SEARCH

SEARCH BY CITATION

Keywords:

  • Cancer stem cell;
  • Epithelial-to-mesenchymal transition;
  • LincRNA;
  • Hotair

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Hotair is a member of the recently described class of noncoding RNAs called lincRNA (large intergenic noncoding RNA). Various studies suggest that Hotair acts regulating epigenetic states by recruiting chromatin-modifying complexes to specific target sequences that ultimately leads to suppression of several genes. Although Hotair has been associated with metastasis and poor prognosis in different tumor types, a deep characterization of its functions in cancer is still needed. Here, we investigated the role of Hotair in the scenario of epithelial-to-mesenchymal transition (EMT) and in the arising and maintenance of cancer stem cells (CSCs). We found that treatment with TGF-β1 resulted in increased Hotair expression and triggered the EMT program. Interestingly, ablation of Hotair expression by siRNA prevented the EMT program stimulated by TGF-β1, and also the colony-forming capacity of colon and breast cancer cells. Furthermore, we observed that the colon CSC subpopulation (CD133+/CD44+) presents much higher levels of Hotair when compared with the non-stem cell subpopulation. These results indicate that Hotair acts as a key regulator that controls the multiple signaling mechanisms involved in EMT. Altogether, our data suggest that the role of Hotair in tumorigenesis occurs through EMT triggering and stemness acquisition. Stem Cells 2013;31:2827–2832


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

The cancer stem cell (CSC) hypothesis postulates that not all cells in a tumor have tumorigenic capacity. Instead, tumors consist of a diverse population of cells encompassing a small fraction of tumor-forming, chemotherapy-resistant, and self-renewal proficient CSC. The CSCs give rise to a large population of rapidly proliferating cells that differentiate into heterogeneous lineages of cancer cells that comprise the tumor mass [1]. Despite the fact that CSCs resemble normal stem cells in several aspects, they differ from their normal counterparts due the lack of control in processes such as proliferation and maintenance of genomic integrity.

In the last decade, great efforts have been made in order to elucidate the molecular pathways that drive CSC establishment. The breakthrough discovery of the connection between epithelial-to-mesenchymal transition (EMT) and CSC generation has proposed a rational explanation by which epithelial cancer cells may achieve self-renewing capacity (or stemness), and become able to invade and disseminate [2, 3]. The recapitulation of EMT by cancer cells as a mechanism to undergo the metastatic cascade is being widely investigated. Although some researchers are skeptical about the occurrence of EMT in human tumors, there is a large amount of evidence corroborating its occurrence in vitro and in vivo, as reported by distinct groups [4-9].

There is increasing evidence that mammalian cells produce thousands of large intergenic regulatory transcripts. Until recently, the functional significance of these transcripts was unclear, it was even suggested that they could represent “leaky” transcription. However, accumulating data from high-throughput genomic approaches provided strong evidence of the existence of large noncoding transcriptional units potentially implicated in diverse biological processes, including stemness control [10-14]. Hotair (Hox transcript antisense intergenic RNA) is a well-studied lincRNA that binds the Polycomb repressive complex 2 (PRC2) and directs it to target genes, promoting gene silencing by histone H3 lysine 27 trimethylation (H3K27me3) [15, 16]. Hotair expression supports metastasis in breast cancer and is associated with poor outcome in different neoplasias [15-18]. Here, we sought to evaluate the role of Hotair in EMT and CSCs. Our data have shown that Hotair expression drives EMT and CSC formation in different cancer cell lineages. These findings suggest that Hotair promotes tumorigenesis by activating the EMT genetic program.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Hotair Is Required for EMT

We tested whether Hotair is required to EMT through transient transfection with siRNAs. Two days after transfection with control or Hotair-directed siRNAs (siHotair), the colon cancer cell lines HT-29 and DLD1, the mammary epithelial cells MCF10a, and the HCC1954 breast cancer cell lines were submitted to TGF-β1 treatment for 3 days. Interestingly, the four cell lines analyzed showed increased levels of Hotair after TGF-β1 treatment. Transfection with siRNAs targeting Hotair efficiently reduced its expression even in the presence of TGF-β1 (Fig. 1A). As expected, mRNA levels of the epithelial marker E-cadherin decreased while the mesenchymal marker vimentin increased greatly after TGF-β1 treatment in cells transfected with siControl. However, knockdown of Hotair prevented E-cadherin reduction, while vimentin expression remained unaltered (Fig. 1B, 1C). In MCF10a and HCC1954 cells, fibronectin levels are greatly increased after TGF-β1 treatment, reaching a 50-fold change in MCF10a cells. When Hotair is silenced, fibronectin upregulation is totally inhibited in HCC1954, while in MCF10a the induction is partially abrogated. Although, in HT-29 and DLD1 cells, TGF-β1 treatment did not induce fibronectin expression, in DLD1 cells Hotair knockdown caused a significant reduction in fibronectin levels (Fig. 1D). Comparable results were observed for β-catenin in DLD1 and HT-29 cells, while in MCF10a cells we observed a modest but significant upregulation (Fig. 1E). Similar results were obtained when using an alternative siRNA against Hotair (Supporting Information Fig. S1). Collectively, these data provide evidence that Hotair is required to trigger EMT.

image

Figure 1. Knockdown of Hotair prevents breast and colon cancer cells to undergo epithelial-to-mesenchymal transition after TGF-β1 treatment. MCF10a, HCC1954, DLD1, and HT29 cell lines previously transfected with siHotair or siControl were treated with TGF-β1 for 3 days. qPCR was performed to evaluated the expression of (A) Hotair, (B) E-cadherin, (C) Vimentin, (D) fibronectin, and (E) β-catenin mRNA. The HPRT gene was used to normalize RNA inputs. Data are shown as mean ± SD from three independent experiments (*, p < .05; **, p < .01; ***, p < .002 compared to the control).

Download figure to PowerPoint

Hotair Is Highly Expressed in Colon CSC

Recent studies have demonstrated that the EMT program provides stem cell-like properties either to normal or tumor cells [2, 3, 9]. In this scenario, TGF-β1 signaling has an important role in the maintenance of stem cell-like properties and tumorigenic activity through induction of EMT. Since Hotair is necessary to promote the EMT program (Fig. 1), we tested whether Hotair has a role in CSC generation. We treated the colon cancer cells, HT-29 and DLD1, with TGF-β1 for 3 days and counted the number of cells expressing the CSC markers CD133+/CD44+ [19-22]. As expected, the TGF-β1 treatment promoted an increase in the number of CD133+/CD44+ (Fig. 2A, 2B).

image

Figure 2. TGF-β1 treatment increases the CD133+/CD44+ cancer stem cell subpopulation and Hotair expression in colon cancer cell lines. (A, B): HT-29 and DLD1 cells were treated with TGF-β1 for 72 hours and stained with anti-CD44 and anti-CD-133. Representative dot-plots are shown in A, while the percentage of cells with each phenotype are shown in B. (C): Enriched populations of CD133+/CD44+ and CD133/CD44 cells were subjected to qPCR analysis of Hotair. The HPRT gene was used to normalize RNA inputs. Data are shown as mean ± SD from three independent experiments.

Download figure to PowerPoint

We next performed cell sorting by flow cytometry after 3 days of TGF-β1 treatment and evaluated Hotair expression in CD133/CD44 and CD133+/CD44+ subpopulations. Gene expression analyses in HT-29 and DLD1 cells revealed that CD133+/CD44+ subpopulation expresses Hotair in levels 10 or 4-fold higher, respectively, when compared with the CD133/CD44 cells (Fig. 2C). Similar results were observed in nonstimulated cells, which contain a reduced number of CD133+/CD44+ cells (data not show).

Hotair Is Required for Stemness Phenotype

Considering that CD133+/CD44+ subpopulation expresses higher levels of Hotair, we hypothesized that Hotair is required to the arising of CSC. To address this issue, we performed sphere formation assay to assess self-renew potential. Two days after Hotair knockdown, DLD1 and MCF10a cells were used in colonosphere and mammosphere assays, respectively. After 15 days of incubation in anchorage-independent conditions, the control cells presented an elevated number of large colonies, while the cells silenced for Hotair formed only a few colonies of small size (Fig. 3A). Counting of colonies revealed a reduction of approximately 78% in colonosphere and mammosphere formation (Fig. 3B). In order to verify the effectiveness of Hotair knockdown during the sphere formation assays, we evaluated its expression 15 days after siRNAs treatment. Although, we have detected a slight recovery in Hotair expression, a significant knockdown was still observed (Fig. 3C). Therefore, our data suggest that Hotair is required for the maintenance of CSC phenotype in colon and breast cell lines.

image

Figure 3. Colonosphere and mammosphere formation are inhibited by Hotair knockdown. Representative images (A) and quantification (B) of colonosphere and mammosphere formed from DLD1 and MCF10a cells, previously transfected with siHotair or siControl. Spheres were counted by visual inspection in light microscopy. Data are shown as mean ± SD from three independent experiments. (C): qPCR was performed to evaluate the expression of Hotair in days 1 and 14 of the sphere assays. The HPRT gene was used to normalize RNA inputs. Data are shown as mean ± SD from three independent experiments (*, p < .05).

Download figure to PowerPoint

To get some insight into the mechanisms by which Hotair induces EMT and supports the CSC phenotype, we analyzed data from Gupta et al. [15] who investigated the role of Hotair in breast cancer metastasis. In this work, the authors globally determined the transcriptional profile, PCR2-members occupancy, and H3K27me3 of MDA-MBA-231 breast tumor cells ectopically expressing Hotair. We explored these results regarding a panel of 240 genes related to EMT and/or stemness, in order to verify the spectrum of gene pathways in which Hotair may be acting. We observed that the majority of EMT/stemness genes are regulated by Hotair modulation (Supporting Information Fig. S2).

Gene expression pattern, PCR2 occupancy, and H3K27me3 for 41 of them, which are master regulators of EMT/stemness process, corroborated the involvement of Hotair in EMT/stemness triggering (Fig. 4). In cells overexpressing Hotair, genes that induces EMT, such as ZEB1, SNAI1, TWIST, CTNNA1 (β-catenin), as well as the mesenchymal markers vimentin (VIM) and fibronectin (FN1), are upregulated. In accordance, e-cadherin (CDH1), GSK3B, BMP7, and ERBB3 were found downregulated. Stemness genes, such as SOX1, SOX10, and POU75F1 (OCT4), were also induced. Interestingly, the membrane markers CD44 and CD24 were found upregulated and downregulated, respectively, in cells overexpressing Hotair, as expected for breast CSCs [2]. Of note, we observed that the embryonic stem cell marker NANOG was repressed, this could be explained by the fact that NANOG is related to pluripotency rather CSC potential (reviewed in [23]).

image

Figure 4. Hotair overexpression promotes a broad alteration in the gene expression profile of EMT and stemness-related genes. We analyzed results of ChIP-chip (left panel) and gene expression (right panel) of MDA-MB-231 breast tumor cells ectopically expressing Hotair, for a selection of 35 EMT and/or stemness-related genes. All data used in this analysis were obtained from Gupta et al. [15]. Cells overexpressing Hotair or transfected with empty vector were subjected to ChIP using anti-EZH2, SUZ12, and H3K27me3 antibodies. Immunoprecipitated chromatin was used for hybridization in a genome-wide promoter array. Gene expression profiling was carried out in cells ectopically expressing Hotair with or without concomitant depletion of PRC2 by the use of shRNA against SUZ12. Abbreviation: EMT, epithelial-to-mesenchymal transition.

Download figure to PowerPoint

The ChIP-chip data of H3K27me3 and the PRC2 members SUZ12 and EZH2 are compatible with the gene expression pattern for the EMT/stemness (Fig. 4). The EMT inducers CTNNA1, TWIST1, SNAI1, ZEB1 and the mesenchymal markers VIM and FN1, display no H3K27me3 or occupancy by PRC2 members. As expected, genes that are downregulated during EMT, such as CDH1, PROM1, BMP7, presented increased PRC2 occupancy. Taken together, this analysis further supported our findings suggesting Hotair as a player in the activation of the genetic program that promotes EMT and supports CSC phenotype.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Here, we have shown that TGF-β1 increased Hotair expression and triggered the EMT program in different tumor cells. More importantly, ablation of Hotair expression prevented EMT induction and CSCs arising. The TGF-β1 pathway induces the EMT program through multiple signaling mechanisms that can act independently or in concert to trigger EMT [24-26]. Once inhibition of Hotair prevents cells to undergo EMT after TGF-β1 stimulation, we can hypothesize that Hotair acts as a key regulator controlling different signaling mechanisms involved in EMT/stemness establishment. These data are consistent with the transcriptional regulatory layer in which Hotair operates [15, 27]. Also, this notion is further supported by in vitro studies where ectopic expression of Hotair leaded to a fourfold increase in the expression of the EMT-inducer Snail [15]. These findings suggest EMT as a possible mechanism by which Hotair promotes metastasis.

Hotair interacts with PRC2 and is required for H3K27me3 in the target sites [28]. E-cadherin has been identified among the many genes repressed by H3K27me3 mark mediated by the PRC2 [29]. Additionally, a component of PRC2, Suz12, has been reported as associated with CSC formation by increasing PRC2 repression of E-cadherin [30]. Thus, the increase in Hotair expression induced by TGF-β1 (Fig. 1A) is possibly a central event that promotes the repression of E-cadherin gene by PRC2. This hypothesis is corroborated by our expanded analysis of gene expression profile, PRC2 occupancy, and H3K27me3 [15] in which we observed activation of EMT and stemness genetic programs in cells overexpressing Hotair. Accordingly, in our experimental model, the acquisition of a mesenchymal state was prevented by ablation of Hotair. Thereby, the regulation carried out by Hotair along with the PRC2 may promote a global chromatin state more permissive to EMT and CSC formation.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Altogether, our data show for the first time that the role of the lincRNA Hotair in tumorigenesis occurs through EMT triggering and stemness acquisition in colon and breast cell lines. Future studies are warranted to further elucidate the in vivo function of Hotair in tumorigenesis.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

We thank Anemari Ramos Dinarte dos Santos, Cristiane Ayres Ferreira, Flow Cytometry Core at National Institute of Science and Technology in Stem Cell and Cell Therapy for the technical support and Claudia Marchini Alves for critical review of the manuscript. This work was supported by grants 2012/00588-5 and 1998/14247-6 , São Paulo Research Foundation (FAPESP), CNPQ, CAPES, and FUNDHERP.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. Disclosure of Potential Conflicts of Interest
  10. References
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
stem1547-sup-0001-suppinfo.doc1792KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.