these conditions. Transcriptional alterations of above mentioned test genes were not induced by TGFb exposure of grade 3 cell lines. In contrast to these effects, treatment of 2 independent grade 1 tumor cell lines with the TGFb inhibitor – SB431542, induced up to PBTZ 169 web 2-fold increase in BST2 . Clinical Outcome Analysis A publicly available primary breast cancer microarray dataset GSE4922 was used for evaluating clinical outcome of BST2 overexpressing patient subsets. The Kaplan-Meier estimates were used to plot survival curves, and the p-value from a log-rank test used to determine the statistical significance of hazard ratios. Disease-free survival was defined as the time interval from surgery until the recurrence or last date of follow up. All survival statistics were performed in the R package. Site-specific Binding of AP2 to the BST2 Promoter Towards a direct analysis of BST2 regulation, we first evaluated the 240 bp region of the human BST2 promoter and exon 1 by Vector NTI Suite 17149874 software and detected the presence of AP2 and STAT binding sites within this region. Since QPCR analysis showed more than 3-fold induction of AP2 expression by TGFb within responsive low grade tumor cells, whereas only marginal to no change was detected in STAT3, and STAT1 transcript levels, thus further analysis of AP2 binding was prioritized. ChIP assays were carried out using anti AP2, and promoter-specific oligonucleotides. Strong AP2 binding to the BST2 promoter was observed in 5/5 grade 1 & 2 tumor cell lines whereas no binding was detected in 2/2 grade 3 lines, and in MCF7 cells. QPCR analysis of ChIP-derived DNA from all test samples confirmed gel based findings of AP2 recruitment to the BST2 promoter. Finally, a direct role for TGFb Results Differential Expression of BST2 in Primary Breast Cancer Immunoperoxidase localization of anti BST2 on 3 tissue arrays comprised of 234 primary breast tumor cores demonstrated strong membrane staining in the majority of intermediate and high grade tumors, while minimal to no immunostaining was observed in low grade 10712926 tumors. BST2 immunolocalization in tissue arrays was verified in full-sized sections of tumor tissue blocks in several cases. BST2 positive primary tumors with a concurrent DCIS component were found to display strong staining in both pre invasive and invasive tumor cells. In our previous expression profiling study of 14 novel primary breast cancer cell lines of varying histological grade developed in our laboratory, TGFb Mediated BST2 Repression 4 TGFb Mediated BST2 Repression effect was not detectable by ChIP likely due to differences in assay sensitivity. However, AP2 binding to the BST2 promoter was effectively disrupted in the presence of the TGFb inhibitor in all test cell lines. Taken together, these observations suggest a model wherein AP2 binding to the BST2 promoter triggers gene suppression and serves as the basis of differential BST2 expression in low vs. high grade tumors. Induction of Apoptosis and Inhibition of Cell Proliferation Promoted by Loss of BST2 Expression To evaluate its functional role in breast cancer, BST2 overexpressing cells were transfected with BST2 siRNA and evaluated as 3-dimensional Matrigel cultures. Reduction in BST2 expression was confirmed both by QPCR and immunofluorescence. As measured by immunofluorescence localization of anti cleaved caspase 3, a decline in BST2 levels detectably enhanced apoptotic cell death induced by 2 independent pro apoptotic drugs – 4-hydo