ASCL1 regulates and cooperates with FOXA2 to drive terminal neuroendocrine phenotype in prostate cancer
Shaghayegh Nouruzi, Takeshi Namekawa, Nakisa Tabrizian, Maxim Kobelev, Olena Sivak, Joshua M Scurll, Cassandra Jingjing Cui, Dwaipayan Ganguli, Amina Zoubeidi
Shaghayegh Nouruzi, Takeshi Namekawa, Nakisa Tabrizian, Maxim Kobelev, Olena Sivak, Joshua M Scurll, Cassandra Jingjing Cui, Dwaipayan Ganguli, Amina Zoubeidi
View: Text | PDF | Corrigendum
Research Article Cell biology

ASCL1 regulates and cooperates with FOXA2 to drive terminal neuroendocrine phenotype in prostate cancer

Abstract

Lineage plasticity mediates resistance to androgen receptor pathway inhibitors (ARPIs) and progression from adenocarcinoma to neuroendocrine prostate cancer (NEPC), a highly aggressive and poorly understood subtype. Neuronal transcription factor ASCL1 has emerged as a central regulator of the lineage plasticity driving neuroendocrine differentiation. Here, we showed that ASCL1 was reprogrammed in ARPI-induced transition to terminal NEPC and identified that the ASCL1 binding pattern tailored the expression of lineage-determinant transcription factor combinations that underlie discrete terminal NEPC identity. Notably, we identified FOXA2 as a major cofactor of ASCL1 in terminal NEPC, which is highly expressed in ASCL1-driven NEPC. Mechanistically, FOXA2 and ASCL1 interacted and worked in concert to orchestrate terminal neuronal differentiation. We identified that prospero homeobox 1 was a target of ASCL1 and FOXA2. Targeting prospero homeobox 1 abrogated neuroendocrine characteristics and led to a decrease in cell proliferation in vitro and tumor growth in vivo. Our findings provide insights into the molecular conduit underlying the interplay between different lineage-determinant transcription factors to support the neuroendocrine identity and nominate prospero homeobox 1 as a potential target in ASCL1-high NEPC.

Authors

Shaghayegh Nouruzi, Takeshi Namekawa, Nakisa Tabrizian, Maxim Kobelev, Olena Sivak, Joshua M Scurll, Cassandra Jingjing Cui, Dwaipayan Ganguli, Amina Zoubeidi

×

Figure 4

ASCL1 and FOXA2 coregulate PROX1.

Options: View larger image (or click on image) Download as PowerPoint
ASCL1 and FOXA2 coregulate PROX1. (A) Dot plot shows the fold-change in ...
(A) Dot plot shows the fold-change in transcript abundance of FOXA2:ASCL1 cobound genes, x axis comparing NCI-H660 (ASCL1+FOXA2+) versus 42DENZR (ASCL1+FOXA–) cell lines and y axis comparing LuCaP 145-1 (ASCL1+FOXA2+) versus LuCaP 93 (ASCL1+FOXA2–) PDXs (left panel). Top upregulated TFs were ranked based on the highest log2 fold-change expression (right panel). (B) IGV tracks show ASCL1, FOXA2, H3K27ac, and H3K27me3 binding at the PROX1 promoter in NCI-H660. (C) The correlation between PROX1 expression with FOXA2 and ASCL1 in patient datasets (3, 9), with each dot representing a patient tumor and the color reflecting PROX1 mRNA expression; P < 0.05, with significance assessed by 2-tailed unpaired t test. (D) PROX1 expression in patient datasets (3, 9), with significance assessed using 2-tailed unpaired t test. Bonferroni’s correction was applied to adjust for multiple comparisons. (E) Relative mRNA expression of ASCL1, FOXA2, and PROX1 in NCI-H660 (CTL) following ASCL1 KD. Data are reported relative to CTL, mean ± SD; with significance assessed by 2-tailed unpaired t test. (F) Relative mRNA expression of ASCL1, FOXA2, and PROX1 in NCI-H660 (CTL) following FOXA2 KD. Data are reported as described in E. (G) Relative mRNA expression of ASCL1, FOXA2, and PROX1 normalized to GAPDH in 16DCRPC (CTL) and following OE of ASCL1 at 10 and 21 days using a Dox-inducible system. Data are reported relative to day 0, and significance is assessed using a 2-tailed unpaired t test. (H) Relative mRNA expression of ASCL1, FOXA2, and PROX1 in 16DCRPC (CTL) and following OE of FOXA2, with significance assessed by 2-tailed unpaired t test. (I) Identification of significant dependencies in the SCLC ASCL1 subtype using a loss-of-function CRISPR/Cas9 screen. Genes within the shaded red region are identified as significant dependencies with a dependency score of < –0.5.