Stem Cells and Cancer
Rebecca J. Morris, Ph.D.
Professor

 The epidermal layer of the skin is composed largely of cells called keratinocytes. Keratinocytes in the basal layer are organized into subpopulations based on their proliferative nature and include stem cells (relatively rare) and transit amplifying cells (comprise most of the proliferating cells). When a stem cell divides, one daughter usually remains a stem cell while the other daughter gives rise to transit amplifying cells with limited proliferative potential. Upon completion of their divisions, transit amplifying cells undergo an orderly maturation process called terminal differentiation that includes their outward displacement through the suprabasal layers, production of high molecular weight keratins, loss of their nuclei, and formation of an impervious outer structure called the cornified envelope. This process is exceptionally orderly and maintains the normal thickness and cellularity, and the normal functions of the epidermis throughout life. Our work focuses on the stem cells of the hair follicles because they not only serve as a reservoir of epidermal cells, they also possess remarkable regenerative potential and are known to be able to reconstitute a graft, to heal wounds, and even to give rise to non-melanoma skin cancer.

Non-melanoma skin cancer is a significant problem for cancer research because, although rarely a cause of death, it occurs more frequently than any other malignancy, and more than five million new cases are diagnosed in the United States annually with a burden of more than eight billion dollars. An estimated one-third to one-half of all human cancers originate in the skin; skin cancers exceed all others combined; and the lifetime risk of skin cancer the United States is one in five. Solar ultraviolet radiation is the major known cause of non-melanoma skin cancers and is directly relevant to the etiology as demonstrated by epidemiological evidence and the tight correlation between non-melanoma skin cancer in humans, and solar radiation- and chemicallyinduced skin carcinogenesis in murine models. These cancers progress through an orderly sequence in which genetic, biochemical, and cellular abnormalities accumulate in target cells over time. Mild alterations initially seen within keratinocytes can only be identified histologically. Increased cellular atypia occurs with further sun damage, and hyperkeratotic, pre-malignant actinic keratoses develop. Of these, one to ten percent will progress to squamous cell carcinomas. Therefore, determining the etiology of nonmelanoma skin cancer is a critical problem in cancer research. Because avoiding exposure to sunlight is far more easily said than done, there is considerable interest in increasing knowledge of skin cancer etiology and finding targets for prevention.

Towards this end, we chose the molecule high mobility group A2, (Hmga2), a transcription factor involved in chromatin architecture, and expressed chiefly during development, where it has many key biological functions. When expressed in adult tissues from in various organs, Hmga2 is always related to cancer development. The role of Hmga2 in skin tumorigenesis is, however, not yet understood. We demonstrated that Hmga2 is found in non-transformed epidermis, specifically located to the membrane of keratinocytes in epidermis. Ex vivo culture of keratinocytes and in vivo development of skin carcinomas was associated with translocation of the Hmga2 protein from the membrane into the nucleus, where Hmga2 induced its own expression by binding to the Hmga2 promoter. Panobinostat, a histone deacetylase inhibitor, down-regulated Hmga2 expression by preventing Hmga2 from binding its own promoter, and thus inhibiting Hmga2 promoter activity. Hmga2 translocation to the nucleus could in part be prevented by an inhibitor for ROCK1. Our findings demonstrate that upon programing of benign papilloma to malignant squamous cell carcinomas, Hmga2 translocates in a ROCK-dependent manner from the membrane to the nucleus, where it serves as an auto-regulatory transcription factor, causing cellular transformation.

In a second project on the etiology of squamous cell carcinoma we focus on cytokeratin expressing cells in the blood and bone marrow and their role in cancer development. Cytokeratins are frequently found in the blood and bone marrow of patients with epithelial cancers and are attributed to metastasis. We wondered whether we could find keratin expression in blood and bone marrow in untreated adult in vivo models. To address this problem, we have used classical immunoreactivity, Krt1-14;mTmG transgenic in vivo models together with fluorescence activated cell sorting, and quantitative reverse transcriptase polymerase chain reaction. We have made several novel findings. First, we discovered, rare but distinctive, keratin-14 and keratin-15 immunoreactive cells the size of small lymphocytes in blood and bone marrow. Second, using Krt1-14;mTmG transgenic in vivo models, we found low (8.6 GFP positive cells per 10^6 cells analyzed), but significant numbers (p<0.0005) of GFP positive cells in bone marrow when compared with negative controls. Third, qRT-PCR demonstrated very low but reproducibly detectable expression of keratin-14 mRNA in blood and bone marrow when compared with epidermal keratinocytes: with blood expressing one thousand times, and bone marrow, one hundred thousand times, less than epidermal keratinocytes. Moreover, flow cytometric analysis of fresh bone marrow disclosed a subpopulation of keratin-14 immunoreactive cells that was negative for hematopoietic lineage markers. We conclude from these observations that keratin-14 protein and mRNA are expressed at low, but detectable levels in the blood and bone marrow of in vivo models. These observations should further our understanding of cutaneous biology, non-melanoma skin cancer, and other epithelia and their cancers.

In summary, research in the Morris laboratory continues to highlight the role of stem cells in the pathogenesis of squamous cell carcinoma, and has documented an unexpected contribution of bone marrow derived cells. Going forward, we will probe the interactions between epidermal stem cells and bone marrow derived cells as tumor initiating- and propagating- cells. 

“Because avoiding exposure to sunlight is far
more easily said than done, there is considerable
interest in increasing knowledge of skin cancer etiology
and finding targets for prevention.”
Dr. Rebecca Morris