TGF-β induces miR-100 and miR-125b but blocks let-7a through LIN28B controlling PDAC progression

TGF-β/Activin induces epithelial-to-mesenchymal transition and stemness in pancreatic ductal adenocarcinoma (PDAC). However, the microRNAs (miRNAs) regulated during this response have remained yet undetermined. Here, we show that TGF-β transcriptionally induces MIR100HG lncRNA, containing miR-100, miR-125b and let-7a in its intron, via SMAD2/3. Interestingly, we find that although the pro-tumourigenic miR-100 and miR-125b accordingly increase, the amount of anti-tumourigenic let-7a is unchanged, as TGF-β also induces LIN28B inhibiting its maturation. Notably, we demonstrate that inactivation of miR-125b or miR-100 affects the TGF-β-mediated response indicating that these miRNAs are important TGF-β effectors. We integrate AGO2-RIP-seq with RNA-seq to identify the global regulation exerted by these miRNAs in PDAC cells. Transcripts targeted by miR-125b and miR-100 significantly overlap and mainly inhibit p53 and cell-cell junctions' pathways. Together, we uncover that TGF-β induces an lncRNA, whose encoded miRNAs, miR-100, let-7a and miR-125b play opposing roles in controlling PDAC tumourigenesis.The authors thank Action Against Cancer (AAC), Pancreatic Cancer UK (PCUK), The Academy of Medical Sciences, The Royal College of Surgeons of Edinburgh, The Colin McDavid Family Trust, No Surrender Cancer Trust (in memory of Jason Boas), Mr Alessandro Dusi, Cheryl Whitehead, BHM, Sir Douglas Myers, and The Ralph Bates Pancreatic Cancer Research Fund for funding this study. Financial support was provided by the Dutch Cancer Society, KWF# 10401 grant to E.G., Italian Association for Cancer Research AIRC/Start-Up grant to E.G., Istituto Toscano Tumouri ITT-grant to N.F. and E.G., and the Regione Toscana "Progetto DIAMANTE"/FAS grant to N.F. and E.G. The authors thank Prof. Matthias Lohr and Dr Rainer Heuchel at Karolinska University Hospital, Stockholm, Sweden, for providing the TGF-beta 1 and empty vector expressing PANC-1 cells. This work used the computing resources of the UK MEDical BIOinformatics partnership-aggregation, integration, visualization, and analysis of large, complex data (UK MED-BIO) which is supported by the Medical Research Council [grant number MR/L01632X/1].S