Supporting Online Materials
Materials and Methods
Cell culture
MCF-7 and other human breast cancer cell lines were obtained from the Breast Center at
Baylor College of Medicine or American Type Culture Collection. 4T1 breast cancer cells were
provided by Fred Miller. Cells were routinely maintained in Dulbecco’s modified Eagle’s
medium (DMEM) supplemented with 10% fetal bovine serum, 2 mM glutamine, 50 IU/ml of
penicillin, and 50 µg/ml of streptomycin. Cells were kept at 37oC in a humidified incubator with
5% CO2. The proteasome inhibitor MG-132 and the IKK small-molecule inhibitors BMS-345541
and BAY-117082 were purchased from Calbiochem (Gibbstown, NJ). For cell proliferation
assays, cells were first incubated separately with BMS-345541 (5 M) and BAY-117082 (20
M). At different time points, MTT assays were conducted. For the experiment with the
translation inhibitor cycloheximide (10 µg/ml), the same procedure was followed. Control cells
were incubated with the vehicle dimethylsulfoxide alone.
FOXC1-knockdown cells
FOXC1 shRNAs and a control shRNA which does not match any known cDNA were from
Sigma. Cells were stably transfected with the FOXC1 or control shRNA construct and selected
with 5 μg/ml puromycin, as previously described (1).
FOXC1-overexpressing cells
1
The full-length human FOXC1 cDNA in the viral vector pLNCX2 (Clontech, Mountain View,
CA) was stably transduced into breast cancer cells, as previously described (1). Stable cell lines
were selected with 800 μg/ml G418. Pooled populations were used for experiments.
Microarray data analysis
Publicly available gene expression microarray data from breast cancer cell lines were
downloaded
from
http://cancer.lbl.gov/breastcancer/data.Php
http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE12777
(3)
(2)
were
and
from
analyzed
using
Genespring GX 11.5 software (Agilent Technologies). Probe-set-based gene expression data
obtained from Affymetrix gene chips were generated using Robust Multi-chip Averaging (RMA)
algorithm. Log2-normalized signal intensity values were directly imported into the GeneSpring
software platform and underwent pre-processing for background correction and normalization
followed by baseline transformation to median of all samples. Average relative mRNA levels
(mean log2 signal intensity) for Pin1 in basal-like and luminal subgroups were compared through
t-test for fold changes, using “luminal” as a reference group. Overall survival (4, 5) and diseasefree survival (5, 6) curves associated with high and low Pin1 mRNA levels were generated using
Kaplan Meier methods and compared using the log-rank test.
Statistical significance was
defined as P < 0.05.
Transient transfection
Breast cancer cells were grown for 48 h to reach 80% confluence before transfection. For cotransfection experiments, 0.1 µg DNA of Pin1 promoter-luc or NF-B-luc (7) (Promega,
Madison, WI) reporter construct and 1 g of FOXC1 or NF-B p65 vector were transfected with
2
Lipofectamine 2000 (Invitrogen, Carlsbad, CA) into cells in 60-mm dishes. The transfected cells
were cultured for 24 h or 48 h (if siRNA was also co-transfected). Transfected cells were then
washed twice with PBS and harvested in 200 l of reporter lysis buffer (Promega). Twenty
nanograms of a -galactosidase expression vector pSV--Gal (Promega) were co-transfected as
an internal control. Luciferase and -galactosidase assays were performed using Promega
reporter assay reagents and the GloMax Multi-detection system.
Real-time quantitative reverse transcription-PCR
Total RNA was extracted from control and FOXC1-overexpressing MDA-MB-231 cells using
the Qiagen RNeasy mRNA preparation kit (Qiagen) according to the manufacturer’s instructions.
All RNA extraction was done in a designated sterile laminar flow hood with RNase-free
plasticware. RNA was quantified and assessed for purity by UV spectrophotometry. RNA
extraction, reverse transcription-PCR (RT-PCR) assay setup, and post–RT-PCR product analysis
were carried out in separate designated rooms to prevent cross-contamination. One microgram of
total mRNA was reverse transcribed using qScript TM cDNA SuperMix (Quanta Biosciences).
The qRT assay was done using an iCycler iQ Real-Time Thermocycler (Bio-Rad Laboratories).
A 5ul cDNA template was used for each reaction in 96-well plates (Fisher Scientific) along with
primers and Perfecta TM SYBR® Green FastMix (Quanta Biosciences). Samples were amplified
with a precycling hold at 95oC for 10 min, followed by specific cycles of denaturation at 95oC
for 1 min, annealing at 58oC for 1 min and extension at 72oC for 1 min. A standard curve was
generated using the threshold cycle (Ct) of nine serial dilutions of plasmid templates (108-100
copies). The Ct of each sample was interpolated from the standard curve, and the number of
3
mRNA copies was calculated by the iCycler iQ Real-Time Detection System Software. Each
assay was done in duplicate. Primers used for all the genes are listed in Supplementary Table 1.
Immunoblot analysis and immunoprecipitation
Whole-cell lysates for western blotting were generated by cell lysis buffer (50 mM Tris-HCl, pH
7.4, 150 mM NaCl, 2 mM EDTA, 1% NP-40, 10% glycerol) supplemented with a protease
inhibitor cocktail (Sigma, St Louis, MO). Equal amounts of protein were separated by 10% SDSPAGE and then transferred onto a nitrocellulose membrane. The remaining steps were conducted
according to a standard immunoblotting protocol (8). Immunoblotting was done with polyclonal
antibodies against p65, Pin1, and FOXC1 (1:200; Santa Cruz Biotechnology); and with
polyclonal antibodies against phospho-p65, p50, and IκBα (1:1000; Cell Signal). Anti-β actin
(Sigma) was used at a 1:10000 dilution. After the primary antibody incubation, the membrane
was again washed with PBST three times (5 min each) and then incubated with a horseradish
peroxidase (HRP)-linked secondary antibody (Amersham, Piscataway, NJ) at a dilution of
1:4000 in blocking solution. The membrane was washed and bands were visualized using
chemiluminescence assays. For immunoprecipitation, cell lysates of 500 µg were pre-cleared by
protein-G agarose beads (Zymed Laboratory, San Francisco, CA) and then incubated with
specific antibodies at a 1:100 dilution overnight at 4oC. The beads were washed with the above
lysis buffer three times and resuspended in protein sample buffer before the immunoprecipitated
protein was subjected to immunoblotting.
To examine p65 protein stability, control and FOXC1-overexpressing MDA-MB-231 cells were
transfected with a ubiquitin plasmid, and treated with 10 M MG-132 or vehicle for 16 h. Then
4
total protein was immunoprecipitated with the p65 antibody and then subjected to anti-ubiquitin
immunoblotting. In addition, the same cells were treated with the protein translation inhibitor
cycloheximide (10 g/ ml) and harvested at different time points. Results of anti-p65
immunoblotting were quantified by densitometry of the immunoblots.
Immunofluorescence staining
Cells were cultured in 8-well chamber slides (Lab-Tek) at 50% to 60% confluency. Cells were
fixed with 4% paraformaldehyde for 15 min and then permeabilized with PBS containing 0.2%
Triton X-100 for another 15 min. Slides were blocked by 5% BSA for 30 min and incubated with
a primary antibody (1:100) at room temperature for 1 h. Slides underwent three 5-min washings
with PBS. Then, cells were stained with the secondary antibody (Alexa Fluor 488 goat antirabbit IgG, 1:500, Invitrogen) for another 30 min. Again slides underwent three 5-min washings
with PBS. Coverslips were mounted onto glass slides using Vectashield mounting medium
containing DAPI. Cells were observed under a high-resolution microscope. As a negative control,
primary specific antibodies were replaced with normal rabbit IgG.
IPA signaling pathway analysis
The Richardson et al. data set (9) was subjected to Ingenuity Pathway Analysis (IPA, Ingenuity
Systems, Redwood City, CA). Briefly, global gene expression profiles of all breast cancer
samples were analyzed according to their molecular subgroup (basal-like, HER2 or luminal) with
respect to their association with a specific canonical pathway in the Ingenuity Pathways
Knowledge Base. The significance of the association between the average global gene
expression profile associated with a particular subgroup and the specific canonical pathway was
5
measured in two ways: 1) A ratio of the average number of genes from a particular subgroup that
mapped to the pathway divided by the total number of genes (having probe representation on the
microarray platform) assigned to the canonical pathway was calculated. 2) Fischer’s exact test
was used to calculate a p-value determining the likelihood that the association between the genes
in any particular subgroup and the canonical pathway could be explained by chance alone. The
negative log of this p-value is the Impact Factor.
NF-κB transcription factor TransAM assay
The activity of NF-κB family proteins was measured using the TransAM NF-κB ELISA kit
(Active Motif, Carlsbad, CA ) according to the manufacturer’s instructions. Briefly, isolated
nuclear proteins were resuspended in extraction buffer (20 mM Hepes pH 7.9, 0.4 M NaCl, 1
mM EDTA). The supernatant containing the nuclear extracts was retained after a second
centrifugation. Samples (10 μg) were added in triplicate to 96-well plates coated with an
oligonucleotide that contains a consensus binding sequence for NF-B components. After 1 h
incubation at room temperature, primary antibodies of distinct NF-B components were added to
the plates, followed by washing steps; subsequent addition of HRP-conjugated secondary
antibody produced a sensitive colorimetric readout quantified by spectrophotometry at the 450nm wavelength with a reference wavelength of 655 nm.
Cell proliferation assay
Cell viability was assessed by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (MTT)
assay. Cells were seeded in 24-well plates at 20-30% confluence and the MTT assay was
performed one, two, three and four days after treatment. For each assay, 50 μl of MTT (5 mg/ml)
6
was added to each well and cells were incubated at 37˚C for an additional 4 h. After
centrifugation, the supernatant was carefully aspirated and 300 μl of DMSO (Sigma) was added
to each well. Immediately after resolubilization, all plates were scanned at 575 nm on a
microplate reader. The absorbance (A) value was indicative of the number of live cells.
Cell cycle analysis
Wild-type and IKK/IKK double–knockout mouse embryonic fibroblasts and their transient
transfectants (2 ×105) were plated in 60-mm dishes and grown to 70% to 80% confluence with
normal culture medium. Cells were harvested and fixed with 3 mL of ice-cold 70% ethanol
overnight. The fixed cells were then centrifuged at 3,000 rpm for 3 min, washed with PBS, and
centrifuged again. Cell pellets were suspended in PBS containing 50µg/mL RNase and incubated
for 1 h at 37°C. DNA was stained with propidium iodide (50µg/mL) for at least 1 hour at 4°C.
The DNA content was determined by flow cytometry (BD FACSCalibur) and cell phases were
analyzed with the BD CellQuestPro software.
Cell migration and invasion assay
To assess cell motility, 2.5 ×104 cells per well in serum-free RPMI medium were placed in the
upper chamber. RPMI plus 10% FBS was added in the lower chamber as a source of
chemoattractants. Cell migration was analyzed by transwell chamber assays. Cell invasion assays
were performed using BD BioCoat™ Matrigel™ Invasion Chambers. Cells were allowed to
migrate or invade through an 8-M pore membrane for 24 hours at 37˚C. Non-migratory or noninvasive cells in the upper chamber were later removed with a cotton-tip applicator. Migratory or
invasive cells on the lower surface were fixed with methanol and stained with hematoxylin,
7
followed by counting under a light microscope. The membrane was then mounted onto a
microscope slide and the migratory cells were counted in 5 different areas using a light
microscope.
Statistical Analysis
The results are presented as mean ± standard deviation (SD) of three independent experiments,
each performed in duplicate or triplicate, unless otherwise indicated. The Student’s t-test was
used to calculate differences between the various experimental groups. The difference was
considered statistically significant with P < 0.05.
References
1.
P. S. Ray et al., Cancer Res 70, 3870 (May 15, 2010).
2.
R. M. Neve et al., Cancer Cell 10, 515 (Dec, 2006).
3.
K. P. Hoeflich et al., Clin Cancer Res 15, 4649 (Jul 15, 2009).
4.
M. J. van de Vijver et al., N Engl J Med 347, 1999 (Dec 19, 2002).
5.
Y. Pawitan et al., Breast Cancer Res 7, R953 (2005).
6.
A. V. Ivshina et al., Cancer Res 66, 10292 (Nov 1, 2006).
7.
A. Ryo et al., Mol Cell Biol 22, 5281 (Aug, 2002).
8.
Y. Qu et al., Breast Cancer Res Treat 121, 311 (Jun, 2009).
9.
A. L. Richardson et al., Cancer Cell 9, 121 (Feb, 2006).
8
Supplementary Table 1. Primers of real-time RT-PCR
gene
forward
reverse
Pin1
5’-TGGGTGCCTTCAGCAGAGGTCAG-3’
5’-CCGGAATCCGTGAACACGGGC-3’
IB
5’-AAGTGATCCGCCAGGTGAAGGGA-3’
5’-AGCCAAGTGGAGTGGAGTCTGCT-3’
IL-6
5′-TCCAGAACAGATTTGAGAGTAGTG-3′
5′ -GCATTTGTGGTTGGGTCAGG-3′
P65
5’-TGTGTGAAGAAGCGGGACCTGG-3’
5’-CCCCACGCTGCTCTTCTATAGGA-3’
actin
5’-GAGCCTCGCCTTTGCCGATCCG-3’
5’-CCTTGCACATGCCGGAGCCGT-3’
9
Supplementary Figure Legends
Figure S1. FOXC1 induces NF-κB activity in breast cancer cells. (A) Expression of NF-κB
and its inhibitor in MCF-7 cells that overexpressed FOXC1 or the vector was examined by
immunoblotting. (B) Expression of p65 in 4T1 breast cancer cells that stably expressed control
or FOXC1 shRNA was examined by immunoblotting. (C) Nuclear proteins were isolated from
MCF-7 cells that overexpressed FOXC1 or the vector. Isolated proteins were immunoblotted for
p65 and the nuclear protein Lamin A/C. (D) Nuclear proteins were isolated from MDA-MB-231
cells that overexpressed FOXC1 or the vector. The binding of p65 and p50 to consensus DNA
oligonucleotides was assessed by TransAM ELISA. Data represent mean ± SD of three
independent experiments. (E) Nuclear proteins were isolated from control and FOXC1knockdown BT-549 cells. The binding of p65 to consensus DNA oligonucleotides was assessed
by TransAM ELISA. Data represent mean ± SD of three independent experiments. (F) MCF-7
cells were transiently transfected with NF-κB-luc or pGL4-luc, FOXC1, and the IκBα
S32A/S36A super-repressor (IκBα-SR). NF-κB activity was assessed by luciferase assays. Bars
represent mean ± SD of three independent experiments. (G) Real-time RT-PCR analysis (Sybr
Green) of IL-6 expression in control and FOXC1-overexpressing MDA-MB-231 cells. (H) Realtime RT-PCR analysis of IL-6 expression in FOXC1-knockdown BT-549 (left), FOXC1knockdown 4T1 cells (right), and corresponding control cells. P < 0.05 versus controls.
Figure S2. FOXC1 increases IB mRNA levels. Real-time RT-PCR analysis (Sybr Green) of
IB and p65 expression in MDA-MB-231 cells overexpressing FOXC1 or the control vector.
10
Figure S3. FOXC1 knockdown reduces Pin1 protein levels. Expression of Pin1 in BT-549
breast cancer cells that stably expressed control or FOXC1 shRNA was examined by
immunoblotting.
Figure S4. Ubiquitination of p65 is increased by FOXC1 overexpression. FOXC1- or vectoroverexpressing MDA-MB-231 cells were transfected with a HA-ubiquitin construct, treated with
10 µM MG-132, and subjected to immunoprecipitation with an anti-HA antibody, followed by
immunoblotting of p65.
Figure S5. Knockdown of Pin1 abolishes the induction of p65 by FOXC1. Parental (left) and
FOXC1-overexpressing (middle) MDA-MB-231 cells were transfected with transiently cotransfected with NF-κB-luc and Pin1 siRNA, followed by luciferase assays. Bars represent mean
± SD of three independent experiments. P < 0.05 versus controls. Expression of Pin1 and p65 in
FOXC1-overexpressing MDA-MB-231 cells transfected with control and Pin1 siRNA was also
examined by immunoblotting (right).
Figure S6. High Pin1 expression is associated with basal-like breast cancer cells and poor
survival in breast cancer. (A) Relative levels of Pin1 mRNA in the Neve et al. and Hoeflich et
al. microarray datasets of human breast cancer cell lines are shown. (B) Immunoblotting of Pin1
in luminal and basal-like triple-negative human breast cancer cell lines. (C) Kaplan-Meier curves
of disease-free survival stratified by Pin1 mRNA levels. Left, the 249-sample Ivshina et al.
dataset (6). Right, the 159-sample Pawitan et al. dataset (5). Survival differences were assessed
by the log-rank test.
11
Figure S7. High Pin1 expression shows a trend towards worse overall survival in breast
cancer. Kaplan-Meier curves of overall survival stratified by Pin1 mRNA levels. Left, the 295sample van de Vijver et al. dataset (4). Right, the 159-sample Pawitan et al. dataset (5). Survival
differences were assessed by the log-rank test.
Figure S8. p65 deficiency nullifies the effect of FOXC1 on cell proliferation. p65 knockout
MEFs were transfected with FOXC1 or the vector. Cell proliferation was measured by MTT
assays at the indicated time points.
12