Accepted Manuscript

Insulin-like growth factor-I induces CLU expression through Twist1 to promote

prostate cancer growth

Ario Takeuchi, Masaki Shiota, Eliana Beraldi, Daksh Thaper, Kiyoshi Takahara,

Naokazu Ibuki, Michael Pollak, Michael E. Cox, Seiji Naito, Martin E. Gleave,

Amina Zoubeidi

PII: S0303-7207(14)00016-1

DOI: http://dx.doi.org/10.1016/j.mce.2014.01.012

Reference: MCE 8752

To appear in: Molecular and Cellular Endocrinology Molecular

and Cellular Endocrinology

Received Date: 19 June 2013

Revised Date: 27 December 2013

Accepted Date: 14 January 2014

Please cite this article as: Takeuchi, A., Shiota, M., Beraldi, E., Thaper, D., Takahara, K., Ibuki, N., Pollak, M., Cox,

M.E., Naito, S., Gleave, M.E., Zoubeidi, A., Insulin-like growth factor-I induces CLU expression through Twist1

to promote prostate cancer growth, Molecular and Cellular Endocrinology Molecular and Cellular

Endocrinology (2014), doi: http://dx.doi.org/10.1016/j.mce.2014.01.012

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Insulin-like growth factor-I induces CLU expression through Twist1 to promote prostate cancer

growth

Ario Takeuchi1, Masaki Shiota1, Eliana Beraldi1, Daksh Thaper1, Kiyoshi Takahara1, Naokazu Ibuki1,

Michael Pollak2, Michael E. Cox1, Seiji Naito3, Martin E. Gleave1, and Amina Zoubeidi1

1The Vancouver Prostate Centre and Department of Urologic Sciences, University of British

Columbia, Vancouver, British Columbia, Canada

2Department of Medicine and Oncology, McGill University, Montreal, Quebec, Canada

3Department of Urology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

Short title: IGF-I induces CLU via Twist1

Keywords: clusterin; insulin-like growth factor-I; prostate cancer; STAT3; Twist1

Corresponding Author:

Dr. Amina Zoubeidi

The Vancouver Prostate Centre and Department of Urologic Sciences,

University of British Columbia,

2660 Oak Street, Vancouver, British Columbia, Canada V6H 3Z6.

Phone: +1-604-875-4818

Fax: +1-604-875-5654

E-mail: azoubeidi@prostatecentre.com

2

ABSTRACT

Clusterin (CLU) is cytoprotective molecular chaperone that is highly expressed in

castrate-resistant prostate cancer (CRPC). CRPC is also characterized by increased insulin-like

growth factor (IGF)-I responsiveness which induces prostate cancer survival and CLU expression.

However, how IGF-I induces CLU expression and whether CLU is required for IGF-mediated growth

signaling remain unknown. Here we show that IGF-I induced CLU via STAT3-Twist1 signaling

pathway. In response to IGF-I, STAT3 was phosphorylated, translocated to the nucleus and bound to

the Twist1 promoter to activate Twist1 transcription. In turn, Twist1 bound to E-boxes on the CLU

promoter and activated CLU transcription. Inversely, we demonstrated that knocking down Twist1

abrogated IGF-I induced CLU expression, indicating that Twist1 mediated IGF-I–induced CLU

expression. When PTEN knockout mice were crossed with lit/lit mice, the resultant IGF-I deficiency

suppressed Twist1 as well as CLU gene expression in mouse prostate glands. Moreover, both Twist1

and CLU knockdown suppressed prostate cancer growth accelerated by IGF-I, suggesting the

relevance of this signaling not only in an in vitro, but also in an in vivo. Collectively, this study

indicates that IGF-I induces CLU expression through sequential activation of STAT3 and Twist1, and

suggests that this signaling cascade plays a critical role in prostate cancer pathogenesis.

3

Introduction

Prostate cancer is the most common solid malignant tumor among males in Western countries

(Jemal et al., 2010). A series of epidemiological and biological studies demonstrate that the

insulin-like growth factor (IGF) axis is a critical regulator of growth, survival, and metastatic potential

in a variety of malignancies and is closely implicated in prostatic carcinogenesis and prostate cancer

progression as well as resistance to castration therapy (Chan et al., 1998; Krueckl et al., 2004;

Nickerson et al., 2001; Wolk et al., 1998). We have previously demonstrated that IGF-I promoted

human prostate cancer cell growth and that increased IGF-I receptor (IGF-IR) expression and

signaling are components of castrate resistant progression (Krueckl et al., 2004; Takahara et al.,

2011).

IGFs bind to the tyrosine kinase IGF-IR, which is a heterotetrameric type I receptor

protein-tyrosine kinase composed of two ligand-binding α-subunits and two transmembrane

β-subunits. The binding of ligand to IGF-IR induces auto-phosphorylation of the β-subunits of the

receptor complex and further activation of the protein-tyrosine kinase activity (Hubbard et al., 1994;

Weiss & Schlessinger 1998). Once activated, IGF-IR recruits and phosphorylates various downstream

targets such as the insulin receptor substrate-1 and -2 which activate many signaling pathways,

including Ras/Raf/mitogen-activated protein kinase (MAPK) and PI3K/Akt, as well as signal

transducer and activator of transcription 3 (STAT3; Zong et al., 2000) resulting in cell growth and

survival.

Clusterin (CLU) is a stress-induced cytoprotective chaperone, and involved in many biological

processes such as sperm maturation, tissue differentiation, tissue remodeling, membrane recycling,

lipid transportation, cell proliferation and cell death. CLU has been shown expressed in many human

cancers (Zhong et al., 2010). Increased levels of CLU have been reported in breast, colon, lung,

bladder, prostate and other cancers (July et al., 2004; Kevans et al., 2009; Miyake et al., 2002; So et

al., 2005; Steinberg et al., 1997). In prostate, CLU levels are low in benign prostate epithelial cells,

but increase in prostate cancers with higher Gleason grade (Steinberg et al., 1997). Furthermore, CLU

4

expression increases as prostate cancers adapt to androgen-deprivation therapy (July et al., 2002).

These data indicate that CLU is also implicated in prostate carcinogenesis and prostate cancer

progression. Similarly, numerous evidences showed that a basic helix–loop–helix transcription factor

Twist1 is also involved in pathogenesis of various cancers (Franco et al., 2011), including castration

resistance in prostate cancer (Shiota et al., 2010).

IGF-I axis induced CLU expression after irradiation via Src-MEK-ERK-EGR1 signaling in

human breast cancer MCF-7 cells (Criswell et al., 2005). However, the mechanism and role of CLU

induction by IGF-I in prostate cancer remain unrevealed. In this study, we set out to define links

between IGF-I signaling and Twist1/CLU expression in prostate cancer, identifying STAT3 as a

downstream effector of IGF-I.

5

MATERIAL AND METHODS

Cell culture and transfection

The human prostate cancer cell line, PC-3, was purchased from the American Type Culture

Collection (ATCC authentication by isoenzymes analysis) and maintained in Dulbecco’s Modified

Eagle’s Medium (DMEM; Thermo Scientific, Waltham, MA, USA) supplemented with 5%

fetal-bovine serum (FBS). The human prostate cancer cell line LNCaP was kindly provided by Dr.

Leland W.K. Chung (Cedars-Sinai Medical Center, Los Angeles, CA, USA), tested and authenticated

by whole-genome and whole-transcriptome sequencing on Illumina Genome Analyzer IIx platform in

2009. LNCaP cells were maintained in RPMI 1640 (Thermo Scientific) supplemented with 5% FBS.

Antibodies and reagents

Antibodies against Myc (sc-815), CLU (sc-6419) and Twist1 (sc-81417) were purchased from

Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-phosphorylated STAT3Tyr705 (p-STAT3Tyr705,

#9131), anti-phosphorylated STAT3Ser727 (p-STAT3Ser727, #9134) and anti-STAT3 (#9139) antibodies

were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-Lamin B1 and

anti-β-actin antibodies were purchased from Abcam (Cambridge, MA, USA) and Sigma (St Louis,

MO, USA), respectively. Human recombinant IGF-I was obtained from Fitzgerald (Acton, MA,

USA).

Plasmids and siRNAs

Twist1-Myc-Flag plasmid expressing C-terminally Myc-Flag-tagged Twist1 protein and the

corresponding mock plasmid (Myc-Flag plasmid) were purchased from OriGene (Rockville, MD,

USA). The Twist1 reporter plasmid (Twist–Luc) was kindly provided from Dr. Wang LH (Mount

Sinai School of Medicine, New York, NY, USA; Cheng et al., 2008). CLU reporter plasmids

(CLU–Luc –1,998/+254, –1,998/–702, –1,116/–702, and –707/+254) containing various lengths of the

promoter and first exon of the wild-type human CLU gene were constructed as described previously

(Shiota et al., 2011).

The following double-stranded 25-bp siRNA oligonucleotides were commercially generated

6

(Invitrogen, Carlsbad, CA, USA): 5’-CUUCCUCGCUGUUGCUCAGGCUGUC-3’ for Twist1

siRNA #1; 5’-UUGAGGGUCUGAAUCUUGCUCAGCU-3’ for Twist1 siRNA #2. The sequence of

siRNA corresponding to the human CLU initiation site in exon II was

5’-GCAGCAGAGUCUUCAUCAU-3’ (Dharmacon Research Inc.). Stealth™ RNAi Negative

Control Medium GC Duplex #2 (Invitrogen) was used as a control siRNA. Cells were transfected

with the indicated siRNA or the indicated plasmid as previously described (Zoubeidi et al., 2010a;

Zoubeidi et al., 2010b).

Quantitative reverse transcription (RT)-PCR

RNA extraction and RT-PCR were performed as described previously (Lamoureux et al., 2011).

Real time monitoring of PCR amplification of cDNA was performed using the following primer pairs

and probes, Twist1 (Hs00361186_m1), CLU (Hs00156548_m1) and GAPDH (Hs03929097_g1)

(Applied Biosystems) on ABI PRISM 7900 HT Sequence Detection System (Applied Biosystems)

with TaqMan Gene Expression Master Mix (Applied Biosystems). Target gene expression was

normalized to GAPDH levels in respective samples as an internal control. The results are

representative of at least three independent experiments.

Western blot analysis

Whole-cell extracts were obtained by lysis of cells in an appropriate volume of ice-cold RIPA

buffer composed of 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.5% sodium deoxycholate, 1% Nonidet

P-40, 0.1% sodium dodecyl sulfate (SDS) containing 1 mM Na3VO4, 1 mM NaF, 1 mM

phenylmethylsulfonyl fluoride and protease inhibitor cocktail tablets (Complete, Roche Applied

Science, Indianapolis, IN, USA). Nuclear and cytoplasmic extracts were obtained using CelLyticTM

NuCLEARTM Extraction Kit (Sigma) according to manufacturer’s protocol. Cellular extracts were

clarified by centrifugation at 13,000 x g for 10 min and protein concentrations of the extracts

determined by a BCA protein assay kit (Thermo Scientific). Thirty micrograms of the extracts were

boiled for 5 min in SDS sample buffer, separated by SDS-PAGE, and transferred onto a

polyvinylidene difluoride (PVDF) membrane. Membranes were probed with dilutions of primary

7

antibodies followed by incubation with horseradish peroxidase-conjugated secondary antibodies.

After extensive washing, proteins were visualized by a chemiluminescent detection system (GE

Healthcare, Buckinghamshire, UK).

Luciferase reporter assay

Prostate cancer cells were transfected with 0.5 μg of the indicated reporter plasmid, expression

plasmid, siRNA, and 0.05 μg of pRL-TK as an internal control. 24 hours post transfection, media was

changed to serum-free media followed by IGF-I treatment. The luciferase activities were measured

using a Dual-Luciferase Reporter Assay System (Promega) and a microplate luminometer (EG&G

Berthold). The Firefly luciferase activities were corrected by the corresponding Renilla luciferase

activities and protein concentration. The results are representative of at least three independent

experiments.

Chromatin immunoprecipitation assay (ChIP assay)

LNCaP cells were seeded and stimulated with 100 ng/ml IGF-I followed by paraformaldehyde

cross-linking and micrococcal nuclease digestion to achieve a DNA smear of 200–1000 bp. ChIP

assay on the indicated genes was performed using SimpleChIPTM Enzymatic Chromatin IP Kit

(Agarose Beads) according to the manufacturer’s protocol (Cell Signaling Technology). Quantitative

RT-PCR assay was performed using ABI PRISM 7900 HT Sequence Detection System with 2 μL of

20 μL DNA extraction, the primer pairs below and RT2 Real-TimeTM SYBR Green/Rox PCR master

mix (Qiagen, Valencia, CA, USA). The results are representative of at least three independent

experiments. The primer pairs for Twist1 promoter were Fw; 5’-TGCCTTTCCCATGGACTGGG-3’

and Rv; 5’-GAGTTCCAAAGGCCAAACCG-3’ as described previously (Cheng et al. 2008). The

primer pairs for CLU –02 (GPH025704(–)02A) targeting around –1,403 bp, CLU –01

(GPH025704(–) 01A) targeting around –417 bp, CLU +07 (GPH025704(+)07A) targeting around

+6,603 bp, and RPL30 gene (exon 3) were described previously (Shiota et al., 2011).

Production and characterization of Ghrhr(lit/lit)/Cre/PTEN(fl/fl)

The production of Ghrhr(lit/lit)/Cre/PTEN(fl/fl) was previously described (Takahara et al.,

8

2013). Briefly, we crossed pbARR2-Cre, PTEN (fl/fl) mice (Wang et al., 2003) with GHRHR (lit/lit)

mice (Yang et al., 1996) to produce lit/lit and lit/+ PTEN-/- mice. Mice serums and prostates were

harvested in accordance with the guidelines of the Canadian Council on Animal Care and with

appropriate institutional certification between 15 and 20 weeks of age. Nineteen mice were collected

for each cohort after genotyping to clarify lit heterozygosity (lit/+) and homozygosity (lit/lit), Cre

recombinase and PTEN fl/fl status from tail clip DNA. Serum IGF-I levels were measured using

ELISA from (R&D Systems, Minneapolis, MN, USA). Murine Twist1 (Mm00442036_m1) and

murine CLU (Mm00442773_m1) transcript levels were measured using RNA from mice prostates by

quantitative RT-PCR normalized to murine GAPDH (Mm99999915_g1) transcript levels.

Cell growth assay

LNCaP cells (2.5 x 104) transfected with 10 nM of the indicated siRNA were plated in 24-well

plates. The following day, media were changed into media supplemented with 1% serum from lit/lit or

lit/+ mice with or without 100 ng/mL IGF-I. After incubation for 96 h, cell growth was measured

using the crystal violet assay as described previously (Shiota et al., 2011). The results are

representative of at least 3 independent experiments.

Immunohistochemistry

Prostate tissues from PTEN knockout lit/lit mice and PTEN knockout lit/+ mice at 15 and 20

weeks of age were obtained in accordance with the guidelines of the Canadian Council on Animal

Care and with appropriate institutional certification. Immunohistochemical staining was conducted as

previously described (Zoubeidi et al., 2010b) using the Ventana Discover XT TM autostainer (Ventana

Medical System, Tuscan, AZ, USA) with enzyme labeled biotin streptavidin system and solvent

resistant DAB Map kit by antibodies against CLU (Santa Cruz Biotechnology), IGF-IR (Sigma),

p-STAT3Tyr705 (Cell Signaling Technology), Twist1 (Sigma).

Statistical analysis

All data were assessed using the Student’s t-test. Levels of statistical significance were set at P

< 0.05.

10

RESULTS

IGF-I induces both Twist1 and CLU expression in prostate cancer cells

It is known that IGF-I activates Twist1 in NIH-3T3 fibroblasts (Dupont et al., 2001) and CLU

expression in breast cancer cell lines (Criswell et al., 2005). Since Twist1 is a transcription factor, we

intended to determine whether IGF-I induced CLU expression via Twist1 in human prostate cancer

cells. First, we examined the appropriate concentration inducing Twist1 and CLU expression in

LNCaP and PC-3 cells. Then, 100 ng/mL IGF-I most induced Twist1 as well as CLU expression both

in LNCaP and PC-3 cells (data not shown). Then, we chose 100 ng/ml IGF-I for treatment thereafter.

Next, LNCaP and PC-3 cell lines were treated with IGF-I in a time dependent manner and both

Twist1 and CLU expression were evaluated at mRNA and protein levels. We found that IGF-I

increased Twist1 as well as CLU expression at transcript level (Fig. 1A) and protein level (Fig. 1B) in

LNCaP and PC-3 cells.

Twist1 binds to CLU promoter region and regulates CLU expression

The finding above prompted us to examine the functional link between Twist1 transcription

factor and CLU. As shown in Fig. 2A, Twist1 knockdown using 2 different Twist1-specific siRNAs

reduced basal CLU expression at both transcript and protein levels in LNCaP cells. Inversely, Twist1

over-expression up-regulated CLU mRNA and protein expression (Fig. 2B).

Since Twist1 is known as an E-box (5’-CANNTG-3’) binding transcription factor (Li et al.,

1995), we investigated whether Twist1 directly regulates CLU transcription. We first searched for

putative E-box binding sites in the CLU promoter region between –2,000 bp and +500 bp from

transcription start site (TSS) and found that are 10 E-boxes in CLU promoter as shown in Fig. 2C. We

next analyzed CLU promoter activity using different truncated regions (Fig. 2C). We found that CLU

promoter activity was highest between –1,998 bp and –702 bp in LNCaP cells which contains 7

E-boxes (Fig. 2D) as a potential Twist1-binding sites, similarly to the result using PC-3 cells (Shiota

et al., 2012). To further evaluate if Twist1 can regulates CLU promoter activity, Twist1 was

overexpressed and CLU transcriptional activity was analyzed using different CLU promoter

11

constructs. As shown in Fig. 2E, Twist1 increased luciferase activity of CLU–Luc –1,998/–702, but

not CLU–Luc –1,116/–702. These findings suggested that the cis-element of CLU promoter region

containing –1,998/–1,116 bp was activated by Twist1. Inversely, Twist1 knockdown reduced CLU

promoter transcriptional activity (Fig. 2F). To confirm Twist1 binding to CLU promoter region, we

performed ChIP assay against CLU gene in LNCaP cells. The results showed that Twist1 bound to

CLU promoter regions around –1,400 bp from TSS represented by CLU –02, while they did not bind

to the CLU gene regions around +6,600 bp from TSS and RPL30 gene (Fig. 2G).

IGF-I induces CLU expression via Twist1

To further evaluate whether Twist1 expression is required for IGF-I–induced CLU expression,

Twist1 was silenced and levels of Twist1 and CLU were evaluated in the absence or presence of

IGF-I. Twist1-specific siRNA successfully down-regulated both basal and IGF-I–induced Twist1

expression (Fig. 3A, left) as well as CLU mRNA at basal level. Interestingly, we found that Twist1

knockdown abrogated IGF-I–induced CLU expression at mRNA levels (Fig. 3A, right) and protein

levels (Fig. 3B) in both LNCaP and PC-3 cells. These data suggest that Twist1 is involved in basal

CLU expression, as well as required for IGF-I–induced CLU expression. In addition, luciferase

reporter assay using CLU reporter plasmid also revealed that Twist1 knockdown ameliorated CLU

induction by IGF-I (Fig. 3C).

IGF-I activates STAT3 transcription factor, resulting in Twist1 up-regulation

To investigate the mechanism of Twist1/CLU induction by IGF-I, we focused on STAT3

transcription factor because it was reported that STAT3 transcriptionally regulated Twist1 expression

(Cheng et al., 2008). We next analyzed the effect of IGF-I on STAT3 phosphorylation. Our data

showed that IGF-I induces STAT3 phosphorylation only on Tyr705 but not on Ser727, resulting in an

activation of STAT3 (Fig. 4A). This result was supported by the finding that IGF-I facilitated STAT3

translocation into nucleus (Fig. 4B). As a result, IGF-I stimulated STAT3 binding to Twist1-promotor

region, resulting in activation of Twist1 transcription (Fig. 4C), thereby augmented Twist1 binding to

CLU-promoter region (Fig. 4D). Consistently, reporter assay using Twist1 reporter plasmid showed

12

an increased transcription of Twist1 gene by IGF-I (Fig. 4E). These data show that IGF-I promoted

STAT3 activation and STAT3 is at least in part needed to mediate the IGF-I–induced Twist1

expression and subsequent CLU expression.

IGF-I/Twist1/CLU signaling plays a critical role in mice prostate cancer proliferation

To investigate the biologic relevance of the above findings, we assessed Twist1 and CLU

expression in prostate tissues from PTEN knockout mice (Wang et al., 2003) crossed with lit/lit mice,

which harbor growth-hormone-releasing hormone receptor (GHRHR) mutation abolishing GHRHR

function (Wang et al., 2003). Lack of growth-hormone-releasing hormone signaling in lit/lit mice

results in marked reduction of serum growth hormone, which in turn leads to reducing serum IGF-I

level (Fig. 5A), known to correlate with IGF-I level in prostate (Wang et al., 2008). Consistently with

the preceding in vitro data, Twist1 and CLU expression were lower in prostate tissues, harvested

between 15 and 20 weeks of age, from PTEN knockout lit/lit mice compared with those of PTEN

knockout lit/+ mice (Fig. 5B). As well, immunohistochemistry against prostate tissues from PTEN

knockout lit/lit mice and PTEN knockout lit/+ mice suggested that prominent decreased levels of

IGF-IR and Twist1 in lit/lit mice (Fig. 5C).

Previously we showed that prostate cancer grew less rapidly in lit/lit mice compared with lit/+

mice in an in vitro as well as an in vivo (Takahara et al., 2011). Subsequently, we examined whether

cell proliferation induced by IGF-I is affected by Twist1 or CLU silencing. As we previously reported,

LNCaP cell growth was promoted by IGF-I, while this growth promotion was almost completely

abolished by either Twist1 or CLU knockdown (Fig. 6A), suggesting that both Twist1 and CLU are

important downstream mediators of IGF-I–induced prostate cancer growth.

13

DISCUSSION

In this study, we identified a novel mechanism by which IGF-I regulates CLU expression in

prostate cancer cells via the STAT3-Twist1 pathway. We went on to demonstrate that once IGF-I

activates STAT3, STAT3 translocates to the nucleus, binds to the Twist1 promoter, resulting in

Twist1 up-regulation. Twist1 subsequently binds to E-boxes on CLU promoter and enhance CLU

expression, thereby creating a feed-forward loop which leads to increase of cell proliferation in

prostate cancer (Fig. 6B).

Our data showed that IGF-I induced STAT3 phosphorylation in prostate cancer cells,

confirming previous reports in human fibroblasts (HEK293T cells; Zong et al., 2000) and suggesting

that the role of IGF-I on STAT3 activation was conserved across different cell lines. Once

phosphorylated, STAT3 translocated to the nucleus and regulated Twist1 expression in breast cancer

cells (Cheng et al., 2008). STAT3 activation positively correlated with Twist1 expression in breast

cancer tissues (Cheng et al., 2008). Similarly, epidermal-growth factor (EGF)–induced Twist1

transcription was reported to be mediated by STAT3 in several cancer cells (Lo et al., 2007).

Moreover, in hepatocellular carcinoma, activated STAT3 and Twist1 expressions were positively

correlated (Zhang et al., 2012). Thus, the connection between STAT3 and Twist1 has been established

in various cancers including prostate cancer (Cheng et al., 2008; Cho et al., 2013; Hsu et al., 2012;

Teng et al., 2013). Additionally, in this study, STAT3 phosphorylation at Tyr705 was induced by

IGF-I concurret with nuclear translocation, which is consistent with the previous report (Wen et al.,

1995). Moreover, we found that STAT3 bound to Twist1 promoter region in prostate cancer, leading

to Twist1 gene expression, which was increased by IGF-I treatment.

Twist1, a basic helix–loop–helix transcription factor, has been described as a proto-oncogene

(Hamamori et al., 1997; Quertermous et al., 1994) that promoted breast cancer metastasis (Yang et al.,

2004). Similar to CLU, Twist1 was also up-regulated in various malignant tumors, including prostate

cancer (Wallerand et al., 2010; Wang et al., 2004). Moreover, we have recently shown that Twist1

was involved in prostate cancer growth (Shiota et al., 2008) as well as resistance to castration through

14

androgen receptor (Shiota et al., 2010). Collectively, Twist1 plays a key role in the development and

progression of prostate cancer similar to that ascribed to CLU. Like Twist1 (Dupont et al., 2001),

CLU was also known to be induced by IGF-I (Criswell et al., 2005). In this study, Twist1 knockdown

decreased basal CLU transcript and protein, as well as IGF-I–induced CLU expression, indicating that

IGF-I–induced CLU expression was mediated by Twist1. Furthermore, we found that Twist1

regulated CLU transcription by reporter assay and ChIP assay. These findings link Twist1 regulation

of CLU expression, under IGF-I stimulation, as a potential pathway that promotes prostate cancer

growth.

The pbARR2-Cre, PTEN (fl/fl) mouse model, which lead to de novo formation of prostate

tumors, is one model that mimicks human prostate cancer from initiation to local invasion and

metastasis (Trotman et al., 2003; Wang et al., 2003). Prostate-specific loss of PTEN expression

resulted in invasive carcinoma with lymphovascular invasion within 12 weeks, which progressed to

lung metastasis (Wang et al., 2003). Deletion or mutation of the tumor suppressor PTEN gene has

been implicated in many human cancers and has been seen in up to 30% of primary prostate cancers

and >64% of prostate metastases, making PTEN an important candidate gene for prostate cancer

development and progression (Majumder & Sellers 2005; Suzuki et al., 1998). Furthermore, CLU

expression was elevated in PTEN knockout mice (Wang et al., 2003). To define links between IGF-I

signaling and CLU in prostate cancer growth, we crossed pbARR2-Cre, PTEN (fl/fl) mice with

GHRHR (lit/lit) mice. It has been known that in lit/lit mice, several proto-oncogenic pathways

including MAPK and PI3K/Akt were down-regulated (Takahara et al., 2011). In addition, in PTEN

knockout lit/lit mice model, Twist1 as well as CLU expression was reduced in IGF-1–deficient lit/lit

mice, which supported our in vitro data that IGF-I induced Twist1 and CLU expression in LNCaP and

PC-3 cells. Furthermore, this study suggested that Twist1 as well as CLU plays key roles in

IGF-1–induced prostate cancer cell proliferation. Since IGF-I has been a well-known promoter of

prostate cancer growth, these data identified Twist1 and CLU as important mediators of

IGF-I–stimulated prostate pathogenesis in this model.

15

In summary, we identified a novel Twist1/CLU pathway stimulated by IGF-I involving STAT3

phosphorylation, and then CLU. Therefore, signaling from IGF-I to CLU provided a molecular

mechanism that might explain at least in part the influence of IGF-I on prostate cancer.

16

Declaration of interest

There are no conflicts of interest.

Funding

This study was supported by the Terry Fox New Frontiers Program, the Canadian Institutes of Health

Research, the Prostate Cancer Foundation USA.

Acknowledgements

We are grateful to Dr. Lu-Hai Wang (Mount Sinai School of Medicine, New York, NY, USA)

providing the Twist–Luc reporter plasmid. We thank Howard Tearle and Ladan Fazli1 for their

excellent technical assistance.

17

REFERENCE

Chan JM, Stampfer MJ, Giovannucci E, Gann PH, Ma J, Wilkinson P, Hennekens CH & Pollak M

1998 Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study. Science 279

563–566

Cheng GZ, Zhang WZ, Sun M, Wang Q, Coppola D, Mansour M, Xu LM, Costanzo C, Cheng JQ &

Wang LH 2008 Twist is transcriptionally induced by activation of STAT3 and mediates STAT3

oncogenic function. J Biol Chem 283 14665–14673

Cho KH, Jeong KJ, Shin SC, Kang J, Park CG & Lee HY 2013 STAT3 mediates TGF-β1-induced

TWIST1 expression and prostate cancer invasion. Cancer Lett 336 167–173

Criswell T, Beman M, Araki S, Leskov K, Cataldo E, Mayo LD & Boothman DA 2005 Delayed

activation of insulin-like growth factor-1 receptor/Src/MAPK/Egr-1 signaling regulates clusterin

expression, a pro-survival factor. J Biol Chem 280 14212–14221

Dupont J, Fernandez AM, Glackin CA, Helman L & LeRoith D 2001 Insulin-like growth factor 1

(IGF-1)-induced twist expression is involved in the anti-apoptotic effects of the IGF-1 receptor. J

Biol Chem 276 26699–26707

Franco HL, Casasnovas J, Rodríguez-Medina JR & Cadilla CL 2011 Redundant or separate

entities?―roles of Twist1 and Twist2 as molecular switches during gene transcription. Nucleic

Acids Res 39 1177–1186

Hamamori Y, Wu HY, Sartorelli V & Kedes L 1997 The basic domain of myogenic basic

helix-loop-helix (bHLH) proteins is the novel target for direct inhibition by another bHLH protein,

Twist. Mol Cell Biol 17 6563–6573

Hsu KW, Hsieh RH, Huang KH, Fen-Yau Li A, Chi CW, Wang TY, Tseng MJ, Wu KJ & Yeh TS 2012

Activation of the Notch1/STAT3/Twist signaling axis promotes gastric cancer progression.

Carcinogenesis 33 1459–1467

Hubbard SR, Wei L, Ellis L & Hendrickson WA 1994 Crystal structure of the tyrosine kinase domain

of the human insulin receptor. Nature 372 746–754

18

Jemal A, Siegel R, Xu J & Ward E 2010 Cancer statistics, 2010. CA Cancer J Clin 60 277–300.

July LV, Akbari M, Zellweger T, Jones EC, Goldenberg SL & Gleave ME 2002 Clusterin expression

is significantly enhanced in prostate cancer cells following androgen withdrawal therapy. Prostate

50 179–188

July LV, Beraldi E, So A, Fazli L, Evans K, English JC & Gleave ME 2004 Nucleotide-based

therapies targeting clusterin chemosensitize human lung adenocarcinoma cells both in vitro and in

vivo. Mol Cancer Ther 3 223–232

Kevans D, Foley J, Tenniswood M, Sheahan K, Hyland J, O'Donoghue D, Mulcahy H & O'Sullivan J

2009 High clusterin expression correlates with a poor outcome in stage II colorectal cancers.

Cancer Epidemiol Biomarkers Prev 18 393–399

Krueckl SL, Sikes RA, Edlund NM, Bell RH, Hurtado-Coll A, Fazli L, Gleave ME & Cox ME 2004

Increased insulin-like growth factor I receptor expression and signaling are components of

androgen-independent progression in a lineage-derived prostate cancer progression model. Cancer

Res 64 8620–8629

Lamoureux F, Thomas C, Yin MJ, Kuruma H, Beraldi E, Fazli L, Zoubeidi A & Gleave ME 2011

Clusterin inhibition using OGX-011 synergistically enhances Hsp90 inhibitor activity by

suppressing the heat shock response in castrate-resistant prostate cancer. Cancer Res 71

5838–5849

Li L, Cserjesi P & Olson EN 1995 Dermo-1: a novel twist-related bHLH protein expressed in the

developing dermis. Dev Biol 172 280–292

Lo HW, Hsu SC, Xia W, Cao X, Shih JY, Wei Y, Abbruzzese JL, Hortobagyi GN & Hung MC 2007

Epidermal growth factor receptor cooperates with signal transducer and activator of transcription

3 to induce epithelial-mesenchymal transition in cancer cells via up-regulation of TWIST gene

expression. Cancer Res 67 9066–9076

Majumder PK & Sellers WR 2005 Akt-regulated pathways in prostate cancer. Oncogene 24

7465–7474

19

Miyake H, Gleave M, Kamidono S & Hara I 2002 Overexpression of clusterin in transitional cell

carcinoma of the bladder is related to disease progression and recurrence. Urology 59 150–154

Nickerson T, Chang F, Lorimer D, Smeekens SP, Sawyers CL & Pollak M 2001 In vivo progression of

LAPC-9 and LNCaP prostate cancer models to androgen independence is associated with

increased expression of insulin-like growth factor I (IGF-I) and IGF-I receptor (IGF-IR). Cancer

Res 61 6276–6280

Quertermous EE, Hidai H, Blanar MA & Quertermous T 1994 Cloning and characterization of a basic

helix-loop-helix protein expressed in early mesoderm and the developing somites. Proc Natl Acad

Sci U S A 91 7066–7070

Shiota M, Izumi H, Onitsuka T, Miyamoto N, Kashiwagi E, Kidani A, Yokomizo A, Naito S &

Kohno K 2008 Twist promotes tumor cell growth through YB-1 expression. Cancer Res 68

98–105

Shiota M, Yokomizo A, Tada Y, Inokuchi J, Kashiwagi E, Masubuchi D, Eto M, Uchiumi T & Naito

S 2010 Castration resistance of prostate cancer cells caused by castration-induced oxidative stress

through Twist1 and androgen receptor overexpression. Oncogene 29 237–250

Shiota M, Zardan A, Takeuchi A, Kumano M, Beraldi E, Naito S, Zoubeidi A & Gleave ME 2012

Clusterin mediates TGF-β–induced epithelial-mesenchymal transition and metastasis via Twist1 in

prostate cancer cells. Cancer Res 72 5261–5272

Shiota M, Zoubeidi A, Kumano M, Beraldi E, Naito S, Nelson CC, Sorensen PH & Gleave ME 2011

Clusterin is a critical downstream mediator of stress-induced YB-1 transactivation in prostate

cancer. Mol Cancer Res 9 1755–1766

So A, Sinnemann S, Huntsman D, Fazli L & Gleave M 2005 Knockdown of the cytoprotective

chaperone, clusterin, chemosensitizes human breast cancer cells both in vitro and in vivo. Mol

Cancer Ther 4 1837–1849

Steinberg J, Oyasu R, Lang S, Sintich S, Rademaker A, Lee C, Kozlowski JM & Sensibar JA 1997

Intracellular levels of SGP-2 (Clusterin) correlate with tumor grade in prostate cancer. Clin

20

Cancer Res 3 1707–1711

Suzuki H, Freije D, Nusskern DR, Okami K, Cairns P, Sidransky D, Isaacs WB & Bova GS 1998

Interfocal heterogeneity of PTEN/MMAC1 gene alterations in multiple metastatic prostate cancer

tissues. Cancer Res 58 204–209

Takahara K, Tearle H, Ghaffari M, Gleave ME, Pollak M & Cox ME 2011 Human prostate cancer

xenografts in lit/lit mice exhibit reduced growth and androgen-independent progression. Prostate

71 525–537

Takahara K, Ibuki N, Ghaffari M, Tearle H, Ong CJ, Azuma H, Gleave ME, Pollak M & Cox ME

2013 The influence of growth hormone/insulin-like growth factor deficiency on prostatic

dysplasia in pbARR2-Cre, PTEN knockout mice. Prostate Cancer Prostatic Dis 16 239–247

Teng J, Wang X, Xu Z & Tang N 2013 HBx-dependent activation of Twist mediates STAT3 control of

epithelium-mesenchymal transition of liver cells. J Cell Biochem 114 1097–1104

Trotman LC, Niki M, Dotan ZA, Koutcher JA, Di Cristofano A, Xiao A, Khoo AS, Roy-Burman P,

Greenberg NM, Van Dyke T, Cordon-Cardo C & Pandolfi PP 2003 Pten dose dictates cancer

progression in the prostate. PLoS Biol 1 E59

Wallerand H, Robert G, Pasticier G, Ravaud A, Ballanger P, Reiter RE & Ferrière JM 2010 The

epithelial-mesenchymal transition-inducing factor TWIST is an attractive target in advanced

and/or metastatic bladder and prostate cancers. Urol Oncol 28 473–479

Wang S, Gao J, Lei Q, Rozengurt N, Pritchard C, Jiao J, Thomas GV, Li G, Roy-Burman P, Nelson PS,

Liu X & Wu H 2003 Prostate-specific deletion of the murine Pten tumor suppressor gene leads to

metastatic prostate cancer. Cancer Cell 4 209–221

Wang X, Ling MT, Guan XY, Tsao SW, Cheung HW, Lee DT & Wong YC 2004 Identification of a

novel function of TWIST, a bHLH protein, in the development of acquired taxol resistance in

human cancer cells. Oncogene 23 474–482

Wang Z, Luque RM, Kineman RD, Ray VH, Christov KT, Lantvit DD, Shirai T, Hedayat S, Unterman

TG, Bosland MC, Prins GS & Swanson SM 2008 Disruption of growth hormone signaling retards

21

prostate carcinogenesis in the Probasin/TAg rat. Endocrinology 149 1366–1376

Weiss A & Schlessinger J 1998 Switching signals on or off by receptor dimerization. Cell 94 277–280

Wen Z, Zhong Z & Darnell JE Jr 1995 Maximal activation of transcription factor by Stat1 and Stat3

requires both tyrosine and serine phosphorylation. Cell 82 241–250

Wolk A, Mantzoros CS, Andersson SO, Bergström R, Signorello LB, Lagiou P, Adami HO &

richopoulos D 1998 Insulin-like growth factor 1 and prostate cancer risk: a population-based,

case-control study. J Natl Cancer Inst 90 911–915

Yang J, Mani SA, Donaher JL, Ramaswamy S, Itzykson RA, Come C, Savagner P, Gitelman I,

Richardson A & Weinberg RA 2004 Twist, a master regulator of morphogenesis, plays an essential

role in tumor metastasis. Cell 117 927–939

Yang XF, Beamer WG, Huynh H & Pollak M 1996 Reduced growth of human breast cancer

xenografts in hosts homozygous for the lit mutation. Cancer Res 56 1509–1511

Zhang CH, Xu GL, Jia WD, Li JS, Ma JL, Ren WH, Ge YS, Yu JH, Liu WB & Wang W 2012

Activation of STAT3 signal pathway correlates with Twist and E-cadherin expression in

hepatocellular carcinoma and their clinical significance. J Surg Res 174 120–129

Zhong B, Sallman DA, Gilvary DL, Pernazza D, Sahakian E, Fritz D, Cheng JQ, Trougakos I, Wei S

& Djeu JY 2010 Induction of clusterin by AKT--role in cytoprotection against docetaxel in

prostate tumor cells. Mol Cancer Ther 9 1831–1841

Zong CS, Chan J, Levy DE, Horvath C, Sadowski HB & Wang LH 2000 Mechanism of STAT3

activation by insulin-like growth factor I receptor. J Biol Chem 275 15099–15105

Zoubeidi A, Ettinger S, Beraldi E, Hadaschik B, Zardan A, Klomp LW, Nelson CC, Rennie PS &

Gleave ME 2010a Clusterin facilitates COMMD1 and I-κB degradation to enhance NF-κB

activity in prostate cancer cells. Mol Cancer Res 8 119–130

Zoubeidi A, Zardan A, Wiedmann RM, Locke J, Beraldi E, Fazli L & Gleave ME 2010b Hsp27

promotes insulin-like growth factor-I survival signaling in prostate cancer via p90Rsk-dependent

phosphorylation and inactivation of BAD. Cancer Res 70 2307–2317

23

FIGURE LEGENDS

Figure 1. IGF-I increases Twist1 expression followed by CLU up-regulation. (A) LNCaP (left) and

PC-3 (right) cells were treated with 100 ng/mL IGF-I. After the indicated duration, quantitative

RT-PCR was performed using the primer pairs and probes for Twist1, CLU and GAPDH. Each

transcript level from non-treated cells was set as 1. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with

no treatment). (B) LNCaP (left) and PC-3 (right) cells were treated with 100 ng/mL IGF-I. After the

indicated duration, whole-cell extracts were analyzed by SDS–PAGE and western blot analysis with

specific antibodies.

Figure 2. Twist1 binds to CLU promoter region and regulates CLU expression. (A) LNCaP cells were

transfected with 40 nM of the indicated siRNA. At 72 h after transfection, quantitative RT-PCR was

performed using the primers and probes for Twist1, CLU and GAPDH. Each transcript level from

cells transfected with control siRNA was set as 1. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with

control siRNA). Whole-cell extracts were analyzed by SDS–PAGE and western blot analysis with

specific antibodies. (B) LNCaP cells were transfected with 1.0 μg/mL of the indicated expression

plasmid. At 72 h after transfection, quantitative RT-PCR was performed using the primers and probes

for CLU and GAPDH. Each transcript level from mock-transfected cells was set as 1. Boxes, mean;

bars, ±s.d. *P < 0.05 (compared with mock). Whole-cell extracts were analyzed by SDS–PAGE and

western blot analysis with specific antibodies. (C) Schematic representation of the promoter region

and 5’ end of the CLU gene. Black box, E-boxes (5’-CANNTG-3’); gray box, CLE; white box,

AP-1–binding site. CLU–Luc plasmids (–1,998/+254, –1,998/–702, –1,116/–702, and –707/+254)

used in (d), (e) and (f) are shown. (D) LNCaP cells were cotransfected with 0.5 μg/mL of the various

CLU–Luc plasmids and 0.05 μg/mL of pRL-TK. The luciferase activity of CLU–Luc –1,998/+254

was set as 1. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with CLU-Luc –1,998/+254). (E) LNCaP

cells were cotransfected with 0.5 μg/mL of the various CLU–Luc plasmids, 0.5 μg/mL of Myc-Flag or

24

Twist1–Myc-Flag expression plasmid and 0.05 μg/mL of pRL-TK. The luciferase activity of

CLU–Luc with mock plasmid was set as 1. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with mock).

(F) LNCaP cells were cotransfected with 0.5 μg/mL of the CLU–Luc –1,998/+254 plasmid, 20 nM of

the indicated siRNA and 0.05 μg/mL of pRL-TK. The luciferase activity of CLU–Luc –1,998/+254

alone was set as 1. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with control siRNA). (G) ChIP

assays were conducted on nuclear extracts from LNCaP cells using the indicated antibodies. The

quantitative RT-PCR was carried out using� immunoprecipitated DNAs, soluble� chromatin, and

specific primer pairs for the CLU and RPL30 genes: CLU –02 targeting around –1,403 bp from TSS,

CLU –01 targeting around –417 bp from TSS, CLU +07 targeting around +6,603 bp from TSS and

RPL30 as negative control. The results of immunoprecipitated samples were corrected for the results

of the corresponding soluble chromatin samples. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with

IgG).

Figure 3. IGF-I induces CLU expression via Twist1. (A) LNCaP (left) and PC-3 (right) cells were

transfected with 40 nM of the indicated siRNA and incubated for 48 h, and then cells were treated

with 100 ng/mL IGF-I. After 1.5 h (Twist1) or 12 h (CLU), quantitative RT-PCR was performed using

the primer pairs and probes for Twist1, CLU and GAPDH. Each transcript level from non-treated cells

was set as 1. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with no treatment). (B) LNCaP (left) and

PC-3 (right) cells were transfected with 40 nM of the indicated siRNA and incubated for 48 h, and

then cells were treated with 100 ng/mL IGF-I. After the indicated duration, whole-cell extracts were

analyzed by SDS–PAGE and western blot analysis with specific antibodies. (C) LNCaP cells were

cotransfected with 0.5 μg/mL of the CLU–Luc –1,998/+254 plasmid, 20 nM of the indicated siRNA

and 0.05 μg/mL of pRL-TK, incubated for 24 h, and then cells were and treated with IGF-I. After 24 h,

the luciferase activity of CLU–Luc –1,998/+254 transfected with control siRNA without IGF-I was

set as 1. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with no treatment).

25

Figure 4. IGF-I activates STAT3, resulting in Twist1 up-regulation. (A) LNCaP cells were treated

with 100 ng/mL IGF-I. After the indicated duration, whole-cell extracts were analyzed by

SDS–PAGE and western blot analysis with specific antibodies. (B) LNCaP cells were treated with

100 ng/mL IGF-I. After 30 min, cells were harvested and fractioned into nuclear and cytoplasmic

extracts, and then extracts were analyzed by SDS–PAGE and western blot analysis with specific

antibodies. (C) ChIP assays were performed on nuclear extracts from LNCaP cells treated with 100

ng/mL IGF-I for 30 min using mouse IgG or anti-STAT3 antibody. The quantitative RT-PCR was

performed using immunoprecipitated DNAs, soluble chromatin and specific primer pairs for the

Twist1-promoter region. Results of immunoprecipitated samples were corrected for the results of the

corresponding soluble chromatin samples. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with no

treatment). **P < 0.05 (compared with treatment with IGF-I). (D) ChIP assays were performed on

nuclear extracts from LNCaP cells treated with 100 ng/mL IGF-I for 6 h using mouse IgG or

anti-Twist1 antibody. The quantitative RT-PCR was performed using immunoprecipitated DNAs,

soluble chromatin and specific primer pairs for the CLU-promoter region. Results of

immunoprecipitated samples were corrected for the results of the corresponding soluble chromatin

samples. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with no treatment). (E) LNCaP cells were

cotransfected with 0.5 μg/mL of Twist–Luc plasmid and 0.05 μg/mL of pRL-TK, incubated for 36 h

(Twist1), and then cells were treated with 100 ng/mL IGF-I. After 4 h (Twist1), the luciferase activity

of Twist–Luc without IGF-I was set as 1. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with no

treatment.

Figure 5. IGF-I/Twist1/CLU signaling in IGF-I-deficient mice. (A) Serum IGF-I concentration was

measured from PTEN knockout lit/+ mice and PTEN knockout lit/lit mice. Boxes, mean; bars, ±s.e.m.

*P < 0.05 (compared with lit/+). (B) Quantitative RT-PCR was performed using the primer pairs and

probes for murine Twist1, murine CLU and murine GAPDH. Each transcript level from lit/+ mice was

set as 1. Boxes, mean; bars, ±s.e.m. *P < 0.05 (compared with lit/+). (C) Immunohistochemistry

26

against the indicated antibodies in prostate tissues from PTEN knockout lit/+ mice and PTEN

knockout lit/lit mice was shown.

Figure 6. IGF-I promotes prostate cancer cell growth in IGF-I-deficient serum through Twist1 and

CLU. (A) LNCaP cells transfected with 10 nM of the indicated siRNA were cultured in 1% serum

from lit/lit or lit/+ mice (left), or 1% serum from lit/lit mice with or without 100 ng/mL IGF-I (right).

After 96 h, growth of cells transfected with control siRNA and cultured in 1% serum from lit/lit mice

was set as 1. Boxes, mean; bars, ±s.d. *P < 0.05 (compared with lit/lit or IGF-I–). **P < 0.05

(compared with control siRNA). (B) Schematic representation of signaling pathway from IGF-I to

CLU in this study.

Figure 1 Takeuchi et al.

0 6 12 24 IGF-I (h)0 6 12 24 IGF-I (h)

B

CLU

LNCaP

�-actin

Twist1

CLU

PC-3

�-actin

Twist1

A

0

0.5

1

1.5

2

2.5

3

3.5

4

0 h 1.5 h 6 h 12 h 24 h

Rela

tive m

RN

A e

xp

ressio

n LNCaP Twist1

CLU

IGF-I

*

*

** *

*

0

0.5

1

1.5

2

2.5

3

0 h 1.5 h 6 h 12 h 24 h

Rela

tive m

RN

A e

xp

ressio

n

PC-3 Twist1

CLU

IGF-I

**

*

**

**

*

Figure 2 Takeuchi et al.

-1998Luciferase+254

CLU-Luc -1998/+254

-707Luciferase+254

CLU-Luc -707/+254

-1998Luciferase

-702CLU-Luc -1998/-702

-1116Luciferase

-702CLU-Luc -1116/-702

+1-2000 +500

CLE E-boxAP-1

ClusterinC

CLU

Twist1

�-actin

LNCaP

A

0

0.2

0.4

0.6

0.8

1

1.2

1.4

Control

siRNA

Twist1

siRNA

#1

Twist1

siRNA

#2

Rela

tive m

RN

A e

xp

ressio

n

LNCaP Twist1

CLU

Twist

**

B

CLU

LNCaP

�-actin

IB;Myc

Twist1-Myc-Flag

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Myc-Flag Twist1-Myc-Flag

Rela

tive m

RN

A e

xp

ressio

n LNCaPCLU*

F

0

0.2

0.4

0.6

0.8

1

1.2

Control

siRNA

Twist1

siRNA #1

Twist1

siRNA #2

Rela

tive lu

cif

era

se a

cti

vit

y

LNCaP

CLU-Luc

**

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

CLU-Luc

-1998/+254

CLU-Luc

-1998/-702

CLU-Luc

-1116/-702

CLU-Luc

-707/+254

Rela

tive lu

cif

era

se a

cti

vit

y LNCaPD

*

0

0.5

1

1.5

2

2.5

CLU-Luc

-1998/+254

CLU-Luc

-1998/-702

CLU-Luc

-1116/-702

CLU-Luc

-707/+254

Rela

tive lu

cif

era

se a

cti

vit

y LNCaP Myc-FlagTwist1-Myc-Flag

E

* *

0

0.2

0.4

0.6

0.8

1

1.2

CLU -02 CLU -01 CLU +07 RPL30

Imm

un

op

rec

ipit

an

ts/in

pu

t

LNCaPIgG

Histone H3

Twist1

G

*

Figure 3 Takeuchi et al.

0 6 12 24 0 6 12 24 IGF-I (h) 0 6 12 24 0 6 12 24 IGF-I (h)

Twist1

�-actin

Control

siRNA

CLU

Twist1

siRNA #1

LNCaPB

Twist1

�-actin

Control

siRNA

CLU

Twist1

siRNA #1

PC-3

C

IGF-I

0

0.5

1

1.5

2

2.5

3

3.5

+

Rela

tive lu

cif

era

se a

cti

vit

y

LNCaP

CLU-Luc

Control siRNATwist1 siRNA #1

*

0

0.5

1

1.5

2

2.5

3

3.5

+

Rela

tive m

RN

A e

xp

ressio

n

LNCaP

Twist1

Control siRNATwist1 siRNA #1

IGF-I

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

+

Rela

tive m

RN

A e

xp

ressio

n

LNCaP

CLU

Control siRNATwist1 siRNA #1

IGF-I

0

0.5

1

1.5

2

2.5

+

Rela

tive m

RN

A e

xp

ressio

n

PC-3

Twist1

Control siRNATwist1 siRNA #1

IGF-I

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

+

Rela

tive m

RN

A e

xp

ressio

n

PC-3

CLU

Control siRNATwist1 siRNA #1

IGF-I

A

* *

**

0

0.5

1

1.5

2

2.5

LNCaP Ctrl LNCaPIGF (6h)

Imm

un

op

rec

ipit

an

t/in

pu

t

LNCaP

CLU -02

D

IGF-I

*

0 15 30 60 IGF-I (min)

A

�-actin

LNCaP

p-STAT3Tyr705

STAT3

p-STAT3Ser727

– + IGF-I

B

N

STAT3

C N C

LNCaP

Lamin B1

�-actin

Figure 4 Takeuchi et al.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

IgG STAT3

Imm

un

op

rec

ipit

an

t/in

pu

t LNCaP

Twist1 promoter

IGF-I

IGF-I+

C

*

0

0.5

1

1.5

2

2.5

+

Rela

tive lu

cif

era

se a

cti

vit

y LNCaP

Twist-Luc

IGF-I

E

*

– +

Figure 5 Takeuchi et al.

0

50

100

150

200

250

300

lit/+ lit/lit

Seru

m IG

F-I

co

ncen

trati

on

(ng

/mL

)

Serum IGF-IA

*

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

lit/+ lit/lit

Rela

tive m

RN

A e

xp

ressio

n

Twist1

CLU

B

*

CLU IGF-IR p-STAT3Tyr705 Twist1

lit/

+ m

ice

lit/

lit

mic

e

C

Figure 6 Takeuchi et al.

A

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

Control

siRNA

Twist1

siRNA #1

CLU

siRNA

Rela

tive c

ell

gro

wth

LNCaP

lit/litlit/+*

** **

Media

0

0.2

0.4

0.6

0.8

1

1.2

1.4

Control

siRNA

Twist1

siRNA #1

CLU

siRNA

Rela

tive c

ell

gro

wth

LNCaP

IGF-I IGF-I+

*

** **

Media

IGF-I

IGF-IR

Nucleus

STAT3 CLUTwist1

STAT3

Twist1

IGF-I IGF-I

Cancer cell

growth

P

P

B

27

Highlights

� Insulin-like growth factor-I (IGF-I) transcriptionally increased Twist1 levels as well as

clusterin (CLU) expression.

� At transcriptional level, Twist1 regulated CLU expression and mediated CLU induction by

IGF-I.

� As upstream pathway of Twist1, STAT3 mediated Twist1/CLU induction by IGF-I.

� Twist1 and CLU expressions were suggested to be lower in IGF-I–deficient mice.

� Both Twist1 and CLU knockdown suppressed prostate cancer cell proliferation promoted

by IGF-I.