| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
INSERM, U-344, Endocrinologie Moléculaire, Faculté de Médecine Necker, 75730 Paris Cedex 15, France
Address all correspondence and request for reprints to: Dr. Marie-Catherine Postel-Vinay, INSERM, U-344, Faculté Necker-Enfants Malades, 156 rue de Vaugirard, 75730 Paris Cedex 15, France. E-mail: postel-vinay{at}necker.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B DNA-binding activity in
Ba/F3 cells. Inhibition of nuclear factor-B through expression of
the mutant IB
(A32/36) abrogated the GH-mediated survival signal,
but did not result in alterations of the cell cycle in Ba/F3 GHR cells
treated with IGF-I. Phosphatidylinositol 3-kinase was required for both
survival and proliferative responses to IGF-I. Transfection of a
dominant negative form of AKT (AH-AKT) resulted in suppression of
IGF-I-mediated cell survival, but not of the antiapoptotic effect of GH
in Ba/F3 GHR cells. Thus, GH and IGF-I are able to promote cell
survival and proliferation through independent and different pathways
in Ba/F3 cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cytokine-dependent cell lines have provided means to study signal events that are involved in cell proliferation and survival. Pro-B Ba/F3 cells are dependent on IL-3 for their growth. However, when the cells are stably transfected with the GH receptor (GHR) complementary DNA (cDNA), they become able to proliferate in response to GH (5). Considering the lack of current availability of a cell line expressing endogenous GHRs and in which a GH response can be measured, Ba/F3 cells expressing the GHR represent a valuable model to identify the signaling molecules involved in the proliferative and antiapoptotic effects of GH (4, 5).
Hemopoietic cells are dependent upon cytokines and growth factors
contained in the serum. Indeed, growth factor withdrawal results in
cell cycle arrest and apoptosis (6). Many cytokines are
known to activate phosphatidylinositol 3-kinase (PI 3-kinase), which
transduces signals for cell cycle entry and proliferation commitment
(7). A downstream target of PI 3-kinase is the
serine/threonine kinase AKT (8), which can play a key role
in cell survival. Activation of the PI 3-kinase/AKT pathway seems to be
essential for interleukin-2 (IL-2), IL-3, nerve growth factor, and
IGF-I proliferative effects in several cellular models
(9, 10, 11, 12, 13, 14). However, it has been shown that inhibitors of PI
3-kinase dramatically block the survival effects of IGF-I, whereas they
do not affect the survival of IL-3-stimulated Ba/F3 cells (15, 16). Indeed, recent studies have demonstrated that the major
pathway of inhibition of apoptosis of Ba/F3 cells by IL-3 involves
nuclear factor-B (NF-B) (17). Likewise, GH-mediated
survival in Ba/F3 cells expressing the GHR depends on NF-B
activation (4). Thus, activation of the PI 3-kinase/AKT
pathway seems to deliver survival signals specific to the cell type and
the extracellular stimulus.
Both direct and indirect actions of GH have been shown in immunocompetent cells (18). In the pro-B Ba/F3 cell line, IGF-I has been shown to inhibit cell death (15). It can be expected that Ba/F3 cells are able to produce and secrete IGF-I; it has been shown that Epstein-Barr virus-transformed human B lymphocytes secrete IGF-I and that GH can enhance the secretion (19). Thus, the question of the mediating role of IGF-I generation in the effects of GH on Ba/F3 cells had to be addressed.
We recently reported that in contrast to parental Ba/F3 cells, cells
expressing the GHR (Ba/F3 GHR cells) do not undergo apoptosis under
IL-3 and serum deprivation; endogenous GH produced by the cells has
been shown to be responsible for cell survival (4). Using
these cells, we sought to better identify the pathway mediating the
proliferation and antiapoptotic signals induced by GH and IGF-I. Ba/F3
cells stably expressing the GHR were used, because both GH and IGF-I
can induce proliferation and survival in the cells. Evidence was
obtained that 1) GH effects are not mediated by IGF-I in this cellular
model, 2) the signaling pathways involved in the actions of the two
hormones are distinct. The transcription factor NF-B, previously
shown to be crucial for signaling the GH antiapoptotic effect, does not
appear to be activated by IGF-I. PI 3-kinase, which is needed for
proliferative and survival IGF-I effects, does not appear to be
necessary for the GH survival effect.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Plasmids
pRCßactin containing mutant IB (A32/36) expression vector (20) was provided by Michael Karin (University of California, San Diego, CA). The deletion mutant of AKT (AH-AKT) retaining only residues 1–147 (21) and its control counterpart were given by Julian Downward (Imperial Cancer Research Fund, London, UK). The thymidine
kinase promoter-driven luciferase reporter plasmid (22),
controlled by six reiterated B sites, was a gift from George Rawadi
(Hoechst-Marion-Roussel, Romainville, France).
Cell culture and stimulation
The parental Ba/F3 cell line is an IL-3-dependent murine pro-B
cell line. As Ba/F3 wild-type (WT) cells do not express endogenous GH
receptors, stable transfectants, Ba/F3 GHR cells, were prepared; they
stably express rat GH receptors (4, 5). Both cell lines
were maintained in RPMI 1640 medium supplemented with 10% FCS, 2
mM L-glutamine, 10 U/ml penicillin, 10 µg/ml
streptomycin, 50 µM 2-mercaptoethanol, and 10% WEHI-3B
cell supernatant as a source of IL-3 (normal medium). For stimulation
experiments, cells were seeded at 0.5 x 106
cells/ml density and starved for 6 h in a serum- and IL-3-free
medium containing 2% BSA (fraction V, Sigma), 2
mM L-glutamine, 10 U/ml penicillin, 10 µg/ml
streptomycin, and 50 µM 2-mercaptoethanol (starvation
medium). Cells were then stimulated with either bGH (1 µg/ml, 50
nM) or IGF-I (350 ng/ml, 50 nM) for the
indicated time intervals. IL-3 (10 ng/ml, 0.7 nM) was also
used in control experiments to stimulate Ba/F3 in the absence of serum.
The cytokine concentrations were chosen on the basis of the
proliferative responses observed in dose-response curves.
RT-PCR and IGF-I assay
Total RNA was prepared from 1 x 106
Ba/F3 GHR cells using the TRIzol reagent method (Life Technologies, Inc., Gaithersburg, MD). Two micrograms of total
RNA were transcribed into cDNA using 200 U reverse transcriptase from
murine leukemia virus (Life Technologies, Inc.,
Gaithersburg, MD) in 20 µl of a reaction mixture containing 50
mM KCl, 20 mM Tris-HCl (pH 8.4), 2.5
mM MgCl2, 0.1 mg/ml BSA, 1
mM of each deoxy-NTP, and 20 U of the ribonuclease
inhibitor RNasin (Promega Corp., Madison, WI). The cDNA
synthesized during incubation for 1 h at 42 C was used as a
template for PCR in a reaction mixture containing 5 U Taq
DNA polymerase (Life Technologies, Inc.), 50
mM KCl, 20 mM Tris-HCl (pH
8.4), 2.5 mM MgCl2, 0.1
mg/ml BSA, 0.2 mM of each deoxy-NTP, and 0.2
mM of the two IGF-I-specific primers, as
previously described (23). After 25 cycles (1 min at 95 C,
1 min at 55 C, and 2 min at 72 C) of PCR, DNA fragments were
electrophoresed on 1% agarose gel and revealed under UV light.
For IGF-I assay, 4 ml culture medium were incubated in acid medium (0.01 M HCl) for 30 min at room temperature to dissociate IGFs from IGF-binding proteins, then ultrafiltered on Centricon 30 (Amicon, Epernon, France) to separate IGFs from IGF-binding proteins, as previously described (24). Eluates were lyophilized and before being assayed were desalted on Sephadex G-25 disposable columns (Pharmacia Biotech, Uppsala, Sweden) in assay buffer. IGF-I was assayed by RIA as previously described (25), using a rabbit anti-IGF-I polyclonal antibody (gift from Dr. J. Closset, Liege, Belgium). After incubation, free and bound IGFs were separated using albumin-coated charcoal. The threshold sensitivity of the assay was 1–2 ng/ml. Intraassay variation was 5%, and interassay variation was 10%. The IGF preparation used for radiolabeling and for standards was recombinant human IGF-I, provided by Ciba-Geigy (Basel, Switzerland).
Cell cycle analyses
Ba/F3 cells were synchronized in
G0/G1 phase by incubation
in starvation medium for 6 h. Cells were stimulated with 10 ng/ml
IL-3, 350 ng/ml IGF-I, or 1 µg/ml GH for 24 or 48 h. Progression
through the cell cycle was monitored by detection of the DNA content.
Cells were harvested by centrifugation, permeabilized with 30 µl
DNA-Prep LPR reagent, followed by addition of 0.5 ml DNA-Prep stain
propidium iodide solution (DNA-Prep reagents, Coulter Corp., Miami,
FL). Samples were analyzed by FACScan (Becton Dickinson and Co., Mountain View, CA).
Immunoprecipitations
For AKT immunoprecipitation, 5 x 106
cells were lysed in a buffer containing 1% Triton X-100, 50
mM Tris-HCl (pH 7.5), 150 mM NaCl, 5
mM EDTA, 1 mM phenylmethysulfonylfluoride, 10
µg/ml leupeptin, 10 µg/ml aprotinin, 10 µg/ml trypsin inhibitor,
and 1 mM NaVO4, pH 7.2. Supernatants
were incubated overnight in the presence of anti-AKT antibody and
protein A Sepharose beads. After extensive washings in lysis buffer,
pellet (20 µl of solid beads) was resuspended in 20 µl sample
buffer containing dithiothreitol. Immunoprecipitates were separated on
10% SDS-PAGE and analyzed by Western blot.
Western blot analyses
Cells were washed in PBS and lysed in sample buffer containing
dithiothreitol. Lysates from 1 x 106 cells
were resolved by 10% SDS-PAGE under reducing conditions. Proteins were
transferred onto nitrocellulose membranes (Bio-Rad Laboratories, Inc., Hercules, CA). Membranes were incubated for 1 h in
TBS-T [50 mM Tris-HCl (pH 7.6), 200 mM NaCl,
and 0.1% Tween 20] with 2% BSA. Proteins were detected by overnight
incubation of membrane with specific antibody in TBS-T with 2% BSA and
subsequent incubation with a horseradish peroxidase-conjugated protein
G (Bio-Rad Laboratories, Inc.) for 1 h. Specific
protein bands were visualized using the enhanced chemiluminescence
detection system (NEN Life Science Products, Boston, MA)
according to the manufacturer’s instructions. The activation of AKT
was assessed by Western blot using the
anti-AKT-Ser473P antibody, which allows the
detection of phosphorylation Ser473
(11).
Transient transfections and luciferase assays
Cells (10 x 106) were transiently
transfected with 30 µg of the expression vectors by electroporation
at 330 V and 1500 µF in a CellJect apparatus (Eurogentec, Seraing,
Belgium). Transfected cells were cultured in growth medium overnight,
starved for 6 h, and subsequently stimulated with the appropriate
cytokine. For luciferase reporter assays, 30 µg NF-B-dependent
luciferase reporter plasmid were transfected into 10 x
106 cells before overnight incubation in normal
medium. Cells were then stimulated for 16 h, and total cell
extracts were prepared for assay of luciferase activity according to
the manufacturer’s instructions (Promega Corp. kit).
Results are expressed as the fold induction of luciferase activity
calculated under stimulation conditions compared with that under
starvation conditions.
Statistical analyses
Results are given as the mean ± SD for the
indicated number of independently performed experiments. A paired
t test was used to calculate differences between means; the
difference was considered significant at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). As expected, apoptosis was manifest
in Ba/F3 WT cells upon a short-term culture (24 h) in either starvation
medium alone or in the presence of GH (Fig. 1). IL-3 or IGF-I treatment
induced progression in S/G2/M phase of 64% and
45% of Ba/F3 WT cells, respectively (Fig. 1). The cytokines, however,
failed to induce cell cycle progression upon long-term culture;
apoptosis or necrosis was apparent in the total cell population on day
7 (Fig. 1). These observations indicate that IL-3 and IGF-I are unable
to maintain survival and growth over the long term in the absence of
serum.
|
). Likewise, IL-3 or IGF-I stimulation for
24 h induced a proliferative response in 60% and 36% of Ba/F3
GHR cells, respectively (Fig. 1). Ba/F3 GHR cells cultured in
starvation conditions for a long period lost their ability to survive,
and apoptosis appeared in 41% of the cell population on day 7 (Fig. 1). Under treatment with IGF-I, a low proportion (10%) of the cells
progressed through S/G2/M phase, whereas 37% of
cells underwent apoptosis on day 7 (Fig. 1). Also, 54% of Ba/F3 GHR
cells treated with IL-3 were in apoptosis on day 7 (Fig. 1). However,
these cells maintained a constant cell cycle when treated with GH for 7
days (Fig. 1). Together, these findings indicate that, in contrast to
IGF-I or IL-3, GH alone was able to promote long-term cell
survival.
GH action is independent of IGF-I
To investigate whether Ba/F3 GHR cells are able to produce IGF-I,
RNA levels of IGF-I were evaluated by RT-PCR using total RNA from Ba/F3
GHR cells either unstimulated or stimulated by GH for 3–24 h. Total
RNA from murine liver homogenate was used as the positive control.
Although, IGF-I messenger RNA (mRNA) was detectable in liver, we failed
to detect IGF-I transcript in unstimulated Ba/F3 GHR cells (data not
shown). Moreover, addition of GH did not enhance transcription of the
IGF-I gene (data not shown). IGF-I RIA in culture medium of Ba/F3 GHR
cells unstimulated or stimulated by GH for 24 and 48 h did not
reveal the presence of the IGF-I protein (data not shown). These
results support the hypothesis that Ba/F3 GHR cells, under the
conditions used, do not produce IGF-I even in the presence of GH.
We also explored the possibility that GH could exert part of its
biological effect in Ba/F3 GHR cells through local production of very
low IGF-I concentrations. The effects of GH were investigated during
blockage of IGF-I action by anti-IGF-I antibody. Cell cycle analyses
showed that the presence of the antibody did not affect cell cycle of
Ba/F3 GHR cells cultured in starvation medium (Fig. 2). The proliferative response induced by
IGF-I alone was completely abolished by addition of the anti-IGF-I
antibody in Ba/F3 GHR cells (Fig. 2). In contrast, when Ba/F3 GHR cells
were cotreated with GH and anti-IGF-I antibody, no difference from
cells treated with GH alone was observed (Fig. 2). These results
indicate that GH does not require IGF-I to induce cell survival and
proliferation, thus suggesting that the effects of GH are IGF-I
independent in Ba/F3 GHR cells.
|
, expression levels of cyclins
D1 and A were hardly detectable in Ba/F3 WT cells
maintained in starvation medium or treated with GH for 24 h (Fig. 3, lanes 1 and 4) or 48 h (Fig. 3, lanes 5 and 8). These low
cyclin levels correlated with those in cells undergoing apoptosis (Fig. 1). In contrast, expression of cyclin D1 was
greatly increased in these cells upon treatment with either IL-3 or
IGF-I for 24 h (Fig. 3, lanes 2 and 3) or for 48 h (Fig. 3, lanes 6 and 7). Cyclin A expression was also increased in Ba/F3 WT
cells stimulated with IL-3 or IGF-I for 24 h (Fig. 3, lanes 2 and
3). After 48 h, the stimulation of cyclin A expression by IL-3
appeared somewhat reduced compared with the effect of IGF-I, which was
able to maintain cyclin A levels (Fig. 3, lanes 2 vs. 6 and
3 vs. 7).
|
, lanes 9–16). Cyclin A expression levels were increased upon
treatment for 24 h with any of the three cytokines (Fig. 3, lanes
10–12) compared with those in unstimulated cells (Fig. 3, lane 9).
However, cyclin A levels were decreased in Ba/F3 GHR cells treated with
IL-3 for 48 h (Fig. 3, lane 10 vs. 14), whereas they
remained constant upon IGF-I or GH stimulation (Fig. 3, lanes 15 and
16).
Modulation of cyclin inhibitor p21waf1/cip1
expression by IGF-I, GH, and IL-3 was also studied in Ba/F3 WT and
Ba/F3 GHR cells. Expression levels of
p21waf1/cip1 were low in starved Ba/F3 WT or
Ba/F3 GHR cells (Fig. 3, lanes 1, 5, 9, and 13). An increase in
p21waf1/cip1 levels was observed in extracts
from Ba/F3 WT and Ba/F3 GHR cells stimulated with IL-3 and was
maintained after 48 h (Fig. 3, lanes 2, 6, 10, and 14). Expression
of p21waf1/cip1 was very low in both cell lines
after 24 h of IGF-I-stimulation (Fig. 3, lanes 3 and 11). However,
IGF-I induced a high expression of p21waf1/cip1
48 h poststimulation in Ba/F3 WT and Ba/F3 GHR cells (Fig. 3, lanes 7 and 15). According to previous data (26),
increased expression levels of the cyclin inhibitor were also found in
extracts from Ba/F3 GHR cells treated with GH for 24 and 48 h
(Fig. 3, lanes 12 and 16) and, as expected, not in extracts from Ba/F3
WT treated with GH (Fig. 3, lanes 4 and 8). Taken together, these
observations indicate that GH, IGF-I, and IL-3 act differently on
proteins governing cell cycle in Ba/F3 cells.
Is NF-B involved in IGF-I-induced cell survival?
The transcription factor NF-B has been shown to be crucial in
the signaling pathway used by IL-3 or GH to protect cells from
apoptosis (4, 17). To test the ability of IGF-I to
activate NF-B, Ba/F3 WT and Ba/F3 GHR cells were transiently
transfected with an NF-B-element driven luciferase construct; after
transfection, cells were treated with IGF-I, GH, or IL-3 for 16 h.
Minimum activity (considered 1-fold) was obtained in extracts from
unstimulated Ba/F3 WT cells (Fig. 4). In
extracts from starved Ba/F3 GHR cells, a 4-fold increase in relative
luciferase activity was measured (Fig. 4); this was previously reported
to be due to the locally produced GH (4, 26). A 14-fold
increase in luciferase activity was measured in extracts from both
IL-3-treated Ba/F3 WT and Ba/F3 GHR cells (Fig. 4). Consistent with our
recent findings (4, 26), extracts from Ba/F3 GHR cells
stimulated by GH showed a 9.5-fold increase in relative luciferase
activity, whereas no effect was detected in Ba/F3 WT cell extracts
(Fig. 4). Under similar conditions, IGF-I treatment did not
significantly modify luciferase activity (Fig. 4). These findings
suggest that, in contrast to GH or IL-3, IGF-I is not able to activate
the NF-B pathway in Ba/F3 cells.
|
B to promote a cellular
response in Ba/F3 cells, cell survival was studied after inhibition of
the NF-B pathway. Transient transfections of a
pRCßactin IB
(A32/36) vector encoding the dominant negative mutant IB (A32/36)
were performed in Ba/F3 GHR cells. The mutated IB (A32/36)
protein contains serine to alanine mutations in amino acids 32 and 36,
preventing its phosphorylation and subsequent degradation
(20). The empty parental
pRCßactin vector DNA was
similarly transfected as a control. Consistent with previous results
(4, 17), Ba/F3 GHR cells expressing the mutant IB
(A32/36) protein and maintained in the absence or presence of IL-3 or
GH showed 66–69% of cells undergoing apoptosis, whereas transfection
of the empty vector did not have any effect on these cells (Fig. 5). Expression of the mutant IB
(A32/36) protein had no effect on cell cycle after IGF-I treatment, as
evidenced by comparison to cells transfected with the empty vector
(Fig. 5). Inhibition of NF-B signaling did not affect the IGF-I
response. Thus, the antiapoptotic and proliferative effects of IGF-I do
not appear to require NF-B.
|
). Parallel
experiments were performed with IL-3 as the positive control. Ba/F3 WT
cells underwent apoptosis upon starvation in the presence of DMSO or Ly
294002, whereas Ba/F3 GHR cells were resistant to apoptosis under the
same conditions (Fig. 6). In agreement with previous studies (15, 16), pretreatment of Ba/F3 WT cells with Ly 294002 caused 96%
of cells to be arrested in
G0/G1 phase in the presence
of IL-3 (Fig. 6). The same effect was observed in Ba/F3 GHR cells
incubated in the presence of Ly 294002 and IL-3 (Fig. 6). Therefore, Ly
294002 treatment provoked an inhibition of the IL-3-induced
proliferation, but did not affect the IL-3-mediated cell survival. As
expected, treatment with exogenous GH did not affect Ba/F3 WT cell
cycle profiles, which appeared similar to those found in starvation
conditions, in the presence of DMSO or Ly 294002 (Fig. 6). As
previously shown (26), PI 3-kinase inactivation inhibited
the proliferative response induced by exogenous GH in Ba/F3 GHR cells
without affecting the survival pathway and resulting in 96% of cells
in growth arrest (Fig. 6). In contrast, Ly 294002 pretreatment
counteracted the IGF-I survival and proliferative responses, inducing
apoptosis in 42% of the Ba/F3 WT cell population (Fig. 6). However,
similar to IL-3 and GH, PI 3-kinase inhibition only blocked the
IGF-I proliferative response in Ba/F3 GHR cells, leaving intact the
survival pathway in these cells (Fig. 6).
|
). The antibody
anti-AKT-Ser473P, which is known to detect the
activated form of AKT, revealed the presence of traces of activated AKT
in starved Ba/F3 GHR cells (Fig. 7). The low activation levels were
increased under stimulation by the three cytokines (Fig. 7). In all
assays performed, GH as well as IL-3 induced maximum AKT activation
after 10-min stimulation, which was sustained throughout 30 min (Fig. 7). Interestingly, activation of AKT by IGF-I was maximal after
2-min stimulation and was decreased after 20 min. Thus, GH and IGF-I
were able to induce AKT activation in Ba/F3 GHR cells, although with
different kinetics.
|
). The extent of
apoptosis was increased in Ba/F3 WT cells in the presence of AH-AKT
protein under starvation or upon GH treatment (Fig. 8). Although IL-3
treatment was still able to rescue Ba/F3 WT cells from apoptosis in the
presence of AH-AKT, IGF-I was not (55% of cells underwent cell death;
Fig. 8). Therefore, AKT activation seemed to be crucial for
IGF-I-mediated survival in Ba/F3 WT cells.
|
). Indeed, cell cycles from cells
transfected with the mutant form AH-AKT showed profiles similar to
those of their counterparts transfected with the control vector upon
IL-3, IGF-I, or GH treatment (Fig. 8). These observations can be
explained by the fact that a survival pathway is sustained in Ba/F3 GHR
cells via locally produced GH (4), indicating that this
signaling pathway is AKT independent. In contrast, IGF-I seems to
require the PI 3-kinase/AKT pathway for its survival effects. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B in Ba/F3 cells; 2) IGF-I and GH are
both able to activate AKT, but with different kinetic patterns; and 3)
AKT activity is absolutely required for the cell survival mediated by
IGF-I in Ba/F3 cells, but is not necessary to promote GH-induced cell
survival. The idea that apoptosis is a consequence of a cytokine deprivation rather than of nutritional deprivation was pointed out previously (6), and it is now established that pro-B cells undergo apoptosis after cytokine deprivation even in the presence of serum. Our study demonstrates that addition to the culture medium of IGF-I or IL-3 in the absence of serum results in the maintenance of cell survival and proliferation, but only for short-term culture; apoptosis is evident on day 7 of culture. Thus, IGF-I and IL-3 need costimulatory signals, present in the serum, to promote cell viability and cell cycle progression of Ba/F3 cells. Moreover, GH seems to emulate costimulatory signals delivered by serum factors that appeared essential for IL-3 and IGF-I to induce complete responses. The ability of Ba/F3 GHR cells to survive for short-term culture was demonstrated recently to be due to locally produced GH (4). Together, these findings can explain why IGF-I as well as IL-3 are able to maintain survival of Ba/F3 GHR cells for a longer time than Ba/F3 WT cells. The present results further suggest that the ability of Ba/F3 GHR cells to enter the cell cycle upon cytokine stimulation is supported by signals mediated by endogenous GH.
Both direct and indirect effects of GH on immunocompetent cells have been reported (27, 28); indirect effects are mediated by locally produced IGF-I. Our data support a direct effect of GH on the survival and the proliferation of Ba/F3 GHR cells. With the use of an antibody able to block IGF-I action, we could show that the cell response to GH was not modified, suggesting that GH does not need IGF-I to exert a mitogenic effect in Ba/F3 GHR cells. Moreover, our data suggest that IGF-I mRNA is not transcribed in Ba/F3 GHR cells, and consistently that IGF-I is not produced by the cells.
Analysis of the expression of proteins that govern the cell cycle indicates that IGF-I and GH act differently on cell survival and proliferation. Cell cycle is driven by sequential expression of cyclins according to the phase of the cell cycle that, in turn, are controlled by cyclin-dependent kinases (CDK). Activation of CDK inhibitors (CKI) inactivates cyclin-CDK complexes, which control cell cycle entry (29). We show that no differences are found in the kinetics of cyclin D1 and A expression induced by GH and IGF-I. However, although both GH and IGF-I induce p21waf1/cip1 CKI in cycling cells, the ability of IGF-I to induce CKI expression appeared to be delayed. The p21waf1/cip1 protein belongs to a CKI family and was described as an effector of cellular quiescence, arresting cells at G1/S phase transition in a wide variety of signaling contexts (29). Nevertheless, increasing evidence is accumulating about the role of the protein in controlling not only G0/G1 phase entry, but also the onset of G2/M phase transition (30). Indeed, p21waf1/cip1 seems to inactivate the cyclin A-CDK2 complex, thus regulating the G2/M phase transition (30). The presence of p21waf1/cip1 in cycling cells could represent a mechanism to attenuate a strong proliferative response induced by a mitogen. Actually, other cytokines, such as IL-2, also induce p21waf1/cip1 expression in cycling cells while inducing cell proliferation (31). The differences in p21waf1/cip1 expression kinetics induced by IGF-I and GH suggest a divergence in signaling pathways for the two hormones to control cell cycle progression.
Of note, the pattern of expression of cyclins also revealed differences
between GH- and IL-3-induced effects. In Ba/F3 cells, GH and IL-3 were
reported to promote their survival signal through NF-B activation
and their proliferative effect essentially through PI 3-kinase
activation (4, 15, 16, 17, 26). Thus, it was expected that
both cytokines would act similarly on Ba/F3 GHR cells. Indeed, GH can
take over from IL-3 dependence in Ba/F3 GHR cells growing in
serum-containing medium (5). However, as shown here, a
difference between GH and IL-3 cellular responses is observed in Ba/F3
GHR cells stimulated over the long term in serum-free conditions. Up-
regulation of p21waf1/cip1 protein coupled
to a small decrease in cyclin A expression after 48 h of IL-3
stimulation would prompt cells to growth arrest. In contrast, exogenous
GH promoted sustained cyclin A levels; this is probably associated with
the ability of Ba/F3 GHR cells to stay in cell cycle and therefore to
maintain cell culture for a longer time. From these observations it
cannot be excluded that GH induces its proliferative effect through a
different signaling pathway than that of IL-3.
It was recently demonstrated that the GH antiapoptotic effect is
transduced through the NF-B pathway (4). Evidence is
now given that NF-B is not required for IGF-I-induced cell survival
and proliferation of Ba/F3 cells; overexpression of the mutated
IB (A32/36), which is known to inhibit NF-B activity
(20), did not affect the IGF-I-mediated responses, whereas
it was shown to result in death of GH- or IL-3-stimulated Ba/F3 GHR
cells (4, 17). Our results suggest that IGF-I depends on
the PI 3-kinase pathway to deliver both survival and proliferative
effect in Ba/F3 cells. Interestingly, IGF-I was shown to activate
NF-B in neurons, and this activation was reported to be dependent on
PI 3-kinase (32). In Ba/F3 cells, NF-B activation by GH
was also reported to be partially dependent on PI 3-kinase
(26). Nevertheless, our results demonstrate that IGF-I is
not able to activate NF-B in Ba/F3 cells and requires the PI
3-kinase/AKT pathway for its survival effect. Thus, the PI 3-kinase
pathway seems to be crucial for the IGF-I survival effect, although
downstream signaling molecules are different depending on the cell
type. Also, recent data show that induction of
p21waf1/cip1 by TNF requires NF-B
activation (33). Therefore, the rapid increase in
p21waf1/cip1 observed under IL-3 or GH
stimulation could be related to their ability to activate the NF-B
pathway. The fact that IGF-I is not able to activate NF-B in these
cells could be linked to its delayed stimulation of
p21waf1/cip1 expression levels observed in Ba/F3
cells.
We have shown that PI 3-kinase is necessary for the mitogenic action of
GH in Ba/F3 GHR cells (26), whereas PI 3-kinase seems to
be required for both antiapoptotic and proliferative effects of IGF-I.
The involvement of AKT in IGF-I- and GH-mediated survival signals was
reported in several cellular models, such as fibroblasts, neurons, and
CHO cells (13, 34, 35). Here, we show that the PI
3-kinase/AKT pathway is crucial for the cell survival induced by IGF-I,
but does not appear to be involved in the antiapoptotic effect of GH.
Our results indicate that although inactivation of AKT accelerates the
apoptosis rate of starved Ba/F3 WT cells, it does not affect starved
Ba/F3 GHR cell viability. This difference could be explained by the
locally produced GH, which was reported to sustain survival of Ba/F3
GHR cells under starvation conditions (4). Moreover, the
effect of GH on cell survival was shown to be dependent on NF-B
signaling (4), and from these results does not seem to be
dependent on AKT activation. Similarly, overexpression of the mutant
form of AKT (AH-AKT) does not inhibit IGF-I effects in Ba/F3 GHR cells;
it is probably due to the local production of GH, which is responsible
for Ba/F3 GHR cell survival. In contrast, the effect of IGF-I on Ba/F3
WT cell survival is inhibited when AH-AKT is overexpressed, strongly
suggesting that AKT is required for IGF-I-induced Ba/F3 cell survival.
It can be concluded that IGF-I and GH transduce their survival effects
through two different signaling pathways, PI 3-kinase/AKT and NF-B,
respectively.
In conclusion, using the pro-B murine Ba/F3 cell model, we show that GH action is not mediated via IGF-I, and that the two hormones do not use the same molecules to signal their effects on the cell cycle. Our findings suggest that GH, IGF-I, as well as IL-3 can use individual pathways to deliver a similar final response in the same cellular model. Thus, these results support the hypothesis that the requirement of the PI 3-kinase/AKT pathway to protect hemopoietic progenitors from cell suicide commitment is dependent upon the extracellular signal (36).
|
Acknowledgments |
|---|
B (A32/36) and AH-AKT expression vectors, respectively. G.
Rawadi is gratefully acknowledged for the gift of NF-B-luciferase
construct. We thank INSERM U-373 for the use of the FACS, and C. Garcia
for technical assistance with the FACS analyses. We are very grateful
to L. Perin and Y. Le Bouc (INSERM U-515) for IGF-I assay, and to G.
Sonenshein for fruitful discussions and help in the preparation of the
manuscript. |
Footnotes |
|---|
2 E.B. and S.J. contributed equally to this work.
3 Current address: Department of Medicine and Liver Unit, Medical
School, University of Navarra, 31 008 Pamplona, Spain.
Received November 30, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B in interleukin-3-dependent Ba/F3 cell line. Mol Endocrinol 14:650–661B in the survival of
hematopoietic cells mediated by interleukin 3 and the oncogenic
TEL/platelet-derived growth factor receptor beta fusion protein. Proc
Natl Acad Sci USA 95:8081–8086B
phosphorylation sites that signal its ubiquitination and degradation.
Mol Cell Biol 16:1295–1304[Abstract]
B- on serines
32 and 36 controls IB- proteolysis and NF-B activation in
response to diverse stimuli. EMBO J 14:2876–2883[Medline]
B and phosphatidylinositol 3-kinase. Endocrinology 142:147–156B. J
Biol Chem 274:9828–9835 requires NF-B activity and
antagonizes apoptosis in Ewing tumor cells. Oncogene 19:61–68[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  F. Bogazzi, D. Russo, F. Raggi, F. Ultimieri, C. Urbani, M. Gasperi, L. Bartalena, and E. Martino Transgenic Mice Overexpressing Growth Hormone (GH) Have Reduced or Increased Cardiac Apoptosis through Activation of Multiple GH-Dependent or -Independent Cell Death Pathways Endocrinology, November 1, 2008; 149(11): 5758 - 5769. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Barclay, S. T. Anderson, M. J. Waters, and J. D. Curlewis Regulation of Suppressor of Cytokine Signaling 3 (SOC3) by Growth Hormone in Pro-B Cells Mol. Endocrinol., October 1, 2007; 21(10): 2503 - 2515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. R. Reddy, M. J. Pushpanathan, R. F. Ransom, L. B. Holzman, F. C. Brosius III, M. Diakonova, P. Mathieson, M. A. Saleem, E. O. List, J. J. Kopchick, et al. Identification of the Glomerular Podocyte as a Target for Growth Hormone Action Endocrinology, May 1, 2007; 148(5): 2045 - 2055. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T.-B. Liu, P. W. M. Fedak, R. D. Weisel, T. Yasuda, G. Kiani, D. A. G. Mickle, Z.-Q. Jia, and R.-K. Li Enhanced IGF-1 expression improves smooth muscle cell engraftment after cell transplantation Am J Physiol Heart Circ Physiol, December 1, 2004; 287(6): H2840 - H2849. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Bogazzi, F. Ultimieri, F. Raggi, D. Russo, R. Vanacore, C. Guida, S. Brogioni, C. Cosci, M. Gasperi, L. Bartalena, et al. Growth Hormone Inhibits Apoptosis in Human Colonic Cancer Cell Lines: Antagonistic Effects of Peroxisome Proliferator Activated Receptor-{gamma} Ligands Endocrinology, July 1, 2004; 145(7): 3353 - 3362. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Dalla Libera, B. Ravara, M. Volterrani, V. Gobbo, M. Della Barbera, A. Angelini, D. D. Betto, E. Germinario, and G. Vescovo Beneficial effects of GH/IGF-1 on skeletal muscle atrophy and function in experimental heart failure Am J Physiol Cell Physiol, January 1, 2004; 286(1): C138 - C144. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Frago, C. Paneda, S. L. Dickson, A. K. Hewson, J. Argente, and J. A. Chowen Growth Hormone (GH) and GH-Releasing Peptide-6 Increase Brain Insulin-Like Growth Factor-I Expression and Activate Intracellular Signaling Pathways Involved in Neuroprotection Endocrinology, October 1, 2002; 143(10): 4113 - 4122. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.-H. Shen, J.-H. Zhou, S. R. Broussard, G. G. Freund, R. Dantzer, and K. W. Kelley Proinflammatory Cytokines Block Growth of Breast Cancer Cells by Impairing Signals from a Growth Factor Receptor Cancer Res., August 15, 2002; 62(16): 4746 - 4756. [Abstract] [Full Text] [PDF]
|
|
| ||||||||
) or Ba/F3 GHR (
) cells (10 x
106) were transiently transfected with a NF-
