Intracellular Signaling by PI3K in Breast Cancer Metastasis

Institution: University of California, Davis
Investigator(s): Earl Sawai, Ph.D. -
Award Cycle: 1998 (Cycle IV) Grant #: 4IB-0099 Award: $43,329
Award Type: IDEA
Research Priorities
Biology of the Breast Cell>Pathogenesis: understanding the disease



Initial Award Abstract (1998)
The spread of breast cancer to distant sites in the body is the most deadly of the many events in tumor progression. Much is known about specific surface receptors and adhesion events in metastasis, and our understanding of angiogenesis is creating clinical opportunities. However, there is a lack of specific information about the intracellular signaling events that potentially regulate metastasis. Signaling proteins inside the cell are the links between the genetic abnormalities (i.e., expression of oncogenes or loss of tumor suppressor genes) that characterize breast cancer and behavior (phenotype) of cancer cells (i.e., whether a certain tumor metastasizes). It may not be possible to correct specific genetic defects through gene therapy or other emerging technologies, but the metastatic phenotype could still be altered by targeting the regulatory factors operating within the cell.

Our focus is on a specific signaling molecule, called phosphatidylinositol 3-kinase (PI3K), a cell membrane-associated protein which is closely linked to critical growth factor receptors (e.g., Her-2). Our hypothesis is that PI3K connects uncontrolled tumor growth to the cell processes necessary for metastasis. We have developed a unique model system that allows us to study breast cancer metastasis. Using genetically altered breast cancer cell lines (called Met-1/Db-7) that vary in their metastatic potential when implanted into mice, we will determine whether activation of PI3K is critical for metastasis. This project will use a combined genetic, biochemical, and cell biology approach that will allow us to dissect the role of PI3K in metastasis. Our specific goals are to (i) determine whether the difference in metastatic potential of Met-1 and Db-7 is due to alterations in PI3K activation, (ii) determine whether disruption of the PI3K activation pathway affects metastatic and non-metastatic phenotypes in cell culture using either inactive or active mutants of PI3K, and (iii) analyze the metastatic potential of the tumor lines that inducibly express either the active or inactive mutants of PI3K in animals. The payoff is that we can compare the effects of different PI3K mutants on the ability of cancer cells to spread in the body.

We believe that further study of the pathways that control metastatic progression should facilitate both the identification of diagnostic markers and the development of novel therapeutic compounds that inhibit tumor dissemination.


Final Report (2000)
Tumor metastasis is a major cause of morbidity and mortality in breast cancer patients. To investigate the changes in intracellular signaling pathways involved in metastais we used a transgenic mouse mammary tumor model. This model system allows a comparative analysis of two distinct breast tumor lines, designated Met- 1 and Db-7, that have high and low metastatic potential, respectively. The aim of our study was to compare the role of a cellular signaling molecule, called phosphatidylinositol-3'-kinase (PI3K). The highly metastatic Met-1 tumors have PI3Kin an 'activated' form, whereas the low metastatic Db-7 tumors do not 'activate' PI3K.

We surveyed several key elements associated with growth/metastasis in the two tumor cell lines. First, we found that a marker of human breast cancer metastasis, osteopontin (OPN), is expressed in the highly metastatic Met-1 tumors, but much less so in the Db-7 tumors. OPN is a secreted growth factor that has been associated with several metastatic cancers in humans. We have determined that OPN is indeed secreted in the Met- 1 tumors. Secondly, because the cellular adhesion molecule CD44 has also been associated with metastatic breast tumors and has been reported to bind OPN, we investigated whether CD44 is differentially expressed in the Met- 1 and Db-7 tumors. But, we determined that there is no difference in the expression of CD44 in Met-1 and Db-7 tumors. Finally, we have also investigated whether signaling pathways that are related to PI3K activation are also affected. We have found that a cellular serine-threonine kinase in the p21-activated kinase pathway, PAK1, is differentially activated in the metastatic Met- 1 tumors compared with the Db-7 tumors. PAK proteins are critical for influencing cytoskeletal protein organization and cellular morphology. Consequently, the activation of this pathway may be important for metastatic progression by increasing tumor cell locomotion and mobility. Interestingly, the activation of PAK is most dramatic in serum starved cells.

In follow-up studies to this CBCRP funding, we are currently investigating the linkage between OPN expression and PAK activation. The identification of signaling molecules that are linked to metastatic progression will be critical for the development of new therapeutic approaches to prevent the progression of metastatic breast tumors as well as identify markers that would facilitate the detection and diagnosis of metastatic breast cancer.