br Conclusions The following are the supplementary

Conclusions The following are the supplementary data relate to this article.
Disclosure of potential conflicts of interest
Acknowledgments We thank Fondazione Roma for continuous support. We also acknowledge Prof. Mara Riminucci, Prof. Luca Pierelli, Dr. C. Murdoch, Dr. Paola Iudicone and Eve Technologies for the analysis of growth factors and cytokines. This study was financially supported by Sapienza University of Rome prot. N. C26A1429M2 granted to Elena De Falco.
(A and B) Cell cycle regulation is not altered by TBX3 overexpression in PDAC cell lines (Panc1 and BxPC3). (C) TBX3-OE does not increase cell growth in Panc1 cells. (D and E) Overexpression of TBX3 does not significantly influence the induction of apoptosis. Introduction Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal malignancies in the western world (Jemal et al., 2010). Due to an increasing incidence it is estimated to be the third leading cause of cancer-related deaths by 2030 (Rahib et al., 2014). While overall survival in many solid malignancies has significantly improved over the last decade (Jemal et al., 2010), there has only been a marginal improvement in PDAC despite improved therapies and surgical techniques (Howlader et al., 2013). At diagnosis, only 15–20% of the patients even qualify for potentially curative resection. Most patients present with already locally advanced or metastatic disease, caused by the usually late diagnosis due to missing early symptoms and early metastatic spread, a hallmark of PDAC (Haeno et al., 2012). 5-year survival ranges around 7% (Jemal et al., 2010; Howlader et al., 2013), and even after surgical resection and subsequent state-of-the-art adjuvant chemotherapy, the overall survival rate does not exceed 20% (Jemal et al., 2010; Haberland et al., 2010). Regrettably few patients survive longer, but the distinguishing features that predict long-term survival are not yet fully understood (Yachida et al., 2012; Russell et al., 2015a). It has been demonstrated that virtually all cancers including PDAC are highly heterogeneous and contain a subset of cancer stem cells, tumour topotecan Supplier with features that distinguish them from the tumour bulk (Al-Hajj et al., 2003; Hermann et al., 2007; Kim et al., 2005; Li et al., 2007; O\'Brien et al., 2007; Ricci-Vitiani et al., 2007; Singh et al., 2004). These tumour cells can self-renew, have an exceptionally high tumour-initiating capacity, and recapitulate the heterogeneity of the parental tumour. Furthermore, these cells have distinct metabolic (Sancho et al., 2015) and cell cycle properties, are highly chemoresistant (Cioffi et al., 2015; Mueller et al., 2009; Hermann et al., 2013; Hermann et al., 2007), and are responsible for metastatic spread (Hermann et al., 2007). Finally, these cells contain a self-renewal machinery driven by up-regulated pluripotency factor expression such as NANOG or OCT3/4 and by re-activation of embryonic pathways such as NODAL/ACTIVIN signalling (Lonardo et al., 2011). Thus, these pancreatic cancer stem cells share certain stemness features with bona fide pluripotent embryonic stem cells (Liebau et al., 2014). The T-box transcription factor 3 (TBX3) belongs to an ancient group of genes that share a highly conserved DNA binding domain, the T-box (Muller and Herrmann, 1997; Minguillon and Logan, 2003; Bertolessi et al., 2015; Weidgang et al., 2016). TBX3 is widely expressed in many tissues, plays a pivotal role during embryonic development, and is tightly linked to early lineage commitment and the pluripotency of embryonic stem cells (Kim et al., 2008; Weidgang et al., 2013; Kartikasari et al., 2013; Russell et al., 2015b; Waghray et al., 2015). Furthermore, TBX3 functions as a transcriptional repressor, is implicated in cell cycle regulation, and in tumour development. TBX3 is overexpressed in human colorectal (Shan et al., 2015), head and neck (Burgucu et al., 2012), bladder (Beukers et al., 2015) and breast cancer (Fan et al., 2004; Yarosh et al., 2008) as well as in melanoma (Peres et al., 2010). Most recently, high TBX3 expression has been shown to negatively correlate with survival in a cohort of 80 resected human PDAC samples (Wang et al., 2015), demonstrating a substantial role of TBX3 in pancreatic cancer. Moreover, TBX3 overexpression was found to be a negative prognostic factor in colorectal cancer (Shan et al., 2015) and hepatocellular carcinoma (Renard et al., 2007). Mechanistically, TBX3 is repressed by miR-206 and miR-17-92 to inhibit proliferation but also to decrease the cancer stem cell population in breast and pancreatic cancer (Cioffi et al., 2015; Amir et al., 2016), while in certain sarcomas it drives proliferation via p21 repression. Its expression qualifies as a biomarker to predict survival (Willmer et al., 2016a; Willmer et al., 2016b). Interestingly, putative cancer driver mutations in the TBX3 gene cause impaired transcriptional repression in breast cancer (Fischer and Pflugfelder, 2015) and correlate with an invasive lobular carcinoma phenotype (Ciriello et al., 2015). Therefore, targeting TBX3 itself or its signalling axis may be of high clinical relevance in the future.