General and Comparative Endocrinology xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

General and Comparative Endocrinology

journal homepage: www.elsevier.com/locate/ygcen

Research paper

The effects of migratory stage and 11-ketotestosterone on the expression

of rod opsin genes in the shortfinned eel (Anguilla australis)
Georgia Thomson-Laing a,⇑, Christine L. Jasoni b, P. Mark Lokman a
a
Department of Zoology, University of Otago, PO Box 56, Dunedin 9054, New Zealand
b
Department of Anatomy, Centre for Neuroendocrinology, University of Otago, PO Box 56, Dunedin 9054, New Zealand

a r t i c l e i n f o a b s t r a c t

Article history: The androgen 11-ketotestosterone (11KT) can induce many of the changes associated with silvering, i.e.,
Received 16 January 2017 the transformation of a non-migrating ‘yellow’ eel into a migrating ‘silver’ eel. We posited that plasticity
Revised 4 June 2017 in spectral sensitivity of the eye, accompanied by expression of different opsins in the retina during sil-
Accepted 24 June 2017
vering, is controlled by 11KT. To test this hypothesis, mRNA levels of freshwater (fwo) and seawater (swo)
Available online xxxx
opsins and of the two androgen receptors (ara and arb) in retinas of wild-caught female shortfinned eels,
Anguilla australis were compared. Swo expression was much higher (3–4 orders of magnitude) and fwo
Keywords:
expression substantially lower in silver than in yellow eels, whereas mRNA levels of both ars did not dif-
Eel
Anguilla australis
fer between stages. Yellow eel retinas exposed to 11KT in vitro exhibited a robust dose-dependent
Silvering increase in swo, but weak decreasing effects on fwo transcript abundance were inconsistent. Similarly,
Migration increased retinal swo expression was seen after in vivo treatment of yellow eels with 11KT implants,
Opsin whereas expression of fwo remained unaffected. Lastly, co-treatment with 11KT and the androgen recep-
Androgen tor blocker flutamide was undertaken to determine whether 11KT exerts its effects through nuclear
11-Ketotestosterone androgen receptors. Flutamide did not block 11KT-affected expression of any target gene, neither
in vivo nor in vitro. We conclude that 11KT greatly increases the abundance of swo, identifying the andro-
gen as an important regulator of the opsin switch during silvering in freshwater eels.
Ó 2017 Elsevier Inc. All rights reserved.

1. Introduction ocean (Aoyama and Miller, 2003) – oceanic life is a prerequisite

Freshwater eels have an extra-ordinary life history, character-
ized by fascinating, long-distance migrations, by movements
between fresh- and seawater environments and by transforma-
tions that make these activities possible (Aoyama and Miller,
2003; Bruijs and Durif, 2009; Schmidt, 1923; Lokman, 2016). In
brief, adults spawn in the ocean and eggs hatch and develop into
leafshape-like, leptocephalus larvae that travel to distant
freshwater habitats on ocean currents (Van Ginneken and Maes,
2005). Prior to entry into fresh water, larvae metamorphose into
glass eels, and then start acquiring pigmentation to become
so-called ‘‘yellow” or non-migrant eels (Bruijs and Durif, 2009).
Upon reaching adult size, they transform – an event known as
‘‘silvering” – into a ‘‘silver” or migrant eel that is pre-adapted
morphologically, behaviorally and physiologically to life in the
⇑ Corresponding author at: Department of Zoology, University of Otago 340, PO
Box 56, Dunedin 9016, New Zealand.
E-mail address: georgia_t-laing@hotmail.co.nz (G. Thomson-Laing).
for successful return to their natal area where the eels will spawn
to complete their life cycle (Bruijs and Durif, 2009; Tesch and Rohlf,
2003).
Silvering involves a suite of changes that typically include color
changes to the pectoral fins and skin (e.g., Todd, 1981; Okamura
et al., 2007; Jessop, 1987), flattening of the head, and slimming
and lengthening of the lower jaw (c.f., Lokman et al., 2003). One
of the most distinctive changes that further occurs during silvering
is an increase in eye size (Durif et al., 2005; Sudo et al., 2011;
Hagihara et al., 2012; Todd, 1981; Jessop, 1987). There are also
changes in retinal function, reflected in the expression of different
visual pigments (e.g., Archer et al., 1995).
Visual pigments, located on photoreceptor cells (rods and
cones), are responsible for light absorbance in the retina. In rods,
the pigments are complexes composed of an opsin, which is a G-
protein linked membrane receptor, and a vitamin A-like chromo-
phore, either Vitamin A1 or A2. The resulting pigments, known
as rhodopsin and porphyropsin, display sensitivity to shorter and
longer light wavelengths, respectively (Beatty, 1975; Bridges,
1972), photic spectral sensitivity ultimately being determined by

http://dx.doi.org/10.1016/j.ygcen.2017.06.025
0016-6480/Ó 2017 Elsevier Inc. All rights reserved.

Please cite this article in press as: Thomson-Laing, G., et al. The effects of migratory stage and 11-ketotestosterone on the expression of rod opsin genes in
the shortfinned eel (Anguilla australis). Gen. Comp. Endocrinol. (2017), http://dx.doi.org/10.1016/j.ygcen.2017.06.025
2 G. Thomson-Laing et al. / General and Comparative Endocrinology xxx (2017) xxx–xxx
the ratio of both in the retina. Yellow eels utilize more porphy-
ropsin (kmax = 523 nm) than rhodopsin (kmax = 501 nm) to yield
an overall retinal kmax ranging between 515 and 518 nm in the
American eel, A. rostrata (Beatty, 1975). It has been suggested that
this provides high sensitivity to yellow/green wavelengths in the
freshwater habitat (Wood et al., 1992).
In most teleosts, only a single rod opsin is typically expressed in
the retina and the vitamin A1/A2 ratio is used to alter spectral
specificity. However, many Anguilla spp. can further alter spectral
specificity by producing a novel, deep-sea opsin in preparation
for migration (Beatty, 1975; Hope et al., 1998; Zhang et al., 2000;
Wang et al., 2014). The retina of the yellow eel is dominated by
the freshwater opsin and following an ‘opsin switch’ the silver
eel retina becomes dominated by the deep-sea opsin in the Amer-
ican eel (Beatty, 1975) and the Japanese eel (Zhang et al., 2000).
Similarly, artificially matured European eels experience this switch
from high freshwater opsin expression to high deep-sea opsin
expression in the retina during development (Hope et al., 1998).
Accordingly, with transition from the yellow to silver stage, there
is a shift in retinal spectral sensitivity from kmax = 523 nm to
482 nm in the European eel (Carlisle and Denton, 1959; Hope
et al., 1998), and from kmax = 515 nm to 484 nm in the American
eel (Beatty, 1975). The change in visual pigments in the rod pho-
toreceptor results in a change from green-light to blue-light sensi-
tivity, which reflects the wavelengths that eels will encounter
during migration from green-lit fresh water to a blue-lit deep-sea
environment (Archer et al., 1995).
In several fish species, including rainbow trout (Oncorhynchus
mykiss) and zebrafish (Danio rerio), opsin gene expression can be
affected by thyroid hormones (TH) (Suliman and Flamarique,
2014); however, no changes in levels of TH were found in European
eel in relation to silvering or migration (Aroua et al., 2005). In stick-
leback (Gasterosteus aculeatus), the androgen
ketoandrostenedione up-regulated the expression of a red-
sensitive opsin, associated with a change in red-light sensitivity
during the breeding period (Shao et al., 2014). Likewise, a change
in opsin expression was induced by treatment with gonadotropin,
which mediates its effects, in part by steroid hormones, in Euro-
pean eel (Archer et al., 1995; Hope et al., 1998; Wood and
Partridge, 1993).
Interestingly, plasma levels of the steroid hormone 11-
ketotestosterone (11KT) are much higher in silver than in yellow
eels (Lokman et al., 1998; Sudo et al., 2011). Moreover, experimen-
tal exposure of yellow eels to 11KT results in dramatic changes in
appearance, similar to those seen during silvering – notably, it
caused a slimming of the snout, blackening of the pectoral fins
and an increase in eye size, alongside a suite of other changes
(Rohr et al., 2001). Consequently, 11KT appears to be a key driver
of silvering in eel. This prompted us to hypothesize that 11KT will
control the ‘opsin switch’ that pre-adapts the silver eel to an ocea-
nic existence.
To test this hypothesis, we sampled wild yellow and silver New
Zealand shortfinned eels, A. australis, to provide a baseline on
expression of the freshwater (fwo) and seawater opsins (swo) in
the retina. Subsequently, we determined the effect of exogenous
treatment with 11KT on the expression of both opsins in the retina
using in vivo and in vitro approaches. Additionally, we determined
the expression of the presumed mediators of 11KT action, i.e.,
androgen receptor-a (Ara/ara) and -b (Arb/arb) (Ikeuchi et al.,
1999), in the retina. Lastly, we employed the androgen receptor
blocker flutamide (c.f., Kwon et al., 2005) to further characterize
the androgen-dependence of regulation of opsin gene expression
in the retina. Our findings provide compelling evidence for a role
of 11KT in opsin switching during silvering of freshwater eels, prin-
cipally by up-regulating retinal swo gene expression.
Please cite this article in press as: Thomson-Laing, G., et al. The effects of migratory stage and 11-ketotestosterone on the expression of rod opsin genes in
the shortfinned eel (Anguilla australis). Gen. Comp. Endocrinol. (2017), http://dx.doi.org/10.1016/j.ygcen.2017.06.025
2. Materials and methods
2.1. Animal collection and experimental design
Female shortfinned eels were supplied by a commercial fishery
(Mossburn Enterprises) and sourced from Lake Ellesmere, New
Zealand during early autumn. Fish were either sampled soon after
capture (Section 2.1.1) or transported to our laboratory facilities at
the University of Otago, Dunedin, for experimental use (Sections
2.1.2 and 2.1.3). Eels kept in captivity were maintained in tanks
with recirculating water at ambient temperature (10–13 °C) and
a salinity of 12 ppt. Prior to experimentation, water temperature
was gradually increased to 16 °C, with light cycles set to a regime
of 12 light:
12 dark hours. Fish were not fed during this time. Collection and
manipulation of eels were approved by the University of Otago
Animal Ethics Committee in accordance with the guidelines of
the Australian & New Zealand Council for the Care of Animals in
Research and Teaching.
2.1.1. Field survey: Rod opsin and androgen receptor mRNA levels in
the retina of wild-caught shortfinned eels in yellow and silver stages
Eels harvested from fyke nets set overnight were assigned,
based on size and migratory status, to one of three categories
designed to represent successive stages of ontogenetic develop-
ment. Categories (N = 7 in each) utilized were small-sized (body
weight, BW, 300–600 g) non-migrant ‘yellow’ eels (YS), medium-
sized (BW 800–1800 g) non-migrant ‘yellow’ eels (YM), and
migrant ‘silver’ eels (S; BW 800–1400 g). Silver eels were distin-
guished from yellow eels using previously established morpholog-
ical differences in coloration and head shape (Todd, 1981).
Eels were euthanized with benzocaine (0.3 g/L) and physical
11-
parameters, i.e., BW, length and eye diameter (vertical and hori-
zontal), were measured; the latter two measures were used to cal-
culate the eye index (eye surface area standardized over body
length) after Pankhurst (1982). The tail was transected and blood
collected for steroid measurement by radioimmunoassay (Sec-
tion 2.2). Thereafter, ovaries were dissected and weighed to allow
the calculation of the gonadosomatic index (GSI: ovary as a per-
centage of body weight). Eyecups, without lens and iris, were
snap-frozen on dry ice, stored at 70 °C and utilized for quantifica-
tion of mRNA levels of swo, fwo, ara, arb, and the reference gene,
ribosomal protein l36 (l36) (see Section 2.3).
2.1.2. In vitro effects of 11KT on mRNA levels of rod opsins and
androgen receptors in retinal tissue of shortfinned eels
Effects of a suite of compounds on rod opsin gene expression in
eel retinas were preceded by several preliminary trials in order to
optimize incubation conditions. Thus, eyes were retrieved from the
eel head, placed in Ringer solution, and connective tissue was care-
fully removed using scissors and forceps. Fine scissors were used to
separate the black eyecup from the front part of the eye by cutting
along the edge of the gold-iridescent iris. The iris, lens and vitreous
humor were discarded and the empty eye cup was then cut open
and the pigment layer was very gently scraped off, leaving a thin
layer of tissue < 200 lm thick (Suppl Fig. 1, Fig. 1). The tissue
was carefully cut into small pieces (around 15 mm2) and these
were placed in wells of a 24-well plate in the presence or absence
of the compounds of interest.
All incubations were maintained under agitation and set up by
submerging the tissues in eel ringer containing 0.1 g/L strepto-
mycin and 100,000 IU/mL penicillin; the use of an agarose support
on which a piece of PVDF membrane was placed to enable retinal
tissue to be incubated at the interface between incubation medium
 G. Thomson-Laing et al. / General and Comparative Endocrinology xxx (2017) xxx–xxx
Fig. 1. Light micrographs of retinal sections from the eye of the shortfinned eel. The retinal pigmented epithelium (RPE) was removed (A), but the photoreceptor (inner) layer
(PL) retained to generate a tissue prep suitable for in vitro experimentation. Intact retinal tissue is shown for comparison (B). Sections were stained with hematoxylin and
eosin. Scale-bar = 50 mm.
(Leibovitz-15) and air proved unsuccessful. Incubation conditions
were optimized by trialing a time course over several days. A sig-
nificant 11KT-dose-dependent increase in swo mRNA levels was
evident after 1 day of incubation, becoming non-significant (Suppl
Fig. 2) after 2 days of exposure due to increasing scatter in the data.
Our standard incubations, therefore, were run overnight on a sha-
ker at 18 °C.
Thereafter, we ran two in vitro trials. The first trial was aimed at
assessing the effects of different doses of 11KT on opsin and andro-
gen receptor mRNA levels (Trial I). The second trial (Trial II) partly
repeated Trial I to test whether observed results were consistent
and repeatable. Two trials were run to confirm that tissue
responses to treatments were repeatable and robust; indeed, there
can be variability in the responses of tissues collected from wild,
undomesticated animals at slightly different times of the year or
in slightly different stages of development, especially when run-
ning experiments in vitro. The second trial further examined the
effects of culture of 11KT in combination with the AR blocker flu-
tamide to test whether its effects are mediated by nuclear andro-
gen receptors.
2.1.2.1. Trial I. Retinas from the right eye from each of 6 eels were
retrieved and cut into fragments that were assigned to 0, 1, 10 or
100 ng/mL 11KT). After incubation, tissue from each well was
retrieved, frozen on dry ice and stored at 70 °C until processing
for quantification of swo, fwo, arb, ara, l36 by qPCR (Section 2.3).
2.1.2.2. Trial II. Retinal tissue was isolated from eight captive yel-
low eels (BW  600 g) and cut into 12 pieces, 4 of which were incu-
bated with or without 11KT (0, 3, 30, 300 nM; equivalent to 0–
90 ng/mL) as a positive control for tissue viability. The remaining
8 retinal pieces were subjected to treatment with flutamide in a
dose-response design (0, 3, 30, 300 nM) in the presence or absence
of 11KT (30 nM). Tissue was incubated for 24 h and subsequently
processed for quantification of mRNA levels of swo, fwo, l36 and
elf after frozen storage at 70 °C (Section 2.3).
11KT doses for both in vitro experiments were chosen based on
11KT plasma concentrations of 20–50 ng/mL found in silver eels
(Lokman et al., 1998; Setiawan et al., 2012). Flutamide doses were
matched to those of 11KT due to limited information on the effec-
tive flutamide dosage required for blocking androgenic action in
eels.
2.1.3. In vivo effects of 11KT on mRNA levels of rod opsins and
androgen receptors in the retina of yellow shortfinned eels
Yellow eels (BW  0.5 kg) were placed in 0.15 g/L benzocaine to
induce anesthesia. Each fish then received a unique passive inte-
Please cite this article in press as: Thomson-Laing, G., et al. The effects of migratory stage and 11-ketotestosterone on the expression of rod opsin genes in
the shortfinned eel (Anguilla australis). Gen. Comp. Endocrinol. (2017), http://dx.doi.org/10.1016/j.ygcen.2017.06.025
3
grated transponder (PIT) tag and a cholesterol-cellulose pellet
(see Lokman et al., 2015; Sherwood et al., 1988) that contained
0.1 mg 11KT (N = 6), flutamide (N = 6), or a combination of 11KT
+ flutamide (N = 6). A control group (N = 6) was implanted with a
pellet containing no added drug. PIT tag and implants were placed
in the body cavity using a Ralgun, after which fish were placed into
experimental tanks, each equipped with its own filtration system
to ensure that any drug leaching from the fish into the water would
not affect eels in a different treatment group. After a five-week
experimental period, eels were terminally sampled as described
above (Section 2.1.1) and blood and tissues were collected for ster-
oid measurement by radioimmunoassay and qPCR analysis,
respectively. The reference gene, elongation factor-1a (elf), was
included in the analyses.
Previous research by Damsteegt et al. (2016) indicated that
implantation of yellow shortfinned eels (BW  0.65kg) with 1 mg
11KT elevated plasma 11KT levels into a supra-physiological range
of around 100–150 ng/mL. Therefore, implants containing 0.1 mg
11KT were used to elevate plasma 11KT to a more biologically rel-
evant concentration. As in the in vitro experiment, flutamide was
used in vivo at the same dose as 11KT. A five-week experimental
period was chosen as previous findings have indicated successful
elevation of plasma 11KT levels in yellow shortfinned eels for at
least four weeks following implantation with 11KT slow-release
pellets (Damsteegt et al., 2016).
2.2. Radioimmunoassay (RIA)
Blood samples from wild eels were processed, stored and
assayed by validated RIA for 11KT as described previously
(Lokman et al., 2015).
2.3. RNA isolation, reverse transcription and qPCR analysis
2.3.1. Isolation of RNA and reverse transcription
Total RNA was extracted from eye or retinal tissue using TRIZOL
Reagent according to product instructions (Ambion). Tissue was
homogenized with a bullet blender (Gentaur) and RNA was quan-
tified using a Nanodrop spectrophotometer (ND1000). Total RNA
was treated with DNase using the Turbo DNA-freeTM kit and proto-
col (Ambion) and then re-quantified (Nanodrop, ND1000). DNase-
treated total RNA (250–500 ng) was reverse-transcribed using ran-
dom hexamer primers to produce complementary DNA (cDNA)
using the High Capacity cDNA Reverse Transcription Kit (Applied
Biosystems). The resultant cDNA was diluted to a concentration
of 10 ng RNA/mL.
4 G. Thomson-Laing et al. / General and Comparative Endocrinology xxx (2017) xxx–xxx
2.3.2. Retrieval of DNA sequence data of candidate genes
One contig that included the full open-reading frame of fresh-
water opsin (contig CL3228; KY974311) and one encoding seawa-
ter opsin (contig CL3228; KY974312) were obtained from an
unpublished multi-tissue shortfinned eel transcriptome. Specific
primers (Suppl Table 1) were designed in Primer 3 (Untergasser
et al., 2012) and acquired from Integrated DNA Technologies for
amplification of partial A. australis fwo (739 bp) and swo (733 bp)
cDNAs by PCR. Total PCR reaction volumes were 20 mL and included
cDNA (1 mL), Taq polymerase (Bioline; 0.2 mL), NH4 x10 buffer
(2 mL), MgCl2 (1 mL, 50 mM), forward and reverse primers (1 mL
each, 10 mM), dNTP (2 mL, 10 mM), and MilliQ (11.8 mL). The PCRs
were run using a temperature profile starting with 94 °C (2 min),
then 35 cycles of denaturation (94 °C for 15 s) then annealing
(30 s) then extension (72 °C for 45 s), ending with 72 °C (3 min).
PCRs were run with gradient annealing temperatures (57.6 °C–
62.1 °C) to maximise primer specificity. Samples were subse-
quently run on an agarose gel containing SYBR gel dye (Invitrogen)
to visualize PCR products. Bands were excised and extracted from
the gel (DNA Gel Elution Kit, NucleoSpin, Macherey-Nagel) and
sequenced (Sanger-sequencing, Genetic Analysis Services, Univer-
sity of Otago). The resulting swo and fwo sequences had 98% and
99% identity, respectively, with corresponding Anguilla japonica
sequences (swo – AJ249203, fwo – AJ249202; data not shown).
Freshwater and seawater opsin amino acid sequences obtained
for A. australis were compared with relevant sequences from A.
anguilla (swo – AAA99297, fwo – AAA99200), A. japonica (swo –
CAB56647, fwo – CAB56646) and A. marmorata (swo – AIJ19435,
fwo – AIJ19436).
2.3.3. Quantitative PCR
During the course of our study, QPCR hardware had to be
replaced, such that all samples were assayed on an MX3000 P ther-
mal cycler (Stratagene), except for Trial II (in vitro), samples of
which were run on a QuantStudio 5 (Thermo Fisher). Primers for
use in qPCR were designed in Primer 3 (Suppl Table 1), using the
partial sequences for swo and fwo obtained earlier (Section 2.3.2),
whereas sequence data for ara, arb, l36 and elf were obtained from
Setiawan et al. (2012) and Damsteegt et al. (2016). Assays for all
target genes were run using final volumes of either 16 mL or
10 mL per well with the reaction mix containing 1 mL of sample
cDNA, standard cDNA or MilliQ water (no-template controls), tar-
get gene-specific forward and reverse primers (Suppl Table 1,
0.5 mL of each from 10 mM primer stock), SYBR master mix (8 mL
or 5 mL; Takara SYBRÒ Premix Ex TaqTM) and purified water (MilliQ)
(6 mL or 3 mL). Standard curves (ranging from 10 – 0.0001 pg
dsDNA/mL) were run with every assay and used to determine the
relative quantity of target gene mRNA in each sample.
QPCR assays were run with a thermal profile of an initial 95 °-
C cycle for two minutes followed by 40 cycles consisting of denat-
uration (95 °C for five seconds) then annealing (10 s, temperature
gene specific, see Suppl Table 1) then extension (72 °C for five sec-
onds). Following the 40 cycles, a dissociation curve analysis was
run consisting of denaturation (95 °C for one minute) then anneal-
ing (55 °C for 30 s) and then 1 °C degree temperature increases
over 30 s increments resulting in a final temperature of 95 °C
(30 s). To validate and optimize the qPCR assays, primers were
run at a suite of temperatures (60–64 °C), melt curve analysis
was undertaken and opsin qPCR amplicons were confirmation-
sequenced (Genetic Analysis Services, University of Otago).
Target gene expression (swo, fwo, ara, arb) was normalized by
the l36 mRNA quantity or the geometric mean of l36 and elf. Refer-
ence genes were chosen from a selection of genes recommended in
a previous study (Setiawan and Lokman, 2010). Selection
depended on the reference gene(s) exhibiting the most constant
expression across treatment groups – we did not detect any effect
Please cite this article in press as: Thomson-Laing, G., et al. The effects of migratory stage and 11-ketotestosterone on the expression of rod opsin genes in
the shortfinned eel (Anguilla australis). Gen. Comp. Endocrinol. (2017), http://dx.doi.org/10.1016/j.ygcen.2017.06.025
of treatment on reference gene mRNA levels (P > 0.05). Reference
gene selection differed between experiments, but never within,
and therefore, samples normalized by different reference genes
were never directly compared.
2.4. Data analysis
Assumptions of homogeneity of variance (Levene’s test) and
normality were evaluated and if not met, then data were log-
transformed and assumptions rechecked. Data from eels in the
field survey were analyzed using one-factor analyses of variance
(ANOVA) to compare both physical parameters, plasma 11KT levels
and gene expression results between migratory categories (YS, YM
and S). Mean retinal gene expression data from in vitro experi-
ments were compared by a one-factor (11KT) or two-factor ANOVA
with interaction (11KT at 2 dose levels, flutamide at 4 dose levels).
Two-factor ANOVAs were also run to determine whether main
effects or their interactions existed following exposure of yellow
eels to drug-containing implants (11KT at 2 dose levels; flutamide
at 2 dose levels). Tukey post hoc analyses were used where appli-
cable to determine differences between sample means in different
categories (P < 0.05).
3. Results
3.1. Field survey: Rod opsin and androgen receptor mRNA levels in the
retina of wild-caught shortfinned eels in yellow and silver stages
3.1.1. Critical amino acid sites in spectral tuning
In comparison to the A. australis freshwater opsin, the seawater
opsin displayed an aspartic acid to asparagine substitution at
amino acid residue 83. Additionally, at residues 124 and 292, ala-
nine was substituted for serine in the seawater opsin. These substi-
tutions at sites 83, 124 and 292 were identical to the substitutions
seen in the sea opsin proteins from A. anguilla, A. japonica and A.
marmorata.
3.1.2. Biometric measures
Relative gonad mass was significantly several-fold higher in sil-
ver eels (3.19 ± 0.09) than in yellow eels (GSI < 1) of both the small
(YS; 0.18 ± 0.03) and medium (YM) size classes (YM; 0.54 ± 0.09;
F2, 18 = 507; P < 0.001; Fig. 2A), consistent with previous reports
(Lokman et al., 1998; Sudo et al., 2011; Setiawan et al., 2012). Eel
eye index was also significantly higher in silver eels (52 ± 2.9) than
in either size of yellow eels (38 ± 1.6 for YS and 42 ± 1.9 for YM; F2,
17 = 17.6; P < 0.001; Fig. 2B). No differences in GSI or eye-index
were evident between the YS and YM groups, although there was
a slight tendency for these measures to increase in the larger eels.
3.1.3. Plasma 11KT
Levels of plasma 11KT differed significantly among the migra-
tory categories analyzed (F2, 18 = 111; P < 0.001). Plasma 11KT in
silver eels (56 ± 11.4 ng/mL) was at least 10-fold higher than
11KT levels in both medium and small yellow eels. Most dramati-
cally, plasma 11KT in silver eels was > 600 fold higher than in small
yellow eels. Additionally, there was a significantly higher level of
plasma 11KT in medium yellow eels (3.9 ± 1.71 ng/mL) when com-
pared to small yellow eels (0.1 ± 0.02 ng/mL).
3.1.4. Gene expression
In silver eels, there was a highly significant decrease in the reti-
nal expression of the fwo gene (F2, 17 = 27.7; P < 0.001) compared to
both yellow eel categories (Fig. 3A). Even more dramatic was the
change in retinal expression of swo (F2, 17 = 681; P < 0.001), which
was nearly four orders of magnitude higher in silver eels than that
 G. Thomson-Laing et al. / General and Comparative Endocrinology xxx (2017) xxx–xxx
Fig. 2. Stage-specific differences in gonadosomatic index (GSI; A) and eye index (B)
of freshly-captured wild female shortfinned eels. Bar graph showing mean ± SEM
comparison of physical measurements among yellow small (YS; N = 7), yellow
medium (YM; N = 7) and silver (S; N = 7) migratory categories. Different lowercase
letters above bars represent significant differences (P < 0.05) between category
means.
observed in yellow eels of either small or medium size (Fig. 3B).
The expression of androgen receptor genes, ara and arb, in the
retina did not differ among migratory categories, but did show a
trend towards increased expression with advancing developmental
stage (ara; F2, 17 = 2.76 P = 0.09, arb; F2, 17 = 2.01; P = 0.17;
Fig. 3C, D).
3.2. In vitro effects of 11KT on mRNA levels of rod opsins and androgen
receptors in retinal tissue of shortfinned eels
3.2.1. Trial I
Increasing concentrations of 11KT from 0 to 100 ng/mL caused a
classical dose-dependent, nearly 20-fold, increase in normalized
transcript abundance of swo in the retina (F3,20 = 49.46; P < 0.001;
Fig. 4B). Increased swo expression compared to the control was evi-
dent for all cultures supplemented with 11KT, the effect being
most pronounced in explants treated with 10 ng/mL 11KT. Effects
of the androgen on normalized mRNA levels of fwo, ara and arb
were not detected (Fig. 4A, C, D).
3.2.2. Trial II
Dose-dependent increases of 11KT on the normalized transcript
abundance of swo in the yellow eel retina were confirmed
(F3,20 = 49.46; P < 0.001; Fig. 5C). Mean swo expression peaked in
retinal fragments exposed to 300 nM at 24 h, but did not differ sig-
Please cite this article in press as: Thomson-Laing, G., et al. The effects of migratory stage and 11-ketotestosterone on the expression of rod opsin genes in
the shortfinned eel (Anguilla australis). Gen. Comp. Endocrinol. (2017), http://dx.doi.org/10.1016/j.ygcen.2017.06.025
5
nificantly from that in incubations supplemented with 30 nM
11KT. Treatment with 11KT at 3 nM led to a significant decrease
(P < 0.01) in fwo gene expression to 0.87 ± 0.07 arbitrary units
(AU) compared to 2.7 ± 0.72 for the control (0 nM 11KT; Fig. 5A).
Higher concentrations of 11KT, also led to reductions, albeit not
significant, in fwo gene expression (1.6 ± 0.42 AU at 30 nM;
1.5 ± 0.20 AU at 300 nM).
In an attempt to test the specificity of the 11KT treatment, an
additional set of experiments was run with the inclusion of the
AR antagonist flutamide (multiple concentrations) in combination
with 30 nM 11KT. Firstly, flutamide alone did not significantly alter
either swo or fwo gene expression (Fig. 5B, D, left panels). None of
the concentrations of flutamide used were sufficient to block the
effects of 11KT on swo gene expression (Fig. 5D). Indeed, addition
of flutamide paradoxically resulted in a weak, non-significant and
dose-dependent increase in swo mRNA levels. Similarly, flutamide
had no effect on fwo expression (Fig. 5B), which in this trial also
was not affected by exposure to 11KT.
3.3. In vivo effects of 11KT on mRNA levels of rod opsins and androgen
receptors in the retina of yellow shortfinned eels
3.3.1. Plasma 11KT
Following a five-week exposure period, levels of 11KT in plasma
were significantly elevated in response to slow-release implants
containing 11KT (F1, 19 = 83.9; P < 0.001). There was no overall
effect of flutamide on plasma 11KT levels, although concentrations
trended higher in the flutamide-only group (2 ± 0.6 ng/mL) in com-
parison to control eels (0.7 ± 0.2 ng/mL). This trend was also seen
in fish treated with 11KT + flutamide (32 ± 9 ng/mL) when com-
pared to eels in the 11KT-only treatment (25 ± 8.3 ng/mL).
3.3.2. Gene expression
Exposure to 11KT (via slow-release implant) for five weeks
induced a significant, approximately 10-fold, increase in expres-
sion of the swo gene (F1, 18 = 18.0; P < 0.001; Fig. 6A) consistent
with stimulation of the retinal gene expression pattern associated
with silvering. However, 11KT had no effect on expression of fwo
(Fig. 6B). The expression of androgen receptors, ara and arb, simi-
larly, was not affected by 11KT (Fig. 6C, D).
In order to test the specificity of the 11KT treatment, an addi-
tional set of experiments was run with the inclusion of 0.1 mg flu-
tamide, in combination with 11KT. For genes whose expression
was not changed by 11KT, flutamide, as expected, had no effects
(Fig. 6A, C, D). By contrast, flutamide paradoxically appeared to
enhance the effects of 11KT on the expression of swo, but its effects
were not significant (Fig. 6B).
Additionally, exposure to 11KT for five weeks in this set of
experiments did not affect GSI (F1, 18 = 1.13; P = 0.30), relative
eye weight (F1, 18 = 0.135; P = 0.72) or eye index, although a weak
trend for an increase was evident for the latter (F1, 18 = 2.20;
P = 0.16; data not shown). Likewise, a change of the pectoral fins
into a pitch-black color was not observed.
4. Discussion
The silvering transformation, which pre-adapts anguillid eels to
an oceanic existence, is associated with dramatic changes in mor-
phology, physiology and behavior. Changes to the eye are particu-
larly noticeable – whilst to the lay observer, the principal change to
the eye is an increase in its size, changes to structure (ratio of cones
to rods; c.f., Braekevelt, 1988) and biochemistry (opsin expression;
Archer et al., 1995) have also been reported. Given that treatment
with the androgen 11KT induces many of the changes associated
6 G. Thomson-Laing et al. / General and Comparative Endocrinology xxx (2017) xxx–xxx

Fig. 3. Relative mRNA levels of freshwater opsin (fwo; A), seawater opsin (swo; B), androgen receptor-a (ara; C) and androgen receptor-b (arb; D) in retinas from freshly-
captured wild female shortfinned eels. Eels were classified into small yellow (YS; N = 7), medium yellow (YM; N = 6) and silver (S; N = 7) migratory categories. Expression is
normalized over that of ribosomal protein L36 (l36). Bars represent means ± 1 standard error. Different letters above bars indicate significant differences (P < 0.05) between
category means.

Fig. 4. Relative mRNA levels of freshwater opsin (fwo; A), seawater opsin (swo; B), androgen receptor-a (ara; C) and androgen receptor-b (arb; D) in retinal tissue from yellow
female shortfinned eels, subjected to in vitro incubation with 11-ketotestosterone (11KT; 0, 1, 10, 100 ng/mL) for one day. Expression is normalized over that of ribosomal
protein L36 (l36). Bars represent means ± 1 standard error of N = 6 for all groups. Different letters above bars indicate significant differences (P < 0.05) between treatment
means.

with silvering (Lokman et al., 2003; Sudo et al., 2012), we posited
that the changes to the eye, the ‘‘opsin switch”, would also be gov-
erned by 11KT. Our findings from in vivo and in vitro experiments
support the notion of a role for 11KT in regulation of the opsin
switch, at least with regard to the swo gene.
Fundamental to our study scope were published findings on
opsin switching – the down-regulation of fwo expression in
exchange for increases in swo expression – in European eel
(Carlisle and Denton, 1959; Hope et al., 1998), American eel
(Beatty, 1975) and A. japonica (Zhang et al., 2000). Prior to testing

Please cite this article in press as: Thomson-Laing, G., et al. The effects of migratory stage and 11-ketotestosterone on the expression of rod opsin genes in
the shortfinned eel (Anguilla australis). Gen. Comp. Endocrinol. (2017), http://dx.doi.org/10.1016/j.ygcen.2017.06.025
 G. Thomson-Laing et al. / General and Comparative Endocrinology xxx (2017) xxx–xxx 7

Fig. 5. Relative mRNA levels of freshwater opsin (fwo; A, B), seawater opsin (swo; C, D) in retinal explants from yellow female shortfinned eels, incubated in vitro in the
presence of 0–300 nM 11-ketotestosterone (11KT; A, C) or the androgen receptor blocker flutamide (B, D). Flutamide exposure was done in the absence or presence of 30 nM
11-KT. For all genes, mRNA quantity was calculated relative to the geometric mean of reference genes elongation factor-1a (elf) and ribosomal protein l36 (l36). Bars represent
means ± 1 standard error of N = 8 for all groups. Different letters above bars indicate significant differences (p-value < 0.05) between treatment means.

Fig. 6. Relative mRNA levels of freshwater opsin (fwo; A), seawater opsin (swo; B), androgen receptor-a (ara; C) and androgen receptor-b (arb; D) in retinas from yellow
female shortfinned eels, exposed for 5 weeks to implants that did, or did not, contain 11-ketotestosterone (11KT; 0.1 mg) and/or the androgen receptor blocker flutamide
(0.1 mg). For all genes, mRNA quantity was calculated relative to the geometric mean of reference genes elongation factor-1a (elf) and ribosomal protein l36 (l36). Bars
represent means ± 1 standard error of N = 4 (11KT + flutamide) or N = 6 (other groups) fish. Different letters above bars indicate significant differences (P < 0.05) between
treatment means.

whether 11KT regulates this switch in shortfinned eels, the pres- spectacular, decrease (around fivefold) in that of fwo; clearly,
ence of such a switch in this fish required confirmation. Here, we shortfinned eels employ opsin switching, which enables them to
report on a dramatic increase in swo expression (several efficiently absorb light of lower wavelengths prior to embarking
thousand-fold) in silver eels and a notable, but not nearly as on their oceanic journey. Support for this change in spectral

Please cite this article in press as: Thomson-Laing, G., et al. The effects of migratory stage and 11-ketotestosterone on the expression of rod opsin genes in
the shortfinned eel (Anguilla australis). Gen. Comp. Endocrinol. (2017), http://dx.doi.org/10.1016/j.ygcen.2017.06.025
8 G. Thomson-Laing et al. / General and Comparative Endocrinology xxx (2017) xxx–xxx
sensitivity was reinforced by differences in the amino acid
sequences between fwo and swo at key spectral tuning positions
(D83N, A124S, A292S). The same amino acid substitutions have
been identified for A. anguilla as critical to the shift in spectral sen-
sitivity of fresh and deep-sea opsins to green/yellow and blue
wavelengths, respectively (Archer et al., 1995).
The opsin switch coincided with changes in morphological hall-
marks of silvering, i.e., silver eels presenting with black pectoral
fins, changed coloration, increased eye size and greatly enlarged
relative gonad size (Petersen, 1896; Todd, 1981; Pankhurst,
1982), reinforcing that fish were correctly assigned to the different
life history categories. Importantly, the opsin switch also coincided
with stage-dependent changes in plasma levels of 11KT, but not
with obvious changes in androgen receptor transcript abundance.
Our hypothesis was that this association between increasing
levels of 11KT during silvering and the observed opsin switch
reflected a causal relationship. To test this, we resorted to conduct-
ing a suite of experiments in which we principally focused on the
effects of 11KT. We first developed an in vitro system, suitable for
incubations of eel retinal fragments. For this purpose, eyes were
promptly retrieved after euthanasia and the contents from within
the eye ball removed prior to very gently removing the black pig-
ment on the inside and the connective tissue on the outside of the
eye cup. The resulting retinal tissue was amenable to short-term
submerged culture in a basal medium, reflected in sufficient yields
of RNA and consistent responses to hormone treatments.
Using this in vitro bioassay system, we detected a strong dose-
dependent increase in mRNA levels of swo, but not fwo, in retinal
tissue after overnight treatment with 11KT. The response of retinal
tissue – an up-to 20-fold higher expression of swo – was consistent,
robust, and highly significant in two experimental trials. In con-
trast, effects of 11KT on transcript abundance of fwo were rather
variable, ranging between clearly absent and weakly inhibitory.
The inhibitory effect was only detectable at 3 nM (0.9 ng/mL),
higher doses of 11KT merely yielding a trend for a decrease in
fwo mRNA levels.
We then sought to examine the effects of 11KT in a more com-
plex system, the live animal. We obtained very comparable results
– thus, treatment of yellow eels with 11KT increased the transcript
abundance of swo in fish treated with 11KT, whereas that of fwo
remained unaffected. Oddly and unexpectedly, other measures,
e.g., fin color, eye size and relative gonad size, all known to be
responsive to 11KT treatment (Rohr et al., 2001; Setiawan et al.,
2012; Sudo et al., 2012; Zadmajid et al., 2015; Damsteegt et al.,
2016), were not affected in this experiment. Levels of 11KT were
elevated to  25–30 ng/mL in response to 11KT treatments, reach-
ing levels of biological relevance (20–50 ng/mL; Lokman et al.,
1998; Setiawan et al., 2012) as opposed to the levels described in
previous experiments. Plasma levels of 11KT in yellow eels after
3–4 weeks of implantation averaged  150 ng/mL (implant con-
taining 1 mg 11KT, BW  0.65 kg; Damsteegt et al., 2016)
and  300 ng/mL (0.2 mg 11KT, BW  0.36 kg; Zadmajid et al.,
2015), respectively. Furthermore, in a recent dose-response exper-
iment (Lokman, unpublished observations), plasma 11KT levels
averaged 1, 52, 96 and 101 ng/mL by 21 days after implantation
with 0, 0.03, 0.1 and 0.3 mg, respectively (mean BW 0.68 kg).
Therefore it is possible that a five-week period with circulating
11KT at  25–30 ng/mL was not long enough to elicit the same
level of response that was seen in experiments with at least four-
fold higher concentrations of plasma 11KT.
We previously posited that the small size of experimental ani-
mals may be responsible for the limited efficacy of 11KT
(Zadmajid et al., 2015), and we reinforce that idea here; however,
seasonal differences in responsiveness to 11KT could also be
important and research is currently underway to address this
notion. Regardless of the reasons behind the limited efficacy of
Please cite this article in press as: Thomson-Laing, G., et al. The effects of migratory stage and 11-ketotestosterone on the expression of rod opsin genes in
the shortfinned eel (Anguilla australis). Gen. Comp. Endocrinol. (2017), http://dx.doi.org/10.1016/j.ygcen.2017.06.025
11KT on a suite of end points, swo mRNA levels proved to be
increased greatly. This finding establishes the expression of the
swo gene as a highly sensitive indicator of exposure to 11KT.
Taken together, the results from our in vivo and in vitro studies
point to strong regulation of swo mRNA levels by 11KT to govern
the production of a novel light-absorbing pigment with different
spectral qualities in the retina. In contrast, a role for 11KT in regu-
lating expression of fwo is not supported. It is difficult to reconcile
these observations with the idea that 11KT would act as a ‘‘master
regulator” that drives a developmental switch by eliminating one
function and inducing that of another in a co-ordinated fashion
(e.g., thyroid hormone-regulated metamorphosis in amphibians;
see Laudet, 2011). Analyses of the fwo and swo promoters could
prove informative in future.
In amphibian metamorphosis, thyroid hormone levels slowly
increase during premetamorphosis, up-regulating the expression
of one of its own receptors, the thyroid hormone receptor-b
(TRb); up-regulation of the receptor allows the dramatic functional
remodeling that subsequently occurs at tremendous pace during
metamorphic climax (see review by Laudet, 2011). With this detail
in mind, we were interested to evaluate how the expression of the
presumed mediators of 11KT action, ara/Ara and arb/Arb, would
change in response to exposure with the ligand – but no effects
of 11KT on Ar expression were found.
We aimed to confirm that the observed effects of 11KT were
mediated by a classical genomic nuclear Ar, and therefore resorted
to blocking these receptors with flutamide, a known AR antagonist.
Interestingly, treatment with flutamide, consistently proved inef-
fective, or indeed, even seemingly weakly agonistic, both in vivo
and in vitro. However, flutamide tended to increase plasma 11KT
levels, an effect that could reflect a reduction in androgen-
mediated negative feedback. We have no explanation for these
observations, but refer to a study that identified similar agonistic
effects of flutamide on primary hippocampal neurons from rat
maintained in vitro – flutamide (10 nM–10 lM) blocked
androgen-dependent gene expressional regulation, but mimicked
its neuroprotective effects through the AR (Nguyen et al., 2007).
Reasons for those and similar observations by others (discussed
by Nguyen and co-workers) were not clear, but they reinforce that
agonistic effects of flutamide are not unheard of; in any case, future
testing of the pharmacological properties of flutamide on the eel
androgen receptor is needed to better understand how flutamide
interacts with the eel Ar.
As alternative explanation, the levels of flutamide that were
used (up to 300 nM) may have been insufficient to exert any effect.
Indeed, Kwon et al. (2005) reported that at 1 lM, flutamide was
ineffective at blocking androgen action on eel hepatocytes
in vitro, and that only at higher doses (10 lM) did it inhibit 11KT
action. These concentrations, however, are nearing those that are
cytotoxic to rat hepatocytes in vitro (LC50  50 lM for 3 h incuba-
tions; Maruf and O’Brien, 2014).
In conclusion, we have demonstrated that the eel retina
responds to androgens, both in vivo and in vitro, by increasing
mRNA levels of seawater opsin; evidence for a decrease in fwo
expression in response to 11KT could not found. These observa-
tions partly explain the ‘‘opsin switch” that occur as eels transform
from yellow into silver eels and further the concept of a major role
of 11KT in pre-adapting freshwater eels to an oceanic existence, a
pre-requisite for these fish to complete their extra-ordinary life
cycle.
Acknowledgments
We express our appreciation to Mr Matthew Downes (Depart-
ment of Zoology, University of Otago) for help with histology.
Thanks are further due to Mr Tuan Nguyen and Miss Jolyn Chia,
 G. Thomson-Laing et al. / General and Comparative Endocrinology xxx (2017) xxx–xxx
who helped to care for the fish as well as assisting with sampling
alongside other members of the REPRONZ lab (Department of Zool-
ogy, University of Otago). This work was supported by a University
of Otago (Zoology) Performance Enhancement Grant, PBRF_ML63
(2015 and 2016) to PML.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ygcen.2017.06.
025.
References
Aoyama, J., Miller, M., 2003. The silver eel. In: Aida, K., Tsukamoto, K., Yamauchi, K.
(Eds.), Eel Biology. Springer, Japan, pp. 107–117.
Archer, S., Hope, A., Partridge, J.C., 1995. The molecular basis for the green-blue
sensitivity shift in the rod visual pigments of the European eel. Proc. Royal Soc.
262B, 289–295.
Aroua, S., Schmitz, M., Baloche, S., Vidal, B., Rousseau, K., Dufour, S., 2005. Endocrine
evidence that silvering, a secondary metamorphosis in the eel, is a pubertal
rather than a metamorphic event. Neuroendocrinology 82, 221–232.
Beatty, D.D., 1975. Visual pigments of the American eel Anguilla rostrata. Vision Res.
15, 771–776.
Braekevelt, C.R., 1988. Retinal fine structure in the European eel Anguilla anguilla. VI.
Photoreceptors of the sexually immature silver eel stage. Anat. Anz. 166, 23–31.
Bridges, C.D.B., 1972. The rhodopsin-porphyropsin visual system. In: Dartnall, H.A.
(Ed.), Photochemistry of Vision. Springer, Berlin Heidelberg, pp. 417–480.
Bruijs, M.M., Durif, C.F., 2009. Silver eel migration and behaviour. In: G. van den
Thillart, S. Dufour, J.C. Rankin (Eds.), Spawning Migration of the European Eel
Springer, Netherlands, 65–95.
Carlisle, D.B., Denton, E.J., 1959. On the metamorphosis of the visual pigments of
Anguilla anguilla (L). J. Mar. Biol. Assoc. U.K. 38, 97–102.
Damsteegt, E.L., Ozaki, Y., McCormick, S.P., Lokman, P.M., 2016. Triacylglyceride
physiology in the short-finned eel, Anguilla australis—the effects of androgen.
Am. J. Physiol. Regul. Integr. Comp. Physiol. 310, R422–R431.
Durif, C., Dufour, S., Elie, P., 2005. The silvering process of Anguilla anguilla: a new
classification from the yellow resident to the silver migrating stage. J. Fish Biol.
66, 1025–1043.
Hagihara, S., Aoyama, J., Limbong, D., Tsukamoto, K., 2012. Morphological and
physiological changes of female tropical eels, Anguilla celebesensis and Anguilla
marmorata, in relation to downstream migration. J. Fish Biol. 81, 408–426.
Hope, A.J., Partridge, J.C., Hayes, P.K., 1998. Switch in rod opsin gene expression in
the European eel, Anguilla anguilla (L.). Proc. Royal Soc. 265B, 869–874.
Ikeuchi, T., Todo, T., Kobayashi, T., Nagahama, Y., 1999. CDNA cloning of a novel
androgen receptor subtype. J. Biol. Chem. 274, 25205–25209.
Jessop, B.M., 1987. Migrating American eels in Nova Scotia. Trans. Am. Fish. Soc.
116, 161–170.
Kwon, H.C., Choi, S.H., Kim, Y.U., Son, S.O., Kwon, J.Y., 2005. Androgen action on
hepatic vitellogenin synthesis in the eel, Anguilla japonica is suppressed by an
androgen receptor antagonist. J. Steroid Biochem. Mol. Biol. 96, 175–178.
Laudet, V., 2011. The origins and evolution of vertebrate review metamorphosis.
Currr. Biol. 21, R726–R737.
Lokman, P.M., 2016. Migration, gamete biology and spawning. In: Arai, T. (Ed.),
Biology and Ecology of Anguillid Eels. CRC Press, Taylor & Francis Group, Boca
Raton, pp. 206–224. etc.
Lokman, P.M., Vermeulen, G.J., Lambert, J.G., Young, G., 1998. Gonad histology and
plasma steroid profiles in wild New Zealand freshwater eels (Anguilla
dieffenbachii and A. australis) before and at the onset of the natural spawning
migration. I. Females. Fish Physiol. Biochem. 19, 325–338.
Lokman, P.M., Rohr, D.H., Davie, P.S., Young, G., 2003. The physiology of silvering in
anguillid eels: Androgens and control of metamorphosis from the yellow to
silver stage. In: Aida, K., Tsukamoto, K., Yamauchi, K. (Eds.), Eel Biology.
Springer, Japan, pp. 331–349.
Lokman, P.M., Wylie, M.J., Downes, M., Di Biase, A., Damsteegt, E.L., 2015. Artificial
induction of maturation in female silver eels, Anguilla australis: the benefits of
androgen pre-treatment. Aquaculture 437, 111–119.
Please cite this article in press as: Thomson-Laing, G., et al. The effects of migratory stage and 11-ketotestosterone on the expression of rod opsin genes in
the shortfinned eel (Anguilla australis). Gen. Comp. Endocrinol. (2017), http://dx.doi.org/10.1016/j.ygcen.2017.06.025
9
Maruf, A.A., O’Brien, P., 2014. Flutamide-induced cytotoxicity and oxidative stress in
an in vitro rat hepatocyte system. Oxid. Med. Cell. Longev. 2014, 398285.
Nguyen, T.-V.V., Yao, M., Pike, C.J., 2007. Flutamide and cyproterone acetate exert
agonist effects: induction of androgen receptor-dependent neuroprotection.
Endocrinology 148, 2936–2943.
Okamura, A., Yamada, Y., Yokouchi, K., Horie, N., Mikawa, N., Utoh, T., Tsukamoto, K.,
2007. A silvering index for the Japanese eel Anguilla japonica. Environ. Biol.
Fishes 80, 77–89.
Pankhurst, N.W., 1982. Relation of visual changes to the onset of sexual-maturation
in the European eel Anguilla anguilla (L). J. Fish Biol. 21, 127–140.
Petersen, C.G.J., 1896. The common eel (Anguilla vulgaris, Turton) gets a particular
breeding-dress before its emigration to the sea. The bearing of this fact on the
classification and on the practical eel-fisheries. Rep. Danish Biol. Sta. V, 5–35.
Rohr, D.H., Lokman, P.M., Davie, P.S., Young, G., 2001. 11-Ketotestosterone induces
silvering-related changes in immature female short-finned eels, Anguilla
australis. Comp. Biochem. Physiol. 130A, 701–714.
Schmidt, J., 1923. Breeding places and migrations of the eel. Nature 111, 51–54.
Setiawan, A.N., Lokman, P.M., 2010. The use of reference gene selection programs to
study the silvering transformation in a freshwater eel Anguilla australis: a
cautionary tale. BMC Mol. Biol. 11, 75.
Setiawan, A.N., Ozaki, Y., Shoae, A., Kazeto, Y., Lokman, P.M., 2012. Androgen-
specific regulation of FSH signalling in the previtellogenic ovary and pituitary of
the New Zealand shortfinned eel, Anguilla australis. Gen. Comp. Endocrinol. 176,
132–143.
Shao, Y.T., Wang, F.-Y., Fu, W.-C., Yan, H.Y., Anraku, K., Chen, I.S., Borg, B., 2014.
Androgens increase iws opsin expression and red sensitivity in male three-
spined sticklebacks. PLoS One 9, 6.
Sherwood, N.M., Crim, L.W., Carolsfeld, J., Walters, S.M., 1988. Sustained hormone
release. I. Characteristics of in vitro release of gonadotropin-releasing hormone
analogue (GnRH-A) from pellets. Aquaculture 74, 75–86.
Sudo, R., Suetake, H., Suzuki, Y., Utoh, T., Tanaka, S., Aoyama, J., Tsukamoto, K., 2011.
Dynamics of reproductive hormones during downstream migration in females
of the Japanese eel, Anguilla japonica. Zool. Sci. 28, 180–188.
Sudo, R., Tosaka, R., Ijiri, S., Adachi, S., Aoyama, J., Tsukamoto, K., 2012. 11-
ketotestosterone synchronously induces oocyte development and silvering-
related changes in the Japanese eel, Anguilla japonica. Zool. Sci. 29, 254–259.
Suliman, T., Flamarique, I., 2014. Visual pigments and opsin expression in the
juveniles of three species of fish (rainbow trout, zebrafish, and killifish)
following prolonged exposure to thyroid hormone or retinoic acid. J. Comp.
Neurol. 522, 98–117.
Tesch, F.W., Rohlf, N., 2003. Migration from Continental Waters to the Spawning
Grounds. In: Aida, K., Tsukamoto, K., Yamauchi, K. (Eds.), Eel Biology. Springer,
Japan, pp. 223–234.
Todd, P.R., 1981. Morphometric changes, gonad histology, and fecundity estimates
in migrating New-Zealand fresh-water eels (Anguilla spp.). N.Z. J. Mar.
Freshwater Res. 15, 155–170.
Untergasser, A., Cutcutache, I., Koressaar, T., Ye, J., Faircloth, B.C., Remm, M., Rozen,
S.G., 2012. Primer3 - new capabilities and interfaces. Nucleic Acids Res. 40,
e115.
van Ginneken, V.T., Maes, G., 2005. The European eel (Anguilla anguilla, Linnaeus),
its lifecycle, evolution and reproduction: a literature review. Rev. Fish Biol. Fish.
15, 367–398.
Wang, F.Y., Fu, W.C., Wang, I.L., Yan, H.Y., Wang, T.Y., 2014. The giant mottled eel,
Anguilla marmorata, uses blue-shifted rod photoreceptors during upstream
migration. PLoS One 9, 11.
Wood, P., Partridge, J.C., 1993. Opsin substitution induced in retinal rods of the eel
(Anguilla anguilla (L.)): a model for G-protein-linked receptors. Proc. Royal Soc.
254B, 227–232.
Wood, P., Partridge, J.C., De Grip, W.J., 1992. Rod visual pigment changes in the elver
of the eel Anguilla anguilla L. measured by microspectrophotometry. J. Fish Biol.
41, 601–611.
Zadmajid, V., Falahatimarvast, A., Damsteegt, E.L., Setiawan, A.N., Ozaki, Y., Shoae,
A., Lokman, P.M., 2015. Effects of 11-ketotestosterone and temperature on
inhibin subunit mRNA levels in the ovary of the shortfinned eel, Anguilla
australis. Comp. Biochem. Physiol. 187B, 14–21.
Zhang, H., Futami, K., Horie, N., Okamura, A., Utoh, T., Mikawa, N., Okamoto, N.,
2000. Molecular cloning of fresh water and deep-sea rod opsin genes from
Japanese eel Anguilla japonica and expressional analyses during sexual
maturation. FEBS Lett. 469, 39–43.