Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cell Cycle. Author manuscript; available in PMC 2010 October 15.
Published in final edited form as:
PMCID: PMC2861502

Retinoid signaling regulates breast cancer stem cell differentiation


The cancer stem cell (CSC) hypothesis implicates the development of new therapeutic approaches to target the CSC population. Characterization of the pathways that regulate CSCs activity will facilitate the development of targeted therapies. We recently reported that the enzymatic activity of ALDH1, as measured by the ALDELFUOR assay, can be utilized to isolate normal and malignant breast stem cells in both primary tumors and cell lines. In this study, utilizing a tumorsphere assay, we have demonstrated the role of retinoid signaling in the regulation of breast CSCs self-renewal and differentiation. Utilizing the gene set enrichment analysis (GSEA) algorithm we identified gene sets and pathways associated with retinoid signaling. These pathways regulate breast CSCs biology and their inhibition may provide novel therapeutic approaches to target breast CSCs.

Keywords: Cancer stem cell, breast cancer, retinoid signalling, ALDH, tumorsphere, gene set enrichment analysis


Since the discovery that leukemias are organized in a cellular hierarchy, many studies have demonstrated the presence of cancer stem cells (CSCs) in solid tumors.1 CSCs drive the different steps of the carcinogenesis process. They retain key properties of normal stem cell such as self-renewal, which initiates and drives tumorigenesis, and differentiation, which albeit aberrant contributes to cellular heterogeneity.2 Moreover, CSCs are relatively resistant to conventional therapies such as radiation or chemotherapy.3 The clinical relevance of these findings in breast cancer was demonstrated by recent neoadjuvant studies that documented an increase in breast cancer stem cells following the administration of neo-adjuvant conventional cytotoxic chemotherapy.4 These observations underscore the importance of the isolation and characterization of the CSC population in elucidating the molecular mechanisms by which these cells mediate tumorigenesis and therapeutic resistance as well as developing new therapeutic approaches.

Few methods have been developed to enrich for CSCs in breast tumors.5 A pioneer study first identified a population of cells in human breast cancer that displayed cancer stem cell properties. This population was defined by the expression of cell surface markers (CD44+/CD24−/low/lin).6 In our previous works we showed that the ALDEFLUOR phenotype is a useful marker to identify and isolate stem cells from the normal and malignant human mammary epithelium.7 The ALDEFLUOR assay measures the aldehyde dehydrogenase (ALDH) enzymatic activity that is present in stem cells from normal or cancerous tissues in different organ sites.

ALDH is a detoxifying enzyme responsible for the oxidation of intracellular aldehydes, and may have a role in early differentiation of stem cells through its role in oxidizing retinol to retinoic acid (RA).8 RA regulates the transcription of different target genes through its interaction with the retinoic acid receptors (RAR) and retinoic X receptors (RXR).9,10 Through its actions on these receptors, RA initiates a program of cellular differentiation. Based on this property, all-trans retinoic acid (ATRA) has been utilized as a therapeutic compound to induce the differentiation of acute promyelocytic leukemia (APML) cells expressing the PML-RARα fusion protein.11 In contrast, inhibition of ALDH and retinoic signaling expands the pool of human hematopoietic stem cells (HSC).12

Given the recent identification of ALDH as a marker that can be used to isolate breast CSCs and the role of retinoic signaling in the control of stem cell differentiation, we postulated that breast CSCs differentiation might depend on retinoid signaling and that modulation of ALDH enzymatic activity could interfere with the self-renewal and differentiation program of breast CSCs. Here we introduce a further characterization of breast CSCs biology based on our recent findings regarding the presence of ALDEFLUOR-positive CSCs in breast cancer cell lines (BCLs) and the identification of a breast cancer stem cell signature.13 We analyzed the effect of modulation of retinoic signaling on breast CSCs from BCLs to characterize the pathways that regulate CSCs self-renewal in order to facilitate the development of targeted therapies.

Effect of retinoid signaling modulation on tumorsphere formation

To evaluate the role of retinoid signaling on the regulation of breast CSCs, we treated BCLs with either diethylaminobenzaldehyde (DEAB), a specific inhibitor of ALDH enzymatic activity, to block retinoid signaling, or with ATRA to induce a constitutive activation of retinoid signaling target genes in an ALDH-independent manner. Data from breast tumors, as well as cell lines, have demonstrated that CSCs can be isolated and propagated as “tumorspheres” in suspension culture.14 In a recent study, we reported that in each breast cancer cell line tested, the ALDEFLUOR-positive population had increased tumorsphere-forming capacity compared to ALDEFLUOR-negative cells.13 To determine whether DEAB or ATRA treatment had an effect on tumorsphere formation, we cultured four different cell lines (184A1, SUM149, SUM159, HCC1954) in adherent conditions and in the presence of 100 µM of DEAB or 1 nM of ATRA during 7 days or 3 days, respectively. After treatment with DEAB or ATRA, cells were detached and cultured in suspension under serum-free conditions for 5 days. DEAB-treated cells presented an increase in primary tumorsphere formation compared to the control, for each cell line tested (Fisher’s exact test, p<0.05) (Figure 1A). In contrast, cells treated with ATRA presented a decrease in primary tumorsphere formation compared to the control (Fisher’s exact test, p<0.05) (Figure 1B). Similar results were observed for secondary tumorsphere formation (Fisher’s exact test, p<0.05) (Supplementary Figure 1). These results suggested that ALDH blockade increases the CSC population. This could occur through inhibition of differentiation of these cells. In contrast, activation of retinoid signaling reduces the CSC population.

Figure 1
Effect of ATRA and DEAB treatment on tumorsphere formation of BCLs

Evaluation of the breast CSC signature enrichment in BCLs after modulation of the retinoid signaling by DEAB and ATRA

To further demonstrate that retinoid signaling plays a role in the regulation of breast CSCs differentiation, we defined and compared the gene expression profiles of DEAB- and ATRA-treated cells and confronted them to our previously described breast CSC gene expression signature.13 This breast CSC signature was established by comparison of gene expression profiles of the ALDEFLUOR-positive and ALDEFLUOR-negative populations from eight BCLs. It consisted of a 413-genes/ESTs list that included genes known to play a role in stem cell function. Five out of the eight BCLs (184A1, HCC1954, MDA-MB-453, SUM149, SUM159) used to establish the signature were treated with DEAB or ATRA, then collected and profiled using Affymetrix whole-genome oligonucleotide microarrays. To determine whether the genes regulated by the modulation of retinoic signaling (comparison of DEAB-treated versus ATRA-treated cells) are related to the genes included in our breast CSC signature, we utilized the gene set enrichment analysis (GSEA) algorithm.15

A portion of the genes overexpressed in the ALDEFLUOR-positive populations (10/30 genes) was highly expressed in DEAB-treated cells (Supplementary Table 1). In contrast, a subset of the ALDEFLUOR-negative-associated genes (103/289 genes) were highly expressed in ATRA-treated cells (Supplementary Table 2). The reciprocal analysis yielded no significant enrichment between the ALDEFLUOR-positive-associated gene set and the ATRA-treated-associated gene set or between the ALDEFLUOR-negative-associated gene set and the DEAB-treated-associated gene set. These data thus showed an enrichment of genes involved in breast CSCs activity after ALDH blockade by DEAB and an association between genes expressed in differentiated cancer cells and the transcriptional profile of ATRA-treated cells. They indicated that retinoid signaling plays a role in the control of the breast CSCs differentiation.

Determination of the pathways regulated by retinoic signaling modulation

To identify the pathways that are associated with retinoic signaling we utilized the GSEA algorithm to screen the pathways and gene signatures from the Broad Institute (MSigDB c2: Curated Gene Sets).15 We evaluated the different pathways specifically enriched in DEAB-treated cells compared to ATRA-treated cells. A total of 57 pathways and gene signatures were differentially regulated between DEAB-treated cells and ATRA-treated cells (all gene sets with an FDR≤0.25) (Table 1). Among these gene sets, three were significantly associated with ALDH blockade and 54 with ATRA-induced differentiation.

Table 1
Gene sets enriched in DEAB-treated compared to ATRA-treated cells

Genes involved in tRNA biosynthesis were preferentially enriched in DEAB-treated cells (TRNA_SYNTHETASES, AMINOCYL_TRNA_BIOSYNTHESIS). This pathway is essential for protein synthesis and cell viability. tRNA synthetases regulate a signaling network including multiple processes such as DNA repair (through the regulation of ATM/ATR/P53), angiogenesis, or wound healing.16

All the gene sets associated with ATRA-treated cells presented an enrichment score (ES) with a negative value (−0.88 ≤ ES ≤ −0.37), indicating downregulation of multiple pathways during ATRA treatment. Among the different pathways downregulated, several play a role in the stem cell self-renewal program such as the polycomb EZH2 network (PARK_HSC_VS_MPP_DN, PARK_MSCS_DIFF) and WNT signaling (LIN_WNT_UP, KENNY_WNT_UP). The AKT/βcatenin signaling was also downregulated by ATRA treatment (BRCA2_BRCA1_UP). It was previously shown that AKT activation regulates normal and malignant breast stem cell self-renewal through phosphorylation of GSK3β resulting in the activation of the WNT pathway.17 These results underline the role of WNT signaling in the regulation of the breast CSCs activity. Moreover, we identified five gene sets whose expression is correlated with P53 function (P21_ANY_DN, P21_P53_ANY_DN, P21_P53_EARLY_DN, P21_EARLY_DN, P21_P53_MIDDLE_DN). The sets are a collection of downregulated downstream targets of P53 induced by the overexpression of P53 or P21 genes in human ovarian cancer cell lines.18 The P21/P53-signaling pathway has been implicated in the regulation of stem cell self-renewal as a molecular switch regulating the cell cycle entry of stem cells.19,20 Thus, P53 may play a crucial role in the regulation of breast CSCs differentiation.

Several other gene sets related to carcinogenesis process, metastatic activity, or drug resistance were downregulated by ATRA treatment. Interestingly, two gene expression signatures that predict poor clinical outcome of patients with hepatocellular carcinoma (HCC_SURVIVAL_GOOD_VS_POOR_DN) or acute myeloid leukemia (YAGI_AML_PROGNOSIS) were lost in ATRA-treated cells compared to DEAB-treated cells. These data are consistent with studies that described the CSC population as the driver of the carcinogenesis process from tumor initiation to metastasis formation.2 A subset of key stem cell properties including self-renewal, which drives tumorigenesis, and differentiation, which generate the bulk of tumor cells, are progressively lost during the differentiation process. This may explain the downregulation of several gene sets related to stem cell activity or tumor aggressiveness. Taken together, our results provide an important basis to identify and understand the different mechanisms that lead to the regulation of breast CSCs biology.

Conclusions and futures directions

Characterization of the pathways that regulate CSCs differentiation will facilitate the development of targeted therapies. In this study, utilizing a tumorsphere assay, we have demonstrated the role of retinoid signaling in the regulation of breast CSCs self-renewal and differentiation. We determined the effect of DEAB and ATRA treatment on different BCLs by gene expression profiling analysis. We previously established a breast CSC signature based on the comparison of the transcriptional profiles of breast CSCs versus differentiated cells from different BCLs.13 GSEA revealed an enrichment of genes overexpressed in the breast CSC signature in DEAB-treated cells, whereas genes underexpressed in our signature were significantly expressed in ATRA-treated cells. These results suggest that modulation of retinoid signaling may be sufficient to promote self-renewal or induce differentiation of breast CSCs. Similar results have been observed in the hematopoietic system with ALDH enzymatic activity inhibition, which induces an expansion of the HSC population.12

ATRA is routinely used as therapeutic agent to induce differentiation of leukemic stem cells in APML.11 Our results indicate that ATRA treatment induces the differentiation of breast CSCs resulting in a significant decrease of the breast CSC population. This suggests that ATRA may be considered as a therapeutic strategy to target the breast CSC population. Previous studies reported that ATRA treatment inhibits proliferation and invasion of breast cancer cells in vitro.21,22 Moreover, ATRA potentiates the cytotoxic effect of taxotere on BCLs.23 Furthermore, a phase I/II clinical trial tested the combination of ATRA and tamoxifen in patients with advanced breast cancer and showed that ATRA treatment reversed tamoxifen resistance in some patients.24 All these studies report encouraging data to further evaluate ATRA treatment as a potential therapeutic approach to improve breast cancer clinical outcome.

The importance of retinoids and their receptors in development has been recognized for years. Retinoic acid regulates several complexes of chromatin remodeling factors through its interaction with retinoid receptors (RARs).25,26 RARs bind to DNA and interact with corepressors such as nuclear receptor corepressor (NCoR) 1 and NCoR2/SMRT (silencing mediator of RAR and thyroid hormone receptor).27 These co-repressors interact directly with histone deacetylases (HDACs) that maintain chromatin in the condensed state that typifies transcriptional repression. HDACs play a central role in the epigenetic regulation of gene expression, and explain the importance of retinoid signaling in the control of CSCs activity. Interestingly, partial inactivation of RARβ in mice induces the formation of lung cancer.28 Moreover, RARβ expression is reduced in breast tumors due to the methylation of the promoter sequences.29 Taken together, these studies suggest that inhibition of retinoid signaling may be an initial step in carcinogenesis through the dysregulation of the self-renewal program leading to an increase of CSCs.

Due to the prominent role of retinoid signaling in the regulation of breast CSCs activity, the elucidation of the different networks related to this signaling will lead to a better understanding of CSCs biology. To determine the CSCs biological processes related to retinoic acid activity we used the GSEA algorithm to measure the enrichment of several curated gene sets (C2) in the list of genes differentially expressed between DEAB-treated cells and ATRA-treated cells. Several gene sets and pathways were associated with retinoid signaling and may constitute key regulatory pathways o f breast CSCs biology and provide new markers and therapeutic targets. Among these pathways, AKT/β-catenin, WNT, or P53 signaling are already known to be involved in CSCs activity.3 To eliminate the CSC population, several specific inhibitors of these pathways are currently tested. Recently, it was demonstrated that the AKT inhibitor, perifosine, is able to target the tumorigenic cell population in breast tumor xenografts.17 Small molecule inhibitors (nutlins, spirooxindales) of MDM2-P53 interactions have been developed to restore P53 function. The clinical use of nutlins selectively enhanced the cytotoxicity of chemotherapeutic agents in leukemic stem cells but not in normal hematopoietic progenitors, raising hope for the design of tailored molecular therapies.30,31 Several other gene sets identified by GSEA represent a potential source of new markers and therapeutic targets specific of the breast CSC population.

We previously showed that ALDH enzymatic activity is a marker of normal and malignant human mammary stem cells.7 We have now shown that ALDH cellular function plays a role in the regulation of breast CSCs biology through its effects on retinoid metabolism. We provide a rationale to study retinoid signaling in order to better understand the self-renewal and differentiation programs that regulates CSCs. Further investigation into the biological processes linking retinoid signaling and CSCs activity may identify new targets for the selective eradication of this cell population.

Supplementary Material

Sup Fig legend

Sup Tab 1

Sup Tab 2

Supp Fig 1


This work was supported by Ligue Nationale Contre le Cancer (Label DB) and Institut National du Cancer (PL2008).


1. Visvader JE, Lindeman GJ. Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nat Rev Cancer. 2008;8:755–768. [PubMed]
2. Wicha MS, Liu S, Dontu G. Cancer stem cells: an old idea--a paradigm shift. Cancer Res. 2006;66:1883–1890. [PubMed]
3. Korkaya H, Wicha MS. Selective targeting of cancer stem cells: a new concept in cancer therapeutics. BioDrugs. 2007;21:299–310. [PubMed]
4. Li X, Lewis MT, Huang J, Gutierrez C, Osborne CK, Wu MF, et al. Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. J Natl Cancer Inst. 2008;100:672–679. [PubMed]
5. Charafe-Jauffret E, Ginestier C, Birnbaum D. Breast cancer stem cells: tools and models to rely on. BMC Cancer. 2009;9:202. [PMC free article] [PubMed]
6. Al-Hajj M, Wicha MS, ito-Hernandez A, Morrison SJ, Clarke MF. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A. 2003;100:3983–3988. [PubMed]
7. Ginestier C, Hur MH, Charafe-Jauffret E, Monville F, Dutcher J, Brown L, et al. ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell. 2007;1:555–567. [PMC free article] [PubMed]
8. Sophos NA, Vasiliou V. Aldehyde dehydrogenase gene superfamily: the 2002 update. Chem Biol Interact. 2003;143–144:5–22. [PubMed]
9. Zechel C. Requirement of retinoic acid receptor isotypes alpha, beta, and gamma during the initial steps of neural differentiation of PCC7 cells. Mol Endocrinol. 2005;19:1629–1645. [PubMed]
10. Collins SJ. The role of retinoids and retinoic acid receptors in normal hematopoiesis. Leukemia. 2002;16:1896–1905. [PubMed]
11. Tallman MS, Andersen JW, Schiffer CA, Applebaum FR, Feusner JH, Ogden A, et al. All-trans-retinoic acid in acute promyelocytic leukemia. N Engl J Med. 1997;337:1021–1028. [PubMed]
12. Chute JP, Muramoto GG, Whitesides J, Colvin M, Safi R, Chao NJ, et al. Inhibition of aldehyde dehydrogenase and retinoid signaling induces the expansion of human hematopoietic stem cells. Proc Natl Acad Sci U S A. 2006;103:11707–11712. [PubMed]
13. Charafe-Jauffret E, Ginestier C, Iovino F, Wicinski J, Cervera N, Finetti P, et al. Breast cancer cell lines contain functional cancer stem cells with metastatic capacity and a distinct molecular signature. Cancer Res. 2009;69:1302–1313. [PMC free article] [PubMed]
14. Ponti D, Costa A, Zaffaroni N, Pratesi G, Petrangolini G, Coradini D, et al. Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer Res. 2005;65:5506–5511. [PubMed]
15. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: A knowledge-based approach fir interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102:15545–15550. [PubMed]
16. Park SG, Schimmel P, Kim S. Aminoacyl tRNA synthetases and their connections to disease. Proc Natl Acad Sci U S A. 2008;105:11043–11049. [PubMed]
17. Korkaya H, Paulson A, Charafe-Jauffret E, Ginestier C, Brown M, Dutcher J, et al. Regulation of mammary stem/progenitor cells by PTEN/Akt/beta-catenin signaling. PLoS Biol. 2009;7:e1000121. [PMC free article] [PubMed]
18. Wu Q, Kirschmeier P, Hockenberry T, Yang TY, Brassard DL, Wang L, et al. Transcriptional regulation during p21WAF1/CIP1-induced apoptosis in human ovarian cancer cells. J Biol Chem. 2002;277:36329–36337. [PubMed]
19. Cheng T, Rodrigues N, Shen H, Yang Y, Dombkowski D, Stier S, et al. Hematopoietic stem cell quiescence maintained by p21cip1/waf1. Science. 2000;287:1804–1808. [PubMed]
20. Molofsky AV, Pardal R, Morrison SJ. Diverse mechanisms regulate stem cell self-renewal. Curr Opin Cell Biol. 2004;16:700–707. [PubMed]
21. Mangiarotti R, Danova M, Alberici R, Pellicciari C. All-trans retinoic acid (ATRA)-induced apoptosis is preceded by G1 arrest in human MCF-7 breast cancer cells. Br J Cancer. 1998;77:186–191. [PMC free article] [PubMed]
22. Van hJ, Wouters W, Ramaekers FC, Krekels MD, Dillen L, Borgers M, et al. All-trans-retinoic acid metabolites significantly inhibit the proliferation of MCF-7 human breast cancer cells in vitro. Br J Cancer. 1998;77:26–32. [PMC free article] [PubMed]
23. Wang Q, Wieder R. All-trans retinoic acid potentiates Taxotere-induced cell death mediated by Jun N-terminal kinase in breast cancer cells. Oncogene. 2004;23:426–433. [PubMed]
24. Budd GT, Adamson PC, Gupta M, Homayoun P, Sandstrom SK, Murphy RF, et al. Phase I/II trial of all-trans retinoic acid and tamoxifen in patients with advanced breast cancer. Clin Cancer Res. 1998;4:635–642. [PubMed]
25. Matsushita H, Scaglioni PP, Bhaumik M, Rego EM, Cai LF, Majid SM, et al. In vivo analysis of the role of aberrant histone deacetylase recruitment and RAR alpha blockade in the pathogenesis of acute promyelocytic leukemia. J Exp Med. 2006;203:821–828. [PMC free article] [PubMed]
26. Segalla S, Rinaldi L, Kilstrup-Nielsen C, Badaracco G, Minucci S, Pelicci PG, et al. Retinoic acid receptor alpha fusion to PML affects its transcriptional and chromatin-remodeling properties. Mol Cell Biol. 2003;23:8795–8808. [PMC free article] [PubMed]
27. Jepsen K, Rosenfeld MG. Biological roles and mechanistic actions of co-repressor complexes. J Cell Sci. 2002;115:689–698. [PubMed]
28. Berard J, Laboune F, Mukuna M, Masse S, Kothary R, Bradley WE. Lung tumors in mice expressing an antisense RARbeta2 transgene. FASEB J. 1996;10:1091–1097. [PubMed]
29. Widschwendter M, Berger J, Hermann M, Muller HM, Amberger A, Zeschnigk M, et al. Methylation and silencing of the retinoic acid receptor-beta2 gene in breast cancer. J Natl Cancer Inst. 2000;92:826–832. [PubMed]
30. Vassilev LT, Vu BT, Graves B, Carvajal D, Podlaski F, Filipovic Z, et al. In vivo activation of the p53 pathway by small-molecule antagonists of MDM2. Science. 2004;303:844–848. [PubMed]
31. Ding K, Lu Y, Nikolovska-Coleska Z, Wang G, Qiu S, Shangary S, et al. Structure-based design of spiro-oxindoles as potent, specific small-molecule inhibitors of the MDM2-p53 interaction. J Med Chem. 2006;49:3432–3435. [PubMed]