phosphatidylinositol 4-phosphate in the golgi apparatus regulates … · 1 phosphatidylinositol...
TRANSCRIPT
1
Phosphatidylinositol 4-phosphate in the Golgi apparatus regulates
cell-cell adhesion, and invasive cell migration in human breast cancer
Emi Tokuda1, Toshiki Itoh2, 3, Junya Hasegawa2, 4, Takeshi Ijuin1, Yukiko Takeuchi1,
Yasuhiro Irino5, Miki Fukumoto1, and Tadaomi Takenawa1
1 Integrated Center for Mass Spectrometry, Kobe University Graduate School of
Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe, Hyogo, 650-0017, Japan, 2 Divisions of
Membrane Biology, Kobe University Graduate School of Medicine, 7-5-1
Kusunoki-cho, Chuo-ku, Kobe, Hyogo, 650-0017, Japan, 3 present address: Biosignal
Research Center, Kobe University, 1-1 Rokkodai-cho, Nada-ku, Kobe, Hyogo,
657-8501, Japan, 4 present address: Laboratory of Intracellular Membrane Dynamics,
Graduate School of Frontier Biosciences, Osaka University, 2-2 Yamadaoka, Suita,
Osaka, 565-0871, Japan, 5 Department of Evidence-based laboratory Medicine, Kobe
University Graduate School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe, Hyogo,
650-0017, Japan
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
2
Correspondence author: Tadaomi Takenawa: Kobe University, 7-5-1 Kusunoki-cho,
Chuo-ku, Kobe, Hyogo, 650-0017, Japan. Phone: 81-78-382-5410; E-mail:
Running title: PI(4)P regulates cancer progression
Precis: Findings suggest that invasive progression of breast cancer cells may be
directed by formation of a procancerous phospholipid in the Golgi apparatus.
Keywords: phosphatidylinositol 4-phosphate, Golgi apparatus, GOLPH3, cell-cell
adhesion, cell invasion
Authors’ Contributions
Conception and design: E. Tokuda, T. Itoh and T. Takenawa
Development of methodology: E. Tokuda and T. Takenawa
Acquisition of data (provided animals, acquired and managed patients, provided
facilities, etc.): E. Tokuda, J. Hasegawa, T. Ijuin, Y. Takeuchi, Y. Irino and M. Fukumoto
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
3
analysis): E. Tokuda
Writing, review, and/or revision of the manuscript: E. Tokuda, T. Itoh, J. Hasegawa, T.
Ijuin, and T. Takenawa
Study supervision: T. Takenawa
Disclosure of Potential conflicts of interest
The authors disclose no potential conflicts of interest.
Grant Support
This study was supported by a Grant-in-Aid for Scientific Research (S) (Grant Number
23227005 to T.T.), and a Grant-in-Aid for Challenging Exploratory Research (Grant
Number 24650621 to E.T.) from Japan Society for the Promotion of Science.
Main text: 4993 words, Abstract: 164 words, 7 figures
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
4
Abstract
Downregulation of cell-cell adhesion and upregulation of cell migration play critical
roles in the conversion of benign tumors to aggressive invasive cancers. In this study,
we show that changes in cell-cell adhesion and cancer cell migration/invasion capacity
depend on the level of phosphatidylinositol 4-phosphate (PI4P) in the Golgi apparatus
in breast cancer cells. Attenuating SAC1, a PI4P phosphatase localized in the Golgi
apparatus, resulted in decreased cell-cell adhesion and increased cell migration in
weakly invasive cells. In contrast, silencing phosphatidylinositol 4-kinase III�
(PI4KIII�), which generates PI4P in the Golgi apparatus, increased cell-cell adhesion
and decreased invasion in highly invasive cells. Furthermore, a PI4P effector, Golgi
phosphoprotein 3 (GOLPH3), was found to be involved in the generation of these
phenotypes in a manner that depends on its PI(4)P-binding ability. Our results provide a
new model for breast cancer cell progression in which progression is controlled by PI4P
levels in the Golgi apparatus.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
5
Introduction
Phosphoinositides are lipid constituents of plasma and organelle membranes
and comprise seven different phosphorylated versions. The levels of various
phosphorylated phosphoinositides within the cell are regulated by multiple
phosphoinositide kinases and phosphoinositide phosphatases (1, 2). PI(4)P is a
relatively abundant phosphoinositide required for the maintenance and function of the
Golgi apparatus, including intracellular trafficking (3). The PI(4)P level at the Golgi
apparatus is controlled by phosphatidylinositol 4-kinases (PI4Ks) and SAC1
phosphoinositide phosphatase. The PI4K isoforms PI4KII� and PI4KIII� are known to
localize to the Golgi apparatus and to be involved in membrane transport from the
trans-Golgi network to the plasma membrane (4, 5). In contrast, SAC1, which
dephosphorylates PI(4)P at the Golgi apparatus, is involved in the same membrane
transport pathway and in protein glycosylation (6).
Several PI(4)P-binding proteins, including FAPP, OSBP, CERT, and GGA,
have been identified (7). One PI(4)P effector, GOLPH3, maintains a tensile force on the
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
6
Golgi apparatus and plays a role in the Golgi secretory pathway (8, 9). Recently,
GOLPH3 was shown to be involved in cancer progression. Human chromosome 5p13,
on which the GOLPH3 gene is located, is amplified in several cancers (10), and
expression of GOLPH3 is increased in several cancer cell lines and in some cancer
patients (11-14). GOLPH3 enhances cell proliferation and tumorigenicity (10, 15).
However, its role in cell migration and invasion remains unknown.
To metastasize from their location within a primary tumor to the surrounding tissue,
tumor cells undergo several processes, including loss of cell-cell adhesion and gains in
motile and invasive potential (16). This temporary and reversible phenotype is referred
to as epithelial-mesenchymal transition (EMT). Following invasion or distal metastasis,
metastatic tumor cells revert to an epithelial phenotype in the so-called
mesenchymal-epithelial transition (MET). EMT/MET conversions are recognized as
critical events in cancer progression (17).
To identify the physiological role of PI(4)P in the Golgi apparatus during
tumor progression, we investigated the effects of PI(4)P metabolism on cell-cell
adhesion and cell migration in breast cancer cells. PI(4)P in the Golgi apparatus was
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
7
found to regulate cell-cell adhesion, cell migration, invasion, and metastasis, which are
essential for breast cancer progression through regulation of GOLPH3 localization to
the Golgi apparatus. Based on these results, we propose a novel mechanism for the
development of cancer progression.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
8
Materials and Methods
Reagents and antibodies
Mouse anti-E-cadherin, �-catenin, PI4KIII�, and GM130 monoclonal antibodies were
purchased from BD Biosciences. Rat anti-E-cadherin monoclonal, mouse anti-GAPDH
monoclonal, rabbit anti-SAC1, rabbit anti-GOLPH3, and anti-cadherin 11 antibodies
were purchased from Takara Bio, Sigma-Aldrich, Proteintech, Epitomics, and Cell
Signaling Technology, respectively. All fluorescent reagents and antibodies were
purchased from Life Technologies. Purified phosphatidylethanolamine (PE) and
phosphatidylcholine (PC) were purchased from Avanti Polar Lipids, Inc. Purified
phosphoinositides were obtained from CellSignals Inc. and human breast cancer tissue
protein lysates from OriGene Technologies.
Plasmids
Human Fapp1 2 × PH, Grp1 PH, PLC�1 PH, and Hrs1 FYVE cDNA fragments were
subcloned into the pGEX-6P1 vector (GE Healthcare) as described previously (18).
GOLPH3 was generated by polymerase chain reaction (PCR) amplification with
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
9
MDA-MB-231 cDNA as a template and subcloned into the pRetroQ-AcGFP1-C1 vector
(Clontech Laboratories, Inc.). The GOLPH3 R90L mutant was generated from
GOLPH3-wt DNA using site-directed mutagenesis.
Cell culture
The human breast cancer cell lines MCF7, MDA-MB-23, and Hs578t and GP2-293
were maintained in DMEM medium supplemented with 10% FBS. T-47D cells were
maintained in RPMI medium supplemented with 10 μg/ml insulin and 10% FBS. The
cell lines were obtained from American Type Culture Collection. The authenticity of all
cell lines was confirmed by short tandem repeat profiling by National Institute of
Biomedical Innovation in Japan.
RNA interference (RNAi)
Stealth siRNAs targeting phosphoinositide phosphatases, SAC1 (HSS117880 and
HSS117881), PI4KIII� (HSS108028 and HSS108029), and GOLPH3 (HSS127330 and
HSS127331) were purchased from Life Technologies. The stealth siRNAs were
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
10
transfected into cells with Lipofectamine RNAiMAX (Life Technologies); experiments
were performed 72 h after transfection.
Production of retrovirus-infected cell lines
Retroviruses with the VSV-G envelope were packaged and produced according to
manufacture’s instruction (Clontech Laboratories, Inc.). Cells were infected with
retroviruses in the presence of 8 μg/m1 polybrene. The infected cells were selected by
adding 1 μg/ml puromycin to the medium for 2 weeks.
Immunofluorescence analysis
Cells grown on coverslips were fixed with 3.7% formaldehyde in PBS and
permeabilized using 0.2% Triton X-100 or 20 μM digitonin in PBS and blocked with
1.5% BSA in PBS before incubation with primary antibodies diluted in blocking buffer.
Immunofluorescence analysis was performed as previously described (19).
Quantification of Golgi PI(4)P levels
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
11
Recombinant GST-conjugated proteins were expressed in the Rosseta-gami
B(DE3)pLyS bacterial strain (Merck) and purified on glutathione-Sepharose 4B (GE
Healthcare) according to the manufacturer’s instructions. GST was removed by
on-column cleavage with PreScission protease (GE Healthcare). Recombinant Fapp1 2
× PH domain protein was labeled using the Alexa Fluor 647 Protein Labeling Kit (Life
Technologies) and used as a fluorescent probe for detecting Golgi PI(4)P. Cells were
fixed with 3.7% formaldehyde in PBS, permeabilized using 20 μM digitonin in PBS,
and stained as described under “immunofluorescence analysis.” PI(4)P levels were
quantified using FluoView software. For quantifying the amount of PI(4)P at the Golgi
apparatus, the fluorescence intensity of AF647-labeled Fapp1 2 × PH was normalized to
that of GM130, a Golgi apparatus marker.
Matrigel invasion assay
The indicated siRNA-transfected or the indicated retrovirus-infected MDA-MB-231
cells were plated in a BioCoat Matrigel invasion chamber (BD Biosciences) and
cultured for 18 h. The cells were fixed with 3.7% formaldehyde in PBS, and the
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
12
invading cells were stained with phalloidin.
Migration assay
The indicated siRNA-transfected MCF7 or T-47D cells, or the indicated
retrovirus-infected MCF7 cells were seeded. Forty-eight hours after transfection or
seeding, the cells were serum-starved for 16 h. The migration assay was performed for
24 h.
Subcellular fractionation
Cells were resuspended in Triton X-100 buffer (20 mM Tris-HCl [pH 7.5], 150 mM
NaCl, 2% glycerol, 0.1% Triton X-100, 1 mM CaCl2, and Complete Mini [Roche
Diagnostics]). The cell lysate was centrifuged at 20,630 g for 15 min at 4°C. The
supernatant was collected as a Triton X-100 soluble fraction. The pellet was sonicated
in SDS buffer (20 mM Tris-HCl [pH 7.5], 1% SDS, and 10% glycerol) and collected as
a Triton X-100 insoluble fraction.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
13
Real-time PCR
Total RNA was isolated from cells using TRIzol (Life Technologies) and used as a
template for cDNA synthesis with the PrimeScript II 1st-strand cDNA Synthesis Kit
(Takara). Real-time PCR was performed using a TaqMan probe in an ABI 7500
real-time PCR system (Life Technologies). The data were normalized to those obtained
for GAPDH.
RhoA activation assay
Cells were transfected with the indicated siRNAs. Seventy-two hours after transfection,
a RhoA activation assay was performed using a G-LISA RhoA Activation Assay
Biochem Kit (Cytoskeleton, Inc.). The data were normalized to total RhoA.
Tumor tail vein injection
Experiments were performed with 7-week-old female BALB/c-nu/nu mice (Charles
River Laboratories). All experiments were approved by the Institutional Animal Care
and Use Committee in Kobe University and were carried out in accordance with the
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
14
institute’s guidelines. Cells suspended in Hanks’ Balanced Salt Solution (Gibco) were
injected intravenously (1 × 106 cells/mouse). Three weeks after injection, the mice were
euthanized and their lungs were fixed with Bouin’s Fluid (Wako Pure Chemical
Industries). The number of lung surface metastatic foci was counted.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
15
Results
Effect of knockdown of various phosphatases on cell-cell adhesion
Non-invasive MCF7 human breast cancer cells normally exhibit tight cell-cell
adhesion. To obtain insights into the role of phosphoinositide metabolism in the
regulation of cell-cell adhesion, we investigated the effect of siRNA-mediated depletion
of all phosphoinositide phosphatase isoforms in MCF7 cells (Fig. 1). Among siRNAs
targeting 22 phosphoinositide phosphatases, transfection of those targeting PTEN,
INPP4B, SAC1, INPP5D (SHIP1), and PIB5PA (PIPP) resulted in decreases in cell-cell
adhesion (Fig. 1, asterisk). PTEN, INPP4B, INPP5D, and PIB5PA have been previously
reported to exhibit tumor-suppressive activity through dephosphorylation of PI(3,4,5)P3
or PI(3,4)P2 and subsequent inhibition of PI3K signaling (20-22).
SAC1 knockdown increases PI(4)P levels in the Golgi apparatus and affects
cell-cell adhesion and cell migration
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
16
SAC1, which dephosphorylates PI(4)P to PI in the Golgi apparatus, has not
been reported to be involved in cancer progression. Therefore, we decided to further
study whether this phosphoinositide phosphatase is involved in cancer progression.
Transfection of MCF7 cells with SAC1-targeting siRNA did not alter E-cadherin
or��-catenin protein levels in these cells (Fig. 2A), although it did result in reduced
cell-cell adhesion (Fig. 2B). Similar results were observed in T-47D human ductal
breast epithelial tumor cells, which normally exhibit cell-cell adhesion (Supplementary
Fig. 1A and B). We therefore used subcellular fractionation to monitor the association of
the actin cytoskeleton with E-cadherin and �-catenin in these cells. The amount of
E-cadherin and �-catenin in the Triton X-100-insoluble fraction decreased in
SAC1-depleted cells (Figure 2C). Consistent with the previous finding that cell
migration increases as a consequence of the disappearance of cell-cell adhesion (23),
SAC1 depletion promoted MCF7 cell (Fig. 2D) and T-47D cell (Supplementary Fig.
1C) migration and caused reduction in cell-cell adhesion in both cell types. SAC1
depletion seemed to induce stress fiber formation at the cell periphery in MCF7 cells
(Supplementary Fig. 3A), with an increase in RhoA activity in these cells (Fig. 2E).
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
17
Furthermore, SAC1 depletion induced expression of SNAI2, an EMT marker in MCF7
cells (Fig. 2F). These results indicate that SAC1 partly regulates EMT, which is
characterized by changes in the localization of E-cadherin and �-catenin at cell-cell
adhesion sites, actin cytoskeletal rearrangement, and expression of EMT markers.
Because Sac1 dephosphorylates PI(4)P in the Golgi apparatus, we quantified
the levels of cellular phosphoinositides using a liposome overlay assay. No change in
the total amount of phosphoinositides was detected in Sac1-depleted cells
(Supplementary Fig. 2A and B). Therefore, we measured relative PI(4)P levels in the
Golgi apparatus using Alexa Fluor 647 (AF647)-labeled Fapp1 PH domains (Fapp1 2 ×
PH) that specifically recognize PI(4)P with high affinity (18). PI(4)P localized at the
Golgi apparatus of HeLa cells can be efficiently detected by this fluorescent probe after
permeabilization of the cells with digitonin (Supplementary Fig. 2C) (24). The
specificity of the fluorescence signals was confirmed by pre-incubation of the probe
with PC/PE/PI(4)P liposomes, which completely blocked PI(4)P detection in the Golgi
apparatus (Supplementary Fig. 2C). Using this method, we found that Sac1 knockdown
increased PI(4)P levels in the Golgi apparatus by approximately two-fold, compared to
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
18
that in control MCF7 cells (Fig. 2G and H). Further knockdown of PI4KIII�, which
generates PI(4)P in the Golgi apparatus, in SAC1-depleted MCF7 cells restored the
formation of cell-cell adhesion (Supplementary Fig. 3B and C), indicating that PI(4)P
promotes cell migration by disrupting cell-cell adhesion.
PI(4)P levels in the Golgi apparatus during breast cancer progression
TGF� and TNF� are known to induce EMT in several cancer cell lines (25,
26). Treatment with TGF� and TNF� efficiently altered the normal epithelial
morphology of MCF7 cells, producing a �migratory phenotype. TGF�- and
TNF�-treated MCF7 cells also exhibited an increase of approximately two-fold in
PI(4)P levels in the Golgi apparatus (Fig. 3A and B). We then compared PI(4)P levels at
the Golgi apparatus in highly invasive (MDA-MB-231 and Hs578t) and weakly
invasive (MCF7 and T-47D) breast cancer cell lines. In highly invasive cell lines, PI(4)P
levels in the Golgi apparatus were significantly higher than those in weakly invasive
cell lines (Fig. 3C and D). Moreover, PI4KIII� expression was higher in late-stage
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
19
human breast cancer tissues (metastatic, stages III and IV) than in early-stage tissues
(non-metastatic, stages I and IIa) (Fig. 3E). In contrast, SAC1 expression decreased in
stages III and IV human breast cancer tissues (metastatic) (Fig. 3E). These data
collectively support the idea that there is a strong correlation between Golgi PI(4)P and
breast cancer malignancy.
PI4KIII� knockdown decreases PI(4)P levels in the Golgi apparatus and affects
cell-cell adhesion and invasion
To further confirm the correlation between PI(4)P levels in the Golgi
apparatus and cell-cell adhesion, we examined the effect of silencing PI4KIII�, which
produces PI(4)P at the Golgi apparatus, in MCF7 cells. PI4KIII� knockdown decreased
PI(4)P levels in the Golgi apparatus by half in MCF7 cells (Supplementary Fig. 3D-F).
Because tight cell-cell adhesion already exhibit in MCF7 cells, further formation of
cell-cell adhesion was not induced by PI4KIII� silencing in MCF7 cells (Supplementary
Fig. 3B). We then used highly invasive MDA-MB-231 human breast adenocarcinoma
cells that do not normally exhibit cell-cell adhesion and examined the four isoforms of
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
20
PI4-kinase that might be involved in these phenotype changes. Of the four PI 4-kinase
isoforms examined, attenuation of only PI4KIII� promoted an increase in cell-cell
adhesion in MDA-MB-231 cells (Supplementary Fig. 4A and B). Because PI4KIII� is
known to produce PI(4)P in the Golgi apparatus, it may be involved in this change. In
MDA-MB-231 cells, PI4KIII� knockdown decreased PI(4)P levels in the Golgi
apparatus to one-third of the control level (Fig. 4A and B).
MDA-MB-231 cells have been reported to utilize cadherin-11 rather than
E-cadherin and N-cadherin as a cell-adhesion molecule (27, 28). PI4KIII�-depleted
MDA-MB-231 cells formed adhesions with cadherin-11 and �-catenin without any
changes in the expression levels of these proteins (Fig. 4C and D). In addition, further
transfection of PI4KIII�-depleted MDA-MB-231 cells with Sac1 siRNA decreased their
cell-cell adhesion (Supplementary Fig. 4C and D). Another line of highly invasive
human ductal breast carcinoma cells, Hs578t cells, does not exhibit cell-cell adhesion
under normal conditions. Depletion of PI4KIII� in these cells resulted in an increase in
cell-cell adhesion (Supplementary Fig. 5A - C). Ht578t cells express N-cadherin and
cadherin-11 but not E-cadherin (27, 28). Membrane localization of these cadherins was
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
21
induced by PI4KIII� silencing without changes in protein levels (Supplementary Fig.
5A - C). MDA-MB-231 cell invasion was significantly suppressed by PI4KIII�
depletion (Fig. 4E), with an increase in RhoA activity in these cells (Fig. 4F).
Furthermore, these cells exhibited decreased expression of SNAI1, an EMT marker (Fig.
4G). These data suggest that decrease in PI(4)P levels in the Golgi apparatus is likely to
increase cell-cell adhesion, thereby inhibiting invasion and promotion of EMT in breast
cancer cells.
GOLPH3 is involved in PI(4)P-dependent changes in the phenotype
GOLPH3 is a downstream target of PI(4)P that is localized in the Golgi
apparatus (9). In PI4KIII�-depleted MDA-MB-231 cells, GOLPH3 was no longer
localized at the Golgi apparatus (Fig. 5A), without any significant change in its
expression (Supplementary Fig. 6A). GOLPH3 expression did not change between
control and SAC1 siRNA-transfected MCF7 cells (Supplementary Fig. 6B). GOLPH3
already localized at the Golgi apparatus in control siRNA-transfected MCF7 cells
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
22
(Supplementary Fig. 6C); however, fluorescence intensity of GOLPH3 increased in
SAC1-depleted MCF7 cells (Supplementary Fig. 6D). These data indicated that the
PI(4)P in the Golgi apparatus regulates GOLPH3 localization but not expression.
Because it has been reported that GOLPH3 is related to tumorigenesis (10, 15), we
hypothesized that it may be involved in the regulation of cell-cell adhesion and cell
migration/invasion. Attenuation of GOLPH3 resulted in increased cell-cell adhesion and
suppression of the invasive phenotype, which was similar to what was observed during
PI4KIII� knockdown (Fig. 5B-D). In addition, further GOLPH3 depletion in
SAC1-depleted MCF7 cells decreased migration ability (Fig. 5E) and promoted cell-cell
adhesion (Supplementary Fig. 6E). These data indicate that GOLPH3 is involved in
cell-cell adhesion, as well as migration and invasion, which are dependent on Golgi
PI(4)P.
PI(4)P-binding ability of GOLPH3 plays an important role in cancer progression
The green fluorescent protein (GFP)-tagged GOLPH3 R90L mutant
(GFP-GOLPH3 R90L), which is unable to bind to PI(4)P (8, 9), also did not localize to
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
23
the Golgi apparatus (Fig. 6A and Supplementary Fig. 6F). In MCF7 cells
overexpressing GFP-GOLPH3, cell-cell adhesion decreased and was not altered by
expression of the GFP-GOLPH3 R90L mutant (Fig. 6A). Cell migration was also
promoted by overexpression of wild-type GOLPH3 in MCF7 cells but not by
overexpression of the GOLPH3 R90L mutant (Fig. 6B). In MDA-MB-231 cells,
expression of the GOLPH3 R90L mutant did not affect cell-cell adhesion or invasion
(Fig. 6C and D), whereas expression of wild-type GOLPH3 abolished cell-cell adhesion
and enhanced invasion (Fig. 6C and D). This enhanced invasion ability in
GOLPH3-expressing cells was abolished by PI4KIII� depletion (Fig. 6E).
Additionally, PI(4)P levels in the Golgi apparatus increased in GOLPH3
wt-expressing MCF-7 and MDA-MB-231 cells but not GOLPH3 R90L-expressing
MCF-7 and MDA-MB-231 cells (Fig. 7A-D). However, in GOLPH3 wt-expressing
MCF7 cells, upregulation of SNAI2 expression was not observed (Fig. 7E). In contrast,
downregulation of SNAI2 expression was observed in GOLPH3 R90L-expressing
MCF7 cells (Fig. 7E). This finding suggests that GOLPH3 is necessary but not
sufficient for EMT induction. Finally, we tested the effect of GOLPH3 on cancer
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
24
metastasis in vivo. GOLPH3 wt-expressing MDA-MB-231 cells showed much higher
levels of metastasized to the lung than GFP-expressing cells (Fig. 7F and G). However,
GOLPH3 R90L, which is a mutant with defective lipid binding, lost its metastasizing
activity. These data indicate that GOLPH3 functions as a regulator of cancer metastasis
in breast cancer through interaction with Golgi PI(4)P.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
25
Discussion
In this study, we have shown that PI(4)P in the Golgi apparatus is a key
regulator of cancer progression. In the Golgi apparatus, SAC1 and PI4KIII� are
predominantly involved in PI(4)P metabolism. However, except for its role in the
regulation of lipid homeostasis and sphingolipid biosynthesis, little is known about the
specific role of SAC1-dependent dephosphorylation of PI(4)P (29-31). We showed
that PI4KIII� and SAC1 expression and PI(4)P levels in the Golgi apparatus were
correlated with cancer progression. According to the NextBio database
(www.nextbio.com), SAC1 and PI4KIII� mRNA expression are correlated with many
cancers, including breast cancer (Supplementary Tables 1 and 2). These results support
our hypothesis that PI(4)P at the Golgi apparatus regulates cancer progression.
The Golgi apparatus is a central organelle that is involved in protein and lipid
biosynthesis and vesicular trafficking. Newly synthesized proteins and lipids are
glycosylated in the Golgi apparatus. Although aberrant glycosylation and
overexpression of glycolipids have been shown to be common features of malignant
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
26
tumors (32, 33), the biological significance of these changes is not well understood.
PI(4)P is enriched at the Golgi apparatus, where PI(4)P-binding proteins that have been
shown to be involved in vesicular transport, lipid transfer, and glycosylation are located.
(1). A Golgi PI(4)P effector, GOLPH3, is involved in Golgi-plasma membrane transport
and in the glycosylation of proteins (8, 34, 35). In yeast, Vps74, the ortholog of human
GOLPH3, binds directly to Sac1 (36). We found that expression of GOLPH3 wt, but not
GOLPH3 R90L, resulted in an increase in PI(4)P levels in the Golgi apparatus. These
data suggest that GOLPH3 overexpression competes with SAC1 for PI(4)P in the Golgi
membrane, or that GOLPH3 senses PI(4)P levels at the Golgi apparatus, thus regulating
SAC1 or PI4KIII� localization or activity. We observed that PI(4)P regulated GOLPH3
localization to the Golgi apparatus. These result suggest that GOLPH3 is localized to
the Golgi apparatus by increasing PI(4)P, leading to aberrant glycosylation and
overexpression of glycolipids.
EMT is an important process in cancer progression. It is characterized by a
decrease in epithelial markers (e.g., E-cadherin) and upregulation of mesenchymal
markers, including vimentin, SNAI1, and SNAI2, which together result in disruption of
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
27
cell-cell adhesion. However, whether EMT/MET occurs in cancer progression remains
controversial (37, 38). We observed that the membrane localization of cadherins and
�-catenin and the expression of SNAI1 and SNAI2 are changed by PI(4)P generation at
the Golgi apparatus, although decreases in the expression of E-cadherin, which is a
characteristic of complete EMT, were not detected. Malignant cancer cells adopt some
mesenchymal features but retain characteristics of epithelial cells (39-41), a condition
that has been described as incomplete EMT (42). Taken together with the finding that
the amount of PI(4)P increases at the Golgi apparatus during EMT, our data suggest that
Golgi PI(4)P is necessary but not sufficient for EMT.
Several studies have shown that GOLPH3 plays important roles in the
progression of several tumor types. For example, GOLPH3 upregulated and promotes
proliferation and tumorigenicity in breast cancer (15). Recent studies have demonstrated
the clinical significance of GOLPH3 in several cancers (13, 14, 43). A very recent study
reported that GOLPH3 contributes to the migration and invasion of glioma cells through
RhoA expression (44). We found that changes in PI(4)P levels at the Golgi apparatus
resulted in alteration of RhoA activity. RhoA activity is required for initial cell-cell
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
28
adhesion formation (45) and is repressed during EMT (45). Taken together with our
results, these findings suggest that Golgi PI(4)P and GOLPH3 regulate the cytoskeletal
rearrangement of actin through RhoA as well as Golgi-plasma membrane transport
during tumor progression.
In conclusion, our data suggest that PI(4)P levels in the Golgi apparatus play a
critical role in cancer progression by regulating the PI(4)P effector GOLPH3
(Supplementary Fig. 6G). In this study, we showed that changes in Golgi PI(4)P levels
and/or Golgi localization of GOLPH3 were likely to result in changes in cell-cell
adhesion and cell migration/invasion in breast cancer cell lines. Furthermore, we
showed that PI(4)P-binding ability of GOLPH3 was essential for metastasis in vivo.
Therefore, molecules that modulate the PI(4)P level in the Golgi apparatus, such as
SAC1 and PI4KIII�, may represent intriguing anticancer therapeutic targets.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
29
Acknowledgements
We are grateful to T. Oikawa (Department of Molecular Biology, Hokkaido University
Graduate School of Medicine) for the discussions and to Y. Murata and T. Yanagita
(Division of Molecular and Cellular Signaling, Kobe University Graduate School of
Medicine) for the technical advice.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
30
References
1. Vicinanza M, D'Angelo G, Di Campli A, De Matteis MA. Function and
dysfunction of the PI system in membrane trafficking. EMBO J. 2008;27:2457-70.
2. Skwarek LC, Boulianne GL. Great expectations for PIP: phosphoinositides as
regulators of signaling during development and disease. Dev Cell. 2009;16:12-20.
3. Graham TR, Burd CG. Coordination of Golgi functions by
phosphatidylinositol 4-kinases. Trends Cell Biol. 2011;21:113-21.
4. Wang YJ, Wang J, Sun HQ, Martinez M, Sun YX, Macia E, et al.
Phosphatidylinositol 4 phosphate regulates targeting of clathrin adaptor AP-1 complexes
to the Golgi. Cell. 2003;114:299-310.
5. Yeaman C, Ayala MI, Wright JR, Bard F, Bossard C, Ang A, et al. Protein
kinase D regulates basolateral membrane protein exit from trans-Golgi network. Nat
Cell Biol. 2004;6:106-12.
6. Mayinger P. Regulation of Golgi function via phosphoinositide lipids. Semin
Cell Dev Biol. 2009;20:793-800.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
31
7. D'Angelo G, Vicinanza M, Di Campli A, De Matteis MA. The multiple roles
of PtdIns(4)P -- not just the precursor of PtdIns(4,5)P2. J Cell Sci. 2008;121:1955-63.
8. Wood CS, Schmitz KR, Bessman NJ, Setty TG, Ferguson KM, Burd CG.
PtdIns4P recognition by Vps74/GOLPH3 links PtdIns 4-kinase signaling to retrograde
Golgi trafficking. The Journal of cell biology. 2009;187:967-75.
9. Dippold HC, Ng MM, Farber-Katz SE, Lee SK, Kerr ML, Peterman MC, et al.
GOLPH3 bridges phosphatidylinositol-4- phosphate and actomyosin to stretch and
shape the Golgi to promote budding. Cell. 2009;139:337-51.
10. Scott KL, Kabbarah O, Liang MC, Ivanova E, Anagnostou V, Wu J, et al.
GOLPH3 modulates mTOR signalling and rapamycin sensitivity in cancer. Nature.
2009;459:1085-90.
11. Li XY, Liu W, Chen SF, Zhang LQ, Li XG, Wang LX. Expression of the
Golgi phosphoprotein-3 gene in human gliomas: a pilot study. J Neurooncol.
2011;105:159-63.
12. Kunigou O, Nagao H, Kawabata N, Ishidou Y, Nagano S, Maeda S, et al. Role
of GOLPH3 and GOLPH3L in the proliferation of human rhabdomyosarcoma. Oncol
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
32
Rep. 2011;26:1337-42.
13. Li H, Guo L, Chen SW, Zhao XH, Zhuang SM, Wang LP, et al. GOLPH3
overexpression correlates with tumor progression and poor prognosis in patients with
clinically N0 oral tongue cancer. J Transl Med. 2012;10:168.
14. Hua X, Yu L, Pan W, Huang X, Liao Z, Xian Q, et al. Increased expression of
Golgi phosphoprotein-3 is associated with tumor aggressiveness and poor prognosis of
prostate cancer. Diagn Pathol. 2012;7:127.
15. Zeng Z, Lin H, Zhao X, Liu G, Wang X, Xu R, et al. Overexpression of
GOLPH3 Promotes Proliferation and Tumorigenicity in Breast Cancer via Suppression
of the FOXO1 Transcription Factor. Clin Cancer Res. 2012;18:4059-69.
16. Nieto MA. The ins and outs of the epithelial to mesenchymal transition in
health and disease. Annu Rev Cell Dev Biol. 2011;27:347-76.
17. Nieto MA, Cano A. The epithelial-mesenchymal transition under control:
global programs to regulate epithelial plasticity. Semin Cancer Biol. 2012;22:361-8.
18. Furutani M, Tsujita K, Itoh T, Ijuin T, Takenawa T. Application of
phosphoinositide-binding domains for the detection and quantification of specific
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
33
phosphoinositides. Anal Biochem. 2006;355:8-18.
19. Hasegawa J, Tokuda E, Tenno T, Tsujita K, Sawai H, Hiroaki H, et al.
SH3YL1 regulates dorsal ruffle formation by a novel phosphoinositide-binding domain.
J Cell Biol. 2011;193:901-16.
20. Lion M, Bisio A, Tebaldi T, De Sanctis V, Menendez D, Resnick MA, et al.
Interaction between p53 and estradiol pathways in transcriptional responses to
chemotherapeutics. Cell Cycle. 2013;12:1211-24.
21. Qiao M, Sheng S, Pardee AB. Metastasis and AKT activation. Cell Cycle.
2008;7:2991-6.
22. Ye Y, Jin L, Wilmott JS, Hu WL, Yosufi B, Thorne RF, et al. PI(4,5)P2
5-phosphatase A regulates PI3K/Akt signalling and has a tumour suppressive role in
human melanoma. Nature communications. 2013;4:1508.
23. Yilmaz M, Christofori G. EMT, the cytoskeleton, and cancer cell invasion.
Cancer Metastasis Rev. 2009;28:15-33.
24. Hammond GR, Schiavo G, Irvine RF. Immunocytochemical techniques reveal
multiple, distinct cellular pools of PtdIns4P and PtdIns(4,5)P(2). The Biochemical
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
34
journal. 2009;422:23-35.
25. Burk U, Schubert J, Wellner U, Schmalhofer O, Vincan E, Spaderna S, et al. A
reciprocal repression between ZEB1 and members of the miR-200 family promotes
EMT and invasion in cancer cells. EMBO Rep. 2008;9:582-9.
26. Ruike Y, Imanaka Y, Sato F, Shimizu K, Tsujimoto G. Genome-wide analysis
of aberrant methylation in human breast cancer cells using methyl-DNA
immunoprecipitation combined with high-throughput sequencing. BMC Genomics.
2010;11:137.
27. Pishvaian MJ, Feltes CM, Thompson P, Bussemakers MJ, Schalken JA, Byers
SW. Cadherin-11 is expressed in invasive breast cancer cell lines. Cancer Res.
1999;59:947-52.
28. Nieman MT, Prudoff RS, Johnson KR, Wheelock MJ. N-cadherin promotes
motility in human breast cancer cells regardless of their E-cadherin expression. The
Journal of cell biology. 1999;147:631-44.
29. Cleves AE, Novick PJ, Bankaitis VA. Mutations in the SAC1 gene suppress
defects in yeast Golgi and yeast actin function. The Journal of cell biology.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
35
1989;109:2939-50.
30. Liu Y, Boukhelifa M, Tribble E, Bankaitis VA. Functional studies of the
mammalian Sac1 phosphoinositide phosphatase. Advances in enzyme regulation.
2009;49:75-86.
31. Whitters EA, Cleves AE, McGee TP, Skinner HB, Bankaitis VA. SAC1p is an
integral membrane protein that influences the cellular requirement for phospholipid
transfer protein function and inositol in yeast. The Journal of cell biology.
1993;122:79-94.
32. Migita T, Inoue S. Implications of the Golgi apparatus in prostate cancer. The
international journal of biochemistry & cell biology. 2012;44:1872-6.
33. Durrant LG, Noble P, Spendlove I. Immunology in the clinic review series;
focus on cancer: glycolipids as targets for tumour immunotherapy. Clinical and
experimental immunology. 2012;167:206-15.
34. Schmitz KR, Liu J, Li S, Setty TG, Wood CS, Burd CG, et al. Golgi
localization of glycosyltransferases requires a Vps74p oligomer. Dev Cell.
2008;14:523-34.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
36
35. Tu L, Tai WC, Chen L, Banfield DK. Signal-mediated dynamic retention of
glycosyltransferases in the Golgi. Science. 2008;321:404-7.
36. Wood CS, Hung CS, Huoh YS, Mousley CJ, Stefan CJ, Bankaitis V, et al.
Local control of phosphatidylinositol 4-phosphate signaling in the Golgi apparatus by
Vps74 and Sac1 phosphoinositide phosphatase. Molecular biology of the cell.
2012;23:2527-36.
37. Tarin D, Thompson EW, Newgreen DF. The fallacy of epithelial mesenchymal
transition in neoplasia. Cancer Res. 2005;65:5996-6000; discussion -1.
38. Drasin DJ, Robin TP, Ford HL. Breast cancer epithelial-to-mesenchymal
transition: examining the functional consequences of plasticity. Breast Cancer Res.
2011;13:226.
39. Wicki A, Lehembre F, Wick N, Hantusch B, Kerjaschki D, Christofori G.
Tumor invasion in the absence of epithelial-mesenchymal transition:
podoplanin-mediated remodeling of the actin cytoskeleton. Cancer cell. 2006;9:261-72.
40. Hidalgo-Carcedo C, Hooper S, Chaudhry SI, Williamson P, Harrington K,
Leitinger B, et al. Collective cell migration requires suppression of actomyosin at
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
37
cell-cell contacts mediated by DDR1 and the cell polarity regulators Par3 and Par6.
Nature cell biology. 2011;13:49-58.
41. Nagaharu K, Zhang X, Yoshida T, Katoh D, Hanamura N, Kozuka Y, et al.
Tenascin C induces epithelial-mesenchymal transition-like change accompanied by
SRC activation and focal adhesion kinase phosphorylation in human breast cancer cells.
The American journal of pathology. 2011;178:754-63.
42. Christiansen JJ, Rajasekaran AK. Reassessing epithelial to mesenchymal
transition as a prerequisite for carcinoma invasion and metastasis. Cancer Res.
2006;66:8319-26.
43. Zhou JX, Xu T, Qin R, Yan Y, Chen C, Chen YY, et al. Overexpression of
Golgi phosphoprotein-3 (GOLPH3) in glioblastoma multiforme is associated with
worse prognosis. J Neuro-Oncol. 2012;110:195-203.
44. Zhou X, Zhan W, Bian W, Hua L, Shi Q, Xie S, et al. GOLPH3 regulates the
migration and invasion of glioma cells though RhoA. Biochemical and biophysical
research communications. 2013;433:338-44.
45. Yilmaz M, Christofori G. EMT, the cytoskeleton, and cancer cell invasion.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
38
Cancer Metastasis Rev. 2009;28:15-33.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
39
Figure Legends
Figure 1. Phosphoinositide phosphatase knockdown.
MCF7 cells were transfected with siRNA targeting the indicated phosphoinositide
phosphatases. Seventy-two hours after transfection, the cells were fixed and stained
with phalloidin. Scale bar = 20 �m.
Figure 2. SAC1 knockdown decreases cell-cell adhesion and promotes cell
migration.
(A) The indicated siRNA-transfected MCF7 cells were lysed and immunoblotted with
the indicated antibodies.
(B) The indicated siRNA-transfected MCF7 cells were fixed and stained with
anti-E-cadherin antibody and phalloidin. Scale bar = 20 �m.
(C) The indicated siRNA-transfected MCF7 cells were fractionated into Triton
X-100-soluble and Triton X-100-insoluble fractions. Each fraction was
immunoblotted with the indicated antibodies.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
40
(D) The indicated siRNA-transfected MCF7 cells were analyzed using a migration
assay. The data shown are the mean (SEM) values of three independent experiments.
*, P < 0.001; **, P < 0.005. Scale bar = 100 �m.
(E) MCF7 cells were transfected with the indicated-siRNAs. RhoA activity was
measured using a G-LISA RhoA activation assay. The data shown are the mean
(SEM) values of four independent experiments. *, P < 0.01.
(F) MCF7 cells were transfected with indicated-siRNAs. The SNAI2 mRNA level was
quantified by real-time PCR. The data shown are the mean (SEM) values of three
independent experiments. *, P < 0.003.
(G) Golgi PI(4)P levels were quantified in the indicated siRNA-transfected MCF7 cells.
The data shown are the mean (SEM) values from more than 100 cells measured. *,
P < 0.001.
(H) The indicated siRNA-transfected MCF7 cells were fixed and stained with the
anti-GM130 antibody and Alexa Fluor 647-labeled Fapp1 2 × PH domain. Scale bar
= 20 �m.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
41
Figure 3. EMT-like phenotypes are affected by PI(4)P levels in the Golgi
apparatus.
(A) MCF7 cells, not treated or treated with 2 ng/ml TGF� + 5 ng/ml TNF�, were fixed
and stained with anti-GM130 antibody together with Alexa Fluor 647-labeled
Fapp1 2× PH domain. Scale bar = 20 �m.
(B) Quantification of Golgi PI(4)P levels in MCF7 cells before and after EMT induction.
The data shown are the mean (SEM) values from more than 100 cells. *, P < 0.001.
(C) Each breast cancer cell line was fixed and stained with anti-GM130 antibody and
the Alexa Fluor 647-labeled Fapp1 2 × PH domain. Scale bar = 20 �m.
(D) Golgi PI(4)P levels were quantified in breast cancer cell lines. The data shown are
the mean (SEM) values from more than 100 cells measured. *, P < 0.001.
(E) Human breast cancer tissue lysates were immunoblotted with the antibodies against
PI4KIII� and SAC1. Lysates isolated from five non-metastatic and metastatic breast
cancer subjects were tested.
Figure 4. PI4KIII� knockdown induces cell-cell adhesion and suppression of
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
42
invasion.
(A) The indicated siRNA-transfected MDA-MB-231 cells were fixed and stained with
anti-GM130 antibody and Alexa Fluor 647-labeled Fapp1 2 × PH domain. Scale bar
= 20 �m.
(B) Golgi PI(4)P levels were quantified in control- or PI4KIII�-silenced MDA-MB-231
cells. The data represent the mean (SEM) values from more than 100 cells. *, P <
0.001.
(C) The indicated siRNA-transfected MDA-MB-231 cells were fixed and stained with
anti-�-catenin, anti-cadherin-11, and phalloidin. Scale bar = 20 �m.
(D) Control- or PI4KIII�-siRNA-transfected MDA-MB-231 cells were lysed and
immunoblotted with the indicated antibodies.
(E) The indicated siRNA-transfected MDA-MB-231 cells were seeded onto Matrigel
invasion chambers. Eighteen hours after seeding, the cells were fixed and stained
with phalloidin (lower panels). The number of invasive cells that passed through
each Matrigel chamber was counted (upper panel). The data represent the mean
(SEM) values of three independent experiments. *, P < 0.001. Scale bar = 100 �m.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
43
(F) MDA-MB-231 cells were transfected with the indicated-siRNAs. RhoA activity was
measured using a G-LISA RhoA activation assay. The data shown are the mean
(SEM) values of four independent experiments. *, P < 0.02; **, P < 0.002.
(G) MDA-MB-231 cells were transfected with indicated-siRNAs. The SNAI1 mRNA
level was quantified by real-time PCR. The data shown are the mean (SEM) of three
independent experiments. *, P < 0.001; **, P < 0.03.
Figure 5. GOLPH3 is required for regulation of cell-cell adhesion and for
migration/invasion ability.
(A) MDA-MB-231 cells were transfected with control or GOLPH3 siRNA and then
fixed and stained with anti-GOLPH3 and anti-GM130 antibodies, and phalloidin.
Scale bar = 20 �m.
(B) The indicated siRNA-transfected MDA-MB-231 cells were lysed and
immunoblotted with GOLPH3 antibodies.
(C) The indicated siRNA-transfected MDA-MB-231 cells were fixed and stained with
anti-�-catenin antibody and phalloidin. Scale bar = 20 �m.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
44
(D) The indicated siRNA-transfected MDA-MB-231 cells were seeded onto Matrigel
invasion chambers. Eighteen hours after seeding, the cells were fixed and stained
with phalloidin (upper panels). The number of invasive cells that passed through
each Matrigel chamber was counted (lower panel). The data shown are the mean
(SEM) values of three independent experiments. *, P < 0.001. Scale bar = 100 �m.
(E) The indicated siRNA-transfected MCF7 cells were analyzed in a migration assay.
The data shown are the mean (SEM) values of four independent experiments. *, P
< 0.01; **, P < 0.005. Scale bar = 100 �m.
Figure 6. Binding of PI(4)P to GOLPH3 is required for regulation of cell-cell
adhesion and cell migration/invasion.
(A) GFP (mock), GFP-tagged wild-type GOLPH3 (GOLPH3 wt), or the GFP-tagged
GOLPH3 R90L mutant (GOLPH3 R90L) was overexpressed in MCF7 cells. Cells
were stained with an anti-E-cadherin antibody and phalloidin. Scale bar = 20 �m.
(B) Mock, GOLPH3 wt, or GOLPH3 R90L-expressing MCF7 cells were used for a
migration assay. The data shown are the mean (SEM) values of six independent
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
45
experiments. *, P < 0.0005. Scale bar = 100 �m.
(C) Mock, GOLPH3 wt or GOLPH3 R90L-expressing MDA-MB-231 cells were
stained with an anti-�-catenin antibody and phalloidin. Scale bar = 20 �m.
(D) Mock, GOLPH3 wt, or GOLPH3 R90L mutant-expressing MDA-MB-231 cells
were seeded onto Matrigel invasion chambers. Eighteen hours after seeding, the
cells were fixed and stained with phalloidin. The number of invasive cells that
passed through the Matrigel chambers was counted. The data shown are the mean
(SEM) values of four independent experiments. *, P < 0.005; **, P < 0.0005. Scale
bar = 100 �m.
(E) Mock or GOLPH3 wt-expressing MDA-MB-231 cells were transfected with the
indicated siRNAs. Forty-eight hours after transfection, the cells were
serum-starved and seeded onto Matrigel invasion chambers. Eighteen hours after
seeding, the cells were fixed and stained with phalloidin. The number of invasive
cells that passed through the Matrigel chambers was counted. The data shown are
the mean (SEM) of four independent experiments. *, P < 0.0005. Scale bar = 100
�m.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
46
Figure 7. Binding of PI(4)P to GOLPH3 is required for metastasis in vivo.
(A)GFP, GOLPH3 wt, or GOLPH3 R90L-expressing MCF7 cells were transfected with
the indicated siRNAs and then fixed and stained with anti-GM130 and the Alexa
Fluor 647-labeled Fapp1 2 × PH domain. Scale bar = 20 �m.
(B) Golgi PI(4)P levels were quantified in GFP, GOLPH3 wt, or GOLPH3
R90L-expressing MCF7 cells. Significant increases were observed following
GOLPH3 wt expression. The data shown are the mean (SEM) values from more
than 100 cells. *, P < 0.001.
(C) GFP, GOLPH3 wt, or GOLPH3 R90L-expressing MDA-MB-231 cells were
transfected with the indicated siRNAs and then fixed and stained with anti-GM130
antibody and the Alexa Fluor 647-labeled Fapp1 2 × PH domain. Scale bar = 20 �m.
(D)Golgi PI(4)P levels were quantified in GFP, GOLPH3 wt, or GOLPH3
R90L-expressing MDA-MB-231 cells. The data shown are the mean (SEM) values
from more than 100 cells. *, P < 0.001.
(E) Total RNA was isolated from GFP or GFP-tagged GOLPH3 stable MCF7 cells. The
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
47
SNAI2 mRNA level was quantified by real-time PCR. The data shown are the mean
(SEM) values of three independent experiments. *, P < 0.0001.
(F) Lung surface metastatic foci were counted. The data shown are the mean (SEM)
values from seven (mock and GOLPH3 wt) or eight (GOLPH3 R90L) mice. *, P <
0.04; **, P < 0.02.
(G)Representative images of the lungs with metastatic foci.
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
cont
rol
PT
EN
INP
P4A
INP
P4B
TM
EM
55A
TM
EM
55B
SA
C1
INP
P5E
INP
P5D
OC
RL
INP
PL1
SK
IP
PIB
5PA
INP
P5B
SY
NJ1
SY
NJ2
INP
P5F
FIG
4
SB
F1
SB
F2
MT
MM
TM
R1
MT
MR
2
Figure 1
*
**
*
*
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
Figure 2
A
E-cadherin
β-catenin
SAC1
GAPDH
siRNA cont
rol
SA
C1-
1S
AC
1-2
cont
rol
SA
C1-
1si
RN
A
E-cadherin actinB C
D E F
GM130 Fapp 2 × PH
cont
rol
SA
C1-
1si
RN
A
E-cadherin
β-catenin
SAC1
GAPDH
siRNA cont
rol
SA
C1-
1
SA
C1-
2
cont
rol
SA
C1-
1
SA
C1-
2
TritonX-100 soluble insoluble
control SAC1-1 SAC1-2
siRNA
PI4
P le
vel (
Fap
p1 2
× P
H/ G
M13
0)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
* *
0
5
10
15
20
25
30
35
40
45
50
cont SAC1-1 SAC1-2siRNA
mig
ratio
n ar
ea (
%) * **
control SAC1-1siRNA
0 h
24 h
G
0
1.0
0.2
1.4
0.4
0.6
1.2
0.8
1.6
1.8
2.0
SN
AI2
mR
NA
rel
ativ
e ex
pres
sion
control SAC1-1 SAC1-2
siRNA
* *
H
control SAC1-1 SAC1-2
siRNA
0
1.0
0.2
0.4
0.6
0.8
activ
e R
hoA
/ tot
al R
hoA * *
1.2
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
Figure 3
A BGM130 Fapp 2 × PHco
ntro
lT
GFβ +
TN
Fα
PI4
P le
vel (
Fap
p1 2
× P
H/ G
M13
0)
0
0.1
0.2
0.9
0.4
0.6
0.8
0.3
0.5
0.7
control TGFβ + TNFα
*
D
PI4
P le
vel (
Fap
p1 2
× P
H/ G
M13
0)
0
1.0
0.2
1.4
0.4
0.6
1.2
0.8
1.6
* *
MCF7
T47D
MDA-M
B-231
Hs578
t
**
GM130 Fapp 2 × PH
MC
F7
T47
DM
DA
-MB
-231
Hs5
78t
cell
linesC
E
SAC1
actin
PI4KIIIβ
non-metastatic metastatic
stage I IIA IIIA IIIC IV
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
Figure 4
A
C
D
E
F
Bsi
RN
Aco
ntro
lP
I4K
IIIβ-1
GM130 Fapp 2 × PH
cadherin-11
β-catenin
PI4KIIIβ
GAPDH
siRNA
cont
rol
PI4
KIIIβ-1
PI4
KIIIβ-2
G
cadherin-11 β-catenin actin
siR
NA
cont
rol
PI4
KIIIβ-2
control PI4KIIIβ-1 PI4KIIIβ-2
siRNA
PI4
P le
vel (
Fap
p1 2
× P
H/ G
M13
0)
0
1
0.2
1.2
0.4
1.4
0.6
0.8
* *
control PI4KIIIβ-1 PI4KIIIβ-2
siRNA
No.
of i
nvat
ed c
ells
0
1600
200
400
600
800
1000
1200
1400 * *
PI4KIIIβ-1controlsiRNA
control PI4KIIIβ-1 PI4KIIIβ-2
siRNA
0
2.5
0.5
1.0
1.5
2.0ac
tive
Rho
A/ t
otal
Rho
A* * *
0
1.0
0.2
0.4
0.6
1.2
0.8
SN
AI1
mR
NA
rel
ativ
e ex
pres
sion * *
control PI4KIIIβ-1 PI4KIIIβ-2
siRNA
*
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
Figure 5
A
CD
B
GOLPH3
GAPDH
siRNA GO
LPH
3-1
GO
LPH
3-2
cont
rol
β-catenin actin
siR
NA
GO
LPH
3-1
cont
rol
GOLPH3 actin GM130 mergesi
RN
Aco
ntro
lP
I4K
IIIβ-1
0
200
400
600
800
1000
1200
1400
1600
cont GOLPH3-1 GOLPH3-2
siRNA
No.
of i
nvat
ed c
ells
* *
GOLPH3-1controlsiRNA
siRNA
0 h
24 h
control controlSAC1-1 SAC1-1
control GOLPH3
0
15
10
25
20
35
30
40
5
mig
ratio
n ar
ea (
%) *
**
siRNA
cont
rol +
cont
rol
cont
rol +
GOLP
H3
SAC1-1
+ co
ntro
l
SAC1-1
+ GOLP
H3
SAC1-2
+ co
ntro
l
SAC1-2
+ GOLP
H3
E
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
Figure 6
AB
C
mock wt R90L
GOLPH3
No.
of i
nvat
ed c
ells
0
1600
200
400
600
800
1000
1200
1400
1800
* **
mock wt R90L
GOLPH3
0
60
10
70
20
80
30
40
50
mig
ratio
n ar
ea (
%)
* *GFP E-cadherin actin
moc
kG
OLP
H3
wt
GO
LPH
3 R
90L
0 h
24 h
mock wt R90LGFP
mock wt R90LGFP
D
GFP β-catenin actin
moc
kG
OLP
H3
wt
GO
LPH
3 R
90L
PI4KIIIβ-1controlsiRNA
moc
kG
OLP
H3
wt
GF
P
mock GOLPH3 wt
No.
of i
nvat
ed c
ells
0
2400
300
600
900
1200
1500
1800
2100
3000
2700
cont
rol
cont
rol
siRNA
GFP
PI4KIIIβ
-1
PI4KIIIβ
-1
PI4KIIIβ
-2
PI4KIIIβ
-2
* ** *E
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
Figure 7
A B
C
GFP Fapp 2 × PHGM130
moc
kG
OLP
H3
wt
GO
LPH
3 R
90L
mock wt R90L
GOLPH3
PI4
P le
vel (
Fap
p1 2
× P
H/ G
M13
0)
0
1.0
0.2
1.4
0.4
0.6
1.2
0.8
* *1.6
1.8
GFP Fapp 2 × PHGM130
moc
kG
OLP
H3
wt
GO
LPH
3 R
90L
D
PI4
P le
vel (
Fap
p1 2
× P
H/ G
M13
0)
0
1.0
0.2
1.4
0.4
0.6
1.2
0.8
* *
mock wt R90L
GOLPH3
E
0
1.0
0.2
0.4
0.6
1.2
0.8
SN
AI2
mR
NA
rel
ativ
e ex
pres
sion
mock wt R90L
GOLPH3
*
F
G
moc
k wt
R90L
0
10
20
30
40
50
GOLPH3
Num
ber
of lu
ng s
urfa
ce m
etas
tatic
foci
* **
mock GOLPH3 wt GOLPH3 R90L
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441
Published OnlineFirst April 4, 2014.Cancer Res Emi Tokuda, Toshiki Itoh, Junya Hasegawa, et al. human breast cancerregulates cell-cell adhesion and invasive cell migration in Phosphatidylinositol 4-phosphate in the Golgi apparatus
Updated version
10.1158/0008-5472.CAN-13-2441doi:
Access the most recent version of this article at:
Material
Supplementary
http://cancerres.aacrjournals.org/content/suppl/2014/04/04/0008-5472.CAN-13-2441.DC1
Access the most recent supplemental material at:
Manuscript
Authoredited. Author manuscripts have been peer reviewed and accepted for publication but have not yet been
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
.http://cancerres.aacrjournals.org/content/early/2014/04/04/0008-5472.CAN-13-2441To request permission to re-use all or part of this article, use this link
on July 23, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from
Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on April 4, 2014; DOI: 10.1158/0008-5472.CAN-13-2441