LncRNA DHRS4-AS1 Inhibits the Stemness of NSCLC Cells by Sponging miR-224-3p and Upregulating TP53 and TET1

Non-small cell lung cancer (NSCLC) is the leading cause of cancer-related death. This study aimed to examine the roles of DHRS4-AS1/miR-224-3p signaling in the cancer cell stemness of NSCLC. Real-time PCR showed that DHRS4-AS1 was downregulated in cancerous tissues, and bioinformatics analysis revealed that high DHRS4-AS1 expression indicated a good prognosis for NSCLC patients. Sphere and colony formation assays showed that DHRS4-AS1 overexpression significantly suppressed NSCLC cell colony formation and stem cell-like properties. DHRS4-AS1 also abrogated the expression of OCT4, SOX2, CD34, and CD133, markedly inhibited the expression of epithelial-mesenchymal transition (EMT)-related factors, N-cadherin, ZEB1, and Vimentin, and increased E-cadherin expression in spheres. Furthermore, luciferase reporter assays and real-time PCR analysis demonstrated that DHRS4-AS1 and miR-224-3p were antagonistically repressed in NSCLC cells. RNA immunoprecipitation (RIP) analysis revealed that DHRS4-AS1 interacted with miR-224-3p. DHRS4-AS1 partially reversed the miR-224-3p-decreased TP53 and TET1, resulting in the inhibition of tumor growth in vivo. Finally, TP53 and TET1 were antagonistically regulated by DHRS4-AS1 and miR-224-3p in NSCLC cells. In conclusion, TP53- and TET1-associated DHRS4-AS1/miR-224-3p axis is an essential mechanism by which NSCLC modulates cancer cell stemness.


INTRODUCTION
Lung cancer is the most common cancer worldwide and ultimately causes the death of many patients (Chen et al., 2016;Siegel et al., 2020). Non-small cell lung cancer (NSCLC) is a common type of lung cancer, accounting for up to 85% of all lung cancer cases, and the 5-year survival rate of NSCLC patients remains poor (Boolell et al., 2015). Although effective treatment has been achieved for NSCLC patients in the clinic, most cancers recur after surgery or radiotherapy (Wood et al., 2015;Duma et al., 2019). The molecular mechanisms of NSCLC have not been fully explored. Undoubtedly, the identification of biomarkers for the prognosis and treatment of patients with NSCLC is urgently needed. To produce better therapeutic strategies, it is essential to elucidate the signaling pathways of metastasis and recurrence of NSCLC progression. The recurrence of NSCLC is associated with cancer stem cells. Cancer stem cells are potentially responsible for cancer cell self-renewal, initiation, proliferation, and differentiation, can produce aggressive and metastatic tumors, as well as multidrug resistance (Kusoglu and Biray Avci, 2019).
Long non-coding RNAs (lncRNAs) were reported to contribute the initiation and development of cancers by our and other groups (Bautista et al., 2018;Chang et al., 2018;Zhao et al., 2018;Huang et al., 2019;Li et al., 2020). HOX transcript antisense RNA (HOTAIR) enhances lung cancer cell invasion and motility (Zhao et al., 2014a). Maternally expressed 3 (MEG3) inhibits cancer cell proliferation and increases p53-mediated cancer cell apoptosis during NSCLC (Lu et al., 2013). Metastasis-associated lung adenocarcinoma transcript 1 (MALAT1) is considered a key marker of metastasis in lung cancer (Gutschner et al., 2013). The lncRNA actin filament-associated protein 1 antisense RNA 1 (AFAP1-AS1) promotes NSCLC tumorigenesis and chemoresistance (Yin et al., 2018;Huang et al., 2019). SBF2-AS1, an oncogenic lncRNA, increases NSCLC cell proliferation (Lv et al., 2016). Most importantly, lncRNAs also play roles in the stemness of lung cancer. For instance, the long non-coding RNA Linc00662 promotes cell invasion and contributes to cancer stem cell-like phenotypes in lung cancer cells (Gong et al., 2018). The lncRNA FENDRR suppresses cancer cell stemness by downregulating multidrug resistance gene 1 (MDR1) expression by competitively binding with HuR, an RNA-binding protein that plays a role in NSCLC progression (Gong et al., 2019). The lncRNA CASC11 increases TGF-beta1-mediated cancer cell stemness, resulting in a poor overall survival rate for NSCLC patients (Fu et al., 2019). The lncRNA linc-ITGB1 suppresses cancer stemness by inhibiting Snail expression in NSCLC . The lncRNA HAND2-AS1 inhibits cancer cell migration and invasion and maintains cancer cell stemness by interacting with TGF-beta1 . According to the databases "Gene Expression Profiling Interactive Analysis (GEPIA, http://gepia.cancer-pku.cn/index.html)" data, DHRS4-AS1 might be an important lncRNA in lung cancer progression. Although the function of DHRS4-AS1 was characterized in other cancers Luan et al., 2019;Utnes et al., 2019), the role of DHRS4-AS1 in NSCLC still needs uncovery. Our study revealed that the expression of DHRS4-AS1 in tumors correlates with the overall survival of NSCLC patients, and DHRS4-AS1 functions as a tumor suppressor by regulating cancer cell colony formation and stemness. We conclude that TP53-and TET1-associated DHRS4-AS1/miR-224-3p signaling plays important roles in NSCLC progression in vitro and in vivo. Consequently, DHRS4-AS1 and miR-224-3p are promising therapeutic targets for NSCLC.

Clinical Sample Collection
Eighty-three pairs of lung cancer and adjacent tissues were collected from NSCLC patients at the Affiliated Cancer Hospital of Nanjing Medical University. None of the patients had been treated clinically before surgery. The collected tissues were immediately stored in liquid nitrogen until further analysis. Clinical specimen collection was approved by the Research Ethics Committee of the Affiliated Cancer Hospital of Nanjing Medical University (ID: 2018RC1D), and a consent form was signed by each participating patient.

Haematoxylin-Eosin (H&E) Staining
Clinical lung tissues were fixed in 4% paraformaldehyde and subjected to sectioning. H&E staining was performed as previously described (Zhao et al., 2014b). All chemicals and agents used were purchased from Sigma (Shanghai, China).

Sphere and Colony Formation of Cancer Stem Cells
Cancer stem cells were sorted from A549 and Calu-3 cells using spherocyte medium. In total, 2 × 10 4 cells were plated onto six-well ultra-low cluster plates (NUNC, Thermo Fisher Scientific) and cultured in DMEM/F12 serum-free medium (Gibco) supplemented with epidermal growth factor (20 ng/mL), b-fibroblast growth factor (20 ng/mL), insulin (4 µg/mL), and B27 (2%) (all from Sigma). The number of spheres was captured and counted under an inverted microscope (Leica, Oskar-Barnack-Straße, Germany) after 10 days. Cancer stem cells were identified by western blot or real-time PCR analysis of the expression of stem cell markers, including OCT4, SOX2, CD34, and CD133.
Cancer stem cells were separated using CD133-binding magnetic bead kit (Becton, Dickinson and Company, New Jersey, USA), and 2 × 10 4 stem cells were then plated into six-well plates with an ultra-low attachment surface (MUNC). These cells were maintained in sphere formation medium for 7 days, and the indicated siRNAs or plasmids were transfected into sphereforming cells using Lipofectamine 3000. Finally, the number of spheres was quantified under a microscope after an additional 20 days.

Western Blot Analysis
Protein samples were prepared with a protease inhibitor cocktail containing RIPA lysis buffer (Beyotime, Haimen, China). The protein lysates were denatured at 98 • C for 10 min. Next, samples (∼50 µg) were separated by 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to 0.22-µm polyvinylidene difluoride (PVDF) membranes (Bio-Rad, Hercules, CA, USA) using the Trans-Blot R Turbo TM Transfer System (Bio-Rad). The protein-containing PVDF membranes were blocked with 1.0% bovine serum albumin (BSA) diluted with PBS. The membranes were incubated with the indicated primary antibody at 4 • C for 8 h and then incubated with a goat anti-mouse or goat anti-rabbit horseradish peroxidase (HRP)-conjugated secondary antibody diluted 1:40,000 (Bioworld Technology, Nanjing, China) for 60 min. The following antibodies were used: anti-

Dual-Luciferase Reporter Assays
The cancer cell lines were seeded into 24-well plates at a density of 6 × 10 4 cells per well. The cells were co-transfected with the pmirGLO-DHRS4-AS1-WT, pmirGLO-DHRS4-AS1mut, pmirGLO-TP53 (or TET1)-3'-UTR WT or pmirGLO-TP53 (or TET1)-3 ′ -UTR-mut reporter plasmid and the scramble miR-224-3p mimic, or anti-miR-224-3p (inhibitor). After 24 h of transfection, luciferase activity was determined using the Dual-Luciferase Assay Kit according to the manufacturer's protocol on a GloMax 20/20 luminometer (Promega, Madison, Wisconsin, USA). In this study, the 3 ′ -UTR sequences of isoform 1 (Transcript ID: NM_000546.6) and TET1 (ID: NM_030625) isoform 1 were obtained from the NCBI database, and the full length DHRS4-AS1 sequence (ID: ENSG00000215256) was obtained from the UCSC database. The corresponding luciferase reporter plasmids were constructed into pmirGLO empty vector by GenePharma using NheI and SbfI as restriction enzyme cutting sites for each plasmid. Three wells were utilized for each treatment group, and the experiments were repeated three times independently.

RIP Assays
To determine the miR-224-3p binding sites on DHRS4-AS1, RIP assay was performed using the Magna RIP RNA-binding protein immunoprecipitation kit (Millipore, Massachusetts, USA) according to the manufacturer's instructions. Anti-Ago2 was diluted 1:500 (Sigma, USA, cat. #: SAB4200085). The cancer cell lysate was incubated with anti-Ago2 buffer, and IgG was used as a negative control. Immunoprecipitated RNA was then used for the real-time PCR analysis of DHRS4-AS1 and miR-224-3p. Three wells were utilized for each group.

Animal Experiments
Stable Calu-3 cells were centrifuged at 80 g × 5 min, and 3 × 10 6 cells were then inoculated into a 5-week-old BALB/c nude mice (Model Animal Research Center of Nanjing University, Nanjing, China). The nude mouse experiments were approved by the Research Ethics Committee of Chengdu Medical College (ID: XN19L38). Tumor volumes were calculated every 3 days for 15 days according to the following formula: tumor volume = (length × width 2 )/2. The mice were sacrificed on the 15th day after xenograft implantation.

Statistical Analysis
The data were analyzed statistically using the SPSS software package (version 17.0, SPSS, Inc., Armonk, NY, USA) and GraphPad Prism 6 (GraphPad Software, La Jolla, CA, USA). The results are expressed as the mean ± standard deviation (SD). Statistical significance between two groups was examined by a two-tailed Student's t-test, and statistical significance among multiple groups was determined by using posthoc analysis contrasts. P-values < 0.05 were considered statistically significant.

DHRS4-AS1 Correlates With NSCLC Survival
To explore the relationship between DHRS4-AS1 expression and NSCLC patients survival, the survival rate was analyzed in GEPIA, which base on The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) samples, high DHRS4-AS1 expression indicated a good overall survival outcome of patients with NSCLC ( Figure 1A). Herein, we collected clinical tissues, and found that malignant proliferation of pulmonary epithelial cells in lung cancer tissues evidenced by H&E staining (Figure 1B). Real-time PCR analysis indicated that DHRS4-AS1 expression was decreased in NSCLC tissues compared with adjacent normal tissues (Figures 1B,C). In addition, DHRS4-AS1 was downregulated in lung cancer cells, including A549, PC-9, Calu-3, SPCA-1, and H1299 cells, compared with immortal human bronchial epithelial cells (BEA-2B) ( Figure 1D). Importantly, real-time PCR analysis Frontiers in Cell and Developmental Biology | www.frontiersin.org demonstrated that DHRS4-AS1 expression was lower in NSCLC cell spheres than in parental cells ( Figure 1E). These findings suggest that DHRS4-AS1 is a tumor suppressor and abrogates cancer stemness in NSCLC progression.
Additionally, DHRS4-AS1 inhibited the expression of epithelialmesenchymal transition (EMT)-related factors, N-cadherin, ZEB1, and Vimentin but increased E-cadherin expression in the cancer stem cell spheres (Figures 2H,I and Supplementary File). Frontiers in Cell and Developmental Biology | www.frontiersin.org

DISCUSSION
Lung cancer has been listed as the leading cause of cancerrelated death in humans. Cancer stem cells play an essential role in the aggressive and destructive behavior of lung cancer (Duma et al., 2019;Siegel et al., 2020). Cancer stem cells are positively associated with NSCLC recurrence, and these cells are closely connected with multidrug resistance and cancer initiation, proliferation, and differentiation (Heng et al., 2019). Importantly, lncRNAs have been reported to have important roles in NSCLC initiation, progression and stemness maintain by our and other groups (Zhan et al., 2017;Lu T. et al., 2018;Zhao et al., 2018;Huang et al., 2019;Li et al., 2020). AFAP1-AS1 promotes NSCLC cell proliferation and drug-resistance, correlates poor prognosis (Deng et al., 2015;Huang et al., 2019). TBILA, controlled by TGF-beta, promotes NSCLC progression in vitro and in vivo by cis-modulating HGAL and positively regulating S100A7/JAB1 signaling (Lu Z. et al., 2018). Linc00662 exerts its oncogenic functions by directly interacting with Lin28, a potential diagnostic and therapeutic target for patients with lung cancer (Gong et al., 2018). CASC11 maintains the cancer cell stemness of NSCLC by increasing TGF-beta1 expression (Fu et al., 2019). HAND2-AS1 suppresses NSCLC cells migration and invasion and contributes to cell stemness by interacting with TGF-beta1 . Consequently, exploring stemness-related lncRNAs as promising targets for the diagnosis and prognosis of NSCLC is necessary and meaningful. Here, high DHRS4-AS1 expression in NSCLC tumors indicated good clinical outcomes, and this lncRNA was found to down-regulated in NSCLC tumors, cancer cells and NSCLC-derived stem cell spheres. Furthermore, DHRS4-AS1 significantly inhibited the cancer stem cell colony formation ability and stemness of NSCLC cells. Our data also revealed that DHRS4-AS1 abrogates the expression of stemness markers, including OCT4, SOX2, CD34, and CD133, suppressed the expression of N-cadherin, ZEB1, and Vimentin, and increased E-cadherin expression in mRNA and protein levels. These findings suggest that DHRS4-AS1 may serve as an NSCLC tumor suppressor by inhibiting cancer cell stemness, and DHRS4-AS1 might be a promising target for NSCLC treatment. Substantial evidence has reported that lncRNAs often function as miRNA sponges (Du et al., 2016;Furio-Tari et al., 2016;Song et al., 2019). Bioinformatics analysis and experiments data shown DHRS4-AS1 directly binds to miR-224-3p to regulate cancer cell stemness in NSCLC. Previous studies reported that miR-224-3p promotes breast cancer development by targeting FUT4 (Feng et al., 2016), promotes cervical cancer by repressing FIP200-mediated autophagy (Fang et al., 2016), and participates in the recurrence of human osteosarcoma (Xu et al., 2018). Nevertheless, the role of miR-224-3p in the stemness of NSCLC FIGURE 5 | Knockdown of DHRS4-AS1 attenuated anti-miR-224-3p-mediated tumor growth inhibition in vivo. Calu-3 cells were transfected with lenti-viruses meidatescramble, anti-miR-224-3p, si-NC, or si-DHRS4-AS1 for 48 h. The infected Calu-3 cells were collected and injected into nude mice (n = 3/group). (A) Tumor volumes were measured at the indicated times; #p < 0.01 vs. scramble+si-NC, ϕ p < 0.01 vs. anti-miR-224-3p+si-DHRS4-AS1. Tumors were collected (B) and measured (C) after injection on the 15th day. #p < 0.01 vs. scramble+si-NC, ϕ p < 0.01 vs. anti-miR-224-3p+si-DHRS4-AS1. (D,E) Western blot detected stemness-related and EMT-related markers in xenograft tumors; the optical densities of proteins were determined with ImageJ software; #p < 0.01 vs. scramble+si-NC. All results are expressed as the mean ± SD.
has not been explored. Here, miR-224-3p expression elevated in NSCLC tissues, cancer stem cells and cancer stem cell spheres. Furthermore, miR-224-3p knockdown inhibited cancer stem cell colony and sphere formation abilities. miR-224-3p decreased the expression of OCT4, SOX2, CD34, and CD133, and silencing miR-224-3p inhibited the expression of EMT-related factors, N-cadherin, ZEB1, and Vimentin but elevated E-cadherin expression in vitro and in vivo. DHRS4-AS1 acts as a tumor suppressor since it blocks the miR-224-3p-mediated silencing of TP53 and TET1, resulting in the inhibition of tumor growth in vivo. Our findings imply that the inhibition of miR-224-3p could improve NSCLC by suppressing cancer cell stemness.
In conclusion, DHRS4-AS1 functions as a tumor suppressor by reducing the cancer cell stemness of NSCLC, while miR-224-3p may serve as an oncogenic miRNA in NSCLC. The TP53and TET1-associated DHRS4-AS1/miR-224-3p axis constitutes NSCLC progression by modulated cancer cell stemness in vitro and in vivo. DHRS4-AS1 and miR-224-3p could be considered potential therapeutic targets for NSCLC.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Research Ethics Committee of the Affiliated Cancer Hospital of Nanjing Medical University. The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by the Research Ethics Committee of Chengdu Medical College.

AUTHOR CONTRIBUTIONS
LS and YW: conceptualization and supervision. FY, WZ, and XX: methodology. FY, WZ, XX, and XL: validation. SL: formal analysis and resources. SL, YW, FY, and WZ: investigation. CL and SL: data. LS: writing-original draft preparation. LS, FY, and WZ: writing-review, editing, and funding acquisition. All authors: have read and agreed to the published version of the manuscript.