Cancer Stem Cells
A Scientific Review

Introduction: The new “Cancer Stem Cell Hypothesis”

For many decades, scientists pushed forward the notion that cancer initiates from multiple genetic mutations that may occur in any type of cell in the body. The old cancer model or the clonal genetic model of cancer, as it is called nowadays, defined cancer as a proliferative disease originating from mutated tumor cells that contribute equally to the tumorigenic activity within the tumor. This scheme has been the center of cancer research for many years and all the existing therapeutic strategies were designed and implemented solely based on this premise. This model, however, fails to address a fundamental difference that exists between tumor cells within a tissue: the various levels of differentiation and self-renewal capacity that exist between tumor cells. Recognition of this fact has put forth a new defining model for carcinogenesis, the “cancer stem cell hypothesis” (28).

Based on this model, cancer is a stem cell disease that places malignant stem cells at the center of its tumorigenic activity. Stem cells, at the top of their hierarchy, have the capacity to undergo self-renewal and have the potential to differentiate into different cell types in a specific lineage. The cancer stem cell hypothesis is gaining widespread acceptance by the mounting evidence of recent research studies that have revealed the likely mechanism of cancer propagation by transformed stem cells that constitute a small fraction of the whole tumor mass. In vivo evidence for the presence of cancer stem cells (CSCs) was first demonstrated in leukemia by Bonnet and Dick during the 1990s. They showed that only a small subpopulation of acute myelogenous leukemia (AML) cells had the ability to repopulate leukemic tumor upon transplantation into non-obese diabetic/severe combined immunodeficient (NOD/SCID) mice, resulting in a similar disease phenotype compared to the donor (2). Recently, it has been shown that these CSCs have the same cell markers, CD34+ CD38-, as the primitive hematopoietic stem cells (HSCs) (3). The role of stem cells in carcinogenesis has also been proven conclusively in breast cancer, brain cancer, and several other solid tumors during the past few years (1, 26).

Besides the demonstration of tumorigenesis in SCID mice, few other important observations and theories support the cancer stem cell hypothesis. First, tumor is a heterogeneous “organ”, consisting of stem cells, rapidly proliferating progenitor cells (also called transit-amplifying cells), and various differentiated cells. The fact that tumor shrinkage by surgery or chemotherapy does not translate to cancer cure signifies the important function of the small subpopulation of cancer stem cells within the tumor. Current cytotoxic chemotherapeutic agents only target the rapidly proliferating cancer cells and neglect the existence of the true tumorigenic cancer stem cells that have slower proliferation rates (17).

Second, stem cells have much longer life spans than their progeny cells and therefore have a higher chance of undergoing genetic mutations (22). Moreover, it is likely that only one or two mutations are needed for stem cells to initiate tumorigenesis, as opposed to the hypothesis that tumor formation requires at least six mutations in any somatic cell, making it a rare event (11). Subsequently, the issue of genetic alterations in committed cells versus stem cells needs to be addressed. Based on the new cancer stem cell model, carcinogenesis could arise from deregulated normal stem cells or through dedifferentiation or transformation of mutated progenitor cells to tumorigenic cells that have similar self-renewing properties as the stem cells (15, 19) (Figure 1 and 2).

Stem Cells Gone Wild: Cancer Stem Cells vs. Normal Stem Cells
The similarity between cancer stem cells and normal stem cells is perhaps the first thing that is noticed in exploring the relationship between the two. Indeed, CSCs share similar properties with the normal stem cells. Self-renewal capacity, differentiating potential, activation of anti-apoptotic pathways and telomerase expression, increased activity of membrane transporters, and the ability to migrate and metastasize are examples of such similarity (28). Nevertheless, such comparison is considered a superficial assessment since CSCs could differ from normal stem cells in terms of their regulatory pathways, their microenvironments, and their genetic selections.

One of the likely key differences between normal stem cells and CSCs could lie in the two different mechanisms of stem cell division. A hallmark of many stem cells is the default asymmetric division that leads to one exact copy of the self-renewing stem cell and one committed cell that could differentiate into different cell types in a lineage. However, during the developmental period and wound healing, a second mechanism of division is utilized by stem cells that result in either two self-renewing stem cells or two committed cells. During development, the symmetric division expands the stem cells and the number of committed cells necessary to produce an organism. During wound healing and injury, the symmetric division replenishes stem cells and differentiated cells in the injured tissue and contributes greatly to the healing process. It is widely accepted that the regulation of the switch between these two types of divisions maintains stem cell homeostasis, keeping a balance between self-renewal and differentiation (13).

The distinctive mechanism of symmetric division that causes increased growth and regenerative potential could indicate that it is the preferred mode of division in cancer stem cells. This hypothesis regards the symmetric division of stem cells as a possible prerequisite for tumor transformation in later stages (13). Present evidence points to some gene products that can both stimulate the symmetric division and also act as oncogenes in mammals. An atypical protein kinase named aPKC is one example of an oncogene that has been identified in human lung cancers (20, 21). It should be noted, however, that the current evidence for this hypothesis is not conclusive and the exact preferred mode of cancer stem cell division remains unknown.

Multiple signaling pathways such as Wnt/β-catenin, BMPs, Notch, PTEN, and Sonic hedgehog, and gene products of the Bmi1 have been proved to be involved in the normal stem cell self-renewal (4). It has been shown in some cases and hypothesized in others that the dysregulation of such pathways could result in the proliferation of cancer stem cells in humans. However, dysregulation does not necessarily mean up-regulation or down-regulation of a certain product. It could also indicate selection and activation of a certain pathway through direct mutation. For instance, many tumors have inactivating mutations in tumor-suppressive genes such as p16INK4A and p14ARF, which are down-regulated by Bmi1 (25). The Bmi1, however, is also expressed in normal stem cells to induce self-renewal. Another example is the genes that encode for APC and β-catenin. These genes are highly expressed in normal stem cells, but they are mutated in colon cancer (23).

It is evident that the molecular mechanisms of self-renewal signaling pathways differ in normal stem cells versus cancer stem cells. Research has also revealed that CSCs might undergo different molecular mechanisms in different tissues. Human cancers of colon, prostate, and ovary involve the dysregulation of Wnt signaling, while over-expression of Notch signaling is implicated in breast cancer (4, 16, and 27).

In essence, the stem cell signaling pathways define and signify the role of the niche, the microenvironment in which the stem cells home to and mature. Tissue regeneration or self-renewal depends on the regulation of proliferation-promoting signals (such as Wnt signaling) versus proliferation-inhibiting signals (such as BMP antigrowth signal) that stem cells receive in the niche. Thus, cancer stem cells may rise from dominant growth-promoting signals in the altered niche, or from intrinsic mutations that make them self-sufficient and independent of the microenvironment (11).

In relation to the niche, invasion or metastasis of cancer stem cells exhibit similar properties seen in normal stem cells during homing or mobilization. SDF1 and the receptor CXCR4 are both essential players in this regard (10). Carcinoma-related fibroblasts secrete SDF1, which interacts with tumor cell receptor CXCR4 to induce tumor growth and angiogenesis (14). Recent studies have also found that SDF1 is involved in metastasis of breast cancer cells (8).

Work in Progress: Identification and characterization of different Cancer Stem Cells
The quest to identify and characterize cancer stem cells in different tissues began immediately after the first discovered evidence of their existence in leukemia. Since then, cancer stem cells have been identified in many cancers, especially in well-established solid tumors of the brain, breast, skin, and prostate. The difference is that scientists know more about normal stem cells than cancer stem cells through years of investigation and research. As a result, more stem cell markers have been recognized in normal stem cells than CSCs.

Identification of cancer stem cells is based on their ability to generate xenograft tumors upon transplantation into NOD/SCID mice. As a basis for the cancer stem cell hypothesis, only a small subpopulation fraction of tumor cell lines exhibit such potential. Thus, the next step would be to characterize this subpopulation fraction by flow cytometry. After finding cancer stem cell markers, Side Population (SP) analysis may be used to identify CSCs further. SP analysis is founded on Hoechst exclusion in flow cytometry, determined largely by expression of ABCG2 transporter [ATP-binding cassette (ABC) subfamily G member 2] and the efflux of the fluorescent dye Hoechst 33342 by normal and cancer stem cells (4, 7).

As described earlier, leukemia cancer stem cells have the same CD34+ CD38- phenotype as the hematopoietic stem cells (3). Other examples include brain cancer stem cells (including neuroblastoma) having CD133+ phenotype (26), human breast cancers with CD44+ CD24low lin- and Oct-4 cell markers (1, 18), and human prostate cancer stem cells characterized by CD44+/α2β1hi/CD133+ (12). It is imperative that further studies identify and characterize cancer stem cells more thoroughly and discover unique markers and other distinctive properties that could distinguish CSCs from normal stem cells and from non-tumorigenic masses (17).

Cancer Stem Cells as Targets: From theory to therapy
A convincing hypothesis regarding the failure of current cancer therapy arises from the fact that tumor regression by cytotoxic therapeutic agents does not eliminate cancer in most cases. According to this hypothesis, current chemotherapy removes highly proliferating non-tumorigenic cells in the tumor mass, and leaves the small fraction of slowly-dividing tumorigenic cancer stem cells intact. This leads to re-emergence of a newly formed tumor mass in future (5) (Figure 3).

For this reason, scientists are questioning whether they are targeting the right type of cells during cancer therapy. It is becoming widely accepted that one potential approach to eradicating cancer is to target cancer stem cells that show resistance to apoptosis induced by cytotoxic agents. For instance, many cancer stem cells are in G0 phase and are resistant to cell-cycle specific chemotherapy drugs (28). Perhaps, by designing better targeting therapies, we can reduce the risk of cancer by reducing the number of cancer stem cells in a cancerous tissue, either by inducing apoptosis or by stimulating differentiation of CSCs with loss of self-renewal potential.

The results of recent research studies propose multiple possible approaches for targeting cancer stem cells. Different signaling pathways, different genes, or even different molecular interactions can be the target of the new therapeutic remedies. Agents targeting signaling pathways such as Bmi1 or Sonic Hedgehog have been shown to have anti-neoplastic activity. For instance, two Hedgehog pathway inhibitors, cyclopamine and HhAntag, are both effective candidate drugs in this regard (9, 24).

Despite of all the extensive cancer research done during the past few decades, the scientific community must adapt itself to a new paradigm shift in thinking and exploring. Role of stem cells in carcinogenesis could lead us one step closer to finding the way to cancer cure. A better understanding of growth-inhibiting signals, growth-promoting signals, and other molecular factors in the niche, along with their relevance in determination of the level of cancer stem cells’ dependence to the microenvironment in various cancers is very much needed (11). Moreover, complete understanding of the regulatory switch between symmetric and asymmetric modes of division could provide new clues as to how normal stem cells are transformed into CSCs. Similarly, examination of genetic and epigenetic changes in cancer stem cells and better understanding of tumor-progenitor genes could bring improved therapeutic outcomes (6). In conclusion, Targeting cancer stem cells with enhanced specificity and decreased toxicity and resistance could shape the future of developing cancer therapies.

References:

1. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF (2003) Prospective identification of tumiregenic breast cancer cells. Proc Natl Acad Sci U S A 100:3983-8
2. Bonnet D, Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nature Med. 3:730-7
3. Bonet D (2005) Normal and leukemic stem cells. Br J Haematol. 130:469-79
4. Burkert J, Wright NA, Alison MR (2006) Stem cells and cancer: an intimate relationship. J Pathol 209:287-97
5. Dalerba P, Cho RW, Clarke MF (2007) Cancer stem cells: models and concepts. Annu. Rev. Med. 58:18.1-18.8
6. Feinberg AP, Ohlsson R, Henikoff S (2006) The epigenetic progenitor origin of human cancer. Nature Rev 7:21-33
7. Goodell MA (2002) Multipotential stem cells and ‘side population’ cells. Cytotherapy 4:507-8.
8. Kang H, Watkins G, Douglas-Jones A, Mansel RE, Jiang WG (2005) The elevated level of CXCR4 is correlated with nodal metastasis of human breast cancer. Breast. 14:360-7
9. Karhadkar SS, Bova GS, Abdallah N, et al. (2004) Hedgehog signaling in prostate regeneration, neoplasia and metastasis. Nature 431:707-12
10. Lapidot T, Kollet O (2002) The essential roles of the chemokine SDF-1 and its receptor CXCR4 in human stem cell homing and repopulation of transplanted immune-deficient NOD/SCID and NOD/SCID/B2m(null) mice. Leukemia 16:1992-03
11. Li L, Neaves WB (2006) Normal stem cells and cancer stem cells: the niche matters. Cancer Res 66:4553-7
12. Collins AT, Berry PA, Hyde C, Stower MJ, Maitland NJ(2005) Prospective identification of Tumorigenic Prostate Cancer Stem Cells. Cancer Res 65:10946-51.
13. Morrison SJ, Kimble J (2006) Asymmetric and symmetric stem-cell divisions in dvelopment and cancer. Nature 441:1068-1074
14. Orimo A, Gupta PB, Sgroi DC, et al. (2005) Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell 121:335-48
15. Passegue E, Jamieson CH, Ailles LE, Weissman IL (2003) Normal and leukemic hematopoiesis: are leukemias a stem cell disorder or reacquisition of stem cell characteristics? Proc Natl Acad Sci USA 100:11842-9.
16. Pece S et al. (2004) Loss of negative regulation by Numb over Notch is relevant to human breast carcinogenesis. J Cell Biol 167:215-21.
17. Polyak K, Hahn WC (2005) Roots and stems: stem cells in cancer. Nat Med 11:296-300
18. Ponti D, Costa A, Zaffaroni N, et al. (2005) Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer Res 65:5506-11.
19. Prindull G (2005) Hypothesis: cell plasticity, linking embryonal stem cells to adult stem cell reservoirs and metastatic cancer cells? Exp Hematol 33:738-46.
20. Regala RP et al. (2005) Atypical protein kinase Ciota plays a critical role in human lung cancer cell growth and tumorigenicity. J Biol Chem 280:31109-15
21. Regala RP et al. (2005) Atypical protein kinase Ciota is an oncogene in human non-small cell lung cancer. Cancer Res. 65: p. 8905-8911
22. Reya T, Morrison SJ, Clarke MF, Weissman IL (2001) Stem cells, cancer, and cancer stem cells. Nature 414:105-11.
23. Reya T, Clevers H (2005) Wnt signaling in stem cells and cancer. Nature. 434:843-50
24. Romer JT, Kimura H, Magdaleno S, et al. (2004) Suppression of the Shh pathway using a small molecule inhibitor eliminates medulloblastoma in Ptc1(+/-)p53(-/-) mice. Cancer Cell 6:229-40.
25. Sherr CJ. (2001) The ink4a/arf network in tumor suppression. Nat Rev Mol Cell Biol 2: 731-7.
26. Singh SK, Hawkins C, Clarke ID, et al. (2004) Identification of human brain tumor initiating cells. Nature 432:396-01.
27. Stylianou S, Clarke RB, Brennan K (2006) Aberrant activation of notch signaling in human breast cancer. Cancer Res 66:1517-25.
28. Wicha MS, Liu S, Dontu G (2006) Cancer stem cells: an old idea—a paradigm shift. Cancer Res 66:1883-90.

Acknowledgements

This review was prepared by the following graduate students in the Stem Cell Biology Class, Graduate School of Biomedical Sciences, University of Medicine and Dentistry of New Jersey:

Babak Baseri, Jennifer Brady, Nadia Senmartin (in alphabetical order) 2006

Teaching Assistant: Kathy Trzaska 2006