Side Population Is Enriched in Tumorigenic, Stem-Like ... · Many adult stem cells also seem to have the ability to trans-differentiate into other cell types, although this phenomenon

Side Population Is Enriched in Tumorigenic, Stem-Like Cancer

Cells, whereas ABCG2+ and ABCG2�� Cancer Cells Are

Similarly Tumorigenic

Lubna Patrawala,1Tammy Calhoun,

1Robin Schneider-Broussard,

1Jianjun Zhou,

1

Kent Claypool,1and Dean G. Tang

1,2

1Department of Carcinogenesis, Science Park-Research Division, The University of Texas M.D. Anderson Cancer Center, Smithville, Texasand 2Program in Environmental and Molecular Carcinogenesis, Graduate School of Biomedical Sciences, Houston, Texas

Abstract

Recently, several human cancers including leukemia andbreast and brain tumors were found to contain stem-likecancer cells called cancer stem cells (CSC). Most of these CSCswere identified using markers that identify putative normalstem cells. In some cases, stem-like cancer cells wereidentified using the flow cytometry-based side populationtechnique. In this study, we first show that f30% of culturedhuman cancer cells and xenograft tumors examined (f30 intotal) possess a detectable side population. Purified sidepopulation cells from two cell lines (U373 glioma and MCF7breast cancer) and a xenograft prostate tumor (LAPC-9) aremore tumorigenic than the corresponding non–side popula-tion cells. These side population cells also possess someintrinsic stem cell properties as they generate non–sidepopulation cells in vivo, can be further transplanted, andpreferentially express some ‘‘stemness’’ genes, including Notch-1 and b-catenin . Because the side population phenotype ismainly mediated by ABCG2, an ATP-binding cassette half-transporter associated with multidrug resistance, we subse-quently studied ABCG2+ and ABCG2� cancer cells with respectto their tumorigenicity in vivo . Although side population cellsshow increased ABCG2 mRNA expression relative to the non–side population cells and all cancer cells and xenograft tumorsexamined express ABCG2 in a small fraction (0.5-3%) of thecells, highly purified ABCG2+ cancer cells, surprisingly, havevery similar tumorigenicity to the ABCG2� cancer cells.Mechanistic studies indicate that ABCG2 expression isassociated with proliferation and ABCG2+ cancer cells cangenerate ABCG2� cells. However, ABCG2� cancer cells canalso generate ABCG2+ cells. Furthermore, the ABCG2� cancercells form more and larger clones in the long-term clonalanalyses and the ABCG2� population preferentially expressesseveral ‘‘stemness’’ genes. Taken together, our results suggestthat (a) the side population is enriched with tumorigenicstem-like cancer cells, (b) ABCG2 expression identifies mainlyfast-cycling tumor progenitors, and (c) the ABCG2� popula-tion contains primitive stem-like cancer cells. (Cancer Res 2005;65(14): 6207-19)

Introduction

Stem cells, which have now been found in multiple adult tissuesand organs, have several fundamental properties. First, they aregenerally very rare. For example, the long-term hematopoietic stemcells (LT-HSC) in mouse bone marrow constitute f0.02% and theshort-term HCSs (ST-HSC) f0.1% of the total cells (1). Second,stem cells in their normal microenvironment (i.e., niche) rarelydivide, although they possess tremendous proliferative potential(2). Third, stem cells can self-renew; that is, they can regeneratethemselves when they divide to give rise to progenitor cells (2).Fourth, stem cells possess multipotential, oligopotential, orunipotential differentiation ability (3). Many adult stem cells alsoseem to have the ability to trans-differentiate into other cell types,although this phenomenon is still being hotly debated (3, 4). Finally,stem cells may express unique markers or properties that can allowtheir enrichment and identification.Indeed, many stem cells are identified by their expression of

unique markers. For example, mouse LT-HSCs, ST-HSCs, andmultipotent progenitors can be identified and separated by theirmarker phenotypes, c-kithiSca-1hiThy1.1loLin�/loFlk�, c-kithiSca-1hiThy1.1loLin�/loFlk+, and c-kithiSca-1hiThy1.1�Lin�/loFlk+, respec-tively (1). Unfortunately, most organ-restricted stem cells orprogenitors lack unique and specific markers. One way to identifythem relies on the slow-cycling properties of stem cells. Whenpulsed by bromodeoxyuridine (BrdU) for a period of time followedby ‘‘chasing’’ for longer time intervals, slow-dividing stem cells willbe identified as label-retaining cells (LRC; refs. 2, 5). The LRCspurified from the stem cell niche (i.e., the bulge in the hair follicle)in transgenic animals possess many of the expected stem cellproperties (2, 5). Another way to identify putative adult stem cellswas pioneered by Goodell et al. (6), who observed that whenbone marrow–derived cells are incubated with Hoechst dye 33342and then analyzed by dual-wavelength flow cytometry, a smallpopulation of cells does not accumulate an appreciable amount ofdye and is thus identified as a Hoechstlo side population.Remarkably, the side population is highly enriched in HSCs (6).Since its initial application in bone marrow HSCs, the sidepopulation technique has been adapted to identify putative stemcells and progenitors in multiple tissues/organs including umbil-ical cord blood (7), skeletal muscle (8–10), mammary glands(11, 12), lung (13–15), liver (16), epidermis (17, 18), forebrain (19),testis (20, 21), heart (22), kidney (23), limbal epithelium (24), andprostate (25).The side population–enriched stem cells are rare (f0.01-5%;

refs. 18, 23, 26) and heterogeneous, varying with tissue types, stagesof development, and methods of preparation (10, 27, 28). Forexample, the bone marrow side population cells contain not only

Note: J. Zhou is currently at the Dermatology Branch, National Cancer Institute,NIH, Building 10, Room 12N262, 10 Center Drive, MSC 1908, Bethesda, MD 20892-1908.

Requests for reprints: Dean Tang, Department of Carcinogenesis, Science Park-Research Division, The University of Texas M.D. Anderson Cancer Center, 1808 ParkRoad 1C, Smithville, TX 78957. Phone: 512-237-9575; Fax: 512-237-2475; E-mail: [email protected].

I2005 American Association for Cancer Research.

www.aacrjournals.org 6207 Cancer Res 2005; 65: (14). July 15, 2005

Research Article

Research. on September 7, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

HSCs but also mesenchymal stem cells (29) and do not capture allHSCs (30) but only a subset of long-term repopulating HSCs (31).The skeletal muscle side population cells are composed of mostlybone marrow–derived cells (8, 9) with only a minor population ofresident muscle stem/progenitor cells (i.e., satellite cells). The lungside population cells are also heterogeneous comprising bothCD45+ (i.e., bone marrow derived) and CD45� cells (13–15).Similarly, the testis side population cells may be enriched inspermatogonial, germinal, as well as mesenchymal (i.e., Leydig)stem cells (20, 21). Although the side population cells in most casesseem enriched in primitive stem cells, there are also reports thatsuggest that the side population cells do not identify stem cells(17, 18, 32).The side population phenotype is mediated by the ABC family of

transporter proteins. One of the major mediators seems to beABCG2 or BCRP (33), which was initially identified in drug-selected MCF7 breast cancer cells and later found to effluxmultiple chemotherapeutic drugs and xenobiotics (34). Thestrongest evidence linking ABCG2 and the side populationphenotype comes from the nearly complete loss of the bonemarrow side population phenotype in abcg�/� mice (35). Othersupporting evidence is that side population cells preferentiallyexpress ABCG2 (13, 16, 18, 20, 22, 24, 36, 37) and that ABCG2expression is detected in known stem/progenitor cells such asHSCs (33), nestin-positive islet-derived progenitors (36), hepaticoval cells (16), limbal basal cells (24), and neural stem cell (38). Onthe other hand, it should be noted that only a fraction of the sidepopulation cells expresses ABCG2 (e.g., ref. 24) and that both sidepopulation and known stem/progenitor cells also express otherABC transporters such as MDR-1 (i.e., ABCB1 or P-glycoprotein),MRP-1 (ABCC1), and ABCA2 (36, 37, 39), suggesting that theselatter molecules may also be involved in mediating the sidepopulation phenotype. In support, enforced expression of MDR-1in murine bone marrow cells leads to the expansion of the sidepopulation (40).Recently, several human cancers including leukemia (1, 41)

and breast (42) and brain (37, 43–46) tumors were found tocontain stem-like cells (SLC) called cancer stem cell (CSCs;ref. 47). Most of these tumorigenic SLCs were identified usingmarkers that identify putative normal stem cells. Interestingly,SLCs have also been identified in immortalized cell lines (39),long-term cultured cancer cells (37, 48), or patient tumorsamples (37) using the side population technique. Theseobservations suggest that even long-term cultured (cancer) cellsmay retain SLCs and that the side population approachrepresents a valid marker-independent method to identify suchcells. In a recent study,3 we have also provided independentevidence for the existence of SLCs in multiple human tumor cellcultures as well as xenograft tumors. In the current study, wefirst seek to confirm the utility of the side population techniqueto identify tumorigenic SLCs in cultured human cancer cells andxenograft tumors. Then we focus on the question whether thehigher tumorigenicity associated with the side population mightbe related to the expression of ABCG2, a major mediator of theside population phenotype in bone marrow cells. Our resultssurprisingly show that in contrast to the tumorigenic differences

between the side population and non–side population cells, theABCG2+ and ABCG2� cancer cells show very similar tumor-igenicities in vivo .

Materials and Methods

Cells, reagents, and animals. Various cancer cell lines includingthose of prostate (PPC-1, Du145, LNCaP, and PC3 cells), breast (T47D,

MCF-7, MDA-MB231, MDA-MB435, and MDA-MB468), colon (COLO 320,

DLD-1, RKO, and HCT116), bladder (RT4, UC14, UC1, and UC3), glioma

(D54, U87, U251, and U373), melanoma (WM266-4 and WM562-4), cervix(HeLa), and ovary (SKOV-3) were obtained from American Type Culture

Collection (Manassas, MA) or collaborators and cultured in the

recommended medium containing 5% to 10% of heat-inactivated fetal

bovine serum (FBS). Xenograft human prostate tumors LAPC-4 andLAPC-9 were kindly provided by C. Sawyers (Department of Medicine,

University of California, Los Angeles, CA) (49) and maintained in

nonobese diabetic/severe combined immunodeficiency (NOD/SCID) mice.Du145 xenograft tumors were established in our lab using early-passage

cells and maintained in NOD/SCID mice. All animals were obtained from

The Jackson Laboratory (Bar Harbor, ME) and maintained in standard

conditions according to the Institutional guidelines. The monoclonal anti-ABCG2, FITC-, and PE-conjugated anti-ABCG2 monoclonal antibodies,

isotype control antibody, and secondary antibodies were all obtained

from Chemicon (Temecula, CA). Monoclonal antibodies against Ber-EP4

and CD31 were obtained from DakoCytomation, Inc. (Carpinteria, CA) andBD PharMingen (San Diego, CA), respectively. All chemicals were obtained

from Sigma (St. Louis, MO) unless specified otherwise.

Side population analysis. The protocol was based on Goodell et al. (6)with slight modifications. Briefly, cells (1 � 106 cells/mL) were incubated in

prewarmed DMEM/5% FBS containing freshly added Hoechst 33342 (5 Ag/mL final concentration) for 90 minutes at 37jC with intermittent mixing. In

some experiments, cells were incubated with the Hoechst dye in thepresence of verapamil (50 Amol/L) or reserpine (100 Amol/L). At the end of

incubation, cells were spun down in the cold and resuspended in ice-cold

PBS. 7-AAD (2 Ag/mL final concentration) was added for 5 minutes before

fluorescence-activated cell sorting (FACS) analysis, which allows for thediscrimination of dead versus live cells. As positive controls, we used HL60

promyelocytic leukemia cells selected by chronic exposure to doxorubicin.

Samples were analyzed on a Coulter Epics flow cytometer. The Hoechst dye

was excited with the UV laser at 351 to 364 nm and its fluorescencemeasured with a 515-nm side population filter (Hoechst blue) and a 608

EFLP optical filter (Hoechst red). A 540 DSP filter was used to separate the

emission wavelengths.Indirect immunofluorescence, flow cytometry analysis of ABCG2

expression, and fluorescence-activated cell sorting of ABCG2+ cells.For fluorescence microscopy (50), 3 to 10 � 103 cells were plated on glass

coverslips. Next day, cells were fixed in 4% paraformaldehyde (10 minutesat room temperature) and blocked in 10% goat whole serum. Coverslips

were incubated sequentially with monoclonal antibody anti-ABCG2,

biotinylated goat anti-mouse IgG, and FITC-conjugated streptavidin with

washings in between. Generally 1,200 to 1,500 cells were counted for eachcell type to quantify the ABCG2+ cells. For flow cytometry, cells were gently

dissociated with Accutase (Innovative Cell Technologies, Inc., San Diego,

CA) and washed (twice) in serum-free medium. Cells were stained live inthe staining solution containing bovine serum albumin and insulin and

FITC- or PE-conjugated monoclonal anti-ABCG2 (15 minutes at 4jC).Samples were analyzed on a Coulter flow cytometer. A minimum of

500,000 viable cells was measured per sample, and cell debris and cellaggregates were electronically gated out. For FACS, 2 to 4 � 107 cells were

similarly stained for ABCG2 and used to sort out ABCG2+ and ABCG2�

cells. For the positive population, only the top 10% most brightly stained

cells were selected. For the negative population, only the bottom 5% to 10%most dimly stained cells were selected. The purity of ABCG2+ cells, as

determined by both post-sorting flow analyses as well as restaining followed

by fluorescence microscopy analyses, was z98% and the purity of theABCG2� cells was z99%.

3 C. Jeter et al. Stem-like cancer cells in culture and xenograft tumors: expressionand roles of stemness genes, submitted for publication.

Cancer Research

Cancer Res 2005; 65: (14). July 15, 2005 6208 www.aacrjournals.org



Xenograft tumor experiments and in vivo tumorigenicity. Xenograftprostate (Du145, LAPC-4, and LAPC-9) and glioma (U373) tumors were

aseptically dissected out from animals and minced into f1 mm3 pieces in

DMEM supplemented with 10% FBS. After rinsing in the same medium

(twice) followed by PBS to wash out serum, tumor tissues were incubated

with 1� Accumax (1,200-2,000 units/mL proteolytic activity containing

collagenase and DNase; Innovative Cell Technologies) at 10 mL/1 g tissue in

DPBS for 20 minutes at 37jC under rotating conditions. At the end, residual

tissues were allowed to precipitate to the bottom of tubes and dissociated

cells collected from the supernatant. When necessary, the residual tumor

pieces were subjected to one or more rounds of Accumax digestion and

dissociated cells pooled. A single cell suspension was obtained by filtering the

supernatant through a 40-Am cell strainer (BD Falcon, Bedford, MA). Upon

viability count using erythrosin B (American Type Culture Collection), cell

suspension was gently loaded onto a layer of Histopaque-1077 gradient

(Sigma-Aldrich, St. Louis, MO) and centrifuged at 400 � g for 30 minutes at

room temperature. RBC, dead cells, and debris were removed from the

bottom of the tube and live nucleated cells collected at the interface. Then the

cell mixture was depleted of lineage-positive host cells using the MACS

Lineage Cell Depletion Kit (Miltenyi Biotec, Auburn, CA). Briefly, cells were

first incubated (10minutes at 4jC) in the staining solution [PBS (pH 7.2), 0.5%

FBS, 0.5 Ag/mL insulin] containing biotinylated antibodies against a panel of

lineage antigens (CD5, CD45R, CD11b, anti-Ly-6G, 7-4, and Ter-119). Cells

were then incubated with the anti-biotin Microbeads (15 minutes at 4jC)and the Lin� cells were eluted using the MS columns. The eluted prostate

cancer cellswere all human epithelial cells as confirmed by their expression of

Ber-EP4, a surfacemarker unique to human epithelial cells, indicating thatwe

had obtained highly purified human tumor cells using this protocol.

For tumor experiments, various numbers of cells, either unsorted or

sorted populations (i.e., side population, non–side population, ABCG2+, andABCG2�) were injected in 40 AL of medium/Matrigel (1:1) s.c. into either

male or female NOD/SCID mice (4-8 weeks old). MCF7 cells were injected

into female mice. Tumor development was monitored starting from thesecond week. The primary tumor sizes were measured with a caliper on a

weekly basis and approximate tumor weights determined using the formula

0.5ab2, where b is the smaller of the two perpendicular diameters.

Tumorigenicity was measured mainly by tumor incidence (i.e., the numberof tumors/number of injections) and latency (i.e., time from injection to

detection of palpable tumors). All animals were terminated at 6 to 9 months

after tumor cell injection. Tumors harvested were fixed in formalin and

paraffin sections were made for H&E staining or immunohistochemistry forCD31.

Relationship between ABCG2 expression and cell proliferation. Twosets of experiments were done. In one, cells undergoing mitotic division

were determined in ABCG2+ and ABCG2� populations of cells. In the other,purified ABCG2+ and ABCG2� cancer cells, cultured for various time

intervals, were pulsed with BrdU (2.5 Amol/L � 3 hours) and processed for

BrdUrd staining (50).In vitro and in vivo self-renewal and clonal analyses. Purified ABCG2+

and ABCG2� cancer cells were plated at clonal density (i.e., 100-400 cells per

well; depending on cell type) in the flat-bottomed 6-well culture dishes. Cells

were cultured for different time periods. The percentage of cells thatinitiated a clone was presented as cloning efficiency. The clone sizes (i.e., the

number of cells/clone) were determined for some time points. Triplicate

samples were run for each cell type and experiments repeated when feasible.

For in vivo self-renewal experiments, tumor cells purified from the tumorsderived from unsorted, ABCG2+, or ABCG2� cells were stained for ABCG2

and used to quantify the ABCG2-expressing cells by flow cytometry.

Preparation of mouse bone marrow and newborn mouse keratino-cytes. To prepare bone marrow, we sacrificed C57BL/6 mice (6-8 weeks old)

by cervical dislocation and removed the skin covering the femurs. The

bones were removed by cutting below the knee and cutting at the hip. The

muscle was then cleaned from both femurs. The bone marrow cells wereflushed out of the femur using a 27-gauge needle in 50 mL of PBS, washed

once, and used in the side population analysis. For keratinocytes, newborn

pups were cleaned and anesthetized on ice for at least 30 minutes. The tail

and limbs were removed and discarded. The skin was removed and floated

on 2 mL trypsin (0.25%) solution with dermal side down overnight at 4jC.Epidermis was removed from dermis, placed in Waymouth’s Medium (Life

Technologies, Gaithersburg, MD) containing 10% FBS and 1% penicillin-

streptomycin and minced. The minced epidermis was gently stirred for

20 minutes in a sterile glass beaker. The resulting solution was filteredthrough a 30-Am mesh and plated at 3 � 106 cells per 35-mm dish for

2.5 hours. The medium was changed to Keratinocyte Growth Medium - 2

(KGM-2; Cambrex BioScience Walkersville, Inc., Walkersville, MD) with

0.5 mmol/L calcium. Cells were either used freshly or cultured for a shortperiod of time (to expand) and used in the side population analysis.

Reverse transcription-PCR analysis. Total RNA was isolated using the

Absolutely RNA Nanoprep Kit (Stratagene, La Jolla, CA) and used in

semiquantitative reverse transcription-PCR (RT-PCR) analysis (50). ThePCR primers included ABCG2 (sense, 5V-CTGAGATCCTGAGCCTTTGG-3V;antisense, 5V-TGCCCATCACAACATCATCT-3V); Notch-1 (sense, 5V-ATCGGG-CACCTGAACGTGGCG-3; antisense, 5V-CACGTCTGCCTGGCTCGG CTC-3V);h-catenin (sense, 5V-ACTGGCAGCAACAGTCTTACC-3V; antisense, 5V-TTT-GAAGGCAGTCTGTC GTAAT-3V); SMO (sense, 5V-ATCTCCACAGGAGA-GACTGGTTCGG-3V; antisense, 5V-AAAGTGGG GCCTTGGGAACATG-3V);Oct-4 (sense, 5V-GTGGAGGAAGCTGCAAACAATGAAA-3V; antisense, 5V-GACCGAGGAGTTACAGTGCAGTGAAG-3V); and glyceraldehyde-3-phos-

phate dehydrogenase (sense, 5V-ACCACAGTCCATGC CATCAC-3V; antisense,5V-TCCACCACCCTGTTGCTGTA-3V).

Results

Some cultured human cancer cells and xenograft tumorshave a side population. Several articles have recently reported thepresence of stem cell–enriched side population in long-termcultured mouse C2C12 myogenic cells (39), rat C6 glioma cells (48),or human brain tumor (i.e., glioma and medulloblastoma; ref. 37)cells. We first sought to determine whether this phenomenon isgenerally applicable to other human tumor cells in culture, inparticular, the human epithelial cancer cells. To that end, we firstestablished the side population protocol on our flow cytometerusing as experimental controls the HL60-Dox cells (courtesy ofDr. M. Andreef, Department of Blood and Marrow Transplantation,UT MD Anderson Cancer Center, Houston, TX); i.e., HL60 leukemiacells chronically exposed to a low concentration of doxorubicin.The HL60-Dox cells overexpressed the multidrug resistanceproteins that allowed them to efflux various drugs and xenobioticsincluding the Hoechst 33342 dye.4 As shown in Fig. 1, unselectedHL60 cells did not show a side population (A) whereas >90% of theHL60-Dox cells were in the side population (B), which presented asa distinct ‘‘tail’’ on the histogram and was completely inhibited byeither verapamil (C) or reserpine (D), chemicals previously shownto block the side population phenotype (e.g., refs. 6, 33). Todetermine the sensitivity of our system, we titrated the HL60-Doxcells into HL60 cultures and then did side population analysis. Theresults (Fig. 1E-H) revealed that our system could reliably detect aside population of f0.01%. Indeed, using these experimentalconditions, we reliably detected f0.01% and 0.5% side populationin mouse bone marrow cells (Fig. 1I) and newborn mousekeratinocytes (Fig. 1L), respectively, and the side populationphenotypes in these cells could also be blocked by verapamil orreserpine (Fig. 1J and K ; data not shown).Using the above protocol, we surveyed f30 cultured human

tumor cell lines of the prostate, breast, colon, glioma, bladder,ovary, cervix, glioma, and melanoma and we reliably detected aside population (0.04-0.2%) in f30% of the cell lines (Table 1; data

4 Unpublished observations.

Side Population, ABCG2, and CSC/Progenitor Cells




not shown). We also analyzed side population in tumor cells freshlypurified from three xenograft human prostate tumors (i.e., Du145,LAPC-4, and LAPC-9), and we only detected a distinct sidepopulation (0.07%) in the LAPC-9 tumor (Table 1; data not shown).These results suggest that only some cultured human cancer cellsand tumors cells freshly purified from xenograft tumors contain adetectable side population and that most cultured human cancercells may have a side population too small (i.e., <0.01%) to bereliably detected.Side population cells are more tumorigenic than the non–

side population cells. The side population cells isolated from therat C6 glioma have been shown to be more tumorigenic than thenon–side population cells (48). To determine whether the sidepopulation cells we identified in human cancer cells might also bemore tumorigenic, we did several small-scale tumor experiments.

We first purified the side population cells from the U373 gliomacells, which had f0.1% side population (Table 1; Fig. 2A , a-b).When 1,000 U373 side population cells were injected into the NOD/SCID mice, a prominent tumor arose within about 1 month(Table 2; Fig. 2A, c). The side population tumor histologicallyresembled clinical samples in that it had palisades-like structures(Fig. 2A, d) and was highly vascularized (Fig. 2A, e). In contrast, notumor was observed in 7 months when 50,000 non–side populationU373 cells were injected (Table 2).We similarly isolated side population cells from MCF7 breast

cancer cells, which had f0.2% side population (Table 1;Fig. 2B, a). When injected into the NOD/SCID mice, we observeda cell number–dependent tumor development (Table 2). Forexample, we observed a 17% (one of six) tumor incidence with1,000 side population cells injected and 50% (three of six) tumor

Figure 1. Side population (SP ) analysis.A-H, determination of the sensitivity of theside population protocol using HL60-Doxcells. Vera, verapamil; Res, reserpine.E-H, HL60-Dox cells were added to HL60cells at the concentrations indicated. I-K,mouse bone marrow (BM ) cells were usedin side population analysis in the absence(I ) or presence of verapamil (J ) orreserpine (K). L, newborn mousekeratinocytes were used in side populationanalysis.

Cancer Research




incidence with 10,000 side population cells injected (Table 2).Tumor latency was also reduced with increased numbers of sidepopulation cells injected (Table 2). In contrast to the sidepopulation MCF7 cells, the non–side population MCF7 at 1,000cells did not give rise to any tumors (zero of six). With 10,000cells injected, we observed one tumor with six injections(Table 2). Even with 250,000 cells injected, we only observedone of two tumor incidence (Table 2). In addition, the tumorsfrom the non–side population MCF7 cells arose with longerlatencies (Table 2).Finally, we purified from the LAPC-9 prostate xenograft tumors

the side population cells, which constituted f0.07% of the totaltumor (Table 1; Fig. 2C , a-b). When injected into the NOD/SCIDmice, as few as 100 cells generated a tumor (25% incidence; Table 2).With 1,000 side population cells injected, we observed a tumorincidence of 75% (three of four; Table 2). With 1,500 side populationcells injected, we observed two of two tumor formation with shorterlatencies (Table 2). Tumors derived from the side population cellshistologically resembled the unsorted LAPC-9 tumors (data notshown). In contrast, no tumor generation was observed with 1,500(zero of six) or 15,000 (zero of four) non–side population LAPC-9cells injected. Even 150,000 non–side population LAPC-9 cellsinjected did not give rise to tumors within 9 months, althoughtumor did arise with 300,000 non–side population cells (Table 2).These results suggest that the side population LAPC-9 tumor cells

are probably >100 times more tumorigenic than the non–sidepopulation LAPC-9 cells.Side population cells have some stem cell properties. To

determine whether the higher tumorigenicity in the side popula-tion cells might be associated with some of the intrinsic stem cellproperties, we first used the purified side population and non–sidepopulation U373 cells in a clonogenic assay, which partiallymeasures the self-renewal capacity of the cells. As shown inFig. 2A, f , whereas f10% of the side population U373 cells couldsustain a clonal growth, <0.01% of the non–side population U373cells were clonogenic. Then we asked whether the side populationcell-generated tumors contain non–side population cells and canbe further passaged in vivo . To that end, we took the U373 sidepopulation tumor obtained from the preceding experiments(above) and purified tumor cells out and did side populationanalysis. The results revealed a side population of f0.1% (data notshown), which was similar to the percentage of side populationcells in the first-generation tumor (Table 1; Fig. 2A, a). Theseobservations suggest that the side population cells can give rise tonon–side population cells in vivo . When injected into the NOD/SCID mice, 100 side population cells generated tumors with f60%(7 of 11) efficiency and 1,000 side population U373 cells alsogenerated a tumor (Table 2). By contrast, no tumors were observedwith 1,000 non–side population cells injected (zero of six). Even200,000 non–side population U373 cells injected did not generate a

Table 1. Side population and ABCG2 in human tumor cells

Cells Characteristics Side population* ABCG2c

Prostate tumor cells

LNCaP AR+, low tumorigenicity; low metastatic UD 1.9LAPC9 AR+, intermediate tumorigenicity; metastatic 0.07 1.2

Du145 AR�, intermediate tumorigenicity; metastatic UD 0.8

PC3 AR�, high tumorigenicity; highly metastatic UD 2.0

PPC-1 AR�, high tumorigenicity; highly metastatic UD 0.2Breast tumor cells

MCF7 ER+, low tumorigenicity 0.2 0.6

T47D ER+, low tumorigenicity ND 1.3MDA-MB468 ER�, intermediate tumorigenicity ND 0.1

MDA-MB231 ER�, tumorigenic and metastatic UD 0.6

MDA-MB435 ER�, tumorigenic and metastatic UD 0.7

Colon tumor cellsCOLO 320 tumorigenic ND 1.3

DLD-1 tumorigenic and invasive ND 0.2

RKO highly tumorigenic; metastatic ND 3.7

HCT116 highly tumorigenic; metastatic ND 3.1Glioma cells

D54 highly tumorigenic and invasive 0.1 0.7

U87 highly tumorigenic and invasive 0.05 2.0

U251 highly tumorigenic and invasive 0.04 1.9U373 highly tumorigenic and invasive 0.1 2.5

Others

WM266-4 melanoma UD 2.3WM562-4 melanoma UD 0.6

HeLa cervical carcinoma UD 0.5

SKOV3 ovarian adenocarcinoma; metastatic 0.05 0.7

Abbreviations: UD, undetectable; ND, not determined.

*The numbers indicate the % and represent the mean values of three to seven independent experiments.cThe numbers represent the mean % derived from two to five independent immunostaining and/or flow cytometry analyses.





tumor (Table 2), suggesting that the U373 side population cellspossess self-renewal capacities in vitro and in vivo and they areprobably >200 more tumorigenic than the non–side populationcells.To determine whether the side population cells have some other

intrinsic properties of stem cells such as preferential expression of‘‘stemness’’ genes, which are important for stem cell self-renewal,proliferative capacity, or fate determination (2, 47, 51),3 we didsemiquantitative RT-PCR analysis of Notch-1, h-catenin (a crucialmolecule in the Wnt signaling pathway), and Smoothened (SMO;the activating receptor for the Hedgehog signaling pathway). Asshown in Fig. 2B, b , the side population MCF7 cells expressedhigher levels of Notch-1 and h-catenin mRNAs than the non–sidepopulation MCF7 cells although both populations expressedsimilar levels of SMO mRNA.Taken together, these results suggest that the side population

cancer cells have some intrinsic properties of stem cells, similar toobservations in various normal stem cell populations (seeIntroduction).Expression of ABCG2 in a small subset of cancer cells:

association of ABCG2 with cell proliferation. Three consid-erations made us to turn our attention to ABCG2. First, ABCG2is one of the primary mediators of the side populationphenotype in mouse bone marrow and some other cells (35).

Second, the majority of the cancer cells we have studied do notpossess a detectable side population under our experimentalconditions therefore an alternate ‘‘marker’’ is needed to identifythe rare tumorigenic cells. Third, one of the potential concernsin the side population analysis is that the chronic accumulationof the low levels of the Hoechst dye in the non–side populationcells might be toxic to these cells, although our post-sort viabilityanalysis as well as culture of multiple non–side population cellsdid not reveal such cytotoxicities (data not shown). Regardless, ifABCG2 could be used as a surrogate marker for the sidepopulation cells, it would completely avoid the potentialcytotoxicity problem. With these considerations, we firstexamined the relationship between side population cells andABCG2 expression. As expected, the side population MCF7 cellsexpressed higher levels of ABCG2 mRNA than the non–sidepopulation cells (Fig. 2B, b). We then examined the ABCG2protein expression, using both immunofluorescence staining andflow cytometry, in the same spectrum of human tumor cell linesand xenograft tumor-derived cells. In every case, we detectedABCG2 expression in a small percentage (0.1-4%) of the cells(Table 1; Fig. 3). A similar ABCG2 expression pattern (i.e., 0.1-2%)was also observed in several bladder cell lines (Fig. 3E, d ; datanot shown) as well as in LAPC-4 and LAPC-9 xenograft tumors(data not shown).

Figure 2. Side population (SP) cells are more tumorigenic and have some stem cell properties. A, U373 side population analysis (a-b) and the side populationcell-initiated tumor (c ) analyzed for histology (d ; �100) or CD31 staining (e ; �100). Clonogenic assays using side population and non–side population cells(f ). B, MCF7 side population analysis (a ) or RT-PCR assays (b). C, LAPC-9 side population analysis (a-b ) and RT-PCR assays (c ).

Cancer Research




Interestingly, we observed that many of the ABCG2+ cancercells were in the process of cell division (Fig. 4). Overall, weobserved that f30% of the ABCG2+ cancer cells were mitotic,whereas only f1% of the ABCG2� cells were mitotic. In clonalanalyses, we found that the majority of divided cells that hadcompleted or were about to complete cytokinesis equallydistributed ABCG2 to both daughter cells (Fig. 4G-I and J-L).However, in f1% of the dividing cells undergoing cytokinesisABCG2 seemed to segregate asymmetrically to mainly onedaughter cell (e.g., Fig. 4M-O). These observations suggest thatABCG2 might preferentially mark proliferating cells and that someABCG2-expressing cancer cells might be undergoing asymmetricalcell division, a cardinal feature of stem cells. To determinewhether ABCG2 expression might be associated with cellproliferation in general, we prospectively purified ABCG2+ andABCG2� cancer cells to near homogeneity and used them in aBrdU-labeling experiment. As shown in Fig. 5, acutely purifiedABCG2+ Du145 prostate (A) and MDA-MB435 breast (B) cancercells that had been BrdU-pulsed and cultured for only 3 hoursshowed significantly more proliferation than the correspondingABCG2� cells.ABCG2+ and ABCG2� tumor cells are similarly tumorigenic.

Next, we purified ABCG2+ and ABCG2� cells and did tumorexperiments to determine whether the ABCG2+ cancer cells mightbe more tumorigenic. Much to our surprise, when the twopopulations of U373 cells were injected into the NOD/SCID mice,we did not observe any major differences with respect to theirtumorigenicities (Table 3). In fact, the ABCG2� cells tended togenerate tumors slightly faster than the ABCG2+ or unsorted cells(Table 3).To determine whether the lack of correlation between ABCG2

expression and tumorigenicity may be a cell type–restricted

phenomenon, we carried out similar tumor experiments usingpurified ABCG2+ and ABCG2� cells from prostate (Du145),breast (MDA-MB435), and colon (HCT116) cancer cell cultures.In every case, we failed to observe any significant difference intumor incidence or latency periods between the two populations(Table 3; data not shown). We also compared the ABCG2+ andABCG2� cells purified from Du145 xenograft tumors and againdid not observe any major differences in their tumorigenicities(Table 3).Evidence that ABCG2 expression marks proliferating

tumor progenitors whereas ABCG2� population containsprimitive cancer stem cells. If the ABCG2+ cells proliferatefaster than the ABCG2� cancer cells, why are they not moretumorigenic? One possibility is that ABCG2+ cells are mostly fastproliferating tumor progenitors (i.e., transit amplifying cells)rather than primitive, slow-cycling CSCs. To test this possibility,we first determined whether the ABCG2+ tumor cells couldgenerate ABCG2� cells and, more importantly, whether ABCG2�

cells could generate ABCG2+ cells. As shown in Table 4, tumorsderived from the ABCG2+ cancer cells all contained only afraction of ABCG2+ cells (Table 4), suggesting that ABCG2+ cellsgenerated ABCG2� tumor cells in vivo . Except for one tumorderived from the 1,000 ABCG2+ Du145 cells, we detected onlysmall percentages of ABCG2+ cells in all other tumors derivedfrom the ABCG2+ tumor cells (Table 4). On the other hand,tumors derived from the ABCG2� cells also contained a smallfraction of ABCG2+ cells (Table 4), suggesting that someABCG2� cells also have the ability to generate ABCG2+ tumorcells.Next, we did a series of clonal analyses to determine the

relationship between ABCG2+ and ABCG2� cancer cells. Althoughfreshly purified ABCG2+ Du145 (Fig. 5A ; 3 hours) or MDA-MB435(Fig. 5B ; 3 hours) cells proliferated faster than their correspondingABCG2� counterparts, culture for as short as 1 day eliminated orreduced this proliferative difference. Continued cultures of thesecells revealed increasing proliferating (i.e., BrdU+) cells in theABCG2� populations (Fig. 5A-B). These results suggest thepossibility that, with time in culture, the ABCG2+ tumorprogenitors gradually lose their proliferative capacity whereasprimitive CSCs in the ABCG2� population give rise to highlyproliferative ABCG2+ tumor progenitors. In support, immunostain-ing revealed the emergence of ABCG2+ cells from the startingABCG2� cancer cells within 1 week (data not shown), consistentwith the ability of some ABCG2� tumor cells to generate ABCG2+

cells in vivo (Table 4).Consistent with the BrdU incorporation assays, clonal analyses

revealed that at earlier time points the ABCG2+ tumor cells hadhigher cloning efficiency; that is, more cells had the ability toestablish a clone (Fig. 5C-D). However, at later time points, theABCG2� tumor cells picked up and formed similar or higher(Fig. 5C-D) percentages of clones. Subsequently, we carried outdifferential clonal analyses in which purified ABCG2+ andABCG2� tumor cells were plated at clonal density and clonalsizes were quantified at a shorter and a longer time point. Asshown in Fig. 5E , 7 days after plating, more ABCG2+ Du145cells formed larger clones. However, at 14 days post plating,there were significantly more large clones derived from theABCG2� Du145 cells (Fig. 5G). The average clonal sizes (cells/clone) of the ABCG2+ versus ABCG2� Du145 cells were 101 and60 cells at 7 days versus 4,716 and 5,658 cells at 14 days,respectively (P < 0.01 in both cases, ANOVA). Similarly, more

Table 2. Side population is enriched with tumorigenic cells

Cell type Cell injected (n) Tumor incidence (latency)

U373-SP-lj* 1,000 1/1 (33 d)

U373-NSP-1j 50,000 0/1 (terminated in 7 mo)

U373-SP-2jc 100 7/11 (34-48 d)1,000 1/1 (39 d)

U373-NSP-2j 50,000 0/1 (terminated in 7 mo)

200,000 0/1 (terminated in 7 mo)MCF7-SP* 1,000 1/6 (160 d)

10,000 3/6 (130-150 d)

MCF7-NSP 10,000 1/6 (194 d)

250,000 1/2 (190 d)LAPC9-SPc 100 2/8 (52 and 108 d)

1,000 3/4 (108 d)

1,500 2/2 (41 and 72 d)

LAPC9-NSP 150,000 0/1 (terminated in 9 mo)300,000 1/1 (92 d)

NOTE: Tumor incidence refers to the number of tumors/the number

of injections. Latency refers to the time from tumor cell injection tothe appearance of a palpable tumor.

*Cultured cells: 1j and 2j refer to the first and second-generation

tumors, respectively.

cXenograft tumor-derived cells.





ABCG2+ MDA-MB435 cells formed larger clones than theABCG2� MDA-MB435 cells at 7 days after plating (Fig. 5F).However, by 11 days after plating, there were significantly morelarge clones (i.e., >500 cells per clone) derived from the ABCG2�

MDA-MB435 cells (Fig. 5H). The average clonal sizes of theABCG2+ versus ABCG2� MDA-MB435 cells were 30 and 16 cellsat 7 days (P < 0.01) versus 346 and 356 cells at 11 days,respectively. Together, the proliferation (Fig. 5A-B) and clonal(Fig. 5C-H) analyses provide strong evidence that the ABCG2+

tumor cells likely represent fast-proliferating tumor progenitorcells, whereas the ABCG2� population contains slow-cycling,primitive CSC cells that, with time, could establish robust clonalgrowth and also generate fast-cycling progenitor cells.Finally, we did RT-PCR analysis to assess the mRNA expression

of several stemness genes (Fig. 6A-B). Consistent with the conceptthat the ABCG2� population contains primitive, stem-like cancercells, we found that purified ABCG2� tumor cells expressed highermRNA levels of Notch-1, h-catenin, and SMO (Fig. 6A-B). Even Oct-4,a transcription factor essential for embryonic stem cell self-renewal

(51) and recently shown to be expressed in some adult stem cells(52), also showed preferential expression in the ABCG2� cells inthree of four cell types (Fig. 6A). Interestingly, the Notch-1 mRNAwas detected nearly exclusively in the ABCG2� tumor cells (Fig. 6A).Similarly, the h-catenin mRNA was observed only in the ABCG2�

Du145 cells (Fig. 6B). These RT-PCR results provide strong supportfor the existence of primitive CSC in the ABCG2� tumor cellpopulation.

Discussion

A long-time puzzle to tumor biologists is the observations thateven with long-term cultured cancer cells, in general sufficientnumbers of cells have to be injected to initiate an orthotopictumor in recipient animals (reviewed in refs. 1, 47), suggestingthat even in long-term tumor cell cultures, not all cells are equaland only a small population of cells is tumorigenic. Indeed, whenmultiple human cancer cells, which have been in culture underdifferent conditions for years or even decades are assessed for

Figure 3. ABCG2 expression in human cancer cells. Cultured glioma (A), or prostate (B), breast (C ), colon (D ), or other (E ; see Table 1) cancer cells were stained forABCG2 using a monoclonal antibody. Original magnifications, �100.

Cancer Research




their clonal growth and clonogenic potentials, we find that onlya small percentage of cells possesses such potentials.3 Further-more, when tumor cell–derived spheres or xenograft tumorsin situ are pulsed with BrdU followed by extended chase, only avery minor population of the cells manifests as the long-termLRCs,3 which are known to preferentially identify stem cells (2, 5).In further support, long-term cultured rat C6 glioma cells (48)and some human brain tumor (i.e., glioma and medulloblastoma;

ref. 37) cells are found to contain a side population, which isknown to be enriched in stem cells (see Introduction).Importantly, the C6 side population cells are more tumorigenicthan and can also generate the non–side population cells (48),providing the first direct evidence for a population of moretumorigenic cells in long-term tumor cell cultures. It is theseobservations (37, 48)3 that have prompted us to first determinewhether the side population technique can be generally applied

Figure 4. Association of ABCG2 expression with cell division. B-C and E-F, cells that have just undergone nuclear division but have not undergone cytokinesis(arrows ). By contrast, cells in (G-I, J-L , and M-O ) have just undergone cytokinesis. Cells in (D-O ) were clonal cultures. Original magnifications, �400.





to other cultured human tumor cells, in particular, the humanepithelial cancer cells.Our results reveal that f30% cultured human cancer cell lines

and xenograft tumor-derived cells possess a side population thatcan be reliably detected under the current experimentalconditions. Several notable points are worthy of mention. First,in most literature reports, the side population is identified oneither a MoFlo or FACS Vantage flow cytometer as a continuoustail of the non–side population. Therefore, the discrimination ofthe side population from the non–side population is, to a certainextent, arbitrary and can vary significantly from experiment to

experiment. In contrast, using a Coulter Epics flow cytometerand our modified protocol, we have in most cases identified theside population as a distinct ‘‘side’’ population separate from themain non–side population. This could potentially give usrelatively pure side population cells. Second, in support of thispossibility, our system seems to identify the side population cellsin a more stringent manner and the percentages of the sidepopulation cells may be more representative of putative CSCs inthe cultures or xenograft tumors. Therefore, only f30% culturedhuman cancer cell lines and xenograft tumor-derived cellspossess a side population of 0.04% to 0.2%. These percentages

Figure 5. Proliferative and self-renewal properties of ABCG2+

and ABCG2- tumor cells. Purified ABCG2+ and ABCG2- Du145 (A)or MDA-MB435 (B ) cells were plated for either 3 hours or culturedfor the time periods indicated. Cells were pulsed with BrdUrd for3 hours before the end of each time point. Columns, mean %BrdUrd+ cells from two experiments with 500 to 1,200 total cellscounted; bars, FSE. *, P < 0.001 (t test). , P < 0.01 (A) or <0.05(B). Purified ABCG2+ and ABCG2- Du145 (C ) or MDA-MB435(D) cells were plated at clone density (400 cells per well) andcultured for the times indicated. At the end, the numbers of cloneswere quantified. % Cloning efficiency. *, P < 0.01. Purified ABCG2+

and ABCG2- Du145 (E and G ) or MDA-MB435 (F and H) cellswere plated at clone density (100 cells per well) and cultured for7 (E and F ), 14 (G), or 11 (H ) days. At the end, the numbers ofcells in each clone were counted and clones were groupedaccording to their sizes. On average 100 to 200 clones wererandomly counted for each condition. Representative of twoindependent experiments. *, P < 0.05.

Cancer Research




are similar to the side population of multiple normal stem cellor progenitor cell populations [i.e., 0.01-5%; refs. 18, 23, 26; e.g.,mouse bone marrow (0.01%; Fig. 1), newborn mouse keratinocyteprogenitors (0.5%; Fig. 1), and human bone marrow (f0.03%;ref. 8)]. The majority of the cancer cell lines or xenograft tumorsexamined seems to possess too small a side population to bereliably detected.Importantly, the side population cells purified from two cell

lines and one xenograft tumor are more tumorigenic thanthe non–side population counterparts. Furthermore, the sidepopulation cells are found to possess several intrinsic propertiesof stem cells: self-renewal, preferential expression of somestemness genes, and an ability to give rise to non–side populationcells. These results thus extend the others studies on rat C6glioma cells (48) and support the concept that the sidepopulation is indeed enriched in stem-like tumorigenic cells. Itshould be noted that with higher numbers of the non–sidepopulation epithelial cancer cells (i.e., MCF7 and LAPC-9)injected, we also observed tumor development. These resultsmay suggest that the non–side population also contains a verysmall percentage of tumorigenic cells, although the results mighthave stemmed from the contamination of small numbers of the

side population cells in the non–side population as epithelialcancer cells are very ‘‘sticky’’.The major focus of the current study is to address whether the

higher tumorigenicity associated with the side population cells isrelated to ABCG2 expression. Because the side populationphenotype in some cell types is mainly mediated by ABCG2, theside population cells (e.g., in MCF7) express higher levels ofABCG2 mRNA, ABCG2 is expressed only in a small subset ofcancer cells, and ABCG2 expression in cancer cells seemsassociated with cell proliferation, it stands to reason that theABCG2+ cancer cells might or should be more tumorigenic thanthe ABCG2� tumor cells. Surprisingly, however, highly purifiedABCG2+ cells from several types of tumor cells are not moretumorigenic than the corresponding ABCG2� cells. Furtherproliferation assays, clonal analyses, self-renewal, and molecularstudies suggest a model in which the ABCG2� populationcontains primitive stem-like cancer cells with higher self-renewal(because of higher levels of stemness genes) and proliferativepotentials but are normally slow cycling (Fig. 6C). These cells thengive rise to ABCG2+ tumor progenitor cells that are more activelyproliferating but possess reduced self-renewal and long-termproliferative capacities (Fig. 6). The ABCG2+ tumor progenitorcells eventually give rise to ABCG2�, partially or even fullydifferentiated tumor cells that constitute the bulk of tumor cellmass (Fig. 6).The side population has been shown to be very heterogeneous

(10, 27, 28). Therefore, the side population detected in cancer cellsmight contain several subsets of cells, one of which expressesABCG2 thus explaining the increased ABCG2 expression in the sidepopulation. The higher tumorigenicity associated with the sidepopulation might be conferred by combined effects of several othersubpopulations of cells in addition to the ABCG2+ cells. Forexample, cells expressing other ABC family members might alsocontribute to the cancer cell side population phenotype. Indeed, ithas been shown that only a fraction of side population cellsexpresses ABCG2 (24) and that both side population and knownstem/progenitor cells also express other ABC transporters such asMDR-1 (i.e., ABCB1 or P-glycoprotein), MRP-1 (ABCC1), and

Table 3. ABCG2 and tumorigenesis

Samples Cell

injected (n)

Incidence Latency

(median), d

U373-unsorted 1,000 3/3 26

10,000 4/4 19U373-ABCG2+ 1,000 8/8 30

10,000 2/2 26

U373-ABCG2� 1,000 4/4 1910,000 3/4 19

100,000 4/4 19

Du145-unsorted 100 3/5 77

1,000 4/6 5310,000 5/6 45

100,000 4/4 35

Du145-ABCG2+ 100 3/8 81

1,000 2/6 80Du145-ABCG2� 100 2/6 73

1,000 1/6 63

10,000 2/6 55

Du145-ABCG2+* 100 2/6 1011,000 1/2 63

Du145-ABCG2� 100 2/6 78

1,000 2/6 5410,000 1/4 54

MDA-MB435-unsorted 100 4/6 67

1,000 4/6 64

10,000 2/4 61MDA-MB435-ABCG2+ 100 7/8 57

1,000 3/4 47

MDA-MB435-ABCG2� 100 6/8 57

1,000 3/4 5710,000 5/6 52

*Cells in this experiment were purified from the Du145 xenograft

tumors. Cells in all other experiments were purified from culturedcells.

Table 4. ABCG2+ and ABCG2� cells in tumors

Tumors derived from* ABCG2+ cells (%) n

Du145

ABCG2+ (100 cells) 0.2, 0.2 2ABCG2+ (1,000 cells) 0.9, 37.4 2

ABCG2� (100 cells) 0.1, 0.3, 5.1 3

ABCG2� (10,000 cells) 3.8, 1.1 2

MDA-MB435ABCG2+ (100 cells) 0.8, 0.6 2

ABCG2+ (1,000 cells) 0.2, 0.1 2

ABCG2� (100 cells) 0.8 1

ABCG2� (1,000 cells) 0.1 1ABCG2� (10,000 cells) 0.5, 5.7, 6.7 3

*Xenograft tumors derived from either unsorted or sorted (i.e.,

ABCG2+ or ABCG2�) cells were harvested to prepare single-cell

human tumor cell suspension (see Materials and Methods), which was

then used in ABCG2 staining and flow cytometry analysis.





ABCA2 (36, 37, 39). In addition, enforced expression of MDR-1 inmurine bone marrow cells is sufficient to expand the sidepopulation (40). We are currently using cultured cancer cells aswell as xenograft and primary human tumor samples to determinethe molecular basis of the higher tumorigenicity associated withthe side population (e.g., amplification of oncogenes and/ormutations of specific tumor suppressors), dissect differentsubpopulations of the side population with respect to theirtumorigenic potentials, and elucidate the interrelationship betweenthe side population and several other tumorigenic populations wehave identified.

Acknowledgments

Received 2/21/2005; revised 4/16/2005; accepted 4/28/2005.Grant support: NIH grants CA90297, AG023374, and P30 CA16672; NIEHS grant

ES07784; American Cancer Society grant RSG MGO-105961; Department of Defensegrant DAMD17-03-1-0137; Prostate Cancer Foundation; and M.D. Anderson CancerCenter (PCRP and IRG).

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

We thank C. Conti, T-J. Liu, M. Andreef, and C. Sawyers for providing cells; C. Carterfor assistance in preparing mouse bone marrow cells; the Histology Core for excellentassistance in tissue processing and immunohistochemistry; the Animal Facility Corefor help in tumor experiments; E. Richie for her insights; and members of the Tang labfor support and helpful discussion.

Figure 6. ABCG2- cancer cells express higherlevels of stemness genes. A-B, RT-PCR analysisof stemness genes using purified ABCG2+ (+)and ABCG2- (�) cancer cells. Negative controls(Neg. CTL) refer to RT using H2O and positivecontrol (Pos. CTL ) in (H ) refers to unsorted Du145cells previously shown to express h-catenin.3

C, a hypothetical model. See text for morediscussion.

References

1. Passegue E, Jamieson CH, Ailles LE, Weissman IL.Normal and leukemic hematopoiesis: are leukemias astem cell disorder or a reacquisition of stem cellcharacteristics? Proc Natl Acad Sci U S A 2003;100:11842–9.

2. Fuchs E, Tumbar T, Guasch G. Socializing with theneighbors: stem cells and their niche. Cell 2004;116:769–78.

3. Raff M. Adult stem cell plasticity: fact or artifact?Annu Rev Cell Dev Biol 2003;19:1–22.

4. Wagers AJ, Weissman IL. Plasticity of adult stem cells.Cell 2004;116:639–48.

5. Tumbar T, Guasch G, Greco V, et al. Defining theepithelial stem cell niche in skin. Science 2004;303:359–63.

6. Goodell MA, Rosenzweig M, Kim H, et al. Dye effluxstudies suggest that hematopoietic stem cells expressinglow or undetectable levels of CD34 antigen exist inmultiple species. Nat Med 1997;3:1337–45.

7. Storms RW, Goodell MA, Fisher A, Mulligan RC, SmithC. Hoechst dye efflux reveals a novel CD7(+)CD34(�)lymphoid progenitor in human umbilical cord blood.Blood 2000;96:2125–33.

8. Jackson KA, Majka SM, Wang H, et al. Regeneration ofischemic cardiac muscle and vascular endothelium byadult stem cells. J Clin Invest 2001;107:1395–402.

9. Asakura A, Seale P, Girgis-Gabardo A, Rudnicki MA.Myogenic specification of side population cells inskeletal muscle. J Cell Biol 2002;159:123–34.

10. Montanaro F, Liadaki K, Schienda J, Flint A, GussoniE, Kunkel LM. Demystifying SP cell purification: viability,yield, and phenotype are defined by isolation parame-ters. Exp Cell Res 2004;298:144–54.

11. Welm BE, Tepera SB, Venezia T, Graubert TA, RosenJM, Goodell MA. Sca-1(pos) cells in the mousemammary gland represent an enriched progenitor cellpopulation. Dev Biol 2002;245:42–56.

12. Clayton H, Titley I, Vivanco M. Growth and differen-tiation of progenitor/stem cells derived from the humanmammary gland. Exp Cell Res 2004;297:444–60.

13. Summer R, Kotton DN, Sun X, Ma B, FitzsimmonsK, Fine A. Side population cells and Bcrp1 expressionin lung. Am J Physiol Lung Cell Mol Physiol 2003;285:L97–104.

14. Giangreco A, Shen H, Reynolds SD, Stripp BR.Molecular phenotype of airway side population cells.Am J Physiol Lung Cell Mol Physiol 2004;286:L624–30.

15. Summer R, Kotton DN, Sun X, Fitzsimmons K, FineA. Translational physiology: origin and phenotype oflung side population cells. Am J Physiol Lung Cell MolPhysiol 2004;287:L477–83.

16. Shimano K, Satake M, Okaya A, et al. Hepatic ovalcells have the side population phenotype defined by

expression of ATP-binding cassette transporter ABCG2/BCRP1. Am J Pathol 2003;163:3–9.

17. Terunuma A, Jackson KL, Kapoor V, Telford WG,Vogel JC. Side population keratinocytes resembling bonemarrow side population stem cells are distinct fromlabel-retaining keratinocyte stem cells. J Invest Derma-tol 2003;121:1095–103.

18. Triel C, Vestergaard ME, Bolund L, Jensen TG, JensenUB. Side population cells in human and mouseepidermis lack stem cell characteristics. Exp Cell Res2004;295:79–90.

19. Kim M, Morshead CM. Distinct populations offorebrain neural stem and progenitor cells can beisolated using side-population analysis. J Neurosci2003;23:10703–9.

20. Lassalle B, Bastos H, Louis JP, et al. ‘Side population’cells in adult mouse testis express Bcrp1 gene and areenriched in spermatogonia and germinal stem cells.Development 2004;131:479–87.

21. Falciatori I, Borsellino G, Haliassos N, et al.Identification and enrichment of spermatogonial stemcells displaying side-population phenotype in immaturemouse testis. FASEB J 2004;18:376–8.

22. Martin CM, Meeson AP, Robertson SM, et al.Persistent expression of the ATP-binding cassettetransporter, Abcg2, identifies cardiac SP cells in thedeveloping and adult heart. Dev Biol 2004;265:262–75.

Cancer Research




23. Iwatani H, Ito T, Imai E, et al. Hematopoietic andnonhematopoietic potentials of Hoechst(low)/side pop-ulation cells isolated from adult rat kidney. Kidney Int2004;65:1604–14.

24. Watanabe K, Nishida K, Yamato M, et al. Humanlimbal epithelium contains side population cellsexpressing the ATP-binding cassette transporterABCG2. FEBS Lett 2004;565:6–10.

25. Bhatt RL, Brown MD, Hart CA, et al. Novel methodsfor the isolation and characterization of the putativeprostatic stem cell. Cytometry 2003;54A:89–99.

26. Tamaki T, Akatsuka A, Okada Y, Matsuzaki Y, OkanoH, Kimura M. Growth and differentiation potential ofmain- and side-population cells derived from murineskeletal muscle. Exp Cell Res 2003;291:83–90.

27. Nadin BM, Goodell MA, Hirschi KK. Phenotype andhematopoietic potential of side population cells through-out embryonic development. Blood 2003;102:2436–43.

28. Uchida N, Dykstra B, Lyons K, Leung F, Kristiansen M,Eaves C. ABC transporter activities of murine hemato-poietic stem cells vary according to their developmentaland activation status. Blood 2004;103:4487–95.

29. Kawada H, Fujita J, Kinjo K, et al. Non-hematopoieticmesenchymal stem cells can be mobilized and differen-tiate into cardiomyocytes after myocardial infarction.Blood 2004;104:3581–7.

30. Chen CZ, Li L, Li M, Lodish HF. The endoglin(positive) sca-1(positive) rhodamine(low) phenotypedefines a near-homogeneous population of long-termrepopulating hematopoietic stem cells. Immunity 2003;19:525–33.

31. Pearce DJ, Ridler CM, Simpson C, Bonnet D.Multiparameter analysis of murine bone marrow sidepopulation cells. Blood 2004;103:2541–6.

32. Kubota H, Avarbock MR, Brinster R. Spermatogonialstem cells share some, but not all, phenotypic andfunctional characteristics with other stem cells. ProcNatl Acad Sci U S A 2003;100:6487–92.

33. Zhou S, Schuetz JD, Bunting KD, et al. The ABCtransporter Brcp1/ABCG2 is expressed in a wide varietyof stem cells and is a molecular determinant of the side-population phenotype. Nat Med 2001;7:1028–34.

34. Doyle LA, Ross DD. Multidrug resistance mediated bythe breast cancer resistance protein BCRP (ABCG2).Oncogene 2003;22:7340–58.

35. Zhou S, Morris JJ, Barnes Y, Lan L, Schuetz JD,Sorrentino BP. Bcrp1 gene expression is required fornormal numbers of side population stem cells in mice,and confers relative protection to mitoxantrone inhematopoietic cells in vivo . Proc Natl Acad Sci U S A2002;99:12339–44.

36. Lechner A, Leech CA, Abraham EJ, Nolan AL,Habener JF. Nestin-positive progenitor cells derivedfrom adult human pancreatic islets of Langerhanscontain side population (SP) cells defined by expressionof the ABCG2 (BCRP1) ATP-binding cassette transporter.Biochem Biophys Res Commun 2002;293:670–4.

37. Hirschmann-Jax C, Foster AE, Wulf GG, et al. Adistinct ‘‘side population’’ of cells with high drug effluxcapacity in human tumor cells. Proc Natl Acad Sci U S A2004;101:14228–33.

38. Cai J, Cheng A, Luo Y, et al. Membrane properties ofrat embryonic multipotent neural stem cells. J Neuro-chem 2004;88:212–26.

39. Benchaouir R, Rameau P, Decraene C, et al. Evidencefor a resident subset of cells with SP phenotype in theC2C12 myogenic line: a tool to explore muscle stem cellbiology. Exp Cell Res 2004;294:254–68.

40. Bunting KD, Zhou S, Lu T, Sorrentino BP. EnforcedP-glycoprotein pump function in murine bone marrowcells results in expansion of side population stem cellsin vitro and repopulating cells in vivo . Blood 2000;96:902–9.

41. Bonnet D, Dick JE. Human acute myeloid leukemia isorganized as a hierarchy that originates from a primitivehematopoietic cell. Nat Med 1997;3:730–7.

42. Al-Hajj M, Wicha MS, Benito-Hernandez A, MorrisonSJ, Clarke MF. Prospective identification of tumorigenicbreast cancer cells. Proc Natl Acad Sci U S A 2003;100:3983–8.

43. Singh SK, Clarke ID, Terasaki M, et al. Identificationof a cancer stem cell in human brain tumors. Cancer Res2003;63:5821–8.

44. Hemmati HD, Nakano I, Lazareff JA, et al. Cancerousstem cells can arise from pediatric brain tumors. ProcNatl Acad Sci U S A 2003;100:15178–83.

45. Galli R, Galli R, Binda E, et al. Isolation andcharacterization of tumorigenic, stem-like neuralprecursors from human glioma. Cancer Res 2004;64:7011–21.

46. Singh SK, Hawkins C, Clarke ID, et al. Identificationof human brain tumour initiating cells. Nature2004;432:396–401.

47. Reya T, Morrison SJ, Clarke MF, Weissman IL. Stemcells, cancer, and cancer stem cells. Nature 2001;414:105–11.

48. Kondo T, Setoguchi T, Taga T. Persistence of a smallpopulation of cancer stem-like cells in the C6 rat gliomacell line. Proc Natl Acad Sci U S A 2004;101:781–6.

49. Klein KA, Reiter RE, Redula J, et al. Progression ofmetastatic human prostate cancer to androgen inde-pendence in immunodeficient SCID mice. Nat Med1997;3:402–8.

50. Bhatia B, Tang S, Yang P, et al. Cell-autonomousinduction of functional tumor suppressor 15-lipoxyge-nase 2 (15-LOX2) contributes to replicative senescenceof human prostate progenitor cells. Oncogene 2005;24:3583–95.

51. Chambers I, Smith A. Self-renewal of teratocarcinomaand embryonic stem cells. Oncogene 2004;23:7150–60.

52. Tai MH, Chang CC, Olson LK, Trosko JE. Oct4expression in adult human stem cells: evidence insupport of the stem cell theory of carcinogenesis.Carcinogenesis 2005;26:495–502.





2005;65:6207-6219. Cancer Res Lubna Patrawala, Tammy Calhoun, Robin Schneider-Broussard, et al. Are Similarly Tumorigenic

Cancer Cells− and ABCG2+Cancer Cells, whereas ABCG2Side Population Is Enriched in Tumorigenic, Stem-Like

Updated version

http://cancerres.aacrjournals.org/content/65/14/6207

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/65/14/6207.full#ref-list-1

This article cites 52 articles, 19 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/65/14/6207.full#related-urls

This article has been cited by 73 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/65/14/6207To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/content/65/14/6207.full#ref-list-1

http://cancerres.aacrjournals.org/content/65/14/6207.full#related-urls

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

mailto:[email protected]



Documents

Side Population Is Enriched in Tumorigenic, Stem-Like ... · Many adult stem cells also seem to have the ability to trans-differentiate into other cell types, although this phenomenon