Journal of Fish Biology (2013) 83, 295–310
doi:10.1111/jfb.12164, available online at wileyonlinelibrary.com
Expression profile of oestrogen receptors and
oestrogen-related receptors is organ specific and sex
dependent: the Japanese medaka Oryzias latipes model
N. K. M. Cheung*†, A. C. K. Cheung*†, R. R. Ye*†, W. Ge†‡,
J. P. Giesy*§!¶** and D. W. T. Au*†††
*Department of Biology and Chemistry, City University of Hong Kong, 83 Tat Chee Avenue,
Kowloon, Hong Kong SAR, †State Key Laboratory in Marine Pollution, City University of
Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong SAR, ‡School of Life Sciences, The
Chinese University of Hong Kong, New territories, Hong Kong SAR, §Department of
Veterinary Biomedical Sciences, Toxicology Centre, 44 Campus Drive, Saskatoon, SK, S7N
5B3, Canada, !Department of Zoology and Center for Integrative Toxicology, Michigan State
University, East Lansing, MI 48824, U.S.A., ¶ School of Biological Sciences, Kadoorie
Biological Sciences Building, The University of Hong Kong, Pok Fu Lam Road, Hong Kong
SAR and **School of Environment, Nanjing University, 22# Hankou Road, Nanjing, Jiangsu,
China
(Received 15 August 2012, Accepted 1 May 2013)
Gene expression of all known subtypes of oestrogen receptor (ER) and oestrogen-related receptor
(ERR) in multiple organs and both sexes of the Japanese medaka Oryzias latipes was profiled and
systematically analysed. As revealed by statistical analyses and low-dimensional projections, the
expressions of ERRs proved to be organ and sex dependent, which is in contrast with the ubiquitous
nature of ERs. Moreover, expressions of specific ERR isoforms (ERRγ 1, ERRγ 2) were strongly
correlated with that of all ERs (ERα, ERβ1 and ERβ2), suggesting the existence of potential
interactions. Findings of this study shed light on the co-regulatory role of particular ERRs in
oestrogen-ERs signalling and highlight the potential importance of ERRs in determining organ and
sex-specific oestrogen responses. Using O. latipes as an alternative vertebrate model, this study
provides new directions that call for collective efforts from the scientific community to unravel the
mechanistic action of ER-ERR cross-talks, and their intertwining functions, in a cell and sex-specific
manner in vivo.
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles
Key words: oestrogen regulation; oestrogenic effects; organs; teleost; vertebrate.
INTRODUCTION
While oestrogen (E2) is generally perceived to be a feminine hormone, its effects are
not restricted to females. Besides functioning in the development of ovary and secondary characteristics of females, oestrogen also drives the development of testes as
well as male secondary characteristics and behaviour (Eddy et al ., 1996; Hess et al .,
1997; Muramatsu & Inoue, 2000). Oestrogen is not only active in the reproductive
axis, but is also involved in regulating the central nervous (McEwen & Alves, 1999)
††Author to whom
bhdwtau@cityu.edu.hk
correspondence
should
be
addressed.
295
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles
Tel.:
+852
3442
9710;
email:
296
N. K. M. CHEUNG ET AL.
and cardiovascular systems (Stampfer et al ., 1991; Mendelsohn & Karas, 1999), cell
growth and differentiation (Brannvall, 2002), immune function (Grossman, 1985)
and skeletal development (Bilezikian et al ., 2001). Oestrogen is thus active in maintaining homeostasis, as well as development and function of the entire spectrum of
organs in vertebrates.
Oestrogen exerts its effects mainly through interactions with oestrogen receptors
(ER), a class of ligand-dependent receptors in the superfamily of nuclear receptors. Once ERs are activated by binding of oestrogen agonists to the ligand binding
domain, the ligand–receptor complex is capable of initiating transcription of over a
thousand target genes (Tang et al ., 2007) by binding to the promoter regions either
directly on oestrogen response elements as homo- or hetero-dimer, or indirectly as
a complex with other transcription factors (Björnström & Sjöberg, 2005).
The oestrogen-related receptors (ERR), a class of orphan nuclear receptors bearing
high structural homology (particularly in the DNA-binding domain) to ERs, has
been postulated to influence E2 signalling by either synergizing or competing with
ERs (Horard & Vanacker, 2003) in regulating multiple shared transcriptional targets
(Vanacker et al ., 1999). Consequently, complete understanding of the mechanisms
and functioning of oestrogen signalling cascade will not be feasible without thorough
consideration of the direct and indirect involvement of ERRs. The importance of
ERRs in E2 responses is already recognized, particularly in clinical applications
(Giguère, 2002).
Teleosts have been extensively utilized as model organisms to elucidate effects of
oestrogenic compounds that can modulate sex hormone responses, such as oestrogenic endocrine disruptors, in vertebrates (Ankley et al ., 1997). While ER sequences
of over 90 teleost species are identified and deposited in the NCBI Genbank, the
sequences of ERRs have been described for only three bony fishes. Such a contrast
reflects a general scarcity of information on teleost ERRs, impeding furtherance of
the understanding on the teleostean oestrogen signalling system and its dynamics in
response to synthetic oestrogenic chemicals.
The Japanese medaka Oryzias latipes (Temminck & Schlegel 1846) is one of
the most frequently utilized teleost models for toxicity testing of xenoestrogens and
for generalization of the biological effects of oestrogenic compounds in vertebrates
(Park et al ., 2008; Zhang et al ., 2008a; Tompsett et al ., 2009). This report
summarizes the first systematic illustration of organ and sex dependent expressions
of all known ER and ERR subtypes in six major organs (brain, gill, gonad, heart,
liver and spleen) of both male and female O. latipes. Given the inconsistency in
nomenclature of ERs and ERRs among teleost species, a phylogenetic tree was
constructed to enable stringent comparisons between the expression profiles of O.
latipes and other teleost species as well as mammalian models. Statistical analyses
were employed to identify potential ER–ERR ‘cross-talks’ for further investigation
of their biological significance in vertebrates.
MATERIALS AND METHODS
F I S H C U LT U R E A N D C A R E
Orange-red strain O. latipes were maintained in aquaria at the City University of Hong
Kong under static conditions (26◦ C, range ±1◦ C; pH 7·3, range ±0·1; 7·2 mg O2 l−1 ,
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
ORGAN AND SEX-SPECIFIC EXPRESSION OF ER AND ERR
297
range ±0·2 mg O2 l−1 ; 14L:10D). Half of the water in each aquarium was replaced daily
with dechlorinated, charcoal-filtered tap water. Fish were fed twice everyday with Otohime
β1 (Marubeni Nisshin Feed Co.; www.mn-feed.com) and supplemented with brine shrimp
Artemia sp. nauplii (Lucky Brand, O.S.I. Marine Lab; www.oceanstarinternational.com) for
3 days every week. Under these husbandry conditions, the fish reached sexual maturity at c.
3 month-post-hatch and laid eggs daily. European Union directive 86/609/EEC was strictly
adhered throughout the study. Animal handling procedures as mentioned in this study were
accepted by the Animal Ethics Committee, City University of Hong Kong.
SAMPLING
Sexually mature O. latipes (8 month-post-hatch; 14 males and 14 females) were collected at
6 h after mating and ovulation. Fishes were completely anaesthetized within 15 s by immersion
in ice-cold aquarium water. Sedated fish was placed on ice bed and decapitated by making a
transverse cut from the gill slit to the spinal cord (immediately caudal to the vagal lobe). The
six target organs isolated from the decapitated fish were liver, spleen, gill (both arches), heart
(atrium and ventricle), brain (rostral-caudal: from telencephalon to the decapitation position;
ventral-most: hypothalamus and inferior lobe) and gonad (i.e. ovary or testis; whole organ
was encapsulated by the ovarian or testicular epithelium). Organs from two individuals of the
same sex were pooled as one replicate, i.e. a total of seven replicates were analysed for each
organ for each sex. Samples were snap-frozen in liquid nitrogen and stored at −80 ◦ C.
T O TA L R N A E X T R A C T I O N A N D F I R S T - S T R A N D c D N A
SYNTHESIS
Total RNA was extracted by the use of TRIzol Reagent (Invitrogen; www.invitrogen.com)
according to the methods in the manufacturer’s manual. Two microlitres of each of the total
RNA samples was electrophoresed on 1·2% agarose gel in 1× tris-acetate-EDTA buffer with
GelRed (Biotium; www.biotium.com). RNA integrity was assured by the presence of sharp
28S and 18S rRNA bands in a ratio of c. 2:1. Another 1 µg of the total RNA was treated
with RQ1 RNase-Free DNase (Promega; www.promega.com) according to the manufacturer’s
protocol. First-strand cDNA was generated from the digestion product following the methods
of Yu et al . (2006).
SELECTION OF REFERENCE GENE
To maximize the use of the expression profiles as basal reference for future mechanistic
studies of oestrogenic compounds, rpl7 , a house-keeping gene deemed very stable upon E2
exposure (Zhang & Hu, 2007), was specifically chosen as the reference gene. To ascertain rpl7
is an appropriate endogenous control gene for the present absolute quantitative real-time PCR
(qPCR) study, expressions of three other commonly used reference genes (including mt-rnr2 ,
actb and gapdh) were also determined in all organs and both sexes of O. latipes (using qPCR
as illustrated below). Expression stability across organs and between sexes was gauged using
NormFinder (Andersen et al ., 2004). rpl7 was proven the most stable reference candidate.
A B S O L U T E Q U A N T I F I C AT I O N O F G E N E E X P R E S S I O N B Y
R E A L - T I M E P C R ( Q RT P C R )
Gene-specific primers were designed using National Center for Biotechnology Information
primer basic local alignment search tool (BLAST). Default parameters were used, except that
specificity check was conducted for ‘Oryzias latipes (taxid:8090)’ against the nr database.
Synthesized primers (1st BASE; www.base-asia.com) were stringently verified for proper
efficiency and specificity under routine reaction setup stated below. Sequences of the validated
gene-specific primer are listed in Table I.
Gene-specific PCR products of the target (ERs and ERRs) and reference (rpl7 ) genes were
produced by use of validated gene-specific primers (Table I). Copy numbers of these PCR
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
Forward
primer (5% –3% )
GAGGAGGAGGAGGAGGAGGAG
GCTGGAGGTGCTGATGATGG
CGCAGTCAAACCACACTCAT
GGCCGAACCTAGTAGTCCAGA
CAGTCCGTCTTCGGTCATC
TCCTACGTGAAGACCGAACC
AAGACAGAGCCGTCAAGTCC
CTCGGACACCAACTCTAGC
GTCGCCTCCCTCCACAAAG
Template
accession
number
D28954.1
NM_001104702.1
NM_001128512.1
NM_001104917.1
NM_001104918.1
NM_001163091.1
NM_001104919.2
NM_001163092.1
NM_001104870.1
Gene
erα
erβ1
erβ2
errα
errβ1
errβ2
errγ 1
errγ 2
rpl7
Reverse
primer (5% –3% )
GTGTACGGTCGGCTCAACTTC
CGAAGCCCTGGACACAACTG
TCTTCAGGTCCTCCATCAGG
GCACCGTCTCCTCGTGTT
CGTTGGAGTGGCTGTTCAT
CGAGCGTCCGATGTAGCC
GAGTCTAAGCCGTTGGGATG
TTGCTGGCCAAGGCTGT
AACTTCAAGCCTGCCAACAAC
Table I. Gene-specific primer sequences
130
116
105
108
101
111
125
102
99
Amplicon
size (bp)
298
N. K. M. CHEUNG ET AL.
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
ORGAN AND SEX-SPECIFIC EXPRESSION OF ER AND ERR
299
products (standards) were quantitated by measuring against the MassRuler Low Range DNA
Ladder (Fermentas; www.thermoscientificbio.com) and, in parallel, serially diluted (six steps
from 10−4 to 10−10 ) and quantitated by qPCR to construct gene-specific standard curves that
allowed mapping between template copy numbers and the corresponding threshold cycles
(Cq ). Expression of each gene in each sample was measured by use of qPCR and represented
as Cq , which was interpolated to absolute copy number according to the aforesaid gene-specific
standard curves and normalized to the number of copies of the reference gene, rpl7 . Triplicates
were run for each sample. Reaction condition and cycling setup followed those of Yu et al .
(2006), except that Applied Biosystems 7500 Fast Real-Time PCR System and Power SYBR
Green PCR Master Mix (Applied Biosystems; www.appliedbiosystems.com) were used.
S TAT I S T I C A L A N A LY S E S
Principal component analysis (PCA) via singular value decomposition (SVD) was
employed to project the ER and ERR expression patterns to lower dimensions after log10 transformation, mean-centring and scaling to unit variance. Scores on the first two-components
were analysed by use of two-way MANOVA to explore interaction effect between the class
variables of organ and sex on patterns of expression. Organ and sex dependent expression
patterns were further investigated by the use of partial least-squares discriminant analysis
(PLS-DA). Optimal numbers of latent variables retained were determined by minimizing
the classification error rates (p error ) in leave-one-out (LOO) cross-validations. Comparisons
between sexes for each individual ER and ERR in every organ were conducted using a
t-test. Welch’s t-test or Mann–Whitney rank-sum test was used in place of a t-test whenever
homoscedasticity or normality assumptions was violated as indicated by Levene’s test on
median centres and Shapiro-Wilk’s test, respectively. The relationship between absolute
expressions of individual ERs and ERRs was investigated by use of Spearman correlation
analysis and bootstrapped hierarchical clustering. P -values from multiple comparisons were
adjusted to control for false discovery rate (FDR) by use of the Benjamini and Hochberg procedure (Benjamini & Hochberg, 1995). Significance level of 0·05 was used for inferential statistics. All statistical analyses were performed in the R environment 2.14.1 (www.r-project.org).
P H Y L O G E N E T I C A N A LY S I S
To facilitate like-to-like comparisons of the ER and ERR expression profiles between
O. latipes and other model species, especially mummichog Fundulus heteroclitus (L. 1766)
and zebrafish Danio rerio (Hamilton 1822), which are the only other teleosts having a complete, well studied array of ERs and ERRs, a phylogenetic tree was constructed. Peptide
sequences of each ER and ERR subtypes from human Homo sapiens, mouse Mus musculus,
O. latipes, F. heteroclitus and D. rerio (Table II) were aligned by use of MAFFT 6.864 (Katoh
& Toh, 2008) with the E-INS-i strategy and default parameters. Poorly aligned or divergent
regions were removed by use of Gblocks 0.91b (Talavera & Castresana, 2007) with default
settings, except for the following changes: −b2 = 17, −b4 = 5 and −b5 = h. The phylogenetic
tree was then built from the curated alignment using PhyML 3.0 (Guindon et al ., 2010) with
settings as follows: LG substitution matrix, modelled equilibrium frequency, estimated proportion of invariable sites, estimated gamma shape parameter and number of substitution rate categories = 4. A starting tree was first constructed using BioNJ, which was then improved upon
by the SPR (subtree pruning and regrafting) algorithm. The constructed tree was mid-point
re-rooted and visualized in iTOL (Interactive Tree of Life) v2.1.1 (Letunic & Bork, 2011).
RESULTS
P H Y L O G E N E T I C A N A LY S I S
The phylogenetic relationship among ERs and ERRs of O. latipes, D. rerio and
F. heteroclitus as well as those of H. sapiens and M. musculus was illustrated by a
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
300
N. K. M. CHEUNG ET AL.
Table II. RefSeq (if not available, then Genbank) accession numbers of the peptide
sequences recruited for the construction of phylogenetic tree. The identity and naming of
the Danio rerio sequences referenced the Zebrafish Information Network (ZFIN; zfin.org;
25-Dec-2011) instead of the original gene cloning publications. Note that genes erβ, errβ
and errγ are duplicated in the listed teleost. Nomenclature of these duplicated genes (in
parentheses) is inconsistent among species
Homo
sapiens
Mus
musculus
Oryzias
latipes
erα
erβ
NP_000116.2
NP_001428.1
NP_031982.1
NP_997590.1
errα
errβ
NP_004442.3
NP_004443.3
NP_031979.2
NP_036064.3
errγ
NP_001429.2
NP_036065.1
errδ
–
–
BAA86925.1
(erβ1)
NP_001098172.1
(erβ2)
NP_001121984.1
NP_001098387.1
(errβ1)
NP_001098388.1
(errβ2)
NP_001156563.1
(errγ 1)
NP_001098389.2
(errγ 2)
NP_001156564.1
–
Fundulus
heteroclitus
AAT72914.1
(erβa)
AAU44352.1
(erβb)
AAU44353.1
ABB80450.1
(errβa)
ABB80451.1
(errβb)
ABB80452.1
(errγ b)
ABB80453.1
–
Danio
rerio
NP_694491.1
(erβa)
NP_851297.1
(erβb)
NP_777287.2
NP_998120.1
XP_001333980.4
(errγ a)
NP_998119.1
(errγ b)
NP_001122150.1
XP_001921093.3
phylogenetic tree (Fig. 1). Despite the fact that nomenclature of various evolutionary
duplicated ERs and ERRs is inconsistent among teleosts (Table II), postfixes of O.
latipes clearly match with those of D. rerio and F. heteroclitus in a scheme relating
‘a’ to ‘1’ and ‘b’ to ‘2’. Namely, the erβ1 of O. latipes is orthologous to the erβa in
D. rerio, and errγ 2 in O. latipes is orthologous to errγ b in D. rerio. This matching
scheme is consistent for most of the concerned receptor genes with the exception of
errβs, where errβ1 and errβ2 of O. latipes are orthologous to errβb and errβa,
respectively, in F. heteroclitus.
O R G A N - D E P E N D E N T PAT T E R N S O F E X P R E S S I O N O F E R S
AND ERRS
In O. latipes, all ERs, as well as errα, were constitutively expressed in all sampled
organs from both sexes, whilst the expression of all other ERR subtypes varied enormously across organs (Fig. 2). Clustering by organ was evident from PCA (Fig. 3).
The four major clusters of organs: brain, heart, gill + gonad and liver + spleen were
distributed along PC1 (Fig. 3), which was the linear combination comprised mostly
of ERRs (loadings: errα, 0·32; errβ1, 0·51; errβ2, 0·37; errγ 1, 0·49; errγ 2, 0·44;
all ERs ≤ |0·20|). Explicit modelling of this seemingly organ-dependent expression
pattern by use of PLS-DA generated another low-dimensional projection [Fig. 4(a)],
which was analogous to that of PCA. It displayed the same clustering by organs on
the first latent vector that was also constructed predominantly from ERRs (loadings:
errα, 0·25; errβ1, 0·52; errβ2, 0·38; errγ 1, 0·50; errγ 2, 0·45; all ERs ≤ |0·19|).
Both PCA and PLS-DA projections clearly demonstrate that expressions of ERRs
varied among organs, whereas that of ERs was relatively constant. Certain ERRs
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
ORGAN AND SEX-SPECIFIC EXPRESSION OF ER AND ERR
0·1
0·99
0·63
Human ERα
Mouse ERα
Zebrafish erα
Mummichog erα
Medaka erα
Human ERβ
1·00
Mouse ERβ
0·91
Zebrafish erβb
0·71
Mummichog erβb
0·95
0·95
Medaka erβ2
Zebrafish erβa
0·90
Mummichog erβa
0·98
Medaka erβ1
Zebrafish errδ
Human ERRα
1·00
Mouse
ERRα
0·97
Zebrafish errα
1·00
Mummichog errα
0·73
Medaka errα
Mummichog
errβa
0·97
Medaka errβ2
0·04
0·88
Human ERRβ
1·00
Mouse ERRβ
0·03
Zebrafish errβ
0·67
Mummichog
errβb
0·40
0·77
Medaka errβ1
0·87 Human ERRγ
Mouse ERRγ
0·92
Zebrafish errγa
0·99 Medaka errγ1
Zebrafish errγb
0·97
Mummichog errγb
0·95
Medaka errγ2
1·00
1·00
301
0·94
Fig. 1. Phylogenetic tree of all known oestrogen receptor (ER) and oestrogen-related receptor (ERR) subtypes
from Homo sapiens (human), Mus musculus (mouse), Oryzias latipes (medaka), Fundulus heteroclitus
(mummichog) and Danio rerio (zebrafish). Numerical figure shown next to each node is the branch
support assessed by approximated likelihood ratio test based on a Shimodaira–Hasegawa-like procedure
implanted in PhyML.
were silent in specific organs, for instance, errβ2 and errγ 1 were not detected in
heart; errβ1, errβ2 and errγ 1 were unexpressed in liver; transcripts of errβ1, errγ 1
and errγ 2 were not detected in spleen (Fig. 2).
ORGAN-SPECIFIC DIFFERENCES IN EXPRESSION OF ER
AND ERR BETWEEN SEXES
Applying PLS-DA to model differences in expression patterns of ERs and ERRs
with sex as a class variable resulted in a coalesced projection [Fig. 4(b)], which
did not feature any distinct grouping of sexes or organs. This classification was
unrelated to dimension reduction as the full model training p error was as low as 0·61.
That is, at best, the classification of sexes was erroneous 61% of time. The fact
that there was no gross difference in expression of ERs and ERRs between males
and females was further supported by symmetry of the expression profiles (Fig. 2).
Two-way MANOVA of the scores on the first two principal components, however,
demonstrated a statistically significant interaction between the two class variables,
sex and organ (Wilks’ lambda = 0·629, approximate F 5,72 = 3·49, P < 0·001). This
was consistent with the conclusion that differences in ER and ERR expressions
between males and females were not general, but rather organ-specific. As observed
in the projections of PCA (Fig. 3) and the PLS-DA that modelled organ-dependent
expressions [Fig. 4(a)], individuals of different sexes could be segregated into distinct
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
302
N. K. M. CHEUNG ET AL.
Brain
erα
erβ1
erβ2
errα
errβ1
errβ2
errγ1
errγ2
Gill
Gonad
erα
erβ1
erβ2
errα
errβ1
errβ2
errγ1
errγ2
erα
erβ1
erβ2
errα
errβ1
errβ2
errγ1
errγ2
Heart
Receptor genes
erα
erβ1
erβ2
errα
errβ1
errβ2
errγ1
errγ2
Liver
erα
erβ1
erβ2
errα
errβ1
errβ2
errγ1
errγ2
Spleen
erα
erβ1
erβ2
errα
errβ1
errβ2
errγ1
errγ2
4
2
0
2
4
Log10 (normalized copy number of transcript)
Fig. 2. Expression levels of each analysed gene in the studied organs from Oryzias latipes of both sexes ( ,
male; , female). The mean ± s.e. normalized number of transcript copies were represented, on a log10 scale, by the bar length and the coloured marks on the opposite side of origin. Note that the expression
profile is grossly bilaterally symmetric.
clusters in liver as well as in gonad (which was unsurprising since testis and ovary
are two entirely different organs). Only erα, erβ1, errβ2 and errγ 2 exhibited sex
difference for some organs (Table III). Whenever a statistically significant difference
in expression of a particular ER or ERR gene was detected between males and
females, expression was always higher in females (Fig. 2). Conversely, exceptional
sex difference was evident for hepatic errγ 2, which was not expressed in females
and detectable only in males (Fig. 2).
C O R R E L AT I O N S B E T W E E N I N D I V I D U A L E R s A N D E R R s
Correlation of every pair of ERs and ERRs showed two distinct expression clusters
[Fig. 5(a)], which were characterized by significant, positive correlations between
receptors of the same class (i.e. among ERs or among ERRs) and the lack of strong
correlations between those from different receptor classes (i.e. between ERs and
ERRs). For across-class correlation, only errγ 1 and erβ2 were correlated, and errγ 2
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
303
ORGAN AND SEX-SPECIFIC EXPRESSION OF ER AND ERR
1·0
erα
erβ2
0·5
erβ1
errβ2
errγ1
PC2
errα
0·0
errβ1
errγ2
–0·5
–1·0
–1·0
–1·5
0·0
0·5
1·0
PC1
Fig. 3. Two dimensional projection of the expression profile to the first two principal components. Scores
(samples) were linearly scaled to have absolute maximum of 1 and overlaid on top of the loading
(receptor genes) plot. Variance explained: 42% for principal component 1 (PC1) and 24% for principal
component 2 (PC2). Male and female samples are denoted as ♂ and ♀. Organs are annotated in different
colours:
spleen.
and
, brain;
and
, gill;
and
, gonad;
, and
heart;
, and
liver;
, and
was the only ERR that exhibited statistically significant correlation with all subtypes
of ER [Fig. 5(a)].
Sub-clusters within the two major clusters (ERs and ERRs) were also evident,
yet, unexpectedly, the duplicated ER or ERR subtypes, with the exception of errβ,
were not clustered together [Fig. 5(b)]. Namely, in terms of profiles of expression,
erβ2 was more similar to erα than erβ1 (bootstrap: 96%); similarly, errγ 2 was
very different from the rest of ERRs (bootstrap: 99%).
DISCUSSION
This is the first report that has profiled and correlated gene expression of the
complete array of ERs and ERRs in multiple organs of both sexes of a vertebrate. Earlier studies have endeavoured profiling ER or ERR gene expression in
other teleosts such as Atlantic croaker Micropogonias undulatus (L. 1766) (Hawkins
et al ., 2000), fathead minnow Pimephales promelas Rafinesque 1820 (Filby & Tyler,
2005), gilthead seabream Sparus aurata L. 1758 (Socorro et al ., 2000), goldfish
Carassius auratus auratus (L. 1758) (Tchoudakova et al ., 1999; Ma et al ., 2000;
Choi & Habibi, 2003), largemouth bass Micropterus salmoides (Lacépède 1802)
(Sabo-Attwood et al ., 2004), the inbred QurtE strain of O. latipes (Chakraborty
et al ., 2011), F. heteroclitus (Tarrant et al ., 2006; Greytak & Callard, 2007), rainbow
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
304
N. K. M. CHEUNG ET AL.
(a)
(b)
1·0
1·0
erβ2
erα
0·5
errβ2
0·5
erβ1
errγ1
errα
0·0
t(2)
t(2)
erβ1
errβ1
errβ1
0·0
errγ2
–0·5
–0·5
–1·0
errα
errγ2 errγ1
errβ2 erβ2
erα
–1·0
–1·0
–0·5
0·0
0·5
1·0
–1·0
–0·5
t(1)
0·0
t(1)
0·5
1·0
Fig. 4. Biplot of the first two latent vectors from partial least-squares discriminant analysis (PLS-DA) classifications in an attempt to model (a) organ-dependent and (b) sex-dependent expression patterns of
oestrogen receptors (ER) and oestrogen-related receptors (ERR) in Oryzias latipes. Scores (samples)
were linearly scaled to have absolute maximum of 1 and overlaid with the loadings (receptor genes).
Male and female samples are denoted as ♂ and ♀. (a) The three-component classification model (the
last latent vector is not shown for the sake of simplicity) performed satisfactorily in modelling the
organ-dependent expression pattern (training p error = 0·05; cross-validated p error = 0·20), while capturing
adequate variation in the expression dataset (R2 X = 0·78). Note that this projection is very similar to
that of principal component analysis. (b) The modelling of sex-related expression pattern was rather
incompetent (two-components: training p error = 0·90, cross-validated p error = 0·97, R2 X = 0·40; Optimal
five-components: training p error = 0·61, cross-validated p error = 0·85, R2 X = 0·83). Organs are annotated
in different colours:
, and
, and
brain;
, and
gill;
, and
gonad;
, and
heart;
, and
liver;
spleen.
trout Oncorhynchus mykiss (Walbaum 1792) (Nagler et al ., 2007) and embryos of
D. rerio (Bardet et al ., 2002, 2004; Tingaud-Sequeira et al ., 2004). Yet, these studies considered individual classes of receptors but never both simultaneously, and the
vast majority of these studies were conducted on ERs only.
These findings and statistical interpretations for O. latipes are probably applicable
to other vertebrates as the absolute quantitative gene expression profiles reported here
Table III. P -values of two-sample statistical tests (t-test – Mann–Whitney U -test) examining the null hypothesis of no sex difference in the average normalized expression of each
gene in each organ. All P -values were adjusted to control for false discovery rate using the
Benjamini & Hochberg procedure
erα
Brain
Gill
Gonad
Heart
Liver
Spleen
0·152
0·800
0·314
0·774
0·008**
0·437
erβ1
erβ2
errα
errβ1
errβ2
errγ 1
errγ 2
0·022*
0·006**
0·027*
0·624
0·861
0·073
0·481
0·481
0·934
0·285
0·437
0·314
0·935
0·314
0·285
0·934
0·861
0·994
0·062
0·934
1·000
0·950
NA
NA
0·086
0·507
0·011*
NA
NA
0·285
0·064
0·212
0·152
NA
NA
NA
0·709
0·074
0·006**
0·800
♂only
NA
NA, undetectable in both sexes. ♂only : Not expressed by females but detectable in all male individuals.
*P < 0·05, **P < 0·01.
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
ORGAN AND SEX-SPECIFIC EXPRESSION OF ER AND ERR
305
are consistent with the results of studies concerning mammals and other teleosts. All
subtypes of ER were expressed by all organs studied in H. sapiens (Kuiper et al .,
1997), M. musculus (Couse et al ., 1997) and teleosts. ERRα was also found in
virtually all organs of H. sapiens, M. musculus (Giguère et al ., 1988; Shi et al .,
1997; Sladek et al ., 1997) and F. heteroclitus (Tarrant et al ., 2006). Previous studies
in H. sapiens, rodents and F. heteroclitus have also shown that ERRβ and ERRγ
expressions are restricted to specific organs. For instance, in rat Rattus sp., ERRβ was
only expressed in brain (cerebellum and hypothalamus), heart and testis, but not liver
(Giguère et al ., 1988); in F. heteroclitus, both errβa and errβb were detected in most
organs studied but absent from liver and spleen (Tarrant et al ., 2006). Likewise, in H.
sapiens, M. musculus and F. heteroclitus, ERRγ was expressed in diverse organs but
at a much lesser magnitude or entirely absent in the liver (Heard et al ., 2000; Süsens
et al ., 2000; Tarrant et al ., 2006). The highly conserved ER and ERR profiles and
oestrogen biology among vertebrates indicate that O. latipes can serve as a viable
alternative vertebrate model for the study of generalized oestrogen responses and
signalling system in vivo.
It has been proven in O. latipes that modulation of gene expression in response
to oestrogenic compounds is cell, tissue and organ-dependent (Inoue et al ., 2002;
Zhang et al ., 2008a, b). Such a phenomenon is commonly attributed to conjectural
cell-specific ‘cross-talk’ between ERs and ERRs that controls transcription of specific arrays of oestrogen target genes. Thus far, no experimental evidence in vivo has
been provided to substantiate such claim (Giguère, 2002). Conceivably, it is timeconsuming and expensive, yet absurd to study speculative interactions between all
combinations of ERs and ERRs without knowing whether such subtype ER–ERR
interactions occur naturally in specific cells, tissues and organs. This study demonstrates that while all ER subtypes and errα were ubiquitously expressed, expressions
of errβ and errγ isoforms (i.e. errβ1, errβ2, errγ 1 and errγ 2) varied enormously
across specific organs. The results lead to reasonable postulation that the observed
organ-specific oestrogen responses are defined by the expressions of selected errβ
and errγ isoforms, which dispose specific pairing and cross-talk with the ERs in
specific organs.
The premise that ERR isoforms may functionally interact with ERs is reinforced
by the systematic statistical analyses exemplified in this study. Peculiar correlation
between expressions of errγ 2 and all ERs, as well as between errγ 1 and erβ2 has
been identified. It is attested that functionally related genes having highly correlated
expression profiles are often being co-regulated (Allocco et al ., 2004) and are likely
to interact with each other physically (Teichmann & Babu, 2002). Conceivably, the
statistically significant correlation between expressions of ERs and errγ subtypes
may indicate possible existence of co-regulatory mechanisms and, potentially,
functional interactions in vivo. Further research on functional interactions between
ERs and errγ s, in particular erβ1-errγ 2, erβ2-errγ 2 and erβ2-errγ 1, is urgently
needed to define biological significance of these ER-ERR cross-talks at conferring
cell, tissue and organ-dependent responses upon receiving oestrogenic stimuli.
It is recognized that whole organ qPCR analysis is unable to demonstrate definite
co-expression of ERs and ERRs within the same cell, which is prerequisite to
potential interaction. Recently, two histological devices: the embryo chip (Cheung
et al ., 2012) and whole adult histoarray (Kong et al ., 2008; Park et al ., 2008, 2009;
Padilla et al ., 2009; Tompsett et al ., 2009), were developed to enable localization
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
306
N. K. M. CHEUNG ET AL.
(a)
erα
erβ1
erβ2
Spearman’s ρ
–1·0
–0·5
0·0
0·5
1·0
errα
errβ1
errβ2
errγ1
errγ2
erα
erβ1
erβ2
errα
errβ1
errβ2
errγ1
errγ2
(b)
0·8
11
errγ2
errα
94
52
69
errβ2
erβ2
0·3
erα
0·4
errβ1
0·2
Cluster 1
errγ1
0·5
99
96
0·6
erβ1
Dissimilarity
0·7
Cluster 2
Fig. 5. Correlogram and hierarchical clustering of the oestrogen receptor (ER) and oestrogen-related receptor
(ERR) expressions. (a) Correlogram showing the correlation coefficient (Spearman’s ρ) of every pairing
of the analysed genes. Significance of the correlation coefficients after controlling for false discovery
rate: *P < 0·05, **P < 0·01, *** P < 0·001. The two prominent green blocks at the upper left and lower
right sectors are the pairings of ERs and ERRs, respectively. (b) Average linkage clustering of the ERs
and ERRs using 1-|Spearman’s ρ| as the measure of dissimilarity. Numerical figures above the nodes
were cluster support (%) based on multiscale bootstrap resampling (B = 10000). ERs and ERRs were
agglomerated into two distinct clusters (annotated as Cluster 1 and Cluster 2) with very high statistical
confidence (i.e. bootstrap: >95%). Note that duplicated ER and ERR subtypes were not necessarily
clustered together.
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
ORGAN AND SEX-SPECIFIC EXPRESSION OF ER AND ERR
307
and quantification of multiple gene and protein expressions at the cellular level
across multiple organs of the same fish simultaneously using in situ hybridization
(ISH) and immunohistochemistry (IHC). Alternatively, qPCR analysis of cells
removed from histologic sections using laser capture microdissection (LCM) can
also be performed to localize gene expression at the tissue and cellular levels with
sufficient resolution. Studies are ongoing to screen antibodies that can be applied
to co-localize cellular expression of the above target ER and ERR proteins in O.
latipes. Concurrently, in vitro assays are being developed to demonstrate possible
interaction between O. latipes ER and errγ proteins on transcriptional regulation
of oestrogen responsive genes.
The existence of a differential profile of erβs (i.e. erβ1 and erβ2) in teleost
species was not uncommon (Filby & Tyler, 2005; Hawkins et al ., 2005; Greytak
& Callard, 2007; Nagler et al ., 2007). This has been attributed to neo- and subfunctionalization processes that ultimately diverged each of the erβ duplicates to
provide distinct functions, and hence was expressed by specific cells and tissues
(Hawkins et al ., 2000, 2005). This is the first study to report divergence in expressions for ERR duplicates (i.e. errβ1 and errβ2; errγ 1 and errγ 2), suggesting
potential neo and sub-functionalization of ERR duplicates in vertebrates. Interestingly, errγ 2 expression in liver exhibited unique sex-specific pattern (i.e. detectable
in males only), which is distinct from that of all other ERs and ERRs (i.e. generally higher in females than in males). Establishment of knockout mutant of errγ 2,
together with other ERRs, would be useful to uncover their novel roles in oestrogen
signalling, especially in a sex-specific manner.
PERSPECTIVES
The data presented by this study highlight the potential importance of ERRs in
defining organ and sex-specific responses to oestrogenic substances in vertebrates.
The findings provide useful insights into functional importance of ERRs in oestrogen
signalling. Possible research directions were put forward to aid revealing biological
significance of ERs and ERRs divergence in O. latipes/vertebrates. In view of technically demanding and laborious nature of functional analyses, communal efforts are
certainly deemed necessary.
The work described in this paper was supported by a grant from the Research Grants
Council of the Hong Kong Special Administrative Region, China (RGC ref no CityU 160009)
and a grant from the State Key Laboratory in Marine Pollution, City University of Hong
Kong and the Area of Excellence Grant (AoE/P-04/04). J.P.G. was supported by the Canada
Research Chair programme, and, at large, at the Department of Biology and Chemistry and
State Key Laboratory in Marine Pollution, City University of Hong Kong, The Einstein
Professor Program of the Chinese Academy of Sciences and the Visiting Professor Program of
King Saud University. We also thank all teammates for providing assistance in fish sampling.
References
Allocco, D. J., Kohane, I. S. & Butte, A. J. (2004). Quantifying the relationship between
co-expression, co-regulation and gene function. BMC Bioinformatics 5, 18.
Andersen, C. L., Jensen, J. L. & Ørntoft, T. F. (2004). Normalization of real-time quantitative
reverse transcription-PCR data: a model-based variance estimation approach to identify
genes suited for normalization, applied to bladder and colon cancer data sets. Cancer
Research 64, 5245–5250.
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
308
N. K. M. CHEUNG ET AL.
Ankley, G. T., Johnson, R. D., Toth, G., Folmar, L. C., Detenbeck, N. E. & Bradbury, S.
P. (1997). Development of a research strategy for assessing the ecological risk of
endocrine disruptors. Reviews in Toxicology 1, 71–106.
Bardet, P.-L., Horard, B., Robinson-Rechavi, M., Laudet, V. & Vanacker, J.-M. (2002). Characterization of oestrogen receptors in zebrafish (Danio rerio). Journal of Molecular
Endocrinology 28, 153–163.
Bardet, P.-L., Obrecht-Pflumio, S., Thisse, C., Laudet, V., Thisse, B. & Vanacker, J.-M.
(2004). Cloning and developmental expression of five estrogen-receptor related genes
in the zebrafish. Development Genes and Evolution 214, 240–249.
Benjamini, Y. & Hochberg, Y. (1995). Controlling the false discovery rate: a practical and
powerful approach to multiple testing. Journal of the Royal Statistical Society B 57,
289–300.
Bilezikian, J. P., Marcus, R. & Levine, M. A. (2001). The Parathyroids: Basic and Clinical
Concepts, 2nd edn. San Diego, CA: Academic Press.
Björnström, L. & Sjöberg, M. (2005). Mechanisms of estrogen receptor signaling: convergence
of genomic and nongenomic actions on target genes. Molecular Endocrinology 19,
833–842.
Brannvall, K. (2002). Estrogen-receptor-dependent regulation of neural stem cell proliferation
and differentiation. Molecular and Cellular Neuroscience 21, 512–520.
Chakraborty, T., Shibata, Y., Zhou, L.-Y., Katsu, Y., Iguchi, T. & Nagahama, Y. (2011). Differential expression of three estrogen receptor subtype mRNAs in gonads and liver from
embryos to adults of the medaka, Oryzias latipes. Molecular and Cellular Endocrinology 333, 47–54.
Cheung, N. K. M., Hinton, D. E. & Au, D. W. T. (2012). A high-throughput histoarray
for quantitative molecular profiling of multiple, uniformly oriented medaka (Oryzias
latipes) embryos. Comparative Biochemistry and Physiology 155, 18–25.
Choi, C. Y. & Habibi, H. R. (2003). Molecular cloning of estrogen receptor alpha and
expression pattern of estrogen receptor subtypes in male and female goldfish. Molecular
and Cellular Endocrinology 204, 169–177.
Couse, J. F., Lindzey, J., Grandien, K., Gustafsson, J.-Å. & Korach, K. S. (1997). Tissue distribution and quantitative analysis of estrogen receptor-α (ERα) and estrogen
receptor-β (ERβ) messenger ribonucleic acid in the wild-type and ERα-knockout
mouse. Endocrinology 138, 4613–4621.
Eddy, E. M., Washburn, T. F., Bunch, D. O., Goulding, E. H., Gladen, B. C., Lubahn, D. B.
& Korach, K. S. (1996). Targeted disruption of the estrogen receptor gene in male mice
causes alteration of spermatogenesis and infertility. Endocrinology 137, 4796–4805.
Filby, A. L. & Tyler, C. R. (2005). Molecular characterization of estrogen receptors 1, 2a, and
2b and their tissue and ontogenic expression profiles in fathead minnow (Pimephales
promelas). Biology of Reproduction 73, 648–662.
Giguère, V. (2002). To ERR in the estrogen pathway. Trends in Endocrinology and Metabolism
13, 220–225.
Giguère, V., Yang, N., Segui, P. & Evans, R. M. (1988). Identification of a new class of
steroid hormone receptors. Nature 331, 91–94.
Greytak, S. R. & Callard, G. V. (2007). Cloning of three estrogen receptors (ER) from
killifish (Fundulus heteroclitus): differences in populations from polluted and reference
environments. General and Comparative Endocrinology 150, 174–188.
Grossman, C. J. (1985). Interactions between the gonadal steroids and the immune system.
Science 227, 257–261.
Guindon, S., Dufayard, J.-F., Lefort, V., Anisimova, M., Hordijk, W. & Gascuel, O. (2010).
New algorithms and methods to estimate maximum-likelihood phylogenies: assessing
the performance of PhyML 3.0. Systematic Biology 59, 307–321.
Hawkins, M. B., Thornton, J. W., Crews, D., Skipper, J. K., Dotte, A. & Thomas, P. (2000).
Identification of a third distinct estrogen receptor and reclassification of estrogen receptors in teleosts. Proceedings of the National Academy of Sciences of the United States
of America 97, 10751–10756.
Hawkins, M. B., Godwin, J., Crews, D. & Thomas, P. (2005). The distributions of the duplicate
oestrogen receptors ER-beta a and ER-beta b in the forebrain of the Atlantic croaker
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
ORGAN AND SEX-SPECIFIC EXPRESSION OF ER AND ERR
309
(Micropogonias undulatus): evidence for subfunctionalization after gene duplication.
Proceedings of the Royal Society B 272, 633–641.
Heard, D. J., Norby, P. L., Holloway, J. & Vissing, H. (2000). Human ERRgamma, a third
member of the estrogen receptor-related receptor (ERR) subfamily of orphan nuclear
receptors: tissue-specific isoforms are expressed during development and in the adult.
Molecular Endocrinology 14, 382–392.
Hess, R. A., Bunick, D., Lee, K.-H., Bahr, J., Taylor, J. A., Korach, K. S. & Lubahn, D. B.
(1997). A role for oestrogens in the male reproductive system. Nature 390, 509–512.
Horard, B. & Vanacker, J.-M. (2003). Estrogen receptor-related receptors: orphan receptors
desperately seeking a ligand. Journal of Molecular Endocrinology 31, 349–357.
Inoue, A., Yoshida, N., Omoto, Y., Oguchi, S., Yamori, T., Kiyama, R. & Hayashi, S. (2002).
Development of cDNA microarray for expression profiling of estrogen-responsive
genes. Journal of Molecular Endocrinology 29, 175–192.
Katoh, K. & Toh, H. (2008). Recent developments in the MAFFT multiple sequence alignment
program. Briefings in Bioinformatics 9, 286–298.
Kong, R. Y. C., Giesy, J. P., Wu, R. S. S., Chen, E. X. H., Chiang, M. W. L., Lim, P. L.,
Yuen, B. B. H., Yip, B. W. P., Mok, H. O. L. & Au, D. W. T. (2008). Development
of a marine fish model for studying in vivo molecular responses in ecotoxicology.
Aquatic Toxicology 86, 131–141.
Kuiper, G. G. J. M., Carlsson, B., Grandien, K., Enmark, E., Häggblad, J., Nilsson, S. &
Gustafsson, J. Å. (1997). Comparison of the ligand binding specificity and transcript
tissue distribution of estrogen receptors alpha and beta. Endocrinology 138, 863–870.
Letunic, I. & Bork, P. (2011). Interactive tree of life v2: online annotation and display of
phylogenetic trees made easy. Nucleic Acids Research 39, W475–W478.
Ma, C. H., Dong, K. W. & Yu, K. L. (2000). cDNA cloning and expression of a novel estrogen
receptor beta-subtype in goldfish (Carassius auratus). Biochimica et Biophysica Acta
1490, 145–152.
McEwen, B. S. & Alves, S. E. (1999). Estrogen actions in the central nervous system.
Endocrine Reviews 20, 279–307.
Mendelsohn, M. E. & Karas, R. H. (1999). The protective effects of estrogen on the cardiovascular system. The New England Journal of Medicine 340, 1801–1811.
Muramatsu, M. & Inoue, S. (2000). Estrogen receptors: how do they control reproductive and
nonreproductive functions? Biochemical and Biophysical Research Communications
270, 1–10.
Nagler, J. J., Cavileer, T., Sullivan, J., Cyr, D. G. & Rexroad, C. (2007). The complete nuclear
estrogen receptor family in the rainbow trout: discovery of the novel ERalpha2 and
both ERbeta isoforms. Gene 392, 164–173.
Padilla, S., Cowden, J., Hinton, D. E., Yuen, B., Law, S., Kullman, S. W., Johnson, R.,
Hardman, R. C., Flynn, K. & Au, D. W. T. (2009). Use of medaka in toxicity testing.
Current Protocols in Toxicology 39, 1.10.1–1.10.36.
Park, J.-W., Tompsett, A., Zhang, X., Newsted, J. L., Jones, P. D., Au, D., Kong, R., Wu, R.
S. S., Giesy, J. P. & Hecker, M. (2008). Fluorescence in situ hybridization techniques
(FISH) to detect changes in CYP19a gene expression of Japanese medaka (Oryzias
latipes). Toxicology and Applied Pharmacology 232, 226–235.
Park, J. W., Tompsett, A. R., Zhang, X., Newsted, J. L., Jones, P. D., Au, D. W. T., Kong,
R., Wu, R. S. S., Giesy, J. P. & Hecker, M. (2009). Advanced fluorescence in situ
hybridization to localize and quantify gene expression in Japanese medaka (Oryzias
latipes) exposed to endocrine-disrupting compounds. Environmental Toxicology and
Chemistry 28, 1951–1962.
Sabo-Attwood, T., Kroll, K. J. & Denslow, N. D. (2004). Differential expression of largemouth
bass (Micropterus salmoides) estrogen receptor isotypes alpha, beta, and gamma by
estradiol. Molecular and Cellular Endocrinology 218, 107–118.
Shi, H., Shigeta, H., Yang, N., Fu, K., O’Brian, G. & Teng, C. T. (1997). Human estrogen
receptor-like 1 (ESRL1) gene: genomic organization, chromosomal localization, and
promoter characterization. Genomics 44, 52–60.
Sladek, R., Bader, J.-A. & Giguère, V. (1997). The orphan nuclear receptor estrogen-related
receptor alpha is a transcriptional regulator of the human medium-chain acyl coenzyme
A dehydrogenase gene. Molecular and Cellular Biology 17, 5400–5409.
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310
310
N. K. M. CHEUNG ET AL.
Socorro, S., Power, D. M., Olsson, P.-E. & Canario, A. V. M. (2000). Two estrogen receptors
expressed in the teleost fish, Sparus aurata: cDNA cloning, characterization and tissue
distribution. Journal of Endocrinology 166, 293–306.
Stampfer, M. J., Colditz, G. A., Willett, W. C., Manson, J. E., Rosner, B., Speizer, F. E.
& Hennekens, C. H. (1991). Postmenopausal estrogen therapy and cardiovascular disease. Ten-year follow-up from the nurses’ health study. The New England Journal of
Medicine 325, 756–762.
Süsens, U., Hermans-Borgmeyer, I. & Borgmeyer, U. (2000). Alternative splicing and expression of the mouse estrogen receptor-related receptor gamma. Biochemical and Biophysical Research Communications 267, 532–535.
Talavera, G. & Castresana, J. (2007). Improvement of phylogenies after removing divergent
and ambiguously aligned blocks from protein sequence alignments. Systematic Biology
56, 564–577.
Tang, S., Zhang, Z., Tan, S. L., Tang, M. H. E., Kumar, A. P., Ramadoss, S. K. & Bajic, V.
B. (2007). KBERG: KnowledgeBase for Estrogen Responsive Genes. Nucleic Acids
Research 35, D732–D736.
Tarrant, A. M., Greytak, S. R., Callard, G. V. & Hahn, M. E. (2006). Estrogen receptor-related
receptors in the killifish Fundulus heteroclitus: diversity, expression, and estrogen
responsiveness. Journal of Molecular Endocrinology 37, 105–120.
Tchoudakova, A., Pathak, S. & Callard, G. V. (1999). Molecular cloning of an estrogen
receptor beta subtype from the goldfish, Carassius auratus. General and Comparative
Endocrinology 113, 388–400.
Teichmann, S. A. & Babu, M. M. (2002). Conservation of gene co-regulation in prokaryotes
and eukaryotes. Trends in Biotechnology 20, 407–410.
Tingaud-Sequeira, A., André, M., Forgue, J., Barthe, C. & Babin, P. J. (2004). Expression
patterns of three estrogen receptor genes during zebrafish (Danio rerio) development:
evidence for high expression in neuromasts. Gene Expression Patterns 4, 561–568.
Tompsett, A. R., Park, J. W., Zhang, X., Jones, P. D., Newsted, J. L., Au, D. W. T., Chen,
E. X. H., Yu, R., Wu, R. S. S., Kong, R. Y. C., Giesy, J. P. & Hecker, M. (2009).
In situ hybridization to detect spatial gene expression in medaka. Ecotoxicology and
Environmental Safety 72, 1257–1264.
Vanacker, J. M., Pettersson, K., Gustafsson, J. Å. & Laudet, V. (1999). Transcriptional targets
shared by estrogen receptor-related receptors (ERRs) and estrogen receptor (ER) alpha,
but not by ERbeta. EMBO Journal 18, 4270–4279.
Yu, R. M. K., Chen, E. X. H., Kong, R. Y. C., Ng, P. K. S., Mok, H. O. L. & Au, D. W. T.
(2006). Hypoxia induces telomerase reverse transcriptase (TERT) gene expression in
non-tumor fish tissues in vivo: the marine medaka (Oryzias melastigma) model. BMC
Molecular Biology 7, 27.
Zhang, Z. & Hu, J. (2007). Development and validation of endogenous reference genes
for expression profiling of medaka (Oryzias latipes) exposed to endocrine disrupting
chemicals by quantitative real-time RT-PCR. Toxicological Sciences 95, 356–368.
Zhang, X., Hecker, M., Tompsett, A. R., Park, J., Jones, P. D., Newsted, J., Au, D., Kong,
R., Wu, R. S. S. & Giesy, J. P. (2008a). Responses of the medaka HPG axis PCR
array and reproduction to prochloraz and ketoconazole. Environmental Science and
Technology 42, 6762–6769.
Zhang, X., Hecker, M., Park, J. W., Tompsett, A. R., Newsted, J., Nakayama, K., Jones, P.
D., Au, D., Kong, R., Wu, R. S. S. & Giesy, J. P. (2008b). Real-time PCR array to
study effects of chemicals on the Hypothalamic-Pituitary-Gonadal axis of the Japanese
medaka. Aquatic Toxicology 88, 173–182.
 2013 The Authors
Journal of Fish Biology  2013 The Fisheries Society of the British Isles, Journal of Fish Biology 2013, 83, 295–310