highly tumorigenic lung cancer cd133+ cells display stem-like features and are spared by cisplatin...

Highly tumorigenic lung cancer CD133 cellsdisplay stem-like features and are sparedby cisplatin treatmentGiulia Bertolinia,1, Luca Roza,1, Paola Peregob, Monica Tortoretob, Enrico Fontanellac, Laura Gattib, Graziella Pratesib,Alessandra Fabbrid, Francesca Andriania, Stella Tinellib, Elena Roze, Roberto Caserinia, Salvatore Lo Vullof,Tiziana Camerinif, Luigi Marianif, Domenico Deliac, Elisa Calabro`g, Ugo Pastorinog, and Gabriella Sozzia,2

aMolecular Cytogenetics Unit, bPreclinical Chemotherapy and Pharmacology Unit, cCell Cycle Control Unit, Department of Experimental Oncology,dDepartment of Pathology, fUnit of Medical Statistics and Biometry, and gUnit of Thoracic Surgery, Fondazione Istituto di Ricovero e Cura a CarattereScientifico, Istituto Nazionale dei Tumori, 20133 Milan, Italy; and ePathology Unit, Casa di Cura La Maddalena, 90136 Palermo, Italy

Edited by Carlo M. Croce, The Ohio State University, Columbus, Ohio, and approved August 10, 2009 (received for review May 29, 2009)

The identification of lung tumor-initiating cells and associatedmarkers may be useful for optimization of therapeutic approachesand for predictive and prognostic information in lung cancerpatients. CD133, a surface glycoprotein linked to organ-specificstem cells, was described as a marker of cancer-initiating cells indifferent tumor types. Here, we report that a CD133, epithelial-specific antigen-positive (CD133ESA) population is increased inprimary nonsmall cell lung cancer (NSCLC) compared with normallung tissue and has higher tumorigenic potential in SCID mice andexpression of genes involved in stemness, adhesion, motility, anddrug efflux than the CD133 counterpart. Cisplatin treatment oflung cancer cells in vitro resulted in enrichment of CD133 fractionboth after acute cytotoxic exposure and in cells with stable cispla-tin-resistant phenotype. Subpopulations of CD133ABCG2 andCD133CXCR4 cells were spared by in vivo cisplatin treatment oflung tumor xenografts established from primary tumors. A ten-dency toward shorter progression-free survival was observed inCD133 NSCLC patients treated with platinum-containing regi-mens. Our results indicate that chemoresistant populations withhighly tumorigenic and stem-like features are present in lungtumors. The molecular features of these cells may provide therationale for more specific therapeutic targeting and the definitionof predictive factors in clinical management of this lethal disease.

ABC transporters cancer stem cells chemoresistance CXCR4 xenografts

Lung cancer is the leading cause of cancer deaths worldwidebecause of its high incidence and mortality, with 5-yearsurvival estimates10% for nonsmall cell lung cancer (NSCLC)(1). Refined investigation on the mechanisms of tumorigenesisand chemoresistance of lung cancer is needed to improvesurvival rate.Recently, the cancer stem cell (CSC) theory has been pro-

posed to explain the tumor heterogeneity and the carcinogenesisprocess (2, 3). According to this model, tumor can be viewed asa result of abnormal organogenesis driven by CSCs, defined asself-renewing tumor cells able to initiate and maintain the tumorand to produce the heterogeneous lineages of cancer cells thatcompose the tumor (4, 5). The existence of CSCs was first provedin acute myeloid leukemia (6), and more recently in glioblastoma(79), melanoma (10, 11), and epithelial cancers (1218).CSCs were identified by using flow cytometry-based cell

sorting of tissue-specific surface markers or sphere-formingassay in selective serum-free medium. CD133 (prominin-1), afive-transmembrane glycoprotein, was initially described as amarker specific for CD34 human hematopoietic progenitorcells (19, 20), normal stem cells of the neural (21, 22), epithelial(23, 24), and endothelial lineages (25), and their tumoral coun-terparts (1416, 26, 27). However, it is still a matter of debate

whether CD133 cells truly represent the ultimate tumorigenicpopulation, particularly in colon (28) and brain (29, 30) cancer.Long-term cultures of sphere-growing cells derived from

human lung tumors were shown to be highly enriched for CD133expression, able to self-renew, and able to be the only tumori-genic population in vivo (18). A recent report further demon-strated that Oct-4 expression plays a critical role in maintenanceof stem-like properties in lung cancer CD133 cells (31).CSCs may be inherently resistant to the cytotoxic effect of

chemotherapy because of their low proliferation rate and resis-tance mechanisms, such as the expression of multidrug trans-porters of the ATP-binding cassette (ABC) superfamily. ABCB5was found to be expressed on a distinct subset of chemoresistantCD133 melanoma cells (32), and its selective targeting causedtumor growth inhibition in xenograft models (11). Expression ofABCG2 (BCRP1), a transporter involved in resistance to mul-tiple drugs (33), was reported in normal lung undifferentiatedcells (34) and lung cancer cell lines (35).By using both in vitro systems and implemented in vivo models

of direct xenografts of human primary lung cancers in mice, weprovide evidence that lung tumor CD133 cells are highlytumorigenic, are endowed with stem-like features, and, impor-tantly, are spared by cisplatin treatment.

ResultsIdentification of CD133 Cells in Lung Cancer. In an attempt toidentify lung cancer-initiating cells, we analyzed 60 primarytissue samples derived from a consecutive series of lung cancerpatients (Table S1). By using the surface marker CD133 aloneor performing double staining with the epithelial-specific anti-gen (ESA) marker to exclude potential contamination by he-matopoietic and endothelial progenitors, we noted a very similarfrequency of CD133 and CD133ESA populations in all casesexcept one (LT23; Table S1), indicating that most CD133 cellswere of epithelial origin.We demonstrated the presence of a rare(mean, 1%) CD133ESA population in normal lung tissuesof the patients, whereas tumor samples consistently showed anoticeable (mean, 5%; P 0.0004) CD133ESA population

Author contributions: G.B., L.R., P.P., L.G., G.P., F.A., U.P., and G.S. designed research; G.B.,M.T., E.F., L.G., A.F., S.T., E.R., R.C., S.L.V., T.C., L.M., and E.C. performed research; E.C. andU.P. contributednewreagents/analytic tools; G.B., L.R., P.P., L.G., G.P., F.A., S.L.V., T.C., L.M.,D.D., and G.S. analyzed data; and G.B., L.R., and G.S. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

Freely available online through the PNAS open access option.

1G.B. and L.R. contributed equally to this work.

2To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/cgi/content/full/0905653106/DCSupplemental.

www.pnas.orgcgidoi10.1073pnas.0905653106 PNAS September 22, 2009 vol. 106 no. 38 1628116286

CELL

BIOLO

GY

(Fig. S1). FACS analysis showed the presence of a variablefraction of CD133ESA cell populations in 47 of 56 (83.9%)tumor samples, varying from 0.02% to a maximum of 35%.However, 60% of the cases showed small amounts (2%) ofCD133 cells. Further investigation of CD133 expression byimmunohistochemistry (IHC) on paraffin-embedded tumor sec-tions confirmed the existence of CD133 cells and identified 32of 58 (55.2%) positive cases (Fig. 1 and Table S1). Positivity oftumor cells was defined as membranous staining or staining ofmembrane and cytoplasm, and the intensity was always specificand strong. Because the immunoreactivity was heterogeneous,no cutoff was applied, and only cases with no CD133 immuno-reactivity were scored as negative. IHC generally confirmed theresults obtained with FACS analysis; however, FACS was moresensitive in detecting positive cases with a very small percentageof CD133 cells. Indeed, by categorizing FACS results into fourclasses (negative or positive, with 2% and 5% positivity cutoffs),the frequency of positive samples at IHC gradually increasedfrom 11% to 92% (test for trend P 0.0005). When data wereanalyzed in terms of discrimination between negative and pos-itive IHC results on the basis of FACS classes, the areas underthe receiver-operating characteristic curve (nonparametric esti-mate) were 0.764 (P 0.0001; or 0.761, P 0.0001, by usingFACS score for computation as a continuous variable), valuebetween 1 (perfect discrimination) and 0.5 (lack of discrimina-tion). Cross-tabulation of CD133 expression data using IHCstaining and clinicopathological features indicated that low-grade (G1G2) tumors (P 0.0356) and adenocarcinomahistology (P 0.0441) were more frequently represented amongCD133-positive cases (Table S2). Follow-up information was toolimited (median follow-up duration, 14 months; eight recordedtumor progressions and two tumor deaths) for investigation ofCD133 expression prognostic value.

Establishment of Xenograft Models of Lung Primary Tumors. Todevelop novel preclinical models, we successfully grew 10 xeno-

grafts in nude mice starting from 29 human lung primary tumors.These xenografts were serially maintained in vivo as tumor linesthat represented a continuous source of tumor tissue for subse-quent functional studies. IHC for CD133 and for a panel of lungcancer antigens showed high similarity between parental tumorsand mouse xenografts (Fig. 2A and Fig. S2) and a similar fractionof CD133ESA cells by FACS (Fig. 2B) that was maintainedover time in serial transplantations. No association was observedbetween the levels of CD133 expression in the original primarytumor and the ability to grow and propagate xenografts in mice.

Tumorigenic in Vivo Potential of Isolated CD133 Cells. Because thedefinition of cancer-initiating cells relies mainly on functionalproperties, like the ability to sustain tumor growth recapitulatingthe original cellular heterogeneity, we investigated the tumori-genic ability of lung tumor-derived CD133 and CD133 cellpopulations.After depletion of mouse cells expressing the H2K-MHCI

antigen by immunomagnetic separation (when necessary),CD133 cells were isolated by FACS or magnetic cell sorting(Miltenyi Biotech) from four xenografts and one primary tumor.

Fig. 1. IHC analysis of CD133 expression in two representative positive (LT74and LT60) and one negative (LT66) adenocarcinoma lung tumor samples atlow and high magnifications. As controls (CTR) for staining specificity, anormal lung sample negative for CD133 (Bottom Left) and a control antibody-stained tumor sample (Bottom Right) are also shown.

Fig. 2. Tumor xenografts resemble the original primary tumor. (A) IHC oflow-molecular weight cytokeratins (CK-LMW), cytokeratin 7 (CK7), surfactantprotein C (SP-C), transcription thyroid factor 1 (TTF-1), CD133, and MIB-1performed in LT45 parental tumor, corresponding xenograft, and xenografttumor derived from injection of CD133 cells. (B) FACS analysis ofCD133ESAcells in xenografts compared with parental tumors.

16282 www.pnas.orgcgidoi10.1073pnas.0905653106 Bertolini et al.

Considerable enrichment of CD133 cells was observed in thepositive fraction (80% purity), as determined by FACS (Fig.S3A). The same numbers of CD133, CD133, and unsortedcells mixed withMatrigel were injected in the flank of SCIDmiceby using limiting doses derived from preliminary experiments(104 down to 102). Palpable tumors were visible after a variabletime interval in the majority of CD133-injected mice, whereasno tumor growth was observed in CD133-injected mice (Fig.S3B and Table S3) except for LT73 (see below). Tumorigenicityof the unsorted population was variable for the different cases,and for the xenografts it appeared to be related to the percentageof CD133 cells in the bulk population; however, tumor forma-tion of enriched CD133 cells was faster and resulted in in-creased tumor take compared with that observed after injectionof CD133 or unsorted cells. Morphological and IHC analysesrevealed that tumors derived from CD133 cells faithfullyreproduced the original tumor (Fig. 2A). Similar results wereobserved after s.c. injection of FACS-isolated CD133 cells fromthe A549 lung adenocarcinoma cell line (Table S3). These datasuggest that the CD133 cell population is enriched in cellscapable of initiating lung cancer in SCID mice.To investigate whether lung cancers originating from CD133

cells possess higher long-term tumorigenic potential comparedwith the rarely observed tumors originating from the CD133fraction, we performed serial transplantation assays in SCIDmice of cells isolated from LT73 tumor xenografts originallyderived from CD133 or CD133 cell injection (Fig. S3C). Cellsderived from CD133 tumors were able to generate tumorsmaintaining the original morphology and proportion of CD133

cells in primary, secondary, and tertiary transplantation,whereas cells from CD133 tumors lost tumorigenic potentialduring serial transplantations (Fig. S3C). In principle, the gen-eration of tumors from CD133 cells could be caused bycontamination of the negative fraction by CD133 cells duringcell sorting. However, it is likely that for this specific highlyaggressive tumor the CD133 fraction also had some tumor-initiating potential at low dose. Here, we demonstrated thatthese CD133 cells could not maintain tumor growth in serialtransplantations, indicating a lower and fading tumorigenicability compared with CD133 cells.

Biological and Molecular Features of Human Lung Cancer-InitiatingCells. Down-regulation of CD133 expression was observed inshort-term adherent cultures (n 7) grown in serum-supplemented medium compared with the original cell suspen-sion (Fig. 3A).Normal and neoplastic stem-like cells from neural and epi-

thelial organs can be expanded as sphere-like aggregates inserum-free EGFbFGF-supplemented medium (9, 15, 36) thatfavors the proliferation of undifferentiated cells. However, es-tablishment of long-term sphere cultures from primary lungtumor samples was unsuccessful. Nevertheless from adherentLT73 and A549 cell lines we were able to consistently expandcells growing as floating spheres (A549/s and LT73/s) thatallowed a variety of cellular and molecular analyses.To verify the clonal origin of A549 and LT73 spheres, we

performed a label-retaining assay using vital red and greenfluorescent dyes (PKH26 and PKH67, as described in SI Text).

Fig. 3. In vitro characterization of CD133 cells. (A) FACS analysis for CD133 expression in freshly dissociated primary LT73 and corresponding adherent cellculture after four passages in vitro. (B) PKH26 staining of LT73 and A549 spheres. (Top) Single cells from dissociated spheres were separately labeledwith PKH26and PKH67 red and green fluorescent dyes, and cultures were harvested by mixing red and green labeled cells. (Middle and Bottom) A549 (Middle) and LT73(Bottom) spheres, derived frommixed cultures, showed a separate green or red fluorescent staining, with a decreasing gradient of fluorescence intensity withinthe spheres. Microscopic images of spheres derived from labeled cells were acquired at 20 magnification in bright field, with rhodamine filter for PKH26fluorescence (Left) and with fluorescein filter for PKH67 fluorescence (Right). (C) Time-course analysis of CD133 cells sorted from A549 spheres. FACS analysisof CD133 after sorting and at the first and second population doubling (PD). (D) Real-time PCR analysis of stemness gene expression in CD133 and CD133

fractions, sorted from LT28 xenograft (Left) and A549 spheres (Right). Unsorted cells were used as calibrator for the relative quantification of gene expression.(E) Real-time PCR analysis of transporters of ABCC-B-G families in CD133 and CD133 cells sorted from A549/s, and ABCC family in CD133 and CD133 cellssorted from LT73/s and LT45 xenograft. Unsorted spheres or LT45 unsorted cells were used as calibrators for the relative quantification of gene expression.

Bertolini et al. PNAS September 22, 2009 vol. 106 no. 38 16283

CELL

BIOLO

GY

Spheres derived from mixed cultures of PKH26-labeled andPKH67-labeled cells showed a distinct red or green fluorescentstaining, indicating that A549/s and LT73/s did not representstochastic cellular aggregates. Moreover, the decreasing gradientof fluorescence intensity of labeled cells within the spheresallowed us to evaluate a difference in the rate of cell divisionsbecause proliferating cells easily dilute the dyes, whereas slowlydividing quiescent cells retain most of the fluorescence (37).A549/s and LT73/s spheres showed a decreasing gradient of redor green fluorescence, indicating that cells within spheres havea clonal origin and are endowed with a differential proliferationrate (Fig. 3B). FACS analysis indicated that A549/s and LT73/sdisplayed an enrichment of CD133 cell fraction to 1%,whereas it was only 0.2% in the standard adherent culture. Atime-course FACS analysis of the sorted CD133 fractionshowed that CD133 cells remained quiescent for 2 weeks,whereas the CD133 fraction and the unsorted A549/s popula-tion showed a constant and sustained proliferative rate. Afterthis time CD133 cells started to proliferate, and CD133 ex-pression decreased along with cell divisions and returned to theoriginal fraction in 50 days (Fig. 3C). In contrast, the CD133population remained CD133 negative during the observationperiod.The expression of genes involved in stem cell pathways [i.e.,

Sonic hedgehog (SHH, Gli-1), Notch (Notch1, Notch2, Hes-1),CD133, Oct4/3, and Nanog] was investigated by real-time PCRin CD133 and CD133 populations sorted from A549/s andLT73/s and lung tumor xenografts (LT56, LT28, LT45). Anincreased expression level of genes involved in the maintenanceof stemness, markedly of Oct4/3 and Nanog, and adhesion andmotility genes, like -6 integrin and CXCR4, was consistentlyobserved in CD133 cells compared with the CD133 counter-part (Fig. 3D; detailed procedures described in SI Text).To better define the molecular features of cancer-initiating

cells with respect to cellular defense mechanisms, we analyzedthe mRNA levels of transporters of the ABC superfamily inCD133 and CD133 cell populations sorted from A549/s,LT73/s, and LT45 xenografts by using TaqMan Micro Fluidiccards (Applied Biosystems). By using this approach, we foundthat 70% of the 50 human ABC transporters were expressedat detectable levels (Fig. 3E). Compared with the CD133fraction, CD133 cells exhibited an increased expression ofdifferent components of the ABCC family that are known to beinvolved in the multidrug-resistant phenotype (i.e., ABCC1).The level of several ABCB family members and ABCG2 mRNAwas also enhanced. Overall, these results indicate that CD133cells are slowly dividing cells, able to give rise to a CD133 cellpopulation, thereby recapitulating the original cellular hetero-geneity, and CD133 cells express high levels of embryonic stem

cell genes, motility genes, and ABC transporters, suggesting thatCD133 cells possess the features of stem-like cancer cells.

In Vitro Resistance of CD133 Cells to Cisplatin. Because of their lowproliferation rate and the activation of defense mechanisms, thesmall population of CD133 lung cancer-initiating cells may thusbe inherently resistant to the cytotoxic effect of chemotherapy.We therefore assayed whether standard chemotherapy treat-ments result in enrichment of cancer-initiating cells in thesurviving cell fraction. In vitro studies were performed inadherent A549 cells that show a low fraction (0.2%) of CD133cells (detailed procedures described in SI Text). Exposure of cellsto a cytotoxic concentration of cisplatin, corresponding to theIC80, resulted in an 8-fold enrichment of CD133 cells (from0.2% 0.05% to 1.6% 0.5%; Fig. 4A).We then extended such an analysis to cells with acquired

resistance to cisplatin. By using the A549/Pt cells, generated bychronic exposure of A549 to increasing concentrations of cis-platin and endowed with a stable drug-resistant phenotype whengrown in the absence of selecting agent (degree of resistance10), we found that in vitro development of drug resistance wasassociated with increased expression of CD133. Indeed, a 13-foldenrichment in CD133 population (from 0.2% 0.05% to2.7% 0.87%) was consistently observed in A549/Pt cells grownas adherent culture (Fig. 4B).

Increased ABCG2 and CXCR4 Expression of Human Lung Cancer-Initiating Cells After in Vivo Cisplatin Treatment.To demonstrate thein vivo relevance of these findings, mice carrying six differentlung cancer xenografts were treated i.v. with cisplatin, accordingto a weekly schedule.Cisplatin treatment resulted in a variable tumor volume

inhibition (TVI) among the xenografts, ranging from 35% to83%, overall confirming the poor responsiveness to chemother-apy observed in the clinical setting of lung cancer (Table S4). Toverify whether cisplatin treatment was able to enrich the fractionof cancer-initiating cells in the residual tumors, at 7 days after thelast treatment and at the time of tumor regrowth, mice werekilled. FACS analysis of cells isolated from the resected tumorsshowed an enrichment of 7 and 35 times in the CD133 fractionin A549 and LT66 tumors, respectively, shortly after chemother-apy (day 7 after last treatment), which reverted to the originalvalues at the time of tumor regrowth (Table S4). In tumors withan original large content of CD133 cells, such as LT45 andLT56 (50% and 15%, respectively), the fraction of CD133 cellstaken as a whole did not change after cisplatin treatment, but asubpopulation of CD133ABCG2 cells showed a remarkableenrichment (Fig. 4C). A similar change in the CD133ABCG2fraction was also noticed in A549 and LT66 treated tumors

Fig. 4. CD133 cells survive cisplatin treatment. (A) FACS analysis of CD133 expression in A549 parental cell line and A549 treated for 1 h with cisplatin (IC80).CD133 expression increased in the A549 cell line 72 h after treatment. (B) FACS analysis of CD133 expression in stable cisplatin-resistant A549/Pt cells comparedwith the A549 parental cell line. (C) Enrichment of CD133ABCG2 (Upper) and CD133CXCR4 (Lower) populations in xenografts LT45 and LT56 after in vivocisplatin treatment. (D) CD133 expression in advancedNSCLCpatients treatedwith platinum-containing regimens. Progression-free survival curvewas calculatedwith the KaplanMeier method and compared by using the log-rank test.


(Table S4). Overall, our findings indicate that chemotherapy,even when reducing tumor burden, may spare CD133 tumorcells endowed with drug-resistance properties.Double staining with CD133/CXCR4 antibodies also revealed

a subpopulation of positive cells that increased after chemother-apy in 2 out of 3 tumors analyzed, and a persistent increase incells expressing CXCR4 was noticed in most treated tumors andin the A549-treated cell line (Fig. 4C and Table S4). Such resultssuggest the survival of CD133 chemoresistant clones alsoendowed with high mobility and metastatic features.Cisplatin-treated tumors also consistently expressed higher

mRNA levels of Nanog, Oct4/3, Notch, -6 integrin, and CXCR4genes than their corresponding untreated controls (Fig. S4A).The same was observed for ABC transporters (Fig. S4B). Thesefindings indicate that the CD133-chemoresistant fraction pref-erentially expresses genes important for the proliferation, self-renewal, homing, and drug resistance of stem cells.

CD133 Expression as a Marker of Chemotherapeutic Response inPatients. Finally, to investigate the possible relationship betweenCD133 status and response to platinum-containing regimens inthe clinical setting, we retrospectively analyzed CD133 proteinexpression by IHC in formalin-fixed biopsies obtained beforetreatment from 42 stage IIIB/IV NSCLC patients receivingcarboplatin and gemcitabine (Table S5). CD133 expression onthese archival biopsies was found in 10 (23.8%) of 42 patients.Partial response to platinum-based chemotherapy was ob-

served in 12 of 32 (37.5%) CD133 patients and 4 of 10 (40%)CD133 patients. Patients with CD133 tumors had a tendencytoward a shorter median progression-free survival than patientswith CD133 tumors (Fig. 4D); however, the difference did notreach statistical significance. Interestingly, 9 of 10 CD133patients relapsed within 9 months (the last patient at 20 months)compared with 22 of 32 (68.7%) relapses in the group of CD133patients, suggesting that NSCLC tumors carrying higher levels ofCD133 cells might be more likely to develop early tumorrecurrence after chemotherapy.

DiscussionThe identification of distinct phenotypic and functional markersassociated with stem-like properties of lung cancer tumor-initiating cells would be instrumental for developing therapeutictargeting strategies in lung cancer patients.Here, we report that a CD133/ESA population is increased

in primary NSCLC samples compared with normal lung tissue ina prospectively collected series of lung cancer patients. Flowcytometry and IHC analyses showed a quite good concordancefor the detection of the CD133 population in surgical tumorsamples, supporting the use of the latter technique for screeningparaffin-embedded samples and also small archival biopsies.CD133 cells isolated from primary lung tumors showed

higher tumorigenic potential than their CD133 counterpartsand were able to reproduce the original tumor heterogeneity inSCID mice. This phenotype was associated with an increasedexpression level of genes involved in stemness, adhesion, andmotility. An interesting finding was the enrichment for severalABC transporters in CD133 cells compared with the CD133fraction. Thus, CD133 cells possess features of cancer-initiatingstem-like cells, including ABCG2 expression and, moreover,they expressed additional ABC transporters that may contributeto determine a drug-resistant phenotype.Previous studies on the putative chemoresistant features of

lung cancer-initiating cells have approached this issue by usingonly a limited number of in vitro models of sphere-growing cellsisolated from tumors or established cell lines (18, 31). Our invitro experiments with cisplatin treatment also resulted in en-richment of CD133 cancer-initiating cells in the survivingfraction both after acute cytotoxic exposure and in an estab-

lished lung cancer cell line model of acquired stable cisplatinresistance. However, in vitro models may not perfectly mirrorwhat is observed in primary cancer cells, whereas in vivo modelsare more conducive to the evaluation of the functional propertiesof tumor-initiating cells, particularly in the assessment of treat-ment responses, because they best reflect the tumor heteroge-neity observed in patients and the interaction with the micro-environment. In novel preclinical models, i.e., xenografts fromprimary lung tumors of patients, serially maintained in vivo astumor lines, we show that cisplatin treatment resulted in remark-able enrichment of the CD133 cell fraction in the residualtumors, which then reverted to the original values at time oftumor regrowth. Interestingly, in tumors with a high content ofCD133 cells, only subpopulations of CD133ABCG2 andCD133CXCR4 cells displayed a clear enrichment after treat-ment, therefore suggesting that lung cancer CD133 cells com-prise populations of cells with a similar phenotype but differentpotential. It is noteworthy that if the subpopulation of tumor-initiating cells has chemoresistant features, conventional mea-sures of treatment efficacy (such as TVI) might only reflect howthe bulk tumor responds to chemotherapeutic treatment, failingto provide relevant information on long-lasting tumor-eradicating potential. In this view, additional and differentendpoints should be considered to evaluate innovative thera-peutic strategies.These chemoresistant cell fractions preferentially expressed

genes relevant for proliferation, self-renewal, differentiation,drug resistance, and homing of stem cells. Such results indicatethat chemotherapy is effective in eliminating drug-sensitiveCD133 differentiated tumor cells, whereas cancer-initiatingcells with the above described phenotypes are spared by thesetreatments and could be responsible for tumor restoration afterchemotherapy cessation. Accordingly, by using archived biopsieswe observed that CD133 protein expression tended to correlatewith early recurrence in a series of advanced-stage NSCLCpatients treated with platinum-containing regimens, althoughmore data are needed to draw definitive conclusions.Our in vivo findings showing coexpression of CD133/ABCG2

markers in the cisplatin-spared cell population are in agreementwith the increase in the side population fraction observed indrug-surviving cells of a H460 lung cancer cell line that can bespecifically depleted by using the ABCG2 inhibitor fumitremor-gin C (35). Our results suggest the interest of CD133/ABCG2expression in relation to treatment effect in patients with lungcancer and a rationale for a new generation of ABCG2 inhibitorsto be used in combination therapy. Indeed, the use of a specificantibody against ABCB5 transporter significantly inhibited tu-mor growth in treated melanoma xenografts (11).An interesting finding was the elevated expression of CXCR4

in CD133 cells compared with the CD133 fraction and aconsistent enrichment of CD133CXCR4 population aftercisplatin treatment of tumor xenografts. Our data are in line withthose reported in pancreatic adenocarcinoma (38), suggestingthe existence of a chemoresistant CD133CXCR4 subpopula-tion with highly tumorigenic and metastatic properties also inlung cancer. Considering that SDF-1, the specific ligand ofCXCR4, is strongly expressed in the lung, targeting with specificinhibitors of the SDF1CXCR4 axis could represent a thera-peutic option to eradicate the chemoresistant CSC population.The demonstration of the presence in lung tumors of highly

tumorigenic cells with stem-like properties and exhibiting fea-tures of chemoresistant cells may be useful, together with the useof carefully selected in vivo models, to evaluate novel pathwaysto be targeted to increase the therapeutic response in the clinicalsetting of this lethal disease.

Bertolini et al. PNAS September 22, 2009 vol. 106 no. 38 16285

CELL

BIOLO

GY

Materials and MethodsTumor and Normal Lung Tissue Dissociation and Cell Cultures. Clinical specimenswere obtained from a consecutive series of consenting patients according to theInternal Review and the Ethics Boards of the Istituto Nazionale Tumori ofMilan.

Single-cell suspensionswere obtained from lung tumor and normal tissues,collected from clinical surgical specimens, and mouse tumor xenografts aftermechanical and enzymatical disaggregation, as described in SI Text. Foradherent cell cultures, cells were plated in conventional medium, RPMI me-dium 1640 supplemented with 10% heat-inactivated FCS (all from Lonza). Toobtain sphere cultures, cells were plated at a density of 104 cells per milliliterin serum-free medium DMEM/F12 (Lonza), supplemented with commercialhormone mix B27 (Gibco Invitrogen), EGF (20 ng/mL; PeproTech), bFGF (10ng/mL; PeproTech), and heparin (2 g/mL). Floating sphere cultures wereexpanded by mechanical dissociation, followed by replating of single cells incomplete fresh medium every 3 days.

For experiments using tumor xenografts, mouse cells were depleted by usingthe Dynal cell-collection Biotin Binder kit (Invitrogen) as detailed in SI Text.

Flow Cytometry Analysis. Single-cell suspensions were washed and incubatedin staining solution containing 1% BSA and 2 mM EDTA with the specificantibodies at appropriate dilutions. For CD133 staining, 106 cells were incu-batedwith 10L of CD133/1-phycoerythrin antibody (50g/mL; AC133 clone;Miltenyi Biotech) diluted in 80 L of staining solution and 20 L of FcRblocking reagent (Miltenyi Biotech) for 10 min at 4 C. Antibodies used werephycoerythrin (PE)-conjugated anti-CD133/1, PE-conjugated anti-CD133/2,and FITC-conjugated anti CD326 (EpCAM), all from Miltenyi Biotech; PE-Cy5-conjugated anti-CXCR4 (Becton Dickinson); and FITC-conjugated anti-BCRP1(Chemicon).

Samples were acquired and analyzed by using a FACSCalibur andCELLQuest Pro software (Becton Dickinson). For time-course experiments,CD133 and CD133 fractions were plated in serum-free medium after sort-ing, and the percentage of CD133 cells was determined at each cell divisionby flow cytometry.

For the protocol of magnetic and cytofluorimetric CD133 cell separation,refer to SI Text.

IHC. IHC was performed on formalin-fixed, paraffin embedded samples.CD133 immunostaining was assessed on whole-tissue sections, and CD133immunoreactivity was evaluated inside the neoplastic epithelial component.All specimens were evaluated independently by two observers (E.R. and G.S.),and interobserver agreement was reached in all cases. Technical details ofimmunostaining are reported in SI Text.

In Vivo Studies of Tumorigenicity.All experimentswere carriedoutwith femaleCD-1 nude mice or SCID mice, 710 weeks old (Charles River Laboratories).Miceweremaintained in laminar flow rooms,with temperature and humidityconstant.Mice had free access to food andwater. Experimentswere approvedby the Ethics Committee for Animal Experimentation of the FondazioneIstituto di Ricovero e Cura a Carattere Scientifico Istituto Nazionale Tumori,according to institutional guidelines.

For establishment of xenograft models and tumor lines, regular smallfragments were obtained by patient surgical specimens as described (39).Fragments were implanted s.c. by trocar gauge in one or both flanks of nudemice. Tumor lines were achieved by serial s.c. passages of fragments fromgrowing tumors into healthy mice (39).

To assess the tumorigenic potential of different cell populations, aftersorting, viable 103 and 104 CD133, CD133, and unsorted cells were sus-pended in Matrigel (BD Biosciences) at a ratio of 1:1, and 200 L of cells wass.c. injected into the right flank of SCIDmice. For serial transplantation assays,103 CD133 and CD133 cells were injected s.c. in SCID mice, and derivedtumorswere dissociated to single cells and serially reinjected inmice at 102 cellnumbers, generating secondary and tertiary tumors.

The in vivo chemotherapy studies to evaluate response to cisplatin wereperformed as described in SI Text.

ACKNOWLEDGMENTS. We thank Dr. Giacomo Cossa for technical assistanceand Prof.MalcolmAlison for helpful discussions. F.A. is a fellowof FondazioneErmenegildo Zegna. This work was supported by the Associazione ItalianaRicercaCancro (toG.S. andU.P.), ItalianMinistry ofHealth (Ricerca Finalizzata)(to G.S.), Compagnia di San Paolo di Torino, and European Community Inte-grated Project 037665 CHEMORES.

1. Jemal A, et al. (2006) Cancer statistics, 2006. CA Cancer J Clin 56:106130.2. Pardal R, Clarke MF, Morrison SJ (2003) Applying the principles of stem-cell biology to

cancer. Nat Rev Cancer 3:895902.3. Reya T, Morrison SJ, ClarkeMF,Weissman IL (2001) Stem cells, cancer, and cancer stem

cells. Nature 414:105111.4. Clarke MF, et al. (2006) Cancer stem cellsperspectives on current status and future

directions: AACR Workshop on cancer stem cells. Cancer Res 66:93399344.5. Clarke MF (2005) Self-renewal and solid-tumor stem cells. Biol Blood Marrow Trans-

plant 11:1416.6. Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy

that originates from a primitive hematopoietic cell. Nat Med 3:730737.7. SinghSK, et al. (2003) Identificationofa cancer stemcell inhumanbrain tumors.Cancer

Res 63:58215828.8. Hemmati HD, et al. (2003) Cancerous stem cells can arise from pediatric brain tumors.

Proc Natl Acad Sci USA 100:1517815183.9. Galli R, et al. (2004) Isolation and characterization of tumorigenic, stem-like neural

precursors from human glioblastoma. Cancer Res 64:70117021.10. Fang D, et al. (2005) A tumorigenic subpopulation with stem cell properties in mela-

nomas. Cancer Res 65:93289337.11. Schatton T, et al. (2008) Identification of cells initiating human melanomas. Nature

451:345349.12. Al Hajj M, et al. (2003) Prospective identification of tumorigenic breast cancer cells.

Proc Natl Acad Sci USA 100:39833988.13. Bapat SA, Mali AM, Koppikar CB, Kurrey NK (2005) Stem and progenitor-like cells

contribute to the aggressive behavior of human epithelial ovarian cancer. Cancer Res65:30253029.

14. Collins AT, et al. (2005) Prospective identification of tumorigenic prostate cancer stemcells. Cancer Res 65:1094610951.

15. Ricci-Vitiani L, et al. (2007) Identification and expansion of human colon-cancer-initiating cells. Nature 445:111115.

16. OBrien CA, Pollett A, Gallinger S, Dick JE (2007) A human colon cancer cell capable ofinitiating tumor growth in immunodeficient mice. Nature 445:106110.

17. Dalerba P, et al. (2007) Phenotypic characterization of human colorectal cancer stemcells. Proc Natl Acad Sci USA 104:1015810163.

18. EramoA,et al. (2008) Identificationandexpansionof the tumorigenic lungcancer stemcell population. Cell Death Differ 15:504514.

19. Yin AH, et al. (1997) AC133, a novel marker for human hematopoietic stem andprogenitor cells. Blood 90:50025012.

20. Miraglia S, et al. (1997) A novel five-transmembrane hematopoietic stem cell antigen:Isolation, characterization, and molecular cloning. Blood 90:50135021.

21. UchidaN, et al. (2000) Direct isolation of human central nervous system stem cells. ProcNatl Acad Sci USA 97:1472014725.

22. Lee A, et al. (2005) Isolation of neural stem cells from the postnatal cerebellum. NatNeurosci 8:723729.

23. Corbeil D, et al. (2000) The human AC133 hematopoietic stem cell antigen is alsoexpressed inepithelial cells and targeted toplasmamembraneprotrusions. J Biol Chem275:55125520.

24. RichardsonGD, et al. (2004) CD133, a novelmarker for human prostatic epithelial stemcells. J Cell Sci 117:35393545.

25. Salven P, et al. (2003) VEGFR-3 and CD133 identify a population of CD34 lymphatic/vascular endothelial precursor cells. Blood 101:168172.

26. Yin S, et al. (2007) CD133 hepatocellular carcinoma cells possess high capacity fortumorigenicity. Int J Cancer 120:14441450.

27. Singh SK, et al. (2004) Identification of human brain tumor initiating cells. Nature432:396401.

28. Shmelkov SV, et al. (2008) CD133 expression is not restricted to stem cells, and bothCD133 and CD133 metastatic colon cancer cells initiate tumors. J Clin Invest118:21112120.

29. Wang J, et al. (2008) CD133 glioma cells form tumors in nude rats and give rise toCD133 cells. Int J Cancer 122:761768.

30. Joo KM, et al. (2008) Clinical and biological implications of CD133 and CD133 cellsin glioblastomas. Lab Invest 88:808815.

31. Chen YC, et al. (2008) Oct-4 expression maintained cancer stem-like properties in lungcancer-derived CD133 cells. PLoS ONE 3:e2637.

32. Frank NY, et al. (2005) ABCB5-mediated doxorubicin transport and chemoresistance inhuman malignant melanoma. Cancer Res 65:43204333.

33. Sarkadi B, Ozvegy-Laczka C, Nemet K, Varadi A (2004) ABCG2: A transporter for allseasons. FEBS Lett 567:116120.

34. Summer R, et al. (2003) Side population cells and Bcrp1 expression in lung.Am J Physiol285:L97L104.

35. Levina V, et al. (2008) Drug-selected human lung cancer stem cells: Cytokine network,tumorigenic and metastatic properties. PLoS ONE 3:e3077.

36. Villa A, Snyder EY, Vescovi A, Martinez-Serrano A (2000) Establishment and propertiesof a growth factor-dependent, perpetual neural stem cell line from the human CNS.Exp Neurol 161:6784.

37. Lanzkron SM, Collector MI, Sharkis SJ (1999) Hematopoietic stem cell tracking invivo: A comparison of short-term and long-term repopulating cells. Blood 93:19161921.

38. Hermann PC, et al. (2007) Distinct populations of cancer stem cells determine tumorgrowth and metastatic activity in human pancreatic cancer. Cell Stem Cell 1:313323.

39. Pratesi G, et al. (1989) Effects of 5-FU and cis-DDP combination on human colorectaltumor xenografts. Tumori 75:6065.


Supporting InformationBertolini et al. 10.1073/pnas.0905653106SI TextLung Tumor Tissue Disaggregation. Solid tissues were finely mincedby razorblade, washed in DMEM/F12 (Lonza), and then incubatedwith Accumax 1 (Innovative Cell Technologies) for 1 h at 37 C.Single-cell suspension was obtained by filtering digested tissuethrough a 70-m cell strainer and then gently loaded onto a layerofHistopaque-1077 gradient (SigmaAldrich).After centrifugationat 400 g for 30 min at room temperature, red blood cells, deadcells, and debris were removed from the bottomof the tube, and livenucleated cells were collected at the interface.

Magnetic and Cytofluorimetric Cell Separation. CD133 separation.CD133 and CD133 cells were isolated from tumor cellsuspension or sphere cultures by FACS or magnetic bead sortingusing the MACS system (Miltenyi Biotech). Magnetic beadseparation was used for the cases with a high content of CD133cells (10%).For magnetic separation, cells were incubated with the mono-

clonal CD133/1 antibody labeled with MicroBeads (MiltenyiBiotech) for 30 min at 4 C, and CD133 cells were selected byusing MS columns (Miltenyi Biotech), which retained positivecells linked by beads. For FACS separation, cells were stainedwith CD133/1-PE antibody (AC133 clone; Miltenyi Biotech)diluted 1:10 in blocking buffer solution for 10 min at 4 C andwere sorted with FACS Vantage-SE cell sorter (Becton Dick-inson). In both instances, the purity of the CD133 and CD133cell populations was evaluated by standard flow cytometryanalysis using a PE-labeled antibody against human CD133/2(clone 293C3; Miltenyi Biotech).Mouse cell depletion. Xenograft tumor cell mixture was incubatedwith anti-mouse biotinylated H-2Kd antibody (clone SF1-1.1; BDBioscience) at 1 g per 106 cells for 10 min at 4 C and thenincubated with 25 g per l06 cells of anti-biotin Dynabeads for20 min at 4 C. Bound human cells were collected by using amagnet (Dynal MPB; Invitrogen).

IHC. Paraffin sections (2 m thick) were dewaxed in xylene andrehydrated with distilled water. CD133 antigen retrieval wasperformed in citrate buffer solution (5 mM/L, pH 6). Afterperoxidase inhibition with a solution of 0.3% H2O2 in methylalcohol for 30 min, the slides were incubated with the followingantibodies: CD133/1 (AC133; Miltenyi Biotech); PE-10, TTF-1,and MIB-1 (Dako); CK-7 (NeoMarkers); and a pool for low-molecular weight cytokeratins (35h11 from Dako and CAM5.2from Becton Dickinson) and high-molecular weight cytokeratins(34e12 from Dako and KS8.12 from Sigma). The primaryantibody detection was performed by using Ultra Vision detec-tion system-HPR polymer (Thermo Fisher Scientific) and dia-minobenzidine substrate chromogen (Dako), followed by coun-terstaining with hematoxylin.

Real-Time PCR. Total RNA extraction from cells and DNasedigestion were carried out with the RNAeasy kit (Qiagen).cDNA synthesis was performed with the High-Capacity cDNAReverse Transcription Kit (Applied Biosystems) using 1 g oftotal RNA in final volume of 20 L.The relative quantification of the stemness-associated genes

mRNA (Notch1, Notch2, Hes1, SHH, Gli-1, CD133, Oct4/3,CXCR4, ITGA6, Nanog) was performed by TaqMan technologyusing the ABI PRISM 7900 HT Sequence Detection System(Applied Biosystems) and ready-to-use Assay-on-Demand (Ap-plied Biosystems).

Human HPRT was used as endogenous control for thenormalization of different samples and relative quantization ofgene expression; the data were analyzed by comparative Ctmethod (Ct). For analysis of xenograft mRNA, we previouslyvalidated the specificity for human transcripts of all used Taq-Man assays (Applied Biosystems) to exclude a possible cross-reactivity with mouse transcripts.Analysis of ABC transporter mRNA levels was performed

with TaqMan Low-Density Arrays: 2 L of cDNA was mixedwith 48 L of nuclease-free water and 50 L of TaqManUniversal PCR Master Mix and then loaded into a sample portof the Micro Fluidic Cards (Human ABC Transporters Panel;Applied Biosystems). The thermal cycling conditions of real-time PCR were 2 min at 50 C and 10 min at 95 C, followed by40 cycles of 30 s at 97 C and 1 min at 59.7 C. Data of RQ assayswere analyzed with RQ Manager 1.2 and Sequence DetectionSystem (SDS) 2.2.2 software (Applied Biosystems).

PKH26 and PKH67 Labeling. A549 and LT73 lung cancer sphereswere labeled with the PKH26 red fluorescent and PKH67 greenfluorescent dyes (Sigma). Briefly, harvested cells were dissoci-ated to single cells, washed in PBS, and resuspended in 1 mL ofthe dilution buffer. The cell suspension was mixed with an equalvolume of the labeling solution containing 4 106 M PKH26or PKH67 in the dilution buffer and was incubated for 5 min atroom temperature. A total of 2 mL of FBS was added to labelingsolution to stop the reaction, and cells were washed three timesin serum-free medium. The labeling ratio was determined byflow cytometry.Red- and green-labeled single cells were mixed and plated at

clonal density of 1 104 cells per mL in six-well culture plates(Corning). A distinct red or green fluorescent staining ofgrowing spheres was monitored through fluorescence micros-copy, indicating a clonal origin from one PKH26- or PKH67-labeled cell.

Cell Lines and Cell Sensitivity to Cisplatin. A549 and A549/Pt cellswere maintained in RPMI medium 1640 plus 10% FCS. Thecisplatin-resistant A549/Pt subline was generated from parentalcells by continuous exposure to increasing concentrations of cis-platin. Resistance was stable up to 6 months when cells were grownin the absence of drug. Cellular sensitivity to drug was evaluated byusing growth inhibition assays. For assessment of modulation ofCD133 levels by cisplatin, exponentially growing A549 cells wereseeded in T75 flasks, and 24 h later they were exposed for 1 h to acisplatin concentration corresponding to IC80. Cells were harvested72 h after treatment for flow cytometric analysis.

In Vivo Evaluation of Response to Cisplatin. Tumor fragments wereimplanted at day 0, and tumor growth was followed by biweeklymeasurement of tumor diameters with a Vernier caliper. Tumorvolume (TV) was calculated according to the formula: TV(mm3) d2D/2, where d andD are the shortest and the longestdiameters, respectively. Cisplatin (Teva Pharma) was adminis-tered i.v. (10 mL/kg) at its optimal dose according to a scheduleof every seventh day for three times (5 mg/kg per injection),starting when tumors were palpable (50 mm3). The efficacy ofdrug treatment was assessed as TV inhibition percentage(TVI%) in treated over control mice, calculated as: 100 (meanTV-treated mice/mean TV control mice 100). For tumorremoval, mice were killed by cervical dislocation under lightanesthesia.

Bertolini et al. www.pnas.org/cgi/content/short/0905653106 1 of 4

0%1%

0.1%1%

0.4% 1%

12%

0.2%

6%

1.5%2% 0.2%

0.5%

30%

3%

0.2%1%

0.5%0%

27%

0%

10%

20%

30%

17 la

mro

N

17 r

om

uT

27 la

mro

N

27 r

om

uT

37 la

mro

N

37 r

om

uT

57 la

mro

N

57 r

om

uT

67 la

mro

N

67 r

om

uT

77 la

mro

N

77 r

om

uT

87 la

mro

N87

ro

muT

08 la

mro

N

08 r

om

uT

18 la

mro

N

18 r

om

uT

38 la

mro

N

38 r

om

uT

/331DC

%+

ASEsllec

331DC

-EP

Primary tumor

30%

6%

ESA-FITC

0.5%

0.2%

Normal lungB

LT76

LT80

A

Fig. S1. Identification of CD133 cells in lung tumors and normal lung tissues. (A) Percentage of CD133ESA cells in 10 primary tumors and correspondingnormal lung tissues. (B) FACS analysis of CD133ESA cells in two representative freshly dissociated lung tumors and corresponding normal tissue.


Parental Tumor Xenograft Negative Ctr

H&E

20X 20X

CK7

TTF-1

SP-C

MIB-1

CD133

20X 20X 20X

20X 20X 20X

20X 20X 60X

20X 20X 20X

60X 20X 60X

CK-LMW

60X 60X 20X

Fig. S2. IHC for low-molecular weight CKs (CK-LMW), cytokeratin 7 (CK7), surfactant protein C (SP-C), transcription thyroid factor 1 (TTF-1), CD133, andMIB-1performed in a large cell carcinoma parental tumor (LT56) and corresponding xenograft. Control antibody-stained tumor sample is also shown (Negative Ctr).


Fig. S3. (A) FACS analysis of CD133 expression before and after sorting of CD133 and CD133 fractions. CD133 cells were isolated from freshly dissociatedlung tumors by FACS (LT28;Upper) or bymagnetic beads (LT45; Lower). (B) In vivo tumorigenicity of CD133 and CD133 cells. Tumorigenic potential of CD133,CD133, and unsorted cells, purified from two xenografts (LT56 and LT28) and one primary tumor (LT45), after s.c. injection in immunocompromised mice. Thetumor take for each subpopulation of injected cells is indicated. (C) In vivo serial transplantation assay. A total of 103 CD133 and CD133 cells, purified fromLT73 xenograft, were injected s.c. into SCID mice. Derived tumor xenografts were dissociated to single-cell suspension and then serially reinjected in mice (102

cells), generating secondary and then tertiary tumors. Tumor growth curves of primary and tertiary tumors are shown.


Fig. S4. (A) Real-time PCR analysis of stemness genes expression in cisplatin-treated A549, LT66, LT73, and LT56 xenografts. Control untreated xenografts wereused as calibrator for the relative quantification of gene expression. (B) Real-time PCR analysis of ABC transporters in cisplatin-treated LT73 and LT56 xenograftsusing TaqMan Micro Fluidic cards. Control untreated xenografts were used as calibrator for the relative quantification of gene expression.


Table S1. Clinicopathological characteristics and CD133/ESA expression in NSCLC patients

ID Histology/grading Stage FACS analysis CD133, % FACS analysis ESA, % FACS analysis CD133/ESA, % IHC analysis CD133

LT22 ADC G3 IB 23 80 20 PositiveLT23 LC G3 IIB 31 57 15 PositiveLT24 SCC G2 IIB 1.2 12 1 PositiveLT25 SCC G3 IIIA 0 10 0 NegativeLT26 ADC G3 IIB 0.3 10 0 NegativeLT27 ADC G2 IA 0 80 0 NegativeLT28 ADC G3 IIB 2 5 / NegativeLT29 ADC G3 IB 0 70 0 NegativeLT30 ADC G2 IB 33 89 29 PositiveLT32 SCC G2 IB 1 1.5 0.3 NegativeLT33 SCC G2 IA / / / NegativeLT34 SCC G2 IB 8 26 7 PositiveLT35 SCC G3 IA 0 42 0 PositiveLT36 ADC G3 IA 3 80 3 NegativeLT37 LC G3 IB / / / PositiveLT38 ADC G2 IA 2 13 0.7 PositiveLT39 ADC G3 IA 0.7 33 0.7 PositiveLT40 ADC G2 IIIB 0.9 70 / PositiveLT42 ADC G3 IB 0 8 0 NegativeLT43 ADC G3 IV 2 12 2 PositiveLT44 SCC G3 IA 2 13 / NegativeLT45 ADC G3 IIIA 36 56 35 PositiveLT46 ADC G3 IIIA 0.5 88 0.5 /LT47 ADC G3 IIIA 2 50 2 NegativeLT48 ADC G3 IB 0.3 63 0.3 NegativeLT49 ADC G3 IIIA 3 48 3 PositiveLT50 ADC G3 IIA 2.5 35 2.5 PositiveLT51 ADC G2 IB 0 90 0 NegativeLT52 SCC G3 IA 6 43 5 PositiveLT53 SCC G3 IA 0.7 50 0.7 NegativeLT54 ADC G3 IIIB 1 78 0.7 PositiveLT55 ADC G1 IB 1.5 75 1.5 NegativeLT56 LC G3 IB 15 87 15 PositiveLT57 ADC G2 IIIA 2 57 0.7 PositiveLT58 ADC G3 IA 1 62 1 PositiveLT59 LC G3 IIIA 2 13 2 PositiveLT60 ADC G3 IA / / / PositiveLT62 SCC G3 IV 1.2 14 0.5 /LT63 ADC G2 IA / / / PositiveLT64 ADC G3 IA 5 50 5 NegativeLT65 ADCG3 IIIB 1 46 0.5 NegativeLT66 ADC G3 IIIA 0.02 72 0.02 NegativeLT67 ADC G3 IIIA 0 70 0 NegativeLT68 SCC G2 IB 3 5 0.6 PositiveLT69 ADC G3 IIIA 1.5 84 / NegativeLT70 SCC G3 IA 0 3 0 NegativeLT71 SCC G2 IA 1.3 6 1 NegativeLT72 ADC G2 IIIA 1 60 1 PositiveLT73 ADC G2 IB 27 93 27 PositiveLT74 ADC G2 IA 2 12 2 PositiveLT75 ADC G2 IA 12 40 12 PositiveLT76 ADC G2 IA 7.5 45 6 PositiveLT77 SCC G3 IIB 2 40 2 NegativeLT78 ADC G2 IA 0.2 10 0.2 PositiveLT79 SCC G3 IIA 3 55 3 NegativeLT80 SCC G3 IB 30 58 30 PositiveLT81 ADC G2 IV 3 20 3 PositiveLT82 ADC G2 IA 5 70 5 PositiveLT83 SCC G3 IB 1 26 1 NegativeLT84 SCC G3 IB 1.8 30 1.6 Negative

ADC, adenocarcinoma; SCC, squamous cell carcinoma; LC, large cell carcinoma. Slash (/) indicates material not available.


Table S2. Correlation of CD133 expression (by IHC) and tumor histology, stage, and grading in NSCLC patients

IHC score

Overall Negative Positive

n % n % n % P

Histology 0.0441ADC 37 63.8 15 57.7 22 68.8SCC 17 29.3 11 42.3 6 18.8LC 4 6.9 4 12.5

Tumor stage 0.7091Ia 20 34.5 8 30.8 12 37.5Ib 16 27.6 8 30.8 8 25.0II 7 12.1 4 15.4 3 9.4IIIa 10 17.2 5 19.2 5 15.6IIIbIV 5 8.6 1 3.8 4 12.5

Grading 0.0356G1G2 22 37.9 6 23.1 16 50.0G3 36 62.1 20 76.9 16 50.0


Table S3. In vivo tumorigenicity of CD133 and CD133 cells

Case Cell dose Stage Histology CD133, % Source (separation technique)

Tumor incidence

CD133 CD133 Unsorted

LT28 1 103 IIb ADC 0.2 Xenograft (FC) 2/2 0/4 0/4LT45 1 104 IIIa ADC 35 Primary tumor (MB) 2/3 0/2 0/2LT56 1 103 Ib LC 18 Xenograft (MB) 3/4 0/4 2/4LT66 1 103 IIIa ADC 0.02 Xenograft (FC) 4/4 1/4 2/4LT73 Ib ADC 30 Xenograft (FC)Primary 1 103 3/4 3/4 3/4Secondary 1 103 4/4 4/4 /

1 102 4/4 2/4 /Tertiary 1 103 2/2 1/2 /

1 102 4/4 1/4 /A549/s 1 103 0.8 Cell line (FC) 3/4 2/4 4/4

1 102 5/6 1/6 4/6A549 WT 1 103 0.1 Cell line / / 2/4

1 102 / / 0/6

FC, flow cytometry separation; MB, magnetic bead separation. CD133 and CD133 cells were sorted by FACS or magnetic beads from xenograft, primarytumors, and A549 spheres. Cells were injected s.c. into the flanks of immunocompromised mice at scalar doses. For tumor incidence, / indicates not done.


Table S4. Effect of chemotherapy [cisplatin (cisPt)] on tumor cell subpopulations

Fraction of cells expressing, %

Xenograft cisPt* Max TVI, % CD133 CXCR4 ABCG2 CD133/ABCG2 CD133/CXCR4

A549 Control 0.3 5 36 0.3 NDcisPt 35 2.15 26 38 1.5 NDRelapse 0.7 30 ND ND ND

LT66 Control 0.02 1.3 33 0 0cisPt 70 0.7 7.8 48 0.7 0.7Relapse 0.15 1.3 50 0.15 0.15

LT28 Control 0.2 9.5 5.4cisPt 83 0.2 19.8 24 ND NDRelapse 0.15 19.5 20

LT45 Control 50 4.5 19 14.5 NDcisPt 64 40 9.5 56 25 ND

LT73 Control 5.3 8.5 13.5 5 5.2cisPt 55 5.7 8 22 5.5 5Relapse 5 9 18 4.5 4.5

LT56 Control 18 7 46 6.5 3.5cisPt 43 17 10 68 15 7.5Relapse 15 7 45 7.5 3.5

ND, not determined.*Regimen cisPt: i.v. 5 mg/kg q7d 3 weeks.


Table S5. CD133 expression as a marker of chemotherapeutic response in advanced-stage NSCLC patients treated withcarboplatin/gemcitabine: Patient characteristics

Clinicopathologic features No. of cases (%); n 42 No. of CD133 cases (IHC); n 10 P

Sex 0.6966Male 29 (69) 6Female 13 (31) 4

Age 0.304961 y 19 (45) 361 y 23 (55) 7

Smoking behavior 0.4164Never/nonsmokers 9 (21.4) 1Smokers 33 (78.6) 9Packs per year, median (range) 18.2 (9.145.6)

Stage at diagnosis 0.4164IV 9 (21.4) 1IV 33 (78.6) 9

Histology 0.1189Adenocarcinoma 25 (59.5) 8Squamous cell carcinoma 14 (33.3) 1


highly tumorigenic lung cancer cd133+ cells display stem-like features and are spared by cisplatin...

Documents