Project 1 - Epigenetic Control of Normal and Malignant Neuroendocrine Differentiation

Dana-Farber Cancer Inst, Boston MA

Investigators

Abstract

PROJECT SUMMARY (Project 1) Neuroendocrine (NE) is the common theme in selected epithelial cancers that are especially difficult to treat, including small cell lung cancer (SCLC), small intestine NE tumors (SI-NETs, previously called carcinoid), Merkel cell carcinoma (MCC), and treatment-emergent NE prostate cancer (NEPC). Despite different tissue origins, these cancers share a convergent cis-regulatory program of NE differentiation and, frequent genetic loss of RB and TP53 tumor suppressor functions. It remains unclear whether these cancers represent transformation of resident NE cell precursors or if certain oncogenic pathways specifically capture the inherent potential of multipotent cells for NE differentiation. Notably, NE cancers (NECs) and their respective normal cell counterparts show highly specific expression and dependency on a core circuit of basic-helix- loop-helix (bHLH) and other transcription factors (TFs), including ATOH1, INSM1, ASCL1, and NEUROD1. The premise of Project 1 is that a deep understanding of this core transcriptional circuitry and its intersections with common oncogenic and tumor suppressor pathways will reveal new avenues to treat NE cancers as a group, irrespective of their tissue origins. Two fundamental advances underlie this goal. First, we have developed a novel human cellular system to study normal differentiation trajectories of enteroendocrine cells (EECs), the dominant NE cell population in intestinal epithelium and the cell type that resembles both SI- NETs and high-grade NE cancers (HGNC). In this unique experimental model of authentic NE differentiation, cells pass through stages characterized first by oscillating ASCL1 expression and ASCL1-bound enhancers, brief co-expression of ASCL1 and NEUROD1, and finally NEUROD1 expression in pre-terminal EECs. Primary human SI-NETs show RNA and chromatin profiles that imply cellular arrest at a NEUROD1+ post- ASCL1 stage. Second, NE phenotypes across different cancers reflect a convergent chromatin state, characterized by shared enhancers that bind core NE-specific TFs ASCL1 and/or NEUROD1. Treatment- emergent NEPCs fall in two groups, characterized by high ASCL1 or high NEUROD1 expression and activation of the corresponding transcriptional programs; primary prostate tumors contain both NEPC cell types, while metastases are enriched for one or the other. Building on the parallel models and findings, Aim 1 dissects intestinal NE circuitry in fine detail, interrogates specific TF functions, and tests the hypothesis that ASCL1 oscillation is a critical bottleneck for EEC quality control. Aim 2 asks how the most frequent genetic defects in SI-NET (CDKN1B mutation) and HGNC (RB and TP53 loss) impinge on the NE differentiation network. Together, this work will define, at single-cell resolution, the regulatory logic of normal NE differentiation and how cancer mutations subvert that logic. Aim 3 explores epigenetic changes that drive NE phenotypes in androgen receptor-independent NEPC and will define, in mechanistic terms, the alternative dependencies that these treatment-emergent cancers acquire.