Overexpression of LINC00551 promotes autophagy-dependent ferroptosis of lung adenocarcinoma via upregulating DDIT4 by sponging miR-4328

Cell culture

Human LUAD PC9 and A549 cells were obtained from the Chinese Academy of Science Cell Bank (Shanghai, China) and cultured in Roswell Park Memorial Institute 1640 medium (RPMI-1640; Gibco, Gaithersburg, MD, USA) supplemented with 10% fetal bovine serum (FBS; Gibco) in 5% CO₂ at 37 °C. HEK-293T cells were maintained in a 10% FBS DMEM medium. RSL-3 and Ferr-1 were purchased from Selleck (TX, USA).

Cell transfection

The plasmids carrying DDIT4 cDNA and shRNA were constructed in GeneChem (Shanghai, China). MiR-4328 Inhibitor was purchased from Rebobio (Guangzhou, China) and Mimics (miR-4328 mimics), and their controls (miR-NC) were synthesized by Sangon (Shanghai, China). All transfection procedures were performed using Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA). For further experiments, the cells were harvested for 48 h after transfection. Lentivirus infection resulted in the production of stable gene-expressing cells. These cells were infected by lentivirus for 72 h and treated with 2 µg/mL puromycin. The overexpression or shRNA efficiency was detected by quantitative real-time PCR (qRT-PCR). The sequences of mimics, inhibitors and shRNAs are listed in Table S2.

Quantitative real-time PCR (qRT-PCR)

Total RNA was isolated from tissue or cell samples using Trizol reagent (Invitrogen, NY, USA). The RNA samples were reverse transcribed into cDNA using RT Master Mix (MCE, NJ, USA). The levels of targeted gene mRNA transcripts relative to the control were quantified in triplicate by qRT-PCR in StepOne System™ (Applied Biosystems, Waltham, MA, USA) using SYBR Green qPCR Master Mix with High ROX (MCE, NJ, USA) and specific primers. The data were analyzed using the 2^−ΔΔCt method. Primers used for qPCR were: LINC00551, 5′-CAGCCTTCAGTTGGAGGAAC-3′ and 5′-TGGCCAATGACGAATACTGA-3′; DDIT4, 5′-GATGCCTAGCCAGTTGGTAAG-3′ and 5′-CTAAACAGCCCCTGGATCTTG-3′; β-Actin, 5′-AAAGACCTGTACGCCAACAC-3′ and 5′-GTCATACTCCTGCTTGCTGAT-3′. This project was approved by the Ethical Committee on Scientific Research of the Second Xiangya Hospital of Central South University, Grant number 2020462.

RNA sequencing

Three paired replicate RNA sequencing libraries were prepared from PC9 vector and PC9 LINC00551 overexpression cells. The six libraries were sequenced separately using the BGISEQ-500 sequencer, paired-end 150-bp read length, at 44 ×  10⁶ reads per sample. The statistical power of this experimental design, as calculated by RNApower (v 3.13), was 0.9. The clean reads were mapped to the reference genome using HISAT2 (v 2.0.4) (Kim, Langmead & Salzberg, 2015). Bowtie2 (v2.2.5) was applied to align the clean reads to the reference coding gene set, and then the expression level of the gene was determined by RSEM (v1.2.12) (Li & Dewey, 2011). The heatmap was drawn using the pheatmap package (v1.0.8), according to the gene expression in different samples. The differential expression analysis was performed using the DESeq2 (v1.4.5) with a Q value ≤ 0.05. Gene Set Enrichment Analysis (GSEA) was performed using the GSEA software. The sequence data was uploaded to NCBI GenBank with the following BioSample accessions: SAMN26553836, SAMN26553837, SAMN26553838, SAMN26553839, SAMN26553840, and SAMN26553841.

Cell viability assays

Corresponding cells (3 ×10³ cells/well) were cultured in 96-well plates for 24 h, and the cell viability was assessed using the Cell Counting Kit-8 (CCK-8, Topscience, Shanghai, China), according to the manufacturer’s instructions. After incubating at 37 °C for 3 h, the absorbance was measured at 450 nm using the TECAN sunrise spectrophotometer (Tecan, Männedorf, Switzerland).

Lipid reactive oxygen species (ROS) assays

The different A549 and PC9 cell groups were treated in triplicate with vehicle DMSO or RSL-3 (2.0 µM) for 48 h to determine the level of lipid ROS. The cells were then harvested and stained with 10 µM C11-BODIPY (581/591, ThermoFisher) for 30 min at 37 °C in the dark. After being washed, the cell samples were analyzed by a flow cytometer (BD, NJ, USA).

Iron assay

The total iron and Fe²⁺ in the indicated cell samples were measured using an Iron Assay Kit (Sigma Aldrich, MO, USA). Briefly, ∼2 × 10⁶ cells were harvested and homogenized with four volumes of iron assay buffer, followed by centrifugation at 13,000 × g for 10 min at 4 °C. The supernatants were collected, and the differently diluted samples (95 µL each) were reacted with 5 µL of iron assay buffer or iron reducer for 30 min at room temperature in the dark to measure ferrous iron or total iron, respectively. Subsequently, the reacted samples in each well were mixed with a 100 µL iron probe and incubated for 60 min at room temperature in the dark. The absorbance of each well was measured at 593 nm using the TECAN sunrise spectrophotometer (Tecan, Männedorf, Switzerland). The total iron levels were quantified using the standard curves based on varying concentrations of standard iron provided. The levels of ferric iron were calculated by deducting ferrous iron from total iron.

Western blot analysis

Cells were harvested, washed, and then lysed in the RIPA buffer (Beyotime, Shanghai, China) containing a cocktail of protease inhibitors (MCE, NJ, USA). After quantifying the protein concentration using a bicinchoninic acid (BCA) protein assay kit (Beyotime, Shanghai, China), the cell lysates (30 µg/lane) were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes. The membranes were blocked in 10% non-fat dry milk in Tris-Buffered Saline and Tween-20 (TBST) for 1 h and incubated overnight with primary antibodies at 4 °C. After being washed three times. The bound antibodies were reacted with a secondary antibody. They were then visualized with a fluorescence imaging system (Sagecreation, Beijing, China) using enhanced chemiluminescence (Advansta, CA, USA). The primary antibodies included anti- β-Actin (20536-1-AP), anti-GPX4 (14432-1-AP), anti-p62 (18420-1-AP), anti-SLC7A11 (26864-1-AP), anti-DDIT4 (10086-1-AP), anti-LC3 (Proteintech, 14600-1-AP), and anti-ATG5 (Zen-bio, R23497).

Transmission electron microscopy (TEM)

The different groups of cells were harvested, washed, and fixed in 2.5% glutaraldehyde and 2% paraformaldehyde (Merck, Darmstadt, Germany) and then post-fixed with 1% Osmium tetraoxide (OsO4). The cells were cut into ultrathin sections (70 nm), stained with uranyl acetate and lead citrate, and examined under a transmission electron microscope (Hitachi, Tokyo, Japan).

Dual-luciferase reporter assays

A dual-luciferase reporter gene assay kit (#11402ES60; Yeasen, Shanghai, China) was used to perform the dual-luciferase assay. The 3′-UTR of DDIT4 wild type (DDIT4-WT) or DDIT4 mutant type (DDIT4-MUT) was cloned into the pGLO luciferase vector and transfected into 293T cells using Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) with miR-NC or miR-4328 mimics. After 48 h, the luciferase activity of 293T was measured using a dual-luciferase reporter luminometer system (Promega Corporation, Madison, WI, USA).

Xenograft lung tumour model in nude mice

The animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) of Second Xiangya Hospital following the Guidelines of the Care and Use of Laboratory Animals issued by the Chinese Council on Animal Research. Briefly, female BALB/c nude mice at six weeks were obtained from Hunan SJA Laboratory Animal Co. Ltd. (Hunan, China) and kept in a specific pathogen-free environment. The mice were injected subcutaneously with 2 × 10⁶ indicated cells into the left or right flank for 21 days (PC9) or 28 days (A549) post-implantation. At the end of the experiment, the tumours were dissected and weighed.

Statistical analyses

All statistical analyses were performed using SPSS 22.0 software, and graphs were made using the GraphPad Prism 8.0 software. The data were presented as mean ± SD. The student’s t-test analyzed the difference between the two groups. A paired t-test (for continuous data) or a chi-square test (for categorical data) were also used. The survival of animals was estimated by the Kaplan–Meier method and analyzed by the log-rank test. Ns represented non-significant (p >  0.05), while^∗p < 0.05,^∗∗p < 0.01, and^∗∗∗p <  0.001 represented different degrees of significance.

Overexpressing LINC00551 promotes the RSL-3-induced ferroptosis and autophagy in LUAD cells

The level of LINC00551 transcripts in the TCGA was analyzed using the online tool GEPIA (http://gepia.cancer-pku.cn/index.html) (Tang et al., 2017). LINC00551 transcripts were significantly down-regulated in LUAD tissues compared to the non-tumour tissues (Fig. 1A). LINC00551 transcript levels were detected in 62 pairs of human clinical LUAD and adjacent normal tissues, and the results revealed that LINC00551 was poorly expressed in LUAD tissues (Fig. 1B). Stratification of 672 LUAD tissues with a cut-off value of median expression revealed that LUAD patients with lower levels of LINC00551 transcripts were associated with worse overall survival (OS) than those with higher levels of LINC00551 transcripts (Fig. 1C). LINC00551 generally displays lower expression in lung adenocarcinoma cell lines than normal lung epithelial cell line BEAS-2B (Fig. 1D). LINC00551 overexpressing PC9 and A549 cells were constructed to examine the impact of LINC00551 on ferroptosis (Fig. 1E). After being exposed to the ferroptosis inducer, RSL-3, the overexpression of LINC00551 significantly reduced the cell viability, but a ferroptosis inhibitor like Ferrostatin-1 (Fer-1) was able to restore this effect (Figs. 1F and 1G). In addition, LINC00551 over-expression significantly increased the concentrations of Fe²⁺ (Figs. 1H and 1I) and lipid ROS (Figs. 1J and 1K) in PC9 and A549 cells. Solute carrier family 7 member 11 (SLC7A11, also known as xCT) is a cystine/glutamate antiporter that imports extracellular cysteine for glutathione biosynthesis, ROS detoxification, and antioxidant activity. Glutathione peroxidase 4 (GPX4) is a lipid repair enzyme that represses ferroptosis by reducing lipid ROS. Both SLC7A11 and GPX4 are key repressors of ferroptosis. The decreased expression of SLC7A11 and GPX4 is generally considered a hallmark of ferroptosis. In this study, we found that overexpressing LINC00551 significantly reduced the protein expression levels of SLC7A11 and GPX4 (Fig. 1L), indicating that LINC00551 promotes ferroptosis in LUAD.

Recent evidence indicates that the molecular machinery of some selective autophagy (such as macroautophagy, a lysosome-mediated degradation process) facilitates ferroptosis (Hou et al., 2016). Excessive activation of autophagy seems to be an important driver of ferroptosis (Kang & Tang, 2017; Liu et al., 2020). The formation of autophagosomes is triggered by LC3 switching from LC3-I to -II (Kabeya et al., 2000). Whereas the autophagy substrate p62, an essential receptor for selective autophagy, functions by simultaneously interacting with LC3 (Gatica, Lahiri & Klionsky, 2018). Interestingly, RSL-3 exhibits broader and more vigorous activity in the upregulation of LC3-II or downregulation of p62 compared to other inducers (Li et al., 2021). We further detected the LC3II/I and p62 levels in RSL-3-induced cells. The results displayed that overexpression of LINC00551 significantly increased the conversion ratios of LC3-II to -I but decreased the relative levels of p62 expression in LINC00551 overexpressing PC9 and A549 cells (Fig. 1L). According to TEM analysis, RSL-3 treated LUAD cells depicted relatively smaller mitochondria with condensed membrane densities, a reduction of mitochondrial crista (red arrow), and an increase in the number of lysosomes (blue arrow) (Fig. 1M). Furthermore, tumor xenograft in nude mice exhibited a pronounced decline in size and weight of subcutaneous tumors in nude mice after overexpression of LINC00551 (Figs. 1N–1Q). These results indicated that the overexpression of LINC00551 promoted the RSL-3-induced ferroptosis and autophagy in LUAD cells.

LINC00551 regulates DDIT4 expression by sponging miR-4328

RNA sequencing was performed to investigate the potential molecular mechanism of LINC00551 in LUAD. The major genes up-regulated by LINC00551 are illustrated in Fig. 2A, most of which were primarily enriched in DNA repair and cell cycle, according to GSEA and KEGG analyses (Figs. 2B and 2C). DNA repair is an important part of the DNA damage response (DDR) network and is also one of the regulatory factors of cell cycle arrest (Li, Pearlman & Hsieh , 2016). Previously, we demonstrated that LINC00551 could regulate autophagy and ferroptosis in LUAD. Autophagy, a physiological self-eating process, is involved in DDR (Roshani-Asl et al., 2020). According to several recent investigations, it has been proven that DDR participates in ferroptosis (Chen, Tseng & Chi, 2020). DDIT4 is one of the main differentially expressed genes. It is a key regulator of rapamycin complex 1 (mTORC1) (Brugarolas et al., 2004), playing an important role in the regulation of DNA repair and RCDs, such as autophagy apoptosis and ferroptosis. RNA sequencing analysis revealed that the expression of DDIT4 mRNA was significantly increased in the LINC00551 overexpressing PC9 cells (Fig. 2A). Notably, the ferroptosis-related gene CHAC1 was increased in LINC00551 overexpressing PC9 cells (Fig. 2A). In LINC00551 overexpressing PC9 cells, the expression of DDIT4 was confirmed at both the mRNA and protein levels (Figs. 2D and 2E). Moreover, a positive correlation was found between LINC00551 and DDIT4 in the TCGA dataset (Fig. 2F). Bioinformatic analysis using Targetscan (http://www.targetscan.org/) and Starbase database (http://starbase.sysu.edu.cn/index.php) suggested that miR-4328 could bind to both LINC00551 and DDIT4 (Fig. 2G). Transfection with miR-4328 mimics significantly reduced the relative levels of DDIT4 mRNA and LINC00551 transcripts, while transfection with miR-4328 inhibitor significantly increased their levels (Figs. 2H and 2I). The dual-luciferase assay revealed that the luciferase activity of wild-type DDIT4 (DDIT4 WT) or wild-type LINC00551 (LINC00551 WT) was reduced by transfecting with miR-4328 mimics than the negative control group (miR-NC), but had no effect on the luciferase activities when mutated DDIT4 (DDIT4 MUT) or mutated LINC00551 (LINC00551 MUT) were transfected with it (Figs. 2J and 2K). PC9 cells co-transfected with miR-4328 mimics and LINC00551 partially restored the loss of DDIT4 gene in cells transfected with miR-4328 mimics only (Fig. 2L). These results indicated that miR-4328 can bind with both LINC00551 and DDIT4 and that LINC00551 regulates DDIT4 by acting as a ceRNA that competes with miR-4328 for binding.

LINC00551 promotes autophagy and ferroptosis by regulating DDIT4

To uncover the contributions of DDIT4 to LINC00551-induced effects, LINC00551 was transfected into PC9 cells with or without DDIT4 silencing (Fig. 3A). The results displayed that DDIT4 silencing significantly recovered the viability inhibition caused by LINC00551 overexpression (Fig. 3B). Moreover, DDIT4 silencing mitigated the levels of Fe²⁺ and lipid ROS caused by LINC00551 overexpression (Figs. 3C and 3D). As DDIT4 has been reported to inhibit mTOR activity (Brugarolas et al., 2004), the total/phospho(p)-mTOR was detected. The activity of mTOR was inhibited in LINC00551 overexpressed PC9 cells, and this effect was abrogated when DDIT4 was knocked down. Moreover, overexpression of LINC00551 remarkably increased the conversion ratios of LC3-II to -I but decreased the relative levels of SLC7A11, GPX4, and p62 expression in PC9 cells; these effects were abolished with the DDIT4 silencing (Fig. 3E). These data suggested that LINC00551 promotes RSL-3-induced autophagy and ferroptosis by regulating DDIT4.

The LINC00551/DDIT4 axis regulates ferroptosis in an autophagy- dependent manner

According to recent research, some types of autophagy may cause ferroptosis, and this process may be dependent on autophagy. However, the connections between the LINC00551/DDIT4 regulated autophagy and ferroptosis remain indistinct. Chloroquine (CQ), an autophagy inhibitor, or Ferr-1, can significantly reverse RSL-3-induced ferroptosis in LINC00551 over-expressed PC9 cells. This suggests that blocking autophagy diminishes the effect of LINC00551 over-expression on RSL3-induced ferroptosis in PC9 cells (Fig. 4A). Additionally, the treatment with CQ abrogated the effect of LINC00551 over-expression on autophagy and ferroptosis markers in PC9 cells (Fig. 4B). Meanwhile, the silencing of ATG5, a key regulator of autophagy, partially rescued the RSL-3-induced ferroptosis and eliminated the effect of DDIT4 overexpression on the RSL-3-induced ferroptosis in PC9 cells (Fig. 4C). ATG5 silencing attenuated the enhanced impact of DDIT4 overexpression on the RSL-3 increased concentrations of Fe²⁺ and lipid ROS in PC9 cells (Figs. 4D and 4E). Simultaneously, ATG5 silencing also abrogated the RSL-3-changed LC3-II formation and expression of p62, SLC7A11, and GPX4 in PC9 cells (Fig. 4F). These findings demonstrated that LINC00551/DDIT4 axis regulates the ferroptosis of LUAD cells in an autophagy-dependent manner.