0013-7227/00/$03.00/0
Endocrinology
Copyright © 2000 by The Endocrine Society
Vol. 141, No. 4
Printed in U.S.A.
Opposing Actions of Two Transforming Growth Factor-␤
Isoforms on Pituitary Lactotropic Cell Proliferation*
S. HENTGES†, M. PASTORCIC, A. DE, N. BOYADJIEVA,
AND
D. K. SARKAR
Department of Animal Sciences, Rutgers, State University of New Jersey (A.D., N.B., D.K.S.), New
Brunswick, New Jersey 08901; and Departments of Veterinary and Comparative Anatomy (S.H., M.P.),
and Pharmacology and Physiology, Washington State University, Pullman, Washington 99164-6520
ABSTRACT
Three transforming growth factor-␤ protein isoforms (TGF␤1,
TGF␤2, and TGF␤3) have been identified in mammals. These isoforms appear to have similar actions on cell growth in various tissues.
In rat pituitary tissue, TGF␤1 is localized in PRL-secreting lactotropes and has been shown to act on lactotropes to inhibit estradiolinduced cell proliferation. The steroid inhibits the production and
secretion of TGF␤1. It is not known whether the other two isoforms
are produced in and/or act on lactotropes. Using immunocytochemical
detection techniques, we determined that, like TGF␤1, TGF␤3 is
colocalized with PRL in the anterior pituitary of Fischer-344 female
rats. Administration of estradiol increased TGF␤3-immunoreactive
cell numbers, TGF␤3 protein, and TGF␤3 messenger RNA levels in
the pituitary. Determinations of TGF␤3 actions in vitro in primary
cultures of pituitary cells indicated that TGF␤3 concentration dependently increases lactotropic cell proliferation. The growthpromoting action of TGF␤3 was potentiated by estradiol. Immunoneutralization studies indicated that although TGF␤1 antibody failed
to prevent estradiol’s mitogenic action, it potentiated the mitogenic
action of TGF␤3. In contrast, TGF␤3-neutralizing antibody inhibited
lactotropic cell proliferation by estradiol. These data indicate that
unlike many other tissues, TGF␤1 and TGF␤3 have opposite actions
on lactotropic proliferation in the pituitary. Furthermore, TGF␤1 and
TGF␤3 may be involved in estradiol’s mitogenic action on lactotropes.
(Endocrinology 141: 1528 –1535, 2000)
T
Recently, the inhibitory action of TGF␤1 on lactotropic
proliferation has been attributed to actions through the type
II TGF␤ receptors (9, 11). Additionally, altered expression of
these receptor subtypes on lactotropes may in part be responsible for the loss of the ability of TGF␤1 to suppress cell
proliferation in transformed lactotropes that result from estradiol exposure (7, 11). The levels of both immunoreactive
TGF␤ receptor type II protein and in situ TGF␤ receptor type
II mRNA hybrids in the pituitary were significantly decreased during estradiol-induced tumorigenesis (7, 9). Determination of [125I]TGF␤1-binding sites in lactotropes by
double immunohistochemistry and receptor autoradiography revealed specific binding sites of TGF␤1 in lactotropes
of the anterior pituitary (9). [125I]TGF␤1 binding in the anterior pituitary was reduced after estradiol treatment. Hence,
it appears that lactotropes are not only the site of TGF␤1
production, but may also be a site of TGF␤1 action. The
expression and actions of the other two TGF␤ isoforms in
lactotropes have not previously been described.
Given the similarity of the TGF␤ proteins to one another and
the fact that they are thought to act via similar complexes of
TGF␤ receptor subtypes, it is not surprising that in many tissues
TGF␤ isoforms elicit similar responses. However, the differential expression (reviewed in Ref. 12) and receptor affinities (13)
of the TGF␤ isoforms indicate that there may be isoformspecific functions for the TGF␤ proteins. Hence, we studied the
production and actions of the various TGF␤ isoforms in lactotropes in the presence and absence of estradiol.
HE TRANSFORMING growth factor-␤ (TGF␤) superfamily of peptide growth factors is highly conserved
through evolution (1, 2). The three isoforms found in mammals (TGF␤1 to -3) share 60 – 80% amino acid homology, and
each isoform is nearly identical between species. TGF␤1,
TGF␤2, and TGF␤3 have been shown to affect cell growth
and gene expression in a number of cells of different embryological origin (1–5). The effects of TGF␤ isoforms on cell
growth can be stimulatory or inhibitory, depending on the
system considered (3). These isoforms are known to inhibit
epithelial cell proliferation and stimulate mesenchymal cell
proliferation. However, the potency of growth stimulation or
inhibition has been shown to vary between isoforms.
We have previously shown that TGF␤1 is produced in the
pituitary of normal rats (6). Immunohistochemical methods
revealed the presence of TGF␤1 in the lactotropes of the
anterior pituitary gland. Pituitary levels of TGF␤1 messenger
RNA (mRNA) and protein decrease during estradiolinduced cell proliferation in the pituitary gland (6 – 8). Determination of the effects of TGF␤1 on estradiol-induced
lactotropic cell proliferation and PRL secretion from pituitary cells in culture indicated that TGF␤1 inhibits the growth
of lactotropes and decreases PRL secretion (7–10). The action
of this growth factor on the secretion of other pituitary hormones was not evident.
Received April 7, 1999.
Address all correspondence and requests for reprints to: Dr. D. K.
Sarkar, Department of Animal Sciences, Rutgers, State University of
New Jersey, New Brunswick, New Jersey 08901.
* This work was supported by NIH Grants CA-56056, AA-11591, and
AA-00220 (to D.K.S.).
† Present address: Vollum Institute, Oregon Health Science University, L-474, Portland, Oregon 97201.
Materials and Methods
Animals
Fischer-344 female rats (175–200 g) were obtained from Simonsen
Laboratories (Gilroy, CA) and housed in a controlled environment (22
1528
TGF␤1 INHIBITS AND TGF␤3 STIMULATES LACTOTROPIC CELL PROLIFERATION
C; lights on, 0500 –1900 h) and provided with certified rodent chow meal
(Ralston Purina Co., St. Louis, MO) and water ad libitum. Vaginal smears
were inspected daily, and only those animals showing three consecutive
4-day estrous cycles were used for studies involving cycling animals.
Some of these animals were ovariectomized bilaterally and sc implanted
with an 17␤-estradiol-filled SILASTIC brand capsule or an empty capsule (Dow Corning Corp., Midland, MI; length, 1 cm; od, 0.125 in.; id,
0.062 in.) under sodium pentobarbital (40 mg/kg, ip) anesthesia. Animal
surgery and care were in accordance with institutional guidelines and
complied with the NIH policy governed by the Principles for Use of
Animals and the Guide for the Care and Use of Laboratory Animals.
Immunohistochemical localization of TGF␤2, TGF␤3,
and PRL
Anterior pituitary tissues were obtained from cyclic female rats on the
day of estrus or from ovariectomized rats with or without estradiol
treatment after perfusion with 4% formalin in 10 mm PBS (pH 7.4). The
pituitaries were postfixed with 4% buffered-formaldehyde, paraffin embedded, and sectioned into 2-␮m sections. The sections were deparaffinized and rehydrated through a descending series of alcohol solutions
to water. The tissues were processed for immunostaining using TGF␤2
or TGF␤3 antibody (gifts from Dr. K. C. Flanders) and PRL antibody
(PRL-S9, NIDDK) and double immunohistochemical procedures as previously described (6, 8). Sections were first treated with hydrogen peroxide in absolute methanol for 30 min at room temperature to block the
endogenous peroxidase activity. This was followed by incubation with
10% normal serum for 30 min to further block nonspecific binding.
Sections were incubated with primary antibodies overnight at 4 C. The
antiserum specificity was characterized (14) and verified by no staining
in tissues when reacted without primary antibody or with primary
antibodies preabsorbed with a 100-fold excess of antigen. The antibodies
were detected using immunocytochemistry kits (Vector Laboratories,
Inc., Gilroy, CA; or BioGenex Laboratories, Inc., San Ramone, CA). The
sections were counterstained with Gill’s hematoxylin to detect nuclei.
Two investigators independently performed cell counts, which involved
counting five separate areas in each tissue and 500 cells/area.
Primary cultures of anterior pituitary cells
Anterior pituitaries from ovariectomized rats implanted with estradiol-containing capsules for 7–10 days were dissociated enzymatically
with Hanks’ Balanced Salt Solution containing collagenase, deoxyribonuclease, and BSA and were grown on poly-l-lysine-coated coverslips
as described previously (7). Cells were maintained in DMEM (1:1; Sigma,
St. Louis, MO; containing 100 U/ml penicillin and 100 ␮g/ml streptomycin) with high serum (10% FCS) for 1 day and then in medium
containing 2.5% FCS and 10% horse serum for another 2 days. Cultures
were maintained in serum-free DMEM containing human transferrin
(100 ␮m), insulin (5 ␮m), putrescine (1 ␮m), and sodium selenite (30 nm)
at 37 C in 7.5% CO2. The lactotropic cell population, as determined by
identifying PRL-immunoreactive (PRL-IR) cells, was 69 ⫾ 2% in these
cultures (7).
Treatment of primary pituitary cell cultures
Cells were plated on poly-l-lysine-coated 24-well plates. For immunostaining studies the cells were plated onto poly-l-lysine-coated coverslips that were placed in the 24-well plates. Cells were plated at 250,000
cells/well. Primary cultures were maintained for 4 days before the onset
of experimentation. During experimentation, cells were maintained in
DMEM-Ham’s F-12 containing serum supplement (described above).
Cultures were treated as indicated in the figures and in Results. Treatments were changed at 48-h intervals for total treatments of 96 h. In
studies involving bromodeoxyuridine (BrdUrd) incorporation assays,
BrdUrd was added for 4 h in fresh medium containing serum supplement and treatment after the initial 96-h treatment.
1529
harvesting. Cells were then fixed with 99% ethanol and treated with
hydrogen peroxide to block endogenous peroxidase activity. Cells were
incubated at 4 C overnight with BrdUrd monoclonal mouse IgG (1:200;
Becton Dickinson and Co., San Jose, CA) and stained using the Vectastain ABC kit (Vector Laboratories, Inc., Burlingame, CA) with diaminobenzidine as the chromagen. The cells were then incubated with
PRL antibody (1:100,000; PRL-S9, NIDDK) at 4 C overnight and stained
using the Vectastain ABC-AP kit (Vector Laboratories, Inc.) with alkaline
phosphatase as the chromogen. As BrdUrd is incorporated into cells
during S phase, cells with immunoreactivity for BrdUrd were considered to be dividing. Cells immunoreactive for both BrdUrd and PRL
were considered to be dividing lactotropes. Two investigators independently performed cell counts, which involved counting 5 separate areas
in each coverslip and 500 cells/area.
Western blot analysis of TGF␤3 protein
Immunoblotting was performed on pituitary tissue homogenates.
One hundred micrograms of total protein from each pituitary, as determined using the DC protein quantification reagents (Bio-Rad Laboratories, Inc., Hercules, CA), were run out on 12% denaturing polyacrylamide minigels and transferred to nitrocellulose in a semidry
transfer chamber (Bio-Rad Laboratories, Inc.). The nitrocellulose membranes were stained with Ponceau S solution (Sigma) to check the efficiency of transfer and to verify equal protein levels in each lane. The
membranes were then placed in 5% milk block for 5 h, followed by
incubation in primary antibody (monoclonal TGF␤3 antibody, R&D
Systems, Minneapolis, MN; 1 ␮g/ml) in blocking buffer at 4 C overnight.
Horseradish perioxidase-conjugated antimouse IgG (1:1500; 30 min at
room temperature) was used for detection of the antibody by enhanced
chemiluminescence (Amersham Pharmacia Biotech, Arlington Heights,
IL; film was exposed to blotted membranes for 10 min).
Assays of TGF␤1 and TGF␤3 mRNA levels
Total cellular RNA was isolated from anterior pituitary tissue by
guanidium isothiocyanate-phenol extraction (15) and used for Northern
blot analysis of TGF␤1 or TGF␤3. RNA samples were resolved on denaturing 1% agarose gels containing formaldehyde and transferred to
MSI nylon filters (MSI, Westboro, MA) by blotting. The RNA was UV
cross-linked to the filter and then hybridized sequentially with DNA
probes. Prehybridizations were performed for 2 h, and hybridizations
were carried out for 24 h in a solution containing 50% formamide, 1 m
NaCl, 10% dextran sulfate, 1 ⫻ Denhardt’s solution, 2% SDS, and 0.1
mg/ml denatured salmon sperm DNA at 42 C. The filters were then
exposed to Kodak XAR-5 film (Eastman Kodak Co., Rochester, NY) at
⫺70 C. Probes were purified gene fragments labeled by random priming
with [32P]deoxy-CTP to a specific activity of more than 2 ⫻ 109 cpm/␮g.
The 985-bp TGF␤1 probe contained the entire rat complementary DNA
(cDNA) and was the XbaI to HindIII fragment from prTGF␤1 provided
by Drs. S. Quian and A. B. Roberts. The TGF␤3 probe included sequences
from position 831-1449 of the mouse cDNA and was provided by Dr. H.
Moses. The cyclophilin probe was the 117-bp PstI to XmnI fragment
including the 5⬘-portion from the rat cDNA. The 18S probe was the
1.5-kb EcoRI fragment from the human 18S gene cloned into HHCSA 65
and obtained from American Type Culture Collection (Manassas, VA).
Statistics
The data shown in the figures and text are the mean ⫾ se. Data were
analyzed using one-way ANOVA. The post-hoc test used was StudentNewmann-Keuls test; P ⬍ 0.05 was considered significant. Comparisons
between points on two different dose-response curves were performed
using the Bonferroni multiple comparison test; P ⬍ 0.01 was considered
significant.
Lactotropic cell proliferation
Results
Immunohistochemical localization of TGF␤ isoforms
Lactotropic cell proliferation was determined by identifying the cells
that had both BrdUrd and PRL immunoreactivities, as described previously (8). In brief, cultures were treated with 0.1 mm BrdUrd 4 h before
We have shown previously that TGF␤1 immunoreactivity
(TGF␤1-IR) is colocalized in the PRL-IR cells of the anterior
pituitary (6). Using double immunohistochemical methods,
1530
TGF␤1 INHIBITS AND TGF␤3 STIMULATES LACTOTROPIC CELL PROLIFERATION
we determined that TGF␤3-IR was colocalized with PRL-IR
(Fig. 1). Together, these observations suggest that both
TGF␤1 and TGF␤3 were produced in the lactotropes.
TGF␤2-IR cells were identified in the anterior pituitary, but
TGF␤2-positive cells were not lactotropes because they did
not show colocalization with PRL-IR (data not shown).
Previously, we demonstrated that estradiol treatment results in a reduction of TGF␤1-IR in the anterior pituitary (7).
Here we determined that estradiol treatment resulted in an
increased intensity of TGF␤3-IR (Fig. 2A) and the number of
TGF␤3-IR cells (Fig. 2B). As TGF␤2 was not localized to the
lactotropes, the effects of estradiol on TGF␤2-IR and mRNA
levels were not determined.
Determination of TGF␤1 and TGF␤3 mRNA and
protein levels
Measurement of TGF␤1 and TGF␤3 mRNA levels in the
anterior pituitary by Northern blot analysis after estradiol
treatment indicated that estradiol decreased TGF␤1 mRNA
levels, but increased TGF␤3 mRNA levels (Fig. 3). Conversely, ovariectomy increased TGF␤1 and decreased TGF␤3
mRNA levels (Fig. 3).
The expression of TGF␤1 and TGF␤3 proteins coincided
with the expression of the mRNA. TGF␤1 protein levels
decreased with estradiol treatment and increased after ovariectomy (6, 10). In the previous study, RIA was used to quantitate TGF␤1 protein levels. There is not yet a reliable RIA
protocol for TGF␤3, so we had to rely on Western blot analysis to give us a general idea of comparative TGF␤3 levels
under different estrogenic conditions. Western blot analysis
indicated that TGF␤3 protein levels were increased with
estradiol treatment and decreased after ovariectomy (Fig. 4).
The Western blot data shown in Fig. 4 are representative of
six blots, all of which show this same pattern. The TGF␤3
band was specific and was the only band that appeared.
Additionally, the specificity of this antibody was checked by
preabsorption with excess TGF␤3 peptide as described for
FIG. 1. Characterization of TGF␤-IR in the anterior pituitary. This
is a representative photograph showing that TGF␤3 (blackish blue)
cells also contain PRL (brown) in anterior pituitary tissue obtained
from a cyclic (C) female rat on the day of estrus. Tissue was processed for immunostaining using TGF␤3, PRL antibodies, and double immunohistochemical procedures. Arrowheads indicate some
cells positive for both PRL and TGF␤3. A corresponding photograph showing colocalization of TGF␤1 and PRL has been published by us previously (6).
Endo • 2000
Vol 141 • No 4
immunohistochemistry. This was not repeated in the Western blot analysis. Figure 4 indicated a more diffuse band for
the TGF␤3 standard than for the treated samples. The reason
for this difference in migration is not fully understood, but
possibly results from some degradation of this peptide. It
may also be that as the TGF␤3-probed test samples were
prepared from tissue, the peptide was modified slightly in
some way that was not present in the purified peptide. Also,
the samples had to be acid activated because we were probing for intracellular peptide, and it may be possible that this
treatment caused slight conformational change, resulting in
slower migrating bands. We have no positive explanation for
the slight migration difference, but we believe that this does
not negate the observed changes in TGF␤3 protein.
Taken together, the immunohistochemical data, the Western blot analysis, and the quantitation of mRNA indicate that
estradiol differentially regulates TGF␤1 and TGF␤3.
Effects of TGF␤1, TGF␤2, and TGF␤3 on lactotropic
cell proliferation
To determine the biological actions of these growth factors
on the lactotropes, we used primary cultures of pituitary cells
to determine the proliferating response of lactotropes to
TGF␤ isoforms in vitro. TGF␤3 increased the number of lactotropes undergoing DNA replication in a dose-dependent
manner (Fig. 5F). The TGF␤3 concentration required for halfmaximal stimulation was approximately 0.01 ng/ml. TGF␤1
alone produced no significant effect on the growth of low
proliferating lactotropes in primary culture (Fig. 5F). TGF␤2
also had no significant effect on lactotrope growth in primary
cultures (data not shown).
Effect of TGF␤ isoforms on estradiol-induced lactotropic
cell proliferation
The actions of TGF␤1 and TGF␤3 were further determined
by examining the effects of TGF␤3 and TGF␤1 on estradiolinduced lactotropic cell proliferation. TGF␤3 enhanced estradiol’s growth stimulatory action on lactotropes in a concentration-dependent manner (Fig. 5F). The cell-proliferating
effects of TGF␤3 appear to be specific, as the related factor,
inhibin (10 ng/ml), did not affect lactotropic proliferation
(85⫾ 18% of the control value; n ⫽ 4). In contrast to the action
of TGF␤3, TGF␤1 concentration-dependently inhibited estradiol-induced lactotropic cell proliferation (Fig. 5F). TGF␤2
had no significant effect on estradiol-induced lactotropic cell
proliferation. Thus, it appears that although TGF␤1 inhibits
estradiol-induced lactotropic cell proliferation, TGF␤3 stimulates lactotropic proliferation.
Effects of TGF␤1- and TGF␤3-neutralizing antibodies on
lactotropic cell proliferation
Determinations of the blocking effects of the TGF␤1 and
TGF␤3 antibodies indicated that the actions of TGF␤1 and
TGF␤3 on lactotropes are specific. Ten micrograms of the
TGF␤3 antibody were able to block the effect of 10 ng TGF␤3
on estradiol-induced lactotropic cell proliferation (Fig. 4, G
and F). The action of TGF␤1 on lactotropes was completely
blocked by a neutralizing antibody (10 ng/ml TGF␤1, 23.2 ⫾
TGF␤1 INHIBITS AND TGF␤3 STIMULATES LACTOTROPIC CELL PROLIFERATION
1531
FIG. 2. Effect of estradiol on TGF␤3-IR in anterior pituitary tissue. A, Representative photographs show TGF␤3 immunoreactivity (brown) in
a 2-␮m section from the anterior pituitary from an estradiol-treated rat (4-week treatment; left) and from an ovariectomized rat (4 weeks
posttreatment; middle). The sections were counterstained with Gill’s hematoxylin, which accounts for the blue nuclear staining in each cell.
A pituitary tissue section of an estradiol-treated rat showed reduced staining when this section was treated with antigen-preincubated TGF␤3
antibody (right). Scale bar, 20 ␮m. B, Mean ⫾ SE percentage of cells reacted with TGF␤3 antibody in anterior pituitary of ovariectomized (OVEX;
4-week treatment) rats and estradiol-treated OVEX rats (E2; 4-week treatment). a, P ⬍ 0.05. n ⫽ 4 rats/group.
7.9% of control; 10 ng/ml TGF␤1 and 10 ␮g/ml anti-TGF␤1,
105 ⫾ 7.9% of control; n ⫽ 6; P ⬍ 0.001).
Studies involving immunoneutralization of endogenous
TGF␤3 using neutralizing antibodies during estradiol-activated cell proliferation indicated that TGF␤3 antibody completely prevented the cell-proliferating action of estrogen on
lactotropes (Fig. 5G). Hence, TGF␤3 was required for estradiol to stimulate lactotropic cell proliferation.
Immunoneutralization of TGF␤1 did not affect the ability
of estradiol to induce lactotropic proliferation (Fig. 5G).
Hence, the TGF␤1 inhibitory influence is absent during estradiol-activated cell proliferation when the production of
TGF␤1 in the cells is lowered. These data together with the
data showing estradiol inhibition of lactotropic production of
TGF␤1 suggest that lactotrope growth is kept under control
in basal conditions by TGF␤1.
In the absence of estrogen, when TGF␤1 activity was immunoneutralized by TGF␤1 antibody in cultured cells, and
the cells were subsequently subjected to exogenous TGF␤3,
lactotropic growth in response to TGF␤3 was markedly increased. The TGF␤3 growth response in the presence of
TGF␤1 antibody was similar to that observed in the presence
of estradiol. Hence, the reduced TGF␤1 production observed
after estradiol exposure is important for lactotropes to respond maximally to TGF␤3.
Discussion
Previously, we have shown that TGF␤1 is a potent inhibitor of lactotropic cell proliferation (10). These findings have
been confirmed (present report and Refs. 7 and 11). In addition, we previously determined that the level of TGF␤1
protein in the anterior pituitary is decreased in response to
estradiol treatment (6). In the current study we verified that
TGF␤1 mRNA levels correspond to protein levels, indicating
that elevated estradiol levels decrease the production of the
growth inhibitory TGF␤1 peptide. Of major interest is the
finding that TGF␤3 protein production, mRNA synthesis,
1532
TGF␤1 INHIBITS AND TGF␤3 STIMULATES LACTOTROPIC CELL PROLIFERATION
Endo • 2000
Vol 141 • No 4
FIG. 3. Characterization of TGF␤3 and
TGF␤1 mRNA levels in response to estradiol and ovariectomy. TGF␤3 and
TGF␤1 mRNA contents were measured
in the anterior pituitary by Northern
blot analysis as described previously
(7). A, Representative Northern blot autoradiographs showing the changes in
TGF␤1 and TGF␤3 mRNA contents in
the anterior pituitaries from cyclic female rats on the day of estrus (C), ovariectomized rats treated for 4 weeks with
estradiol (E), or ovariectomized rats not
treated with estradiol (O). Eighteen micrograms of total RNA from each pituitary sample were electrophoresed
through a 1.3% agarose gel and transferred to nylon filters. Blots were hybridized with 32P-labeled probes for rat
TGF␤1, mouse TGF␤3, and human 18S
RNA. Blots were autoradiographed for
24 – 48 h at ⫺70 C. B and C, Summary
of mRNA quantification data carried
out by densitometric analysis of the autoradiogram by laser scanner. The values were normalized to the 18S RNA.
n ⫽ 5– 6 rats/group.
and action on proliferation in lactotropes oppose those of
TGF␤1 in response to estradiol. To the best of our knowledge
this is the first time that two isoforms of TGF␤ have been
shown to elicit opposing actions on a single cell type. Many
prior studies by other investigators have revealed differential
expression of the TGF␤ isoforms in a time- and tissue-specific
manner, indicating that the isoforms may have varied functions. Furthermore, the TGF␤ isoforms display differential
binding affinities to the receptor subtypes in certain cells,
including those of the anterior pituitary (11), which may
account for some of the specificity of the actions of the TGF␤s.
In the anterior pituitary, estradiol appears to be a key
regulator of TGF␤1 and TGF␤3 expression and, possibly,
action. Our finding that estradiol increases TGF␤3 production in the pituitary is in agreement with the findings of other
investigators that TGF␤3 mRNA levels are up-regulated by
estradiol treatment in bone. Yang et al. demonstrated that in
bone, TGF␤3 expression (but not TGF␤1 or TGF␤2) is increased with estradiol treatment and is dependent on estrogen receptor mediation (16). The TGF␤3 gene does not contain a typical palindromic estrogen response element. Yang
et al. suggest that estrogen bound to its receptor recognizes
an alternative responsive element on the TGF␤3 gene. Further studies are needed to determine the nature of the estrogen-induced TGF␤3 increase that we observed in the anterior pituitary.
Estradiol stimulated lactotrope proliferation and decreased TGF␤1, but increased TGF␤3 levels in anterior pituitary tissue. Therefore, it may be that estradiol has a dual
action involving down-regulation of TGF␤1 and up-regulation of TGF␤3. The combined effect of releasing an inhibitory
control and increasing a stimulatory factor could result in the
potent mitogenic effect that estradiol has only on lactotropes
in the pituitary. Estrogen potentiation of the TGF␤3 response
can be explained by the fact that the steroid also inhibits
TGF␤1 growth inhibitory influences. Indeed, the suppression of TGF␤1 levels appears to be necessary for TGF␤3 to
exert the maximal proliferative effect; TGF␤3 alone elicits
only moderate stimulation of lactotropic proliferation,
whereas neutralizing TGF␤1 during the addition of TGF␤3
results in a robust proliferative response. Additionally, neutralization of TGF␤3 diminished the ability of estradiol to
stimulate lactotropic proliferation in primary cultures of anterior pituitary cells. Although estradiol acts to increase
TGF␤3 production, it also functions to increase TGF␤3 action,
as demonstrated by the enhanced response observed with
TGF␤3 in the presence of estradiol. Thus, it appears that the
lactotrope is a unique cell type where, under estrogenic conditions, TGF␤ isoforms act as both positive and negative
growth regulators.
The mechanisms by which TGF␤s regulate cell proliferation have not been well established. Our data indicate the
need to investigate the mechanisms of TGF␤ action in an
isoform-specific manner. Primary cultures of anterior pituitary cells from estradiol-treated Fischer-344 rats contained
approximately 69% lactotropic cells. It is feasible that TGF␤3
could have a paracrine action on neighboring cells to induce
the release of other growth-stimulating factors. In aortic
smooth muscle cells, TGF␤ stimulates cell proliferation by
increasing platelet-derived growth factor (PDGF) synthesis
(3). For endothelial cells, factors other than PDGF would be
expected to be indirect mitogenic factors, because these cells
TGF␤1 INHIBITS AND TGF␤3 STIMULATES LACTOTROPIC CELL PROLIFERATION
1533
FIG. 4. Western blot analysis of TGF␤3
protein in anterior pituitary tissue of
cyclic (on the day of estrus), ovariectomized (4 weeks postovariectomy), and
estradiol-treated (4-week treatment)
rats. Pituitaries were homogenized,
and samples were acid activated and
neutralized before loading. One hundred micrograms of protein were loaded
from each sample. A, A representative
photograph of a Western blot incubated
with antibody specific for TGF␤3 (0.2
␮g/ml). Lane A, Estradiol; lane B, cyclic;
lane C, ovariectomized; lane D, 10 ng
recombinant human TGF␤3. B, The
same membrane as that blotted in
A was stained with Ponceau S solution
to check transfer efficiency and equal
protein loading. C, The relative protein amount in each band was determined by densitometric analysis of the
enhanced chemiluminescence-exposed
film by laser scanner. For each blot the
density of the band corresponding to
TGF␤3 in the treatment groups was
compared with that in the control (cyclic) tissue. a, P ⬍ 0.05. n ⫽ 6 rats/group.
usually lack PDGF receptors (3). In fibroblasts, the stimulation of cell growth by TGF␤ involves the ability of TGF␤ to
transiently increase cellular responses to a variety of exogenously added growth-promoting factors, such as epidermal
growth factor, bombesin, and vasopressin (17). TGF␤ may
also participate in the stimulation of fibroblast proliferation
by increasing collagen production (18). These data suggest
that mitogenesis by TGF␤ may be mediated indirectly
through other growth stimulatory peptide factors; we have
studied this possibility (18a). Whether the action of TGF␤3 on
lactotropes is mediated by other growth factors as well as the
receptor system used by TGF␤3 will need to be addressed in
future studies.
Estradiol enhances cell proliferation in tissues such as
bone, kidney, uterus, and mammary gland (19, 20). These
estradiol-responsive tissues have also been identified as
1534
TGF␤1 INHIBITS AND TGF␤3 STIMULATES LACTOTROPIC CELL PROLIFERATION
Endo • 2000
Vol 141 • No 4
FIG. 5. TGF␤1 and TGF␤3 regulation of lactotropic cell proliferation. Primary cultures of anterior pituitary cells were prepared as described
in Materials and Methods. A–E, Photographs show PRL-stained (red), BrdUrd-stained (brown), and hematoxylin-stained (blue) cells in
representative cultures treated with vehicle (A; 4 mM HCl and 1% BSA; control), estradiol (B; 10 nM), estradiol and TGF␤1 (C; 10 nM estradiol
and 10 ng/ml TGF␤1), estradiol and TGF␤3 (D; 10 nM estradiol and 10 ng/ml TGF␤3), and estradiol and anti-TGF␤3 (E; 10 nM estradiol and
10 ␮g/ml TGF␤3 antibody). Arrowheads indicate some double stained cells. Cells were plated for 4 days in culture, then treated for 4 days with
the various peptides and/or antibodies. Four hours before fixation of the cells, cultures received 0.1 mM BrdUrd, and using double immunocytochemistry, we determined the percentage of lactotropes proliferating by colocalizing BrdUrd and PRL immunoreactivities in a single cell.
Bar, 20 ␮m. F, Graph depicting the dose-response effects of TGF␤1 and TGF␤3 in estradiol’s presence and absence on lactotropic proliferation.
n ⫽ 5–12/group. ANOVA indicated significant dose-response effects for TGF␤3 (F ⫽ 5.1; P ⬍ 0.004), TGF␤1 with estradiol (F ⫽ 3.9; P ⬍ 0.01),
TGF␤3 with estradiol (F ⫽ 7.2; P ⬍ 0.0005), and TGF␤1 antibody with TGF␤3 (F ⫽ 3.7; P ⬍ 0.02). There was a significant difference between
groups treated with TGF␤3 alone and those treated with TGF␤3 and estradiol (P ⬍ 0.001 for each dose of TGF␤3). Immunoneutralization of
TGF␤1 significantly increased the ability of TGF␤3 to stimulate lactotropic cell proliferation compared with that of TGF␤3 alone (P ⬍ 0.01 for
all TGF␤3-treated groups). G, Histogram summarizing the effects of anti-TGF␤1 (10 ␮g/ml) and anti-TGF␤3 (10 ␮g/ml) on estrogen’s ability
to induce proliferation in lactotropes. 17␤-Estradiol (10 nM) was used. The control and estradiol groups received 10 ␮g/ml rabbit ␥-globulin
(Calbiochem, La Jolla, CA). n ⫽ 5– 6/group.
TGF␤ target tissues (21). The estrogen agonist/antagonist
drug raloxifene acts as an antagonist to estrogen’s actions in
some tissues, such as breast and uterus, while acting to mimic
estrogen’s protective actions in bone. The selective action of
raloxifene has been correlated to an isoform-specific increase
in TGF␤ expression, where TGF␤3 is selectively increased;
estradiol has a similar action, but at higher concentrations
than those needed with raloxifene (16). The pituitary of F-344
rats appears to be another tissue where estrogen may exert
its actions by increasing TGF␤3 expression.
The results presented here suggest that under basal conditions lactotropic growth is halted or maintained at a low
rate by TGF␤1 produced locally in lactotropes and that estradiol, by reducing the inhibitory influence of TGF␤1 and
TGF␤1 INHIBITS AND TGF␤3 STIMULATES LACTOTROPIC CELL PROLIFERATION
enhancing the production of stimulatory TGF␤3, activates
DNA replication and lactotropic proliferation. Whether this
bifunctional regulation of growth by two closely related
TGF␤ proteins is unique to lactotropes or exists in other
estradiol-responsive cells will be an interesting question to
further the understanding of the mitogenic actions of
estradiol.
Acknowledgments
The authors thank the NIDDK and Dr. K. C. Flanders for supplying
the PRL and TGF␤ antibodies, Drs. S. Quian and A. B. Roberts for
providing the TGF␤1 c DNA, and Dr. H. Moses for providing the TGF␤3
cDNA.
References
1. Roberts AB, Sporn MB 1990 Peptide growth factors and their receptors. In:
Sporn MB, Roberts AB (eds) Handbook of Experimental Pharmacology.
Springer-Verlag, Heidelberg, vol 95:419 – 472
2. Massague J 1990 The TGF-␤ family. Annu Rev Cell Biol 6:597– 641
3. Battegay EJ, Raines EW, Seifert RA, Bowen-Pope DF, Ross R 1990 TGF-␤
induces bimodal proliferation of connective tissue cells via complex control of
an autocrine PDGF loop. Cell 63:515–524
4. Reiss M, Saetorelli AC 1987 Regulation of growth and differentiation of
human keratinocytes by type b transforming growth factor and epidermal
growth factor. Cancer Res 47:6705– 6709
5. Shipley GD, Pittelkow MR, Wille JJ, Scott RE, Moses HL 1986 Reversible
inhibition of normal human prokeratinocytes proliferation by type b transforming growth factor-growth inhibitor in serum free medium. Cancer Res
46:2068 –2071
6. Burns G, Sarkar DK 1993 Transforming growth factor-␤1-like immunoreactivity in the pituitary gland of the rat: effect of estrogen. Endocrinology
133:1444 –1449
7. Pastorcic M, De A, Boyadjieva N, Vale W, Sarkar DK 1995 Reduction in the
expression and action of transforming growth factor-␤1 on lactotropes during
estrogen-induced tumorigenesis. Cancer Res 55:4892– 4898
8. De A, Boyadjieva N, Pastorcic M, Sarkar DK 1995 Potentiation of estrogen’s
mitogenic effect on the pituitary gland by alcohol consumption. Int J Oncol
7:643– 648
9. De A, Morgan TE, Speth RC, Boyadjieva N, Pastorcic M, Sarkar DK 1995
1535
Pituitary lactotrope expresses transforming growth factor ␤ (TGF␤) type II
receptor mRNA and protein and contains 125I-TGF␤1 binding sites. J Endocrinol 149:19 –27
10. Sarkar DK, Kim KH, Minami S 1992 Transforming growth factor-␤1 mRNA
and protein expression in the pituitary gland and its action on PRL secretion
and lactotropic growth. Mol Endocrinol 6:1825–1833
11. Sarkar DK, Pastorcic M, De A, Engel M, Moses H, Ghasemzadeh B 1998 Role
of transforming growth factor (TGF)-␤ type I and TGF-␤ type II receptors in
the TGF-␤1-regulated gene expression in pituitary prolactin-secreting lactotropes. Endocrinology 139:3620 –3628
12. Roberts AB, Kim SJ, Noma T, Glick AB, Lafyatis R, Lechleider R, Jakowlew
SB, Geiser A, O’Reilly M, Danielpour D, Sporn MB 1991 Multiple forms of
TGF-␤: distinct promoters and differential expression. In: Bock GR, Marsh J
(eds) Clinical Applications of TGF-␤. Wiley & Sons, West Sussex, vol 157:7–15
13. Lyons RM, Miller DA, Graycar JL, Moses HL, Derynck R 1991 Differential
binding of transforming growth factor-␤1, -␤2, and –␤3 by fibroblasts and
epithelial cells measured by affinity cross-linking of cell surface receptors. Mol
Endocrinol 5:1887–1896
14. Zhao Y, Chegini N, Flanders KC 1994 Human fallopian tube expresses transforming growth factor (TGF) ␤ isoforms, TGF-␤ type I-III receptor messenger
ribonucleic acid and protein, and contains [125I]TGF-␤-binding sites. J Clin
Endocrinol Metab 79:1177–1184
15. Chomczynski P, Sacchi M 1987 Single step method for RNA isolation by acid
guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem
162:156 –159
16. Yang NN, Venugopalan M, Sushant H, Glasebrook A 1996 Identification of
an estrogen response element activated by metabolites of 17␤-estradiol and
raloxifene. Science 273:1222–1224
17. Newman MJ 1993 Transforming growth factor ␤ and the cell surface in tumor
progression. Cancer Metastasis Rev 12:239 –254
18. Ignoltz RA, Massaugé J 1986 Transforming growth factor stimulates the
expression of fibronectin and collagen and their incorporation into the extracellular matrix. J Biol Chem 261:4337– 4345
18a.Hentges S, Boyadjieva N, Sarkar DK 2000 Transforming growth factor-␤3
stimulates lactotrope cell growth by increasing basic fibroblast growth factor
from folliculo-stellate cells. Endocrinology 141:859 – 867
19. Wicklund JA, Wertz N, Gorski J 1981 A comparison of estrogen effect on
uterine and pituitary growth and prolactin synthesis in F-344 and Holtzman
rats. Endocrinology 109:1700 –1707
20. Llyod HM, Meares JD, Jacobi J 1975 Effects of oestrogen and bromocryptine
on in vivo secretion and mitosis in prolactin cells. Science 255:497– 498
21. Soto AM, Sonnenschein C 1987 Cell proliferation of estrogen-sensitive cells:
the case for negative control. Endocr Rev 8:44 –52