Insulin receptor substrates (IRS) are central integrators of hormone, cytokine, and growth factor signaling. IRS proteins can be phosphorylated by a number of signaling pathways critical to normal mammary gland development. Studies in transgenic mice that overexpress IGF-I in the mammary gland suggested that IRS expression is important in the regulation of normal postlactational mammary involution. The goal of these studies was to examine IRS expression in the mouse mammary gland and determine the importance of IRS-1 to mammary development in the virgin mouse. IRS-1 and -2 show distinct patterns of protein expression in the virgin mouse mammary gland, and protein abundance is dramatically increased during pregnancy and lactation, but rapidly lost during involution. Consistent with hormone regulation, IRS-1 protein levels are reduced by ovariectomy, induced by combined treatment with estrogen and progesterone, and vary considerably throughout the estrous cycle. These changes occur without similar changes in mRNA levels, suggesting posttranscriptional control. Mammary glands from IRS-1 null mice have smaller fat pads than wild-type controls, but this reduction is proportional to the overall reduction in body size. Development of the mammary duct (terminal endbuds and branch points) is not altered by the loss of IRS-1, and pregnancy-induced proliferation is not changed. These data indicate that IRS undergo complex developmental and hormonal regulation in the mammary gland, and that IRS-1 is more likely to regulate mammary function in lactating mice than in virgin or pregnant mice.

Genomic instability is a hallmark of human cancers. Spindle assembly checkpoint (SAC) is a critical cellular mechanism that prevents chromosome missegregation and aneuploidy by blocking premature separation of sister chromatids. The process of sister-chromatid separation occurs when separase cleaves Scc1 of the cohesin complex holding sister-chromatids together. The cell prevents this event from taking place untimely by establishing regulatory mechanisms on separase. Thus, the SAC, much like the DNA damage checkpoint, is essential for genome stability.Although a large body of evidence supports an intimate relationship between defects in the DNA damage checkpoint and tumorigenesis, such evidence is lacking for the spindle assembly checkpoint. Attempts to disable SAC by inactivating individual component, such as Mad2, or BubR1, yielded only unviable mice.Securin was the first identified inhibitor of separase in budding yeast as PDS1 (1, 2). Deletion of PDS1 caused genome instability and lethality when treated with nocodazole (2). Based on these results, it was hypothesized that deletion of the securin gene would accelerate tumorigenesis in mammals through inducing aneuploidy.However, when securin was knocked out in mice, the mutant mice were essentially normal and the cells derived from these mice show an intact SAC (3, 4). Although, securin null cells are compromised in their ability to grow in the presence of sub-lethal doses of nocodazole (4), indicating that securin does contribute to the proper functioning of the checkpoint. The over-expression of securin has been associated with metastasis in human breast cancer and therefore my securin knockout mice have not developed tumors.Apparently securin gain-of-function or loss-of-function does not have strong enough oncogenic activity to initiate tumor development, but perhaps in combination with a stronger oncogene tumors will develop faster and more aggressively. To test this, I am combining securin knockout with MMTV-ErbB2 transgenic mice, which develop adenocarcinomas in the mammary tissue by approximately 6 months old (5).Given the association of securin expression with metastasis, it is possible that securin may have a role in tumor cell metastasis. To test that possibility, I am taking advantage of the MMTV-ErbB2 mice again as these mice show lung metastasis by approximately 8 months old (5). Therefore, my hypothesis is that loss of securin facilitates tumor formation by destabilizing the genome as a result of mitotic errors, but may hinder metastasis because of its cell cycle unrelated functions.References:1. Yamamoto, A., Guacci, V., and Koshland, D. (1996a). J. Cell Biol. 133, 85–97.2. Yamamoto, A., Guacci, V., and Koshland, D. (1996b). J. Cell Biol. 133, 99–110.3. Mei, J., Huang, X., and Zhang, P. (2001). Curr. Biol. 11, 1197–1201.4. Huang, X., Hatcher, R., York, P., Zhang, P.(2005). Mol. Bio. of the Cell. 16, 4725–4732.5. Muller WJ; Sinn E; Pattengale PK; Wallace R; Leder P. 1988. Cell 54(1):105-15. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 6172.

Selective estrogen receptor down-regulators (SERDs), such as ICI 182780 (Fulvestrant), comprise a class of drugs that has been developed to inhibit estrogen receptor alpha (ER) activity in breast cancer. ICI is a pure antagonist that competitively binds to the ligand-binding domain of ER. The binding of ICI is thought to reduce the receptor's half-life by increasing protein turnover. However, the exact mechanism of ICI action, including whether or not receptor degradation is actually required for its antiestrogenic action, is currently unknown. Our laboratory has shown that a cytoplasmic ER mutant (ERΔNLS) lacking a nuclear localization sequence (NLS) is completely resistant to ICI-mediated degradation in both MCF-7 breast cancer cells and their ER-negative derivatives, C4-12 cells. Immunofluorescence and confocal microscopy were performed to assess localization of all ER proteins. These cells were also treated for different lengths of time with varying concentrations of both estradiol (E2) and ICI. Immunoblotting was performed to determine levels of ER protein. A competitive radioligand binding assay was also conducted to assess the ligand-binding capability of the ERΔNLS mutant receptor. Immunofluorescence and confocal imaging of stably transfected C4-12 cells has confirmed distinct subcellular localization of each ER protein. Wild-type ER localizes mainly to the nucleus, while ERΔNLS is cytoplasmic. As expected, treatment of either wild-type MCF-7 cells or C4-12 cells expressing wild-type ER with E2 or ICI caused degradation of ER protein. Treatment of MCF-7/ ERΔNLS or C4-12/ ERΔNLS cells with E2 caused degradation of ERΔNLS protein. In stark contrast, increasing concentrations of ICI failed to cause degradation of ERΔNLS. Furthermore, experiments performed in the presence of cycloheximide have shown that ICI does not modulate ERΔNLS half-life. The inability of ICI to degrade ERΔNLS is not due to an inability of ERΔNLS to bind ligand, since results from competitive radioligand binding assays show that the affinity of ERΔNLS for both E2 and ICI is equal to the affinity of wild-type ER for both ligands. We have recently cloned the SV40-NLS onto ERΔNLS, and confocal microscopy has confirmed that the SV40-NLS is sufficient to force ERΔNLS into the nucleus. I am currently using this construct to determine if it is the NLS domain itself or merely subcellular localization of ER that is important for ICI-mediated degradation. In conclusion, ER is a critical marker for response to antiestrogen therapy. Treatment with ICI is not without its shortcomings. A better understanding of the mechanism by which ICI degrades ER may lead to the development of better antiestrogen therapies to inhibit ER action with less toxicity and greater specificity. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 4137.