DAMD 17 - 02 - 1 - 0388 TITLE : X - Box Binding Protein - 1 in Breast Cancer PRINCIPAL INVESTIGATOR :

RESISTANCE TO ENDOCRINE THERAPY IN BREAST CANCERKerrie Bouker, Yuelin Zhu, Rebecca Riggins, Alan Zwart, Ruchi Nehra, BiancaGomez, and Robert ClarkeLombardi Comprehensive Cancer Center, Department of Oncology, GeorgetownUniversity, Washington D.C. 2007clarker@ georgetown.eduThe precise mechanisms of resistance to endocrine therapies are unclear. Antiestrogen-stimulated growth can occur but is a minor phenotype when compared with antiestrogenunresponsiveness. Absence of estrogen receptor (ER) expression is the most common denovo resistance mechanism but is uncommon in acquired resistance. We have developeda series of variant cell lines that exhibit differences in their responsiveness to estrogensand antiestrogens. Using these variants and both genomic and proteomic approaches, wehave begun to test our hypothesis that endocrine responsiveness is conferred by acomplex signaling network that exhibits components of ER-dependent and ER-independent signaling, protein interactivity, and signaling redundancy. Genomic (gene expression microarray and serial analysis of gene expression) studiesidentified several candidate genes including interferon regulatory factor-1 (IRF-1),nuclear factor kappa B (NFκB), and the human X-box binding protein-1 (hXBP-1).Proteomic (2-dimensional gel electrophoresis) analyses implicated nucleophosmin(NPM). We have observed changes in both basal expression and the ability of ER-mediated events to regulate the expression of these genes. Our recent mechanistic studiesstrongly implicate signaling through IRF-1, and its two protein partners NPM and NFκB,in functionally affecting the ability of breast cancer cells to induce apoptosis in responseto antiestrogens. For example, the proapoptotic effects of ICI 182,780 (Faslodex;Fulvestrant) are blocked by a dominant negative IRF-1 (dnIRF-1), a function alsoperformed by endogenous NPM expression. We show that IRF-1 acts as a tumorsuppressor gene in breast cancer: overexpression reduces tumorigenicity in nude micewhile dnIRF-1 increases tumorigenicity. A small molecule inhibitor of NFκB(parthenolide) restores sensitivity to ICI 182,780 in resistant cells. Overexpression ofhXBP-1 confers resistance to both ICI 182,780 and Tamoxifen. We detect each of these proteins in breast tumor tissue microarrays (all are detectable in>50% of breast cancers), often in patterns consistent with our network. For example,NFκB may regulate hXBP-1 expression and these are coexpressed (p=0.018); an inverserelationship is seen between the prosurvival NFκB and the tumor suppressor IRF-1(p=0.034). Thus, our mechanistic and translational studies are consistent with a novelgene expression network, regulated by ER-mediated events, that affects signaling toapoptosis. Network components may be useful as biomarkers and/or molecular targets. Award Number: DAMD17-02-1-0388 Pg. 18Robert Clarke, Ph.D., D.Sc. X-BOX BINDING PROTEIN-1 IN BREAST CANCERBianca Gomez, Yuelin Zhu, Rebecca Riggins, Alan Zwart and Robert ClarkeLombardi Comprehensive Cancer Center, Department of Oncology, GeorgetownUniversity, Washington D.C. 2007clarker@ georgetown.eduIn studies of acquired Tamoxifen (TAM) and Faslodex (ICI 182,780) resistance, weidentified a member of the ATF/CRE family, human X-Box binding protein-1 (hXBP-1),as a potential contributor to endocrine resistance and cell survival in breast cancer. XBP-1 activates genes with cyclic AMP responsive elements (CRE) and is directly implicatedin both protein refolding and degradation in endoplasmic reticulum stress responses andas an antiapoptotic factor. Some published gene microarray studies have implicatedhXBP-1 as being associated with estrogen receptor (ER)-positive tumors. Thus, we havebegun to establish the role of hXBP-1 in breast cancer. We first introduced the hXBP-1 cDNA into MCF7 cells, which are antiestrogen sensitiveand estrogen dependent for growth in vitro and in vivo. Cells were transfected with apcDNA 3.1 expression vector (Invitrogen) containing the XBP-1 cDNA (MCF7/hXBP-1). The empty vector (same construct but without the XBP-1 cDNA) also was transfectedinto MCF-7 cells to generate control cell populations. We assessed whetheroverexpression of hXBP-1 could confer on MCF-7 cells the ability to grow in vitro in theabsence of estrogenic supplementation. The data show that hXBP-1 enabled the cells tosurvive and proliferate in the absence of estrogens, whereas the control cells did notproliferate and began to die between days 3 and 6. In the absence of estrogens, 6% ofMCF-7 cells are in S-phase, compared with almost 20% of the MCF7/hXBP-1 cells;showing an effect of hXBP-1 on cell cycle transition. The ability to maintain estrogen-independent growth would be expected to confer resistance to an aromatase-inhibitor,although not necessarily to an antiestrogen. The antiestrogen crossresistance phenotypeof MCF7/hXBP-1 cells is reported elsewhere at this meeting. While the precisemechanisms of resistance in these cells are under investigation, consistent with thepresence of a CRE in the aromatase gene promoter, MCF7/hXBP-1 cells expressincreased rates of aromatase mRNA transcription. Thus, hXBP-1 may be a key regulatorof this gene in breast cancer. We also have confirmed that hXBP-1 binds to ER. To determine expression of hXBP-1 in breast tumors, we used immunohistochemicalanalysis of breast tumor tissue microarrays. We found moderate or strong hXBP-1 in79% of breast tumors but could not confirm its coexpression with ER as reported in genemicroarray studies. However, we did detect a significant coexpression of hXBP-1 withNFκB p65 (p=0.018; Spearman’s rank correlation analysis). This observation isconsistent with evidence from other cell systems implicating NFκB in the transcriptionalregulation of hXBP-1 expression. These data provide evidence of an important role forhXBP-1 in breast cancer. Award Number: DAMD17-02-1-0388 Pg. 19Robert Clarke, Ph.D., D.Sc.