The Cell of Origin of Breast Cancer -- Stem Cells, Progenitor Cells or More Differentiated Cells?

Tumor cells are similar to stem cells in many aspects, including incomplete differentiation status, high proliferation potential, capacity for self-renewal, ability to migrate, and expression of stem cell markers; consequently, cancers were thought to arise from stem cells that are blocked from normal differentiation. This theory has been supported in the case of teratocarcinomas, and it is generally accepted for colorectal cancer. In the breast, it has been hypothesized that hormone-induced cell differentiation of stem cells is a contributing factors for the reduced risk of breast cancer in women who have their first full-term pregnancy early in life. However, the exact cell of origin of breast cancer is still unclear. The goal of this project is introduce oncogenic mutations selectively into specific cell subsets (stem cells, various progenitors, and various differentiated cells) in the mouse mammary gland, and to learn about their responses to these oncogenic mutations.  Besides cell subsets in the normal mammary tissue, these mutated genes will also be introduced into different cell subsets in precancerous early lesions to gain insight into how the cellular heterogeneity in the early lesions contributes to the eventual malignancy. The data from these studies may have a profound impact on understanding breast cancer evolution, cellular and genetic heterogeneity, and cancer spread. The results may also suggest novel approaches for cancer prevention and may provide better preclinical models for screening new targeted therapeutics.

The issue of cancer cell of origin cannot be studied conclusively in cultured cells and has been addressed in vivo. However, conventional transgenic and knockout mouse models do not usually have the precision for expression of oncogenic mutations in selected cell subsets. However, as detailed in the last section of this page, the Li lab has pioneered the RCAS-TVA technology for expression of oncogenes in distinct mammary cell subsets. This technique is used widely in the Li lab to address questions in this project and several projects that are described below.

Wnt Signaling and Breast Cancer

Wnt Signaling and Breast Cancer (320x240)
Wnt Signaling and Breast Cancer

Wnt signaling is activated in many cancer types including breast cancer, especially the basal-subtype (which overlap heavily with the triple-negative subtype), which is most difficult to treat. However, how Wnt signaling causes transformation of breast cells remain unclear. The focus of this project to delineate the cell subtype (stem cells, progenitor cells, or more differentiated cells) that is most susceptible to transformation by Wnt signaling and to investigate the signaling components and key collaborating oncogenic pathways that are crucial in breast cell transformation. One potential collaborating partner is members of the EGFR signaling pathway.

LGR4 Signaling in Breast Cancer Metastasis and Cancer Stem Cells

LGR4, LGR5, and LGR6 comprise a group of proteins in the leucine-rich repeat-containing G-protein-coupled receptors (LGR). They are known to play important and non-redundant roles in development, stem cells, and cancer. LGR4 is highly expressed in the basal-like breast cancer (BLBC), the most aggressive subtype of breast cancer that lacks targeted therapy. There is evidence that LGR4 participates in the formation and progression of BLBCs; however, how LGR4 is causally involved in BLBC and cancer stem cells is unclear. The goal of this project is to establish the molecular mechanisms by which LGR4 regulates breast cancer metastasis and cancer stem cells.  Cutting edge technologies include mass-spectrometry, reverse phase protein array, RNA-seq, and advanced mouse models will be used. This work has important implications in understanding molecular regulation of the progression of basal-like breast cancer and may lead to novel therapeutics in treating basal-like tumors with high LGR4 expression.

Formation and Breakdown of Anticancer Barriers (Apoptosis and Senescence) in Breast Tumorigenesis

Premalignant lesions (such as atypia) of the breast sometimes but not always progress to invasive cancer. What causes this small subset of premalignant lesions to progress is poorly understood. Studies in several tissue types indicate that apoptosis and senescence are activated in human premalignant lesions as a result of oncogene over-expression and/or oncogene-induced aberrant proliferation, providing barriers to progression to malignancy. These barriers must be overcome for early lesions to develop into full-blown cancer. However, how these barriers are initially formed in precancerous early lesions and how they are broken in the progression to cancer remain unclear. The goal of this project is to elucidate these barriers at the molecular levels and to identify the key molecular pathways underlying the formation and regulation of these barriers.  Some of the key molecules studied in this project include ATM, CHK2, p53, ARF, JAK, STAT, and BCL family members. The impact of this project includes a better understanding of cancer initiation and potential new approaches in breast cancer prevention. (Please read the next two projects for this translational implication.)

JAK-STAT5 Signaling in Breast Cancer Risk and Prevention

Li Lab photo (320x240)
JAK-STAT5 Signaling

Reducing breast cancer incidence can theoretically have a profound impact on saving lives and reducing the huge cost of treatment. Antiestrogens can prevent breast cancer, but they require prolonged treatment and can have significant side effects; women who are yet cancer-free are often reluctant to participate in this chemopreventive treatment or discontinue the treatment prematurely. Therefore, new preventive therapy that does not require years of continuous treat­ment is needed.  Premalignant lesions of the breast sometimes but not always progress to invasive cancer. The Li lab previously found that Jak2-STAT5 signaling may be a key pathway that can break the apoptosis anticancer barrier in these early lesions. View publication.

The goal of this project is testing whether transient or intermittent inhibition of this pathway in early lesions could devitalize them and lower the risk of breast cancer, while the possible adverse effects, cost, and inconvenience to women would be small. The research approach includes rodent models, human tissue studies, and a window-of-opportunity clinical trial in the Breast Center Clinic that is testing this concept of cancer prevention.

Preventing Breast Cancer in Women on Antipsychotics

There is a large population of breast cancer high-risk women who are also on antipsychotics (neuroleptics) in the U.S. Antiestrogen treatment is the only FDA-approved therapy for preventing breast cancer. It requires prolonged and continuous treatment and can have significant side-effects. Therefore, it is not widely used in this vulnerable population of women. The overall goal of this project is to identify an effective chemoprevention approach that requires only a short-term treatment for women on antipsychotics.

The use of antipsychotics is often associated with an abnormal increase of serum prolactin which can activate JAK2-STAT5 signaling. Therefore, the specific goal of this project is to test whether antipsychotics activate JAK2-STAT5 signaling in preexistent early precancerous lesions in the breast, and to investigate whether intermittent treatment to block JAK2-STAT5 activity can effectively prevent breast cancer in high-risk women on antipsychotics. This proposed study will establish the roles of PRL-JAK2-STAT5 in breast cancer risk in women on antipsychotics, and may lead to a realistic prevention method for a large population of high-risk women on antipsychotics. 

The RCAS-TVA Technology for Precision Oncogene Delivery and Advanced Cancer Modeling

TVA Gene Delivery System (320x240)
TVA Gene Delivery System

The great majority of human breast cancers initiate from one or a few mutated cells in a “field” of normal breast cells, but transgenic and knockout models (including inducible, conditional and even knock-in models) express the initiating oncogene in the great majority of the mammary epithelial cells, and thus are not ideal for modeling cancer initiation (though proven valuable for studying the behaviors of established cancer). The RCAS-TVA method uses a sub-group A avian leukosis virus vector (RCAS) to carry exogenous genes into specific somatic mouse cells that are made susceptible to infection by the expression of TVA, which encodes the receptor for RCAS. View publication.

In mammalian cells, RCAS produces only the exogenous gene product, but not viral structural proteins; thus, this virus does not spread or induce immune rejection. This method has allowed members of the Li lab to introduce oncogenes into a small num­ber of mammary cells in mice, and to study their effects in these cells in the surrounding completely normal mammary tissue, recapitulating the initiation of human sporadic can­cers. Furthermore, by expressing the TVA gene under the control of promoters selectively active in stem, progenitor, or more differentiated cells, cancer-related genes could be introduced into these individual cell subsets to investigate their distinct responses to oncogenic mutations. Several of projects described above take advantages of various features of this powerful technique.