|Home | About | Journals | Submit | Contact Us | Français|
Identification of the cell types capable of initiating and sustaining growth of the neoplastic clone in vivo is a fundamental problem in cancer research. It is likely that tumor growth can be sustained both by rare cancer stem-like cells and selected aggressive clones and that the nature of the mutations, the cell of origin, and its environment will contribute to tumor propagation. Genomic instability, suggested as a driving force in tumorigenesis, may be induced by genetic and epigenetic changes. The feature of self-renewal in stem cells is shared with tumor cells, and deviant function of the stem cell regulatory networks may, in complex ways, contribute to malignant transformation and the establishment of a cancer stem cell-like phenotype. Understanding the nature of the more quiescent cancer stem-like cells and their niches has the potential to develop novel cancer therapeutic protocols including pharmacological targeting of self-renewal pathways. Drugs that target cancer-related inflammation may have the potential to reeducate a tumor-promoting microenvironment. Because most epigenetic modifications may be reversible, DNA methylation and histone deacetylase inhibitors can be used to induce reexpression of genes that have been silenced epigenetically. Design of therapies that eliminate cancer stem-like cells without eliminating normal stem cells will be important. Further insight into the mechanisms by which pluripotency transcription factors (e.g., OCT4, SOX2, and Nanog), polycomb repressive complexes and microRNA balance selfrenewal and differentiation will be essential for our understanding of both embryonic differentiation and human carcinogenesis and for the development of new treatment strategies.
The classic clonal evolution theory where malignant transformation results from random mutations and subsequent clonal selection and the cancer stem cell hypothesis arguing that tumor growth is driven by a rare subpopulation of cells with stem cell-like properties are continuously discussed as models of carcinogenesis. Recent advances in stem cell technology and cancer biology have influenced progress in the understanding of tumor initiation and progression and elucidated that additional layers outside the DNA can be involved in the regulation of gene expression. However, this knowledge also underlines the heterogeneity and complexity in biology and response to treatment existing between and within different tumor types. This review focuses on the mechanisms and the cells involved in tumor initiation and propagation. It aims to mention aspects related to the debate on tumorigenic potential and models, to address the regulation of self-renewal and differentiation, and to point to valuable knowledge that can contribute to new therapeutic strategies.
The cellular origin of the cancer-initiating cell has been focused and debated for more than 100 years. Cancer, as a heterogeneous group of disorders with marked different biologic properties, is thought to arise from a single cell, where series of genetic alterations in tumor-suppressor genes and oncogenes are responsible for continued clonal selection and tumor cell heterogeneity, resulting in tumor proliferation, invasion, metastasis, and drug resistance .
The clonal genetic model of cancer has been successful in predicting mutations that seem to be necessary for the earliest stages of tumor growth. However, pathologic epigenetic changes (changes in gene expression that do not involve a change in DNA sequence) are increasingly considered as alternatives to mutations and chromosomal alterations in disrupting gene function . Genomic instability has been suggested as a driving force in tumorigenesis [3,4]. There are, however, indications that epigenetic changes such as global DNA hypomethylation and promoter-specific hypermethylation can be early events in the loss of cellular homeostasis and may precede genetic mutations and genomic instability in some instances . Genetic instability may be induced by deregulated epigenetic mechanisms and then result in the acquisition of genetic mutations that can induce silencing of tumor-suppressor genes and activation of oncogenes. Chromosomal instability generation, reactivation of transposable elements, and loss of imprinting are proposed mechanisms of how DNA hypomethylation can contribute to cancer development .
Several epigenetic mechanisms such as DNA methylation, histone modification, and microRNA expression may change the genome function through both endogenous and environmental factors [6–8]. MicroRNA (single-stranded noncoding RNA molecules of 20–22 nucleotides) are now known to have a central role in the regulation of gene expression and to be implicated in an increasing number of biologic processes (reviewed in Bartel , He and Hannon , and Zhang et al. ). They are important in developmental processes, including differentiation, proliferation, and apoptosis . Expression of microRNA is frequently decreased or increased in cancerous tissues, and many microRNA are located at or close to genomic sites that are commonly deleted or amplified in various cancers . There are also reports indicating that altered expressions of specific microRNA genes contribute to the initiation and progression of cancer [14,15]. Profiles of microRNA expression are shown to differ between normal tissues and tumor tissues and among tumor types [14–19].
Infection, inflammation, and tissue repair processes are increasingly recognized as important factors that may contribute to tumor development [20–24]. Interestingly, cancer-related inflammation is suggested to contribute to genetic instability in tumor cells. Inflammatory mediators may by a variety of mechanisms influence DNA repair pathways and thereby induce microsatellite instability. Such mediators may also induce double-strand breaks, defective mitotic checkpoint, and deregulated homologous recombination leading to chromosomal instability . An uncontrolled “wound-healing process” during tissue repair might possibly attract and/or initiate mesenchymal stem cells or other cells resulting in stimulated or deregulated self-renewal processes. In this context, cell fusion and/or horizontal gene transfer events may play a role [26,27].
The multistep tumorigenesis theory focuses on the nature and number of mutations rather than on the properties of the cell in which they occur. Identification of the cell types capable of initiating and sustaining growth of the neoplastic clone in vivo is a fundamental problem in cancer research.
A tumor can be viewed as an aberrant organ initiated by a tumorigenic cancer cell that has acquired the capacity for indefinite proliferation through accumulated mutations. The balance between cell proliferation, differentiation, and apoptosis has influence on the total amount of cells in the body. An aberrant regulation of these processes can induce cancer development. During development, cellular differentiation of the unspecialized pluripotent embryonic stem cells in the inner mass of early embryos occurs. All tissues in the body are derived from organ-specific stem cells that are defined by their capacity to undergo self-renewal as well as to differentiate into the cell types that comprise each organ. These tissue-specific stem cells are distinguished from embryonic stem cells in that their differentiation is largely restricted to cell types within a particular organ. There are also adult stem cells that are involved in the regeneration of tissues during the lifetime of the individual. Some stem cells are continuously active to replace cells as they mature and die off as, for instance, skin cells. Other stem cells remain dormant until a physiological signal is received; for example, some breast stem cells respond strongly to pregnancy hormones and, to a lesser extent, to hormones within the monthly cycle. It has become more and more evident that normal development and cancer share many properties. The feature of self-renewal in stem cells is shared with tumor cells, and this common property has led to the proposal that self-renewal provides increased opportunities for carcinogenic changes to occur. It has also been suggested that altered regulation of self-renewal directly underlies carcinogenesis . Stem cells live longer than many differentiated cells and may therefore be exposed to damaging agents for a longer time, allowing accumulation of mutations that can result in transformation [29,30].
That cancer might arise from a rare population of cells with stem cell properties was proposed approximately 150 years ago (references in Wicha et al. ). It has been also more than 40 years since tissue-specific stem cells were postulated to be the cell of origin in cancer . The concept of a less differentiated, more pluripotent cell forming cancer was taken a step further when it was proposed that cancers represented a maturation arrest of stem cells . Development of fluorescence-activated cell sorting, recent advances in stem cell biology, and the development of new xenotransplantation animal models have improved the possibilities to validate the cancer stem cell hypothesis, stating that only a minority of cells within a tumor are responsible for tumor initiation and maintenance. During the last years, it has been possible to isolate and characterize these cells from hematological malignancies [34,35] based on the use of markers that are thought to distinguish tumorigenic from nontumorigenic cells. A number of recent articles have reported that cancer stem cells can also be isolated from solid human cancers such as breast, brain, bone, lung, melanomas, prostate, colon, pancreas, liver, and head and neck [36–48].
The term cancer stem cell is still an operational term defined as a cancer cell that has the ability to self-renew to give rise to another malignant stem cell as well as a cell that will give rise to the phenotypic diverse cancer cells that are known as transient amplifying cells [31,49–51]. The transient amplifying cells are believed to be the cells that are responsible for the bulk tumor cell proliferation that is responsive to conventional therapy, whereas the cancer stem-like cells can be left intact and will eventually repopulate the tumor [51,52] (Figure 1).
Traditionally, many cancer cells have been considered to have tumorigenic potential, but the cancer stem cell model has suggested that only small subpopulations of cancer cells have tumorigenic potential based on experiments in which human cancer cells were transplanted into nonobese diabetic/severe combined immunodeficiency (NOD/SCID) mice [34,35]. Although xenotransplantation animal models have improved the possibility to study the cancer stem cell hypothesis, the critical role of tumor growth interactions with the local and extended microenvironment complicates the interpretation of such studies. An alternative explanation of the low frequency of tumor-producing cells from, for example, human acute myeloid leukemia (AML) in NOD/SCID mice may be that these rare tumor cells are the cells that can adapt to growth in a foreign environment . Xenotransplantation may therefore fail to reveal several growthsustaining cells because the foreign microenvironment hinder essential interactions with human support cell such as fibroblasts, endothelial cells, macrophages, mast cells, and mesenchymal stem cells . The nontransplantable human AML cell populations may lack critical features for obtaining stromal support, for instance, through cytokine and chemokine factors and receptors in the foreign microenvironment, whereas the transplantable population may have acquired features that enable those cells to survive in a foreign milieu. There are also concerns whether the lack of an appropriate microenvironment of xenografts in the immunocompromised mouse models underestimates the frequency of tumor-initiating and -sustaining cells. Such frequencies may best be tested by transferring mouse tumor cells into nonirradiated histocompatible recipient mice . This is underlined by the recent report by Kelly et al. where they show that a large proportion of murine lymphoid and myeloid tumors in syngeneic mouse tumor models were able to initiate tumor formation.
The role of the microenvironment in the development of solid tumors are thought to be even more complex than that of hematopoietic cancers because solid tumors are challenged by tissue barriers and the need for angiogenesis (for review, see Visvader and Lindeman ). The differentiation pathways in such tumors are also less well understood. However, Quintana et al.  have shown that, by modifying xenotransplantation assay conditions, including the use of more highly immunocompromised mice, they can increase the detection of tumorigenic melanoma cells by several orders of magnitude. Such studies indicate that cells with tumorigenic potential are likely to be much more frequent in many human cancers than estimates based on pure xenotransplantations. These studies also indicate that the immune system may play a role in the selective growth of tumor-initiating cells.
Whether cells with tumorigenic potential are common or rare within human cancers is still a fundamental question in cancer biology. The degree to which the cancer stem cell and clonal evolution models are mutually exclusive and whether a single cancer stem cell hierarchy exists in patients with cancer also remain open questions . Although the cancer-initiating cells are still not fully identified or characterized, they seem to display different phenotypes, which can possibly be derived from both embryonic and adult stem cell pools, as well as from progenitor or differentiated cells . Evidence for tumor initiation from embryonic stem cells is revealed from studies showing that transplantation of undifferentiated human embryonic stem cells into immunodeficient mice form teratomas, large tumor-like formations containing tissues belonging to all three germ layers . Teratomas can also appear spontaneously in mice and humans and are defined as teratocarcinomas when they contain a core of malignant undifferentiated cells . Evidence from leukemia suggests that cancers can be initiated from multiple cell types in developmental hierarchies .
Tumor growth can most likely be sustained either by rare cancer stem-like cells or by selected aggressive clones or both, and the nature of the mutations, the cell of origin, and its environment can also contribute to tumor propagation. Tumors may be initiated in a way that follows the stem cell paradigm and further progress to acquire additional mutations that resemble the clonal evolution model. Such tumors could then exhibit features of both models [54,55]. Stem-like cells in a hierarchical organized tumor are not necessarily static, and multiple clones of stem-like cells may evolve in response to treatment and changes in the microenvironment. There are no clear-cut answers, and events from several models are most likely involved during cancer development and propagation. Proving or disproving a single specific hypothesis or model will probably not provide the best basis for new generation of successful therapies.
Distinction between the cells that give rise to the first form of the tumor, which may be stem cells, progenitor cells, or differentiated cells depending on the tumor phenotype, and cells that propagate the tumor and exhibit self-renewal and differentiation capacity would be of great value . There are, however, several challenges for establishing such a distinction. For example, recent studies have shown that CD133, previously thought to be a robust brain tumor stem cell marker [39,62], does not consistently distinguish tumorigenic from nontumorigenic cells [63–66]. The CD133 has also been used as a phenotypic marker of colon cancer stem cells [44,45], but also in this cancer type, both CD133-negative and -positive populations have been reported to induce tumor growth in vivo . The markers used to isolate stem cells in normal and cancerous tissues are not expressed exclusively by stem cells. Neither can markers used for identification of stem cells in one organ directly be used for identification in other organs. Such conditions underline the importance of linking phenotypic markers with functional assays. The activity of aldehyde dehydrogenase isoform 1 that can be assessed by the ALDEFLUOR assay has been identified as a common functional marker of normal and malignant human breast stem cells . Aldehyde dehydrogenase isoform is also expressed in hematopoietic, neuronal, and colonic stem cells, and aldehyde dehydrogenase isoform 1 may have a promising potential as a universal functional marker for identification and isolation of stem cells from multiple sources [69,70]. CD44 and CD133 have been reported as cell surface cancer stem cell markers in several solid tumor types . Interestingly, it has recently been reported that CD44, in contrast to CD133, seems to be a robust marker of functional importance for colorectal cancer initiation .
Cell plasticity properties such as epithelial-to-mesenchymal transitions (EMT) also threat the identification of tumor-initiating cells. During development, multicellular organisms are dependent on the EMT of polarized epithelial cells to motile cells. Interactions between epithelial cells and mesenchymal cells with a migratory morphology are essential in organ formation, and such transitions can also be activated during cancer invasion and metastasis (for review, Thiery et al. ). Recently, EMT has been reported to generate cells with stem cell-like properties [74,75].
Disruption of genes and pathways involved in the regulation of stem cell self-renewal seems to be important in the stem cell model of cancer [27-29,76-79]. Whether stem cells or cancer stem-like cells continue to self-renew or start to undergo differentiation is thought to be influenced by the microenvironment [55,80,81]. It has been suggested that cancer can be initiated by a loss of polarity in stem cells that may lead to impairment of asymmetric cell division, rendering the daughter cells unable to respond to normal homeostasis mechanisms that regulate cell proliferation . Interestingly, recent data demonstrate that the tumor-suppressor p53 regulates polarity of self-renewing divisions in mammary stem cells and suggest that loss of p53 favors symmetric divisions of cancer stem-like cells contributing to tumor growth in a transgenic model of breast cancer .
An emerging role of microRNA in the regulation of stem cell self-renewal and differentiation [12,84] has been revealed, indicating that they are crucial for proper stem cell function and maintenance. During development, microRNA expression seems to be tissue specific, and this may implicate that microRNA are essential to establish and maintain cell type and tissue identity [85,86]. Coordinated transcription factor networks involving OCT4, SOX2, and Nanog have currently emerged as the master regulatory mechanisms of stem cell pluripotency and differentiation , and microRNA to OCT4, SOX2, and Nanog coding regions are found to modulate embryonic stem cell differentiation . Interestingly, miR-302 that can target OCT4/SOX2/NANOG has been shown to have a role in converting differentiated cells to induced pluripotent stem cells [89,90]. MiRNA loci and promoter regions related to pluripotency and self-renewal regulating transcription factors are shown in Table 1. Knowledge of predicted microRNA that target different regulators of pluripotency can be of value in the design of future experimental studies (knockout/knockdown/overexpression) to gain insight and target master regulators of pluripotency (Table 2). Available experimental data related to microRNA and pluripotency genes (Table 3) reveal biologic knowledge and basis for interfering with pluripotency, self-renewal, and differentiation by microRNA.
It has been hypothesized  that several microRNA associated with glioblastomas may have a normal function in regulating neural stem cell self-renewal and differentiation during development. However, their dysfunction in cancer may contribute to the maintenance of an undifferentiated proliferative phenotype by preventing the expression of differentiation targets and allowing the expression of targets promoting stem cell renewal [18,19].
There are still challenges with respect to identification and control of tumor initiating and propagating cells. Increased mechanistic knowledge of the cellular processes that establish and/or maintain such cells is needed. Embryonic stem cells are known to rely on polycomb group proteins to reversibly repress genes required for differentiation . Interestingly, it has been hypothesized that acquisition of promoter DNA methylation at these repressed genes could permanently silence them and thereby lock the cell into a perpetual state of self-renewal [101–103] predisposing to subsequent malignant transformation and cancer development. Epigenetic repressive pathways involving polycomb complexes, DNA methylation, and microRNA seem to cooperate to reduce transcriptional noise and to prevent aberrant induction of differentiation. Deviant function of the stem cell regulatory network may therefore, in complex ways, contribute to malignant transformation and the establishment of a cancer stem cell-like phenotype .
A key challenge in cancer biology is to target the most aggressive tumor cells within a cancer. Opinions of how many and which tumor cells must be eliminated for cancer treatment to be successful are divided. Depending on the circumstances, there are now reports supporting both the perspectives that every tumor cell can propagate human cancer and that this property is exclusive to a subpopulation of particular cells . Effective therapy will probably require targeting of all the tumor cell populations by combinational approaches , and a deeper understanding of tumor-initiating and -propagating mechanisms is therefore required. For safety, the goal for any therapy should be the elimination of all malignant cells because of their potential to expand to form tumors and to disseminate . Increasing evidence suggests that nonmalignant host cell populations play a significant role in tumor growth, indicating requirements for a better understanding of how to target effects on tumor initiation and propagation also from these cells . Drugs that, for instance, target cancer-related inflammation may have the potential to reeducate a tumor-promoting microenvironment to become a tumor-inhibiting environment .
Cytokine networks play important roles in tumorigenesis, and there is also evidence that cytokines may regulate stem cell behavior. Colon carcinoma stem-like cells identified by expression of the cell surface marker CD133 are found to produce and use interleukin 4 (IL-4) in apoptosis protection . Treatment with inhibitors of IL-4 revealed enhanced antitumor chemotherapy efficacy through selective sensitization of the CD133-positive cells. IL-6 can induce malignant properties in breast cancer stem-like cells by triggering the Notch pathway , and the up-regulation of IL-6 is indicated as a possible inflammatory or hypoxia stress response. Charafe-Jauffret et al.  used the ALDEFLUOR assay to isolate cancer stem cell populations from normal and malignant mammary tissues. Increased expression of CXR1/IL-8RA was revealed by oligonucleotide microarray in the ALDEFLUOR-positive population. Furthermore, recombinant IL-8 was found to increase the expression of ALDEFLUOR-positive cells and the tumor sphere-forming capacity in breast cancer cell lines. Interestingly, it has also recently been reported that breast cancer stem-like cells can be targeted and selectively decreased through blockage of the IL-8 receptor CXCR1 in vitro and in NOD/SCID xenograft models .
Targeting proteins involved inEMTmay also be an option to include in the therapeutic strategy to eliminate detrimental cells [73,111] because chemoresistance in tumors undergoing EMT has been indicated  and depletion of the EMT-inducing transcription factor Twist1 improve chemotherapy efficacy in breast cancer cells .
The concept of cancer stem-like cells has significant clinical implications as cancer stem-like cells are thought to be more resistant to chemotherapy and radiotherapy. Increased drug efflux, metabolic alterations, cell cycle kinetics, and enhanced DNA repair as well as deregulated differentiation are suggested to be involved [114–119]. Understanding the nature of the more quiescent cancer stem-like cells and their niches has the potential for the development of novel cancer therapeutic protocols . The elucidation of self-renewal pathways has already identified some novel pharmacological targets, and novel compounds are being developed with preclinical activity and hope for clinical utility [121–124]. Recently, an approach to identify selective inhibitors of cancer stem cells by high-throughput screening has been described . Furthermore, low proteasome activity has been suggested as a general feature of cancer-initiating cells, which can be exploited to identify and target such cells both in vitro and in vivo . Cancer stem cell-directed therapeutic approaches to target stem-like cells in different organ-specific tumors has been suggested as relevant strategies to improve clinical cancer therapy (overview in Schatton et al. , Frank et al. , and Winquist et al. ). A recent study  shows the importance of specifically targeting the tumorigenic population of breast cancer cells. Conventional chemotherapy in a neoadjuvant setting to both HER2-positive and -negative human tumors was found to increase the proportion of a highly tumorigenic subpopulation (CD44+/CD24-) of breast cancer cells. In contrast, after the more specific targeted treatment of HER2-positive tumors with the epidermal growth factor receptor/HER2 inhibitor lapatinib, no increase in the tumorigenic subpopulation in the biopsy samples was found. The study underlines that targeting both tumorigenic and dividing daughter cells is essential in preventing cancer relapse (Figure 1).
CD133-positive glioma cells with tumor-initiating properties have been reported to promote radioresistance by preferential activation of DNA damage response . This resistance was reversed with specific inhibitors of Chk1 and Chk2 checkpoint kinases. Furthermore, Notch signaling that is involved in self-renewal regulation and stem cell fate determination was also found to be important in the regulation of radioresistance of such cells . The Notch promoted radioresistance was indicated to be regulated through activation of different cellular survival mechanisms as the phosphatidylinositol-3-OH kinase/Akt pathway and up-regulation of prosurvival proteins and also through p53 interaction. Tumor-initiating cells from p53 null mouse mammary tumors have also recently been shown to repair radiation-induced DNA damage more efficiently than the tumor bulk cells . Down-regulation of PTEN (phosphatase and tensin homolog deleted on chromosome 10) increased Akt signaling, and activation of the canonical Wnt/β-catenin signaling pathway was observed in these cells. Furthermore, by pharmacological inhibition of the Akt pathway, Wnt signaling andDNAdamage repair were inhibited, and the tumorinitiating cells were sensitized to radiation.
Hypoxia is also known to play a role in tumor radioresistance, and local tumor control after radiotherapy is found inversely correlated with tumor hypoxia. In that connection, it has been suggested that hypoxia might affect stem cell generation and maintenance in tumors through activation of hypoxia-inducible factors (HIFs) . Interestingly, activation of HIF-2α may induce the expression of OCT4, and HIF-1α can interact with Notch signaling and thereby contribute to the maintenance of an undifferentiated cell state [133–135].
Induction of differentiation of cancer stem-like cells by different methods is suggested as a potential treatment strategy. For instance has bone morphogenic proteins been reported to inhibit tumorigenic potential of brain tumor-initiating cells by triggering Smad signaling, reducing proliferation, and promoting differentiation . Epigenetic therapy is suggested as a potential strategy to induce differentiation of cancer stem-like cells [137–139]. In contrast to geneticmutations, most epigenetic modifications may be reversible and preventable, and this knowledge expands the therapeutic possibilities. The findings that epigenetic mechanisms such as DNA methylation and histone modifications can affect gene expression enable the potential to interfere with gene expression by the use of pharmacologicals as DNA methylation and histone deacetylase inhibitors to induce reexpression of genes that have been silenced epigenetically. DNA methylation and histone deacetylase inhibitors have been used both in preclinical and in clinical studies [140,141], and combined treatment strategies are, in some cancers, shown to be more effective than individual treatment approaches [142–144].
Given that microRNA regulate gene expression and easily enter the cell, they may also have a therapeutic potential against cancer . The recent demonstration that human breast cancer stem-like cells compared with nontumorigenic cancer cells differentially express a microRNA cluster that specifically targets BMI1, a critical promoter of stem cell self-renewal belonging to the polycomb group complex, indicates the potential to target self-renewal and differentiation regulation in cancer stem-like cells by microRNA . The fact that many microRNA are expressed at low levels in tumors but at high levels in normal tissues indicates possibilities for differential targeting and that they can be tolerated in normal cells .
The similarity of cancer stem-like cells to normal stem cells indicates that they inherit or acquire stem cell properties. An important question is whether it will be possible to identify therapies that eliminate cancer stem-like cells without eliminating normal stem cells in the same tissues to avoid intolerable adverse effects. Recent reports suggest that cancer stem-like cells can be selectively targeted without altering normal stem cell function. PTEN is a phosphatase that attenuates proliferation and survival signals through the negative regulation of the phosphatidylinositol-3-OH kinase pathway. It is a frequently mutated gene in human cancers and can be inactivated by different mechanisms [147,148]. Alterations in the PTEN pathway are associated with leukemogenesis, and it has been shown that PTEN plays a critical role in controlling hematopoietic stem cell proliferation and differentiation . Yilmaz et al.  have shown that PTEN deletion causes generation of transplantable leukemia-initiating cells but also depletion of normal hematopoietic stem cells; however, there seems to be a mechanistic difference between the maintenance of normal stem cells and leukemia-initiating cells. These studies indicate that compounds that promote stem cell quiescence might have different effects on normal stem cells and cancer stem-like cells and that it can be possible to identify and target pathways that have distinct effects on normal stem cells and cancer stem-like cells within the same tissue. Another example of potentially targeting cancer stem-like cells specifically is provided by recent studies showing that embryonic carcinoma cells exhibit a higher sensitivity to the histone deacetylase inhibitor apicidin compared with embryonic stem cells in down-regulation of Nanog, which is likely to act as a master transcription regulator controlling pluripotency .
The acquisition and maintenance of cell fate or “identity” regulated by a strict coordination between genetic and epigenetic programs are essential for growth and development. Cancer cells possess traits reminiscent of those ascribed to normal stem cells. It has been shown that histologically poorly differentiated tumors show preferential overexpression of genes normally enriched in embryonic stem cells combined with repression of polycomb-regulated genes and overexpression of activation targets of Nanog, OCT4, SOX2, and c-Myc . In this way, poorly differentiated aggressive tumors show an embryonic stem cell-like gene expression signature. An epigenetic stem cell signature showing DNA promoter hypermethylation of polycomb group targets has also been found in human colorectal tumors as well as in ovarian and breast cancers . These findings reveal that understanding the mechanisms by which pluripotency transcription factors, polycomb repressive complexes, and microRNA balance between selfrenewal and cellular proliferation and differentiation is essential for our understanding of both embryonic differentiation and human carcinogenesis [87,92,153] (Figure 2).
Assembly of available and future knowledge should pave the way for increased understanding and development of multifaceted approaches of ex vivo and in vivo models, leading to further characterization of the cancer-initiating and -propagating cells and their niches. Such knowledge should also lead to possibilities to decipher chemoresistance and treatment sensitivity mechanisms and to development of more efficient therapy.
Comments from Rolf Bjerkvig, Karl-Henning Kalland, Janice Nigro, and Kai-Ove Skaftnesmo are highly appreciated.
1This work was supported by the Norwegian Cancer Society, the Norwegian Research Council, Helse Vest, Haukeland University Hospital, The Bergen Translational Research Program, CRP-Santé Luxembourg, and the European Commission 6th framework program, contract no. 504743.