232

232 GEORGE A. HUNTER

GEORGE A. AND EDWARD
HUNTER AND M.. DONALDSON
EDWARD M DONALDSON

primary
primary role
role in
in the
the process
process ofof sex
sex differentiation.
differentiation. The
The demonstration
demonstration H-YH-Y
antigen
antigen expression
expression inin the
the ovaries
ovaries of
of homozygous
homozygous chickschicks and
and amphibians
amphibians sex
sex
inverted
inverted with
with estradiol
estradiol suggests
suggests aa major
major rolerole of
of the
the antigen
antigen in
in the
the process
process of
of
hormonal
hormonal sex
sex inversion
inversion inin these
these groups.
groups. Examination
Examination of of the
the role
role of
of the
the
antigen
antigen system
system inin the natural or
the natural or hormonally
hormonally induced sex inversion
induced sex inversion of
of both
both
hermaphroditic
hermaphroditic andand gonochorist
gonochorist fish
fish is
is aa promising
promising area
area of
of study.
study.

2. THE
2. THECORTICOMEDULLARY
CORTICOMEDULLARYINDUCTOR
INDUCTORMODELS
MODELS
The corticomedullary inductor model was proposed by Witschi (1929) (1929)to
explain sex differentiation in amphibians. Witschi observed that the primor­
primor-
dium of the amphibian gonad, like those of of most vertebrates, is comprised of of
both
both an
an outer
outer cortex
cortex and
and inner medulla
medulla ultimately
ultimately derived
derived from
from the
the germinal
germinal
epithelium. During differentiation, either the cortex or the medulla devel­ devel-
ops at the expense of the other resulting in the development of an ovary or
testis, respectively. Witschi theorized
theorized that the genetic male and female
factors
factors embodied
embodied byby the balance
balance theory
theory of of sex
sex determination
determination were
were phe­
phe-
notypically manifested in the dualistic character of the primordial gonad.
The action of these genetic factors results in the production of an embryonic
cortexin or medullarin, which in turn initiated ovarian and testicular differ­
differ-
entiation, respectively. The existence of of these hypothetical inductors re­ re-
mains to be demonstrated. As a result, consideration of the suitability of of
Wistchi's
Wistchi’s model has remained a conceptual debate.
In later publications, Witschi (1965,
(1965, 1967)
1967) modifi ed his theory to include
modified
an
an antagonist action of the inductors. The theory of antagonism has been
illustrated in the amphibians by ablation of the dominant component of the
gonad which results in differentiation of the remaining component (Haffen,(Haffen,
1977).
1977). Witschi (1967)
(1967) has suggested that the interaction ofof the inductors is
similar to an immune reaction. Therefore, each has the capability of inhibit­
inhibit-
ing or destroying the other.
other. Reinboth (1982)
(1982) has recognized the similarity
between Witschi's
Witschi’s (1967)
(1967)model and the model of mammalian sex differentia­
differentia-
tion involving the H-Y antigen system and a presumptive ovarian factor. The
existence of the ovarian factor and the nature of its interaction with the H-Y
antigen system remain a subject of study and debate (Wachtel and Koo,
1981).
1981).
With regard to fish, the debate surrounding the dual-inductor concept
has centered on reconciliation of the model with the proposed unitary origin
of the teleost gonad and the common occurrence within the teleosts of
various forms of hermaphroditism.
various
The discrete topographical division of of the primoridal amphibian gonad
provided Witschi with strong support for his dual-inducer concept. This dual
embryonic nature appeared ideally suited to providing the separate chemical
 HORMONAL SEX
5. HORMONAL SEX CONTROL
CONTROL AND
AND ITS
ITS APPLICATION
APPLICATION TO
TO FISH
FISH CULTURE
CULTURE 233

environments necessary
environments necessary for for the
the divergent
divergent development
development of of either
either male
male or
or
female gonia. Within the fish, a dualistic structure structure of of the primordial gonad
has
has been reported in the elasmobranchs (Chieffi, (Chieffi, 1959).
1959). However, unlike
this group and the rest of of the vertebrates the gonads of of cyclostomes and
teleosts have reported to develop from a unitary
have been reported unitary primordium homolo-homolo­
gous to the cortex of of other vertebrates
vertebrates (D’Ancona,
(D'Ancona, 1941,1941, 1949).
1949). D’Ancona’s
D'Ancona's
observations have
observations have received
received general support (Hoar,
general support (Hoar, 1969).
1969). However,
However, the the
reconciliation of of a dual-inductor
dual-inductor system with with a single primordial source has
been difficult.
been difficult. First,
First, neither D'Ancona nor
neither D’Ancona later investigators
nor later have been
investigators have been able
able
to explain how
to explain the development
how the development of aa single primordium could
single primordium could provide
provide aa basis
basis
for
for two distinct cell lines lines producing
producing two antagonistic
antagonistic inductors. Second,Second,
discrete
discrete male
male and
and female
female territories
territories are
are found
found as early as
as early as the
the first
first indication
indication
of
of sex
sex differentiation
differentiation in in many hermaphroditic species.
many hermaphroditic species. Further,
Further, although the
although the
classification
classification of many hermaphroditic
of many hermaphroditic species remains in
species remains in doubt, it is
doubt, it is clear
clear
that
that they
they are not restricted
are not restricted to to aa few isolated advanced
few isolated advanced families
families as
as suggested
suggested
by
by D'Ancona (1949). Smith
D’Ancona (1949). Smith (1975)
(1975)hashas provided
provided an an excellent
excellent review
review ofof their
their
distribution.
distribution. Because
Because of of these
these difficulties
difficulties and
and aa recognition
recognition of of the problems
problems
associated
associated with
with the
the examination
examination of of the
the peritoneal
peritoneal embryonic
embryonic differentiation
differentiation
of
of interrenal,
interrenal, nephric, and and gonadal
gonadal elements as as described
described by by Hardisty
Hardisty
(1965), Harrington
(1965), Harrington (1974,
(1974, 1975)
1975) has suggested that
has suggested that aa reevaluation
reevaluation of of the
single
single primordium
primordium hypothesis
hypothesis may may be in in order.
order.
An additional problem presentedpresented by the hermaphrodites is the method
by which
which the
the proposed
proposed antagonistic
antagonistic inductors
inductors maymay be compartmentalized
compartmentalized in in
aa gonad
gonad containing
containing bothboth testicular
testicular and
and ovarian
ovarian areas.
areas. The gonadal
gonadal organiza­
organiza-
tion
tion of
of some
some species
species offers
offers aa certain
certain degree
degree of of spatial
spatial isolation
isolation in
in the case
case of
of
protogynous
protogynous or or protandrous
protandrous serranid
serranid or or sparid
sparid species
species (Atz,
(Atz, 1964)
1964) and
and the
the
synchronous
synchronous hermaphrodite
hermaphrodite Rivulus marmoratus (Harrington,
Riuulus mannoratus (Harrington, 1971).
1971). How­
How-
ever,
ever, in
in the grouper Epinephelus, there
the grouper there appears
appears to to be
be nono discrete
discrete segrega­
segrega-
tion
tion of
of male
male and
and female
female territories.
territories.

3.
3. SEX
SEX STEROIDS
STEROIDS
Although
Although the two two presumptive
presumptive inducers
inducers have
have never
never been
been identified,
identified,
Yamamoto
Yamamoto (1969)
(1969)concluded
concluded thatthat based
based on
on his
his work
work with Oryzias latipes the
with Oryzias the
two
two inductors,
inductors, which
which hehe termed
termed the
the gynotermone
gynotermone andand androtermone,
androtermone, were were
in
in fact
fact estrogens
estrogens andand androgens,
androgens, respectively.
respectively. Yamamoto's
Yamamoto’s synthesis
synthesis of
of the
the
hormone
hormone andand inductor
inductor models
models has
has dominated
dominated the
the work
work on
on sex
sex control
control in
in fish.
fish.
However,
However, thethe origins
origins ofof the
the hormonal
hormonal model
model are
are much
much older.
older.
In
In the
the six
six decades
decades since
since Lillie
Lillie (1917)
(1917)postulated
postulated sexsex steroid
steroid involvement
involvement
in
in his
his explanation
explanation of of the
the free-martin
free-martin effect
effect in
in cattle,
cattle, exhaustive
exhaustive studies
studies have
have
been
been conducted
conducted in in an
an attempt
attempt toto determine
determine thethe specific
specific role
role of
of hormones
hormones in in
the
the process
process ofof sex
sex differentiation.
differentiation. Numerous
Numerous experiments
experiments involving
involving classic
classic
234 GEORGE A. HUNTER
GEORGE A. AND EDWARD
HUNTER AND EDWARD M
M.. DONALDSON
DONALDSON

castration
castration and
and replacement
replacement have
have confirmed
confirmed that
that the
the sex
sex hormones
hormones mediate
mediate
the development of of secondary
secondary sexual
sexual characteristics
characteristics inin mammals
mammals andand birds
birds
Gost,
(Jost, 1965;
1965; Goldstein and Wilson, 1975).
1975). In both mammals (Price, 1970)1970)and
birds (Haffen, 1975), biological,
(Haffen, 1975), biological, biochemical, and histochemical tests indi­ indi-
cate hormonal activity in the indifferent gonad of the dominant sex. sex. Howev­
Howev-
er, evidence for a role of the sex steroids in sex differentiation resulting from
steroid administration has been inconclusive. In the marsupials, Burns
(1950)
(1950) working with opposum, Didel phys virginiana,
Didelphys virginiuna, was able to demon­
demon-
strate the formation of ovotestes under the influence of of estradiol administra­
administra-
tion. However, in eutherian mammals, numerous experiments involving the
tion.
uioo or in vitro
in vivo uitro administration of exogenous sex steroids have been ineffec­
ineffec-
tive (Burns,
(Burns, 1961;
1961; McCarrey and Abbott, 1979).1979).
Feminization as a result of estrogen administration to the male embryo
has been achieved in the chick (Narbaitz
(Narbaitz and De Robertis, 1970)1970) and quail
(Haffen, 1969); however, the inversions are often transitory. The administra­
(Haffen, 1969); administra-
tion of a variety of androgens have produced either simple masculinizing
effects
effects or masculinizing and feminizing effects.
effects. Testosterone and its esters
act similarily to the natural hormones producedproduced by the embryonic male
gonad masculinizing the genital ducts but not influencing the female gonads. gonads.
Some androgens including androstanedione, androstenedione, androstene­ androstene-
diol, and trans-hydroandrosterone
truns-hydroandrosterone masculinize female genital ducts but fem­ fem-
inize male genital ducts and gonads (Haffen and Wolff, 1977). 1977). Numerous
studies have demonstrated that (1) (1)feminized male gonads can secrete a
hormone similar to the sex reversing hormone, (2) (2) embryonic gonadal secre­
secre-
tions from the medulla which have the same effect as steroid hormones, and
(3)
(3) indifferent avian gonads synthesize and secrete steroids, steroids, all of which
Haffen and Wolff
Wolf€claimed support the steroid-inductor model. Somatic sex
inversion has been achieved in cultured embryonic left testes administered
exogenous androgens or estrogens (Carlon and Erickson, 1978). 1978). Carlon and
Erickson suggested that it is the absence of steroidogenesis during the indif­ indif-
ferent period which is necessary for testicular development.
Similarily,
Similarily, androgen administration to several species of reptiles has re­ re-
sulted in variable results (Haffen
(Haffen and Wolff, 1977). Estrogen administration
Wolff, 1977).
to the green lizard,
lizard, Lacerta
Lacertu viridis resulted in partial or complete inhibition
of testicular development in some individuals to produce an ovotestis and
complete inhibition in others to produce an ovary (Raynaud,(Raynaud, 1967).
1967).
Within amphibians a relatively large number of studies involving hor­ hor-
monally induced sex inversion have demonstrated that the administration of
exogenous estrogens and androgens results in functional feminization in
urodeles and masculinization of ranid anurans, respectively. However, para­ para-
doxical
doxical actions of steroid treatment have been reported in several species
(Burns,
(Burns, 1961;
1961; Haffen and Wolff,
Wolff, 1977).
1977).
 HORMONAL SEX
5. HORMONAL SEX CONTROL
CONTROL AND
AND ITS
ITS APPLICATION
APPLICATION TO
TO FISH
FISH CULTURE
CULTURE 235

amphibia, sex steroids

In teleost fish as in the amphibia, steroids are capable of of influencing
the course of of sex differentiation.
differentiation. Yamazaki (1983) (1983) reported
reported at least 15 15 species
in which functional sex inversion has been been achieved.
achieved. However, these species
are primarily gonochoristic teleosts within a small number number of of families.
families .
Chieffi. (1967)
Chieffi (1967) reported the effects of of androgens and estrogens on the elas- elas­
mobranchs, primarily Scyliorhinus canicula.
mobranchs, canicula. Both estrogens and androgens
influence the differentiation
influence differentiation of the genital ducts; however, only estrogen
influences the gonad in this species producing prodUcing ovotestes.
ovotestes.
The small number of of hermaphroditic species which have been treated
with steroids to manipulate natural sex inversion have responded inconsis- inconsis­
tently. Reinboth (1962,(1962, 1975)
1975) used a single 2-mg injection of of testosterone to
mimick sex inversion in several species of of protogynous hermaphrodite
(Labridae). He concluded that, in general, androgens induced pre-
wrasses (Labridae). pre­
inversion in protogynous species, but this evidence alone did not
cocious sex inversion
support the steroid-inductor model (Reinboth, (Reinboth, 1970).
1970). Chen et al. (1977) (1977)
reported thatthat two of Epinephelus tauvina, also a pro­
of three groupers, Epinephelus pro-
togynous
togynous hermaphrodite, fed 80 mg methyltestosterone over 30 days initi- initi­
ated sex inversion.
inversion. Subsequently,
Subsequently, all 25 fish given a dosage of of 11 mg meth­
meth-
yltestosterone/kg diet 3 3 times per week over a 2-month 2-month period underwent
sex
sex inversion.
inversion.
The
The rice field eel, Monopterus
Monopttwus albus, albus, has perhaps been the most inten­ inten-
sively studied protogynous hemaphrodite (Chan, (Chan, 1977;
1977; Chan et al. al.,, 1977).
1977).
Biochemical
Biochemical and histochemical techniques have demonstrated 3j3-hy­ 3P-hy-
droxysteroid dehydrogenase an
droxysteroid dehydrogenase and d 17j3-hydroxysteroid
17s-hydroxysteroiddehydrogenase
dehydrogenase activity
and
and aa large increase in
large increase in the
the production
production of of androgens
androgens during
during natural
natural sexsex inver­
inver-
sion (Chan
(Chan et al.al.,, 1975;
1975; Tang et al.al.,, 1975).
1975). However,
However, attempts to manipulate
manipulate
the
the course
course of natural
natural sex inversion by
sex inversion by the
the administration
administration of of steroids
steroids have
have
been ineffective
ineffective (Tang(Tang et al. 1974; Chan et al.
al.,, 1974; al.,, 1977).
1977). These studies are
examined in detail in C hapter 4, this
Chapter this volume.
The achievement
achievement of functional
functional sex inversion in severalseveral gonochorist
gonochorist spe­ spe-
cies remains the most compelling evidence for steroid involvement in the
sex differentiation
differentiation of fish. fish. However,
However, as Reinboth (1970) (1970)has noted, it is not
possible to rule out a pharmacological
possible pharmacologicalrather than physiological
physiological action in this
process.
process. No studies
studies have examined
examined the mode of action of the steroids steroids in the
induced sex inversion
inversion of teleosts.
teleosts. In this regard,
regard, Vannini et al. al. (1975)
(1975)exam­
exam-
ined the influence
influence of of actinomycin
actinomycin D D and
and puromycin,
puromycin, inhibitors
inhibitors of DNA­ DNA-
dependent RNA transcription RNA-dependent protein synthesis
transcription and RNA-dependent synthesis re­ re-
spectively,
spectively, on testosterone-induced sex sex inversion
inversion in Rana dalmatina
dalmatina tad­ tad-
poles.
poles. Both
Both antibiotics
antibiotics suppressed testosterone action. action. Vannini
Vannini and co-work­
co-work-
ers
ers also
also reported that testosterone-induced sex sex inversion
inversion isis linked with RNA
synthesis.
synthesis. They
They havehave proposed that that the mechanisms
mechanisms of testosterone
testosterone action
action
involve
involve the derepression of latent male male genes
genes in the nuclei
nuclei of somatic
somatic tissue
tissue
236
236 GEORGE A. HUNTER
GEORGE A. AND EDWARD
HUNTER AND M .. DONALDSON
EDWARD M DONALDSON

which
which inin turn
turn causes
causes them
them to
to proliferate
proliferate as
as gonadal
gonadal medullary
medullary tissue.
tissue. Evi­
Evi-
dence for the direct action of steroids and their specific
specific receptor molecules
on DNA and, thereby, transcriptional processes has been established
(O'Malley
(O’Malley and Schrader, 1976).
1976). Further, no studies have examined the po­ po-
tential sites of genetic control of hormonal action, specifically
specifically the partially
common biosynthetic pathway of the sex sex steroids or sex-specific
sex-specific receptor
sites on undifferentiated germ cells.
Therefore, additional supportive evidence for Yamamoto's
Yamamoto’s conclusions
conclusions
was confined to the specificity of sex steroids as exogenous sex inductors, the
very low effective dosage of sex steroids,
steroids, and the selective incorporation of
sex steroids into the differentiating gonad. However, based on similar evi­
sex evi-
dence, later studies
studies have added uncertainty rather than support to Yamamo­
Yamamo-
to's
to’s (1969)
(1969) hypothesis of a sex-steroid-inductor model.
Yamamoto based the specificity of sex-steroid action on the absence of
paradoxical
paradoxical effects and inactivity of corticoids as sex inducers in Oryzias
latipes. To date,
latipes. date, no studies have reported effective sex inversion using cor­ cor-
ticoids, although other chemicals such as N,N-dimethylformamide
N,N-dimethylformamide can modi­ modi-
fy gonadal differentiation (van den Hurk and Slof, Slof, 1981).
1981).
More recent studies have reported paradoxical
paradoxical effects
effects of
of sex steroid ad­
ad-
ministration. Gresik and Hamilton (1977)
(1977) citing unpublished work by J. B.
Hamilton and D. D. Kantor report that the injection of eggs containing XX
or XY Oryzias latipes embryos with either methyltestosterone, estrone ace­ ace-
tate, or the synthetic progestin ethynodiol diacetate at low concentrations
favors testicular differentiation. High concentrations favor ovarian differ­ differ-
entiation. They also report that paradoxical
paradoxical sex inversion is more difficult to
achieve in YY
YY individuals than in XY individuals. Paradoxical
Paradoxical feminization in
Hemihaplochromis multicolor fry reared in
the cichlids was first observed in Hemihaplochromis
water containing testosterone propionate or methyltestosterone
methyltestosterone (Muller,
(Muller,
1969).
1969). Subsequently,
Subsequently, the addition of these steroids at 500 j.Lg/1 kg/l was also
found to have a paradoxical effect in other cichlids including Chichlasoma
Chichlasoma
heudeloti, and
biocellatium, Tilapia heudeloti,
biocellatium, and Oreochromis mossambicus (Hack­ (Hack-
mann, 1974).
1974). The effect was again demonstrated in Oreochromis mossam­ mossam-
bicus fed methyltestosterone
methyltestosterone at 1000
1000 mg/kg diet for a period of 50 days after
hatching
hatching (Nakamura,
(Nakamura, 1975).
1975). In
In genetic
genetic females,
females, oogenesis
oogenesis progressed,
progressed, al­al-
though at a slower rate than control ovaries. The ovarian cavity was formed
paradoxically
paradoxically in genetic males. Following the end of treatment, these fish
developed intersexual gonads. Advanced spermatogenesis was observed in
the inner periphery of the efferent ducts. Maturing oocytes and a definite
ovarian
ovarian cavity
cavity were
were observed
observed on
on the outer
outer side
side of
of the
the ducts.
ducts. In
In the
the same
same
study, administration of methyltestosterone at 50 mg/kg diet resulted in
gonadal masculinization. Nakamura (1975)(1975) noted that, in contrast to his ob-
 5.
5. HORMONAL
HORMONAL SEX
SEX CONTROL AND
CONTROL A N D ITS
ITS APPLICATION
APPLICATION TO
TO FISH CULTURE 237
FISH CULTURE 237

servations, Reinboth (1969) (1969) had reported that in Hemihaplochromis multi­ multi-
color a long-term treatment at extremely high androgen dosages (30-50 (30-50 g/kg
diet)
diet) was required
required to cause gonadal masculinization, but a short-term treat­ treat-
ment at similar dosages resulted in gonadal feminization. He suggested
interspecies differences as a possible explanation but acknowledged that the
major reason for these variations was unknown. More recently, paradoxical paradoxical
actions ofof testosterone and ethynyltestosterone have been reported in rain­ rain-
bow trout (V. (V. ]J . Bye,
Bye, 1980,
1980, personal communication) and channel catfish,
Zctalurus punctatus (Goudie
Ictalurus (Goudie et al. al.,, 1983).
1983).
Recent work with the amphibians has demonstrated that care must be
taken in the interpretation of paradoxical steroid action. action. Chieffi
Chief3 (1965)
(1965) noted
the paradoxical effects found in urodele gonads following following high dosages of
androgens. For example,
example, in Pleurodeles waltlii, waltlii, testosterone propionate in­ in-
duced complete feminization at all dosages administered (Gallien, (Gallien, 1950).
1950).
Chieffi
ChiefE concluded
concluded that that the
the nonspecific
nonspecific actionaction of
of the steroids
steroids argued
argued against
against
their putative roles as sex inducers. The administration of of estrogens to sever­
sever-
al ranid species results in feminization,
feminization, intersexuality, or masculinization
with increasing dosages (Padoa, (Padoa, 1942; 1942; Gallien, 1941). Hso et al.
1941). Hsii al. (1978a,
(1978a,b)b)
have
have also
also demonstrated
demonstrated that that high
high dosages
dosages of of estradiol
estradiol result
result in in paradoxical
paradoxical
masculinization of Rana catesbeiana tadpoles. However, this treatment also
inhibits
inhibits Ll5-3[3-hydroxysteroid
A5-3P-hydroxysteroid dehydrogenase,
dehydrogenase, aa key key enzyme
enzyme in in ovarian
ovarian
steroidogenesis.
steroidogenesis. Hso Hsu andand co-workers
co-workers suggested
suggested thatthat if
if the paradoxical
paradoxical action
action
of
of this steroid arose
arose through an an inhibition
inhibition of of ovarian
ovarian steroidogenesis,
steroidogenesis, the the
paradoxical
paradoxical effects would lend
effects would lend support
support to to the
the steroid-inductor
steroid-inductor theory.theory.
The data from the limited number of studies employing steroidogenic
inhibitors
inhibitors or or antiandrogens
antiandrogens have have not not supported
supported specific
specific steroid
steroid action.
action.
Cyproterone acetate is the most commonly used antiandrogen in studies
with fish.
fish. In mammals, it has been shown to be a potent inhibitor of endoge­ endoge-
nous or exogenous testosterone (Hamada (Hamada et al. al.,, 1963).
1963).The antiandrogen has
been partially effective in blocking the development of secondary sexual
characteristics
characteristics in in the
the three-spined
three-spined stickleback
stickleback (Rouse
(Rouse et al. al.,, 1976),
1976), the
the Indi­
Indi-
an
an catfish,
catfish, Heteropneustes fossilis fossilis (Sundararaj
(Sundararaj and and Nayyar,
Nayyar, 1969),1969), and
and the
the
guppy, reticulata (Smith,
guppy, Poecilia reticulata (Smith, 1976).
1976). Schreck
Schreck (1974),
(1974), citing
citing the un­un-
published
published workwork of of Irons
Irons and
and Schreck,
Schreck, reported
reported aa possible
possible sex sex inversion
inversion in in
male
male Oryzias latipes fed fed the
the antiandrogen,
antiandrogen, cyproterone
cyproterone acetate,
acetate, at at 50-500
50-500
mg/kg
mg/kg dietdiet for
for 12
12 weeks.
weeks. Hopkins
Hopkins et al. al. (1979)
(1979) fed
fed cyproterone
cyproterone acetateacetate inin
combination with various natural and synthetic estrogens to Oreochromis Oreochromis
au reus (Tilapia
aureus aurea) fry.
(Tilapia aurea) fry. The
The results indicated that
results indicated that rather
rather than
than potentiating
potentiating
the effect of of the estrogens it actually lessened their effectiveness. Rastogi
and
and Chieffi
Chief3 (1975)
(1975)demonstrated
demonstrated that that the
the same antiandrogen
antiandrogen did not not inhibit
inhibit
the
the masculinizing
masculinizing effectseffects of
of testosterone
testosterone propionate
propionate or or lll-ketotestosterone
l-ketotestosterone
238
238 GEORGE
GEORGE A.
A. HUNTER AND EDWARD
HUNTER AND M.. DONALDSON
EDWARD M DONALDSON

in the swordtail Xiphophorus helleri even though it inhibited the binding of of

the steroid to testosterone-sensitive target tissues.
The mechanism of action of cyproterone acetate in the lower vertebrates
is not clear. Chieffi al. (1974)
Chief3 et al. (1974) administered cyproterone acetate to Rana
esculenta tadpoles. They hypothesized that if the androgens were indeed the
esculenta
sex inductors, the administration of cyproterone acetate would result in
competition for androgen-receptor
androgen-receptor sites and subsequent abnormal testicular
development.
development. Instead Instead the
the ovaries
ovaries were masculinized. Chieffi
were masculinized. Chief3 and
and co-workers
co-workers
suggested that these results indicated
suggested that these results indicated that that the sex
sex inductors were not
inductors were not struc­
struc-
turally similar to sex steroids.
steroids. However, Hsii et al. (1979)
Hsu al. (1979) have also demon­
demon-
strated
strated that
that cyproterone
cyproterone acetate
acetate administered
administered inin the rearing
rearing water
water atat 1500
1500
IJ.g/l for a 7 -month period can transform
pg/l for a 7-month period can transform ovaries of ovaries of Rana catesbeiana into
catesbeiana into
testes. They suggested that the masculinizing effect of the antiandrogen was
attributable to inhibition of ovarian steroidogenesis, specifically A5_3f3- A5-3p- and
possibly 17f3-hydroxysteroid
l7a-hydroxysteroid dehydrogenase. They also hypothesized that
this inhibition of young ovaries would result initially in ovarian degeneration
and subsequent masculinization of the gonads with prolonged treatment.
Further, HsiiHsu and co-workers maintain that these results support the steroid­ steroid-
inductor theory.
The observation ofparodoxical steroid action also suggests the possibility
of a dominant-neutral
dominant-neutral sex mechanism of dimorphism. Studies involving the
culture of isolated mammalian and avian cortex and medulla suggest that
sexual dimorphism occurs as a result of the presence or absence of a single
sexual
dominant sex. sex. In the absence of the dominant sex the neutral sex develops
(McCarrey and Abbott, 1979). 1979). As a general rule, the dominant sex is corre­corre-
lated with heterogamety (Jost, Gost, 1965).
1965). Hackmann and Reinboth (1974) (1974) noted
that paradoxical sex inversion in the lower vertebrates occurs only in one
direction. Specifically
Specifically in the urodeles and cichlids, high doses of androgen
result
result in in feminization,
feminization, butbut inin the
the families
families Ranidae and Hylidue estrogen
and Hylidae estrogen
treatment results in masculinization. Hackmann and Reinboth proposed that
the evidence from the cichlids could support a hypothesis in which the
female hormone
hormorie induces the female sex type while the absence of the hor­ hor-
mone
mone promotes
promotes the development
development of of male
male sex
sex type.
type. The
The evidence
evidence from
from the
breeding experiments conducted by Hackmann and Reinboth (1974) (1974) with
Hemihaplochromis multicolor indicated male heterogamety. Therefore, the
Hemihaplochromis
association between heterogamety and the dominant sex observed in birds
association
and
and mammals
mammals could could not
not be
be supported.
supported.
Similarly, the results presented by Gresik and Hamilton (1977) (1977) and van
den Hurk and Slof (1981) (1981) suggested a dominant-neutral
dominant-neutral sex hypothesis in
which interrupted testicular development results in the formation of an
ovary
ovary.... Van den Hurk and Slof (1981) (1981) obtained feminization in in rainbow trout
5.
5. HORMONAL SEX
SEX CONTROL
CONTROL AND
AND ITS
ITS APPLICATION
APPLICATION TO
TO FISH
FISH CULTURE 239

by administering progesterone in the rearing water at 300 �gll pg/l over the
period
period of of sexual
sexual differentiation.
differentiation. Van Van den den Hurk
Hurk and Slof suggested
and Slof suggested that that inin
addition to a possible direct feminizing effect of progesterone, there also also
exists
exists the
the possibility
possibility that
that progesterone
progesterone may may actact as
as aa steroid-blocking
ster'oid-blocking agent agent ofof
the
the androgenic
androgenic pathway,
pathway, resulting
resulting in in ovarian
ovarian differentiation.
differentiation. An An indirect
indirect
action
action of of the
the progestins
progestins via via aa blockage
blockage of of androgen
androgen biosynthesis
biosynthesis has has been
been
previously
previously described
described in in mammals
mammals by by Gower (1975). However,
Gower (1975). However, Van Van denden Hurk
Hurk
and
and Slof
Slof also
also noted
noted that
that progesterone
progesterone has has not
not been
been demonstrated
demonstrated to to affect
affect sex
sex
differentiation in Oryzias latipes (Yamamoto (Yamamoto and Matsuda, 1963). 1963). Recently,
van
van den Hurk and
den Hurk and Lambert
Lambert (1982)(1982) have
have also
also reported
reported thatthat progesterone
progesterone does does
not
not influence
influence sexsex differentiation
differentiation in in rainbow
rainbow trout
trout when
when administered
administered in in the
the
diet.
diet. The evidence
evidence produced
produced by van van den Hurk et al.
den Hurk (1982), demonstrating
al. (1982), demonstrating
the
the steroidogenic capabilities of
steroidogenic capabilities of rainbow
rainbow trouttrout testes
testes but
but notnot ovaries
ovaries at at the
the
time
time ofof sex
sex differentiation,
differentiation, also also supports
supports aa dominant-neutral
dominant-neutral sex sex hypothesis.
hypothesis.
The limited evidence from these studies is clearly not sufficient sufficient to deter­
deter-
mine
mine whether
whether aa dominant-neutral
dominant-neutral system system of of sexual
sexual dimorphism is is present in in
fish. Presumably the unitary origin of the teleost gonad has prevented dis­
fish. dis-
sociation-recombination
sociation-recombination experiments experiments which which would
would provide
provide moremore substantial
substantial
evidence for a dominant-neutral
dominant-neutral sex hypothesis. The discrete male and
female
female territories
territories inin hermaphroditic
hermaphroditic species species may provide an
may provide an ideal
ideal vehicle
vehicle forfor
such
such an an examination.
examination.
Hishida
Hishida (1962,
(1962, 1965)
1965)by by demonstrating
demonstrating the the selective
selective incorporation
incorporation of of sex
sex
steroids into
steroids into the
the developing
developing gonads, gonads, provided
provided Yamamoto
Yamamoto with with considerable
considerable
support
support for his hypothesis.
for his hypothesis. Administration
Administration of of 4-[ 14Cltestosterone
14C]testosteronepropionatepropionate
to
to larval
larval Oryzias latipes resultedresulted in in the
the accumulation
accumulation of of the labeled
labeled steroid
steroid
only
only byby the
the actively
actively differentiating
differentiating gonads gonads (Hishida, 1962). In
(Hishida, 1962). In aa later study,
later study,
Hishida (1965)
(1965) fed
fed larval
larval Oryzias latipes 16-[ 16-[14C]estrone
14C]estrone and diethylstilbes­
and diethylstilbes-
trol
trol l-[14Clmonoethyl).
l-[14C]monoethyl). A 4- 4- to
to lO-fold
10-fold concentration
concentration of of the
the steroids
steroids was was
found
found in in developing
developing gonads,
gonads, againagain indicating
indicating active
active accumulation
accumulation of of the
the
steroids.
steroids. Conversion
Conversion of of estrone
estrone to to estradiol
estradiol was also demonstrated.
was also demonstrated. Further,Further,
based
based on on recovered counts, the
recovered counts, the effective
effective oral
oral dosages
dosages forfor 100%
100% sex sex inversion
inversion
were
were calculated
calculated to to be 1.8 1. 8 X X 10 - 2�g estrone
10-2pg estrone and and 1. 1.11 X X 10 - 2 �g pg di­di-
ethylstilbestrol.
ethylstilbestrol. Based
Based on on this
this evidence,
evidence, Yamamoto
Yamamoto (1969) (1969) asserted
asserted thatthat the
the
levels
levels ofof steroids
steroids required
required for for manipulation
manipulation of of sex
sex differentiation
differentiation were were within
within
the
the physiological
physiological capabilities
capabilities of of the
the species.
species. Further
Further evidence
evidence that that phys­
phys-
iological
iological levels
levels of
of steroid
steroid can can influence
influence sexual
sexual differentiation
differentiation was was provided
provided by by
Satoh
Satoh (1973).
(1973). InIn this
this study,
study, trunk
trunk regions
regions of of newly hatched Oryzias latipes
newly hatched
fry
fry were
were transplanted
transplanted into into the
the anterior
anterior chamber
chamber of of the
the eyes
eyes of of adult
adult fish.
fish. The
The
gonads
gonads of genetic females
of genetic females transplanted
transplanted into into male hosts did
male hosts did not
not differentiate
differentiate
into
into anan ovary,
ovary, but
but did
did form
form spermatogenic
spermatogenic cells. cells. Satoh
Satoh suggested
suggested that that the
the
results supported the
results supported the actions
actions of of androgens
androgens as as andro-inducers.
andro-inducers. AlthoughAlthough the the
240 GEORGE A. HUNTER
GEORGE A. AND EDWARD
HUNTER AND EDWARD M .. DONALDSON
M DONALDSON

results
results provide
provide strongstrong supportive
supportive evidence,
evidence, they they dodo notnot conclusively
conclusively demon­
demon-
strate a role for the steroids in primary gonadal
strate a role for the steroids in primary gonadal differentiation. differentiation.
Studies
Studies correlating
correlating the development of
the development of steroidogenic
steroidogenic and and differentiating
differentiating
tissue
tissue or identifying steroid synthesis in sexually differentiating gonads
or identifying steroid synthesis in sexually differentiating gonads areare
rare.
rare. Dildine (1936) (1936)first
first noted
noted the correlation
correlation between
between the the development
development of of aa
stromal
stromal region
region and and the initiation
initiation of of male
male influences
influences on on the
the sex
sex differentiation
differentiation
in
in Poecilia reticulata. More recently, Takahashi (1975a) observed
reticulata. More recently, Takahashi ( 1975a) observed that,that, in
in
embryonic
embryonic Poecilia reticulata, the appearance
appearance of aggregations of stromal
of aggregations of stromal
cells
cells in
in the
the gonadal
gonadal hilus
hilus 18 18 days
days after
after the lastlast parturition
parturition are are indicative
indicative of of
testicular differentiation. Oral administration of methyltestosterone methyltestosterone to grav­ grav-
id
id females
females results
results in in somatic
somatic aggregation
aggregation in in the hilar
hilar region
region of of embryonic
embryonic
ovaries.
ovaries. Females
Females affected
affected in in this
this manner
manner contain
contain developing
developing oocytes
oocytes andand
malelike
malelike stromal
stromal aggregations
aggregations in in the hilar
hilar region.
region. Within
Within 20 daysdays ofof birth
birth the
ovaries completely degenerate
ovaries completely degenerate and and areare replaced
replaced by by testes
testes which
which display
display
precociously
precociously differentiating
differentiating spermsperm ducts,ducts, testicular
testicular interstitium,
interstitium, and and aa con­
con-
comitant
comitant initiation
initiation of of spermatogenesis.
spermatogenesis. The period period thatthat is most sensitive
is most sensitive toto
change
change in in the
the developing
developing ovaries
ovaries is is at 18 days
at 18 days after
after the last
last parturition,
parturition, andand itit
is synchronous with the period of stromal aggregation in the hilar region of
developing testes.testes. Takahashi concluded that masculinization of the somatic
element
element is is essential
essential for for functional
functional sex sex inversion
inversion of of females,
females, and and that
that the
the
somatic
somatic differentiation
differentiation may may occur occur priorprior to to germinal
germinal differentiation.
differentiation.
Nakamura
Nakamura (1978) (1978) hashas reported
reported similar
similar stromal
stromal aggregations
aggregations in in the
the hilar
hilar re­
re-
gion
gion prior
prior to to male
male differentiation
differentiation in in Oryzias latipes and and the
the mosquito
mosquito fish fish
Gambusia af finis. However,
affinis. However, Satoh Satoh and and Egami
Egami (1972)(1972) diddid not
not observe
observe sex sex
differences
differences in in the
the histological
histological structure
structure of of the
the gonadal
gonadal somatic
somatic tissue
tissue prior
prior to
to
gametogenesis in Oryzias latipes.
Yoshikawa
Yoshikawa and and Oguri
Oguri (1979)
(1979) reported
reported that that inin Oryzias latipes the the differ­
differ-
entiation of somatic cells into interstitial cells always precedes the transfor­ transfor-
mation
mation fromfrom spermatogonia
spermatogonia to to spermatocytes.
spermatocytes. Further, Further, interstitial
interstitial cells
cells are
are
always
always found in the vicinity of germ cells in meiosis, suggesting that intersti­ intersti-
tial
tial cells
cells areare responsible
responsible for for the
the differentiation
differentiation of germ germ cells.
cells. However,
However, in in
the
the same
same species,
species, Satoh
Satoh ((1974)
1974) reported
reported the appearance
appearance of steroidogenic
steroidogenic
cells
cells after
after the onset of
the onset of gonadal
gonadal sex sex differentiation.
differentiation. Similarily,
Similarily, van
van denden Hurk
Hurk
al. (1982)
et al. (1982)foundfound no steroid-synthesizing structures
no steroid-synthesizing structures in in indifferent
indifferent gonads
gonads of of
rainbow
rainbow trouttrout 50 50 days
days postfertilization
postfertilization.. In this study,
In this study, steroidogenic
steroidogenic (Leydig)
(Leydig)
cells
cells were detected in differentiated testes at 100 100 days postfertilization.
Takahashi
Takahashi and and Iwasaki
Iwasaki (1973)
(1973) provided
provided the the first
first examination
examination of of the onset
onset
of
of steroidogenesis
steroidogenesis in in the
the differentiating
differentiating teleostteleost gonad.
gonad. These
These researchers
researchers
detected
detected �5-3�-hydroxysteroid
A5-3P-hydroxysteroid dehydrogenasedehydrogenase activity activity in in gonadal interstitial
gonadal interstitial
cells in Poecilia reticulata 7 days
cells in days postpartum.
postpartum. The The enzyme
enzyme activity
activity was de­
was de-
tected concurrent with
tected concurrent with thethe multiplication
multiplication of spermatogonia
spermatogonia and and differentia-
differentia-
5.
5. HORMONAL
HORMONAL SEX
SEX CONTROL AND
CONTROL A N D ITS
ITS APPLICATION
APPLICATION TO
TO FISH
FISH CULTURE
CULTURE 241

tion
tion of
of cells
cells in
in the
the testicular
testicular hilus
hilus into
into the
the sperm
sperm ductduct analogues
analogues and and testic­
testic-
ular stroma. The enzyme
ular stroma. enzyme was was detected
detected in in the cells
cells of
of the
the stroma
stroma butbut not
not inin
the
the sperm ductduct analogues. Further, enzyme
analogues. Further, enzyme activity
activity could
could notnot be
be detected
detected in in
fish
fish sampled
sampled at at 33 or
or 5 days
days postpartum.
postpartum. Takahashi
Takahashi and and Iwasaki
Iwasaki concluded
concluded
that
that the enzyme
enzyme activity
activity appears
appears and increases simultaneously
and increases simultaneously with with thethe dif­
dif-
ferentiation
ferentiation of of the
the testicular
testicular duct
duct system rather rather than
than the the germ
germ cells.
cells. Re­
Re-
cently, van den Hurk et al. al. (1982)
(1982) have conducted in vitro assays of Salmo
gairdneri gonadal
gairdneri gonadal homogenates
homogenates at 50, 100,
at 50, 100, and
and 200
200 days
days postfertilization.
postfertilization.
The
The histologically
histologically indifferent
indifferent gonads
gonads at at 50 days
days postfertilization
postfertilization contained
contained
3J3-hydroxysteroid
3P-hydroxysteroid dehydrogenase,
dehydrogenase, 5, 4-isomerase, and
5,4-isomerase, and l7a-hydroxylase,
l7au-hydroxylase,in­ in-
dicating
dicating thethe capacity
capacity to to synthesize
synthesize progestins.
progestins. Developing
Developing testes testes atat 100
100 days
days
postfertilization
postfertilization also also contained
contained 17a,17a, 20-desmolase
20-desmolase and and 11J3-hydroxylase,
11P-hydroxylase, in­ in-
dicating
dicating thethe capacity
capacity to to synthesize
synthesize androgens.
androgens. However,
However, sex sex differentiation
differentiation
had
had commenced
commenced between between 50 and and 100
100 days
days postfertilization.
postfertilization. Therefore,
Therefore, it it was
was
not possible to determine whether these androgen-synthesizing
not possible to determine whether these androgen-synthesizing capabilities capabilities
initiated
initiated or or occurred
occurred as as aa result
result ofof testicular development. Both
testicular development. Both testes
testes andand
ovaries
ovaries possessed 17P-hydroxysteroid dehydrogenase at 200 days postfertil­
possessed 17J3-hydroxysteroid dehydrogenase at 200 days postfertil-
ization. Further, the
ization. Further, the ovaries
ovaries possessed
possessed aromatase,
aromatase, indicating
indicating the the capability
capability
to
to synthesize estrogen. Therefore, although androgen- or estrogen-syn­
synthesize estrogen. Therefore, although androgen- or estrogen-syn-
thesizing capabilities could not be demonstrated demonstrated in indifferent gonads, a
capability
capability for for progestin
progestin synthesis was was noted. Van Van den
den Hurk
Hurk and and Slof
Slof (1981)
(1981)
had
had previously demonstrated that progesterone, administered at the time of
previously demonstrated that progesterone, administered at the time of
sex
sex differentiation,
differentiation, results
results inin gonadal feminization.
feminization. As As they
they note,
note, the the ab­
ab-
sence of
sence of estrogen-synthesizing
estrogen-synthesizing capabilities
capabilities in in differentiated
differentiated gonads
gonads 100 100 days
days
postfertilization
postfertilization is is substantial
substantial evidence
evidence against
against aa steroidal
steroidal ovarian
ovarian inductor.
inductor.
In
In aa later
later study,
study, vanvan den
den Hurk
Hurk andand Lambert
Lambert (1982)(1982)demonstrated
demonstrated the the conver­
conver-
sion
sion ofof 11J3-hydroxyandrostenedione
lip-hydroxyandrostenedione to to 11-ketoandrostenedione
11-ketoandrostenedione using using testic­
testic-
ular
ular homogenates
homogenates from from 100-day-old
100-day-old rainbow
rainbow trout.
trout. Therefore,
Therefore, the the presence
presence
of
of 1llp-dehydroxysteroid
1 J3-dehydroxysteroid dehydrogenase
dehydrogenase was was demonstrated.
demonstrated. Further Further admin­
admin-
istration
istration of of 11 J3-hydroxyandrostenedione at
llp-hydroxyandrostenedione at 60
60 mg/kg
mg/kg for for 88 weeks
weeks fromfrom first
first
feeding
feeding resulted
resulted in in an
an all
all male
male population.
population. Van den Hurk
Van den Hurk and and Lambert
Lambert
concluded
concluded that that this steroid is
this steroid is endogenous
endogenous to to rainbow
rainbow trout
trout andand responsible
responsible
for
for the
the initiation
initiation of of testicular
testicular development.
development.

D. Summary
Summary
No
No single
single model
model ofof sex
sex determination
determination and
and differentiation
differentiation may
may be recon­
recon-
ciled
ciled with
with all
all the
the present
present evidence
evidence from
from the fish. The
the fish. The sex-determining
sex-determining mech­
mech-
anism
anism is assumed to
is assumed to have
have aa genetic
genetic basis.
basis. Although
Although thethe majority
majority of
of fish
fish do
do not
not
have cytologically distinguishable sex chromosomes, both male and female
digametic systems
digametic systems have been demonstrated.
have been demonstrated. Nonheterosomal
Nonheterosomal chromosomal
chromosomal
242 GEORGE A. HUNTER
GEORGE A. HUNTER AND EDWARD M. DONALDSON
AND EDWARD DONALDSON

sex-determining systems have also been reported for some species. Even
within species with well developed heterosomal systems,
systems, sex determination
has been found to be labile to extrinsic factors.
factors. Of the c).Irrently available
cprrently
models for sex determination, the polygenic system described by Kallman
(1965)
(1965) and Yamamoto (1969)
(1969) appears to fit most closely the available data.
Sfnmarily,
Similarily, the mechanism of sex differentiation in fish has not been
determined. The sex-specific
sex-specific H-Y antigen system found in higher verte­ verte-
brates has been reported in several fish species.
species. Whether it plays a signifi­
signifi-
cant role in the process of of Se1(
sex differentiation
daerentiation in fish remains to be deter­
deter-
mined. The dual-sex-inductor model of d sex differentiation continues to
dominate the conceptual approach to research in fish. fish. However, debate
continues over the ability of a unitary primordium, reported for the teleosts,
to elaborate separate inducers. Evidence supporting Yamamoto's
Yamamoto’s (1969)
(1969) hy­
hy-
pothesis that the sex steroids are in fact the presumptive inducers remains
inconclusive.
inconclusive. The steroids are capable of influencing sex determination in
the majority
majority ofof gonochoristic
gonochoristic teleost
teleost species, but
but have
have been
been applied
applied with
with
variable success to hermaphroditic species.
species. The evidence for steroid action
based on the identification of steroid activity, the juxtaposition of steroido­
steroido-
genic tissues to differentiating germ cells, and the effects of anti steroidal
antisteroidal
compounds remains equivocal. However, regardless of whether the action of
steroids is pharmacological or physiological,
physiological, steroids can play a decidedly
influential role in the process of sex differentiation. Therefore, they are a
valuable
valuable tool
tool for
for both
both further
further exploration
exploration of
of the
the process
process of
of sex
sex differentiation
differentiation
and
and the
the manipulation
manipulation of of gonadal
gonadal sex
sex for
for purposes
purposes of
of fish
fish culture.
culture.

III. HORMONAL SEX CONTROL

III. HORMONAL CONTROL

Although there are numerous studies concerned

concerned with the application of of
hormones to control gonadal sex, most of of them have remained within the
framework for effective treatment
treatment described by Yamamoto (1969).
(1969). Based on
his work with Oryzias Zutipes, Yamamoto suggested that, for effective treat­
Oryzius latipes, treat-
ment, the species-speciijc
species-specific optimal dosage of a particular steroid should be
administered from the stage of the undifferentiated gonad through the time
of morphological differentiation. Although the majority of studies have con­
con-
firmed these criteria as excellent general rules, exceptions have been re­ re-
ported. Further, within these criteria considerable scope for variation exists.
exists.
Therefore, to facilitate consideration of the various components of these
studies a generalized model is presented
presented (Fig.
(Fig. 1).
1).
Hormonal sex-control studies consist of of three chronologically
chronologically ordered
phases: management, treatment,
treatment, and evaluation. Within the management
phase, both the species and the gonadal sex appropriate to attaining manage-