Hormonal and pheromonal control of spawning behavior in the

Fish Physiology and Biochemistry 26: 71–84, 2002.
© 2003 Kluwer Academic Publishers. Printed in the Netherlands.
Hormonal and pheromonal control of spawning behavior in the goldfish
Makito Kobayashi1,∗ , Peter W. Sorensen2,∗ & Norm E. Stacey3,∗
1 Department
of Biology, Division of Natural Sciences, International Christian University, Mitaka, Tokyo 1818585, Japan; 2 Department of Fisheries, Wildlife, and Conservation Biology, University of Minnesota, St. Paul,
MN 55108 USA; 3 Department of Biological Sciences, University of Alberta, Edmonton, Alberta, T6G 2E9, Canada
Accepted: December 1, 2002
Species that employ sexual reproduction must synchronize gamete maturity with behavior within and between genders. Teleost fishes solve this challenge by using reproductive hormones both as endogenous signals to synchronize
sexual behavior with gamete maturation, and as exogenous signals (pheromones) to synchronize spawning interactions between fish. This dual role of hormonal products is best understood in the goldfish, an external fertilizer
with a promiscuous mating system. Female gonadal growth and vitellogenesis is stimulated by 17β-estradiol (E2)
which also evokes release of a recrudescent pheromone. At the completion of vitellogenesis, ovarian E2 production
drops and plasma testosterone increases, sensitizing the female gonadotropin II (luteinizing hormone; LH) system
to environmental cues (temperature, spawning substrate, pheromones). These cues eventually trigger a LH surge
that alters steroidogenesic pathways to favor the production of progestins including 17,20β-dihydroxy-4-pregnen3-one (17,20β-P). Plasma 17,20β-P stimulates oocyte maturation but is also released to the water along with
sulfated 17,20β-P and androstenedione to serve as a preovulatory pheromone. This pheromone stimulates male
behavior, LH release, and sperm production. At the time of ovulation, females become sexually active in response
to prostaglandin F2α (PGF2α) synthesized in the oviduct. PGF2α and its metabolites are released as a postovulatory pheromone that induces male spawning behavior which further increases male LH and sperm production.
Androgenic hormones are required for male behavior and LH release. Although goldfish are gonochorists, hormone
treatments can induce heterotypical functions in adults. Similar findings in other fish demonstrate that a sexually
bipotential brain is not restricted to hermaphroditic fishes.
Spawning, the controlled release of gametes by fish,
requires synchronization of behavior with gamete maturation both within and between the sexes. Not surprisingly, control of this key reproductive event is
complex and involves both internal hormonal signals
and external cues including pheromones (external signals that pass between member of the same species),
photoperiod, temperature, and spawning substrate.
Despite considerable interest in fish reproduction, the
physiological control of spawning is well understood
in only a few species, a frustrating situation given
the enormous diversity of reproductive strategies pursued by the approximately 24,000 species of extant
∗ All authors share equal responsibility for this manuscript.
fishes. Physiological control of spawning is probably
best understood in the goldfish, Carassius auratus,
an unremarkable member of the Family Cyprinidae
(the largest family of fishes) which has been domesticated for at least a few thousand years. Because the
goldfish is a suitable size for use in the laboratory
where it readily matures and spawns, its reproductive endocrinology and behavior have been intensively
studied (Van Der Kraak et al. 1998; Kobayashi et al.
1989a, 1997; Stacey and Sorensen 2002), often as a
model for the edible carps (Sokolowska et al. 1984).
Goldfish studies have confirmed basic aspects of
hypothalamo-hypophyseal-gonadal function common
to vertebrate reproductive systems, and illustrated the
value of nonmammalian models by revealing novel
features of reproductive control. Early discoveries
that gonadotropin II (homologous to luteinizing hor-
mone [LH]; Yoshiura et al. 1997) release is inhibited
by hypothalamic dopamine (Chang and Peter 1983),
and that prostaglandin F2α (PGF2α) acts as a hormone to induce female sexual behavior (Stacey 1976),
have fundamentally changed thinking about physiological events regulating ovulation and spawning.
More recently, our understanding has been profoundly
influenced by the discovery that hormonal products released by goldfish act as ‘hormonal pheromones’ that
affect conspecific physiology and behavior (Dulka et
al. 1987b; Sorensen et al. 1987, 1998; Stacey and
Sorensen 2002). The dual and simultaneous actions
of hormonal compounds as both internal and external
signals obliges us to reconsider the classical notion
that sex hormones synchronize reproductive processes
only within individuals. Finally, because goldfish sex
behaviors and LH release are sexually dimorphic and
readily elicited in the laboratory, the goldfish serves
as a valuable model for studying both the hormonal
control of sexually dimorphic function, and the potential for heterotypical (opposite gender) function in a
gonochoristic (separate gender) fish.
Here, we review information on three aspects of
spawning in goldfish: (1) hormonal control of behavior and gonadal maturation, (2) pheromonal control of
behavior and sperm production, and (3) plasticity that
exists within these reproductive systems. Space limitations preclude reference to any but the most recent
and relevant information on other species.
Hormonal control of spawning
Gonadal maturation and endocrine events proceeding
Females. Because male goldfish remain in a state of
spawning readiness throughout the spawning season,
the timing of spawning is determined by the female’s
ovulatory LH surge. The female LH surge follows
an extended period of vitellogenesis during which females release a recrudescent pheromone that attracts
males (Female ‘recrudescent’ pheromone, see below).
The early and mid-vitellogenic period is characterized by elevated plasma 17β-estradiol (E2), whereas
completion of vitellogenesis is marked by reduced
E2 and increased levels of 19 carbon (‘C19’) sex
steroids including testosterone (T) (Kagawa et al.
1984; Kobayashi et al. 1988). This postvitellogenic
increase in T appears to prime the ovulatory LH
surge system to respond to environmental cues that
trigger LH release. For example, if sexually regressed, ovariectomized or vitellogenic females are
implanted with T and moved into warm water, they exhibit female-typical LH surges, whereas females with
blank implants, or sexually mature males with high
plasma T, do not (Kobayashi et al. 1986a, 1989b; M.
Kobayashi, unpublished). The positive influence of
T on the LH surge in postvitellogenic female goldfish differs from the positive feedback effect of E2
on the LH surge in mammals (Kalra and Kalra 1997).
Thus, whereas preovulatory E2 in mammals typically
induces an LH surge after a predictable latency, preovulatory T in goldfish can remain high for months
until they encounter exogenous ovulatory cues such as
warm water, spawning substrate, or the odor of ovulating conspecifics (Stacey et al. 1979a, b; Sorensen and
Stacey 1987). These environmental factors also appear
to synchronize ovulation among females (Kobayashi
et al. 1988).
Although the ovulatory LH surge is facilitated by
T and triggered by environmental and social cues, a
circadian clock set by ambient photoperiod determines
its precise timing. At typical spawning temperatures
(18–20 ◦ C) on long (14–16 L) days, the surge commences late in photophase and terminates late in the
following scotophase with the occurrence of ovulation (Stacey et al. 1979a, b; Kezuka et al. 1989).
Goldfish therefore spawn synchronously in low light
of early morning, likely a strategy to reduce predation. In addition to determining the timing of ovulation
(and therefore spawning), onset of the LH surge determines the timing of release of the hormonally-derived
preovulatory pheromone which is comprised of a
changing mixture of steroids: androstenedione (AD),
the maturation inducing steroid 17,20β-dihydroxy-4pregnen-3-one; (17,20β-P; Kobayashi et al. 1987;
Stacey et al. 1989; Scott and Sorensen 1994), and
17,20β-dihydroxy-4-pregnen-3-one-20-sulfate in particular (17,20β-P-S; Sorensen et al. 1995a; see below).
The goldfish LH surge appears remarkably similar to
that of the common carp, Cyprinus carpio (Santos et
al. 1986).
Males. Prior to spawning, males increase LH if
they make contact with the female preovulatory
pheromone; thus, males and females that interact shortly before ovulation experience synchronous
LH increase. The pheromone-induced LH increase
in males stimulates testicular synthesis of 17,20βP (Kobayashi et al. 1986a, b; Dulka et al. 1987b),
in turn increasing the number and motility of sperm
in the sperm ducts (DeFraipont and Sorensen 1993)
and male fertility at spawning (Zheng et al. 1997).
Male-typical LH release occurs within minutes of exposure to the preovulatory pheromone regardless of
time of day (Dulka et al. 1987a, b), and appears to
be regulated by plasma androgens; i.e., it occurs consistently in males but rarely in females unless they are
implanted with 11-ketotestosterone (KT) (Kobayashi
et al. 1997a). A similar male-typical, pheromonal
17,20β-P induced LH increase occurs in both common
carp (Barry et al. 1990; Stacey et al. 1994) and crucian
carp, Carassius carassius (Bjerselius et al. 1995).
Endocrine control of spawning behavior
Overview of goldfish spawning behavior. The goldfish mating system is typical of many other nonterritorial, non-parental cyprinids, such as the common carp, with which it will hybridize (Taylor and
Mahon 1977). Females commence vitellogenesis during winter, exhibit group synchronous oocyte development, and ovulate one to several times over a protracted spring-summer spawning season (Kobayashi
et al. 1986b, c). Starting several hours prior to ovulation (which occurs near dawn) small, disorganized
groups of males actively ‘follow’ females, frequently
initiating physical contact (‘nudging’) and inspecting
their urogenital and gill regions where pheromones
are released (Sorensen et al. 2000). Following behavior typically leads to active ‘chasing’, the intensity of
which increases until the time of ovulation as males
appear to compete to be closest to the female(s),
at times ‘pushing’ each other away (DeFraipont and
Sorensen 1993). Once ovulated, females become sexually receptive and initiate spawning acts by entering aquatic vegetation (Figure 1). Typically, one/few
male(s) will enter vegetation with a receptive female
and then swim rapidly with the female through a small
arc, at which time the female will release (oviposit)
eggs and the male(s) will release (ejaculate) sperm to
complete a ‘spawning act’. Released eggs adhere to
the vegetation. Female spawning behavior will continue until all ovulated eggs are released, and may
involve a hundred or more oviposition acts over several hours with various males. In this manuscript,
the female spawning act is referred to as ‘femaletypical behavior’ or ‘female sex behavior’, and chasing, nudging, and male spawning acts are referred to
as ‘male-typical behaviors’ or ‘male sex behaviors’
(Figure 1).
Female behavior. PGF2α is the key hormone regulating female sex behavior in goldfish (Stacey 1987;
Sorensen et al. 1995b). At ovulation, oocytes stimulate the oviduct to increase synthesis of PGF2α,
which rises dramatically in the blood, and travels to
the brain where it induces female receptivity (Stacey
and Peter 1979; Sorensen et al. 1988, 1995b, and
unpublished) (Figure 2). PGF2α appears to play a critical role synchronizing female-typical behavior with
ovulation. Thus, spawning behavior of ovulated females is terminated by manually stripping ovulated
oocytes (Yamazaki 1965; Stacey and Liley 1974) or
by injecting the prostaglandin synthesis inhibitor, indomethacin (Stacey 1976; Sorensen unpublished), and
is rapidly induced by injecting nonovulated females
(or males) with PGF2α (Stacey 1976, Stacey and Peter
1979) (Figure 2).
In addition to functioning as a behavioral hormone in female goldfish, PGF2α and one of its
major metabolites, 15-keto-prostaglandinF2α (15KPGF2α), are released to function as a potent
pheromone that triggers male courtship and induces
rapid endocrine and testicular responses (Sorensen
et al. 1988, 1989, 1995b, unpublished; Zheng
and Stacey 1997) (see Postovulatory Prostaglandin
Pheromone below). It is not yet clear exactly how
common this ‘hormonal pheromonal’ role for PGF2α
might be amongst the fishes. However, PGF2α injection is known to induce spawning behaviors in
nonovulated females of a wide variety of teleosts
including other cypriniforms (lampan jawa, Puntius
gonionotus; Liley and Tan 1985: loach, Misgurnus anguillicaudatus; Kitamura et al. 1994) and more highly
derived species (two-spot cichlid, Cichlasoma bimaculatum; Cole and Stacey 1984: paradise fish, Macropodus opercularis; Villars and Burdick 1986: dwarf
gourami, Colisa lalia; Yamamoto et al. 1997). Also,
studies of the sensitivity of the olfactory epithelium
to water-borne PGF2α and related compounds have
found that many ostariophysan and salmonid fishes
detect prostaglandins (Stacey and Sorensen 2002),
suggesting widespread use of prostaglandins as both
sex hormones and pheromones by freshwater fishes.
Unlike the situation in tetrapods, where female sex
behaviors typically are dependent on ovarian estrogen
(Blaustein and Erskine 2002), female-typical behavior of goldfish appears to be independent of ovarian
steroids. For example, PGF2α-induced behavior is unaffected by ovarian maturity, ovariectomy, or T or
E2 replacement therapy and is exhibited by juvenile
females tested prior to their first ovarian maturation
Figure 1. Diagrammatic representation (ethogram) of the spawning behavior of male and female goldfish. See text for details.
(Kobayashi and Stacey 1993). However, in the internally fertilizing guppy (Poecilia reticulata), as in
internally fertilizing tetrapods, E2 induces normal
female-typical behaviors (Liley 1972).
Gonadotropin-releasing hormone (GnRH) appears
to play a role in spawning of female goldfish, because
PGF2α-induced female spawning activity is increased
by both the salmon (sGnRH) and chicken-II (cGnRHII) forms of the peptide, and decreased by a GnRH antagonist (Volkoff and Peter 1999). Unlike female goldfish, GnRH antagonist did not reduce spawning behavior of males. It is unclear which of the three neuronal
GnRH populations in the goldfish brain (cranial nerve
0 or ‘terminal nerve’; TN; preoptic; mid-brain; Kim
et al. 1995) might be involved with female behavior.
However, because female goldfish ovulate and spawn
after axonal transport of GnRH from TN to other brain
areas is blocked by olfactory tract section (Stacey and
Kyle 1983; Kobayashi et al. 1994; Kim et al. 2001),
GnRH of TN origin may not be essential for display
of female behavior. In the dwarf gourami, lesions of
TN GnRH neurons affects male nest-building behavior, but does not impair spawning behavior per se
(Yamamoto et al. 1997). In summary, the brain mechanism(s) underlying PGF2α-induced female-typical
behavior of goldfish is functional in juveniles, appears independent of estrogen, and may be modulated
by GnRH. Similar PGF-mediated mechanisms synchronizing female-typical behavior with ovulation are
expected to be widespread among oviparous fish.
Male behavior. As in tetrapods (Pfaff et al. 2002a),
testicular androgens regulate male-typical behavior
in fishes (Borg 1994). This role for androgens has
been demonstrated by castration and androgen re-
placement in some species (Liley and Stacey 1983;
Borg 1994), but in goldfish and other species where
testicular regeneration makes castration impractical,
androgen effects have been examined in intact males
and females (Borg 1994; Stacey and Kobayashi 1996).
Non-aromatizable 11-oxygenated androgens, such as
11-ketoandrostendione and 11-ketotestosterone (KT),
are more effective than testosterone in inducing maletypical behavior in intact female goldfish (Stacey
and Kobayashi 1996) (Figure 2), castrated stickleback (Gasterosteus aculeatus; Borg 1994) and intact
male bluegill (Lepomis macrochirus; Kindler et al.
1991). It perhaps is not surprising that male-specific
11-oxygenated androgens are more potent than T in
induction of male-typical behavior, because T is produced by both testis and ovary (Borg 1994). However,
treatment with exogenous T can nevertheless induce
male-typical behaviors in female fish (Borg 1994;
Kindler et al. 1991; Stacey and Kobayashi 1996), so
it is not clear why ovulatory females with equivalent,
endogenously generated concentrations of T do not
exhibit male behaviors.
Estrogenic endocrine disruptors in natural habitats
may affect reproduction of male fishes (Kime 1998).
Indeed, E2 administration or exposure suppresses
male-typical sexual behaviors of goldfish (Bjerselius
et al. 2001; Schoenfuss et al. 2002), guppy, Poecilia
reticulata (Bayley et al. 1999), and medaka, Oryzias
latipes (Oshima et al. 2002). E2 seems to directly
inhibit testicular androgen synthesis (Trudeau et al.
1993) without affecting gene expression of pituitary
gonadotropin subunits (M. Kobayashi unpublished).
The common vertebrate paradigm of androgenregulated male-typical behaviors may not apply to
all fish. For example, in the protogynous bluehead
Figure 2. Diagrammatic representation of the hormonal control and plasticity of sex behavior and luteinizing hormone (LH) secretion in
goldfish. PGF (prostaglandin F2α); 11-KT (11-ketotestosterone); T (testosterone); 17,20β-P (water-borne 17,20β-dihydroxy-4-pregnen-3-one).
(Thalassoma bifasciatum; Godwin et al. 1996) and
cleaner wrasses (Labroides dimidiatus; Nakashima
et al. 2000), behavioral sex change can precede gonadal sex change. Indeed, because ovariectomized
bluehead wrasse will perform male spawning behavior
when placed in the social conditions that normally induce female-to-male sex change (Godwin et al. 1996),
gender-typical behaviors in some protogynous hermaphrodites may be influenced more by social factors
than by endocrine milieu.
As seems to be the case in at least some other
teleosts (Liley and Kroon 1995), LH secretion in male
goldfish is strongly stimulated by behavioral interaction with sexually active females (Sorensen et al.
1989; Zheng and Stacey 1997). Unlike the female preovulatory LH surge, this LH secretion in males can be
induced at any time of the day (Sorensen et al. 1989;
Zheng and Stacey 1997; Kobayashi et al. 1997), although the largest and most consistent responses occur
in scotophase (Dulka et al. 1987a). It is not known
what function this male-typical LH increase serves,
or if it is androgen-dependent. In summary, studies
in goldfish indicate that non-aromatizable androgens
such as KT induce the potential for male-typical behaviors, which are triggered by hormonally derived
Pheromonal control of spawning
Pheromones are chemical signals that transmit information between conspecifics and which do not
require overt learning (Sorensen and Stacey 1999).
Most fish including goldfish appear to rely heavily on
pheromones, presumably because of the wealth of information that chemical cues can convey in poorly lit
environments. Fish have evolved to use pheromones
to mediate predator avoidance, aggregation, and reproduction. Reproductive pheromones are best understood in the goldfish which use at least three female cues and one produced by males, all four of
which are comprised of mixtures of hormonal products (Figure 3). Pheromones of the goldfish’s most
immediate relatives, the common and crucian carps,
Figure 3. Model of the goldfish hormonal pheromone system. In vitellogenic females, plasma E2 stimulates urinary release of a recrudescent pheromone that attracts males. In postvitellogenic females, exogenous cues induce an evening surge in LH that increases release of
androstenedione (AD), suppressing potential male response to the small amounts of 17,20β-P released. The 17,20β-P: AD ratio soon increases,
enabling the preovulatory pheromone to increase LH in males. Several hours later, 17,20β-P-S, a urinary metabolite of 17,20β-P dominates the
prevoulatory pheromone mixture, enhancing its behavioral and endocrine effectiveness. Prior to ovulation, LH increase in pheromone-exposed
males has increased both the quantity and quality of releasable sperm. At ovulation, eggs are released into the oviduct were they induce synthesis
of prostaglandin F2α (PGF2α) which acts in the brain to stimulate female sex behavior and is cleared to the water with its principal metabolite
(15K-PGF2α) to function as a postovulatory pheromone that stimulates male sex behavior, LH release and testicular responses.
appear remarkably similar to those of the goldfish.
Here, we discuss the cues released by females in the
order of their release, then the male pheromone, neuroendocrine mediation of pheromone responses, and
Female sex pheromones
The female recrudescent pheromone. Goldfish, common, and crucian carp have no prominent sexual
dimorphisms, typically live in turbid waters and spawn
only a few times a year at somewhat unpredictable
times and sites, factors that may have favored the
evolution of several chemical cues which allow males
to identify and remain close to reproductively mature females. Amongst pre-spawning, maturing goldfish, male attraction appears to be facilitated by a
‘recrudescent ’ pheromone released by vitellogenic fe-
males (previously termed a ‘maturation’ pheromone;
Stacey and Sorensen 2002). This cue was first described in the urine of females with high plasma E2
(Yamazaki and Watanabe 1979; Yamazaki 1990) (Figure 3). Recently, we (P.W. Sorensen, M. Kobayashi
and R. Kihslinger) have confirmed this observation
by showing that the odors of vitellogenic females
and E2-treated ovariectomized females will induce
large, equivalent increases in male physical contact/
inspection behavior. In contrast, the odors of mature males and ovariectomized, untreated females are
without effect on male behavior (see Figure 4 for
details). The chemical identity of the recrudescent
pheromone is not known; however, as measured by
extracellular electrophysiological (electroolfactogram
or ‘EOG’) recording, the goldfish olfactory organ does
not detect E2 or its common conjugates (Sorensen
et al. 1987 and unpublished), suggesting it may be
Figure 4. Behavioral evidence for a recrudescent pheromone in
goldfish (Sorensen et al., unpublished). Groups (N = 10) of five
male goldfish were held in aerated 70 l aquaria overnight and exposed to one of five odor treatments (100 ml/min for 15 min): (1)
Well water control; (2) Spermiated males; (3) Vitellogenic females;
(4) Ovariectomized females that had been implanted (3 months
post-surgery) for 30 days with silicone capsules containing 50 µl
castor oil; (5) Ovariectomized females that had been implanted with
capsules containing 200 ng of E2 dissolved in oil. Conspecific odors
were holding water from pairs of fish held overnight in 2 l of well
water. Male-male nudging (physical contact/inspection, a behavior
typical of courtship; Poling et al. 2001) was observed for the 15 min
immediately preceding (clear bars) and during (black bars) odor
addition. ∗ ; P < 0.05: Wilcoxon matched pairs test.
a novel steroidal metabolite. Further, because E2 induces morphological changes in the goldfish kidney
(Yamazaki and Watanabe 1979) and the pheromone
appears to be in the urine (Yamazaki and Watanabe
1979), it seems likely that some E2-induced specialization of the kidney is responsible for its production.
The preovulatory steroid pheromone. When exposed
to appropriate stimuli (spawning substrate, increased
temperature, pheromones) (Stacey et al. 1979a, b,
1989; Kobayashi et al. 1987), postvitellogenic female
goldfish exhibit a dramatic preovulatory LH surge
that lasts about 15 h, terminating with ovulation and
spawning. This surge induces rapid changes in steroid
synthesis and a resumption of meiosis/oocyte maturation (Patiño et al. 2001). Plasma C19 steroids increase
soon after the onset of the LH surge, decline as C21
steroids (including the oocyte maturation-inducing
steroid, 17,20β-P) increase dramatically during the
middle portion of the surge, and increase again when
C21 steroids decline at its end when fish ovulate and
spawn (Figure 3). Steroid release closely parallels
Figure 5. Water-borne 17,20β-P increases the proportion of female goldfish that ovulate. Postvitellogenic female goldfish were
removed from 12 ◦ C stock tanks, placed as groups of 4 into 70 l
glass aquaria lacking spawning substrate in the early morning. Temperatures of incoming water were then increased to 18 ◦ C by late
afternoon. At 18.00 h either 17,20β-P (50 µg; to create about 1 nM)
(n = 12 groups of 4) or methanol carrier alone (100 µl) (n = 11
groups of 4) was then added to each aquarium and this treatment
was repeated at 20.00 h. The next morning all fish were checked
at 09.00 h to determine if they had ovulated (Sorensen and Stacey
1987; unpublished). ∗∗∗ , P < 0.001; Fisher exact test.
steroid levels in the plasma, with the exception that
conjugated metabolites are also released to the water
(Scott and Sorensen 1994; Sorensen and Scott 1994).
Of the several dozen reproductive steroids released
by goldfish, only AD, 17,20β-P, and its metabolite
17,20β-P-S are detected with high specificity and sensitivity by the goldfish olfactory system (Sorensen
et al. 1990, 1995a). Accordingly, we consider these
steroids to be ‘primary’ components of the preovulatory pheromone and all have activity when tested alone
and in combination with other steroids (see below).
Interestingly, these pheromonal cues are released by
different routes: the free steroids (AD and 17,20β-P)
are released across the gills whereas glucuronidated
and sulfated forms (17,20β-P-S) are released as controlled pulses in urine (Sorensen et al. 2000). In
addition to these primary components, goldfish release
at least a few other sex steroids that bind with moderate affinity to the same olfactory receptors as the
primary components. We term these ’redundant’ components (Sorensen and Stacey 1999) because although
these cues have some biologcial activity, it generally
appears to mirror that of the primary components.
As the mixture of steroids in the preovulatory
pheromone changes during the female’s LH surge,
the pheromone exerts different effects on the male.
Initially, AD released early in the LH surge causes
increases in aggressive interactions between males
(Poling et al. 2001). However, as the surge progresses,
17,20β-P comes to be the dominant component within
about 6 h. Even brief exposure to 17,20β-P, or
to steroid mixtures dominated by 17,20β-P (Stacey
1991), trigger immediate increases in male LH (Dulka
et al. 1987b; Zheng and Stacey 1996, 1997; Sorensen,
unpublished) as well as increased swimming and inspection behavior amongst males (DeFraipont and
Sorensen 1993; Poling et al. 2001). These behavioral
changes are persistent and accompanied by increased
sperm production and motility by the time of ovulation/spawning, 4–8 h later (Dulka et al. 1987b;
DeFraipont and Sorensen 1993). Dramatic increases in
male fertility and spawning success are the final consequence (Zheng et al. 1997). The actions of water-borne
17,20β-P on LH release and sperm production have
been tested in both the common carp (Stacey et al.
1994) and crucian carp (Bjerselius et al. 1995), and
appear identical to those in goldfish. Possible behavioral effects and roles of steroid mixtures have not yet
been examined in these species, however.
Late in the female’s ovulatory LH surge, females
start to release significant quantities of 17,20β-P-S in
their urine. This steroid has somewhat different actions than 17,20β-P, stimulating intense bouts of chasing and following and small increases in LH (Sorensen
et al. 1995a; Poling et al. 2001). Mixture composition appears important to the preovulatory signal with
17,20β-P and 17,20β-P-S synergizing each other’s behavioral activity (P. Sorensen, unpublished). Although
it is now clear that the steroidal pheromone stimulates and synchronizes male reproductive behavior and
physiology with that of the female, its precise composition and actions in the natural environment have yet
to be fully resolved.
In addition to its pheromonal effects on males,
waterborne 17,20β-P appears to act as a ‘female’
pheromone to induce ovulation amongst other female
goldfish, perhaps facilitatory synchrony amongst entire populations. This possibility is suggested both
by the observation the female goldfish olfactory system detects sex steroids as well as males (Sorensen
et al. 1987), and by a single experiment in which
females experienced higher ovulation rates in the absence of spawning substrate when exposed to 17,20βP overnight (Sorensen and Stacey 1987; see Figure 5
for details). This preliminary finding is consistent with
reports of ovulatory synchrony among groups of goldfish held in the laboratory (Kobayashi et al. 1988).
Presumably the effects of the 17,20β-P pheromone on
female LH levels are modulated by other environmental variables in a more complex manner than is the case
with males but this must be studied.
The postovulatory prostaglandin pheromone. At the
time of ovulation, female LH titres decrease dramatically (Stacey et al. 1989), and steroid synthesis
decreases dramatically (Scott and Sorensen 1994). Coincidentally, female goldfish become sexually active
as a result of PGF2α produced in response to ovulated eggs in the oviducts (Stacey 1976; Stacey and
Peter 1979; Sorensen and Goetz 1993; Sorensen et al.
1995b; Sorensen in preparation). Additionally, PGF2α
and its metabolites are rapidly released as a postovulatory pheromone that attracts the male to the female
while triggering male courtship, and activating a variety of unique (e.g., different from the processes
stimulated by the 17,20β-P pheromone) physiological
mechanisms that further increase milt stores (Sorensen
et al. 1988, 1989; Zheng and Stacey 1996, 1997).
So important is the F prostaglandin (PGF) pheromone
for males that they typically fail to spawn if rendered
anosmic (Stacey and Kyle 1983).
The postovulatory PGF pheromone appears to be
a mixture of PGF2α and 15K-PGF2α, with the latter being the primary component. These compounds
are released in great quantity (>50 ng/h) by recentlyovulated goldfish and by nonovulated fish that have
either been injected with PGF2α or had ovulated
eggs placed into their oviducts (Sorensen et al. 1988,
1995b, unpublished). The PGF pheromone may be
a specialized signal because PGFs are released exclusively in urine pulses which coincide with female spawning activity (Appelt and Sorensen 1999).
EOG recording from the goldfish olfactory epithelium shows that PGF2α and 15K-PGF2α are detected
with great sensitivity (nM and pM thresholds, respectively) and specificity by the male olfactory epithelium
which possesses separate receptor mechanisms for
these compounds (Sorensen et al. 1988). Interestingly,
unlike the situation with steroids, the olfactory epithelium of male goldfish and carp is more sensitive
to PGFs than is that of females (Irvine and Sorensen
1993; Sorensen and Goetz 1993), a sexual dimorphism
mediated by androgen (Cardwell et al. 1995; Stacey
et al. in press).
EOG sensitivity and specificity in goldfish is consistent with results of behavioral assays (Sorensen
et al. 1988) in which low concentrations (0.1 nM) of
PGFs in aquarium water induce mature male goldfish to inspect and chase each other (Sorensen et al.
1988, 1989). No difference has yet been noted in the
functions of PGF2α or 15K-PGF2α. Because PGF exposure does not induce LH release in isolated males
(Sorensen et al. 1989), it is thought that the LH
The male sex pheromone
Male goldfish direct most of their behavior towards
females, but also exhibit agonistic behaviors towards
males that likely are associated with sperm competition. Although mature males release little 17,20β-P
or 17,20β-P-S, they release great quantities of androgenic steroids including AD (Sorensen and Scott
1994, unpublished), which increase agonistic behaviors when added to aquaria (Poling et al. 2001). AD
released by males also appears to suppress milt levels
in grouped males (Stacey 1991; Stacey et al. 2001;
Fraser and Stacey 2002). Thus, water-borne AD appears to have several context-dependent functions; it
serves (apparently by itself) as a male pheromone with
both releaser and primer effects on other males, and
as a component of a complex female preovulatory
Endocrine and neuroendocrine mediation of sex
pheromone response
Figure 6. Diagrammatic representation of sexual bipotentiality of
the brain in fish (adapted from Kobayashi et al. 2000). In the rat,
the undifferentiated brain sex is believed to normally be female
but in the presence of androgen and estrogen during the perinatal
period, the brain develops neural systems which regulate male functions and inhibit female functions (crossed out area). Teleost fish,
on the other hand, appear to possess a sexually bipotential brain.
When a protogynous hermaphroditic fish is in the female phase,
the female portion of the brain is active and the male portion is
quiescent (shaded area). At the time the individual starts to behave
like a male, the male portion of the brain is activated and the female
portion becomes quiescent (shaded area): External factors (age, social status, etc.) that regulate the sex change vary among species.
Although gonochoristic and gynogenetic teleosts normally use only
brain areas controlling homotypical behaviors during their lifetime,
brain areas controlling heterotypical behaviors can be activated by
hormonal treatments.
increases measured in males actually result from PGFinduced socio-sexual interactions (Sorensen et al.
1989; Zheng and Stacey, 1996, 1997). Thus, unlike
endocrine response to 17,20β-P, which can occur in
isolated males (Fraser and Stacey 2002), the behavioral and endocrine responses exhibited to PGFs are
highly dependent on the social context in which the
cue is encountered.
The physiological mechanisms that regulate pheromone responsiveness have been examined only in
males, where they appear to be androgen-dependent
secondary sex characters. For example, androgentreated female goldfish and crucian carp exhibit maletypical LH increase when exposed to pheromonal
17,20β-P (Kobayashi et al. 1997; Kobayashi and
Nakanishi 1999). Masculinizing effects are seen both
at the peripheral and central levels of the nervous
system. Evidence for peripheral effects comes from
EOG studies of several cyprinids including goldfish
(Sorensen and Goetz 1993) showing that the olfactory epithelium of males is more sensitive to PGFs
than is that of females, and that androgen treatment
increases EOG responses to PGFs (Cardwell et al.
1995; Sorensen, unpublished; Stacey et al. in press).
On the other hand, indirect evidence for central effects
comes from the round goby (Neogobius melanostomus), where steroid pheromones induce sexually
isomorphic EOG responses but dimorphic, androgendependent behavioral responses (Murphy et al. 2001;
Murphy and Stacey 2002).
Neuroendocrine mediation of pheromone action
in goldfish has been examined for male-typical LH
responses to pheromonal 17,20β-P and to spawning
stimuli (stimuli received as a consequence of spawning). Pheromonal 17,20β-P appears to increase LH
and releasable sperm through a pituitary-dependent
mechanism involving tonic dopamine inhibition of the
gonadotropes (Dulka et al. 1992; Zheng and Stacey
1997). In contrast, spawning stimuli appear to affect
sperm through two mechanisms (Zheng and Stacey
1996, 1997): (1) an unidentified LH release mechanism that appears not to involve dopamine, and (2) a
more rapid, pituitary-independent mechanism. Thus,
injection of a GnRH antagonist does not affect the
rapid increases in sperm supply to spawning stimuli, but does block LH responses to both pheromonal
17,20β-P and spawning stimuli (Zheng and Stacey
1997); however, it is not known whether tonic GnRH
secretion plays an essentially permissive role, or
whether 17,20β-P and spawning stimuli increase LH
by increasing GnRH release.
Species specificity
The issue of species specificity is of particular relevance to hormonal pheromones, which are widely
used by fishes yet appear to have limited chemical
diversity (Stacey and Sorensen 2002). Although tests
of species-specificity are few and the results varied,
some behavioral studies using conspecific and heterospecific holding water (i.e., not synthetic hormonal
products) of related species suggest sex pheromones
can be species-specific (Sorensen and Stacey 1999).
Although future studies of this important topic should
target undomesticated species, the best available data
are from goldfish, where the pheromonal components are commonly produced and detected by many
fish (Sorensen and Stacey 1999; Stacey and Sorensen
2002). Therefore, pheromone specificity, if it exists in
goldfish, is expected to reside in: (1) the use of/or discrimination of specific chemical mixtures, which may
include non-hormonal components, and (2) the use of
controlled pheromone release that determines the location of the signal’s active space (Sorensen et al. 2000).
Future studies of specificity might be most profitable
in species such as the gobies, which appear to employ
more specialized signals, perhaps because of female
sexual selection (Colombo et al. 1982; Murphy et al.
2001), a situation that may facilitate the evolution of
species specificity (Sorensen and Stacey 1999).
Sexual plasticity of spawning behavior
Plasticity of behavior and LH secretion
In mammals and birds, the potential for gender-typical
gonadotropin secretion and behavior typically is determined by early organizational actions of steroids
that induce sexual differentiation of the brain. Later,
hormones may exert activational effects direcly on the
adult brain (Pfaff et al. 2002b). Although gonochoristic (separate gender) teleost fish exhibit gender-typical
reproductive functions and normally do not exhibit
heterotypical patterns, heterotypical patterns of behavior and LH secretion can be induced in some species
including the adult goldfish by hormone treatment.
For example, PGF2α injection induces female spawning behavior within minutes in male goldfish and at a
rate similar to that seen in females (Stacey and Kyle
1983; Figure 2) but does not inhibit performance of
male-typical behaviors. Thus, PGF2α-injected males
placed with both PGF2α-injected females and mature males alternate between engaging in male- and
female-typical behaviors, the same pattern seen in
simultaneous hermaphrodites (e.g., Serranus subligarius; Cheek et al. 2000). Similarly, androgen implants
induce male-typical behaviors in mature females without reducing their capacity to exhibit PGF-induced
female-typical behaviors (Stacey and Kobayashi 1996,
Figure 2). Interestingly, such androgen treatments prevent neither ovarian development nor ovulatory LH
surges (Stacey and Kobayashi 1996; Kobayashi et al.
1997). Finally, although female goldfish do not usually exhibit a male-typical LH surge in response to
pheromonal 17,20β-P (at least outside the normal
time for the preovulatory LH surge), they do if implanted with KT (Kobayashi et al. 1997, Figure 2).
It is not yet known if androgen-treated females also
exhibit the male-typical LH surge induced by spawning stimuli, or if hormone treatment can induce a
female-typical ovulatory LH surge in males. Potential for heterotypical reproductive function also occurs
in adult gynogenetic crucian carp, Carassius auratus langsdorfii, where KT treatment induces maletypical behaviors and 17,20β-P-induced LH secretion
in females without inhibiting PGF2α-induced female
behaviors (Kobayashi and Nakanishi 1999).
Adult goldfish and gynogenetic crucian carp exhibit heterotypical reproductive functions in response
to steroid and PGF2α treatments, and maintain homotypical function during these treatments. This sexual
plasticity extends to the gonadal level: i.e., androgen
treatment can induce development of testicular tissue within the ovary (Kobayashi et al. 1991). Such
bipotentiality is comparable to that of hermaphroditic
fishes (Strüssmann and Nakamura 2003).
Gonochoristic teleosts other than goldfish and crucian carp also exhibit bipotentiality as adults. For
instance, androgen treatment induces male-typical behavior in adult female stickleback (Gasterosteus ac-
uleatus; Wai and Hoar 1963), guppy (Landsman et al.
1987), and gobies (Murphy and Stacey 2002), and exposure to paper mill effluents containing androgenic
chemicals causes female mosquitofish, Gambusia affinis holbrooki, to display male sex behavior (Howell
et al. 1980). Adult teleosts thus appear to retain
relatively more bisexual potential than mammals (Yamanouchi 1997; Pfaff et al., 2002b), where normal
processes of male development typically involve loss
of female-typical potential.
in fish, but rather as the opportunistic expression of
latent behavioral, endocrine and gonadal bisexuality in
ecological situations where hermaphroditism is more
adaptive than gonochorism. From this perspective, the
interesting questions pertain not to adult teleost bisexuality but to the evolutionary and physiological
mechanisms by which this bisexuality has been lost
in other vertebrate groups.
Sexual bipotentiality in the fish brain
Although the neuroanatomical basis for the sexual
bipotentiality of teleost behavior is unclear, studies of
hermaphroditic fish suggest that brain GnRH and arginine vasotocin (AVT) are involved. After sex change,
the number and size of GnRH-immunoreactive neurons change in bluehead wrasse (Grober et al. 1991),
ballan wrasse, Labrus berggylta (Elofsson et al. 1999),
and dusky anemone fish, Amphiprion melanopus
(Elofsson et al. 1997). Similar changes are seen in
the number and size of AVT neurons in bluehead
wrasse (Godwin et al. 2000) and marine goby, Trimma
okinawae (Grober and Sunobe 1996) at the time of
sex change. It is unknown, however, whether these
peptidergic cell changes are the cause or result of
behavioral sex change. When female goldfish were
implanted with KT and induced to perform male sex
behavior, no change was observed in GnRH- and
AVT-immunoreactive cells (Parhar et al. 2001).
The mechanisms mediating display of heterotypical behaviors are not known, but short response latencies suggest activation of preexisting mechanisms
rather than development of new ones (Figure 6). In
protogynous hermaphrodites, for example, display
of heterotypical behavior occurs rapidly in response
to social cues (1–2 h in cleaner wrasse, Nakashima
et al. 2000; 1–2 days in bluehead wrasse, Godwin
et al. 1996), and in the gonochoristic goldfish, KT
can induce male-typical behavior in females within 3
days (Kobayashi, unpublished data) whereas PGF2α
induces female behavior in males within minutes
(Stacey and Kyle 1983).
Although examples of adult sexual bipotentiality
in teleost behavior admittedly are limited, we suggest bisexuality is typical of teleosts regardless of their
gender system (hermaphroditism, gonochorism, gynogenesis). From this perspective, fish hermaphroditism
could be viewed not as the result of specializations that
depart dramatically from normal reproductive function
Norm Stacey was supported the Natural Sciences and
Engineering Council of Canada, Peter Sorensen by
the National Science Foundation (IBN9723798) and
NIH (DCO3792-01A1), and Makito Kobayashi by the
Ministry of Education, Culture, Sports, Science, and
Appelt, C.W. and Sorensen, P.W. 1999. Freshwater fish release
urinary pheromones in a pulsatile manner. In: Advances in
Chemical Signals in Vertebrates. pp. 247–256. Edited by R.E.
Johnston, D. Muller-Schwarze and P.W. Sorensen. Kluwer Academic/Plenum Publishers, New York.
Barry, T.P., Santos, A.T.G., Furukawa, K., Aida, K. and Hanyu, I.
1990. Steroid profiles during spawning in male common carp.
Gen. Comp. Endocrinol. 80: 223–231.
Bayley, M., Nielsen, J.R. and Baatrup, E. 1999. Guppy sexual behavior as an effect biomarker of estrogen mimics. Ecotoxicol.
Environ. Safety 43: 68–73.
Bjerselius, R., Olsen, K.H. and Zheng, W. 1995. Endocrine, gonadal
and behavioral responses of male crucian carp to the hormonal
pheromone 17α,20β-dihydroxy-4-pregnen-3-one. Chem. Senses
20: 221–230.
Bjerselius, R., Lundstedt-Enkel, K., Olsen, H., Mayer, I. and
Dimberg, K. 2001. Male goldfish reproductive behavior and
physiology are severely affected by exogenous exposure to
17β-estradiol. Aquatic Toxicol. 53: 139–152.
Blaustein, J.D. and Erskine, M.S. 2002. Feminine sexual behavior:
Cellular integration of hormonal and afferent information in the
rodent brain. In: Hormones, Brain and Behavior. pp. 139–214.
Edited by D.W. Pfaff, A.P. Arnold, A. Etgen, S. Fahrbach and R.
Rubin. Academic Press, New York.
Borg, B. 1994. Androgens in teleost fishes. Comp. Biochem.
Physiol. 109C: 219–245.
Cardwell, J.R., Stacey, N.E., Tan, E.S.P., McAdam, D.S.O. and
Lang, S.L.C. 1995. Androgen increases olfactory receptor response to a vertebrate sex pheromone. J. Comp. Physiol. A 176:
Chang, J.P. and Peter, R.E. 1983. Effects of dopamine on gonadotropin release in female goldfish, Carassius auratus. Neuroendocrinology 36: 351–357.
Cheek, A.O., Thomas, P. and Sullivan, C.V. 2000. Sex steroids
relative to alternative mating behaviors in the simultaneous
hermaphrodite Serranus subligarius (Perciformes: Serranidae).
Horm. Behav. 37: 198–211.
Cole, K.S. and Stacey, N.E. 1984. Prostaglandin induction of
spawning behavior in Cichlasoma bimaculatum (Pisces Cichlidae). Horm. Behav. 18: 235–248.
Colombo, L., Belvedere, P.C., Marconato, A. and Bentivegna, F.
1982. Endocrinology of teleost reproduction. In: Proceedings of
the Second International Symposium on the Reproductive Physiology of Fish. pp. 84–94. Edited by C.J.J. Richter and H.J.Th.
Goos. Pudoc, The Netherlands.
DeFraipont, M. and Sorensen, P.W. 1993. Exposure to the
pheromone 17α, 20β-dihydroxy-4-pregnen-3-one enhances the
behavioral spawning success, sperm production, and sperm
motility of male goldfish. Anim. Behav. 46: 245–256.
Dulka, J.G., Sorensen, P.W. and Stacey, N.E. 1987a. Sociallystimulated gonadotropin release in male goldfish: differential
circadian sensitivities to a steroid pheromone and spawning stimuli. In: Proceedings of the Third International Symposium on
the Reproductive Physiology of Fish. p. 160. Edited by D.R.
Idler, L.W. Crim and J.M. Walsh. Memorial University Press,
St. John’s, Newfoundland.
Dulka, J.G., Stacey, N.E., Sorensen, P.W. and Van Der Kraak,
G.J. 1987b. A sex steroid pheromone synchronizes male-female
spawning readiness in goldfish. Nature 325: 251–253.
Dulka, J.G., Sloley, B.D., Stacey, N.E. and Peter, R.E. 1992.
A reduction in pituitary dopamine turnover is associated with
sex pheromone-induced gonadotropin increase in goldfish. Gen.
Comp. Endocrinol. 86: 496–505.
Elofsson, U., Winberg, S. and Francis, R.C. 1997. Number of preoptic GnRH-immunoreactive cells correlates with sexual phase
in a protandrously hermaphroditic fish, the dusky anemone fish
(Amphiprion melanopus). J. Comp. Physiol. A 181: 484–492.
Elofsson, U., Winberg, S. and Nilsson, G.E. 1999. Relationships
between sex and the size and number of forebrain gonadotropinreleasing homone-immunoreactive neurons in the ballan wrasse
(Labrus berggylta), a protogynous hermaphrodite. J. Comp.
Neurol. 410: 158–170.
Fraser, E.J. and Stacey, N.E. 2002. Isolation increases milt production in goldfish. J. Exp. Zool. 293: 511–524.
Godwin, J., Crews, D. and Warner, R.R. 1996. Behavioural sex
change in the absence of gonads in a coral reef fish. Proc. Roy.
Soc. Lond. B. 263: 1683–1688.
Godwin, J., Sawby, R., Warner, R.R., Crews, D. and Grober, M.S.
2000. Hypothalamic arginine vasotocin mRNA abundance variation across sexes and with sex change in a coral reef fish. Brain
Behav. Evol. 55: 77–84.
Grober, M.S., and Sunobe, T. 1996. Serial adult sex change involves rapid and reversible chnages in brain neuro-chemistry.
Neuroreport 7: 2945–2949.
Grober, M.S., Jackson, I.M.D. and Bass, A.H. 1991. Gonadal
steroids affect LHRH preoptic cell number in a sex/role changing
fish. J. Neurobiol. 22: 734–741.
Howell, W.M., Black, D.A. and Bortone, S.A. 1980. Abnormal expression of secondary sex characters in a population
of mosquitofish, Gambusia affinis holbrooki: evidence for
environmentally-induced masculinization. Copeia 1980: 676–
Irvine, I.A.S. and Sorensen, P.W. 1993. Acute olfactory sensitivity of wild common carp, Cyprinus carpio, to goldfish sex
pheromones is influenced by gonadal maturity. Can. J. Zool. 71:
Kagawa, H., Young, G. and Nagahama, Y. 1984. In vitro 17βestradiol and testosterone production by ovarian follicles of
the goldfish, Carassius auratus. Gen. Comp. Endocrinol. 54:
Kalra, S.P. and Kalra, P.S. 1997. Interactive hypothalamic neural circuitry in the control of reproduction: a bird’s eye view. In: Neural
Control of Reproduction. pp. 3–16. Edited by K.-I. Maeda, H.
Tsukamura and A. Yokoyama. Japan Sci. Soc. Press, Tokyo.
Kezuka, H., Kobayashi, M., Aida, K. and Hanyu, I. 1989. Effects of photoperiod and pinealectomy on the gonadotropin surge
and ovulation in goldfish Carassius auratus. Nippon Suisan
Gakkaishi 55: 2099–2103.
Kim M.H., Oka, Y., Amano, M., Kobayashi, M., Okuzawa, K.,
Hasegawa, Y., Kawashima, S., Suzuki, Y. and Aida, K. 1995.
Immunohistochemical localization of sGnRH and cGnRH-II in
the brain of goldfish, Carassius auratus. J. Comp. Neurol. 356:
Kim, M.H., Kobayashi, M., Oka, Y., Amano, M., Kawashima, S.
and Aida, K. 2001. Effects of olfactory tract section on the immunohistochemical distribution of the brain GnRH fibers in the
female goldfish, Carassius auratus. Zool. Sci. 18: 241–248.
Kime, D.E. 1998. Endocrine disruption in fish. Kluwer Academic
Publishers, Boston.
Kindler, P.M., Bahr, J.M. and Philipp, D.P. 1991. The effects of
exogenous 11-ketotestosterone, testosterone, and cyproterone
acetate on prespawning and parental care behaviors of male
bluegill. Horm. Behav. 25: 410–423.
Kitamura, S., Ogata, H. and Takashima, F. 1994. Activities of Ftype prostaglandins as releaser sex pheromones in cobitide loach,
Misgurnus anguillicaudatus. Comp. Biochem. Physiol. 107A:
Kobayashi, M. and Stacey, N. E. 1993. Prostaglandin-induced female spawning behavior in goldfish (Carassius auratus) appears
independent of ovarian influence. Horm. Behav. 27: 38–55.
Kobayashi, M. and Nakanishi, T. 1999. 11-Ketotestosterone induces
male-type sexual behavior and gonadotropin secretion in gynogenetic crucian carp, Carassius auratus langsdorfii. Gen. Comp.
Endocrinol. 115: 178–187.
Kobayashi, M., Aida, K. and Hanyu, I. 1986a. Gonadotropin surge
during spawning in male goldfish. Gen. Comp. Endocrinol. 62:
Kobayashi, M., Aida, K. and Hanyu, I. 1986b. Pheromone from
ovulatory female goldfish induces gonadotropin surge in males.
Gen. Comp. Endocrinol. 63: 451–455.
Kobayashi, M., Aida, K. and Hanyu, I. 1986c Annual changes in
plasma levels of gonadotropin and steroid hormones in goldfish.
Bull. Japan. Soc. Sci. Fish. 52: 1153–1158.
Kobayashi, M., Aida, K. and Hanyu, I. 1987. Hormone changes
during ovulation and effects of steroid hormones on plasma
gonadotropin levels and ovulation in goldfish. Gen. Comp. Endocrinol. 67: 24–32.
Kobayashi, M., Aida, K. and Hanyu, I. 1988. Hormone changes
during the ovulatory cycle in goldfish. Gen. Comp. Endocrinol.
69: 301–307.
Kobayashi, M., Aida, K. and Hanyu, I. 1989a. Involvement of
steroid hormones in the preovulatory gonadotropin surge in
female goldfish. Fish Physiol. Biochem. 7: 141–146.
Kobayashi, M., Aida, K. and Hanyu, I. 1989b. Induction of gonadotropin surge by steroid hormone implantation in ovariectomized and sexually regressed female goldfish. Gen. Comp.
Endocrinol. 73: 469–476.
Kobayashi, M., Aida, K. and Stacey, N. E. 1991. Induction of testis
development by implantation of 11-ketotestosterone in female
goldfish. Zool. Sci. 8: 389–393.
Kobayashi, M., Amano, M., Kim, M. H., Furukawa, K., Hasegawa,
Y. and Aida, K. 1994. Gonadotropin-releasing hormones of terminal nerve origin are not essential to ovarian development and
ovulation in goldfish. Gen. Comp. Endocrinol. 95: 192–200.
Kobayashi, M., Furukawa, K., Kim, M. H. and Aida, K. 1997. Induction of male-type gonadotropin secretion by implantation of
11-ketotestosterone in female goldfish. Gen. Comp. Endocrinol.
108: 434–445.
Kobayashi, M., Stacey, N.E., Aida, K. and Watabe, S. 2000. Sexual plasticity of behavior and gonadotropin secretion in goldfish
and gynogenetic crucian carp. In: Proceedings of the Sixth International Symposium on the Reproductive Physiology of Fish.
pp. 117–125. Edited by B. Norberg, O.S. Kjesbu, G.L. Taranger,
E. Andersson and S.O. Stefansson. Institute of Marine Research
and University of Bergen, Bergen.
Landsman, R.E., David, L.A. and Drew, B. 1987. Effects of 17αmethyltestosterone and mate size on sexual behavior in Poecilia
reticulata. In: Proceedings of the Third International Symposium on the Reproductive Physiology of Fish. p. 133. Edited by
D.R. Idler, L.W. Crim and J.M. Walsh. Memorial University of
Newfoundland, St. John’s.
Liley, N.R. 1972. The effects of estrogens and other steroids on the
sexual behavior of the female guppy, Poecilia reticulata. Gen.
Comp. Endocrinol. Suppl. 3: 542 552.
Liley, N.R. and Stacey, N.E. 1983. Hormones, pheromones and
reproductive behaviour. In: Fish Physiology. Volume 9B. pp. 1–
63. Edited by W.S. Hoar, D.J. Randall, and E.M. Donaldson.
Academic Press, New York.
Liley, N.R. and Tan, E.S.P. 1985. The induction of spawning
behavior in Puntius gonionotus (Bleeker) by treatment with
prostaglandin PGF2α. J. Fish Biol. 26: 491–502.
Liley, N.R. and Kroon, F.J. 1995. Male dominance, plasma hormone
concentrations, and availability of milt in male rainbow trout
(Oncorhynchus mykiss). Can. J. Zool. 73: 826 –836.
Murphy, C.A. and Stacey, N.E. 2002. Methyl-testosterone induces male-typical ventilatory behavior in response to putative steroidal pheromones in female round gobies (Neogobius
melanostomus). Horm. Behav. 42: 109–115.
Murphy, C.A., Stacey, N., and Corkum, L.D. 2001. Putative
steroidal pheromones in the round goby, Neogobius melanostomus: olfactory and behavioral responses. J. Chem. Ecol. 27:
Nakashima, Y., Sakai, Y., Karino, K. and Kuwamura, T. 2000.
Female-female spawning and sex change in a haremic coral-reef
fish, Labroides dimidiatus. Zool. Sci. 17: 967–970.
Oshima, Y., Kang, I.J., Kobayashi, M., Nakayama, K., Imada, N.
and Honjo, T. 2002. Suppression of sexual behavior in male
Japanese medaka (Oryzias latipes) exposed to 17β-estradiol.
Chemosphere 50: 429–436.
Parhar, I., Tosaki, H., Sakuma, Y. and Kobayashi, M. 2001. Sex
differences in the brain of goldfish: gonadotropin-releasing hormones and vasotocinergic neurons. Neurosci. 104: 1099–1110.
Patiño, R., Yoshizaki, G., Thomas, P., and Kagawa, H. 2001.
Gonadotropic control of ovarian follicle maturation: the twostage concept and mechanisms. Comp. Biochem. Physiol. B.
129: 427–439.
Pfaff, D.W., Arnold, A.P., Etgen, A., Fahrbach, S. and Rubin, R.
2002a. Series editors. Hormones, Brain and Behavior, Volume 2.
Academic Press, New York.
Pfaff, D.W., Arnold, A.P., Etgen, A., Fahrbach, S. and Rubin, R.
2002b. Series editors. Hormones, Brain and Behavior, Volume 4.
Academic Press, New York.
Poling, K.R., Fraser, E.J. and Sorensen, P.W. 2001. The three
steroidal components of the goldfish preovulatory pheromone
signal evoke different behaviors in males. Comp. Biochem.
Physiol. B. 129: 645–651.
Santos, A.J.G., Furukawa, K., Kobayashi, M., Bando, K., Aida,
K. and Hanyu, I. 1986. Plasma gonadotropin and steroid hor-
mone profiles during ovulation in the carp Cyprinus carpio. Bull.
Japan. Soc. Sci. Fish. 52: 1159–1166.
Schoenfuss, H.L., Levitt, J.T., Van Der Kraak, G. and Sorensen,
P.W. 2002. Ten week exposure to treated sewage discharge has
relatively minor, variable effects on reproductive behavior and
sperm production in goldfish. Environ. Toxicol. Chem. 21: 2185–
Scott, A.P. and Sorensen, P.W. 1994. Time course of release of
pheromonally active steroids and their conjugates by ovulatory
goldfish. Gen. Comp. Endocrinol. 96: 309–323.
Sokolowska, M., Peter, R.E., Nahorniak, C.S., Pan, C.H., Chang,
J.P., Crim, L.W. and Weil. C. 1984. Induction of ovulation in
goldfish, Carassius auratus, by pimozide and analogues of LHRH. Aquaculture 36: 71–83.
Sorensen, P.W. and Stacey, N.E. 1987. 17α,20β-dihdroxy-4pregnen-3-one functions as a bisexual priming pheromone in
goldfish. Am. Zool. 27: 412.
Sorensen, P.W. and Goetz, F.W. 1993. Pheromonal and reproductive
function of F-prostaglandins and their metabolites in teleost fish.
J. Lipid Mediators 6: 385–393.
Sorensen, P.W. and Scott, A.P. 1994. The evolution of hormonal sex
pheromones in teleost fish: poor correlation between the pattern
of steroid release by goldfish and olfactory sensitivity suggests
that these cues evolved as a result of chemical spying rather than
signal specialization. Acta Scand. Physiol. 152: 191–205.
Sorensen, P.W. and Stacey, N.E. 1999. Evolution and specialization
of fish hormonal pheromones. In: Advances in Chemical Signals
in Vertebrates. pp. 15–47. Edited by R.E. Johnston, D.MullerSchwarze and P.W. Sorensen. Kluwer Academic/Plenum Publishers, New York.
Sorensen, P.W., Hara, T.J. and Stacey, N.E. 1987. Extreme olfactory
sensitivity of mature and gonadally-regressed goldfish to a potent steroidal pheromone, 17α,20β-dihydroxy-4-pregnen-3-one.
J. Comp. Physiol. A 160: 305–313.
Sorensen, P.W. Stacey, N.E. and Chamberlain, K.J. 1989. Differing behavioral and endocrinological effects of two female sex
pheromones on male goldfish. Horm. Behav. 23: 317–332.
Sorensen, P.W., Christensen, T.A. and Stacey, N.E. 1998. Discrimination of pheromonal cues in fish: emerging parallels with
insects. Curr. Opin. Neurobiol. 8: 458–467.
Sorensen, P.W., Scott, A.P., and Kihslinger, R.L. 2000. How
common hormonal metabolites function as relatively specific
pheromonal signals in the goldfish. In: Proceedings of the Sixth
International Symposium on the Reproductive Physiology of
Fish. pp. 125–128. Edited by B. Norberg, O.S. Kjesbu, G.L.
Taranger, E. Andersson and S.O. Stefansson. Institute of Marine
Research and University of Bergen, Bergen.
Sorensen, P.W., Hara, T.J., Stacey, N.E. and Goetz, F.W. 1988. F
prostaglandins function as potent olfactory stimulants that comprise the postovulatory female sex pheromone in goldfish. Biol.
Reprod. 39: 1039–1050.
Sorensen, P.W., Hara, T.J., Stacey, N.E. and Dulka, J.G. 1990. Extreme olfactory specificity of the male goldfish to the preovulatory steroidal pheromone 17α,20β-dihydroxy-4-pregnen-3-one.
J. Comp. Physiol. A. 166: 373–383.
Sorensen, P.W., Scott, A.P., Stacey, N.E. and Bowdin, L. 1995a.
Sulfated 17α,20β-dihydroxy-4-pregnen-3-one functions as a potent and specific olfactory stimulant with pheromonal actions
in the goldfish. General and Comparative Endocrinology 100:
Sorensen, P.W., Brash, A.R., Goetz, F.W., Kellner, R.G., Bowdin,
L. and Vrieze, L. A. 1995b. Origins and functions of F
prostaglandins as hormones and pheromones in the goldfish.
In: Proceedings of the Fourth International Symposium on the
Reproductive Physiology of Fish. pp. 252–254. Edited by F.W.
Goetz and P. Thomas. FishSymp 95, Austin, Texas, U.S.A.
Stacey, N.E. 1976. Effects of indomethacin and prostaglandins on
the spawning behavior of female goldfish. Prostaglandins 12:
Stacey, 1987. Roles of hormones and pheromones in fish reproductive behavior. In: Psychobiology of Reproductive Behavior,
an Evolutionary Perspective. pp. 28–60. Edited by D. Crews.
Prentice-Hall, Englewood Cliffs, NJ, U.S.A.
Stacey, N.E. 1991. Hormonal pheromones in fish: status and
prospects. In: Proceedings of the Fourth International Symposium on the Reproductive Physiology of Fish. pp. 177–181.
Edited by A.P. Scott, J.P. Sumpter, D.S. Kime and M.S. Rolfe.
FishSymp 91, Sheffield.
Stacey, N.E. and Liley, N.R. 1974. Regulation of spawning behavior
in the female goldfish. Nature 247: 71–72.
Stacey, N.E. and Peter, R.E. 1979. Central action of prostaglandins
in spawning behavior of female goldfish. Physiol. Behav. 22:
Stacey, N.E. and Kyle, A.L. 1983. Effects of olfactory tract lesions
on sexual and feeding behavior in the goldfish. Physiol. Behav.
30: 621–628.
Stacey, N.E. and Kobayashi, M. 1996. Androgen induction of male
sexual behaviors in female goldfish. Horm. Behav. 30: 434–445.
Stacey, N.E. and Sorensen, P.W. 2002. Fish hormonal pheromones.
In: Hormones, Brain, and Behavior. Volume 2. pp. 375–435.
Edited by D.W. Pfaff, A.P. Arnold, A. Etgen, S. Fahrbach and
R. Rubin. Academic Press, New York.
Stacey, N.E., Cook, A.F. and Peter, R.E. 1979a. Ovulatory surge
of gonadotropin in the goldfish Carassius auratus. Gen. Comp.
Endocrinol. 37: 246–249.
Stacey, N.E., Cook, A.F. and Peter, R.E. 1979b. Spontaneous
and gonadotropin induced ovulation in the goldfish, Carassius
auratus. L.: Effects of external factors. J. Fish Biol. 15: 305–316.
Stacey, N.E., Zheng, W.B. and Cardwell, J.R. 1994. Milt production
in common carp (Cyprinus carpio): stimulation by a goldfish
steroid pheromone. Aquaculture 127: 265–276.
Stacey, N.E., Sorensen, P.W., Van Der Kraak, G.J., and Dulka, J.G.
1989. Direct evidence that 17α,20β-dihydroxy-4-pregnen-3-one
functions as a goldfish primer pheromone: Preovulatory release
is closely associated with male endocrine responses. Gen. Comp.
Endocr. 75: 62–70.
Stacey, N.E., Fraser, E.J., Sorensen, P.W. and Van Der Kraak, G.J.
2001. Milt production in goldfish: regulation by multiple social
stimuli. Comp. Biochem. Physiol. C 130: 467–476.
Stacey, N.E., Chojnacki, A., Narayanan, A., Cole, T.B. and Murphy,
C.A. Hormonally-derived sex pheromones in fish: exogenous
cues and signals from gonad to brain. Can. J. Physiol. Pharmacol.
(in press).
Strüssmann, C.A. and Nakamura, M. 2003. Morphology, endocrinology, and environmental modulation of gonadal sex
differentiation in teleost fishes. Fish Physiol. Biochem., in press.
Taylor, J. and Mahon, R. 1977. Hybridization of Cyprinus carpio
and Carassius auratus, the first two exotic species in the lower
Laurentian Great Lakes. Environ. Biol. Fishes 1: 205–208.
Trudeau, V.L., Wade, W.G., Van Der Kraak, G. and Peter, R.E. 1993.
Effects of 17β-estradiol on pituitary and testicular function in
male goldfish. Can J. Zool. 71: 1131–1135.
Van Der Kraak, G., Chang, J.P. and Janz, D.M. 1998. Reproduction. In: The Physiology of Fishes, Second Edition. pp. 465–488.
Edited by D.H. Evans. CRC Press, New York.
Villars, T.R. and Burdick, M. 1986. Rapid decline in the behavioral response of paradise fish (Macropodes opercularis)
to prostaglandin F2α treatment. Gen. Comp. Endocrinol. 63:
Volkoff, H. and Peter, R.E. 1999. Actions of two forms of gonadotropin releasing hormone and a GnRH antagonist on spawning behavior of the goldfish Carassius auratus. Gen. Comp.
Endocrinol. 116: 347–355.
Wai, E.H. and Hoar, W.S. 1963. The secondary sex characters and
reproductive behavior of gonadectomized sticklebacks treated
with methyl testosterone. Can. J. Zool. 41: 611–628.
Yamamoto, N., Oka, Y. and Kawashima, S. 1997. Lesions of
gonadotropin-releasing hormone-immunoreactive terminal nerve
cells: Effects on the reproductive behavior of male dwarf
gouramis. Neuroendocrinol. 65: 403–412.
Yamanouchi, K. 1997. Brain mechanisms inhibiting the expression
of heterotypical sexual behavior in rats. In: Neural Control of Reproduction. pp. 291–235. Edited by K.-I. Maeda, H. Tsukamura
and A. Yokoyama. Japan Sci. Soc. Press, Tokyo.
Yamazaki, F. 1965. Endocrinological studies on the reproduction
of the female goldfish, Carassius auratus L., with special reference to the function of the pituitary gland. Mem. Fac. Fisher.
Hokkaido Univ. 13: 1–64.
Yamazaki, F. 1990. The role of urine in sex discrimination in the
goldfish Carassius auratus. Bull. Fac. Fish. Hokkaido Univ. 41:
Yamazaki, F. and Watanabe, K. 1979. The role of sex hormones
in sex recognition during spawning behavior of the goldfish,
Carassius auratus L. Proc. Indian Natl. Sci. Acad. Part B 45:
Yoshiura, Y., Kobayashi, M., Kato, Y. and Aida, K. 1997. Molecular
cloning of the cDNAs encoding two gonadotropin β subunits
(GTH Iβ and IIβ) from the goldfish. Gen. Comp. Endocrinol.
105: 379–389.
Zheng, W. and Stacey, N.E. 1996. Two mechanisms for increasing
milt volume in male goldfish. J. Exp. Zool. 276: 287–295.
Zheng, W. and Stacey, N.E. 1997. A steroidal pheromone and
spawning stimuli act via different neuroendocrine mechanisms
to increase gonadotropin and milt volume in male goldfish
(Carassius auratus). Gen. Comp. Endocrinol. 105: 228–235.
Zheng, W., Strobeck, C. and Stacey, N.E. 1997. The steroid
pheromone 4-pregnen-17α,20β-diol-3-one increases fertility and
paternity in goldfish. J. Exp. Biol. 200: 2833–2840.