significant energetic resources in lipid (Stim-
son 1987) can support coral metabolism for almost 3 mo
(Montipora verrucosa, Spencer Davies 1991). Coral
energetic budget deficits imply some degree of trade-
off in resource allocation to metabolic activities and
other functions including defense (Cronin 2001). 
Many soft corals maintain symbioses with zooxan-
thellae, and during bleaching events these organisms
are often severely stressed (Michalek-Wagner & Willis
2001). We observed bleaching in about half the popu-
lation of the soft coral Sinularia maxima on a shallow
reef off Guam during November 1994. This event
appeared to be much more influenced by solar radia-
tion than temperature (i.e. bleaching occurred at a
depth ?1 m following a 2 wk period of sunny days with
exceptionally clear and still water, even though water
temperature never exceeded 28?C; this study).
Bleached soft corals exhibited an increase in the num-
ber of predation events (as measured by bite scars,
Slattery et al. 2001) compared to unbleached con-
specifics. This was surprising since S. maxima contains
various cembranoid diterpene feeding deterrent com-
pounds (Wylie & Paul 1989, Slattery et al. 1998),
although these can vary in time and space (Slattery et
al. 2001). S. maxima also sequesters a mycosporine-
like amino acid (MAA; this study) which has the poten-
tial to absorb in the UV spectral range thus acting as a
natural sunscreen (see references in Shick 2004, for
review). The purpose of this study was to (1) describe
the biochemical differences (i.e. quantitative and qual-
itative variation in primary and secondary metabolites
between bleached and unbleached soft corals,
(2) characterize the effects of solar radiation on the soft
coral?zooxanthellae association, and (3) determine
whether bleached soft coral colonies are more palat-
able to predators than are unbleached conspecifics. 
MATERIALS AND METHODS
Study site. The soft coral Sinularia maxima is a ubi-
quitous member of shallow-water reefs of the Indo-
Pacific, and it is one of the most abundant soft corals
on the leeward reefs of Guam (13? 25? N; 144? 55? E)
(Slattery et al. 2001). At Cocos Lagoon, the southwest-
ern-most reef of Guam, S. maxima co-occurs with S.
paulae and S. polydactyla on back reef flats at depths
of ?2 m (although these latter 2 species exhibited no
signs of bleaching during this study). In addition, a
population of S. maxima is located on the northern
slope of Marc?s Reef, adjacent to the boat channel at a
depth of 15 m. 
Naturally-bleached soft coral collections. On 13
November 1994 we collected replicate (n = 15 each)
colonies of bleached and unbleached Sinularia max-
ima samples from a depth of 1 m, and unbleached
colonies from a depth of 15 m. Unbleached S. maxima
specimens have a characteristic pinkish-tan pigmen-
tation; the bleached colonies exhibited a total lack of
pigment across the colony surface. Bite marks were
recorded on bleached and unbleached soft coral
colonies using procedures described in Slattery et al.
(2001). An additional 45 bleached and 45 unbleached
colonies from 1 m, and 45 unbleached colonies from
15 m were tagged for subsequent sampling in each of
following 3 mo (n = 15 per treatment each mo) in order
to assess recovery from bleaching. These soft corals
were returned to the University of Guam Marine Lab-
oratory (UOGML) for subsequent processing of zoo-
xanthellae and biochemical constituents, and for feed-
ing assays (see below). Once each month of the study
(on collection days: 11 December 1994, 8 January
1995, and 5 February 1995; see below), the light qual-
ity and quantity were recorded at this site hourly
between 09:00 and 16:00 h using an IL1400A radiome-
ter/photometer equipped with UVA/B and PAR detec-
tors (International Light). UVA and PAR measurements
were collected with UV-stabilized silicon photodiode
170
Slattery & Paul: Effects of bleaching on soft coral defenses
detectors that included flat-response filtration in
the PAR (400?750 nm) and UVA (315?390 nm)
wavelengths at noise equivalent irradiances of
5.00e?10 W cm?2 and 2.67e?10 W cm?2, respectively.
UVB measurements were collected with a solar blind
vacuum photodiode detector that included filtration in
the UVB (185?320 nm; ?max = 240 nm) wavelength at a
noise equivalent irradiance of 6.67e?6 W cm?2. These
survey days were generally sunny with very limited
cloud cover and no rain. While these days represented
the best conditions noted throughout the course of the
study, winter tends to be the dry season on Guam, so
irradiance is typically not affected by atmospheric con-
ditions. Thus our data are considered representative of
the light environment supporting these soft corals.
Quantification of zooxanthellae and soft coral
chemical constituents. The soft coral samples were
subdivided into 4 pieces for the following analyses.
Zooxanthella densities were determined by homoge-
nizing soft coral tissue samples of 1 g wet wt in sterile
seawater (SSW) and counting the number of symbiont
cells using a hemacytometer; care was taken to assure
that tissues were collected from the same region of
each soft coral colony, representing similar surface
areas (ca. 1 cm2). Triplicate subsamples from each soft
coral were counted and the mean used as a represen-
tative of the total density of soft coral zooxanthellae in
host tissue. The protein and lipid concentrations of the
soft coral?zooxanthellae association were determined
from a second tissue subsample of approximately 2 g
wet wt (mass is the typical unit of measurement for
alcyonacean soft corals due to extensive variability in
their hydrostatic skeletons, and consequently volume
[Slattery & McClintock 1995]; this study duplicated the
aforementioned procedures). This subsample was
lyophilized, ground and extracted in 1 N NaOH for
protein analyses (using a colorimetric assay), or 2:1
chloroform/methanol for lipid analyses (using a gravi-
metric technique). A third subsample of approximately
5 g wet wt of soft coral tissue was extracted in a 1:1
mixture of dichloromethane/methanol to provide a
crude extract containing the soft coral defensive
metabolites. These extracts were subsequently used in
feeding assays (see below) and for quantification of the
defensive metabolites pukalide and 11?-acetoxy-
pukalide using analytical HPLC techniques described
in Slattery et al. (1998, 2001). A fourth subsample of
0.3 g dry wt of soft coral tissue was extracted in a 4:1
mixture of methanol and water to recover MAAs. The
polar extract was examined spectrophotometrically;
MAAs were separated and quantified in comparison
with known standards using analytical HPLC, as
described in Karentz et al. (1991). 
Artifical bleaching experiment. To determine the
effect of solar radiation on soft coral symbiont and
biochemical constituents, we conducted a combina-
tion transplant/shading study. We reasoned that the
differences in susceptibility to bleaching within the
shallow-water Sinularia maxima population were due
to differing clades of light-acclimatized zooxanthellae
(in accordance with Rowan et al. 1997 and Michalek-
Wagner et al. 2001). We postulated further that the
population of S. maxima from 15 m would exhibit low
tolerance to the solar radiation conditions of the
shallow reef, resulting in bleaching. Thus, we trans-
planted replicate soft corals from the 15 m site to the
1 m site, and into each of 4 treatment conditions on
11 December 1994. One treatment was exposed to full
sunlight and the remaining 3 treatments were under
different Plexiglas? shades. The 3 shaded conditions
included (1) UV-transparent Plexiglas? II filter (UVT:
allows passage of all UV and visible light >275 nm)
(2) UV-opaque Plexiglas? UF3 filter (UVO: allows
passage of all visible light >450 nm), and (3) Plexi-
glas? II filter painted black (PARO/UVO: blocks pas-
sage of all UV and visible light). The Plexiglas?
shades were 30.5 ? 30.5 ? 0.3 cm; these were cable
tied to 4 pieces of 0.9 cm diameter reinforcing iron bar
hammered into the reef flat (n = 10 shades per treat-
ment condition). Three soft corals (approximate size:
5 cm3) were positioned directly under the middle of
each replicate shade, and since these shades were
typically no more than 15 cm above the substratum,
minimal direct incident sunlight reached the soft coral
colonies throughout the day. This experiment was
conducted for 3 mo, and at 4 wk intervals (i.e. 8 Janu-
ary 1995, 5 February 1995, and 4 March 1995) one of
the 3 transplants under each of the shade treatments,
as well as one from the full sunlight condition, were
harvested and transported to UOGML for processing
as above (n = 10 [replicates] treatment?1 mo?1). These
shades were cleaned of fouling organisms at least
once per week through the course of the experiment.
In addition to these shade treatments, we trans-
planted soft corals from the shallow reef to the deep
reef, and we back-transplanted soft corals from both
sites into their site of origin as a procedural control for
handling effects (Slattery et al. 2001). These shallow
to deep transplants (n = 5 mo?1), and back-transplants
(n = 5 mo?1 from each depth) were collected during
each of our 3 subsequent sampling dates. In this
transplant study, replicates refer to unique individuals
collected from the larger Sinularia maxima population
throughout Cocos Lagoon. However given their clon-
ality (Slattery et al. 2001) it is possible that some of the
soft corals used in this project might represent ramets
of a specific genotype. The effect on our analyses
should be to reduce variation, thus the data presented
here represent a conservative estimate of stress-
related responses in Sinularia maxima. 
171
Mar Ecol Prog Ser 354: 169?179, 2008
Feeding deterrence assays. Feeding assays were
conducted in the laboratory at UOGML using the
crude extracts and pure metabolites from the bleached
and unbleached colonies of Sinularia maxima collected
immediately following a natural bleaching event and
immediately following an experimentally-induced
bleaching event. The omnivorous pufferfish Canthi-
gaster solandri was used as a model predator (e.g. Slat-
tery et al. 1998) since it feeds occasionally on soft coral
polyps in the field. This laboratory assay design was
described thoroughly by Slattery et al. (1998). Briefly,
pufferfish were maintained in flow-through containers
and offered pre-weighed paired 1 cm3 food cubes con-
sisting of a control and a treatment diet. The control
food cubes were formulated with 2.5 g agar in 50 ml of
tap water, mixed with 12 g wet wt of Prime Reef?
Aquarium food that contained protein levels similar to
those of the soft coral (?14%) but lipid levels about half
natural levels (?30%). To increase lipid levels within
the food cubes, we extracted unbleached hard corals
(Porites rus, apparently lacking chemical defenses: M.
Slattery pers. obs.) collected at the study site using the
extraction schemes of Slattery & McClintock (1995),
and added the gravimetric equivalent lipid to our arti-
ficial diet. Lipid level was adjusted lower by reducing
the concentration of Prime Reef? in our food cubes by
about one half, and then amending the diet with squid
paste. Since squid mantle has much higher levels of
protein and lower levels of lipid than the aquarium for-
mulation, mathematical adjustment allowed us to
effectively maintain protein concentrations while de-
creasing lipid concentrations in our control diet. Treat-
ment cubes were formulated in a similar manner but
included an ecologically-relevant concentration of the
extract or metabolites (see below: ?Results?). A second
set of control and treatment cubes was produced iden-
tically, but maintained in flow-through containers
without pufferfish to control for changes in the food
cube mass due to hydration (see Cronin & Hay 1996a,
Slattery et al. 1998). Feeding preference or deterrence
was assessed by calculating differences in the adjusted
mass of the food cubes consumed. 
The interactive effects of defensive metabolite
(pukalide) and lipid variability on feeding preference
or deterrence were assessed in a similar set of bio-
assays. Four different combinations of these 2 bio-
chemical constituents were formulated as combination
diets in the aforementioned food cubes, representing
the extremes of these constituents recorded through-
out the study: (1) high concentration of pukalide
(= 0.5%: feeding deterrent levels reported by Slattery
et al. 2001) and high concentration of lipid (= 30%: see
Fig. 2B); (2) low concentration of pukalide (= 0.125%:
see Fig. 2D) and high concentration of lipid; (3) high
concentration of pukalide and low concentration of
lipid (= 10%: see Fig. 2B); and (4) low concentration of
pukalide and low concentration of lipid. Since we were
interested in potential additivity or synergisms with
respect to feeding preference or deterrence, we also
conducted single constituent pair-wise assays (i.e. con-
trol diet amended with pukalide, or control diet
amended with lipid, only). These 4 ? 3 blocked pair-
wise comparisons (i.e. 2 single constituent and one
combination constituent assays) were all conducted
for identical periods of time (1.75 h) so that results of
the individual assays could be compared across the
block in a higher-order statistical manner (Hay et al.
1994).
Statistical analyses. The changes in soft coral sym-
biont densities and biochemical constituent concentra-
tions following a natural bleaching event and an
experimentally-induced (transplant/shading) bleach-
ing event were analyzed using factorial ANOVA with
date and soft coral condition as fixed factors. The data,
presented as percentages, were arcsin transformed
prior to analyses, and tested for assumptions of normal-
ity and equal variance. Post hoc tests for multiple com-
parisons included Sheffe?s S and Dunnett?s tests. Field
bite mark data on bleached and unbleached soft corals
were compared with a 2-sample t-test. Feeding assays
using naturally- and experimentally-bleached soft
coral extracts were tested using a paired t-test be-
tween control and treatment cubes based on mass
differences adjusted for hydration. The blocked single
constituent/combination constituent assays were ana-
lyzed first by paired t-tests, and then (using the differ-
ence between control and treatment consumption for
172
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
0.0
0.1
0.2
0.3
08:00 10:00 12:00 14:00 16:00
Time of day
U
VA
 ir
ra
di
an
ce
 a
t 1
 m
 
(?
W
 c
m
?2
)
U
VB
 ir
ra
di
an
ce
 a
t 1
 m
 
(?
W
 c
m
?2
)
Fig. 1. Representative UV quality and quantity measurements
at Cocos Lagoon, Guam. Presented are the means of 4
monthly measurements from November 1994 to February
1995; error bars have been removed for visual clarity. UVA
(s) and UVB (h) measurements were collected hourly using
an International IL1400A radiometer/photometer at a depth of
1 m. These data were compared to readings collected at the
surface, under UV transparent screens, under UV opaque
screens, under PAR-UV opaque screens, and at 15 m depth 
(see ?Results?)
Slattery & Paul: Effects of bleaching on soft coral defenses
each assay) the data were analyzed by ANOVA with
diet condition as a fixed factor (see Hay et al. 1994 for
description of technique). Data are presented as mean
? SD.
RESULTS
The surface irradiance in Guam during the late fall to
early winter (i.e. November 1994 to February 1995)
averaged 26.34 ? 17.84 ?W cm?2 PAR, 1.85 ? 1.27 ?W
cm?2 UVA and 0.20 ? 0.26 ?W cm?2 UVB over the
course of an average day (Fig. 1). Surface irradiance
was measured at 11:00 in the morning each month and
was typically in the range of 27.8 ?W cm?2 PAR, 1.6 ?W
cm?2 UVA and 0.2 ?W cm?2 UVB. Solar irradiance
attenuation at 1 m and 15 m averaged about 11% and
26%, respectively. UV transparent shades (UVT)
caused an insignificant drop in ambient UV (?1.5%:
1.76 ?W cm?2 UVA and 0.155 ?W cm?2 UVB), while UV
opaque (UVO) and blackened shades (PARO/UVO)
reduced ultraviolet irradiance to levels consistent with
dusk (0.3 ?W cm?2 UVA and 0.01 ?W cm?2 UVB). 
Naturally-bleached soft coral
Sinularia maxima colonies consistently contained 3
defensive metabolites: 2 feeding deterrent compounds
(pukalide and 11?-acetoxypukalide) and one natural
sunscreen (the MAA palythine, ?max = 320 nm); these
were used as biomarkers of bleaching effects. Follow-
ing a natural bleaching event, the soft corals exhibited
significant variability in zooxanthella densities and
lipid, pukalide and palythine concentrations through
time (Fig. 2A,B,D,F). Each of these parameters was
173
1 m unbleached
15 m unbleached
1 m bleached
0
10
20
30
40
Zo
ox
an
th
el
la
 d
en
si
tie
s 
(x
 1
06
) g
?1
 w
et
 w
t
13 Nov 94
Date
P
er
ce
nt
 li
pi
d 
co
nc
.
g?
1  
dr
y 
w
t 
5
10
20
30
35
15
25
P
er
ce
nt
 p
ro
te
in
 c
on
c.
g?
1  
dr
y 
w
t 
15
10
11
12
13
14
C
A
B
11 Dec 94 8 Jan 94 5 Feb 95
0.10
0.15
0.25
0.20
0.30
P
er
ce
nt
 p
uk
al
id
e 
co
nc
.
g?
1  
dr
y 
w
t 
P
er
ce
nt
 1
1?
-a
ce
to
xy
- 
pu
ka
lid
e 
co
nc
. g
?1
 d
ry
 w
t
1.30
1.20
1.10
1.00
P
al
yt
hi
ne
 c
on
c.
 
(n
m
ol
) m
g?
1  
pr
ot
ei
n 
30
20
10
0
D
E
F
13 Nov 94
Date
11 Dec 94 8 Jan 95 5 Feb 95
Fig. 2. Changes in zooxanthella densities and biochemical concentrations in the soft coral Sinularia maxima following a natural
bleaching event in Guam. Compared are bleached soft corals from 1 m depth (s), unbleached soft corals from 1 m depth (h), and
unbleached soft corals from 15 m depth (n) each month from November 1994 (T0) to February 1995 (T3). Data points represent
the mean (?SD; n = 15 replicates per treatment) for (A) zooxanthella densities, (B) % lipid concentration, (C) % protein concen-
tration, (D) %pukalide concentration, (E) %11? acetoxypukalide concentration, and (F) palythine concentration. Note difference 
in scales (y-axis) for each parameter 
Mar Ecol Prog Ser 354: 169?179, 2008
significantly lower in bleached than in unbleached
corals collected from the same shallow-water popula-
tion in November 1994 (ANOVAzooxanthellae: MS = 6.464,
F3,168 = 4.518, p = 0.0151; ANOVAlipid: MS = 371.493,
F3,168 = 9.031, p = 0.0004; ANOVApukalide: MS = 0.008,
F3,168 = 4.086, p = 0.0220; ANOVApalythine: MS =
2316.078, F3,168 = 3.447, p = 0.0386). The recovery of
these soft coral parameters to pre-bleaching levels
occurred within a 1 to 2 mo period (Scheffe?s S : p <
0.05 at T0 [13 Nov 94] and T1 [11 Dec 94] for zooxan-
thella densities and lipid concentration, and p < 0.05 at
T0 only for pukalide and palythine concentration).
Palythine and protein concentrations were the only
parameters that exhibited significant differences
between the shallow- and deep-water populations,
with increased concentrations directly correlated to
solar irradiance (Scheffe?s S : p < 0.05 at T0 to T3
[Nov 94 to Feb 95]). Bite scars on bleached soft corals
exceeded those on unbleached soft corals from the
same depth by almost 4:1 (19.6 ? 8.3, n = 37 and 6.1 ?
2.2, respectively, n = 24; p = 0.023). 
Artificially-bleached soft coral
Solar radiation can cause bleaching in the soft coral
Sinularia maxima (Fig. 3A). The presence of UV-light
resulted in significant losses of zooxanthellae, as well
as declines in lipid, protein and pukalide concentra-
tions (Fig. 3B?D). When a deep-water population of
S. maxima was transplanted to shallow water, corals
exhibited a reduction (within 1 mo) in zooxanthellae
similar to that of naturally-bleached soft corals
(Fig. 3A; open circles). However, at least 2 of our bio-
174
Zo
ox
an
th
el
la
 d
en
si
tie
s 
(x
 1
06
) g
?1
 w
et
 w
t
0
10
20
30
40
P
er
ce
nt
 li
pi
d 
co
nc
. 
g?
1  
dr
y 
w
t 
5
10
20
30
35
15
25
P
er
ce
nt
 p
ro
te
in
 c
on
c.
 
g?
1  
dr
y 
w
t 15
12
16
14
13
Control
Sun
UVT
UVO
PARO/UVO
A
B
C
4 Mar 95
Date
11 Dec 94 8 Jan 95 5 Feb 95
P
er
ce
nt
 p
uk
al
id
e 
co
nc
.
g?
1  
dr
y 
w
t 
0.10
0.15
0.25
0.20
0.30
P
er
ce
nt
 1
1?
?a
ce
to
xy
- 
pu
ka
lid
e 
co
nc
. g
?1
 d
ry
 w
t
1.20
1.10
1.00
0.90
P
al
yt
hi
ne
 c
on
c.
 
(n
m
ol
) m
g?
1  
pr
ot
ei
n 
22
20
18
16
14
12
D
E
F
4 Mar 95
Date
11 Dec 94 8 Jan 95 5 Feb 95
Fig. 3. Changes in zooxanthella densities and biochemical concentrations in the soft coral Sinularia maxima following an artifi-
cial bleaching experiment. Compared are soft corals transplanted from 15 m depth to 1 m depth into 4 treatments: full sunlight
(s), shaded with UV transparent plexiglass (UVT: h), shaded with UV opaque plexiglass (UVO: m), and shaded with PAR/UV
opaque plexiglass (PARO/UVO: j) each month from November 1994 (T0) to February 1995 (T3). Control soft corals (n) were col-
lected at 15 m and then returned to their site of origin (i.e. back-transplant procedural controls). Data points represent the mean
(?SD; n = 10 replicates per treatment) for (A) zooxanthella densities, (B) % lipid concentration, (C) % protein concentration, (D) %
pukalide concentration, (E) % 11? acetoxypukalide concentration, and (F) palythine concentration. Note difference in scales 
(y-axis) for each parameter
Slattery & Paul: Effects of bleaching on soft coral defenses
markers (protein and palythine) displayed slightly dif-
ferent responses under artificial bleaching conditions.
Specifically, naturally bleached soft corals initially had
significantly lower levels of palythine prior to recovery,
while artificially bleached colonies appeared to have
stable levels of this sunscreen prior to recovery. Protein
profiles were similar, although the artificially bleached
colonies recovered faster. Back-transplants into the
deep reef (15 m, Fig. 3 open triangles) and the shallow
reef (1 m) exhibited no significant handling effects on
zooxanthellae or biochemical constituents, however,
shallow to deep transplants also bleached and failed to
recover, much like the PARO/UVO samples (data not
shown). The recovery of soft coral lipid and pukalide to
pre-bleaching levels occurred within a 1 to 2 mo period
(Scheffe?s S : p < 0.05 at T2 [8 Jan 95] and T3 [5 Feb 95]
for lipid and pukalide concentration), as noted in natu-
rally-bleached soft corals (Fig. 2B,D). The deep to
shallow transplants and UVT treatments resulted
in a significant induction of palythine over time
(Fig. 3F: Scheffe?s S : p < 0.05 at T2 and T3). The reduc-
tion of solar radiation (PARO/UVO: see Fig. 1) also
resulted in bleaching, albeit more extreme; in addition
to the loss of zooxanthellae, lipid, and pukalide
(ANOVAzooxanthellae: MS = 15.252, F3,280 = 29.586, p <
0.0001; ANOVAlipid: MS = 706.590, F3,280 = 19.120, p <
0.0001; ANOVApukalide: MS = 0.048, F3,280 = 35.982, p <
0.0001; Scheffe?s S : p < 0.05 at T1 to T3 [Jan to
Mar 95] for zooxanthella densities, lipid and pukalide
concentration), protein and 11?-acetoxypukalide
also exhibited significant declines in concentration
(ANOVAprotein: MS = 8.390, F3,280 = 3.649, p = 0.0162;
ANOVA11?-acetoxypukalide: MS = 0.039, F3,280 = 3.882, p =
0.0122; Scheffe?s S: p < 0.05 at T3 for protein and 11?-
acetoxypukalide concentrations, respectively). More-
over, the PARO/UVO shaded soft corals did not
recover with time (Fig. 3A). 
Feeding deterrence assays
Crude extracts of the bleached and unbleached soft
corals from shallow- and deep-water populations var-
ied in their palatability to the omnivorous pufferfish
Canthigaster solandri (Fig. 4A). The extracts from the
unbleached soft corals were a deterrent to these
predators, while the bleached soft coral extract appar-
ently stimulated feeding (Fig. 4A: significantly more
treatment cubes consumed relative to control). Feed-
ing assays with the extracts of light-shaded soft corals
provided 3 separate outcomes: (1) UVO-shaded soft
corals contained feeding deterrent extracts, (2) UVT-
shaded soft corals contained feeding stimulant ex-
tracts, and (3) PARO/UVO-shaded soft corals con-
tained non-deterrent extracts (Fig. 4B). In 4 ? 3 blocked
pairwise comparisons of amended diets, pukalide
alone was deterrent only at the highest concentration
(Fig. 5A,C), while lipid alone appeared to lose nutri-
tional attractiveness at about 10% (Fig. 5C,D). When
the lipid and pukalide content were simultaneously
manipulated within the artificial food cubes, fish were
more influenced by pukalide concentration than by
lipid concentration (Fig. 5A,D: ANOVAHigh:High: p <
0.0001; ANOVALow:Low: p = 0.0042). A high concentra-
tion of pukalide mixed with a low concentration of lipid
(i.e. Fig. 5C: P+L) was the least attractive diet option to
the fish, but the constitutive effects were additive only
(ANOVA: p = 0.0753).
DISCUSSION
Naturally-induced Sinularia maxima bleaching re-
sulted in significant reductions in lipid, pukalide and
175
0.00
0.25
0.50
0.75
Control
Treatment
1 m-unbleached
15 m-unbleached
1 m-bleached
p<0.0001 p<0.0001p=0.0052
0.00
0.25
0.50
0.75
M
as
s 
lo
st
 (g
)
UVO UVT PARO/UVO
p<0.0001 p=0.6049p<0.0001
A
B
Fig. 4. Feeding effects of Sinularia maxima extracts on the
generalist pufferfish Canthigaster solandri in the laboratory.
Histograms represent the mean food cube mass lost (+SD; n =
15 replicates per treatment) to feeding pufferfish standard-
ized to food cube hydration via no-fish controls. Compared
are: (A) extracts from unbleached soft corals at 1 m and 15 m
depth, and bleached soft corals at 1 m depth, and (B) extracts
from soft corals treated to 3 different light conditions: UV 
opaque, UV transparent, and UV/PAR opaque shading
Mar Ecol Prog Ser 354: 169?179, 2008
palythine concentrations, but not protein or 11?-ace-
toxypukalide concentrations. However, experimen-
tally-bleached soft corals also exhibited a significant re-
duction in protein, but no change in palythine
concentration. These differences in biomarker re-
sponse may represent some of the inherent differences
between the shallow (naturally-bleached: Nov 94) and
deep (experimentally-bleached: Dec 94) populations of
S. maxima; protein is especially labile over environ-
mental gradients (Slattery & McClintock 1995). Lipid,
an important energy reservoir in cnidarians (Stimson
1987), can be supplied to host tissues by translocation
from zooxanthellae symbionts (Falkowski et al. 1984).
The loss of zooxanthellae-derived nutrients following
symbiont loss (i.e. bleaching) can have a negative im-
pact on coral health and growth (Muller-Parker &
D?Elia 1997, Baker 2001). Significant reduction in S.
maxima lipid following bleaching implies that zooxan-
thellae-derived nutrients are an important component
of the overall energetic budget of this soft coral and that
potential compensatory trade-offs in metabolic activi-
ties will be requisite with the loss of this re-
source. In general, energetic resources
should be allocated among growth, repro-
duction, maintenance, acquisition and de-
fense (Cronin 2001). When resources be-
come limiting, stressed organisms often
favor production of defensive metabolites
over the other metabolic requirements
(Gershenzon 1994). On the other hand,
stress can also result in biochemical
changes that increase prey attractiveness to
predators and/or reduce defenses (EST;
White 1984, Rhoades 1985, Cronin & Hay
1996a). We noted a decrease in levels of the
defensive metabolite pukalide during
bleaching, as predicted by the EST, but no
change in 11?-acetoxypukalide concentra-
tion. Michalek-Wagner & Bowden (2000)
argued that increases in 3 defensive
metabolites by 2 species of soft corals in
Australia following a bleaching experiment
maintained protection against bacterial in-
fections during this period of physiological
stress. Their findings were particularly
relevant since the antibacterial metabolite
flexibilide was produced preferentially
over the anti-predator metabolite sinulari-
olide. Pukalide and 11?-acetoxypukalide
are both important predator deterrent com-
pounds for Sinularia maxima (Wylie & Paul
1989, Slattery et al. 2001), and both have
modest antimicrobial activity (Slattery et al.
2001), so it is unclear why the latter com-
pound was favored over the former. How-
ever, 11?-acetoxypukalide appears to be far less labile
than pukalide (Slattery et al. 2001), and maintenance of
either defensive metabolite during a period of environ-
mental stress could be adaptive (Michalek-Wagner &
Bowden 2000) since both compounds share overlap-
ping ecological functions. The reduction of pukalide
during bleaching, despite the fact that cembranoid
diterpenes are apparently biosynthesized by octocorals
(Kokke et al. 1984, Michalek-Wagner et al. 2001), sug-
gests that either translocated energetic resources from
symbionts are used to fuel pukalide biosynthesis, or
symbiont-derived translocated products act as the
biosynthetic precursors to this molecule. 
In contrast, mycosporine-like amino acids (including
palythine) are sunscreen products of the shikimate
biosynthetic pathway that is present in bacteria and
primary producers but not in animals (Shick et al.
1999), although there is increasing evidence that meta-
zoan consumers can convert bioaccumulated precursor
MAAs into secondary MAA products (Shick 2004). The
decline in palythine concentrations in Sinularia max-
176
High (5.0%) ? Pukalide content ? Low (0.125%)
Lo
w
 (1
0%
) ?
 L
ip
id
 c
on
te
nt
 ?
 H
ig
h 
(3
0%
)  
A
C
B
D
0.75
0.50
0.25
0.0
MS = 0.088
F = 27.84
p < 0.0001
* *
M
as
s 
lo
st
 (g
)
M
as
s 
lo
st
 (g
)
0.75
0.50
0.25
0.0
MS = 0.235
F = 175.623
p < 0.0001
* *
*
0.75
0.50
0.25
0.0
MS = 0.001
F = 1.022
p = 0.4110
P L P+L P L P+L
0.75
0.50
0.25
0.0
MS = 0.003
F = 3.793
p = 0.0042
*
Fig. 5. Feeding effects of the Sinularia maxima defensive compound
pukalide, and lipid on the generalist pufferfish Canthigaster solandri in
the laboratory. Histograms represent the mean food cube mass lost (+SD;
n = 15 replicates per treatment) to feeding pufferfish standardized to food
cube hydration via no-fish controls. Open bars designate results from con-
trol food cubes, while shaded bars designate results from the treatment
food cubes. Compared are assays which combined: (A) high levels of
pukalide with high levels of lipid, (B) low levels of pukalide with high lev-
els of lipid, (C) high levels of pukalide with low levels of lipid, and (D) low
levels of pukalide with low levels of lipid. All feeding assays were con-
ducted for identical periods of time in order to allow ANOVA comparisons 
between assays. P = pukalide, L = lipid
Slattery & Paul: Effects of bleaching on soft coral defenses
ima following bleaching (Fig. 2F) and subsequent
increases in palythine following zooxanthella recov-
ery, supports sunscreen synthesis by the symbionts.
Translocation of MAAs to host tissue is a common phe-
nomenon (e.g. Gleason & Wellington 1995, Adams &
Shick 1996), and some MAAs are maintained or
increased by host tissues even when symbiont popula-
tions decline (Shick et al. 1999). The relatively rapid
change in palythine concentration (in direct correla-
tion with zooxanthellae abundance) implies the sun-
screen is largely sequestered within symbiont rather
than host cells in S. maxima. Significantly, our data
indicate that palythine production may be an inducible
response to increases in UVR (Fig. 3F), presumably to
limit photo-damage (Gleason & Wellington 1995,
Adams & Shick 1996). Inducible chemical defenses
have been well documented in terrestrial plant-herbi-
vore associations (Karban & Baldwin 1997), and rarely
in marine algae-grazer interactions (Cronin & Hay
1996b, Toth et al. 2007). However, inducible chemical
defenses may be a relatively important mechanism of
coping with light stress (e.g. Zangerl & Berenbaum
1987, Torregiani & Lesser 2007). Prior work with zoo-
xanthellae-host associations also indicated that light
acclimatization is due to genetic variability of the sym-
bionts (Rowan et al. 1997), although the phenotypic
characteristics that are responsible for increased resis-
tance to light quality and/or quantity were not eluci-
dated.  
Recovery of the soft corals following bleaching was a
relatively rapid event; within 2 mo of a natural or
experimentally-induced bleaching event, symbiont
populations had returned to pre-bleached levels in
Sinularia maxima. Likewise, soft corals transplanted
from depth to the shallow reef flat lost about 85% of
their symbiont populations in a month, and within
another month, zooxanthellae concentration had
recovered to about 80% of the original levels. These
recovery rates are about 3 to 4 mo faster than those
observed by Michalek-Wagner & Willis (2001) in the
soft coral Lobophytum compactum. Some of these dif-
ferences might be due to species-specific and/or sym-
biont clade differences (Hofmann 1999, Rowan et al.
1997, respectively). Alternatively, degree of stress
appears to be an important consideration in the sever-
ity of bleaching and the rate of recovery. We subjected
S. maxima to differential light stress only, while
Michalek-Wagner & Willis (2001) induced bleaching
via a combination of light and temperature extremes.
In studies with Lobophytum compactum and Sinularia
flexibilis, artificial bleaching procedures resulted in
biochemical responses different from those in the same
species collected in the field during the 1998 GBR
(Great Barrier Reef, Australia) bleaching event.
Michalek-Wagner & Bowden (2000) observed signifi-
cant increases in defensive metabolite levels (sinulari-
olide, isolobophytolide, and flexibilide) during the nat-
ural bleaching event, but noted decreased levels of
these metabolites when soft corals were experimen-
tally bleached. These researchers concluded that dura-
tion of stress and possibly the added stress of freshwa-
ter run-off in the field were responsible for these
differences between natural and experimentally-
induced bleaching. Evidence has accumulated to sug-
gest that bleaching is not the harbinger of coral reef
death (Muller-Parker & D?Elia 1997, Rowan et al.
1997), and bleaching may represent a mechanism to
ensure that hosts periodically turn over their symbiont
populations in favor of better stress-adapted individu-
als that can tolerate environmental changes (Baker
2001, but see Hoegh-Guldberg et al. 2002). Our study
assessed the ?holobiont? so it is unclear whether recov-
ery represents an infection by new symbionts or sur-
vival and replication of some of the original symbiont
population (Barneah et al. 2004). Bleaching due to
increased UV radiation has been previously docu-
mented (e.g. Gleason & Wellington 1993), but the
importance of incident light quality to zooxanthellae
stability within cnidarian hosts has rarely been dis-
cussed. For example, the near absence of all solar irra-
diance (PARO/UVO shades: Fig. 3) resulted in zooxan-
thellae loss in Sinularia maxima within 1 mo. As noted,
under natural bleaching conditions these soft corals
exhibited significant reductions in pukalide and paly-
thine concentrations. However, soft corals shaded in
this manner did not recover symbionts, and this ulti-
mately took its toll on other biochemical constituents
(i.e. declines in protein and 11?-acetoxypukalide), fur-
ther demonstrating the importance of the host?sym-
biont association (Muller-Parker & D?Elia 1997). Our
results imply that S. maxima cannot sustain popula-
tions of zooxanthellae symbionts without light, and
that extended symbiont-free periods are detrimental
to the health of this soft coral (Baker 2001, McClana-
han 2004) and, possibly, the coral reef community
(McClanahan et al. 2001). 
Sinularia maxima feeding deterrence was signifi-
cantly impacted by bleaching, either natural or experi-
mentally-induced. Naturally-bleached soft corals were
more susceptible to predation than unbleached soft
corals by about 4:1 (based on our field observations of
bite scars produced by a suite of coral reef fishes, e.g.
butterflyfishes, angelfishes, and pufferfishes: M. Slat-
tery pers. obs.). Extracts of bleached individuals actu-
ally stimulated feeding by an omnivorous pufferfish
Canthigaster solandri (Fig. 4A). These results support
the EST hypothesis in that stressed soft corals became
more attractive to predators than unstressed (i.e.
unbleached) individuals. Renaud et al. (1990) and
Cronin & Hay (1996a) were the only previous marine
177
Mar Ecol Prog Ser 354: 169?179, 2008
studies to assess the effect of abiotic stressors on prey
chemical defenses. They examined differences in
palatability of the temperate alga Dictyota menstrualis
subjected to variable dessication regimes and noted
higher levels of grazer preference for the more
severely stressed thalli. Additional studies have
demonstrated reduced food preference due to bio-
chemical factors other than chemical defenses (e.g.
Goranson et al. 2004). Our work provided somewhat
puzzling results on the biochemical stress-related
changes that lead to increased soft coral palatability.
We expected, a priori, to see reductions in all of the
defensive metabolites to account for the observed
increased levels of predation on naturally-bleached
individuals. Our expectations were based on prior
work that indicated minimal concentrations of
pukalide and 11?-acetoxypukalide (0.5% and 1.1%,
respectively) deter predation by Canthigaster solandri
(Slattery et al. 1998, 2001), and that lipid levels often
dictate feeding behavior (Stimson 1987). However,
while pukalide levels decreased there was no signifi-
cant change in 11?-acetoxypukalide levels following
bleaching. Thus to counteract the feeding deterrent
effects of 11?-acetoxypukalide, lipid and pukalide lev-
els would have to rise and fall, respectively. That lipid
levels also dropped, despite an increase in prey palata-
bility, suggests that specific constituents of the lipid
were disproportionately lost resulting in an increased
percentage in specific feeding attractant constituents
(Bachok et al. 2006). For example, lipid constituents
typically include polyunsaturated fatty acids; these
would be susceptible to degradation by light or radical
oxidation (an indirect effect of UVR) and would there-
fore change the ratio of this biochemical component
and might account for changes in attractiveness. More-
over, these results support previous data that indicate
11?-acetoxypukalide is not as bioactive as pukalide
(Slattery et al. 2001), and seem to indicate that other
nutritional qualities of the crude extract (e.g. insoluble
proteins, carbohydrates etc) mask the deterrent effects
of this compound (Simpson & Raubenheimer 2001). It
is clear that further work is required in order to ascer-
tain the specific mechanisms by which bleaching
impacts feeding susceptibility in Sinularia maxima,
and the role of the zooxanthellae symbionts will need
to be considered (e.g. Michalek-Wagner et al. 2001).
However, our results support the EST model and indi-
cate that the indirect consequences of bleaching might
have a greater impact on the health of the soft coral
(e.g. tissue damage due to predation) than the tempo-
rary loss of symbionts. As consequences of global cli-
mate change become more pronounced in shallow
tropical marine ecosystems, it is likely that indirect
effects of these abiotic stressors will have significant
impacts on community dynamics. 
Acknowledgements. We thank C. Avila, D. Ginsburg, D.
Nagle, B. Novack, M. Puglisi, P. Schupp, J. Starmer, and R.
Thacker for their tireless efforts in the field. Y. Benayahu pro-
vided taxonomic verification of the species examined in this
study; for his help and insights we are extremely grateful. M.
Lesser provided MAA standards for analytical analyses, for
which we are indebted. Assistance with extractions and ana-
lytical chemistry were provided by B. Avery, D. Comfort Jr., B.
Laird, W. Walker, M. Warren, and K. Wesson. M. Lesser, D.
Gochfeld and D. Nagle provided comments that improved
this manuscript. This research was supported by grants from
the National Science Foundation OCE 9321533 & OCE
9528570, MS-AL Sea Grant R/AT-3-PD, and NOAA
NA16RU1496 to MS, and the National Institutes of Health GM
38624 to VJP. This is contribution number 615 from the Uni-
versity of Guam Marine Laboratory and contribution number
720 from the Smithsonian Marine Station at Fort Pierce.
LITERATURE CITED
Adams NL, Shick JM (1996) Mycosporine-like amino acids
provide protection against ultraviolet radiation in eggs of
the green sea urchin Strongylocentrotus droebachiensis.
J Photochem Photobiol 64:149?158
Anthony KRN, Connolly SR, Willis BL (2002) Comparative
analysis of energy allocation to tissue and skeletal growth
in corals. Limnol Oceanogr 47:1417?1429
Bachok Z, Mfilinge P, Tsuchiya M (2006) Characterization of
fatty acid composition in healthy and bleached corals from
Okinawa, Japan. Coral Reefs 25:545?554
Baker AC (2001) Reef corals bleach to survive change. Nature
411:765?766
Barneah O, Weis VM, Perez S, Benayahu Y (2004) Diversity of
dinoflagellate symbionts in Red Sea soft corals: mode of
symbiont acquisition matters. Mar Ecol Prog Ser 275:
89?95
Baruch R, Avishai N, Rabinowitz C (2005) UV incites diverse
levels of DNA breaks in different cellular compartments of
a branching coral species. J Exp Biol 208:843?848
Brown BE (1997) Coral bleaching: causes and consequences.
Coral Reefs 16:S129?138
Cronin G (2001) Resource allocation in seaweeds and marine
invertebrates: chemical defense patterns in relation to
defense theories. In: McClintock JB, Baker BJ (eds)
Marine chemical ecology. CRC Press, Boca Raton, FL,
p 325?354
Cronin G, Hay ME (1996a) Seaweed-herbivore interactions
depend on the recent history of both the plant and the ani-
mal. Ecology 77:1531?1543
Cronin G, Hay ME (1996b) Induction of seaweed chemical
defenses by amphipod grazing. Ecology 77:2287?2301
Cruz-Rivera E, Hay ME (2003) Prey nutritional quality inter-
acts with chemical defenses to affect consumer feeding
and fitness. Ecol Monogr 73:483?506
Douglas AE (2003) Coral bleaching?how and why? Mar Poll
Bull 46:385?392
Duffy JE, Paul VJ (1992) Prey nutritional quality and the
effectiveness of chemical defenses against tropical reef
fishes. Oecologia 90:333?339
Falkowski PG, Dubinsky Z, Muscatine L, Porter JW (1984)
Light and bioenergetics of a symbiotic coral. Bioscience
34:705?709
Gershenzon J (1994) The cost of plant chemical defense
against herbivory: a biochemical perspective. In: Bernays
EA (ed) Insect?plant interactions. CRC Press, Boca Raton,
FL, p 105?173
178
Slattery & Paul: Effects of bleaching on soft coral defenses
Gleason DF, Wellington GM (1993) Ultraviolet radiation and
coral bleaching. Nature 365:836?838
Gleason DF, Wellington GM (1995) Variation in UVB sensitiv-
ity of planula larvae of the coral Agaricia agaricites along
a depth gradient. Mar Biol 123:693?703 
Glynn PW (1993) Coral reef bleaching: ecological perspec-
tives. Coral Reefs 12:1?17
Goranson CE, Ho CK, Pennings SC (2004) Environmental
gradients and herbivore feeding preferences in coastal
salt marshes. Oecologia 140:591?600
Grottoli AG, Rodrigues LJ, Juarez C (2004) Lipids and stable
carbon isotopes in two species of Hawaiian corals, Porites
compressa and Montipora verrucosa, following a bleach-
ing event. Mar Biol 145:621?631
Hay ME, Kappel QE, Fenical W (1994) Synergisms in plant
defenses against herbivores: interactions of chemistry,
calcification, and plant quality. Ecology 75:1714?1726
Hoegh-Guldberg O, Jones RJ, Ward S, Loh WK (2002) Is coral
bleaching really adaptive? Nature 415:601?602
Hofmann GE (1999) Ecologically relevant variation in induc-
tion and function of heat shock proteins in marine organ-
isms. Am Zool 39:889?900
Karban R, Baldwin IT (1997) Induced responses to herbivory.
University of Chicago Press, Chicago IL
Karentz D, McEuen FS, Land KM, Dunlap WC (1991) Survey
of mycosporine-like amino acid compounds in Antarctic
marine organisms: potential protection from ultraviolet
exposure. Mar Biol 108:157?166
Kokke WCMC, Epstein S, Look SA, Rau GH, Fenical W,
Djerassi C (1984) On the origin of terpenes in symbiotic
associations between marine invertebrates and algae
(zooxanthellae). J Biol Chem 259:8168?8173
Lesser MP (1997) Oxidative stress causes coral bleaching dur-
ing exposure to elevated temperatures. Coral Reefs 16:
187?192
McClanahan TR (2004) The relationship between bleaching
and mortality of common corals. Mar Biol 144:1239?1245
McClanahan T, Muthiga N, Mangi S (2001) Coral and algae
changes after the 1998 coral bleaching: interaction with
reef management and herbivores on Kenyan reefs. Coral
Reefs 19:380?391
Michalek-Wagner K, Bowden BF (2000) Effects of bleaching
on secondary metabolite chemistry of alcyonacean soft
corals. J Chem Ecol 26:1543?1562 
Michalek-Wagner K, Willis BL (2001) Impacts of bleaching on
the soft coral Lobophytum compactum. II. Biochemical
changes in adults and their larvae. Coral Reefs 19:240?246
Michalek-Wagner K, Bourne DJ, Bowden BF (2001) The
effects of different strains of zooxanthellae on the sec-
ondary-metabolite chemistry and development of the soft
coral host Lobophytum compactum. Mar Biol 138:753?760 
Muller-Parker G, D?Elia CF (1997) Interactions between
corals and their symbiotic algae. In: Birkeland C (ed) Life
and death of coral reefs. Chapman & Hall, New York,
p 96?113
Renaud PE, Hay ME, Schmitt TM (1990) Interactions of plant
stress and herbivory: intraspecific variation in the suscep-
tibility of a palatable versus an unpalatable seaweed to
sea urchin grazing. Oecologia 82:217?226 
Rhoades DF (1985) Offensive-defensive interactions between
herbivores and plants: their relevance to herbivore popu-
lation dynamics and ecological theory. Am Nat 125:
205?238
Rowan R, Knowlton N, Baker A, Jara J (1997) Landscape ecol-
ogy of algal symbionts creates variation in episodes of
coral bleaching. Nature 388:265?269
Shick JM (2004) The continuity and intensity of ultraviolet
irradiation affect the kinetics of biosynthesis, accumula-
tion, and conversion of mycosporine-like amino acids
(MAAs) in the coral Stylophora pistillata. Limnol
Oceanogr 49:442?458
Shick JM, Romaine-Lioud S, Ferrier-Pages C, Gattuso JP
(1999) Ultraviolet-B radiation stimulates shikimate path-
way-dependent accumulation of mycosporine-like amino
acids in the coral Stylophora pistillata despite decreases
in its population of symbiotic dinoflagellates. Limnol
Oceanogr 44:1667?1682
Simpson SJ, Raubenheimer D (2001) The geometric analysis
of nutrient-allelochemical interactions: a case study using
locusts. Ecology 82:422?439
Slattery M, McClintock JB (1995) Population structure and
feeding deterrence in three Antarctic soft corals. Mar Biol
122:461?470
Slattery M, Avila C, Starmer J, Paul VJ (1998) A sequestered
soft coral diterpene in the aeolid nudibranch Phyllo-
desmium guamensis. J Exp Mar Biol Ecol 226:33?49
Slattery M, Starmer J, Paul VJ (2001) Temporal and spatial
plasticity in chemical constituents of tropical Pacific soft
corals Sinularia maxima and S. polydactyla. Mar Biol 138:
1183?1193
Spencer Davies P (1991) Effects of daylight variations on
the energy budgets of shallow-water corals. Mar Biol
108:137?144
Stimson J (1987) The location, quantity and rate of change in
quantity of lipids in tissues of Hawaiian hermatypic corals.
Bull Mar Sci 41:889?904
Torregiani JH, Lesser MP (2007) The effects of short-term
exposures to ultraviolet radiation in the Hawaiian coral
Montipora verrucosa. J Exp Mar Biol Ecol 340:194?203
Toth GB, Karlsson M, Pavia H (2007) Mesoherbivores reduce
net growth and induce chemical resistance in natural sea-
weed populations. Oecologia 152:245?255
Warner ME, Chilcoat GC, McFarland FK, Fitt WK (2002) Sea-
sonal fluctuations in the photosynthetic capacity of photo-
system II in symbiotic dinoflagellates in the Caribbean
reef-building coral Montastrea. Mar Biol 141:31?38
White TCR (1984) The abundance of invertebrate herbivores
in relation to the availability of nitrogen in stressed food
plants. Oecologia 63:90?105
Wylie CR, Paul VJ (1989) Chemical defenses in three species
of Sinularia (Coelenterata, Alcyonacea)?effects against
generalist predators and the butterflyfish Chaetodon uni-
maculatus Bloch. J Exp Mar Biol Ecol 129:141?160
Zangerl AR, Berenbaum MR (1987) Furanocoumarins in wild
parsnip: effects of photosynthetically active radiation,
ultraviolet light, and nutrients. Ecology 68:516?520
179
Editorial responsibility: Joseph Pawlik,
Wilmington, North Carolina, USA
Submitted: March 22, 2007; Accepted: August 7, 2007
Proofs received from author(s): January 26, 2008