Normal stem cells: We are continuing to advance the recapitulation of human breast epithelial morphogenesis by humanizing the mammary fat pad of NOD/ SCID mice as described by Kuperwasser et al. (2004). We have been successful in using alternative approaches to enhance epithelial cell growth in the humanized model system. We have optimized a protocol for the isolation and differentiation of human macrophages. The addition of primary human macrophages has significantly increased the formation of human epithelial lobular structures. The paracrine signaling interactions remain to be determined;however, it is clear that the addition of human macrophages provides a more permissive niche for the growth of human cells, and potentially human mammary stem cells, in the in vivo model. To begin to understand the paracrine signaling between the macrophages and epithelial cells, we have identified the cytokines secreted by the differentially activated macrophages via quantitative real time PCR array analysis. To gain insight into the role of the microenvironment of breast stem cell growth and tumorigenicity, we have investigated the role of the stromal fibroblasts on the tumorigenicity of primary metastatic breast cancer cells. We chose to examine the effect of fibroblasts derived from pre-menopausal African- or Caucasian-American patients given the high incidence of aggressive (i.e. triple negative), metastatic breast cancer development in pre-menopausal African-American women. Results show that when the gland is humanized with Caucasian-American fibroblasts then injected with pleural effusion cells derived from metastatic breast cancer patients, the resultant tumors are significantly larger compared to the glands humanized with African-American fibroblasts. These results were also repeated using the well-characterized breast cancer T47D cell line. Quantitative real-time PCR array analysis has indicated that the Caucasian-American fibroblasts have increased cytokine production, potentially leading to the more permissive environment for tumorigenesis. Using our newly optimized protocol for isolating human breast extracellular matrix, we have shown that human breast ECM can alter the tumorigenicity of breast cancer cells, depending on the source of the ECM (i.e. age and race of the patient) and the hormone receptor status of the breast cancer cells. We have used proteomics to begin to identify the components of the breast ECM and will use this knowledge for future studies to identify the role of the breast microenvironment and the stem cell niche. Current studies are being conducted to determine if ECM molecules have the ability to re-direct stem cell fate and tumorigenesis. Development of a standard human xenograft breast cancer model: Breast cancer studies often utilize xenograft mouse models by implanting human cancer cells under the renal capsule, subcutaneously, or orthotopically into the mammary gland. However, these procedural differences and the discretionary use of estrogen and immune suppressants (etoposide) can make seemingly relatable results incomparable. To standardize the in vivo breast cancer model, we directly compared the effects of orthotopic, transdermal, and subcutaneous sites of injection on tumorigenicity, in the presence or absence of estrogen and/or etoposide pretreatment. Cells derived from a human breast cancer pleural effusion were injected into the inguinal or thoracic mammary gland of NOD/SCID mice either after surgical incision (orthotopic) or through the skin (transdermal). Tumors in inguinal glands had significantly decreased latency and increased volume and incidence compared to those in thoracic glands, independent of method of injection. Within the inguinal gland, no statistically significant difference in tumor volume was found between groups treated with and without estrogen or etoposide. In addition, athymic NCr-nu/nu mice treated estradiol were injected orthotopically or subcutaneously with well-characterized human breast cancer cell lines (MCF7, ZR-75-1, MDA MB231, or MCF10A1h). Increased tumor volume was observed with all four cell lines injected orthotopically. These studies strongly suggests that the inguinal gland is the preferred site for a xenograft model of breast cancer and illustrates the influence of the local microenvironment on in vivo models. Cancer stem cells: Our work has been focused on developing ways to enrich and/or identify tumor initiating cells (TIC) in both established breast cancer cell lines and clinical samples. We demonstrated that in ER negative clinical samples and resulting xenografts, cells lacking CD44 were largely non-tumorigenic while CD44+CD24+ and CD44+CD24- cells were equally tumorigenic. Subsequently, we have focused on segregating CD44+ cells into two distinct fractions: one tumorigenic and one non-tumorigenic. We have identified a novel extracellular marker profile that effectively enrichs TIC from human ER- breast cancer. FACS sorted TIC cells from primary tumors and their resulting xenografts give rise to tumors in NOD/SCID mice with as few as 50 cells. Contrary to this, 2,500 cells lacking the TIC marker profile fail to yield tumors. Furthermore, the novel TIC cells faithfully recapitulate tumor molecular sub-type and immunophenotypic heterogeneity through multiple passages. At the transcript level, our novel TIC profile enriches a population of cells with elevated expression genes known to play a role in stem cell self-renewal including BMI1, nanog, and sox2. We currently are planning to evaluate global mRNA, miRNA, and promoter methylation status of the TIC and non-tumor initiating cells. Additionally, we demonstrated that in the MCF10 Ca1a cell line, CD44+CD24- cells were more invasive than CD44+CD24+ cells. Furthermore, we demonstrated that both expression of CD24 and the epithelial-like phenotype associated with its expression was under dynamic regulation. Specifically, epithelial-like CD44+CD24+ cells readily give rise to invasive mesenchymal-like CD44+CD24- cells and vice versa. We demonstrated similar events with MCF7, SUM159, and MDA MB231 cells. Also, we demonstrated that CD24 promoter CpG methylation was similar between CD44+CD24+ and CD44+CD24- cells, supporting the observation that this gene is susceptible to dynamic regulation. Using siRNA methodologies, we found that CD24 depletion increased a cells ability to invade through Matrigel. However, CD24 knockdown failed to yield the mesenchymal phenotype observed in cells where CD24 is down-regulated endogenously.
| Wolf, Ronald; Voscopoulos, Christopher; Winston, Jason et al. (2009) Highly homologous hS100A15 and hS100A7 proteins are distinctly expressed in normal breast tissue and breast cancer. Cancer Lett 277:101-7 |
| Vonderhaar, Barbara K; Smith, Gilbert H (2008) Stem cells and breast cancer. Introduction. Breast Dis 29:1 |
| Fleming, J M; Long, E L; Ginsburg, E et al. (2008) Interlobular and intralobular mammary stroma: genotype may not reflect phenotype. BMC Cell Biol 9:46 |
