Phorbol 12-Myristate 13-Acetate (PMA; cat. no. HY-18739), FerroOrange dye (cat. no. HY-D1913), BODIPY 581/591 C11 dye (cat. no. HY-D1301), tin protoporphyrin IX dichloride (SnPPIX, cat. no. HY-101194) and deferoxamine (DFO; cat. no. HY-B1625) were purchased from MedChemExpress. RPMI-1640 medium (cat. no. 11875093), Trypsin-EDTA (cat. no. 25200056), FBS (cat. no. A5670701), penicillin-streptomycin (cat. no. 15070063), 7-aminoactinomycin D (7-AAD, cat. no. 00-6993), PE-anti-CD206 (cat. no. 12-2069-42), APC-anti-CD163 (cat. no. 17-1639-42), Alexa Fluor 568-goat anti-rabbit IgG (cat. no. A-11011), CD163 polyclonal antibody (cat. no. PA5-109340) and formic acid (cat. no. 036617.36) were purchased from Thermo Fisher Scientific, Inc. Alexa Fluor 647-anti-heme oxygenase 1 (HO-1; cat. no. ab237268) was purchased from Abcam. The cytokines IL-4 (cat. no. 200-04-5UG) and IL-13 (cat. no. 200-13-10UG) were purchased from PeproTech Inc. (Thermo Fisher Scientific, Inc.). Hb (cat. no. CSB-NP002401h) was purchased from Cusabio Technology LLC.

The Gene Expression Profiling Interactive Analysis (GEPIA) database (gepia.cancer-pku.cn) was used to examine the correlation between the expression levels of ALB, A2M, LTF, IL6, HP, CP, APOA1 and SERPINA1 and the prognosis of patients with LUSC based on data from The Cancer Genome Atlas (TCGA; portal.gdc.cancer.gov). Additionally, pan-cancer analysis was performed on the HP gene. Differentially expressed genes (DEGs) between LUSC and normal lung tissues were analyzed using TCGA and Genotype-Tissue Expression data, along with a comparison of HP expression between tumor and normal tissues. A GeneCards database (genecards.org) search for ‘acute phase proteins’ yielded 107 APP genes with relevance scores ≥1. The R package intersect function (version 1.2.0; CRAN.R-project.org/package=dplyr) was used to analyze the intersection between LUSC DEGs and APP genes. Gene interactions were examined using the STRING database (cn.string-db.org), focusing on the top 481 genes interacting with HP. Key node genes were identified using the cytoHubba plugin (version 0.1; Institute of Information Science at Academia Sinica) with six topological network algorithms [Maximal Clique Centrality (MCC), Maximum Neighborhood Component (MNC), radiality, closeness, Edge Percolated Component (EPC) and degree] in Cytoscape software (version 3.9.0; Cytoscape Consortium). Pearson tests in the LinkedOmics database (linkedomics.org) were performed on LUSC samples from TCGA to statistically analyze HP-associated genes, identifying 3,763 genes with a false discovery rate (FDR) <0.001. Using the intersect function from the R package, the intersection between HP-associated genes and HP-interacting genes was assessed. The enrichGO and enrichKEGG functions (version 4.18.4; bioconductor.org/packages/release/bioc/html/clusterProfiler.html) in R (version 4.2.1; R Core Team) were used to determine the Gene Ontology (GO) terms and perform Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis associated with the HP gene in LUSC. Ferroptosis-related genes were obtained from the KEGG database (genome.jp/kegg). For GO analysis, significance was defined as q-value <0.001 for BP and q-value <0.01 for CC and MF after Benjamini-Hochberg correction, while for KEGG analysis, a q-value <0.05 was considered significant. The LUSC datasets GSE191881 and GSE74706 from Gene Expression Omnibus (GEO; ncbi.nlm.nih.gov/geo), along with gene expression profiles of LUSC tissues and normal lung tissues from TCGA, and relative HP protein expression levels from The Human Protein Atlas database (version 25.0; proteinatlas.org/ENSG00000257017-HP/cancer/lung+cancer#cptac_lung_sqcc), were obtained for analysis. The LM22 gene signature matrix, which contains characteristic genes of M2 macrophages, was obtained from a previous study (23). The infiltration levels of M2 macrophages in LUSC tissues were evaluated using the CIBERSORT (https://github.com/Moonerss/CIBERSORT; accessed in September 2025) algorithm.

Venn diagrams were drawn using the VennDetail function (version 1.26.0; bioconductor.org/packages/VennDetail) in R, along with the Venn function (version 1.12; CRAN.R-project.org/package=venn) and ggvenn function (version 0.1.19; CRAN.R-project.org/package=ggvenn). GO and KEGG enrichment circle plots were generated using OmicShare Tools (omicshare.com; accessed in September 2025).

Human normal lung epithelial cells BEAS-2B were obtained from Jennio Biotech Co., Ltd. and were verified using STR analysis. The lung squamous cell lines (NCI-H226, NCI-H520, NCI-H1703 and NCI-H2170) and the human pro-monocyte cell line THP-1 were purchased from Procell Life Science & Technology Co., Ltd. and were verified using STR analysis. BEAS-2B cells were cultured in bronchial epithelial cell-specific medium (Jennio Biotech Co.), which was already supplemented with dThe NCI-H226, NCI-H520, NCI-H1703, NCI-H2170 and THP-1 cells were cultured in RPMI-1640 (Thermo Fisher Scientific, Inc.) medium supplemented with 1% penicillin-streptomycin and 10% FBS (Thermo Fisher Scientific, Inc.). All cells were maintained at 37°C and 5% CO₂. THP-1 cells were differentiated into M2 macrophages as described previously (24). Briefly, THP-1 cells were treated with 150 nM Phorbol 12-Myristate 13-Acetate for 24 h, followed by culture in fresh RPMI-1640 for 24 h. The macrophages were polarized into M2 macrophages by incubating them with IL-4 (20 ng/ml) and IL-13 (20 ng/ml) for 24 h. The cells were stained with APC-anti-CD163 and PE-anti-CD206 for flow cytometry analysis using the CyFlow Cube 6 system (Sysmex Europe SE) to confirm M2 macrophage polarization.

Total RNA was extracted from BEAS-2B and LUSC cell lines (NCI-H226, NCI-H520, NCI-H1703, and NCI-H2170) using TRIzol^® reagent (cat. no. 15596018CN; Thermo Fisher Scientific, Inc.) and treated with DNase I to eliminate DNA contamination. Subsequently, single-stranded cDNA was synthesized from RNA using reverse transcriptase M-MLV (cat. no. 6110B; Takara Bio, Inc.). The following primer sequences used for amplification of HP and GAPDH were as follows: HP forward (F), 5′-CAGCACAGTCCCCGAAAAGAA-3′ and reverse; 5′-CAGTCGCATACCAGGTGTCC-3′; and GAPDH F, 5′-GGAGCGAGATCCCTCCAAAAT-3′ and R, 5′-GGCTGTTGTCATACTTCTCATGG-3′; primers were synthesized by Takara Bio, Inc. qPCR analysis was performed using UltraSYBR Green Master Mix (cat. no. CW0957H; Jiangsu Kangwei Century Biotechnology Co., Ltd.) to measure the mRNA expression levels of HP. Cq values for HP were normalized to GAPDH, and quantitative analysis was performed using the 2^−ΔΔCq method (25). The PCR amplification conditions were based on a published report (26). The thermocycling program consisted of an initial denaturation at 95°C for 1 min (1 cycle), followed by 40 cycles of denaturation at 95°C for 5 sec, annealing at 60°C for 10 sec and extension at 72°C for 15 sec, with a final extension at 72°C for 10 min (1 cycle). RT-qPCR analysis was performed using rTaq (Takara Bio, Inc.) in a DNA thermal cycler (Maxygen).

HP levels in BEAS-2B and LUSC cell lines (NCI-H226, NCI-H520, NCI-H1703, and NCI-H2170) were determined using an HP assay kit (cat. no. E-EL-H6199; Wuhan Elabscience Biotechnology Co., Ltd.) according to the manufacturer's protocol. Culture supernatants from cells in the logarithmic growth phase were collected and diluted 1:10,000. The diluted test samples, standard solutions and blank diluent were added into microplate wells at 100 µl per well, and then incubated at 37°C for 90 min. The intracellular fluid was then discarded, and 100 µl biotinylated antibody working solution was added to each well, followed by incubation at 37°C for 60 min. The plate was then washed three times with detergent, and 100 µl HRP enzyme conjugate working solution was added to each well. The plate was incubated at 37°C for 30 min, followed by five washes. Then, 90 µl 3,3′,5,5′-Tetramethylbenzidine substrate solution was added to each well and incubated at 37°C in the dark for 15 min before adding 50 µl stop solution to terminate the reaction. Finally, the optical density (OD) value of each well was measured at a wavelength of 450 nm using an ELISA reader (ELX 808IU; Omega Bio-Tek, Inc.), and protein concentrations were calculated using a standard curve.

Recombinant lentiviral vectors overexpressing HP (LV-HP; GV492-Ubc-HP-3FLAG) and a control vector (LV-Ctrl; GV492-Ubc-MCS-3FLAG) were designed and synthesized by Shanghai GeneChem Co., Ltd. A third-generation self-inactivating lentiviral packaging system was used. HEK293T cells (Shanghai GeneChem Co., Ltd.) were co-transfected with 20 µg of GV vector plasmid, 15 µg of pHelper 1.0 and 10 µg of pHelper 2.0 (mass ratio of 20:15:10). The transfection mixture was incubated at room temperature for 15 min and added to the cells. After 6–8 h at 37°C, the medium was replaced. Lentiviral particles were harvested 48 h post-transfection, concentrated by ultracentrifugation at 53,000 × g for 2 h at 4°C, and resuspended in PBS.

To establish HP overexpression cells, NCI-H226 and NCI-H520 cells were infected with LV-HP or LV-Ctrl (multiplicity of infection, 50). The cells were plated at a density of 5×10⁴ cells/well in a 6-well plate and cultured until reaching 30% confluence. Subsequently, the cells were infected with 200 µl lentivirus and 40 µl transfection enhancer HiTransG A for 16 h. Following this, the medium was replaced with fresh complete culture medium for further cultivation. After 72 h post-infection, the efficiency of HP overexpression was assessed in cells and culture supernatant using RT-qPCR and ELISA, respectively.

An LDH cytotoxicity assay kit (cat. no. HY-K1090; MedChemExpress) was used to assess LDH levels in the supernatant of LUSC cell lines (NCI-H226 and NCI-H520 cells) and M2 macrophages, according to the manufacturer's protocol. LUSC cells infected with LV-HP or LV-Ctrl at a concentration of 1×10⁵ cells/ml were seeded in a 96-well plate at 100 µl per well and incubated with or without Hb treatment for 24 h. Similarly, M2 macrophages at a concentration of 1×10⁵ cells/ml were plated in a 96-well plate at 100 µl per well and treated with or without Hb in the presence of conditioned media (CM) from LUSC cells infected with LV-HP or LV-Ctrl for 24 h. After washing with PBS, 20 µl 0.4 mol/l lactic acid solution, 20 µl 4 mmol/l 2-p-iodophenyl-3-p-chloronitrobenzenetetrazole solution and 20 µl reaction solution were added to the cells. The samples were incubated at room temperature for 30 min, and the OD value was measured using an ELISA reader with a detection wavelength of 492 nm and a reference wavelength of 650 nm. The fold change for each sample (including the samples in the control group) was calculated by dividing the value of each sample by the mean value of the control group.

Flow cytometry was used to quantify 7-AAD-positive dead cells, intracellular Fe²⁺ and lipid peroxide levels, CD163- and CD206-double-positive cells and HO-1-expressing cells. To assess the proportion of 7-AAD-positive cells, and the levels of intracellular Fe²⁺ and lipid peroxides in NCI-H226 and NCI-H520 cells, cells infected with LV-HP or LV-Ctrl were seeded at a concentration of 2×10⁵ cells per well in 6-well plates, with or without the addition of Hb, and incubated at 37°C for 24 h. To determine the percentage of 7-AAD-positive cells, NCI-H226 and NCI-H520 cells treated as aforementioned were washed with PBS. Subsequently, the cells were resuspended in PBS containing 1 µg/ml 7-AAD (cat. no. 00-6993; Thermo Fisher Scientific, Inc.) and incubated at room temperature in the dark for 10 min. Following trypsin digestion, the cells were washed with PBS, centrifuged at 4°C at 200 × g for 5 min to collect the cells, and the percentage of 7-AAD-positive cells was analyzed using flow cytometry. To determine the concentration of Fe²⁺, NCI-H226 and NCI-H520 cells, treated as aforementioned, were washed with PBS and then incubated at room temperature in the dark in 1 µmol/l FerroOrange (cat. no. HY-D1913; MedChemExpress) working solution for 30 min. Subsequently, the fluorescence intensity of FerroOrange-stained cells was measured using flow cytometry. The mean fluorescence intensity of each sample was then calculated and compared with the control group. To assess lipid peroxidation levels, NCI-H226 and NCI-H520 cells were treated with LV-HP or LV-Ctrl, washed with PBS, and then incubated in the dark at room temperature with fresh culture medium containing 2 µm BODIPY581/591C11 dye (cat. no. HY-D1301; MedChemExpress) for 30 min. Following trypsin digestion, cells were washed with PBS, collected by centrifugation at 4°C, 200 × g for 5 min, and the fluorescence intensity of BODIPY581/591C11-stained cells was measured using flow cytometry.

To identify the induction of M2 macrophages, THP-1 cells were treated with 150 nM Phorbol 12-Myristate 13-Acetate (cat. no. HY-18739; MedChemExpress) for 24 h, cultured in fresh RPMI-1640 for 24 h, and then stimulated with IL-4 (20 ng/ml; cat. no. 200-04-5UG; Thermo Fisher Scientific, Inc.) and IL-13 (20 ng/ml; cat. no. 200-13-10UG; Thermo Fisher Scientific, Inc.) for an additional 24 h, all at 37°C. Subsequently, the cells were washed with PBS and then incubated with APC-anti-CD163 (1:500; cat. no. 17-1639-42; Thermo Fisher Scientific, Inc.) and PE-anti-CD206 (1:500; cat. no. 12-2069-42; Thermo Fisher Scientific, Inc.) at 4°C in the dark for 30 min. Following a final wash with PBS and centrifugation, flow cytometry was used to analyze CD163- and CD206-double-positive cells. To assess the proportion of 7-AAD-positive cells and the levels of intracellular Fe²⁺ and lipid peroxides in M2 macrophages, culture supernatants from NCI-H226 and NCI-H520 cells infected with LV-HP or LV-Ctrl were used as CM. M2 macrophages were cultured with the CM in combination with other reagents for 24 h. Subsequently, the M2 macrophages were stained with 7-AAD, FerroOrange or BODIPY581/591C11 dyes, followed by flow cytometry analysis. To assess the expression of HO-1 in M2 macrophages, 2×10⁶ M2 macrophages were fixed in 0.25% cold paraformaldehyde (cat. no. P0099; Beyotime Biotechnology) at 4°C for 1 h, followed by lysis with 0.1% Triton X-100 (cat. no. HFH10; Thermo Fisher Scientific, Inc.) for 10 min at room temperature in the dark. The cells were incubated with an Alexa Fluor 647-anti-HO-1 antibody (cat. no. ab237268; Abcam) at room temperature for 1 h prior to flow cytometry analysis. All flow cytometry analyses were conducted on a CyFlow Cube 6 system. FlowJo (version 10; BD Biosciences) was used to analyze the data, employing gating strategies to exclude debris and select single cells based on the forward- and side-scatter characteristics.

M2 macrophages were inoculated into a 6-well plate and co-cultured with CM and Hb for 24 h. The cells were then fixed in 2.5% glutaraldehyde in PBS at 4°C for 4 h and subjected to three 15-min washes with PBS. Post-fixation was performed using 1% OsO₄ in PBS at 4°C for 1.5 h, followed by three additional 15-min washes with PBS. Dehydration was performed using sequential immersions in ethanol solutions (30, 50, 70 and 80%), followed by acetone solutions (90 and 95%) for 15 min each. Absolute acetone dehydration was then performed twice for 20 min each. Subsequently, the cells were incubated in a 1:1 mixture of absolute acetone and the final Spurr resin for 1 h at room temperature, followed by a 1:3 mixture for 2 h, and finally in a mixture of the final Spurr resin and absolute acetone overnight. Following embedding, ultrathin sections, sectioned at 60 nm thickness were prepared, stained and imaged using a transmission electron microscope (cat. no. H-7650; Hitachi, Ltd.).

A total of 5×10³ M2 macrophages per well was inoculated into a 96-well plate and co-cultured with CM and Hb, either alone or in combination with a CD163 polyclonal antibody (1:50-100; cat. no. PA5-109340; Thermo Fisher Scientific, Inc.), for 24 h. Intracellular heme content was determined following established protocols (27). Briefly, cells were washed three times with PBS, centrifuged at 200 × g for 5 min and then resuspended in 180 µl concentrated formic acid. The heme concentration of the formic acid solution was determined spectrophotometrically at 400 nm.

Experimental data were analyzed using GraphPad Prism software version 6.0 (Dotmatics). Survival curves were generated using the Kaplan-Meier method, and differences between groups were assessed by the log-rank test. Correlation analysis was performed using the Pearson correlation coefficient. A one- or two-way ANOVA with a Tukey's post hoc test was used to compare differences among multiple groups. To determine the difference between two groups, an unpaired t-test was used. All data are presented as the mean ± SD. P<0.05 was considered to indicate a statistically significant difference.

To investigate the expression of APPs in LUSC and its correlation with prognosis, a total of 107 APP genes were initially identified with a relevance score of ≥1 from the GeneCards database (Table SI). From the GEPIA database, 5,944 DEGs were identified between LUSC tumors and normal lung tissues (Table SII). Intersection analysis revealed 46 overlapping APP and DEG genes (Fig. 1A). STRING database analysis demonstrated significant network interactions among 44 of these 46 APP genes (Fig. 1B). Using six topological network algorithms, including MCC, MNC, degree, EPC, closeness and radiality from the Cytoscape plugin cytoHubba, eight consistently top-ranked genes were identified: ALB, A2M, LTF, IL6, HP, CP, APOA1 and SERPINA1 (Table SIII and Fig. 1C). Among the eight genes, Kaplan-Meier survival analysis using GEPIA data showed that only HP (P=0.049) and A2M (P=0.0043) were significantly correlated with the prognosis of patients with LUSC, while the remaining six genes did not demonstrate statistically significant prognostic associations (Fig. 1D). A previous study identified A2M as a potential therapeutic target for NSCLC (13). However, the specific role of the HP gene in predicting prognosis in NSCLC remains unexplored. Through pan-cancer analysis using the GEPIA database, it was found that HP was associated with the prognosis of LUSC and five other cancer types: Stomach adenocarcinoma (STAD), skin cutaneous melanoma (SKCM), brain lower grade glioma (LGG), kidney renal papillary cell carcinoma (KIRP) and kidney renal clear cell carcinoma (KIRC) (Fig. 1E). Analysis of the GEPIA database revealed a significant differential expression in HP gene expression in LUSC tissues compared with that of normal lung tissues, with no significant differences observed in the other five tumor types (Fig. 1F). These findings collectively indicated that HP was a hub gene in LUSC progression.

To investigate HP expression in LUSC, gene expression profiles were obtained from GEO and TCGA. Findings from GEO (Fig. 2A) and TCGA (Fig. 2B) revealed a significant downregulation of HP gene expression in LUSC tumor tissues compared with that of normal lung tissues. Subsequent RT-qPCR analysis of HP gene expression in the normal lung epithelial cell line BEAS-2B and LUSC cell lines (NCI-H226, NCI-H520, NCI-H1703, NCI-H2170) showed significant decreases in the LUSC cell lines compared with that of BEAS-2B (Fig. 2C). Examination of the Human Protein Atlas database showed a notable decrease in HP protein expression in LUSC tumor tissues compared with that of normal lung tissues (Fig. 2D). ELISA analysis also confirmed a decrease in HP protein secretion levels in LUSC cell culture supernatant compared with that from the normal lung epithelial cell line BEAS-2B (Fig. 2E). These results indicated a significant downregulation of the HP gene and protein in LUSC tumor tissues and cells.

An examination of potential signaling pathways influenced by HP in LUSC was conducted using TCGA data in the LinkedOmics database. The analysis identified 3,763 genes linked to HP with an FDR threshold of <0.001 (Table SIV). In the STRING database, by setting the maximum number of genes in the interactome to 500 and filtering for interactions with a score >0.4, 481 genes, directly or indirectly, interacted with HP (Table SV). When comparing the 3,763 HP-associated genes with the 481 interacting genes, it was found that 191 genes were both associated with and interacted with HP (Fig. 3A). GO enrichment analyses (Table SVI), visualized using the OmicShare platform, revealed 20 enriched GO terms across biological processes, cellular components and molecular functions (Fig. 3B, upper panel). The biological processes included lipid metabolism (‘neutral lipid metabolic process’, ‘triglyceride metabolic process’ and ‘arachidonic acid secretion process’) and oxidation processes (‘response to oxidative stress’ and ‘ROS metabolic process’). Cellular component enrichment was observed in lipid components, specifically the outer mitochondrial membrane. Molecular functions included ‘oxygen binding’, enzyme activities (‘oxidoreductase’ and ‘peroxidase’), ‘lipoprotein lipase activator activity’ and ‘antioxidant activity’. The GO enrichment analysis indicated that the 191 identified genes were predominantly involved in oxidative stress responses associated with ferroptosis. This finding was corroborated by the KEGG pathway enrichment analysis, which showed a significant enrichment of these genes in the ferroptosis pathway (Fig. 3B, lower panel). Analysis of the KEGG database using the keyword ‘Ferroptosis’ identified 33 genes involved in the ferroptosis pathway (Table SVII). By comparing the 191 genes associated with and interacted with HP and the 33 ferroptosis pathway genes, four intersecting genes were identified: HMOX1, SLC40A1, CP and TF (Fig. 3C). Pearson correlation analysis revealed significant associations between HP and these four genes, with correlation coefficients of 0.23 (P=2.19×10⁻⁷), 0.31 (P=1.47×10⁻¹²), 0.48 (P=3.98×10⁻³⁰) and 0.27 (P=6.87×10⁻¹⁰), respectively. These results suggested that these four genes may serve essential roles as regulators in the HP-induced ferroptosis pathway.

Given the low expression of HP in LUSC cells, lentiviral vectors were constructed to overexpress HP in the NCI-H226 and NCI-H520 LUSC cell lines. After 72 h, cells and culture supernatants were collected to quantify HP mRNA and protein expression using RT-qPCR and ELISA, respectively. Compared with the LV-Ctrl group, LV-HP infection significantly increased HP mRNA (Fig. 4A) and protein (Fig. 4B) expression levels, confirming the successful generation of lentiviral vectors expressing HP for subsequent experiments. Ferroptosis is a form of cell death triggered by iron-dependent lipid peroxidation, which increases membrane permeability (28). To study the influence of HP on membrane permeability in LUSC NCI-H226 and NCI-H520 cells, LDH release and 7-AAD assays were performed in these cells treated with or without LV-HP in the presence or absence of Hb. It was observed that there were no significant differences in LDH release (Fig. 4C) and 7-AAD positive cells (Fig. 4D) among the LV-Ctrl, LV-HP, LV-Ctrl/Hb, and LV-HP/Hb groups. Flow cytometry analysis showed no significant differences in intracellular Fe²⁺ (Fig. 4E) and lipid peroxides (Fig. 4F) among the cells in the LV-Ctrl, LV-HP, LV-Ctrl/Hb and LV-HP/Hb groups. Thus, it was demonstrated that the overexpression of HP did not induce ferroptosis in the LUSC cells.

Previous studies have elucidated the roles of HP, HMOX1, SLC40A1, CP and TF in regulating iron recycling in M2 macrophages, with CD163 serving as the receptor (29,30). To assess the correlation between HP expression and M2 macrophage abundance, LUSC samples from the TCGA database were stratified into high- and low-HP expression groups based on median HP expression levels (cut-off value, 50th percentile). An analysis of immune infiltration in the LUSC gene expression profile was then conducted to assess M2 macrophage levels in these groups. The high-HP expression group exhibited a significantly higher abundance of M2 macrophages compared with that of the low-HP expression group (Fig. S1), indicating a positive correlation between HP expression and M2 macrophage abundance. Furthermore, LM22 data containing characteristic genes of M2 macrophages from a previous study was obtained (23) and analyzed the correlation between HP and characteristic genes of M2 macrophages. The results showed significant associations between HP and marker genes of M2 macrophages, such as ACP5, AIF1 and C3AR1, with a Pearson correlation of 0.30 (P=8.16×10⁻¹²), 0.28 (P=8.15×10⁻¹¹), and 0.21 (P=1.37×10⁻⁶), respectively. These results demonstrated a positive correlation between HP gene expression and the abundance of M2 macrophages.

It was next investigated whether the overexpression of HP in LUSC cells induced ferroptosis in M2 macrophages. THP-1 cells were differentiated into M2 macrophages, as confirmed by flow cytometry demonstrating a CD206⁺CD163⁺ phenotype in ~80% of cells (Fig. 5A). Subsequently, NCI-H226 and NCI-H520 cells were transfected with lentivirus LV-HP or LV-Ctrl, and the CM was collected. M2 macrophages were exposed to the CM with or without Hb added for 24 h to assess induction of ferroptosis, marked by enhanced membrane permeability attributed to iron-dependent lipid peroxidation. The LDH release assay revealed that the LDH levels were not significantly altered in the supernatant of M2 macrophages exposed to CM from LV-HP-transfected tumor cells, LV-Ctrl-transfected cells and LV-Ctrl-transfected cells with Hb (LV-Ctrl/Hb) (Fig. 5B). Conversely, M2 macrophages cultured in CM from LV-HP with Hb (LV-HP/Hb) exhibited a significant increase in LDH release levels compared with that of the other three groups. Flow cytometry analysis of 7-AAD-positive cells indicated no statistically significant variances among the LV-Ctrl, LV-Ctrl/Hb and LV-HP groups, but a significantly higher percentage of 7-AAD-positive cells in the LV-HP/Hb group (Fig. 5C). Intracellular Fe²⁺ and lipid peroxide levels in M2 macrophages, measured using FerroOrange (Fig. 5D) and C11-BODIPY respectively (Fig. 5E), showed no significant differences among the LV-Ctrl, LV-Ctrl/Hb and LV-HP groups. However, the LV-HP/Hb group demonstrated a significantly higher fluorescence intensity for both markers, indicating increased free Fe²⁺ (Fig. 5D) and lipid peroxide (Fig. 5E) levels. Transmission electron microscopy showed ferroptotic characteristics in M2 macrophages from the LV-HP/Hb group, such as decreased mitochondrial size, increased mitochondrial membrane density and a reduction in mitochondrial cristae, with no similar morphological alterations observed in the other three groups (Fig. 5F). Taken together, these findings suggest the CM from lung squamous carcinoma cells overexpressing HP effectively triggered ferroptosis in M2 macrophages in the presence of Hb.

To investigate the role of HP in intracellular Fe²⁺ generation and ferroptosis induction in M2 macrophages, NCI-H226 and NCI-H520 cells were transfected with LV-HP or LV-Ctrl. The resulting supernatants were used as CM. M2 macrophages were exposed to different combinations of CM, Hb and an anti-CD163 antibody. After 24 h, heme and HO-1 protein levels were measured, along with markers of ferroptosis. Spectrophotometry analysis showed that in the absence of HP, there was no significant difference in heme levels between the LV-Ctrl/Hb and LV-Ctrl/Hb/anti-CD163 groups. However, in the presence of HP, heme levels in the LV-HP/Hb/anti-CD163 group were significantly decreased when compared with that of the LV-HP/Hb group (Fig. 6A). These findings suggested that blocking CD163 receptors with an anti-CD163 antibody impeded Hb entry into M2 macrophages, thereby reducing intracellular heme production.

The results of flow cytometry analysis of intracellular HO-1 protein levels were similar to those of heme detection (27,31). There was no significant difference in HO-1 levels between the LV-Ctrl/Hb and LV-Ctrl/Hb/anti-CD163 groups; however, the HO-1 levels in the LV-HP/Hb/anti-CD163 group were significantly lower compared with those in the LV-HP/Hb group (Fig. 6B), consistent with previous research findings (27,31). To confirm the impact of CD163, HO-1, and Fe²⁺ on M2 macrophage ferroptosis, M2 macrophages were treated with an anti-CD163 antibody, the HO-1 inhibitor SnPPIX or the Fe²⁺ chelator DFO. Ferroptosis markers were then assessed using flow cytometry, including 7-AAD-positive membrane-permeable dead cells, Fe²⁺ and lipid peroxide. In the absence of HP, the addition of the anti-CD163 antibody did not affect 7-AAD-positive death cells, Fe²⁺ levels and lipid peroxide in the LV-Ctrl/Hb and LV-Ctrl/Hb/anti-CD163 groups (Fig. 6C-E). However, in the presence of both HP and Hb, treatment with the anti-CD163 antibody significantly reduced the levels of 7-AAD-positive dead cells, Fe²⁺ and lipid peroxide compared to the groups without antibody treatment (Fig. 6C-E). Similar results were found in the corresponding M2 macrophages treated with SnPPIX (Fig. 6F-H) and DFO (Fig. 6I-K). Taken together, these findings indicated that HP promoted ferroptosis in M2 macrophages by inducing intracellular Fe²⁺ generation through the CD163/HO-1 signaling pathway.