| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Pathology and Cell Biology, Faculty of Medicine, University of Montréal, Montréal, Québec, Canada H3T 1J4
Address all correspondence and requests for reprints to: Dr. María L. Vitale, Département Pathologie et Biologie Cellulaire, Faculté de Médecine, Université de Montréal, 2900 boulevard Édouard-Montpetit, Montréal, Quebéc, Canada H3T 1J4. E-mail: vitalem{at}ere.umontreal.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-actinin, but not of vinculin. DA
enhanced association of talin to cell membranes. Increased
talin-membrane interaction may be implicated in DA-induced maintenance
of a round phenotype in lactotrope cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DA and the dopaminergic agonist bromocriptine not only block PRL secretion and synthesis, but in addition they reduce the size of human (8) and rat (9) prolactinomas and modify the morphology of cultured anterior pituitary cells (10). It has been demonstrated that reduction in the size of prolactinomas implicates lactotrope cell shrinkage (11). The microtubules and the microfilaments are the major cytoskeletal components involved in cell shape regulation. The microtubules play an important role in PRL secretion, where they are involved in trafficking of PRL-containing secretory granules (12). Depolymerization of microtubules with colchicine blocks PRL secretion (13). Inhibition of PRL secretion by DA and bromocriptine is accompanied by a decrease in the expression of the microtubule-associated proteins, such as MAP-2 and Tau (14). Conversely, stimulation of PRL secretion is accompanied by an increased polymerization of tubulin (15, 16) and phosphorylation of various microtubule-associated proteins (17). Less is known regarding the involvement of the actin cytoskeleton in the control of PRL secretion and lactotrope morphology. In other cell types, cortical actin filament (F-actin) disassembly occurs during stimulation of secretion (18, 19, 20), allowing secretory granules to reach exocytotic sites at the plasma membrane (21). In a recent study, we have shown that stimulation of PRL secretion from normal, differentiated rat lactotropes triggers cortical F-actin disassembly (22). DA blocks secretagogue-induced cortical actin disassembly and also induces the reassembly and stabilization of lactotrope’s cortical actin filaments (22). Moreover, in lactotropes, as in rat pancreatic acinar cells (19), disassembly of actin filaments by cytochalasin D is sufficient to stimulate secretion (22). Interestingly, DA inhibits cytochalasin D-induced PRL secretion (22). The results suggest that stabilization of the cortical actin cytoskeleton may be one of the mechanisms involved in DA-induced inhibition of PRL secretion.
In the present study, we investigated DA-induced morphological
responses in lactotrope cells and the signalling transduction pathways
that may be implicated in DA-evoked effects. We showed that DA induced
cortical F-actin reassembly, cytoplasmic actin cable disassembly, and
inhibition of cell flattening via activation of the D2 receptor. The
effects were reversible and specific to the lactotropes. Inhibition of
cAMP synthesis, reduction of intracellular Ca2+ levels, and
inactivation of the small GTP-binding protein Rho induced morphological
responses similar to those evoked by DA. On the other hand, elevated
intracellular cAMP levels antagonized DA-induced morphological changes.
In addition, we found that DA enhanced the expression of the actin
filament-anchoring proteins -actinin and talin and increased the
association of talin to cell membranes. Increased talin-membrane
interactions may be related to DA-induced stability of cortical actin
filaments and cell shrinkage.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Fluorescence microscopy and cell rounding up assay
After a 24-h period of recovery, cells were challenged with
culture medium containing different compounds for different periods
according to the specific protocol. For treatments longer than 24
h the media were removed after 24 h and replaced with fresh media
containing the corresponding compound to be tested. After the
treatments, dishes were removed from the incubator, rinsed with regular
Locke’s solution (Locke’s solution containing 1.2 mM
MgCl2, 2.2 mM CaCl2, and 100
µM ascorbic acid), immediately fixed with 3.7%
formaldehyde, permeabilized with acetone, and processed for
fluorescence microscopy as previously described (22). Coverslips were
thoroughly rinsed with PBS (137 mM NaCl, 3 mM
KCl, 8 mM Na2HPO4, and 1.5
mM KH2PO4, pH 7.4) and incubated
for 1 h at room temperature with either 3% nonfat milk in PBS or
1% BSA in PBS to block unspecific labeling. Antibody dilutions were
prepared in 1% nonfat milk in PBS or in 0.5% BSA in PBS,
respectively. To analyze talin distribution in cultured lactotropes, we
double labeled the cells with talin monoclonal antibody (1:20 dilution)
and fluorescein isothiocyanate (FITC)-conjugated antimouse IgG (1:400
dilution), and subsequently with PRL antibody (1:1500 dilution)
followed by tetra-methylrhodamine isothiocyanate
(TRITC)-conjugated antirabbit IgG (1:400 dilution). To study
morphological changes in lactotropes during stimulation or inhibition
of PRL secretion, anterior pituitary cell cultures or
lactotrope-enriched cultures were double labeled for PRL and F-actin.
After blocking, cells were incubated with rat PRL antibody (1:1500
dilution) followed by FITC-antirabbit IgG (1:400 dilution). After these
incubations, coverslips were washed with PBS and incubated for 45 min
at room temperature in the dark with rhodamine-labeled phalloidin
(1:200 dilution in PBS). Cells were observed with a Leitz
Ortholux II fluorescent microscope equipped with an I-filter block for
fluorescein and an M-filter block for rhodamine. To evaluate the impact
of the treatments on cell morphology, the shape (round phenotype
vs. flat phenotype) of lactotropes subjected to different
treatments was recorded. This procedure, which was performed without
knowing the type of treatment applied to the cells (single blind
design), was performed in 100 PRL-immunopositive or PRL-immunonegative
cells per coverslip. Therefore, every final value for a given
experimental condition is the result of the observation of not less
than 500 lactotropes or nonlactotrope cells from 2–3 different
experiments. Photographs were taken with Technical Pan Kodak films
(Eastman Kodak Co., Rochester, NY).
Preparation of the membrane fraction
Lactotrope-enriched cell cultures were rinsed with cold PBS.
Next, lysis buffer [50 mM Tris-HCl (pH 7.4), 1
mM MgCl2, 1 mM EDTA, 1
mM PMSF, 2 µg/ml leupeptin, 1 mM DTT, and 2
µg/ml aprotinin] was added to the dishes, and cells were scraped off
and homogenized. Homogenates were centrifuged at 600 rpm (GS-6R,
Beckman Coulter, Inc., Mississauga, Canada) for 4 min.
Pellets were discarded, and supernatants (S1) were centrifuged at
15,000 x g for 25 min (microfuge E, Beckman Coulter, Inc.). Pellets (P2) that correspond to the membrane
fraction were rinsed with lysis buffer and resuspended in
electrophoresis buffer. Proteins in the supernatants (S2) were
precipitated with 10% trichloroacetic acid, washed with acetone, and
resuspended in electrophoresis buffer (24).
Preparation of the cytoskeletal fraction
Cell cultures were rinsed with cold PBS. Next, PBS containing
1% Triton X-100, 10 mM MgCl2, 5 mM
EGTA, 600 mM KCl, 1 mM PMSF, 2 µg/ml
leupeptin, and 2 µg/ml aprotinin was added to the dishes, and the
cells were scraped off and homogenized. Homogenates were centrifuged at
15,000 x g for 25 min (microfuge E, Beckman Coulter, Inc.). Pellets (cytoskeletal fraction) were rinsed with
the same buffer without detergent (25). Proteins in the supernatant
were precipitated with cold methanol. Proteins in each fraction were
resuspended in electrophoresis buffer.
Protein measurement
Proteins in the samples were measured by the dye binding assay
(Bio-Rad Laboratories, Inc., Mississauga, Canada) or by
the DOTMETRIC protein assay (Chemicon, Temecula, CA).
Electrophoresis and immunoblotting
Proteins were separated by SDS-PAGE and transferred onto
nitrocellulose membranes. Membranes were saturated with BLOTTO medium
and tested for the presence of actin-binding proteins with specific
antibodies diluted in PBS containing 5% skim milk: monoclonal
antibodies against talin (1:500 dilution), vinculin (1:500 dilution),
and -actinin (1:500 dilution). The secondary antibody was coupled to
horseradish peroxidase. Antigen-antibody complexes were detected by
enhanced chemiluminescence.
Statistics
Differences between groups were statistically analyzed using
Student’s t test.
Sources of chemicals and antibodies
Enzymes for anterior pituitary cell dispersion and
chemiluminescence kits were purchased from Boehringer Mannheim (Laval, Canada). Antibiotics, soybean trypsin
inhibitor, poly-L-lysine, BSA (fraction V), dopamine,
forskolin, bromocriptine, 8-bromo-cAMP (8BrcAMP), and
monoclonal antibodies against talin, vinculin, filamin, spectrin, and
FITC- and TRITC-antirabbit IgG were purchased from Sigma Chemical Co. (St. Louis, MO). -Actinin monoclonal antibody was
purchased from Chemicon (Temecula, CA). Protein kinase A (PKA)
inhibitor KT5720 and the Ca2+ channel blocker nifedipine
were obtained from Calbiochem (La Jolla, CA).
Clostridium botulinum C3 exoenzyme was obtained from List
Biological Laboratories (Campbell, CA). Antibodies against anterior
pituitary hormones were provided by the NIH. Rhodamine-phalloidin was
purchased from Molecular Probes, Inc. (Eugene, OR). DA
solutions were prepared in 100 µM ascorbic acid. When
drugs were prepared as stock solutions in dimethylsulfoxide, the final
concentration of the solvent in the working dilutions was less than
0.1%.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Most lactotrope cells incubated in
the presence of 100 nM DA for 48 h were rounded and
smaller than nontreated lactotropes (Fig. 1B). When viewed at higher
magnification, lactotropes incubated in DA-free medium showed clusters
of PRL-containing secretory granules in the cytoplasm and cell
periphery (Fig. 2A, open
arrows). Although cultured lactotrope cells do not possess actin
stress fibers as do fibroblasts, actin cables running through the
cytoplasm of untreated lactotropes were apparent (Fig. 2A', open
arrowhead). As shown previously (22), lactotropes cultured in the
absence of DA displayed a discontinuous cortical staining for F-actin
(Fig. 2A', arrowhead). Instead, DA-treated lactotropes not
only were rounded, but cytoplasmic actin filaments disappeared, and
cortical staining for F-actin was continuous and stronger than that in
nontreated cells (Fig. 2B', arrowhead). The absence of
cytoplasmic actin cables was typical of rounded lactotropes.
|
|
). Cultures were
incubated with increasing concentrations of DA for 48 h. After
that period of treatment, cells were classified as being rounded or
flat, and the percentage of round cells was calculated for each
experimental condition. The results show that DA concentrations as low
as 50 nM, which is close to DA levels in
hypothalamic-hypophysis postal vessels (26, 27), induced cortical actin
cytoskeleton reorganization and cell rounding up in 80 ± 6% of
the cells observed (n = 800 cells). Higher concentrations of DA
did not further increase this percentage. Twenty to 30% of lactotropes
did not respond to DA treatment with a modification of their
morphology. The effect of DA on lactotrope cell shape was mimicked by
the DA analog bromocriptine (Fig. 3A), indicating that
lactotrope D2 receptors were involved in DA-induced cell rounding. The
effect of DA was exclusive to lactotropes. Anterior pituitary cultures
consist of five types of hormone-secreting cells. Most nonlactotrope
cells (PRL-immunonegative cells) present in our cultures were rounded.
Their morphology was not affected by either DA or
bromocriptine treatment even at high concentrations (Fig. 3B). Because anterior pituitary ACTH-secreting cells, the corticotrope
cells, respond to glucocorticoids by modifying their actin cytoskeleton
(28) and because ACTH secretion is modulated by DA (29), we evaluated
whether DA induced actin reorganization and cell rounding up in these
cells, an effect that we could have missed because corticotropes
represent less than 10% of anterior pituitary cells in our cultures.
Accordingly, anterior pituitary cell cultures were incubated in the
absence or presence of DA for 48 h. After the treatments,
ACTH-immunopositive cells were observed and classified as being round
or flat. In the absence of DA, most corticotropes were rounded (74
± 3% round corticotropes; n = 600). A 48-h incubation with 50 or
100 nM DA did not affect that percentage [74 ± 3%
round corticotropes (n = 600) and 75 ± 4% round
corticotropes (n = 600), respectively]. DA morphological
responses were not affected by culturing the cells on other
extracellular matrixes, such as rat tail collagen.
|
, 
). Treatment of the
cells with 100 nM DA prevented the appearance of the flat
phenotype (Fig. 4A, 
). The effect was evident after a 4-h incubation
in the presence of 100 nM DA. In contrast, most
nonlactotrope cells present in the same cultures always remained
rounded regardless of the absence or presence of DA in the incubation
medium (Fig. 4B, and ). Paracrine regulation of anterior
pituitary hormone secretion, particularly PRL, is very important (30);
thus, we prepared lactotrope-enriched cultures to evaluate whether
DA-induced inhibition of cell flattening involved the participation of
anterior pituitary cell types other than lactotropes. Lactotrope cells
in these enriched cultures also responded to DA by maintaining a round
phenotype (Fig. 4C, ). The effect of DA was reversible; after
removal of DA from the incubation medium (DAw, arrow),
lactotrope cells recovered the flat phenotype (Fig. 4C, 
). DA was
also able to induce cell rounding in lactotropes that were already flat
(Fig. 4C, •, DA, arrow).
|
, few lactotrope cells present in
lactotrope-enriched cultures and incubated for 24 h in the absence
of any drug possessed a rounded phenotype, most lactotrope cells were
flat. Treatment of the cultures with 100 nM DA for 24
h significantly increased the percentage of round lactotropes (Fig. 5A, DA). As DA inhibits the synthesis of cAMP in lactotropes, thus
decreasing the activity of the PKA, we tested whether a cAMP-dependent
pathway was involved in DA-induced inhibition of cell flattening. After
a 24-h incubation with increasing concentrations of the specific PKA
inhibitor KT5720 (31, 32), we observed an increase in the percentage of
round lactotropes (Fig. 5A, KT5720). The maximal increase was around
30% [KT5720 - control (C) = 30%]. Simultaneous
incubation of the cultures with DA and the PKA inhibitor yielded a
percentage of round lactotropes of 59% [(KT5720 + DA) - C
= 59%] that was larger than the one observed with DA alone (DA
- C = 48%) or KT5720 alone (KT5720 - C = 25%), but
lower than the arithmetical addition of DA and KT5720 effects (48% +
25% = 73%). To further study the involvement of cAMP-dependent
pathways in DA-induced morphological responses, at least in some
lactotrope cells we analyzed whether high levels of cAMP blocked
DA-induced morphological effects. Treatment of the cells with the
adenylate cyclase activator forskolin or with the membrane-permeant
cAMP analog, BrcAMP, induced cell flattening (Fig. 5B), showing that
increased intracellular cAMP had effects opposite those of DA on
lactotrope morphology. In cultures preincubated with the PKA inhibitor,
exposure to forskolin and BrcAMP failed to induce cell flattening (Fig. 5B), which is consistent with the fact that PKA is downstream the
increase in intracellular cAMP. However, when cell rounding up was
induced by incubating the cultures with DA, a subsequent exposure of
the cells to 1 µM forskolin or 2.5 mM BrcAMP
in the presence of DA rescued the cells from DA-induced morphological
alterations (Fig. 5B), suggesting that the DA-controlled step is
upstream of the cAMP-sensitive process.
|
, Nif). Incubation of the cultures with DA together with nifedipine did
not further increase the percentage of round lactotropes (Fig. 6A, Nif
+ DA), suggesting that a reduction of Ca2+ entry is
involved in DA-evoked morphological responses in all lactotropes
responsive to DA. Moreover, simultaneous incubation of the cells with
nifedipine and the PKA inhibitor KT5720 increased KT5720-induced
inhibition of cell flattening up to DA- and nifedipine-induced levels
(compare Fig. 5A
, KT5720, with Fig. 6A, Nif + KT5720). Forskolin (1
µM) and BrcAMP (2.5 mM) blocked
nifedipine-induced inhibition of cell flattening (Fig. 6A, Nif + F and
Nif + BrcAMP), and nifedipine inhibited cAMP-stimulated cell flattening
[compare Fig. 5B, F and BrcAMP, with Fig. 6A, Nif + F and Nif +
BrcAMP; P < 10-5: F vs. (Nif +
F) and BrcAMP vs. (Nif + BrcAMP)].
|
, incubation
of cells with C3 for 72 h caused an increase in the percentage of
round lactotropes with respect to nontreated cells. This increase was
lower that that caused by DA treatment, perhaps due to the fact that
not all of the cells were able to incorporate C3 or because some
lactotropes did not respond to C3. When cultures were treated with C3
for 72 h and during the last 24 h exposed to 100
nM DA, there was an increase in the percentage of round
lactotropes up to a level similar to that obtained after DA treatment
alone (Fig. 6B, C3 + DA). If during the last 24 h, KT5720 or
nifedipine was added to the incubation medium, C3-induced cell rounding
up was not further enhanced (Fig. 6B, C3 + KT5720 and C3 + Nif).
C3-induced morphological responses were blocked by forskolin (Fig. 6B, C3 + F), but C3 did not affect forskolin-stimulated cell flattening
(compare Fig. 5B, F, with Fig. 6B, C3 + F).
Effect of DA on the expression and intracellular localization of
actin filament-anchoring proteins
DA-induced cell retraction and reorganization of actin filaments
may indicate the participation of several actin regulatory proteins.
Accordingly, we investigated whether DA affects the expression and/or
localization of actin filament-anchoring proteins and/or actin
filament-bundling proteins in lactotropes. Cultured lactotropes
expressed the actin-anchoring proteins talin, vinculin, and
-actinin, but not spectrin (not shown); the cells did not express
the actin filament-bundling protein filamin (not shown). The effects of
DA treatment on the levels and intracellular localization differed for
each the three proteins studied. Lactotropes incubated in the presence
of DA for 24–48 h showed increased levels of talin and -actinin,
but not of vinculin. When membrane enriched and cytosolic fractions
were prepared, we observed that DA induced a dose-response increase in
talin association to the membrane fraction (Fig. 7A). The antibody we used against talin
recognizes not only the whole talin molecule (230 kDa), but also the
190-kDa peptide that generally originates from calpain-catalyzed
cleavage of talin (38). As shown in Fig. 7A, immunoblottings of
lactotrope membrane fractions revealed the presence of two bands with
molecular masses of 230 and 195 kDa. DA treatment inhibited talin
proteolysis into the 195-kDa fragment. The amount of -actinin
recovered in the membrane fraction was not affected by DA treatment;
however, increased levels of -actinin were found in the cytosolic
fraction after incubation of the cells with DA (Fig. 7A).
|
, a, arrowhead). Sometimes the
spotty staining was concentrated in a region of the cell close to the
nucleus (Fig. 7B, b, arrow). After treatment with 100
nM DA for 48 h, talin staining was stronger at the
cell periphery in some lactotrope cells (Fig. 7B, c, open
arrowhead), other cells revealed a punctuated staining stronger
than that observed in nontreated cells (Fig. 7B, d, arrow;
compare to b). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Exposure of cultured lactotropes to DA induces cortical actin assembly (22), dissolution of cytoplasmic actin cables, and maintenance of a rounded phenotype (this paper). We observed that after seeding, lactotrope cells slowly started to flatten. The presence of DA in the incubation medium inhibited lactotrope cell flattening and induced the appearance of the round phenotype in cells that were already flat. Modification of cell morphology in fibroblasts, neuronal cells, and macrophages after treatment with growth factors was reported to occur in minutes (33, 34, 42). Instead, DA effects on lactotrope morphology were slow. A significant decrease in the number of flat lactotrope cells was evident only after a 4-h incubation with DA. The difference could be due to a distinct responsiveness of normal cells compared with that of cell lines or to the fact that intracellular mechanisms involved in both events differ. Twenty percent of lactotropes did not respond to DA and remained flat. In our previous study, we also noticed that 20% of cultured lactotropes were refractory to DA-induced stabilization of the cortical actin cytoskeleton (22). It is possible that this percentage of unresponsive lactotropes corresponds to the subpopulation of lactotropes that are galanin secretors and that do not react to DA by a decrease in PRL secretion (43).
DA-evoked morphological changes were exclusive to lactotrope cells. Anterior pituitary cell cultures include six types of cells, five of which are hormone-secreting cells. Under our experimental conditions, DA treatment only affected the morphology of lactotrope cells (PRL-immunopositive cells). Nonlactotrope cells (PRL-immunonegative cells) remained unaffected despite increasing DA concentrations and longer exposure periods. As corticotrope cells represent a small proportion of anterior pituitary cells in our cultures (<10%), and the corticotrope’s actin cytoskeleton is sensitive to substances that control ACTH secretion (28), we evaluated the possibility that DA not only affected lactotrope cell morphology but corticotrope cell morphology as well. Our results showed that, contrarily to what occurs in lactotropes, corticotrope cells cultured in the absence of DA remained round, and DA had no effect on the actin cytoskeleton of corticotrope or these cell shapes. DA-induced morphological responses in lactotropes were reproduced by bromocriptine, implicating the D2 receptor. Moreover, because the presence of anterior pituitary cell types other than lactotropes did not affect the final outcome of DA on lactotrope shape and actin cytoskeleton organization, we conclude that DA directly stimulated D2 receptors on lactotropes to inhibit lactotrope flattening and assembly of cytoplasmic actin cables.
Dopaminergic stimulation of the D2 receptor, via coupling to distinct
Gi subunits, inhibits cAMP production and
decreases Ca2+ currents (4, 5, 44). cAMP-dependent pathways
are considered the main control of lactotrope secretory activity and
growth. However, a decrease in intracellular cAMP alone is not
sufficient to inhibit PRL synthesis (45). Therefore, Ca2+
may also contribute to the control of lactotrope secretory activity
(46, 47), although the correlation between Ca2+ dynamics
and PRL gene expression is not a simple one (7). Our previous study
showed that increased cAMP and Ca2+ were involved in
secretagogue-induced cortical actin disassembly. Here we show that
inhibition of cAMP synthesis and of Ca2+ entry plays a role
in DA-induced morphological alterations in lactotrope cells. Blocking
PKA activity with the specific inhibitor KT5720 induced the disassembly
of cytoplasmic actin cables and cell rounding up, indicating that an
inhibition of cAMP-dependent pathways mimicked DA-evoked morphological
responses. However, the percentage of cells that were responsive to the
inhibition of PKA was lower than the percentage of round cells observed
after DA treatment, suggesting that in some lactotropes, DA-mediated
inhibition of cell flattening does not implicate a cAMP-dependent
pathway. The percentage of rounded cells after DA treatment was lower
than the percentage of rounded cells observed after incubation with DA
in combination with KT5720, indicating that there exists another subset
of lactotropes that is unresponsive to DA but in which inhibition of
cAMP-dependent pathways induce cell rounding up. Furthermore, the
effects of DA and KT5720 were not additive, suggesting that there is a
third subset of lactotropes in which cAMP-controlled pathways may
contribute to DA-evoked morphological. Elevation of intracellular cAMP
levels by stimulation of the adenylate cyclase with forskolin or with
the membrane-permeant cAMP analog, BrcAMP, yielded a flat phenotype.
Forskolin- and BrcAMP-induced cell flattening were inhibited by the PKA
inhibitor, which is consistent with the fact that PKA is downstream
from the increase in cAMP. cAMP-stimulated cell flattening was
antagonized by nifedipine, indicating that entry of Ca2+ is
required for cAMP-induced cell flattening. However, cAMP-induced cell
flattening was not blocked by DA, suggesting that the DA-sensitive step
must be upstream from the cAMP-dependent mechanisms.
Lactotrope cell rounding up was also induced by the Ca2+ channel blocker nifedipine. Nifedipine was as effective as DA in evoking these morphological responses. Moreover, simultaneous inhibition of Ca2+ channels and PKA slightly enhanced PKA-induced inhibition of cell flattening, and the overall response was similar to that observed after treatment of the cells with DA alone. The results suggest that DA-induced maintenance of a round phenotype is dependent on a reduction of Ca2+ entry. Moreover, there is subset of lactotropes in which inhibition of Ca2+ entry induces cell rounding up independently of cAMP-dependent pathways.
A good body of evidence has shown that growth factor-induced morphological changes are mediated by the Rho subfamily of small GTP-binding proteins. In particular, Rho-related guanosine triphosphatases are known to modulate the dynamics of the actin-based cytoskeleton and thereby cell shape (48). Under our experimental conditions, treatment of lactotrope cells with the C. botulinum C3 exoenzyme, a procedure that inactivates Rho (35), led to cytoplasmic actin cable disassembly and cell retraction, two morphological effects similar to those evoked by DA. The C3 exoenzyme was less effective than DA. In cells already treated with C3, blockade of Ca2+ entry or inhibition of PKA did not enhance C3-evoked increase in the percentage of round lactotropes. In fact, as the cells must be treated with C3 for longer periods of time than with nifedipine or KT5720 to allow the uptake of the toxin and to observe an effect, it is difficult to quantitatively evaluate the interaction among the intracellular signaling pathways affected by C3, KT5720, and nifedipine. As we observed in the case of DA-evoked morphological responses, C3-induced effects were also blocked by forskolin. Therefore, increased cAMP antagonizes the effect of Rho-controlled pathways on actin remodeling in lactotropes. Rho-sensitive steps seem to be upstream from cAMP-regulated mechanisms. C3-mediated inactivation of Rho had similar effects in lactotropes as it does in fibroblasts, cardiomyocytes, and osteoblast-like cells, where C3 has also been reported to provoke loss of stress fibers and cell rounding up (34, 49, 50, 51). Our results suggest that DA-induced reorganization of the actin cytoskeleton of fully differentiated lactotropes, may involve the inhibition of Rho. This is consistent with the fact that growth factors induce cells to spread by activating Rho-dependent pathways. In lactotropes, DA, which inhibits the secretory activity and causes cells to shrink, could be regarded as an antigrowth factor. In these cells, DA could inhibit Rho to stop cell growth. Preliminary experiments from our laboratory demonstrated that treatment of lactotropes with DA released RhoA from the membrane to the cytosol, suggesting DA-induced inactivation of RhoA in lactotropes (52).
DA-induced lactotrope cell shrinkage was accompanied by cytoplasmic
actin cable disassembly and, as we have previously shown (22), by
reassembly and stabilization of cortical actin filaments. DA-induced
actin remodeling may result from modifications in the expression and/or
association to the membrane of several actin cytoskeleton associated
proteins to control filament bundling and membrane filament attachment.
Under our experimental conditions, cultured lactotropes expressed the
actin-anchoring protein vinculin, but not spectrin. The cells express
the anchoring and bundling proteins talin and -actinin, but not the
actin filament bundling protein filamin. Talin, -actinin, and
vinculin are implicated in anchorage of actin filaments to the plasma
membrane (53, 54, 55). Interestingly, DA affected each of these proteins
differently. Neither the vinculin concentration nor its intracellular
localization was affected by DA. However, talin and -actinin seem to
be downstream targets of DA in cultured lactotrope cells. DA had two
distinct effects on talin. Firstly, DA inhibited talin breakdown. It is
generally accepted that the formation of the 190-kDa proteolytic
product of talin is catalyzed by calpain, a Ca2+-dependent
cysteine protease (38). Inhibition of cleavage of lactotrope talin by
DA may be a consequence DA-induced reduction of intracellular
Ca2+ levels causing a decrease in calpain activity.
Secondly, DA increased the association of talin with the membrane
fraction. Increased talin-membrane association may accommodate
anchorage of a larger number of actin filaments. Translocation of talin
from the cytosol to the plasma membrane has been reported after
platelet stimulation, leading to cortical actin polymerization (56).
Talin cross-links actin filaments into networks and bundles (57) and
has the ability to bind directly to membranes to promote actin
polymerization (58). We suggest that increased talin concentration near
the plasma membrane will contribute to cross-link, and therefore
rigidify, actin filaments in the cortical region of the lactotrope
cells. DA treatment also up-regulated -actinin. In sharp contrast to
talin, -actinin association with membranes was not affected by DA,
and increased -actinin levels were mostly found in the cytosolic
fraction. Increased cytosolic levels of -actinin by DA are difficult
to explain. -Actinin cross-links actin bundles to each other (59),
to focal adhesions (54), and to secretory vesicles (60, 61). Therefore,
our results indicate that anchorage of cortical actin filaments in
lactotrope cells after DA treatment does not require -actinin. On
the other hand, as DA disassembled cytoplasmic actin cables in
lactotropes, the increased concentration of -actinin in the
cytosolic fraction may be a consequence of the breakdown of the
filaments. The increase could be also related to the accumulation of
secretory vesicles in the cytosol after DA treatment.
In conclusion, in the present study we have shown that dopaminergic treatment of anterior pituitary cells has important morphological effects exclusively on lactotrope cells that include inhibition of cell flattening and remodeling of the actin cytoskeleton. Negative regulation of cAMP synthesis, Ca2+ entry, and Rho activity may be involved in DA-induced morphological responses. DA-induced morphological alterations were accompanied by cytoplasmic actin cable disassembly (this study) and cortical actin filament reassembly and stabilization (22). Moreover, DA enhances the association of talin to cell membrane, suggesting that DA increased the availability of actin-binding sites on the plasma membrane to facilitate the anchorage, and thus the stability, of cortical actin filaments. A thicker and more stable cortical actin would be a better barrier to prevent exocytosis of PRL-containing granules. Furthermore, DA-induced reorganization of the actin cytoskeleton may be at the origin of the known reduction of macroprolactinoma size that follows DA agonists administration.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Supported by a scholarship from Fonds de la Santé du
Québec.
Received October 30, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-ergocryptine, estradiol, and thyrotropin-releasing hormone in
rat anterior pituitary cells in culture. J Cell Biol 86:377–387-actinin with cytoskeletal
proteins. FEBS Lett 220:291–294[CrossRef][Medline]
-actinin depends on its molecular lenght. J Cell Biol 110:2013–2024-actinin-like
protein in chromaffin granules from adrenal medulla: evidence for
localization on the cytoplasmic surface and binding to actin filaments.
Neuroscience 8:165–181[CrossRef][Medline]
-granular pool of the cytsokeletal protein
-actinin in human platelets that redistributes with the adhesive
glycoprotein thrombospondin-1 during the exocytotic process.
Arteriocler Thromb Vasc Biol 17:2293–2305This article has been cited by other articles:
 |
|
 |
|
  F. M. Stanley Insulin-Increased Prolactin Gene Expression Requires Actin Treadmilling: Potential Role for P21 Activated Kinase Endocrinology, December 1, 2007; 148(12): 5874 - 5883. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Kabbaj, S. R. Yoon, C. Holm, J. Rose, M. L. Vitale, and R.-M. Pelletier Relationship of the Hormone-Sensitive Lipase-Mediated Modulation of Cholesterol Metabolism in Individual Compartments of the Testis to Serum Pituitary Hormone and Testosterone Concentrations in a Seasonal Breeder, the Mink (Mustela vison) Biol Reprod, March 1, 2003; 68(3): 722 - 734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Ben-Jonathan and R. Hnasko Dopamine as a Prolactin (PRL) Inhibitor Endocr. Rev., December 1, 2001; 22(6): 724 - 763. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
50 nM vs. C).
