GENERAL AND COMPARATIVE

ENDOCRINOLOGY
General and Comparative Endocrinology 137 (2004) 9–18
www.elsevier.com/locate/ygcen

Spawning and gamete follicle rupture in the cnidarian Renilla

koellikeri: effects of putative neurohormones
 Tremblay, Josee Henry, and Michel Anctil*
Marie-Eve
Departement de sciences biologiques and Centre de recherche en sciences neurologiques, Universite de Montreal, C.P. 6128, Succ. Centre-Ville,
Montreal, Que., Canada H3C 3J7
Received 9 December 2003; revised 17 February 2004; accepted 23 February 2004

Abstract

The neuroendocrine control of spawning (release of intact gamete follicles) and of the ensuing exfoliation (freeing of gametes by
follicle epithelium rupture) was investigated in colonies of the sea pansy Renilla koellikeri, an octocorallian of the sea pen family.
Polyps of male colonies produce substantially more sperm follicles than female colonies do egg follicles, and significantly more
sperm follicles are expelled than egg follicles during the summer spawning season. Spawning is accompanied by strong peristaltic
contractions across the colony. Serotonin, a positive modulator of peristalsis in the sea pansy, induced spawning of either sperm or
egg follicles, increasing both the proportion of spawning colonies and the number of expelled gamete follicles per colony in a dose-
dependent manner. The serotonin antagonist 1-(1)naphthylpiperazine greatly reduced both spontaneous and serotonin-induced
spawning. Antho-RFamide, a neuropeptide found in ciliated neurons within follicle epithelia, induced the exfoliation of the follicle
epithelium from spawned follicles. Exposure of follicles to light enhanced the potency of Antho-RFamide. The actin-binding toxin
phalloidin substantially reduced the incidence of Antho-RFamide-induced exfoliation and phalloidin-FITC staining was localized in
the muscle feet of follicle epithelial cells. These results provide the first experimental evidence of neuroendocrine functions involved
in cnidarian spawning.
Ó 2004 Elsevier Inc. All rights reserved.

Keywords: Cnidaria; Spawning; Serotonin; Antho-RFamide; Sea pansy; Reproduction

1. Introduction diploblastic animals such as cnidarians have hindered

Although the processes of gamete maturation and
release and their control by environmental factors have
been extensively investigated in cnidarians (Fautin,
2002; Fautin et al., 1989; Harrison and Wallace, 1990
for reviews), very little is known of the internal factors
involved in the mediation of these processes. Neuroen-
docrine pathways constitute one category of factors of
paramount importance in the control of invertebrate
reproduction (Adams, 1997). However, neuroendocrine
circuitry is difficult to unravel in the absence of a con-
ventional circulatory system, and both the apparent lack
of centralized nervous systems and of gonadal organs in
*
Corresponding author. Fax: +1-514-343-2293.
E-mail address: michel.anctil@umontreal.ca (M. Anctil).
the search for neural bases of reproductive control.
In the gonochoric sea pansy Renilla koellikeri, as in
other octocorals, there are eight mesenteries in each
autozooid polyp and the two lateral pairs of mesenteries
bear the egg or sperm follicles inside the colonial mass
(rachis). During the spawning season (May to late July)
along the Southern California coastline, gamete follicles
are released in the wake of accelerated and robust ra-
chidial peristalsis, then expelled through the mouth by
peristalsis of the pharynx (Satterlie and Case, 1979;
Wilson, 1883). Although spawning may occur through-
out the day, Wilson (1883) observed that it occurred
primarily at dawn in Renilla reniformis while Satterlie
and Case (1979) noted that small spawns could be ob-
tained in R. koellikeri by altering the light regime, thus
suggesting that light plays a role in inducing spawning
events. Spawning usually occurs synchronously among
the central autozooid polyps within a colony (Wilson,

0016-6480/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.ygcen.2004.02.009
10  Tremblay et al. / General and Comparative Endocrinology 137 (2004) 9–18
M.-E.
1883), but synchronization of spawning activities be-
tween neighbouring colonies may also occur (Satterlie
and Case, 1979) without matching the extent of mass
synchronous spawning of certain reef corals (Babcock
et al., 1986; Harrison et al., 1984).
In order to find clues for the possible involvement of
neurohormones in spawning, a knowledge of the
chemical neuroanatomy of the sea pansy is necessary,
particularly as it relates to the reproductive tissues. With
the possible exception of RFamide-immunoreactive
neurons innervating the reproductive tissue of Hydra-
ctinia echinata (Grimmelikhuijzen, 1985), the sea pansy
appears to be the only cnidarian to date for which a
comprehensive neurochemical mapping includes repro-
ductive tissues. First, serotonin and melatonin immu-
noreactivities were co-localized in neurons with long
processes coursing around the egg and sperm follicles
(Mechawar and Anctil, 1997). Second, Antho-RFamide-
containing neurons were immunolocalized in the follicle
epithelium of both egg and sperm follicles (Pernet et al.,
2004). Last, gonadotropin-releasing hormone(GnRH)-
immunoreactive neurons were localized in the mesen-
teries bearing gametes (Anctil, 2000).
Physiological evidence also supports a possible in-
volvement of these transmitters in sea pansy spawning.
First, serotonin positively and melatonin negatively
modulate peristalsis (Anctil, 1989; Anctil et al., 1991), an
activity greatly enhanced during spawning. Second, the
native peptide Antho-RFamide appears to act as a
neuromuscular transmitter but it can also enhance
peristalsis (Anctil and Grimmelikhuijzen, 1989). Last,
mammalian GnRH and partially purified sea pansy ex-
tracts with high GnRH-like immunoreactivity modulate
peristalsis negatively (Anctil, 2000). Recently, a GnRH-
like peptide was identified in the cephalopod Octopus
vulgaris and it was localized in neurons and gland cells
of the optic gland that regulates reproductive activity
(Iwakoshi et al., 2002). Thus as in vertebrates GnRHs
may be involved in the control of reproduction in in-
vertebrates.
The evidence pointing to serotonin, melatonin, An-
tho-RFamide, and GnRHs as potential neurohormonal
factors involved in sea pansy spawning prompted the
present study in which the effects of these substances
were examined on two short-term reproductive activities
of sea pansies: the release and expulsion of gamete fol-
licles (spawning) and, after expulsion, the freeing of
gametes by follicle rupture (exfoliation).
2. Materials and methods
2.1. Animal handling and maintenance
Colonies of R. koellikeri Pfeffer were collected by
divers in subtidal waters of the Southern California
coast and shipped to Montreal by Marinus (Long
Beach, California). They were maintained in filtered and
aerated artificial sea water (ASW, Instant Ocean) at
18 °C and pH 8.0, in an aquarium with a coarse sand
bed. The unfed colonies were exposed to a light/dark
cycle similar to that prevailing in their natural envi-
ronment during the time of the study (May–August of
2002 and 2003). During June when the majority of
samplings and manipulations occurred, the daylight
period was set between 05:45 and 21:00 EST. Experi-
ments were conducted on batches of colonies from
successive arrivals to the Montreal laboratory at inter-
vals of two weeks. Prior to observations and experi-
ments, colonies were kept in darkness for 16 h. Colonies
under observation were placed in individual finger bowls
covered with cheese gauze and put inside the aquarium
so as to be exposed to the same conditions as those of
the free-standing colonies. The restrained colonies were
checked several times daily for spawning events between
09:00 and 16:00 EST. In addition, some restrained col-
onies were removed from the aquarium and transferred
from darkness to light (100 W incandescent light bulb
intensities of 2000–3000 lx and colour temperature of
2850 K) or from ASW temperature of 16–24 °C and the
incidence of spawns was recorded. Temperatures above
24 °C caused deterioration of sea pansy tissues.
2.2. Quantitative assessment of reproductive tissue and
observations of spawning
To assess the level of reproductive maturity in the
colonies, egg, and sperm follicles were counted from
early May–late July. The colonial mass (rachis) of 75
colonies (50 colonies in 2002 and 25 in 2003) was sliced
into strips 5 mm wide and the follicles from all the strips
were teased apart with dissecting forceps. They were
spread on the bottom of a petri dish placed over a
Jencons Easi-Grid transparency and viewed with a
magnifying glass to facilitate counting. Male and female
colonies were identified by the color of their follicles.
Eggs have a tan or orange color while sperm balls ap-
pear whitish with streaks spread fan-wise on them. Be-
cause it was often difficult to separate the smaller
follicles from surrounding tissue of similar color, they
were stained by injecting methylene blue into the gas-
trovascular channels of the rachis through cannulation
of a retracted autozooid polyp. The number of autozo-
oid (reproductive) polyps was also recorded for each
colony sampled in 2002. The means and standard error
of means of the counts of egg follicles, sperm follicles
and autozooids per colony were computed and the re-
sults analyzed using the unpaired t test with Welsh’s
correction (GraphPad Software).
Prior to observations of induced spawning, sponta-
neous spawning events (gamete shedding and expulsion)
were recorded by videography. Colonies were removed
  Tremblay et al. / General and Comparative Endocrinology 137 (2004) 9–18
M.-E.
in the morning from the darkened aquarium and im-
mersed in ASW-filled finger bowls for observation.
When follicles first became visible inside the polyps,
filming of the polyps with a Sony Handycam TR57 vi-
deo camera was carried out for the next 2–3 h to view an
entire spawning sequence. The sides of the bowls were
covered with black cloth and a lamp with a 100 W
electric bulb was placed at a distance of 30 cm from the
bowl to optimise contrast between polyps and back-
ground. The videotapes from 10 such episodes were
digitized with an ATI Technologies All-in-Wonder Ra-
deon graphics card and still images from the digital re-
cordings were obtained with ATI Multimedia Center
software.
For observations of the spontaneous or induced ex-
foliation process, egg and sperm follicles were surgically
removed from reproductive tissues and were placed in a
petri dish filled with ASW. Exfoliation was observed
with a Wild M400 photomicroscope and photographed
at a magnification of 64
using Kodak T-MAX 100
black-and-white film.
2.3. Effect of putative neurohormones on spawning
Chemicals were tested for their capacity to induce
spawning by adding them with a syringe to 500 ml of
ASW into the finger bowl in which the colony was im-
mersed, to obtain the desired final concentration. The
micromolar/millimolar range of concentrations of test
chemicals in this study are commonly used for cnidari-
ans, and the neuropeptide Antho-RFamide, which is
native to the sea pansy, was shown to have maximal
effects on muscle contraction at 1–10 lM (Anctil and
Grimmelikhuijzen, 1989). Colonies were allowed to re-
lax, becoming turgescent and displaying peristaltic ac-
tivity, before exposure to test chemicals. As controls,
some colonies were exposed to the equivalent volume of
ASW prior to adding the chemicals.
In each experiment, 10 colonies were exposed to the
test chemicals and the number of colonies undergoing
spawning was recorded for up to 2 h after adding
chemicals. Colonies were scored as undergoing spawn-
ing if two or more follicles were expelled from the mouth
of at least one autozooid. The tested neurohormones
were the indolamines serotonin creatinine sulfate and
melatonin (both from Sigma Chemicals), and mamma-
lian GnRH and Antho-RFamide (both from Peninsula
Laboratories). In addition, two serotonin antagonists,
mianserin HCl and 1-(1-naphthyl)piperazine HCl (both
from Sigma–Aldrich), were used. In experiments with
the antagonists, the latter were added in the finger bowls
holding the colonies 15 min prior to adding serotonin. In
the serotonin dose–response experiments the means of
the incidence of spawning were calculated from each set
of experiments and the values were normalized as per-
centage of the maximal response. Different sets of 10
11
colonies were used for each serotonin concentration.
From the data nonlinear regression curve fits were ex-
trapolated using GraphPad software.
From dose–response experiments conducted in 2002
on the effect of serotonin on spawning, data on the
number of expelled follicles per spawning episode were
also recorded for both sexes. For each serotonin con-
centration the means were calculated from the counts of
follicles and the results were expressed as percentage of
maximal numbers of expelled follicles.
2.4. Effect of putative neurohormones on exfoliation of
follicles
For exfoliation studies, 20 gamete follicles extracted
from one single colony were transferred to a petri dish
containing 1 ml of ASW, and 1 ml of the four test neu-
rohormones tried in spawning experiments (see above)
was added to obtain a final concentration of 5 lM. The
follicles were then kept under observation using the
dissecting microscope. Controls were performed with 20
follicles from the same colony as the tested follicles,
placed in a petri dish containing 2 ml of ASW. The
number of exfoliated follicles was counted 15, 45, and
75 min after adding test chemicals. These experiments
were first performed with the preparations exposed to
light. The follicles were exposed to a 100 W lamp placed
30 cm away from the preparation dish (illuminating the
water surface at an intensity of 3000 lx) which was
maintained at 18 °C over a cooling plate. The experi-
ments were then repeated in complete darkness with
other sets of 20 follicles from the same colonies.
In experiments testing the contribution of contractile
elements to the exfoliation response, the actin-binding
toxin phalloidin (Sigma–Aldrich) was made up as a
1 mM stock solution in dimethyl sulfoxide (DMSO). An
aliquot from this solution was added to the ASW in the
finger bowls in each of which 40 follicles from one col-
ony lie, to obtain a final concentration of either 0.1 or
10 lM. After a 15 min incubation in phalloidin, Antho-
RFamide was added for a final concentration of 10 lM,
and the incidence of exfoliations scored over the next
75 min. For controls the pre-incubation in phalloidin
was omitted and the preparations were exposed to
10 lM Antho-RFamide and 1% DMSO, or to 1%
DMSO alone in ASW. These experiments were con-
ducted under artificial light and at 18 °C as specified
above.
2.5. Localization of phalloidin binding
Sea pansy colonies were sliced in 5 mm thick strips
from which clusters of follicles were teased out and
immersed in 4% paraformaldehyde in phosphate-buf-
fered saline (PBS) for 1 h at 18 °C. The clusters were
rinsed in PBS, then in PBS with 15% and 30% sucrose
12  Tremblay et al. / General and Comparative Endocrinology 137 (2004) 9–18
M.-E.
until they sank at the bottom of the vials. The clusters
were embedded in O.C.T. Compound (Tissue-Tek) and
stored at )80 °C. Frozen sections 16 lm-thick were cut
with a Leica CM3050 cryostat and deposited on gelatin-
coated slides. Sections were rinsed in PBS with 0.3%
Triton-X-100 (PBST) for 20 min, then covered with a
solution of phalloidin-FITC (Sigma–Aldrich) diluted
1:250 or 1:100 in PBST for 1 h in darkness. After rinsing
the sections in PBST, they were counterstained with 1%
Evans Blue in PBS, rinsed again in PBS and mounted in
glycerol diluted 1:2 in PBS. The sections were examined
with a Nikon Eclipse TE 300 microscope and photo-
graphed with a Nikon Coolpix 4500 digital camera.
Images were cropped and contrasted with Adobe
Photoshop. Controls consisted of slides in which phal-
loidin-FITC was replaced with FITC-conjugated goat
anti-rabbit immunoglogulin (Jackson Immunological
Products) diluted 1:80 in PBST.
3. Results
3.1. Egg and sperm follicles during the spawning season
To evaluate the extent of gamete production up to
and during the reproductive season, 75 sea pansy colo-
nies were sampled in May–July of 2002 and 2003 and
the number of their follicles were systematically coun-
ted. Of these 75 colonies, 34 were males, 27 were fe-
males, and the sex of the remaining 14 could not be
determined because their follicles (gametes with overly-
ing epithelium) were few and small. Sperm and egg
Fig. 1. Average number of gamete follicles (light and medium-dark bars) and of reproductive (autozooid) polyps (dark bars) per male versus female
colony.
follicles become distinguishable on the basis of color
(tan for eggs, white for sperm balls) and texture (yolk
granules for eggs, fan-spread streaks for sperm balls)
only when a critical size is reached, 100 lm in diame-
ter. Both male and female follicles reached diameters
400 lm at maturity in our study.
In samples from 2002, there were over 3400 follicles/
colony in males and over 1400 egg follicles/colony in
females (Fig. 1), and this difference is significant
(P < 0:0001, unpaired t test with Welch’s correction).
Given that an average of 194 autozooid polyps were
counted in male colonies and 182 in female colonies
(Fig. 1), these numbers represent 18 sperm follicles/au-
tozooid and 7–8 egg follicles/autozooid. A similar dis-
tribution pattern obtained in colonies sampled in 2003
(Fig. 1). The highest number of sperm follicles in a
colony (5201) was recorded in June 2002 and that for a
female colony (1565) in May 2002.
Fig. 2 shows images illustrating the sequence of
events leading to gamete follicle release from autozooid
polyps during spawning episodes. When follicles are
shed from their mesenteries they are carried by circu-
lating sea water up the coelenteron of the polyps
(Fig. 2A) where they may remain for several hours or a
few days. Some polyps may carry large numbers of
follicles while neighbour polyps are almost devoid of
follicles. The follicles access the lower lip of the pharynx
(Fig. 2B) and start ascending the pharynx by peristalsis
(Fig. 2C). When the follicles reach the mouth opening,
the tentacles bend over the mouth and the body column
of the polyp bends as the follicles are being expelled
(Fig. 2D). It required as little as 35 s from the entry of
  Tremblay et al. / General and Comparative Endocrinology 137 (2004) 9–18
M.-E.
Fig. 2. Still frames from video sequences of spawning sea pansy col-
onies. (A) Released gamete follicles (arrow), 400 lm in diameter,
moving freely in the coelenteron of an autozooid polyp. (B) Bottom of
an autozooid where follicles (small arrow) aggregate before entering
the lumen of the pharynx. Note also the cluster of non-reproductive
siphonozooid polyps (large arrow). (C) One follicle (arrow) is seen to
move up by peristalsis inside the lumen of the pharynx. (D) Follicles
(arrows) are being expelled out of the mouth of a partially contracted
autozooid. (E) Expelled follicles are stuck to a mucus ball connected to
the mouth (not shown) of an autozooid by a mucus thread (arrow).
follicles into the pharynx to their expulsion through the
mouth. Some polyps of spawning colonies were ob-
served to expel copious amounts of mucus with (Fig. 2E)
or without follicles. Changes in exposure to artificial
light (from complete darkness to luminance of 3000 lx)
did not affect the rate of spontaneous or induced
spawning (not shown), whereas spawning occurred only
in ASW temperatures between 18 and 22 °C.
3.2. Effects of putative neurohormones on spawning
Colonies were first screened during the summer of
2002 for their ability to spawn when exposed to putative
transmitters. Ten colonies were used for each experi-
mental condition except for exposure to serotonin alone
in which 20 colonies were tested. In order to determine
whether unresponsive colonies lacked mature follicles,
post-experiment dissections were conducted to examine
13
the state of maturity of follicles in reproductive tissues.
Exposure of mature colonies to Antho-RFamide,
mammalian GnRH and melatonin (all at 10 lM) failed
to induce spawning. In contrast, serotonin induced
spawning in all colonies tested except for six colonies
(30%) devoid of mature follicles. The response delay of
the effect varied between 5 and 45 min after serotonin
addition.
In separate experiments conducted over two spawn-
ing seasons (2002 and 2003), serotonin was found to
induce spawning in a dose-dependent manner (Fig. 3A).
The response appeared to saturate at around 0.1 mM
serotonin in which 75% of the tested colonies were in-
duced to spawn (Fig. 3B). The incidence of spawning
appears relatively high at the lowest serotonin concen-
tration tested because there is a basal 15% of mature
colonies that spawn spontaneously (Fig. 3B). When the
serotonin antagonist 1-(1-naphthyl)piperazine was ad-
ded with serotonin, the dose–response curve shifted to
higher serotonin concentrations by a tenfold (Fig. 3A),
representing a drop to less than half the number of
spawns induced by 0.1 mM serotonin (Fig. 3B). The
basal rate of spawning was also reduced significantly by
1-(1-naphthyl)piperazine (Fig. 3A), from 15 to 5% of
colonies tested. Mianserin, another serotonin antago-
nist, failed to reduce serotonin-induced spawning (Figs.
3A and B).
The number of follicles expelled during each spawn-
ing episode also increased with increasing serotonin
concentrations (Fig. 4). The average number of expelled
follicles for 10 colonies rose from 6 at 1 lM to 17 at
0.1 mM. The number of expelled sperm follicles greatly
exceeded the number of expelled egg follicles (Fig. 4,
inset; P < 0:05, unpaired t test), in keeping with the
large difference between male and female colonies in the
number of stored follicles (Fig. 1).
3.3. Effects of putative neurohormones on follicle
exfoliation
A small number of released follicles were exfoliated
before they were expelled during spawning. Thirty
minutes or more after expulsion, the follicles that had
remained intact underwent exfoliation. To minimize
exfoliation before exposing the follicles to transmitter
candidates, unreleased follicles were removed directly
from their mesenteries. When allowed to settle in petri
dishes in darkness or dim light, no exfoliation of these
follicles was observed during the first 60 min, and only a
few follicles spontaneously exfoliated in the following
2 h. Follicles from 22 colonies exposed to serotonin,
melatonin or mammalian GnRH did not undergo ex-
foliation beyond the rate of spontaneous exfoliation in
ASW (not shown).
In contrast, Antho-RFamide induced exfoliation
(Fig. 5). No difference of potency was observed between
14  Tremblay et al. / General and Comparative Endocrinology 137 (2004) 9–18
M.-E.

Fig. 3. (A) Dose–response curves of the spawn-inducing effect of serotonin (N ¼ 6), alone (data from years 2002 and 2003), and in combination with
mianserin (N ¼ 3) or 1-(1)naphthylpiperazine (N ¼ 5) (both from year 2003), on sea pansy colonies. Data points represent means  SEM of the
percentage of sampled colonies that spawned. N ¼ number of separate experiments, each experiment conducted with ten colonies. (B) Histogram
showing the means  SEM of the number of colonies (out of 10 colonies per experiment) spawning spontaneously (control) or induced to spawn by
0.1 mM serotonin alone or in the presence of the serotonin antagonists mianserin and 1-(1-naphthyl)piperazine. N ¼ as in (A).

male and female colonies, and between colonies sampled
in 2002 and 2003. Follicles from each of eight colonies
were exposed to three experimental conditions: (1) ASW
under artificial light (control), (2) Antho-RFamide in
darkness, and (3) Antho-RFamide under artificial light.
There were significant differences in the mean number of
exfoliated follicles for each colony (P ¼ 0:0001, Krus-
kal–Wallis test) 75 min after adding Antho-RFamide
between the three conditions (Fig. 5A). The number of
exfoliated follicles was five times greater in preparations
exposed to Antho-RFamide with light than in control
preparations, with intermediate values for preparations
exposed to Antho-RFamide in darkness. Exfoliations
started to occur 15 min after treatment, then progressed
steadily for the next hour, except for preparations ex-
posed to Antho-RFamide in darkness where the values
levelled off between 45 and 75 min after treatment
(Fig. 5B).
Sperm follicles exposed to Antho-RFamide are nearly
spherical and have a smooth surface at first (Fig. 6A),
then they become slightly pear-shaped as a protrusion
appears at one pole (Fig. 6B). The bulge becomes larger,
exerting tension on the follicle epithelium (Figs. 6C and
D). Finally the follicle epithelium is ruptured and sperm
starts oozing out (Fig. 6E), expanding into a cloud
(Fig. 6F). This process unfolded over a period of
  Tremblay et al. / General and Comparative Endocrinology 137 (2004) 9–18
M.-E. 15

Fig. 4. Dose–response curve of the effect of serotonin on the number of gamete follicles expelled during each spawning episode (pooled data for 10
colonies). Data points represent means  SEM (N ¼ 3 separate experiments). Inset: histogram illustrating the large difference between male and
female colonies in the number of gamete follicles expelled in response to 10lM serotonin. All data from experiments conducted in 2002.

8 min. Eggs change their shape momentarily when
freed from the follicle epithelium, but otherwise the ex-
foliation process was similar to that of sperm follicles
(not shown).
Phalloidin, a toxin that binds and stabilizes muscle
actin (Wieland, 1977), reduced the Antho-RFamide-in-
duced incidence of follicle exfoliation considerably and
in a dose-dependent manner (Fig. 7). The response to
10 lM Antho-RFamide dropped by 80% in the presence
of 0.1 lM phalloidin, and even under the basal rate
(spontaneous exfoliations) with 10 lM phalloidin. When
formaldehyde-fixed gamete follicles were treated with
phalloidin-FITC in phosphate-buffered saline contain-
ing Triton X-100, the basal part of the epithelium
forming the follicle envelope around the gametes was
stained (Fig. 7, inset). This is the zone of the endoderm
where the contractile feet of myoepithelial cells are lo-
cated. The staining in this zone appeared as a diffuse,
thin sheet overlying the gametes in whole-mounts of
follicles (not shown) or as a ring of punctate profiles in
sectioned mesenteries (Fig. 7, inset).
4. Discussion
In this study, we present new findings on the
spawning process of the sea pansy, in particular a sharp
sexual dimorphism in the quantity of gamete follicles
produced and released, and we provide the first experi-
mental evidence in a cnidarian of the involvement of
putative neurohormones, namely serotonin and Antho-
RFamide, in the control of the key reproductive events
of spawning and follicle exfoliation. These results pave
the way for an eventual understanding of cnidarian
reproductive endocrinology.
The large difference in the number of follicles between
male and female sea pansy colonies (Fig. 1) is a phe-
nomenon that appears to have gone unnoticed in other
sea pens and in octocorals in general, with the exception
of a report by Orejas et al. (2002) for the gorgonian
Ainigmaptilon antarcticum. However, while the number
of male follicles was more than twice that of female
colonies in the sea pansy, the reverse relationship ob-
tained in A. antarcticum, female follicles being twice as
numerous as male follicles. It is possible that the
broadcast of intact sperm follicles in the water during
spawning allows sperm follicles to settle along with egg
follicles of similar mass before exfoliation and thus in-
crease the probability of fertilization. In the swift cur-
rents of sublittoral waters inhabited by the sea pansy
(Kastendiek, 1976), the relatively large number of male
follicles may help to offset the threat of excessive sperm
dispersal, and therefore of diminished egg-sperm en-
counters, by producing large quantities of sperm. Our
report of the expulsion of intact sperm follicles agrees
with a similar finding by Eckelbarger et al. (1998) in the
sea pen Pennatula aculeate, but contradicts a previous
observation of Satterlie and Case (1979) that sperm balls
disintegrate before expulsion in the sea pansy. The rea-
son for the latter discrepancy is unclear.
Of the four putative neurohormones for which immu-
noreactivity was previously observed in the innervation
16  Tremblay et al. / General and Comparative Endocrinology 137 (2004) 9–18
M.-E.
Fig. 5. Effect of light and Antho-RFamide 5 lM on exfoliation of
gamete follicles. (A) Number of observed exfoliations [means  SEM,
N ¼ 8 (4 male + 4 female colonies)] 75 min after first exposure to light,
Antho-RFamide, or the two combined. (B) Time course of recruitment
of exfoliations (pooled data from colonies in A). All data from ex-
periments conducted in 2002.
of reproductive tissues (Anctil, 2000; Mechawar and
Anctil, 1997; Pernet et al., 2004), only serotonin was
effective in inducing spawning in the sea pansy. Sero-
tonin induced spawning in a dose-dependent manner
(Fig. 3) and caused the expulsion of an increasing
number of egg or sperm follicles during each spawning
episode (Fig. 4). The specificity of the serotonin effect is
supported by the efficacy of the serotonin antagonist
1-(1-naphthyl)piperazine in reducing the induced
spawning while another antagonist, mianserin, was in-
effective (Fig. 3). It is indicative of the potency of sero-
tonin that at or near saturating concentrations, only
colonies lacking mature follicles were not induced to
spawn. These observations should not be surprising in
view of the widely known inducing effect of serotonin on
spawning in several invertebrates, especially in bivalve
molluscs (Smith and Croll, 1997 for review). Our find-
ings suggest that the role of serotonin in the control of
spawning is a phylogenetically ancient one.
Both Chia and Crawford (1973) for the sea pen
Ptilosarcus gurneyi and Satterlie and Case (1979) for the
Fig. 6. Sequence of photomicrographs (A–F) showing the typical
exfoliation process of a sperm follicle induced by Antho-RFamide. See
Section 3 for explanations.
sea pansy reported that spawning was accompanied by
enhanced peristaltic waves across the colonies. Strong
peristaltic waves moving over the colony are likely to aid
in releasing mature follicles from their attachment in
mesenteries, and we noted in this study that a peristaltic
wave also accompanied the progress of follicles up the
pharynx toward the mouth of autozooid polyps (Fig. 2).
Therefore it is likely that serotonin induced spawning
through its previously documented enhancing effect on
peristaltic contractions of the sea pansy (Anctil, 1989).
The involvement of peristalsis in serotonin-induced
spawning is supported by the finding that the serotonin
antagonist 1-(1-naphthyl)piperazine, which was highly
effective in reducing spontaneous and serotonin-induced
peristaltic contractions in the sea pansy (Anctil, 1989), is
  Tremblay et al. / General and Comparative Endocrinology 137 (2004) 9–18
M.-E.
Fig. 7. Histogram showing the effect of increasing concentrations of
phalloidin on the occurrence of exfoliation in follicles exposed to
10 lM Antho-RFamide. Control ¼ number of exfoliations occurring
without exposure to Antho-RFamide, but in the presence of light.
Values are means  SEM Inset: photomicrograph showing punctate
fluorescence in the follicle epithelium of an egg from a section of co-
lonial mass treated with phalloidin-FITC. Note that staining is more
dense in the layer of the epithelium closer to the egg membrane (ar-
rows). Nu, egg nucleus. All data from experiments conducted in 2003.
also effective in reducing spontaneous spawning as well
as in antagonising serotonin-induced spawning in the
present study. Further support is obtained by the lack of
effect of another serotonin antagonist, mianserin, on
either peristalsis (Anctil, 1989) or spawning.
A few serotonin-immunoreactive neurons were ob-
served to encircle the follicles in the sea pansy (Mecha-
war and Anctil, 1997). A similar arrangement has also
been reported on egg-producing ootypes and in the
muscular wall of the uterus of the parasitic platyhel-
minth Fasciola hepatica in which serotonin appears to be
involved in egg production and release (Fairweather,
1997; Fairweather et al., 1987 for review). It is possible
that as the follicles increase in diameter with maturation,
these neurons stretch, and as a threshold is reached,
serotonin is released followed by enhancement of peri-
stalsis. However, numerous other serotonin-immunore-
active neurons have also been reported in somatic tissues
of the autozooid polyps and in the rachis of the sea
pansy (Umbriaco et al., 1990) and they are as likely to
affect peristalsis colony-wide. The alternative possibility
that serotonin mediates a light signal to trigger spawn-
ing is challenged by our inability to induce spawning by
changing light conditions.
The observation that gamete follicles were intact
when expelled during spawning prompted us to ask
whether another factor, separate from serotonin, was
involved in follicle exfoliation. We found that the pep-
tide Antho-RFamide consistently induced exfoliation of
both male and female follicles within 15 min of treat-
ment, even though follicles removed from the gonad
tissues inside the colonies, not spawned follicles, were
17
used for experiments. In addition, light was sufficient to
induce a few exfoliations and necessary to sustain re-
cruitment of large numbers of exfoliations over 75 min
after treatment with Antho-RFamide (Fig. 5). This ap-
parent synergy of light and Antho-RFamide suggests
that exfoliation is mediated by photoreceptive cells.
That these cells may also contain and release Antho-
RFamide is further suggested by observations of RFa-
mide immunoreactive ciliated cells within the gamete
follicle epithelia (Pernet et al., 2004).
These cells extend a process to the contractile feet of
the adjacent myoepithelial cells that are also part of the
follicle epithelia (Pernet et al., 2004). Thus the change in
shape of both egg and sperm follicles from spherical to
pear-shaped and the development of tension around a
bulge at one pole of the follicles noted during exfoliation
(Fig. 6) are consistent with a contraction of the myo-
epithelial cells that would exert a distorting pressure on
the gametes. This is supported by the finding that
phalloidin, known to bind and stabilise muscle actin
(Wieland, 1977), not only binds to the presumptive
contractile layer of the follicle epithelium but also lar-
gely prevents Antho-RFamide-induced exfoliations
(Fig. 7). Therefore we propose that the Antho-RFa-
mide-containing cells in the follicles act as light-sensitive
cells as well as motoneurons to mediate the exfoliation
process directly.
It is not clear why exfoliation does not occur inside the
colony prior to spawning. In view of the apparent sensi-
tivity to light of the exfoliation process, one possibility is
that insufficient light would be available to the follicles
because of the screening effect of tissue layers in the sur-
face of the colonial mass (rachis) overlying the mesen-
teries. In both the dorsal and ventral walls of the rachis as
well as in the walls of the chambers enclosing the follicle-
bearing mesenteries, numerous deep purple calcareous
spicules are present (see Fig. 1 in Pernet et al., 2004) and
are likely to act as an effective filter to the incoming light.
Acknowledgments
We thank Louise Pelletier for her technical assistance
with photomicroscopy. This research was supported by
a grant from the Natural Science and Engineering
Research Council (NSERC) of Canada to M.A. and by
a NSERC Summer Research Fellowship to M.E.T.
References
Adams, T.S. (Ed.), 1997. Reproductive Biology of Invertebrates, Vol.
VIII: Progress in Reproductive Endocrinology. Wiley, Chichester,
UK.
Anctil, M., 1989. Modulation of a rhythmic activity by serotonin via
cyclic AMP in the coelenterate Renilla kollikeri. J. Comp. Physiol.
159, 491–500.
18  Tremblay et al. / General and Comparative Endocrinology 137 (2004) 9–18
M.-E.
Anctil, M., 2000. Evidence for gonadotropin-releasing hormone-like
peptides in a cnidarian nervous system. Gen. Comp. Endocrinol.
119, 317–328.
Anctil, M., Grimmelikhuijzen, C.J.P., 1989. Excitatory action of the
native neuropeptide antho-RFamide on muscles in the pennatulid
Renilla kollikeri. Gen. Pharmacol. 20, 381–384.
Anctil, M., Pani, A.K., Ali, M.A., 1991. Modulation of rhythmic
contractions by melatonin via cyclic GMP in the coelenterate
Renilla koellikeri. J. Comp. Physiol. 161, 569–575.
Babcock, R.C., Bull, G.D., Harrison, P.L., Heyward, A.J., Oliver,
J.K., Wallace, C.C., Willis, B.L., 1986. Synchronous spawnings of
105 scleractinian coral species on the great barrier reef. Mar. Biol.
90, 379–394.
Chia, F.-S., Crawford, B.J., 1973. Some observations on gametogen-
esis, larval development and substrate selection of the sea pen
Ptilosarcus gurneyi. Mar. Biol. 23, 73–82.
Eckelbarger, K.J., Tyler, P.A., Langton, R.W., 1998. Gonadal
morphology and gametogenesis in the sea pen Pennatula aculeata
(Anthozoa: Pennatulacea) from the Gulf of Maine. Mar. Biol. 132,
677–690.
Fairweather, I., 1997. Platyhelminthes. In: Adams, T.S. (Ed.), Repro-
ductive Biology of Invertebrates, Vol. VIII: Progress in Reproduc-
tive Endocrinology. Wiley, Chichester, UK, pp. 1–31.
Fairweather, I., Maule, A.G., Mitchell, S.H., Johnston, C.F., Halton,
D.W., 1987. Immunocytochemical demonstration of 5-hydroxy-
tryptamine (serotonin) in the nervous system of the liver fluke,
Fasciola hepatica (Trematoda, Digenea). Parasitol. Res. 73, 255–
258.
Fautin, D.G., 2002. Reproduction in Cnidaria. Can. J. Zool. 80, 1735–
1754.
Fautin, D.G., Spaulding, G.J., Chia, F.-S., 1989. Cnidaria. In:
Adiyodi, K.G., Adiyodi, R.G. (Eds.), Reproductive Biology of
Invertebrates. vol. IVA. Fertilization, Development, and Parental
Care. IBH Publishing Co, Oxford, New Delhi, pp. 43–62.
Grimmelikhuijzen, C.J.P., 1985. Antisera to the sequence Arg–Phe-
amide visualize neuronal centralization in hydroid polyps. Cell
Tissue Res. 241, 171–182.
Harrison, P.L., Wallace, C.C., 1990. Reproduction, dispersal and
recruitment of scleractinian corals. In: Dubinski, Z. (Ed.), Coral
Reefs. Elsevier Science Publishers, Amsterdam, pp. 133–207.
Harrison, P.L., Babcock, R.C., Bull, G.D., Oliver, J.K., Wallace, C.C.,
Willis, B.L., 1984. Mass spawning in tropical reef corals. Science
223, 1186–1189.
Iwakoshi, E., Takuwa-Kiroda, K., Fujisawa, Y., Hisada, M., Ukena,
K., Tsutsui, K., Minakata, H., 2002. Isolation and characterization
of a GnRH-like peptide from Octopus vulgaris. Biochem. Biophys.
Res. Commun. 291, 1187–1193.
Kastendiek, J., 1976. Behavior of the sea pansy Renilla kollikeri Pfeffer
(Coelenterata: Pennatulacea) and its influence on the distribution
and biological interactions of the species. Biol. Bull. 151, 518–
537.
Mechawar, N., Anctil, M., 1997. Melatonin in a primitive metazoan:
seasonal changes of levels and immunohistochemical visualization
in neurons. J. Comp. Neurol. 387, 243–254.
Orejas, C., Lopez-Gonzalez, P.J., Gili, J.M., Teixido, N., Gutt, J.,
Arntz, W.E., 2002. Distribution and reproductive ecology of the
Antarctic octocoral Ainigmaptilon antarcticum in the Weddell Sea.
Mar. Ecol. Progr. Ser. 231, 101–114.
Pernet, V., Anctil, M., Grimmelikhuijzen, C.J.P., 2004. Antho-RFa-
mide-containing neurons in the primitive nervous system of the
anthozoan Renilla koellikeri. J. Comp. Neurol. 472, 208–220.
Satterlie, R.A., Case, J.F., 1979. Development of bioluminescence and
other effector responses in the pennatulid coelenterate Renilla
kollikeri. Biol. Bull. 157, 506–523.
Smith, S.A., Croll, R.P., 1997. Mollusca. In: Adams, T.S. (Ed.),
Reproductive Biology of Invertebrates, vol. VIII: Progress in
Reproductive Endocrinology. Wiley, Chichester, UK, pp. 61–151.
Umbriaco, D., Anctil, M., Descarries, L., 1990. Serotonin-immunore-
active neurons in the cnidarian Renilla koellikeri. J. Comp. Neurol.
291, 167–178.
Wieland, T., 1977. Interaction of phallotoxins with actin. Adv.
Enzyme Regul. 15, 285–300.
Wilson, E.B., 1883. The development of Renilla. Philos. Trans. R. Soc.
Lond. B 174, 723–815.