Cutaneous Stem Cells and Cancer

In the last several years, there has been a great deal of interest in the relationship between stem cells and cancer. In this report, we will consider several aspects of stem cell biology that apply to cancer, while focusing on the skin. First, normal adult stem cells possess the self-renewal properties of cancer cells and thus can serve as a model system for study. Second, stem cells may be targets of carcinogenic pathways. Lastly, cancers, as abnormal tissues, appear to contain a subpopulation of cells with stem cell qualities, termed "cancer stem cells".

Cutaneous Stem Cells

In the past decade, several important studies have defined the cutaneous stem niches, identified different stem cell populations, and characterized key stem cell properties. Most importantly, stem cells divide infrequently to produce another stem cell (self-renewal) and daughter transit-amplifying cells that undergo proliferation and differentiation to replenish the cells that are lost to the environment after terminal differentiation. Many tumor cells undergo apoptosis as well as terminal differentiation, and thus tumors also depend on a process of self-renewal to sustain growth. 

Better knowledge of the normal mechanisms of asymmetric cell division utilized by stem cells will lead to a greater understanding of cancer cell growth and will likely identify additional targets of intervention. For example, recent evidence of mitotic asymmetries in DNA strand segregation and centrosome inheritance challenges the hypothesis that cell division creates two identical cells, and helps to explain how stem cells may maintain their molecular integrity.

Another significant feature of stem cells is their relative quiescence or slowly-cycling nature. On the human scalp, stem cells present in the hair follicle bulge region are quiescent during the 3-5 years of the anagen, catagen, and telogen phases and only divide at the onset of a new anagen growth phase. This cellular quiescence may be one mechanism utilized by cancer stem cells to avoid the effects of chemotherapy and radiation. Indeed, cancer patients treated with chemotherapy and radiotherapy lose their hair because rapidly dividing cells in the hair matrix are destroyed; however, hair follicle stem cells survive to regenerate a new follicle once treatment is finished.

Other biologic properties of cutaneous stem cells also mimic the behavior of cancers. While stem cells are permanent residents of the stem cell niche, early daughter cells must migrate away and invade the dermis during hair follicle cycling, similar to tumor cell invasion. Many of the cell signaling pathways and molecular effectors involved in stem cell activation during this process are dysregulated during carcinogenesis and tumor progression. The Hedgehog, β-catenin/LEF/TCF and BMP signaling pathways that control stem cell behavior are well-known players in cancer biology. Other stem cell regulatory pathways and downstream effectors are rapidly being identified and characterized during advances in stem cell research.

Tumor-Initiating Cells

An important question in cancer biology is whether adult stem cells are tumor-initiating cells during carcinogenesis. If so, what molecular pathways are involved in the transformation of stem cells? As adult stem cells reside in the body for your lifetime, it is believed that we accumulate multiple genetic mutations as we age, which eventually leads to cancer.

While difficult to prove, a growing body of evidence suggests that this is the case, at least for some tumors. Rebecca Morris has shown that DNA alterations during tumor initiation persist in cutaneous stem cells of mouse skin and that these cells give rise to skin tumors. Also, signaling pathways controlling stem cell behavior are known to induce a variety of skin tumors. Tony Oro and others have shown that activation of the hedgehog pathway induces mouse basal cell carcinomas. Elaine Fuchs' lab demonstrated that activated β-catenin produced pilomatrical tumors in mice and found activating mutations in human pilomatrixomas. Fiona Watt's group blocked β-catenin signaling by expressing a dominant-negative Lef1 transcription factor in mouse skin and the mice developed sebaceous tumors. In our subsequent collaboration, we identified mutations within the LEF1 gene in approximately 1/3 of human sebaceous tumors. These results suggest that the level of signaling through β-catenin drives the fate determination of stem cells. Promotion of one cell lineage at the expense of others may be an early event in tumorigenesis.

"Cancer Stem Cells"

Do cancers, as abnormal tissues, contain a subpopulation of cells with stem cell qualities - 'so-called' cancer stem cells? If so, what are their properties and how can this be exploited therapeutically? As mentioned above, cancer stem cells, like normal adult stem cells, may survive conventional radiation and chemotherapeutic approaches, leading to recurrences and possibly driving tumor growth and metastasis. Much weight has been placed on the ability of small subpopulations of cells isolated from cancers to form tumors in mice. It is believed that these "tumor-initiating cells" are equivalent to and defining property of cancer stem cells.

Supporting evidence for this theory is the utilization of cell-surface markers for normal stem cell populations. CD133 is one well-known example that is expressed in normal neural stem cells and can be used to sort cancer cells enriched for tumor-initiating capacity. CD20 has been used to isolate fractions of melanoma cells that form tumors in mice. Some authorities, however, question whether tumor initiation in mice is a valid measure of cancer stem cell behavior.

Since an important property of stem cells is a native quiescent, slow-cycling state, it is unclear how the ability to form tumors in mice within weeks would correlate with the in vivo behavior of cancer stem cells. In addition, the tumor microenvironment created when single-cell suspensions are injected into mice should differ dramatically from that of in vivo human tumors. Label-retaining analyses or other means of identifying slowly-cycling cells may help to address this issue but have not been reported. In addition to cell-surface epitopes, other markers of normal stem cells may help to identify and extract cancer stem cells for study. We recently determined that a subset of human cutaneous sebaceous tumors express the keratin 15 markers found in hair follicle stem cells. Whether keratin 15-positive tumor cells represent slow-cycling tumor stem cells is still unclear.

Summary 

The application of advances in stem cell biology to cancer research can lead to a better understanding of what drives carcinogenesis, tumor progression, recurrence, and metastasis. The localization of keratinocyte, melanocyte, and neural crest-derived stem cells in the bulge region of the skin provides an excellent model system to study the relationship between stem cells and cancer.