omental adipose tissue derived stromal cells promote … · tribution of adipose tissue in the...

Cancer Therapy: Preclinical

Omental Adipose Tissue–Derived Stromal Cells PromoteVascularization and Growth of Endometrial Tumors

Ann H. Klopp1, Yan Zhang5, Travis Solley1, Felipe Amaya-Manzanares5, Frank Marini2, Michael Andreeff2,Bisrat Debeb1, Wendy Woodward1, Rosemarie Schmandt3, Russell Broaddus4, Karen Lu3, andMikhail G. Kolonin5

AbstractPurpose: Adipose tissue contains a population of tumor-tropic mesenchymal progenitors, termed adipose

stromal cells (ASC), which engraft in neighboring tumors to form supportive tumor stroma.We hypothesized

that intra-abdominal visceral adipose tissue may contain a uniquely tumor-promoting population of ASC to

account for the relationship between excess visceral adipose tissue and mortality of intra-abdominal cancers.

Experimental Design: To investigate this, we isolated and characterized ASC from intra-abdominal

omental adipose tissue (O-ASC) and characterized their effects on endometrial cancer progression as

comparedwith subcutaneous adipose-derivedmesenchymal stromal cells (SC-ASC), bonemarrow–derived

mesenchymal stromal cells (BM-MSC), and lung fibroblasts. To model chronic recruitment of ASC by

tumors, cells were injected metronomically into mice bearing Hec1a xenografts.

Results: O-ASC expressed cell surface markers characteristic of BM-MSC and differentiated into mes-

enchymal lineages. Coculture with O-ASC increased endometrial cancer cell proliferation in vitro. Tumor

tropismofO-ASC and SC-ASC for humanHec1a endometrial tumor xenografts was comparable, butO-ASC

more potently promoted tumor growth. Compared with tumors in SC-ASC–injected mice, tumors in

O-ASC–injected mice contained higher numbers of large tortuous desmin-positive blood vessels, which

correlated with decreased central tumor necrosis and increased tumor cell proliferation. O-ASC exhibited

enhanced motility as compared with SC-ASC in response to Hec1a-secreted factors.

Conclusions: Visceral adipose tissue contains a population of multipotent MSCs that promote endo-

metrial tumor growthmore potently thanMSCs fromsubcutaneous adipose tissue.Wepropose thatO-ASCs

recruited to tumors express specific factors that enhance tumor vascularization, promoting survival and

proliferation of tumor cells. Clin Cancer Res; 18(3); 771–82. �2011 AACR.

Introduction

Endometrial cancer is the most common gynecologicmalignancy in the United States and is strongly linked withobesity (1, 2). Women with a body mass index (BMI,calculated as kg/m2) greater than 40 have a greater than

6-fold relative risk of death from uterine cancer than leanwomen (2, 3). Furthermore, estimates suggest that 39% ofcases of endometrial cancer can be attributed to obesity (4).The relationship between obesity and endometrial cancerhas generally been attributed to estrogen and other para-crine factors secreted by adipose tissue (5).

However, a more direct relationship between adiposetissue and malignant tumors is suggested by recent studiesshowing that adipose tissue contains a population of mes-enchymal progenitor cells that migrate into tumors and canfacilitate tumor progression (6). These adipose stromal cells(ASC) have many features of bone marrow–derived mes-enchymal stromal cells (BM-MSC), including cell surfacemarker expression, plastic adherence, and the capacity todifferentiate into cells of mesenchymal lineage: osteoblasts,chondrocytes, and adipocytes (7, 8). Accumulating evi-dence indicates that cancer-associated fibroblasts (CAF),which provide a trophic tumor microenvironment, are, atleast to some extent, derived fromMSCs (9–13).MSCs havebeen reported to impact tumor progression, increasingtumor initiation and promoting metastasis in somemodels(14–19). This evidence suggests that the strong linkbetween

Authors' Affiliations: Departments of 1Radiation Oncology, 2Leukemia,3Gynecologic Oncology, and 4Pathology, The University of Texas MDAnderson Cancer Center; and 5The Brown Foundation Institute of Molec-ular Medicine, The University of Texas Health Science Center at Houston,Houston, Texas

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Current address for F. Marini: Wake Forest Institute of RegenerativeMedicine, Winston Salem, North Carolina.

Corresponding Author: Mikhail G. Kolonin, Institute of Molecular Medi-cine, Center for Stem Cell and Regenerative Medicine, University of TexasHealth Science Center at Houston, 1825 Pressler St., Room 630-G,Houston, TX 77030. Phone: 713-500-3146; Fax: 713-500-2424; E-mail:[email protected]

doi: 10.1158/1078-0432.CCR-11-1916

�2011 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 771

on January 29, 2020. © 2012 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst December 13, 2011; DOI: 10.1158/1078-0432.CCR-11-1916

http://clincancerres.aacrjournals.org/

obesity and endometrial cancermaybe attributed, at least inpart, to increased availability of ASCs which are recruited toform a supportive tumor microenvironment.

Excess intra-abdominal adipose tissue further increasesthe risk, and in some cases, mortality of intra-abdominalcancers such as prostate, colon, pancreatic, and endometrialcancer (3, 20, 21). We hypothesized that a tumor-promot-ing population of multipotent MSCs in visceral adiposetissue could explain this relationship. To investigate this, weisolated a populationofASCs from theomentum, themajordepot of visceral intra-abdominal adipose tissue. The effectof omental-derived ASC (O-ASC) was then compared withbone marrow–derived MSC (BM-MSC), subcutaneous ASC(SC-ASC), and control fibroblasts by using ex vivo and in vivoassays.

Materials and Methods

Cell cultureCells were grown in a-minimum essential medium

(a-MEM) containing 20% FBS, L-glutamine, and penicil-lin–streptomycin. Hec1a and WI38 cells were obtainedfrom American Type Culture Collection. Hec1a cells werestably transfected using Lipofectamine, with pEF-Luc.SC-ASC from a cancer-free female donor (BMI¼ 33.0) werereported previously (22). BM-MSCs were isolated fromnormal individuals undergoing bone marrow harvest forallogeneic transplantation following informed consent,according to institutional guidelines under the approvedprotocol, as describedpreviously (23). Briefly,mononuclearcells were separated by centrifugation over a Ficoll-Hypaque

gradient (Sigma-Aldrich), suspended in a-MEM/10% FBS,and plated on 180 cm2 dishes. After 3 days, nonadherentcellswere removedbywashingwith PBS, andmonolayers ofadherent cells were cultured until they reached confluence.Cells were then trypsinized (0.25% trypsin with 0.1%EDTA) and subcultured at densities of 5,000 to 6,000cells/cm2. Cell passages 3 to 10 were used for the experi-ments (24). Cultured cells were routinely tested for viabilitywith trypan blue exclusion and maintained high viability(>95%). For in vivo experiments, theminute fractionof deadcells was predominantly removed by washing before tryp-sinization. For flow cytometric analysis, cells were routinelypre-gated to exclude cell clumps, contaminating polymor-phonuclear cells, red blood cells, platelets, endothelialmicroparticles, debris, as well as dead cells based on 7-AADstaining.

Isolation of omental ASCUnder an Institutional Review Board-approved protocol,

grossly normal appearing human omentum was obtainedfrom the staging procedures. O-ASC1 was isolated from apatient (BMI ¼ 25.3) with recurrent adenocarcinoma ofendometrium and ovary (omentumnot involved). O-ASC2was isolated from a patient (BMI ¼ 30.4) with recurrentgranulosa cell tumor of the ovary (omentum involved).ASCs were isolated according to published protocols (24).Tissue was subjected tomechanical disruption, followed by0.5 mg/mL collagenase type I (Worthington Biochemical)and 50 U/mL dispase (Becton Dickinson) digestion accord-ing to published protocols (6). Digested adipose tissue wascentrifuged at 1,000 rpm. Supernatant containing adipo-cytes was moved and cell pellet was resuspended in EGM-2MV (EBM-2 media with EGM-2MV growth factor supple-mentation from Lonza). Resuspended cells were filteredthrough 100 mm cell strainer (Becton Dickinson) andfollowed by 40 mm cell strainer. Red blood cells were lysedusing in 5 mL of RBC Lysis Buffer (eBioscience) withincubation in 37�C water bath for 10 minutes. Remainingmononuclear cells were resuspended and plated in EGM-2MV medium.

Characterization of O-ASC linesAfter isolation, cells were expanded in vitro, and then

characterized with flow cytometry to evaluate cell surfacemarker expression. O-ASC1 was characterized at passage 3and O-ASC2 at passage 5 with antibodies against the fol-lowingmarkers: CD34,CD44,CD45,CD29,CD90, EpCam(from Becton Dickinson), and CD105 (from BioLegend).Cell differentiation studies were conducted as describedpreviously (25, 26). For osteoblast differentiation, conflu-ent cells in 6-well plates were cultured in NH OsteoDiffMedia (Miltenyi) for 3 weeks with the media changed 2times a week. After 3 weeks,media were removed, cells werewashed 3 times with PBS and fixed with precooled 100%methanol. Cells were stained with Alizarin Red S solutionfor 2 to 5 minutes at room temperature to visualize orangered calcium deposits, or with von Kossa. For adipocytedifferentiation, confluent cells in 6-well plate were cultured

Translational Relevance

Both incidence and progression of endometrial cancerare positively associated with obesity. Preferential dis-tribution of adipose tissue in the abdomen furtherincreases the risk of endometrial cancer, as well as therisk and mortality of other intra-abdominal cancers. Wehypothesized that visceral adipose tissue contains apopulation of tumor-promoting stromal progenitors toaccount for the link between visceral obesity and malig-nancy. To investigate this, we isolated and characterizedadipose stromal cells (ASC) from the omentum of twopatients with gynecologic cancer (O-ASC). Using anendometrial cancer xenograft model, we compared theeffects of O-ASC and subcutaneous ASC (SC-ASC) oncancer progression. O-ASC more potently facilitatedtumor growth and enhanced tumor vascularization. Ourdata indicate that O-ASC and SC-ASC are equally effi-cient in homing to tumors but are markedly different intheir function. These findings suggest that excess visceraladipose tissue promotes malignancy by providing mes-enchymal stromal cells to intra-abdominal tumors andthus may represent a new therapeutic target.

Klopp et al.

Clin Cancer Res; 18(3) February 1, 2012 Clinical Cancer Research772




in adipogenic induction media, Dulbecco’s ModifiedEagle’s Media (DMEM; Mediatech) supplemented with10% FBS, 1% penicillin/streptomycin, 10 mg/mL insulin,500 mmol/L 3-isobutyl-1-methylxanthine (Sigma-Aldrich),1mmol/L dexamethasone (Sigma-Aldrich), and 200mmol/Lindomethacine (Sigma-Aldrich). After cells were main-tained in induction media for 48 to 72 hours, media werechanged to adipogenic maintenance media supplemen-ted with 10% FBS, 1% penicillin/streptomycin, and 10mg/mL insulin for 24 hours. After 24 hours, media werechanged back to induction media and cycle was repeated3 times. After the third induction step, cells were inmaintenance media for 10 days changing media twice aweek. Cells were then fixed in 10% formalin for 1 to 2hours followed by 60% isopropanol. Oil Red O (Sigma-Aldrich) stain was applied to visualize red lipid vacuoles.For chondrocyte differentiation, one million cells in a 15mL conical were pelleted and resuspended in 2 mL ofchondrocyte media; DMEM 1� (Mediatech) supplemen-ted with 1% penicillin/streptomycin, 50 mg/mL ascorbicacid (Sigma-Aldrich), 100 nmol/L dexamethasone (Sig-ma-Aldrich), and 10 ng/mL TGF-b3 (Invitrogen). Cellswere then incubated under 5% CO2 as a pellet at 37�Cwith cap loosened to allow for gas exchange. TGF-b3 wasadded daily to a concentration of 10 ng/mL. Media werechanged 3 times a week for 21 days. Cell pellet was rinsedin PBS and fixed with Gendre’s Fluid, after which paraffin-embedded blocks were sectioned and stained with anti-bodies against collagen II.

Proliferation assayLuciferase-labeled Hec1a cells were plated at a density of

1,000 cells per well in a BD Falcon 96-well plate alone or in1:1 coculture with O-ASC1, O-ASC2, BM-MSC, SC-ASC, orWI38. Cocultures were conducted in either a-MEM (Med-iatech) with 250 mg/mL G418. After 24 hours, luciferaseexpression wasmeasured in 0.15mg/mL D-luciferin. After 1hour incubation at 37�C, luminescence was measured withspectrophotometer (FLUOstar Omega, BMG Labtech).Media were then replaced andmeasurements were repeatedevery other day until cells reached confluence.

Generation of luciferase-GFP- andRFP-expressing cellsCell lines were labeled with the GFP using a lentiviral

vector, pFUGW (27) that was generously provided by Dr.Baltimore (Caltech). The titers of virus stocks were deter-minedby calculating the percentage ofGFP-positive (GFPþ)293T cells transduced with serially diluted virus suspen-sions. For transduction, the O-ASC and SC-ASC cell lineswere seeded at moderate density overnight. Two hoursbefore transduction, the medium was changed, and thentransductions were carried out for 24 hours in the presenceof 8 mg/mL polybrene (Sigma-Aldrich). The cells were thenfluorescence-activated cell sorted and further expanded forinjections. The MSCs were labeled similarly but with a redfluorescent protein (RFP) generously obtained from Tsien’slaboratory (UCSD) that was cloned into pFUGW constructin place of GFP.

Migration assayStromal cellmigration includingO-ASC,WI38, BM-MSC,

and SC-ASC were assayed in conditioned media (CM)generated from Hec1a cells. CM was collected from a 6-well plate which was plated with 300,000 cells per wellincubated in serum-free media for 48 hours. Migrationassays were conducted in a 24-well Transwell (CorningInc.) with 8 mm pores. CM was placed in lower portion ofTranswell and cell lines of interest on the upper portion ofTranswell. All samples were assayed at least in quadrupli-cate. Twenty-four hours later, the number of cellmigrated tothe bottom of the Transwell was measured by stainingmembranes with Hema 3 Stat Pack (Thermo Fisher Scien-tific) and mechanically removing residual cells on uppermembrane. The number of cells on themembranewas thenquantitated per high-power field.

In vivo studiesAnimal experimentation was approved by the Animal

Welfare Committee of theUniversity of Texas, Houston, TX.Ten-week-old mice NU/NU-foxn1nu (Jackson Laboratory)were injected subcutaneously with 1.5� 106 Hec1a cells in150 mL of PBS info the right flank (5 mice per group). WI38fibroblasts, MSC, O-ASC2, and SC-ASC were grown ina-MEM containing 20% FBS, L-glutamine before injection.O-ASC1was grown in eithera-MEMcontaining 20%FBS orEGM-2MV. Starting the next day, mice were injected with1.5� 105 corresponding cells (or PBS) every third day ontolower back, alternating the side of injection. Tumor size wasmeasured with a caliper and volume was calculated aslength � width2 � 0.52. Before tumors reached 1 cm insize,micewere euthanized and tumorswere recovered fromanesthetized mice after heart perfusion with 10 mL PBS.Immunofluorescence on formalin-fixed, paraffin-embed-ded tissue sections was carried out after antigen retrieval,washing with 0.2% Triton X-100, blocking in Serum-FreeProtein Block (DAKO), followed by incubation with pri-mary antibodies (4�C, 12 hours) and secondary antibodies(room temperature, 1 hour) in PBS containing 0.05%Tween-20. The primary antibodies used were as follows:goat anti-GFP from GeneTex (1:100), rabbit anti-RFP fromAbcam (1:100), rabbit anti-Ki-67 from Thermo Scientific(1:200), rabbit (1:200), or goat (1:100) anti-CD31 fromSanta Cruz Biotechnology, and rabbit anti-desmin fromAbcam (1:200). Secondary donkey FITC-conjugated (1:150) and Cy3-conjugated (1:300) IgG were from JacksonImmunoResearch. Nuclei were stainedwithHoechst 33258(Invitrogen). Masson’s Trichrome staining of tumor sec-tions was carried out at UTHealth histology CORE facility.Images were acquired with Olympus IX70 inverted fluores-cence microscope/MagnaFire software.

To assess tumor composition, 3 random internal tumorfield views proximal to the capsule were inspected for eachtumor at 100� magnification. In each field views, theproportion of the area positive for necrosis (Mason’s tri-chrome anuclear pink staining), proliferation (containingKi-67þ cells), vascularization (containing CD31þ vessels),and mature vasculature (containing desminþ vessels) was

Omental Adipose Stromal Cells Promote Endometrial Cancer

www.aacrjournals.org Clin Cancer Res; 18(3) February 1, 2012 773




scored by two investigators blinded to the experimentalgroup of the samples. Data were averaged from individualfield views for each group.

Statistical analysiswas conductedwithunpaired Student ttest. Normal distribution of data was confirmed by theKolmogorov–Smirnov test.

Results

Characterization of O-ASCCells were isolated from omental adipose tissue of 2

patients with gynecologic cancer: one with localized ade-nocarcinoma of endometrium andovary and the other withgranulosa cell tumor of the ovary infiltrating the omentum.Tissue was subjected tomechanical disruption, followed bycollagenase and dispase digestion (6). Adipocytes wereremoved with centrifugation and remaining mononuclearcells were plated and cultured in standard MSC expansionconditions. Morphologically, O-ASC from patients 1 and 2(O-ASC1 and O-ASC2) exhibited a mesenchymal pheno-type, similar to that previously reported for BM-MSC andSC-ASC (Fig. 1A).

To characterize cell surface marker expression on O-ASC,we expanded O-ASC1 and 2 and compared cell surfacemarker expression with flow cytometry to BM-MSC (Fig.1B). O-ASC expressed CD29, CD44, CD73, CD90, andCD105, which are characteristically expressed on MSCs(7, 28). BM-MSCs expressed slightly higher levels of CD44and CD105 than O-ASC, whereas O-ASC expressed slightlyhigher levels of CD29 than BM-MSCs. O-ASCs were nega-tive for an epithelial marker EpCam, a monocyte markerCD11b, and for CD34 and CD45. This excludes malignant,monocytic, and endothelial cells as a possible origin ofO-ASC.

To test whether O-ASCs exhibit multipotency typical ofMSCs (7) in vitro, we subjected them to differentiationassays (Supplementary Fig. S1). On adipogenic induction,O-ASC formed lipid droplets, which were detected with OilRed O stain showing lipogenesis typical of adipocyte dif-ferentiation. Similar to BM-MSCs, O-ASC1 cells were capa-ble of accumulating large lipid droplets, whereas O-ASC2cells tended to uniformly accumulate smaller lipid droplets.Following osteogenic induction, O-ASC1, and to a lesserdegree O-ASC2, formed mineral deposits detected withAlizarin red (Supplementary Fig. S1) and expressed alkalinephosphatase indicating osteoblastic differentiation. Onchondrogenic induction, Toluidine Blue staining of sec-tions from pellets grown in suspension indicated the capac-ity of O-ASC to differentiate into cartilaginous lineage.Expression of collagen II, although detectable in O-ASC1and O-ASC2, was lower than in BM-MSC differentiationcultures, suggesting that O-ASC formed primarily fibrocar-tilage. Combined, these results confirm that O-ASCs areindeed MSCs from the omentum.

ELISA for cytokinesStromal cells (O-ASC1, BM-MSC, SC-ASC, and WI38)

were cultured in 10% FBS a-MEM media. Media were

removed at intervals from 1 to 7 days, centrifuged andfrozen. ELISA assays were then done on supernatants byusing kits from R&D Systems.

O-ASC2O-ASC1

BM-MSCSC-ASC

A

O-ASC1 O-ASC2 BM-MSC

B

CD34

CD45

CD11b

EpCam

CD44

CD29

CD73

CD90

CD105

Figure 1. Characterization of O-ASC. A, morphology of adherent cellsshowing similarity of O-ASCs to SC-ASCs and BM-MSCs. B, flowcytometric characterization of O-ASC based on surface markerexpression indicates their similarity to BM-MSC. Light gray lines, isotypecontrols.

Klopp et al.





Homing of O-ASC to tumors and in vivo effects ontumor growthTo evaluate in vivo effects ofO-ASCon endometrial cancer

growth, immunodeficient nude mice with human endo-metrial adenocarcinoma Hec1a xenografts in the upperflank were injected into the lower flanks with low doses(105 cells 3 times weekly) of MSCs. In this metronomicmodel based on our previous study (6), the mesenchymalcells are remote from the xenografted tumor and must thusbe recruited into the tumors, simulating the off-site recruit-ment of ASC from regional adipose tissue in human tumors.The impact of O-ASC1 and O-ASC2 on Hec1a xenograftgrowth was then compared with SC-ASC and BM-MSC aswell as control lung fibroblasts (WI38) and a PBS controlgroup in thismetronomicmodel. Stromal cells injected intolower flank were labeled with GFP or RFP to track engraft-ment of cells into Hec1a tumors.Weekly tumor volumemeasurements showed that Hec1a

xenograft growth was significantly (P < 0.05) accelerated byadministration of O-ASC1, O-ASC2, and BM-MSC, as com-

paredwithPBS controls (Fig. 2A). In contrast, tumors grownin mice injected with SC-ASC or WI38 did not grow signif-icantly more rapidly than PBS-injected control tumors.

To determine whether the effect of O-ASC on tumorgrowth could be attributed to more efficient migration ofstromal cells into the tumor microenvironment, we evalu-ated tissue sections to assess biodistribution of injected cellswith immunofluoresence to RFP and GFP. Consistent withpublished studies (6), no GFPþ or RFPþ cells were detectedin liver, pancreas, spleen, andother control organs (data notshown). In contrast, comparable numbers of GFPþ or RFPþ

cells were detected in tumors of mice administered withO-ASC1,O-ASC2, BM-MSC, and SC-ASC but notWI38. Themarker-expressing cells were detected both in the internalzone of the tumor and, more commonly, in the stromalcapsule at the tumor periphery (Fig. 2B). On the basis of thenumbers of cells counted in the tumor section area, and theinferred number of such sections composing each tumor,we estimated that on the average 10,000 to 15,000 cellsweredetectable within a tumor, which corresponds to 10% to

Figure 2. Experimental endometrialtumor growth promotion by O-ASCs.A, nudemice xenografted (at week 0)with Hec1a tumors have beensubcutaneously injected (lower back)starting 1 day postxenografting withO-ASC, SC-ASC, BM-MSC, WI38fibroblasts, or PBS on alternatingsides 3 times a week. N ¼ 5 pergroup. O-ASC1, O-ASC2, andBM-MSCs significantly increasexenograft growth (�, P < 0.05compared with PBS group, Student ttest). Error bars, SEM. B, homing ofinjected cells detected byimmunofluorescence analysis ofparaffin sections of tumors collectedin the end of the injection cycle withanti-GFP antibodies. Green arrowsindicate cells engrafted in tumorperipheral and centralcompartments. Nuclei are stainedblue. Scale bar, 50 mm.

*

**

O-ASC2

O-ASC1

BM-MSC

SC-ASC

WI38

PBS

A

B Tumor edge Tumor center

Weeks postinjection

Tu

mo

r volu

me

(m

m3)

1

250

200

150

100

50

02 3 4

GFPGFP






15%of each injection dose. These numbers, consistent withthose reported previously for MSC tumor homing (29),indicate efficient recruitment of administered cells bytumors but not by normal organs. Lack of tumor growthpromotion by SC-ASC despite their capacity for tumorhoming indicate that properties unique for O-ASC andBM-MSC are responsible the tumor growth advantageobserved upon recruitment of these cells.

We tested that the administered cells contribute to thetumor vasculature. Immunofluorescence of tumor sectionswith antibody recognizing human and mouse CD31 showthat engrafted GFPþ cells are CD31-negative, althoughoften associated with CD31-positive endothelium (Supple-mentary Fig. S2). This indicates that mesenchymal cells donot transdifferentiate into endothelial cells in this experi-mental setting.We alsodidnot detect expressionofCD34 inengrafted GFPþ cells by using antibodies recognizinghumanCD34. This is consistent with in vitro flow cytometryanalysis of O-ASC showing they were negative for CD34(Fig. 1).

Effect of O-ASC on tumor matrix organization and cellviability

To investigate the mechanism through which O-ASCpromote tumor growth, we analyzed the morphology offixed tumors recovered in the end of the study. Masson’sTrichrome staining for ECMproteins revealed differences inorganization of tumors with recruited O-ASC1, O-ASC2,and BM-MSC as compared with the other groups andcontrol tumors in PBS-injected mice (Fig. 3). The formerhad thicker peripheral tumor capsule with defined extra-cellularmatrix (ECM) representation (blue) aswell as denseinner tumormass all theway to the center of the tumor core.In contrast, tumors in mice injected with PBS had a thinnerperipheral tumor capsule (Fig. 3). Histologic analysis alsoshowed that tumors in mice injected with O-ASC1 andO-ASC2 had no detectable necrosis at the tumor peripheryand only limited areas of necrosis in the center (Fig. 3). Incontrast, the other groups had tumors displaying vast areasof necrosis (detectable as pink areasmissing nucleated cells)both at the periphery and particularly in the tumor center.The percentage of tumor containing necrotic areas wasquantified to be significantly lower (average of 22% and9% necrotic tumor area in Hec1a xenografts recruitingO-ASC1 andO-ASC2 as comparedwith 48%necrotic tumorarea in PBS control tumors, P < 0.05; Supplementary Fig.S3). These results indicate that O-ASC support the forma-tion of the fibrous tumor capsule and decrease the extent ofintratumoral necrosis.

Effect of O-ASC on cancer cell proliferationTo determine whether O-ASCs promote malignant cell

proliferation, we carried out immunofluorescence to detectKi-67, a protein expressed in dividing cells. Immunofluo-rescence showed that malignant cells throughout thetumors ofmice recruitingO-ASC1 andO-ASC2were almostuniformly positive for Ki-67, whereas in WI38 and PBScontrol groups, proliferative tumor cells were limited to the

subcapsular region (Fig. 4A). Quantification of tumor sec-tion area containing proliferating cells revealed a higherpercentage of proliferative tumor mass in O-ASC1 andO-ASC2 groups, as compared with PBS-injected controlmice (average of 56% and 88% proliferative tumor massin O-ASC1 and O-ASC2 recruiting tumors respectively ascompared with 22% for control PBS tumors, P < 0.05;Supplementary Fig. S3). The frequency of proliferating cellsfor tumors recruiting SC-ASCs was not significantly higherthanPBScontrol tumors (Fig.4AandSupplementaryFig. S3).

To test whether the cells recruited by tumors have a directmitogenic effect on malignant cells, we analyzed prolifer-ation of Hec1a culture ex vivo (Supplementary Fig. S4). Inthis experiment, luciferase-labeled Hec1a cells were grownwith and without O-ASC1, O-ASC2, BM-MSC, SC-ASC, orWI38 fibroblasts and tumor cell proliferation was moni-tored with luciferase imaging. Although coculture witheither O-ASC1 or O-ASC2 significantly increased HEC1acell proliferation, SC-ASC and WI38 control cells had acomparable effect, whereas BM-MSCs on the contraryinhibited Hec1a proliferation. This lack of correlationbetween the ex vivo and in vivo effects of administered cells

O-ASC

BM-MSC

WI38

PBS

Tumor edge Tumor center

*

SC-ASC

Figure 3. Histologic tumor analysis. Masson's Trichrome staining ofsections of tumors collected in the end of the injection cycle revealsincreased capsular ECM deposition (blue), higher density of viable cells(dark purple nuclei) and lack of anucleated necrotic areas (�) in bothperipheral and inner compartments of tumors recruiting O-ASCs. This iscontrasted with control tumors, which exhibited less capsular ECMdeposition in the capsule with increased central tumor necrosis. Scalebar, 500 mm.

Klopp et al.





on Hec1a suggested that increased proliferation in tumorswas likely because of indirect effects of O-ASC.

Effect of O-ASC on tumor vascularizationBlood vessels are essential for tumor oxygenation and

nutrient supply, and their deficiency or inefficiency canlimit tumor perfusion and growth (30). We therefore inves-tigated whether enhanced angiogenesis could account forthe effect of O-ASC on xenograft growth. Interestingly,immunofluorescence with antibodies against CD31(PECAM-1) or VCAM-1 revealed only weak level of expres-sion of these endothelial makers on many Hec1a xenograftstructures that morphologically appeared as vasculature.Nevertheless, CD31 immunofluorescence signal was suffi-cient for us to detect higher density of vasculature in tumorsobserved formice administeredwithO-ASC1,O-ASC2, andBM-MSC as compared with controls and the other groups(Fig. 4). According to quantification studies, in O-ASC2groups, the percentage of tumor area containing CD31þ

blood vessels was significantly higher than in PBS controls,and there was also a trend for increased vascularization inthe O-ASC1 and MSC groups (Supplementary Fig. S3).Although large tortuous blood vessels were numerous bothat the periphery of the tumor closer to the capsule (Fig. 4A)and in the core of the tumor (Fig. 4B) for the O-ASC1,O-ASC2, and BM-MSC groups, virtually no CD31þ vesselswere detected in the tumor center for SC-ASC, WI38, andPBS control groups. Similar results were obtained by immu-nohistochemistry with antibodies against CD34, whichstain both endothelial and someof the perivascular/stromalcells (data not shown).

According to the histologic evaluation of tumors, thetumors that recruited O-ASC1, O-ASC2, and BM-MSC hada more extensive fibrovascular network and appeared tocontain more mature vasculature. The mesenchymal cellswithin this network are known to be a mixed population ofleukocytes and mesenchymal cells, some of which func-tionally integrate into tumor blood vessels as pericytes and

Figure 4. Tumor cell proliferationand vascularization.Immunofluorescence analysis ofsections of tumors collected in theend of the injection cycle. A, edge ofthe tumor (left) and proximal layersstained with anti-CD31 (green) andanti-Ki-67 (red) antibodies revealingincreased density of proliferatingcells residing in deeper tumor layersand increased presence of largetortuous blood vessels (arrows). B,increasedvascularizationassociatedwith O-ASC and BM-MSC injectionconfirmed for inner compartments oftumors. Nuclei are blue. Scale bar,500 mm.

CD31 Ki-67 Nuclei

O-ASC1 BM-MSC

SC-ASC PBSWI38

O-ASC2

A

B

BM-MSCO-ASC2

SC-ASC PBSWI38

CD31 Nuclei

Tumor edge

Tumor center

O-ASC1






contribute to vascular perfusion capacity (31). To comparethe extent of vasculature maturation in tumors recruitingdifferent types of MSCs, we analyzed tumor sections forexpression of desmin, a marker of pericytes (32). Anti-desmin immunofluoresence revealed robust pericytecoverage throughout the tumor vasculature in mice admin-istered with O-ASC1 and O-ASC2, and to a slighter lessextent with BM-MSC and SC-ASC (Fig. 5). Quantificationanalysis enumerating both perivascular and stromal des-min-positive cells indicated a significant increase in theirrecruitment, for all groups receiving cell injections, as com-pared with PBS-injected mice. Mice that received O-ASC1and O-ASC2 injections had significantly more desmin-pos-itive cells in tumors (Supplementary Fig. S3), consistentwith their increased vascularization.Desmin expressionwasabundant on perivascular cells in large tortuous bloodvessels both in subcapsular layers (Fig. 5A) and in the center(Fig. 5B) of the tumors in mice injected with O-ASC1,O-ASC2, or BM-MSC. In contrast, tumors in mice injectedwith for SC-ASC and WI38 contained predominantly

disseminated desmin-positive cells, whereas only someareas of the tumor contained rare desmin-positive cellsin the PBS group (Fig. 5A and B). Because no clear endo-thelial networks were revealed with CD31 (Fig. 4) orVCAM-1 (not shown) immunofluorescence in these 3groups, it is not clear whether these individual desmin-positive cells are associated with abnormal microvascula-ture weakly expressing endothelial markers or are simplydisseminated throughout the stroma. Combined, theseobservations suggest that O-ASCs promote formation ofpericyte coverage and, therefore, likely functional bloodvessels throughout the tumor upon recruitment, whichexplains improved cell survival and proliferation.

Cross-talk between O-ASC and tumor cellsTo investigate the potential differences in distinct ASC

types to secrete factors known to promote cancer progres-sion, we carried out ELISA assays (Supplementary Fig. S5).CMwere analyzed for the concentrationof secretedVEGF-A,fibroblast growth factor (FGF)-2, and hepatocyte growth

Desmin Nuclei

Tumor edge

B Tumor center

SC-ASC PBSWI38

BM-MSCO-ASC1 O-ASC2

O-ASC1 BM-MSC

SC-ASC WI38 PBS

O-ASC2

A

O-ASC1

Figure 5. Tumor vasculaturematurity analysis.Immunofluorescence analysis ofsections of tumors collected in theend of the injection cycle with anti-desmin (green) antibodies revealsincreased density of large tortuousvessels covered with pericytes inboth peripheral (A) and innercompartments (B) of tumorsrecruiting O-ASCs or BM-MSCs.Serial tumor sections shown in Bcorrespond to those stained withCD31 in Fig. 4B. Nuclei are blue.Scale bar, 500 mm.

Klopp et al.





factor (HGF). SDF1-a secretion was analyzed in parallel, asHEC1a tumors secrete CXCR4 and had been shown torespond to SDF1-a. Our data show that significantly higherlevels of VEGF, FGF, and SDF1-a are secreted by BM-MSCand O-ASC, as compared to SC-ASC and WI38. Thesefindings suggest thatO-ASCs and BM-MSCs promote tumorgrowth above the level observed for SC-ASC because of thecombined activity of several factors that can directly orindirectly enhance angiogenesis.Finally, to determine whether O-ASCs exhibit enhanced

in vitromigration in response to Hec1a-secreted factors, wecarried out an ex vivo migration assay. The effect of Hec1a-conditioned medium on the motility of cells used in thestudy was tested by using the Boyden chamber Transwellassay (Supplementary Fig. S6). O-ASC1, O-ASC2, and BM-MSC, but not SC-ASC and WI38 fibroblasts, exhibitedsignificant in vitro migration activation in response toHec1a-secreted factors (P< 0.05). Because the cells respond-ing to tumor signals also appear to be comparatively moreefficient in supporting tumors, our data collectively suggestthat the cross-talk between malignant and MSCs is animportant component of cancer progression. To gaininsight into the potential mechanisms of ASC traffickingto the tumor, we profiled chemokine secretion by tumorcells. Among a panel of 31 human chemokines tested, weidentified IL8 and CXCL1, as chemotactic proteins secretedby Hec1a cells (Supplementary Fig. S7). To determinewhether these factors could be mechanistically involved inO-ASCmigration,we investigated expressionof the IL-8 andCXCL1 receptors, CXCR1 and CXCR2, on stromal cells.High levels of CXCR1 expression were detected onO-ASC1,O-ASC2, and BM-MSC, but not on SC-ASC or WI38 fibro-blasts (Supplementary Fig. S8) whereas CXCR2 wasexpressed on all stromal cells investigated. These findingssuggest that IL-8 and CXCL1 secretion by Hec1a cells maytrigger chemotaxis of stromal cells toward tumors viaCXCR1 and CXCR2. Because all cell types displayed tumorhoming, we propose that the ubiquitously expressedCXCR2 is responsible for engraftment of the injected cellsin our experiment.

Discussion

Mechanism of endometrial tumor growth promotionby O-ASCActivation and proliferation of stromal cells recruited

by tumors leads to their conversion into myofibroblasts/reactive stroma (12, 33, 34). The resulting remodeling ofthe ECM executed by mesenchymal tumor cells and theresulting desmoplasia is an integral component of cancerprogression (35). However, the origins of these stromalcells, which compose the tumor microenvironment andtheir individual roles in cancer progression, are not wellunderstood. This is due largely to the absence of anyspecific marker to identify the origins of stromal cellswithin tumors.Studies in animal models have showed that on tumor

infiltration, MSCs contribute to the CAF pool and establish

a trophic microenvironment (9, 11). In previous studies byus and others, MSCs have been shown to traffic to tumors(19, 29) and participate in forming the supportive fibro-vascular network (13, 32, 36, 37). Although there is evi-dence that cells composing tumor stroma can be recruitedfrom the bone marrow, their possible extramedullary ori-gins have been less well explored. Adipose tissue is partic-ularly rich in MSCs and could thus serve as primary MSCsource for tumors (7, 38). Although systemic mobilizationof MSC from adipose tissue (ASC) has been proposed (25,26, 39) and recruitment of ASC from remote depots isconceivable and has been shown to occur in mouse cancermodels (6), it is likely that adipose tissues adjacent to thetumor may be a more significant contributor of ASC.Moreover, the differences reported for ASCs residing insubcutaneous and visceral adipose depots (40) suggest thatASCs from the omentum could have a distinct role in theprogression of intraperitoneal cancers.

Here, we analyzed omental adipose tissue, which is theprimary intraperitoneal reservoir of adipose tissue which isroutinely removed during staging procedures for ovarianand some endometrial cancers (41). This is the first studyinvestigating the effects of O-ASCs on cancer progression.We found that O-ASCs from different patients exhibitmorphology, cell surface marker expression, and differen-tiation capacity comparable with those observed for BM-MSCs and previously reported for SC-ASCs (42). O-ASC,SC-ASC, and BM-MSC exhibited tumor tropism wheninjected into tumor-bearing mice distantly from the endo-metrial xenograft. Despite these similarities, we found thatO-ASCs had more potent effect than SC-ASCs on endome-trial cancer cell proliferation and survival in vivo, thusresulting in expedited tumor growth. O-ASCs and SC-ASCswere isolated from donors with comparable BMI, suggest-ing that the differences in their cancer promotion capacitycannot be attributed to obesity-associated changes in ASCproperties. Consistent with promoted angiogenesis beingthe foundation of increased tumor cell mass (30), O-ASCrecruitment correlatedwith higher abundance of both smalland large tortuous blood vessels suggesting that O-ASCtumorhomingpromotes the vascularizationof endometrialcaners. In addition,we observed increased pericyte coverageof vasculature in tumors recruiting O-ASCs. Accumulatingevidence indicates that ASCs serve as pericytes in adiposetissue (43), which could be their function uncoupled frommultipotency and capacity to serve as progenitor cells.Although our previous studies show that ASCs can acquireperivascular localization upon tumor homing, the numbersof administered O-ASCs observed in the tumors cannotaccount for the increased numbers of desmin-positive cells.Therefore,O-ASCsmay play a role in recruiting endogenouspericyte progenitors that contribute to the formation offunctional tumor vessels.

Increased microvascular density (MVD) has been associ-atedwith advanced stage andpoorprognosis in endometrialcancer (44, 45). In our study, there was a clear correlationbetween the capacity of an administered mesenchymal celltype to enhance tumor vascularization and its mitogenic






and necrosis-protective effects. These findings suggest thatO-ASCs promote tumor growth through stimulation ofangiogenesis. Although these findings are consistent withthe previously reported proangiogenic ASC properties (43),for the first time they suggest that visceral adipose tissuecontains a population ofMSCswith unique tumor-promot-ing qualities. Our data indicate VEGF, FGF, and SDF1-a ascandidatemolecules responsible for the superior capacity ofO-ASC to promote tumor growth. Future studies systemat-ically surveying secretomesofASC fromdistinctWATdepotsmay identify new tumorigenic factors synergizing withVEGF, FGF, and SDF1-a. Although the vasculogenic func-tion of O-ASC may be sufficient to explain their tumorgrowth-promoting effects in our study, it is not clear whyBM-MSCs had a comparable effect on tumor growth despitetheir comparatively lower effect on the promotion of tumorvascularization. The mechanisms through which stromalcells regulate tumor physiology are complex and multifac-eted, and their functions independent of angiogenesis reg-ulation are likely to play a role. In addition to their directproangiogenic properties, MSCs execute their functionthrough the production of collagens and other ECM pro-teins, which provide the framework of the tumor microen-vironment, and in resolving scarring during consecutivestages of inflammatory processes (46). Although we couldnot detect quantifiable differences in the ECM depositionbetweendifferent groups, stromatogenesis anddesmoplasiaconcomitant with angiogenesis are likely to be importantas mechanisms through which O-ASCs regulate cancerprogression. In addition, recent studies show the capacityof MSC to induce the epithelial-to-mesenchymal transition(EMT) in malignant cells and promote metastasis (14, 18).The EMT has been shown to activate both proliferation andmigration of malignant cells, which makes it interesting toconsider possible role of O-ASCs in promoting metastaticcancer dissemination. Finally, like other MSCs, ASCs havebeen shown to suppress inflammation and mute T-cellfunction (46), which could at least partially account for theincreased tumor cell survival observed. There is a possibilitythat in addition to inducing pericyte recruitment or differ-entiation, O-ASCs could affect the immune responsethrough modulating infiltration of other cell populationsfrom the bone marrow. Because monocytes/macrophagesand various lymphocyte populations are known to influ-ence tumorgrowth,wemeasured their content in the tumorsby immunofluorescence. Although we did not detect sig-nificant differences inmacrophage and lymphocyte contentbetween different animal groups (data not shown), a moredetailed analysis may be needed in future studies to com-prehensively analyze the role of ASCs in shaping the com-position of tumor microenvironment.

We identified interleukin (IL)-8 as a potential chemotac-tic signaling pathway involved in O-ASC recruitment. Che-mokine profiling showed that CXCL8 (IL-8) andCXCL1 aresecreted by Hec1a endometrial carcinoma cells. The IL-8receptor, CXCR1, was detected on O-ASC1, O-ASC2, andBM-MSC but not SC-ASC orWI39 cells, suggesting a role forIL-8 signaling in O-ASC recruitment and possibly tumorpromotion. IL-8 has previously been reported to supportMSC recruitment in a glioma model and to mediate radi-ation-enhanced tumor tropism of MSC (47–49). Futurestudies will investigate the therapeutic potential of inhibit-ing O-ASC recruitment by suppressing signaling throughchemotactic pathways such as IL-8.

In summary, our results show that the omentum containsa population of MSCs which resemble BM-MSCs and SC-ASCs with regard to cell surface marker expression andmultipotency but have a unique effect on tumor progres-sion, suggesting their role in etiology of human intraperi-toneal cancers. Our combined data indicate that O-ASCssupport tumor vascularization, which leads to enhancedsurvival and proliferation of malignant cells and may con-tribute to O-ASC effects on tumor growth. We propose thatthe cross-talk between the tumor and the surroundingadipose tissue could contribute to ASC "education" andtheir adaptation for the tumor benefit, thus resulting in a"vicious circle" driving the progression of cancer. Ourfindings suggest a novel mechanism accounting for theincreased risk andmortality of endometrial cancer in wom-en with excess visceral adipose tissue. Preventing migrationof O-ASCs into endometrial cancers by disrupting interac-tions of O-ASCs with malignant cells, tumor endothelium,and other tumor stromal cells may be a novel treatmentstrategy.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

The authors thank Chieh Tseng for technical help in tissue analysis.

Grant Support

Support for K. Lu and R. Broaddus was provided from NIH 2P50CA098258-06 SPORE; for F. Marini from RC1CA146381, P50CA083639,R01CA109451, and R01NS06994; and for M.G. Kolonin from KomenKG080782 and NIH P50 CA140388 SPORE.

The costs of publication of this article were defrayed in part by the pay-ment of page charges. This article must therefore be hereby marked advertise-ment in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received July 27, 2011; revisedNovember 30, 2011; acceptedDecember 2,2011; published OnlineFirst December 13, 2011.

References1. Flegal KM, Graubard BI, Williamson DF, Gail MH. Cause-specific

excess deaths associated with underweight, overweight, and obesity.JAMA 2007;298:2028–37.

2. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight,obesity, andmortality from cancer in a prospectively studied cohort ofU.S. adults. N Engl J Med 2003;348:1625–38.

Klopp et al.





3. FriedenreichC,CustA, LahmannPH,Steindorf K, Boutron-RuaultMC,Clavel-Chapelon F, et al. Anthropometric factors and risk of endome-trial cancer: the European prospective investigation into cancer andnutrition. Cancer Causes Control 2007;18:399–413.

4. Bergstrom A, Pisani P, Tenet V, Wolk A, Adami HO. Overweight asan avoidable cause of cancer in Europe. Int J Cancer 2001;91:421–30.

5. Basen-Engquist K, Chang M. Obesity and cancer risk: recent reviewand evidence. Curr Oncol Rep 2011;13:71–6.

6. Zhang Y, Daquinag A, Traktuev DO, Amaya F, Simmons PJ, March KL,et al. White adipose tissue cells are recruited by experimental tumorsand promote cancer progression in mouse models. Cancer Res2009;69:5259–66.

7. Dominici M, Le Blanc K, Mueller I, Slaper-Cortenbach I, Marini F,Krause D, et al. Minimal criteria for defining multipotent mesenchymalstromal cells. The International Society for Cellular Therapy positionstatement. Cytotherapy 2006;8:315–7.

8. Daquinag AC, Zhang Y, Kolonin MG. Vascular targeting of adiposetissue as an anti-obesity approach. Trends Pharmacol Sci 2011;32:300–7.

9. Galie M, Konstantinidou G, Peroni D, Scambi I, Marchini C, Lisi V, et al.Mesenchymal stemcells sharemolecular signaturewithmesenchymaltumor cells and favor early tumor growth in syngeneicmice. Oncogene2008;27:2542–51.

10. Hu M, Yao J, Carroll DK, Weremowicz S, Chen H, Carrasco D, et al.Regulation of in situ to invasive breast carcinoma transition. CancerCell 2008;13:394–406.

11. Orimo A, Weinberg RA. Stromal fibroblasts in cancer: a novel tumor-promoting cell type. Cell Cycle 2006;5:1597–601.

12. Baglole CJ,RayDM,Bernstein SH, FeldonSE, Smith TJ, SimePJ, et al.More than structural cells, fibroblasts create and orchestrate the tumormicroenvironment. Immunol Invest 2006;35:297–325.

13. Hall B, Dembinski J, Sasser AK, Studeny M, Andreeff M, Marini F.Mesenchymal stem cells in cancer: tumor-associated fibroblasts andcell-based delivery vehicles. Int J Hematol 2007;86:8–16.

14. Karnoub AE, Dash AB, Vo AP, Sullivan A, Brooks MW, Bell GW, et al.Mesenchymal stem cells within tumour stroma promote breast cancermetastasis. Nature 2007;449:557–63.

15. Zhu W, Boachie-Adjei O, Rawlins BA, Frenkel B, Boskey AL, IvashkivLB, et al. A novel regulatory role for stromal-derived factor-1 signalingin bone morphogenic protein-2 osteogenic differentiation of mesen-chymal C2C12 cells. J Biol Chem 2007;282:18676–85.

16. Djouad F, Plence P, Bony C, Tropel P, Apparailly F, Sany J, et al.Immunosuppressive effect of mesenchymal stem cells favors tumorgrowth in allogeneic animals. Blood 2003;102:3837–44.

17. Djouad F, Mrugala D, Noel D, Jorgensen C. Engineered mesenchymalstem cells for cartilage repair. Regen Med 2006;1:529–37.

18. Klopp AH, Lacerda L, Gupta A, Debeb BG, Solley T, Li L, et al.Mesenchymal stem cells promote mammosphere formation anddecrease E-cadherin in normal and malignant breast cells. PLoS One2010;5:e12180.

19. Klopp AH, Spaeth EL, Dembinski JL, Woodward WA, Munshi A, MeynRE, et al. Tumor irradiation increases the recruitment of circulatingmesenchymal stem cells into the tumor microenvironment. CancerRes 2007;67:11687–95.

20. Prizment AE, Flood A, Anderson KE, Folsom AR. Survival of womenwith colon cancer in relation to precancer anthropometric character-istics: the IowaWomen's Health Study. Cancer Epidemiol BiomarkersPrev 2010;19:2229–37.

21. Balentine CJ, Enriquez J, Fisher W, Hodges S, Bansal V, Sansgiry S,et al. Intra-abdominal fat predicts survival in pancreatic cancer.J Gastrointest Surg 2010;14:1832–7.

22. Nie J, Chang B, Traktuev DO, Sun J, March KL, Chan L, et al.Combinatorial peptides identify a5b1 integrin as a receptor for thematricellular protein SPARC on adipose stromal cells. Stem Cells2008;26:2735–45.

23. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, MoscaJD, et al. Multilineage potential of adult human mesenchymal stemcells. Science 1999;284:143–7.

24. Bakker AH, Van Dielen FM, Greve JW, Adam JA, Buurman WA.Preadipocyte number in omental and subcutaneous adipose tissueof obese individuals. Obes Res 2004;12:488–98.

25. Kolonin MG, Simmons PJ. Combinatorial stem cell mobilization. NatBiotechnol 2009;27:252–3.

26. Bellows CF, Zhang Y, Simmons PJ, Khalsa AS, Kolonin MG. Influenceof BMI on level of circulating progenitor cells. Obesity 2011;19:1722–6.

27. Lois C, Hong EJ, Pease S, Brown EJ, Baltimore D. Germline trans-mission and tissue-specific expression of transgenes delivered bylentiviral vectors. Science 2002;295:868–72.

28. Bianco P, RobeyPG, Simmons PJ.Mesenchymal stem cells: revisitinghistory, concepts, and assays. Cell Stem Cell 2008;2:313–9.

29. Kidd S, Spaeth E, Dembinski JL, Dietrich M, Watson K, Klopp A, et al.Direct evidence of mesenchymal stem cell tropism for tumor andwounding microenvironments using in vivo bioluminescent imaging.Stem Cells 2009;27:2614–23.

30. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation.Cell 2011;144:646–74.

31. Du R, Lu KV, Petritsch C, Liu P, Ganss R, Passegue E, et al. HIF1alphainduces the recruitment of bone marrow-derived vascular modulatorycells to regulate tumor angiogenesis and invasion. Cancer Cell2008;13:206–20.

32. Spaeth EL, Dembinski JL, Sasser AK, Watson K, Klopp A, Hall B, et al.Mesenchymal stem cell transition to tumor-associated fibroblastscontributes to fibrovascular network expansion and tumor progres-sion. PLoS One 2009;4:e4992.

33. Dvorak HF. Tumors: wounds that do not heal. Similarities betweentumor stroma generation and wound healing. N Engl J Med 1986;315:1650–9.

34. Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420:860–7.

35. Bissell MJ, Radisky D. Putting tumours in context. Nat Rev Cancer2001;1:46–54.

36. Nakamizo A, Marini F, Amano T, Khan A, Studeny M, Gumin J, et al.Human bone marrow-derived mesenchymal stem cells in the treat-ment of gliomas. Cancer Res 2005;65:3307–18.

37. Studeny M, Marini FC, Dembinski JL, Zompetta C, Cabreira-HansenM, Bekele BN, et al. Mesenchymal stem cells: potential precursors fortumor stroma and targeted-delivery vehicles for anticancer agents.J Natl Cancer Inst 2004;96:1593–603.

38. Zuk PA, Zhu M, Mizuno H, Huang J, Futrell JW, Katz AJ, et al. Multi-lineage cells from human adipose tissue: implications for cell-basedtherapies. Tissue Eng 2001;7:211–28.

39. Bellows CF, Zhang Y, Chen J, Frazier ML, Kolonin MG. Circulation ofprogenitor cells in obese and lean colorectal cancer patients. CancerEpidemiol Biomarkers Prev 2011;20:2461–8.

40. JoeAW,Yi L, EvenY, Vogl AW,Rossi FM.Depot-specificdifferences inadipogenic progenitor abundance and proliferative response to high-fat diet. Stem Cells 2009;27:2563–70.

41. Usubutun A, Ozseker HS, Himmetoglu C, Balci S, Ayhan A. Omen-tectomy for gynecologic cancer: how much sampling is adequatefor microscopic examination? Arch Pathol Lab Med 2007;131:1578–81.

42. Suga H, Shigeura T, Matsumoto D, Inoue K, Kato H, Aoi N, et al. Rapidexpansion of human adipose-derived stromal cells preserving multi-potency. Cytotherapy 2007;9:738–45.

43. Traktuev D, Merfeld-Clauss S, Li J, Kolonin M, Arap W, Pasqualini R,et al. A population of multipotent CD34-positive adipose stromal cellsshare pericyte and mesenchymal surface markers, reside in a perien-dothelial location, and stabilize endothelial networks. Circ Res2008;102:77–85.

44. Kamat AA, Merritt WM, Coffey D, Lin YG, Patel PR, Broaddus R, et al.Clinical and biological significance of vascular endothelial growthfactor in endometrial cancer. Clin Cancer Res 2007;13:7487–95.

45. Espinosa I, Jose Carnicer M, Catasus L, Canet B, D'Angelo E,Zannoni GF, et al. Myometrial invasion and lymph node metastasisin endometrioid carcinomas: tumor-associated macrophages,microvessel density, and HIF1A have a crucial role. Am J SurgPathol 2010;34:1708–14.






46. Caplan AI. Adultmesenchymal stemcells for tissue engineering versusregenerative medicine. J Cell Physiol 2007;213:341–7.

47. Halpern JL, Kilbarger A, Lynch CC. Mesenchymal stem cells promotemammary cancer cell migration in vitro via the CXCR2 receptor.Cancer Lett 2011;308:91–9.

48. Birnbaum T, Roider J, Schankin CJ, Padovan CS, Schichor C, Gold-brunner R, et al. Malignant gliomas actively recruit bone marrow

stromal cells by secreting angiogenic cytokines. J Neurooncol2007;83:241–7.

49. Kim SM, Oh JH, Park SA, Ryu CH, Lim JY, Kim DS, et al. Irradiationenhances the tumor tropism and therapeutic potential of tumor necro-sis factor-related apoptosis-inducing ligand-secreting human umbil-ical cord blood-derived mesenchymal stem cells in glioma therapy.Stem Cells 2010;28:2217–28.

Klopp et al.





2012;18:771-782. Published OnlineFirst December 13, 2011.Clin Cancer Res Ann H. Klopp, Yan Zhang, Travis Solley, et al. Vascularization and Growth of Endometrial Tumors

Derived Stromal Cells Promote−Omental Adipose Tissue

Updated version

10.1158/1078-0432.CCR-11-1916doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2011/12/13/1078-0432.CCR-11-1916.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/18/3/771.full#ref-list-1

This article cites 49 articles, 11 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/18/3/771.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]

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/18/3/771To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-11-1916

http://clincancerres.aacrjournals.org/content/suppl/2011/12/13/1078-0432.CCR-11-1916.DC1

http://clincancerres.aacrjournals.org/content/18/3/771.full#ref-list-1

http://clincancerres.aacrjournals.org/content/18/3/771.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/18/3/771


omental adipose tissue derived stromal cells promote … · tribution of adipose tissue in the...

Documents