Pathogenic mechanisms of sinonasal sarcoma, a novel gender dimorphic cancer

点击翻译按钮获取中文摘要

PROJECT SUMMARY Bi-phenotypic sinonasal sarcoma (SNS) is a newly identified sarcoma that affects women three times more frequently than men. The only treatment for this highly invasive lesion is disfiguring facial surgery. We recently identified a novel fusion in SNS, which creates a novel chimera PAX3-MAML3 that fuses the DNA-binding domain of the PAX3 transcription factor with the Mastermind-like 3 (MAML3) transcriptional co-activator. The transcriptional program that PAX3-MAML3 triggers to elicit malignant transformation and invasive growth, and the basis of the gender dimorphism of SNS is completely uncharacterized. Preliminary data demonstrate that expression of PAX3-MAML3 is sufficient to drive formation of tumors that histologically mimic SNS, and to induce gene expression patterns observed in SNS. Notably, PAX3-MAML3 induced expression of estrogen receptor β (ERβ) and phosphorylation of ERα, and stimulated ER activity. Furthermore, tumors formed by xenografted PAX3-MAML3-expressing cells were significantly larger in female than male recipient mice. Transcriptome analysis of primary SNS tumors confirmed activation of an estrogen response signature in vivo. These results lead us to hypothesize that the ER plays an essential role in PAX3-MAML3's pathogenic activity in SNS, and that this underlies the gender dimorphic nature of this cancer. We further posit that anti- estrogens might serve as a complementary or even alternative approach to disfiguring surgery for the treatment of this highly invasive cancer. This will be tested through the following: Aim I: Determine role of ER in PAX3-MAML3 pathogenesis in vitro and in vivo We will examine the effects of ER inactivation, using CRISPR and anti-estrogenic agents (i.e. tamoxifen, fulvestrant, and aromatase inhibitor), on multiple aspects of PAX3-MAML3-induced transformation: (A) In vitro, we will monitor effects on cell proliferation and survival, anchorage-independent growth, and invasiveness. (B) In vivo, we will test the contribution of ER and estrogen signaling to PAX3-MAML3-mediated tumorigenesis. We will assess the effects of ER inactivation (by CRISPR- mediated depletion of ERα/ERβ or by treatment of mice with anti-estrogenic agents), comparing tumor growth in male vs. female mice. Aim II: Determine how PAX3-MAML3 and ER coordinate gene expression to drive invasive growth We will: (A) define the transcriptional program induced by PAX3-MAML3 by RNA-seq, and determine how it is modulated by ER (through use of ER agonists, and CRISPR-mediated deletion of ERα/β); and (B) define the PAX3-MAML3 and ER chromatin occupancy landscapes to gain a mechanistic understanding of how PAX3-MAML3 and ER coordinate this transcriptional program. Transcriptome and ChIP- seq data sets will be intersected to distinguish direct from indirect target genes of PAX3-MAML3 and ER, and to gain insight into cooperative effects on transcription. A longterm goal of this integrated analysis is to facilitate identification of PAX3-MAML3/ER targets and their downstream pathways that are essential for the transformed, invasive phenotype, and may ultimately serve as additional therapeutic targets. !