Abstract
The cancer stem-like cell (CSC) hypothesis postulates that a small population of cells in a cancer has self-renewal and clonal tumor initiation properties. These cells are responsible for tumor initiation, growth, recurrence and for resistance to chemotherapy and radiation therapy. CSCs can be characterized using markers such as SSEA-1, SSEA-4, CD44, CD24, ALDEFLUOR and others. CSCs form spheres when they are cultured in serum-free condition in low attachment plates and can generate tumors when injected into immune-deficient mice. During epithelial to mesenchymal transition (EMT), cells lose cellular adhesion and polarity and acquire an invasive phenotype. Recent studies have established a relationship between EMT and increased numbers of CSCs in some solid malignancies. Non-coding RNAs such as microRNAs and long non-coding RNAs (lncRNAs) have been shown to have important roles during EMT and some of these molecules also have regulatory roles in the proliferation of CSCs. Specific lncRNAs enhanced cell migration and invasion in breast carcinomas, which was associated with the generation of stem cell properties. The tumor microenvironment of CSCs also has an important role in tumor progression. Recent studies have shown that the interaction between tumor cells and the local microenvironment at the metastatic site leads to the development of premetastatic niche(s) and allows for the proliferation of the metastatic cells during colonization. The role of exosomes in the microenvironment during the EMT program is currently a major area of research. This review examines CSCs and the relationship between EMT and CSCs in solid tumors with emphasis on thyroid CSCs. The role of non-coding RNAs and of the microenvironment in EMT and in tumor progression are also examined. This review also highlights the growing number of studies that show the close association of EMT and CSCs and the role of exosomes and other elements of the tissue microenvironment in CSC metastasis. A better understanding of these mechanisms will lead to more effective targeting of primary and metastatic malignancies.
Similar content being viewed by others
Main
The cancer stem-like cell (CSC) hypothesis was first proposed for human leukemia based on the observation that a small fraction of cells could generate leukemia in severe combined immunodeficient mice, while the majority of tumor cells failed to engraft and cause disease.1 A few years later Bonnet et al2 isolated leukemia cells expressing CD34 but lacking CD38. When injected into immunodeficient mice they were able to reproduce leukemic transformation. Subsequently, isolation of CSCs from human brain gliomas, which were shown to be clonogenic with sphere-forming abilities by Ignatova et al3 and characterization of CSCs from breast cancer tissues by Al-Hajj et al4 who showed that breast carcinoma CSCs had a CD44+/CD24– phenotype and formed tumors in immunodeficient mice demonstrated that solid tumors in addition to hematologic malignancies also contained CSCs (Figure 1). Various methods permitting isolation and characterization of CSCs have emerged (Table 1). It appears that a small population of cells in cancer that is capable of self-renewal is responsible for tumor initiation, growth and recurrence in a variety of cancers.5, 6, 7, 8, 9, 10, 11, 12
Numerous biomarkers to characterize CSCs have been reported. However, the specificity of many of these markers in terms of their ability to differentiate CSCs vs other cancer cells has not been completely demonstrated7, 8, 9, 10, 11 (Table 1). Non-neoplastic stem cells, both embryonic and somatic, have the ability for self-renewal by continuous division and can be induced to become tissue/organ-specific cells with specialized functions. Non-neoplastic stem cells include embryonic stem cells characterized by pluripotency and can give rise to the three traditional germ layers—ectoderm, endoderm and mesoderm, and somatic or adult stem cells, which are tissue-specific cells with a limited life-span that give rise to cells in a specific lineage. The role of embryonic stem cells and adult stem cells in the development of CSCs remains controversial.6, 13, 14, 15 Recent studies have suggested that the stem cell programs operating in tumor-initiating cells (TICs) and non-neoplastic stem cells of the corresponding normal tissues are likely to differ significantly.6
Solid tumors are usually heterogeneous at the histological, as well as the molecular level. Tumor heterogeneity can be explained by two possible models: the stochastic model and the hierarchical model. The stochastic model assumes that cancer results from random mutations within oncogenes and tumor-suppressor genes leading to excessive proliferation.16, 17, 18 In contrast, the hierarchical model suggests that tumor cells originate from a small population of cells (CSCs) that have similar characteristics to adult stem cells. In this model, tumor development occurs when an adult stem cell bypasses regulation or terminally differentiated cells undergo dedifferentiation. Although some investigators have proposed only one model to explain CSCs in the development of thyroid cancers,19, 20 it is very likely that both models are operational and contribute to cancer development.
The CSC hypothesis has been challenged by some investigators.17, 21, 22 Kern and Shibata21 critiqued the conflicting evidence regarding the presence of CSCs by re-evaluating the published CSC studies and stem cell-related claims in US-awarded patents from a mathematical perspective. They argued that the phenomenon of treatment response and relapse is difficult to explain mathematically when assuming a consistent small population (CSCs) as the only source of tumor regeneration and resistance. Rather, tumorigenic property is more likely to be a common function that most tumor cells are able to assume under certain conditions. Antoniou et al22 suggested that there was no convincing evidence to support the CSC concept and that the phenotypic properties ascribed to CSCs represented the result of independent, linked or congruent genetic, epigenetic or other signaling programs.22 However, in spite of these and other objections, the continued accumulation of experimental data provides increasing support for the CSC hypothesis.17, 23, 24, 25, 26, 27, 28
Epithelial to mesenchymal transition
During epithelial to mesenchymal transition (EMT), cells lose cellular adhesion and polarity and acquire an invasive phenotype.29, 30 EMT is associated with loss of E-cadherin and increased expression of other biomarkers including Snail, Slug, ZEB1, ZEB2 and N-cadherin (Figures 1 and 2; Table 2). EMT is first observed during embryogenesis in the developing fetus, but is also present during wound healing, tumor progression and tumor invasion.29, 30, 31, 32, 33 In thyroid tumors, EMT has been linked to papillary thyroid carcinoma (PTC) progression with overexpression of vimentin and has been associated with invasion and lymph node metastasis.34 EMT has been associated with dedifferentiated cancers such as anaplastic thyroid carcinomas (ATCs) with decreased expression of E-cadherin and membranous β-catenin and increased expression of Slug and Twist1.29, 30, 31, 32, 33 Using a breast cancer model, Mani et al23 showed that acquisition of mesenchymal traits was linked to the acquisition of stem cell markers. They further showed that transformed mammary epithelial cells that had undergone EMT formed spheres, soft agar colonies and formed in vivo tumors more efficiently suggesting that the EMT program that enabled cancer cells to spread from a primary tumor may also promote their self-renewal properties.23 The studies of the Weinberg laboratory with a breast cancer model and primary human breast cancers demonstrated that during the EMT program there was an increase in the CSC population in breast cancers.25 Similar findings have been shown in the thyroid.34, 35, 36, 37 Hardin et al36, 37 used a model of cultured thyroid cancer cells treated with TGFβ1, which led to the development of EMT. These findings supported the earlier report of Buehler et al35 showing that primary thyroid carcinoma specimens underwent EMT during thyroid cancer progression from PTCs to ATCs. The studies of Hardin et al36, 37 showed that when thyroid tumor cell lines, including well-differentiated PTC cell lines, as well as ATC cell lines, undergo EMT, there is also an increase in the number of stem cells in the tumor population. Similar findings were reported in colorectal cancer under the influence of Notch-1 expression.38
Mesenchymal to epithelial transition (MET) is the reverse process of EMT and also has a central role in embryogenesis and tumor progression.29, 30, 31 The studies of Tsai et al39 provided some insight into the mechanism by which EMT and MET were implicated in converting epithelial tumor cells into motile mesenchymal cells during metastases. They used a mouse model to show that Twist1 expression was sufficient to promote squamous cell carcinoma to undergo EMT and disseminate into the circulation. Then, at distant sites Twist1 was turned off and allowed reversion of EMT to MET, which was important for the disseminated tumor cells to proliferate and form metastases.39 Ocaña et al40 recently described a new EMT regulator, paired homeobox factor 1 transcription factor (PRRX1), which regulates migratory and invasive properties during EMT. Cells that successfully colonized after metastasis usually go through MET to colonize the metastatic site.40 This group showed that PRRX1 was increased during EMT and subsequently decreased during MET, which facilitated colonization and secondary tumor formation, therefore, PRRX1 appears to have a role in invasion and migration of cancer cells as they metastasize.40 This observation was supported by using forced expression of PRRX1, which blocked the capacity of metastasis-competent cells to produce metastatic tumors and supported the concept that PRRX1 suppression was needed for MET.40 Recent studies by Hardin et al36 showed that PRRX1 was important for EMT in thyroid cancers. Using a TGFβ1 model of EMT induction in PTC cell lines, it was shown that PRRX1 was overexpressed along with Slug and Twist during tumor progression. Analysis of low-grade and highly aggressive human thyroid cancers showed that only the highly aggressive cancers overexpressed PRRX1.36
The plasticity of CSCs and the EMT and MET programs suggest that the therapeutic targeting of CSCs may be very challenging. A great deal of research on EMT and MET is directed at inhibiting cancer cells in their mesenchymal state to prevent metastatic disease. However, such approaches may have stimulatory effects on existing metastatic foci or may even activate dormant cancer cells.41, 42 The concept of metastatic dormancy during the metastatic cascade is an important feature of metastatic disease.41 Metastatic dormancy may include physical barriers provided by extracellular matrices or protein expression encoded by suppressor genes. The goals of understanding the biology of EMT and MET should include designing better drugs to target cancer cells including CSCs more effectively.41, 42
Signaling pathways in cscs
Specific signaling pathways have been described in embryonic stem cells and more recently in CSCs. These pathways are critical in the regulation of self-renewal and survival for both stem cells and CSCs. These pathways include, among others, the Notch, Wnt/β-catenin and hedgehog (HH) pathways. They have been analyzed in various types of cancers including those in the thyroid.7, 43, 44, 45, 46, 47, 48, 49, 50
Notch pathway
Mammalian cells express four transmembrane Notch receptors including NOTCH-1, NOTCH-2, NOTCH-3 and NOTCH-4.43, 44, 45, 46, 47, 48, 49, 50 Notch signaling has been implicated in self-renewal in CSCs in a variety of cancers. Notch receptors are expressed during the development of thyrocytes in experimental systems and the expression levels of these receptors are aligned with thyroid differentiation markers. Studies have shown that overexpression of NOTCH-1 in thyroid cancer cells restores differentiation, reduces cancer cell growth and stimulates differentiation markers such as sodium iodide transporter (NIS).48 In contrast, other studies using clinicopathologic correlation have suggested that Notch-1 receptor may be a predictor of lymph node metastasis and may be a poor prognostic marker in patients with PTCs.50 A recent study by Yu et al47 using primary thyroid cancers showed that Notch-1 expression levels were downregulated in differentiated thyroid cancer tissues compared with benign thyroid tissues. Decreased Notch-1 was more common in more aggressive tumors with extrathyroidal invasion. Restoration of Notch-1 intracellular domain in a stable doxycycline-inducible metastatic differentiated thyroid carcinoma cell line led to a reduction in cell growth and tumor cell migration.47
Wnt/β-catenin
Wnt/β-catenin or canonical Wnt pathway leads to the accumulation of β-catenin in the cell cytoplasm, which is then translocated into the nucleus and acts as a transcriptional co-activator of the TCF/LEF family of transcription factors. Experimental evidence has shown that Wnt2/2b and β-catenin signaling are necessary and sufficient to specify lung progenitors in the foregut.51 The role of the Wnt pathway in thyroid cancer and CSCs has been studied using primary thyroid cancer cells.7 CSCs isolated from ATCs were most aggressive and tumorigenic compared with CSCs isolated from well differentiated papillary and follicular cancers. Constitutive activation of β-catenin, together with downstream regulation of E-cadherin in CSCs derived from ATCs, correlated with higher migration and metastatic rates compared with well-differentiated thyroid cancers.7
Hedgehog
The HH pathway is important for the maintenance of CSCs in a variety of tumors including gliomas, colorectal and gastric cancers.52, 53, 54 In vitro activation of the HH pathway has been associated with increased expression of the key HH signaling components such as SMO and GLI1.52 Inhibition of the HH pathway has been associated with decreased expression of genes associated with cellular growth and increased apoptosis. Sonic HH (SHH) signaling pathway was activated in benign thyroid tumors such as follicular adenomas.54 Heiden et al53 reported that SHH signaling maintained the CSC self-renewal of ATCs by inducing Snail expression. When SHH inhibitors were used to knockdown Gli1 in an ATC cell line, there was decreased CSC self-renewal and increased radiation sensitivity. Subsequent studies showed that SHH signaling pathway stimulated ATC cell motility and invasiveness by activating Akt and c-Met.52
Epigenetic regulation in cscs
Epigenetic regulation has been proposed to explain the phenotypic plasticity that has been reported in CSCs.13, 32, 55 Phenotypic plasticity may allow non-CSCs to acquire CSC traits and replenish the CSC pool as has been reported in malignancies such as malignant melanomas56, 57 and prostate cancers.58 Roesch et al57 used the H3K4 demethylase JARID1B as a biomarker and characterized a small sub-population of slow-cycling melanoma cells that cycled with doubling times of >4 weeks within the rapidly proliferating main population. Isolated JARID1B-positive melanoma cells gave rise to highly proliferative progeny. Knockdown of JARID1B led to an initial acceleration of tumor growth followed by exhaustion, which suggested that JARID1B-positive sub-population was essential for continuous growth. Lee et al58 reported that reduced expression of DNA methyltransferase 1 (DNMT1) had an important role in the induction of EMT and CSC phenotype in prostate cancer cells. After reducing DNMT1 by 5-azacytidine, there was promotion of EMT induction, as well as CSC sphere formation. Reduction of DNMT1 was associated with suppression of H3K9me3 and H3K27 on the Zeb2 and KLF4 promoter region of the prostate cancer cells. In other cancers, such as colorectal cancer, aberrant methylation was reported to have important roles in accumulation of damages to cell function and led to malignant transformation.59 Similarly, DNA methylation and histone acetylation have been reported to have important roles in the epigenetic regulation in heterogeneity of lung cancer.60 Thus, epigenetic regulation appears to have a major role in explaining the plasticity observed in CSCs and non-CSCs in vitro.
Non-coding rnas and emt
MicroRNAs and EMT
MicroRNAs are small 18–24 nucleotide molecules that regulate mRNA at a posttranslational level by binding to an eight-base seed sequence at the 3′-UTR of mRNAs. They have a critical role during development and are linked to various diseases including cancers.61, 62, 63, 64, 65 MicroRNAs have been shown to have a role in tumor development and progression, as they may function as oncogenes and/or tumor-suppressor genes in thyroid and other cancers.64 MicroRNAs are increasingly recognized as important regulators of EMT and of CSCs.65, 66, 67, 68, 69, 70 Recent studies have shown that the miR-200 family has a major regulatory role in EMT.65 The miR-200 family has been shown to regulate EMT induced by EGF/EGFR66 and also by targeting TGFβR1.65 Braun et al65 used ATC cell lines to show that overexpression of miR-200 and miR-30 in mesenchymal ATC-derived cells reduced their invasive potential and induced MET. Expression of SMAD2 and TGFβR1, which were upregulated in ATC cell lines, was controlled by members of the miR-30 and/or miR-200 families. These findings suggested that microRNA signatures are potent markers for ATCs that leads to EMT and invasion of these cancers. Zhang et al66 used ATC cell lines to examine the relationship between EGF/EGFR, miR-200 and the induction of EMT or MET. They reported that EGFR silencing in ATC cell lines led to upregulation of E-cadherin and downregulation of vimentin indicating transition toward MET. Hebrant et al67 found that specific miRNAs in ATCs had a regulatory role in EMT. They reported that the LOX gene, which is a principal factor in EMT, was an upregulated mRNA target in ATCs associated with downregulation of miR-29a.
MiRNAs also have regulatory roles in well-differentiated thyroid cancers during EMT.68, 69, 70 MiR-146b-5p has been shown to regulate PTC cells during EMT.69 Hardin et al36 used a TGFβ-induced model of EMT to show that miR-146b-5p was initially upregulated during the first fourteen days of TGFβ treatment, which was followed by downregulation of this microRNA with subsequent treatment for two weeks. Deng et al69 later showed that miR-146b-5p targeted ZNRF3, increased the cell surface Wnt receptors Frizzled-6 and LRP6 and decreased Wnt/β-catenin signaling, as an inhibitor of Wnt/β-catenin suppressed the effect of miR-146b-5p on migration, invasiveness and EMT induction in PTC cell lines. Other microRNAs are also important in the regulation of thyroid cancer. A recent study by Wang et al70 examined the role of miR-663 in PTCs. This miRNA has been detected in a large variety of tumor types and has been shown to have both an oncogenic and a tumor-suppressor role in different tumors. In this study, miR-663 expression was inhibited in PTC tissue samples and PTC cell lines as assessed by qPCR. Upregulation of miR-663 inhibited cell invasion and migration in PTC cell lines.70 The authors suggested that by targeting TGFβ1 miR-663 regulated EMT, as well as matrix metalloproteinases.70
Long non-coding RNAs and EMT
Long non-coding RNAs (lncRNAs) are transcripts 200 nucleotides in length or larger that do not contain a protein-coding sequence. LncRNAs are very abundant with estimates of >30 000 lncRNAs transcripts in the human genome.71, 72, 73, 74, 75, 76, 77 One group of investigators estimated that there were close to 60 000 lncRNAs including 8000 oncologic lncRNAs.74 A growing number of these lncRNAs are being isolated and characterized from thyroid and other cancers.75 Some lncRNAs are upregulated, whereas others are downregulated in thyroid cancers, especially in PTCs, which has been the most frequent type of thyroid cancers used to study the role of lncRNAs in thyroid tumors to date. An interesting finding was that some of these lncRNAs have important roles in EMT and in stem cells, as well as in development.75, 76, 77 Examples of three lncRNAs associated with tumor progression, EMT and CSCs are summarized below.
Metastasis-associated lung adenocarcinoma transcript 1 (malat1)
MALAT1 is a widely expressed lncRNA that is highly conserved. MALAT has been found to be highly overexpressed in several human malignancies and has been associated with tumor invasion and metastasis in various cancers.78, 79, 80, 81, 82, 83, 84
Jiao et al78 reported that MALAT1 could increase the proportion of pancreatic CSCs, maintain self-renewal capacity, decrease chemosensitivity to anticancer drugs and promote angiogenesis in vitro in pancreatic cancer cell lines. Downregulation of MALAT1 reduced self-renewal-associated factors, such as SOX2, Bmi and Nanog. Treatment with siRNA led to less tumor growth in xenografted pancreatic cancer tumor cells.
Recent studies by Zhang et al79 using a papillary thyroid cancer cell line (TPC-1) showed that MALAT1 expression was markedly increased in vitro when the tumor cells were treated with TGFβ1, which induced EMT and led to an increase in expression of EMT markers. As analogous studies by Hardin et al36 with these same cell lines had previously showed that EMT is associated with an increase in thyroid CSCs, it is possible that there was also an increase in thyroid CSCs in experiments by Zhang et al.79
Hox transcript antisense intergenic rna (hotair)
HOTAIR is a well-studied lncRNA that binds the polycomb repressive complex 2 and directs it to target genes, and promotes gene silencing by histone H3 lysine 27 trimethylation.85, 86, 87, 88, 89 HOTAIR expression supports metastasis in breast cancer and has been linked to poor outcome in several malignancies.85 Padua Alves et al86 examined breast and colon cancer cell lines in studies of HOTAIR regulation and found that this lncRNA was increased after TGFβ1 treatment in both breast and colon cancers. This led to an increase in the EMT program, as well as colony-forming capacity of the colon and breast cancer cells. These findings support a role for HOTAIR in EMT and in stem cell formation. SiRNA experiments prevented the EMT program stimulated by TGFβ1 and the colony-forming capacity in the two cell lines.86 HOTAIR has also been implicated in the regulation of CSCs in cervical cancer and endometrial cancer.87, 88, 89
Linc-ror
Linc-ROR was first discovered in induced pluripotent stem cells (iPSCs) where it was regulated by the key pluripotency factors Oct4, Sox2 and Nanog.90 Linc-ROR has a principal role in maintenance of iPSCs and embryonic stem cells by preventing the activation of cellular stress pathways including the p53 response.90 Our laboratory has localized linc-ROR in ATC cell lines and in a CSC clone by in situ hybridization (Figures 3c and d) and by RT-PCR (data now shown). The levels of linc-ROR in the CSC clone was noted to be higher than in the parental ATC cell line (Figures 3c and d) from which the CSC clone was derived. Linc-ROR has a regulatory role in multiple cancer types.91, 92, 93, 94, 95, 96 Recent studies have shown that linc-ROR may function as a molecular sponge for miR-145 targets, such as OCT4, SOX2 and Nanog.92 Hou et al91 examined linc-ROR in breast tissue specimens and breast cancer cell lines. They found that linc-ROR enhanced cell migration and invasion, which was associated with the generation of stem cell properties. Linc-ROR was also found to function as a competing endogenous RNA to miR-205 and prevented the degradation of miR-205 target genes including the EMT inducer. In nude mice, silencing of linc-ROR expression inhibited breast cancer tumor growth and lung metastasis.90
Role of the microenvironment and exosomes in emt and CSCs
A growing body of evidence indicates that the interaction between tumor cells and the local microenvironment at the metastatic site leads to the development of the premetastatic niche(s), which allows for the proliferation of metastatic cells during colonization.97 Tumor cells can produce various factors leading to the establishment of the premetastatic niche(s). The microenvironment consist of the tumor stroma, the vasculature including endothelium, smooth muscle cells and pericytes, inflammatory and immune cells such as macrophages, lymphocytes and dendritic cells, cancer-associated fibroblasts (CAFs) and the extracellular matrix non-cellular components. Interactions with adhesion molecules, such as integrins and CD44, proteins, such as thrombospondin and fibronectin, and soluble cytokines all comprise the tumor microenvironment.98
Exosomes, which are an important part of the microenvironment, range in size from 40 to 150 nm and represent a subset of multi-vesicular bodies (MVB).97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107 They are formed by inward budding of the limiting membrane of MVB and are released from the cell into the extracellular environment upon fusion of the MVB and the plasma membrane. Exosomes contain a concentrated group of functional molecules including proteins, lipids, mRNAs, non-coding RNAs and DNA molecules. Exosomes provide protection for the transported molecules and also serve as intercellular communications locally and systemically.98 Exosomes interact with the microenvironment in different ways: (a) they can activate target cells through the transfer of ligands such as different growth factors and (b) they can transfer receptors from one cell to another by fusion with the plasma membrane of recipient cells. Exosomes get to their targets by endocytosis and transfer the cargo directly into the cytosol of recipient cells. Thus, exosomes are central mediators of cell–cell communication (Figure 1).98, 103 Prokaryotic and eukaryotic cells release exosomes including cancer cells from a variety of malignancies.
Vesicle-based cell–cell communication by exosomes has been reported during EMT and cancer progression.99, 100, 101 These molecules can be internalized directly by stromal and cancer cells and can induce functional changes in the recipient cells. Positive feedback mechanisms can lead to changes in exosome production during tumorigenesis. Deregulation of exosome secretion pathways in tumor cells could lead to an altered microenvironment and modulation of tumor progression.107 Interactions between cancer cells and components of the tumor stroma have been shown to facilitate EMT induction and drive tumor progression. It has been shown that CAFs facilitate aggressive phenotype of breast cancer cell lines by induction of EMT by TGFβ in a paracrine manner.99 Other studies have shown that metastatic lung cells were regulated by miR-200 expression and were responsive to TGFβ.100, 101 Metastatic lung cancer cells alternated between epithelial and mesenchymal states and formed polarized epithelial spheres in 3D culture in the presence of TGFβ.101 In these experiments, forced expression of miR-200 inhibited the tumor cells from undergoing EMT-associated invasion and metastasis.100 The regulatory role of exosomes in thyroid cancer has recently been examined. Lee et al102 studied exosomes from the PTC cell line TPC-1 and compared these with a non-neoplastic thyroid cell line NTHY-ori 3-1 (NTHY). Exosomes derived from TPC-1 contained miR-146b-5p and miR-222 as cargo. Both miRNAs showed significantly higher expression in TPC-1 cells compared with NTHY control cells. They found similar miRNA composition within exosomes and in cells suggesting that these miRNAs were actively and selectively packaged into exosomes for secretion.102
Other studies have shown that exosomes derived from cancer cells have an important role in premetastatic niche formation (Figure 1).103, 104, 105, 106, 107 Peinado et al103 showed in a series of elegant experiments that melanoma exosomes could educate bone marrow progenitor cells into a prometastatic phenotype through MET. The accumulating data would suggest that exosomes mediate communication at both the primary and metastatic tumor sites and that the content of the exosomes can induce a pro-tumorigenic niche and support cell transformation.104 However, the mode of cargo selection and regulation remain major challenges in understanding the effects of exosomes on CSCs and on the microenvironment.
Summary
The CSC hypothesis has been supported by a great deal of experimental evidence over the past decade. Many biomarkers that recognized CSCs have facilitated their isolation and characterization in vitro, but much work is needed to demonstrate the specificity of these biomarkers for CSCs. The close relationship between EMT and CSCs has provided further insights into tumor progression and contribution of CSCs to tumor metastasis and colonization. Recent studies have also established roles for microRNAs and lncRNAs in EMT and in the regulation of CSCs. A growing body of evidence has increased our understanding of the interaction between tumor cells and the local microenvironment, including the role of exosomes, leading to tumor metastasis and colonization. A better understanding of these mechanisms will lead to more effective targeting of primary and metastatic malignancies.
References
Lapidot T, Sirard C, Vormoor J et al, A cell initiating human acute myeloid leukemia after transplantation in SCID mice. Nature 1994; 367: 645–648.
Bonnet D, Dick JE . Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 1997; 3: 730–737.
Ignatova TN, Kukekov VG, Laywell ED et al, Human cortical glial tumors contain neural stem-like cells expressing astroglial and neuronal markers in vitro. Glia 2002; 39: 193–206.
Al-Hajj M, Wicha MS, Benito-Hernandez A et al, Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci USA 2003; 100: 3983–3988.
Clevers H . The cancer stem cell: premises, promises and challenges. Nat Med 2011; 17: 313–319.
Ye X, Tam WL, Shibue T et al, Distinct EMT programs control normal mammary stem cells and tumour-initiating cells. Nature 2015; 525: 256–260.
Guo Z, Hardin H, Lloyd RV . Cancer stem-like cells and thyroid cancer. Endocr Relat Cancer 2014; 21: T285–T300.
Montemayor-Garcia C, Hardin H, Guo Z et al, The role of epithelial mesenchymal transition markers in thyroid carcinoma progression. Endocr Pathol 2013; 24: 206–212.
Sato R, Semba T, Saya H et al, Concise review: stem cells and epithelial-mesenchymal transition in cancer: biological implications and therapeutic targets. Stem Cells 2016; 34: 1997–2007.
Findlay VJ, Wang C, Watson DK et al, Epithelial-to-mesenchymal transition and the cancer stem cell phenotype: insights from cancer biology with therapeutic implications for colorectal cancer. Cancer Gene Ther 2014; 21: 181–187.
Creighton CJ, Gibbons DL, Kurie JM . The role of epithelial-mesenchymal transition programming in invasion and metastasis: a clinical perspective. Cancer Manag Res 2013; 5: 187–195.
Deng J, Wang L, Chen H et al, Targeting epithelial-mesenchymal transition and cancer stem cells for chemoresistant ovarian cancer. Oncotarget 2016; 7: 55771–55788.
Scheel C, Weinberg RA . Phenotypic plasticity and epithelial-mesenchymal transitions in cancer and normal stem cells? Int J Cancer 2011; 129: 2310–2314.
Fierabracci A, Puglisi MA, Giuliani L et al, Identification of an adult stem/progenitor cell-like population in the human thyroid. J Endocrinol 2008; 198: 471–487.
Fierabracci A . Identifying thyroid stem/progenitor cells: advances and limitations. J Endocrinol 2012; 213: 1–13.
Alison MR, Lin WR, Lim SM et al, Cancer stem cells: in the line of fire. Cancer Treat Rev 2012; 38: 589–598.
Fanali C, Lucchetti D, Farina M et al, Cancer stem cells in colorectal cancer from pathogenesis to therapy: controversies and perspectives. World J Gastroenterol 2014; 20: 923–942.
Wang T, Shigdar S, Gantier MP et al, Cancer stem cell targeted therapy: progress amid controversies. Oncotarget 2015; 6: 44191–44206.
Takano T . Fetal cell carcinogenesis of the thyroid: theory and practice. Semin Cancer Biol 2007; 17: 233–240.
Takano T . Fetal cell carcinogenesis of the thyroid: a modified theory based on recent evidence. Endocr J 2014; 61: 311–320.
Kern SE, Shibata D . The fuzzy math of solid tumor stem cells: a perspective. Cancer Res 2007; 67: 8985–8988.
Antoniou A, Hebrant A, Dom G et al, Cancer stem cells, a fuzzy evolving concept: a cell population or a cell property? Cell Cycle 2013; 12: 3743–3748.
Mani SA, Guo W, Liao MJ et al, The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell 2008; 133: 704–715.
Tam WL, Weinberg RA . The epigenetics of epithelial-mesenchymal plasticity in cancer. Nat Med 2013; 19: 1438–1449.
Yang J, Weinberg RA . Epithelial-mesenchymal transition: at the crossroads of development and tumor metastasis. Dev Cell 2008; 14: 818–829.
Bhatia P, Tsumagari K, Abd Elmageed ZY et al, Stem cell biology in thyroid cancer: insights for novel therapies. World J. Stem Cells 2014; 6: 614–619.
O’Connor ML, Xiang D, Shigdar S et al, Cancer stem cells: a contentious hypothesis now moving forward. Cancer Lett 2014; 344: 180–187.
Todaro M, Iovino F, Eterno V et al, Tumorigenic and metastatic activity of human thyroid cancer stem cells. Cancer Res 2010; 70: 8874–8885.
Thiery JP . Epithelial-mesenchymal transitions in tumour progression. Nat Rev Cancer 2002; 2: 442–454.
Thiery JP, Acloque H, Huang RY et al, Epithelial-mesenchymal transitions in development and disease. Cell 2009; 139: 871–890.
Nieto MA, Huang RY, Jackson RA et al, EMT: 2016. Cell 2016; 166: 21–45.
Kalluri R, Weinberg RA . The basics of epithelial-mesenchymal transition. J Clin Invest 2009; 119: 1420–1428.
Massague J, Obenauf AC . Metastatic colonization by circulating tumour cells. Nature 2016; 529: 298–306.
Hardy RG, Vicente-Duenas C, Gonzalez-Herrero I et al, Snail family transcription factors are implicated in thyroid carcinogenesis. Am J Pathol 2007; 171: 1037–1046.
Buehler D, Hardin H, Shan W et al, Expression of epithelial-mesenchymal transition regulators SNAI2 and TWIST1 in thyroid carcinomas. Mod Pathol 2013; 26: 54–61.
Hardin H, Guo Z, Shan W et al, The roles of the epithelial-mesenchymal transition marker PRRX1 and miR-146b-5p in papillary thyroid carcinoma progression. Am J Pathol 2014; 184: 2342–2354.
Hardin H, Yu XM, Harrison AD et al, Generation of novel thyroid cancer stem-like cell clones: effects of resveratrol and valproic acid. Am J Pathol 2016; 186: 1662–1673.
Fender AW, Nutter JM, Fitzgerald TL et al, Notch-1 promotes stemness and epithelial to mesenchymal transition in colorectalcancer. J Cell Biochem 2015; 116: 2517–2527.
Tsai JH, Donaher JL, Murphy DA et al, Spatiotemporal regulation of epithelial-mesenchymal transition is essential for squamous cell carcinoma metastasis. Cancer Cell 2012; 22: 725–736.
Ocaña OH, Corcoles R, Fabra A et al, Metastatic colonization requires the repression of the epithelial-mesenchymal transition inducer Prrx1. Cancer Cell 2012; 22: 709–724.
Phay JE, Ringel MD . Metastatic mechanisms in follicular cell-derived thyroid cancer. Endocr Relat Cancer 2013; 20: R307–R319.
van Denderen BJ, Thompson EW . Cancer: the to and fro of tumour spread. Nature 2013; 493: 487–488.
Pannuti A, Foreman K, Rizzo P et al, Targeting Notch to target cancer stem cells. Clin Cancer Res 2010; 16: 3141–3152.
Park HS, Jung CK, Lee SH et al, Notch1 receptor as a marker of lymph node metastases in papillary thyroid cancer. Cancer Sci 2012; 103: 305–309.
Blaumueller CM, Qi H, Zagouras P et al, Intracellular cleavage of Notch leads to a heterodimeric receptor on the plasma membrane. Cell 1997; 90: 281–291.
Yamashita AS, Geraldo MV, Fuziwara CS et al, Notch pathway is activated by MAPK signaling and influences papillary thyroid cancer proliferation. Transl Oncol 2013; 6: 197–205.
Yu XM, Jaskula-Sztul R, Georgen MR et al, Notch1 signaling regulates the aggressiveness of differentiated thyroid cancer and inhibits SERPINE1 expression. Clin Cancer Res 2016; 22: 3582–3592.
Ferretti E, Tosi E, Po A et al, Notch signaling is involved in expression of thyrocyte differentiation markers and is down-regulated in thyroid tumors. J Clin Endocrinol Metab 2008; 93: 4080–4087.
Malanchi I, Santamaria-Martinez A, Susanto E et al, Interactions between cancer stem cells and their niche govern metastatic colonization. Nature 2012; 481: 85–89.
Park HS, Jung CK, Lee SH . Notch 1 receptor as a marker of lymph node metastasis in papillary thyroid cancer. Cancer Sci 2012; 103: 305–309.
Goss AM, Tian Y, Tsukiyama T et al, Wnt2/2b and beta-catenin signaling are necessary and sufficient to specify lung progenitors in the foregut. Dev Cell 2009; 17: 290–298.
Williamson AJ, Doscas ME, Ye J et al, The sonic hedgehog signaling pathway stimulates anaplastic thyroid cancer cell motility and invasiveness by activating Akt and c-Met. Oncotarget 2016; 7: 10472–10485.
Heiden KB, Williamson AJ, Doscas ME et al, The sonic hedgehog signaling pathway maintains the cancer stem cell self-renewal of anaplastic thyroid cancer by inducing Snail expression. J Clin Endocrinol Metab 2014; 99: E2178–E2187.
Nelson KK, Gattuso P, Xu X et al, Expression of the sonic hedgehog pathway molecules in synchronous follicular adenoma and papillary carcinoma of the thyroid gland in predicting malignancy. Surgery 2010; 148: 654–660.
Easwaran H, Tsai HC, Baylin SB . Cancer epigenetics: tumor heterogeneity, plasticity of stem-like cell states and drug resistance. Mol Cell 2014; 54: 716–727.
Sharma SV, Lee DY, Li B et al, A chromatin-mediated reversible drug tolerant state in cancer cell subpopulations. Cell 2010; 141: 69–80.
Roesch A, Fukunaga-Kalabis M, Schmidt EC et al, A temporarily distinct subpopulation of slow-cycling melanoma cells is required for continuous tumor growth. Cell 2010; 141: 583–594.
Lee E, Wang J, Yumoto K et al, DNMT1 regulates epithelial-mesenchymal transition and cancer stem cells which promotes prostate cancer metastasis. Neoplasia 2016; 18: 553–566.
Song L, Li Y . The role of stem cell DNA methylation in colorectal carcinogenesis. Sem Cell Rev Rep 2016; 12: 573–583.
Dong N, Shi L, Eang DC et al, role of epigenetics in lung cancer heterogeneity and clinical implication. Semin Cell Dev Biol 2016; pii: S1084-9521, 30263–30264.
Mendell JT, Olson EN . MicroRNAs in stress signaling and human disease. Cell 2012; 148: 1172–1187.
Gregory PA, Bracken CP, Bert AG et al, MicroRNAs as regulators of epithelial-mesenchymal transition. Cell cycle 2008; 7: 3112–3118.
Peng F, Xiong L, Tang H et al, Regulation of epithelial-mesenchymal transition through microRNAs: clinical and biological significance of microRNAs in breast cancer. Tumour Biol 2016; 37: 14463–14477.
de la Chappelle A, Jazdzewski K . MicroRNAs in thyroid cancer. J Clin Endocrinol Metab 2011; 96: 3326–3336.
Braun J, Hoang-Vu C, Dralle H et al, Downregulation of microRNAs directs the EMT and invasive potential of anaplastic thyroid carcinomas. Oncogene 2010; 29: 4237–4244.
Zhang Z, Liu ZB, Ren WM et al, The miR-200 family regulates the epithelial-mesenchymal transition induced by EGF/EGFR in anaplastic thyroid cancer cells. Int J Mol Med 2012; 30: 856–862.
Hebrant A, Floor S, Saiselet M et al, miRNA expression in anaplastic thyroid carcinomas. PLoS ONE 2014; 9: e103871.
Fuziwara CS, Kimura ET . MicroRNA deregulation in anaplastic thyroid cancer biology. Int J Endocrinol 2014; 2014: 743450.
Deng X, Wu B, Xiao K et al, MiR-146b-5p promotes metastasis and induces epithelial-mesenchymal transition in thyroid cancer by targeting ZNRF3. Cell Physiol Biochem 2015; 35: 71–82.
Wang Z, Zhang H, Zhang P et al, MicroRNA-663 suppresses cell invasion and migration by targeting transforming growth factor beta 1 in papillary thyroid carcinoma. Tumor Biol 2016; 37: 7633–7644.
Mattick JS, Rinn JL . Discovery and annotation of long noncodingRNAs. Nat Struct Mol Biol 2015; 22: 5–7.
Perry RB, Ulitsky I . The functions of long noncoding RNAs in development and stem cells. Development 2016; 143: 3882–3894.
Dhamija S, Diederichs S . From junk to master regulators of invasion: Inc RNA functions in migration, EMT and metastasis. Int J Cancer 2016; 139: 269–280.
Evans JR, Feng FY, Chinnaiyan AM . The bright side of dark matter: lncRNAs in cancer. J Clinical Invest 2016; 126: 2775–2782.
Zhang R, Hardin H, Chen J et al, Non-coding RNAs in thyroid cancer. Endocr Pathol 2016; 27: 12–20.
Yuan JH, Yang F, Wang F et al, A long noncoding RNA activated by TGF β promotes the invasion-metastasis cascade in hepatocellular carcinoma. Cancer Cell 2014; 25: 666–681.
Wang KC, Chang HY . Molecular mechanisms of long noncoding RNAs. Mol Cell 2011; 43: 904–914.
Jiao F, Hu H, Han T et al, Long noncoding RNA MALAT-1 enhances stem cell-like phenotypes in pancreatic cancer cells. Int J Mol Sci 2015; 16: 6677–6693.
Zhang R, Hardin H, Huang W et al, MALAT1 long non-coding RNA expression in thyroid tissues: analysis by in situ hybridization and real-time PCR. Endocr Pathol 2016; 28: 7–12.
Liu J, Peng WX, Mo YY et al, MALAT1-mediated tumorigenesis. Front Biosci 2017; 22: 66–80.
Jadaliha M, Zong X, Malakar P et al, Functional and prognostic significance of long non-coding RNA MALAT1 as a metastasis driver in ER negative lymph node negative breast cancer. Oncotarget 2016; 7: 40418–40436.
Li Q, Dai Y, Wang F et al, Differentially expressed long non-coding RNAs and the prognostic potential in colorectal cancer. Neoplasma 2016; 63: 977–983.
Mendell JT . Targeting a long noncoding RNA in breast cancer. N Engl J Med 2016; 374: 2287–2289.
Qi Y, Ooi HS, Wu J et al, MALAT1 long ncRNA promotes gastric cancer metastasis by suppressing PCDH10. Oncotarget 2016; 7: 12693–12703.
Oh EJ, Kim SH, Yang WI et al, Long non-coding RNA HOTAIR expression in diffuse large B-cell lymphoma: in relation to polycomb repressive complex pathway proteins and H3K27 trimethylation. J Pathol Transl Med 2016; 50: 369–376.
Padua Alves C, Fonseca AS, Muys BR et al, Brief report: the lincRNA Hotair is required for epithelial-to-mesenchymal transition and stemness maintenance of cancer cell lines. Stem Cells 2013; 31: 2827–2832.
Kim HJ, Lee DW, Yim GW et al, Long non-coding RNA HOTAIR is associated with human cervical cancer progression. Int J Oncol 2015; 46: 521–530.
Dou J, Ni Y, He X et al, Decreasing IncRNA HOTAIR expression inhibits human colorectal cancer stem cells. Am J Transl Res 2016; 8: 98–108.
Luczak A, Supernat A, Lapinska-Szumczyk S et al, HOTAIR in relation to epithelial-mesenchymal transition and cancer stem cells in molecular subtypes of endometrial cancer. Int J Biol Markers 2016; 31: e245–e251.
Loewer S, Cabili MN, Guttman M et al, Large intergenic non-coding RNA-RoR modulates reprogramming of human induced pluripotent stem cells. Nat Genet 2010; 42: 1113–1117.
Hou P, Zhao Y, Li Z et al, LincRNA-ROR induces epithelial-to-mesenchymal transition and contributes to breast cancer tumorigenesis and metastasis. Cell Death Dis 2014; 5: e1287.
Wang Y, Xu Z, Jiang J et al, Endogenous miRNA sponge lincRNA-RoR regulates Oct4, Nanog, and Sox2 in human embryonic stem cell self-renewal. Dev Cell 2013; 25: 69–80.
Yang P, Yang Y, An W et al, The long non-coding RNA-ROR promotes the resistance of radiotherapy for human colorectal cancer cells by targeting the P53/miR-145 pathway. J Gastroenterol Hepatol 2016; 32: 837–846.
Zhan HX, Wang Y, Li C et al, LincRNA-ROR promotes invasion, metastasis and tumor growth in pancreatic cancer through activating ZEB1 pathway. Cancer Lett 2016; 374: 261–271.
Chen YM, Liu Y, Wei HY et al, Linc-ROR induces epithelial-mesenchymal transition and contributes to drug resistance and invasion of breast cancer cells. Tumor Biol 2016; 37: 10861–10870.
Rezaei M, Emadi-Baygi M, Hoffmann MJ et al, Altered expression of LINC-ROR in cancer cell lines and tissues. Tumor Biol 2016; 37: 1763–1769.
Kaplan RN, Riba RD, Zacharoulis S et al, VEGFR1-positive hematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 2005; 438: 820–827.
Kosaka N . Decoding the secret of cancer by means of extracellular vesicles. J Clin Med 2016; 5: E22.
Yu Y, Xiao CH, Tan LD et al, Cancer-associated fibroblasts induce epithelial-mesenchymal transition of breast cancer cells through paracrine TGF-β signaling. Br J Cancer 2014; 110: 724–732.
Schliekelman MJ, Gibbons DL, Faca VM et al, Targets of the tumor suppressor miR-200 in regulation of the epithelial-mesenchymal transition in cancer. Cancer Res 2011; 71: 7670–7682.
Gibbons DL, Lin W, Creighton CJ et al, Contextual extracellular cues promote tumor cell EMT and metastasis by regulating miR-200 family expression. Genes Dev 2009; 23: 2140–2151.
Lee JC, Zhao JT, Gundara J et al, Papillary thyroid cancer-derived exosomes contain miRNA-146b and miRNA-222. J Surg Res 2015; 196: 39–48.
Peinado H, Aleckovic M, Lavotshkin S et al, Melanoma exosomes educate bone marrow progenitor cells toward a pro-metastatic phenotype through MET. Nat Med 2012; 18: 883–891.
Greening DW, Gopal SK, Mathias RA et al, Emerging roles of exosomes during epithelial-mesenchymal transition and cancer progression. Semin Cell Dev Biol 2015; 40: 60–71.
Zhang J, Liu SC, Luo XH et al, Exosomal long noncoding RNAs are differentially expressed in the cervicovaginal lavage samples of cervical cancer patients. J Clin Lab Anal 2016; 30: 1116–1121.
Melo SA, Luecke LB, Kahlert C et al, Glypican-1 identifies cancer exosomes and detects early pancreatic cancer. Nature 2015; 523: 177–182.
Vella LJ . The emerging role of exosomes in epithelial-mesenchymal-transition in cancer. Front Oncol 2014; 66–70.
Ma R, Minsky N, Morshed SA et al, Stemness in human thyroid cancers and derived cell lines: the role of asymmetrically dividing cancer stem cells resistant to chemotherapy. J Clin Endocrinol Metab 2014; 99: E4000–E4409.
Xu J, Hardin H, Zhang R et al, Stage-specific embryonic antigen-1(SSEA-1) expression in thyroid tissues. Endocr Pathol 2016; 27: 271–275.
Cheung SK, Chuang PK, Huang HW et al, Stage-specific embryonic antigen-3 (SSEA-3) and β3GalT5 are cancer specific and significant markers for breast cancer stem cells. Proc Natl Acad Sci USA 2016; 113: 960–965.
Acknowledgements
We thank the Translation Research Initiatives in Pathology Laboratory and the Experimental Pathology Laboratory for technical assistance. The work was supported by the Carbone Cancer Center support grant P30CA014520. Dr R Zhang received a grant from the Department of Pathology and Laboratory Medicine at the University of Wisconsin School of Medicine and Public Health.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors declare no conflict of interest.
Additional information
The cancer stem-like cell (CSC) hypothesis postulates that a small population of cancerous cells has self-renewal and clonal tumor-initiation properties. There is a relationship between epithelial to mesenchymal transition (EMT) and increased numbers of CSCs in some malignancies. Non-coding RNAs have been shown to play important roles during EMT and proliferation of CSCs. The tumor microenvironment of CSCs is also a factor in tumor progression. A better understanding of these mechanisms will lead to more effective therapies.
Rights and permissions
About this article
Cite this article
Hardin, H., Zhang, R., Helein, H. et al. The evolving concept of cancer stem-like cells in thyroid cancer and other solid tumors. Lab Invest 97, 1142–1151 (2017). https://doi.org/10.1038/labinvest.2017.41
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/labinvest.2017.41
This article is cited by
-
HOXD9/miR-451a/PSMB8 axis is implicated in the regulation of cell proliferation and metastasis via PI3K/AKT signaling pathway in human anaplastic thyroid carcinoma
Journal of Translational Medicine (2023)
-
Early stage evaluation of cancer stem cells using platinum nanoparticles/CD133+ enhanced nanobiocomposite
Cancer Nanotechnology (2023)
-
TRAF7-targeted HOXA5 acts as a tumor suppressor in prostate cancer progression and stemness via transcriptionally activating SPRY2 and regulating MEK/ERK signaling
Cell Death Discovery (2023)
-
RETRACTED ARTICLE: LncRNA XIST inhibits ovarian cancer cell growth and metastasis via regulating miR-150-5p/PDCD4 signaling pathway
Naunyn-Schmiedeberg's Archives of Pharmacology (2021)
-
Akt isoform-specific effects on thyroid cancer development and progression in a murine thyroid cancer model
Scientific Reports (2020)