The human epithelial ovarian cancer cell lines HEY and COV318 were purchased from the American Type Culture Collection (ATCC, Rockville, MD, USA). COV318 and HEY cells were grown in DMEM medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% (v/v) fetal bovine serum (FBS) (Invitrogen, San Diego, CA), L-glutamine (0.3 g/l) and 1% (v/v) penicillin and streptomycin (Sigma-Aldrich, St. Louis, MO, USA) at 37°C in a humidified incubator with 5% CO₂ atmosphere. The cisplatin-resistant ovarian cancer cell lines (PEO4 and A2780CP) were obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA). These cells were cultured in RPMI 1640 (Sigma-Aldrich, St. Louis, MO, USA), supplemented with 10% (v/v) fetal bovine serum (FBS) (Invitrogen, San Diego, CA), L-glutamine and 1% (v/v) penicillin and streptomycin (Sigma-Aldrich, St. Louis, MO, USA) in adherent cultures at 37°C in a humidified incubator with 5% CO₂ atmosphere. To maintain the resistance, 1 µM cisplatin was added to the medium of A2780CP and PEO4 cells every two to three passages. All cell lines were tested for mycoplasma contaminations and STR profiled for authentication.

Ferlixit (62.5 mg/5 ml, sodium ferric gluconate complex in sucrose, SANOFI) and cisplatin (50 mg/100 ml, SANDOZ) were obtained from the outpatient pharmacy at the Unit of Gynecologic Oncology, Magna Graecia University of Catanzaro; deferoxamine mesylate salt (DFO), erastin, ferrostatin-1 (Fer-1), hemin and cloroquine (Cq) were purchased from Sigma Aldrich (Sigma-Aldrich, St. Louis, MO, USA); the antioxidant (±)-6-hydroxy-2,5,7,8-tetra-methylchromane-2-carboxylic acid (trolox) was ordered from Cayman Chemical (Cayman Chemical Company, Ann Arbor, USA). Cells were seeded in a 6-well plate in serum-free medium. Each compound was used at the following final concentrations: ferlixit at 100 µM and 250 µM for 1, 8, 24 or 48 h; cisplatin at 150 μM for 24 h; DFO at 200 µM for 24 h; erastin at 8 µM and 25 µM for 1, 8 or 24 h; Fer-1, hemin and Cq at 10 µM for 8, 24 or 48 h; and trolox at 200 μM for 8 h. Treatments were performed at least three times on independent biological replicates.

3-[4,5-Dimethyl-2-thiazolyl]-2,5-diphenyltetrazolium bromide (MTT) (Sigma Aldrich, St. Louis, MO, USA) assay was performed to detect proliferation of HEY and COV318 cells untreated or treated with erastin, cisplatin and ferlixit. A total of 10⁴ cells/well were seeded into a 96-well flat-bottom plate and were subjected to drug treatment. There were quintuplicates for each cell type. Fresh MTT (5  mg/ml, Sigma Aldrich, St. Louis, MO, USA), re-suspended in phosphate-buffered saline (PBS) was added to each well for a final concentration of 0.5 mg/ml. After 4 h incubation, culture medium was discarded and replaced with 200 μl of isopropanol to solubilize formazan crystals. Optical density was measured at 595 nm in a spectrophotometer. Each experiment was performed in triplicate.

Growth rate of HEY and COV318 cells was obtained using the trypan blue dye exclusion method. Cells were counted after 24, 48, and 7h. Each experiment was performed in triplicate.

Approximately 1 × 10⁶ cells/well were seeded in 6-well plates overnight followed by the various treatment. Cells were centrifuged and the relative pellets were incubated with PI staining in the dark at 37°C for 15 min. Samples were then washed twice with PBS. The cellular fluorescence was analyzed by FACS BD LSRFortessaTM X-20 cytofluorometer (BD Biosciences). A total of 2 × 10⁴ event was acquired for each sample from three independent experiments. Fluorescence was measured using FlowJo software program (Tree Star, Inc.). Each experiment was performed in triplicate.

Cells were seeded in 6-well plates at density of 4 × 10⁵ cells/well and grown overnight. Following treatment with erastin, ferlixit, hemin, and DFO, morphological changes of “ballooning” phenotype were observed by light microscopy. Images were captured with Leica DM IL LED inverted phase contrast microscope (Leica Microsystems, Wetzlar, Germany).

HEY and COV318 cells (2 × 10⁶ cells/well) were plated in 100 mm culture dishes. Cells were then treated with erastin, DFO and/or ferlixit. Cells were centrifuged and the relative pellet were fixed for 3 h with 3% glutaraldehyde solution in 0.1 M phosphate buffer (pH 7.4). After washing in PBS for 15 min, samples were post-fixed in osmium tetroxide (1%) for 2 h, dehydrated in graded acetone, and then progressively embedded in acetone/resin with final embedment in pure resin (Araldite‐Fluka). Ultrathin sections (60–90 nm in thickness) were cut with a diamond knife, collected on copper grids (G300 Cu), and then examined with a Jeol JEM 1400‐Plus electron microscope operating at 80 kV.

Intracellular ROS amounts were determined by incubating cells for 10 min at 37°C with the redox-sensitive probe 2′-7′-Dichlorodihydrofluorescein diacetate (CM-H2DCFDA; Thermo Fisher Scientific, Waltham, USA), according to the instructions of the manufacturer. CM-H2DCFDA fluorescence was analyzed by flow cytometry using a FACS BD LSRFortessaTM X-20 cytofluorometer (BD Biosciences) and data were processed with FlowJo software (Tree Star, Inc.). Each experiment was performed in triplicate.

Lipid peroxidation was investigated by flow cytometry using BODIPY™ 581/591 C11 dye (Thermo Fisher Scientific, Waltham, USA). Briefly, cells were seeded in 6-well plates at a density of 4 × 10⁵ cells/well and grown overnight. After treatments, cells were loaded with 2.5 μM BODIPY™ 581/591 C11 for 30 min at 37°C. After 30 min of loading, unincorporated dye was removed by washings twice with PBS. Samples were then centrifuged at 1,000 r.p.m. for 3 min and the pellets were resuspended in 500 μl of PBS. The cell suspension was subjected to the flow cytometry analysis to analyze the amount of lipid ROS within cells. Oxidation of BODIPY-C11 resulted in a shift of the fluorescence emission peak from ~590 to ~510 nm proportional to lipid ROS generation (22). The fluorescence intensities of cells per sample were determined by flow cytometry using the FACS BD LSRFortessaTM X-20 cytofluorometer (BD Biosciences). A minimum of 20,000 cells were analyzed per condition. Fluorescence of each probe was measured using FlowJo software program (Tree Star, Inc.). Each experiment was performed in triplicate.

Cellular MDA levels were determined by using the lipid peroxidation assay kit (Sigma Aldrich, MAK085, Missouri, USA) according to the protocol of the manufacturer. Briefly, cells were seeded in a 6-well plate (5 × 10⁵ cells per plate) and treated with erastin or ferlixit. Cells were washed with ice-cold PBS and homogenized on ice in 300 μl of the MDA lysis buffer with 3 μl BHT (100×), then centrifuged (13,000×g, 10 min) to remove insoluble material. Approximately 200 μl of the supernatant from each homogenized sample were placed into a microcentrifuge tube; then 600 μl of the thiobarbituric acid (TBA) solution were added into each vial to form the MDA-TBA adduct. The mixture was incubated for 1 h at 95°C followed by ice-cooling for 10 min. Lipid peroxidation was determined by the reaction of MDA with thiobarbituric acid (TBA) to form a colorimetric (532 nm)/fluorometric adduct, proportional to the MDA present. The absorbance was measured in a spectrophotometer. Each experiment was performed in triplicate.

Generation of mitochondrial ROS was measured by flow cytometry with the use of MitoSOX Red Mitochondrial Superoxide Indicator (Thermo Fisher Scientific Inc.). After treatments, cells were incubated with 5 µM MitoSOX Red for 10 min at 37°C and then analyzed by flow cytometry using a FACS BD LSRFortessaTM X-20 cytofluorometer (BD Biosciences). A minimum of 20,000 cells was analyzed per condition. Fluorescence was measured using FlowJo software program (Tree Star, Inc.). Each experiment was performed in triplicate.

Changes in the mitochondrial membrane potential were analyzed by staining the cells with TMRE (tetramethylrhodamine ethyl ester) dye (Thermo Fisher Scientific, Waltham, USA). Briefly, cells were cultured in 6-well plates and, upon treatments, were incubated with 100 nM TMRE dye for 30 min at 37°C and then washed with PBS. Samples were then centrifuged at 1,000 r.p.m. for 3 min and the pellets were resuspended in 500 μl of PBS. TMRE fluorescence was analyzed by flow cytometry using a FACS BD LSRFortessaTM X-20 cytofluorometer (BD Biosciences). A minimum of 20,000 cells was analyzed per condition. Fluorescence was measured using FlowJo software program (Tree Star, Inc.). Each experiment was performed in triplicate.

The intracellular labile iron pool (LIP) was quantified by monitoring the recovering of calcein fluorescence induced by iron chelators after the calcein fluorescence was quenched by intracellular LIP. Briefly, cancer cells were seeded in 6-well plates at a density of 4 × 10⁵ cells/well and grown overnight. Then cells were loaded with 0.25 μM calcein acetoxymethyl ester (CA-AM; calcein-AM) (Sigma-Aldrich, Missouri, USA) for 30 min at 37°C. After washing twice with PBS, cells were treated with 200 μM 3-hydroxy-1,2-dimethyl-4(1H)-pyridone (deferiprone or L1) (Sigma-Aldrich, Missouri, USA) or left untreated. Following staining, and washing PBS, cells were analyzed by a FACS BD LSRFortessaTM X-20 cytofluorometer (BD Biosciences). The difference in the mean fluorescence index between chelator-treated and untreated cells (Δ mean fluorescence intensity, ΔMFI) reflects the amount of LIP. Each experiment was performed in triplicate.

To obtain total protein extracts cells were washed once with PBS (1×) and total cell lysates were prepared using RIPA buffer containing 1 M Tris HCl, Triton X-100, 3 M NaCl, 0.5 M EDTA, 10% SDS supplemented with cOmplete™ Protease Inhibitor Cocktail provided in EASYpacks (Roche Diagnostics, Mannheim, Germany) as described by Biamonte et al. (23). Briefly, extracts were centrifuged at 12,000×g for 30 min at 4°C to eliminate insoluble fragments. For nuclear and cytoplasmic protein extraction, 2 × 10⁷ cells were washed once with PBS (1×) and cell pellets were resuspended in 200 μl of Lysis Buffer w/NP40. After 5 min incubation in ice, all tubes were centrifuged at 3,500 r.p.m. for 5 min at 4°C. The supernatant, containing cytoplasmic protein fraction, was transferred in a new eppendorf tube, while the pellet, containing cell nuclei, was washed in PBS (1×). To obtain nuclear protein content, the latter was resuspended in 1 ml of Lysis Buffer w/o NP40 and centrifuged twice at 3,500 r.p.m. at 4°C for 5 and 10 min, respectively. After these two centrifugations, the pellet was resuspended in 300 μl of Lysis Buffer w/o NP40 and 300 μl of Cushing Buffer were added. This mix solution was centrifuged at 6,000 r.p.m. for 10 min at 4°C and then resuspended in 300 μl of Resuspension Buffer. Three freeze-thaw cycles were performed according to protocol. Following a last centrifuge at 9,500 r.p.m. for 15 min at 4°C, the supernatant, containing nuclear proteins was collected. For recipes of buffers used for selective nuclear and cytoplasmic protein extraction refer to Chirillo et al. (24).

Protein amount was determined using the Protein Assay Dye Reagent Concentrate (Bio-Rad Laboratories, Hercules, California, USA). Each protein sample (40–50 μg) was separated by 10–15% SDS–PAGE and then transferred to nitrocellulose membranes (Sigma-Aldrich, St. Louis, MO, USA). After blocking with 5% milk, incubation with primary antibody was performed overnight at 4°C. Antibodies against FtH (1:200, sc-376594), NCOA4 (1:500, sc-373739) were purchased from Santa Cruz Biotechnology (Santa Cruz Biotechnology, Dallas, Texas); antibodies against VDAC2 (1:500, ab37985) and GPX4 (1:500, ab41787) were purchased from Abcam (Cambridge, United Kingdom). After incubation with peroxidase-conjugated secondary antibodies (Peroxidase AffiniPure Sheep Anti-Mouse IgG, 1:10,000; Peroxidase AffiniPure Donkey Anti-Rabbit IgG, 1:10,000; Peroxidase AffiniPure Donkey Anti-Goat IgG, 1:10,000; Jackson ImmunoResearch Europe Ltd. Cambridge House) for 1 h at room temperature, the signals were visualized via using the ECL western blotting detection system (Santa Cruz Biotechnology, Dallas, Texas) and acquired by Uvitec Alliance Mini HD9 (Uvitec Cambridge, UK). To ensure equal loading of proteins a goat polyclonal anti-γ-Tubulin (γ-TUB) antibody (1:3,000; sc-17787; Santa Cruz Biotechnology) was used. The protein band intensity on western blots was quantified and normalized to that of γ-TUB by using ImageJ software (http://rsb.info.nih.gov/ij/).

Total RNA was extracted using the Trizol method (Life Technologies, Carlsbad, CA, USA) according to the instructions of the manufacturer (25, 26). Then, 1 µg of total RNA were retrotranscribed using High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific, Waltham, Massachusetts, USA). qRT-PCR was performed using the SYBR Green qPCR Master Mix (Thermo Fisher Scientific, Waltham, Massachusetts, USA). Analysis was performed on QuantStudio 3 Applied Biosystems by Thermo Fisher Scientific. The relative mRNA expression level of p53 and BAX were calculated by the 2^−ΔΔCt method and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as the housekeeping gene. Each experiment was performed in triplicate.

Overall data are represented as mean ± standard deviation (SD) of at least three biological replicates. When appropriate, data were analyzed by performing a simple comparison between two groups using Student’s t-test. We were interested in determining whether the means of more than two groups were equal or not, thus, we performed an analysis of variance (ANOVA). A p-value <0.05 was considered statistically significant.

Ferroptosis has been proposed as a new potential therapeutic option for both chemotherapy-sensitive and -resistant OVCA (5, 21), but its effectiveness is still currently limited, also due to a lack of insights on the genetic and metabolic determinants of ferroptosis sensitivity. Here, we first assessed whether erastin is a bona fide ferroptosis inducer in OVCA cell lines. Taking into account that erastin was originally selected as an agent able to trigger ferroptosis exclusively in RAS-mutated cancer cells, we selected four OVCA cell lines, three of which are KRAS^wt (COV318, PEO4, and A2780CP) and one is KRAS^G12D (HEY). Furthermore, this panel includes both cisplatin-sensitive (HEY and COV318) and cisplatin-resistant (PEO4 and A2780CP) cells (see cell viability assay in Supplementary Figure S1). HEY, COV318, PEO4, and A2780CP cells were treated with two increasing concentrations of erastin (8 and 25 μM) for 8 h. As shown in Figure 1A, HEY and A2780CP cell viability significantly decreased upon treatment with 8 μM (~35–55%, p <0.05) and further broke down with 25 μM erastin for 8 h (~1–20%, p <0.05). Conversely, COV318 and PEO4 cells are insensitive to both erastin concentrations, with cell viability nearly of 90% upon treatments. Similar results were obtained by using propidium iodide (PI) staining. Flow cytometry analysis highlighted a significant increase in cell death in HEY and A2780CP cells treated with both 8 μM (25.1% PI⁺ HEY and 29.3% PI⁺ A2780CP cells) and 25 μM erastin (91.8% PI⁺ HEY and 59.1% PI⁺ A2780CP cells) for 8 h (Figure 1B) while no significant differences in cell death were observed in COV318 and PEO4 cells upon treatment. In agreement, morphological observation under inverted phase-contrast microscope identified in treated-HEY and A2780 cells the ferroptosis distinct “ballooning” phenotype (Figure 1C). This morphological feature was undetectable in COV318 and PEO4 cells. These results indicate that erastin is not always effective in OVCA cells and that the erastin-sensitivity is after all independent of either KRAS mutational status or cisplatin sensitivity.

Iron homeostasis and intracellular amount of LIP are both critical for ferroptosis (27, 28). Indeed, high LIP content promotes Fenton reactions thus causing ROS production and ferroptosis; on the other hand, the use of iron chelators protects cancer cells from ferroptosis by restricting the intracellular free iron levels (29–33). We decided to evaluate the LIP content of erastin-sensitive HEY cells and erastin-resistant COV318 cells upon the administration of 8 μM erastin for 8 h. Treatment with 25 μM erastin was excluded due to the excessive cytotoxic effects exerted in HEY cells. Flow cytometry analysis with Calcein-AM staining highlighted two major findings: i) HEY cells showed higher LIP (ΔMFI:3046) compared to COV318 cells (ΔMFI:578) already at baseline conditions, ii) treatment with erastin exacerbates this difference by causing a significant increase of LIP in HEY (ΔMFI:5165) but not in COV318 cells (ΔMFI:545) (Figure 2A and Supplementary Figure S2). Western blotting analyses suggest that in HEY cells this phenomenon might occur via ferritinophagy, since erastin treatment caused a consistent reduction of FtH and its cargo NCOA4 (Figure 2B). Conversely, in COV318 cells erastin administration left NCOA4 expression unaltered but rather enhanced FtH expression, thus preventing the overload of intracellular free iron (Figure 2B). In parallel, HEY showed a higher baseline content of both total ROS and mitochondrial superoxide species (CM-H2DCFDA MFI:70019; MitoSOX Red MFI:673) compared to COV318 cells (CM-H2DCFDA MFI:40150; MitoSOX Red MFI:484) and erastin treatment strengthened this gap (HEY: CM-H2DCFDA MFI:115691; MitoSOX Red MFI:1067) (COV318: CM-H2DCFDA MFI:39288; MitoSOX Red MFI:641) (Figure 2C and Supplementary Figure S2).

Then, we investigated the molecular basis underlying erastin-induced ferroptosis in HEY cells. Erastin treatment may alter mitochondrial membrane potential (ΔΨ), thus ultimately leading to mitochondrial dysfunction (34). Hence, we measured the effects of 8 μM erastin on the ΔΨ by loading HEY cells with the ΔΨ fluorescent indicator TMRM. Flow cytometry analysis revealed that TMRM fluorescence increased in HEY cells within 1h after erastin treatment (MFI:11666) compared to untreated cells (MFI:4821), which is indicative of mitochondrial hyperpolarization, and then significantly decreased after 8 h (MFI:1451) suggesting mitochondrial depolarization (Figure 3A and Supplementary Figure S3A). Moreover, WB analysis showed that erastin promoted a consistent decrease of the voltage-dependent anion channel 2 (VDAC2), already identified as indicator of ferroptosis-induced mitochondrial dysfunction (35) (Figure 3B). In parallel, by using TEM, we observed that erastin determined damage of mitochondria architecture characterized by loss, fragmentation, and disorganization of cristae and a sparse (rarefied) or often vacuolated matrix (red arrowheads in Figure 3C). Conversely, in the untreated HEY cells, mitochondria appeared well-preserved with an ordered distribution of cristae (green arrowheads in Figure 3C). The same panel of analyses was performed in COV318 cells where the erastin treatment is unable to induce mitochondrial dysfunction (Figures 3A–C, Supplementary Figure S3A). Finally, albeit a slight reduction of intracellular GPX4 protein levels, lipid peroxidation, measured by both C11-BODIPY and Malondialdehyde (MDA) assays, did not significantly increase neither in HEY nor in COV318 cells treated with erastin (Supplementary Figures S3B–D). Nevertheless, in order to evaluate the role of ROS in erastin-induced cell death, we performed a co-treatment with erastin (8 μM for 8 h) and the antioxidant compound trolox (200 μM for 8 h). As reported in Supplementary Figure S3E, trolox significantly repressed the erastin-dependent accumulation of both total ROS and mitoROS and, in parallel, consistently abolished the erastin-induced cell death (Supplementary Figure S3F).

To further deepen the role of baseline LIP amount in erastin-triggered ferroptosis, HEY cells were first treated with the iron chelator deferoxamine mesylate (DFO) (200 μM) for 24 h; then, culture medium was replaced, and cells were treated with erastin (8 μM for 8 h). First, DFO treatment alone significantly reduced the LIP (HEY^DFO ΔMFI:2315 vs HEY^untreated ΔMFI:4780), as well as both total and mitoROS (HEY^DFO CM-H2DCFDA:24508 vs HEY^untreated CM-H2DCFDA:53563; HEY^DFO MitoSOX Red:139 vs HEY^untreated MitoSOX Red:192). Albeit leading to a slight mitochondrial membrane depolarization (HEY^DFO TMRM MFI:1501 vs HEY^untreated TMRM MFI:2152) and reduction of VDAC2 protein levels, DFO alone did not affect mitochondrial ultrastructure or HEY cell death (3.7% PI⁺ HEY^DFO vs 2.02% PI⁺ HEY^untreated) (Figures 4A–E). Then, we observed that the reduction of intracellular free iron strongly protected HEY cells from erastin-induced cell death (HEY^untreated ΔMFI:4780, HEY^erastin ΔMFI:6374, HEY^DFO/erastin ΔMFI:5712) (2.02% PI⁺ HEY^untreated, 46.4% PI⁺ HEY^erastin, 4.9% PI⁺ HEY^DFO/erastin) (Figures 4A and Supplementary Figure S4). Accordingly, HEY cells pre-treated with DFO did not show the “ballooning” phenotype upon administration of erastin (Figure 4B), and mitochondrial functionality appeared overall preserved. Indeed, i) total and mitoROS were inhibited by the iron chelation (HEY^untreated CM-H2DCFDA:53563, HEY^erastin CM-H2DCFDA:81276, HEY^DFO/erastin CM-H2DCFDA:42444) (HEY^untreated MitoSOX Red:192, HEY^erastin MitoSOX Red:332, HEY^DFO/erastin MitoSOX Red:198) (Figures 4C and Supplementary Figure S4); ii) DFO significantly counteracted the mitochondrial membrane hyperpolarization induced by 1 h erastin and then left the ΔΨ almost unaltered upon 8 h erastin (HEY^untreated TMRM MFI:2152; HEY^{erastin (1 h)} TMRM MFI:6294; HEY^{erastin (8 h)} TMRM MFI:1195; HEY^{DFO/erastin (1 h)} TMRM MFI:1541; HEY^{DFO/erastin (8 h)} TMRM MFI:1143) (Figures 4C and Supplementary Figure S4); iii) erastin-mediated VDAC2 downregulation was consistently counteracted by DFO while FtH and NCOA4 levels continued to decrease (Figure 4D); iv) mitochondrial ultrastructure was maintained (Figure 4E).

Overall, these results indicate that ferroptosis sensitivity of HEY cells is associated with the baseline LIP and that a sufficient increase of intracellular free iron upon erastin treatment is mandatory to trigger mitochondrial dysfunction.

Then, to demonstrate that a sufficient baseline LIP is essential to induce ferroptosis, we treated the erastin-resistant COV318 cells with growing concentrations of ferlixit (100 and 250 μM for 24 and 48 h). Plus, the lowest dose of ferlixit (100 μM) was used as pre-treatement for 24 h followed by administration of 8 μM erastin (8 h). We observed that the sole administration of ferlixit at 250 μM for 48 h reduced COV318 cell viability to ~45% (p <0.05) and that the combination (100 μM for 24 h) with erastin (8 μM for 8 h) broke down cell viability to less than 2% (p <0.05) (Figure 5A). In agreement, cytofluorimetric assays show that PI⁺ COV318 cells upon 250 μM for 48 h were 33.7% while PI⁺ COV318 cells treated with the combined therapy reached 97.4% (Figure 5B). Consistently with these effects, the ballooning phenotype increased along with the growing concentrations and time of ferlixit treatment, and then reaches its maximum upon the combination of ferlixit and erastin (Figure 5C). Since the synergy of ferlixit (100 μM, 24 h) with erastin (8 μM, 8 h) led to excessive cytotoxic effects, the analyses of the molecular mechanisms underlying the combination therapy approach were performed shortening the time of ferlixit treatment (100 μM, 8 h). First, we demonstrated that ferlixit/erastin combined treatment caused a greater increase of LIP compared to ferlixit alone (250 μM, 48 h) (COV318^untreated ΔMFI:9315, COV318^{250 µMferlixit} ΔMFI:18616, COV318^{ferlixit/erastin} ΔMFI:45345) (Supplementary Figure S5A). In parallel, total and mitochondrial ROS already rise upon ferlixit alone and underwent a further increase after the combined therapy (COV318^untreated CM-H2DCFDA:32099, COV318^erastin CM-H2DCFDA:31793, COV318^{100 µMferlixit} CM-H2DCFDA:30597, COV318^{250 µMferlixit} CM-H2DCFDA:46532, COV318^{ferlixit/erastin} CM-H2DCFDA:77019) (COV318^untreated MitoSOX Red:181, COV318^erastin MitoSOX Red:274, COV318^{100 µMferlixit} MitoSOX Red:236, COV318^{250 µMferlixit} MitoSOX Red:647, COV318^{ferlixit/erastin} MitoSOX Red:1432) (Figures 5D and Supplementary Figure S5B). Notably, when ferlixit was used alone at 250 μM for 48 h, the ΔΨ significantly increased while when used in combination with erastin mitochondria faced membrane depolarization (COV318^untreated TMRM:2750, COV318^erastin TMRM:2505, COV318^{100 µMferlixit} TMRM:2368, COV318^{250 µMferlixit} TMRM:5544, COV318^{ferlixit/erastin} TMRM:1061) (Figures 5D and Supplementary Figure S5B). The WB analysis revealed that while the sole ferlixit administration (250 μM, 48 h) caused a slight decrease of VDAC2, this reduction became consistent only upon the combined treatment. The effects of both therapeutic approaches on FtH and NCOA4 deserve more detailed comments: i) FtH resulted always upregulated because ferlixit, as any other iron compound, promotes its translation through the IRP/IRE iron regulatory system; ii) NCOA4 is unaffected by ferlixit alone, but it appears downregulated upon ferlixit/erastin treatment (Figure 5E). At last, TEM analysis highlights that ferlixit promoted an initial mitochondrial morphology alteration while the combined treatment provoked evident mitochondria disorganization (Figure 5F). Once again, no increase in lipid peroxidation has been detected (Supplementary Figures S5C–E).

Finally, we wondered whether in the ferroptosis-sensitive HEY cells, characterized by a consistent baseline LIP, the sole administration of ferlixit might totally replace the use of erastin. Thus, we treated HEY cells with ferlixit (100 and 250 μM for 24 and 48 h) and we found that already at the lowest concentration cell viability decreased to ~40% (p <0.05) and the relative PI⁺ HEY cells were 50.2%. Notably, at 100 and 250 μM for 48 h, ferlixit caused massive cytotoxic effects (PI⁺ HEY^{100 μM ferlixit (48 h)}: 97.5%; PI⁺ HEY^{250 μM ferlixit (48 h)}: 95.4%) (Figures 6A, B). The ballooning phenotype clearly mirrored the above-mentioned effects (Figure 6C). Once we demonstrated the increase in the amount of intracellular LIP (Supplementary Figure S6A), we selected the lowest dose of ferlixit (100 μM for 24 h) to explore the underlying molecular mechanisms. Flow cytometry analysis revealed that TMRM fluorescence increased in HEY cells within 1 h after ferlixit treatment (MFI:6851) and then decreased after 24 h (MFI:2185) compared to untreated cells (MFI:3785), which is indicative of the two main step of mitochondrial hyperpolarization and depolarization. In agreement, i) total and mitochondrial ROS increased upon ferlixit administration (HEY^untreated CM-H2DCFDA:24046 vs HEY^ferlixit CM-H2DCFDA:63302) (HEY^untreated MitoSOX Red:261 vs HEY^ferlixit MitoSOX Red:1104) (Figures 6D and Supplementary Figure S6B), ii) VDAC2 was downregulated (Figure 6E). Here too, ferlixit induced FtH upregulation while NCOA4 resulted unaltered (Figure 6E). Mitochondria morphological analysis by TEM highlights that in this cell type the sole ferlixit was able to promote mitochondria swelling (Figure 6F). Once again, no increase in lipid peroxidation has been observed (Supplementary Figures S6C–E).

In order to stress the potential use of iron compounds as single agents or combination therapy with erastin to sensitizes OVCA cells to ferroptosis, we treated both HEY and COV318 cells with 10 μM hemin. As shown in Supplementary Figure S6F, 10 μM hemin for 24 h was sufficient to trigger ferroptosis in HEY cells (78.1% PI⁺ HEY) while treatment with 10 μM for 48 h moderately promoted cell death in COV318 cells (34.6% PI⁺ COV318). Therefore, we combined hemin (10 μM for 48 h) with erastin (8 μM for 8 h) and we observed a significant impact on cell death also in COV318 (94.7% PI⁺ COV318). The analysis of the “ballooning phenotype” further supported the role of hemin in ferroptotic cell death (Supplementary Figure S6G).