The brain-pituitary-gonadal axis of the three-spined

stickleback, Gasterosteus aculeatus

Yi Ta Shao

Department of Zoology

Stockholm University

2012

The brain-pituitary-gonadal axis of the three-spined stickleback, Gasterosteus aculeatus

Doctoral dissertation 2012

Yi Ta Shao yita.shao@zoologi.su.se ; itshao@gate.sinica.edu.tw Department of Zoology Stockholm Universirty SE-106 91 Stockholm Sweden

© Yi Ta Shao, Stockholm 2012. ISBN 978-91-7447-568-5 Cover illustration by Bertil Borg Printed in Sweden by US-AB, Stockholm 2012 Distributor: Department of Zoology, Stockholm University

To

My family

0

The brain-pituitary-gonadal axis of the three-spined

stickleback, Gasterosteus aculeatus

Doctoral dissertation 2012 Yi Ta Shao Department of Zoology Stockholm University 106 91 Stockholm

Abstract

The seasonal reproduction of the three-spined stickleback is stimulated by long day photoperiod. As in other vertebrates, the reproductive system of stickleback is regulated by the brain-pituitary-gonadal (BPG) axis which is largely controlled by feedback effects. Both negative and positive feedback effects on the BPG axis have been found in fish. So far, the roles feedback effects on the BPG axis play in the photoperiodic regulation of seasonal reproduction are still unclear. This thesis focused on the photoperiodic regulation and gonadal feedback effects on the gene expressions of gonadotropin (GtH) and gonadotropin-releasing hormones (GnRH) in the brain and pituitary, and how gonadal feedback regulated the steroid homeostasis in stickleback. Both GnRH2 and GnRH3 mRNA was found in the hypothalamus. Higher expression levels of both GnRH2 and 3 in breeding than in post-breeding males suggested that they are both involved in seasonal reproduction. There was no evidence for a role of GnRH3, which may be the dominating form, in the photoperiodic control of reproduction. However, the polarity of the feedback effect on gnrh3 gene expression may turn from positive to be negative when the males went into post-breeding state. Tapeworm, Schistocephalus solidus, infection inhibited the reproduction of sticklebacks. However, the infection caused higher expression levels of both GnRHs and GtHs genes, which may be due to feedback effect on the BPG axis.

Under short day, both lh-β and fsh-β were suppressed by low androgen levels. This negative feedback may inhibit maturation completely, unless a rise of androgens triggers positive feedback under long day. The change in feedback polarity may result in all or nothing maturation. Furthermore, the androgen inhibitory effect on lh-β and fsh-β under short day could be abolished by aromatase inhibitor, which means the estrogen may cause negative feedback in males under short day.

There was no compensation effect on plasma androgen level in fully mature hemi-castrated fish. However, both testosterone and 11-ketoandrostenedione treatments increased plasma levels much less in sham-operated fish than in castrated ones, indicating that homeostatic mechanisms are nevertheless present.

1

TABLE OF CONTENTS

LIST OF PAPERS 2

INTRODUCTION 3

Reproductive rhythms of fish 3

Brain-pituitary-gonadal axis 4

Gonadotropins (GtH) 5

Gonadotropin-releasing hormones (GnRH) 6

Reproduction of the stickleback 9

Infection with the parasite, Schistocephalus solidus 11

AIM OF THE THESIS 12

MATERIAL AND METHODS 13

Experiment animal 13

Treatments and sampling 13

Implants 13

Operation 14

Sampling 14

Experimental protocol 15

Cloning and bioinformatic analysis 18

Real time quantitative PCR (q-PCR) 19

GnRH 19

GtH 20

In situ hybridization 20

Radioimmunoassay (RIA) 22

Statistics 22

RESULTS AND DISCUSSION 22

GnRH gene expression in stickleback (paper I, II) 22

The feedback effect on GtH gene expression (paper III) 25

Homeostasis of circulating androgens levels (paper IV) 27

SUMMARY 28

ACKNOWLEDGEMENTS 30

REFERENCES 31

2

LIST OF PAPERS

Paper I Yi Ta Shao, Yung Che Tseng, Chia Hao Chang, Pung Pung Hwang and

Bertil Borg. 2012. GnRH mRNA levels in the male three-spined

stickleback, Gasterosteus aculeatus; effects of photoperiod,

gonadectomy and breeding-postbreeding condition. MS

Paper II Yi Ta Shao, Yung Che Tseng, Susanne Trombley, Pung Pung Hwang,

Monika Schmitz and Bertil Borg. 2012. Schistocephalus solidus

infections increase gonadotropin and salmon gonadotropin releasing

hormone mRNA levels in the three-spined stickleback, Gasterosteus

aculeatus. Parasitol. Int. 61: 470-474.

Paper III Yi Ta Shao, Mia Arvidsson, Susanne Trombley, Rüdiger Schulz, Monika

Schmitz and Bertil Borg. 2012. Androgen feedback effects on LH and

FSH, and photoperiodic control of reproduction in male three-spined

sticklebacks, Gasterosteus aculeatus. MS

Paper IV Yi Ta Shao, Rüdiger Schulz and Bertil Borg. 2012. Homeostasis of

circulating androgen levels in the male three-spined stickleback

(Gasterosteus aculeatus). MS

3

INTRODUCTION

Reproductive rhythms of fish

Reproduction is often a cyclic phenomenon in fish. In many tropical fishes, the

relatively constant environment allows them to have extended spawning seasons

which may include multiple breeds throughout the year (Pauly, 1998; Petersen &

Warner, 2002). Other fishes, however, especially those living in the temperate

zones or other places with clear seasonal climatic variability, have a predictable and

restricted spawning period. The reproduction periods of these species usually last

only several weeks or even just a couple of days. The specific reproductive and

maturational timing of fishes is a result of adaption and evolution. It is critical for

the male and female to be ready to breed at the same time and to ensure the young

hatch and commence feeding in the right season (Billard and Breton, 1978).

In order to spawn at the right time, fishes use both environmental cues and

endogenous cycles for timing of growth and maturation. Various teleosts, due to

their unique evolution and adaptations, can use almost all occurring environmental

events as exogenous timing factors. Many environmental cues, such as

photoperiod, temperature, salinity, lunar phase, pH, food abundance and social

interactions, are known to regulate spawning and sex maturation, (Stacey, 1984).

Nevertheless, how these external factors regulate the occurrence of maturation and

spawning is not clear.

The photoperiod is the most well-studied factor that synchronizes seasonal

reproduction timing. It is the major cue used in many temperate species such as

salmons, sticklebacks and seabasses (Begtashi et al. 2004; Bromage et al., 1994;

Howell et al. 2003; Randall et al., 1998 Borg 2010). The earliest recorded

photoperiodic experiment on fishes was conducted in 1937 and used the brook trout

(Salvelinus fontinalis). The experiment demonstrated that when the trout were

exposed to increasing light phase early in the year followed by rapidly decreasing

light phase, their second breeding season would appear 3 months earlier than that of

the control group (Hoover & Hubbard, 1937). Changing photoperiods have been

4

observed to accelerate maturation or advance spawning time also in many other

species of Salmonidae, such as rainbow trout (Oncorhynchus mykiss, Bromage, 1992),

coho salmon (O. kisutch, Withler, et al., 1998), masu salmon (O. masou, Amano et al.,

1995) and Atlantic salmon (Salmo salar, Taranger, et al., 1998). In the three-spined

stickleback (Gasterosteus aculeatus), sexual maturation is stimulated by long

photoperiod, but inhibited by short photoperiod (Baggerman, 1985; 1989; Borg,

1982a; Borg, 2010).

The use of artificial photoperiod manipulation has increasingly become a

standard operating practice in the aquaculture industry to control maturation or

breeding of a number of species, such as Atlantic salmon (Porter et al. 1999; Endal et

al. 2000), striped trumpeter (Latris lineate) (Morehead et al. 2000) and European

seabass (Dicentrachus labrax) (Begtashi et al. 2004). It is also effective in Atlantic

cod (Gadus morhua) (Davie et al. 2003; Hansen et al. 2001; Almeida et al. 2009) and

in other species, such as gilthead seabream (Sparus aurata) and white seabass

(Atractoscion nobilis) (Aalbers, 2008; Perdikaris, 2001).

However, the mechanism of how photoperiods influence the reproductive

process of fishes is still unclear. Yet it is known that fishes synchronize their

physiological functions to the light/dark cycle, and that the endogenous periodicities

of physiological processes are responsible in part for seasonal reproduction.

Brain-pituitary-gonadal axis

As in other vertebrates, the reproductive system of fishes is regulated by the

brain-pituitary-gonadal endocrine axis which is regulated by feedback effects (Fig 1).

Both positive (stimulating) and negative (suppressing) feedback effects on the

BPG-axis have been reported in fishes (Crim and Evans, 1979, Billard et al., 1977;

1978). For example, gonadal hormones, i.e. testosterone (T), 11-ketotestosterone

(11KT) or estradiol (E2) acting on the pituitary and/or brain level, may act on the

pituitary and/or brain level to suppress or stimulate gonadotropin (GtH) secretion.

The negative feedback effects on the BPG-axis are important mechanisms to keep the

levels of sex steroids stable; and the positive feedback may play an important role in

5

controlling the onset of reproductive events in fishes.

Figure 1. A proposed model of the Brain-pituitary- gonadal axis. Feedback loops control the synthesis and concentrations of sex hormones. GtHs: gonadotropins; GnRH: gonadotropin-releasing hormone; GnIF: gonadotropin inhibitory factors; T: testosterone; E2: estradiol.

Gonadotropins (GtHs)

Teleosts have two types of gonadotropins - luteinizing hormone (LH) and

follicle-stimulating hormone (FSH) as in tetrapods. They share a similar α peptide

chain but have different β chains. The two gonadotropins stimulate different

receptors, due to the specific β chain.

The gonadotropins contents in the blood or pituitary can be measured by

radioimmunoassy (RIA). However, a specific antibody is required for this method,

and it is unpractical to be made in a small species. On the other hand, the synthesis

rate of GtHs can be estimated by measuring the lh-β or fsh-β mRNA content of the

pituitary, which makes it possible to study the activity of GtH in relative small species,

such as three-spined stickleback (e.g. Hellqvist, et al., 2006).

Gonadotropin-releasing hormones (GnRH)

The secretion of GtHs is controlled by the hypothalamus, which can be

stimulated by gonadotropin-releasing hormone (GnRH) (Goos, 1991; Peter, et al.,

6

1991), or inhibited by other substances, such as the peptide -

gonadotropin-inhibitory hormone (GnIH) (Tsutsui et al. 2012) and dopamine

(Goos, 1991; Peter, et al., 1991). The primary structure of GnRHs contains an

amino-terminus (pGlu), a NH2 terminus and in all 10 amino acids. The first

GnRH isoform characterized was isolated from mammalian hypothalamii and

was named mammalian GnRH (mGnRH). mGnRH is known as a hormone that

regulates the reproductive system by stimulating release of gonadotropic

hormones from the anterior pituitary gland (Burgus et al., 1972). mGnRH is the

most widespread form in mammals, and is also found in European eel (Anguilla

anguilla) (King et al., 1990). In birds, two GnRH molecular variants have been

isolated from chicken brains, cGnRH-I (King and Millar, 1982) and cGnRH-II

(Miyamoto et al., 1984). cGnRH-II is not only present in birds, but also in many

fishes such as European eel (King et al., 1990), African catfish (Clarias gariepinus)

(Goos, 1989) and goldfish (Carassius auratus) (Yu et al., 1988). The first form

characterized from fish is sGnRH which was isolated from chum salmon

(Oncorhynchus keta) by Sherwood et al. (1983).

At present, more than 11 structural variants of GnRH have been isolated

and characterized from fishes and lamprey (Table 1). Almost all fish species

investigated have at least two GnRH forms in the central nervous system; and

most bony fishes have three forms of GnRH (review in Somoza, et al., 2002).

Fernald and White (1999) proposed that the multiple GnRH forms evolved from

3 different evolutionary lineages; the mesencephalic, telencephlic and

hypothalamic lineages.

According to the features of the peptide sequences and the dispersal

pattern of GnRH neurons, those isoforms can be divided into different groups. In

teleost, they are GnRH1, GnRH2 (cGnRH II) and GnRH3 (sGnRH) (review in Okubo

& Nagahama 2008). GnRH2 has been found in all studied teleost species, both in

Chondrichthyes and Osteichthyes (Somoza et al., 2002). GnRH3 is also

widespread amoung fishes, but is absent in Chondrichthyes, sturgeons,

lungfishes and in some teleost species, i.e. African sharptooth catfish (Clarias

gariepinus) and broadhead catfish (Clarias macrocephalus) (Bogerd et al., 1994;

7

Ngamvongchon et al., 1992). The gene of hypothalamic GnRH (GnRH1) lineage

can be missing in some species, such as goldfish (Carassius auratus) (Lin and

Peter, 1996 [22]) and masou salmon (Oncorhynchus masou) (Suzuki et al., 1992

[23]). Moreover, almost the whole genome of zebrafish (Danio rerio) has been

sequenced, but the gene of GnRH1 has not been found in this species (Steven et

al. 2003; Palevitch et al. 2007). A possible reason is that the gene of this GnRH

lineage has been lost or become a pseudogene during evolution [review in

Okubo & Nagahama 2008]. In the species which do not have hypothalamic

GnRH, such as salmon, zebrafish or goldfish, telencephalic GnRH (GnRH3) may

take on the role [review in Okubo & Nagahama 2008]. The GnRH3 expressing

neuron were seen in the pre-optic area, as well as in the terminal nerve ganglion,

which project their axons into the pituitary, indicating that sGnRH would act as

the main stimulator of the pituitary hormones in the absence of hypothalamic

GnRH (Okubo and Nagahama, 2008).

Table 1. The GnRHs found in fish, included lamprey (Partly summarized from Somoza, et al.

2002)

GnRH form Species Authors sGnRH chum salmon Sherwood et al., 1983 (Oncorhynchus keta) lGnRH I sea lamprey Sherwood et al., 1986 (Petromyzon marinus) lGnRH II sea lamprey Sower et al., 1993 cfGnRH broadhead catfish Ngamvongchon et al., 1992 (Clarias macrocephalus) dfGnRH spiny dogfish Lovejoy et al., 1992 (Squalus acanthias) snGnRH snook Sherwood et al., 1993 (Centropomus undecimalis) sbGnRH seabream Powell et al., 1994 (Sparus aurata) hrGnRH herring Carolsfeld et al., 2000 (Clupea harengus pallasi) wfGnRH whitefish Adams et al., 2002 (Coregonus clupeaformis) mdGnRH medaka Okubo et al., 2000 (Oryzias latipes) pjGnRH pejerrey Montaner et al., 2001 (Odontesthes bonariensis) ?GnRH Tilipa Cao & Ding, 2006 (Oreochromis aurea) The GnRHs listed below are present in fishes,

but were first identified from other species mGnRH rat Burgus et al., 1972

8

(Rattus norvegicus) cGnRH I chicken King and Millar, 1982 (Gallus gallus) cGnRH II chicken Miyamoto et al., 1984

sGnRH is also widespread among fishes, but is absent in Chondrichthyes,

sturgeons, lungfishes and in some teleost species, i.e. African sharptooth catfish

(Clarias gariepinus) and broadhead catfish (Clarias macrocephalus) (Bogerd et al.,

1994; Ngamvongchon et al., 1992) (Fig 2).

Figure 2. Hypothetical scheme for the evolution of gonadotropin-releasing hormone (GnRH) in vertebrates. hypothalamic lineage (GnRH 1), mesencephalic lineage (GnRH 2) and telencephlic lineage (GnRH 3). Mya: million years ago. (Okubo and Nagahama, 2008)

In teleost, the GnRH1 vary between different taxa, the seabream form

(sbGnRH) is limited to some orders in Acanthopterygii (review in Somoza et al.,

2002), and the catfish form (cfGnRH) has only been found in the genus Clarias

(Ngamvongchon et al., 1992; Bogerd et al., 1994). However, the gene of

hypothalamic GnRH lineage may be missing in some species, like goldfish

(Carassius auratus) (Lin and Peter, 1996) and masou salmon (Oncorhynchus

masou) (Suzuki et al., 1992) (Fig 3).

rhesus

9

Figure 3. Schematic diagram illustrating the molecular evolution of hypothalamic lineage GnRH in vertebrates. Its ancestral form was certainly mGnRH. It underwent great structural diversification in non-mammalian lineages, leading to 6, 2, 1 and 1 structural variant in the teleost, amphibian, reptilian and avian lineages respectively. Representative species having the respective molecular forms are shown. (Okubo and Nagahama, 2008)

Reproduction of the stickleback

The three-spined stickleback, Gasterosteus aculeatus, is one of the most

studied fishes as regards the control of seasonal reproductive cycles (Borg, 2010).

The natural breeding season of the species is late spring-early summer when

photoperiods are long and temperature high. Breeding male sticklebacks have bright

breeding colours, with blue eyes and red belly, and show nest building behavior. The

nest is built of algae glued together with threads of the protein spiggin produced by

the kidney, which hypertrophies under androgen stimulation (Jakobsson et al. 1999;

Borg, 2007).

After the breeding season, these characters disappear. In the yearly cycle, the

highest kidney development is found in the breeding season and the lowest in the

post-breeding period in late summer (Borg, 1982). A similar pattern is also found in

plasma levels of the androgen 11-ketotestosterone (Mayer et al., 1990) and in

pituitary lh-β and fsh-β mRNA levels (Hellqvist et al., 2006). Long photoperiod, but

not short photoperiod, stimulates maturation in the stickleback (Baggerman, 1957;

Borg, 1982). Both lh-β and fsh-β mRNA level are higher in males kept under long

photoperiod (L:D 16:8) than under short photoperiod (L:D 8:16) in winter (Hellqvist,

et al. 2008).

Andersson et al. (1995) found three distinct groups of GnRH-ir perikarya in the

10

stickleback brain; in the nucleus olfactoretinalis, the nucleus anterioris

periventricularis and in nucleus lateralis tuberis. The proximal pars distalis of the

pituitary, where the gonadotropic cells are located, was innervated by GnRH-ir fibers.

Via high performance liquid chromatography (HPLC) combined with

radioimmunoassay (RIA), at least two peaks with sGnRH-immunoreactive material

using of whole stickleback brain extracts has been identified. One form was

tentatively identified as sGnRH (GnRH3), whereas the other(s) are still unclear

(Andersson et al., 1995). In genome databases (Ensembl Genome Browser system),

two types of GnRHs, GnRH2 (cGnRH II) and GnRH3 (sGnRH) have been found in DNA

level (Fig 4a, b). However, GnRH1 is absent in stickleback genome. Even though the

neighbor genes of GnRH1 were found on chromosome XII, no possible GnRH gene

was present in the appropriate area (Fig 4c) (Tostivint 2011).

a.

b.

11

Figure 5. The parasite S. solidus and its host. The numbers indicate the totally 8 parasites found in this stickleback (photo by Yi Ta Shao)

c.

Figure 4. Synteny map comparing the orthologues of the (a) GnRH2, (b) GnRH3 and (c) GnRH3 locus and the genes flanking it in humans (Homo sapiens), chicken (Gallus gallus), Clawed frog (Xenopus tropicalis), medaka (Oryzias latipes), stickleback (Gasterosteus aculeatus), pufferfish (Tetraodon nigroviridis) and zebrafish (Danio rerio). (Tostivint 2011).

Infection with the parasite, Schistocephalus solidus

The three-spined stickleback is an intermediate host for the cestode S. solidus

(Fig 5). The life cycle of S. solidus includes a free swimming larval stage (coracidium),

a second larval stage hosted in cyclopoid copepods (procercoid), a third larval stage

hosted in the stickleback (plerocercoid) and an adult stage in birds (Fig 6). When the

fish is eaten by a bird, the plerocercoid will mature quickly and start to reproduce

within a few weeks (Schultz et al. 2006). The parasitized fish often has a distended

abdomen which contains one or multiple worms. The total mass of those parasites

may be close to, or even exceeding, the

mass of their host fish. S. solidus can

have a devastating effect on stickleback

sexual maturation in some populations,

but less in others (McPhail and Peacock,

1983). In the cases of Scout Lake (Heins

and Baker, 2003) and Walby Lake, Alaska

(Heins and Baker, 2003; Heins, et al.,

2010), most of the host females

matured eventually, despite the eggs

mass of those females may be lower than in uninfected fish. The infected males

2 1

3 4

5

8

7

6

12

collected from Victoria Park pond had lower kidney somatic index (KSI) and were less

aggressive in breeding behavior than uninfected males. However, no difference in KSI

or reproductive behavior was found between the infected and uninfected males from

Kendoon Loch (MacNab et al., 2009),). Sexual maturation of infected sticklebacks

may be related to the mass of parasites, with more heavily infected fish being less

like to mature (review in Barber, 2007), or due to the genotypes of the host

populations (MacNab, et al. 2009).

Figure 6. Life cycle of S. soldus. a. free swimming larvae, b. 2nd stage larvae host in copepods, c. 3rd stage larvae in stickleback, d.final host in birds. (Dubinina, 1980)

AIM OF THE THESIS

The main objective of this thesis is to understand the BPG axis response to the

photoperiodic seasonal reproduction by using three-spined stickleback as the model

animal. This thesis focused on the photoperiodic regulation and gonadal feedback

effects on the genes expressions of GnRH and GtH in the levels of brain and pituitary.

This study aimed to find out:

Which GnRH forms are present in the three-spined stickleback? What role do

they play in the control of seasonal reproduction and in gonadal feedback?

How does Schistocephalus solidus influence the host’s GnRH and GtH genes

expression?

Does photoperiod change the feedback effects on lh-β and fsh-β and cause

all-or-nothing maturation?

Are androgen levels in male stickleback regulated by homeostasis?

a. b.

c. d.

13

MATERIAL AND METHODS

Experimental animals

Except for paper II, the sticklebacks were caught in the Ö resund, southern Sweden

during the winter and transported to Stockholm University. The animals were stored

in aquaria containing artificial brackish water (0.5% salinity) which was aerated and

filtered. The bottom was covered with sand; ceramic pots and tubes provided hiding

places. The fish were feed daily with frozen bloodworms, Artemia or mysids.

Treatments and sampling

Implants

Eight types of implants were used in paper III and IV to carry drugs into the

fish body (Table 2). All implants were made from medical grade Silastic tube (ID

0.64mm; OD 1.19mm) which was cut to 5mm in length and filled with crystalline

T (Fluka), crystalline 11-ketoandrostenedione (11KA) (4- Androstene- 3,11,17-

trione; Sigma) or aromatase inhibitor (AI, fadrozole; CGS 16949 A, 4-(5,6,7,8-

tetrahydroimidazo[1,5-α]pyridin-5-yl) benzonitrile; a gift from Novartis). Both

ends were sealed with silicone glue. For high dosage of 11KA (11KA(H)), the

ordinary (low dose) 11KA implants (11KA(L)) were perforated three times with a

31G syringe needle. 11KA is a non-aromatisable androgen, whereas it can be

converted to 11-ketotestosterone (11KT) in the fish.

Other types of T implants were made from 5 mm Silastic tubes, but filled

with different concentrations of T (0.10; 0.20; 0.25 or 0.50 %) dissolved in cacao

butter, and not sealed at the ends. Those implants made it possible to give a

lower dosage than by using crystalline implants. The cacao butter-implants were

stored at 4 ⁰C before use. Empty implants or the implants filled with cacao butter

alone were used as controls.

The experiments were carried out with the permission of the Stockholm

Northern Animal Experiment Ethical Committee

14

Operations

Fish were either castrated or sham-operated and were given different types

of implants. The fish were anaesthesized with c. 0.1% 2-phenoxyethanol (up to

2009) or 0.025% MS-222 (Ethyl 3-aminobenzoate, methanesulfonic acid salt)

solution (0.25 g MS-222 buffered with 0.50 g sodium bicarbonate dissolved in

1.0L water) (from 2010), c. 1.5 mm long incisions were made into the abdominal

cavity on each side and the testes were excised with fine forceps. Sham-operated

fish were treated similarly, but the testes were not removed. The incisions were

closed with BV-2 (0.4 Ph. Eur) suture. In some categories of fish were marked

with spine-clipping. After the operation, the fish were put into aquaria and kept

under different photoperiods.

Sampling

All fish were euthanized with 0.1% 2-phenoxyethanol (up to 2009), or with

0.025% buffered MS-222 solution (from 2010) before dissection. The body and

kidney were weighed, and the breeding color and the kidney maturity status

were recorded. Kidney-somatic index (KSI, kidney weight / (body weight-

parasite weight)) or gonadal-somatic index (GSI, ovary weight/(body weight-

parasite weight)) and parasitation index (PI, total parasite weight/(body weight-

parasite weight)).

The brain and pituitary were excised separately after decapitation and

removal of the skullcap. Whole brains with 2-3 mm of myelencephalon were

used, including the olfactory lobes, which is one of the areas where

GnRH-immunoreactive perikarya are distributed in the stickleback (Andersson et

al. 1995). The pituitary was separated from the brain after decapitation and

removal of the skullcap, and both of them were then immersed in RNAlater®

(Ambion) (50μl for pituitary; 350μl for brain) respectively kept in 4 ⁰C for 8 hour,

before storage at -70 ⁰C until analysis.

15

The blood was collected from the caudal peduncle with micro haematocrit

tubes (Na-hep. Cat No. 7493 11; BRAND). After centrifugation for 2 min, the

plasma samples were removed to pre-weighted Eppendorf tubes. The plasma

volume was measured by weighing and plasma was stored at -70 ⁰C.

Experimental protocols

Paper I

Three experiments were done in paper I (Table 4). Moreover, there is one

additional post-breeding castration experiment only presented in the thesis

summary. Male sticklebacks were kept under long (L16:D8) or short (L8:D16)

photoperiod for 4 weeks and dissected in Feb 2009. In Nov 2009 – Jan 2010,

male fish were kept under long photoperiod to reach breeding state (with bright

breeding colour and nest), or an even longer time to post-breeding state (no

more breeding color). Gonadectomy experiments were carried out in 2010 Mar

(breeding fish) and 2011 Oct (post-breeding fish), which were dissected two

weeks after operation (breeding fish) or after 10 weeks when those males lost all

breeding colour (post-breeding fish). gnrh2 (chicken II GnRH) and gnrh3 (salmon

GnRH)) mRNA levels were measured.

Table 4. Protocols of paper I.

Group (n) LD

Long photoperiod (20) 16:8 Short photoperiod (20) 8:16 Breeding (18) 16:8 Post-breeding (17) 16:8 Castrated breeding (10) 16:8 Sham-operated breeding (10) 16:8 <The experiment below was not included in paper I> Castrated post-breeding (7) 24:0 Sham-operated post-breeding (8) 24:0

16

Paper II

Male and female three-spined sticklebacks were collected by drop nets at

the Askö Laboratory in northwestern Baltic Proper (58°49′N, 17°38′E). The

samplings were done on April 28 2010, a few weeks before the onset of the

breeding season. The fish were anaesthesized and dissected after capture. The

sample sizes of each groups are shown in Table 5. The gnrh2 and gnrh3 gene

expression levels in the brain and lh-β and fsh-β levels in the pituitary were

measured.

Table 5. Fish in paper II.

PI (%) KSI (%) GSI (%)

Males With parasite (n=8) 22.02 ± 3.59 1.16 ± 0.12 – Without parasite (n=6) – 1.88 ± 0.23 – Female With parasite (n=10) 24.38 ± 1.37 – 3.85 ± 0.42 Without parasite (n=9) – – 6.43 ± 0.56 Parasite index, kidney somatic index and Gonadosomatic index calculated as organ or total parasite weight x (body weight without parasite)-1 x 100 *p<0.05

Paper III

Four sets of experiments were done during 2007 to 2011. In general, males

were either castrated or sham-operated and treated with different implants.

After operations fish were transferred to different aquaria under different

photoperiods (Table 6). The fish were dissected approx. 1 month after operation.

The mRNA levels of lh-β and fsh-β in the pituitary were then measured. In the

experiment 2011, the plasma samples of each group were collected for T levels

measurement.

*

*

17

Table 6. Protocols of paper III. 2007 (35) 12:12 Control# Control 2009 (16) 8:16 Sham Control (10) Castration Control -10 (10) Castration Control (20) Castration T 0.25 (10) Castration T 0.25 (35) 13:11 Control# Control (10) Castration T cryst (10) Castration Control (10) Castration AI (20) Castration T 0.25 (10) Castration T 0.25 (35) 14:10 Control# Control + AI (10) Castration Control (10) Castration T cryst (20) Castration T 0.25 + AI 2008 (10) 16:8 Sham Control 2011 (16) 16:8 Sham Control (10) Castration Control (16) Castration Control (10) Castration T 0.25 (15) Castration T 0.10 (10) Castration T cryst (15) Castration T 0.25 (10) Castration 11KA (15) Castration T 0.50 (10) 8:16 Sham Control (16) 8:16 Sham Control (10) Castration Control (16) Castration Control (10) Castration T 0.25 (15) Castration T 0.10 (10) Castration T cryst (15) Castration T 0.25 (10) Castration 11KA (15) Castration T 0.50 Sample sizes (n), Sham-operated (Sham), Implants: Control (empty or cacaobutter alone), T cryst (Crystalline T), T 0.10/ 0.25/ 0.50 (0.10%/0.25%/0.50% T in cacaobutter), 11KA (Crystalline 11KA). #Controls includes both unoperated and sham-operated fish.

Paper IV

Breeding fish were either completely castrated, hemi-castrated, or

sham-operated and treated with different types of implants. Several sets of

experiments, each using 10-25 individuals were carried out in 2008, and 2009

spring (11KA and AI treatments) and in 2011 (T and AI treatments). In total, 206

fish were used. The numbers in each treatment are shown in Table 7. Plasma

samples were collected, and the levels of 11KT and T were measured via

Radioimmunoassay (RIA).

18

Table 7. Protocols of paper IV.

2008-2009 2011

Operation Implant Sample N(N)* Operation Implant Sample N(N)*

Sham Empty 29 (29) Sham Empty 13 (13) 11KA(L) 12 (12) T(1%) 10 (10) 11KA(H) 12 (12) T(2%) 11 (11) AI 12 (10) AI 10 (9) Hemi Empty 16 (16) Castr Empty 10 (0) 11KA(L) 4 (4) T(1%) 10 (0) 11KA(H) 13 (13) T(2%) 8 (6) Castr Empty 15 (0) 11KA(L) 6 (0) 11KA(H) 5 (3)

Sham operated (Sham), hemi-castrated (Hemi) or completely castrated (Castr). Empty implants or implants with 1% or 2% of testosterone (T) in cacaobutter; 11KA(H)/(L): high/low dosages of 11 ketoandrostenedione; AI: the aromatase inhibitor flutamide. *Sample number, N, and number of males with breeding colours, (N).

Cloning and bioinformatic analysis

In paper I, peptide sequences from other species (teleosts were given a highest

priority) were used to BLAST the genome databases (Ensembl Genome Browser

system) for stickleback GnRH homologs. In silico predicted full-length of stickleback

GnRH homologs obtained from the genome were carefully confirmed by the

GenBank database. Specific primers (Table. 8) were designed for the

reverse-transcriptase polymerase chain reaction (RT-PCR) analysis. PCR products thus

obtained were subcloned into a pGEM-T Easy vector (Promega, Madison, WI, USA),

and the nucleotide sequences were determined with an ABI 377 sequencer (Applied

Biosystems, Warrington, UK). Sequence analysis was conducted with a BLASTx

program (NCBI). To verify the membership of identified candidates in the core GnRH

family, the deduced amino-acid sequences of cloned stickleback GnRHs were aligned

with ClustalX together with all known GnRH protein sequences of teleost available in

public databases.

Table 8. RT-PCR primer for cloning. Gene Primer sequence Product size(bp) gnrh2 F 5’-ATGTGTGTATCTCGGCTGGTTTTG-3’ 258 R 5’-TCACTTTCTTTTCTGAAGCTCTCT-3’ gnrh3 F 5’-ATGGAAGCGAGCAGCAGAGCGATG-3’ 273 R 5’-AAGAGAAAGTTCCAAAATAATTGA-3’

Physical gene maps of verified GnRH loci were scaled based on assemblies of the

Ensembl Genome Browser. Genes located up- and downstream of GnRH genes in

19

these loci were blasted against mammalian genomes to determine the highest score.

In addition, up- and downstream genes located in the conserved syntenies of the

mammalian GnRH were also examined in stickleback genome databases to

determine the precision of in silico cloning.

Real time quantitative PCR (q-PCR)

GnRHs

Two types of GnRHs; chicken GnRH II (cGnRH II, a GnRH 2) and salmon

GnRH (sGnRH, a GnRH 3) had been found in the stickleback genome when GnRH

sequences from other species were used to BLAST the genome databases

(Ensembl Genome Browser system) for Stickleback GnRH homologs. No other

previously known vertebrate GnRH was found and though the neighbor genes of

GnRH 1 were found on chromosome 12, but no GnRH was present in the

appropriate area, as also not found by Tostivint (2011). By means of

reverse-transcriptase polymerase chain reaction (RT-PCR), the complete cDNA

sequences of about 255~270 bp of the stickleback GnRH coding regions were

confirmed, and used to design the primers for ribosomal protein (RPL8), GnRH 2

and GnRH 3 (Table 3) specifically for three-spined stickleback.

Total RNA was extracted by TriZol reagent (Invitrogen) and

reverse-transcribed from the brains of sticklebacks according to the methods in a

previous study (Wu et al. 2009). The RNA samples were cleaned up with DNase

(2U) using TURBO DNA-free kit according to the manufacturer’s instruction

(Ambion) to remove possible genomic DNA contamination and re-concentrated

by RNeasy MinElute Cleanup Kit® (Qiagen), before being reverse-transcribed.

The enzyme, SuperScript® III (Invitrogen), was used in the mRNA - cDNA reverse

transcription in this and the following experiment. The mRNA expressions of

target genes were measured by Q-PCR with the Roche LightCycler® 480 System

(Roche Applied Science, Mannheim, Germany). PCRs contained 40 ng of cDNA,

50 nM of each primer, and the LightCycler® 480 SYBR Green I Master (Roche) in a

final volume of 10 μl. All Q-PCR reactions were performed as follows: 1 cycle of

50 ⁰C for 2 min and 95 ⁰C for 10 min, followed by 45 cycles of 95 ⁰C for 15 sec

and 60 ⁰C for 1 min using 0.05 μM of specific primers for stickleback GnRHs

20

(Table 3) PCR products were subjected to a melting-curve analysis, and

representative samples were electrophoresed to verify that only a single product

was present. Control reactions were conducted with RNA free water to

determine the background levels. The standard curve for each gene was

confirmed to be in a linear range with the reference gene (RPL8) while series

dilutions of cDNA as the template.

GtHs

Total RNA was extracted from stickleback’s pituitaries through

homogenization using TriZol reagent according to the methods in a previous

study (Hellqvist et al. 2008). The extracted RNA was treated with DNase (2U)

using TURBO DNA-free kit according to the manufacturer’s instruction (Ambion)

to remove possible genomic DNA contamination. SuperScript® III (Invitrogen),

was used in the mRNA - cDNA reverse transcription. After reverse-transcription,

the samples were diluted 1/10 for lh-β and fsh-β, or 1/100 for 18S

measurements. The relative quantity of lh-β, fsh-β and 18S mRNA expression

was measured by Q-PCR with the Stratagene MX3000P™ using Stratagene’s

MxPro™ Q-PCR Software with brilliant II Fast SYBR® Green Q-PCR Master Max

according to the manufacturer’s instructions (Stratgene, Agilent Technologies).

Specific primers for lh-β, fsh-βand 18s rRNA (0.05 μM) (Table 3) were used. PCRs

contained 40 ng (FSH-β, LH-β) or 4 ng (18s) of cDNA , 50 nM of each primer, and

same volume of Fast SYBR Green in a final volume of 10 μl.The parameters for

the real-time PCR were set at 95 ⁰C for 2 min followed by 40 cycles (at 95 ⁰C for 5

sec and 60 ⁰C for 20 sec). One last cycle was run to produce a melting curve to

show the amplified products specificity. lh-β and fsh-β mRNA expression data

were normalized against 18s as a reference gene.

21

Table 3. Sequences of the primers used in qRT-PCR. The RPL8 and 18S were used as reference genes in the measurement of cGnRH II / sGnRH and lh-β / fsh-βrespectively.

Gene Forward primer Reverse primer

cgnrh II 5’-CCGTCGGAGATTTCAGAGGAGATT-3’ 5’-TCTGAAGCTCTCTGGCTAAGGCAT-3’ (ENSGACG00000011943) sgnrh 5’-AGCGATGGTGCAGGTGTTGATGTT-3’ 5’-TTAAGTCTCTCTTGGGTCTGGGCA-3’ (ENSGACG00000012674) rpl8 5’-CGACCCGTACCGCTTCAAGAA-3’ 5’-GGACATTGCCAATGTTCAGCTGA-3’

(ENSGACG00000002035)

fsh-β 5’-CATCGAGGTGGAGGTCTGTG-3’ 5’-GGGGCTGATGGCTGCTGT-3’ (EMBL Accession No.AJ534871) lh-β 5’-GGTCACTGCCTCACCAAGGA-3’ 5’-GGAGCGCGATCGTCTTGTA-3’ (EMBL Accession No.AJ534969) 18S 5'-CTCAACACGGGAAACCTCAC-3' 5'-AGACAAATCGCTCCACCAAC-3'

(NCBI Accession No.BT026885.1)

In situ hybridization

Fragments of GnRH homologues were obtained by PCR and subcloned into the

pGEM-T easy vector. After the PCR with T7 and SP6 primers, digoxigenin

(DIG)-labeled (Perkin-Elmer, Boston, MA, USA) RNA antisense probes were

synthesized by in vitro transcription with proper (T7 or SP6) RNA polymerase (Roche,

Penzberg, Germany). RNA probes were examined by checking on commercial RNA

gels (Amresco, Solon, Ohio, USA).

Two sets of breeding males were used in Oct 2011 (n=2) and Jul 2012 (n=4). The

stickleback brains with pituitary were fixed with 4% paraformaldehyde (PFA) at 4 °C

for 3 h for cryosections (2011) or overnight for paraffin sectioning (2012). For

cryosection, the diencephalon (including pituitary) were separated from the brain

and immersed in phosphate buffer solution (PBS) containing 30% sucrose overnight,

and embedded with compound embedding medium (OCT) (Sakura, Tokyo, Japan) at

-20 °C. Transverse cryosections (about 10 μm) were made crossing the pituitary and

hypothalamus with a cryostat (CM 1900, Leica, Heidelberg, Germany), and were

attached to amino silane (APS) adhesive slides (Matsunami, Osaka, Japan). The

paraffin sections were made from the whole brain which was dehydrated following

the standard protocol, where DEPC treated water was used to wash the samples

before dehydration and to dilute alcohol in dehydration. Serial sagittal sections (5μm)

were made between c. 1.5mm right and left from the center line, which covered the

most of hypothalamus and telencephalon. The paraffin sections were attached to

APS adhesive slides. Prepared slides were washed with phosphate buffer solution

tween-20 (PBST) and then hybridized with probe at 65 °C for overnight incubation.

The concentration of DIG-labeled RNA probe examined by dot blot assay (control 100

22

ng/µL RNA as a reference) was about 750 ng/ml. On the next day, slides were

washed in a hybridization buffer series in 2x SSC at 65 °C. After another serial

washing with 0.2x SSC and PBST, samples were incubated with 5% sheep serum in 2

mg/mL bovine serum albumin (BSA) blocking solution. After the blocking step, slides

were incubated with an alkaline phosphatase (AP)-conjugated anti-DIG antibody

(Roche, 1:5,000 dilution in blocking solution) at 4 °C and then incubated in staining

buffer (0.1 M Tris , pH 9.5, 50 mM MgCl2, 0.1 M NaCl, and 0.1% Tween-20). Staining

was conducted with a mixture of nitrotetrazolium blue chloride (100 mg/mL) and

5-bromo-4-chloro-3-indolyl phosphate (BCIP) (50 mg/mL) in buffer. The reaction was

stopped with PBST and methanol. Images were obtained using a microscope

(Axioplan 2 Imaging, Carl Zeiss, Oberkochen, Germany).

Radioimmunoassay (RIA)

Each plasma sample was diluted with 100 μl RIA buffer (Schulz 1985) and

vortexed for 20 sec. Following heat treatment at 80 ⁰C for 60 min, the tubes were

centrifuged at 13000 rpm for 20 min in 4 ⁰C. The supernatants were pipetted into a

new set of tubes, which were stored in -70 ⁰C until measurement.

Both T and 11KT levels of the samples were measured via the

radioimmunoassay (RIA) developed by (Schulz 1985) to quantify levels directly from

heat-treated samples following the protocol in Schulz et al. (1994).

Statistics

The data were analysed with two tailed t-test for independent samples after

normality test (p<0.05) for comparisons between two groups and with one way

ANOVA for comparisons between multiple groups. The interaction effects between

treatment dosages and operations were analysed using two-way ANOVA by SPSS

v.14.

RESULTS AND DISCUSSION

GnRH gene expression in stickleback (paper I, II)

23

Both GnRH2 (cGnRH II) and GnRH3 (sGnRH) cDNA were identified in sticklebacks,

whereas there was no sign of any possible GnRH1 cDNA sequence. Via in situ

hybridization, the cells expressing GnRH3 were found in the hypothalamus and

telecephalon, but barely in other brain areas (Fig 7). The GnRH2 expressing neurons

were mainly located in the vental mesocephalon and thalamus, but they were found

in the hypothalamus as well (Fig 7). The distribution pattern of GnRHs neurons in the

stickleback brain was largely consistent with that of masu salmon (Oncorhynchus

masou) (review in Amano et al. 1997), whereas the GnRH2 expressing neurons were

not present in the hypothalamus of masu salmon. In Atlantic croaker (Micropogonias

undulatus), the hypothalamic GnRH-expressing neurons (GnRH1, sea bream GnRH)

and its mRNA level have also been detected in pituitary (Mohamed et al. 2005).

GnRH2 and 3 expressing cells were found in the pituitary of stickleback as well.

Figure 7. GnRHs expression in the brain of stickleback labeled via mRNA in situ hybridization. A. GnRH3 expressing cells in the whole brain sagittal section. B. GnRH3 expressing cells in the hypothalamus and pituitary, transverse section. C. GnRH2 expressing cells in the whole brain, sagittal section. D. GnRH2 expressing cells in the hypothalamus and pituitary, transverse section. Arrows indicated the expressing cells. T, telecephalon; OT, optic tectum-thalamus; C, cerebellum; Th, thalamus; HT, hypothalamus; D (-r/ -l), diencephalon (right/ left); POA, preoptic area; PIT, pituitary; MT, midbrain tegmentum; VT, ventral telencephalon. Bars = 600μm (A), (C); 300 μm (B), (D).

Dramatically higher expression levels of both gnrh2 and gnrh3 in breeding, than

24

in post-breeding sticklebacks, suggest that both gnrh2 and gnrh3 play roles in

stickleback reproduction. However, photoperiodic effects on GnRHs mRNA

expressions were less clear. There was a significantly higher expression level of

gnrh2 found under long day, but no difference was noticed in gnrh3 which appears

to be the dominant form in three-spined stickleback. Furthermore, in another study

the mRNA levels of gnrh3 in the stickleback has been found to be significantly

decreased, rather than increased, when the fish had been transferred to long day

photoperiod for 5 day (O’Brien et al 2012). The effects of light signal may influence

the BPG axis via other mechanism(s). Furthermore, the expression levels of gnrh2

and gnrh3 were regulated by gonadal feedbacks. Castration had inhibitory effect on

both gnrh2 and gnrh3 gene expressions in breeding males, whereas castration

casued stimulatory effect on gnrh3 expression in post-breeding males (Fig 8, this

result was not included in paper I).The polarity of the feedback effect on gnrh3 gene

expression may turn from positive to be negative when the males left the breeding

state.

Figure 8. GnRHs mRNA

levels in sham-operated

(n=9) or castrated (n=10)

post-breeding male

sticklebacks. Means±S.E.

shown, GnRH3: p < 0.05;

GnRH2: p=0.324.

The mRNA levels of lh-β, fsh-β or gnrh2, gnrh3 were not suppressed in

stickleback infected with S. solidus, but rather the contrary. The results indicated

that suppression of gonads and secondary sexual characters in stickleback was not

due to an inhibition by parasite on the brain-pituitary level. These results are in

contrast to the effects found in the common roach (Rutilus rutilus), where the lh-β

mRNA levels, and also pituitary LH content, were diminished in Ligula intestinalis

infected fish (Carter et al. 2005). Trubiroha et al. (2009) found a suppressive effect of

Ligula on both lh-β and fsh-β mRNA in both sexes. The mechanism which caused the

25

increase of the expressions of GtHs and GnRHs in infected stickleback is still unclear.

Since the infected males may have lower androgen levels than uninfected males

(MacNab et al. 2011), this stimulatory effects may be due to the feedback on the

host’s BPG axis.

However, how the parasite infection up-regulated GnRHs and GtHs gene

expressions is still unclear. This study suggested that the stimulatory effects on

sexual endocrine system in the brain and pituitary levels may be one of the

mechanisms to compensate the impacts of parasite, which provide sticklebacks the

potential to mature, at least at a minimum level.

Feedback control of GtH gene expression (paper III)

The sham-operated fish kept under long day (LD 16:8) have higher levels of lh-β

mRNA and fsh-β mRNA than those kept under under short day (LD 8:16). The

negative feedback effects were more marked under short days than under long days,

whereas positive feedback could be more obvious under long days (Table 9). Under

LD 8:16, low doses of T suppressed both lh-β and fsh-β mRNA. However, with high

doses of T treatments or under longer photoperiods, the negative effects on lh-β and

fsh-β mRNA often diminished or turned to be positive (Table 9). In mammals and

birds, the photoperiod can influence the strength of feedback effect on GtH

(Kriegsfeld and Bittman 2010, Wilson 1984), whereas it does not alter the polarity of

feedback. In those cases, the inhibitory photoperiod can enhance the negative

feedback in both ewe (Rosa & Bryant 1985) and male hamsters (Turek 1977).

The inhibitory effect of low T on lh-β and fsh-β under LD 8:16 were abolished by

AI, whereas the stimulatory effect of high T was not, suggesting that estrogen may

play a role in suppressing male maturation under short day. This is consistent with

that the feedback effects of the non-aromatisable androgen 11KA are similar to T

cryst both in the present and in previous studies (Hellqvist et al. 2008). However,

those results were in contrast to the extensively studied positive feedback effects in

salmonids, where previous studies have shown that pituitary LH levels in salmonids

can be stimulated by aromatizable androgens and estrogens (Crim et al. 1981). A

26

positive feedback has also been found in European eel pituitary cells in vitro, where

the GtH II (LH) content and GtH II – β (lh-β) mRNA level increased with treatments

both with T and the non-aromatizable androgen dihydrotestosterone (DHT), but not

with estrogen (Huang et al. 1997).

Table 9. Effect of photoperiod, operation and implants on GtH mRNA levels feedbacks

Year Comparison Photoperiod (L:D) lh-β feedback# fsh-β feedback#

2007 Sham vs. Cast. 11:13 0.15 * 0.04 **

12:12 9.32 *** 0.07 **

13:11 53.57 *** 0.32 *

14:10 2.54 ** 0.36 *

Cast. vs. T0.25 11:13 0.27 * 0.15 *

12:12 0.29 * 0.26 *

13:11 0.22 * 0.62 ns

14:10 0.21 * 2.06 ns

2008 Sham vs. Cast. 8:16 / 16:8 0.26 * / 19.51 * 0.05 * / 0.71ns

Cast. vs. T0.25 8:16 / 16:8 0.06 ** / 8.4 ns 0.02 ** / 0.27 *

Cast. vs. Tcryst 8:16 / 16:8 5.12 * / 26.33 ** 0.02 ** / 2.09 ns

Cast. vs. 11KA 8:16 / 16:8 1.06 ns / 20.15 * 0.11 * / 2.11ns

2010 Sham vs. Cast. 8:16 0.33 * 0.12 *

Cast. vs. T0.25 8:16 0.23 * 0.16 **

Cast. vs. Tcryst 8:16 2.08 * 0.18 *

Cast. vs. T0.25 8:16 0.83 ns 0.35 *

+ AI

Cast. vs. Tcryst 8:16 1.86 * 0.23 *

+ AI

Cast. vs. AI 8:16 0.32 * 0.68 *

2011 Sham vs. Cast. 8:16 / 16:8 0.50 * / 5.67 * 0.26 * / 2.88 **

Cast. vs. T0.10 8:16 / 16:8 0.30 * / 0.39 ns 0.08 * / 0.14 *

Cast. vs. T0.25 8:16 / 16:8 0.29 * / 0.34 ns 0.08 * / 0.12 *

Cast. vs. T0.50 8:16 / 16:8 0.18 ** / 2.11 * 0.18 * / 0.63 ns

Castration (Cast.), Sham-operated (Sham) Feedback#: The mRNA levels of sham operated fish are compared to the mRNA levels of castrated control fish in the same photoperiod and experiment. The castrated controls are set to 1 and thus ratio <1 indicates negative feedback and ratios >1 indicate positive feedback.. Implants: Control (empty or cacaobutter alone), T cryst (Crystalline testosterone), T 0.10/ 0.25/ 0.50 (0.10%/0.25%/0.50% T in cacaobutter), 11KA (Crystalline 11-ketoandrostenedione), AI (aromatase inhibitor). ns not significan; * p<0.05; ** p<0.01; *** p<0.001.

27

The suppression of both lh-β and fsh-β by low androgen levels, especially under

short day, may inhibit maturation completely unless a rise of androgens above a

threshold level would allow complete maturation (Fig 8).

Figure 9. The proposed model of the polarity shift in feedback effect. The androgen

threshold for positive feedback decreased as the light hours increasing. Then, the androgens

feedback may turn from negative to be positive under long photoperiod.

Homeostasis of circulating androgens levels (paper IV)

Androgens plasma levels in the hemi-castrated fish were only half of the levels

in sham-operated fish, which showed that there was no compensatory

steroidogenesis from the remaining testis, which is in agreement with Hellqvist et al.

(2002).

28

Lower doses of 11KA and T implants increased plasma androgen levels in hemi-

or completely castrated but not in sham-operated fish. Both low and high doses of

11KA increased plasma 11KT levels in hemi-castrated fish, whereas only the high

dose of 11KA did so in sham-operated fish. Kurtz et al. (2007) implanted intact

breeding stickleback males with high and low doses of 11KA. In contrast to our

results, plasma 11KT rose significantly after the treatment of both dosages. The

difference may be due to that they used larger capsules and smaller fish, thus their

low dose was probably in effect higher than the one we used.

The circulating GtH levels may increase after castration in breeding fish,

indicating a negative feedback (Schulz et al 2012). However, not only negative but

also positive feedbacks on the BPG axis are known from fishes, including the

stickleback (Hellqvist et al. 2008). Steroidogenesis in testis of male stickleback can be

suppressed by treating the fish with a high dose of methyltestosterone via the water

(Borg et al. 1985). In the present study, it is doubtful if androgen-treatments led to a

dramatic decrease in steroid production, since 11KT levels were not changed by T

treatments and T levels not by 11KA treatments. However, the homeostatic control

may be exerted not only by androgen-treated fish producing less androgens in the

testes but perhaps also by higher clearance of circulating androgens.

In case aromatization plays a role in homeostatic mechanisms, androgen levels

would be expected to rise in fish treated with inhibitors. However, this was not the

case. Reduced plasma androgen levels in hemi-castrated fish were not compensated

in fully mature male stickleback, which suggested that the remaining testis’

steroidogenic activity had been fully activated already. However, both 11KA and T

treatments increased plasma levels much less in sham-operated fish than in

castrated ones, indicating that homeostatic mechanisms dampen increases of both

11KT and T plasma level.

Summary

Both GnRH2 (cGnRH II) and GnRH3 (sGnRH) neurons were present in the

stickleback’s hypothalamus, and may be involved in the seasonal reproduction.

29

There was no evidence for a role of gnrh3 in the photoperiodic control of

reproduction. However, the polarity of the feedback effect on gnrh3 gene

expression may turn from positive to be negative when the males left the

breeding state.

S. solidus infection can up-regulate both GnRHs and GtHs gene expressions,

which may be due to feedback effects on the BPG axis.

The polarity of feedback on lh-β and fsh-β was regulated by both photoperiod

and plasma sex steroids levels.

The suppression of both lh-β and fsh-β by low androgen levels, especially under

short day, may inhibit maturation completely unless a rise of androgens above a

threshold levels would allow complete maturation. This may cause all or

nothing maturation.

The androgen inhibitory effect on lh-β and fsh-β under short day could be

abolished by aromatase inhibitor, which may explain how estrogen suppresses

male maturation under short day.

The reduction of plasma androgen levels in fully mature hemi-castrated fish may

suggest that the remaining testis is not able to increase its steroidogenesis

further. However, both 11KA and T treatments increased plasma levels much

less in sham-operated fish than in castrated ones, indicating that homeostatic

mechanisms are nevertheless present.

30

ACKNOWLEDGEMENTS

I would like to thank my supervisor Bertil Borg for his excellent scientific guidance,

and my co-supervisor, Dick Nässel, my committees, Mention Sören Nylin and Bodil

Elmhagen ,and Heinrich Dircksen for discussions and comments. Your advices are

always valuable. I also thank my coauthors, Monika Schmitz and Susanne Trombley

in Uppsala University, and Pung Pung Hwang and Yung Che Tseng in Academia Sinica

for kind hospitality and guidance, and Rüdiger Schulz and Wytschke van Dijk for the

helps in experiments and writing. Thanks Jaingnan Lou, Oleh Lushchak, Neval Kapan

and Erik Hoffmann, my friends, for always being there whenever I needed your help.

I would also like to thank people at the department, Anette Lorents, Minna

Miettinen, Ulf Norberg, Berit Strand and Siw Gustafsson for making the environment

pleasant to be and work in. The former people at the department, Johannes Strauss,

Agata Jolodziejczyk, Jeannette Söderberg, Lina Enell, Lily Tesfai, Kristin Eriksson,

thank you for your kindness and help. Always! I would like to thank all of the

“customers” of our Golden Stickleback pub. We enjoyed many wonderful Friday

afternoons sharing beers and dreams.

I am thankful to all of my friends in Taiwan and Sweden, for always encouraging and

helping me. I also want to thank my family, especially my wife, Evy, my study could

not have been finished without you.

31

REFERENCES

Aalbers SA (2008) Seasonal, diel, and lunar spawning periodicities and associated

sound production of white seabass (Atractoscion nobilis). Fish Bull 106: 143-151.

Almeida FFL, Taranger GL, Norberg B, Karlsen Ø , Bogerd J, Schulz RW (2009)

Photoperiod-Modulated Testis Maturation in Atlantic Cod (Gadus morhua, L.).

Biol Reprod 80: 631–640

Amano M, Hyodo S, Kitamura S, Ikuta K, Suzuki Y, Urano A, Aida K (1995) Short

photoperiod accelerates preoptic and ventral telencephalic salmon GnRH

synthesis and precocious maturation in underyearling male masu salmon. Gen

Comp Endocrinol 99:22–27.

Amano M, Urano A, Aida K (1997) Distribution and function of

gonadotropin-relesasing hormone (GnRH) in teleost brain. Zool Sci 14:1-11.

Andersson E, Bogerd J, Borg B, Sharp P J, Sherwood NM, Goos HJTh (1995)

Characterization and localization of gonadotropin-releasing hormone in the

brain and pituitary of the three-spined stickleback, Gasterosteus aculeatus.

Cell Tissue Res 279:485-494.

Andersson E, Borg B, Goos HJTh (1992) Temperature, but not photoperiod, influences

gonadotropin-releasing hormone binding in the pituitary of the three-spined

stickleback, Gasterosteus aculeatus. Gen Comp Endocrinol 88:111-116.

Antonopoulou E, Mayer I, Berglund I, Borg B (1995) Effects of aromatase inhibitors on

sexual maturation in Atlantic salmon, Salmo salar, male parr. Fish Physiol

Biochem 14:15-24

Baggerman B (1957) An experimental study on the timing of breeding and migration

in the three-spined stickleback (Gasterosteus aculeatus L.). Arch Neerl Zool

12:105-318.

Baggerman B (1985) The role of biological rhythms in the photoperiodic regulation of

seasonal breeding in the stickleback Gasterosteus aculeatus. Neth J Zool

35:14– 31.

32

Baggerman B (1989) On the relationship between gonadal development and

response time to photostimulation of sticklebacks living under natural

conditions and under constant short-day conditions for long periods of time. Can

J Zool 67:126– 135.

Barber I (2007) Host-parasite interactions of the Three-spined stickleback. In:

Huntingford, F.A., Ö stlund-Nilsson, S., Mayer, I. (Eds.) Biology of the

Three-spined Stickleback. CRC press, pp 271-318.

Begtashi I, Rodriguez L, Moles G, Zanuy S, Carrillo M (2004) Long-term exposure to

continuous light inhibits precocity in juvenile male European sea bass

(Dicentrarchus labrax, L.). I. Morphological aspects. Aquaculture 241: 539–559.

Billard R, Breton B (1978) Rhythms of reproduction in teleost fish. In: Thorpe, J. E.

(Ed.) Rhythmic Activity of Fishes. Academic Press, pp. 31-53.

Billard R, Richard M, Breton B (1977) Stimulation of gonadotropin secretion after

castration in rainbow trout. Gen Comp Endocrinol 33: 163-165.

Bogerd J, Zandbergen T, Andersson E, Goos H (1994) Isolation, characterization and

expression of cDNAs encoding the catfish-type and chicken-II-type

gonadotropin-releasing hormone precursors in the African catfish. Eur J

Biochemi 222: 541–549.

Borg B (1982) Seasonal effects of photoperiod and temperature on spermatogenesis

and male secondary sexual characters in the threespined stickleback,

Gasterosteus aculeatus. Can J Zool 60:3377–3386.

Borg B (2007) Reproductive physiology of sticklebacks, In: S. Ö stlund-Nilsson, Mayer,

I., Huntingford, F.A. (Eds.) Biology of the Three-spined Stickleback. CRC press,

pp. 225-248.

Borg B (2010) Photoperiodism in fishes. In: R. J. Nelson, Denlinger D.L., Somers D.E.

(Eds) Photoperiodism: The Biological Calender. Oxford University Press. pp

371-398.

Bromage NR, Jones J, Randall CF, Thrush M, Davies B, Springate J, Duston J, Barker G

(1992) Broodstock management, fecundity, egg quality and timing of egg

33

production in the rainbow trout (Oncorhynchus mykiss). Aquaculture 100:

141–166.

Bromage N R, Randall CF, Duston J, Thrush M, Jones J (1994) Environmental control of

reproduction in salmonids. In: Muir, J. F. and Roberts, R. J. (Eds.) Recent advances

in Aquaculture IV. Blackwell, pp. 55-65.

Burgus R, Butcher M, Amoss M, Ling N, Monahan M, Rivier J, Fellows R, Blackwell R,

Vale W, Guillemin R (1972) Primary structure of ovine luteinizing

hormone-releasing factor (LRF). Proc Nat Acad Sci USA 69:278–282.

Borg B, Reschke M, Peute J, van den Hurk R (1985) Effects of castration and

androgen-treatment on pituitary and testes of the three-spined stickleback,

Gasterosteus aculeatus L., in the breeding season. Acta. Zool. (Stockh.) 66:47-54.

Fernald RD, White RB (1999) Gonadotropin-releasing hormone genes: Phylogeny,

structure, and functions. Front Neuroendocrin 20:224–240.

Carter V, Pierce R, Dufour S, Arme C, Hoole D (2005) The tapeworm Ligula intestinalis

(Cestoda: Pseudophyllidea) inhibits LH expression and puberty in its teleost

host, Rutilus rutilus. Reproduction 130:939-945.

Crim LW, Evans DM (1979) Stimulation of pituitary gonadotropin by testosterone in

juvenile rainbow trout (Salmo gairdneri). Gen Comp Endocrinol 37: 192-196.

Crim LW, Peter RE, Billard R (1981) Onset of gonadotropic hormone accumulation in

the immature trout pituitary gland in response to estrogen or aromatizable

androgen steroid hormones. Gen Comp Endocrinol 44:374-381.

Davie A, Porter MJR, Bromage NR (2003) Photoperiod manipulation of maturation

and growth of Atlantic cod (Gadus morhua). Fish Physiol Biochem 28:399-401.

Dubinina MN (1980) Tapeworms ( Cestoda, Ligulidae ) of the fauna of the USSR.

(Online version, translated from the Russian text of 1966). New Delhi: Amerind

Publishing Co., 320 pp.

Endal HP, Taranger GL, Stefansson SO, Hansen T (2000) Effects of continuous

additional light on growth and sexual maturity in Atlantic salmon, Salmo salar,

34

reared in sea cages. Aquaculture 191: 337–349.

Goos HJTh (1991) Fish gonadotropins and GnRH’s: fundamental aspects and practical

applications. Bulletin of the Institute of Zoology, Academia Sinica, Monograph.

16: 119-137.

Goos HJTh, Joy KP, De Leeuw R, Van Oordt PGWJ, Van Delft AML, Gielen JTh. (1987)

The effect of luteinizing hormone-releasing hormone analogue (LHRHa) in

combination with different drugs with anti-dopamine and anti-serotonin

properties on gonadotropin release and ovulation in the African catfish,

Clarias gariepinus. Aquaculture 63: 143-156.

Hansen T, Karlsen Ø , Taranger GL, Hemre GI, Holm JC, Kjesbu OS (2001) Growth,

gonadal development and spawning time of Atlantic cod (Gadus morhua)

reared under different photoperiods. Aquaculture 203:51-67.

Huang YS, Schmitz M, Belle NL, Chang CF, Quérat B, Dufour S (1997) Androgens

stimulate gonadotropin-II β-subunit in eel pituitary cells in vitro. Mol Cell

Endocrinol 131: 157-166.

Heins DC, Baker JA (2003) Reduction of egg size in natural populations of threespined

stickleback infected with a cestode macroparasite. J Parasitol. 89: 1-6.

Heins DC, Baker JA, Toups MA, Birden EL (2010) Evolutionary significance of

fecundity reduction in threespine stickleback infected by the

diphyllobothriidean cestode Schistocephalus solidus. Biol J Linn Soc

100:835-846.

Hellqvist A, I Mayer, B Borg (2002) Effect of hemi-castration on plasma steroid levels

in two teleost fishes; the three-spined stickleback, Gasterosteus aculeatus, and

the Atlantic salmon, Salmo salar. Fish Physiol Biochem 26:107-110.

Hellqvist A, Schmitz M, Borg B (2008) Effects of castration and androgen-treatment

on the expression of FSH-β and LH-β in the threespine stickleback,

Gasterosteus aculeatus –feedback differences mediating the photoperiodic

maturation response? Gen Comp Endocrinol 158:678-682.

Hellqvist A, Schmitz M, Mayer, Borg B (2006) Seasonal changes in expression of LH-β

35

and FSH-β in male and female three-spined stickleback, Gasterosteus

aculeatus. Gen Comp Endocrinol 145:263-269.

Hoover EE, Hubbard HE (1937) Modification of the sexual cycle in trout by control of

light. Copeia 4:206-210.

Howell RA, Berlinsky DL, Bradley TM (2003) The effects of photoperiod manipulation

on the reproduction of black sea bass, Centropristis striata. Aquaculture

218:651–669.

Jakobsson S, Borg B, Haux C, Hyllner SJ (1999) An 11-ketotestosterone induced

kidney-secreted protein: the nest building glue from male three-spined

stickeback, Gasterosteus aculeatus. Fish Physiol Biochem 20:79-85.

King JA, Dufour S, Fontaine YA, Millar RP (1990) Chromatographic and immunological

evidence for mammalian GnRH and chicken GnRH II in eel (Anguilla anguilla)

brain and pituitary. Peptides 11:507–514.

King JA, Millar RP (1982) Structure of chicken hypothalamic luteinizing

hormone-releasing hormone II. Isolation and characterization. J Biol Chem

257:10729–10732.

Kriegsfeld LJ, Bittman EL (2010) Photoperiodism in mammals. In: R.J. Nelson,

Denlinger D.L., Somers D.E. (Eds), Photoperiodism: The Biological Calender,

Oxford University Press., pp 462-503.

Kurtz J, Kalbe M, Langefors Å, Mayer I, Milinski M, Hasselquist D (2007) An

experimental test of the immunocompetence handicap hypothesis in a teleost

fish: 11-ketotestosterone suppresses innate immunity in three-spined

sticklebacks. Am Nat 170:509-519.

Lin XM, Peter RE (1996) Expession of salmon gonadotropin-releasing hormone

(GnRH) and chicken GnRH-II precursor messenger ribonucleic acid in the brain

and ovary of goldfish. Gen Comp Endocrinol 101:282–296.

MacNab V, Katsiadaki I, Barber I (2009) Reproductive potential of Schistocephalus

solidus infected male three-spined stickleback Gasterosteus aculeatus from

two U.K. populations. J Fish Biol 75:2095-2107.

36

MacNab V, Scott AP, Katsiadaki I, Barber I (2011) Variation in the reproductive

potential of Schistocephalus infected male sticklebacks is associated with

11-ketotestosterone titre. Horm Behav 60:371-379.

Maugars G, Schmitz M (2008) Gonadotropin and gonadotropin receptor expression

during the onset of sexual maturation in early maturing Atlantic salmon male

parr, Salmo salar. Mol Reprod Dev 75:403-413.

Mayer I, Borg B, Schulz R (1990) Seasonal changes in and effect of

castration/androgen replacement on the plasma levels of five androgens in

the male three-spined stickleback, Gasterosteus aculeatus L. Gen Comp

Endocrinol 79:23-30.

McPhail JD, Peacock SD (1983) Some effects of the cestode (Schistocephalus solidus)

on reproduction in the threespine stickleback (Gasterosteus aculeatus):

evolutionary aspects of a host-parasite interaction. Can J Zool 61:901.

Miranda LA, Montaner AD, Ansaldo M, Affanni JM, Somoza GM (1999)

Characterization of brain gonadotropin-releasing hormone (GnRH) molecular

variants in brain extracts from different perciform fishes from Antarctic waters.

Polar Biol 21:122-127.

Miyamoto K, HasegawaY, Nomura M, Igarashi M, Kangawa K, Matsuo H (1984)

Identification of the second gonadotropin releasing hormone in chicken

hypothalamus: Evidence that gonadotropin secretion is probably controlled by

two distinct gonadotropin-releasing hormones in avian species. Proc Nat Acad

Sci USA 81: 3874–3878.

Mohamed JS, Thomas P, Khan IA (2005) Isolation, cloning, and expression of three

prepro-GnRH mRNAs in Atlantic croaker brain and pituitary. J Comp Neurol

488:384–395.

Montaner AD, Miranda LA, Vizziano D, López A, Okuzawa K, Somoza GM (2001)

Gonadotropin-releasing hormone in two perciform fishes: Micropogonias

furnieri and Pagrus pagrus. Fish Physiol Biochem 24:243-246.

37

Morehead DT, Ritar AJ, Pankhurst NW (2000) Effect of consecutive 9- or 12-month

photothermal cycles and handling on sex steroid levels, oocyte development

and reproductive performance, in female striped trumpeter Latris lineate

(Latrididae). Aquaculture 189: 293–305.

Ngamvongchon S, Lovejoy DA, Fischer WH, Craig AG, Nahorniak CS, Peter RE, Rivier

JE, Sherwood NM (1992) Primary structures of two forms of

gonadotropin-releasing hormone, one distinct and one conserved, from

catfish brain. Mol Cell Neurosci 3: 7–22.

O'Brien CS, Bourdo R, Bradshaw WE, Holzapfel CM, Cresko WA (2012) Conservation

of the photoperiodic neuroendocrine axis among vertebrates: Evidence from

the teleost fish, Gasterosteus aculeatus. Gen Comp Endocrinol 178: 19-27

Okubo K, Nagahama Y (2008) Structural and functional evolution of

gonadotropin-releasing hormone in vertebrates. Acta Physiologica 193: 3–15.

Palevitch O, Kight K, Abraham E, Wray S, Zohar Y, Gothilf Y (2007) Ontogeny of the

GnRH systems in zebrafish brain: in situ hybridization and promoter-reporter

expression analyses in intact animals. Cell Tissue Res 327:313–322.

Pauly D (1998) Tropical fishes: patterns and propensities. J Fish Biol 53:1-17.

Perdikaris C (2001) Photoperiodic control of early maturation. Applicability of the

method to Mediterranean seabass (Dicentrarchus labrax) and seabream (Sparus

aurata) culture. 10th Proc. Panell. Congr. Ichthyol. pp: 161-164.

Peter RE, Trudeau VL, Sloley BD (1991) Brain regulation of reproduction in teleosts.

Bulletin of the Institute of Zoology, Academia Sinica, Monograph 16: 89-118

Petersen CW, Warner RR (2002) The ecological context of reproductive behavior. In:

Sale P. F. (Ed) Coral Reef Fishes. Academic Press, pp. 103-120.

Porter MJR, Duncan NJ, Mitchell D, Bromage NR (1999) The use of cage lighting to

reduce plasma melatonin in Atlantic salmon Salmo salar and its effects on the

inhibition of grilsing. Aquaculture 176:237–244.

Randall CF, Bromage NR, Duston J, Symes J (1998) Photoperiod induced phase-shifts

38

of the endogenous clock controlling reproduction in the rainbow trout: a

circannual phase-response curve. J Reprod Fertil 112:399-405.

Rosa HJD, Bryant MJ (2003) Seasonality of reproduction in sheep. Small Ruminant

Res 48:155-171.

Schulz R (1985) Measurement of five androgens in the blood of immature and

maturing male rainbow trout, Salmo Gairdneri (Richardson). Steroids 46:717-726.

Schultz ET, Topper M, Heins DC (2006) Decreased reproductive investment of female

threespined stickleback Gasterosteus aculeatus infected with cestode

Schistocephalus solidus: parasite adaptation, host adaptation, or side effect?

Oikos 114: 303-310.

Schulz RW, W van Dijk, E Chaves-Pozo, Á García-López, LR de França, J Bogerd (2012)

Sertoli cell proliferation in the adult testis is induced by unilateral

gonadectomy in African catfish. Gen Comp Endocrinol 177:160-167.

Sherwood NM, Eiden L, Brownstein M, Spiess J, Rivier J, Vale W (1983)

Characterization of a teleost gonadotropin-releasing hormone. Proc Nat Acad

Sci USA 80:2794–2798.

Somoza GM, Miranda LA, Strobl-Mazzulla P, Guilgur LG (2002) Gonadotropin-

releasing hormone (GnRH): from fish to mammalian brains. Cell Mol Neurobiol

22:589-609.

Stacey NE (1984) Control of the timing of ovulation by exogenous and endogenous

factors. In: Potts, G.W., Wootton, R.J. (eds) Fish reproduction: strategies and

tactics. Academic Press pp. 207-222.

Steven C, Lehnen N, Kight K, Ijiri S, Klenke U, Harris WA, Zohar Y (2003) Molecular

characterization of the GnRH system in zebrafish (Danio rerio): cloning of

chicken GnRH-II, adult brain expression patterns and pituitary content of

salmon GnRH and chicken GnRH-II. Gen Comp Endocrinol 133:27–37.

Suzuki M, Hyodo S, Kobayashi M, Aida K, Urano A (1992) Characterization and

localization of mRNA encoding the salmon-type gonadotropin-releasing

hormone precursor of the masu salmon. J Mol Endocrinol 9:73–82.

39

Tsutsui K, Ubuka T, Bentley GE, Kriegsfeld LJ (2012) Gonadotropin-inhibitory hormone

(GnIH): Discovery, progress and prospect. Gen Comp Endocrinol 177:305-314.

Taranger GL, Haux C, Stefansson SO, Björnsson BT, Walther B, Hansen T (1998) Abrupt

changes in photoperiod affect age at maturity, timing of ovulation and plasma

testosterone and oestradiol-17β profiles in Atlantic salmon, Salmo salar.

Aquaculture 162: 85-98.

Tostivint H (2011) Evolution of the gonadotropin-releasing hormone (GnRH) gene

family in relation to vertebrate tetraploidizations. Gen Comp Endocrinol

17:575-581.

Trubiroha A, Wuertz S, Frank SN, Sures B, Kloas W (2009) Expression of gonadotropin

subunits in roach (Rutilus rutilus, Cyprinidae) infected with plerocercoids of

the tapeworm Ligula intestinalis (Cestoda). Int J Parasotol 39:1465-1473.

Turek FW (1977) The interaction of photoperiod and testosterone in regulating

serum gonadotropin levels in castrated male hamsters. Endocrinology

101:1210-1215.

Yu KL, Sherwood NM, Peter RE (1988) Differential distribution of two molecular

forms of gonadotropin-releasing hormone in discrete brain areas of goldfish

(Carassius auratus). Peptides 9: 625–630.

Voet D, Voet J (2004) Biochemistry, 3rd ed. Wiley. USA.

Wilson FE (1985) Androgen feedback-dependent and –independent control of

photoinduced LH secretion in male tree sparrows (Spizella arbora). J Endocrinol

105:141-152.

Withler RE, Clarke WC, Blackburn J, Baker I (1998) Effect of triploidy on growth and

survival of pre-smolt and post-smolt coho salmon (Oncorhynchus kisutch).

Aquaculture 168: 413-422.

Wu SC, Horng JL, Liu ST, Hwang PP, Wen ZH, Lin CS, Lin LY (2009)

Ammonium-dependent sodium uptake in mitochondrion-rich cells of medaka

(Oryzias latipes) larvae. Am J Physiol-Cell Ph 298:237-250