5-Japanese Puffer Fish - aromatase inhibitor induced masculinization

Original Article
Sex Dev 2007;1:311–322
DOI: 10.1159/000108935
Received: August 14, 2007
Accepted: September 13, 2007
Fugu (Takifugu rubripes) Sexual Differentiation:
CYP19 Regulation and Aromatase Inhibitor
Induced Testicular Development
H. Rashid
a
H. Kitano a K. Hoon Lee a S. Nii a T. Shigematsu a K. Kadomura b
A. Yamaguchi a M. Matsuyama a
a
Laboratory of Marine Biology, Faculty of Agriculture, Kyushu University, Hakozaki, Higashi-Ku, Fukuoka ;
b
Nagasaki Prefectural Institute of Fisheries, Nagasaki , Japan
Key Words
Aromatase CYP19 Estradiol-17 Fadrozole Fugu
Ovarian cavity formation Sexual differentiation
Testicular differentiation Testosterone
Abstract
In order to assess the involvement of aromatase CYP19 isoforms
and endogenous sex steroids in gonadal sex differentiation
and development of the Japanese fugu (Takifugu
rubripes) , an aromatase inhibitor (AI, fadrozole) was administered
to developing fishes from the ‘first feeding’ till the
100th day after hatching. It was observed that ovarian cavity
formation was inhibited by fadrozole at doses of 500 and
1000 g/g diet, which was followed by testicular differentiation
in all treated fugu. In the non-treated fugu, CYP19A was
predominantly expressed in the ovary and CYP19B in the
brain (in both sexes), although both were expressed interchangeably
at low levels. An exceptionally high expression
of CYP19B was also evident in testis throughout the study
period. Both forms of CYP19 mRNA showed low levels of expression
in brain and gonad with no significant differences
between the two AI treatments. AI treatment inhibited
CYP19A mRNA in trunk during the crucial period of ovarian
cavity formation and CYP19B in gonad and brain by the end
of gonadal sex differentiation. An elevation of testosterone
and 11-ketotestosterone was observed which can be associated
with the down-regulation of the circulating 17-estradiol
production during the AI treatment period. After stopping
AI treatment, both circulating estrogen and androgen
were normalized. The current results suggest that suppression
of CYP19A before and during morphological sex differentiation
inhibits ovarian cavity formation in fugu. Furthermore,
non-detectable limits of 17-estradiol and high
testosterone levels by the end of the gonadal differentiation
period can be ascribed to inhibition of CYP19B, suggesting
that conversion of 17-estradiol from testosterone is plausibly
regulated by CYP19B, and that this factor (CYP19B) may
play an important role in AI-induced testicular development
after gonadal sex differentiation through regulation of the
testosterone–17-estradiol balance in fugu.
Copyright © 2007 S. Karger AG, Basel
Like in all other lower vertebrates, gonadal sex differentiation
in fish is susceptible to any environmental and
steroidogenic stimulation or inhibition. Alteration of the
sex differentiation process is possible in many fishes
through the manipulation of fish environment and steroid
function [Rubin, 1985; Kitano et al., 1999; D’Cotta et
Fax +41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com
© 2007 S. Karger AG, Basel
1661–5425/07/0015–0311$23.50/0
Accessible online at:
www.karger.com/sxd
Michiya Matsuyama
Laboratory of Marine Biology
Faculty of Agriculture, Kyushu University
6-10-1, Hakozaki, Higashi-Ku, Fukuoka 812-8581 (Japan)
Tel. +81 92 642 2887, Fax +81 92 642 2888, E-Mail rinya_m@agr.kyushu-u.ac.jp

al., 2001; Piferrer et al., 2005; Goto-Kazeto et al., 2006].
Some of the chemicals, which have androgenic or estrogenic
functions, are reported to alter gonadal sex in fish
from female to male or vice versa if administered during
the period of gonadal sex differentiation [Papoulias et al.,
2000; Arslan and Phelps, 2004; Hirai et al., 2006] or even
at maturity [Chang et al., 1995; Yeh et al., 2003]. Other
than the direct androgenic or estrogenic effects, there are
chemicals which are capable of disrupting the enzymatic
pathway of natural androgen-estrogen balance in fish
[McAllister and Kime, 2003; Hallgren et al., 2006; Villeneuve
et al., 2006]. Among the non-steroidal aromatase
inhibitors (AI), fadrozole is usually used as an effective
human drug in the treatment of estrogen-dependent disease
including breast cancer and was found to be effective
in suppression of all estrogens after oral administration
[Dowsett et al., 1994]. It has been reported that fadrozole
treatment also results in reduced 17 -estradiol (E2) production
in mammals, viz. monkey, hamster, rabbit and rat
[Shetty et al., 1995; Moudgal et al., 1996]. Induction of
male sex determination has also been reported in reptiles
(e.g., lizard and turtle) administered with fadrozole [Wibbels
and Crews, 1994]. In fish, fadrozole is reported to result
in reduced E2 production [Bhandari et al., 2004; Mandiki
et al., 2005], atresia of ovarian follicles in masculinizing
genetic females [Ankley et al., 2002] and complete
irreversible masculinization [Kitano et al., 2000; Kwon et
al., 2000; Afonso et al., 2001; Fenske and Segner, 2004;
Tzchori et al. , 2004]. In all cases, estrogen synthesis was
found to be significantly reduced in fadrozole-treated
fish.
The balance between androgen (inducers for maleness)
and estrogen (inducers for femaleness) hormones in
higher and lower vertebrates is regulated via an aromatization
reaction. Cytochrome P450 aromatase (P450arom),
a microsomal enzymatic complex, is responsible for this
aromatization reaction which results in irreversible conversion
of androgens into estrogens in most vertebrates
[Simpson et al., 2002]. In most fishes P450arom is a product
of two CYP19 isoforms, predominantly expressed in
ovary (ovarian type aromatase, CYP19A) and brain
(brain type aromatase, CYP19B). Other than local conversion
of androgen into estrogen, CYP19 isoforms have
been found to play important roles in gonadal differentiation
and development in fish. CYP19A has been found
to contribute to ovarian cavity formation and ovarian development
and CYP19B to testicular development and
germ cell maturation in many fishes [reviewed by Devlin
and Nagahama, 2002]. The multifunctional ability of
aromatase CYP19 genes in regulating steroid hormone
balance and gonadal differentiation and development has
made them a key parameter for maintaining sexuality in
fish. Thus, a disruption in aromatase gene function using
an AI (fadrozole) can inhibit both the CYP19 isoforms
and suppress E2 production, resulting in oocyte atresia,
formation of intermediate gonads or testicular differentiation
of genetically female fish.
The current experiment was performed on the Japanese
fugu (Takifugu rubripes) , a gonochoristic tetraodontiform
fish, which lives in the marine waters of Japan,
China and Korea. It is the second vertebrate (the first being
human) whose genome has been sequenced completely
and which is proposed as a useful model for annotating
the human genome due to similarity in the repertoire
of genes [Brenner et al., 1993; Aparicio et al. 2002].
The factors involved in gonadal sex determination and
differentiation of such an important fish species have not
yet been studied well. However, by performing genomewide
linkage analyses, Kikuchi et al. [2007] revealed that
the sex of fugu is determined by a single region on linkage
group 19 in an XX/XY system. As a part of understanding
the mechanism of gonadal sex determination,
differentiation and development of fugu, histological investigation
of the process of gonadal sex differentiation
has been carried out previously in our laboratory and the
course of ovarian and testicular development in this species
was identified [Matsuura et al., 1994; Yamaguchi et
al., 2006]. In a recent study with fugu Dmrt ( Doublesex/
mab3 related transcription factor 1), a transcription factor
that regulates early differentiation of Sertoli cells in
the testes of vertebrates, we have found that it is involved
in gonadal development rather than sex differentiation
and that its expression is correlated with the proliferation
of spermatogonia [Yamaguchi et al., 2006]. So far there is
no detailed information on the effects of different environmental
parameters on the sex determination and differentiation
of fugu. However, a recent study (conducted
in our lab) on the effects of high temperature (29 ° C, environmental
water temperature being 17 ° C) on developing
fugu revealed no effect of high temperature on sexuality
of this fish (unpublished data). Hence, to get a better
understanding of the process of fugu gonadal sex differentiation
together with the role of two CYP19 isoforms in
this process, the current study was aimed at investigating
whether gonadal sex differentiation of this fish is affected
by AI fadrozole which is reported to suppress CYP19 gene
function and disrupt production of E2 in many animals.
312
Sex Dev 2007;1:311–322
Rashid et al.

Materials and Methods
Experimental Fish
Fertilized eggs of fugu were bought from a local fish hatchery
and kept in a tank at the Fisheries Research Laboratory of Kyushu
University situated at Tsuiyazaki, Japan. Most of the eggs hatched
out 12 to 13 days after fertilization. Hatched out larvae were kept
in the hatching tank for 3 days without any supplementary feed.
After the absorption of the yolk sack (3 days after hatching, dah),
they were transferred to a larvae rearing tank. From 4 dah, the
larvae were supplied with live rotifer plus formulated feed (with
or without fadrozole). From 19 dah, decapsulated and nutrified
Artemia were supplied to larvae together with rotifer and formulated
feed. Rotifer supply was stopped on 33 dah and formulated
feed, together with Artemia , was continued until 38 dah. From 39
dah, larvae were supplied only with formulated feed up to the end
of the experiment (9 months after hatching, mah). The fishes were
reared under flow-through seawater and aeration at environmental
temperature throughout the experimental period. The water
temperatures in the experimental tanks were around 14 ° C during
hatching out of fugu larvae, and rose to 24 ° C during the fadrozole
treatment period (up to 100 dah). Formulated feed was supplied
ad libitum.
This experiment was carried out under the guidelines for animal
experiments in the Faculty of Agriculture and Graduate
Course of Kyushu University, and according to the laws (No. 105)
and notifications (No. 6) of the Japanese Government.
AI Treatment
Aromatase inhibitor used in this experiment was fadrozole (4-
(5,6,7,8-tetrahydroimidazole[1,5- ]pyridin-5-yl)-benzonitrile
monohydrochloride hemihydrate), purchased from Novartis
Pharma K.K. (Japan). After dilution in ethanol (100%), the desired
concentration of fadrozole was sprayed over formulated feed
as uniformly as possible. The fadrozole mixed diets were then
dried properly inside a draft. After complete drying, the diets
were stored at 4 ° C in sealed glass vessels for the entire period of
the experiment. Two doses of fadrozole were used for feeding fugu
larvae, i.e. 500 g/g diet (treatment A) and 1000 g/g diet (treatment
B). Control (treatment C) food was prepared by spraying
only ethanol over formulated diet and drying accordingly.
Sampling Procedure
Both AI-treated and control fish were sampled on 35, 42, 70,
100 dah, and 6 and 9 mah. Brain (or head) and gonad (or trunk)
samples of 20 to 30 fish at every stage were collected from anesthetized
fish. Both brain (or head) and gonad (or trunk) samples
for RNA extraction were either preserved in RNA later TM (QIA-
GEN GmbH, Germany) or quick-frozen in liquid nitrogen, later
preserved at –80 ° C. Gonad (or trunk) samples for histological
investigations were fixed in Bouin’s fluid. After fixation for 24 h,
the gonad samples were run through an ethanol series to dehydrate
the tissue. After dehydration, the gonad samples were preserved
at –20 ° C (keeping in 100% ethanol) until they were embedded
in paraffin. Blood samples were collected on 100 dah and
9 mah from the caudal vein using injection syringes. Soon after
collection, blood samples were centrifuged at 4000 rpm for 25
min to separate the serum. Later the serum samples were preserved
in separate test tubes at –20 ° C until assay.
G o n a d Hi s t o l o g y
Histological observations of fadrozole-treated and non-treated
gonads were done on 42 dah (morphological differentiation of
gonad), 70 dah, 100 dah (termination of fadrozole treatment),
6 mah and 9 mah (end of investigation). In every stage, around 20
to 25 gonad samples from each treatment were studied. For histological
observations, gonad samples in 100% ethanol were first
cleared by a lemosole wash for 25 to 30 min and then embedded
in paraffin. Embedded tissues were sectioned at 5 m thickness
and stained by hematoxylin-eosin for observation under light microscope.
RNA Extraction and Reverse Transcription
Total RNA from frozen or RNA later TM -pre s e r ve d br a i n (or
head) and gonad (or trunk) samples was extracted with Isogen
(Nippon Gene Co., Ltd., Japan) following the manufacturer’s protocol.
First-strand cDNA (10 l reaction volume) for each sample
was synthesized from 1 g of total RNA using SuperScript TM III
Reverse Transcriptase (Invitrogen, USA) according to the manufacturer’s
protocol. Oligo-d(T) 16 primer (5 -GGCCACGCGTC-
GAC TAGTACT(T) 16 -3 ) (50 M ) was used for reverse transcription
of mRNA.
Semi-quantitative PCR: Optimization Experiment
Optimization experiments for PCR conditions for fugu CYP-
19A, CYP19B and -actin primers were initially done. Gene-specific
primers for fugu CYP19A and CYP19B were designed from
the partial (pfCYP19A) and full (pfCYP19B) cDNA sequences that
have been isolated in our laboratory (to be published elsewhere).
Primers for -actin, the housekeeping gene that was used as internal
control, were designed from the gene-specific sequence described
in Venkatesh et al. [1996] (GenBank accession no. U37449).
Standard PCR reactions were performed using these gene-specific
primers ( table 1 ). The PCR reactions were performed in 10- l reaction
mixtures containing 10 m M Tris-HCl (pH 8), 1.5 m M MgCl 2 ,
50 m M KCl, 100 M dNTP, 0.5 U Ampli Taq Gold, and 0.5 M of
each primer. The cycling conditions for PCR reactions were 1 min
at 94 ° C, 1 min at 64 ° C, and 2 min at 72 ° C (for pfCYP19A) and
1 min at 94 ° C, 1 min at 58 ° C, and 2 min at 72 ° C (for CYP19B and
-actin). To get the optimal cycle number for each gene-specific
primer set, PCR was conducted using cDNA collected from 9 mah
fugu brain and ovary tissue (tissues predominantly expressing
CYP19B and CYP19A, respectively). Seven cycle numbers (20, 25,
30, 35, 40, 45 and 50 cycles) were used for PCR with each sample.
Products amplified by CYP19A versus -actin and CYP19B versus
-actin primers were run together on the same gel (1.5% TBE agarose
gel stained with ethidium bromide). Gel photographs were
taken by a CCD camera under UV illumination. The signals for
PCR bands were calculated by NIH Image . The peak scores calculated
by NIH image were plotted for CYP19A/-actin and CYP 19B/
-actin to draw standard curves. PCR cycle number for CYP 19A,
CYP19B and -actin were then chosen from the linear line of the
standard curves to be used for the respective gene throughout the
investigation of all stages.
Semi-quantification of CYP19 mRNA Expression in Fugu
Brain and Gonad
For a single tissue, PCR was conducted in a 10- l reaction volume
for all three sets of primers using their best cycling conditions
obtained from optimization experiments. Five microliters
CYP19 in Fugu Sexual Differentiation Sex Dev 2007;1:311–322 313

Table 1. Primer sets used for semiquantitative
RT-PCR analysis of brain
type and ovarian type aromatase
(CYP19B and CYP19A) of T. rubripes
using the -actin gene as internal control
Primer Sequence (5]3) Length of amplified
fragment
pfCYP19A F TCCTGGCTTCGGGGTTCTCATGG
pfCYP19A R CTGTCCAGATGCGCCTGAGTGGAG 353 bp
pfCYP19B F GCTCTGGAGGACGACACCATC
pfCYP19B R GGGATGAACCTCATGGCCAG 579 bp
pf-actin F GCCAACAGGGAGAAGATGACCCAGA
pf-actin R CCAGGGAGGAAGAGGAGGCAGC 370 bp
pf: Pufferfish (fugu); F: forward; R: reverse; bp: base pairs.
Table 2. Sex ratio of control and
AI-treated T. rubripes at different
developmental stages
Stage of
fish a
Fadrozole
treatment b
No. of samples
studied
Sex composition
- U Intermediate
42 dah A 23 11 (48%) 12 (52%)
B 25 11 (44%) 14 (56%)
C 25 13 (52%) 12 (48%)
70 dah A 23 13 (57%) 10 (43%)
B 20 10 (50%) 10 (50%)
C 25 12 (48%) 13 (52%)
100 dah A 24 14 (58%) 10 (42%)
B 24 15 (63%) 9 (37%)
C 24 11 (46%) 13 (54%)
6 mah A 25 25 (100%)
B 23 25 (100%)
C 24 12 (50%) 12 (50%)
9 mah A 21 25 (100%)
B 22 25 (100%)
C 25 14 (56%) 11 (44%)
a
dah: Days after hatching; mah: months after hatching.
b
A: 500 g/g diet; B: 1000 g/g diet; C: control (no fadrozole treatment).
of products from each of the PCR reactions were run on gel. To
avoid gel to gel variation of results, PCR products from a single
cDNA for CYP19A, CYP19B and -actin were run together on the
same gel. Calculation of PCR band intensity was performed in a
way similar to that described in the previous section. Semi-quantification
of CYP19A and CYP19B for every cDNA sample was
done by calculating the ratio of PCR gel band intensity for CYP-
19A/-actin and CYP19B/-actin. Sample number ranged from
15 to 20 for every tissue at every stage.
Serum Steroid Measurement
Serum steroids were measured using ELISA protocol. Estradiol
enzyme immunoassay kit (Cayman Chemical Company, MI,
USA) was used for measuring serum E2, whereas serum testosterone
(T) and 11-ketotestosterone (11-KT) were measured using the
ELISA protocol previously practiced in our laboratory for measuring
T and E1 [Ohta et al. , 2001]. Blood serum collected on 100
dah and 9 mah was measured for steroids and sample size for every
treatment on each stage was 12 to 15.
D a t a An a ly s i s
All the results for semi-quantitative RT-PCR and serum steroid
measurements were analyzed by one-way ANOVA followed
by Tukey’s multiple comparison test up to 5% level of significance
(p ! 0.05).
Results
Gonad Histology
Histological investigations of AI-treated and control
gonads at different stages of development revealed a gradual
trend of AI-treated gonads towards testicular development.
Detailed descriptions on stage-specific histological
structure of treated and control gonads are presented
in the following section and data on the number of inves-
314
Sex Dev 2007;1:311–322
Rashid et al.

A
B
C
D
Fig. 1. Haematoxylin-eosin stained sections
of fugu gonads from 42 dah ( A – C )
and 70 dah ( D – F ). A , D Control testis (arrowheads).
B , E Control ovary. C , F Intermediate
gonad from fadrozole treated fish.
OC, ovarian cavity; DOC, deformed ovarian
cavity; OG, oogonium; SG, spermatogonium;
G, gonium; EL, empty lumina of
tubule. Scale bars, 20 m; in insets 50
m.
E
F
tigated samples and sex ratios of all the stages are compiled
in table 2 .
42 dah: Morphological gonadal sex differentiation of
T. rubripes occurs on 42 dah as investigated in a previous
study at our laboratory [Yamaguchi et al., 2006]. Nontreated
control fish at this stage were composed of female
and male individuals as identified by ovarian cavity (OC,
ovary) formation and no ovarian cavity (testis) ( fig. 1 A,
B). In both treatments, A and B, some of the gonads contained
deformed ovarian cavities (DOC) ( fig. 1 C) compared
to the well-developed OC in control ovary ( fig. 1 B),
and the rest of the gonads resembled control testis (arrowheads
in fig. 1 A).
CYP19 in Fugu Sexual Differentiation Sex Dev 2007;1:311–322 315

A
B
C
D
Fig. 2. Haematoxylin-eosin stained sections
of fugu gonads from 100 dah ( A – C ),
6 mah ( D , E ) and 9 mah ( F ). A Control testis.
B, E Control ovary. C Intermediate gonad
from fadrozole treated fish. D , F Testis
from fadrozole treated fish. OC, ovarian
cavity; DOC, deformed ovarian cavity;
OG, oogonium; OCT, oocyte; PNO, perinucleolar
oocyte; PVO, previtellogenic oocyte;
SG, spermatogonium; SC, spermatocyte;
ST, spermatid; SZ, spermatozoon.
Scale bars, 20 m; in insets 50 m.
E
F
70 dah: At this stage, the control ovary had a well-developed
OC and proliferating oogonia ( fig. 1 E). Control
testes were composed of mostly spermatogonia ( fig. 1 D).
On the other hand, treated fishes (A and B) had testes similar
to controls or intermediate gonads having DOC and
empty lumina (EL) of tubules ( fig. 1 F). The intermediate
gonads were filled with gonia which were histologically
not distinguishable from spermatogonia and oogonia.
100 dah: Control ovaries contained cavities and were
composed of oogonia and perinucleolar oocytes (PNO)
( fig. 2 B). Control testes, on the other hand, had proliferating
spermatogonia ( fig. 2 A). AI-treated fishes exhibited
testes similar to control and intermediate gonads containing
DOC and remains of oocytes (OCT) ( fig. 2 C).
6 mah: Control ovaries consisted mostly of perinucleolar
oocytes (PNO) and previtellogenic oocytes (PVO)
316
Sex Dev 2007;1:311–322
Rashid et al.

Fig. 3. Standard curves and relevant gel
photographs (1.5% TBE agarose gel stained
with ethidium bromide) for semi-quantitative
RT-PCR of fugu CYP19A ( A ) and
CYP19B ( B ) using -actin as an internal
control. M, molecular marker; numbers
(20 to 50) indicate number of PCR cycles;
bp, b a s e p a i r s .
Peak score
700
600
500
400
300
200
100
0
M 20
A
CYP19A
-Actin
20 25 30 35 40 45 50
PCR cycle number
25 30 35 40 45 50
Peak score
M
M
579 bp
(CYP19A)
370 bp
(-Actin)
B
1,200
1,000
800
600
400
200
0
20
CYP19B
-Actin
20 25 30 35 40 45 50
PCR cycle number
25 30 35 40 45 50
M
353 bp
(CYP19B)
370 bp
(-Actin)
( fig. 2 E). On the other hand, all treated gonads resembled
control testis. All testes (both treatment and control)
were found to be undergoing spermatogenesis. Spermatogonia,
spermatocytes and spermatids were observed in
histological sections ( fig. 2 D).
9 mah: In the control ovary, cellular stages were almost
similar to those observed on 6 mah ovary (figure not
shown). In both control and treated testis, the final stage
of spermatogenesis, spermatozoa, were observed in histological
sections together with spermatogonia and spermatocytes
( fig. 2 F).
Optimization of PCR Conditions for
Semi-Quantitative Experiments
The standard curves for CYP19A and CYP19B show
that PCR products up to 40 cycle numbers fall into the
linear line for CYP19A, up to 35 cycle numbers for CYP19B
and up to 35 cycle numbers for -actin (fig. 3 ). From the
linear line of the standard curve we selected 35, 35 and 30
cycle numbers for CYP19A, CYP19B and -actin, respectively.
These cycle numbers were then used for conducting
PCR reactions throughout the investigation of all the
stages.
Semi-quantification of CYP19 mRNA Expression in
Fugu Brain and Gonad
CYP19A mRNA expression was high in non-treated
female ovary throughout the investigation and the highest
expression was found in 9 mah ovary (composed of
PNO and PVO) ( fig. 4 D). In all the stages studied, CYP19A
showed very low expression in brain (or head) of both
sexes ( fig. 4 ). Its expression was even not detectable (ND)
in treated brain on 100 dah ( fig. 4 C). In addition, ‘malelike’
low expression of CYP19A was observed in treated
gonads (or trunks) throughout the study period ( fig. 4 ).
An ND level of CYP19A expression was observed in treated
trunk on 35 dah (before morphological gonadal differentiation)
( fig. 4 A).
CYP19B mRNA was highly expressed in brain (or
head) of non-treated fish (both male and female) throughout
the study period. Together with high expression of
CYP19B in non-treated brain, an elevated expression was
observed in testis at relatively adult stages, compared to
low level in ovary ( fig. 4 C, D). In AI-treated brain (or
head), CYP19B expression was low compared to control
but was still higher than that of CYP19A. CYP19B was
also low in treated gonads (or trunks) in all stages compared
to control testis but was expressed at higher level
compared to CYP19A. CYP19B was upregulated (towards
normalization) in gonads after withdrawing fadrozole
application (measured on 9 mah) but was maintained at
low level compared to control testis ( fig. 4 ).
Serum Steroid Level
ELISA (E2, 11-KT and T) experiments revealed low
circulating estrogen (E2) and higher androgen (11-KT
and T) levels throughout the experimental period. In
treatment A and B on 100 dah (termination of fadrozole
exposure), a massive suppression of E2 production was
observed evident by non-detectable limit (NDL) of E2 in
this stage compared to detectable levels in control female
and male (27 pg/ml in female and 16 pg/ml in male). After
9 mah, there were no significant differences of E2
levels between control and treatment although the highest
level was detected in control female (23 pg/ml)
( fig. 5 A). By the end of the fadrozole exposure period
CYP19 in Fugu Sexual Differentiation Sex Dev 2007;1:311–322 317

0.6
a
0.6
a
0.5
0.5
x
CYP19/ -actin
A
0.4
0.3
0.2
0.1
0
z
A
x
y
z
ND ND
B C A B C
Head Trunk
CYP19A
b
A
b, c
c
d
d
B C A B C
Head Trunk
CYP19B
CYP19/ -actin
B
0.4
0.3
0.2
0.1
0
y
A
y
y y y
y y
B CM CF A B CM CF
Head Trunk
CYP19A
c
A
b
b, c
c
c
c
c
B CM CF A B CM CF
Head Trunk
CYP19B
CYP19/ -actin
C
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
y, z y, z y
x
ND ND z z
A B CM CF A B CM CF
Brain Gonad
CYP19A
c
A
a
a
b
c
d
B CM CF
d
A
d
B CM CF
Brain Gonad
CYP19B
CYP19/ -actin
D
1.2
1.1
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
z
A
x
y
y, z
z
z
z
z
B CM CF A B CM CF
Brain Gonad
CYP19A
a
A
a
a
a
a
b
b
c
B CM CF A B CM CF
Brain Gonad
CYP19B
Fig. 4. Relative expression of CYP19A and CYP19B in fugu brain/head and gonad/trunk on 35 dah ( A ), 42 dah
( B ), 100 dah ( C ), and 9 mah ( D ). A , f a d ro z ole t re at me nt A (5 0 0 g/g diet); B, fadrozole treatment B (1000 g/g
diet); C, control; CM, control male; CF, control female; ND, not detected. Values above bars indicate significant
differences (p ! 0.05) compared within the data. Significant values are plotted separately for CYP19A (x, y, z)
and CYP19B (a, b, c, d).
(100 dah), 11-KT levels in treated fishes were almost
twice as much as in control males, the highest level being
in treatment A (494 pg/ml). The 11-KT level decreased in
9 mah AI treatment when it had no significant difference
with control male ( fig. 5 B). The highest level of steroid
detected in treated fish during the AI exposure period
was T, as observed in treatment B (1727 pg/ml) on 100
dah. This T value was almost double the value in control
male at this stage. After withdrawing fadrozole application,
there was a significant drop of T level in both AI
treatments towards normalization (measured on 9 mah)
(fig. 5C).
Discussion
Oocyte atresia, formation of intermediate gonads and
finally testicular development of genetic females have
been observed in both gonochoristic and hermaphroditic
fishes by the effect of AI fadrozole [Kwon et al., 2000;
Afonso et al., 2001; Fenske and Segner, 2004; Suzuki et
al., 2004; Tzchori et al., 2004; Komatsu et al., 2006]. Not
only fadrozole but also other AIs and exogenous androgens
were found to influence testicular differentiation in
fishes such as to lead to masculinization [Piferrer et al.,
1994; Feist et al., 1995; Kitano et al., 2000]. The mecha-
318
Sex Dev 2007;1:311–322
Rashid et al.

Serum E2 (pg/ml)
40
30
20
10
b
a
b
b
b
a, b
Serum 11-KT (pg/ml)
600 2,000
a
a
a
a
450 1,500
c, d
150 500
c
d
Serum T (pg/ml)
300 1,000
b
b, c
b, c b, c
b
c
b, c
c, d
d
A
0
NDL NDL
A B CM CF A B CM CF
100 dah 9 mah
B
0 0
A B CM CF A B CM CF
A B CM CF A B CM CF
100 dah 9 mah C
100 dah 9 mah
Fig. 5. Serum steroid levels of fadrozole treated and control fugu on 100 dah and 9 mah. A Serum E2, B serum
11-KT, C s e r u m T. A , f a d ro z ole t re at me nt A (5 0 0 g/g diet); B, fadrozole treatment B (1000 g/g diet); CM, control
male; CF, control female. Values above the bars indicate significant differences (p ! 0.05) compared within
the data.
Fig. 6. Diagrammatic representation of the interaction between
AI, CYP19 and sex steroids in the process of testicular differentiation
of T. rubripes . Top arrow shows different stages of gonadal
sex differentiation, considering 42 dah as the point of morphological
gonadal sex differentiation. ‘Pre-differentiation’ refers to
the stage where no morphological distinction between testis and
ovary is possible; ‘differentiation’ refers to the period of gonadal
sex differentiation and development from 42 dah; and ‘post-differentiation’
refers to the remaining stage after gonadal sex differentiation
is completed (i.e., premature stage). From the highest
to lowest degree of expression, small arrows inside the text boxes
indicate: t upregulated level; r normalized level; w lowered level;
y suppressed level; X inhibited level. dah, days after hatching;
m a h , mont h s a f t e r h at c h i n g .
nism of AI-induced testicular differentiation through
disruption of aromatase function has been found to be
controlled in different ways in different fishes. In zebrafish,
for example, irreversible masculinization was
achieved by manipulation of the aromatase system by
fadrozole during the critical period of sexual differentiation
[Fenske and Segner, 2004]. AI-induced sex inversion
in red-spotted grouper was attributed to the inhibition of
P450 gene expression and aromatase activity and the resultant
decrease in the biosynthesis of endogenous E2
[Li et al., 2006]. In the case of golden rabbitfish, estrogen
was predicted to be involved in ovarian differentiation
and AI-induced suppression of estrogen was found to be
an essential prerequisite for testicular differentiation
CYP19 in Fugu Sexual Differentiation Sex Dev 2007;1:311–322 319

[Komatsu et al., 2006]. In fugu, similar to the studies
mentioned above, AI (fadrozole) treatment during the
period of gonadal sex differentiation was found to induce
testicular differentiation towards irreversible masculinization.
In the current study, the AI-treated masculinizing
fugu had deformed ovarian cavities (DOC), empty
lumina of tubules (EL) and remains of oocytes (OCT) inside
intermediate gonads at different stages of development.
They were predicted to be sex-changing (from genetic
female to morphological male) fishes and were
clearly distinguished from those containing testis resembling
that of control male, a case similar to that in AItreated
masculinization of rabbitfish. Moreover, appearance
of DOC in some of the AI-treated gonads on 42 dah
(time when morphological gonadal sex differentiation is
first evident in fugu) suggests that AI started inhibiting
ovarian cavity (OC) formation in female fugu before gonadal
sex differentiation. The inhibition of OC formation
during this crucial period of gonadal sex differentiation
and OC formation (35 to 42 dah) can be correlated with
a massive suppression of CYP19A mRNA in AI-treated
trunk at this stage (ND level on 35 dah ( fig. 4 A) and suppressed
level on 42 dah ( fig. 4 B) of treated trunk compared
to higher level in relevant female trunk). This correlation
suggests that OC formation in AI-treated fugu
may have been inhibited by the suppression of CYP19A
mRNA due to AI treatment. The involvement of CYP19A
in ovarian differentiation and development in several
fish species has been related to its higher expression in
ovary during the period of gonadal sex differentiation
compared to the low level in testis, and the subsequent
high aromatase activity responsible for the synthesis of
estrogen [reviewed by Devlin and Nagahama, 2002]. For
instance, elevation of aromatase expression was reported
in rainbow trout ovaries before histological sex differentiation
[Guiguen et al., 1999]. CYP19A expression was
found to be higher in females compared to males in Nile
tilapia and European sea bass during the early stage of sex
differentiation [Kwon et al., 2001; Piferrer et al., 2005]. In
medaka, aromatase mRNA was detected in ovary before
and during gonadal sex differentiation and treatment
with AI (fadrozole) during ovarian development was
found to suppress OC formation [Suzuki et al., 2004].
W h i le CYP19A is predominantly expressed in ovary
and predicted to play an important role in ovarian differentiation
of fugu, CYP19B showed high level of expression
in control brain/head (of both sexes) throughout the
study period. Moreover, other than high CYP19B expression
in the brain of fugu, as found in most of the teleosts
[Fukada et al., 1996; Trant et al., 1997; Kishida and Callard,
2001; Fenske and Segner, 2004; Chang et al., 2005],
an exceptionally high expression of CYP19B was also evident
in non-treated fugu testis throughout the study period,
compared to the low level of CYP19A ( fig. 4 ). In addition,
in our recent study (to be published elsewhere) we
have found positive signals (using in situ hybridization
protocol) of pfCYP19B mRNA in Sertoli cells (juvenile
fugu; 1 year old) and spermatids (mature fugu; more than
2 years old) of fugu testis which is very similar to the situation
described for rat testis [Carpino et al., 2001].
CYP19 was also found to be localized in Leydig and Sertoli
cells and maturing spermatocytes of testis in other
mammals [Bourguiba et al., 2003; Carreau et al., 2003].
Until now, there is no report of such a high expression of
CYP19B in the testis of any teleost, nor even any report
on different sites of action at different stages of development
and maturity. Up to now, PCR-based investigations
found only lower levels of CYP19B expression in teleost
testes [Kwon et al., 2001; Fenske and Segner, 2004; Chang
et al., 2005; Choi et al., 2005]. The exceptionally high expression
of CYP19B in the testis of juvenile and adult
fugu, together with specific expression in Sertoli cells of
juvenile and in spermatids of mature testis, indicates a
possible role in regulating endogenous steroid levels (T-
E2 balance) and probable involvement in the development
and maturation of testicular cells like in the case of
mammals.
In AI-treated gonad and brain of fugu, a massive suppression
of CYP19B mRNA was observed by the end of
the gonadal sex differentiation period (measured on 100
dah). This suppression can be correlated with high serum
levels of T and 11-KT and non-detectable limits (NDL) of
E2 at this stage. The highest serum T level at this stage
may have been achieved by little or no conversion of T
into E2, due to the suppressed level of CYP19B mRNA
both in gonad and in brain of AI-treated fish. It is remarkable
here that withdrawal of AI-treatment resulted
in the normalization of CYP19B in gonads (all testes;
measured on 9 mah) and brain associated with the normalization
of serum androgen (T and 11-KT) and E2 levels.
In agreement with our study, a reduced aromatase
activity in brain associated with a decrease in plasma E2
and preovulatory follicle atresia was observed in adult female
of fathead minnow treated with fadrozole [Ankley
et al., 2002]. And in developing zebrafish [Fenske and
Segner, 2004], treated with fadrozole and methyl-testosterone
(MT), a strong correlation between aromatase
gene expression, E2 production and sexual differentiation
was observed, suggesting that gonadal differentiation
in zebrafish depends on the androgen-estrogen bal-
320
Sex Dev 2007;1:311–322
Rashid et al.

ance catalyzed by aromatase. In our study, the correlated
expression of CYP19B and serum steroids in fugu, similar
to the above exemplified phenomena, suggests that aromatase
CYP19B isoform may be involved in the regulation
of T-E2 balance in this fish; and the suppression of
CYP19B by AI-treatment at the end of the gonadal sex
differentiation process and resultant imbalance in circulating
sex steroids was probably responsible for AI-induced
testicular differentiation of developing fugu
(fig. 6 ).
Co n c l u s i o n
The present investigation is the first in vivo study on
a gonochoristic marine teleost to compare and compile
histological evidences, CYP19 mRNA expressions and serum
steroid profiles after administering AI fadrozole. We
have found that treatment of fugu with AI, fadrozole,
from their ‘first feeding’ stage through the period of gonadal
sex differentiation can block OC formation of female
fish followed by testicular differentiation of all
treated fish. During the fadrozole treatment period, suppression
of CYP19A before gonadal sex differentiation resulted
in the inhibition of OC formation. Suppression of
CYP19B by the end of gonadal sex differentiation and the
resulting imbalance in circulating sex steroids (E2, T, 11-
KT) contributed to the testicular differentiation of genetically
female fish. Withdrawal of AI application after
the gonadal sex differentiation period resulted in the normalization
of CYP19 isoforms and serum steroid levels.
These findings suggest that gonadal sex differentiation
and development of fugu are differently regulated by
CYP19 isoforms together with T-E2 balance as catalyzed
by aromatase enzyme. Moreover, the molecular and endocrine
mechanism for generation of all-male populations
of fugu, one of the priced aquaculture species of
East Asia, has been uncovered. However, extensive in
vivo and in vitro studies on aromatase gene function and
related genetic, molecular and endocrine parameters are
required, to understand detailed mechanisms of gonadal
sex determination and differentiation of this species.
Acknowledgements
The authors are grateful to the students of the Laboratory of
Marine Biology, Kyushu University, for assistance during the experimental
period. Thanks to Dr. Yoshimura at the Fukuoka Prefectural
Institute for Fish Stock Enhancement, Fukuoka, for providing
fertilized eggs and technical advise for raising fugu.
This study was supported in part by a Grant-in-Aid (19580207)
for Scientific Research from the Ministry of Education, Science,
Sports and Culture of Japan.
The nucleotide sequence data for pfCYP19A and pfCYP19B
reported in this paper will appear in the DDBJ/EMBL/GenBank
nucleotide sequence databases with the accession numbers
AB330136 and AB330137, respectively.
References
Afonso LOB, Wassermann GJ, Oliveira RTD:
Sex reversal in Nile tilapia (Oreochromis niloticus)
using a nonsteroidal aromatase inhibitor.
J Exp Zool 290: 177–181 (2001).
Ankley GT, Kahl MD, Jensen KM, Hornung
MW, Korte JJ, et al: Evaluation of the aromatase
inhibitor fadrozole in a short-term reproduction
assay with the fathead minnow
(Pimephales promelas) . Toxicol Sci 67: 121–
130 (2002).
Aparicio S, Chapman J, Stupka E, Putnam N,
Chia JM, et al: Whole-genome shotgun assembly
and analysis of the genome of Fugu
rubripes . Science 297: 1301–1310 (2002).
Arslan T, Phelps RP: Production of monosex
male black crappie, Pomoxis nigromaculatus
, populations by multiple androgen immersion.
Aquaculture 234: 561–573 (2004).
Bhandari RK, Higa M, Nakamura S, Nakamura
M: Aromatase inhibitor induces complete
sex change in the protogynous honeycomb
grouper (Epinephelus merra) . Mol Reprod
Dev 67: 303–307 (2004).
Bourguiba S, Genissel C, Lambard S, Bouraima
H, Carreau S: Regulation of aromatase gene
expression in Leydig cells and germ cells.
J Steroid Biochem Mol Biol 86: 335–343
(2003).
Brenner S, Elgar G, Sandford R, Macrae A, Venkatesh
B, et al: Characterization of the pufferfish
(Fugu) genome as a compact model
vertebrate genome. Nature 366: 265–268
(1993).
Carpino A, Pezzi V, Rago V, Bilinska B, Ando S:
Immunolocalization of cytochrome P450
aromatase in rat testis during postnatal development.
Tissue Cell 33: 349–353 (2001).
Carreau S, Lambard S, Delalande C, Denis-Galeraud
I, Bilinska B, et al: Aromatase expression
and role of estrogens in male gonad: a
review. Reprod Biol Endocrinol 1: 35–40
(2003).
Chang CF, Lau EL, Lin BY: Estradiol-17 suppresses
testicular development and stimulates
sex reversal in protandrous black porgy,
Acanthopagrus schlegeli . Fish Physiol Biochem
14: 481–488 (1995).
Chang X, Kobayashi T, Senthilkumaran B, Kobayashi-Kajura
H, Sudhakumari CC, et al:
Two types of aromatase with different encoding
genes, tissue distribution and developmental
expression in Nile tilapia (Oreochromis
niloticus) . Gen Comp Endocrinol
141: 101–115 (2005).
Choi JY, Park JG, Jeong HB, Lee YD, Takemura
A, et al: Molecular cloning of cytochrome
P450 aromatases in the protogynous wrasse,
Halichoeres tenuispinis . Comp Biochem
Physiol B 141: 49–59 (2005).
D’Cotta H, Fostier A, Guiguen Y, Govoroun MS,
Baroiller JF: Aromatase plays a key role during
normal and temperature-induced sex
differentiation of tilapia Oreochromis niloticus.
Mol Reprod Dev 59: 265–276 (2001).
Devlin RH, Nagahama Y: Sex determination and
sex differentiation in fish: An overview of genetic,
physiological, and environmental influences.
Aquaculture 208: 191–364 (2002).
CYP19 in Fugu Sexual Differentiation Sex Dev 2007;1:311–322 321

Dowsett M, Smithers D, Moore J, Trunet PF,
Coombes RC, et al: Endocrine changes with
the aromatase inhibitor fadrozole hydrochloride
in breast cancer. Eur J Cancer
30A:1453–1458 (1994).
Feist G, Yeoh CG, Fitzpatrick MS, Schreck CB:
The production of functional sex-reversed
male rainbow trout with 17 -methyltestosterone
and 11-hydroxyandrostenedione.
Aquaculture 131: 1–2 (1995).
Fenske M, Segner H: Aromatase modulation alters
gonadal differentiation in developing
zebrafish (Danio rerio) . Aquat Toxicol 67:
105–126 (2004).
Fukada S, Tanaka M, Matsuyama M, Kobayashi
D, Nagahama Y: Isolation, characterization,
and expression of cDNAs encoding the medaka
(Oryzias latipes) ovarian follicle cytochrome
P-450 aromatase. Mol Reprod Dev
45: 285–290 (1996).
Goto-Kazeto R, Abe Y, Masai K, Yamaha E, Adachi
S, et al: Temperature-dependent sex differentiation
in goldfish: Establishing the
temperature-sensitive period and effect of
constant and fluctuating water temperatures.
Aquaculture 254: 617–624 (2006).
Guiguen Y, Baroiller J-F, Ricordel M-J, Iseki K,
McMeel OM, et al: Involvement of estrogens
in the process of sex differentiation in two
fish species: the rainbow trout (Oncorhynchus
mykiss) and a tilapia (Oreochromis niloticus)
. Mol Reprod Dev 54: 154–162
(1999).
Hallgren SLE, Linderoth M, Olsén KH: Inhibition
of cytochrome P450 brain aromatase reduces
two male specific sexual behaviours in
the male Endler guppy (Poecilia reticulata) .
Gen Comp Endocrinol 147: 323–328 (2006).
Hirai N, Nanba A, Koshio M, Kondo T, Morita
M, et al: Feminization of Japanese medaka
(Oryzias latipes) exposed to 17 -estradiol:
Formation of testis–ova and sex-transformation
during early-ontogeny. Aquat Toxicol
77: 78–86 (2006).
Kikuchi K, Kai W, Hosokawa A, Mizuno N, Suetake
H, et al: The sex-determining locus in
the tiger pufferfish, Takifugu rubripes . Genetics
175: 2039–2042 (2007).
Kishida M, Callard GV: Distinct cytochrome
P450 aromatase isoforms in zebrafish (Danio
rerio) brain and ovary are differentially
programmed and estrogen regulated during
early development. Endocrinology 142: 740–
750 (2001).
Kitano T, Takamune K, Kobayashi T, Nagahama
Y, Abe SI: Suppression of P450 aromatase
gene expression in sex-reversed males produced
by rearing genetically female larvae at
a high water temperature during a period of
sex differentiation in the Japanese flounder
(Paralichthys olivaceus) . J Mol Endocrinol
23: 167–176 (1999).
Kitano T, Takamune K, Nagaham Y, Abe S: Aromatase
inhibitor and 17 -methyltestosterone
cause sex-reversal from genetical females
to phenotypic males and suppression
of P450 aromatase gene expression in Japanese
flounder (Paralichthys olivaceus) . Mol
Reprod Dev 56: 1–5 (2000).
Komatsu T, Nakamura S, Nakamura M: Masculinization
of female golden rabbitfish Siganus
guttatus using an aromatase inhibitor
treatment during sex differentiation. Comp
Biochem Physiol C 143: 402–409 (2006).
Kwon JY, Haghpanah V, Kogson-Hurtado LM,
McAndrew BJ, Penman DJ: Masculinization
of genetic female Nile tilapia (Oreochromis
niloticus) by dietary administration of an
aromatase inhibitor during sexual differentiation.
J Exp Zool 287: 46–53 (2000).
Kwon JY, McAndrew BJ, Penman DJ: Cloning of
brain aromatase gene and expression of
brain and ovarian aromatase genes during
sexual differentiation in genetic male and female
Nile tilapia Oreochromis niloticus . Mol
Reprod Dev 59: 359–370 (2001).
Li G-L, Liu X-C, Zhang Y, Lin H-R: Gonadal development,
aromatase activity and P450 aromatase
gene expression during sex inversion
of protogynous red-spotted grouper Epinephelus
akaara (Temminck and Schlegel)
after implantation of the aromatase inhibitor,
fadrozole. Aquacult Res 37: 484–491
(2006).
Mandiki SNM, Babiak I, Bopopi JM, Leprieur F,
Kestemont P: Effects of sex steroids and their
inhibitors on endocrine parameters and gender
growth differences in Eurasian perch
(Perca fluviatilis) juveniles. Steroids 70: 85–
94 (2005).
Matsuura S, Naito T, Shincho M, Yoshimura K,
Matsuyama M: Gonadal sex differentiation
in tiger puffer, Takifugu rubripes [in Japanese,
with English abstract]. Suisanzoshoku
42: 619–625 (1994).
McAllister BG, Kime DE: Early life exposure to
environmental levels of the aromatase inhibitor
tributyltin causes masculinisation and
irreversible sperm damage in zebrafish (Danio
rerio) . Aquat Toxicol 65: 309–316 (2003).
Moudgal NR, Shetty G, Selvaraj N, Bhatnagar
AS: Use of a specific aromatase inhibitor for
determining whether there is a role for oestrogen
in follicle/oocyte maturation, ovulation
and preimplantation embryo development.
J Reprod Fertil Suppl 50: 69–81
(1996).
Ohta K, Mine T, Yamaguchi A, Matsuyama M:
Steroidogenic pathway of estradiol-17 synthesis
in the ovarian follicles of the protogynous
wrasse, Pseudolabrus sieboldi . Zool Sci
18: 937–945 (2001).
Papoulias DM, Noltie DB, Tillitt DE: Effects of
methyl testosterone exposure on sexual differentiation
in medaka, Oryzias latipes. Mar
Env Res 50: 181–184 (2000).
Piferrer F, Zanuy S, Carrillo M, Solar II, Devlin
RH, et al: Brief treatment with an aromatase
inhibitor during sex differentiation causes
chromosomally female salmon to develop as
normal, functional males. J Exp Zool 270:
255–262 (1994).
Piferrer F, Blazques M, Navarro L, Gonzalez A:
Genetic, endocrine, and environmental
components of sex determination and differentiation
in the European sea bass ( Dicentrarchus
labrax L.). Gen Comp Endocrinol
142: 102–110 (2005).
Rubin DA: Effect of pH on sex ratio in cichlids
and a poeciliid (Teleostei). Copeia 1: 233–235
(1985).
Shetty G, Bhatnagar AS, Moudgal NR: Blockade
of estrogen synthesis with an aromatase inhibitor
affects luteal function of the pseudopregnant
rat. J Steroid Biochem Mol Biol 55:
347–353 (1995).
Simpson ER, Clyne C, Rubin G, Boon WC, Robrtson
K, et al: Aromatase – a brief overview.
Annu Rev Physiol 64: 93–127 (2002).
Suzuki A, Tanaka M, Shibata N: Expression of
aromatase mRNA and effects of aromatase
inhibitor during ovarian development in
the medaka, Oryzias latipes . J Exp Zool
301A:266–273 (2004).
Trant JM, Lehrter J, Gregory T, Nunez S, Wunder
J: Expression of cytochrome P450 aromatase
in the channel catfish, Ictalurus punctatus
. J Steroid Biochem Mol Biol 61: 393–397
(1997).
Tzchori I, Zak T, Sachs O: Masculinization of genetic
females of the common carp ( Cyprinus
carpio L.) by dietary administration of an
aromatase inhibitor. Israeli J Aquacult –
Bamidgeh 56: 239–246 (2004) .
Venkatesh B, Tay BH, Elgar G, Brenner S: Isolation,
characterization and evolution of nine
pufferfish (Fugu rubripes) actin genes. J Mol
Biol 259: 655–665 (1996).
Villeneuve DL, Knoebl I, Kahl MD, Jensen KM,
Hammermeister DE, et al: Relationship between
brain and ovary aromatase activity
and isoform-specific aromatase mRNA expression
in the fathead minnow (Pimephales
promelas) . Aquat Toxicol 76: 353–368 (2006).
Wibbels T, Crews D: Putative aromatase inhibitor
induces male sex determination in a female
unisexual lizard and in a turtle with
temperature-dependent sex determination. J
Endocrinol 141: 295–299 (1994).
Yamaguchi A, Lee KH, Fujimoto H, Kadomura
K, Yasumoto S, et al: Expression of the DMRT
gene and its roles in early gonadal development
of the Japanese pufferfish Takifugu rubripes
. Comp Biochem Physiol D 1: 59–68
(2006).
Yamamoto T: Sex differentiation, in Hoar W,
Randall D (eds): Fish Physiology, vol. III, pp
117–175 (Academic Press, New York 1969).
Yeh S-L, Kuo C-M, Ting Y-Y, Chang C-F: The effects
of exogenous androgens on ovarian development
and sex change in female orangespotted
protogynous grouper, Epinephelus
coioides . Aquaculture 218: 729–739 (2003).
322
Sex Dev 2007;1:311–322
Rashid et al.