Breast Cancer Research and Treatment 58: 183–186, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands.
Workshop summary
Research potential of a unique xenograft model of human proliferative breast disease Gloria H. Heppner1 , Sandra R. Wolman2 , Jeffrey Rosen3 , David Salomon4 , Gil Smith4 , and Suresh Mohla4 1 Karmanos
Cancer Institute, Wayne State University, Detroit, MI; 2 Uniformed Services University of the Health Sciences, Bethesda, MD; 3 Baylor College of Medicine, Houston, TX; 4 National Cancer Institute, Bethesda, MD, USA Key words: xenograft model, preneoplasia, MCF10AT Summary A workshop on the ‘Research potential of a unique xenograft model of human proliferative breast disease’ was held at the Karmanos Cancer Institute, Wayne State University, Detroit, Michigan, in November of 1998. The accumulated information and current experimental findings on the MCF10AT model of preneoplastic, proliferative breast disease were reviewed. Discussions focused on the relevance of the model to clinical breast cancer and on the most profitable lines of further research to strengthen its utility.
Introduction Experimental analysis of the early development of human breast cancer has been hampered by the lack of reproducible models that mimic the morphological events believed to be indicative of an increased risk of the disease. Page and Dupont have defined histologically abnormal states, collectively known as proliferative breast disease (PBD), in the breasts of women with a high risk for subsequent development of invasive cancer. Two of these lesions, atypical hyperplasia (AH) and carcinoma-in-situ (CIS), confer a 4-fold and an 11-fold risk, respectively, of developing invasive breast cancer within 15 years [1]. Investigators at the Karmanos Cancer Institute, and their colleagues elsewhere, have initiated a system of cell lines, the MCF10AT system, that can be propagated as xenografts. When injected into nude/beige mice the MCF10AT lines produce a variety of histological elements and patterns, including ‘simple’ ducts, hyperplasia, AH, CIS, and invasive tumors. About 25% of the grafts develop invasive carcinoma at times ranging from 50 days to 2 years after injection [2]. Estrogen (E2) supplementation of MCF10AT implant-bearing nude/beige mice markedly accelerates the frequency and progression of PBD lesions.
The histology of the lesions in E2-supplemented mice suggests that the degree of angiogenesis is greater than in lesions from non-supplemented mice [3]. The MCF10AT system has recently been reviewed in detail [4]. Following presentations on the characterization and applications of the MCF10AT model, a series of recommendations was developed:
Validation of the model in regard to human breast cancer (1) The MCF10AT system of cell lines displays morphological stages that mimic the early phases of breast cancer evolution from hyperplasia to invasive disease. Some details differ from the norms of human breast histology; for example, the MCF10AT simple tubules are generally larger than their normal counterparts and although structurally similar, display cellular features more characteristic of hyperplasia. The CIS lesions are generally low grade and the comedo type of necrosis is rarely seen. The spectrum of MCF10AT cancers is morphologically heterogeneous, and the histologic distribution is unusual in the frequency of squamous differentiation in the cancers. The invasive tumors
184
GH Heppner et al.
range from well differentiated adenocarcinomas to completely undifferentiated lesions. Thus, the model appears to be a valid representation of the ADH-CIS spectrum in vivo. (2) The physical three-dimensional relationships of the various lesions to each other has not yet been fully delineated. Little evidence is available for one type of lesion actually giving rise to another, in part because of restrictions imposed by the small size, and limited inclusion of different lesions within the same nodule. Serial sectioning which would permit threedimensional reconstruction of entire xenografts, and sampling at different times and stages after inoculation, were recommended to address this issue. (3) A detailed biologic characterization of the xenograft lesions, including immunohistochemical studies of hormone receptors (ER, PR), p53, c-erbB2, and other markers correlated with the individual types of cells and lesions is under way. The goal is to achieve molecular signatures of each stage in the progression of tumor development. However, there is an inherent problem in examination of these markers – evidence from the clinical counterpart lesions indicates that there is considerable heterogeneity of their expression. Thus, although it may be possible to define molecular signatures within the model, it must be recognized that unique signatures do not characterize the clinical lesions. In response to the evidence of conformationally altered, wild type p53 expression in the model [5], the suggestion was made to study the phosphorylation status of specific carboxy-terminal sites in the tumor suppressor in MCF10AT cells in the hopes of determining whether p53 is disrupted functionally. In addition, determination of changes in various CDK inhibitors and cell cycle components was recommended. Stem cell characterization Mammary disease as well as normal regional mammary development in situ have been shown to be clonal by direct analysis of DNA isolated from paraffin blocks from the affected breast [6–8]. Given current interest in tissue-specific stem cells in cancer and the demonstrated clonal nature of contiguous regions of both normal and diseased mammary epithelium, the MCF10AT model provides an excellent opportunity to study human mammary stem cells and their potential for development of neoplastic disease. Bi-directional differentiation of both epithelial and myopithelial pathways has been recognized in
MCF10AT cells [9]. Further validation of the stem cell characteristics of the line will include testing for the presence of epithelial-specific surface markers, both those associated with myoepithelial cells (CALLA/CD10) and those with luminal epithelial cells (MUC-1 glycoprotein and ESA (epithelialspecific antigen)). Flow cytometry with single cell sorting could permit separation of the cell types which could then be evaluated for their ability to produce colonies in 2-D and 3-D (collagen gel) culture systems. Although a stem cell might express markers of both or neither type, the progeny of a clone is expected to include both cell types. The progression of individual clones within the implants should be examined with respect to the clonal or polyclonal nature of the various histopathological lesions. Suitable approaches include marking cells in culture with retroviral vectors containing two different selectable markers such as green fluorescence protein (GFP) and E. coli β galactosidase. Both expression of the selection markers and the location of the retroviral gene insertion within the somatic DNA could demonstrate the clonal or non-clonal nature of the lesions obtained. Functional genomics A comprehensive description of expressed genetic alterations accompanying the morphologic progression is being developed using the combination of laser microdissection and microchip array technologies. The MCF10AT lines show promise as a system to test gene function and to define predictive events in breast cancer progression. Studies using retrovirally transduced gene constructs will be followed by evaluation in both the Matrigel culture environment and in xenografts. The latter should preferably be performed in the cleared fat pad of appropriate immuno-compromised mice, perhaps, (if possible) the RAG-1 and RAG-2 deficient mice developed in Baltimore’s laboratory [10] which are reported to have more normal E2 levels than do nude or SCID mice. The importance of demonstrating that the phenotypic properties of the transduced cells are altered reproducibly by the expression of the transduced construct was emphasized. Role of estrogen in breast cancer development The demonstration that E2 enhances progression in MCF10AT xenografts and growth in a newly de-
Research potential of a unique xenograft model veloped short-term Matrigel culture system creates new opportunities to exploit the model. Previous characterizations of the frequency and phenotypes of the lesions that develop in xenografts are being repeated in the presence of E2 supplementation. The possibility that the parental (non-transfected) MCF10A cells may produce outgrowths in E2-supplemented mice is also being investigated. Additional studies will examine the status of ER and PR in these cells and the effects of these hormones on authentic steroid receptor gene targets, not only the artificial ERE-tkCAT constructs that have been tested. In light of the in vivo pathology, further development of the in vitro system was seen as leading to a promising, practical way to screen new antiestrogens. The potential for identification of molecular markers that can be used as diagnostic tools and of intermediate biomarkers for eventual chemoprevention studies was also noted. Epithelial stromal interactions Another attractive feature of the MCF10AT model is its utility in investigating the role of stroma in breast cancer development. The perception that E2 increases angiogenesis in the xenografts was mentioned earlier; efforts to model this effect in epithelial–endothelial cell co-cultures will allow explorations of the molecular and cellular mechanisms that underlie the relationships between angiogenesis and breast cancer development. Expansion of the co-culture methodology to include human breast fibroblast lines was suggested as a way to address additional autocrine– paracrine interactions, and to investigate the role of aromatase, the rate-limiting enzyme responsible for estrogen biosynthesis, in early breast cancer development. The role of mouse stroma in the development of the xenograft lesion deserves further attention. Histologically, mouse-derived stroma closely invests the successfully growing implants, and the biological contribution of the mouse cells needs to be characterized. Questions to be addressed include: Is the stromal component of successful outgrowths itself proliferative? Does it result from chemotaxis? Does it enhance human epithelial cell proliferation?
have been generated and how these derivatives can be utilized to address specific biological questions. For simplification purposes, however, it was recommended that further developmental work on the model be focused on three lines: The parental, non-transformed and immortalized MCF-10A, the RAS-transformed MCF-10AneoT cells, and the first passage MCF-10AT cell line that progresses in vivo to AH, CIS, and carcinoma. The model should be defined as a model for early breast cancer development and progression. Investigators who may wish to utilize the MCF10AT model in their own laboratories are encouraged to contact Dr. Robert Pauley, Director, Cell Lines Core, Karmanos Cancer Institute, 110 E. Warren Ave, Detroit, MI 48201. Phone: 313-833-0715 x2456; email at:
[email protected] Additional efforts to optimize and standardize the model were suggested, e.g., the lines could be incorporated into Matrigel that contains a potent angiogenic factor such as VEGF and inoculated into SCID mice that have been previously immunosuppressed with etoposide [11]. Theoretically, this should increase the number of lesions and decrease the latency period after which they arise. The simultaneous administration of E2 pellets to the SCID mice should further accelerate the time line for such changes. The in vivo system also offers the possibility of transducing candidate breast-cancer-associated genes into the MCF-10A, MCF-10AneoT, and MCF-10AT cells and studying their effects on the frequency and temporal development of ADH, CIS, and carcinoma. Acknowledgements The workshop was supported by The Division of Cancer Biology, National Cancer Institute through a supplement to NIH grants: CA40104 (S. Haslam) and CA66898 (G. Heppner).
References 1.
2.
Simplification and optimization of the MCF10AT model Participants in the workshop commented repeatedly on the large number of MCF10AT derivatives that
185
3.
Page DL, Dupont WD: Anatomic markers of human premalignancy and risk of breast cancer. Cancer 66: 1326–1335, 1990 Dawson PJ, Wolman SR, Tait L, Heppner GH, Miller FR: MCF10AT: A model for the evolution of cancer from proliferative breast disease. Am J Path 148: 313–319, 1996 Shekhar PVM, Nangia-Makkar P, Wolman SR, Tait L, Heppner GH, Visscher DW: Direct action of estrogen on sequence of progression of human preneoplastic breast disease. Am J Path 152: 1129–1132, 1998
186 4.
GH Heppner et al.
Heppner GH, Wolman SR: MCF-10AT: A model for human breast cancer development. The Breast J 5: 1–8, 1999 5. Shekhar PVM, Welte R, Christman JK, Wang H, Werdell J: Altered p53 conformation: a novel mechanism of wild-type P53 functional inactivation in a model for early human breast cancer. Int J Oncol 11: 1087–1094, 1997 6. Tsai YC, Lu Y, Nichols PW, Zlotnikov C, Jones PA, Smith HS: Contiguous patches of normal human mammary epithelium derived from a single stem cell: implications for breast carcinogenesis. Cancer Res 56: 402–404, 1996 7. Deng G, Lu Y, Zlotnikov G, Thor AD, Smith HS: Loss of heterozygosity in normal tissue adjacent to breast carcinomas. Science 274: 2057–2059, 1996 8. Rosenberg CL, Larson PS, Romo JD, De Las Morenas A, Faller DV: Microsatellite alterations indicating monoclonality in atypical hyperplasias associated with breast cancer. Human Path 28: 214–219, 1997
9.
10.
11.
Pauley RJ, Soule HD, Tait L, Miller FR, Wolman SR, Dawson PJ, Heppner GH: The MCF10 family of spontaneously immortalized human breast epithelial cell lines and models of neoplastic progression. Eur J Cancer Prevention 2: 67–76, 1993 Roman CA, Cherry SR, Baltimore D: Complementation of V (D) J recombination deficiency in RAG-1 (-/-) B cells reveals a requirement for novel elements in the N-terminus of RAG-1. Immunity 7: 13–24, 1997 Visonneau S, Cesano A, Torosian MH, Miller EJ, Santoli D: Growth characteristics and metastatic properties of human breast cancer xenografts in immunodeficient mice. Am J Path 152: 1299–1311, 1998
Address for offprints and correspondence: Gloria Heppner, Karmanos Cancer Institute, 4100 John R., 2nd Floor, Detroit, MI 48201, USA; Tel: 313.745.1301; Fax: 313.966.0823
