Microenvironment and Immunology

Stromal Progenitor Cells from Endogenous Adipose TissueContribute to Pericytes and Adipocytes That Populate theTumor Microenvironment

Yan Zhang, Alexes C. Daquinag, Felipe Amaya-Manzanares, Olga Sirin, Chieh Tseng, and Mikhail G. Kolonin

AbstractEpidemiologic studies associate cancer with obesity, but the pathophysiologic connections remain obscure. In

this study, we show that obesity facilitates tumor growth in mice irrespective of concurrent diet, suggesting adirect effect of excess white adipose tissue (WAT). When transplanted into mice, adipose stromal cells (ASC) canserve as perivascular adipocyte progenitors that promote tumor growth, perhaps helping explain the obesity–cancer link. In developing this hypothesis, we showed that ASCs are expanded in obesity and that they traffic fromendogenousWAT to tumors in severalmousemodels of cancer. Strikingly, a comparison of circulating and tumor-infiltrating cell populations in lean, and obese mice revealed that cancer induces a six-fold increase of ASCfrequency in the systemic circulation.We obtained evidence that ASCsmobilized in this way can be recruited intotumors, where they can be incorporated into blood vessels as pericytes and they can differentiate into adipocytesin an obesity-dependent manner. Extending this evidence, we found that increased tumor vascularization(reflected by changes in tumor vascular morphology and a two-fold increase in vascular density) was associatedwith intratumoral adipocytes and elevated proliferation of neighboring malignant cells. Taken together, ourresults suggest that ASCs recruited from endogenous adipose tissue can be recruited by tumors to potentiate thesupportive properties of the tumor microenvironment. Cancer Res; 72(20); 5198–208. �2012 AACR.

IntroductionObesity involves abnormal accumulation of body fat. As a

result of a prolonged positive energy balance, adipocytes inwhite adipose tissue (WAT) accumulate lipid droplets, whichhave systemic repercussions. Accompanying dyslipidemia,insulin resistance, and other systemic metabolic changes areimportant long-term consequences of obesity (1, 2). Epide-miology revealed that obesity, being a component of themetabolic syndrome, is associated with accelerated progres-sion of several types of cancer (3, 4). The state of chronicinflammation ensuing in both cancer (5) and obesity (1) mayplay a key role in linking obesity and cancer (2). It has alsobeen proposed that WAT has a direct effect on tumor growth(6, 7); however, convincing evidence has been lacking (8, 9).WAT is a potent endocrine organ secreting adipokines, suchas cytokines and growth factors (1, 10). Leptin, insulin-likegrowth factors (IGF), and steroid hormones have been

studied as adipokines potentially implicated in cancer (8–11). For example, IGF-1, systemic levels of which are elevatedin obesity, is sufficient to accelerate tumor growth in cancermodels (12).

While adipokines are secreted by adipocytes, importantangiogenic, immunomodulatory, and survival factors areproducts of infiltrating monocytic cells and of perivascularadipocyte progenitors, termed adipose stromal cells (ASC;refs. 13–15). ASCs are similar to mesenchymal stromal cells(MSC) originally isolated as bone marrow colony-forming unit(CFU-F) fibroblasts (16, 17). Accumulating evidence indicatesthat MSCs serve as progenitors of cells contributing to thetrophic tumor microenvironment (18–21). Administered bonemarrow MSC and ASC engraft tumors in animal models resultin accelerated cancer progression (7, 22–24); however, thephysiologic relevance of these results has been unclear. Thecapacity of mesenchymal progenitors to sense cancer as a siteof hypoxia/inflammation has been proposed to explain theirtumor homing (17, 25). ASC proliferation accompanies WATexpansion (26) and number of ASCs per gramofWAT increasesin obese individuals (27, 28). Mesenchymal progenitors arenormally undetectable in the peripheral circulation; however,mobilization of cells with the ASC immunophenotype inobesity, further elevated in patients with cancer (29, 30),suggests the possibility of their trafficking. On the basis ofthese observations, we have reasoned that ASC mobilizationand their recruitment by tumors could be increased in obesity,leading to increased stimulation of tumor growth through theparacrine action of adipokines.

Authors' Affiliation: Center for Stem Cell and Regenerative Medicine,Institute of Molecular Medicine, University of Texas Health Science Centerat Houston, Houston, Texas

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Mikhail G. Kolonin, Institute of Molecular Medi-cine, University of Texas Health Science Center at Houston, 1825 PresslerSt. Room630-G,Houston, TX77030.Phone: 713-500-3146; Fax: 713-500-2424; E-mail: Mikhail.G.Kolonin@uth.tmc.edu

doi: 10.1158/0008-5472.CAN-12-0294

�2012 American Association for Cancer Research.

CancerResearch

Cancer Res; 72(20) October 15, 20125198

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

To this day, the role of WAT in cancer progression has notbeen proven. Studies showing that tumor growth is acceleratedby diet-induced obesity (DIO) have not successfully uncoupledeffects of diet from effects of WAT (11, 31). Here, we show thatexcess WAT promotes tumor growth irrespective of the diet inmice. To investigate migration of cells from WAT duringcancer progression, we used a competitive repopulationmodelthat does not rely on invasive cell injections. We show thatrecruitment of endogenous ASCs in obesity is associated withincreased vascularization and adipogenesis accompanied byproliferation of malignant cells.

Materials and MethodsAnimal experimentsMouse studies were carried out under the Animal Welfare

Committee of the University of Texas (Houston, TX). Mousestrains C57BL/6, C57BL/6-Tg(UBC-GFP)30Scha/J (termedGFP mice), B6.Cg-Tg(ACTB-mRFP1)1F1Hadj/J (termed RFPmice), and B6.129S7-Rag1tm1Mom/J (termed RAG-1 mice)were from Jackson. For DIO induction (32), high-fat diet (HFD)D12492 (60 kcal% fat) and low-fat diet LFD (LFD) D12450B (10kcal% fat) from Research Diets were used. Body compositionwas measured by EchoMRI-100T (Echo Medical Systems) asdescribed (33). For tumor grafting, 106 cells were injected witha 21-gauge needle onto upper back (LLC and ID8) or intomammary fat pad (E0771 and MDA-231). Tumor size wasmeasured with a caliper; volume was calculated as length �width2 � 0.52. Tissues were recovered from Avertin-anesthe-tizedmice. Cells were isolated from tissues as described (7, 14).

Cell lines and primary cell cultureE0771 (fromF.M. Sirotnak), ID8 (fromF.C.Marini), and other

cancer lines (from American Type Culture Collection) werecultured in Dulbecco's Modified Eagle's Media containing 10%FBS and authenticated by animal grafting and subsequenttumor histology. Blood was recovered by heart perfusion with10 mL PBS/EDTA, and peripheral blood mononuclear cells(PBMC) were isolated as described (7, 30). Tissue suspensionswere prepared as described (7, 14). Adipogenesis induction andOil red O staining was conducted as described (14).

Flow cytometryFor fluorescence-activated cell sorting (FACS), cells were

pregated to exclude debris, cell clumps, contaminating poly-morphonuclear cells, red blood cells, as well as dead cells basedon 7-aminoactinomycin D (7-AAD) staining. Tissue cell sus-pensions, WAT stromal/vascular fraction (SVF) with adipo-cytes removed (14), or PBMCs were sorted into populationswith FACSAria/FACSDiva software (BD Biosciences) based onred fluorescent protein (RFP) fluorescence (Texas Red chan-nel), GFP fluorescence [fluorescein isothiocyanate (FITC)channel], and the following IgG clones: APC-anti-CD34(RAM34), PE-Cy7-anti-CD31 or PE-anti-CD31 (MEC 13.3),APC-Cy7-CD45 (30-F11), and the corresponding isotype con-trols (BD Biosciences). Isotypes and the positions of previouslycharacterized hematopoietic and endothelial populations onthe plots (7, 14) were used to set gate cutoffs.

Tissue analysisFormalin-fixed, paraffin-embedded tissues were sectioned

and analyzed by immunofluorescence as described (14).Primary antibodies used were goat anti-GFP (GeneTex,1:100); rabbit anti-RFP (Abcam, 1:100); rabbit anti-Ki-67(Thermo Scientific, 1:100); goat or rabbit anti-CD31 (SantaCruz Biotechnology, 1:100); rabbit anti-desmin (Abcam,1:200); and rabbit anti-perilipin (Cell Signaling Technology,1:100). Secondary donkey Alexa 488–conjugated (1:150) IgGwas from Invitrogen and Cy3-conjugated (1:300) IgG wasfrom Jackson ImmunoResearch. Nuclei were stained withHoechst 33258 or TO-PRO-3 (Invitrogen). Images wereacquired with a confocal Leica TCS SP5 microscope/LASAF software or Olympus IX70 inverted fluorescence micro-scope/MagnaFire software. For quantifications, at least 10random �100 magnification fields were blindly scoredand/or measured using microscope grid.

Statistical analysis was conducted by using the one-tailedhomoscedastic Student t test.

ResultsDiet-independent effect of obesity on tumor growth

First, we wished to uncouple the direct effect of obesity oncancer progression from indirect effect of the diet. We set upcohorts by rendering mice obese (body mass > 45 g) or lean(body mass < 30 g) by prefeeding with HFD or LFD. Toexclude diet as a variable, lean and obese mice have beenplaced on regular chow for 1 month before cancer initiation.HFD prefed mice slightly lost weight upon diet normaliza-tion, however, still had higher body mass due to increasedbody fat (Supplementary Fig. S1A and S1B). After dietnormalization, mice were isografted with tumor cells. Obesemice displayed accelerated kinetics of Lewis lung carcinoma(LLC) growth, as compared with lean mice (Fig. 1A). Astumors grew in size, obese mice gradually lost weight,whereas lean mice maintained it (Fig. 1A). Fat, but not lean,body mass decreased in response to cancer (SupplementaryFig. S1B), indicating that weight loss was specifically due tothe reduction in WAT amount. Human breast MDA-231adenocarcinoma orthotopically xenografted into the mam-mary fat pad of RAG-1 mice also grew faster in DIO mice,although the effect was subtle, coincidentally with obesitybeing less pronounced in this strain (Supplementary Fig.S1C). Using an ovarian ID8 adenocarcinoma model, wesegregated animals into obese, lean, and DIO-resistant micethat had become only moderately overweight on HFD beforetumor implantation. Despite HFD prefeeding, the kinetics oftumor growth in DIO-resistant mice was not significantlydifferent from that in the lean LFD controls (SupplementaryFig. S1C), indicating that prolonged positive energy balancedoes not fully account for the effect of obesity. Combined,these findings show that tumor growth can be promoted byobesity irrespective of diet at the time of tumor implantationand growth. Our data indicate excess WAT as an importantcomponent in the obesity–cancer relationship. However, thecontribution of systemic physiologic obesity consequences,remaining post-diet normalization (34), cannot be excluded.

Adipose Progenitor Cell Trafficking in Cancer

www.aacrjournals.org Cancer Res; 72(20) October 15, 2012 5199

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Obesity results in ASC expansionWe next set out to test whether recruitment of cells from

excessWATby tumors is implicated in obesity effect on cancer.

Obese mice displayed a 7-fold increase in body fat amount(Supplementary Fig. S1B). To investigate accompanying ASCexpansion, we compared ASC quantities in lean and obese

Figure 1. ASC expansion and promoted tumor growth irrespective of diet in obesity. A, tumor growth in lean and obese (DIO) C57BL/6mice graftedwith LLC cellsatweek 0 upondiet normalization (left) and concomitant changes in bodymass (right). B, flowcytometric enumeration of ASC (%of viable cells) among cells fromi.p.WATbygatingSVF on viableCD31�CD45� cells and thenonCD34þ cells. SSC-A, side scatter. C, relative net numbers of ASCs in i.p.WATof lean and obesemice quantified on the basis of flow cytometric ASC frequency and total i.p. WAT mass. D, phase contrast micrograph of adherent India ink–stained cellsfromi.p.WATafter 1day inculture featuringASCs(arrowheads)andmonocytes (arrows). E,adherentASCrecovery fromi.p.WATofobesemicenormalized to leanmice (from equal WAT amount). F, enumeration of ASC as CD34þCD31�CD45� cells (% of viable) in PBMCs of lean and obese mice bearing tumors. G,micrographs of adherent CD34þCD31�CD45� cells sorted fromWAT or from PBMCs of an obese mouse (left). H, oil red O staining of lipid droplets (arrows) in acolony formed by obese mouse PBMC–derived CD34þCD31�CD45� cell upon adipogenesis induction. Scale bar, 50 mm. Error bars, SEM. �, P < 0.05.

Zhang et al.

Cancer Res; 72(20) October 15, 2012 Cancer Research5200

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

animals. Unlike bone marrow MSCs, ASCs express CD34 (15),which enables their enumeration by flow cytometry as CD34þ

CD31�CD45� cells (7, 14). For intraperitoneal (i.p.) WAT, themajor ASC reservoir (26), the ASC frequency was found to be30.4% in lean and 32.6% in obese mice, respectively (Fig. 1B;Supplementary Fig. S2). On the basis of the total amounts of i.p.WAT recovered (1.0 � 0.23 g from lean and 6.19 � 0.25 g fromobesemice), we calculated that the net number of i.p. ASCs is 6times higher in obese mice (Fig. 1C). Frequencies of CD31�

CD45�CD34þ cells were very low in the bone marrow, lungs,and liver of both lean and obese animals (SupplementaryFig. S2). Upon plating WAT stromal/vascular cells fraction,adherent ASCs appear as large fibroblasts with well-definednuclei and nucleoli and can be clearly distinguished fromsmaller myelomonocytic cells internalizing India ink (Fig.1D). Quantification of adherent ASCs from i.p. WAT uponenzymatic tissue digestion showed increased efficiency in theirrecovery from obese mice (Fig. 1E). Combined, these dataindicate that WAT expanded in obesity serves as a reserve ofextra ASCs.

Obesity-associated cell mobilization and tumorinfiltrationTo test whether ASCs traffic from WAT through the

systemic circulation, we conducted a comparative analysisof PBMCs from lean and obese mice bearing tumors. Anal-ysis of PBMCs (Supplementary Fig. S3A and S3B) showedthat while circulating CD34þCD45� cells were rare in leananimals (0.06%), their frequency increased 6-fold (to 0.37%)in obesity (Fig. 1F). The majority of these cells had theCD34þCD31�CD45� ASC phenotype: the endothelial markerCD31 was expressed only by 5.9% of CD34þCD45� cells inobese mice (Supplementary Fig. S3A). Analysis of individualCD34þCD31�CD45� cells isolated by FACS showed mor-phology undistinguishable from that of ASCs sortedfrom WAT in parallel (Fig. 1G). Blood-derived CD34þ

CD31�CD45� cells formed colonies and accumulated lipiddroplets upon adipogenic induction (Fig. 1H). The obesity-associated egress of CD34þCD31�CD45� adipocyte progeni-tors strongly suggests their ASC identity. To obtain evidencethat ASCs may be recruited by tumors, we subjected tumorcell suspension to flow cytometric analysis. Indeed, tumorscontained CD34þCD31�CD45� cells (Supplementary Fig.S2), and the increase in their frequency associated withobesity was consistent with their possible WAT origin.

Abonemarrow transplantationmodel for trackingWAT-derived cellsTo enable tracking of hematopoietic tumor stroma (17,

18, 35, 36) in parallel with ASCs, we designed an in vivocompetitive repopulation assay. This obesity/cancer model isbased on 2 syngeneic mouse strains: host ubiquitously expres-sing GFP and donor expressing RFP. In chimeric GFP/RFPmice generated through bone marrow transplantation (Fig.2A), it is possible to distinguish hematopoietic (RFPþ) from thehost (GFPþ) cells. We reasoned that, if ASC traffic in cancer ashypothesized, obesity should be associated with increasedfrequency of GFPþ cells in tumors. Upon DIO induction and

RFP bone marrow transplantation followed by 1-month recov-ery on LFDandHFD, GFP/RFP chimeraswere subjected to dietnormalization and then orthotopically grafted into the mam-mary padwith E0771 cells, a line derived from syngeneic breastadenocarcinoma.

As expected, RFPþ and GFPþ cells were observed in allmouse tissues (Supplementary Fig. S4A and S4B). RFPþ cellswere confirmed as CD45þ leukocytes, and less than 0.5% ofbone marrow cells were GFPþ (Supplementary Fig. S4A).Femur sections revealed hematopoietic RFPþ cells relative toGFPþ vascular, bone, muscle, and WAT cells (Fig. 2B). Micro-scopic analysis of adherent WAT-derived cells showed thetypical ASC morphology of GFPþ cells and RFP fluorescenceof adherent leukocytes (Fig. 2B). In contrast, in peripheralblood, only rare adherent GFPþ cells were observed amongRFPþ leukocytes (Fig. 2B). The majority of small adherentfibroblastoid RFPþ monocytes, also observed in no-color mice(Fig. 1D), weremacrophages, as evident from India ink stainingand F4/80 expression (Supplementary Fig. S3B).

Consistent with the data in no-color mice (Fig. 1F), weobserved the frequency of circulating GFPþ cells being higherin obesemice (1.2%) than in leanmice (0.3%) by flow cytometry(Fig. 2C). In PBMCs of obese mice, 0.3% of GFPþ cells had theASC CD34þCD31�CD45� immunophenotype, whereas themajority were CD45þ leukocytes (Fig. 2C). Adherent cell anal-ysis confirmed that while in lean mice, virtually all circulatingGFPþ cells had the monocytic appearance, larger GFPþ fibro-blasts were present in PBMCs of obese mice (Fig. 2D). Theirtypical ASC appearance has become pronounced after 4 days inculture (Fig. 2E). Collagen-I, alpha-smooth muscle actin(a-SMA), and decorin, proteins expressed by ASCs (7, 14, 23)were expressed in colonies of GFPþCD34þCD31�CD45�

cells (Supplementary Fig. S5). Accumulation of lipid dropletsupon adipogenesis induction of the expanded GFPþ

CD34þCD31�CD45� cells as adipocyte progenitors (Fig. 2F)confirmed them as ASCs.

Engraftment of hematopoietic and WAT-derived cells intumors

Tumors grew significantly faster in obese GFP/RFP chi-meras concomitantly with fat mass loss (Fig. 3A). Confocalimmunofluorescence on tumor sections (Fig. 3B) revealed thatlarge GFPþ cells disseminated throughout both the peripheraland central tumor zones (Fig. 3B). They were observed at ahigher frequency in obese mice, whereas RFPþ cell frequencyshowed the opposite trend (Supplementary Fig. S4B). Analysisof tumor cell suspensions in culture revealed adherent GFPþ

cells with the ASC morphology (Fig. 3C). As expected, we alsoobserved GFPþ and RFPþmonocytes (Fig. 3C). More adherentGFPþ cells were recovered from tumors grown in obese mice(Fig. 3D). Approximately 44%of adherent GFPþ tumor–derivedcells had the ASC morphology in obese mice, compared with23% in lean (Fig. 3E). To exclude tumor size as a confounder inassessing the effect of WAT excess on ASC numbers in tumors,we have taken advantage of gender-specific differences intumor growth rate. Upon grafting E0771 tumors into chimericGFP/RFP lean females and age-matched obese (HFD prefed)males, we observed comparable tumor growth curves in the 2

Adipose Progenitor Cell Trafficking in Cancer

www.aacrjournals.org Cancer Res; 72(20) October 15, 2012 5201

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Figure 2. In-parallel tracking ofWAT andmedullary cells. A, transplant scheme. Bonemarrow fromRFP (red) mice is transplanted into lethally irradiated lean orobese (DIO) GFPmice, which are grafted with tumors after diet normalization. In obesemice, trafficking of GFPþASCs (green) fromWAT (yellow) is increased.B, GFP (green) and RFP (red) fluorescence of tissues and cells (1 day post-plating) isolated from an obese GFP/RFP chimera. Leukocytes (arrows)of host (green) and donor (red) origin and host ASCs (arrowheads) are indicated. C, flow cytometric enumeration of circulating GFPþ (FITC channel) cellsamong viable PBMCs from representative lean and obese E0771 tumor–grafted mice (left) and gating of the GFPþ cells from PBMCs of the obesemouse to enumerate CD34þ ASCs as a percentage of GFPþ cells (right). SSC-A, side scatter. D, total GFPþ cells from PBMCs of a lean and an obesemouse (C) were plated in culture for 1 day. GFP-fluorescent (green) adherent monocytes (arrows) and cells with the ASC morphology (arrowheads) areindicated. E, high magnification of obese mouse PBMC–derived GFPþ cells 4 days post-plating. F, lipid droplet formation (�) in a colony formed by an obesemouse PBMC–derived GFPþ cell ASCs 7 days after adipogenesis induction. Scale bar, 50 mm.

Zhang et al.

Cancer Res; 72(20) October 15, 2012 Cancer Research5202

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

groups. Flow cytometry indicated that tumor GFPþ cell fre-quency was still higher in obese mice (9.8%) than in lean mice(3.0%), despite similar tumor size (Fig. 3F). Importantly, 20.0%of tumor GFPþ cells in lean mice and 30.1% in obese mice hadthe CD34þCD31�CD45� phenotype (Fig. 3F) indicating themas ASCs.We tested whether a pericyte marker desmin could be used

to further validate GFPþ cells as ASCs. Our data show that91.3% of ASCs are indeed desmin-positive in WAT (Supple-mentary Fig. S6A). All PBMCCD34þCD31�CD45� cells and themajority (52.9%) of CD34þCD31�CD45� cells recovered fromtumors also were desmin-positive (Supplementary Fig. S6A),suggesting that mobilized ASCs retain their pericyte identity.Four-color confocal immunofluorescence on tissue sectionsfrom GFP/RFP chimeras showed co-expression of CD34 anddesmin on perivascular GFPþ cells in bothWAT and in tumors(Supplementary Fig. S6B). Finally, immunoblotting revealedthe presence of delta-decorin (DDCN), a recently identifiedmarker of ASC (14), in tumor protein extracts (SupplementaryFig. S6C), confirming ASC recruitment by tumors.

Tumor vascularization and cell proliferation associatedwith ASC recruitmentMasson's trichrome staining revealed comparable deposi-

tion of collagen in tumors from lean and obese animals(Supplementary Fig. S7A). This analysis also revealed lessextensive areas of hemorrhage and necrosis in obese mice.Immunofluorescence analysis revealed comparable fibronec-tin deposits typically devoid of GFPþ cells in tumors for eachgroup (Supplementary Fig. S7B). These data show that thedesmoplastic reorganization of the internal tumor matrix isnot significantly influenced by obesity in the models used.Blood vessels are essential for oxygen and nutrient delivery,

and vascular patency predetermines tumor growth (35, 37).Wetherefore investigated whether ASC recruitment is associatedwith vascular remodeling. In obesemice, a higher proportion ofendothelial cells were from the host, as revealed by luminal co-localization of CD31 with GFP (Fig. 4A and B). In lean mice,blood vessels were sparse in certain tumor areas, whereas inobese mice, they were abundant throughout the tumor mass(Fig. 4A). Quantification revealed a 2-fold higher vasculardensity in tumors from obese mice (Fig. 4C). Tumor vesselsin leanmicewere compressed and slim, whereas in obesemice,they were more hyperdilated (Fig. 4C) and filled with circu-lating blood cells (Fig. 4A and B). Enumeration of vesselspositive for desminþGFPþ cells (Fig. 4D) indicated a trend forincreased vessels maturation in tumors from obese mice (Fig.4C). In addition, expression of a-SMA, a perivascular markerexpressed on ASCs (7, 14, 23), was more abundant on perivas-cular GFPþ cells in obese mice (Fig. 4E). Combined, theseresults reinforce the notion that ASCs contribute to the pool ofperivascular cells in tumors in an obesity-dependent manner.An observation consistently made for tumors grown in

obese animals was that tumor capsules were notably thickerthan in lean mice (Figs. 3B and 5A and B). Tumor capsulecomposition also appeared different in obese animals (Sup-plementary Fig. S8A). Visually obvious increased adiposity wasconfirmed by immunofluorescence identifying perilipin, a

marker of mature lipid droplets (Fig. 5A). Numerous largeGFPþ cells were also observed dispersed throughout the tumorand contained unilocular perilipin-positive lipid droplets indi-cating them as adipocytes. While the presence of adipocytes atthe tumor periphery could potentially be explainable byingrowth of the surrounding connective tissue, the presenceof separate adipocytes abundant within the tumor core indi-cates their differentiation from engrafting progenitors. Inter-estingly, the average size of adipocytes was notably larger intumors grown in obese animals (Figs. 3 and 5), despite dietexcluded as a variable.

To test whether recruitment of ASCs is associated withchanges in malignant cell proliferation, we conducted immu-nofluorescence analysis to detect Ki-67 (Supplementary Fig.S8B). Tumors in obese mice were found to contain compar-atively more widespread areas populated with proliferatingcells (Fig. 5B and C). Importantly, up to 10% of tumor cells werepositive for Ki-67 in the vicinity of intratumoral adipocytes, asopposed to only 2% in areas devoid of adipocytes for obesemice (Fig. 5D). Cell proliferation was even more strikinglyassociated with the vasculature containing GFPþ cells (Fig.5C). In obesemice, up to 30% of tumor cells were Ki-67þ next tolarge GFPþ blood vessels and less than 5% in poorly vascular-ized areas devoid of GFPþ cells (Fig. 5D). This trend was alsoobserved for tumors in lean mice, although the frequency oftumor Ki-67þ cells was overall 4-fold lower (Fig. 5D). Becauseadipocytes themselves are found in association with bloodvessels in tumors, at this point, it is unclear whether adipocytessupport malignant cell proliferation and survival independent-ly of the vasculature. Combined, our data from independentmouse models suggest that WAT excess leads to functionalASC engraftment in tumor stroma.

DiscussionOur results indicate that obesity can accelerate tumor

growth irrespective of concurrent diet. While there are mul-tiple systemic effects through which obesity may promotecancer (34), in this study, we focused on the potential role ofexcess WAT. We hypothesized that in addition to systemicallysecreting adipokines, WAT serves as a source of cells recruitedby tumors and stimulating cancer through locally secretedparacrine factors.While bonemarrow is the bonafide source ofhematopoietic cells contributing to tumor microenvironment(18, 35, 36), our results indicate that mesenchymal progenitorsare recruited to tumors, at least partly, fromWAT. Evidence forthis phenomenon has surfaced in other recent reports (7, 22–24, 38). However, preceding studies have been based on datafrom invasive nonphysiologic models. Here, we provide evi-dence for ASC trafficking from endogenous WAT in vivo.

Our data suggest that a combination of elevated ASC avail-ability and their trafficking signaling results in the net increaseof ASC recruitment to tumors in obesity. Our findings areconsistent with the previously observed cellular tumor micro-environment composition changes in obesity (9, 39). Theobserved obesity-associated mobilization of ASCs, concomi-tant with their accumulation in tumor stroma/vasculature,indicates that WAT contributes to the pool of mesenchymal

Adipose Progenitor Cell Trafficking in Cancer

www.aacrjournals.org Cancer Res; 72(20) October 15, 2012 5203

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Figure 3. Recruitment of ASCs bytumors increased in obesity. A,tumor growth in lean and obeseGFP/RFP chimeras grafted withE0771 cells at week 0 upon dietnormalization (left) andconcomitant changes in fat bodymass measured by EchoMRI(right). B, confocalimmunofluorescence on sectionsfrom E0771 tumors with anti-GFP(green) and anti-RFP (red)antibodies identifies host anddonor cells, respectively. Shownfor lean and obese mice areperipheral and internal tumor areasas indicated. RFPþ cells within thestroma (red arrows), GFPþ cellswithin the stroma (green arrows),GFPþ vasculature-associatedcells (green arrowheads), andtumor capsule (brackets) areindicated. Nuclei are stained withTO-PRO-3 (blue). C, adherentGFPþ cells with ASC morphology(arrowheads) and RFPþ

monocytes (arrows) observed inculture 1 day post-plating of E0771tumor cell suspension from anobese GFP/RFP chimera. D,adherent GFPþ cells FACS-sortedfrom E0771 tumors grown in leanand obesemice 1 day post-plating.Cells with ASC morphology(arrowheads) and other cell types(arrows) are indicated. E,percentages of cells with ASCmorphology among GFPþ cellsfrom D. F, flow cytometric analysisof cell suspensions from size-matched E0771 tumors grown inlean and obese mice. Viable cellswere gated to separate RFPþ

(Texas Red channel) and GFPþ

(FITC channel) cells frommalignantcells (blue). The combined GFPþ

and RFPþ cells were then gated tovisualize CD31þ and CD45þ cells.Finally, GFPþCD31�CD45� cellswere gated to enumerate GFPþ

cells with the ASC CD31�

CD45�CD34þ phenotype (as apercentage of total GFPþ cells).�, P < 0.001. Error bar, SEM. Scalebar, 50 mm.

Zhang et al.

Cancer Res; 72(20) October 15, 2012 Cancer Research5204

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Figure 4. Pericyte recruitment in obesity andtumor vascularization. A and B, confocalimmunofluorescence analysis of sections fromE0771 tumors with anti-GFP (green) and anti-CD31 (red) antibodies. Shown for lean andobese GFP/RFP chimeras are internal tumorareas at low (A) and high (B) magnification.Blood vessels (red) contain luminal GFPþ

CD31þ endothelial cells (yellow arrows) andperivascular/stromal GFPþCD31� cells (greenarrows). Note increased pericyte coverage anddilation of tumor vessels in obese mice. C,quantitative vasculature analysis in E0771tumors from lean and obese mice. Vasculardensity was assessed as mean number ofvessels per �100 view field. Blood vessel sizewas assessed as mean lumen width for allvessels scored. Blood vessel maturity wasassessed as mean percentage of vesselsassociated with desmin-positive pericytesamong all vessels scored. Error bar, SEM.D, confocal immunofluorescence tumoranalysiswith anti-GFP (green) and anti-desmin(red) antibodies. Yellow signal upon digitalchannel merging indicates GFPþ pericytes,which is confirmed by Z-stack projections ofmedian series for individual cells in theindicated magnified area (bottom). E, confocalimmunofluorescence analysis of internal areasfrom E0771 tumors grown in lean and obeseGFP/RFP chimeras with anti-GFP (green) andanti-a-SMA (red) antibodies. Nuclei arestained blue (TO-PRO-3). Scale bar, 100 mm.

Adipose Progenitor Cell Trafficking in Cancer

www.aacrjournals.org Cancer Res; 72(20) October 15, 2012 5205

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

tumor cells. Our data argue against the possibility that prolif-eration of infiltrating ASCs significantly adds to their increasedabundance in tumors. While it is likely that ASCsmay infiltratethe tumor from adjacent surrounding WAT by migratingthrough solid tissues, recent reports on mobilization of mes-enchymal progenitors in patients with cancer (29, 40) inde-pendently indicate the bloodstream as a contributing route.

Our recent data suggest that CXCL1 and interleukin (IL)8secreted by tumor cells and signaling via receptors CXCR1 orCXCR2 is implicated inmigration of human omental ASCs (24),and future studies will establish the underlying mechanismsfurther.

The specific molecular pathways through which ASCsmay contribute to cancer progression are yet to be established.

Figure 5. Adipogenesis andproliferation associated with tumorvascularization in obesity. A,analysis of sections from tumors byimmunofluorescence withantibodies against GFP (green) andperilipin (red). a, GFPþ adipocytescontaining perilipinþ unilocularlipid droplets; b, GFPþperilipin�

blood vessels. B and C, analysis ofsections from tumors withantibodies against GFP (green) andKi-67 (red). Shown are proliferatingtumor cells (arrowheads) and rareproliferating GFPþ cells (arrow).Shown are peripheral (A and B) andinternal (C) E0771 tumor areas.Tumor capsule indicated withbrackets. Scale bar, 100 mm. D,effect of proximity to clusters ofintratumoral adipocytes (left) and toclusters of dilated blood vessels(right) on tumor cell proliferation.From 15 representative �100 viewfields, Ki-67þ nuclei were countedamong 100 Hoechst 33258–stained nuclei. Shown are mean �SEM. �, P < 0.05; ���, P < 0.001.

Zhang et al.

Cancer Res; 72(20) October 15, 2012 Cancer Research5206

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Our data indicate that ASCs recruited by tumors becomeperivascular or differentiate into intratumoral adipocytes.Mesenchymal progenitors, including ASCs, had been previous-ly shown to modulate cell survival, angiogenesis, and immuneresponse (25), which could account for their tumor-promotingeffects. While recent reports on tumor adipocytes in indepen-dent mouse models and in clinical specimens (41, 42) are inagreement with our findings, we have not observed accumu-lation of lipids in malignant cells reported in a recent study(43), which suggests that mechanisms of tumor lipid accumu-lation and metabolism may be cancer type–specific. Our dataon obesity-associated increase in pericyte coverage of tumorvessels suggest increased vascular patency, resulting inincreased malignant cell survival and proliferation, as a can-didate mechanism of ASC effect. This possibility is consistentwith reports on the pro-angiogenic effects of factors moleculessecreted by ASCs (15, 44). On the basis of our combined data,we propose that increased adiposity and vascularization asso-ciated with obesity-associated ASC recruitment may be func-tional predeterminants of tumor cell survival and proliferation,both contributing to tumor growth (Fig. 6).In summary, this study establishes recruitment of WAT-

derived cells by tumors as a potential contributor to thestimulatory effects of obesity on cancer progression. We pro-pose that several distinct ASC functions contribute to tumorgrowth induction observed in obesity. The apparent activity ofWAT-derived cells in tumors raises a question about the safetyof lipotransfer procedures in patients with cancer (45). Recentstudies suggest the role of WAT-derived stroma in cancermetastasis (22), and development of approaches to targetedinactivation of ASCs will enable identification of their specific

roles at distinct stages of cancer progression.We propose ASCsas a potential therapy target in cancer.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.The funders had no role in study design, data collection and analysis, decision

to publish, or preparation of the manuscript.

Authors' ContributionsConception and design: M.G. KoloninDevelopment of methodology: Y. Zhang, A.C. Daquinag, M.G. KoloninAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Y. Zhang, A.C. Daquinag, F. Amaya-Manzanares,O. Sirin, C. Tseng, M.G. KoloninAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): Y. Zhang, A.C. Daquinag, F. Amaya-Manzanares,O. Sirin, C. Tseng, M.G. KoloninWriting, review, and/or revision of the manuscript: Y. Zhang, A.C. Daqui-nag, F. Amaya-Manzanares, O. Sirin, C. Tseng, M.G. KoloninAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): Y. Zhang, F. Amaya-Manzanares,O. Sirin, C. Tseng, M.G. KoloninStudy supervision: Y. Zhang, M.G. Kolonin

AcknowledgmentsThe authors thankM. Andreeff, Rolph Brekken, B.R. Davis, A.J.Marian, and P.J.

Simmons for insightful comments as well as Fernando Florez for technical help.

Grant SupportThe work has been supported by grant CNE-119003 from the American

Cancer Society, as well as grants RP100400 and RP110776 from the CancerPrevention and Research Institute of Texas (CPRIT).

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received January 25, 2012; revised August 13, 2012; accepted August 17, 2012;published online October 15, 2012.

References1. Park J, Euhus DM, Scherer PE. Paracrine and endocrine effects of

adipose tissue on cancer development and progression. Endocr Rev2011;32:550–70.

2. Cowey S, Hardy RW. The metabolic syndrome: a high-risk state forcancer? Am J Pathol 2006;169:1505–22.

3. Flegal KM, Graubard BI, Williamson DF, Gail MH. Cause-specificexcess deaths associated with underweight, overweight, and obesity.JAMA 2007;298:2028–37.

4. Calle EE, Kaaks R. Overweight, obesity and cancer: epidemiologicalevidence and proposed mechanisms. Nat Rev Cancer 2004;4:579–91.

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

6. Vona-Davis L, Howard-McNattM,RoseDP. Adiposity, type 2 diabetesand the metabolic syndrome in breast cancer. Obes Rev 2007;8:395–408.

Figure 6. A working modelproposed for ASC trafficking incancer. In addition to the bonemarrow, WAT is a source of cellsrecruited by tumors. While tumorleukocytes are recruited mainly fromthe bone marrow, mesenchymaltumor stroma, pericytes, andadipocytes can be derived fromASCs migrating from WAT.

Adipose Progenitor Cell Trafficking in Cancer

www.aacrjournals.org Cancer Res; 72(20) October 15, 2012 5207

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

7. 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.

8. Roberts DL, Dive C, Renehan AG. Biological mechanisms linkingobesity and cancer risk: new perspectives. Annu Rev Med 2010;61:301–16.

9. Khandekar MJ, Cohen P, Spiegelman BM. Molecular mechanisms ofcancer development in obesity. Nat Rev Cancer 2011;11:886–95.

10. Sirin O, Kolonin MG. Treatment of obesity as a potential complemen-tary approach to cancer therapy. Drug Discov Today. 2012 May 22.[Epub ahead of print].

11. GrossmannME,RayA,NkhataKJ,MalakhovDA,RogozinaOP,DoganS, et al. Obesity and breast cancer: status of leptin and adiponectin inpathological processes. Cancer Metastasis Rev 2010;29:641–53.

12. HurstingSD,NunezNP,Varticovski L,VinsonC.Theobesity-cancer link:lessons learned from a fatless mouse. Cancer Res 2007;67:2391–3.

13. 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.

14. Daquinag AC, Zhang Y, Amaya-Manzanares F, Simmons PJ, KoloninMG. An isoform of decorin is a resistin receptor on the surface ofadipose progenitor cells. Cell Stem Cell 2011;9:74–86.

15. Gimble JM, Katz AJ, Bunnell BA. Adipose-derived stem cells forregenerative medicine. Circ Res 2007;100:1249–60.

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

17. Kolonin MG, Evans KW, Mani SA, Gomer RH. Alternative origins ofstroma in normal organs and disease. Stem Cell Res 2012;8:312–23.

18. Chantrain CF, Feron O, Marbaix E, DeClerck YA. Bone marrow micro-environment and tumor progression. Cancer Microenviron 2008;1:23–35.

19. 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. Oncogene2007;27:2542–5.

20. 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.

21. Bhowmick NA, Neilson EG, Moses HL. Stromal fibroblasts in cancerinitiation and progression. Nature 2004;432:332–7.

22. Martin-Padura I, GregatoG,Marighetti P,MancusoP,Calleri A, CorsiniC, et al. The white adipose tissue used in lipotransfer procedures is arich reservoir of CD34 þprogenitors able to promote cancer progres-sion. Cancer Res 2012;72:325–34.

23. KiddS, Spaeth E,Watson K, Burks J, LuH, Klopp A, et al. Origins of thetumor microenvironment: quantitative assessment of adipose-derivedand bone marrow-derived stroma. PLoS One 2012;7:e30563.

24. Klopp AH, Zhang Y, Solley T, Amaya-Manzanares F, Marini F, AndreeffM, et al. Omental adipose tissue-derived stromal cells promote vas-cularization and growth of endometrial tumors. Clin Cancer Res2012;18:771–82.

25. Bergfeld SA, DeClerck YA. Bone marrow-derived mesenchymal stemcells and the tumor microenvironment. Cancer Metastasis Rev2010;29:249–61.

26. 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.

27. Maumus M, Sengenes C, Decaunes P, Zakaroff-Girard A, Bourlier V,Lafontan M, et al. Evidence of in situ proliferation of adult adiposetissue-derived progenitor cells: influence of fat mass microenviron-ment and growth. J Clin Endocrinol Metab 2008;93:4098–106.

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

29. 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.

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

31. Nunez NP, Perkins SN, Smith NC, Berrigan D, Berendes DM, Varti-covski L, et al. Obesity accelerates mouse mammary tumor growth inthe absence of ovarian hormones. Nutr Cancer 2008;60:534–41.

32. Kolonin MG, Saha PK, Chan L, Pasqualini R, Arap W. Reversal ofobesity by targeted ablation of adipose tissue. Nat Med 2004;10:625–32.

33. Taicher GZ, Tinsley FC, Reiderman A, Heiman ML. Quantitative mag-netic resonance (QMR)method for bone andwhole-body-compositionanalysis. Anal Bioanal Chem 2003;377:990–1002.

34. De Angel RE, Conti CJ, Wheatley KE, Brenner AJ, Otto G, Degraffen-ried LA, et al. The enhancing effects of obesity on mammary tumorgrowth andAkt/mTORpathway activation persist after weight loss andare reversed by RAD001. Mol Carcinog. 2012 Jan 30. doi: 10.1002/mc.21878. [Epub ahead of print].

35. Du R, Lu KV, Petritsch C, Liu P, Ganss R, Passegue E, et al. HIF1ainduces the recruitment of bone marrow-derived vascular modulatorycells to regulate tumor angiogenesis and invasion. Cancer Cell2008;13:206–20.

36. DawsonMR, Chae SS, Jain RK, Duda DG. Direct evidence for lineage-dependent effects of bonemarrow stromal cells on tumor progression.Am J Cancer Res 2011;1:144–54.

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

38. Lin G, Yang R, Banie L, Wang G, Ning H, Li LC, et al. Effects oftransplantationof adipose tissue-derivedstemcells onprostate tumor.Prostate 2010;70:1066–73.

39. Park EJ, Lee JH, Yu GY, He G, Ali SR, Holzer RG, et al. Dietary andgenetic obesity promote liver inflammation and tumorigenesis byenhancing IL-6 and TNF expression. Cell 2010;140:197–208.

40. Mancuso P, Martin-Padura I, Calleri A, Marighetti P, Quarna J, Rabas-cio C, et al. Circulating perivascular progenitors: a target of PDGFRinhibition. Int J Cancer 2011;129:1344–50.

41. Dirat B, Bochet L, Dabek M, Daviaud D, Dauvillier S, Majed B, et al.Cancer-associated adipocytes exhibit an activated phenotype andcontribute to breast cancer invasion. Cancer Res 2011;71:2455–65.

42. Zyromski NJ, Mathur A, Pitt HA, Wade TE, Wang S, Nakshatri P, et al.Obesity potentiates the growth and dissemination of pancreatic can-cer. Surgery 2009;146:258–63.

43. Nieman KM, Kenny HA, Penicka CV, Ladanyi A, Buell-Gutbrod R,Zillhardt MR, et al. Adipocytes promote ovarian cancer metastasisand provide energy for rapid tumor growth. Nat Med 2011;17:1498–503.

44. Rehman J, Traktuev D, Li J, Merfeld-Clauss S, Temm-Grove CJ,Bovenkerk JE, et al. Secretion of angiogenic and antiapoptotic factorsby human adipose stromal cells. Circulation 2004;109:1292–8.

45. Bertolini F, Lohsiriwat V, Petit JY, Kolonin MG. Adipose tissue cells,lipotransfer and cancer: a challenge for scientists, oncologists andsurgeons. Biochim Biophys Acta 2012;1826:209–14.

Zhang et al.

Cancer Res; 72(20) October 15, 2012 Cancer Research5208

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

2012;72:5198-5208. Cancer Res   Yan Zhang, Alexes C. Daquinag, Felipe Amaya-Manzanares, et al.   MicroenvironmentContribute to Pericytes and Adipocytes That Populate the Tumor Stromal Progenitor Cells from Endogenous Adipose Tissue

  Updated version

  http://cancerres.aacrjournals.org/content/72/20/5198

Access the most recent version of this article at:

  Material

Supplementary

  http://cancerres.aacrjournals.org/content/suppl/2012/10/11/72.20.5198.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://cancerres.aacrjournals.org/content/72/20/5198.full#ref-list-1

This article cites 43 articles, 9 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/72/20/5198.full#related-urls

This article has been cited by 8 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

  .pubs@aacr.org

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://cancerres.aacrjournals.org/content/72/20/5198To request permission to re-use all or part of this article, use this link

on November 14, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from