The creation and maintenance of

The creation and maintenance of subcutaneous patient-derived xenografts (PDX) from fresh or cryopreserved ACC tissue (Moskaluk et al., 2011) provided a renewable source of ACC A 779 manufacturer for validation of our CSC hypothesis. As we previously demonstrated, these PDX models reproduced ACC morphology and maintained the SOX10 gene signature (Ivanov et al., 2013a; Moskaluk et al., 2011). To isolate SOX10+ CSC from grafted ACC tissue, we used a ROCK inhibitor-based cell culture protocol (Liu et al., 2012) and profiled our cell cultures for expression of SOX10 and other ACC-intrinsic stem cells markers as well as for expression of CD133, a CSC cell surface marker. In line with our hypothesis, we isolated from patient-derived xenografts (PDX) and clinical ACC specimens a new population of SOX10+/CD133+ cancer stem cells, which we named ACC-CSC. These cells co-expressed SOX10 with activated NOTCH1 and met basic CSC criteria by producing spheroids in vitro and generating aggressive ACC-like tumors in nude mice. In addition, ACC-CSC activated stem cell signaling with involvement of SOX10 and NOTCH1 (Panaccione et al., 2016). In this study, we authenticate cultured ACC-CSC using ACC-intrinsic MYB-NFIB fusions, investigate their signaling pathways, and test new tools, markers, and assays for their purification.
Materials and methods
Results
Discussion Development of targeted therapies for ACC has been stalled by lack of reliable cell lines/cultures and approaches for CSC isolation. Unfortunately, most ACC cell lines generated 20–30years ago and shared between laboratories turned out to be grossly contaminated or misidentified (). These cell lines are now blacklisted (http://iclac.org/databases/cross-contaminations/). However, of >100 manuscripts that relied on these cell lines only two have been retracted (), and new studies with same cell lines have been recently published (Sumida et al., 2016; Maruyama et al., 2010; Wu et al., 2015; Fang et al., 2015; Sun et al., 2015; Sumida et al., 2013; Wang et al., 2016). While the most plausible explanation for the ACC cell line identity crisis is lack of proper authentication (Yu et al., 2015), ACC cells are also notoriously difficult to culture as they rapidly undergo senescence or die in regular cell culture media (our unpublished observations). In this study, we continue characterization of a novel population of CSC, ACC-CSC, that we recently purified from ACC using a ROCK inhibitor and mouse cell feeder layer-based approach (Panaccione et al., 2016). To our knowledge, this report is the first study to detect MYB fusions in ACC cultures, despite the fact that MYB-NFIB fusions have been described in 86% of ACC patients (Brill et al., 2011). In a fraction of ACC tumors, MYB fusions with other genes, such as TGFBR3 and RAD51B, as well as MYBL1 fusions with NFIB have been also reported (Drier et al., 2016; Mitani et al., 2016). Here, we identified a novel MYB fusion with RP11-54515.3, an lncRNA gene from chromosome 6q24.3. Intriguingly, this lncRNA is a cis natural antisense transcript (cis-NAT) for SHPRH, an E3 ubiquitin ligase. While exact functions of cis-NAT are not known, they may inhibit expression of corresponding sense transcripts via transcription interference or post-transcriptionally (Rosikiewicz and Makalowska, 2016). Interestingly, SHPRH functions as a tumor suppressor protein that induces degradation of β-catenin, a CSC driver, and has been recently recognized as a target of axitinib, a small molecule inhibitor of WNT/β-catenin signaling (Qu et al., 2016). It may be hypothesized therefore that the MYB-RP11-54515.3 fusion may inhibit SHPRH translation and stimulate ACC-CSC via WNT/β-catenin signaling. This hypothesis will be validated on Accx33 cells that harbor this fusion. Overall, generation of ACC cells with oncogenically activated MYB provided long-awaited tools for studies focused on the MYB role in ACC. Our next goal was to perform a more detailed characterization of ACC-CSC to identify therapeutically amenable pathways of their propagation. To this end, we used CyTOF, an innovative flow cytometry technology, which is based on time-of-flight mass-spectrometry (Giesen et al., 2014). CyTOF analysis demonstrated that CD133+ AC-CSC are highly enriched with β-catenin and STAT3, two major targets actively explored in novel cancer-targeting therapies (Qu et al., 2016; Xiang et al., 2016). We and others previously implicated both STAT3 and Wnt/β-catenin signaling in ACC (Ivanov et al., 2013a; de Araujo et al., 2008; Daa et al., 2004; Frierson et al., 2002), and this study, for the first time, links both pathways to CSC suggesting their cooperation and providing targets for ACC-CSC eradication.