Implications of the cancer stem cell hypothesis for the early detection and prevention of breast cancer.

PL03-01

Recent research in a variety of tumor systems has given support to the "Cancer Stem Cell Hypothesis". Two important components of this hypothesis are that tumors originate in tissue stem or progenitor cells as a result of disregulation of the normally tightly regulated process of self-renewal. As a result of this, tumors contain and are driven by a cellular subcomponent that retains key stem cell properties including self-renewal which drives tumorigenesis and differentiation that contributes to cellular heterogeneity. The "Cancer Stem Cell Hypothesis" has profound implications for the early detection, prevention, and treatment of cancer. Our laboratory has developed techniques for the isolation and characterization of stem cells from human mammary glands and mammary carcinomas. Mammary stem and progenitor cells can be cultured in vitro as floating spherical colonies which we have termed "mammospheres" which are composed of progenitor cells capable of multi-lineage differentiation, as well as stem cells capable of mammosphere formation. These more primitive cells express the stem cell markers OCT-4, Bmi-1, and ALDH-1. Mammosphere initiating cells which are estrogen receptor negative give rise to estrogen receptor positive proliferating progenitor cells. We have utilized these in vitro systems as well as a humanized NOD-SCID mouse model to study the developmental pathways which regulate stem cell self-renewal and lineage specific commitment. Both Notch and Hedgehog signaling are involved in stem cell self-renewal as evidenced by the observation that activation of these pathways increases mammosphere number upon serial passage. These effects are mediated by the polycomb gene Bmi-1. Over expression of the Hedgehog component Gli-2 in mammosphere initiating cells results in the generation of ductal hyperplasias when these cells are implanted into the mammary fat pads of humanized NOD-SCID mice. In contrast, inhibition of Bmi-1 utilizing a siRNA lenti virus inhibits mammosphere formation in vitro and mammary gland development in these mice. These results suggest that disregulation of normal stem cell self-renewal results in expansion of stem cell pools which may then serve as targets for further transformational events. We have previously reported the isolation of a subpopulation of cells within human breast cancers which display stem cell characteristics. These cells characterized as CD44+ CD24 low lin- display activated Hedgehog signaling and increased expression of Bmi-1. In order to define markers which might prove useful for in situ characterization of normal and malignant mammary stem cells, we have examined the expression of aldehyde dehydrogenase (ALDH) an enzyme previously found to be expressed in hematopoietic and neuronal stem cells. ALDH may be assessed by the Aldefluor assay which utilizes a fluorescent probe to detect cells displaying aldehyde dehydrogenase activity. We have found that the Aldefluor positive population of normal mammary epithelium has stem cell characteristics as demonstrated by mammosphere formation and generation of ductal alveolar structures in humanized NOD-SCID mice. Furthermore, in human mammary carcinoma xenografts, Aldeflour positive cells display properties of "cancer stem cells". As few as 200 Aldefluor positive cells, which comprise between 1-10% of the tumor cell population, produce tumors in NOD-SCID mice while 2x10³ Aldefluor negative cells failed to generate tumors. Consistent with the Aldefluor positive population representing cancer stem cells, tumors generated from this population recapitulate the phenotypic heterogeneity of ALDH expression seen in the original tumor. Based on the above studies, we utilized immunochemistry to detect ALDH-1 as a marker of normal and malignant mammary stem/progenitor cells in situ. ALDH-1 antibodies detect a rare cell population in normal human mammary gland and in mammospheres generated from these cells. Expanded ALDH-1 positive populations are found in pre-malignant lesions such as ductal carcinoma in situ (DCIS) and at areas of microinvasion. Furthermore, utilizing tumor tissue microarrays, we have determined that expression of ALDH-1 is associated with a poor clinical outcome. These studies demonstrate that ALDH-1 may be utilized to detect normal and malignant human mammary stem/progenitor cells providing a new tool for the early detection of beast cancer. Furthermore, since these cells may drive carcinogenesis, strategies aimed at their elimination, represents a rational approach for cancer prevention.