Journal of Cancer Therapy, 2013, 4, 1290-1297
http://dx.doi.org/10.4236/jct.2013.48152 Published Online October 2013 (http://www.scirp.org/journal/jct)
Signal Transduction Pathways Mediated by Secreted and
Non-Secreted Forms of Intact Insulin-Like Growth Factor
Binding Protein-3 (IGFBP-3) and Its 1-97 N-Terminal
Fragment in PC-3 Human Prostate Cancer Cells*
Hanief Mohammad Shahjee1#, Benjamin Kefas2, Nisan Bhattacharyya1, Mohamed K. Radwan3
1Diabetes Branch, NIDDK, National Institutes of Health, Bethesda, USA; 2Division of Neuro-Oncology, Neurology Department,
University of Virginia Health System, Charlottesville, USA; 3Department of Biochemistry and Biophysics, University of Rochester
Medical Center, Rochester, USA.
Received August 6th, 2013; revised September 3rd, 2013; accepted September 10th, 2013
Copyright © 2013 Hanief Mohammad Shahjee et al. This is an open access article distributed under the Creative Commons Attribu-
tion License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
ABSTRACT
Our previous results indicated that both the secreted and the intracellular form of full length and 1-97 N-terminal frag-
ment of IGFBP-3 induce apoptosis in PC-3 human prostate cancer cells in an IGF-dependent and independent manner.
This study was undertaken to delineate possible down-stream signaling pathways that are involved in this process. In-
tact IGFBP-3 and its N-terminal 1-97 fragments with or without a signal propeptide were fused to YFP and expressed in
PC-3 human prostate cancer cells. In some cases, the putative IGF-binding site was presented in full length IGFBP-3
and its N-terminal fragment was also mutated. Extent of apoptosis was quantified using FACS. Up-regulation of total
Stat-1 and activation of phospho-Stat-1 were shown by western blot. TGF-β signal was measured by luciferase reporter
assay. Results from inhibitor studies indicated that both the Caspase 8 and caspase 9 pathways are involved in IGFBP-3
(non-secreted form) which induced apoptosis in PC-3 cells. Exogenous addition of IGFBP-3 to PC-3 cells increased
Stat-1 protein expression/tyrosine phosphorylation. Interestingly, results also showed that knockdown of Stat-1 by
siRNA potentiated the IGFBP-3 induced apoptosis in PC-3 cells. In addition, both full-length IGFBP-3 and its 1-97 N-
terminal fragments inhibited TGF-β signaling in these cells. This is the first report that compares the signal transduction
pathways involved in apoptotic pathways mediated by IGFBP-3 in PC-3 human prostate cancer cells. Non-secreted
form of full length IGFBP-3 and its N-terminal fragments induced apoptosis in PC-3 cells via activation of caspase 8
and caspase 9. Although, only non-secreted form of IGFBP-3 is involved in inducing apoptosis in PC-3 cells via cas-
pase 8 and caspase 9 activation pathways but both secreted and non-secreted forms of IGFBP-3 are involved in modu-
lating Stat-1 and TGF-β pathways to induce apoptotic actions in PC-3 cells. Non-secreted intact IGFBP-3 and its
N-terminal fragments induced apoptosis in PC-3 cells via activation of caspase 8 and caspase 9 pathways. Modulation
in STAT-1 and TGF-β pathways may also be important for IGFBP-3 induced apoptosis in PC-3 cells in general. These
studies clearly demonstrate that secreted and non-secreted FL and 1-97 N-terminal fragments induce apoptosis in PC-3
cells by regulating different mechanistic pathways.
Keywords: N-Terminal Fragment; Apoptosis; Caspases; Human Prostate Cancer Cells
1. Introduction
Insulin-like growth factor binding protein-3 (IGFBP-3),
the most abundant protein of the six IGFBPs, is an im-
portant modulator of insulin-like growth factor (IGF) bio-
logical activity. IGFBP-3 binds to IGF-I and IGF-II with
high affinities and restricts their bioavailability by form-
ing stable ternary complexes with an acid-labile subunit.
These complexes have been reported to be stable in plas-
ma [1]. These complexes play major roles by preventing
the IGFs from activating IGF-I receptors on target cells
*This work was supported by an intramural grant from NIDDK, NIH.
Authors’ contributions: HMS carried out all experiments for these
studies. NB supervised the research and inter-pretation of the data was
carried out by HMS and NB. HMS, BK, NB and MKR made the final
draft for the paper.
#Corresponding author.

Signal Transduction Pathways Mediated by Secreted and Non-Secreted Forms of Intact Insulin-Like Growth Factor
Binding Protein-3 (IGFBP-3) and Its 1-97 N-Terminal Fragment in PC-3 Human Prostate Cancer Cells
1291
and stimulating cell proliferation and survival [2,3]. Sev-
eral reports have indicated that IGFBP-3 induces cell
death and inhibits cell proliferation in various cell types,
including prostate cancer cells (PC-3 cells), and it’s in-
dependent of its IGF-I binding [4-6]. IGFBP-3 is a se-
creted protein but it can be internalized to the nucleus
when it’s added to cells. Other reports have indicated that
the presence of IGFBP-3 in the cytoplasm [5,7] or in the
mitochondria [8] can induce cell death. Although the
IGFBP-3 receptor has not been identified, several signal
transduction mechanisms responsible for biological ac-
tions of IGFBP-3 have been implicated [9-12]. IGFBP-
3’s availability to cells and tissues is also regulated by
proteolysis. Proteolysis of IGFBP-3 protein was initially
demonstrated in serum of pregnant women, where it cir-
culates primarily in low molecular weight forms [13].
These fragments bind to IGF with lower affinities, there-
by increasing the limited availability of IGF to target re-
ceptors. Interestingly, various IGFBP-3 fragments have
been reported to mediate direct stimulatory or inhibitory
actions at the target cells [14].
Results from our previous study have demonstrated
that when full-length IGFBP-3 is added exogenously to
PC-3 cells, it can be processed into a small N-terminal
fragment (amino acids 1-97). Our results also demon-
strated that expression of 1-97 N-terminal IGFBP-3
fragments induces apoptosis in these cells [15]. Irrespec-
tive of the presence of a signal peptide, expression of
either the wild-type intact IGFBP-3 and N-terminal 1-97
fragment or an IGF-nonbinding mutant 6m (mutations in
six IGF-binding amino acid residues) fusion proteins
caused a loss of cell viability indicative of apoptosis.
However, the signal transduction mechanisms specific to
either the full-length or the 1-97 N-terminal fragment of
IGFBP-3 have not been studied.
In this study, not only we confirmed the involvement
of secreted as well as non-secreted forms of IGFBP-3 in
inducing apoptosis in PC-3 cells, but also we studied the
possible signaling pathways modulated by these various
forms of IGFBP-3 during the process. We examined the
induction of apoptosis in the presence or absence of cas-
pase-8 and -9 inhibitors. Surprisingly, we found that in-
tracellular form of intact and 1-97 N-terminal fragments
of IGFBP-3 (wild-type or mutant) could induce apoptosis
in a caspase-8 and -9 dependent manner in comparison to
secreted form of IGFBP-3. Our results also demonstrate
that exogenous addition of IGFBP-3 up-regulates Stat-1
protein levels and its tyrosine phosphorylation in a time-
dependent fashion. To better understand the role of
STAT-1 in IGFBP-3 induced apoptosis in PC-3 cells, we
knocked down the Stat-1 expression by using STAT-1
siRNA technique in PC-3 cells. In contrast to a previous
report [10], our results (Figure 1) showed that STAT1
siRNA treatment further potentiates IGFBP-3 (secreted
(a)
(b)
Figure 1. siRNA knockdown of Stat-1 protein in PC-3 cells.
(a) Stat-1 protein expression levels were reduced after
transfection with Stat-1 siRNA. PC-3 cells were transfected
with no DNA (Untreated; Lane 1), scrambled siRNA (Lane
2), non-target (NT) siRNA (lane 3), and Stat-1 siRNA (Lane
4) for 24, 48, and 72 h. Whole cell extracts were prepared
and western blot analyses were performed to examine the
Stat-1 protein expression levels by using an anti-human
Stat-1 antibody. Stat-1 protein expression level is decreased
after 48 and 72 h of transfection. Blot is representative re-
sult of three independent experiments. Transfection of Stat-
1 siRNA potentiates IGFBP-3-mediated apoptosis in PC-3
cells. (b) PC-3 cells were first transfected with either scram-
bled (open bars) or Stat-1 siRNA (black bars) for 24 h and
subsequently with various constructs expressing the full-
length and/or N-terminal fragments of IGFBP-3 protein for
another 48 h (total 72 h). Results from FACS analyses are
shown. Percentages of non-viable cells are indicated. Each
experiment was performed in duplicate on three separate
occasions. Data is expressed as mean ± SEM.
and non-secreted), which induced apoptosis in these cells,
thereby suggesting that Stat-1 may be an important target
molecule for IGFBP-3, which induced PC-3 cell death. In
addition, when we transfected a PC-3 stable cell line ex-
pressing CAGA-TGF-β reporter luciferase gene with vari-
ous IGFBP-3 constructs (Figure 2), we observed that
both secreted as well as non-secreted forms of IGFBP-3
constructs inhibited TGF-β signal in these cells, thereby
providing evidence that IGFBP-3 also exploits the TGF-β
signaling pathways for its apoptotic action in PC-3 cells.
2. Materials and Methods
2.1. Materials
The caspase 8-selective inhibitor Z-IETD-fmk, the cas-
pase 9-selective inhibitor Z-LEHD-fmk, and annexinV-

Signal Transduction Pathways Mediated by Secreted and Non-Secreted Forms of Intact Insulin-Like Growth Factor
Binding Protein-3 (IGFBP-3) and Its 1-97 N-Terminal Fragment in PC-3 Human Prostate Cancer Cells
1292
Figure 2. Transient transfection of IGFBP-3 constructs
reduces TGFβ signaling in PC-3 cells. PC-3 cell-line stably
expressing (CAGA) 12 tkluc (a TGFβ-responsive promoter
construct) was used for this experiment. These cells were
transfected with the indicated IGFBP-3 constructs and were
subsequently treated with TGFβ (10 ng/ml) or carrier for 24
h. Dual luciferase (renilla luciferase was used as the trans-
fection control) assays were performed. Fold activation
(luciferase activity) values (+/−TGFβ added) obtained for
each sample are shown. There is decreased TGFβ signaling
in cells transfected with constructs that are expressing in-
tracellular full-length and N-terminal 1-97 IGFBP-3. Each
experiment was repeated three times in triplicates. Data is
expressed as mean ± SEM.
APC were purchased from BD Pharmingen (San Diego,
CA). Recombinant glycosylated human IGFBP-3 was ob-
tained from R & D Systems (Minneapolis, MN). Lipo-
fectamine plus transfection reagent was obtained from Invi-
trogen (Carlsbad, CA). TGF-β and mouse β-actin anti-
body were purchased from Sigma Aldrich (St Louis, MO).
Stat-1 siRNA was purchased from Dharmacon (Lafayette,
CO). Rabbit anti-human antibodies against total and ty-
rosine phosphorylated (Y701) Stat-1 proteins were ob-
tained from Cell Signaling Technology (Danvers, MA).
2.2. Transfection
PC-3 cells (1.5 × 105 cells/6-well) were grown to 70% -
80% confluence in F12K medium containing 10% fetal
bovine serum (FBS). Cells were transfected (37˚C, 3 h)
in 1 ml of serum-free medium containing 500 ng of dif-
ferent plasmid constructs using Lipofectamine Plus (In-
vitrogen). The cells were recovered by adding 1.7 ml of
F12K medium containing 0.3 ml of FBS and were incu-
bated at 37˚C for 48 h unless otherwise specified.
2.3. Whole Cell Extract Preparation
PC-3 cells were incubated with noted concentrations of
recombinant IGFBP-3 (R & D Systems, Minneapolis,
MN) for indicated periods. After incubation, cells were
washed with phosphate-buffered saline (PBS) and were
lysed with 100 µl per 20-µl cellpellet of whole cell ex-
tract buffer (10 mM HEPES, pH 7.4, 10% glycerol, 250
mM NaCl, 1 mM EDTA, 0.1% Nonidet P-40, 1 mM di-
thiothreitol, and 1× protease inhibitor mixture) at 4˚C C
for 20 min. Lysates were vortexedat high speed for 15 s
and placed on ice for 5 - 10 min. After centrifugation
(12,000 rpm, 5 min, 4˚C), whole cell extracts (superna-
tants) were stored at −70˚C.
2.4. Western Blot
Protein concentrations were measured using the DC pro-
tein assay (Bio-Rad). Proteins were fractionated using
4% - 20% SDS-PAGE (Bio-Rad) under reducing condi-
tions. For whole cell extracts, 50 µg of protein was load-
ed per lane. Specific proteins were identified by western
blotting using Rabbit anti-Stat-1 antibody or Rabbit pStat-
1 antibody (Cell Signaling Technology, Danvers, MA).
Equal loading of the protein was confirmed by using
mouse anti-human β-actin antibody (Sigma, St. Louis, MO).
Protein bands were detected by using SuperSignal West
Pico Chemiluminescent Substrate (Pierce, Rockford, IL).
2.5. siRNA Transfection
PC-3 cells were plated in antibiotic-free F12K medium
(1.0 × 105 cells/ml for 12-well plate) for transfection.
Cells were transfected with either Stat-1, non-targetor
scrambled siRNA following a procedure provided by the
supplier (Dharmacon). Cells were incubated for 24 - 96 h
for protein analysis. For cell-death analysis experiments,
cells were first transfected with STAT-1 siRNA or
scrambled siRNA for 24 h and then were transfected with
the indicated IGFBP-3 constructs in serum-free media for
3 h and incubated in serum containing media for 48 h for
FACS analysis.
2.6. Fluorescence Activated Cell Sorting
(FACS) Analysis
After transfecting PC-3 cells with IGFBP-3 constructs
for 3 hrs, 20 µM each of caspase 8 and caspase 9 inhibi-
tor was added before cells were incubated at 37˚C in se-
rum containing media for recovery. For analysis of apop-
tosis, both attached and floating cells were collected. The
cell pellets were washed with 1× binding buffer (10 mM
HEPES, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2) (BD
Pharmingen, San Diego, CA) and were stained using 300
- 500 µl of binding buffer containing 5 µl of annexin
V-APC at room temperature for 15 - 30 min. Cells were
analyzed using a CyAnLX flow cytometer equipped with
Summit software (DakoCytomation, Fort Collins, CO).
2.7. Luciferase Reporter Assay
PC-3 cells were transfected with the TGF-β-responsive
reporter construct (CAGA) 12 tkluc plasmid DNA for 3
hrs. Cells were recovered in serum-containing media and
cells were incubated for overnight at 37˚C. Selective
marker G418 was added and cells that stably expressed
this construct were selected after 3 weeks of incubation
with G418. One of the stable cell (CAGA 11), was tran-

Signal Transduction Pathways Mediated by Secreted and Non-Secreted Forms of Intact Insulin-Like Growth Factor
Binding Protein-3 (IGFBP-3) and Its 1-97 N-Terminal Fragment in PC-3 Human Prostate Cancer Cells
1293
siently transfected with YFPN1, PrewtFL, Prewt (1-97),
Pre6m (1-97), WtFL, Wt (1-97), and 6m (1-97) along
with Renilla luciferase construct (transfection control) for
3 h followed by the recovery in 10% serum containing
F12K media for 24 h. Cells were washed and TGF-β (10
ng/ml) was added to fresh serum-free media overnight at
37˚C. Cells were washed, lysed and luciferase assay was
performed using a dual luciferase assay system (Promega,
Madison, WI). Luciferase activities were normalized by
using renilla luciferase amounts present in each extract.
2.8. Statistical Analysis
All data were presented as means ± standard deviation
(SD). The significance of the difference between mean
values was determined by an analysis of variance with p
< 0.05 considered significant
3. Results
3.1. Addition of Inhibitors Specific for Caspase-8
and Caspase-9 Reduces IGFBP-3-Induced
Apoptosis in PC-3 Cells
Results from our previous study demonstrated that the
expression of various IGFBP-3 fusion protein constructs
induced apoptosis in PC-3 cells. [5,15]. In order to un-
derstand the mechanism(s) by which full-length IGFBP-
3 and its N-terminal 1-97 fragment induce apoptosis, we
transfected PC-3 cells with various IGFBP-3 constructs
with or without a signal peptide. Apoptosis was deter-
mined in untreated cells (red bars) and in the presence of
20 μm of caspase-8 (IETD; yellow bars) and 20 μm of
caspase-9 (LEHD, green bars) specific inhibitors. The
number of apoptotic cells was determined by Annexin-V
staining (FACS). Results indicated that the addition of
either caspase 8 or 9 inhibitor significantly inhibited
IGFBP-3-induced apoptosis in PC-3 cells (Figure 3).
These results clearly shows that Intracellular form of full
length and 1-97 N-terminal fragments of WT and 6m
IGFBP-3 induce apoptosis in PC-3 cancer cells using
both caspase-8 and caspase-9 pathway which is not the
case with secreted form of Intact and 1-97 N-terminal
fragment of IGFBP-3. These results also confirm that
IGFBP-3 does not need to depend on the availability of
IGF or be secreted to induce apoptosis in PC-3 cells and
interestingly secreted form of IGFBP-3 is not involved in
inducing apoptosis via caspase 8 and caspase 9 activation
pathways.
3.2. Addition of Exogenous IGFBP-3 Increases
the Expression of Stat-1 Protein Level and
Its Tyrosine Phosphorylation (Y701)
in PC-3 Cells
A previous report [10] has shown that the addition of
Figure 3. IGFBP-3 induced apoptosis in PC-3 cells is inhib-
ited by inhibitors specific for Caspase-8 and Caspase-9.
PC-3 cells were transfected with plasmid constructs direct-
ing the expression of either the full-length IGFBP-3 or its
N-terminal 1-97 fragments. Cells were subsequently treated
with either DMSO (vehicle; Red bars) or 10 μM IETD
(caspase-8 inhibitor; yellow bars) and LEHD (caspase-9
inhibitor; green bars) for 48 h. Results from FACS analyses
(Annexin-V+ cells) are shown. Percentages of non-viable
cells are indicated. Figure 3 showed significant inhibition of
IGFBP-3 induced apoptosis in PC-3 cells in the presence of
caspase-8 and/or caspase-9 inhibitors. Experiments were
performed in triplicate twice and error bars represent mean
± SEM.
IGFBP-3 induces apoptosis and differentiation via STAT-
1 in rat chondroprogenitor cells. In order to investigate
the roles of STAT-1 signaling pathways in IGFBP-3 in-
duced apoptosis in PC-3 cells, we incubated the cells
with different concentrations of exogenous recombinant
IGFBP-3. Expression of total Stat-1 levels in PC-3 cells
were increased in a dose-dependent manner, peaking
after 50 ng/ml of IGFBP-3 (Figure 4(a) ) addition. There
was also a transient increase in tyrosine-phosphorylated
Stat-1 (pY701 STAT-1) levels within 10 min of incuba-
tion with 50 ng of IGFBP-3 (Figure 4(b)). These results
demonstrate that addition of IGFBP-3 can elevate the
expression levels as well as pY701 Stat-1 in PC-3 cells
indicating the involvement of Stat-1 signaling pathways.
3.3. Introduction of Stat-1 siRNA Potentiates the
IGFBP-3-Induced Apoptosis in PC-3 Cells
In order to better understand the roles of STAT-1 signal
pathways in IGFBP-3 induced apoptosis in PC-3 cells,
we first tested the effectiveness of a siRNA to block the
expression of Stat-1 in these cells. Western blot analysis
showed marked decrease in Stat-1 protein levels after 48
and 72 h of siRNA transfection in PC-3 cells (lanes
marked as Stat-1siRNA) compared to cells with no
siRNA treatment or transfected either with a scrambled
siRNA or a non-target siRNA (Figure 1(a)).There was
also a significant decrease in Stat-1 mRNA levels after
24 hrs of siRNA transfection in PC-3 cells (data not shown

Signal Transduction Pathways Mediated by Secreted and Non-Secreted Forms of Intact Insulin-Like Growth Factor
Binding Protein-3 (IGFBP-3) and Its 1-97 N-Terminal Fragment in PC-3 Human Prostate Cancer Cells
1294
(a)
(b)
Figure 4. Addition of exogenous IGFBP-3 induced Stat-1
protein expression as well as its tyrosine phosphorylation
(pY701) in PC-3 cells. (a) PC-3 cells were treated with ex-
ogenous recombinant IGFBP-3 (in ng/ml are indicated be-
low each lane) for 1 h. Whole cell extracts were prepared
and a western blot analysis was performed. Anti-human
Stat-1 antibody was used to examine the Stat-1 protein (91
kDa) (upper panel) and β-actin (42 kDa) (lower panel) ex-
pression levels in each sample. There is a dose-dependent
increase of Stat-1 protein expression level in the presence of
exogenously added recombinant IGFBP-3. This data is a
representative result of three independent experiments. (b)
IGFBP-3 treatment causes transient increase in Stat-1 tyro-
sine phosphorylation (pY701) PC-3 cells were treated with
50 ng/ml of IGFBP-3 for indicated periods of time. Whole
cell extracts were prepared and western blot was performed
by using an antibody against pY701 STAT-1(upper panel;
indicated by a 91 kDa) (upper panel) and unmodified
STAT-1 (middle panel, 91 kDA). The experiment was re-
peated three times. The blot was stripped and was reused to
detect for the β-actin (lower panel; indicated by a 42 kDa
protein) level using an anti-human β-actin antibody and
that served as the loading control. Blot is a representative
result of three independent experiments.
shown) Secondly, cell viability was determined in PC
cells transfected first with either STAT-1 siRNA (black
bars) or a scrambled siRNA (white bars) and then with
full length and various 1-97 N-terminal IGFBP-3 con-
structs. FACS analysis results showed that the IGFBP-
3-induced apoptosis was potentiated in cells transfected
with Stat-1 siRNA (Figure 1(b)), a result that is in con-
tradiction with a previous observation [10] but this report
confirms that Stat-1 plays an important role in IGFBP-3
induced apoptosis in PC-3 cells.
3.4. Expression of Intracellular IGFBP-3
Constructs Inhibits TGF-β SIgnaling
in PC-3 Cells
TGF-β has been implicated in several cellular processes
including apoptosis and proliferation [16,17]. In order to
determine and delineate the involvements of TGF-β sig-
naling pathways, we compared the effects of intact
IGFBP-3 or its N-terminal 1-97-IGFBP-3-YFP fragments
in PC-3 cells. First, several PC-3 cell-lines that stably ex-
pressed (CAGA) 12 tkluc construct, a TGF-β-responsive
luciferase reporter system, were created. One of these
clonal cell-lines was selected for transfection with vari-
ous constructs (WTFL IGFBP-3/1-97 N-terminal frag-
ment and 6m FL IGFBP-3/ 1-97 N-terminal fragment)
that express IGFBP-3 with or without its signal peptide.
Extent of TGF-β signals were measured by luciferase
output with without the addition of TGF-β. There was
significant inhibitions of TGF-β-mediated (CAGA) 12
luciferase activities in cells transfected with YFP-WT-
FL-IGFBP-3, YFP-WT-1-97-IGFBP-3, or Pre-WT-1-97-
IGFBP-3-YFP in comparison to cells transfected with
YFP empty vector or untransfected control (Figure 2).
Similar results were observed with YFP-6m-1-97-IGFBP-
3 and Pre-6m-1-97-IGFBP-3-YFP in which the IGF-bind-
ing site had been mutated, indicating that the inhibition
of TGF-β signaling occurs through an IGF-independent
mechanism. Interestingly, cells transfected with Pre-WT-
FL-IGFBP-3-YFP had a little effect. Whole cell extracts
and media collected from these transfected cells were
examined and equivalent expression of IGFBP-3 fusion
protein was observed in each sample (data not shown).
Full-length IGFBP-3 as well as its 1-97 N-terminal frag-
ment that expressed only intracellular protein inhibited
TGF-β signaling in PC-3 cells. This indicates that TGF-β
signaling pathways are involved in cellular actions of
IGFBP-3.
4. Discussion
IGFBP-3, the most abundant IGFBP in human serum,
acts not only as a carrier of IGFs prolonging their half-
lives in circulation, but also functions as a modulator of
IGF activity by regulating their availability to interact
with IGF receptors [18]. However, there is increasing
evidence that IGFBP-3 may have its own biological ac-
tions. For example, IGFBP-3 can inhibit cell proliferation
in an IGF-independent manner [4,19,20]. Conceptually,
IGFBP-3 can exert its actions on target cells by: 1) in-
ducing apoptosis, 2) regulating cell cycle and prolifera-
tion, and 3) inducing cross-talk with major signal trans-
duction pathways. Mechanisms for inducing apoptosis, in
part, have been demonstrated by studies showing that
IGFBP-3 increases the ratio of proapoptotic to antiapop-
totic proteins in breast cancer cells [21]. As a regulator of
the cell cycle, IGFBP-3 has been shown to modulate the
induction of the cyclin-dependent kinase inhibitory pro-
tein p21/WAF/CIP1 in LNCaP prostate cancer cells [22].
Previous studies demonstrated that caspase-8 can acti-
vate caspase-3 and -7 directly and/or through triggering
the activation of caspase-9 by the release of mitochon-

Signal Transduction Pathways Mediated by Secreted and Non-Secreted Forms of Intact Insulin-Like Growth Factor
Binding Protein-3 (IGFBP-3) and Its 1-97 N-Terminal Fragment in PC-3 Human Prostate Cancer Cells
1295
drial cytochrome c [23]. Being consistent with our pre-
vious studies [5], current data suggest that non-secreted
IGFBP-3 induced apoptosis in PC-3 cells, is mediated by
caspase-8 as well as the caspase-9 pathways. Our results
indicate that apoptosis induced by non-secreted IGFBP-3
as compared to secreted form could be blocked after ad-
dition of caspase-8 and -9 inhibitors, indicating that both
initiator caspase pathways are involved in this process.
Several studies have demonstrated that full-length
IGFBP-3 can induce various signaling pathways [6,17,24,
25]. In this regard, PC-3 cells were transfected with dif-
ferent IGFBP-3 constructs and then treated with inhibi-
tors specific for PI-3K and p38 MAPK. Results from
FACS analyses indicated that there was little or no differ-
ence in the extent of cell death (data not shown), indicat-
ing that the PI-3K and p38 MAPK signaling pathways
may not be involved in IGFBP-3 mediated cell death in
PC-3 cells.
Another signaling molecule (Stat-1) has been impli-
cated in IGFBP-3 induced apoptosis in rat chondropro-
genitor cells [10], this study led us to demonstrate the
role of Stat-1 in IGFBP-3 induced apoptosis in PC-3 hu-
man prostate cancer cells. Dose response experiment show-
ed increased Stat-1 protein expression at higher concen-
trations of exogenous IGFBP-3. Tyrosine phosphoryla-
tion of Stat-1 was also enhanced at higher concentrations
of exogenous IGFBP-3 in a time dependent manner (Fig-
ures 2(a) and (b)). In contrast to this previous report, our
results indicated that STAT-1 knock-down potentiated
the apoptotic effects of various IGFBP-3 constructs in
PC-3 human prostate cancer cells (Figures 3(a) and (b)).
The data suggest that Stat-1 protein might be playing a
protective role in PC-3 cells and the data also suggest
that the involvement of STAT-1 in apoptosis and prolif-
eration might be cell type dependent.
IGFBP-3 has shown to interact with transforming
growth factor-β (TGF-β) cell surface receptors. In mink
lung epithelial cells and other cell-types, IGFBP-3 binds
to TGFβ-receptor typeV (TGFβ-RV) [12]. An IGFBP-3
mediated inhibitory effect on growth involving this par-
ticular receptor has been proposed, but mechanisms of
action have not been fully elucidated. In this pathway,
IGFBP-3 also binds to and activates intracellular signal-
ing via the TGF-β RII and TGF-β RI heteromeric com-
plex [24,25]. Addition of IGFBP-3 activated Smad2 pho-
sphorylation and inhibited cell growth in these cells.
Studies have also indicated that the addition of TGF-β
can potentiate IGFBP-3-induced cell death in PC-3 cells
(6). In that regard, our results demonstrated that the ex-
pression of IGFBP-3 constructs could significantly in-
hibit TGF-β signaling (Figure 2) in PC-3 cells. This im-
plies that IGFBP-3 might interfere with TGF-β signaling
pathways. It is also interesting to note that this inhibition
was more prominent in cells where IGFBP-3 is expressed
intra-cellularly. Results from our previous studies [5,15]
clearly indicated that the IGFBP-3-YFP fusion proteins
expressed from Pre wt 1-97, Pre6m 1-97 WtFL, Wt 1-97,
and 6 m 1-97 constructs are present predominantly inside
the cells whereas Prewt FL protein was present outside in
the media. How these intracellular proteins can interfere
with TGF-β signals is not clear yet. Although we did not
measure differential Smad activation levels in these cells,
our data clearly suggest that TGF-β signaling pathways
are playing an important role in IGFBP-3 induced bio-
logical actions in these cells. Future studies would be
needed to elucidate the roles of IGFBP-3 in TGF-β sig-
naling by comparing the gene and protein expression
profile in these cells. Our data indicate that full length
IGFBP-3 and its N-terminal fragments induced the apo-
ptosis of PC-3 cells through the modulations of caspases
8, 9. STAT-1 and TGF-β signal pathways are also in-
volved. It also suggests the modulation of STAT-1 and
TGF-β pathways as possible therapy for prostate cancer.
5. Acknowledgements
We gratefully acknowledge Drs. William Chong (NIDCR,
NIH) and Maria Mochin-Peters (University of Maryland,
Baltimore) for comments and suggestions.
REFERENCES
[1] T. Laursen, A. Flyvbjerg, J. O. L. Jorgensen and R. C
Baxter, “Stimulation of the 150-Kilodalton Insulin-Like
Growth Factor-Binding Protein-3 Ternary Complex by
Continuous and Pulsatile Patterns of Growth Hormone
(GH) Administration in GH-Deficient Patients,” The Jour-
nal of Clinical Endocrinology & Metabolism, Vol. 85, No.
11, 2000, pp. 4310-4314.
http://dx.doi.org/10.1210/jc.85.11.4310
[2] M. M. Rechler and D. R. Clemmons, “Regulatory Actions
of Insulin-Like Growth Factor-Binding Proteins,” Trends
Endocrinology Metabolism, Vol. 9, No. 5, 1998, pp. 176-
183. http://dx.doi.org/10.1016/S1043-2760(98)00047-2
[3] S. M. Firth and R. C. Baxter, “Cellular Actions of the
Insulin-Like Growth Factor Binding Proteins,” Endocri-
nology Review, Vol. 23, No. 6, 2002, pp. 824-854.
http://dx.doi.org/10.1210/er.2001-0033
[4] J. Hong, G. Zhang, F. Dong and M. M. Rechler, “Insu-
lin-Like Growth Factor (IGF)-Binding Protein-3 Mutants
That Do Not Bind IGF-I or IGF-II Stimulate Apoptosis in
Human Prostate Cancer Cells,” Journal of Biological
Chemistry, Vol. 277, No. 12, 2002, pp. 10489-10497.
http://dx.doi.org/10.1074/jbc.M109604200
[5] N. Bhattacharyya, K. Pechhold, H. Shahjee, et al., “Non-
Secreted Insulin-Like Growth Factor Binding Protein-3
(IGFBP-3) Can Induce Apoptosis in Human Prostate Can-
cer Cells by IGF-Independent Mechanisms without Be-
ing Concentrated in the Nucleus,” Journal of Biological

Signal Transduction Pathways Mediated by Secreted and Non-Secreted Forms of Intact Insulin-Like Growth Factor
Binding Protein-3 (IGFBP-3) and Its 1-97 N-Terminal Fragment in PC-3 Human Prostate Cancer Cells
1296
Chemistry, Vol. 281, No. 34, 2006, pp. 24588-24601.
http://dx.doi.org/10.1074/jbc.M509463200
[6] R. Rajah, B. Valentinis and P. Cohen, “Insulin-Like Growth
Factor (IGF)-Binding Protein-3 Induces Apoptosis and
Mediates the Effects of Transforming Growth Factor-Be-
ta1 on Programmed Cell Death through a p53- and IGF-
Independent Mechanism,” Journal of Biological Chemis-
try, Vol. 272, No. 18, 1997, pp. 12181-12188.
http://dx.doi.org/10.1074/jbc.272.18.12181
[7] A. J. Butt, K. A. Fraley, S. M. Firth and R. C. Baxter,
“IGF-Binding Protein-3-Induced Growth Inhibition and
Apoptosis Do Not Require Cell Surface Binding and Nu-
clear Translocation IN Human Breast Cancer Cells,” Endo-
crinology, Vol. 143, No. 7, 2002, pp. 2693-2699.
http://dx.doi.org/10.1210/en.143.7.2693
[8] K.W.Lee, L. Ma, X. Yan, et al., “Rapid Apoptosis Induc-
tion by IGFBP-3 Involves an Insulin-Like Growth Factor-
Independent Nucleomitochondrial Translocation of RXR-
alpha/Nur77,” Journal of Biological Chemistry, Vol. 280,
No. 17, 2005, pp. 16942-16948.
http://dx.doi.org/10.1074/jbc.M412757200
[9] L. J. Schedlich, T. F. Young, S. M. Firth and R. C. Baxter,
“Insulin-Like Growth Factor-Binding Protein (IGFBP)-3
and IGFBP-5 Share a Common Nuclear Transport Path-
way in T47D Human Breast Carcinoma Cells,” Journal of
Biological Chemistry, Vol. 273, No. 29, 1998, pp. 18347-
18352. http://dx.doi.org/10.1074/jbc.273.29.18347
[10] A. Spagnoli, M. Torello, S. R. Nagalla, et al., “Identifica-
tion of STAT-1 as a Molecular Target of IGFBP-3 in the
Process of Chondrogenesis,” Journal of Biological Chemis-
try, Vol. 277, No. 21, 2002, pp. 18860-18867.
http://dx.doi.org/10.1074/jbc.M200218200
[11] G. Zappalà and M. M. Rechler, “IGFBP-3, Hypoxia and
TNF-Alpha Inhibit Adiponectin Transcription,” Bioche-
mical and Biophysical Research Communications, Vol.
382, No. 4, 2009, pp. 785-789.
http://dx.doi.org/10.1016/j.bbrc.2009.03.112
[12] S. M. Leal, S. S. Huang and J. S. Huang, “Interactions of
High Affinity Insulin-Like Growth Factor-Binding Pro-
teins with the Type V Transforming Growth Factor-Be-
tareceptor in Mink Lung Epithelial Cells,” Journal of
Biological Chemistry, Vol. 274, No. 10, 1999, pp. 6711-
6717. http://dx.doi.org/10.1074/jbc.274.10.6711
[13] L. C. Giudice, E. M. Farrell, H. Pham, et al., “Insulin-
Like Growth Factor Binding Proteins in Maternal Serum
throughout Gestation and in the Puerperium: Effects of a
Pregnancy-Associated Serum Protease Activity,” The
Journal of Clinical Endocrinology & Metabolism, Vol. 71,
No. 4, 1990, pp. 806-816.
http://dx.doi.org/10.1210/jcem-71-4-806
[14] P. Angelloz-Nicoud, C. Lalou and M. Binoux, “Prostate
Carcinoma (PC-3) Cell Proliferation Is Stimulated by the
22 - 25-kDa Proteolytic Fragment (1-160) Andinhibited by
the 16-kDa Fragment (1 - 95) of Recombinant Human Insu-
lin-Like Growth Factor Binding Protein-3,” Growth Hor-
mone & IGF Research, Vol. 8, No. 1, 1998, pp. 71-75.
[15] H. Shahjee, N. Bhattacharyya and G. Zappala, “An N-Ter-
minal Fragment of Insulin-Like Growth Factor Binding
Protein-3 (IGFBP-3) Induces Apoptosis in Human Pros-
tate Cancer Cells in an IGF-Independent Manner,” Growth
Hormone & IGF Research, Vol. 18, No. 3, 2008, pp. 188-
197.
[16] Y. Oh, H. L. Muller, L. Ng and R. G. Rosenfeld, “Trans-
forming Growth Factor-Beta-Induced Cell Growth Inhi-
bition in Human Breast Cancer Cells Is Mediated through
Insulin-Like Growth Factor-Binding Protein-3 Action,”
Journal of Biological Chemistry, Vol. 270, No. 23, 1995,
pp. 13589-13592.
http://dx.doi.org/10.1074/jbc.270.23.13589
[17] Z. S. Gucev, Y. Oh, K. M. Kelley and R. G. Rosenfeld,
“Insulin-Like Growth Factor Binding Protein 3 Mediates
Retinoic Acid- and Transforming Growth Factor Beta 2-
Induced Growth Inhibition in Human Breast Cancer
Cells,” Cancer Research, Vol. 56, 1996, pp. 1545-1550.
[18] G. R. Devi, D. L. Graham, Y. Oh and R. G. Rosenfeld,
“Effect of IGFBP-3 on IGF and IGF-Analogue-Induced
Insulin-Like Growth Factor-I Receptor (IGFIR) Signal-
ling,” Growth Hormone & IGF Research, Vol. 11, No. 4,
2001, pp. 231-239.
http://dx.doi.org/10.1054/ghir.2001.0231
[19] C. Lalou, C. Lassarre and M. Binoux, “A Proteolytic
Fragment of Insulin-Like Growth Factor (IGF) Binding
Protein-3 That Fails to Bind IGFs Inhibits the Mitogenic
Effects of IGF-I and Insulin,” Endocrinology, Vol. 137,
No. 8, 1996, pp. 3206-3212.
http://dx.doi.org/10.1210/en.137.8.3206
[20] K. W. Colston, C. M. Perks, S. P. Xie and J. M. Holly,
“Growth Inhibition of Both MCF-7 and Hs578T Human
Breast Cancer Cell Lines by Vitamin D Analogues Is As-
sociated with Increased Expression of Insulin-Like Growth
Factor Binding Protein-3,” Journal of Molecular Endo-
crinology, Vol. 20, 1998, pp. 157-162.
[21] A. J. Butt, S. M. Firth, M. A. King and R. C. Baxter, “In-
sulin-Like Growth Factor-Binding Protein-3 Modulates
Expression of Bax and Bcl-2 and Potentiates p53-Inde-
pendent Radiation-Induced Apoptosis in Human Breast
Cancer Cells,” Journal of Biological Chemistry, Vol. 275,
No. 50, 2000, pp. 39174-39181.
http://dx.doi.org/10.1074/jbc.M908888199
[22] B. J. Boyle, X. Y. Zhao, P. Cohen and D. Feldman, “Insu-
lin-Like Growth Factor Binding Protein-3 Mediates 1 Al-
pha, 25-Dihydroxyvitamin d(3) Growth Inhibition in the
LNCaP Prostate Cancer Cell Line through p21/ WAF1,”
Journal of Urology, Vol. 165, No. 4, 2001, pp. 1319-1324.
http://dx.doi.org/10.1016/S0022-5347(01)69892-6
[23] D. R. Green, “Apoptotic Pathways: The Roads to Ruin,”
Cell, Vol. 94, No. 6, 1998, pp. 695-698.
http://dx.doi.org/10.1016/S0092-8674(00)81728-6
[24] S. Fanayan, S. M. Firth, A. J. Butt and R. C. Baxter,
“Growth Inhibition by Insulin-Like Growth Factor-Bind-
ing Protein-3 in T47d Breast Cancer Cells Requires Tran-
sforming Growth Factor-Beta (TGF- Beta) and the Type
II TGF-Beta Receptor,” Journal of Biological Chemistry,
Vol. 275, No. 50, 2000, pp. 39146-39151.
http://dx.doi.org/10.1074/jbc.M006964200
[25] S. Fanayan, S. M. Firth and R. C. Baxter, “Signaling

Signal Transduction Pathways Mediated by Secreted and Non-Secreted Forms of Intact Insulin-Like Growth Factor
Binding Protein-3 (IGFBP-3) and Its 1-97 N-Terminal Fragment in PC-3 Human Prostate Cancer Cells
1297
through the Smad Pathway by Insulin-Like Growth Fac-
tor-Binding Protein-3 in Breast Cancer Cells. Relationship
to Transforming Growth Factor-Beta 1 Signaling,” Jour-
nal of Biological Chemistry, Vol. 277, No. 9, 2002, pp.
7255-7261. http://dx.doi.org/10.1074/jbc.M108038200