Alterations in the expression of PDCD4 in ductal carcinoma of the breast

Abstract

Programmed cell death 4 gene (PDCD4), an in vivo repressor of transformation, was originally isolated from a human glioma library by screening it with an antibody against a nuclear antigen in proliferating cells. PDCD4 functions as a transformation repressor by inhibiting the activity of the RNA helicase, eIF4A. We previously showed that retinoids, anti-estrogens and HER2/neu antagonist induce PDCD4 expression in human breast cancer cell lines. Very little is known about the expression of PDCD4 in human breast cancer tissues or the significance of the PDCD4 expression in breast cancer. To gain insight into the pattern of the PDCD4 expression in breast tissues, we performed an immunohistochemical analysis of the PDCD4 expression in 80 archived, normal and ductal breast carcinoma tissues (invasive and carcinoma in situ) (DCIS) and correlated PDCD4 expression with expression of known prognostic markers in breast cancer (ER, PR and HER2/neu). To assess the role of methylation on PDCD4 expression in breast cancer cells, breast cancer cell lines were treated with the demethylating agent 5-deoxy-azacytidine and analyzed for PDCD4 expression. We observed primarily nuclear localization of PDCD4 in ductal carcinoma in situ compared to normal breast tissues where the PDCD4 expression was predominantly cytoplasmic. This was seen more frequently in DCIS cases that were ER positive and HER2/neu negative samples. PDCD4 expression was markedly decreased in the invasive ductal carcinoma. We did not observe any significant relationship between PDCD4 expression and the expression of RAR or PR. In T-47D, MDA-MB-435 and MDA-MB-231 cells, treatment with 5-deoxy-azacytidine did not result in an increased expression of PDCD4. The present study demonstrated altered cellular localization of PDCD4 when comparing normal breast to neoplastic breast tissues. In addition, there was a decreased expression of PDCD4 in breast cancer when compared with normal breast tissue. A loss of the PDCD4 expression in breast cancer cell lines does not appear to result from hypermethylation of the PDCD4 promoter.

Full text

Abstract. Programmed cell death 4 gene (PDCD4), an in vivo
repressor of transformation, was originally isolated from a
human glioma library by screening it with an antibody against
a nuclear antigen in proliferating cells. PDCD4 functions as a
transformation repressor by inhibiting the activity of the
RNA helicase, eIF4A. We previously showed that retinoids,
anti-estrogens and HER2/neu antagonist induce PDCD4
expression in human breast cancer cell lines. Very little is
known about the expression of PDCD4 in human breast cancer
tissues or the significance of the PDCD4 expression in breast
cancer. To gain insight into the pattern of the PDCD4
expression in breast tissues, we performed an immunohisto-
chemical analysis of the PDCD4 expression in 80 archived,
normal and ductal breast carcinoma tissues (invasive and
carcinoma in situ) (DCIS) and correlated PDCD4 expression
with expression of known prognostic markers in breast
cancer (ER, PR and HER2/neu). To assess the role of
methylation on PDCD4 expression in breast cancer cells,
breast cancer cell lines were treated with the demethylating
agent 5-deoxy-azacytidine and analyzed for PDCD4
expression. We observed primarily nuclear localization of
PDCD4 in ductal carcinoma in situ compared to normal
breast tissues where the PDCD4 expression was predo-
minantly cytoplasmic. This was seen more frequently in DCIS
cases that were ER positive and HER2/neu negative samples.
PDCD4 expression was markedly decreased in the invasive
ductal carcinoma. We did not observe any significant relation-
ship between PDCD4 expression and the expression of
RAR or PR. In T-47D, MDA-MB-435 and MDA-MB-231
cells, treatment with 5-deoxy-azacytidine did not result in an
increased expression of PDCD4. The present study demon-
strated altered cellular localization of PDCD4 when comparing
normal breast to neoplastic breast tissues. In addition, there
was a decreased expression of PDCD4 in breast cancer when
compared with normal breast tissue. A loss of the PDCD4
expression in breast cancer cell lines does not appear to result
from hypermethylation of the PDCD4 promoter.
Introduction
The programmed cell death 4 gene (PDCD4) was originally
isolated from a human glioma library by screening it with an
antibody against a nuclear antigen in proliferating cells (1).
PDCD4 (H731L) is homologous to the mouse Pdcd4 gene
(also known as MA-3/TIS/A7-1) (2) and it codes for a protein
of 469 amino acids with a predicted size of 62 kDa (3,4). The
two human PDCD4 homologs, H731L and H731, are respec-
tively, 96 and 93% identical to the mouse Pdcd4 gene (5).
H731L and H731 are alternative transcripts of the same gene
with H731 lacking 11 amino acids present in the N-terminal
region of the mouse Pdcd4 and H731L (5). The human PDCD4
gene was localized to chromosome 10q24 (6) but its function
is not well defined. The deduced amino acid sequence of
PDCD4 suggests that the protein contains two N-terminal
basic domains, which may function as nuclear localization
signals and two nuclear export sequences, suggesting that
PDCD4 is capable of shuttling back to the cytoplasm under
certain conditions (2,7). Pdcd4 is now known to prevent eIF4A
from binding to eIF4G, resulting in the inhibition of cap-
dependent translation (5). PDCD4 has an intrinsic RNA-
binding activity (7) and it was postulated that it may also be
involved in RNA-processing events such as splicing and
nucleo-cytoplasmic transport. Expression of the mouse Pdcd4
gene was shown to be associated with apoptosis in several
systems (8,9). However, its role in apoptosis is yet to be
elucidated.
Very little is known about the upstream regulators and
downstream targets of PDCD4 in normal cells. Yang et al,
reported that Pdcd4 inhibited the activation of AP-1-dependent
transcriptional activity in a dose-dependent manner in colon
cancer cell lines (10). The serine/threonine kinase Akt is now
reported to be an upstream modulator of the Pdcd4 activity.
Akt (protein kinase B) was observed to specifically phos-
ONCOLOGY REPORTS 18: 1387-1393, 2007
Alterations in the expression of PDCD4 in
ductal carcinoma of the breast
YONG HANNAH WEN1, XIUQUAN SHI2, LUIS CHIRIBOGA1,
SACHIKO MATSAHASHI3, HERMAN YEE1and OLUBUNMI AFONJA2
1Departments of Pathology and 2Pediatrics, New York University School of Medicine, 550 First Avenue, New York,
NY 10016, USA; 3Department of Internal Medicine, Saga Medical School, Saga University, Saga 849-8501, Japan
Received April 20, 2007; Accepted August 16, 2007
_________________________________________
Correspondence to: Dr Olubunmi Afonja, Department of
Pediatrics, New York University School of Medicine, 550 First
Avenue, New York, NY 10016, USA
E-mail: afonjo01@med.nyu.edu
Key words: programmed cell death 4, estrogen receptor, HER2/neu,
apoptosis, methylation, differentiation
1387-1393 7/11/07 17:57 Page 1387

phorylate Ser67 and Ser457 of Pdcd4 in vitro and in vivo
resulting in a significant decrease in the ability of Pdcd4 to
interfere with the transactivation of an AP-1-responsive
promoter by c-Jun (11). Additionally, RKO cells, stably
transfected with Pdcd4, had a reduced invasive capability
which was associated with a significant reduction in
Mitogen-Activated Protein kinase kinase kinase kinase 1
(MAP4K1) activity, thereby suggesting that Pdcd4 targets
the Jun N-terminal kinase (JNK) pathway by modulating
upstream targets such as MAP4K1 (12). Carbonic anhydrase II
was recently identified as a PDCD4 target in endocrine tumor
cell lines (13).
PDCD4 is ubiquitously expressed at low levels in normal
tissues (2,8,9,14). There are conflicting data about its primary
cellular localization and this appears to be dependent on the
tissue being analyzed and the cellular microenvironment (7).
The PDCD4 expression was detected primarily in nuclear
extracts from the T-47D breast cells (15). It was, however,
detected within the nucleus and cytoplasm of normal breast
epithelial cells (2,4). There is experimental evidence
suggesting that PDCD4 is primarily a nuclear protein which is
capable of shuttling into the cytoplasmic compartment of
cultured cells upon serum withdrawal or stress (7). Together,
these data suggest that low levels of functional PDCD4 may be
essential for cell proliferation and other physiological processes
yet to be identified (4). A loss of the PDCD4 expression or
accumulation of protein in the nuclei may positively regulate
cell proliferation and this may in part contribute to the
tumorigenic process in various malignancies. In our initial
studies of the PDCD4 expression in cultured cancer cell lines,
PDCD4 expression was not detected in breast cancer cells
that lacked ER and RAR expression (15). Our observation of
the lack of the basal expression of PDCD4 in most of the ER-
breast cancer cell lines suggests that a loss of the PDCD4
expression may contribute to the development of an aggressive
phenotype of breast cancer (15).
In this study, we performed an immunohistochemical
analysis of the PDCD4 expression in ductal carcinoma of the
breast to determine if a loss of the PDCD4 expression
correlates with an aggressive histological phenotype.
Additionally, we analyzed the effects of a demethylating
agent on the PDCD4 expression in cultured breast cancer
cells.
Materials and methods
Breast tissue samples. Formalin-fixed paraffin-embedded
ductal carcinomas of the breast and normal breast tissues
were retrieved from the archival Surgical Pathology files of the
Department of Pathology at Bellevue Hospital/New York
University School of Medicine and was approved by the
Institutional Review Board. A total of 80 cases were selected,
including 65 cases of invasive ductal carcinoma, 5 of which
contain associated ductal carcinoma in situ (DCIS) and 10 cases
of pure DCIS. Adjacent benign breast epithelium was present
in 28 cases. Additionally, 5 cases of normal breast (from
reduction mammoplasty) were selected and served as normal
controls.
Cell lines. T-47D, MDA-MB-231 and MDA-MB-435 cells
were obtained from the American Type Tissue Culture
Collection (ATCC; Rockville, MD). T-47D cells were
maintained in RPMI-1640 (Mediatech Inc., USA) supple-
mented with 10% fetal calf serum (FCS). MDA-MB-231 and
MDA-MB-435 cells were maintained in Leibowitz's L-15
medium supplemented with 10% FCS.
Immunohistochemistry. Tissue microarrays were constructed
with duplicate 3-mm cores taken from different areas of each
case to ensure reproducible staining with a final array of 36
cores.
Five micron thick microarray sections were prepared onto
charged glass slides and deparaffinized in three washes of
xylene and then through graded alcohol (100 to 90 to 70%) to
deionized water. Antigen retrieval consisted of incubating the
tissue array sections in boiling 0.01 M citrate buffer, pH 6.0
for the indicated time and allowing the sections to cool to room
temperature before staining. Staining was performed by a
computer-controlled automated immunostainer (NexES,
Ventana Medical Systems, Tucson, AZ) for which the staining
protocols were optimized and programmed into the computer.
The antibodies, their dilutions and antigen retrieval (AR)
times (in min) used in this study were as follows: rabbit
antihuman PDCD4 polyclonal antibody [1:100 dilution, AR
(10)] (16), antihuman ER mouse monoclonal antibody - clone
6F11 [prediluted, AR (20), Ventana Medical Systems],
antihuman PR mouse monoclonal antibody - clone 1A6
[prediluted, AR (20), Ventana Medical Systems], antihuman
c-ErbB2 mouse monoclonal antibody - clone CB11
[prediluted, AR (10), Ventana Medical Systems] and rabbit
antihuman RARαpolyclonal antibody [1:40 dilution, AR (20),
Santa Cruz, USA]. The chromogen used in these stains is 3,3-
diaminobenzidine (DAB) resulting in a brown precipitate.
Negative controls consisted of incubating the tissue section
with isotype-matched serum without a primary antibody. The
stained sections were then scored by HW and HY.
Evaluation of PDCD4+, ER+, PR+, HER2/neu+and RAR+cells.
Assessment of immunohistochemical staining was performed
jointly by two investigators (Y.H.W. and H.Y) using a double
objective microscope. Homogeneous cytoplasmic or nuclear
staining for PDCD4 was accepted as positive. The percentage
of positive cells for PDCD4 was calculated by counting 2,000
tumor cells in the most positive areas and at least 10 high-
power fields (HPFs; 0.16 mm2) in each case. Positive staining
was graded as a percent of positive staining: 0=0-10%,
1+=10-25%, 2+=26-50% and 3+=>50%. Similar grading
criteria were used to grade staining for ER, PR and RARα. The
staining intensity for PDCD4 was assessed as being either
weak or strong when compared to the staining intensity
observed in normal breast epithelial cells. The mean intensity
of staining was calculated as the mean intensity of staining
observed in duplicate 3-mm cores taken from different areas
of each case.
The established Dako scoring for erb-B2 was used:
0=<10% and weak with less than full membrane staining,
1=>10% but weak and less than full membrane staining,
2=>10% and weak with full membrane staining, 3=>10% and
strong with full membrane staining.
Drugs. 5-deoxy-azacytidine (5-cd) was obtained from Sigma
(St. Louis, MO, USA). T-47D, MDA-MB-231 and MDA-
WEN et al: EXPRESSION OF PDCD4 IN DUCTAL CARCINOMA OF THE BREAST
1388
1387-1393 7/11/07 17:57 Page 1388

MB-435 cells were incubated with 0.1 μM or 1 μM 5-cd for
72 h.
Western blotting. Cells were plated in 60-mm tissue culture
dishes at a density of 0.5x106cells/dish in 5 ml of 10% FCS
supplemented Leibowitz's L-15 medium (for MDA-231 and
MDA-MB-435 cells) or RPMI-1640 (for T-47D cells)
containing either DMSO (vehicle used to dissolve 5-cd), or 5-cd
at concentrations previously mentioned. Following incubation
for 72 h, the medium was removed and the cells were washed
twice with PBS. Whole cell extracts were prepared by
suspending the cells in lysis buffer containing 60 mM Tris
(pH 7.5), 7.5% glycerol, 5% 2-mercaptoethanol and 1% SDS
and the protein concentration was determined using the BCA
assay (Bio-Rad Laboratories, Hercules, CA, USA). Extracts
containing 25 μg of protein were electrophoresed in 10%
polyacrylamide-SDS denaturing gels and transferred to
nitrocellulose membrane (Osmonics, Westborough, MA,
USA) over 3 h by electroblotting. Ponceau S staining of the
membrane and Coomassie blue staining of the SDS-PAGE
gel were used to confirm that equal amounts of protein were
transferred from each lane. The membrane was blocked with
5% non-fat milk in TBS [50 mM Tris (pH 7.5) and 150 mM
NaCl] for 1 h at room temperature. The membrane was then
incubated with a rabbit polyclonal PDCD4 antibody at a
1:1000 dilution overnight at 4˚C. After washing with TBS, the
membrane was incubated with goat anti-rabbit IgG
conjugated with horseradish peroxidase (1:2000 dilution) for
1.5 h at room temperature. The membrane was then washed
with TBS twice, followed by two washes with TBS
containing 0.05% Tween-20 and developed using enhanced
chemiluminescence (ECL) (Pierce, Rockford, IL, USA).
Statistical analysis. Statistical analysis was performed using
the Prism Graphpad software (Graphpad Prism v4.0b, San
Diego, CA, USA). ANOVA was used to compare multiple
groups and the Bonferroni's multiple comparison tests were
also calculated. P<0.05 was considered statistically
significant.
Results
Expression of PDCD4 in normal and malignant breast cells.
PDCD4 was abundantly expressed as a cytoplasmic protein in
all 5 cases of normal breast tissue (Fig. 2C, left panel).
Benign epithelial cells adjacent to the tumor also showed an
abundant PDCD4 expression in all 28 cases analyzed (Fig. 2C,
middle and right panels). In contrast, PDCD4 was expressed at
varying levels in 13 (87%) of 15 DCIS and in only 36 (55%) of
65 invasive ductal carcinoma cases (Fig. 1A). Some
differences in the distribution and intensity of staining were
observed between DCIS, DCIS with invasive ductal carcinoma
and pure invasive ductal carcinoma. Specimens were assigned
scores 0-3 according to the intensity of the staining and the
number of stained cells. The difference of the PDCD4
expression in normal breast tissue compared with in situ and
the invasive breast carcinomas is shown in Fig. 1B. When
compared to normal breast tissue and DCIS, the invasive
carcinoma showed a significantly reduced expression of
PDCD4 (Fig. 1B). This is statistically significant with p<0.001.
Subcellular localization of PDCD4 staining in breast cells.
PDCD4 was present in both the cytoplasmic and nuclear
subcellular compartments. In the benign breast epithelium,
PDCD4 was predominantly localized within the cytoplasm
(25 out of 28, 89%) (Fig. 2A-C). Nuclear staining for PDCD4
was noted in 10 (36%) out of the 28 cases. Nuclear staining
was not as intense as cytoplasmic staining in these cases. Weak
cytoplasmic (1+) but stronger nuclear staining (2-3+) for
PDCD4 was observed in 3 out of the 28 cases of the benign
breast epithelium adjacent to the tumor that was analyzed. A
greater proportion of DCIS showed nuclear staining for
PDCD4. A nuclear PDCD4 expression was evident in 9 (60%)
out of the 15 DCIS. This observation was unique in DCIS
(Figs. 2B and D). In invasive ductal carcinoma, both nuclear
and cytoplasmic immunostaining for PDCD4 were reduced
(Fig. 2D). Unlike DCIS, the localization of PDCD4 in the
invasive carcinoma was observed more often in the cytoplasm
(34 out of 65, 52%) rather than in the nucleus (14 out of 65,
22%) (Figs. 2B and D).
Correlation between PDCD4, ER, PR, HER2/neu and
RAR expression. In our previous study, looking at PDCD4
expression in cultured cells, we observed PDCD4 expression
in all ER positive breast cancer cell lines analyzed.
Therefore, we wanted to see if this relationship holds true in
human breast cancer tissue samples. We analyzed the
relationships between PDCD4 expression, estrogen receptor
(ER) and HER2/neu status. Overall, 15 (19%) cases were
ONCOLOGY REPORTS 18: 1387-1393, 2007 1389
Figure 1. Graphical representation of the expression of PDCD4 (A) and
relative intensity of PDCD4 staining (B) in archived breast samples (normal
and malignant). The number of positive cases is shown as a percent of the
total number of cases analyzed. (IDC, invasive ductal carcinoma; DCIS,
ductal carcinoma in situ). The mean intensity of staining was calculated as
the mean intensity of staining observed in duplicate 3-mm cores taken from
different areas of each case.
A
B
1387-1393 7/11/07 17:57 Page 1389

ER-/HER2/neu+, 22 (27%) cases were ER+/HER2/neu+, 21
(26%) cases were ER-/HER2/neu-and 22 (27%) cases were
ER+/HER2/neu-. There was a significant association between
the expression of ER, HER2/neu and PDCD4 nuclear
localization. PDCD4 was expressed as a nuclear protein in all
DCIS cases that were ER+/HER2/neu-(Fig. 3A), whereas only
50% of ER-/HER2/neu-DCIS tumors expressed a PDCD4
protein (Fig. 3A). An increased nuclear expression observed
in ER+/HER2/neu-DCIS cases was statistically significant
when compared with ER+/HER2/neu-IDC cases with p<0.05
(Fig. 3A). When compared to ER+/HER2-tumors (DCIS and
IDC), a decreased expression of PDCD4 was observed in
most ER-/HER2-tumors and this was statistically significant
with p<0.03 (Fig. 3C). We did not identify any cases with
exclusive nuclear staining for PDCD4.
We did not observe any significant relationship between
PDCD4 expression, RAR and PR status of breast cancer cells
(data not shown). There were 5 cases of DCIS with IDC that
were strongly positive for ER and PR (3+) and negative
for HER2/neu. PDCD4 was expressed as a nuclear and/or
WEN et al: EXPRESSION OF PDCD4 IN DUCTAL CARCINOMA OF THE BREAST
1390
Figure 2. PDCD4 is expressed as a cytoplasmic and/or nuclear protein in normal and transformed breast cells. (A) a bar chart showing the number of cases
(expressed as a percentage of the total number of cases analyzed) with cytoplasmic and/or nuclear expression of PDCD4. (B) a representation of the relative
intensity of the cytoplasmic or nuclear expression of PDCD4 in normal breast epithelial cells (norm), ductal carcinoma in situ (DCIS) and invasive ductal
cancer (IDC). (C) and (D) representative pictures of the immunohistochemical staining pattern for PDCD4 in normal (C) and transformed (D) breast cells. (C)
[low magnification (A-C, upper panel) and high magnification (D-F, lower panel)] shows normal breast epithelial cells (left panel), normal breast epithelium
adjacent to DCIS (middle panel) and normal breast epithelium adjacent to IDC (right panel). (D) representative cases of normal breast epithelial (A, left), DCIS
with predominant nuclear staining (B, middle) and low staining of PDCD4 observed in IDC (C, right). The insert shows each section at a high magnification.
AB
C
D
1387-1393 7/11/07 17:57 Page 1390

cytoplasmic protein in areas of DCIS in all of these tumors,
as opposed to the absence or very weak expression of PDCD4
in the IDC component of these tumors. The nuclear staining
observed in the DCIS components was similar or greater to
that observed within the cytoplasm of these cells. A weak
cytoplasmic expression of PDCD4 was observed in 2 of the 3
ER+/PR-/HER2-IDC cases analyzed.
Effect of demethylating agent on PDCD4 expression in T-47D
and MDA-MB-435 cells. DNA demethylation plays a critical
role in transcriptional regulation in differentiated somatic
cells and a loss or decreased gene expression in tumor cells
can result from hypermethylation of CpG dinucleotides
within the promoter region of the gene. The PDCD4 promoter
is yet to be identified. An analysis of the 5' sequences
immediately upstream of the start of transcription revealed
several potential CpG methylation sites. Therefore, we sought
to see if 5-aza-2'-deoxycytidine (5-cd) would induce PDCD4
expression in breast cancer cells. T-47D, MDA-MB-231 and
MDA-MB-435 cells were treated with 100 nm and 1 μM 5-cd
for 48 h. Total cell lysate was isolated from these cells and
analyzed by immunoblotting using a specific PDCD4
polyclonal antibody that was previously reported (2).
PDCD4 protein expression was not induced by 5-cd in T-47D
and MDA-MB-435-cells as seen in Fig. 4. Similar results were
obtained in the MDA-MB-231 cells (data not shown).
Discussion
The focus of this study was to analyze PDCD4 expression
in breast tissues and determine the clinical significance of
variations in PDCD4 expression in ductal carcinoma of the
breast. Consideration of the in vivo pattern of the expression of
PDCD4 in breast tissues and a better understanding of the
mechanism of the action of PDCD4 will help in understanding
its role in breast physiology and carcinogenesis.
The present study demonstrated that PDCD4 is abundantly
expressed as a cytoplasmic protein in normal breast epithelial
cells. In addition to the predominant cytoplasmic expression
observed in normal breast epithelial cells, a weak nuclear
expression was observed in 10% of cases analyzed. These
findings are in agreement with those previously reported by
Yoshinaga et al (2). The expression of PDCD4 was observed
within the epithelial cells of normal breast tissue and it was
observed in both the nucleus and the cytoplasm (2). These
findings suggest that PDCD4 may play a role in normal breast
physiology. We observed a marked decrease in the PDCD4
expression in transformed breast cells. In ductal carcinoma
in situ, the expression of PDCD4 was observed in 87% of
tumor samples analyzed, whereas only 55% of the invasive
ductal carcinoma samples weakly expressed PDCD4. More
importantly, the PDCD4 expression was significantly reduced
in the invasive breast carcinoma when compared to normal
breast epithelium and this was statistically significant
(p<0.001). Our findings suggest that a loss of the PDCD4
expression may contribute to breast carcinogenesis and may be
associated with a more aggressive phenotype. These findings
are similar to those recently reported by Zhang et al (16).
In their series of 18 hepatocellular carcinoma samples, Zhang
et al, observed reduced levels of the PDCD4 protein in
hepatocellular carcinoma tissues when compared with their
corresponding non-cancerous tissues (16). Chen et al also
reported a decreased expression of PDCD4 in human lung
cancers and this correlates with tumor progression and poor
prognosis (17). Loss of the PDCD4 expression has also been
implicated in the pathogenesis of glioblastoma multiforme (18).
ONCOLOGY REPORTS 18: 1387-1393, 2007 1391
Figure 3. Correlation of PDCD4 expression with ER and HER2/neu
expression. (A) shows a number of cases (represented as a percentage of the
total cases analyzed) expressing PDCD4 either as a cytoplasmic and/or
nuclear protein in relation to the expression of ER or HER2/neu. (B) the mean
intensity of staining for PDCD4 in DCIS and IDC in relation to the ER (a)
or Her2/neu (b) status.
Figure 4. A lack of induction of the PDCD4 expression in breast cancer cells
treated with 5-aza-deoxycytidine. T-47D (lanes 1-3), MDA-MB-435
(lanes 4-6) and MDA-MB-231 (not shown) were incubated with 100 nM
(lanes 2 and 5) or 1 μM (lanes 3 and 6) 5-aza-cytidine for 72 h.
Immunoblotting for PDCD4 protein expression was performed on whole cell
extract. Induction of PDCD4 expression was not detected in T-47D (lanes
2 and 3), MDA-MB-435 (lanes 5 and 6) and MDA-MB- 231 (not shown)
breast cancer cells. Coomassie blue staining of the SDS-PAGE gel confirmed
equal loading of protein.
A
B
1387-1393 7/11/07 17:57 Page 1391

It was previously reported that PDCD4 shuttles between
the nucleus and cytoplasm and may have a role in RNA
metabolism in addition to its role in protein translation (7). In
our study, we observed that PDCD4 is expressed primarily as a
cytoplasmic protein in normal breast epithelial cells, with an
increased nuclear expression seen in most cases of DCIS
analyzed and a significantly decreased nuclear and cytoplasmic
expression in most cases of IDC analyzed. In cultured breast
cancer cells, PDCD4 is expressed primarily as a nuclear
protein and an enforced expression of PDCD4 in these cells
results in apoptosis. These observations suggest that changes in
the cellular sub-localization of PDCD4 may be one of the
contributing events in the process of breast carcinogenesis. It is
tempting to speculate that the PDCD4's main physiological
role in breast cancer cells involves the regulation of protein
synthesis and signals resulting in uncontrolled cell proliferation
that may result in an increased nuclear import to enhance its
role in RNA metabolism, thereby resulting in a decreased
synthesis of proteins that support uncontrolled growth and
survival. It will be important to identify key proteins that
regulate PDCD4 expression and its subcellular localization
and how aberrations in these processes contribute to
carcinogenesis.
We observed a significant correlation between the PDCD4
expression and ER. This is particularly of interest as it was
previously shown that the antisense PDCD4-transfected MCF 7
breast cancer cells were less sensitive to the anti-proliferative
effects of tamoxifen and geldanamycin (19). Additionally,
PDCD4 was recently identified as an ER status reporter gene in
a ‘gene expression signature’ that was associated with a better
outcome (20). We speculate that PDCD4 may be involved in
cellular responses mediated via ER. Although we had
previously identified PDCD4 as a retinoid induced gene, we
did not demonstrate any significant association between the
PDCD4 and RAR expression in our series. One potential
explanation for the findings observed in the current study is
that we analyzed the basal expression of PDCD4 and not the
induction of PDCD4 by RAR agonists. It should be noted
that some RAR+ breast cancer cell lines had low basal or
absent levels of the expression of PDCD4. However, when
incubated with RAR agonists, PDCD4 expression was induced
(15).
Unlike other classic tumor suppressor genes such as p53
and retinoblastoma genes, there are no published studies to
show that PDCD4 is inactivated by methylation, deacetylation
or genetic mutations in human cancers. Our study is the first
to suggest that hypermethylation of CpG dinucleotides within
the promoter region of the PDCD4 gene is not responsible for
the lack of PDCD4 expression in breast cancer cells and
this may apply to other cancers in which decreased expression
of PDCD4 was documented. In cultured breast cancer cells
treated with the histone deacetylase inhibitor, Trichostatin A,
we did not observe any changes in the PDCD4 expression
levels (data not shown), suggesting that chromatin remodeling
does not directly influence PDCD4 expression in breast
cancer cells. Other mechanisms of down-regulation of
classic tumor suppressor genes include point mutations, a
loss of heterozygosity and microdeletion of the region of
DNA in which the gene is contained. PDCD4 is located on
chromosome 10q24 and there is no published report of
chromosomal aberrations involving this locus. There are
published studies documenting allelic damage involving 10q in
columnar cell changes with atypia, ductal carcinoma in situ
and invasive carcinoma (21). The gene most frequently
involved is the PTEN gene located on 10q23. Further studies
specifically looking at the PDCD4 locus are required to
determine if this region is also damaged during the progression
from atypia to invasive carcinoma.
Our results indicate that PDCD4 is mainly expressed in
the cytoplasm of normal breast cells. There is increased
nuclear expression of PDCD4 in DCIS and a marked decrease
in cytoplasmic and nuclear expression in IDC. PDCD4 is
expressed in most of the ER positive tumors analyzed. Our data
correlating cellular localization of PDCD4 with ER/HER2
status stem from the small number of DCIS cases we were able
to obtain from the archival Surgical Pathology files. We had
limited information as to the metastatic status of the archived
tumors analyzed in this study. It will be of interest to know if a
loss of the PDCD4 expression correlates with metastatic
disease. Our findings support a role for PDCD4 in normal
breast physiology and a role for the loss of the PDCD4
expression in breast carcinogenesis. Our findings also indicate
that PDCD4 shuttles between the nuclear and cytoplasmic
compartments of the cells and its subcellular localization will
determine its role in cellular physiology. In addition, in cultured
breast cancer cells, absence of the induction of PDCD4
expression by 5-cd suggests that hypermethylation of the
PDCD4 promoter does not contribute to a decreased expression
observed in breast cancer cells. Our future studies are aimed at
identifying proteins that regulate PDCD4 and define the role of
PDCD4 in the growth and biological behavior of normal and
transformed breast cells.
Acknowledgements
We are grateful to Dr Angel Pellicer (Department of Pathology,
NYU School of Medicine) for the critical review of this
manuscript. This study was supported by the Robert Wood
Johnson Foundation Grant (038398) (OA), NIH Grant
K12CA01713 (OA) and an Irma T. Hirschl Charitable Trust
Award (C015710) (OA).
References
1. Matsuhashi S, Yoshinaga H, Yatsuki H, Tsugita A and Hori K:
Isolation of a novel gene from human cell line with Pr-28 MAb
which recognizes a nuclear antigen involved in the cell cycle.
Res Commun Biochem Cell Mol Biol 1: 109-120, 1997.
2. Yoshinaga H, Matsuhashi S, Fujiyama C and Masaki Z: Novel
human PDCD4 (H731) gene expressed in proliferative cells is
expressed in the small duct epithelial cells of the breast as
revealed by an anti-H731 antibody. Pathol Int 49: 1067-1077,
1999.
3. Cmarik JL, Min H, Hegamyer G, Zhan S, Kulesz-Martin M,
Yoshinaga H, Matsuhashi S and Colburn NH: Differentially
expressed protein Pdcd4 inhibits tumor promoter-induced
neoplastic transformation. Proc Natl Acad Sci USA 96:
14037-14042, 1999.
4. Yoshinaga H, Matsuhashi S, Ahanek J, Masaki Z and Hori K:
Expression and identification of H731 gene product in HeLa
cells. Res Commun Biochem Cell Mol Biol 1: 121-131, 1997.
5. Yang HS, Jansen AP, Komar AA, Zheng X, Merrick WC,
Costes S, Lockett SJ, Sonenberg N and Colburn NH: The
transformation suppressor Pdcd4 is a novel eukaryotic
translation initiation factor 4A binding protein that inhibits
translation. Mol Cell Biol 23: 26-37, 2003.
WEN et al: EXPRESSION OF PDCD4 IN DUCTAL CARCINOMA OF THE BREAST
1392
1387-1393 7/11/07 17:57 Page 1392

6. Soejima H, Miyoshi O, Yoshinaga H, Masaki Z, Ozaki I,
Kajiwara S, Niikawa N, Matsuhashi S and Mukai T: Assignment
of the programmed cell death 4 gene (PDCD4) to human
chromosome band 10q24 by in situ hybridization. Cytogenet
Cell Genet 87: 113-114, 1999.
7. Bohm M, Sawicka K, Siebrasse JP, Brehmer-Fastnacht A,
Peters R and Klempnauer KH: The transformation suppressor
protein Pdcd4 shuttles between nucleus and cytoplasm and
binds RNA. Oncogene 22: 4905-4910, 2003.
8. Shibahara K, Asano M, Ishida Y, Aoki T, Koike T and Honjo T:
Isolation of a novel mouse gene MA-3 that is induced upon
programmed cell death. Gene 166: 297-301, 1995.
9. Jurisicova A, Latham KE, Casper RF and Varmuza SL:
Expression and regulation of genes associated with cell death
during murine preimplantation embryo development. Mol
Reprod Dev 51: 243-253, 1998.
10. Yang HS, Jansen AP, Nair R, Shibahara K, Verma AK,
Cmarik JL and Colburn NH: A novel transformation suppressor,
Pdcd4, inhibits AP-1 transactivation but not NF-kappaB or
ODC transactivation. Oncogene 20: 669-676, 2001.
11. Palamarchuk A, Efanov A, Maximov V, Aqeilan RI, Croce CM
and Pekarsky Y: Akt phosphorylates and regulates Pdcd4 tumor
suppressor protein. Cancer Res 65: 11282-11286, 2005.
12. Yang HS, Matthews CP, Clair T, Wang Q, Baker AR, Li CC,
Tan TH and Colburn NH: Tumorigenesis suppressor pdcd4 down-
regulates mitogen-activated protein kinase kinase kinase kinase
1 expression to suppress colon carcinoma cell invasion. Mol
Cell Biol 26: 1297-1306, 2006.
13. Lankat-Buttgereit B, Gregel C, Knolle A, Hasilik A, Arnold R and
Goke R: Pdcd4 inhibits growth of tumor cells by suppression of
carbonic anhydrase type II. Mol Cell Endocrinol 214: 149-153,
2004.
14. Azzoni L, Zatsepina O, Abebe B, Bennett IM, Kanakaraj P and
Perussia B: Differential transcriptional regulation of CD161 and a
novel gene, 197/15a, by IL-2, IL-15 and IL-12 in NK and T cells.
J Immunol 161: 3493-3500, 1998.
15. Afonja O, Juste D, Das S, Matsuhashi S and Samuels HH:
Induction of PDCD4 tumor suppressor gene expression by RAR
agonists, Anti-estrogen and HER2/neu antagonist in breast cancer
cells. Evidence for a role in apoptosis. Oncogene 23: 8135-8145,
2004.
16. Zhang H, Ozaki I, Mizuta T, Hamajima H, Yasutake T, Eguchi Y,
Ideguchi H, Yamamoto K and Matsuhashi S: Involvement of
programmed cell death 4 in transforming growth factor-beta1-
induced apoptosis in human hepatocellular carcinoma. Oncogene
00: 1-12, 2006.
17. Chen Y, Knosel T, Kristiansen G, Pietas A, Garber ME,
Matsuhashi S, Ozaki I and Petersen I: Loss of PDCD4 expression
in human lung cancer correlates with tumor progression and
prognosis. J Pathol 200: 640-646, 2003.
18. Gao F, Zhang P, Zhou C, Li J, Wang Q, Zhu F, Ma C, Sun W
and Zhang L: Frequent loss of PDCD4 expression in human
glioma: possible role in the tumorigenesis of glioma. Oncol Rep
17: 123-128, 2007.
19. Jansen AP, Camalier CE, Stark C and Colburn NH:
Characterization of programmed cell death 4 in multiple human
cancers reveals a novel enhancer of drug sensitivity. Mol Cancer
Ther 3: 103-110, 2004.
20. van't Veer LJ, Dai H, van de Vijver MJ, et al: Gene expression
profiling predicts clinical outcome of breast cancer. Nature 415:
530-536, 2002.
21. Dabbs DJ, Carter G, Fudge M, Peng Y, Swalsky P and
Finkelstein S: Molecular alterations in columnar cell lesions of the
breast. Mod Pathol 19: 344-349, 2006.
ONCOLOGY REPORTS 18: 1387-1393, 2007 1393
1387-1393 7/11/07 17:57 Page 1393