Total cellular iron as a marker of cancer stem cells and uses thereof

Abstract

The present invention relates to a novel use of total cellular iron, preferably under the form of ferrous iron (Fe.sup.2+), as a marker of cancer stem cells (CSCs). The invention also relates to methods using said iron marker, in particular for metastatic cancer diagnosis or treatment, for screening for compounds of interest, as well as for killing CSCs.

Claims

1. A method for selectively killing CSCs in a mammal, comprising administering to said mammal an iron-chelating pharmaceutical composition, comprising at least one component selected from 20-alkyl-amino derivatives of salinomycin of formula (I′) ##STR00003## wherein: X is selected from the group consisting of OH and ═O, Y is selected from the group consisting of —NR.sub.1R.sub.2; —NR.sub.3—(CH.sub.2).sub.n—NR.sub.4R.sub.5; —O—(CH.sub.2).sub.n—NR.sub.4R.sub.5; —NR.sub.3—(CH.sub.2).sub.n—N.sup.+R.sub.6R.sub.7R.sub.8; and —O—(CH.sub.2).sub.n—N.sup.+R.sub.6R.sub.7R.sub.8; and R.sub.1 and R.sub.2, identical or different, are selected from the group consisting of H; (C.sub.1-C.sub.16)-alkyl; (C.sub.3-C.sub.16)-alkenyl; (C.sub.3-C.sub.16)-alkynyl; aryl; heteroaryl; (C.sub.1-C.sub.6)-alkyl-aryl; (C.sub.1-C.sub.6)-alkyl-heteroaryl; or R.sub.1 represents H and R.sub.2 represents OR.sub.9, where R.sub.9 is H, (C.sub.1-C.sub.6)-alkyl, aryl and (C.sub.1-C.sub.6)-alkyl-aryl; R.sub.3 is selected from the group consisting of H; (C.sub.1-C.sub.6)-alkyl; (C.sub.1-C.sub.6)-alkyl-aryl; R.sub.4 and R.sub.5, identical or different, are selected from the group consisting of H; (C.sub.1-C.sub.6)-alkyl; aryl; (C.sub.1-C.sub.6)-alkyl-aryl; R.sub.6, R.sub.7 and R.sub.8, identical or different, are selected from the group consisting of (C.sub.1-C.sub.6)-alkyl; aryl; (C.sub.1-C.sub.6)-alkyl-aryl; n=2, 3, 4, 5 or 6, Z is a functional group capable of chelating iron salts such as OH: NHNR.sub.9R.sub.10 (hydrazine), NHOC(O)R.sub.11 (O-Acyl hydroxylamine), N(OH)—C(O)R.sub.11 (N-acyl hydroxylamine), OOH, SR.sub.12; 2-aminopyridine; 3-aminopyridine; —NR.sub.3—(CH.sub.2).sub.n—NR.sub.4R.sub.5; —NR.sub.3—(CH.sub.2).sub.n—OH; where: R.sub.9 and R.sub.10, identical or different, are selected from the group consisting of H, (C.sub.1-C.sub.6)-alkyl, aryl and (C.sub.1-C.sub.6)-alkyl-aryl; R.sub.11 is selected from the group consisting of H; (C.sub.1-C.sub.16)-alkyl; (C.sub.3-C.sub.16)-alkenyl; (C.sub.3-C.sub.16)-alkynyl; aryl; heteroaryl; (C.sub.1-C.sub.6)-alkyl-aryl; (C.sub.1-C.sub.6)-alkyl-heteroaryl; R.sub.12 is selected from the group consisting of H; (C.sub.1-C.sub.16)-alkyl; (C.sub.3-C.sub.16)-alkenyl; (C.sub.3-C.sub.16)-alkynyl; aryl; heteroaryl; (C.sub.1-C.sub.6)-alkyl-aryl; (C.sub.1-C.sub.6)-alkyl-heteroaryl, n=0, 2, 3 or 4, AM5, AM23, AM23S, and combinations thereof.

2. The method of claim 1, wherein said iron-chelating pharmaceutical composition binds total cellular iron under the form of ferrous (Fe.sup.2+) and ferric (Fe.sup.3+) iron in said mammal, thereby inducing ferritinophagy and reactive oxygen species (ROS) production responsible for specific and/or selective CSC death in said mammal.

3. The method of claim 1, wherein said iron-chelating pharmaceutical composition binds total cellular iron under the form of ferrous (Fe.sup.2+) and ferric (Fe.sup.3+) iron in said mammal while preventing drug-resistance in said mammal by selectively targeting CSCs.

4. The method of claim 1, wherein said iron-chelating pharmaceutical composition binds total cellular iron under the form of ferrous (Fe.sup.2+) and ferric (Fe.sup.3+) iron in said mammal while preventing drug-resistance in said mammal by selectively targeting CSCs.

5. A method according to claim 1, wherein the mammal having CSCs is selected according to an in vitro method for diagnosing cancer having a high risk of recurrence, a high risk of metastasis, and/or a cancer with resistance to therapy, in a subject, comprising at least: a) measuring the amount of total cellular iron, preferably under the form of ferrous iron (Fe.sup.2+), in a biological sample from a subject; and b) comparing said amount measured in step a) to a reference value range for total cellular iron, wherein an amount of total cellular iron as measured in step a) is higher than said reference value range is indicative of the presence of CSCs, thereby indicating that said subject has a cancer.

6. A method according to claim 5, wherein the amount of total cellular iron as measured in step a) of the in vitro method, which is indicative of the presence of CSCs, is superior or equal to 0.05 pg/cell, preferably superior to 0.06 pg/cell, preferably superior to 0.07 pg/cell, preferably superior to 0.08 pg/cell, preferably superior to 0.09 pg/cell, preferably superior to 0.10 pg/cell, preferably superior to 0.11 pg/cell, preferably superior to 0.12 pg/cell, preferably superior to 0.13 pg/cell, preferably superior to 0.14 pg/cell, preferably superior to 0.20 pg/cell, preferably superior to 0.24 pg/cell, more preferably higher than 0.30 pg/cell.

7. A method according to claim 1, wherein said composition is co-administered with radiation therapy or chemotherapy.

8. A method according to claim 1, for selectively killing breast CSCs in a mammal.

Description

FIGURES

(1) FIG. 1. Salinomycin and AM5 sequester iron in lysosomes and trigger ferritin degradation. a, Chemical strategy to visualize small molecules in cells. b, Fluorescence microscopy images of U2OS cells showing the subcellular localization of labeled Sal derivatives AM4, AM5, and AM9. Lysotracker Deep Red stains the lysosomes and 4′,6-diamidino-2-phenylindole (DAPI) stains nuclear DNA. Scale bar, 10 μm. Zoom corresponds to ×6. c, Live cell fluorescence microscopy images showing the subcellular localization of iron(II) using the fluorogenic reduction of RhoNox-1 (green) in HMLER CD24.sup.low cells treated with Sal or AM5 or AM9 (0.5 μM) for 48 h. Lysotracker deep red stains the lysosomes (red) and DAPI stains nuclear DNA (blue). Scale bar, 10 μm. d, Fixed cell fluorescence microscopy images showing the subcellular localization of iron(II) using the fluorogenic reduction of RhoNox-1 (green) in HMLER CD24.sup.low cells treated with Sal or AM5 or AM9 (0.5 μM) for 48 h. e, Quantification of lysotracker-positive vesicles colocalizing with RhoNox-1 in fixed cells was carried out by means of visual inspection. At least 75 cells were counted per condition. Data represent three independent biological replicates (n=number of lysotracker vesicles). Bars and error bars correspond to mean values and s.d. of three biological replicates, respectively. f, Immunoblotting showing levels of iron homeostasis regulatory proteins in HMLER CD24.sup.low cells treated as indicated. g, Fluorescence microscopy images showing the subcellular localization of ferritin in HMLER CD24.sup.low cells treated as indicated for 6 h. Scale bar, 10 μm. Zoom corresponds to ×6. h, Immunoblotting showing levels of ferritin in HMLER CD2.sup.low cells treated as indicated for 24 h. i, Detection of iron(III) with Perl's reagent in tumor tissues of mice treated as indicated. Scale bar, 50 μm. j, .sup.1H-NMR spectra of Sal+/−FeCl.sub.2, anthracene and bipyridine. Spectra were recorded in CD.sub.3CN. Arrows indicate proton signals affected by the presence of iron. Coordination of Sal to iron is arbitrary. k, .sup.1H-NMR spectra of I. AM5 (2 mM) and Napht (1.0 mol equiv.), II. AM5 and Napht in the presence of FeCl.sub.2 (0.5 mol equiv.), III. AM5, Napht and Bipy (1.6 mol equiv.) in the presence of FeCl.sub.2 (Bipy added after FeCl.sub.2), IV. AM5 and Bipy. I, .sup.1H-NMR spectra of I. Sal (2 mM) and Napht (1.0 mol equiv.), II. Sal and Napht, in the presence of FeCl.sub.2 (0.5 mol equiv.), III. Sal, Napht and Bipy (1.6 mol equiv.) in the presence of FeCl.sub.2 (Bipy added after FeCl.sub.2), IV. Sal and Bipy. For k and I, Samples prepared in CD.sub.3OD, spectra recorded at 298 K, 5 min following sample preparation (600 MHz). Stars indicate proton signals shielded by iron(II). Boxed regions highlight signals of free Napht (boxed region of I, II, and III) and free/bound Bipy, (boxed region of III and IV), respectively.

(2) FIG. 2. Salinomycin and AM5 promote iron-dependent lysosomal cell death via ROS production. a and b, Percentage of cell death of HMLER CD24.sup.low cells treated as indicated for 48 h. Measurements and quantification performed FACS using Annexin V FITC and PI fluorescence. **P<0.01, ***P<0.001, Student's t-test. c, Quantification of cellular iron by electrothermal atomic absorption spectrometry in HMLER cells. Error bars represent s.d. (n=3). **P<0.01, Student's t-test. d, Comparative immunoblotting analysis of endogenous levels of EMT markers, TfR and cathepsin B in HMLER cells.

(3) FIG. 3. Schematic illustration of salinomycin targeting iron homeostasis. Transferrin (Tf) and transferrin receptor (TfR) mediate iron endocytosis. Acidification of endocytic vesicles releases iron from transferrin, which is then reduced and exported to the cytosol. Iron can be stored as a non-toxic iron/ferritin (FT) complex. The iron chelator DFO and iron conjugated base Sal/AM5 can sequester iron in the lysosome and deplete the pool of cytosolic iron. Iron depletion triggers ferritin degradation by autophagy including NCOA4-mediated ferritinophagy and/or chaperone mediated autophagy (CMA) and/or endosome microautophagy. Although mechanisms through which ferritin was delivered to late endosomes/lysosomes remain to be fully characterized, microautophagy appears to be the central mechanism. Indeed, ferritin relocalization to the lysosomal compartment was also observed in absence of the ferritin cargo receptor NCOA4. Lysosome/autophagosome fusion or multivesicle body/lysosome fusion leads to the cathepsin-mediated degradation of ferritin to replenish the available pool of iron, an event that can be prevented by the cathepsin inhibitor CA-074 (CA). Iron promotes the production of ROS through Fenton-type chemistry, which can be prevented by iron chelation with DFO or scavenged by N-acetyl-L-cysteine (NAC). ROS induce lysosome membrane permeabilization and cell death. Black arrows indicate normal iron homeostasis. Red arrows indicate the response to Sal and AM5.

(4) FIG. 4. Salinomycin interacts with iron without altering its reactivity.

(5) a, .sup.1H-NMR spectra of 5 mM Sal+/−FeCl.sub.3 (0.5 mol equiv.). Spectra were recorded in CD.sub.3CN at 298 K. Arrows indicate proton signals affected by the presence of iron. The signal at 7.4 ppm is characteristic hydrogen bonding and a more rigid supramolecular assembly of Sal with iron. b, Photographs of methanolic solutions of the indicated small molecules+/−iron(II). Color changes indicate the formation of distinct iron complexes. Picture was taken 1 minute after addition of DFO or 2,2′-bipyridyl (Bipy) when these compounds were added to vials already containing Sal and FeCl.sub.2. The appearance of a color in the middle and far right vials indicates that Bipy and DFO outcompete Sal to form a complex with iron. c, Schematic illustration of the fluorogenic reduction of RhoNox-1 by iron (II) that can be inhibited by DFO. d, Quantification of the fluorogenic reduction of RhoNox-1 in the presence of iron (II) showing that DFO but not Sal poisons iron.

(6) FIG. 5. Immunoblotting showing levels of E-Cad and Ferritin in parental and derivative HMLER cells.

(7) FIG. 6. Schematic representation of EMT

(8) FIG. 7. Knock down of ferritin expression interferes with the OSM-mediated induction of EMT in MCF-7 cells. left, Immunoblotting showing levels of EMT markers in MCF-7 cells treated or not with 100 ng/ml of OSM for 72 h. right Immunoblotting showing levels of ferritin and fibronectin proteins in MCF-7 cells transfected with siRNA control (scramble) or siRNA targeting specifically ferritin mRNA (FTH1) during 48 h and then the cells were incubated with OSM (100 ng/ml) for 48 h. Beta-tubulin is used as loading control. MCF-7 were transfected.

(9) FIG. 8. Knock down of ferritin expression interferes with the OSM-mediated induction of CD44.sup.highCD24.sup.low/− in MCF-7 cells. MCF-7 were transfected with siRNA control or siRNA specifically targeting ferritin mRNA (FTH1) during 48 h and then the cells were incubated with OSM (100 ng/ml) for 48 h. a. Analysis of CD44 and CD24 expression by flow cytometry. Left panel, CSC indicates the percent of CD44.sup.high/CD24.sup.Low or negative (Low/−) population gated. Right panels, histogram representations of CD44 or CD24 expression (dark grey), or by the isotype control (light grey). b. Graphs representing the percent mean of CD44.sup.high/CD24.sup.Low/− population gated. c. Graphs representing the mean fluorescence intensity (IF) of CD24 and CD44, respectively. d. Quantification of the ALDH+ population in MCF-7 cells treated as indicated above, and measured by flow cytometry. Bars and error bars correspond to mean values and s.d. of two biological replicates, respectively. **P<0.01, ns, not significant, Student's t-test.

(10) FIG. 9. Sal and AM5 effect on non-CSC and CSC cells. Immunoblotting showing levels of iron homeostasis regulatory proteins in HMLER CD44.sup.Low CD24.sup.high (HMLER_ID2) and iCSCA2 (another cancer stem cell) treated as indicated. CA 074 is cathepsin B inhibitor.

(11) FIG. 10. HMLER CD24.sup.Low cells uptake transferrin more than its isogenic HMLER ID2 counterpart. Briefly, HMLER CD24.sup.LowCD44.sup.high cells and its HMLER CD24.sup.high CD44.sup.Low/− ID2 counterpart are co-cultured at a ratio of 1:1 for 24 h and then are incubated with Transferrin-FITC (50 μg/mL) for the indicated times. Transferrin-FITC uptake was assessed by flow cytometry. a. CD24.sup.LowCD44.sup.high HMLER cells were identified by CD44 expression (here, CD44-PE) to discriminate from ID2 cells which are CD44.sup.Low/−. b. CD24.sup.LowCD44.sup.high HMLER cells were stained by celltracker deep red dye (here, APC) to discriminate ID2 cells which are not stained before co-culture. c. Graphs representing the fluorescence intensity median of Transferrin-FITC for each population for the time indicated.

(12) FIG. 11. Transferrin potentiates the TGF-beta-mediated induction of CD44.sup.highCD24.sup.Low/− in HMLER cells. Briefly, HMLER cells were treated as indicated with TGF beta in the presence or absence of transferrin. CD44 and CD24 expression were assessed by flow cytometry. CD44.sup.+ indicates the percent of CD44.sup.high/CD24.sup.Low/− population gated.

(13) FIG. 12. Sal and AM5 induce biochemical features characteristic of ferroptosis. a, Flow cytometry analysis of lipid ROS in cells treated with Sal (0.5 μM) or AM5 (0.5 μM) for 48 h. Flow cytometry analysis of 510 Annexin V-FITC (A) and Propidium Iodide (PI) fluorescence in b, HMLER CD24.sup.low and c, iCSCL-10A2 cells treated with Sal (0.5 μM) or AM5 (0.5 μM) for 24, 48 or 72 h. d, Primary data of the quantification of the flow cytometry analysis of e. e, Flow cytometry analysis of Annexin V-FITC (A) and Propidium Iodide (PI) fluorescence in iCSCL-10A2 cells treated with Sal (0.5 μM) or AM5 (0.5 μM) for 72 h, in the presence or absence of the indicated inhibitors. Living cells are A−/PI− and ferroptotic cells (regulated necrosis) exhibit a positive PI+ staining. Bars and error bars correspond to mean values and s.d. of two biological replicates, respectively. f, Endogenous levels of GSH in cells treated as indicated, measured as described in the Materials and Methods. Bars and error bars, mean values and s.d. of two biological replicates. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, one-way ANOVA.

(14) FIG. 13. Sal and AM5 effect on an ALDH+CSC cell population in vitro. Flow cytometry measurement of the effect of 0.5 μM Sal and AM5 against ALDH+ iCSCL-10A2 cell subpopulation after treated for 48 h. DEAB is the diethylaminobenzaldehyde ALDH inhibitor.

(15) FIG. 14. Sal and AM5 alter the maintenance of CSCs in vivo in the patient derived xenograft (PDX) model. Quantification of the proportion of residual ALDH+ cells in PDX treated by means of intra-peritoneal injections (n>4 per condition per PDX in NOD/scid mice) measured by flow cytometry. Bars and error bars correspond to mean values and s.d., respectively. *P<0.05, **P<0.01, ***P<0.001, Student's t-test.

(16) FIG. 15. In vivo antitumor effect of Sal and AM5 against PDX in nude mice. Tumor-seeding capacity of cells treated in vivo by means of intra-peritoneal injections (n>4 per condition per PDX in NOD/scid mice) and estimated number of CSCs calculated by extreme limiting dilution analysis (ELDA) software. P values, χ.sup.2 pairwise test.

(17) FIG. 16. Effect of AM5 and/or Taxol on mammosphere formation. a. Representative phase contrast photomicrographs of mammospheres formed after 7 or 14 days in the absence of any added compound (Not treated) or in the presence of a defined amount of salinomycin, AM5, Taxol or the combination of AM5 and Taxol. A smaller mass indicates cell death and regression of the mammosphere. b. Quantification of the number and the size of the mammospheres. The combination of AM5 at 15 nM and Taxol at 5 nM decreases the number and the size of mammospheres with a higher efficacy than AM5 alone at 15 nM or 5 nM.

EXAMPLES

(18) Material and Methods:

(19) Cell culture. Dulbecco's Phosphate-Buffered Saline (14190-094, 500 mL, Gibco), DMEM/F12 (31331-028, 500 mL, Gibco), DMEM high glucose with UltraGlutamine (BE12-604F/U1, BioWhittaker, Lonza), McCoy's 5A (Modified) Medium (26600-023, Gibco), RPMI 1640 with L-glutamine (BE12-702F/U1, BioWhittaker, Lonza), Fetal Bovine Serum (FBS, 10270-106, Gibco), Hydrocortisone (H0888, Sigma), Insulin (10516, Sigma or 19278, Sigma), BD human recombinant Epidermal growth factor (hEGF, 354052, BD Biosciences), PEN-STREP (DE17-602E, BioWhittaker, Lonza), Puromycin dihydrochloride (A11138-02, Life Technologies). Human osteosarcoma U2OS cell line (ATCC®, HTB-96™) was cultured in McCoy's 5A medium supplemented with 1×PEN-STREP and 10% FBS. MCF-7 (ATCC®, HTB-22™) was maintained in RPMI medium supplemented with 1×PEN-STREP and 10% FBS. The human mammary epithelial cell line infected with a retrovirus carrying hTERT, SV40 and the oncogenic allele HrasV12, named HMLER CD44.sup.high/CD24.sup.low cells, not expressing E-cadherin and expressing Vimentin (referred to as HMLER CD24.sup.low) was a generous gift from A. Puisieux (INSERM). See ref 5 of the main text. HMLER CD24.sup.low cells or isogenic non-stem HMLER CD24.sup.high cells were cultured in DMEM/F12 supplemented with 10% FBS, 10 μg/mL insulin, 0.5 μg/mL hydrocortisone, 10 ng/mL hEGF, and 0.5 μg/mL puromycin. A mycoplasma test was performed using PCR mycoplasma detection kit (G238, Applied Biological Materials).

(20) Cell viability assay. Cell viability assay was carried out by plating 1000 cells/well in 96-well plates. Cells were treated as indicated for 72 h. CellTiter-Blue® Reagent (G8081, Promega) was added after 72 h treatment and cells were incubated for 1 h before recording fluorescence intensities (Excitation, 560/20 nm; Emission, 590/10 nm) using a Perkin Elmer Wallac 1420 Victor2 Microplate Reader.

(21) Clonogenic assay. HMLER CD24.sup.low cells were plated in 6-well plates and incubated with various concentrations of Sal and derivatives for 72 h. Single-cell suspensions were mixed with an equal volume of 0.7% soft agar and plated in 6-well plates (2500 cells/well) for 10 days. After staining with 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT, M2128, Sigma). Colonies with a diameter of more than 0.5 mm were counted.

(22) Chemical labeling of Sal derivatives and fluorescence microscopy. U2OS cells were cultured at ˜80% confluence and were treated for 6 h with 2 M compounds unless stated otherwise. LysoTracker® Deep Red (L12492, Molecular Probes®) and MitoTracker Red CMXRos (M7512, Molecular probes) were added 1 h prior cell fixation. Cells were fixed 12 min with 2% PFA/PBS, washed with PBS and permeabilized for 10 min with 0.1% Triton X-100/PBS, then washed three times with 1% BSA/PBS. Methanol fixation was used instead of formaldehyde fixation for specific antibodies according to the manufacturer protocol. Briefly, cells were incubated in absolute methanol (HPLC grade) at −20° C. for 15 min prior permeabilization. The click reaction cocktail was prepared from Click-iT® EdU Imaging kits (C10337, Life Technologies) according to the manufacturer's protocol. Briefly, mixing 430 μL of 1× Click-iT® reaction buffer with 20 μL of CuSO.sub.4 solution, 1.2 μL Alexa Fluor® azide, 50 μL reaction buffer additive (sodium ascorbate) to reach a final volume of ˜500 μL. Cover-slips were incubated with the click reaction cocktail in the dark at RT for 30 min, then washed three times with PBS. For immunofluorescence, cells were blocked with 5% BSA, 0.2% Tween-20/PBS (blocking buffer) for 10 min at RT. Cover-slips were incubated with 100 μL of diluted primary antibodies in blocking buffer (e.g., ERp72, RCAS1, LC3, Ferritin) overnight at 4° C. Cover-slips were then washed three times with blocking buffer and incubated as described above with the appropriate secondary antibodies for 30 min. Cover-slips were washed three times with PBS and mounted using Vectashield Mounting Medium with DAPI (H-1200, VECTOR Laboratories). High resolution fluorescence images were acquired using a Deltavision real-time microscope (Applied Precision). 60×/1.4NA and 100×/1.4NA objectives were used for 2D and 3D acquisitions that were deconvoluted using SoftWorx software (Ratio conservative—15 iterations, Applied Precision). ImageJ was used for further image processing.

(23) Tumorsphere assay. HMLER CD24.sup.low cells were plated as single-cell at 10.sup.3 cells/mL in ultra-low attachment culture dishes using serum-free DMEM/F12 supplemented with B-27 (17504044, Invitrogen, 1:50), 20 ng/mL hEGF, 4 μg/mL insulin and 0.5 μg/mL hydrocortisone. After 7 days, tumorspheres were enzymatically dissociated with 0.05% trypsin (15090, Gibco) for 15 minutes at 37° C. to obtain a single-cell suspension. Sphere formation was assessed 7 days after seeding cells individually in 96-well ultralow attachment plates (Corning), which were treated as indicated. The number and the size of tumorspheres were analyzed under a light microscope.

(24) Xenograft tumor formation experiments. MCF-7 cell cultures were collected, enzymatically dissociated, washed with PBS, and re-suspended in a PBS/Matrigel mixture (1:1 v/v). The mixture (0.1 mL) was then implanted in the mammary fat pad of 5-week-old female AthymicNude-Fox1nu mice bilaterally (Harlan, France). Mice were maintained in individually-ventilated cages (Tecniplast, France) under constant temperature and humidity. All experiments were performed under laminar flow (Tecniplast France). Mice received estradiol supplementation (0.4 mg/kg) the same day and 7 days from cell injection, and were observed and palpated for tumor appearance. Mice were treated with AM5 (3 mg/kg body weight/day) by means of intraperitoneal injections every 5 opened days of the week. Tumor growth was measured weekly using calipers. Tumor volume was determined using the standard formula: L×W.sup.2×0.52, where L and W are the longest and shortest diameters, respectively. All animal studies were approved by the Direction des services Vétérinaires, Prefecture de Police, Paris, France (authorization number A75-14-08) and the ethical committee (number 34) of Paris Descartes University. No randomization was used and experimenters were blinded to drug treatments and tissue analyses.

(25) Patient derived xenografts (PDXs). PDXs were established as previously described in Charafe-Jauffret et al., Cancer Res. 73, 7290-7300 (2013) from tissue samples collected prospectively at the Institut Paoli-Calmettes (IPC). Samples of human origin and associated data were obtained from the IPC/CRCM Tumor Bank that operates under authorization #AC-2007-33 granted by the French Ministry of Higher Education and Research. Before scientific use of samples and data, patients were appropriately informed and asked to consent in writing, in compliance with French and European regulations. The project was approved by the IPC Institutional Review Board. Cells from 2 different patient-derived xenografts (PDX 1 and 2) were transplanted orthotopically into fat pads of 4-week-old female non-obese diabetic/severe combined immunodeficient (NOD/SCID) mice (NOD.CB17-Prkdcscid/J, Charles River France) after mechanical treatment and enzymatic digestion using collagenase/hyaluronidase (StemCell Technologies) to generate single-cell suspension suitable for implantation in vivo. PDXs were primarily established as previously described in Caraft-Jauffret et al., and evaluated for ALDEFLUOR-positive CSCs. We injected single cancer cells into fat pads of NOD/SCID mice and monitored tumor growth. When tumor size reached approximately 150 mm.sup.3, Sal and AM5 treatments were performed as indicated above for FIGS. 14 and 15. Tumor growth was compared with that of docetaxel and placebo-treated controls. After 4 weeks of treatment, the animals were sacrificed and the proportion of ALDEFLUOR-positive CSCs was measured in each residual tumor. All animal studies were approved by the French Ministry of Higher Education and Research (authorization number D130554) and ethical committee (number 14), of Aix-Marseille Université. No randomization was used and experimenters were blinded to drug treatments and tissue analyses.

(26) Aldefluor assay. The analysis was processed on single-cell suspension from the PDXs obtained as described above. The ALDEFLUOR Kit (Stemcell™ Technologies) was used to isolate the population with high aldehyde dehydrogenase enzymatic activity using an LSR2 cytometer (Becton Dickinson Biosciences), as previously described in Charafe-Jauffret et al. To eliminate cells of mouse origin from the PDXs, we used staining with an anti-H2Kd antibody (BD Biosciences, 1:200, 20 min on ice) followed by staining with a secondary antibody labeled with phycoerythrin (PE; Jackson Laboratories, 1:250, 20 min on ice).

(27) Limiting dilution assay. Cells from 4-week treated mice were re-implanted into one to four secondary NOD/SCID mice with injection of 10, 50, 500 or 5000 cells for each treated tumor to functionally evaluate the proportion of residual CSCs in each treatment group (Unt., Doc, Sal, AM5) for the PDX 1 and 2. Each mouse that developed a tumor reaching a size of 10 mm was considered as a tumor-bearing mouse.

(28) Estimation of number of CSCs in vivo. The number of mammary gland outgrowths obtained in a fat pad after cell re-implantation is currently used to evaluate the number of stem cells able to repopulate this fat pad. It is based on self-renewal and differentiation abilities, two hallmarks of stem cells. In the same manner, estimation of the number of CSCs can be obtained from tumor outgrowth data collected after limited dilution re-implantation of each group of treatments, based on tumor-initiating capacities of CSCs. For calculation, an online tool is available using Extreme Limiting Dilution Analysis (ELDA) (http://bioinf.wehi.edu.au/software/elda/) for limiting dilution analysis based on the Poisson single-hit model according to the number of outgrowths observed and the number of pat pad re-implanted for each cell dilution.

(29) Histology. Organs from mice were removed at time of sacrifice. For morphological analyses, organs were fixed with 4% paraformaldehyde, paraffin embedded, and 4-μm sections were stained with hematoxylin and eosin (H&E). Sections were scanned at high resolution using a slide scanner (NanoZoomer 2.0-HT, Hamamatsu, Massy, France).

(30) Localization of intracellular iron(II). RhoNox-1 was synthesized according to the original procedure described in Naujokat and Steinhart, J. Biomed. Biotechnol. 2012. Cells were seeded on p-Dishes, then treated as indicated in FIG. 1. Cells were washed with 1× Hank's Balanced Salt Solution (HBSS, 14025-092, Life technologies) three times, then incubated with 5 μM of RhoNox-1/HBSS for 1 h at 37° C., 5% CO.sub.2 and washed with HBSS. Co-staining was performed by adding 100 nM LysoTracker Deep Red and Hoechst 33342 (H1399, Sigma, final concentration of 1 μg/mL) in HBSS. Experiments were performed with a 60×APO TIRF oil immersion objective of Nipkow Spinning Disk confocal system.

(31) Immunoblotting. Cells were treated as indicated, then washed twice with PBS and lysed with 2× Laemmli buffer. Cell extracts were heated at 95° C. for 5 min, sheared through a 26-gauge needle and quantified with a Nanodrop 2000 (Thermo Scientific). Protein lysates (˜100 μg) were resolved by SDS-PAGE electrophoresis and transferred onto PVDF membranes (Amersham). Membranes were blocked with 5% BSA, 0.1% Tween-20/TBS for 1 h. The blots were then probed with the relevant primary antibodies in 5% BSA, 0.1% Tween-20/TBS at 4° C. overnight with gentle agitation. Membranes were washed with 0.1% Tween-20/TBS for 30 min and were incubated with secondary antibodies for 1 h at RT. Antigens were detected by ECL (Amersham). Imaging was performed using a ChemiDoc™ XRS+ System (Biorad).

(32) Perl's reaction. Tumors from mice were removed at time of sacrifice. Tumors were fixed with 4% PFA and embedded in paraffin. To monitor accumulation of ferric iron, 4-μm sections were processed for standard Prussian blue staining. Briefly, slides were deparaffinized in xylenes, rehydrated through a graded ethanol series, and subjected to a 1:1 potassium ferrocyanide solution (2%)/hydrochloric acid solution (2%) for 20 min. Then, slides were washed with distilled water (5 min). Nuclear fast red solution was used to stain nuclei. Images were acquired by light microscopy using an inverted microscope (Eclipse Ti—S, Nikon) and 10×/0.30, 20×/0.50 or 40×/0.785 Plan Fluor objectives (Nikon). Images were captured using a Super high definition cooled colour camera head DS-Ri1 (Nikon) and NIS Elements software (Nikon).

(33) This allows the detection of iron III in tumor tissue.

(34) Lysosomal membrane permeabilization assay. Lysosomal membrane permeabilization (LMP) was assessed by monitoring the release of FITC-Dextran (FD10S, 10 kDa, Sigma) from lysosomes. In brief, cells were incubated with 1 mg/mL FITC-Dextran for 2 h at 37° C. Cells were washed, chased with culture medium for 2 h and treated as indicated for 48 h or with CQ (positive control) for 3 h. Cells were then fixed with 2% formaldehyde/PBS, washed with PBS, mounted and acquired with a 60×APO TIRF oil immersion objective of Nipkow Spinning Disk confocal system.

(35) Annexin V/PI assay. Cells were treated as indicated. After treatment, cell death was quantified using Annexin V-FITC/Propidium Iodide (PI) assay according to the manufacturer's protocol (FITC Annexin V Apoptosis Detection Kit II, 556570, BD Pharmingen™) and analyzed by a LSRFortessa™ flow cytometer (BD Bioscience, San Jose, Calif.). The data were processed using Cell Quest software (BD Biosciences).

(36) Lipid ROS measurements. Cells were treated with Sal or AM5 (0.5 μM) as indicated. Then, cells were trypsinized and washed with PBS. Subsequently, cells were incubated with Bodipy-C11® (D3861, Thermo Fisher Scientific, 2 μM) at 37° C. for 60 min. Next, cells were washed twice with PBS. Oxidation of Bodipy-C11 resulted in a shift of the fluorescence emission peak from 590 nm to 510 nm proportional to lipid ROS generation that was analyzed by LSRFortessa™ flow cytometer (BD Biosciences, San Jose, Calif.). The data were processed using Cell Quest software (BD Biosciences) and FlowJo software (FLOWJO, LLC). For cell imaging, cells were treated as indicated in the main figure (Bodipy-C11®, 1 μM, 1 h), then fixed with formaldehyde (2% in PBS, 12 min) and analyzed by fluorescence microscopy (ex. 488 nm).

(37) Intracellular GSH level quantification. Cells were treated with Sal or AM5 (0.5 μM) as indicated, then harvested and counted. The intracellular GSH level was measured using a commercial kit (ab205811, Abcam) according to the manufacturer's protocol.

(38) RhoNox-1 reduction assay. Experiments were carried in 96-well plate dishes. Reagent stock solutions were freshly prepared using Milli-Q water. 50 μL of solutions of Sal (800 μM), DFO (800 μM) and FeSO4 (80 μM) were mixed prior to addition of 50 μL of RhoNox-1 (40 μM) and completed with the appropriate volume of water to reach a final volume of 200 μL per well. Measurements were performed 30 min after addition of RhoNox-1 to the mix and were recorded using the PARADIGM™ Microplate Detection Platform (ex. 492 nm; em. 580 nm).

(39) Electrothermal atomic absorption spectrometry. Cells were harvested, the supernatant was removed, cell pellets were dried and mineralized by adding 100 μL of concentrated nitric acid (PlasmaPure® 67-69% HNO.sub.3, SCP Science, Baie-d'Urfé, Canada) at 80° C. and further diluted with 400 μL of ultrapure water (Milli-Q®, Millipore, Molsheim, France). Iron was determined in cell mineralizates by means of Electrothermal Atomic Absorption Spectrometry (ETAAS) on a PinAAcle® 900Z spectrometer (Perkin Elmer, Les Ulis, France). This method allows the quantification of cellular iron (total iron).

(40) Small interfering RNA transfection. Suitable small interfering RNAs (siRNA) were designed with the Qiagen RNA interference designer tool for specific down-regulation of FTH1. The sequence used for FTH1 was 5′-GUCCAUGUCUUACUACUUUTT-3′ (S100300251) targeting the sequence following 5′-CTGTCCATGTCTTACTACTTT-3′ and a negative control siRNA for FTH1 (5′-CAUUAGUUUGGGCAGUAUATT-3′, (S103089212). Subconfluent cells were transfected with siRNA in Opti-MEM using the Oligofectamine™ reagent (Invitrogen) for 48 h. Then, cells were treated with OSM (100 ng/mL for 48 h) prior to protein or RNA extraction and flow cytometry analysis for markers as indicated.

(41) Statistical analysis. Data were compared using a two-tailed Student's t-test or a one-way ANOVA as indicated. Data are presented as mean values. Two groups were considered to be significantly different if P<0.05.

Example 1: Lysosomal Iron Localization

(42) The inventors functionalized alkynes in cells by means of in situ click chemistry to detect otherwise invisible compounds, a strategy virtually applicable to any molecule (FIG. 1a). Sal surrogates co-localized with a marker of lysosome in U2OS cells, demonstrating that these compounds physically accumulated in the lysosomal compartment irrespective of the overall charge and without altering the lysosomal pH (FIG. 1b).

(43) In particular, the closely related derivative AM4, devoid of a protonable amine, also accumulated in lysosomes lending strong support to the notion that Sal targets lysosomes. In comparison, AM5 did not target the other organelles studied. Since Sal can interact with alkali metals, and given that intracellular iron is tightly regulated and transits through the lysosomal compartment, we explored the effect of Sal on iron homeostasis starting with the subcellular localization of iron(II) using the specific RhoNox-1 fluorogenic probe. While the staining of iron(II) was diffuse in both untreated HMLER CD24.sup.low cells and in cells treated with AM9, treatment with Sal or AM5 led to a staining that was restricted to the lysosomal compartment (FIG. 1c and FIG. 1d). Indeed, approximately 80% of lysotracker-positive vesicles (e.g. lysosomes) colocalized with the RhoNox-1 probe following treatment with Sal or AM5, whereas only 20-30% of lysotracker-positive vesicles showed colocalization following treatment with AM9 or no treatment (FIG. 1e). Thus, Sal and AM5 sequester iron in lysotracker-positive vesicles (i.e. lysosomes). Treatment with Sal and AM5 also induced a response characteristic of a cytosolic depletion of iron, including increased levels of iron-responsive element-binding protein 2 (IRP2) and transferrin receptor (TfR) along with reduced levels of ferritin (FIG. 1f). A similar response was also observed when cells were treated with the iron(III) chelator deferoxamine (DFO).

(44) These results were consistent with the idea that Sal prevented the release of iron(II) from lysosomes. Remarkably, AM5 promoted a re-localization of ferritin to the lysosomal compartment (FIG. 1g), whose degradation was prevented by pharmacological inhibition of the lysosomal protease cathepsin B using CA-074 (FIG. 1h). The protein NCOA4 has been identified as a cargo receptor of ferritin and is required for its degradation by a selective form of autophagy named ferritinophagy. In strong agreement with the notion that Sal and AM5 activated ferritinophagy, NCOA4 could be detected by western blotting when cells were treated with these drugs in conjunction with an inhibitor of cathepsin B (FIG. 1h). Additionally, Sal and AM5 increased the level of autophagy marker protein LC3-II, which was more pronounced upon inhibition of cathepsin B (FIG. 1h), and ferritin co-localized with LC3 in cells co-treated with AM5 and CA-074 (data not shown). Furthermore, the accumulation of iron(III) in MCF-7 tumors treated with AM5 reflected a cellular response to the targeting of iron homeostasis (FIG. 1i). In line with Sal interacting with iron(II) in cells, nuclear magnetic resonance (NMR) revealed that proton signals of Sal were shifted upon titration with iron(II), whereas signals of the internal standard anthracene remained unaffected, demonstrating that Sal interacted with iron(II) ex cellulo. Conversely, the presence of bipyridine as a competitor prevented Sal from interacting with iron(II) (FIG. 1j). NMR further revealed that addition of 0.5 mol equiv. of FeCl.sub.2 to a methanolic solution of AM5 induced broadening and flattening of specific proton signals (FIG. 1k), including that of the C18-C19 vinylic protons at 6.6 and 6.3 ppm. This data, characteristic of the effect of a paramagnetic metal on the relaxation of protons in close proximity, indicated interactions between AM5 and iron(II) in solution. Interestingly, previous X-ray crystallography studies have revealed that the vinylic protons are topologically close to the sodium ion inside the cavity formed by Sal around the metal in a co-crystal. Thus, the pronounced effect of iron(II) onto the signals of these two protons suggested that iron(II) may occupy a similar position inside the folded molecule. The sub-stoichiometric amount of FeCl.sub.2 required to promote this effect was consistent with a 2:1 AM5:iron(II) stoichiometry. However, this data could also reflect a time-averaged set of signals between bound and free AM5, indicating a fast exchange that cannot be resolved within the timescale of the NMR experiment. In comparison, the proton signals of naphthalene (Napht), an organic small molecule devoid of heteroatoms and therefore unable to chelate iron, remained unaffected, and thus could be used as internal standard (e.g. unaltered signal at 7.8 ppm) to compare the intensity of signals of AM5 between samples. Strikingly, addition of a slight excess of the iron(II) chelator 2,2′-bipyridine (Bipy) to a mixture of AM5, Napht and FeCl.sub.2, led to the occurrence of new proton signals of free and bound Bipy along with the concomitant rescue of the signals previously observed for the unbound AM5. These data indicated that under these conditions, Bipy displaces the metal from AM5, in line with the fact that AM5 is a looser iron interacting partner compared to Bipy. It is noteworthy that a similar trend was observed for Sal (FIG. 11), although the shielding effect of iron(II) was more pronounced on the proton signals of AM5. Moreover, while iron(II) promoted the formation of byproducts of Sal over time, AM5 was found to be stable under these conditions. These properties of AM5 provides a rationale for the higher potency of this synthetic derivative compared to Sal.

(45) Altogether, these data supported a model whereby the lipophilic natural product accumulated in the lysosomal compartment and interacted with iron, thereby preventing release of the metal into the cytosol and initiating the appropriate response to replenish the available pool of iron.

Example 2: Iron Concentration, Iron Chelation and Iron Related Biomarkers

(46) Iron can catalyze the production of reactive oxygen species (ROS) via Fenton chemistry. Treatment of U2OS, HMLER CD24.sup.low, and iCSCL-10A2 cells with Sal and AM5 led to a significant increase in lysosomal ROS in all cell lines, which mirrored the accumulation of iron in this organelle (data not shown).

(47) The implication of iron and ROS in the phenotype induced by Sal and AM5 hinted towards the activation of a regulated form of necrotic cell death, termed ferroptosis, detected in HMLER CD24.sup.low and iCSCL-10A2 cells after 72 h treatment (FIGS. 12b and c). In support of this, Bodipy-C11 staining indicated the presence of lipid peroxidation in HMLER CD24.sup.low and iCSCL-10A2 cells treated with Sal or AM5 (FIG. 12a). In addition, cell death induced by Sal or AM5 could be partially prevented by the ferroptosis inhibitor ferrostatin-1, whereas the apoptosis and necrosis inhibitors Z-VAD-FMK and necrostatin-1, respectively, had no effect on the cell death profiles (FIGS. 12d and e). Furthermore, diminution of endogenous levels of the ROS scavenger glutathione (GSH), a hallmark of ferroptosis, could be detected in HMLER CD24.sup.low and iCSCL-10A2 cells treated with Sal or AM5 (FIG. 12f).

(48) Sal and AM5 also induced a permeabilization of the lysosomal membrane of HMLER CD24.sup.low cells as observed from the release of bulky lysosomal dextran into the cytosol (data not shown). Co-treatment with the ROS scavenger N-acetyl-L-cysteine (NAC) partly prevented Sal and AM5 from killing these cells (data not shown), and DFO reduced ROS levels in treated cells (data not shown), exhibiting a protective effect against Sal and AM5 (FIG. 2a). Finally, cathepsin B inhibition decreased the ability of AM5 to induce the production of ROS (data not shown) and rescued cell viability (FIG. 2b). These data indicated that the accumulation of iron in lysosomes promoted by Sal and AM5 led to the production of lysosomal ROS followed by lysosomal dysfunction and cells death. Strikingly, HMLER CD24.sup.low cells contained significantly higher levels of iron, iron-uptake protein TfR and active cathepsin B compared to control cells (FIGS. 2c and 2d). These findings underlie the selective effect of Sal on HMLER CD24.sup.low cells and raise a putative role of iron in the maintenance of CSCs.

(49) Wnt1 protein level was higher in HMLER CD24.sup.low compared to control cells pointing towards iron as a potential driver of CSCs (FIG. 2d).

Example 3: Sal and Sal Analogs Alter CSC Maintenance in the In Vivo PDX Model

(50) AM5 selectively targeted the ALDH+ subpopulation of another model of CSCs, namely iCSCL-10A2 cells, more effectively than Sal (FIG. 13), and exhibited little toxicity against primary breast cells (data not shown). These data illustrate the general susceptibility of AM5 to selectively target CSCs. Sal and AM5 were further shown to reduce tumor growth in two early passage patient-derived xenografts (PDXs), where the clinically approved drug docetaxel (Doc) was less effective (data not shown). This effect was associated with a reduced ratio of ALDH+ cells (FIG. 14), and a decreased tumor-seeding capacity of tumor cells treated in vivo without detectable toxicity at effective doses, with AM5 being more potent than both Sal and Doc (FIG. 15). In particular, treatment with Doc alone shows a lower response than treatment with either AM5 or AM5+Doc. These data further provide solid evidence that Sal analogs, such as AM5, selectively target CSCs in vivo.

Example 4: Effect of AM5, Taxol and Combination Thereof on the Proliferation of HMLER Cd24-Cells

(51) AM5, Taxol and a combination of AM5 and Taxol were also assessed for their capability to inhibit cell proliferation and formation of mammospheres (FIG. 16). The combination of AM5 at 15 nM and Taxol at 5 nM inhibits cell proliferation and mammosphere formation with an improved efficacy as compared to AM5 alone at either 15 nM or 5 nM.

Example 5: Iron Concentration in Various Breast Cells Lines and Tumorsphere Formation

(52) Table 1 below represents the measurement of iron in a wide range of cell lines. Iron concentration is measured as in Example 2.

(53) TABLE-US-00001 TABLE 1 Iron Total Iron Total Iron per 3 × 10.sup.6 cells per cell Sensibility Tumor cell Subtype Cell model (ng) (pg) Sphere to AM5 Non-tumoral HBL100 114 0.04 + First Immortalized HMLE W2 99 0.03 +++ h Tert + Tand t SV40 + ras HMLER CD24.sup.low 290 0.10 +++ h Tert + T and t de HMLER shECAD 150 0.05 ++ SV40 + ras + sh ECAD h Tert+T and t HMLER shGFP 115 0.04 ++ SV40 + ras + sh gfp CSTN BT549 116 0.04 ++ Resistant MDA-MB-361 120 0.04 +++ Second h Tert + T and t SV40 + ras HMLER CD24.sup.high 69 0.02 ++ First H+ MCF-7 83 0.03 ++ First MDA-MB-134 65 0.02 ++ Second CSTN MDA-MD-157 79 0.03 +/− Second BT474 60 0.02 + Second H+ T47D 97 0.03 ++ Resistant H+ Zr75.1 57 0.02 +/− First CSTN MDA-MB-231 34 0.01 − Second Colon SW620 38 0.01 + Third Colon SW480 52 0.02 ++ Third Iron Total iron Total iron per 7 × 10.sup.6 cells per cell Subtype Cell model (ng) (pg) h Tert + Tand t SV40 + ras HMLER CD24l.sup.ow 553 ± 29 0.08 h Tert + T and t SV40 + ras HMLER CD24.sup.high 143 ± 20 0.02

(54) TABLE-US-00002 TABLE 2 Name Essential specificities HBL100 Human mammary epithelial cell line obtained from primary cultures of cells derived from an early lactation sample of human milk (from ATCC). HMLE W2 Human mammary epithelial cell line infected with a retrovirus carrying hTERT, SV40 (R. A. Weinberg, Whitehead Institute, Massachusetts Institute of Technology, USA) HMLER ID2 Human mammary epithelial cell line infected with a retrovirus carrying hTERT, SV40 and the oncogenic allele HrasV12 (R. A. Weinberg, Whitehead Institute, Massachusetts Institute of Technology, USA) HMLER CD24.sup.low HMLER CD44.sup.high/CD24.sup.low not expressing E-cadherin and expressing Vimentin (was obtained from A. Puisieux INSERM) HMLER shGFP HMLER cells expressing a control shRNA (shCtrl). (ctrl) Generated by infection with retrovirus encoding the pWZL-GFP plasmid. (R. A. Weinberg, Whitehead Institute, Massachusetts Institute of Technology, USA) HMLER shECAD Transformed HMLER breast cancer cells displaying a short hairpin RNA (shRNA)-mediated inhibition of the human CDH1 gene, which encodes E-cadherin. Generated by infection with retrovirus encoding the pWZL-GFP plasmid. (R. A. Weinberg, Whitehead Institute, Massachusetts Institute of Technology, USA) MCF-7 Human ductal breast epithelial tumor cell line classified in Estrogen/Progesterone Receptor (ER/PR) positive group and luminal A (from ATCC). Zr75.1 Human ductal breast epithelial tumor cell line, classified in Estrogen/Progesterone Receptor (ER/PR) and HER-2 positive group and luminal A (from ATCC). MDA-MB-361 Human ductal breast epithelial tumor cell line, classified in Progesterone Receptor (PR) and HER-2 positive group and luminal B (from ATCC). These cells were isolated from a metastatic site in the brain. MDA-MB-134 Human ductal breast epithelial tumor cell line classified in Estrogen/Progesterone Receptor (ER/PR) positive group and luminal B (from ATCC). MDA-MD-157 Human ductal breast epithelial tumor cell line, classified in Estrogen/Progesterone Receptor (ER/PR) and HER-2 negative group and Basal (from ATCC). MDA-MB-231 Human ductal breast epithelial tumor cell line, classified in Estrogen/Progesterone Receptor (ER/PR) and HER-2 negative group and Basal (from ATCC). BT474 Human ductal breast epithelial tumor cell line, classified in Progesterone Receptor (PR) and HER-2 positive group and luminal B (from ATCC). Hs578T Human ductal breast epithelial tumor cell line, classified in Estrogen/Progesterone Receptor (ER/PR) and HER-2 negative group and Basal (from ATCC). BT20 Human ductal breast epithelial tumor cell line, classified in Estrogen/Progesterone Receptor (ER/PR) and HER-2 negative group and Basal (from ATCC). SW620 Colon tumor cells; derived from metastatic site: lymph node (from ATCC). SW480 Colon tumor cells; derived from a primary adenocarcinoma of the colon (from ATCC). BT549 Human ductal breast epithelial tumor cell line, classified in Estrogen/Progesterone Receptor (ER/PR) and HER-2 negative group and Basal (from ATCC). T47D Human ductal breast epithelial tumor cell line classified in Estrogen/Progesterone Receptor (ER/PR) positive group and luminal A (from ATCC).
These results highlight the link between the concentration of iron and tumorsphere formation.

Example 6: Iron Homeostasis would be a Driver of Cancer Stem Cell Phenotype

(55) 1. Wnt1 protein level was higher in HMLER CD24.sup.low compared to control cells (FIG. 2d). 2. E-cadherin (E-cad) is essential to maintenance of epithelial layers. E-cad expression is a marker of EMT. E-cad protein level was higher in W1, W2 and HMLER ID2 compared to HMLER stem-like CD24.sup.low cells (FIG. 5). In contrast, ferritin protein level was higher in HMLER stem-like CD24.sup.low cells compared to parental isogenic cells (W1, W2 and HMLER ID2). The loss of E-cadherin and gain of ferritin may be associated with cancer stem cell phenotype. 3. The connection between epithelial-mesenchymal (E-M) plasticity and CSC properties has been paradigm-shifting, linking tumor cell invasion and metastasis with therapeutic recurrence. Cytokines top the list of tumor-associated secreted factors, and are likely to have important effects on the pathways that govern CSC and E-M plasticity. In addition to TGFβ, additional cytokines and growth factors including Oncostatin M (OSM) have also been implicated in inducing EMT (FIG. 6) and by extension, CSC properties. The ability of these cytokines and growth factors to induce CSC properties concomitant with EMT may explain why their presence in the TME correlates with poor patient outcomes. To investigate whether iron homeostasis play a role in inducing CSC properties concomitant with EMT, the Inventors performed OSM treatment to promote EMT in breast cancer MCF-7 cells. EMT induction was examined by both the increase of two EMT markers Fibronectin and Snail, and the decrease of E-cad protein levels (FIG. 7 left). It could be observed that OSM also induces an increase in ferritin protein level in MCF7 cells (FIG. 7 right). The knock-down of Ferritin expression by RNAi slows down the OSM-induced increase of fibronectin protein level. These data indicate that ferritin expression is associated with EMT. 4. The Inventors next performed OSM treatment to increase the CD44.sup.highCD24.sup.Low/− cancer stem cell (CSC) population in breast cancer MCF7 cells (FIGS. 8a & 8b: % of CSC shift from 0.42% to 4.84%). Whereas the % of CD44.sup.high population increases from 33% to 76%, the % of CD24.sup.high population remains unchanged (FIG. 8a). Similar to these data, it was observed that the mean fluorescence intensity (IF) increases for CD44 in the OSM treated cells but not for the CD24 population. Knock down of Ferritin expression by RNAi alone slightly increases the IF of the CD24 population, but not for the CD44 population (FIG. 8a). Interestingly, it was observed that the Knock down of Ferritin expression interferes with the OSM-mediated induction of CD44.sup.highCD24.sup.Low/− in MCF-7 cells (FIGS. 8a and 8b). Furthermore, knocking down ferritin impaired the ability of cytokine oncostatin M (OSM) to induce a stem-like phenotype as defined by the percentage of ALDH+cells (FIG. 8d), indicating that CSC generation in inhibited. 5. To investigate functional difference between non-CSCs and CSCs, we first examined the effect of Sal and AM5 treatment on matched isogenic non-CSCs (HMLER CD24.sup.high) cells, CSCs (HMLER CD24.sup.low) cells and a new CSC model available in the laboratory. In HMLER CD24.sup.low (see FIG. 1f) and iCSCA2 cells, Sal and AM5 induced a response characteristic of a cytosolic depletion of iron, including increased levels of iron-responsive element-binding protein 2 (IRP2) and transferrin receptor (TfR) along with reduced levels of ferritin (FIG. 1f and FIG. 9). Although Sal and AM5 reduced the level of ferritin in non-CSC (HMLER CD24.sup.high ID2) cells, we did not observe a response characteristic of a cytosolic depletion of iron, including increased levels of IRP2 and TfR. 6. The Inventors next examined the transferrin uptake difference between non-CSC and CSC. To do this, HMLER CD44.sup.highCD24.sup.Low cells and their counterpart (HMLER CD44.sup.Low/− CD24.sup.high ID2) are co-cultured and transferrin uptake assessed. HMLER CD44.sup.high CD24.sup.Low stem like cells were identified by CD44 expression to discriminate ID2 cells. Interestingly, CD44.sup.high cells uptake transferrin more than CD44.sup.Low (FIGS. 10a and c left). They next used cell tracker method staining to discriminate HMLER CD44.sup.high CD24.sup.Low stem-like cells from ID2 cells (FIGS. 10a and c left). APC positive cells uptake transferrin more than APC negative cells. These data indicate that HMLER CD44.sup.high CD24.sup.Low stem-like cells uptake transferrin more than their isogenic HMLER ID2 counterpart. 7. The Inventors then investigated the effect of transferrin to induce CSC properties. It could be demonstrated that TGF-31 increases the percent of CD44.sup.high/CD24.sup.Low/− population (8% VS 21%). Transferrin uptake potentiates the TGF-beta-mediated induction of CD44.sup.highCD24.sup.Low/− in HMLER cells. These data indicate that iron homeostasis would be a driver of cancer stem cell phenotype. 8. The Inventors finally investigated the specific targeting of CSCs. It was shown that AM5 selectively targets ALDH+ CSCs, leading to a decreased ratio of ALDH+ cells and a decreased tumor-seeding capacity (FIG. 13-15). Of note, AM5+ Doc treatment generates a higher response than treatment with Doc alone. Similarly, AM5+ Taxol generates a higher response than treatment with Taxol alone, as determined according to number and size of in vitro mammosphere formation (FIG. 16).

REFERENCES

(56) Perou et al. 2000, Molecular portraits of human breast tumours, Nature 406, 747-752. Gusterson et al. 2005, Immunohistochemistry: Basics and Methods, Cancer treatment: the combination of vaccination with other therapies. Andersen et al. 2008, Cancer Immunology Immunotherapy, 57(11): 1735-1743. Horan P K et al, Fluorescent cell labeling for in vivo and in vitro cell tracking Methods Cell Biol. 1990. Quoix et al. 2011, Therapeutic vaccination with TG4010 and first-line chemotherapy in advanced non-small-cell lung cancer: a controlled phase 2B trial., Lancet Oncol. 12(12): 1125-33. Liseth et al. 2010, Combination of intensive chemotherapy and anticancer vaccines in the treatment of human malignancies: the hematological experience. Hirooka et al. 2009, A combination therapy of gemcitabine with immunotherapy for patients with inoperable locally advanced pancreatic cancer, Pancreas 38(3): e69-74. Baskar et al., Cancer and Radiation Therapy: Current Advances and Future Directions, Int. J Med Sci. 9(3): 193-199. Gupta et al., 2009, Identification of Selective Inhibitors of Cancer Stem Cells by High-Throughput Screening, Cell, vol. 138, p. 645-659. Charafe-Jauffret et al., 2013, ALDH1-positive cancer stem cells predict engraftment of primary breast tumors and are governed by a common stem cell program. Cancer Res. 73, 7290-7300. Naujokat and Steinhart, 2012, Salinomycin as a drug for targeting human cancer stem cells. J. Biomed. Biotechnol. DOI: 10.1155/2012/950658. Miller et al., 2011, An Iron Regulatory Gene Signature Predicts Outcome in Breast Cancer, Cancer Res. 71(21): 6728-6737.