Fifty-two pairs of HCC tumor tissues and para-tumor tissues were collected from the Second Affiliated Hospital, School of Medicine, Zhejiang University. This study was approved by the Ethics Committee of the Second Affiliated Hospital, School of Medicine, Zhejiang University. Written informed consent was obtained from HCC patients who underwent surgery. Accession numbers of RNA, DNA and protein sequences used in the manuscript should be provided.

Huh-7, Hep3B, and SUN449 cells were provided and authenticated by ATCC (Manassas, VA, USA). Huh-7 cells were cultivated in DMEM, Hep3B cells were maintained in MEM, and SNU-449 cells were cultivated in RPMI media 1640. These media were purchased from Gibco (Gibco, Carlsbad, CA, USA). All of the cells were maintained in a medium containing 10% fetal bovine serum (FBS, Gibco), 1% penicillin and 1% streptomycin.

RNA isolation and RT‑qPCR were performed as described earlier (Chen et al., 2020). The total RNA was isolated from HCC tissue and cells using TRIzol agent. Next, the cDNA was generated using PrimeScript™ RT reagent Kit with gDNA Eraser (RR047A, Takara, Dalian, China) for lncRNA and MiR-X miRNA First-Strand Synthesis Kit (638315, Takara) for miRNA. RT‑qPCR was conducted using TB Green® Premix Ex Taq™ (RR420A, Takara) according to the specification. The relative gene expression was evaluated by the 2^−ΔΔCt method (Livak and Schmittgen, 2001). GAPDH and U6 acted as the internal control, respectively. The sequences of POIR and miR-182 primers utilized in this study are listed in Table 1. Rest primers sequences were listed in Table S1.

Small interfering (si)RNA lnc POIR or scrambled siRNA was synthesized by GenePharma Company (Shanghai, China). miR-182 mimic/inhibitor or a negative mimic/inhibitor control were synthesized by Ribobio, respectively (Ribo, China). The transfection was transfected using Lipofectamine® 2000 (Invitrogen; Carlsbad, CA, USA) according to the instructions. The sequences used in this study are listed in Table 2.

Cytotoxicity assay was assayed using Cell Counting Kit-8 (CCK-8) (Dojindo, Japan). The CCK-8 assay was performed after the seeding of HCC cells (5,000 cells/well) in 96-well plates. Afterward, the cells were treated with 0, 1.25, 2.5, 5, 10, or 20 μM sorafenib for 48 h. The supernatant was discarded and replaced with a fresh 100 µL serum-free medium (containing 10 µL CCK-8 reagents). Absorbance was measured at 450 nm using a microplate reader (BioTek).

Western blotting was performed as previously described. Proteins were lysed in a RIPA buffer (Beyotime, China). Next, the protein concentrations were determined using a bicinchoninic acid protein assay (Beyotime). The lysed protein (40 µg) was electrophoresed with 10% SDS-PAGE and transferred to PVDF membranes (Millipore, USA), followed by incubation with 5% non-fat milk for 2 h. Subsequently, the membranes were incubated with primary antibodies specifically for LC3B (#3868S, Cell Signaling Technology, 1:1000 dilution), P62 (#88588S, CST, 1:1000 dilution), GAPDH (#2118S, CST, 1:2000) at 4°C overnight, followed by an incubation with HRP secondary antibodies (anti-mouse, #7076S; anti-Rabbit, #7074S; 1:2,000, CST) for 2 h at room temperature. The chemiluminescence intensity was evaluated using ECL reagents (Applygen, Beijing, China).

Cell proliferation was analyzed by employing a Click-iTEdU Imaging kit (Invitrogen; Thermo Fisher Scientific, Inc.) as previously described. Briefly, the fixed cells were incubated with 100 µL EdU to detect the positive cells. The cells were counterstained with 100 µL Hoechst 33342 to label the nuclei. Immunofluorescence was observed with a fluorescence microscope at 200× magnification.

mRFP-GFP-LC3 adenovirus was bought from Hanbio (Hanbio, China). The cells that were incubated with an mRFP-GFP-LC3 adenovirus for 24 h were exposed to sorafenib for another 24 h. Autophagic flux was observed with a fluorescent microscope (Olympus, Tokyo, Japan).

The apoptosis assay was performed using an Annexin V-FITC/propidium iodide (PI) apoptosis detection kit according to the user’s guide (Beyotime, Shanghai, China). The harvested cells were suspended and incubated with Annexin V and PI (1:1) in the dark. The analysis of the apoptotic cells was performed using flow cytometry (BD).

Data are presented as the mean ± SD. Statistical analyses were carried out using SPSS software (19.0 revision, IBM, Chicago, IL, USA) and GraphPad Prism (version 7; GraphPad Software, Inc.). Student’s t-test was used to calculate the two group differences, while one-way ANOVA was applied to compare the multiple group differences. P < 0.05 was considered as a statistically significant difference.

To identify the effect of lncRNAs on the response of HCC cells to sorafenib, we explored the expression profiles of lncRNAs in three pairs of tumor tissues and adjacent tissues from patients with HCC using RT-qPCR (Fig. 1A). Among the 25 lncRNAs, we were interested in POIR. Furthermore, we used RT-qPCR to detect the level of POIR expression in 52 liver cancer tissues. The results showed that POIR expression in the HCC tissues was significantly higher than that in the adjacent tissues (Fig. 1B). The level of POIR expression in each of the 52 cases was shown in Fig. 1C, of which 41 cases exhibited increased POIR expression. These data suggest that POIR may be an oncogene. In addition, we detected the cytotoxicity of sorafenib in HCC cells. The IC50 of SNU449 cells (14.82 ± 0.66 μM), which exhibit mesenchymal phenotype (Xue et al., 2016), was higher than that of Huh7 (6.81 ± 0.25 μM) and Hep3B cells (9.89 ± 0.69 μM), which have epithelial phenotype (Figs. S1A–B). At the same time, we also found that POIR expression was significantly higher in HCC cells than that in the LO2 cells, with POIR expression of SNU449 being the highest, followed by Hep3B and Huh-7 (Fig. S1C). This suggests that POIR may be associated with HCC resistance to sorafenib.

To assess the effects of POIR on HCC resistance to sorafenib, we constructed POIR siRNA to achieve POIR knockdown. The efficiency of these three HCC cells with POIR siRNA was measured by RT-qPCR (Fig. S1D). When the cells were transfected with POIR siRNA, the cell viability was dramatically inhibited in all three HCC cell lines compared with NC siRNA (NC-si) (Fig. S1E and Table S2). Compared with control, sorafenib could inhibit cell proliferation, as shown by the decrease of the EDU-positive cell ratio (Fig. 1D). POIR siRNA could enhance the inhibitory effect of sorafenib on proliferation (Fig. 1D). Moreover, POIR siRNA could further potentiate the level of cell apoptosis induced by sorafenib (Fig. 1E). Overall, these data suggest that silencing POIR promoted the sensitization of HCC cells to sorafenib.

Previous studies have linked autophagy to drug resistance in human cancers (Niu et al., 2017). Therefore, we hypothesized that autophagy contributes to POIR silencing-mediated sorafenib sensitivity. To test this hypothesis, we used 3-methyladenine (3-MA), an autophagy inhibitor. Compared with control, 3-MA treatment enhanced the sensitivity of HCC cells to sorafenib (Fig. 2A). The result indicated that autophagy is involved in sorafenib resistance. To determine whether POIR knockdown mediates sorafenib sensitivity by regulating autophagy in HCC, HCC cells were transfected with or without POIR siRNA. We performed confocal microscopy to observe the change of autophagy flux in transfected HCC cells exposed to sorafenib. The results showed that POIR silencing decreased the autophagy flux induced by sorafenib (Figs. 2B and 2C). The results of the western blot analyses were consistent with the confocal microscopy assay (Fig. 2D), indicating POIR might mediate sorafenib resistance via autophagy.

To further investigate that POIR siRNA increased cell sensitivity to sorafenib by regulating autophagy, we used 3-MA to interfere the autophagic process. Interestingly, in the presence of 3-MA, POIR inhibition no longer influence the cell viability (Fig. 3). Therefore, our findings confirmed that inhibiting POIR increases cell sensitivity to sorafenib by regulating autophagy.

It has been reported that POIR acted as a sponge of miR-182-5p and promoted the osteogenesis of PDLSCs (Wang et al., 2016). However, whether POIR mediates sorafenib resistance by regulating miR-182-5p is unknown. Therefore, we hypothesized that POIR siRNA regulates sorafenib sensitivity in HCC by inhibiting autophagy through interacting with miR-182-5p. To further verify the above hypothesis, we first performed miRNA target site prediction using a DIANA software analysis (http://carolina.imis.athena-innovation.gr/diana_tools/web/index.php). We found that POIR contains miR-182-5p binding sites (Fig. 4A). We also found that inhibiting POIR by POIR siRNA upregulated miR-182-5p expression in HCC cells (Fig. 4B). Next, we synthesized miR-182-5p mimic and miR-182-5p inhibitor to observe the effect of miR-182-5p on sorafenib resistance. We used CCK-8 and EDU assays to assess their effect. The results demonstrated that miR-182-5p overexpression significantly enhanced sorafenib sensitivity and suppressed cell proliferation compared with NC. Conversely, the downregulation of miR-182-5p inhibited sorafenib sensitivity and promoted the EDU-positive cell ratio (Figs. 4C–4E). The efficiency was detected by RT-qPCR (Fig. 4F).

To further explore the relationship between POIR, miR-182-5p, and autophagy, SNU449 cells were co-transfected with NC siRNA, miR-182-5p inhibitor, POIR siRNA, and POIR siRNA and miR-182 inhibitor, respectively. The CCK-8 assay showed that POIR downregulation increased the sorafenib sensitivity in SNU449 cells, whereas the miR-182-5p inhibitor abolished these effects (Fig. 5A). Furthermore, flow cytometry assay revealed that miR-182-5p inhibitor reversed the effects of POIR knockdown on SNU449 cell apoptosis (Figs. 5B and 5C). Confocal microscopy and Western blot assay showed that the role of POIR siRNA on the autophagy of SNU449 cells could be reversed by a miR-182-5p inhibitor (Figs. 5D–5F). The efficiency of 449 cells with POIR siRNA, miR-182-5p mimic or miR-182-5p inhibitor was measured by RT-qPCR (Fig. 5G). These data indicate that silencing POIR enhanced sorafenib sensitivity via regulating miR-182-5p and autophagy.