Stem Cells in Breast Cancer Metastasis

Institution: Scripps Research Institute
Investigator(s): John Yates, M.D., Ph.D. - Brunhilde Felding, Ph.D. - Evan Snyder, M.D., Ph.D. -
Award Cycle: 2004 (Cycle 10) Grant #: 10YB-0202 Award: $621,289
Award Type: SPRC Full
Research Priorities
Biology of the Breast Cell>Pathogenesis: understanding the disease

This is a collaboration with: 10YB-1202 -

Initial Award Abstract (2004)
The concept of a reservoir of stem cells, persisting throughout life and capable of multi-lineage differentiation is firmly established and well characterized for blood cells (i.e., RBCs, WBCs & platelets). Less well understood, but compelling in its rationale, is the idea that most normal human tissues are maintained through the differentiation of a resident population of slowly cycling stem cells. These tissue specific stem cells regenerate themselves and produce more differentiated progeny, to yield the complex array of cell types that define most bodily organs. In the mammary gland, cyclical changes of estrus and pregnancy, repeated though a lifetime, require the balanced maintenance of a diversity of cell types in response to a variety of external hormonal stimuli. This renders the mammary gland susceptible to the introduction of genetic errors that may accumulate in its stem cell population and ultimately lead to malignancy and metastasis. Recent studies have identified a small fraction of primary tumor cells with early progenitor phenotypes and genetic fingerprints as the tumor initiators. A key property of stem cells is their extreme migratory responsiveness to cytokinetic and hormonal signals. We therefore propose that a population of aberrant stem-like cells within a breast tumor is critically involved in the initiation of metastatic disease.

The CBCRP-supported collaboration involves expertise in the biology of breast cancer metastasis (Drs. Felding-Habermann and O'Sullivan), expertise in proteomic analysis (Dr. Yates), and expertise in stem cells in neuron biology (Dr. Snyder). Our background from these three disciplines will be integrated into the project aims. Our central hypothesis is that metastatic breast cancers arise when mutated cells with the capacity for self-renewal and in situ differentiation arrive at distant organs and, in response to signals from their new environment, differentiate along pathways which allow them to adapt to and ultimately thrive in the new environment. We will expand upon data implicating epithelial stem cell-like tumor initiator populations as the source of human solid tumors to show that these progenitor cells are also responsible for metastatic disease. First, we aim to isolate and characterize progenitor cell populations from normal breast tissue, breast tumors and metastases. In this aim, we will fully work out the parameters by which we can identify tumor progenitor cells from patient samples and human breast cancer cell models. Because no single method has yet been universally adopted to identify epithelial stem/progenitor cells, we plan to use dye exclusion, cell surface markers, bromodeoxyuridine incorporation, cytoskelton staining, and tumorigenicity experiments as our initial approaches to defining and isolating progenitor populations.

Our second aim is to analyze whether progenitor cell populations from human breast tumors give rise to distant metastases. In this aim, we will take the isolated progenitor populations and test their ability to colonize target organs of breast cancer metastasis in immune deficient mice both from the circulation and from primary tumors established in the mammary fat pads. We will use cell labeling, quantitative PCR and immuno histochemistry to identify and quantify tumors and metastases. In our third aim, we will explore the possibility of harnessing normal stem cells for guided targeting of distant metastases, particularly those that develop in the brain. One of us (Snyder) found that neural stem cells have a stunning tropism to diseased areas in the brain. We will establish brain metastases in immune deficient mice by introducing human breast cancer cells that home to the brain as a target organ, and then test whether normal neural stem cells seek out areas of tumor growth. Genetically tagged with distinct variants of luciferase, tumor cells and neural stem cells can be detected and distinguished by non-invasive imaging. This will allow us to follow both cell types during disease progression in individual animals, and to test the concept that neural stem cells may potentially serve to deliver anti-tumor gene products or pro-drugs and eventually help to repair the tumor-afflicted brain.

In our final aim, we will analyze protein profiles of stem cells from normal breast tissue, breast tumors and metastases. We will establish protein expression "signatures" of matched samples from the same patient using multidimensional proteomic (MudPIT) analysis, and compare progenitor cells with differentiated cell populations from each tissue type. With this approach, we hope to reveal a comprehensive spectrum of differences between the progenitor cells and their progeny and the influence of the site of metastasis on gene and protein expression.

Understanding the biology of the tumor-initiating cell population in breast cancer would have a profound impact on how we design therapies, particularly those targeting metastatic disease in the brain. It would also greatly enhance our ability to predict the metastatic potential of individual tumors and help patients and their physicians to select their most promising treatment options.


Final Report (2006)
Introduction: The central hypothesis of this project is that metastatic breast cancers arise when mutated stem-like cells with the capacity for self renewal and in situ differentiation disseminate to distant organs and, in response to signals from their new environment, differentiate along pathways which allow them to adapt to and thrive within their new surrounding. Our hypothesis includes that normal stem cells may serve to target breast cancer metastasis, based on the ability of normal stem cells to seek out diseased areas in the body.

Progress: Our goals were to:
1. Identify subsets of human breast cancer cells with far-reaching metastatic potential and characterize their functional phenotype;
2. Investigate the proteome profile of metastatic subsets of human breast cancer cells; and
3. Explore the possibility of harnessing normal stem cells for guided targeting of distant metastases, particularly those that develop in the brain. We have made progress in each of these aims and established the foundation for in-depth analyses of each of these research directions.

The results from our study indicate that the majority of cells from malignant pleural effusions of breast cancer patients can display several of the properties previously associated with small subpopulations of cells isolated from primary tumors. We showed that these tumor cells have multiple metastatic organ homing properties in the mouse model and recapitulate a situation often seen in late stage breast cancer patients. We further demonstrated that circulating breast cancer cells from patients with metastatic breast cancer contain cells with a clear potential for widespread metastasis. As the blood stream is a major route for breast cancer cell dissemination during breast cancer metastasis, these cells are likely responsible for a significant proportion of metastatic burden seen in the clinic. A comparison of breast cancer cell variants, derived from circulating tumor cells after homing to diverse target organs in the mouse model, revealed that breast cancer cells which seed brain metastases have a propensity, or can adapt, to the unique microenvironment in the brain. Large scale proteomic analysis indicated specific modifications in the energy metabolism of brain metastatic cells, which suggest that these tumor cells derive their energy predominantly from glucose oxidation. This feature is essentially a hallmark of neuronal cell metabolism. For balance, the brain homing breast cancer cells also activate pathways that minimize production of reactive oxygen species, which result form their enhanced oxidative metabolism. The overall metabolic alterations are associated with strongly enhanced tumor cell survival and proliferation in the brain microenvironment.

In search for a new therapeutic approach against breast cancer metastases, we followed the hypothesis that normal stem cells may be harnessed to reach metastatic breast cancer lesions. We found evidence that neural stem cells can efficiently localize to breast cancer brain metastases, and we are now studying if these cells can be harnessed to deliver inhibitory mechanism to metastatic brain lesions. This portion of the project was performed as a collaboration between Scripps (Felding-Habermann and Yates, co-PIs) and The Burnham Institute for Medical Research (Evan Snyder, co-PI).

Future direction and impact: The results from this study and our development of human brain homing breast cancer cell models and tractable neural stem cell models allow us to investigate in detail, which mechanisms control breast cancer cell survival and proliferation in the brain microenvironent, and whether normal human neural stem cells may serve as a basis for the development of a novel therapeutic approach against breast cancer brain metastasis. The outcome from our studies, triggered and made feasible through this initial project, could directly lead to the development to new therapeutic approaches to prevent and treat metastatic breast cancer in the clinic.

The biology of metastasis to a sanctuary site.
Periodical:Clinical Cancer Research
Index Medicus: Clin Cancer Res
Authors: Palmieri D, Chambers AF, Felding-Habermann B, Huang S, Steeg PS.
Yr: 2007 Vol: 13 Nbr: 6 Abs: Pg:1656-62

Adaptation of energy metabolism in breast cancer brain metastases.
Periodical:Cancer Research
Index Medicus: Cancer Res
Authors: Chen EI, et al, Yates JR 3rd, and Felding-Habermann B
Yr: 2007 Vol: 67 Nbr: 4 Abs: Pg:1472-86

Large scale protein profiling by combination of protein fractionation and multidimensional protein identification technology (MudPIT).
Periodical:Molecular and Cellular Proteomics
Index Medicus: Mol Cell Proteomics
Authors: Chen EI, Hewel J, Felding-Habermann B, Yates JR 3rd.
Yr: 2006 Vol: 5 Nbr: 1 Abs: Pg:53-6