Vorapaxar enhanced mitochondria-associated ferroptosis primes cancer immunotherapy via targeting FOXO1/HMOX1 axis

Abstract

Immunotherapy has revolutionized cancer treatment, yet challenges persist, such as resistance and lethal thromboembolism, necessitating dual-purpose strategies. Targeting ferroptosis emerges as a promising strategy to enhance immunotherapy efficacy, prompting our investigation of antiplatelet agents that simultaneously promote ferroptosis and mitigate thromboembolic risks. Through systematic screening of 20 Food and Drug Administration (FDA)-approved antiplatelet agents, we identify vorapaxar as a potent pro-ferroptotic drug. Mechanistically, vorapaxar binds forkhead box O1 (FOXO1), inhibits its phosphorylation at Ser256, and facilitates nuclear translocation to upregulate heme oxygenase 1 (HMOX1), promoting mitochondrial iron overload and mitochondria-associated ferroptosis. Vorapaxar enhances immunotherapy-induced tumor ferroptosis and antitumor immunity across diverse melanoma models, including B16F10 tumor-bearing mice, Braf/Pten-driven spontaneous melanoma mice, and peripheral blood mononuclear cell (PBMC)-humanized mice. Clinically, high FOXO1/HMOX1 co-expression correlates with improved immunotherapy response and progression-free survival. These findings position vorapaxar as a promising adjunct to immunotherapy, offering a dual benefit for cancer patients requiring both antithrombotic therapy and immunotherapy.

Keywords: FOXO1; HMOX1; ferroptosis; immunotherapy; vorapaxar.

Graphical abstract

Figure 1

Screening of the antiplatelet drugs to identify vorapaxar as pro-ferroptotic agent in cancer cells (A) Heatmap depicting A375 cells viability following treatment with 10 μM antiplatelet agents, 2.5 μM RSL3, or their combinations for 6 h. (B) Molecular formula of vorapaxar. (C) 6 × 6 matrix screening of cell viability (left) and drug synergy assessment using Chou-Talalay combination indices (CIs), with x axis representing fraction affected (Fa) values (right). (D) Cell morphology and viability at various time points post-treatment. Scale bars, 100 μm. (E) Viability of the A375 and SK28 cells following treatment with vorapaxar, RSL3, or their combination (com), with or without cell death inhibitors (ZVAD-FMK [5 μM], necrostatin-1s [Nec-1s, 10 μM], chloroquine [CQ, 10 μM], N-acetylcysteine [NAC, 100 μM] and deferoxamine [DFO, 100 μM]) for 6 h. (F) Cell death analysis in A375 and SK28 cells after 6 h of indicated treatment, assessed by propidium iodide (PI) staining and flow cytometry. Fer-1, 4 μM. (G) Relative malondialdehyde (MDA) levels in A375 and SK28 cells treated as indicated for 3 h. (H) Flow cytometry analysis of lipid ROS in A375 and SK28 cells treated with vorapaxar, RSL3, or their combination, with or without Fer-1 for 3 h. (I) sgCtrl and sgGPX4 A375 and SK28 cells were treated with vorapaxar in the absence or presence of Fer-1 or DFO for 6 h. (J) Spheroids generated from A375 and SK28 cells were cultured for 48 h and received the indicated treatment for 6 h. Dead cells were stained by PI. Scale bars, 200 μm. Data are presented as mean ± SD, n = 3. p values were calculated using one-way ANOVA analysis. ∗∗∗p < 0.001. See also Figure S1.

Figure 2

Vorapaxar combined with RSL3 induce mitochondria-associated ferroptosis primarily through upregulating mitochondrial HMOX1 (A) Dotplot of pathway enrichment analysis comparing the vorapaxar and RSL3 combination (Com) versus DMSO treatment based on gene ontology (GO) database. (B) Ultrastructure images of A373 cells treated as indicated for 3 h were examined by transmission electron microscopy. Red arrow, mitochondria. Scale bars, upper: 2 μm and lower: 500 nm. (C) Mitochondrial membrane potential assessment in A375 cells after 1 h treatments, using rhodamine 123 (Rh123), with MitoTracker Red (mitochondrial marker). Scale bars, 20 μm. (D) Cellular ATP levels of A375 and SK28 cells after indicated treatment for 3 h. (E) Visualization of mitochondrial ferrous iron levels in A375 or SK28 cells with indicated treatment for 1 h. Green, Mito-FerroGreen; red, Mito-Tracker; and blue, DAPI. Scale bars, 20 μm. (F and G) Mitochondrial ROS (F) and mitochondrial lipid ROS levels (G) in A375 and SK28 cells after 3 h indicated treatments were assessed by flow cytometry. TEMPOL, 10 μM and Mito-TEMPO, 10 μM. (H) Dose response of RSL3-induced death of indicated cells in the presence of DMSO, or vorapaxar with or without TEMPOL or Mito-TEMPO after a 6 h treatment. (I) Heatmap of the top 20 upregulated genes in the combination group compared to the DMSO group. (J) GSEA showing enrichment of the heme degradation pathway (left) and endoplasmic reticulum stress pathway (right) between the combination group and DMSO group in RNA sequencing. (K) Mitochondrial and cytoplasmic protein expression levels of HMOX1 in A375 and SK28 cells with indicated treatments for 3 h. (L and M) Flow cytometry analysis of mitochondrial ferrous iron (L) and mitochondrial lipid ROS (M) in A375 and SK28 cells with indicated treatment. (N) Cell viability in sgCtrl and sgHMOX1 A375 and SK28 cells with the combination of RSL3 and vorapaxar treatment for 6 h. Data are presented as mean ± SD, n = 3. p values were calculated using one-way ANOVA analysis in (D, F, G, L, and M) and two-tailed unpaired Student’s t test in (N).∗∗p < 0.01 and ∗∗∗p < 0.001. See also Figures S2 and S3.

Figure 3

Vorapaxar facilitates mitochondria-associated ferroptosis by targeting the FOXO1/HMOX1 axis (A) HMOX1 mRNA levels in A375 and SK28 cells treated with DMSO, 2.5 μM RSL3, 5 μM vorapaxar, or the combination (Com) in the absence or presence of CHX (5 μg/mL) for 3 h. (B) Barplot showing enrichment scores of transcription factor pathways in the combination group compared to the RSL3 group. (C) Genome browser visualization of HMOX1 promoter binding peaks from FOXO1 ChIP-seq data (GEO: GSE128635) in control and nuclear activation mutation FOXO1^A3 cells, with input controls. (D) HMOX1 mRNA levels in A375 and SK28 cells treated as indicated for 3 h. (E and F) Total lipid ROS (E), and mitochondrial lipid ROS (F) in tumor cells with indicated treatment. AS1842856 (AS), 10 μM. (G) Representative immunofluorescence images of mitochondrial ferrous iron levels in A375 cells with indicated treatments for 1 h. Scale bars, 20 μm. (H) Dose response of RSL3-induced death of indicated cells in the presence of DMSO, vorapaxar, AS, or the combination after a 6 h treatment. (I) A375 and SK28 cells cultured as spheroids and treated with indicated treatments for 6 h, followed by PI staining to assess cell death. Scale bars, 200 μm. (J) FOXO1 and HMOX1 protein levels in shCtrl and shFOXO1 tumor cells. (K) Cell viability in shCtrl and shFOXO1 tumor cells with the combination treatment for 6 h. (L and M) Prediction of the interaction between vorapaxar and FOXO1 proteins. (N and O) Bio-layer interferometry (BLI) assay of vorapaxar binding to recombinant FOXO1 proteins. Sensorgram depicting the binding interaction of vorapaxar with FOXO1 (N). Equilibrium response units (RU) are plotted against vorapaxar concentrations (O). (P) Cellular thermal shift assay (CETSA) of FOXO1 proteins levels in tumor cells treated with DMSO, vorapaxar (10 μM) or AS (10 μM) for 1 h. (Q) FOXO1 and phosphorylated FOXO1 Ser256 protein levels in A375 and SK28 cells treated with DMSO or vorapaxar for 3 h. (R) Representative immunofluorescence images of A375 cells with indicated treatments and stained for FOXO1. Red, FOXO1 and blue, DAPI. Scale bars, 20 μm. Data are presented as mean ± SD, n = 3. p values were calculated using two-way ANOVA analysis in (A), one-way ANOVA analysis in (D, E and F), and two-tailed unpaired Student’s t test in (K). ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. See also Figure S4.

Figure 4

Vorapaxar-enhanced ferroptosis inhibits tumor growth through the FOXO1/HMOX1 axis (A) Schedule of tumor growth assessment in sgCtrl or sgGPX4 A375 tumor-bearing mice following indicated treatment. n = 5 per group. (B and C) Tumor weight (B) and tumor growth curve (C) of each group. (D and E) Immunohistochemistry staining for 4HNE or GPX4 in tumors of each group. Representative images (D). Scale bars, 50 μm. Quantification (E). (F) Treatment schedule of A375 tumor-bearing mice for indicated drug administration. n = 5 per group. (G and H) Tumor weight (G) and tumor growth curve (H) of each group. (I) Treatment schedule of B16F10 tumor-bearing mice for indicated drug administration. n = 5 per group. (J–L) Tumor weight (J), tumor growth curve (K), and survival curves (L) of each group. (M) Schematic administration of IKE or vorapaxar in genetically engineered Braf/Pten-driven melanoma mice. n = 5 per group. (N) Representative photographs of 4-OHT-treated regions on the backs and inguinal lymph nodes of mice at the indicated days post-treatment. Scale bars, 5 mm. (O and P) Tumor growth curves (O) and survival curves (P) of Braf/Pten-driven melanoma mice with indicated treatment. (Q and R) Immunohistochemistry staining for 4HNE in tumors. Representative images (Q). Scale bars, 50 μm. Quantification (R). Data are presented as mean ± SD. p values were calculated using one-way ANOVA analysis in (J) and two-way ANOVA analysis in (K and O). Log rank test was used in (L and P), and two-tailed unpaired Student’s t test in (R).∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. See also Figure S5.

Figure 5

Vorapaxar potentiates anti-PD-1 immunotherapy in suppressing melanoma by promoting ferroptosis (A) Schematic of vorapaxar or anti-PD-1 mAb administration in B16F10 tumor-bearing C57BL/6 mice. n = 5 per group. (B–D) Tumor weight (B), tumor growth curves (C), and body weight (D) of each group. (E) Representative flow cytometry plots and quantification of IFN-γ⁺ and GZMB⁺ proportions of tumor-infiltrating CD8⁺ T cells from B16F10 tumor-bearing mice with indicated treatment. (F and G) Immunofluorescence staining for CD8⁺ T cells and GZMB⁺ cells, and immunohistochemistry staining for 4HNE in tumors. Representative images (F). Red: CD8⁺ T, green: GZMB⁺ cells, and blue: DAPI. Scale bars, 50 μm. Quantification (G). (H) Schematic of vorapaxar or anti-PD-1 mAb administration in genetically engineered Braf/Pten-driven melanoma mice. n = 5 per group. (I) Representative photographs of 4-OHT-treated regions on the backs and inguinal lymph nodes of mice at the indicated days after treatment. Scale bars, 5 mm. (J and K) Tumor growth curves (J) and survival curves (K) of each group. (L) Representative images of immunofluorescence staining for CD8⁺ T cells and GZMB⁺ cells, and immunohistochemistry staining for 4HNE in tumors. Red: CD8⁺ T, green: GZMB⁺ cells, and blue: DAPI. Scale bars, 50 μm. Data are presented as mean ± SD. p values were calculated using one-way ANOVA analysis in (B, E, and G) and two-way ANOVA analysis in (C and J). Log rank test was used in (K). ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. See also Figure S5.

Figure 6

Upregulation of FOXO1/HMOX1 axis correlated with the efficacy of improved immunotherapy (A) Schematic flow diagram of A375 tumor-bearing PBMC-humanized mice. n = 6 per group. (B–E) Photography of isolated tumors (B), tumor weight (C), tumor growth curve (D), and body weight (E) of each group. (F) Immunohistochemistry staining for FOXO1, HMOX1, and 4HNE in tumors. Scale bars, 50 μm. (G) Heatmap showing the normalized WP_ferroptosis score and immune infiltration pathway enrichment scores of each sample, calculated by single-sample gene set enrichment analysis (inhouse cohort, n = 62). (H) WP_ferroptosis pathway scores, CD8⁺ T cell infiltration levels, T cell inflammatory gene expression profile (GEP), and CYT levels in the indicated groups. (I and J) Pie plot showing the proportion of responses to immunotherapy among the four groups from the inhouse cohort (I) or Ribas’s cohort (J). (Ribas’s cohort n = 437). (K) Schematic summary for the findings in the present study. Data are presented as mean ± SD. p values were calculated using two-tailed unpaired Student’s t test in (C and F), two-way ANOVA analysis in (D), and one-way ANOVA analysis in (H). Fisher’s exact test/chi-square test was used in (I and J). ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001. See also Figure S6.