Signal Transduction and Targeted Therapy | 2021
SOX9-transactived long non-coding RNA NEAT1 promotes the self-renewal of liver cancer stem cells through PKA/Hippo signaling
Abstract
Dear Editor, Hepatocellular carcinoma (HCC), the most common pathological type of primary liver cancer, ranks as the third deadliest cancer. Despite the progress of surgical resection in recent years, the 5-year survival of HCC patients is still unsatisfactory due to the frequent relapse and chemoresistance. Accumulating evidence has demonstrated that liver cancer stem cells (CSCs) are critical for HCC chemoresistance and recurrence. Nevertheless, the molecular mechanisms of liver CSC regulation remain unclear, which hampers the development of the therapeutic strategy that targets liver CSCs. Long non-coding RNAs (LncRNAs) are defined as RNA transcripts >200 nucleotides (nt) in length without proteincoding potential, which have been reported to be involved in organ development, stemness maintenance, and cell differentiation in physiological processes, but the role of lncRNAs in liver CSC expansion remains unclear. Since self-renewal and chemoresistance are distinct characteristics of CSCs, we enriched liver CSCs by inducing hepatoma spheroid formation in the presence of cisplatin, and an lncRNA microarray analysis identified 87 lncRNAs upregulated in liver CSCs (Supplementary Fig. s1a and Table s1), among which the nuclear paraspeckle assembly transcript 1 (NEAT1) is the most prominent lncRNA in cancer research. We then compared the expression of NEAT1 in HCC spheres and attached HCC cells by fluorescence in situ hybridization and observed the evident upregulation of NEAT1 in HCC spheres (Fig. 1a). Epithelial cell adhesion molecule (EpCAM), cluster of differentiation 24 (CD24), and oval cell 6 (OV6) have been reported to serve as surface biomarker of liver CSCs. Herein upregulated NEAT1 expression was detected in EpCAM-, CD24-, or OV6-positive HCC cells, which further suggested that NEAT1 is highly expressed in liver CSCs (Figs. 1b and s1b). In addition, a positive correlation between NEAT1 and liver CSC markers was also observed in patient HCC tissues (Supplementary Fig. s1c). High NEAT1 expression was observed in the majority of patient HCCs (56/81) compared to the cancer adjacent tissues (cohort 2), while no population distribution difference in gender or age was found between the NEAT1 high and NEAT1 low patient groups (data not shown). Moreover, the patients with high NEAT1 expression had a higher recurrence rate than did patients with low NEAT1 expression (Supplementary Fig. s1d). Although NEAT1 has been demonstrated to be highly expressed in various cancer types, the mechanism underlying its upregulation is not clear. Bioinformatics analysis revealed two conserved binding sites of stemness-associated transcription factor SOX9 (SRY-box transcription factor) in the promoter region of NEAT1 (Fig. 1c). Chromatin immunoprecipitation assay demonstrated that SOX9 was enriched on both two predicted SOX9-binding sites in NEAT1 promoter (Fig. 1c). We then constructed luciferase reporter plasmids of NEAT1 promoter containing wild-type SOX9-binding sites (NEAT1-WT) or mutant SOX9-binding sites (NEAT1-Mut). As expected, the NEAT1-Mut luciferase reporter could not be activated by SOX9 overexpression (Fig. 1d). As shown in Fig. s2a, transfection of SOX9-expressing plasmid robustly increased the NEAT1 transcripts, which further confirm the transactivating role of SOX9 in NEAT1 expression. Consistently, a positive correlation between NEAT1 expression and SOX9 levels was observed in patient HCC tissues (Supplementary Fig. s2b). Considering NEAT1−/− mice was reported to express less Sox9 compared with WT control, there could be a feedback regulation between SOX9 and NEAT1, which is worthy of future investigation. To explore the role of NEAT1 in liver CSCs, NEAT1-silenced HCCLM3 and Huh7 cell lines were established (Supplementary Fig. s3a). Flow cytometric analysis demonstrated that interference of NEAT1 dramatically reduced the proportion of liver CSCs in HCC cells (Supplementary Fig. s3b). Moreover, NEAT1-silenced HCC cells formed fewer and smaller spheroids than did the control cells (Supplementary Fig. s3c, d). An in vitro limiting dilution assay illustrated that NEAT1 knockdown dramatically decreased the CSC population in hepatoma cells (Fig. s3e). Moreover, hepatoma cells expressing shNEAT1 displayed an attenuated tumor-initiation capacity in NOD-SCID mice compared with control cells (Figs. 1e and Supplementary Fig. s3f), further indicating the role of NEAT1 in liver CSC self-renewal. To further confirm the role of NEAT1 in liver CSCs, NEAT1-expressing transfectants were established (Supplementary Fig. s4a) and the consistent results were achieved (Supplementary Fig. s4b–g). Since certain drugs targeting RNA have been approved for clinic application, we propose that NEAT1 might be a promising therapeutic target in HCC therapy. To explore the molecular mechanism underlying NEAT1mediated liver CSC self-renewal, a luciferase reporter assay was performed. Overexpression of NEAT1 activated the Hippo signaling pathway but did not influence the signal transducer and activator of transcription 3, SMAD family member 3, or β-catenin signaling cascades (Supplementary Fig. s5a). Hippo pathway plays an inimitable role in liver development, regeneration, and carcinogenesis. Herein phosphorylation of large tumor-suppressor kinase (LATS) and yes-associated protein (YAP), two key nodes in the Hippo pathway, was decreased by NEAT1 overexpression but increased by NEAT1 interference in HCC spheroids (Figs. 1f and Supplementary Fig. s5b), while phosphorylation of macrophage stimulating 1/2 was not altered (data not shown). Consistently, YAP downstream target genes cysteine rich 61 and connective tissue growth factor were induced by NEAT1 overexpression but reduced by NEAT1 interference (Supplementary Fig. s5c). NEAT1 overexpression significantly promoted the nuclear translocation of YAP in HCC cells (Supplementary Fig. s5d), which was further confirmed by nucleus YAP analysis in NEAT1-overexpressing or NEAT1silenced HCC cells (Supplementary Fig. s5e). As expected,