SUN-LB135 Estrogen Receptor Beta Inhibits NFkB Signaling to Inhibit Triple Negative Breast Cancer

Abstract

Triple Negative Breast Cancer (TNBC) affects approximately 15-20% of BC patients, yet accounts for a disproportionately higher rate of BC morbidity and mortality, in part due to lack of targeted therapies. We have shown that estrogen receptor beta (ERβ) is expressed in approximately 20% of TN breast tumors and that ligand-mediated activation of ERβ with estradiol (E2) or ERβ-selective agonists decreases tumor cell proliferation, invasion and migration in vitro and in vivo. Therefore, we aimed to elucidate the mechanisms by which ERβ elicits its anti-cancer effects in TNBC. RNAseq analysis of ERβ-expressing MDA-MB-231 cells demonstrated that ERβ significantly downregulates NFκB signaling in the presence of E2. ChIPseq for ERβ in these cells revealed that ERβ primarily associated with estrogen response elements, but 12% of all ERβ binding sites were located at NFκB consensus motifs. Using an NFκB reporter construct and qPCR, ERβ was shown to block TNFα-mediated induction of NFκB signaling and NFκB target gene expression. RNAseq analysis of MDA-MB-231-ERβ cells treated with TNFα or E2+TNFα revealed substantial global inhibition of TNFα regulated genes in the presence of E2. ChIPseq for NFκB demonstrated that ERβ significantly alters NFκB’s cistrome whereby it can both diminish NFκB binding and redistribute NFκB throughout the genome. ChIPseq also demonstrated that ligand-mediated activation of ERβ significantly diminished an activating histone mark (H3K27Ac) at many of these NFκB target genes while enhancing a repressive mark (H3K27Me3). The addition of H3K27Me3 at these loci was shown to occur through the recruitment of the histone methyltransferase, EZH2. Drug-mediated blockade of EZH2 activity reversed suppression of NFκB target gene expression by ERβ. Knockdown of NFκB or Mutation of ERβ’s DNA binding domain rendered ERβ incapable of associating with DNA, recruiting EZH2, methylating NFκB target gene loci, repressing NFκB target gene expression and inhibiting proliferation. Interestingly, ERβ was shown to elicit more potent anti-cancer effects in TNBC cells expression a constitutively active form of NFκB. These finding suggest that a primary mechanism by which ERβ functions as a tumor suppressor is through inhibition of NFκB pathway activity. Our studies have also revealed that ERβ functions as a molecular switch for EZH2 and repurposes it for tumor suppressive activities, as EZH2 has previously been reported to enrich NFκB signaling in TNBC. These findings could address the paradox that high EZH2 expression is associated with worse TNBC patient outcomes, while high H3K27Me3 expression is associated with improved patient outcomes. Currently, a Mayo Clinic Breast Cancer SPORE prospective phase II clinical trial is underway to investigate the efficacy of estradiol for the treatment of metastatic ERβ+TNBC and to further evaluate the cross-talk between ERβ, EZH2 and NFκB signaling.

Ensemble learning model for identifying the hallmark genes of NFκB/TNF signaling pathway in cancers

“…These findings suggest that patients with larger weights in the model may be sensitive to ER and PR status and the NFκB/TNF pathway dysregulation. This conclusion is supported by previous studies indicating the mutual suppression between NFκB/TNF and ER/PR in breast cancer [ 60 – 63 ]. Moreover, the ER + /PR + patients receiving endocrine therapy could have a better prognosis by repressing NFκB [ 60 – 63 ].…”

“…This conclusion is supported by previous studies indicating the mutual suppression between NFκB/TNF and ER/PR in breast cancer [ 60 – 63 ]. Moreover, the ER + /PR + patients receiving endocrine therapy could have a better prognosis by repressing NFκB [ 60 – 63 ]. This recapitulates the association between TNBC and the NFκB/TNF pathway.…”

“…To further show the model's applicability in cancer medicine, we selected triple-negative breast cancer (TNBC) as a case study. This selection was motivated by the fact that TNBC has been reported to be strongly associated with the dysregulation of NFκB/TNF [ 60 – 63 ]. Within the ensemble learning model, patients bearing larger absolute weights have more influence on predicting NFκB/TNF hallmark genes, indicating a potential connection to the NFκB/TNF pathway.…”