Sunteți pe pagina 1din 11

Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.

CAN-10-1722

Cancer
Microenvironment and Immunology Research

Cancer Exosomes Trigger Fibroblast to Myofibroblast


Differentiation
Jason Webber1,3, Robert Steadman2,3, Malcolm D. Mason1, Zsuzsanna Tabi1, and Aled Clayton1

Abstract
There is a growing interest in the cellcell communication roles in cancer mediated by secreted vesicles
termed exosomes. In this study, we examined whether exosomes produced by cancer cells could transmit
information to normal stromal fibroblasts and trigger a cellular response. We found that some cancer-derived
exosomes could trigger elevated a-smooth muscle actin expression and other changes consistent with the
process of fibroblast differentiation into myofibroblasts. We show that TGF-b is expressed at the exosome
surface in association with the transmembrane proteoglycan betaglycan. Although existing in a latent state,
this complex was fully functional in eliciting SMAD-dependent signaling. Inhibiting either signaling or
betaglycan expression attenuated differentiation. While the kinetics and overall magnitude of the response
were similar to that achieved with soluble TGF-b, we identified important qualitative differences unique to
the exosomal route of TGF-b delivery, as exemplified by a significant elevation in fibroblast FGF2 production.
This hitherto unknown trigger for instigating cellular differentiation in a distinctive manner has major
implications for mechanisms underlying cancer-recruited stroma, fibrotic diseases, and wound-healing
responses. Cancer Res; 70(23); 962130. 2010 AACR.

Introduction in association with the exosome membrane. Examples


include expression of tumor necrosis factor-alpha (TNF-a;
Exosomes are nanometer-sized vesicles secreted by diverse ref. 8), epidermal growth factor (EGF; ref. 9), fibroblast
cell types that play complex roles in intercellular communica- growth factor (FGF; ref. 10), and others. Collectively, such
tion (1). They comprise a ceramide- and cholesterol-rich lipid data suggest a possible physiologic role for exosomes as a
bilayer membrane (2), an array of membrane and cytosolic novel means of disseminating growth factors and other
proteins (3) and selected RNA species (4). This molecular mediators within the extracellular milieu.
complexity suggests that exosomes may mediate a variety of Our own previous studies have demonstrated the expression
physiologic functions, perhaps through several different of TGF-b by cancer-derived exosomes and linked this with their
mechanisms. There is currently considerable interest in under- immunosuppressive properties (11, 12). These findings were
standing the complex communicative roles exosomes play in confirmed recently by others (13). Mechanistic studies of
physiology and the molecular mechanisms underlying these. exosomal TGF-b remain incomplete, and the potentially broad
Some mechanisms by which exosomes mediate their influence exosomal TGF-b could play in other, non-immune,
biological functions have been characterized. Exosomes biological systems has not yet been explored. To gain more
may act as natural vehicles for delivering protein (5), mRNA insight into these aspects of exosome function, we chose to
(4), or microRNA (6) to recipient cells. A further mechanism focus on a biological function of TGF-b that is of major
involves the direct engagement of receptors/ligands at the importance in health and disease conditions: the induction
recipient cell surface during exosome interactions (7). In of fibroblast differentiation to myofibroblasts. We hypothesize
addition, some recent reports have described the expression that cancer exosomes can alter the function of healthy fibro-
of growth factors, or other usually soluble mediators, found blast present within the tumor stroma, leading to alterations
that may promote tumor growth and progression.
Myofibroblasts are a contractile cell type, characterized by
Authors' Affiliations: 1Department of Pharmacology, Oncology & Radia- the expression of a-smooth muscle actin (a-SMA; ref. 14).
tion and 2Institute of Nephrology, School of Medicine, and 3Institute of During normal wound healing, myofibroblasts are transiently
Tissue Engineering and Repair, Cardiff University, Cardiff, UK
present, responsible for closure of the wound and formation of
Note: Supplementary material for this article is available at Cancer the collagen-rich scar matrix (15). When persistent in tissues,
Research Online (http://cancerres.aacrjournals.org/).
they are a well-established early histologic marker of progres-
Corresponding Author: Aled Clayton, Department of Pharmacology,
Oncology & Radiology, School of Medicine, Cardiff University, Velindre sive organ fibrosis, acting inappropriately to modulate the
Cancer Centre, Whitchurch, Cardiff CF14 2TL, UK. Phone: 44-29-2019- extracellular milieu and altering tissue architecture (15).
6148. Fax: 44-29-2052-9625; E-mail: aled.clayton@wales.nhs.uk. Myofibroblasts are also relevant in solid cancers in which
doi: 10.1158/0008-5472.CAN-10-1722 an altered stroma, rich in myofibroblastic cells, can support
2010 American Association for Cancer Research. tumor growth, vascularization, and metastasis (1618). TGF-b

www.aacrjournals.org 9621

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Webber et al.

remains among the key factors responsible for the develop- Determine exosome density
ment of a myofibroblastic phenotype from a variety of pre- A pellet obtained from 70,000  g ultracentrifugation of
cursor cells, including fibroblasts (15). TGF-b binds culture medium was overlaid onto a prepoured sucrose
sequentially to type II receptor and then type I receptor gradient (0.22.5 mol/L of sucrose). Specimens were
(ALK5), resulting in the formation of an activated receptor centrifuged at 4 C overnight at 210,000  g, using an
complex (19). Type III TGF-b receptors, known as betaglycan MLS-50 rotor in an Optima-Max ultracentrifuge (Beckman
(20), do not play a direct role in TGF-b signaling but may act to Coulter). The refractive index of 16 collected fractions was
facilitate binding of TGF-b to type II receptors (21). Activation measured at 20 C, using an automatic refractometer
of ALK5 triggers a signaling cascade, involving phosphoryla- (J57WR-SV; Rudolph Research). The refractive index was
tion of SMAD2 and 3, which subsequently mediate binding to converted to density, using the conversion table published
SMAD4. This SMAD complex translocates to the nucleus to in the Beckman Coulter ultracentrifugation manual as
initiate transcription of a host of genes, including a-SMA (22). described (25, 26).
SMAD-independent signaling pathways also exist (23) and
contribute to other aspects of a myofibroblastic phenotype Enzyme-linked immunosorbent assay
such as alterations in the pericellular milieu (24). Quantitation of TGF-b1 and FGF2 proteins was performed
In this report, we examined whether cancer exosomes can using the DuoSet ELISA Development System (R & D Systems)
trigger TGF-b signaling pathways and thereafter initiate a and manufacturer's protocol.
program of differentiation of fibroblasts toward a myofibro-
blastic phenotype. The data demonstrate exosomal delivery of SMAD3 reporter assay
TGF-b as a previously unknown mechanism capable of driving Assessment of TGF-b signaling through an SMAD3-depen-
cellular differentiation of fibroblasts, resulting in a myofibro- dent pathway was performed as described (27) by transfect-
blast phenotype that is similar but not identical to that ing HK2 cells (from ATCC) with an SMAD3-responsive
achieved with soluble TGF-b. The study has broad implica- promoter construct (SBE)4-Lux (0.9 mg) and a Renilla vector
tions for how growth factors are disseminated in the extra- (0.1 mg) to monitor transfection efficiency, using the trans-
cellular environment during health and disease conditions fection reagent Lipofectamine 2000 (Invitrogen) at a ratio of
and highlights a previously unknown function for cancer 3:1 (microliters of lipofectamine:micrograms of DNA) in
exosomes as novel modulators of stromal cell differentiation. serum-free and antibiotic-free medium. Twenty-four hours
after transfection, cells were washed with PBS and the
Materials and Methods medium was replaced with serum-free medium containing
purified exosomes or rhTGF-b1 for a further 6 hours. Luci-
Cell culture and reagents ferase activity was measured using the Dual-Glo luciferase
A panel of cancer cell lines, as exosome producers, is main- activity kit (Promega Ltd.) as per the manufacturer's
tained in bioreactor culture flasks, as described (12). Cells protocol. Luciferase activity was normalized to Renilla
included a mesothelioma cell line (established within the activity.
department from pleural effusion of a mesothelioma patient);
prostate cancer cells (LNCAP, DU145, and PC3) from American Immunohistochemistry of a-SMA
Type Culture Collection; a bladder cancer line, HT1376, and a Indirect immunohistochemistry was performed following
colorectal cancer line (CaCo2), from Cancer Research UK; and a fixation in ice-cold acetone:methanol (1:1). Blocking and anti-
breast cancer line (MCF7), from the European Collection of Cell body (anti-a-SMA; Santa Cruz) incubations were performed in
Cultures. Primary fibroblasts (AG02262) from Coriell Institute Hank's balanced salt solution (Sigma). For microscopic visua-
for Medical Research (Camden, NJ), were used at passages 7 to lization of cells, a secondary goat anti-mouse Alexa-488-con-
9 as exosome responder cells, following growth arrest in serum- jugated antibody (Invitrogen) was used. To quantify changes
free media for 48 hours. All purchased cells were obtained in a-SMA, this was substituted for a goat anti-mouse biotin
during 2008 to 2009 and passaged for less than 6 months in conjugate (Perkin Elmer), and detected using a streptavidin-
culture. Cell lines were authenticated by the supplier cell bank europium conjugate. Expression levels were measured on a
by combinations of cytogenetic, isoenzymatic, and DNA profile Wallac Victor2 1420 plate reader (Perkin Elmer).
analyses. All cultures were maintained in 5% to 10% FBS that
had been depleted of bovine exosomes by overnight ultracen- Quantification of hyaluronic acid coat thickness
trifugation at 100,000  g and serial filtration (0.2 and 0.1 mm). Hyaluronic acid (HA)-dependent, cell-associated pericellu-
rhTGF-b1 was from Peprotech. lar matrices were visualized and measured, as previously
described (28).
Exosome purification
Cell culture medium was subject to serial centrifugation, Flow cytometry of exosome-coated beads
(300  g for 10 minutes, 2,000  g for 15 minutes, and 10,000  g Exosomes were coupled to the surface of aldehyde sulfate
for 30 minutes), and exosomes were purified from the super- latex beads (Interfacial Dynamics) in an MES buffer (at a ratio
natant by the 30% sucrose/D2O cushion method (11). Exosome of 1 mg of exosomes:1 mL of stock beads) and analyzed by flow
quantity was determined by the micro BCA protein assay cytometry as previously described (25). Flow cytometric ana-
(Pierce/Thermo Scientific). lysis was performed using mouse monoclonal antibodies

9622 Cancer Res; 70(23) December 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Cancer Exosomes Trigger Fibroblast Differentiation

specific to human CD9 (R & D Systems) and human betagly- performing a sucrose gradient, the density of collected frac-
can (Santa Cruz), corresponding mouse IgG isotype controls tions was measured, and thereafter fractions were analyzed by
(eBioscience), and an Alexa Fluor 488 conjugated goat anti- Western blot, TGF-b-ELISA, and an SMAD3 reporter assay.
mouse IgG (Invitrogen). The Western blot revealed that the multivesicular body mar-
ker TSG101, used as a marker for exosomes, was principally
Pervanadate-induced cleavage of betaglycan present in fractions 9 to 14 (density of 1.0961.23 g/mL;
Exosomes were added to mesothelioma cells and incubated Fig. 1A). This is broadly consistent with similar studies of
at 37 C for 4 hours in either the absence or presence of exosomes from other cellular sources (7). Performing a TGF-b
pervandate (200 mmol/L), previously shown to instigate clea- ELISA on these fractions revealed little TGF-b present outside
vage of betaglycan (29). Exosomes were then repurified and the exosome density range, with fractions 9 to 14 comprising
quantified, as described previously, prior to further analysis. 84.2% of the total TGF-b present. This demonstrates that
the TGF-b present within the 70,000  g pellet of culture
RNA extraction medium is predominantly present at exosomal densities
RNA was extracted from cultured cells, using the Purelink (Fig. 1A, black bars), with the remainder at hyperdense frac-
RNA Mini kit (Invitrogen) as per the manufacturer's protocol. tions (which contained trace amounts of TSG101), which may
be due to exosome aggregates and/or nonexosomal constitu-
Reverse transcription ents. However, to be certain that soluble TGF-b cannot float at
Reverse transcription was performed using a high-capacity these densities as well, an additional gradient run was per-
cDNA reverse transcriptase kit (Applied Biosystems), in a final formed, loading rhTGF-b instead of concentrated exosomes.
volume of 20 mL per reaction, as described in the man- On this occasion, the majority (92.8%) of TGF-b was localized
ufacturer's protocol, using an Applied Biosystems ''Gene within the first 4 fractions (density 1.021.04 g/mL) and not at
Amp PCR System 2400'' thermocycler. exosomal densities (Fig. 1A, line graph). The data show that
TGF-b has a molecular association with exosomes and that
Real-time quantitative PCR this association does not provide information regarding exo-
Quantitative PCR (qPCR) was carried out in a final volume somal TGF-b function. We therefore examined whether exo-
of 20 mL per reaction, containing 1 mL of cDNA, 10 mL of somes were capable of activating TGF-b signaling pathways.
TaqMan Universal Master Mix (20) (Applied Biosystems), To do this, we transfected cells with an SMAD3 promoter
8 mL of H2O, and 1 mL of a TaqMan gene expression assay element upstream of luciferase and measured the activation of
primer and probe mix (Applied Biosystems). A negative con- SMAD3-dependent transcriptional activity by chemilumines-
trol (PCR) was prepared with H2O substituted for the cDNA. cence (27). The sucrose gradient fractions (as described ear-
The PCR amplification was performed in a StepOne Real-Time lier) were used and showed a weak background SMAD3
PCR System thermocycler (Applied Biosystems). Amplifica- activity (11.5% of total activity) at hypodense fractions (1.0
tion was carried out by heating the samples to 50 C for 1.08 g/mL). The majority of transcriptional activity was found
2 minutes, then at 95 C for 10 minutes, followed by repeating at exosomal densities (65.3% of total) (Fig. 1A, gray bars).
cycles of 95 C for 15 seconds and 60 C for 1 minutes, for a total Again, there was some activity present at hyperdense frac-
of 40 cycles. The comparative Ct method was used for relative tions, consistent with the TGF-b ELISA and trace TSG101
quantification of target gene expression against that of a levels by Western blot. We conclude, therefore, that the cancer
standard reference gene (GAPDH). Data were analyzed using exosomes that express functional TGF-b1 are capable of
StepOne software from Applied Biosystems UK Ltd. triggering the SMAD3-dependent intracellular signaling path-
way. To determine whether this was applicable to other
Results cancers, we purified exosomes from a diverse panel of cancer
cell lines, using a sucrose cushion method (11), and, as
Cancer exosomes express latent TGF-b but can trigger described earlier, examined TGF-b expression by ELISA and
the SMAD3-related signaling pathway function by SMAD3 reporter assay. The data show that not all
We have previously shown by ELISA that cancer exosomes, cells produce TGFb-positive exosomes. Exosomes isolated
isolated from cultured mesothelioma cells, express TGF-b1 from a prostate cancer cell (LNCAP) and from a colorectal
(11). It is possible that during the exosome purification cancer cell line (CaCo2) had low or undetectable levels of TGF-
process involving high-speed ultracentrifugation, TGF-b b, respectively (Fig. 1B, black bars). The sensitivity of the assay
may be co-pelleted with exosomes and hence may not be was 20 pg/mL. Other cell types (DU145 or PC3, prostate cancer
genuinely expressed in association with the vesicle. To clarify cells), however, produced exosomes with similar levels of TGF-
this issue, we performed a classical purification of exosomes, b1 compared with the mesothelioma cell line (Meso). When
using a sucrose gradient, as described (26). This technique comparing TGF-b1 expression levels and SMAD3 transcrip-
separates exosomes from nonexosomal soluble contaminants tional activity, there was a good correlation: exosomes highest
by virtue of the biophysical nature of vesicles, which float on in TGF-b1 eliciting the strongest activation of SMAD3-depen-
sucrose at particular densities (typically around 1.11.2 g/mL dent signaling (Fig. 1B, gray bars). The ELISA approach
for most exosome types). requires the liberation of TGF-b1 from a nonfunctional latent
Exosomes were initially concentrated from mesothelioma complex by brief acid treatment to allow the capture and
cell culture medium by spinning at 70,000  g, and on detection antibodies to bind TGF-b. This measurement

www.aacrjournals.org Cancer Res; 70(23) December 1, 2010 9623

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Webber et al.

A B

C D

Figure 1. Cancer exosomes express functional TGF-b1. A, pellet (centrifuged at 70,000  g) from cultured mesothelioma cellconditioned medium was
overlaid on a sucrose gradient (0.22.5 mol/L) and spun at 210,000  g for 20 hours. Fractions (116) were collected, and the density of each was measured
by refractometry. Each fraction was analyzed by Western blot (staining for the multivesicular body marker TSG101), SMAD3 reporter assay, and TGF-b1
ELISA. Graphs show mean  SD of duplicates and depict percentage of total SMAD3 activity (gray bars) or percentage of total TGF-b1 (black bars)
against fraction density. Inset line graph, ELISA data from fractions taken from a sucrose gradient loaded with soluble rhTGF-b1, highlighting soluble rhTGF-b1
(white circles) and exosomal TGF-b1 (black circles) floats at distinct densities. B, exosomes (200 mg/mL) purified from various cancer cell lines (as indicated)
were subject to analysis by SMAD3 activity assay (gray bars), showing mean  SD of triplicate measurements, and by TGF-b ELISA (black bars),
showing mean  SD of 3 preparations from each line. C, TGF-b ELISA was performed on exosomes (10 mg) prepared from 3 cell lines as indicated, without
(active) and with (total) prior activation with acid, revealing exosomal TGF-b to be predominantly in a latent form, showing mean  SD of 4 replicates.
D, similarly, the analysis of exosomes by SMAD3 activity assay, without (active) and with (total) prior acid activation (bars, mean  SD of 4 replicates), revealing
comparable levels of exosomal TGF-b function. In addition, an SMAD3 activity assay using soluble rhTGF-b (02,000 pg/mL) was performed showing similar
levels of functional activity to exosomal TGF-b (line graph, mean  SD of 4 replicates).

(Fig. 1A and B, black bars) is of total TGF-b1 within the is therefore a measure of biologically active/available TGF-b.
specimen rather than a measure of constitutively active TGF- To determine what proportion of exosomal TGF-b was in a
b1. In the SMAD3 reporter assay (Fig. 1A and B, gray bars), we constitutively active conformation, we compared active (not
did not acid activate exosomes prior to functional testing and acid-treated) with acid-activated (total) TGF-b in exosome

9624 Cancer Res; 70(23) December 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Cancer Exosomes Trigger Fibroblast Differentiation

A B C

Figure 2. Cancer exosomes trigger a-smooth muscle (a-SAM) actin expression in fibroblasts. A, primary fibroblasts were growth arrested for 72 hours prior
to stimulation with exosomes isolated from various cell lines as specified (at a dose of 200 mg/mL). As a positive control, fibroblasts were treated with
rhTGF-b1 (5 ng/mL). After a further 72 hours, the fibroblasts were examined by immunohistochemistry for the expression of a-SAM. Scale bar, 100 mm.
B, similarly, fibroblasts were stimulated with exosomes (mesothelioma exosomes shown) or rhTGF-b1 in the presence or absence of a TGF-b1neutralizing
antibody (at 10 mg/mL) or an irrelevant isotype-matched antibody (at 10 mg/mL). Also, selected regions exemplify the a-SMApositive filamentous structures
along the longitudinal axis of the fibroblast cell body that are apparent on rhTGF-b1 (red box) or exosome treatment (blue box). Scale bar, 100 mm.
C, comparison of the kinetics of induced a-SMA expression over a 14-day period, with growth-arrested fibroblasts left untreated (gray circles) and treated with
200 mg/mL of mesothelioma-derived exosomes (black circles) or with 1.5 ng/mL of rhTGF-b1 (white circles). Graph depicts expression levels relative to
untreated fibroblasts at day 1, mean  SD, of triplicates.

preparations. By ELISA, data revealed that approximately 2% cells exhibiting the classical filamentous structures of a-SMA
of the exosomally associated TGF-b is constitutively active, typical of myofibroblasts (Fig. 2B, blue and red boxes). To
hence 98% exist in a latent form (Fig. 1C). We next performed determine whether this effect was dependent on exosomal
this comparison by using the SMAD3 assay, revealing that expression of TGF-b, or perhaps due to some other exosomal
there was a little functional difference between active and constituent(s), we repeated the experiments in the presence of
total (acid-treated) exosomal TGF-b (Fig. 1D, left), and the a TGF-b neutralizing antibody. This was effective in abrogating
magnitude of response to exosomal TGF-b in this assay was a-SMA induction by both treatments (Fig. 2B), whereas an
80% to 95% of that achieved with rhTGF-b (in which 100% is in isotype-matched monoclonal antibody had no effect. TGF-b is
an active form) (Fig. 1D, right). Together, the data show for the known to drive long-lasting phenotypic change in fibroblasts;
first time that some cancer cells produce exosomes that we therefore compared the kinetics of a-SMA induction trig-
express latent TGF-b1, which is presented to recipient cells gered by rhTGF-b with that triggered by exosomes (Fig. 2C).
in a fully biologically available manner, eliciting an SMAD3 The data reveal that changes elicited by each treatment were
signaling response of similar magnitude to that achieved using similar. Both triggered rapid elevation in expression (2-fold
nonlatent rhTGF-b. increase by 24 hours), reaching a peak of expression by day 3,
although the exosome-mediated increase was less pronounced
TGF-bhigh cancer exosomes trigger a-SMA expression than that driven by rhTGF-b under these conditions. There-
We next investigated the effects of TGF-b-expressing exo- after, a-SMA levels reverted slightly by day 7 (16.9% decrease in
somes on fibroblasts. It is well established that soluble TGF-b rhTGF-b, 17% decrease in exosome groups) but remained well
induces the key phenotypic marker of myofibroblasts, a-SMA, above untreated levels for at least 14 days. In conclusion, cancer
which involves its assembly into contractile stress fibers along exosomes that express relatively high levels of TGF-b can drive
the longitudinal axis of the cell body. Using immunohistochem- fibroblast to myofibroblast differentiation. This effect is long-
istry, we visualized the expression of a-SMA in growth-arrested lasting and comparable with that achieved using rhTGF-b.
primary fibroblasts that had been left untreated for 72 hours, or
stimulated either with rhTGF-b or with exosomes (from var- TGF-bhigh cancer exosomes trigger the formation of a
ious cellular sources). As expected, by 72 hours, rhTGF-b hyaluronic acid pericellular coat
strongly induced a-SMA expression. Exosomes that expressed An additional key feature of myofibroblast differentiation is
low or intermediate levels of TGF-b (as shown in Fig. 1B, black the alteration in the immediate pericellular microenvironment.
bars) did not trigger a-SMA expression above that of untreated One classical indicator of this is the formation of a pericellular
cells (Fig. 2A, left column). In contrast, exosomes expressing 6 coat comprised largely of HA surrounding the myofibroblast
pg or more of TGF-b/mg of exosomes were effective at inducing cell body (28). Such HA coats are barely detectable around
this phenotypic alteration (Fig. 2A, right column). These fibroblasts. We visualized HA coats by using fixed erythrocytes
changes, induced by TGF-bhigh exosomes, were morphologi- that are electrostatically repulsed by HA, revealing a region of
cally comparable with the changes triggered by rhTGF-b, with exclusion around the cell body. Growth-arrested fibroblasts

www.aacrjournals.org Cancer Res; 70(23) December 1, 2010 9625

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Webber et al.

A B

Figure 3. Cancer exosomes trigger deposition of a hyaluronic acid pericellular coat. A, growth-arrested fibroblasts were treated for 72 hours with 200 mg/mL
of either CaCo2-derived exosomes (which are TGF-b1low) or with PC3-derived exosomes (TGF-b1high). As a positive control, fibroblasts were stimulated
with rhTGF-b1 (5 ng/mL). A type I TGF-b-RI inhibitor (SB431542, at 10 mmol/L) was added to some wells as indicated. After a further 72 hours,
hyaluronidase (HAidase) was added (at 200 mg/mL) to some wells as indicated. After washing, formalin-fixed horse erythrocytes were added and
allowed to settle (for at least 15 minutes, at 37 C). Scale bar, 100 mm. B, regions of exclusion (examples indicated by arrows) were measured for 20
representative cells for each treatment condition, and the pericellular thickness (calculated as described in Materials and Methods) is shown. Graph
represents mean  SD of pericellular coat thickness (n 20 cells). ***P < 0.0001, using 1-way ANOVA followed by Tukey's honest significant
difference method.

exhibited very slender HA coats (Fig. 3), whereas rhTGF- or isotype-matched control antibodies and analyzed by flow
bmediated myofibroblast differentiation demonstrated sig- cytometry. Betaglycan showed strong staining at the exo-
nificantly broader regions of exclusion. This phenomenon was some surface; this was true of multiple preparations from 3
also apparent when stimulating growth-arrested fibroblasts different cancer cell lines that had high levels of TGF-b
with cancer exosomes, but again this was related to the TGF-b (mesothelioma, PC3, and DU145 cells) (Fig. 4A and B).
content of exosomes (Fig. 3). HA coat formation, in response to However, exosomes from cancer cells that had low levels
either cancer exosomes or rhTGF-b, was shown to be TGF-b of TGF-b (CaCo2 and HT1376 cells) exhibited markedly lower
dependent, as it was attenuated by the addition of the TGF-b (4- to 5-fold) levels of betaglycan (Fig. 4A and B). Plotting
receptor I (TGF-b-RI) inhibitor SB431542. The region of ery- exosomal surface betaglycan expression versus exosomal
throcyte exclusion was confirmed to be composed of HA by TGF-b levels reveals a significant correlation (r2 0.8894,
digestion of the coat with protease-free hyaluronidase. We P < 0.0001; Fig. 4C). In contrast, plotting TGF-b levels against
conclude that exosomal TGF-b can cause alterations, including another exosomal surface protein, CD9, revealed no such
modulations in pericellular microenvironment, in fibroblasts correlation (r2 0.1224, with the slope not significantly
that exhibit multiple hallmarks of myofibroblasts. different from zero, P 0.201). The data suggest that beta-
glycan may, at least, be partly involved in binding TGF-b at
Exosomes require betaglycan expression for the exosome surface. We therefore attempted to show
tethering TGF-b whether or not this was so, using a siRNA approach. While
The aforementioned data have posed the question: How is selective knockdown of betaglycan mRNA in cancer cells was
TGF-b physically associated with exosomes? TGF-b may be achievable with moderate efficiency (50%60% decrease in
rendered membrane associated, such as through interactions betaglycan mRNA), it was difficult to scale up these experi-
with certain integrins (avb1, avb6, a8b1), with endoglin, with ments to provide sufficient quantity of exosomes for func-
thrombospondin, or by binding to the type III receptor tional assays. Nevertheless, this approach supported the
betaglycan (30). The exosomal expression of such molecules premise of betaglycan involvement, as the small amounts
with roles in TGF-b tethering function(s) has not been of exosomes available exhibited approximately 30% reduc-
shown. We initially focused on possible expression of beta- tion in TGF-b levels compared with scrambled siRNA con-
glycan by cancer exosomes, as previous pilot data from our trols (Supplementary Figure S1). As an alternative approach,
allied studies had suggested proteoglycans might be we utilized pervanadate as a reagent that induces the release
expressed by exosomes. We purified exosomes from several of betaglycan from cell surfaces (29). Purified mesothelioma
cancer cell lines, including those high and those low in TGF-b exosomes were exogenously added to mesothelioma cells in a
expression. Following coupling to latex beads (25), the exo- normal (without pervanadate) or a betaglycan-cleaving
somebead complexes were stained with betaglycan-specific microenvironment (with pervanadate). After 4 hours, the

9626 Cancer Res; 70(23) December 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Cancer Exosomes Trigger Fibroblast Differentiation

B C

D E F

Figure 4. Exosomal betaglycan and TGF-b binding. A, exosomes purified from cancer cell cultures were coupled to the surface of latex beads and analyzed by
flow cytometry for betaglycan expression. Representative flow cytometric histograms (left) showing isotype versus betaglycan-stained exosomebead
complexes from a TGF-bhigh source (Meso-Exo) or a TGF-blow source (CaCo2-Exo), showing the median fluorescence intensity (MFI) of the peak. B, summary
data (bars) from 3 exosome preparations from each cell line as indicated, emphasizing a difference in betaglycan expression between TGF-bhigh exosomes
(dark gray bars) and TGF-blow exosomes (light gray bars), showing mean  SD, (n 3). C, in addition, linear regression reveals correlation between exosomal
TGF-b expression and betaglycan expression (line graph, white circles) which is significant. In comparison, a typical exosome surface protein, CD9, does
not correlate with TGF-b1 expression levels (black circles). DF, mesothelioma-derived exosomes were added to mesothelioma cells and incubated in either
the absence or presence of pervanadate (200 mmol/L) for 4 hours. After this time, conditioned media were collected and exosomes were purified. D, flow
cytometry was used to confirm pervanadate-induced cleavage of betaglycan (black bars) from exosomes, while CD9 expression was used as a control to show
that pervanadate treatment did not result in nonspecific cleavage of exosomal proteins. E, TGF-b1 expression of exosomes (10 mg), incubated with
mesothelioma cells in either the absence or presence of pervanadate, was analyzed by ELISA. F, comparison of a-SMA induction in growth-arrested
fibroblasts treated with exosomes incubated in either the absence or presence of pervanadate. Results in panels D, E, and F rerpresent mean  SD of
triplicates. **P < 0.001, ***P < 0.0001, Student's t-test.

www.aacrjournals.org Cancer Res; 70(23) December 1, 2010 9627

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Webber et al.

exosomes were retrieved and washed. Flow cytometric ana-


lysis of exosomebead complexes showed that pervanadate A
treatment had little effect on exosomal CD9 expression
(Fig. 4D, white bars) yet reduced exosomal betaglycan by
around 40% to 50% (Fig. 4D, black bars). In addition, analysis
by ELISA revealed a similar reduction in exosomal TGF-b
levels by pervanadate treatment (Fig. 4E); similarly, the
induction of a-SMA expression by fibroblasts was reduced
by around 50% (Fig. 4F). We conclude that TGF-b is asso-
ciated with betaglycan at the exosome surface and that the
loss of exosomal betaglycan expression can attenuate its
capacity to induce myofibroblastic differentiation.

TGF-bhigh cancer exosomes induce transcription of


mRNAs that is similar but not identical to that induced
by rhTGF-b
An issue that remains unknown is whether the fibroblast
response to exosomal TGF-b differs from that of soluble
TGF-b. To answer this question, we used qPCR to examine a
panel of transcripts, known to be elevated in fibroblasts
following stimulation with rhTGF-b. The effects of exosomes B
from mesothelioma cells (TGF-bhigh), from CaCo2 cells
(TGF-blow), and rhTGF-b were compared, and we also
included the TGF-b-RI inhibitor (SB431542) to evaluate
the contribution of exosomally expressed TGF-b to any
observed effects. Autocrine production of TGF-b is an
important feature of myofibroblastic phenotype (27). Sti-
mulating with rhTGF-b (5 ng/mL) induced around 4-fold
elevated mRNA for TGF-b at 72 hours, and this effect was
abrogated fully in the presence of SB431542, as expected
(Fig. 5A). Mesothelioma-derived exosomes (at a dose of Figure 5. Induction of transcripts by cancer exosomes is similar but not
200 mg/mL of exosomes), broadly equivalent to 1.5 ng/mL identical to that induced by rhTGF-b. A, growth-arrested fibroblasts were
of exosomal TGF-b, resulted in an enhanced induction of treated with rhTGF-b1 (5 ng/mL) or with exosomes from mesothelioma cell
line (Meso) or from a colorectal cancer cell line (CaCo2) each at 200 mg/mL,
TGF-b-mRNA (accounting for TGF-b dose correction). This corresponding to 1.5 ng/mL and less than 20 pg/mL of exosomal TGF-b1,
effect was also abrogated by the inhibitor and was not seen respectively. An inhibitor of TGF-b signaling (SB431542 at 10 mmol/L) was
in fibroblasts stimulated with equal dose of CaCo2 exosomes added, as indicated, and after 72 hours, total RNA was extracted and
(Fig. 5A). Some other altered transcripts, including tran- processed for qPCR. Graphs depict results from 4 independent
experiments, expressing mRNA levels for TGF-b1, cTGF, EGF, and FGF2
scripts of cTGF and EGF, were also induced by exosomes in
(as indicated) relative to untreated fibroblasts (mean  SD, n 4). B,
a manner comparable with rhTGF-b. Any induction seen similarly, culture medium from fibroblasts stimulated with exosomes
here was attenuated by the TGF-b-RI inhibitor in all cases, (200 mg/mL) or rhTGF-b (1.5 ng/mL) was analyzed for FGF2 levels by ELISA
and again CaCo-2derived exosomes had little effect on (mean  SD of triplicates). ***P < 0.0001, 1-way ANOVA followed by
these mRNAs. Examining FGF2 mRNA, however, highlighted Tukey's honest significant difference.
that TGF-bhigh exosomes may well trigger a cellular response
that is not identical to that driven by rhTGF-b. Where
mesothelioma exosomes stimulated a strong induction secreted FGF2 levels, requiring a very high dose of 10 ng/mL
(12- to 16-fold) in FGF2 mRNA, rhTGF-b treatment only to elicit an induction in FGF2 protein. Again, CaCo2 exo-
modestly (2-fold) induced FGF2 mRNA even at a dose that somes did not affect the levels of FGF2. In summary, many
was 3 times higher than the exosomal TGF-b equivalent. classical TGF-b responses are triggered by TGF-bhigh exo-
Again, CaCo2 exosomes had no effect on FGF2 mRNA. This somes in a manner similar to that achieved with soluble
robust stimulation was dependent on exosomal TGF-b, as nonvesicular TGF-b. Importantly, however, there may be
the TGFb-RI inhibitor completely abrogated the response. some differences in the overall response, as exemplified by
We examined by measuring FGF2 in cell culture media by changes in FGF2 expression, in which exosomes exert a
ELISA whether this observation related to the actual FGF2 significantly more potent influence.
protein secreted by these fibroblasts. This revealed superior
induction of FGF2 protein by mesothelioma exosomes com- Discussion
pared with an equivalent dose of rhTGF-b (1.5 ng/mL
shown), which was highly significant (Fig. 5B). Stimulation We present, for the first time, evidence that exosomes from
with 1.5 or 5 ng/mL of rhTGF-b had little or no effect on some cancer cells can drive multiple changes in fibroblasts.

9628 Cancer Res; 70(23) December 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Cancer Exosomes Trigger Fibroblast Differentiation

These alterations are consistent with full differentiation into shows that the exosomal mechanism for TGF-b delivery is
myofibroblasts, a significant and physiologically important not limited to SMAD-related pathways and may well be
process with major implications in health and pathologic equal to the diversity of responses generated using soluble
conditions. We also establish that exosomally expressed TGF-b. HA is a major component of the pericellular matrix
TGF-b1 is responsible for this differentiation. surrounding tumor cells and has previously been linked to
Exosomal expression of TGF-b has been previously tumor progression and dissemination (32, 33). Our data
reported by us (11, 31) and subsequently by others (13), show exosomes as an additional mechanism contributing
but it has always remained possible that this apparent toward a modulated stromal-extracellular matrix, as a con-
exosomal TGF-b may simply be due to soluble TGF-b that sequence of fibroblast differentiation.
co-pellets with exosomes during high-speed ultracentrifuga- Exosomes or soluble TGF-b appeared to drive these
tion. Using ultracentrifugation on continuous sucrose gra- effects with comparable efficacy. We therefore examined
dients, as previously described (26), reveals that TGF-b floats other candidate responses to highlight any distinction in
at characteristic exosomal densities. The fractions contain- the cellular response to exosomal versus soluble TGF-b.
ing this vesicular TGF-b were also the fractions that har- This was essentially not apparent when examining mRNA
bored the capacity to activate TGF-b signaling in an SMAD3 for cTGF or EGF, although there may be a slight advantage
reporter assay. Hence, TGF-b is genuinely expressed by in terms of exosome-mediated autocrine production of
cancer cellderived exosomes and this vesicular form of TGF-b mRNA. However, induction of FGF2 by tumor exo-
TGF-b is biologically active in driving SMAD-dependent somes was an aspect that was clearly distinct between the
signaling. Surprisingly, exosomal TGF-b appears mostly treatment conditions, with significant elevation in mRNA
(98%) in a latent form, yet, when administered to cells, and protein seen with exosomes but not triggered at
is utilized almost fully in terms of signal transduction, giving comparable doses with soluble TGF-b. This shows that
levels of SMAD3 activity around 80% to 90% of that achieved the fibroblast response to tumor-derived exosomes differs
using rhTGF-b. in this regard to that of soluble TGF-b. Considering the role
While the precise mechanism of how latent exosomal of FGF2 in directly promoting tumor growth/survival,
TGF-b is delivered and activated at the cell surface is not migration and metastasis, matrix remodeling, and angio-
known, we now present evidence that implicates betaglycan genesis [reviewed in (34)], our data suggest a physiologic
in a TGF-b-tethering role. This proteoglycan plays important role for exosomal TGF-b in biasing the fibroblast response
functions in binding TGF-b and increases its affinity for toward such tumor-promoting functions. Given the mole-
binding to type II receptor when expressed at the cell surface cular complexity of exosomes, however, we do not know
(30). We demonstrate that exosomes with low levels of how such vesicles mediate this preferential FGF2 respon-
surface betaglycan expression exhibit low TGF-b levels siveness. It requires exosomal TGF-b together with 1 or
and do not induce fibroblast differentiation. In contrast, more additional exosomally delivered constituents; these
some exosomes stained very strongly for surface betaglycan factors are not present when treating with soluble TGF-b
and it was only these exosomes that had sufficient TGF-b alone.
levels for driving fibroblast differentiation. Partial cleavage The importance of this exosomal mechanism in vivo is
of betaglycan from the exosome surface was achieved by currently unknown. Recently, Rab GTPases, Rab27a and
pervanadate treatment, which leads to shedding of betagly- Rab27b, have been shown to play important roles in regulating
can from the cell surface (29). Using this approach, abrogat- exosome secretion (35). These are therefore good candidate
ing exosomal betaglycan expression resulted in decreased target proteins to achieve a global blockade of exosome
exosomal TGF-b levels and a comparable decrease in its secretion by tumor cells in vivo. Furthermore, we propose
capacity to activate a-SMA expression by fibroblasts. The strategies, such as targeting betaglycan, to modify the exo-
data show exosomal betaglycan expression as a relevant some phenotype that could be employed to test the impor-
factor in the tethering of TGF-b to the exosome surface, and tance of exosomal TGF-b specifically in modulating the tumor
unlike soluble betaglycan, exosomes deliver functional TGF- stroma in vivo.
b to recipient cells. While much of the research within the exosome field
In terms of biological function, the impact of cancer continues to focus on immunomodulatory functions (5, 7,
exosomes on primary fibroblasts is profound. Stimulating 8, 11, 31, 36), a small number of reports show that cancer
with exosomes, high in TGF-b, triggered sustained changes exosomes may well be capable of altering the functions of
in the actin cytoskeleton, with a dramatic upregulation of nonimmune cells within the tumor microenvironment. For
structural a-SMA remaining elevated for at least 14 days. example, cancer exosomes can exert a proangiogenic influ-
The nature and kinetics of this classic feature of differen- ence on endothelial cells (37). To date, however, our report is
tiated myofibroblasts were comparable with that triggered the first to highlight the capacity of cancer exosomes to
by rhTGF-b and were absolutely dependent on exosomal profoundly modulate fibroblast phenotype and function and
TGF-b rather than other exosomal components, as it could show exosomal TGF-b as the mechanism for this. The data
not be reproduced by exosomes that were poor in TGF-b. implicate cancer exosomes as mechanistic participators in
However, SMAD-dependent changes of this nature are also the establishment and persistence of cancer-altered stroma
accompanied by SMAD-independent effects, such as the and suggest designing therapies targeted toward attenuating
deposition of a pericellular coat composed of HA (28). This cancer exosomes may prove fruitful.

www.aacrjournals.org Cancer Res; 70(23) December 1, 2010 9629

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Webber et al.

Disclosure of Potential Conflicts of Interest Grant Support

The authors declare that they have no conflict of interest. The study was funded through a pilot study grant from Cancer Research
Wales and thereafter by the June Hancock Mesothelioma Research Fund and
Acknowledgments Velindre Cancer Centre's Charities Committee.

The authors are grateful to Dr. J.P. Mitchell, Miss J. Welton, and Mrs L. Court
for their assistance in cell culture maintenance, exosome purifications and Received 05/13/2010; revised 07/28/2010; accepted 09/03/2010;
analyses. published OnlineFirst 11/23/2010.

References
1. Denzer K, Kleijmeer MJ, Heijnen HF, Stoorvogel W, Geuze HJ. Exo- lymph node metastasis in early-stage invasive colorectal carcinoma.
some: from internal vesicle of the multivesicular body to intercellular Anticancer Res 2005;25:270512.
signaling device. J Cell Sci 2000;113(Pt 19):336574. 19. Piek E, Heldin CH, Ten Dijke P. Specificity, diversity, and regulation in
2. Trajkovic K, Hsu C, Chiantia S, Rajendran L, Wenzel D, Wieland F, TGF-beta superfamily signaling. FASEB J 1999;13:210524.
et al. Ceramide triggers budding of exosome vesicles into multi- 20. Lopez-Casillas F, Cheifetz S, Doody J, Andres JL, Lane WS, Massague
vesicular endosomes. Science 2008;319:12447. J. Structure and expression of the membrane proteoglycan betagly-
3. Simpson RJ, Lim JW, Moritz RL, Mathivanan S. Exosomes: proteomic can, a component of the TGF-beta receptor system. Cell 1991;67:785
insights and diagnostic potential. Expert Rev Proteomics 2009;6:267 95.
83. 21. Brown CB, Boyer AS, Runyan RB, Barnett JV. Requirement of type III
4. Valadi H, Ekstro m K, Bossios A, Sjo strand M, Lee JJ, Lo tvall JO. TGF-beta receptor for endocardial cell transformation in the heart.
Exosome-mediated transfer of mRNAs and microRNAs is a novel Science 1999;283:20802.
mechanism of genetic exchange between cells. Nat Cell Biol 22. Massague J. How cells read TGF-beta signals. Nat Rev Mol Cell Biol
2007;9:6549. 2000;1:16978.
5. Zeelenberg IS, Ostrowski M, Krumeich S, Bobrie A, Jancic C, 23. Zhang YE. Non-Smad pathways in TGF-[beta] signaling. Cell Res
Boissonnas A, et al. Targeting tumor antigens to secreted membrane 2009;19:12839.
vesicles in vivo induces efficient antitumor immune responses. Can- 24. Gu L, Zhu Y, Yang X, Guo Z, Xu W, Tian X. Effect of TGF-beta/Smad
cer Res 2008;68:122835. signaling pathway on lung myofibroblast differentiation. Acta Phar-
6. Pegtel DM, Cosmopoulos K, Thorley-Lawson DA, van Eijndhoven MA, macol Sin 2007;28:38291.
Hopmans ES, Lindenberg JL, et al. Functional delivery of viral miRNAs 25. Welton JL, Khanna S, Giles PJ, Brennan P, Brewis IA, Staffurth J, et al.
via exosomes[Epub ahead of print]. Proc Natl Acad Sci U S A 2010. Proteomics analysis of bladder cancer exosomes. Mol Cell Proteo-
7. Raposo G, Nijman HW, Stoorvogel W, Liejendekker R, Harding CV, mics 2010;9:132438.
Melief CJ, et al. B lymphocytes secrete antigen-presenting vesicles. 26. Thery C, Amigorena S, Raposo G, Clayton A. Isolation and character-
J Exp Med 1996;183: 116172. ization of exosomes from cell culture supernatants and biological
8. Zhang H-G, Liu C, Su K, Yu S, Zhang L, Zhang S, et al. A membrane fluids. Curr Protoc Cell Biol2006;chapter 3:unit 3.22.
form of TNF-{alpha} presented by exosomes delays T cell activation- 27. Webber J, Meran S, Steadman R, Phillips A. Hyaluronan orchestrates
induced cell death. J Immunol 2006;176:738593. transforming growth factor-beta1-dependent maintenance of myofi-
9. Sanderson MP, Keller S, Alonso A, Riedle S, Dempsey PJ, Altevogt broblast phenotype. J Biol Chem 2009;284:908392.
P. Generation of novel, secreted epidermal growth factor receptor 28. Webber J, Jenkins RH, Meran S, Phillips A, Steadman R. Modulation
(EGFR/ErbB1) isoforms via metalloprotease-dependent ectodo- of TGFbeta1-dependent myofibroblast differentiation by hyaluronan.
main shedding and exosome secretion. J Cell Biochem Am J Pathol 2009;175:14860.
2008;103:178397. 29. Velasco-Loyden G, Arribas J, Lopez-Casillas F. The shedding of
10. Claudia S, Carolin S, Walter N. Unconventional secretion of fibroblast betaglycan is regulated by pervanadate and mediated by membrane
growth factor 2 and galectin-1 does not require shedding of plasma type matrix metalloprotease-1. J Biol Chem 2004;279:772133.
membrane-derived vesicles. FEBS Lett 2008;582:13628. 30. Lo pez-Casillas F, Wrana JL, Massague  J. Betaglycan presents ligand
11. Clayton A, Mitchell JP, Court J, Mason MD, Tabi Z. Human tumour- to the TGFb signaling receptor.Cell 1993;73:143544.
derived exosomes selectively impair lymphocyte responses to inter- 31. Clayton A, Mitchell JP, Court J, Linnane S, Mason MD, Tabi Z. Human
leukin-2. Cancer Res 2007;67:745866. tumor-derived exosomes down-modulate NKG2D expression. J
12. Mitchell JP, Court J, Mason MD, Tabi Z, Clayton A. Increased exo- Immunol 2008;180:724958.
some production from tumour cell cultures using the Integra CELLine 32. Li H, Guo L, Li JW, Liu N, Qi R, Liu J. Expression of hyaluronan
Culture System. J Immunol Methods 2008;335:98105. receptors CD44 and RHAMM in stomach cancers: relevance with
13. Wang G-J, Liu Y, Qin A, Shah SV, Deng ZB, Xiang X, et al. Thymus tumor progression. Int J Oncol 2000;17:92732.
exosomes-like particles induce regulatory T cells. J Immunol 33. Misra S, Obeid LM, Hannun YA, Minamisawa S, Berger FG, Markwald
2008;181:52428. RR, et al. Hyaluronan constitutively regulates activation of COX-2-
14. Hinz B, Celetta G, Tomasek JJ, Gabbiani G, Chaponnier C. Alpha- mediated cell survival activity in intestinal epithelial and colon carci-
smooth muscle actin expression upregulates fibroblast contractile noma cells. J Biol Chem 2008;283:1433544.
activity. Mol Biol Cell 2001;12:273041. 34. Kwabi-Addo B, Ozen M, Ittmann M. The role of fibroblast growth
15. Tomasek JJ, Gabbiani G, Hinz B, Chaponnier C, Brown RA. Myofi- factors and their receptors in prostate cancer. Endocr Relat Cancer
broblasts and mechano-regulation of connective tissue remodelling. 2004;11:70924.
Nat Rev Mol Cell Biol 2002;3:34963. 35. Ostrowski M, Carmo NB, Krumeich S, Fanget I, Raposo G, Savina A,
16. Olumi AF, Grossfeld GD, Hayward SW, Carroll PR, Tlsty TD, Cunha et al. Rab27a and Rab27b control different steps of the exosome
GR. Carcinoma-associated fibroblasts direct tumor progression of secretion pathway. Nat Cell Biol 2010;12:1930; sup 113.
initiated human prostatic epithelium. Cancer Res 1999;59:500211. 36. Xiang X, Poliakov A, Liu C, Liu Y, Deng ZB, Wang J, et al. Induction of
17. Orimo A, Gupta PB, Sgroi DC, Arenzana-Seisdedos F, Delaunay T, myeloid-derived suppressor cells by tumor exosomes. Int J Cancer
Naeem R, et al. Stromal fibroblasts present in invasive human breast 2009;124:262133.
carcinomas promote tumor growth and angiogenesis through ele- 37. Folkins C, Shaked Y, Man S, Tang T, Lee CR, Zhu Z, et al. Glioma
vated SDF-1/CXCL12 secretion. Cell 2005;121:33548. tumor stem-like cells promote tumor angiogenesis and vasculogen-
18. Liang PIN, Hong J-W, Ubukata H, Liu G, Katano M, Motohashi G, et al. esis via vascular endothelial growth factor and stromal-derived factor
Myofibroblasts correlate with lymphatic microvessel density and 1. Cancer Res 2009;69:724351.

9630 Cancer Res; 70(23) December 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.
Published OnlineFirst November 23, 2010; DOI: 10.1158/0008-5472.CAN-10-1722

Cancer Exosomes Trigger Fibroblast to Myofibroblast


Differentiation
Jason Webber, Robert Steadman, Malcolm D. Mason, et al.

Cancer Res 2010;70:9621-9630. Published OnlineFirst November 23, 2010.

Updated version Access the most recent version of this article at:
doi:10.1158/0008-5472.CAN-10-1722

Supplementary Access the most recent supplemental material at:


Material http://cancerres.aacrjournals.org/content/suppl/2010/11/23/0008-5472.CAN-10-1722.DC1

Cited articles This article cites 35 articles, 18 of which you can access for free at:
http://cancerres.aacrjournals.org/content/70/23/9621.full#ref-list-1

Citing articles This article has been cited by 31 HighWire-hosted articles. Access the articles at:
http://cancerres.aacrjournals.org/content/70/23/9621.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Subscriptions Department at pubs@aacr.org.

Permissions To request permission to re-use all or part of this article, contact the AACR Publications
Department at permissions@aacr.org.

Downloaded from cancerres.aacrjournals.org on July 22, 2017. 2010 American Association for Cancer
Research.

S-ar putea să vă placă și