Hepatic nuclear factor 4 alpha promotes the ferroptosis of lung adenocarcinoma via transcriptional activation of cytochrome P450 oxidoreductase

Cell culture

The Chinese National Collection of Authenticated Cell Cultures (Shanghai, China) provided the lung adenocarcinoma cell lines A549 and H23, and human embryonic kidney cell line 293T. Cells were grown in 10% fetal bovine serum-containing DMEM medium (with high glucose) (KeyGEN, Nanjing, Jiangsu, China).

Lentiviruses and compounds

GENECHEM (Shanghai, China) provided the negative control virus, sh-HNF4A-1 (5′- GCTTCTCTCCAAGGCTAAACT -3′), sh-HNF4A-2 (5′- GGACTAGCAAACTCTACAAAT-3′), and HNF4A overexpression viruses. (1S,3R)-RSL3 was purchased from MedChemExpress (NJ, USA), while imidazole ketone erastin (IKE) and deferoxamine (DFO) were both purchased from TOPSCIENCE (Shanghai, China).

Cytotoxicity assay

To measure the cell sensitivity to ferroptosis inducers, a total of 4,000 cells were seeded into each well of 96-well plates and subsequently treated with 0, 1, 2, 5, and 10 uM RSL3, or 0, 2, 5, 10, and 20 uM IKE. Then 10 ul alamaBlue (YEASEN, Shanghai, China) was added 48 h after treatments in each well. After 2 h of incubation, the fluorescence was monitored at 545 nm excitation and 590 nm emission wavelengths to reflect the cells’ sensitivity as previously described (Liang et al., 2023).

Lipid peroxidation assay

Cells were pretreated with 8 uM RSL3 or 16 uM IKE for 24 h, and then digested to single cell suspensions, incubated with 4 mM BODIPY 581/591 C11 (Thermo Fisher, Waltham, MA, USA) for 30 min at 37 °C. Cells’ fluorescence intensity was detected within the FITC channel using FACSAria III flow cytometry (BD, Franklin Lakes, NJ, USA), and then analyzed using FlowJo software (version VX, OR, USA).

Quantitative real-time PCR (qRT-PCR)

TRNzol Universal (TIANGEN, Beijing, China), Hifair^® III 1st Strand cDNA Synthesis SuperMix for qPCR (YEASEN, Shanghai, China), and Hieff^® qPCR SYBR Green Master Mix (Low Rox Plus) (YEASEN, Shanghai, China) were used to perform qRT-PCR as previously described (Liang et al., 2022). The expression was calculated using the 2^−ΔΔCT method, with β-actin serving as the reference gene. These were the primers were as follows: HNF4A-F:5′- CGAAGGTCAAGCTATGAGGACA -3′, HNF4A-R:5′- ATCTGCGATGCTGGCAATCT -3′ POR-F: 5′- CGGAACCACGCACTTTCATTTCTC -3′, and POR-R: 5′- TGCCAGCGGTGAGTGCTATCT-3′, β-actin-F: 5′- TGACGTGGACATCCGCAAAG-3′, β-actin-R: 5′- CTGGAAGGTGGACAGCGAGG-3′.

Western blot assay

Cells were lysed using RIPA Buffer (Beyotime, Jiangsu, China) containing 1 mM PMSF (Beyotime, Jiangsu, China). BCA Protein Quantification Kit (YEAEN) was used to determine the protein concentration. A total of 20 ug protein was separated using SDS-PAGE and then transferred to PVDF films. We utilized the HNF4A (1:1,000, CY10632; Abways, Shanghai, China) and POR (1:1,000, ab180597; Cambridge, UK) antibodies to investigate the protein expression, and GAPDH (1:3,000, AG019; Beyotime, Jiangsu, China) as the reference.

Chromatin immunoprecipitation (ChIP) assay

To examine the binding area in the POR’s promoter, we used ChIP-Seq data of HNF4A from the Encode database (https://www.encodeproject.org), and analyzed the data using the Integrative Genomics Viewer software (version 2.13.1, https://software.broadinstitute.org/software/igv/). Then, we employed ChIP-qPCR to validate the binding of HNF4A in the promoter of POR using SimpleChIP^® Plus Enzymatic Chromatin IP Kit (Magnetic Beads) (CST, Boston, MA, USA) and HNF4A Antibody (1:50, CY10632; Abways, New York, NY, USA). The primers used in ChIP-qPCR have the following sequences: primer_F: 5′- CGGAACCACGCACTTTCATTTCTC -3′, and primer_R: 5′- TGCCAGCGGTGAGTGCTATCT-3′.

Dual-luciferase reporter assay

The −1,000 to 200 bp of the POR’s promoter sequence (NM_001367562.3) was cloned into the firefly luciferase reporter plasmid after being retrieved from the National Center for Biotechnology Information database (https://www.ncbi.nlm.nih.gov/gene/). The probable HNF4A binding sequences in the POR’s promoter region were predicted using the JASPAR website (https://jaspar.genereg.net). Subsequently, another firefly luciferase reporter plasmid was cloned with the mutant POR’s promoter sequence, which had both probable HNF4A binding sites mutated. We transfected these plasmids into 293T cells with or without HNF4A overexpression to perform the dual-luciferase reporter assay using the Lipo8000 Transfection Reagent (Beyotime, Jiangsu, China) and the Luciferase Reporter Gene Assay Kit (Beyotime, Jiangsu, China) according to the manufacturer’s instruction. Renilla luciferase reporter plasmid was employed as the reference. The creation of each plasmid was handled by CENECHEM.

Statistical analyses

Student’s t-test and ANOVA tests were run using R software (version 4.0.3) to compare variables. A two-tail of p < 0.05 was regarded as significant. Bonferroni adjustment is used where multiple hypothesis testing is performed.

HNF4A expression significantly decreased in ferroptotic A549 cells

We discovered that the expression of HNF4A decreased by more than 50% in ferroptosis based on previously reported RNA-Seq data of A549 cells treated with the ferroptosis inducers RSL3 or IKE (Bi et al., 2022). We treated A549 cells with RSL3 and IKE with or without the ferroptosis inhibitor DFO, and then utilized qRT-PCR and western blot to detect the changes in HNF4A expression, in order to further confirm the decrease in HNF4A expression brought on by ferroptosis. The reduction in HNF4A expression brought on by ferroptosis inducers was prevented by DFO administration, as shown in Figs. 1B and 1C, indicating that HNF4A may be directly engaged in the cellular molecular changes in ferroptosis.

HNF4A promoters ferroptosis and POR’s expression in lung adenocarcinoma

To explore the function of HNF4A in ferroptosis, we sought to modify HNF4A expression in lung adenocarcinoma cells. Gene expression profiles of multiple commonly used lung adenocarcinoma cell lines were downloaded from the Cancer Cell Line Encyclopedia (CCLE) database (https://sites.broadinstitute.org/ccle/). As shown in Fig. 2A, HNF4A was strongly expressed in A549 and CALU3 cell lines and rarely expressed in H23, H358, H1650, H1975, HCC827, and PC9 cell lines.

Therefore, we knocked down the expression of HNF4A in A549 cells and overexpressed it in H23 cells (Figs. 2B and 2C), and investigated whether or not the cells’ susceptibility to inducers of ferroptosis changed. As shown in Fig. 2D, HNF4A knockdown significantly decreased the cytotoxicity of RSL3 and IKE to A549 cells, whereas HNF4A overexpression increased the lethality of ferroptosis inducers on H23 cells. In parallel, HNF4A knockdown increased the level of cellular lipid peroxidation in lung adenocarcinoma cells treated with ferroptosis inducers, while HNF4A overexpression had the opposite effect (Fig. 2E).

HNF4A, as a transcript factor, may promote ferroptosis by regulating the expression of its downstream target genes. Therefore, we obtained the data about HNF4A’s potential target gene from the gene transcription regulation database (GTRD) (http://gtrd.biouml.org/), and looked for ferroptosis-related genes among the potential target genes of HNF4A. We discovered that HNF4A most likely regulated the expression of POR, one of the key ferroptosis-promoting genes, with the highest peak count data in the GTRD data. Also, our qRT-PCR and western blot results demonstrated that HNF4A knockdown significantly reduced POR expression whereas HNF4A overexpression promoted POR expression in lung adenocarcinoma cells (Figs. 2B and 2C).

HNF4A promoted POR expression via binding to POR’s promoter

We analyzed the ChIP-seq data for HNF4A in the Encode database, and found that there were several binding peaks of HNF4A in the promoter region of POR (Fig. 3A). Using JARSPAR, we identified two potential binding sites for HNF4A within 100 bp before the POR transcription start site (TSS) (Fig. 3B), and designed a pair of primers around the two binding sites. Our ChIP-qPCR data confirmed that HNF4A was bound to the POR’s promoter region in A549 and H23 cells, as shown in Fig. 3C.

Dual luciferase assays were subsequently used to validate the regulation of POR’s expression by HNF4A. We created two different firefly luciferase plasmids: one with the POR’s promoter region and the other lacking different HNF4A binding sites (Fig. 3D). According to our findings, the expression of firefly luciferase with the POR’s promoter was considerably suppressed by HNF4A knockdown and significantly enhanced by HNF4A overexpression (Fig. 3E). The promoting effect of HNF4A on POR expression was mostly restored when both two HNF4A binding sites were mutated (Fig. 3E).

POR restoration blocked HNF4A’s promotion on ferroptosis in lung adenocarcinoma

To demonstrate that HNF4A promoted ferroptosis via POR in lung adenocarcinoma, we overexpressed POR in HNF4A-knockdown A549 cells, and knocked down POR expression in HNF4A-overexpressed H23 cells (Figs. 4A and 4B). We found that POR overexpression considerably prevented the influence of HNF4A-knockdown on the cytotoxicity of ferroptosis inducers as well as the level of cellular lipid peroxidation in A549 cells (Figs. 4C and 4D). Similar results were also observed in experiments in H23 cells (Figs. 4C and 4D). These results validated that HNF4A promoted ferroptosis via POR in lung adenocarcinoma.