Hye-Been Yoo^1#, Jin Woo Moon^2#, Hwa-Ryeon Kim^1#, Hee Seung Lee^3#, Koji Miyabayashi^4#, Chan Hee Park³, Sabrina Ge^5,6, Amy Zhang^5,7, Yoo Keung Tae³, Yujin Sub⁸, Hyun-Woo Park¹, Heon Yung Gee⁸, Faiyaz Notta^5,6,7, David A. Tuveson⁹, Seungmin Bang³, Mi-Young Kim^2,10, Jae-Seok Roe^1,10

¹ Department of Biochemistry, Yonsei University, Seoul, Korea
² Department of Biological Sciences, Korea Advanced Institute of Science and Technology, Daejeon, Korea
³ Division of Gastroenterology, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
⁴ Department of Gastroenterology, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan
⁵ Princess Margaret Cancer Centre, Toronto, Ontario, Canada
⁶ Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada
⁷ PanCuRx Translational Research Initiative, Ontario Institute for Cancer Research, Toronto, Ontario, Canada
⁸ Department of Pharmacology, Brain Korea 21 PLUS Project for Medical Sciences, Yonsei University College of Medicine, Seoul, Korea
⁹ Lustgarten Foundation Dedicated Laboratory at Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, USA
¹⁰ Corresponding authors

^# Author names in bold designate shared co-first authorship

BACKGROUND AND AIMS
Pancreatic ductal adenocarcinoma (PDA), with its highly metastatic propensity, is one of the most lethal subtypes of pancreatic cancer. Although recent large-scale transcriptomic studies have demonstrated that heterogeneous gene expressions play an essential role in determining molecular phenotypes of PDA, biological cues for and consequences of distinct transcriptional programs remain unclear.

METHODS
We developed an experimental model that enforces the transition of PDA cells toward a basal-like subtype. We combined epigenome and transcriptome analyses with extensive in vitro and in vivo evaluations of tumorigenicity to demonstrate the validity of basal-like subtype differentiation in association with endothelial-like enhancer landscapes via TEAD2. Finally, we used loss-of-function experiments to investigate the importance of TEAD2 in regulating reprogrammed enhancer landscape and metastasis in basal-like PDA cells.

RESULTS
Aggressive characteristics of the basal-like subtype are faithfully recapitulated in vitro and in vivo, demonstrating the physiological relevance of our model. Further, we showed that basal-like subtype PDA cells acquire a TEAD2-dependent pro-angiogenic enhancer landscape. Genetic and pharmacological inhibitions of TEAD2 in basal-like subtype PDA cells impair their pro-angiogenic phenotypes in vitro and cancer progression in vivo. Lastly, we identify CD109 as a critical TEAD2 downstream mediator that maintains constitutively activated JAK-STAT signaling in basal-like PDA cells and tumors.

CONCLUSIONS
Our findings implicate a TEAD2-CD109-JAK/STAT axis in the basal-like differentiated pancreatic cancer cells and as a potential therapeutic vulnerability.