Cancer-associated Myofibroblasts Possess Various Factors ... · embedded into Tissue-Tek O.C.T. Compound (Sakura Finetech-nical Co., Ltd., Tokyo, Japan). Then serial 4-m frozen sections

Cancer-associated Myofibroblasts Possess Various Factors toPromote Endometrial Tumor Progression1

Akira Orimo,2 Yasuhiro Tomioka,Yoshihiko Shimizu, Miyuki Sato, Satoko Oigawa,Koichi Kamata, Yasuhisa Nogi, Satoshi Inoue,Masakazu Takahashi, Toshio Hata, andMasami MuramatsuDepartments of Biochemistry [A. O., M. S., Y. N., S. I., M. M.],Obstetrics and Gynecology [Y. T., S. O., T. H.], Pathology [Y. S.,K. K.], and Future Program Research Division [M. M.], SaitamaMedical School, 38 Moro-Hongo, Moroyama-machi, Iruma-gunSaitama 350-0495; Department of Pathology, Sasaki Institute, 2-2Kanda-Surugadai, Chiyoda-ku, Tokyo 101-0062 [M. T.]; andCREST, Japan Science and Technology Corporation, [A. O.] Tokyo170-0013, Japan

ABSTRACTMyofibroblastic invasion associated with malignant epi-

thelial cells of endometrial cancer as well as other cancers isoften found in the interstitium. To assess the myofibroblastic-epithelial interaction, frozen sections from a total of 10 endo-metrial cancers with or without invasive myofibroblasts wereimmunohistochemically examined. Interestingly, the invasivemyofibroblasts adjacent to malignant epithelial cells showedfrequently intensive positive staining of several growth factorssuch as vascular endothelial growth factor (VEGF), insulin-likegrowth factor I, and epidermal growth factor, the cognatereceptors such as Fetal liver kinase-1/Kinase Insert Domain-containing receptor/VEGF receptor-2, fms-like tyrosine ki-nase-1/VEGF receptor-1, and epidermal growth factor recep-tor, several cell cycle regulators such as cyclins and cyclindependent kinases, and estrogen receptor �. Moreover, weindicated that the majority of the myofibroblasts as well ascancer epithelial cells are proliferating because of their positivestaining of proliferating cell nuclear antigen and Ki-67. Fur-thermore, the myofibroblasts were also positive of hypoxia-inducible factor 1 �, which is a marker protein of hypoxia,probably followed by activation of VEGF-Flk-1 and VEGF-fms-like tyrosine kinase-1 signals, which could initiate angio-genesis. These findings suggest directly that the myofibroblasts

might participate in the progression of tumor cells in terms ofcancer cell growth stimulation and also activated initiation ofangiogenesis.

INTRODUCTIONTumorigenesis is a multistep process accompanied by

genetic alterations of precancerous cells and simultaneouslyby building up the microenvironment that promotes transfor-mation (1–3). Consequently, epigenetic contributions fromstroma cells surrounding cancer cells play important roles forformation of progressive neoplasm (4 – 6). The features ofstromal invasion in carcinomas are frequently shown in manycancers including breast, endometrium, prostate, lung, colon,and stomach. It is also known that the stroma is histopatho-logically composed of several supportive coconspirators suchas fibroblasts, inflammatory cells, immune cells, smoothmuscle cells, and endothelial cells (3, 4). Interestingly, fibro-blasts constitute a major stromal compartment, and many ofthem associated with tumor cells are seen significantly dif-ferentiated into so-called “myofibroblasts” (7–9), which arealso found in the process of wound healing, possessing morereactive and plastic properties than normal fibroblasts. Thetrait of myofibroblasts containing both fibroblasts andsmooth muscle cell lineages is defined by positive expressionof stromal cell type markers such as prolyl 4-hydroxylase(10, 11), integrin �2�1, vimentin, and CD34, and by positiveexpression of smooth muscle cell markers such as �-smoothmuscle actin, smooth muscle myosin, calponin, and �1-inte-grin (9). However, the differences between myofibroblastsand fibroblasts involved in the oncogenic ability are quiteunknown. It was postulated previously that several tumori-genic features could be enhanced by coexistence of thefibroblasts and malignant epithelial cells. In brief, several invitro coculture (12–15) and in vivo xenografts studies (16 –19), using both cells with a combination of various cell typesamong normal, initiated, and/or tumorigenic stages, havesuggested that the fibroblasts could accelerate progression ofmalignant epithelial cells because of activated proteolysis(20, 21), increased cancer cell proliferation (14, 15), andattenuation of cancer cell death (18). These consequencesmight be feasible in assessing the cancer cell progressionmechanisms, although the molecular signaling of the fibro-blasts-epithelial interaction is also still quite obscure.

We have tried to perform immunostaining of frozen sec-tions from 10 independent endometrial cancer samples usingdifferent antibodies, which specifically recognize several keyproteins involved in cell growth, cell cycle regulation, sexhormone activation, hypoxia, and angiogenesis. Significant pos-itive staining of various factors in interstitial myofibroblasts hasraised an interesting possibility that the myofibroblasts mightactually support tumor progression.

Received 3/12/01; revised 7/23/01; accepted 7/23/01.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.1 Supported by grants from the Ministry of Education, Science andCulture Japan, Sandoz Foundation, and Maruki Memorial Prize bySaitama Medical School (to A. O.).2 To whom requests for reprints should be addressed, at Department ofBiochemistry, Saitama Medical School, 38 Moro-Hongo, Moroyama-machi, Iruma-gun, Saitama, 350-0495, Japan. Phone: 81-492-76-1490;Fax: 81-492-94-9751. E-mail: [email protected] (to A. O.) [email protected] (to M. M.R).

3097Vol. 7, 3097–3105, October 2001 Clinical Cancer Research

Research. on May 28, 2020. © 2001 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

PATIENTS AND METHODSSelection of Patients. Between November 1998 and De-

cember 1999, 10 patients of all of the endometrial cancer pa-tients (mean age, 60.3 years; range, 38–79 years) to whomuterectomy was performed were enrolled in the study underinformed consent, according to no complication of chronicdiseases. Additionally, any hormonal drugs such as GnRH ag-onist and progestogen had never been prescribed before theoperation.

Antibody and Immunohistochemistry. A small portionof each endometrial cancer mass was cut down and promptlyembedded into Tissue-Tek O.C.T. Compound (Sakura Finetech-nical Co., Ltd., Tokyo, Japan). Then serial 4-�m frozen sectionswere generated and stained with H&E. Immunohistochemistrywas performed sequentially as described previously (22). Cov-erslides were fixed, preblocked in 10% normal goat or rabbitserum, and reacted using each first antibody. The reaction byfirst antibody was performed in 1:100–300 for mAb and in1:400–600 for polyclonal antibody, respectively. Anti-prolyl4-hydroxylase mAb (5B5; DAKO Corp., Carpinteria, CA), anti-integrin �2�1 mAb (P1E6; DAKO Corp.), antivimentin mAb(V9; Santa Cruz Biotechnology, Inc.), and anti-CD34 mAb(QBEND10; Immunotech, Marsielle, France) were used as apositive marker of stromal cell lineage. In addition, anti-�-smooth muscle actin mAb (1A4; Sigma Chemical Co.) was usedas a positive marker of smooth muscle cells, and anticytokeratin8, 18, and 19 mAb (NCL-5D3; PROGEN, Heidelberg, Ger-many) was used as a positive marker of epithelial cells. Addi-tionally, anti-VEGF3 polyclonal antibody (C-1; Santa Cruz Bio-technology, Inc.), anti-IGF-I polyclonal antibody (G-17; SantaCruz Biotechnology, Inc.), anti-EGF mAb (Chemicon Interna-tional Inc., Temecula, CA), anti-EGFR polyclonal antibody(1005)-G (Santa Cruz Biotechnology, Inc.), anti-Fetal liverkinase-1/Kinase Insert Domain-containing receptor/VEGFR-2mAb (A-3; Santa Cruz Biotechnology, Inc.), and anti-fms-liketyosine kinase-1/VEGF receptor-1 polyclonal antibody (C-17;Santa Cruz Biotechnology, Inc.) were used as a marker ofgrowth factors and the cognate receptors. As a marker of cellcycle regulators, anticyclin D1 mAb (R-124; Santa Cruz Bio-technology, Inc.), anticyclin D3 mAb (Upstate BiotechnologyInc., Waltham, MA), anticyclin A polyclonal antibody (A-3;Santa Cruz Biotechnology, Inc.), anti-cdk4 polyclonal antibody(C-22)-G (Santa Cruz Biotechnology, Inc.), and anti-cdk6 poly-clonal antibody (C-21; Santa Cruz Biotechnology, Inc.) werealso used. Moreover, anti-PCNA mAb (PC10; Santa Cruz Bio-technology, Inc.) and anti-Ki-67 polyclonal antibody (C-20;Santa Cruz Biotechnology, Inc.) were used as a marker ofproliferating cells. In addition, anti-ER � polyclonal antibody(PA1–308; Affinity Bioreagents, Inc., Golden, CO), anti-PRmAb (MA1–410; Affinity Bioreagents, Inc.), and anti-EFP

3 The abbreviations used are: VEGF, vascular endothelial growth factor;IGF-I, insulin-like growth factor I; mAb, monoclonal antibody; EGF,epidermal growth factor; TGF, transforming growth factor; EGFR,epidermal growth factor receptor; cdk, cyclin-dependent kinase; ER �,estrogen receptor �; PR, progesterone receptor; EFP, estrogen respon-sive finger protein; PCNA, proliferating cell nuclear antigen; Flt-1,fms-like tyrosine kinase-1; HIF1 �, hypoxia-inducible factor 1 �.

Tab

le1

Sum

mar

ized

resu

ltsof

sem

iqua

ntiti

veim

mun

ohis

toch

emic

alst

aini

ngof

tum

or-a

ssoc

iate

dst

rom

ain

10en

dom

etri

alca

ncer

sa

Sam

ple

Patie

nts

Age

Gra

de

Mes

ench

ymal

cell

mar

kers

Gro

wth

fact

ors

Cel

lcy

cle

regu

lato

rsSt

eroi

dho

rmon

eA

ngio

gene

sis

Prol

yl4-

hydr

oxyl

ase

�SM actin

Inte

grin

�2�

1V

imen

tinV

EG

FIG

F-I

EG

FE

GFR

Cyc

linD

1C

yclin

Acd

k4cd

k6PC

NA

Ki-

67E

R�

EFP

PRH

IF1

�Fl

k-1

Flt-

1

1K

.Y.

58G

1ns

bns

nsns

nsns

nsns

nsns

nsns

nsns

nsns

nsns

nsns

2L

.K.

68G

1ns

nsns

nsns

nsns

nsns

nsns

nsns

nsns

nsns

nsns

ns3

M.K

.68

G1

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�4

N.Y

.38

G2

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�5

A.U

.68

G2

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�6

Y.U

.55

G2

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

7F.

K.

73G

2�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

8K

.Y.

53G

3�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

9Y

.D.

48G

3�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

10S.

T.

79G

3�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�a

End

omet

rial

canc

ersa

mpl

es(1

0)w

ere

clas

sifi

edhi

stol

ogic

ally

into

thre

epa

thol

ogic

algr

ades

(G1–

G3)

,re

sulti

ngin

3ca

ses

with

G1

grad

e,4

case

sw

ithG

2gr

ade,

and

3ca

ses

with

G3

grad

e,re

spec

tivel

y.A

ctua

lly,

8sa

mpl

es(f

rom

3to

10)

show

edsi

gnif

ican

tst

rom

ace

llin

vasi

onin

the

inte

rstit

ium

,w

hile

ther

ew

ere

2sa

mpl

es(1

and

2)w

ithfe

wst

rom

ace

llin

vasi

on.

Res

ults

ofim

mun

ohis

toch

emic

alan

alys

isus

ing

vari

ous

antib

odie

sin

the

stro

ma

of10

endo

met

rial

canc

ers

are

sum

mar

ized

usin

ga

pred

efin

edar

bitr

ary

scal

e.Si

gnif

ican

tly,t

here

sults

of4

sam

ples

(3,6

,7,a

nd8)

acco

mpa

nied

bym

uch

myo

fibr

obla

stic

inva

sion

,in

dica

ted

bysh

adin

gm

arks

,in

clin

edto

poss

ess

mor

epo

sitiv

ece

llpo

pula

tion

with

mor

ein

tens

esi

gnal

.b

ns,n

othi

ngof

stro

mal

inva

sion

;��

�,s

tron

gpo

sitiv

ein

tens

ityan

d/or

posi

tive

stai

ning

in�

50%

ofce

lls;�

�,m

oder

ate

posi

tive

and/

orpo

sitiv

est

aini

ngin

10–5

0%of

cells

;�,w

eak

posi

tive

stai

ning

and/

orpo

sitiv

est

aini

ngin

1–10

%of

cells

.

3098 Cancer-associated Myofibroblasts As Proliferating Cells



polyclonal antibody (23) were used as detection of sex hormonereceptors and downstream target genes. Lastly, anti-HIF-1�mAb (clone H1�67; Novus Biologicals, Inc., Littleton, CO) wasused as a marker of hypoxia. After biotinylated antirabbit IgGantibody, antimouse IgG antibody, or antigoat IgG antibody wasreacted selectively, horseradish peroxidase-streptavidin and sub-strate solution were added. Then these sections were slightlystained by hematoxylin.

RESULTSImmunohistochemical Identification of Myofibroblasts

in Endometrial Cancer. Serial frozen sections from 10 en-dometrial cancer samples were stained by H&E to classify themhistologically into three pathological grades (G1–G3). We foundthat there were 3 cases with G1 grade, 4 cases with G2 grade,and 3 cases with G3 grade, respectively. Actually, 8 samples(from 3 to 10 in Table 1) showed significant stroma cell inva-sion in the interstitium, whereas 2 samples (1 and 2) showedlittle stroma cell invasion (data not shown). Herein, a view ofthe sample 6 (Y.U.) was representatively shown. Morphologi-cally there were a number of fibroblasts with spindle-like ap-pearance in the stroma invasion on the specimen stained byH&E (Fig. 1, panel 1). To confirm them immunohistochemi-cally, immunostaining was done using positive markers ofstroma cell lineage including fibroblasts such as prolyl 4-hydroxylase (Fig. 1, panel 2), integrin �2�1 (Fig. 1, panel 3),

vimentin (Fig.1, panel 4), and CD34 (data not shown), and usinga strict positive marker of smooth muscle cell lineage such as�-smooth muscle actin (Fig. 1, panel 5), which never reacts withfibroblasts. In the markers of stroma cell lineage, the anti-prolyl4-hydroxylase antibody was reported as more specific to thefibroblasts than others showing no cross-reactivity, at least withother mesenchymal cell lineages such as lymphocytes, mono-cytes, dendritic cells, granulocytes (10, 11), and smooth musclecells (data not shown). It was found that a number of cells in theinterstitium were positive with respect to each marker on serialsections from the sample 6 (Fig. 1, panels 2–5), suggesting thatthe majority of invasive stroma appears to be composed of agreat many myofibroblasts, because positive signals of bothmarkers of fibroblastic and smooth muscle cell lineage werecolocalized. In contrast, anticytokeratin 8, 18, and 19 antibody,strictly specific to the epithelial cells, stained only the epithelialcells strongly but no stroma in the interstitium as was expected(Fig. 1, panel 6).

Immunohistochemical Staining of Several GrowthFactors and Cell Cycle Regulators in Invasive Carcinoma-associated Myofibroblasts. First, we hypothesized thatgrowth factors would be more feasible as the factor that medi-ates the myofibroblastic-epithelial interaction. A number ofsections from 10 endometrial cancer samples were used forimmunostaining. In Fig. 2A, the representative views of invasivemyofibroblasts in interstitial stroma (Fig. 2A, panels 2, 4, 6, and

Fig. 1 Immunohistochemical identification of myofibroblasts in endometrial cancer. A representative view of a sample (6 shown in Table 1) of 10endometrial cancer samples is shown. A frozen section was stained by H&E (panel 1) and then serial sections were immunostained using followingantibodies: panel 2, anti-prolyl 4-hydroxylase mAb; panel 3, anti-integrin �2�1 mAb; panel 4, antivimentin mAb; panel 5, anti-�-smooth muscle actinmAb; panel 6, anticytokeratin 8, 18, and 19 mAb. Respective positive signals are shown as brown, and incidentally nuclear staining of whole cellsis shown as blue or violet by hematoxylin. The colocalized intense staining of positive markers of stromal cells including fibroblasts and smoothmuscle cells shows that almost all interstitial cells are occupied with myofibroblasts. EP, epithelial cells; CAF, carcinoma-associated myofibroblast;prolyl 4-h, prolyl 4-hydroxylase; �SMactin, �-smooth muscle actin. Scale bar, 100 �m.

3099Clinical Cancer Research



Fig. 2 Immunohistochemical staining of several growth factors and cell cycle regulators in invasive carcinoma-associated myofibroblasts. A,representative views of 2 samples (1 and 6, shown in Table 1) of 10 endometrial cancer samples are shown. Panels 1, 3, 5, and 7 are from sample1, and panels 2, 4, 6, 8, 9, and 10 are from sample 6. The benign regions (panels 9 and 10) from sample 6 are also shown. Each serial frozen sectionwas immunostained using the following antibodies: panels 1, 2, and 9, anti-VEGF polyclonal antibody; panels 3, 4, and 10, anti-IGF-I polyclonalantibody; panels 5 and 6, anti-EGF mAb; and panels 7 and 8, anti-EGFR polyclonal antibody. EP, epithelial cells; CAF, carcinoma-associatedfibroblast; BE, benign region. Scale bar, 100 �m. B, representative views of two samples (1 and 6) are shown as in (A). Similarly, here are viewsfrom sample 1 (panels 1, 3, 5, 7, and 9) and those from sample 6 (panels 2, 4, 6, 8, and 10). Panels 1 and 2, immunostaining with anticyclin D1 mAb;panels 3 and 4, anticyclin D3 mAb; panels 5 and 6, anticyclin A polyclonal antibody; panels 7 and 8, anti-cdk4 polyclonal antibody; and panels 9and 10, anti-cdk6 polyclonal antibody. Positive signals are shown in brown, and incidental nuclear staining of whole cells is shown as blue or violetby hematoxylin. Abbreviations and scale bars are the same as in A.




8), the vicinal tumor epithelial cells (Fig. 2A, panels 2, 4, 6, and8), and the portion of benign epithelial cells (Fig. 2A, panels 9and 10) were presented from the serial sections of the sample 6that had also been shown in Fig. 1, whereas malignant epithelialcells with few invasive myofibroblasts from sample 1 were alsoshown in Fig. 2A, panels 1, 3, 5, and 7.

Indeed, the myofibroblasts (Fig. 2A, panels 2, 4, and 6)showed immunohistochemically significant positive staining ofVEGF, IGF-I, and EGF, whereas vicinal malignant epithelial

cells (Fig. 2A, panels 2, 4, and 6) were quite less stained. Incontrast, the malignant epithelial cells without invasive myofi-broblasts (Fig. 2A, panels 1, 3, and 5) showed strong staining ofVEGF, IGF-I, and EGF, and the stain of VEGF (Fig. 2A, panel9) and IGF-I (Fig. 2A, panel 10) in the benign epithelium wasalso shown strongly, indicating the down-regulation of thesefactors in malignant epithelial cells adjacent to invasive myofi-broblasts, although there were differences among respectivesamples (data not shown). In addition, EGFR staining was

Fig. 2. Continued




strongly positive in invasive myofibroblasts (Fig. 2A, panel 8),vicinal tumor epithelial cells (Fig. 2A, panel 8), and tumorepithelial cells with few invasive myofibroblasts (Fig. 2A, panel 7).

Next, we attempted immunostaining using various antibod-ies specific to cell cycle regulators such as cyclin D1, D3, cyclinA, cdk4, and cdk6. These regulators were significantly positivein invasive myofibroblasts even more than vicinal tumor epi-thelial cells (Fig. 2B, panels 2, 4, 6, 8, and 10), whereas tumorepithelial cells with few invasive myofibroblasts also showedstrong positive signals (Fig. 2B, panels 1, 3, 5, and 7) of all ofthese factors. However, cdk6 signals appeared predominant instroma (Fig. 2B, panel 9), indicating that other interstitial cellsalso produce cdk6 in this sample. Then we performed immuno-staining using each antibody that recognizes PCNA and Ki-67 toassess whether these cells are proliferating. There were positivePCNA and Ki-67 signals in both tumor epithelial cells andinvasive myofibroblasts (Fig. 3, panels 2 and 4) and also intumor epithelial cells with few invasive myofibroblasts (Fig. 3,panels 1 and 3). These results show that both the myofibroblastsand tumor epithelial cells are actually proliferating.

Immunohistochemical Analysis of Sex Hormone Recep-tors and a Few Factors Implicated in Hypoxia and Angio-genesis in Cancer-associated Myofibroblasts. Next, immu-nostaining of ER � was examined because of the feature ofestrogen-dependent growth of endometrial cancer cells. Theexpression of EFP (22, 23) and PR (24), which are downstreamtarget genes of ER � in the uterus, was also examined.

Invasive myofibroblasts showed significantly overlappedstaining of ER � (Fig. 4, panel 2), EFP (Fig. 4, panel 4), and PR(Fig. 4, panel 6), whereas respective signals of vicinal tumorepithelial cells were moderately positive in ER � (Fig. 4, panel2) and PR (Fig. 4, panel 6) and almost negative in EFP (Fig. 4,panel 4), suggesting the possible regulation by sex hormoneseven in the invasive myofibroblasts. In contrast, tumor epithelialcells with few invasive myofibroblasts also showed strong stain-

ing in ER � (Fig. 4, panel 1), EFP (Fig. 4, panel 3), and PR (Fig.4, panel 5). Lastly, we have attempted immunostaining of HIF1�, which is a marker protein of hypoxia and also that of Flk-1and Flt-1, which are markers for initiation of angiogenesis as thereceptors of VEGF. In Fig. 5, the immunoreactivity of HIF1 �(Fig. 5, panel 1), Flk-1 (Fig. 5, panel 2), and Flt-1 (Fig. 5, panel3) was significantly positive in invasive myofibroblasts,whereas vicinal tumor epithelial cells were also moderatelypositive in HIF1 � and Flk-1 but negligible in Flt-1. Thesignificance of positive staining in invasive myofibroblasts isdiscussed later.

DISCUSSIONThe supportive role of stroma surrounding cancer cells,

which is composed of majority of myofibroblasts, to tumor isoriginally demonstrated (4, 5). Interestingly, new reports areshowing currently that myofibroblasts have oncogenic ability topromote not only tumor initiation but also the progression in theprostate cancer (19, 25). However, the molecular mechanism ofa sort of tumorigenic ability in the myofibroblasts is still un-known.

Actually, it was reported previously that 83% of the wholecuretteting samples from 204 endometrial cancers with G1–G3grades showed significant stromal invasion adjacent to cancerepithelial cells (26). Therefore, we have evaluated the immuno-reactivity of several factors regarding cancer promotion instroma surrounding the endometrial cancer cells.

This time, we have demonstrated that almost all of the cellsof the invasive stroma in 4 (samples 3, 6, 7, and 8, shown inTable 1) of 8 endometrial cancer samples (from 3 to 10) arelikely to be real myofibroblasts according to highly positivesignals of prolyl 4-hydroxylase and �-smooth muscle actin,whereas the interstitium of the remaining 4 samples (4, 5, 9, and10) probably consists of smooth muscle cells and other mesen-

Fig. 3 Immunohistochemicalrepresentation of proliferatingcells in carcinoma-associatedmyofibroblasts. Photo fromtwo samples (1 and 6) isshown. There are views (panels1 and 3) from sample 1 andthose (panels 2 and 4) fromsample 6. Panels 1 and 2, im-munostaining using anti-PCNAmAb; panels 3 and 4, usinganti-Ki-67 polyclonal antibody.Positive signals are indicated asbrown, and incidentally nuclearstaining of whole cells isshown as blue or violet by he-matoxylin. EP, epithelial cells;CAF, carcinoma-associated fi-broblast. Scale bar, 100 �m.




chymal cells containing fewer myofibroblasts, because the cellpopulation is less positive of prolyl 4-hydroxylase comparedwith more cell population positive of �-smooth muscle actin,integrin �2�1 and vimentin (Table 1). Consistently, it wasreported previously that myofibroblasts were composed of�80% of breast cancer stroma cells (27). In future, doubleimmunostaining using anti-prolyl 4-hydroxylase antibody andanti-�-smooth muscle actin antibody may be required to actu-ally evaluate the correct ratio of myofibroblasts in the wholestroma.

The role of invasive myofibroblasts during tumorigenesishas been examined by several immunohistochemical studies andcoculture systems with malignant cells showing elevated pro-duction of extracellular matrix proteins such as tenascin (28)and fibronectin (29) and altered secretion of metalloproteinases

such as stromelysin-3 (20, 21)and MMP-2 (30), which areproteolytic enzymes. These findings suggested that the myofi-broblasts might support cancer-invasive properties through fa-cilitation of the initial attachment and also through infiltrativemovement of cancer epithelial cells. In addition, regarding cellproliferation, previous reports have shown altered expression ofmany growth factors such as hepatic growth factor (15), IGF-Iand II (12, 21), TGF-�1 (31), connective tissue growth factor(31), and platelet-derived growth factors receptor (32) in myo-fibroblasts cocultured with malignant cells, supporting the ideathat these growth factors might help accelerate cell growth ofmalignant epithelial cells.

In this study, we have succeeded to immunohistochemi-cally detect predominant positive signals in invasive myofibro-blasts associated with malignant epithelial cells in actual endo-

Fig. 4 Immunohistochemical analysis of sex hormone receptors in carcinoma-associated myofibroblasts. The photo from two samples (1 and 6) isshown. There are views (panels 1, 3, and 5) from sample 1 and those (panels 2, 4, and 6) from sample 6. Panels 1 and 2, immunostaining using anti-ER� polyclonal antibody; panels 3 and 4, anti-EFP polyclonal antibody; and panels 5 and 6, anti-PR mAb. Positive signals are indicated as brown, andincidentally nuclear staining of whole cells is shown as blue or violet by hematoxylin. EP, epithelial cells; CAF, carcinoma-associated fibroblast. Scalebar, 100 �m.

Fig. 5 Immunohistochemical analysis of a few factors implicated in hypoxia and angiogenesis in carcinoma-associated myofibroblasts. The photofrom sample 6 is shown. Panel 1, immunostaining using anti-HIF-1� mAb; panel 2, anti-Flk-1/KDR/VEGFR-2 mAb; and panel 3, anti-Flt-1/VEGFreceptor-1 polyclonal antibody. Positive signals are indicated as brown, and incidentally nuclear staining of whole cells are shown as blue or violetby hematoxylin. EP, epithelial cells; CAF, carcinoma-associated fibroblast. Scale bar, 100 �m.




metrial cancer samples using respective antibodies of severalfactors involved in cell growth, cell cycle regulation, sex hor-mone activation, hypoxia, and angiogenesis. Interestingly, theinvasive myofibroblasts showed significant positive signals ofgrowth factors, such as VEGF, IGF-I, and EGF, and those of thecognate receptors, such as Flk-1 and Flt-1, which are receptorsof VEGF, and EGFR, which was positively detected also in thevicinal epithelial cells. Therefore, it is likely that the invasivemyofibroblasts take a share in stimulation of malignant epithe-lial cells through growth factor receptor signals in a paracrinemanner and also stimulate themselves in an autocrine manner.Moreover, the invasive myofibroblasts were actually proliferat-ing as exemplified by the significant positivity of cyclin D1, D3,cyclin A, cdk4, and cdk6, and also of PCNA and Ki-67.

From these observations, we strongly suggest the possibil-ity that these growth factors would be important molecules,which mediate the myofibroblastic-epithelial interaction in car-cinogenesis. Summarized results in Table 1 with predefinedarbitrary scale show that the high expression of these factors inthe invasive myofibroblasts is a common feature among mostendometrial cancer samples (8 of 10). Notably, the more thesefactors are expressed, the more the cells appeared to infiltrate asmyofibroblasts (see samples 3, 6, 7, and 8).

The present study is the first to actually demonstrate pre-dominant existence of various growth factors, cell cycle regu-lators, and cell proliferating markers in the invasive myofibro-blasts of endometrial cancer. Incidentally, it was reportedpreviously that most myofibroblasts in nodular palmar fibroma-tosis were positive for Ki-67 and PCNA (33). Also reported areseveral data indicating expression and existence of a few growthfactors such as IGF-II (21), TGF-�1 (31), and ConnectiveTissue Growth Factor (31) in invasive myofibroblasts of breastcancer, although their positive signals are not so dramatic.

On the other hand, nearly one-half of endometrial carci-noma has the feature of estrogen-dependent growth and, hence,estrogen is well known as one of the risk factors that stimulatesthe endometrial cancer progression (34, 35). As shown in Fig. 4,the positive staining of ER � colocalized with EFP (22, 23) andPR (24), which are known as downstream genes of ER �, in themyofibroblasts is remarkable and suggests that ER � might beligand-independently activated even in some specimens frompostmenopausal women.

Next, the existence of a few angiogenic factors in myofi-broblasts was examined, referring to previous reports, whichhad indicated the possible role of myofibroblasts implicated inangiogenesis (36, 37). It is currently believed that enhancedHIF1 � expression correlates with tumor growth and angiogen-esis (38, 39), of which the mechanism is likely to be throughup-regulation of VEGF expression (40). Then the significantstaining of HIF1 � and VEGF in the invasive myofibroblasts inthis study lets us imagine the possibility of increased initiationof angiogenesis through activation of VEGF-Flk-1 and VEGF-Flt-1 signals in the interstitium.

Finally, it is recapitulated that our immunohistochemicaldata supported the cooperative role of the myofibroblasts in-volved in endometrial cancer cell progression because of accel-erated cell proliferation and initiation of angiogenesis in vivo.Significantly, a previous clinical report about endometrial can-cer has indicated the importance of stromal invasion in predict-

ing prognosis, because many atypical hyperplasias with signif-icant stromal invasion were more frequently accompanied bymore invasive residual carcinomas compared with the one withless stromal invasion (26). Hence, the evaluation of the patho-logical grades of endometrial cancers might be done seriouslybased on the insight into the existence of stromal invasion aswell as malignant epithelial cells. In future, correlation amongthe extent of stromal invasion, existence of several growthfactors, and the prognosis should be more positively evaluated,and it would also be worthy of examining carefully whetherdrug intervention into the paracrine factors and other signals ofthe stromal-epithelial interaction is applicable to therapeuticapproach of cancer progression.

ACKNOWLEDGMENTSWe thank members of the Masami Muramatsu laboratory and the

Tosiho Hata laboratory for valuable advice.

REFERENCES1. Murray, M. J., Cunningham, J. M., Parada, L. F., Dautry, F., Leb-owitz, P., and Weinberg, R. A. The HL-60 transforming sequence: a rasoncogene coexisting with altered myc genes in hematopoietic tumors.Cell, 33: 749–757, 1983.

2. Fearon, E. R., and Vogelstein, B. A genetic model for colorectaltumorigenesis. Cell, 61: 759–767, 1990.

3. Hanahan, D., and Weinberg, R. A. The hallmarks of cancer. Cell,100: 57–70, 2000.

4. Ronnov-Jessen, L., Petersen, O. W., and Bissell, M. J. Cellularchanges involved in conversion of normal to malignant breast: impor-tance of the stromal reaction. Physiol Rev., 76: 69–125, 1996.

5. Kinzler, K. W., and Vogelstein, B. Landscaping the cancer terrain.Science (Wash. DC), 280: 1036–1037, 1998.

6. Skobe, M., and Fusenig, N. E. Tumorigenic conversion of immortalhuman keratinocytes through stromal cell activation. Proc. Natl. Acad.Sci. USA, 95: 1050–1055, 1998.

7. Desmouliere, A., Geinoz, A., Gabbiani, F., and Gabbiani, G. Trans-forming growth factor-� 1 induces �-smooth muscle actin expression ingranulation tissue myofibroblasts and in quiescent and growing culturedfibroblasts. J. Cell Biol., 122: 103–111, 1993.

8. Grinnell, F. Fibroblasts, myofibroblasts, and wound contraction.J. Cell Biol., 124: 401–404, 1994.

9. Ronnov-Jessen, L., Petersen, O. W., Koteliansky, V. E., and Bissell,M. J. The origin of the myofibroblasts in breast cancer. Recapitulationof tumor environment in culture unravels diversity and implicates con-verted fibroblasts and recruited smooth muscle cells. J. Clin. Investig.,95: 859–873, 1995.

10. Konttinen, Y. T., Nykanen, P., Nordstrom, D., Saari, H., Sandelin,J., Santavirta, S., and Kouri, T. DNA synthesis in prolyl 4-hydroxylasepositive fibroblasts in situ in synovial tissue. An autoradiography-immunoperoxidase double labeling study. J. Rheumatol., 16: 339–345,1989.11. Janin, A., Konttinen, Y. T., Gronblad, M., Karhunen, P., Gosset, D.,and Malmstrom, M. Fibroblast markers in labial salivary gland biopsiesin progressive systemic sclerosis. Clin. Exp. Rheumatol., 8: 237–242,1990.12. Cullen, K. J., Smith, H. S., Hill, S., Rosen, N., and Lippman, M. E.Growth factor messenger RNA expression by human breast fibroblastsfrom benign and malignant lesions. Cancer Res., 51: 4978–4985, 1991.13. Singer, C., Rasmussen, A., Smith, H. S., Lippman, M. E., Lynch,H. T., and Cullen, K. J. Malignant breast epithelium selects for insulin-like growth factor II expression in breast stroma: evidence for paracrinefunction. Cancer Res., 55: 2448–2454, 1995.




14. Atula, S., Grenman, R., and Syrjanen, S. Fibroblasts can modulatethe phenotype of malignant epithelial cells in vitro. Exp. Cell. Res., 235:180–187, 1997.

15. Nakamura, T., Matsumoto, K., Kiritoshi, A., Tano, Y., and Naka-mura, T. Induction of hepatocyte growth factor in fibroblasts by tumor-derived factors affects invasive growth of tumor cells: in vitro analysisof tumor-stromal interactions. Cancer Res., 57: 3305–3313, 1997.

16. Chung, L. W., Chang, S. M., Bell, C., Zhau, H. E., Ro, J. Y., andvon Eschenbach, A. C. Co-inoculation of tumorigenic rat prostate mes-enchymal cells with non-tumorigenic epithelial cells results in the de-velopment of carcinosarcoma in syngeneic and athymic animals. Int. J.Cancer, 43: 1179–1187, 1989.

17. Camps, J. L., Chang, S. M., Hsu, T. C., Freeman, M. R., Hong, S. J.,Zhau, H. E., von Eschenbach, A. C., and Chung, L. W. Fibroblast-mediated acceleration of human epithelial tumor growth in vivo. Proc.Natl. Acad. Sci. USA, 87: 75–79, 1990.

18. Olumi, A. F., Dazin, P., and Tlsty, T. D. A novel coculture tech-nique demonstrates that normal human prostatic fibroblasts contribute totumor formation of LNCaP cells by retarding cell death. Cancer Res.,58: 4525–4530, 1998.

19. Olumi, A. F., Grossfeld, G. D., Hayward, S. W., Carroll, P. R.,Tlsty, T. D., and Cunha, G. R. Carcinoma-associated fibroblasts directtumor progression of initiated human prostatic epithelium. Cancer Res.,59: 5002–5011, 1999.

20. Basset, P., Bellocq, J. P., Wolf, C., Stoll, I., Hutin, P., Limacher,J. M., Podhajcer, O. L., Chenard, M. P., Rio, M. C., and Chambon, P.A novel metalloproteinase gene specifically expressed in stromal cellsof breast carcinomas. Nature (Lond.), 348: 699–704, 1990.

21. Singer, C. F., Rasmussen, A., Lippman, M. E., and Cullen, K. J.Coexpression of stromelysin-3 and insulin-like growth factor II intumors of ectodermal, mesodermal, and endodermal origin: indicator ofa fetal cell phenotype. J. Clin. Endocrinol. Metab., 82: 1917–1922,1997.

22. Orimo, A., Inoue, S., Minowa, O., Tominaga, N., Tomioka, Y.,Sato, M., Kuno, K., Hiroi, H., Shimizu, Y., Suzuki, M., Noda, T., andMuramatsu, M. Underdeveloped uterus and reduced estrogen respon-siveness in mice with disruption of the estrogen responsive fingerprotein (efp) gene, which is a direct target of estrogen receptor � (ER �).Proc. Natl. Acad. Sci. USA, 96: 12027–12032, 1999.

23. Inoue, S., Orimo, A., Hosoi, T., Kondo, S., Toyoshima, H., Kondo,T., Ikegami, A., Ouchi, Y., Orimo, H., and Muramatsu, M. Genomicbinding-site cloning reveals a novel estrogen responsive gene that en-codes a RING finger protein. Proc. Natl. Acad. Sci. USA, 90: 11117–11121, 1993.

24. Kraus, W. L., and Katzenellenbogen, B. S. Regulation of proges-terone receptor gene expression and growth in the rat uterus: modulationof estrogen actions by progesterone and sex steroid hormone antago-nists. Endocrinology, 132: 2371–2379, 1993.

25. Tlsty, T. D., and Hein, P. W. Know thy neighbor: stromal cells cancontribute oncogenic signals. Curr. Opin. Genet. Dev., 11: 54–59, 2001.

26. Kurman, R. J., and Norris, H. J. Evaluation of criteria for distin-guishing atypical endometrial hyperplasia from well-differentiated car-cinoma. Cancer (Phila.)., 49: 2547–2559, 1982.

27. Sappino, A. P., Skalli, O., Jackson, B., Schurch, W., and Gabbiani,G. Smooth-muscle differentiation in stromal cells of malignant andnon-malignant breast tissues. Int. J. Cancer, 41: 707–712, 1988.28. Chiquet-Ehrismann, R., Kalla, P., and Pearson, C. A. Participationof tenascin and transforming growth factor-� in reciprocal epithelial-mesenchymal interactions of MCF7 cells and fibroblasts. Cancer Res.,49: 4322–4325, 1989.29. Serini, G., Bochaton-Piallat, M. L., Ropraz, P., Geinoz, A., Borsi,L., Zardi, L., and Gabbiani, G. The fibronectin domain ED-A is crucialfor myofibroblastic phenotype induction by transforming growth factor-�1. J. Cell Biol., 142: 873–881, 1998.30. Poulsom, R., Pignatelli, M., Stetler-Stevenson, W. G., Liotta, L. A.,Wright, P. A., Jeffery, R. E., Longcroft, J. M., Rogers, L., and Stamp,G. W. Stromal expression of 72 kda type IV collagenase (MMP-2) andTIMP-2 mRNAs in colorectal neoplasia. Am. J. Pathol., 141: 389–396,1992.31. Frazier, K. S., and Grotendorst, G. R. Expression of connectivetissue growth factor mRNA in the fibrous stroma of mammary tumors.Int. J. Biochem. Cell Biol., 29: 153–161: 1997.32. Ponten, F., Ren, Z., Nister, M., Westermark, B., and Ponten, J.Epithelial-stromal interactions in basal cell cancer: the PDGF system.J. Investig. Dermatol., 102: 304–309, 1994.33. Berndt, A., Kosmehl, H., Mandel, U., Gabler, U., Luo, X., Celeda,D., Zardi, L., and Katenkamp, D. TGF � and bFGF synthesis andlocalization in Dupuytren’s disease (nodular palmar fibromatosis) rela-tive to cellular activity, myofibroblast phenotype and oncofetal variantsof fibronectin. Histochem J., 27: 1014–1020, 1995.34. Herbst, A. L., Anderson, S., Hubby, M. M., Haenszel, W. M.,Kaufman, R. H., and Noller, K. L. Risk factors for the development ofdiethylstilbestrol-associated clear cell adenocarcinoma: a case-controlstudy. Am. J. Obstet. Gynecol., 154: 814–822, 1986.35. Weiderpass, E., Baron, J. A., Adami, H. O., Magnusson, C.,Lindgren, A., Bergstrom, R., Correia, N., and Persson, I. Low-potencyoestrogen and risk of endometrial cancer: a case-control study. Lancet,353: 1824–1828, 1999.36. Seemayer, T. A., Lagace, R., Schurch, W., and Tremblay, G.Myofibroblasts in the stroma of invasive and metastatic carcinoma: apossible host response to neoplasia. Am. J. Surg. Pathol., 3: 525–533,1979.37. Nicosia, R. F., and Tuszynski, G. P. Matrix-bound thrombospondinpromotes angiogenesis in vitro. J. Cell Biol., 124: 183–193, 1994.38. Jiang, B. H., Agani, F., Passaniti, A., and Semenza, G. L. V-SRCinduces expression of hypoxia-inducible factor 1 (HIF-1) and transcrip-tion of genes encoding vascular endothelial growth factor and enolase1:involvement of HIF-1 in tumor progression. Cancer Res., 57: 5328–5335, 1997.39. Maxwell, P. H., Dachs, G. U., Gleadle, J. M., Nicholls, L. G.,Harris, A. L., Stratford, I. J., Hankinson, O., Pugh, C. W., and Ratcliffe,P. J. Hypoxia-inducible factor-1 modulates gene expression in solidtumors and influences both angiogenesis and tumor growth. Proc. Natl.Acad. Sci. USA, 94: 8104–8109, 1997.40. Liu, Y., Cox, S. R., Morita, T., and Kourembanas, S. Hypoxiaregulates vascular endothelial growth factor gene expression in endo-thelial cells. Identification of a 5� enhancer. Circ. Res., 77: 638–643,1995.




2001;7:3097-3105. Clin Cancer Res Akira Orimo, Yasuhiro Tomioka, Yoshihiko Shimizu, et al. Promote Endometrial Tumor ProgressionCancer-associated Myofibroblasts Possess Various Factors to

Updated version

http://clincancerres.aacrjournals.org/content/7/10/3097

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/7/10/3097.full#ref-list-1

This article cites 39 articles, 18 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/7/10/3097.full#related-urls

This article has been cited by 13 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/7/10/3097To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/content/7/10/3097.full#ref-list-1

http://clincancerres.aacrjournals.org/content/7/10/3097.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Cancer-associated Myofibroblasts Possess Various Factors ... · embedded into Tissue-Tek O.C.T. Compound (Sakura Finetech-nical Co., Ltd., Tokyo, Japan). Then serial 4-m frozen sections