TIMP-3, an early indicator of breast cancer?

Institution: University of California, San Francisco
Investigator(s): Susan Hawkes, Ph.D. -
Award Cycle: 1996 (Cycle II) Grant #: 2RB-0122 Award: $184,537
Award Type: Research Project Awards
Research Priorities
Detection, Prognosis and Treatment>Imaging, Biomarkers, and Molecular Pathology: improving detection and diagnosis



Initial Award Abstract (1996)
The long-term objective of this research is to define how cells in a tissue within the body interact with one another and what happens to these interactions when some of the cells become cancerous. Several years ago my laboratory discovered a novel protein which has the potential to play several roles in the development of cancer. This protein, called TIMP-3, is produced in significant amounts at various stages in the development of a fetus but not by most normal cells in an adult. Production of this protein appears to be switched on again during the development of cancer. We propose that TIMP-3 may be a useful marker for breast cancer.

TIMP-3 has several functions. It is capable of helping to prevent the destruction of structural components of tissues which sometimes accompanies the development of a tumor. On the other hand, it is also capable of stimulating cells to divide and increase in number. Furthermore, it encourages cells to break away from one another. Both of the last two properties facilitate the development of a tumor, whereas its protective effect on tissue architecture would have the opposite effect. One aim of the present grant application is to unravel this apparent paradox and to determine the exact role that TIMP-3 plays in cancer progression. This will be accomplished by use of different types of cells from the human breast that will grow under laboratory conditions.

Cancer is a multi-stage process and cells which display properties characteristic of five different stages of the disease (from early to late) will help us to determine at what stage the production of TIMP-3 may be important. Regardless of its precise role in this process, we already know the synthesis of TIMP-3 is turned on early in the conversion of normal cells to cancer cells derived from animal tissues other than breast. We have also studied sections of human breast tissue from women with malignant and benign disease and those who have undergone surgery for breast reduction. Using an immunological reagent (antibody) which binds specifically to TIMP-3, we have probed 26 samples for the presence of this protein. All malignant cancers (17 cases) contained TIMP-3, whereas none of the normal tissues or benign tumors (9 cases) showed the presence of this protein. A second major aim of this proposal is to study a very large number of pathology specimens encompassing a wide spectrum of breast disease to determine if the presence of TIMP-3 can be used as a diagnostic tool for a pre-malignant or early stage of the disease in tissue biopsies.

A particularly interesting observation that we have made is that TIMP-3 is present in breast fluid within apparently normal ducts adjacent to invasive tumors. This raises the exciting possibility of monitoring nipple aspirates for the presence of TIMP-3. Nipple aspiration is a relatively simple, non-invasive means of obtaining small volumes of fluid from women who are neither pregnant nor lactating. We have developed a very sensitive test for measuring the activity of the protein and have demonstrated the feasibility of this approach by showing that nipple aspirates from healthy women volunteers (6 studied) contain predominantly TIMP-1, some TIMP-2 (related proteins), but no TIMP-3. We predict that TIMP-3 activity will be detectable in nipple aspirates from women with invasive cancer and possibly earlier stages of the disease as well. Thus, a third aim of this application is to screen nipple aspirates from healthy women, those at particular risk for developing cancer and those who already have evidence of malignant and benign tumors. We propose to determine if the presence of TIMP-3 in this biological fluid correlates with a particular diagnosis or stage of breast cancer. This, we believe, could provide a means of detecting breast cancer at an early stage employing a novel biomarker in a very sensitive, non-invasive and inexpensive assay.


Final Report (1998)
TIMP-3 is a protein that helps to prevent enzymatic destruction of tissue components. Most normal adult cells produce little, if any, TIMP-3. Its synthesis is stimulated during fetal development and carcinogenic transformation in cell culture, where it also stimulates cells to divide and to break away from their extracellular matrix. The Specific Aims for this project are to: 1) evaluate polyclonal antibodies for localizing TIMP-3 in sections of human breast tissue; 2) determine the role that the protein plays in progression of breast cancer; and, 3) evaluate TIMP-3 as an early marker for breast cancer.

We have screened all antisera and developed a strategy for purifying antibody that does not cross react with other tissue proteins. This antibody is particularly useful because it recognizes TIMP-3 in material that has been chemically treated for long term storage, We have genetically engineered a cell culture system to produce large quantities of TIMP-3 that can be used to provide a superior method of antibody purification. In order to study TIMP-3 production in three-dimensional models of breast cancer progression we have developed sensitive confocal microscopy techniques that can detect TIMP-3 in extremely small quantities. The protein is not detectable in normal or benign breast tissue sections by immunohistochemical techniques and light microscopy, but can be detected in tissue containing carcinoma. In cancerous tissues TIMP-3 is uniquely localized in the basement membrane (a set of proteins that provide support for cells) surrounding ducts that contain precancerous cells. It is also located in basement membrane surrounding ducts that look normal but are adjacent to invasive carcinomas. Faint traces of TIMP-3 are found in cells of "early" cancers (ductal carcinoma in situ). All samples of carcinoma in situ analyzed to date contain TIMP-3, whereas approximately half of all invasive carcinomas are TIMP-3 negative.