| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Articles |
Unité de Dynamique des Systèmes Neuroendocriniens, INSERM U-159, Centre Paul Broca, Paris, France
Address all correspondence and requests for reprints to: Dr. Claude Kordon, INSERM U-159, Unité de Dynamique des Systèmes Neuroendocriniens, Centre Paul Broca, 2ter rue dAlésia, Paris F-75014, France.
| Abstract |
|---|
|
|
|---|
, and insulin-like growth factors I and II), only the first
two exhibited properties similar to those of the conditioned medium.
EGF stimulated binding with an EC50 of 3.6 ± 0.8
pM. Immunoprecipitation of EGF, but not transforming growth
factor-
, inactivated the conditioned medium. The effects of both
conditioned medium and EGF were inhibited by herbimycin, a tyrosine
kinase inhibitor; U73122, a phospholipase C inhibitor; and prior
desensitization of protein kinase C. In contrast, both were insensitive
to pertussis toxin pretreatment. In parallel, EGF did not increase LH
secretion by itself, but potentiated its response to GnRH in a
concentration range of 1 pM to 1 nM, resulting
in a shift of the curve toward lower values of GnRH. It is concluded
that EGF is able to control the accessibility of binding sites to GnRH
and to potentiate the responsiveness of gonadotropes to the
decapeptide. | Introduction |
|---|
|
|
|---|
In the present work, we investigated whether paracrine factors endogenous to the pituitary could play a role in up-regulating GnRH-binding sites. We thus tested the capacity of medium conditioned by long term incubation with pituitary cells to affect GnRH binding in a freshly washed population of adenohypophyseal cells and attempted to identify endogenous factors involved in the effect by combining molecular sieve filtration and immunological methods. Epidermal growth factor (EGF) could be identified as the endogenous substance accounting for the effect of the conditioned medium (CM).
| Materials and Methods |
|---|
|
|
|---|
Cell preparation and culture
Anterior pituitary glands were removed and dispersed as
previously described (1). Briefly, anterior lobes were cut into small
cubes and successively incubated after initial enzymatic digestion with
trypsin (Sigma Chemical Co., St. Louis, MO; 5 mg/ml for 20 min) and
deoxyribonuclease (Sigma; 2 mg/ml for 2 min), with a trypsin inhibitor
(Sigma; 1 mg/ml for 6 min), and finally in a
Ca2+/Mg2+-free medium supplemented with 0.6
mg/ml EDTA (pH 7.4 for 16 min). Fragments were mechanically dispersed
with a siliconed Pasteur pipette for 30 min, and cells were plated in
35-mm diameter plastic culture wells (Nunc, Roskilde, Denmark). The
volume was completed at a 3-ml final concentration with DMEM
(Boehringer Mannheim, Mannheim, Germany) containing phenol red (15
mg/ml), 1% glutamine (0.29 mg/ml), antibiotics (penicillin G sulfate;
83 U/ml), streptomycin sulfate (37 U/ml; Sigma), and 8% FCS treated
with 1% Norit-A charcoal (Sigma) and 0.1% Dextran T-70 (Pharmacia,
Uppsala, Sweden) for 12 h at 4 C to remove steroids. Cell cultures
were maintained in a humidified atmosphere containing 7%
CO2 at 37 C for 4 days.
Ligand iodination
The GnRH agonist
des-Gly10-(D-Ala6)-LHRH ethylamide
(GnRHa; Peninsula Laboratories, Belmont, CA) was used as ligand. The
compound was labeled with [125I]MS-30 (Amersham) by
chloramine-T (7). Specific activity ranged from 11001500 Ci/mmol, and
binding to pituitary membranes reached 40%.
Binding assay
Binding was performed on intact attached cells (1). Culture
wells were washed three times with 1 ml Krebs-Ringer medium (KRM)
containing NaCl (120 mM), KCl (3 mM),
CaCl2 (2.6 mM), MgCl2 (0.67
mM), KH2PO4 (1.2 mM),
glucose (1.2 mM), and HEPES (25 mM), pH 7.4, at
21 C to remove unattached cells. Cells were subsequently incubated for
25 min at 21 C in the presence of final radioligand concentrations
ranging from 0.39 nM. The final volume was 650 µl
unless otherwise indicated in the figure legends. Each experimental
point was run in triplicate. Nonspecific binding was assessed in the
presence of 0.3 µM unlabeled GnRHa. After incubation, the
incubation medium was discarded, and the wells were washed three times
with 1 ml cold KRM supplemented with 0.3% BSA and twice with cold KRM
alone. One milliliter of 0.5 N NaOH with 0.1% SDS was then
added. After 1-h digestion at room temperature, the solution was
transferred to 3-ml plastic tubes for determination of radioactivity by
-spectroscopy (Rack gamma 1270, LKB, Uppsala, Sweden). After
determination of binding, the digested cell solution was assayed for
protein content.
PKC desensitization
Cells were preincubated with phorbol 12-myristate 13-acetate (1
µM; Sigma) in 1% (wt/vol) BSA-DMEM for 24 h before
the GnRHa binding test, as described by Strulovici et al.
(8).
Drugs and inhibitors
U73122, a phospholipase C (PLC) inhibitor (2 mM
final dilution; Biomolecular Research, Plymouth Meeting, PA) was added
without preincubation. Herbimycin A, a tyrosine kinase inhibitor (0.1
mM final dilution; Sigma), was preincubated with the cells
for 24 h before experimentation (9). Pertussis toxin (Sigma; 15
ng/ml) was also added to the culture medium 24 h before the
experiment.
Fractionation procedure
Separation was performed by ultrafiltration with Centri/Por
centrifuge concentrators (Spectrum, Medical Industries) with
Mr cut-offs of 30,000, 10,000, and 5,000. After
centrifugation of CM at 50,000 x g for 20 min to
sediment cellular fragments, the supernatant was successively filtered
through the concentrators. Fractions retained the filter containing
moieties with Mr higher than either 30,000 or 10,000 were
recovered, adjusted to the initial volume with DMEM, and filtered
again. The process was repeated three times for eliminating molecules
of lower Mr. The whole procedure yielded four CM fractions:
CM >30,000, 30,000 < CM > 10,000, CM <10,000, and CM
<5000.
Proteolysis of CM <10,000
CM <10,000 fractions were incubated for 60 min at 30 C in the
presence or absence of thermolysine type X (200 mg/ml; Sigma) and
pronase (200 mg/ml; Boehringer) and subsequently filtered across a
10,000 Mr cut-off centriport to eliminate proteolytic
enzymes. Unconditioned medium (UCM) incubated with proteolytic enzymes
and filtered was used as the control.
Immunoprecipitation
Fractions derived from CM and UCM were treated in the same way.
The CM <30,000 Mr fraction was divided into four aliquots
and incubated either with or without specific antibodies: rabbit
antimouse EGF IgG (Becton Dickinson, Mountain View, CA; 2.4 µl/ml
corresponding to a 417-fold dilution), mouse antihuman transforming
growth factor-
(TGF
) Mab IgG1 (Santa Cruz Biotechnology, Santa
Cruz, CA; 18 µl/ml corresponding to a 56-fold dilution), or both for
90 min at 21 C. The UCM <30,000 Mr fraction was incubated
with both antibodies as a control. After incubation, fractions were
filtered though <10,000 Mr Centri/Por concentrators to
eliminate antibodies. The capacity of both antibodies to cross-react
with rat EGF and TGF
was tested. At a 1:5,000 dilution, rat
[125I]EGF bound to antimouse EGF IgG was displaced by rat
EGF with an IC50 of 0.6 nM. No cross-reaction
was observed with rat TGF
up to 100 nM. Rat TGF
,
which only differs from human TGF
by four amino acid residues, was
recognized by mouse antihuman TGF
on a Western blot. It did not
cross-react with rat EGF.
LH release
Half a million cells per well were incubated at 37 C for 2
h in KRM (1 ml) with or without GnRHa and/or EGF. Each point was
assessed on seven wells. Incubates were sampled, centrifuged at
1500 x g for 10 min, assayed for LH by double antibody
RIA according to the method of Niswender et al. (10), and
expressed as milligrams of rat LH RP-1 standard per ml.
Statistics
Dissociation constants at equilibrium (Kd)
and the maximal number of binding sites (Bmax)
were calculated from specific binding curves by nonlinear regression
using a one-site cooperative model (11). Specific binding was
represented on a Scatchard plot (12). For determination of
EC50 values, all points were analyzed by the computerized
iterative least square method.
LH responses to GnRH and EGF were calculated as a percentage of the maximal response amplitude (Vmax) obtained for saturating concentrations of GnRH within each experiment and submitted to parametric ANOVA.
| Results |
|---|
|
|
|---|
|
|
,
insulin-like growth factor I (IGF-I), and IGF-II] were selected on the
basis of the following criteria: of a proteic nature, corresponding to
a Mr in the range of 5,00010,000, and endogenously
produced by adenohypophyseal cells (13, 14, 15, 16, 17, 18). Two substances
corresponding to all criteria, EGF and TGF
, were able to increase
the number of GnRHa-binding sites, whereas IGF-I and IGF-II were
ineffective (Fig. 3
|

was able to account for the
effect of the CM, we attempted to eliminate them from the medium by
immunoprecipitation. Immunoprecipitation of EGF and TGF
in the CM
with a mixture of antibodies directed against both growth factors
abolished the activity of the filtrate (Fig. 4
did not prevent the increase in GnRH
binding.
|
|
|
| Discussion |
|---|
|
|
|---|
are the most likely
candidates to account for the GnRH binding-stimulating activity
recovered from CM of pituitary cells. Both factors are able to mimic
the effect of CM on GnRH binding; their Mr (between
5,00010,000) is compatible with that of active fractions obtained
after sieve filtration of the medium. Their equipotency on GnRH binding
up-regulation is not surprising because they are known to share the
same receptor (21, 22).
Both growth factors are endogenous to the adenohypophysis. The presence
of EGF has been documented in somatotropes, gonadotropes, thyrotropes,
and corticotropes (13, 14), whereas TGF
seems mostly present in
somatotropes and lactotropes (15). The growth factors may not be
actually produced by all these cell types, as EGF mRNA has only been
detected in somatotropes and gonadotropes, and TGF
mRNA has only
been found in somatotropes, gonadotropes, and lactotropes (23). Under
our experimental conditions, however, only EGF is present in amounts
compatible with the action of CM on GnRH binding, as
immunoprecipitation of both factors or of EGF alone, but not of TGF
alone, was able to block it.
We also tested two other factors endogenous to the pituitary, IGF-I and IGF-II (16, 17, 18), the Mr of which correspond to that of the active fraction of the CM. Neither was able to affect GnRH binding, despite the fact that the IGF-I (but not the IGF-II) receptor also exhibits intrinsic tyrosine kinase activity (24).
Identification of EGF with the factor responsible for up-regulating
GnRH binding in the CM is further substantiated by the finding that
both treatments involve a comparable mode of action. The activity of CM
as well as that of EGF is blocked by herbimycin, an observation
consistent with the presence of a tyrosine kinase domain in the
EGF/TGF
receptor (25, 26, 27). In addition, the intrinsic tyrosine
kinase activity of the growth factors is positively coupled to PLC
(28, 29, 30, 31); abolition of the response to EGF and the medium by U73122, an
inhibitor of PLC (20), thus provides an additional argument that EGF
accounts for the effect of the medium. Finally, activation of PLC
and subsequent release of diacylglycerol result in PKC stimulation (32, 33). This can explain why treatment with CM or EGF failed to unmask
GnRH receptors when PKC was down-regulated by prior exposure of the
cells to large concentrations of phorbol esters; under these
conditions, the effects of both EGF and CM were blocked to a comparable
extent as after treatment with a PLC inhibitor.
NPY, another factor endogenous to the pituitary (34), has been
shown to be able to unmask cryptic GnRH-binding sites (6). When added
to adenohypophyseal cells in culture, NPY also results in binding
up-regulation. The effect of NPY, however, is mediated by a different
mechanism; in contrast to our observation concerning CM and EGF, its
action persists after PKC desensitization, but disappears after
pretreatment with pertussis toxin, an agent that ADP-ribosylates and
subsequently inactivates G
i and
G
o proteins. We also report here that binding
stimulation by NPY is insensitive to herbimycin and U73122, in contrast
to the effect observed under the influence of EGF or CM. This also
rules out participation of NPY in binding up-regulation after exposure
to the CM.
Taken together, these results identify EGF as the active endogenous
factor present in the CM and able to up-regulate GnRH-binding sites.
This does not necessarily preclude a possible additional effect of
TGF
under different physiological or experimental conditions, but
suggests that effective concentrations of TGF
were not achieved
under our incubation conditions.
As previously observed (1, 3, 6), other agents unmasking GnRH receptors also induce a parallel increase in Kd and Bmax. The effect on the Kd cannot be accounted for by distinct affinities of naive and cryptic receptors, as binding isotherms best fit a single binding site without cooperativity. Although we have no interpretation for the change in Kd, increased Bmax can be considered more relevant than increased Kd from a functional point of view, because it is accompanied by a greater sensitivity of the GnRH response; the contrary should have been expected if the response was determined by changes in Kd.
By itself, EGF was not able to induce a significant stimulation of LH release. This observation contrasts with that by Przylipiak et al. (35), who reported stimulation by EGF of basal LH release from dispersed pituitary cells. In the latter study, however, the concentration of EGF used (10 nM) was significantly higher than that in our study (100 pM).
In contrast, the gonadotropin response to GnRHa is potentiated by EGF. Addition of EGF does not affect the maximal amplitude of the LH response, so it does not further stimulate LH release induced by maximally active concentrations of GnRH, but at lower than submaximal doses of GnRHa, the growth factor shifts the response toward higher sensitivities to GnRHa, an effect that results in a 2-fold decrease in the EC50.
These data point to an independent, potentiating, but not additive role of EGF on GnRHa-induced LH release. Under different experimental conditions, Mikaye et al. (36) also reported potentiation by low concentrations of EGF (100 pM) of estradiol-induced LH release from pituitary fragments, suggesting that the actions of EGF on gonadotropes may depend on the hormonal status of the pituitary. In this respect, it is of interest to note that cryptic receptors are spontaneously unmasked in pituitaries sampled from castrated male rats (1, 3).
In conclusion, endogenous EGF can be assumed to regulate GnRH receptor numbers and, subsequently, the responsiveness of gonadotropes to submaximal decapeptide concentrations, suggesting that the cryptic subpopulation of GnRH receptors described in the pituitary can be functionally relevant. Paracrine and possibly also autocrine processes intrinsic to the pituitary may thus participate in the adaptation of gonadotropes to their predominantly episodic mode of stimulation (37).
Received July 15, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. J Winters and J. P Moore Intra-pituitary regulation of gonadotrophs in male rodents and primates Reproduction, July 1, 2004; 128(1): 13 - 23. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. V. Childs and G. Unabia Epidermal Growth Factor and Gonadotropin-Releasing Hormone Stimulate Proliferation of Enriched Population of Gonadotropes Endocrinology, February 1, 2001; 142(2): 847 - 853. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Sosnowski, P. L. Mellon, and M. A. Lawson Activation of Translation in Pituitary Gonadotrope Cells by Gonadotropin-Releasing Hormone Mol. Endocrinol., November 1, 2000; 14(11): 1811 - 1819. [Abstract] [Full Text] |
||||
![]() |
J. Schwartz Intercellular Communication in the Anterior Pituitary Endocr. Rev., October 1, 2000; 21(5): 488 - 513. [Abstract] [Full Text] |
||||
![]() |
J. J. Evans Modulation of Gonadotropin Levels by Peptides Acting at the Anterior Pituitary Gland Endocr. Rev., February 1, 1999; 20(1): 46 - 67. [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 |