![Preferential Killing of Breast Tumor Initiating Cells by N,N-Diethyl-2- [4-(Phenylmethyl)Phenoxy]Ethanamine/Tesmilifene Tao Deng,1Jeff C](http://data.docslib.org/img/94d3c5a5bc79cf6dd72ddbf4d911dddd-1.webp)
CancerTherapy: Preclinical
Preferential Killing of Breast Tumor Initiating Cells by N,N-Diethyl-2- [4-(Phenylmethyl)Phenoxy]Ethanamine/Tesmilifene Tao Deng,1Jeff C. Liu,1Kathleen I. Pritchard,2,3 Andrea Eisen,2,3 and Eldad Zacksenhaus1, 3,4, 5
Abstract Purpose: N,N-Diethyl-2-[4-(phenylmethyl)phenoxy]ethanamine (DPPE; tesmilifene) is thought to potentiate the antineoplastic effect of cytotoxic drugs. In a phase III randomized trial for metastatic breast cancer using doxorubicin with or without DPPE, addition of the latter resulted in a significant improvement in overall survival and a trend toward a difference in progression-free survival but, paradoxically, no difference in objective tumor response. Here we tested the hypothesis that DPPE targets breast tumor-initiating cells (TICs). Experimental Design: Human breast TICs from pleural effusions were identified as CD44+:CD24-/low cells by flow cytometry and functionally by their ability to form nonadherent spheres in culture. Mouse mammary TICs from two different models of breast cancer were identified as cells capable of initiating spheres in culture and secondary tumors following transplantation into the mammary gland of syngeneic mice. Results: We show that at physiologically attainable concentrations, treatment with DPPE alone reduced tumorsphere formation and viability of CD44+:CD24-/low breast cancer cells.The kinetics of killing varied for the different breast tumor cells and required continuous exposure to the drug. Whereas doxorubicin killed CD44+:CD24-/low and CD44-:CD24+ cells equally well, DPPE induced apoptosis preferentially in CD44+:CD24-/low cells. Treatment of Her2/Neu+ mammary tumor cells with DPPE in vitro efficiently killed TICs, as determined by flow cytometry and transplantation assays; DPPE further cooperated with doxorubicin to completely eradicate tumorigenic cells. Conclusions: Our results show that continuous treatment with DPPE alone directly targets breast TICs, and provide rationale to test for cooperation between DPPE and known drugs with efficacy toward breast cancer subtypes.
Breast cancer mortality is declining in Western countries (e.g., doxorubicin) and the taxanes (e.g., docetaxel; ref. 3). primarily due to improvements in breast cancer screening and However, at best, only 50% of women treated with these agents in the adjuvant treatment of early stage disease (1, 2). will respond to therapy (4). Thus, there is an urgent need for Metastatic breast cancer remains an incurable illness for novel approaches to combat metastatic breast cancer. which few treatments have actually been shown to prolong There is growing evidence that cancer is organized in survival. The most active chemotherapeutic agents in the a hierarchy in which relatively rare tumor-initiating cells treatment of metastatic breast cancer include the anthracyclines (TICs; also known as cancer stem cells) are capable of disseminating cancer, whereas the majority of tumor cells have lost this potential (5–9). This hierarchy has been first Authors’ Affiliations: 1Toronto General Research Institute-University Health documented in acute myeloid leukemia (10, 11). More 2 3 4 Network; SunnybrookOdette Cancer Centre; and Medicine, Medical recently, TICs (cancer stem cells) were identified in breast Biophysics, and 5Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada (12) and brain (13) cancers as well as other tumor types Received 7/3/08; revised 9/8/08; accepted 9/9/08. (14–21). Breast TICs express early epithelial markers such as Grant support: YMBiosciences, Inc., a matching grant from the Canadian Institute the cell surface marker CD44, whereas nontumorigenic cells for Health Research, the National Cancer Institute of Canada-Terry Fox Breast express CD24 (12). CD44+:CD24-/low cells, but not CD24+ Cancer Group Grant 13005, and the Ontario Institute for Cancer Research. The costs of publication of this article were defrayed in part by the payment of page cells, sorted from malignant pleural effusions give rise to new charges. This article must therefore be hereby marked advertisement in accordance tumors when injected into the fat pad of immunocompromised with 18 U.S.C. Section 1734 solely to indicate this fact. NOD/SCID mice (12). Note: Supplementary data for this article are available at Clinical Cancer Research Mouse models of cancer provide the power of mouse Online (http://clincancerres.aacrjournals.org/). genetics, the availability of ample samples of genetically Requests for reprints: Eldad Zacksenhaus, Division of Cell and Molecular Biology,Toronto General Research Institute-University Health Network, 67 College defined tumors and syngeneic transplantation assays. Several Street, Room 407,Toronto, Ontario, Canada M5G 2M1. Phone: 416-340-4800, ext. groups have recently identified TICs in mouse models of breast 5106; Fax: 416-340-3453; E-mail: [email protected]. Andrea Eisen, cancer (22–25). Our group has shown that TICs derived from Preventive Oncology,Toronto SunnybrookRegional Cancer Centre, 2075 Bayview Her2/Neu+ as well as Wnt1-induced mammary tumors are Avenue, Toronto, ON M4N 3M5, Canada. Phone: 416-480-4617; E-mail: andrea. [email protected]. capable of forming spheres under nonadherent conditions in a F 2009 American Association for Cancer Research. defined medium containing epidermal growth factor and doi:10.1158/1078-0432.CCR-08-1708 fibroblast growth factor (23). The tumorsphere-forming units
www.aacrjournals.org 119 Clin Cancer Res 2009;15(1) January 1, 2009 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. CancerTherapy: Preclinical
prednisone in the treatment of hormone refractory prostate Translational Relevance cancer (39). Understanding the mechanism by which DPPE affects chemotherapy and overall survival may lead to Previous models implicated N,N-diethyl-2-[4-(phenyl- improved therapy (39). Using four different breast cancer methyl)phenoxy]ethanamine (DPPE; tesmilifene) in the models, we now show that treatment with DPPE alone potentiation of cytotoxicity by antineoplastic drugs. In a preferentially kills TICs. phase IIIrandomized trial for metastatic breast cancer using doxorubicin with or without DPPE, addition of the latter resulted in a significant improvement in overall survival Materials and Methods and a trend toward a difference in progression-free survival but, paradoxically, no difference in objective tumor re- Patient recruitment and consent. Subjects with metastatic breast sponse. We have now shown that continuous exposure to cancer and malignant pleural effusions were recruited from the clinical practices of a large, tertiary care cancer center under an Research pharmacologically attainable doses of DPPE effectively Ethics Board–approved protocol. All patients provided written kills tumor-initiating cells in four different models of breast informed consent. Pleural fluid samples were obtained by standard cancer. Our results suggest that DPPE-based therapies thoracocentesis. should involve continuous or frequent administration of Enrichment of lineage-depleted breast cancer cells from pleural the drug. Furthermore, more potent DPPE derivatives effusions. Pleural effusions were obtained and processed within 2 to with better specificity may be developed using the tumor- 4 h after thoracocentesis. The samples were centrifuged at 1,000 rpm initiating cell ^ specific assays presented in our study. for 10 min and cell pellets were washed with HBSS plus 2% Finally, using these assays, combinatorial drug regimens heat-inactivated fetal bovine serum (HIBS) and 2 mmol/L EDTA. with DPPE plus other antineoplastic agents with efficacy Enrichment of epithelial cells was achieved using a negative selection kit A Â 7 toward breast cancer subtypes should be assessed. StemSep (StemCell Technologies, Inc.). Briefly, 100 L per 5 10 cells of a cocktail containing biotinylated antibodies against human hematopoietic cells (CD2, CD14, CD16, CD38, CD45, and CD66) and glycophorin A were added. After 30-min incubation on ice, 60 AL (TFU) were indistinguishable by several criteria from TICs. of magnetic colloid were added and incubated for another 30 min on Specifically, TFUs and TICs were identified in the same fraction ice. Finally, the cells were passed trough magnetic columns and the of sorted cells, and cultured tumorspheres retained the supernatant containing epithelial cells was collected. The purified cells differentiation capacity, cell surface markers, and tumorigenic were cultured either as adherent monolayer with DMEM plus 10% fetal potential of freshly isolated TICs. Thus, TFUs can be used as a bovine serum or under nonadherent conditions (see below) plus surrogate for TICs in vitro. penicillin/streptomycin. Implicit in the cancer stem cell model is that targeted killing Mice and PCR-based genotyping. MMTV-Wnt1 (40) and MMTV-Neu of TICs, rather than the bulk of nontumorigenic cells, may be (41) transgenic mice, both on pure FvB background, were maintained in accordance with the Canadian Animal Care Council. The mice were curative (5, 6, 26). However, TICs are not only rare but, in some genotyped by PCR analysis of DNA extracted from tail biopsies instances, are also shown to be more resistant to radiation and using the following primers: Wnt1, forward 5¶-GGACTTGCTTCTCTTCT- cytotoxic therapy (27–31). The identification of agents that CATAGCC and reverse 5¶-CCACACAGGCATAGAGTGTCTGC; Neu, enhance the specificity of antineoplastic drugs toward TICs or forward 5¶-CTAGGCCACAGAATTGAAAGATCT and reverse 5¶-gTAGG- that preferentially kill TICs is therefore of great interest. TGGAAATTCTAGCATCATCC. Mice were sacrificed when tumors were The tamoxifen analogue N,N-diethyl-2-[4-(phenylmethyl)- 0.5 to 1 cm in diameter. phenoxy]ethanamine (DPPE; tesmilifene) has been implicated Enrichment of lineage-depleted epithelial cells from mouse mammary in the inhibition of histamine binding to CYP 3A4, a P450 tumors. Mammary tumors were resected and minced by sterile razor isozyme that metabolizes several antineoplastic agents, as well blades, washed with HBSS plus 2% HIBS, and digested with 100 units/ as in the inhibition of p-glycoproteins involved in multidrug mL collagenase/hyaluronidase (Stem Cell Technologies) for 30 min at 37jC. The resulting samples were passed through a 40-Am cell strainer resistance (32–34). It is therefore thought that DPPE exerts its (BD falcon) and centrifuged at 450 Â g for 5 min. The cell pellets effect by potentiating the cytotoxicity of various antineoplastic were washed with HBSS + 2% HIBS, and lineage-negative (Lin-) drugs (35–37). The exact mechanism of action and the epithelial cells were isolated using a negative selection EasySep Kit biological targets of DPPE are not known. (StemCell Technologies). Briefly, 10 AL of FcR blocker were added In a phase III randomized trial for metastatic breast cancer per 1 Â 108 cells to prevent nonspecific interaction. Biotinylated using doxorubicin with or without DPPE, it was observed that antibodies (0.5 Ag/106 cells) against hematopoietic (anti-CD45 and the addition of the latter resulted in a significant improvement anti-TER119; StemCell Technologies), stromal (anti-CD140a; Bioscien- in overall survival and a trend toward a difference in ces), and endothelial (anti-CD31; BD PharMingen) cells were added progression-free survival but, paradoxically, no difference in and incubated for 15 min at room temperature. Antibody conjugation objective tumor response (38). The effect of adding DPPE was to magnetic nanoparticles was achieved through 15-min incubation with EasySep Biotin Selection cocktail followed by 10 min with EasySep not reproduced in the most recent trial of epirubicin plus B Magnetic Nanoparticles. An EasySep Magnetic device was used for cyclophosphamide +/- DPPE in na ve, previously untreated 5 min to separate the Lin+ cells that attached to the tube surface from patients with metastatic breast cancer, possibly because of the the Lin- epithelial cells that remained in the supernatant. 6 excellent response by the EC arm alone, although results from In vitro differentiation and immunocytostaining. Sorted long-term follow-up are not yet available. Another trial with CD44+:CD24-/low cells were plated on collagen 1–coated coverslips docetaxel +/- DPPE is ongoing (YM Biosciences, Inc.). DPPE (BD Biocoat) in growth medium (DMEM/F-12, 10% fetal bovine has also been shown to have activity with mitoxantrone and serum, 5 Ag/mL insulin, 1 Ag/mL hydrocortisone, 5 ng/mL epidermal growth factor, and penicillin/streptomycin) for 5 d and then differentiated for 24 h in medium containing DMEM/F-12, 5 Ag/mL 6 YM Biosciences, unpublished. insulin, 1 Ag/mL hydrocortisone, 3 Ag/mL prolactin, and penicillin/
Clin Cancer Res 2009;15(1) January 1, 2009 120 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. Preferential Killing of Breast TICs by DPPE/Tesmilifene streptomycin. After fixation in 3.5% paraformaldehyde, cells were Colorimetric reading at 570 nm was done using a microplate reader permeabilized in OPAS buffer (100 mmol/L PIPES, 1 mmol/L EGTA, (Molecular Devices). For MTT assay of tumorsphere cells, the protocol 100 mmol/L KOH, 4% polyethylene glycol 8000, 0.1% Triton was modified as follows. Briefly, MTT was added to sphere cells and X-100). Rabbit anti–smooth muscle actin (1:200 dilution; Novus incubated for 2 h at 37jC. The supernatant including spheres was Biologics), mouse anti–keratin 18 (1:200 dilution; Fitzgerald), rabbit transferred to tubes and centrifuged at 500 Â g for 5 min. After anti–keratin 14 (1:200 dilution; Panomics), and secondary goat aspirating the supernatant, the sphere cells were solubilized by adding anti-rabbit Alexa 488 (green; 1:200 dilution; Molecular Probes) or goat 100 AL DMSO to liberate the dye. anti-mouse Alexa 568 (red, 1:200 dilution, Molecular Probes) Transplantation. Dissociated tumorsphere cells were resuspended were used for immunocytostaining. Nuclei were visualized with in 10 AL serum-free DMEM/F-12 Ham medium, mixed at 1:1 ratio with 4¶,6¶-diamidino-2-phenylindole (Sigma). 10 AL Matrigel (BD Bioscience), placed on ice, and then injected into # Drug treatment. A stock solution of DPPE in sodium acetate was 4 mammary glands of syngeneic FvB mice or immunocompromised obtained from YM Biosciences. Doxorubicin (Sigma) was dissolved in Rag1-/- mice anesthetized with isoflurane. Sites of needle injection and N,N-dimethylformamide at 1 mg/mL. Before experiments, drugs were incision were sealed with liquid bandage (NewSkin). further diluted in serum-free DMEM/F-12 Ham culture medium. For Statistical analysis. All data are represented as mean F SD. adherent cultures, 50,000 MCF7 or pleural effusion cells were seeded Two-tailed distribution Student’s t test was carried out. P < 0.05 (*) onto 24-well plates. For nonadherent cultures, dissociated cells were and P < 0.01 (**) were considered statistically significant. seeded onto 24-well ultra-low attachment plates (Corning, Costar) at a cell density of 50,000 per well. Tumorsphere formation assay. Enriched tumor cells from pleural Results effusions or MCF7 cells were seeded onto 6- or 24-well ultra-low attachment plates (Corning, Costar) in DMEM/F-12 Ham medium DPPE inhibits the growth of MCF7 cells when cultured (Sigma) containing 10 ng/mL basic fibroblast growth factor (Sigma), as tumorspheres but not as monolayer, whereas doxorubicin 20 ng/mL epidermal growth factor (Sigma), 5 Ag/mL insulin (Sigma), kills both. We first tested the effect of DPPE on the breast j and 0.4% bovine serum albumin and cultured at 37 C, 5% CO2. For cancer cell line MCF7. The cells were plated under adherent - mouse tumorspheres, Lin mammary epithelial cells were plated on conditions; treated with DPPE, doxorubicin, or both drugs; and ultra-low attachment plates (Corning, Costar) in DMEM/F-12 Ham the effect of cell viability was assessed by the colorimetric medium containing 20 ng/mL basic fibroblast growth factor (Sigma), MTT assay, which measures mitochondrial reductase activity. 20 ng/mL epidermal growth factor, 4 Ag/mL heparin (Sigma), and B-27 Exposure of MCF7 cells to clinically attainable concentration of supplement (1:50 dilution; Life Technologies). Medium was refreshed A twice a week. Spheres were mechanically and enzymatically dissociated DPPE (5 mol/L; refs. 42, 43) had only a minor effect, with a every 2 wk in 1Â (0.25%) trypsin-EDTA solution (Life Technologies) 20% reduction in total cell survival relative to cells treated with for 1 min at room temperature, followed by passing through 25-gauge vehicle alone (dimethyformamide; Fig. 1A). Under these needles seven times. The dissociated tumorsphere cells were added to conditions, clinically relevant concentrations of doxorubicin 24-well ultra-low attachment plates at a cell density of 50,000 per well (0.05 and 0.1 Ag/mL) led to about 45% and 75% reductions in and treated as indicated in each figure. live MCF7 cells, respectively. Loss of viability was not enhanced Flow cytometry analysis and sorting. Human tumor cells were by combined treatment with DPPE and doxorubicin. In this A Â 5 washed, counted, and resuspended in 50 L (per 5 10 cells) of and other experiments, treatment with vehicle alone (dimethy- HBSS plus 2% HIBS. Ten microliters of anti-CD44 [allophycocyanin formamide) did not have any significant effect on cell viability (APC)] and antihuman CD24 (FITC; both from BD Biosciences) were relative to no treatment (data not shown). added for 30 min on ice followed by two washes with HBSS/2% HIBS. The stained cells were washed again and resuspended in HBSS/2% HIBS We next tested the effect of DPPE on the formation of MCF7 containing the viability dye 7-amino-actinomycin D (BD Pharmingen) tumorspheres cultured in ultra-low attachment plates in and passed through a cell strainer. For mouse mammary tumors, defined medium containing epidermal growth factor and we used anti-CD49f conjugated to R-phycoerythrin (CD49f-PE) and fibroblast growth factor. Under these nonadherent conditions, anti-CD24 conjugated to FITC (CD24-FITC; BD Pharmingen). Flow tumorspheres are formed by stem cells or TICs, whereas cytometry analysis was done on a FACSCalibur (Becton Dickinson) and more differentiated cells tend to die by anoikis (44). Thus, cell sorting was done on a FACSAria-13 color sorter (Becton Dickinson) MCF7 cells were trypsinized to obtain single cells, plated under at 30 psi, gating on the 7-amino-actinomycin D–negative (live) cells. nonadherent conditions, and treated with the indicated drugs. Annexin V and cell cycle analysis. Apoptosis was detected using The number of spheres in each culture was determined 7 days PE-Annexin V (BD Biosciences). For pleural effusion cells, triple later. In contrast to the modest effect on MCF7 cells grown as staining was carried out by first adding CD44-APC and CD24-FITC A for 20 min on ice. The cells were washed with HBSS/2% HIBS and monolayer (Fig. 1A, top), DPPE (5 mol/L) treatment reduced F F resuspended in Annexin V binding buffer. Annexin V-PE was added and tumorsphere formation by 79.5 3.5% (mean SD) relative incubated for 15 min at room temperature followed by flow cytometry. to vehicle-treated cultures (Fig. 1A, bottom). Treatment with Cell cycle analysis was done on cells stained with propidium iodide/ both DPPE and doxorubicin had a moderate additive effect, RNase (BD Pharmingen). One-milliliter aliquots of cell suspensions leading to 88.6 F 0.86% suppression. To test whether the were pelleted at 400 Â g for 5 min and fixed in 70% ethanol at 4jC for inhibition of MCF7 tumorsphere formation was cytostatic or 5 at least 60 min, and then f1 Â 10 cells were washed with 3 mL PBS cytotoxic, the cells were treated for 2 weeks under nonadherent A (PBS) and stained with 500 L propidium iodide/RNase. Samples were conditions with the various drugs. The tumorspheres were then incubated for 15 min in the dark at room temperature and then counted, trypsinized, and replated under the same conditions analyzed by flow cytometry. without drugs. As shown in Fig. 1B, the number of tumor- 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. Cell viability was measured by 3-(4,5-dimethylthiazol-2-yl)- spheres treated with DPPE continuously (four doses) dropped F 2,5-diphenyltetrazolium bromide (MTT) assay. For MCF7 monolayer to 2 1.4% after 2 weeks of treatment (top) and the remaining cells, a total of 1,000 cells per well were seeded into 96-well plates and cells did not form spheres when further cultured in drug free grown overnight in duplicate. After drug treatment, MTT (1 mg/mL) media (bottom). Thus, DPPE seems to exert a cytotoxic effect on was added for 2 h at 37jC, followed by 100 AL DMSO overnight. TFUs. Notably, continuous exposure to DPPE was critical. A
www.aacrjournals.org 121 Clin Cancer Res 2009;15(1) January 1, 2009 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. CancerTherapy: Preclinical
Fig. 1. Specific killing of MCF7 TFUs by DPPE. A, MCF7 cells seededunder adherent (top) or nonadherent (bottom) conditions were treated with doxorubicin (Dox), DPPE, or combinations of both drugs for 7 d.The number of live adherent cells was assessed by the MTT assay (top) and the number of tumorspheres in the nonadherent cultures was counted microscopically (bottom). B, MCF7 were plated in tumorsphere conditions and treated with single or four doses (Cont.; twice per week) of the indicated drugs for 14d.The number of spheres formed after 2 wkis indicated (top).The cultures were trypsinized and reseeded in the absence of drugs.The number of spheres formed after 2 additional weeks is presented (bottom). Columns, mean number of tumorspheres from an experiment done in duplicate; bars, SD. Similar results were obtained in three independent experiments. *, P < 0.05, compared with dimethyformamide (DMF ) control. C, CD24-CD44 flow cytometry profiles of MCF7 cells grown as monolayer andT11 pleural effusion cells cultured as monolayer for1wkor under nonadherent conditions for 2 wk. Note that under nonadherent conditions, about half the population became CD44+:CD24+.T9 pleural effusion cells cultured as monolayer over an 8-wkperiod. Note the gradual conversion of CD44 +:CD24-/low cells into CD44+:CD24+ cells.
Clin Cancer Res 2009;15(1) January 1, 2009 122 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. Preferential Killing of Breast TICs by DPPE/Tesmilifene single dose treatment reduced the number of spheres by only However, the rare TFUs that formed tumorspheres in non- 55 F 2.8%, compared with 98 F 2.1% following continuous adherent conditions were highly sensitive to DPPE. treatment. CD44+:CD24-/low breast tumor cells from pleural effusions give The number of MCF7 tumorspheres formed in untreated rise to nonadherent tumorspheres. The analysis of MCF7 breast cultures under nonadherent conditions was f1/500 to 1/1,000 cancer cells is limited by the low percentage of TICs in the plated cells. This low percentage of tumorsphere-forming culture (Fig. 1C) and the fact that the cells have been clonally cells coincided with the CD24-CD44 profile of adherent maintained in culture for an extended period of time. To MCF7cells.AsshowninFig.1C,thepercentageof corroborate our results, we analyzed the effect of DPPE on CD44+:CD24-/low cells in MCF7 cells was 0.82%, with the vast primary metastatic breast tumor cells from pleural effusions. majority of the cells (99.11 F 1.2%) being CD44-:CD24+. Thus, TICs were previously identified in pleural effusions at a DPPE may have little discernable effect on MCF7 cells grown as frequency of 1/100 to 1/1,000 in sorted CD44+:CD24-/low cells monolayer because of the low frequency of TFUs in this culture. but at <1/10,000 in the CD44+:CD24+ fraction (12). We
Fig. 2. Sorted CD44+:CD24-/low, but not CD24+, pleural effusion cells form spheres under nonadherent conditions. A, flow cytometry profile of hematopoietic-depleted human pleural effusion cells and the gating for cell sorting (left), post-sorting profile (middle), and the flow cytometry profile of the sorted cells 2 wkafter seeding onto adherent plates (right). Note that the sorting (left and middle) and the flow cytometry profiles (right) were done on different flow cytometry machines.The sorted CD44+:CD24-/low cells gave rise to CD24+/- cells expressing intermediate levels of CD24, whereas the sorted CD24+ cells did not convert to CD44+:CD24-/low. B, spheres were formed after culturing the sorted CD44+:CD24-/low cells in ultra-low attachment plates; no spheres were formed from the CD44+:CD24+ population.
www.aacrjournals.org 123 Clin Cancer Res 2009;15(1) January 1, 2009 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. CancerTherapy: Preclinical
Fig. 3. DPPE treatment inhibits tumorsphere formation and preferentially targets the CD44+:CD24-/low population in pleural effusion cells. A, tumorsphere formation assay (top) and cell survival MTTassay (bottom) were carried out on nonadherent hematopoietic-depleted pleural effusion cells after 2 wkof continuous (two doses per week) treatment with doxorubicin (0.05) Ag/mL, DPPE (5 Amol/L), DPPE plus doxorubicin, or dimethyformamide. Columns, mean number of tumorspheres from an experiment done in duplicate; bars, SD. Similar results were obtained in three independent experiments. B, pleural effusion cells (1 Â10 4) were seeded onto 24-well plates as adherent cultures overnight, treated with a single dose of the indicated drugs for 2 d, and then processed for flow cytometry with FITC-CD24 and APC-CD44. Representative data of three independent experiments. C, left, average of three experiments done as in B and presented as the ratio of CD44+:CD24-/low in drug-treated relative to dimethyformamide-treated cultures. Bars, SD. Right, average of three experiments done as in B and presented as the absolute number of CD44+:CD24-/low cells. D, DPPE induces apoptotic cell death in the CD44+:CD24-/low cell population. Plural effusion cells were treated with the indicated drugs for 2 d and then processed for triple flow cytometry analysis with FITC-CD24, APC-CD44, and PE-AnnexinV.The CD44+:CD24-/low and CD44+:CD24+ cells were first gated as indicated and then the percentage of Annexin V+ cells in the gated fractions was determined (n = 3 per treatment). Note that doxorubicin induced apoptosis in both cell types whereas DPPE preferentially targeted the CD44+:CD24-/low population. *, P <0.05;**,P < 0.001, compared with dimethyformamide control. obtained pleural effusions from 11 breast cancer patients with which were larger in size, represented a more homogenous metastatic disease (12 samples in total; Supplementary Table population of tumor epithelial cells (Supplementary Fig. S1B). S1). The samples varied in their content of RBC, hematopoietic Following purification, tumor cells were cultured in adherent or cells, and tumor epithelial cells. The hematopoietic fraction nonadherent conditions. Five of the 12 samples had high RBC varied from f60% to >95% in different pleural effusion content, and attempts to recover tumor epithelial cells were samples. To remove the hematopoietic cells, samples were unsuccessful. Of the remaining seven samples, two samples did passed through StemSep columns (StemCell Technologies) not survive in culture; three samples (ER+) survived as tumor- containing antihuman antibodies to the following antigens: spheres for about 8 to 12 weeks and were used to establish CD2, CD14, CD16, CD38, CD45, CD66, and glycophorin A. conditions for cell sorting (not shown); and two samples, T9 Flow cytometry analysis revealed efficient depletion of hema- (ER+) and T11 (ER-), gave rise to monolayer cultures that topoietic cells (Supplementary Fig. S1A); the remaining cells, survived for several months and were further analyzed as
Clin Cancer Res 2009;15(1) January 1, 2009 124 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. Preferential Killing of Breast TICs by DPPE/Tesmilifene
described below. Ponti et al. (45) reported similar results Flow cytometry analysis of the T9 and T11 primary with plural effusions; they managed to propagate tumorspheres pleural effusions grown as monolayer cultures revealed that from only 3 [all estrogen receptor positive (ER+)] of 16 both contained mostly CD44+:CD24-/low cells (Fig. 1C). breast tumors. When plated in ultra-low attachment plates, the cells formed
Fig. 4. DPPE suppresses growth of tumorspheres derived from MMTV-Wnt1and MMTV-Neu mouse mammary tumor cells. A and B, tumorsphere formation assays (top) for MMTV-Wnt1or MMTV-Neu tumor cells treated with the indicated drugs for 2 wk.The tumorspheres were dissociated and replated in the absence of drugs for additional 2wkandthencounted(middle) or subjected to MTTcell survival assays (bottom). Representative data of three independent experiments done in duplicates. *, P < 0.05, relative dimethyformamide control. C and D, micrographs of MMTV-Wnt1or MMTV-Neu spheres (arrows) 2 wkafter replating in drug free media. Original magnification, Â20.
www.aacrjournals.org 12 5 Clin Cancer Res 2009;15(1) January 1, 2009 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. CancerTherapy: Preclinical
Fig. 5. DPPE induces apoptotic death in MMTV-Wnt1and MMTV-Neu mouse mammary tumor cells. A, dose response of MMTV-Neu tumorsphere formation following treatment with the indicated doses and drug concentrations. Representative data of three independent experiments done in duplicates. B, representative flow cytometry profiles of AnnexinV ^ stained MMTV-Neu tumorspheres treated for 2 wkwith the indicated drugs. C, kinetics of apoptotic response of MMTV-Neu tumor cells to the indicated drugs. Primary MMTV-Neu tumorspheres were dissociated into single cells, seeded onto ultra-low attachment plates, and continuously treated with the indicated drugs. After 24 h, 72 h, 1wk, or 2 wk, the treated cells were dissociated, stained with 7-amino-actinomycin D (7AAD) and Annexin V, and analyzed by flow cytometry. Points, mean AnnexinV staining of three independent experiments (n = 3) for each drug and time point; bars, SD. RelativeAnnexinV was calculated by normalizing the data to the levels of intrinsic apoptosis in dimethyformamide-treated cells. D, average levels of AnnexinV staining following 2-wktreatment of Wnt1tumor cells with the indicated drugs. Note moderate effect of DPPE on Wnt1tumor cells compared with Neu tumors (B and C). E, representative CD24-FITC and CD49f-PE flow cytometry profiles of MMTV-Wnt1tumor cells treated with the indicated drugs for 2 wkunder adherent conditions. F, average of three experiments showing the percentage of CD24+:CD49f+ cells after 2-wktreatment with the indicated drugs. *, P < 0.05, relative to dimethyformamide control.
Clin Cancer Res 2009;15(1) January 1, 2009 126 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. Preferential Killing of Breast TICs by DPPE/Tesmilifene
Fig. 6. Targeted killing ofTICs by DPPE. Representative FvB female mice, 3 mo after injections with 250 live Her2/Neu tumorsphere cells pretreated for 2 wk with the indicated drugs. Arrows, secondary tumors. tumorspheres that consisted primarily of CD44+:CD24-/low were trypsinized to obtain single-cell suspension, plated in cells. Following extended culturing under adherent conditions, ultra-low attachment plates, and treated with vehicle alone there was a gradual loss of CD44+:CD24-/low cells and a (dimethyformamide) or physiologic attainable concentrations concomitant increase in the CD44-/low:CD24+ population; by of doxorubicin (0.05 Ag/mL), DPPE (5 Amol/L), or both. 2 to 4 weeks, both cell populations were observed, and by Two weeks later, tumorsphere number and cell survival 8 weeks (passage 4), most PE-T9 cells became CD44-:CD24+ (MTT assay) were determined. Doxorubicin treatment reduced cells (Fig. 1C). To directly test the relationship between tumorsphere growth relative to vehicle-treated cultures by only CD44+:CD24-/low and CD24+ cells and identify the cells 2-fold (Fig. 3A). In contrast, cultures treated with DPPE or capable of forming spheres, T9 and T11 primary cells grown DPPE plus doxorubicin exhibited a 6-fold reduction in sphere as monolayer were isolated by fluorescence-activated cell formation and viability. sorting on the basis of CD24 and CD44 expression and then To determine which cell population was targeted by DPPE, plated under adherent or nonadherent conditions (Fig. 2A). the cultures were treated with the various drugs for 2 days The sorted CD44+:CD24-/low cells grew well as monolayer and stained with fluorescent-conjugated CD24 and CD44 cultures and formed spheres in nonadherent conditions antibodies. Flow cytometry analysis revealed that doxorubicin (Fig. 2B). In contrast, sorted CD24+ cells grew poorly under either conditions and did not form spheres. Similar observa- Table 1. Summary of transplantation experiments tions with immortalized human mammary epithelial cells have with MMTV-Wnt1 and MMTV-Neu tumorspheres recently been reported (46). Importantly, flow cytometry analysis done 1 week after plating the sorted cells under Treatment Tumor/injections adherent conditions revealed that the CD44+:CD24-/low cells 250 cells 100 cells gave rise to CD44+:CD24-/low as well as CD44low:CD24+/- cells, expressing intermediate levels of CD24 (Fig. 2A, right). By 3 mo 5 mo 3 mo 5 mo + + contrast, the CD44 :CD24 cell population lost viability MMTV-Wntl and remained CD24+. Finally, to test the differentiation DMF 2/6 3/6 3/6 4/6 capacity of the sorted CD44+:CD24-/low tumor cells, they were Dox 1/6 1/6 0/6 1/6 seeded onto collagen 1–coated coverslips, induced to differ- DPPE 1/6 2/6 0/6 2/6 DPPE/Dox 0/6 0/6 0/6 0/6 entiate, and immunostained with antibodies against smooth muscle actin, cytokeratin 14 (myoepithelial markers), and MMTV-Neu cytokeratin 18 (luminal marker). In four different sorting/ DMF 12/13 13/13 3/9 4/7 differentiation experiments, the CD44+:CD24-/low fractions Dox 4/12 4/12 0/8 0/6 DPPE 0/12 5/12 0/8 0/6 differentiated to express one or more of these markers in DPPE/Dox 0/12 0/12 0/8 0/6 >50% of the tumor cells (Supplementary Fig. S2). Thus, the CD44+:CD24-/low cell population exhibited limited self-renewal + NOTE: Dissociated tumorsphere cells were seeded onto ultra-low capacity under the conditions we used, gave rise to CD44 : attachment plates and treated with the indicated drugs for 2 wk + CD24 cells, formed tumorspheres in culture, and had the (four changes). The cultures were then trypsinized, and live cells ability to differentiate into distinct cell types. were counted by trypan blue exclusion assay. The indicated DPPE suppresses CD44+:CD24-/low breast tumor cells number of live cells was mixed with Matrigel and transplanted into the #4 inguinal mammary glands of recipient female mice. from pleural effusions by inducing apoptotic death. We next Incidence of tumors after 3and 5 mo is indicated. investigated the effect of DPPE on the T9 and T11 primary Abbreviations: DMF, dimethyformamide; Dox, doxorubicin. pleural effusion cultures described above. The tumor cells
www.aacrjournals.org 12 7 Clin Cancer Res 2009;15(1) January 1, 2009 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. CancerTherapy: Preclinical treatment had only a moderate effect on the CD44+:CD24-/low tumorsphere formation even more efficiently (not shown). At cell populations [1.2-fold reduction (P > 0.01) relative to lower drug concentrations or doses, the effect of DPPE on dimethyformamide control group; Fig. 3B and C]. By contrast, tumorsphere formation was significantly reduced. To determine DPPE reduced the ratio of CD44+:CD24-/low to CD24+ by 3.3- whether DPPE induces apoptotic cell death in mouse fold (P < 0.01) and the absolute number of CD44+:CD24-/low mammary tumor cells as it does in human breast tumor cells cell population by 3.8-fold (P < 0.01), indicating that DPPE from pleural effusions (Fig. 3), tumorspheres were dissociated, preferentially targets the CD44+:CD24-/low cell population. replated under nonadherent conditions, and treated with the To investigate the mechanisms by which DPPE suppressed indicated drugs for different time periods. The cells were then cell viability, the cultures were again treated with the various collected and stained with 7-amino-actinomycin D, which is drugs for 2 days, and then unfixed cells were stained with FITC- excluded by live cells, plus Annexin V, which specifically labels CD24 and APC-CD44 as well as PE-Annexin V. Annexin V apoptotic cells, and analyzed by flow cytometry. Continuous detects phosphatidylserine residues that translocate to the outer exposure to DPPE induced massive cell death in Her2/neu cell membrane during early stages of apoptosis. The intrinsic tumors, whereas doxorubicin, at the concentrations we level of cell death for the CD44+:CD24-/low cells (13.96 F used, had only a moderate effect (Fig. 5B). Kinetic analysis 0.48%) was slightly lower than that for the CD44+:CD24+ cells revealed that the induction of apoptosis by DPPE was modest (18.06 F 3.99%; Fig. 3D). Doxorubicin treatment increased (50 F 5.8%) in the first 3 days of treatment and dramatically apoptotic cell death by f2-fold in both cell populations. In increased >5-fold after 2 weeks of treatment (Fig. 5C). contrast, DPPE increased the rate of apoptosis in CD44+:CD24-/ In parallel experiments with Wnt1-induced tumor cells, low cells nearly 2-fold (26.38 F 5.01%; Fig. 3D, right) but did DPPE treatment for 2 weeks increased the level of apoptosis not increase apoptosis above background level in the CD24+ by only 2.2-fold, compared with 1.5-fold induced by doxoru- population (Fig. 3D, left). Treatment with DPPE plus doxoru- bicin (F0.19; Fig. 5D). In MMTV-Wnt1–induced mammary bicin slightly increased overall cell death in the CD44+:CD24-/ tumors, TICs and TFUs are found in the CD24+:CD49f+ double- low cell population (29.84 F 1.77%). Thus, DPPE induces positive cell population (23, 24).7 To test the effect of DPPE on apoptotic cell death in CD44+:CD24-/low cells. tumor cells by marker analysis, monolayer MMTV-Wnt1 tumor DPPE inhibits survival and cell cycle progression of Her2/neu cells were treated for 2 weeks with the various drugs and then and Wnt1 mammary tumor cells. We next analyzed mammary subjected to flow cytometry analysis with CD24 and CD49f. tumor cells derived from MMTV-Wnt1 and MMTV-Her2/neu The percentage of CD24+:CD49f+ cells in primary tumors transgenic mice, in which TICs can be functionally analyzed was initially f38% (data not shown) but was reduced to f3% by transplantation into syngeneic mice (23). We first tested after 2 weeks in monolayer culture (Fig. 5E), indicating strong the effect of DPPE on tumorsphere formation, apoptosis, selection against TICs. Treatment with doxorubicin increased and cell cycle progression in vitro. Primary tumors from the percentage of CD24+:CD49f+ cells by f4-fold, suggesting MMTV-Wnt1 and MMTV-Neu were dissociated into single-cell that this drug preferentially kills the CD24-:CD49f- non-TIC suspension by enzymatic treatment and then immunodepleted population. By contrast, treatment with DPPE reduced the for hematopoietic (CD45/Ter119 antibodies), endothelial percentage of CD24+:CD49f+ cells by 2-fold (Fig. 5E and F). (CD31 antibody), and stromal (CD140a antibody) cells Thus, the tumorsphere formation assay, Annexin V staining, (Supplementary Fig. S1C). The Lin- tumor cells were then and marker analysis indicate that (a) Neu tumor cells are more seeded into ultra-low attachment plates as described (23). sensitive than Wnt1 tumor cells to DPPE; (b) both Neu and Primary tumorspheres from each tumor type were dissociated Wnt1 tumor cells are significantly more sensitive to DPPE than by light trypsinization and replated in the presence of drugs to doxorubicin; and (c) continuous treatment with DPPE is or vehicle alone. After 2 weeks of continuous treatment required for targeted killing of TFUs/TICs. (four doses), the number of tumorspheres in each culture was To test the effect of DPPE on cell cycle progression, assessed. As shown in Fig. 4A and B (top), DPPE treatment tumorspheres treated for 1 week with the indicated drugs were alone reduced the number of tumorspheres in both the Neu stained with propidium iodide to label DNA content and and WNt1 cultures. The effect of DPPE was more pronounced analyzed by flow cytometry. The percentage of cells in S/G2/M in the Neu-induced tumors (<2 F 0.71% relative to dimethy- phases was f15.92 F 2.94% and 8.96 F 0.21% for Wnt1 and formamide control) than in the Wnt1-induced tumors (39.5 F Neu tumorspheres, respectively (Supplementary Fig. S3). 1.41%). The tumorspheres were then dissociated and replated Exposure to 0.05 Ag/mL doxorubicin had little effect on cell in the absence of drugs; 2 weeks later, tumorspheres were either cycle distribution of either culture. In contrast, treatment with counted or subjected to MTT assay to measure viability. Very DPPE suppressed the S/G2/M phases by 33 F 3.35% and 59 F few tumorspheres arose in the DPPE-treated Neu cultures 13.43% in Wnt1 and Neu tumorspheres, respectively, suggest- whereas f23% of Wnt1 tumorspheres survived (relative to ing that DPPE preferentially eliminates cycling cells. vehicle-treated cultures). Similar inhibition of tumorsphere cell Direct killing of Neu and Wnt1 TICs by DPPE. Whereas viability was observed with the MTT assay (Fig. 4A and B, tumorsphere formation can be used as a surrogate for tumor bottom). Examples of the cultures treated with the various drugs initiation, the ultimate test for TICs is their ability to form are shown in Fig. 4C and D. secondary tumors after transplantation into recipient mice. We next determined the sensitivity of Neu tumorspheres Thus, to directly test the effect of DPPE on TICs, Neu or Wnt1 to different doses and concentrations of DPPE over a tumorspheres were dissociated and replated in the presence of 2-week period. Continuous treatment with physiologically the various drugs under nonadherent conditions. After 2 weeks, relevant concentration of DPPE (i.e., four changes over 2 weeks; 5 Amol/mL) produced the strongest effect (Fig. 5A). At a higher concentration (10 Amol/mL), DPPE suppressed 7 Our unpublished data.
Clin Cancer Res 2009;15(1) January 1, 2009 128 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. Preferential Killing of Breast TICs by DPPE/Tesmilifene the cultures were trypsinized and counted by trypan blue MMTV-Wnt1 mammary tumors as tested by both in vitro and exclusion assay; 50, 100, or 250 live cells were mixed with in vivo assays. In aggregate, our data suggest that DPPE directly Matrigel and injected into the mammary glands (#4) of targets TICs. This model may explain the significant improve- syngeneic FvB female mice. Tumor incidence after 3 and ment in overall survival and progression-free survival observed 5 months is depicted in Figure 6 and Table 1. Treatment in the doxorubicin plus DPPE arm despite no objective of Wnt1 tumorspheres with DPPE or doxorubicin reduced difference in tumor response (38), as targeted killing of rare tumor incidence to 4/12 and 2/12, respectively, relative to TICs may reduce additional metastasis and improve overall vehicle-treated cultures (7/12 tumors after injections of 250 or survival without significantly affecting the bulk of metastatic 100 cells). Treatment with both drugs completely abolished tumors. tumor initiation (0/12; Table 1). The most parsimonious model for the cooperation between With Neu tumorsphere cells, doxorubicin treatment alone DPPE and doxorubicin is that their individual effect on TICs is reduced but did not completely eliminate tumor incidence simply accumulative. Each drug kills TICs, albeit with different 3 months after transplantation (4/18 total injections of 250 or efficacy, and combined treatment with both drugs has an 50 cells) as compared with vehicle-treated cells (15/20). In additive effect. Another level of cooperation is suggested from contrast, DPPE-treated tumorsphere cells did not give rise data in Table 1 showing that treatment with DPPE increases to tumors at this stage (0/18; Table 1). Figure 6 shows the latency for tumor formation by 2.6-fold relative to representative recipient mice 3 months after transplantation doxorubicin. The late onset of these tumors may also be with Neu tumor cells pretreated with the indicated drugs. After explained if the cancer stem cell hierarchy is not as stable as is 2 additional months (i.e., 5 months posttransplantation), five commonly thought (i.e., if non-TICs can revert to TICs by tumors appeared in the DPPE-treated cells. The latency of genetic or epigenetic changes after a prolonged period of time tumor formation following DPPE treatment was 2.6-fold following transplantation; ref. 50). Thus, DPPE and doxoru- longer than following doxorubicin treatment. Specifically, the bicin may cooperate by targeting different cells: DPPE latencies for tumor formation in the dimethyformamide-, preferentially targets TICs; doxorubicin preferentially kills doxorubicin-, and DPPE-treated groups were 63.6 F 13.7, non-TICs (with the potential to revert); and combined 51.7 F 3.5, and 137.2 F 11.7 days, respectively. Combined treatment eliminates both cell types, hence any tumorigenic treatment of the Neu tumor cells with DPPE plus doxorubicin potential. Additional analysis will be required to address the completely abolished TICs (0/18 injections in total). We stability of the cancer stem cell hierarchy and its effect on conclude that DPPE effectively kills TICs and that combined chemotherapy in the contexts of drugs, such as DPPE, that treatment with DPPE plus doxorubicin cooperates to eradicate target TICs. both Wnt1 and Neu TICs. We have shown that continuous exposure to high, pharma- cologically attainable doses of DPPE (5 Amol/L) effectively kills Discussion TICs in vitro. The need for continuous exposure may be the hallmark of drugs that target TICs as these cells are thought to Here we show that DPPE (tesmilifene) can act as a single undergo slow self-renewal. Our results suggest that DPPE-based agent to target TICs. It induces apoptotic death in TICs when therapies should involve continuous or frequent administration administered alone and, in certain contexts, cooperates with of the drug. This is notably a challenge given the side effect of doxorubicin to further kill these tumorigenic cells. Our the drugs at the current dose (38), but may be overcome with a observation that DPPE can directly target TICs questions better understanding of the pharmacokinetics of DPPE. previous models that implicated DPPE in the potentiation of Alternatively, more potent DPPE derivatives with better cytotoxicity by antineoplastic drugs (33). These models were specificity may be developed. The identification of the cellular strengthened by the observation that the molecular structure of target(s) of DPPE and the affected biological process/pathway DPPE includes a tertiary amine, phenyl rings, and carbonyl are therefore of paramount importance. group often found in p-glycoprotein inhibitors (47–49). The variation in tumor response to DPPE likely reflects both However, DPPE also enhances cytotoxicity of multidrug the percentage of TICs in a given tumor and the biology of TICs resistance–negative tumor cells, and the enhancement (or from each individual breast cancer subtype. Breast carcinomas cooperation) with other drugs is not accompanied by increased represent several subtypes including ER+, Her2/Neu+,and intracellular concentration of these drugs (42). In fact, it Basal-like (Her2/Neu-:PR-:ER-; ref. 51). We have found that was hypothesized that DPPE may activate, rather than inhibit, the metastatic breast tumors from pleural effusions and HER2/ p-glycoproteins (33). We propose that the effect of DPPE as Neu+ mammary tumors were highly sensitive to DPPE, Wnt1 a single agent has been overlooked in previous experiments TICs were less sensitive, whereas MCF7 sensitivity depended on because TICs were not assayed directly. As we observed, DPPE culture conditions (very low as monolayer; high as spheres). alone had a minor effect on MCF7 cells grown as monolayer, in The high sensitivity of HER2/Neu+ mammary tumors to DPPE which only a very small fraction of the cells are TICs. By is consistent with the high frequency of TICs in this tumor testing for TICs directly, for example by quantifying the ability subtype (23, 31, 52). Consistent with the differential sensitivity of MCF7 breast tumor cells to form tumorspheres under of different tumor types to DPPE, stratification of the clinical nonadherent conditions, we were able to show efficient and results from the National Cancer Institute of Canada MA.19 direct killing of TICs by DPPE. In addition to MCF7 cells, trial revealed that ER-negative tumors responded better than the specific killing of TICs by DPPE was observed with two entire tumor set (33). Given the high sensitivity of the Neu+ independent metastatic breast tumors from breast cancer tumor cells to DPPE (Figs. 4–6; Table 1), it would be of interest patients with malignant pleural effusions. Finally, we showed to determine the overall survival difference for Her2/Neu+ that DPPE specifically killed TICs from MMTV-Neu and breast cancer patients treated with DPPE. In addition, the effect
www.aacrjournals.org 12 9 Clin Cancer Res 2009;15(1) January 1, 2009 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. CancerTherapy: Preclinical of DPPE on HER2+ and ER+ tumors may be tested in DPPE may serve as a prototype TIC-specific inhibitor with combinations with anthracyclin plus herceptin, lapatinib plus available clinical data. capecitabine, or hormonal therapies, respectively. In conclusion, using four different breast tumor models, we showed that DPPE directly kills TICs and, in certain Disclosure of Potential Conflicts of Interest contexts, cooperates with doxorubicin to further eliminate A K.I. Pritchard is a member of the speakers bureau of Aventis, Pfizer, AstraZeneca, these tumorigenic cells. A continuous exposure to 5 mol/L and Pharmacia. She is a consultant for Aventis, Roche, Pharmacia, Ortho-Biotech, concentration of the drug is required for optimal results. Thus, Pfizer, GSK Abraxis,YM Biosciences, and Biomera.
References 1. Pritchard KI. The best use of adjuvant endocrine et al. Highly purified CD44+ prostate cancer cells SG. Membrane topology of cytochrome P450 2B4 treatments. Breast 2003;12:497 ^ 508. from xenograft human tumors are enriched in tu- in Langmuir-Blodgett monolayers. Arch Biochem 2. Hery C, FerlayJ, Boniol M, Autier P. Quantification of morigenic and metastatic progenitor cells. Onco- Biophys 1998;359:82^ 8. changes in breast cancer incidence and mortality gene 2006;25:1696 ^ 708. 38. Reyno L, Seymour L,Tu D, et al. Phase III study of since 1990 in 35 countries with Caucasian-majority 22. Shackleton M,Vaillant F, Simpson KJ, et al. Genera- N,N-diethyl-2-[4-(phenylmethyl) phenoxy]ethan- populations. Ann Oncol 2008;19:1187 ^ 94. tion of a functional mammary gland from a single stem amine (BMS-217380-01) combined with doxorubicin 3. Levine M, Eisen A. Anthracycline adjuvant chemo- cell. Nature 2006;439:84 ^ 8. versus doxorubicin alone in metastatic/recurrent therapy: how much is enough? J Clin Oncol 2001;19: 23. Liu JC, Deng T, Lehal RS, Kim J, Zacksenhaus E. breast cancer: National Cancer Institute of Canada 599^601. Identification of tumorsphere- and tumor-initiating Clinical Trials Group Study MA.19. J Clin Oncol 2004; 22:269 ^ 76. 4. Sledge GW, Neuberg D, Bernardo P, et al. Phase III cells in HER2/Neu-induced mammary tumors. Cancer trial of doxorubicin, paclitaxel, and the combination of Res 2007;67:8671 ^ 81. 39. Raghavan D, Brandes LJ, Klapp K, et al. Phase II doxorubicin and paclitaxel as front-line chemotherapy 24. Cho RW, Wang X, Diehn M, et al. Isolation and trial of tesmilifene plus mitoxantrone and prednisone for metastatic breast cancer: an intergroup trial molecular characterization of cancer stem cells in for hormone refractory prostate cancer: high subjec- (E1193). J Clin Oncol 2003;21:588^92. MMTV-Wnt-1 murine breast tumors. Stem Cells tive and objective response in patients with symp- tomatic metastases. J Urol 2005;174:1808^13; 5. DickJE. Breast cancer stem cells revealed. Proc Natl 2008;26:364 ^ 71. discussion 1813. Acad Sci U S A 2003;100:3547 ^ 9. 25.Wright MH, Calcagno AM, Salcido CD, Carlson MD, Ambudkar SV,Varticovski L. Brca1breast tumors con- 40. Li Y, Hively WP,Varmus HE. Use of MMTV-Wnt-1 6. Al-Hajj M, Becker MW,Wicha M,Weissman I, Clarke + - + transgenic mice for studying the genetic basis of MF. Therapeutic implications of cancer stem cells. tain distinct CD44 /CD24 and CD133 cells with cancer stem cell characteristics. Breast Cancer Res breast cancer. Oncogene 2000;19:1002^ 9. Curr Opin Genet Dev 2004;14:43 ^ 7. 2008;10:R10. 41. Guy CT,Webster MA, Schaller M, ParsonsTJ, Cardiff 7. Pardal R, Clarke MF, Morrison SJ, et al. Applying the 26. ReyaT, Morrison SJ, Clarke MF,Weissman IL. Stem RD, Muller WJ. Expression of neu proto-oncogene in principles of stem-cell biology to cancer. Nat Rev cells, cancer, and cancer stem cells. Nature 2001;414: the mammary epithelium of transgenic mouse induces Cancer 2003;3:895 ^ 902. 10 5 ^ 11. metastatic disease. Proc Natl Acad Sci U S A1992;89: 8. Dontu G, Al-Hajj M, Abdallah WM, et al. Stem cells 27. BaoS,WuQ,McLendonRE,etal.Gliomastem 10578 ^ 82. in normal breast development and breast cancer. cells promote radioresistance by preferential activa- 42. Brandes LJ, Simons KJ, Bracken SP, Warrington Cell Prolif 2003;36 Suppl 1:59 ^ 72. tion of the DNA damage response. Nature 2006; RC. Results of a clinical trial in humans with refrac- 9. Smalley M, Ashworth A. Stem cells and breast can- 444:756^60. tory cancer of the intracellular histamine antagonist, cer: a field in transit. Nat Rev Cancer 2003;3:832^ 44. 28. PhillipsTM, McBride WH, PajonkF. The response of N,N-diethyl-2-[4-(phenylmethyl)phenoxy]ethan- 10. Bonnet D, DickJE. Human acute myeloid leukemia is CD24-/low/CD44+ breast cancer-initiating cells to amine-HCl, in combination with various single anti- organized as a hierarchy that originates from a primi- radiation. J Natl Cancer Inst 2006;98:1777 ^85. neoplastic agents. J Clin Oncol 1994;12:1281 ^ 90. tive hematopoietic cell. Nat Med 1997;3:730 ^ 7. 29. TodaroM,PerezAleaM,ScopellitiA,MedemaJP, 43.YM Biosciences data in house. 11. Lapidot T, Sirard C,Vormoor J, et al. A cell initiating Stassi G. IL-4-mediated drug resistance in colon can- 44. Dontu G, Abdallah WM, Foley JM, et al. In vitro human acute myeloid leukaemia after transplantation cer stem cells. Cell Cycle 2008;7:309 ^ 13. propagation and transcriptional profiling of human into SCID mice. Nature1994;367:645^8. 30.Yu F,Yao H, Zhu P, et al. let-7 regulates self renewal mammary stem/progenitor cells. Genes Dev 2003;17: 12. Al-Hajj M, Wicha MS, Benito-Hernandez A, and tumorigenicity of breast cancer cells. Cell 2007; 1253 ^ 70. Morrison SJ, Clarke MF. Prospective identification 131:110 9 ^ 2 3. 45. Ponti D, Costa A, Zaffaroni N, et al. Isolation and of tumorigenic breast cancer cells. Proc Natl Acad 31. Li X, Lewis MT, HuangJ, et al. Intrinsic resistance of in vitro propagation of tumorigenic breast cancer Sci U S A 2003;100:3983 ^8. tumorigenic breast cancer cells to chemotherapy. cells with stem/progenitor cell properties. Cancer 13. SinghSK,HawkinsC,ClarkeID,etal.Identification J Natl Cancer Inst 2008;100:672 ^9. Res 2005;65:5506 ^ 11. of human brain tumour initiating cells. Nature 2004; 32.Tan B, Piwnica-Worms D, Ratner L. Multidrug resis- 46. Mani SA, Guo W, Liao MJ, et al. The epithelial- 432:396^401. tance transporters and modulation. Curr Opin Oncol mesenchymal transition generates cells with proper- 14. O’Brien CA, Pollett A, Gallinger S, DickJE. A human 2000;12:450 ^ 8. ties of stem cells. Cell 2008;133:704 ^ 15. colon cancer cell capable of initiating tumour growth 33.Vincent M. Tesmilifene may enhance breast can- 47. Brandes LJ, LaBella FS, Warrington RC. Increased in immunodeficient mice. Nature 2007;445:106 ^ 10. cer chemotherapy by killing a clone of aggressive, therapeutic index of antineoplastic drugs in combina- 15. Yang ZF, Ho DW, Ng MN, et al. Significance of multi-drug resistant cells through its action on the tion with intracellular histamine antagonists. J Natl + CD90 cancer stem cells in human liver cancer. p-glycoprotein pump. Med Hypotheses 2006;66: Cancer Inst 1991;83:1329^36. Cancer Cell 2008;13:153 ^ 66. 715 ^ 31. 48. Kudoh K, Kikuchi Y, Hiramatsu H, et al. Enhance- 16. WuC,WeiQ,UtomoV,etal.Sidepopulationcells 34. Brandes LJ, Queen GM, LaBella FS. N,N-Diethyl-2- ment of antitumour activity of cisplatin by N,N- isolated from mesenchymal neoplasms have tumor [4-(phenylmethyl)phenoxy]ethanamine (DPPE) a diethyl-2-[4-(phenylmethyl)phenoxy]ethanamine. initiating potential. Cancer Res 2007;67:8216^ 22. chemopotentiating and cytoprotective agent in clinical HCl in human ovarian cancer cells with intrinsic or ac- 17. Hansford LM, McKee AE, Zhang L, et al. Neuroblas- trials: interaction with histamine at cytochrome P450 quired resistance to cisplatin. Eur J Cancer 1997;33: toma cells isolated from bone marrow metastases 3A4 and other isozymes that metabolize antineoplas- 12 2 ^ 8 . contain a naturally enriched tumor-initiating cell. tic drugs. Cancer Chemother Pharmacol 2000;45: 49. Ramu A, Ramu N. Reversal of multidrug resistance Cancer Res 2007;67:11234 ^ 43. 298 ^ 304. by phenothiazines and structurally related com- 18. SchattonT, Murphy GF, FrankNY, et al. Identification 35. SzincsakN, Hegyesi H, Hunyadi J, Falus A, Juhasz pounds. Cancer Chemother Pharmacol 1992;30: of cells initiating human melanomas. Nature 2008; I. Different h2 receptor antihistamines dissimilarly 165 ^ 73. 451:345 ^9. retard the growth of xenografted human melanoma 50. Campbell LL, PolyakK. Breast tumor heterogene- 19. Li C, Heidt DG, Dalerba P, et al. Identification of cells in immunodeficient mice. Cell Biol Int 2002;26: ity: cancer stem cells or clonal evolution? Cell Cycle pancreatic cancer stem cells. Cancer Res 2007;67: 833^6. 2007;6:2332 ^ 8. 1030 ^ 7. 36. Brandes LJ, Bracken SP. The intracellular hista- 51. Perou CM, SorlieT, Eisen MB, et al. Molecular por- 20. SzotekPP, Pieretti-VanmarckeR, MasiakosPT, mine antagonist, N,N-diethyl-2-[4-(phenylmethyl)- traits of human breast tumours. Nature 2000;406: et al. Ovarian cancer side population defines cells phenoxy]ethamine.HCL, may potentiate doxorubicin 747 ^ 52. with stem cell-like characteristics and Mullerian in the treatment of metastatic breast cancer: results 52. Korkaya H, Paulson A, Iovino F, Wicha MS. HER2 Inhibiting Substance responsiveness. Proc Natl of a pilot study. Breast Cancer Res Treat 1998;49: regulates the mammary stem/progenitor cell popula- Acad Sci U S A 2006;103:11154 ^ 9. 61 ^ 8. tion driving tumorigenesis and invasion. Oncogene 21. Patrawala L, Calhoun T, Schneider-Broussard R, 37. Shank-Retzlaff ML, Raner GM, Coon MJ, Sligar 2008;27:6120^ 30.
Clin Cancer Res 2009;15(1) January 1, 2009 130 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research. Preferential Killing of Breast Tumor Initiating Cells by N,N -Diethyl-2-[4-(Phenylmethyl)Phenoxy]Ethanamine/Tesmilifen e
Tao Deng, Jeff C. Liu, Kathleen I. Pritchard, et al.
Clin Cancer Res 2009;15:119-130.
Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/15/1/119
Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2009/01/07/15.1.119.DC1
Cited articles This article cites 51 articles, 14 of which you can access for free at: http://clincancerres.aacrjournals.org/content/15/1/119.full#ref-list-1
Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/15/1/119.full#related-urls
E-mail alerts Sign up to receive free email-alerts related to this article or journal.
Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].
Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/15/1/119. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.
Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2009 American Association for Cancer Research.