| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Zoology, Tel-Aviv University (P.M., G.G., H.R., Z.Y.), Ramat Aviv 69978, Israel; and the National Center for Mariculture, Israel Oceanographic and Limnological Research (H.R., A.E.), Eilat, Israel
Address all correspondence and requests for reprints to: Dr. Zvi Yaron, Department of Zoology, Tel Aviv University, Ramat Aviv 69978, Israel. E-mail: yaronz{at}post.tau.ac.il
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Studies of the circulating levels of these hormones also suggest some correlation in their regulation; GH levels appear to increase with GtH II at the time of ovulation and spawning, whereas levels of GtH I, which is predominant during vitellogenesis, decrease at this time (8, 9, 10, 11, 12). Apart from these changes in circulating hormone levels, changes in the sizes of these cell populations have been noted in rainbow trout in which the GtH I gonadotrophs are found in greater number in immature fish, whereas in mature fish the GtH II gonadotrophs predominate (13, 14). At the time of spermiation, the somatotroph population is also notably larger in these fish (15).
Studies in tilapia suggest that the direct effects of steroids on the expression of the GtH ß-subunits may explain some of the changing patterns in their circulatory levels. No direct effects of the gonadal steroids were seen on expression of the tilapia GH gene, although they did appear to increase the sensitivity of the somatotrophs to some of the hypothalamic GH-releasing hormones. The effects of testosterone could be mimicked by estradiol (E2), but were not mimicked by the nonaromatizable 11-ketotesoterone, suggesting that the testosterone is aromatized before eliciting these effects (3, 16, 17).
Despite the apparent overlap in the regulation and levels of activity
of gonadotrophs and somatotrophs, little attention has been paid to the
possible interactions between them. In mammals, locally produced
paracrine factors have been reported to regulate the synthesis and
release of gonadotropins, GH and PRL, and also the differentiation of
the cells. These paracrine factors include locally produced cytokines
and growth factors, neurohormones produced in pituitary cells such as
GnRH, and also the pituitary hormones themselves or parts thereof, such
as the glycoprotein 
-subunit (18, 19, 20, 21, 22). In the goldfish pituitary,
activin and inhibin subunits have also been implicated, as these are
found in the somatotrophs and stimulate the release of both GtH and GH
(23, 24). Little information is available regarding the location or
actions of other cytokines and growth factors in the teleost pituitary.
However, after the finding that insulin-like growth factor I (IGF-I) is
produced locally in multiple organs in tilapia, it has been
hypothesized that it may be involved in autocrine or paracrine actions
in organ-specific functions (25). The effects of IGF-I in the teleost
pituitary, depressing GH release and messenger RNA (mRNA) levels, have
been shown (3, 26).
The aim of this study was to examine possible paracrine interactions operating between the gonadotrophs and somatotrophs in the pituitary of tilapia and to investigate the roles of E2 and IGF-I. This required the initial establishment of a technique for separation of the pituitary cell populations.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Separation of pituitary cells
Pituitaries were collected aseptically, and the cells were
dispersed by trypsinization as described previously (27). After the
addition of FCS (final concentration, 20%) to stop the reaction, the
cells were counted, and the cell suspension (5 ml) was loaded onto the
density gradient (10 ml). The gradient was made up of four
concentrations (35%, 45%, 55%, and 65%) of Universal Separation
Media (Sigma Chemical Co., St. Louis, MO) with densities
of 1.0544–1.0985 g/ml. The cells were centrifuged through the gradient
at 1200 x g for 15 min at 18–20 C. The majority of
cells were found to segregate into three clear fractions (fractions
1–3), whereas some cells failed to penetrate the gradient, and a few
(mostly red blood cells) were found below the densest part of the
gradient. Each fraction was collected by gentle aspiration and rinsed
in 10 ml Hanks’ Balanced Salt Solution (5 min, 800 x
g, 18–20 C). The cells of each fraction were then
resuspended in 1 ml Hanks’ Balanced Salt Solution and counted. The
viability of the cells in each of three major fractions was greater
than 95%. Repeated experiments showed that an average of 59.14 ±
6.57% of the initial number of cells were recovered (n = 8), with
49.47 ± 2.47% of these found in the least dense fraction 1,
29.24 ± 2.31% in fraction 2, and 20.3 ± 0.97% in the
densest fraction 3 (n = 9). The validation of this technique was
performed on numerous occasions, using fish at different reproductive
stages.
Culture of separated cells
After separation, cells were plated in Corning 24- or 96-well
tissue culture plates (Corning, Corning, NY), at a density of 2.5
x 105 cells/well in 1 ml or a density of 6.25 x
104 cells/well in 200 µl medium [medium 199, 10% FCS,
10 mM HEPES, and 1% antibiotic suspension
(Pen-strep-nystatin suspension, Biological Industries, Bet HaEmek,
Israel)]. Previous experiments on GH and GtH release have shown these
to be optimal densities for hormone release. Additional preliminary
experiments were carried out in the present study in which the cells
from fraction 3 were plated at as much as twice this density. On
finding that the increased density had no effect on cell morphology,
subsequent experiments employed the cell densities stated above. The
cells were incubated for 4 days at 28 C under 5% CO2.
On the fourth day, cultured cells were photographed using an inverted microscope (Olympus Corp., Melville, NY) and phase contrast optics. In various experiments, the cells were exposed to recombinant human IGF-I (rhIGF-I; dissolved in 0.1 M acetic acid; Life Technologies, Grand Island, NY), recombinant salmon IGF-I (provided by GroPep, Adelaide, Australia; dissolved in 10 nM HCl), E2, or testosterone (Sigma Chemical Co., dissolved in ethanol). The final concentration of the solvents comprised less than 0.1% of the culture medium. The undiluted FCS contained 7.2 nM IGF-I determined as human IGF-I (Silbergeld, A., personal communication).
Immunocytochemistry (ICC)
ICC was carried out on paraffin-embedded pituitary sections, as
described previously (2), after initial blocking with 10% normal goat
serum. The primary antisera used were anti salmon IGF-I (1:2500; a gift
from GroPep, Adelaide, Australia), antirecombinant tilapia GH (1:8000)
(2), and antiserum produced against the recombinant ß-subunit of
tilapia GtH II (1:4000; Zmora, N., and A. Elizur, unpublished).
The second antibody (goat antirabbit) was used at a dilution of 1:600,
and peroxidase-antiperoxidase (Sigma Chemical Co.) was
used at a dilution of 1:200. The reaction was detected using
diaminobenzidine (DAB) and peroxidase.
Cells were fixed in the culture wells (after plating as described above) overnight by the addition of formaldehyde to the incubation medium to a final concentration of 4%. The ICC reactions employed the GH and GtH IIß antisera as described above, and the technique was essentially the same.
4,6-Diaminidino-2-phenyl-dihydrochloride (DAPI) staining of fixed
cells
The cells were fixed as described above and rinsed in phosphate
buffer solution (1 M PBS, pH 7.4) before the addition of
DAPI (1 µg/ml PBS; Sigma Chemical Co.) for 5 min. After
the removal of DAPI, the cells were rinsed again in PBS before being
examined under an inverted microscope fitted with a U-MNU filter
(Olympus Corp.) and photographed using Fuji Photo Film Co. Ltd. TMX 400 film (Tokyo, Japan).
In the case of double staining, nuclei were stained first with DAPI, and consequently, ICC reactions were performed as described above. The cells were photographed with and without the U-MNU filter, using Fuji Photo Film Co. Ltd. TMX 400 film.
RIAs and measurement of aromatase activity
The cellular hormone content of the different fractions was
measured using RIAs specific for tilapia GtH or GH. The RIA for GH,
using the recombinant hormone, has been described previously (2). The
RIA for GtH was based on native GtH that was purified from tilapia
pituitaries harvested during the spawning season (28).
For the measurement of aromatase activity in cells of the various
fractions, the level of E2 secreted into the incubation
medium was measured after 24-h exposure to graded concentrations of
testosterone (Sigma Chemical Co.). This RIA employed a
previously validated antiserum and an iodinated standard (Diagnostics
Products Corp., Los Angeles, CA). The cross-reaction of the assay was
0.79% with estrone, 0.34% with estriol, and less than 0.0001% with
testosterone, 17,20ß-dihydroxyprogesterone and cortisol
(29). The sensitivity of this assay was 0.66 pg/tube.
Measurement of mRNA levels
Comparative levels of GH, GtH Iß, and GtH IIß mRNA were
measured in 1 x 106 cells from each of the fractions.
Total RNA was extracted from the cells using a modification of the
guanidinium-phenol-chloroform method, as previously described (27). The
samples were run on a 1.2% agarose gel and transferred to nylon
membranes (GeneScreen Plus, New England Nuclear Research Products,
Boston, MA) by Northern blotting. RNA on the membranes was hybridized
with complementary DNA probes for GH and GtH IIß and a DNA probe for
GtH Iß, as described previously (17). After rinsing, the membranes
were exposed to the image plate of a phosphorimager (BAS 1000,
Fuji Photo Film Co., Ltd.) for 1 h.
Statistical analysis
Quantitative analysis of apoptotic cells was performed by
counting the number of normal nuclei in a field (6.4 x 0.95 mm)
that spanned the diameter of the well; this was then compared with the
situation in control (untreated) cells. In addition, the degree of
aggregation of the cells was assessed by counting the number of
aggregates with a diameter greater than 0.15 mm in a similar field.
These measurements were repeated in three or four wells in three
separate experiments.
Statistical analysis employed ANOVA followed by the least significant difference test. All experiments were carried out numerous times.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GH, GtH Iß, and GtH IIß mRNA levels in cells of the different
fractions. Pituitary cells from sexually regressed fish were
dispersed and separated as described above. RNA was extracted from
1 x 106 cells in each fraction and hybridized with
probes for GH, GtH Iß, and GtH IIß mRNAs. GtH Iß and GtH IIß
transcripts were present in fractions 1 and 2 and also in cells of the
top fraction that did not penetrate the gradient. In this particular
experiment, there appeared to be a greater proportion of GtH I cells in
the uppermost fractions compared with cells producing GtH II, although
such differences were not noted in cells from fish at other
reproductive stages (not shown). Neither GtH ß-subunit mRNA was found
in fraction 3. In contrast, GH mRNA was found primarily in the densest
fractions, whereas levels in the least dense fraction were considerably
lower than those in the unsorted cells (Fig. 1
).
|
, A and B),
whereas only isolated cells stained with the GH antiserum in this
fraction (not shown). In contrast, cells in the third fraction stained
primarily with GH antiserum (Fig. 2C), and although a few unstained
cells were also seen, these had notably different morphology and the
ir-somatotrophs appeared to attach to and intertwine with them (Fig. 2D). Differences in the morphology of the cells were noted, as in the
live cells (see below).
|
, A–C). In contrast, the majority of cells in fraction 3 appeared to be
smaller, lacked cytoplasmic extensions, and showed considerably less
aggregation than the other fractions or the unsorted cells (Fig. 3D).
|
, A–C). In
contrast, most of the cells in the third fraction had smaller nuclei
that were round in shape and stained brightly with DAPI. Some of these
contained multiple nuclear fragments within a single cell (Fig. 4D).
|
). ICC studies on these same treated
cells confirmed that they were indeed somatotrophs (Fig. 6, A and B). Similar results were
obtained when salmon IGF-I was used in place of the recombinant human
peptide (not shown).
|
|
ICC localization of IGF-I in the tilapia pituitary
Numerous ICC reactions on sections of tilapia pituitaries failed
to show ir-IGF-I in the region of the somatotrophs (e.g.
Figs. 7 and 8). In some, but not all, pituitaries
from mature fish, IGF-I immunoreactivity was seen in areas of the
proximal pars distalis corresponding to those also reacting with
antiserum to GtH IIß (Fig. 7, A and C). In another pituitary,
ir-IGF-I was also apparent in the rostral pars distalis in a thin layer
of epithelial cells lining the junction of the RPD and the
neurohypophysis (Fig. 8A).
|
|
) and female (Fig. 9B) pituitary cells,
the highest levels of E2 were found in cells of fraction 1.
Cultures of fraction 2 cells contained lower levels of E2,
similar to the levels in unsorted cells. In cells not exposed to
testosterone and those from fraction 3 after all levels of testosterone
exposure, E2 was undetectable (Fig. 9, A and B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the separated tilapia pituitary cell cultures, marked differences in morphology were noted in cells of the various fractions. Notably, cells in the somatotroph-enriched fraction had a peculiar appearance and failed to aggregate. Unsorted cells or those from the least dense fractions examined after 4 days in culture show numerous cytoplasmic extensions. ICC studies revealed that the extensions were present in gonadotrophs as well as in other unidentified cells in these fractions. However, the ir-somatotrophs in fraction 3 were devoid of such extensions. This contrasts with the situation in goldfish, where small cytoplasmic extensions were seen on pituitary cells identified as somatotrophs (31). In fact, when cultured with other cell types, the tilapia somatotrophs appeared to be attached to the spreading cells.
After staining the nuclei with DAPI, a reason for the peculiar morphology of the separated somatotrophs became evident. The nuclei of isolated somatotrophs had undergone condensation and fragmentation typical of apoptosis. Also typical of this kind of cell death was a decrease in the cytoplasmic volume of the cells leading to a shrunken appearance. In contrast, tilapia somatotrophs in mixed culture, as shown in previous studies, retained their viability without showing signs of nuclear damage or cell death (17, 27) (our unpublished observations).
To test the hypothesis that substances secreted from other cell types prevent the peculiar appearance of the isolated somatotrophs, the somatotroph-enriched culture was incubated with conditioned medium from the uppermost fraction, but the appearance of the cells was not altered. However, the addition of either rhIGF-I or E2 to the somatotroph-enriched cultures increased the numbers of normal appearing nuclei and increased the degree of cell aggregation. Previous studies on tilapia pituitary cells have shown that somatotrophs respond to treatment with rhIGF-I by reducing the GH mRNA levels (3), whereas studies in mammals have demonstrated the presence of IGF-I receptors on these cells (32). The present study has demonstrated that in the tilapia pituitary, ir-IGF-I is not found in the somatotrophs but was seen in the same region of the PPD reacting with antisera to GtH IIß in some, but not all, pituitaries. The reason for these variations is not yet apparent. In addition, cells in the RPD, in a location typical of the corticotrophs (33), also reacted with the IGF-I antiserum. Although the type of pituitary cells producing IGF-I in mammals remains in dispute, most evidence points to the folliculostellate cells, with other types, including the somatotrophs, possibly also being involved (20, 21, 32).
Aromatase activity was found only in the least dense fractions of separated cells where the majority of gonadotrophs are located. It is suggested, therefore, that the source of pituitary E2 in tilapia is the gonadotrophs, although the possibility that it originates from other unidentified cells in this fraction cannot be excluded. This is opposed to previous studies of the Oreochromis mossambicus, in which it was suggested that aromatase activity, predominant in the PPD, originates from the somatotrophs (34). Also in another teleost, the longhorn sculpin, and in mammals, somatotrophs have been implicated as the location of this enzyme (35, 36). Aromatase activity was seen in separated gonadotrophs of the African catfish, although it was also present in other cell types (37). It appears that in teleosts, brain and pituitary aromatase levels far exceed those in mammals and also exceed those found in the gonads, suggesting that in situ synthesis may be the major source of estrogen within these tissues (38). In the present study, the location of the aromatizing ability in tilapia pituitary cells was similar in cells from sexually mature males and females, although it was more potent in cells from the male fish. This difference presumably arose because of slight differences in the reproductive state of the fish.
The additive effects of IGF-I and E2 treatment suggest that these factors activate different mechanisms to preserve the somatotroph population. In mammals, endogenously produced IGF-I prevents apoptosis of the ovarian follicles and also slightly increased the production of E2 by cultured preovulatory follicles (39, 40). In granulosa cells of early or preantral follicles, apoptosis increased after estrogen withdrawal, and this was completely prevented by replacement with diethylstilbesterol or estradiol benzoate, whereas androgens had the opposite effect (41, 42).
The reason for the failure of conditioned medium from cells of the upper fractions to prevent the morphological changes in the isolated somatotrophs is not entirely clear. It has been suggested, however, that the control of IGF-I in the pituitary is mediated by GH (21). If a similar situation prevails in tilapia, then a reduction in the number of somatotrophs in the upper fractions could have abated the release of IGF-I. It should be noted that the basal E2 levels in all fractions were below the detection limit (i.e. <6.6 pg/ml was secreted during 24 h) even though these cells were from sexually mature fish. It is probable, therefore, that the conditioned medium (taken from cells of sexually regressed fish) simply did not contain enough E2 or IGF-I to elicit an effect.
There is direct evidence in certain teleosts that the numbers of somatotrophs and type II gonadotrophs increase during the breeding season and are lower in sexually regressed fish (13, 14, 15). In tilapia, cyclical circulating GH levels and their response to GH-releasing hormones are greatest at the height of the reproductive season, whereas GtH IIß mRNA levels peak with maximal gonadal development (16, 17). These changes could arise from either an increase in the size of specific cell populations or simply from increased activity of the same cells. Results from the present study suggest that a decrease in the number of cells producing GtH II, for whatever reason, could lead to a reduction in both the aromatizing ability of the pituitary and the output of IGF-I. This coupled with a decrease in circulating steroid levels (occurring in the sexually regressing fish) could also lead to a decrease in the number of somatotrophs by allowing programmed cell death.
Thus, IGF-I appears to have paradoxical effects on the somatotrophs in tilapia: on the one hand, reducing GH release and synthesis (3) and, on the other, preventing a reduction in the somatotroph population. Such actions mean that although negative feedback operates to control the effects of GH on IGF-I, the positive effects on the viability of the cells ensure the capacity of the GH-IGF-I axis to operate optimally. In contrast, the ability of E2 to have similar positive effects on the condition of the somatotrophs suggests that gonadal steroids may act indirectly to increase this population of cells in reproductively active fish. There is evidence to suggest that both of these actions are of a paracrine nature.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Department of Laboratory Medicine and Pathobiology,
University of Toronto, 100 College Street, Room 351, Toronto, Ontario,
Canada M5G 1L5.
3 The Norman and Rose Lederer Chair of Experimental Biology.
Received June 23, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
subunit of glycoproteins from the bullfrog pituitary:
possible effect on prolactin cell function. Gen Comp Endocrinol 102:141–146[CrossRef][Medline]
-reductase in the teleost brain, spinal cord, and pituitary gland.
Gen Comp Endocrinol 60:244–251[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  J. Aizen, H. Kasuto, M. Golan, H. Zakay, and B. Levavi-Sivan Tilapia Follicle-Stimulating Hormone (FSH): Immunochemistry, Stimulation by Gonadotropin-Releasing Hormone, and Effect of Biologically Active Recombinant FSH on Steroid Secretion Biol Reprod, April 1, 2007; 76(4): 692 - 700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y Kazeto and J M Trant Molecular biology of channel catfish brain cytochrome P450 aromatase (CYP19A2): cloning, preovulatory induction of gene expression, hormonal gene regulation and analysis of promoter region J. Mol. Endocrinol., December 1, 2005; 35(3): 571 - 583. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Schwartz Intercellular Communication in the Anterior Pituitary Endocr. Rev., October 1, 2000; 21(5): 488 - 513. [Abstract] [Full Text]
|
|
|
|
|
|
D. M. Baker, B. Davies, W. W. Dickhoff, and P. Swanson Insulin-Like Growth Factor I Increases Follicle-Stimulating Hormone (FSH) Content and Gonadotropin-Releasing Hormone-Stimulated FSH Release from Coho Salmon Pituitary Cells In Vitro Biol Reprod, September 1, 2000; 63(3): 865 - 871. [Abstract] [Full Text]
|
|
|
|
 |
|
 H. A. Hofmann and R. D. Fernald Social Status Controls Somatostatin Neuron Size and Growth J. Neurosci., June 15, 2000; 20(12): 4740 - 4744. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Windeatt, T. D. Southgate, R. A. Dewey, F. Bolognani, M. J. Perone, A. T. Larregina, T. C. Maleniak, I. D. Morris, R. G. Goya, D. Klatzmann, et al. Adenovirus-Mediated Herpes Simplex Virus Type-1 Thymidine Kinase Gene Therapy Suppresses Oestrogen-Induced Pituitary Prolactinomas J. Clin. Endocrinol. Metab., March 1, 2000; 85(3): 1296 - 1305. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
