| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Graduate School for Developmental Biology, Research Group for Comparative Endocrinology, Utrecht University, 3584 CH Utrecht, The Netherlands
b Fish Culture and Fisheries Group, Department of Animal Sciences, Wageningen University, Wageningen, The Netherlands
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cortisol, luteinizing hormone, puberty, steroid hormones, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our previous studies demonstrated that in common carp, repeated temperature-induced stress retarded the first waves of spermatogenesis and an elevation of cortisol levels following application of the stressor [2]. Long-term cortisol treatment resulted in a similar effect on spermatogenesis, accompanied by a decrease in plasma 11-ketotestosterone (11KT) [3]. Although direct effects of temperature on the testis cannot be excluded (as has been shown in homeotherms), such an effect has never been described in poikilotherms.
Several studies indicate that 11KT has an important function during sexual maturation. It has been shown to stimulate spermatogenesis in African catfish (Clarias gariepinus) [4], the common carp (Cyprinus carpio) [5], and Japanese eel (Anguilla japonica) [6].
The reduction of plasma sex steroids due to stress or cortisol has been demonstrated in a variety of vertebrate species, including mammals [7, 8], reptiles [9, 10], amphibians [11], and fish [12–14]. In mammals, the steroid-producing cells of the testis, the Leydig cells, contain glucocorticoid receptors [15], and therefore, corticosteroids may exert a direct effect on steroidogenesis. Results of in vitro experiments suggest that stress or corticosteroids decrease the Leydig cell sensitivity to gonadotropins [8, 16], either by reducing the LH receptor content [17] or by inhibiting the 17
-hydroxylase and/or C17,20-lyase activity [18]. In fish, data on the direct effect of cortisol on steroidogenesis are less consistent compared to mammals. Carragher and Sumpter [19] and Pankhurst et al. [20] found a reduction of 17ß-estradiol and testosterone secretion by cultured ovarian follicles. In other species, including goldfish (Carassius auratus), common carp, and the sparid Pagrus auratus, however, Pankhurst et al. [21] found no evidence that the inhibitory effects of stress on reproduction are mediated by the action of cortisol on ovarian steroidogenesis directly.
The aim of this study was to investigate if the observed decrease in plasma 11KT levels in maturing male common carp is caused by a direct effect of cortisol on the steroid-producing capacity of the testis or via an alteration of the LH secretion. Furthermore, we were interested if the effects of cortisol are correlated with age and development of the fish.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Isogenic male common carp (designated as strain E4xR3R8) were produced by crossing a homozygous gynogenetic E4 female [22] with an unrelated homozygous androgenetic YY-male R3R8 [23]. Fry were produced and raised in the facilities of the Fish Culture and Fisheries Group (Wageningen University, Wageningen, The Netherlands) and transported at 21 days post hatching (dph) to the fish facilities at the Utrecht University.
During the experiment, the fish were kept at 25°C in a flow-through system, exposed to a 12L:12D photoperiod, and fed pelleted dry food (composition as provided by the manufacturer: 54% protein, 18% fatty acids, 1% cellulose, and 18% ashes; Trouw, Putten, The Netherlands) at a daily ration of 20 g/kg-0.8. Immature fish were allowed to acclimatize until 63 dph, after which the experiment started, or were kept until adolescence.
Experiment 1: Pubertal Fish
Cortisol (Steraloids, Inc., Wilton, NH)-treated food (100 mg/kg food) was prepared as described by Pickering et al. [24]. In total, 120 animals were equally divided over two groups. One group received control food, and the other group received the cortisol-treated food from 63 dph onward as described previously [3]. This treatment causes significantly increased plasma cortisol levels over a 6-h period, with peak levels up to 150 ng/ml plasma [3] (Fig. 1).
|
Testicular development in the maturing animals is highly uniform and predictable, with meiosis of spermatogonia starting at approximately 90 dph [25]. Fish from both groups (n = 20) were sampled at several time-intervals during the pubertal development: at 94 dph (early puberty), 100 dph (late puberty), and 120 dph (first wave of spermatogenesis completed). The fish were anesthetized in Tricaine Methane Sulfonate (Crescent Research Chemicals, Phoenix, AZ). Body weight was determined, and blood was collected by puncturing the caudal vasculature. After blood sampling, fish were immediately decapitated, and testes were removed to determine the gonadosomatic index (GSI = testes weight x 100/[body weight - testis weight]) and in vitro incubation for determination of the steroid-synthesizing capacity.
Experiment 2: Adolescent Fish
Forty-eight adolescent fish (males with a completed first wave of spermatogenesis [23, 25]) were equally divided over two groups and fed, starting at 138 dph, either control food or cortisol-containing food, similar to the maturing fish. At 165, 183, and 197 dph, fish were sampled (n = 8) according to the same procedure as in experiment 1.
Androgen Secretion In Vitro
In vitro determination of the steroid-secretory capacity of testicular tissue of maturing fish in experiment 1 was performed as described by Cavaco et al. [4]. In short, the left and right testes of each male were divided into equal halves. Each half-testis was weighed separately and transferred to a separate well of a 24-well Costar plate (Cambridge, MA) containing 0.5 ml of Hepes-buffered L-15 medium (15 mM HEPES and 100 000 U/L of penicillin/streptomycin, pH 7.4). The testis halves were then cut into fragments of approximately 2 mm3. The culture medium of the four half-testes was taken off and replaced by 0.5 ml of medium, with or without dexamethasone (150 ng/ml medium; Sigma, St. Louis, MO) and containing increasing amounts of LH (0, 10, 30, and 100 ng/ml medium). The nonmetabolizable dexamethasone (cortisol agonist) was used to investigate the direct effect of corticosteroids on steroid production. Carp pituitary extract, calibrated for LHß content by LHß RIA, was used as the LH source. After incubation for 20 h at 25°C, the medium was removed, heated for 1 h at 80°C (to avoid any remaining enzymatic activity that may have leaked from the testis tissue), and centrifuged at 10 000 x g for 30 min at room temperature. The supernatant was stored at -20°C until steroid hormone measurement by RIA.
Testicular tissue from each adolescent fish was separately prepared for in vitro incubation with carp pituitary extract as described by Schulz et al. [26]. For each fish, five wells of a 24-well Costar plate containing 0.5 ml of L-15 medium were filled with 100 mg of testicular tissue. The medium was taken off and replaced by 1 ml of medium containing increasing amounts of LH (carp pituitary extract; 0, 10, 30, 100, and 300 ng/ml medium) in the absence or presence of dexamethasone (150 ng/ml medium; Sigma). After an incubation of 20 h at 25°C, the medium was treated as in experiment 1.
Pituitary Extract and Plasma LH
Luteinizing hormone was quantified in carp pituitary extract and in plasma using a homologous RIA [27, 28]. Purified carp LHß subunit (a gift from Dr. E. Burzawa-Gerard) was used for the preparation of standards and for 125I-labeling. The anti-LHß (internal code no. 6.3) was used as first antibody. Although LHß subunit was measured, this is considered to reflect the LH content of pituitary extract and plasma.
Plasma LH levels were measured in all animals. In common carp, as in many species, presence of an FSH has also been demonstrated [29]. However, an FSH-specific assay is not available.
Steroid RIA
The steroid levels in both plasma (11KT) and medium (11KT and 11-ketoandrostenedione [OA]) were determined by RIA as described previously [30]. In most male teleosts, 11KT is considered to be the most dominant androgen in the plasma [31]. Also, in the male common carp, 11KT has been found to be the major androgen produced by the testes [32, 33]. However, in immature common carp, OA is the main androgen produced by the testes (unpublished results).
Statistics
All results are expressed as mean ± SEM. Plasma levels of 11KT and LH are given as nanograms per milliliter of plasma. In vitro data are given as nanograms of steroid secreted, corrected for total testis weight. All results on the treatment effect of cortisol were processed for statistical analysis by Student t-test (P < 0.05). In vitro data were processed by one-way ANOVA, followed by Fisher least significant difference test (P < 0.05).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In maturing control animals, the GSI increased during the experimental period from 94 to 120 dph. This reflects the testicular growth that normally occurs during pubertal development. In contrast, prolonged treatment with cortisol-containing food resulted in impaired testicular development, as follows from the significantly lower GSI at 100 and 120 dph (Fig. 2A). In adolescent fish, relative testicular growth has slowed down, and only at 197 dph was the GSI significantly different from 165 dph. Consequently, the effect of cortisol treatment in adolescent fish was less pronounced compared to that in pubertal fish, but retardation of testicular growth could still be observed at 197 dph (Fig. 2B).
|
Plasma Hormone Levels
Similar to the GSI, plasma 11KT levels in maturing control fish increased during pubertal development. Plasma 11KT levels of cortisol-treated animals were significantly lower compared to control animals (at 100 dph, the difference is not significant) (Fig. 3A). In adolescent control fish, plasma 11KT levels further increased, and they remained at the same level during the experimental period. In cortisol-treated adolescent fish, the 11KT levels at 165 dph were still behind the control values. However, during the experimental period, the 11KT levels in cortisol-treated fish became equal to the control values (Fig. 3B).
|
Prolonged feeding with cortisol did not result in consistent changes in plasma LH levels in both maturing and adolescent fish. A slight increase was only found at the end of the cortisol treatments (for maturing animals at 120 dph and for adolescents at 183 dph; data not shown).
Androgen Secretion In Vitro
Data on in vitro androgen secretion are expressed as micrograms of steroid corrected for the total testes weight (µg per two testes). Schulz et al. [34] have shown differences in steroid secretion per gram of testis tissue depending on the stage of testicular development during puberty. Because our previous results [3] have shown that cortisol treatment retards the first cycle of spermatogenesis, we expressed our data as total secretion per two testes.
In maturing control fish, homologous pituitary extract stimulated the in vitro steroid secretion dose dependently. Previous in vivo cortisol treatment, from 63 dph onward, reduced significantly the pituitary extract-induced OA secretion at 94 dph (Fig. 4A). Similar results were found for 11KT secretion (Fig. 4B), although at 94 dph, OA is the main product produced by the testes. In vitro treatment with dexamethasone resulted in reduced carp pituitary extract-induced androgen secretion as well (Fig. 4, A and B).
|
At 100 dph, OA is still the main androgen produced by the testes. However, both OA and 11KT production have increased compared to 94 dph (Fig. 4, C and D). Cortisol treatment in vivo resulted in significantly lower pituitary extract-induced OA and 11KT secretion in vitro. The in vitro treatment with dexamethasone reduced the secretion of OA and 11KT, but this reduction was not significant due to the somewhat larger variation.
At 120 dph, when 11KT in control animals is becoming the main steroid produced by the testes, both basal and carp pituitary extract-induced 11KT secretion are significantly reduced after prolonged in vivo cortisol treatment (Fig. 4E). The in vitro dexamethasone treatment had a comparable effect; both basal and pituitary extract-stimulated 11KT secretion were affected. Similar results were observed for the OA secretion (data not shown). The OA:11KT ratio (determined for basal secretion) showed that cortisol treatment affected androgen production quantitatively, and also its pattern. In control animals, a shift toward 11KT secretion occurred at 120 dph, whereas cortisol treatment caused the relatively high production of OA to be maintained at this age (Fig. 4F).
In contrast, in adolescent fish, no effect of either in vivo cortisol treatment or in vitro dexamethasone treatment was found on in vitro basal and carp pituitary extract-induced androgen secretion throughout the experiment (Fig. 5).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
As previously observed, cortisol treatment of maturing fish retarded pubertal testicular development as reflected by the lower GSI and lower plasma 11KT levels. In the present experiments, we did not observe a clear effect of cortisol on plasma LH levels (only at the end of the long-term treatments were they slightly elevated), but plasma 11KT levels remained lower during the entire experimental period compared to those in control fish. These results suggest that the decrease in plasma androgen levels is not caused by an effect of cortisol on LH plasma content. Pankhurst and Van Der Kraak [35] also found evidence that the inhibitory effect of stress on plasma sex steroids is independent of plasma LH levels. In the European eel, cortisol treatment increased plasma LH levels and also pituitary LHß mRNA content [36]. These observations are diametrically opposite to the situation in carp: no consistent effects on LH plasma levels (present study), and a decrease in LHß pituitary mRNA [3].
In contrast, in adolescent fish, we observed no effect of cortisol treatment on the testicular development at 165 and 183 dph. Only at 197 dph did an inhibitory effect of cortisol treatment become apparent. At 165 dph, plasma 11KT levels were still lower in cortisol-treated adolescent fish compared to those in controls, but during the experimental period, they increased to the same values as those in the controls. From these observations, we conclude that fish become less sensitive to cortisol during sexual maturation, but at present, we do not have an explanation for this. However, related phenomena have been observed in other species. Indeed, Pankhurst and Van Der Kraak [35] demonstrated that in female rainbow trout, the effect of cortisol on ovarian steroidogenesis depends on the stage of the reproductive cycle.
In mammals, cortisol may have a direct effect on the Leydig cells, because they possess glucocorticoid receptors [15]. Studies by Charpenet et al. [8] and by Orr and Mann [16] demonstrate that stress decreases the sensitivity of the Leydig cell to gonadotropins. This may be caused by reducing the LH-receptor content [17] or by inhibiting the 17-hydroxylase and/or C17,20-lyase activity [18].
Our results demonstrate that prolonged exposure to cortisol reduced the androgen-secreting capacity of the testis. Both OA and 11KT secretion in vitro are significantly reduced. Additionally, the difference in the OA:11KT ratio shows once more that the testicular development in cortisol-treated animals is retarded, because the ratio still reflects a more immature pattern. Our results are not appropriate to reveal the precise mechanism via which cortisol affects the testicular androgen production. It is, however, unlikely that prolonged exposure to cortisol decreases LH receptor content, because the sensitivity to homologous pituitary extract is unchanged. At 100 and 120 dph, the stimulation factor (data not shown) of pituitary extract is similar for both control and cortisol-treated animals. We therefore hypothesize that cortisol affects the enzyme activity involved in the androgen production. In a successive study, we will investigate this hypothesis, as well as the possibility that cortisol competitively inhibits the conversion of 11ß-hydroxyandrostenedione into OA.
In contrast, in vitro treatment with dexamethasone does appear to affect the sensitivity to homologous pituitary extract. At 94 and 120 dph, testes taken from control animals do not show an increase in androgen production on pituitary extract stimulation in the presence of dexamethasone. In several studies, corticosteroids have been suggested (e.g., [35, 37]) and shown (reviewed by Borski [38]) to mediate their inhibiting effect by interfering with the LH signal transduction. In rat Leydig cells, chronic treatment with corticosterone diminished the production of testosterone as well as the basal and LH-stimulated cyclic AMP production [39]. Based on these results, we hypothesize that the in vitro effect of dexamethasone in our experiments may also be caused by an interference of corticosteroids with the LH signal transduction, thereby blocking the LH response and, thus, the LH-induced secretion of 11KT and OA. However, our experiments were not designed to prove this hypothesis.
In summary, we showed that cortisol has a direct inhibitory effect on the testicular androgen secretion and not via plasma LH levels. Furthermore, our results demonstrate that cortisol sensitivity depends on the maturational status of the animal. The underlying mechanism may involve an inhibitory effect on expression of the steroid-producing enzymes, but substrate inhibition of enzymes that have a function in the conversion of cortisol as well as androgen precursors may also be involved. However, a direct interference with the LH signal transduction cannot be excluded.
|
FOOTNOTES |
|---|
1 Supported by grant 805-33.103P from the Netherlands Organization for Scientific Research (NWO). 
2 Correspondence: H.J.Th. Goos, Graduate School for Developmental Biology, Research Group for Comparative Endocrinology, Utrecht University, Padualaan 8, 3584 CH Utrecht, The Netherlands. FAX: 31 30 253 2837; h.j.th.goos{at}bio.uu.nl
Accepted: August 20, 2001.
Received: April 16, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P P de Waal, D S Wang, W A Nijenhuis, R W Schulz, and J Bogerd Functional characterization and expression analysis of the androgen receptor in zebrafish (Danio rerio) testis Reproduction, August 1, 2008; 136(2): 225 - 234. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Overli, S. Winberg, and T. G. Pottinger Behavioral and Neuroendocrine Correlates of Selection for Stress Responsiveness in Rainbow Trout--a Review Integr. Comp. Biol., June 1, 2005; 45(3): 463 - 474. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
