FOXJ1, a novel oncosupressor transcription factor in a subgroup of PDAC patients

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is one of the most prevalent and deadliest cancers. It is predicted to become the second leading cause of cancer mortality within a decade, with an overall 5-year survival of 9%. The existing literature extensively supports the notion that epigenetic modifications and alterations in transcriptional regulation influence the development of PDAC. Therefore, understanding and deciphering epigenetic mechanisms implies the possibility of interfering with relevant processes and developing new therapies. Here, I combined chromatin accessibility (ATAC-seq) and enrichment for H3K27 acetylation measures to explore the epigenetic landscape of a cohort of mouse primary pancreatic tumor (PPT) cell lines to define super-enhancer (SEs) associated genes and to discover master regulators that could play a significant role in the PDAC progression. Through these studies, I found that the transcription factor (TF) Foxj1, an evolutionarily conserved Forkhead gene family member, is implied in the modulation of SEs and enhancers in PI3K-mutant PPT cell lines. This transcription factor is the master regulator of ciliogenesis in various species and is highly dysregulated in several tumors, such as pancreatic cancer. However, the molecular role of FoxJ1 in PDAC is yet to be unveiled. Therefore, by genome editing I developed Foxj1 KO PI3K-mut and Foxj1 inducible pancreatic cancer cell lines to address the role of this TF. For the first time, I have found Foxj1 binding sites and the direct target genes in mammals. This defines a Foxj1 signature that is not connected to ciliogenesis. Moreover, the epigenomic analysis showed the involvement of Rfx transcription factors in modulating Foxj1. This family of TFs was previously described as a stabilizer of Foxj1 and DNA interaction in physiological conditions. These results indicate that Foxj1 and Rfx interaction could be relevant to tumor progression in PDAC. I discovered that inducing the expression of Foxj1 in Kras-mutant mouse PPT stopped cells from growing and partially reversed the mesenchymal phenotype. Finally, to translate those findings, I have explored several RNA-seq datasets of human PDAC, defining a subgroup of patients with a better prognosis based on the Foxj1 signature. At the same time, the positive nuclear staining for FOXJ1 protein in 23% of cases of a cohort of PDAC patients with increased overall survival confirmed the potential role of this TF as an oncosupressor. Here, I have defined the gene regulatory network of FOXJ1 and described its role as a novel master regulator in a subgroup of pancreatic cancer. In conclusion, this research has shed light on FOXJ1, a poorly studied TF, as a novel oncosupressor in PDAC. My results encourage the analysis of FOXJ1 expression as routinary test to define PDAC patients with a better prognosis.