Fig 2: DEG expression analysis by RT-qPCR and in TCGA. (A) Expression of DEGs in the HIF1A-KO and LACZ-control HepG2 cells. (B) Expression of DEGs in tumor and normal tissues in TCGA (*p < 0.05, **p < 0.01, and ****p < 0.0001). (C) Heat map of the correlation between multiple genes and immune cells in HCC. The horizontal and vertical coordinates represent genes, different colors represent correlation coefficients (blue represents positive correlation and red represents negative correlation), and the darker color represents the two stronger correlations (*p < 0.05 and **p < 0.01).

Fig 3: CISD2 silencing accelerates iron accumulation in A549 cells. (A) The activity of cytoplasmic aconitase in CISD2 knockdown cells, **P < 0.01 vs. the control group. (B) The activity of mitochondrial aconitase in CISD2 knockdown cells, *P < 0.05 vs. the control group. (C) Protein levels of iron regulatory protein 2 (IRP2), transferrin receptor protein (TFR), and hypoxia-inducible factor 1-alpha (HIF1-α) in indicated A549 cells. β-Actin was used as the loading control. (D) Intracellular Fe2+ was measured by RPA staining and flow cytometry analysis. Representative images (left) and statistical graphs (right) are shown; *P < 0.05, compared to the control group. (E) The representative images of mitochondrial morphology. Control and CISD2 silenced A549 cells were treated with or without DFO (100 μM) for 12 h, then labeled with MitoTracker Red and subjected to a confocal microscope for the observation of changes in mitochondrial morphology. Scale bars: 5 µm. (F) Quantification of mitochondrial morphology alteration according to the classification of mitochondrial category I–III. (G) The morphological changes of mitochondria in control and CISD2 silenced cells were detected by transmission electron microscopy (TEM) in the presence or absence of DFO (100 μM) treatment. Comparable results were obtained in three independent experiments.

Fig 4: Identification of TGM2-expressing luminal cells in mouse and human PCa and association with HIF1A.(A) UMAP of luminal clusters in prostates of Pten(i)pe-/- mice at 3, 6, 9, and 15 months AGI. (B) Cellular trajectory colored by luminal subset. Color gradients distinguish cells at the various time points AGI. (C) KEGG pathway analysis of the top 500 genes in the trajectory inference analysis. (D) Hif1a transcript levels in the various luminal subsets at the indicated time points AGI. (E) Plot of gene importance depicting HIF1A signaling–related genes among the top 500 genes (dashed vertical line). (F) Violin plots depicting the transcript levels of Hif1a and its target genes in luminal-A, -C1, and -C2 cells. Data were obtained from samples at 3, 6, 9, and 15 months AGI. ***P < 0.001 calculated using the Wilcoxon rank sum test. (G) Analysis of the correlation between HIF1A and TGM2 transcript levels in human PCa transcriptomes on the TCGA. Pearson’s correlation coefficient was used to determine the association between the two genes. (H) Violin plot depicting the transcript levels of Tgm2 in luminal-A, -C1, and -C2 cells. (I) Immunohistochemical staining of TROP2 and TGM2 in prostatic sections of PtenL2/L2 and Pten(i)pe-/- mice at 10 and 15 months AGI. Representative images of the DLP are shown. N = 3 mice per condition. Scale bars, 100 µm. (J) Representative immunohistochemical staining of TGM2 in human prostate adenocarcinoma in TMA sections (n = 56 patients). Scale bars, 100 µm. (K) Kaplan-Meier plots demonstrating the PFS of patients with high prostatic TGM2 expression (n = 32 patients) and those with low levels of TGM2 (n = 24 patients). *P < 0.05, calculated using the Gehan-Breslow-Wilcoxon test.

Fig 5: Transcriptional and post‐translational analysis of HIF1α. (A) PLC5 cells were transfected with corresponding vectors, AMPKα1, or shAMPKα1 for 72 h, and the HIF1A expression was analyzed by qPCR. Columns, mean; error bars, SD (n = 3), N.S., not significant (Welch’s t‐test). (B) PLC5 cells were transfected with the corresponding vector and AMPKα1 for 72 h. Cell lysates were collected for the immunoprecipitation assay. IgG antibody was used as a negative control of immunoprecipitation assay. The experiments were performed in three independent replicates.