BIOLOGY OF REPRODUCTION 65, 1807–1812 (2001)
Testosterone Inhibits 11-Ketotestosterone-Induced Spermatogenesis in African
Catfish (Clarias gariepinus)1
J. Eduardo B. Cavaco,3 Jan Bogerd, Henk Goos, and Rüdiger W. Schulz2
Utrecht University, Faculty of Biology, Research Group Comparative Endocrinology, 3584 CH Utrecht, The Netherlands
ABSTRACT
Male fish produce 11-ketotestosterone as a potent androgen
in addition to testosterone. Previous experiments with juvenile
African catfish (Clarias gariepinus) showed that 11-ketotestosterone, but not testosterone, stimulated spermatogenesis,
whereas testosterone, but not 11-ketotestosterone, accelerated
pituitary gonadotroph development. Here, we investigated the
effects of combined treatment with these two types of androgens on pituitary gonadotroph and testis development. Immature fish were implanted for 2 wk with silastic pellets containing
11-ketotestosterone, testosterone, 5a-dihydrotestosterone, or
estradiol-17b; cotreatment groups received 11-ketotestosterone
in combination with one of the other steroids. Testicular weight
and pituitary LH content were higher (two- and fivefold, respectively) in the end control than in the start control group,
reflecting the beginning of normal pubertal development. Treatment with testosterone or estradiol-17b further increased the
pituitary LH content four- to sixfold above the end control levels. This stimulatory effect on the pituitary LH content was not
modulated by cotreatment with 11-ketotestosterone. However,
the stimulatory effect of 11-ketotestosterone on testis growth
and spermatogenesis was abolished by cotreatment with testosterone, but not by cotreatment with estradiol-17b or 5a-dihydrotestosterone. Also, normal pubertal testis development was
inhibited by prolonged (4 wk) treatment with testosterone. The
inhibitory effect of testosterone may involve feedback effects on
pituitary FSH and/or on FSH receptors in the testis. It appears
that the balanced production of two types of androgens, and the
control of their biological activities, are critical to the regulation
of pubertal development in male African catfish.
luteinizing hormone, pituitary, puberty, spermatogenesis, steroid
hormones, testis
INTRODUCTION
Fish constitute a group of interesting vertebrates for
studying the role of androgens in the regulation of male
reproduction. In fish, 11-oxygenated androgens, such as 11ketotestosterone (KT), are important androgens in addition
to testosterone (T) [1], and two types of androgen receptors,
which differ in their ligand-binding characteristics and tranJ.E.B.C. was supported by Project ‘‘PRAXIS XXI’’ J.N.I.C.T.-Portugal, grant
BD/2603/93.
Correspondence: R.W. Schulz, Utrecht University, Faculty of Biology, Research Group Comparative Endocrinology, Padualaan 8, 3584 CH
Utrecht, The Netherlands. FAX: 31 30 2532837;
e-mail: [email protected]
3
Current Address: Molecular and Comparative Endocrinology, Universidade do Algarve, Faculdade de Ciências do Mar e Ambiente-CCMar,
Campus de Gambelas, P-8000-810 Faro, Portugal.
1
scriptional activities, have been characterized by molecular
[2, 3] and biochemical techniques [4].
Previous work in juvenile male African catfish (Clarias
gariepinus) showed that treatment with T or estradiol-17b
(E 2) elevated the pituitary LH content and LH b-subunit
steady-state mRNA level, whereas 5a-dihydrotestosterone
(DHT) had no such effect [5]. Inhibiting aromatase activity
in primary pituitary cell cultures of adult catfish abolished
the stimulatory effect of T on LH b-subunit expression [6].
This enzyme activity is also present in the pituitary of immature catfish [5], so T-induced LH expression may be mediated via an estrogen-dependant process, which is a notion
supported by data in other species [7]. Hence, the increase
in pituitary LH levels during puberty in catfish [8] may
reflect increased production by the testes of T, which is
converted to E 2 in the pituitary.
Stimulation of spermatogenesis, however, appears to be
a specific effect of KT in catfish, because neither T, DHT,
nor E 2 has such an effect [9]. A direct stimulatory effect
of KT on spermatogenesis has been described in Japanese
eel (Anguilla japonica) testicular explants [10]. In the present study, male catfish at the beginning of spermatogenesis
were subjected to a cotreatment with KT and T in an attempt to induce precocious stimulation of both pituitary and
testis development.
MATERIALS AND METHODS
Animals and Hormones
African catfish were bred and raised as described previously [11], except that catfish pituitary extract instead of hCG was used to induce ovulation. The fish were kept in a copper-free recirculation system at a water
temperature of 258C, exposed to a photoperiod of 14L:10D, and fed with
Trouvit pellets (Trouw, Putten, The Netherlands). Fish in experiments A,
B, and C were 10 wk of age (i.e., only spermatogonia present in the testis),
with an average body weight of 12.5 6 1.4 g at the beginning of the
experiment. In experiment D, 15-wk-old fish were used, with an average
body weight of 26.4 6 4.3 g. All procedures involving experimental animals were conducted in accordance with Dutch national regulations; the
Life Sciences Faculties Animal Care and Use Committee approved the
protocols.
All steroids and salmon GnRH analogue ([D-Arg6, Trp7, Leu8, Pro9NEt]GnRH, sGnRHa) were purchased from Sigma Chemical Company
(St. Louis, MO). The catfish LH preparation used for the stimulation of
testicular androgen secretion in vitro has been described previously [12].
2
Received: 13 April 2001.
First decision: 21 May 2001.
Accepted: 31 July 2001.
Q 2001 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363. http://www.biolreprod.org
Experimental Protocol
African catfish were implanted with silastic pellets containing different
steroids at a dose of 30 mg/g body weight as described previously [9]. An
end control group received steroid-free pellets. A start control group was
sampled at the day of pellet implantation.
In experiment A, three groups of 40 fish received pellets containing
KT, T, or androstenedione (A); two other groups received combined treatments (KT 1 T or KT 1 A). The sex of juvenile catfish cannot be determined externally, and at the end of experiments, the treatment groups were
found to contain 15–25 males. Two weeks after implantation, blood plasma was collected and stored at 2208C until quantification of LH and
androgens by RIA. The pituitaries were removed and used to quantify LH
1807
1808
CAVACO ET AL.
TABLE 1. Plasma levels of 11-ketotestosterone (KT) and testosterone (T)
in steroid-implanted (30 mg/g body weight), immature, male African catfish 2 wk after implantation.a
Treatmentb
Experiment A
C
KT
T
A
KT 1 T
KT 1 A
Experiment D
T2T
KT (ng/ml)c
T (ng/ml)d
6
6
6
6
6
6
n.d.
n.d
5.3 6 0.8e
4.6 6 0.2e
3.4 6 0.3e
4.9 6 0.4e
0.9
8.3
1.0
0.9
7.4
9.9
0.2e
0.6f
0.1e
0.1e
0.7f
0.7f
n.a.
13.9 6 1.2f
Values are the mean 6 SEM.
A, Androstenedione; C, control; T 2 T, second implant of T after 2 wk.
c n.a., Not analyzed.
d n.d., Not detectable.
e–f Means sharing the same letter in the same column do not differ significantly.
a
b
content or basal and GnRH-stimulated LH release in a tissue culture system. The presence of secondary sexual characteristics (i.e., seminal vesicles and urogenital papilla, as described previously [9]) was recorded and
expressed as the percentage of animals in which the structures were developed. Testis and body weight were determined to calculate the gonadosomatic index (GSI; testis weight 3 100/body weight). The stage of
spermatogenesis was determined histologically.
In experiment B, groups of 10–15 males were implanted with KT,
DHT, E 2, KT 1 DHT, or KT 1 E 2, respectively. In this experiment, only
the GSI and the pituitary LH content were analyzed 2 wk after implantation of steroid-containing pellets.
In experiments C and D, groups of 11 or 14–18 males received T
implants at 10 (experiment C) or 15 (experiment D) wk of age, respectively. Two weeks later, another T pellet was implanted, and after another
2 wk, blood and pituitary samples were collected and used for LH quantification. Testis samples were collected for histological analysis of spermatogenesis.
Pituitary Tissue Culture
Determination of basal and sGnRHa-stimulated LH release was carried
out essentially as described previously [13], except that only the maximally effective dose (1 nM sGnRHa) was used. The sGnRHa is a highaffinity ligand for the catfish GnRH receptor, and it effectively stimulates
LH secretion in vitro [14]. Briefly, the pituitaries were preincubated for
18 h in 0.5 ml of Earle balanced salt solution (Life Technologies-Gibco,
Grand Island, NY). After one rinse, 0.5 ml of fresh medium was added,
and the incubation was continued for 3 h. The medium was collected for
the determination of basal LH secretion, and pituitaries were rinsed once
with fresh medium. Finally, 0.5 ml of medium containing 1 nM sGnRHa
was added for another 3 h of incubation to determine the sGnRHa-stimulated LH release. After incubation, the medium was collected and stored
at 2208C until assayed for LH.
Testis Histology
Testes were removed and weighed to calculate the GSI. The tissue was
then processed for light microscopic analysis of the stage of spermatogenesis using 1-mm sections stained with 1% (v/v) methylene blue in 1% (v/
v) borax as described previously [9].
In the African catfish, pubertal spermatogenesis has been subdivided
into four histological stages [9]. Stage I testis contains spermatogonia only;
no meiotic or postmeiotic germ cells are present. Stage II testis shows
spermatogonia and spermatocytes, but no postmeiotic germ cells. Stage III
testis contains spermatogonia, spermatocytes, and spermatids, but no spermatozoa are present. In stage IV testis, all germ cell stages, including
spermatozoa, are present. Testis development in treatment groups is indicated as the percentage of testicular stages (number of animals in a certain
stage 3 100/total number of animals per treatment group).
Stages I–IV do not correspond to stages of the germinal epithelium,
such as those known from mammalian testis. In fish, a group of Sertoli
cells forms an intratubular space, or a cyst, housing germ cells that all
belong to one germ cell clone. In this way, the descendants of a given
stem cell that proceed synchronously through spermatogenesis are taken
care of by ‘‘their’’ group of Sertoli cells. Fish Sertoli cells, therefore,
contact germ cells that are all in the same stage of spermatogenesis. Hence,
recurrent associations along the Sertoli cell surface of different germ cell
clones at different stages of development, which constitute the stages of
the germinal epithelium in mammals (and other amniotes), do not occur
in fish (and other anamniotes).
Radioimmunoassay
The LH concentrations in plasma, pituitary incubation media, and pituitary homogenates were quantified by RIA according to the method described by Goos et al. [15], but with the modifications as described previously [13], using an antiserum raised against the b subunit of LH. The
limit of detection for LH is 0.25 ng/ml, with intra- and interassay variability at median effective concentrations of 6% and 11%, respectively.
The plasma levels of KT and T were determined as described previously
[16]. The limit of detection for the androgen RIAs is 0.3 ng/ml, with
intraassay variabilities of 4% and 9% and interassay variabilities of 5%
and 9% in the KT and T assays, respectively.
Statistics
Hormone levels are expressed as the mean 6 SEM. The LH levels are
given as nanograms per milliliter of plasma or incubation medium or as
nanograms per pituitary. Steroid levels are given as nanograms per milliliter of plasma. For statistical analysis of hormone levels, data were log10
transformed and subjected to analysis of variance followed by the Fisher
least significant difference test. For analysis of the stimulatory effect of
sGnRHa on the release of LH, a one-tailed, paired Student t-test was used.
Secondary sexual characteristics and histological data were analyzed by
the Fisher exact test, followed by the Kruskal-Wallis H-test. In all cases,
differences were considered to be statistically significant at P # 0.05.
RESULTS
Androgen Plasma Levels
In all groups implanted with KT pellets, plasma levels
of KT increased to 7–10 ng/ml (Table 1). Such concentrations are typically found in 10- to 12-mo-old adult males
[17]. The KT plasma levels remained unchanged in T- and
A-treated groups, whereas elevated levels of T were found
in T- and A-treated groups. Fish receiving a second implant
of T after 2 wk showed higher plasma T levels than those
receiving a single implant. Also, the elevated plasma levels
of T found in fish carrying T or A pellets were similar to
those levels recorded in adult males [17].
Testicular Histology, GSI, and Secondary
Sexual Characteristics
An increase in GSI was observed when comparing the
start and end control groups in experiments A and B (Fig.
1, a and c), reflecting the onset of testis development during
the experimental period of 2 wk. Treatment with KT increased the GSI above end control levels and promoted
spermatogenesis, as indicated by the higher proportion of
males with spermatocytes (Fig. 1b). Treatment with T, A,
DHT, or E 2 did not affect the GSI and/or spermatogenesis
compared to the end control group. However, T slightly
promoted seminal vesicle, but not urogenital papilla, development. Unexpectedly, the stimulatory effect of KT on
testis development was abolished by cotreatment with T or
A, which also attenuated the stimulatory effect of KT on
seminal vesicle, but not on urogenital papilla, development
(Fig. 1b). Cotreatment with DHT or E 2, on the other hand,
failed to inhibit the stimulatory effect of KT on testis
growth (Fig. 1c).
In experiment C (4 wk of T treatment beginning at 10
wk of age), the end control group presented immature (n
5 4) and maturing (n 5 5) subgroups, reflecting individual
differences in the timing of maturation. The data obtained
ANDROGENS AND PUBERTY IN MALE AFRICAN CATFISH
1809
FIG. 1. Gonadosomatic index (GSI 5 gonad weight 3 100/body weight) (a, c, d,
and f), development of secondary sexual
characteristics (b), and stage of spermatogenesis (b, e, and g) 2 wk after implantation of different steroids either alone or in
combination with KT. Columns sharing the
same letter are not significantly different (P
. 0.05; ANOVA followed by Fisher least
significant difference test for a, c, d, and f;
Fisher exact test followed by Kruskal-Wallis H-test for b, e, and g). The percentages
given in b (groups sharing the same symbol are not significantly different) indicate
the fraction of animals in experiment A
showing development of seminal vesicles
(S.V.) or urogenital papilla (U.P.). A, Androstenedione; DHT, 5a-dihydrotestosterone; E 2, estradiol-17b; EC, end control;
im, immature; KT, 11-ketotestosterone; m,
maturing; SC, start control; T, testosterone.
from these two subgroups were distributed in two combinations: low GSI, absence of postmeiotic germ cells, and
low pituitary LH content; or high GSI, presence of postmeiotic germ cells, and high pituitary LH content. The end
control group, therefore, was split in two subgroups, which
showed statistically significant differences in all parameters
analyzed. The GSI values and stages of spermatogenesis
(Fig. 1, d and e) were similar in the T-treated fish and in
the immature end control group, whereas the maturing controls were clearly advanced in spermatogenesis.
In experiment D (conducted as experiment C, except that
T treatment started at 15 wk of age), the end control group
was well advanced as regards testis growth and germ cell
development (Fig. 1, f and g) compared to the start control
group. Treatment with T clearly attenuated testis growth.
Because in this experiment the treatment started at a more
advanced stage, the histological analysis also provided evidence for a fallback of spermatogenesis behind the start
control group levels, indicating that more advanced germ
cell stages that had been present, as in the start control
group, were lost during the 4 wk of exposure to T.
Pituitary LH Content, Basal and sGnRHa-Stimulated LH
Secretion In Vitro, and Plasma LH Levels
The pituitary LH content had increased significantly
when the start and the end control groups were compared
1810
CAVACO ET AL.
FIG. 2. Pituitary LH content (a, d, and e),
in vitro secretion (b), and plasma levels of
LH (c and f) 2 wk after implantation of
different steroids either alone or in combination with KT. The pituitary LH content is
given as nanograms per pituitary (mean 6
SEM); columns sharing the same letter are
not significantly different (P . 0.05; ANOVA, followed by Fisher least significant difference test). The plasma levels of LH are
given as nanograms per milliliter of plasma (mean 6 SEM); bars sharing the same
letter are not significantly different (P .
0.05; ANOVA followed by Fisher least significant difference test). In vitro secretion
of LH is given as nanograms per milliliter
of medium (mean 6 SEM) in the absence
(open columns) or presence of 1 nM sGnRHa (cross-hatched columns), which
significantly increase LH release over basal in all groups (P , 0.05; one-tailed,
paired Student t-test). An * indicates significantly higher LH release than the respective control, whereas a # indicates a significantly higher LH release than that in the
other steroid-treated groups. Refer to Figure 1 for further details.
(Fig. 2, a, d, and e), reflecting normal development during
the experimental period of 2 wk. Treatment of 10-wk-old
fish with aromatizable androgens (T and A) or E 2 elevated
the pituitary LH content four- to sixfold above the end control levels, and cotreatment with KT had no effect in this
regard. However, after treatment with KT alone, pituitary
LH contents were lower than in the end control groups.
Implantation of DHT alone had no effect on pituitary LH
levels (Fig. 2d), but it abolished the inhibitory effect of KT
in the respective cotreatment study (experiment B).
Also with regard to pituitary LH levels, two subgroups
(,70 ng/pituitary [immature] and . 1110 ng/pituitary [maturing]) were found in the end control group of experiment
C (Fig. 2e). Different from the shorter treatment of 10-wkold fish, in experiments involving longer treatment with T,
this androgen did not elevate pituitary LH levels above
those found in the maturing end control fish in experiment
C (Fig. 2e) or in experiment D, in which similar pituitary
LH amounts were found for end control (3.3 6 0.8 mg/
pituitary) and T-treated fish (3.1 6 0.5 mg/pituitary). Elevated plasma T levels (Table 1) and the inhibitory effects
on testis growth and spermatogenesis (Fig. 1, d–g) indicate,
however, that T treatment in these fish was effective.
The pituitary release of LH was significantly stimulated
by 1 nM sGnRHa in all groups (Fig. 2b, experiment A).
Whereas basal LH release was similar in most groups (except for the KT 1 T treatment), the amount of LH released
in response to sGnRHa was higher in those groups showing
a higher pituitary LH content than in the end control group.
The reduced LH content in the KT-treated group, however,
was not reflected in a reduced LH release in response to
sGnRHa compared to that in the end control group (Fig.
2b).
The plasma LH levels increased between the start and
end control groups (Fig. 2, c and f). Treatment with KT
had no effect on circulating LH levels, whereas the T- and
A-treated groups showed levels similar to those in the start
control group. Cotreatment with KT had no effect on the
T- or A-induced reduction of plasma LH levels. In experiment C, similar plasma LH values were found in T-treated
and maturing end control fish, which had lower levels than
those found in immature controls (Fig. 2f). In experiment
D, the plasma LH levels did not differ significantly between
the end control (0.7 6 0.1 ng/ml) and T-treated groups (0.8
6 0.2 ng/ml).
DISCUSSION
Previous studies showed that T treatment in juvenile African catfish stimulated gonadotroph activity [5], whereas
KT treatment drove spermatogenesis [9]. This triggered the
present attempt to induce precocious stimulation of both
pituitary gonadotrophs and spermatogenesis by cotreatment
with these two androgens. Unexpectedly, the stimulatory
effect of KT on spermatogenesis was abolished by cotreatment with T or A; a further metabolism of T to DHT or
E 2 is not required for this inhibitory effect. Cotreatment
ANDROGENS AND PUBERTY IN MALE AFRICAN CATFISH
with KT, however, did not interfere with T- or E 2-induced
stimulation of pituitary development.
The results of single-hormone treatments are in agreement with those of previous studies [5], confirming that
aromatizable androgens and E 2 can elevate pituitary LH
levels and sGnRHa-stimulated LH secretion while reducing
plasma LH levels. Treatment with KT, on the other hand,
suppressed pituitary LH levels compared to those in the
end control group. This effect of KT depends on the presence of the testis [5]. Because KT treatment inhibits Leydig
cell activity [18], we propose that the resulting, largely reduced testicular production of T limited the increase in pituitary LH among KT-treated fish. This notion is supported
by the observations that cotreatment with an aromatizable
androgen (or E 2) compensates for the inhibitory effect of
KT on pituitary LH levels. That pituitary LH levels in these
cotreatment groups were indistinguishable from those in
treatments with T or A alone suggests that KT has no major,
direct effect on LH production. Aromatizable androgens,
however, play a key role in stimulating gonadotroph activity in immature male catfish. Also, DHT abolished the inhibitory effect of KT on pituitary LH levels. This may relate to the slight (i.e., twofold) stimulatory effect of DHT
on the steady-state mRNA levels of both LH subunits [5].
However, because DHT has no effect on primary gonadotroph cultures, at least in adult males [6], DHT may not
have a direct effect on the pituitary level.
No difference was found in pituitary LH levels between
end control and T-treated fish during experiment D, in
which the fish were 5 wk older than those used for the other
experiments. The decreased impact of exogenous T in older
fish might reflect increased endogenous stimulation, so that
treatment of juvenile males with T might induce precocious
attainment of a predetermined level of LH in the pituitary.
Unexpectedly, aromatizable androgens inhibited the
stimulatory effect of KT on catfish spermatogenesis. The
initial period of spermatogenesis appears to consist of at
least two phases, and the inhibitory effect of T or A seems
to be restricted to the second phase. This assumption is
based on the observation that neither T nor A inhibited the
initial increase in testis growth from the start to the end
control group, which mainly reflects spermatogonial proliferation [9]. Experiments in the Japanese eel also suggested
that the first five (of ten) mitotic cell cycles of the spermatogonia are regulated differently from the second five
cell cycles [19]. Similarly, in the medaka (Oryzias latipes),
the spermatogonial generations appear to be subdivided
into at least two subgroups, with the first one being characterized by a bisexual potential that is lost after the fourth
mitotic cell cycle [20]. In rodents, spermatogonial generations are subdivided in two subgroups (i.e., undifferentiated
and differentiating) that differ in their morphological and
physiological aspects [21]. We assume that the initial phase
of spermatogonial proliferation in catfish has a low sensitivity toward inhibition by exogenous T or A. However, T
appears to prevent the development, and/or to induce the
loss, of further advanced stages of germ cell development.
Germ cell proliferation and development and, hence,
growth of the spermatogenic cyst are associated with proliferation of the cyst-forming Sertoli cells in the adult fish
testis [22]. In mammals, Sertoli cell proliferation is a prominent regulatory function of FSH, affecting the spermatogenic capacity of the testis [23]. Further experiments will
show if the inhibitory effects of T target, for example, Sertoli cell FSH-receptor expression and/or FSH blood levels.
The latter decreased in response to treatment with T in ju-
1811
venile salmon [24]. With regard to FSH-receptor expression, the inhibitory effect of T on KT-stimulated seminal
vesicle development is interesting, because the seminal vesicles in catfish develop from the caudal part of the testis
anlage. The epithelium of the vesicles is derived from Sertoli cells [25], and we recently showed a high level of FSHreceptor expression in seminal vesicle tissue [26]. One possible explanation for the inhibitory effect of T on KT-stimulated development of both testis and seminal vesicles,
therefore, might be a shared dependency on regulatory processes involving FSH or its receptor. We are presently examining if FSH-receptor expression in testis and seminal
vesicle tissue behaves similarly under different experimental conditions.
Although reduced plasma LH levels were found after
treatment with aromatizable androgens or E 2, indicating
that these steroids inhibited LH secretion, reduced plasma
LH levels are unlikely to have been responsible for the
detrimental effect of T and A on KT-stimulated spermatogenesis, considering that treatment with E 2 also reduced
circulating LH levels [5] but did not impair KT-stimulated
spermatogenesis. Another possible explanation of the inhibitory effect of T on KT-stimulated spermatogenesis relates to the observation that the inhibitory effect of aromatizable androgens on the testis was not mimicked by E 2
or DHT, and that the latter steroids were also ineffective in
compromising Leydig cell function and morphology [18].
This opens the possibility that the inhibitory effect of T on
spermatogenesis might be mediated via suppression of Leydig cell functions that are relevant for spermatogenesis but
distinct from supplying 11-oxygenated androgens. Experiments involving long-term testis tissue cultures (e.g., [10,
27]) should be able to address some of these questions.
Regardless of the exact mode of action of T, its effects
are clearly distinct from those of KT. The recent detection
of two androgen receptors in fish [2–4] with different ligand-binding characteristics opens the possibility that T and
KT use different receptors [28], making fish a unique model
system for studying the role of androgens in the regulation
of spermatogenesis.
In a gonadotropin-deficient mice model, T always had
stimulatory effects on spermatogenesis [29], with the extent
depending on the androgen dose administered. Under certain conditions in the rat, however, such as after interruption
of spermatogenesis following radiation or a cytotoxic treatment, a transient decrease of intratesticular T levels promoted the proliferation and differentiation of A spermatogonia [30]. Experimental models based on fish spermatogenesis may provide interesting aspects for future research
in this regard, considering that the stimulatory effects on
spermatogenesis appear to be linked to KT, with T and A
showing inhibitory effects.
Testosterone is needed to fully stimulate gonadotroph activity in juvenile fish [5]. Although it can have detrimental
effects on spermatogenesis under experimental conditions,
such effects are not apparent during normal puberty in catfish, when stimulation of gonadotrophs provides evidence
for testicular production and biological activity of aromatizable androgens [8] at the time when testis growth and
spermatogenesis progress swiftly. We therefore speculate
that the biological activity of endogenous T is limited by
binding to a hepatic sex steroid-binding globulin and/or a
testicular androgen-binding protein. It is interesting to note
that catfish sex steroid-binding globulin shows high binding
affinities for T and A, but not for KT [31].
In summary, the present study showed that cotreatment
1812
CAVACO ET AL.
of juvenile male African catfish with T and KT did not
induce further stimulation of pubertal maturation compared
to treatments with single hormones. Instead, T inhibited the
stimulatory effect of KT on spermatogenesis. It appears that
a balanced production of T and KT and/or a control of their
biological activities, possibly involving two types of androgen receptors as well as nonreceptor androgen-binding proteins, enable these androgens to regulate testis and gonadotroph development.
ACKNOWLEDGMENTS
The authors wish to thank Mrs. Wytske van Dijk for technical help
and the students Eline Dubois, Colin de Haar, Susanne Janssen, and António Iniesta Martı́n (Faculty Biology, Utrecht University) for their help
with the RIAs and histology of the study.
REFERENCES
1. Borg B. Androgens in teleost fishes. Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 1994; 109:219–245.
2. Ikeuchi T, Todo T, Kobayashi T, Nagahama Y. cDNA cloning of a
novel androgen receptor subtype. J Biol Chem 1999; 274:25205–
25209.
3. Takeo J, Yamashita S. Two distinct isoforms of cDNA encoding rainbow trout androgen receptors. J Biol Chem 1999; 274:5674–5680.
4. Sperry TS, Thomas P. Characterization of two nuclear androgen receptors in Atlantic croaker: comparison of their biochemical properties
and binding specificities. Endocrinology 1999; 140:1602–1611.
5. Cavaco JEB, van Baal J, van Dijk W, Hassing GAM, Goos HJTh,
Schulz RW. Steroid hormones stimulate gonadotrophs in juvenile male
African catfish (Clarias gariepinus). Biol Reprod 2001; 64:1358–
1364.
6. Rebers FEM, Hassing GAM, Zandbergen MA, Goos HJTh, Schulz
RW. Regulation of steady-state luteinizing hormone messenger ribonucleic acid levels, de novo synthesis, and release by sex steroids in
primary pituitary cell cultures of male African catfish, Clarias gariepinus. Biol Reprod 2000; 62:864–872.
7. Xiong F, Liu D, Le Dréan Y, Alsholtz HP, Hew CL. Differential recruitment of steroid hormone response elements may dictate the expression of the pituitary gonadotropin II-subunit gene during salmon
maturation. Mol Endocrinol 1994; 8:782–793.
8. Schulz RW, Zandbergen MA, Peute J, Bogerd J, van Dijk W, Goos
HJTh. Pituitary gonadotrophs are strongly activated at the beginning
of spermatogenesis in African catfish, Clarias gariepinus. Biol Reprod
1997; 57:139–147.
9. Cavaco JEB, Vilrokx C, Schulz RW, Goos HJTh. Sex steroids and the
initiation of puberty in male African catfish, Clarias gariepinus. Am
J Physiol 1998; 44:R1793–R1802.
10. Miura T, Yamauchi K, Takahashi H, Nagahama Y. Hormonal induction of all stages of spermatogenesis in vitro in the male Japanese eel
(Anguilla japonica). Proc Natl Acad Sci U S A 1991; 88:5774–5778.
11. De Leeuw R, Resink JW, Rooyakkers EJM, Goos HJTh. Pimozide
modulates the luteinizing hormone-releasing hormone on gonadotrophin release in the African catfish, Clarias lazera. Gen Comp Endocrinol 1985; 58:120–127.
12. Schulz RW, van der Wind F, Janssen-Dommerholt C, Peute J, Mylonas
CC, Zohar Y, Swanson P, Goos HJTh. Modulation of testicular androgen production in adolescent African catfish (Clarias gariepinus).
Gen Comp Endocrinol 1997; 108:56–66.
13. Schulz RW, Renes IB, Zandbergen MA, van Dijk W, Peute J, Goos
HJTh. Pubertal development of male African catfish (Clarias gariepinus). Pituitary ultrastructure and responsiveness to gonadotropinreleasing hormone. Biol Reprod 1995; 53:940–950.
14. Schulz RW, Bosma PT, Zandbergen MA, van der Sanden MCA, van
Dijk W, Peute J, Bogerd J, Goos HJTh. Two gonadotropin-releasing
hormones in the African catfish, Clarias gariepinus: localization, pituitary receptor bonding and gonadotropin release activity. Endocrinology 1993; 133:1569–1577.
15. Goos HJTh, de Leeuw R, Burzawa-Gérard E, Terlou M, Richter CJJ.
Purification of gonadotropin hormone from the pituitary of the African
catfish (Clarias gariepinus) and the development of a homologous
radioimmunoassay. Gen Comp Endocrinol 1986; 63:162–170.
16. Schulz RW, Lubberink K, Zandbergen MA, Janssen-Dommerholt C,
Peute J, Goos HJTh. Testicular responsiveness to gonadotropic hormone in vitro and Leydig and Sertoli cell ultrastructure during pubertal development of male African catfish (Clarias gariepinus). Fish
Physiol Biochem 1996; 15:243–254.
17. Schulz RW, van der Corput L, Janssen-Dommerholt J, Goos HJTh.
Sexual steroids during puberty in male African catfish (Clarias gariepinus): serum levels and gonadotropin-stimulated testicular secretion
in vitro. J Comp Physiol 1994; 164B:195–205.
18. Cavaco JEB, Blijswijk van B, Leatherland JF, Goos HJTh, Schulz RW.
Androgen-induced changes in Leydig cell ultrastructure and steroidogenesis in juvenile African catfish, Clarias gariepinus. Cell Tissue
Res 1999; 297:291–299.
19. Miura T, Kawamura S, Miura C, Yamauchi K. Impaired spermatogenesis in the Japanese eel, Anguilla japonica: possibility of the existence of factors that regulate entry of germ cells into meiosis. Dev
Growth Differ 1997; 39:685–691.
20. Shibata N, Hamaguchi S. Evidence for the sexual bipotentiality of
spermatogonia in the fish Oryzias latipes. J Exp Zool 1988; 245:71–
77.
21. De Rooij DG, Russell LD. All you wanted to know about spermatogonia but were afraid to ask. J Androl 2000; 21:776–798.
22. Billard R. La spermatogenèse de Poecilia reticulata I. Estimation du
nombre de générations goniales et rendement de la spermatogenèse.
Ann Biol Anim Biochim Biophys 1969; 9:251–271.
23. Meachem SJ, McLachlan RI, de Kretser DM, Robertson DM, Wreford
NG. Neonatal exposure of rats to recombinant follicle stimulating hormone increases adult Sertoli and spermatogenic cell numbers. Biol
Reprod 1996; 54:36–44.
24. Dickey JT, Swanson P. Effects of sex steroids on gonadotropin (FSH
and LH) regulation in coho salmon (Oncorhynchus kisutch). J Mol
Endocrinol 1998; 21:291–306.
25. Loir M, Cauty C, Planquette P, LeBail PY. Comparative study of the
male reproductive tract in seven families of South American catfishes.
Aquat Living Resour 1989; 2:45–56.
26. Bogerd J, Blomenröhr M, Andersson E, van der Putten HHAGM,
Tensen CP, Vischer H, Granneman JCM, Janssen-Dommerholt C,
Goos HJTh, Schulz RW. Discrepancy between molecular structure and
ligand selectivity of a testicular gonadotropin receptor of the African
catfish, Clarias gariepinus. Biol Reprod 2001; 64:1633–1643.
27. Yazawa T, Yamamoto T, Abé S-I. Prolactin-induced apoptosis in the
penultimate spermatogonial stage of the testes in Japanese red-bellied
newt (Cynops pyrrhogaster). Endocrinology 2000; 141:2027–2032.
28. Sperry TS, Thomas P. Androgen-binding profiles of two distinct nuclear androgen receptors in Atlantic croaker (Micropogonias undulatus). J Steroid Biochem Mol Biol 2000; 73:93–103.
29. Singh J, O’Neill C, Handelsman DJ. Introduction of spermatogenesis
by androgens in gonadotropin-deficient (hpg) mice. Endocrinology
1995; 136:5311–5321.
30. Shuttlesworth GA, de Rooij DG, Huhtaniemi I, Reissmann T, Russell
LD, Shetty G, Wilson G, Meistrich ML. Enhancement of A spermatogonial proliferation and differentiation in irradiated rats by gonadotropin-releasing hormone antagonist administration. Endocrinology 2000; 141:37–49.
31. Rebers FEM, Schulz RW, Goos HJTh. Plasma sex steroids binding
activity in the African catfish, Clarias gariepinus. In: Proceedings of
the Fourth International Symposium on the Reproductive Physiology
of Fish; 1991; Norwich, UK. Abstract 129.