Ferroptosis is an iron-dependent form of program cell death caused by toxic accumulation of lipid peroxides in cell membranes. Ferroptotic cell death has morphological and biochemical characteristics that distinguish it from other forms of cell death (1). Ferroptosis is mainly characterized by a rupture of the mitochondrial outer membrane, a small mitochondria volume, and a decrease in mitochondrial cristae (2). Biologically, ferroptotic cell death is different from apoptosis in that it is caused by changes in the System Xc−/GSH/GPX4 axis, leading to Fe^2＋ accumulation and increased lipid peroxidation (1, 3). Recent studies has linked ferroptosis to a number of pathological ailments and disorders. It is well known that ferroptosis is a key tumor suppressor mechanism and that defective ferroptosis can promote the growth of tumors (4, 5).

Mitochondria are organelles that not only produce energy, but also function as important mediators of cell death by producing reactive oxygen species (ROS) and maintain iron and metabolic homeostasis within cells (6). Mitochondria play a critical role in regulating cellular metabolic plasticity and many aspects of programmed cell death (6, 7). Treatment with erastin can enhance the formation of ROS in the mitochondria, leading to opening of mitochondrial permeability transition (mPTP), lowering of ∆Ψm, and depletion of ATP (8, 9). Cell death is significantly influenced by mitochondrial ROS (10, 11). According to a recent study, ferroptosis caused by erastin and its analogues might increase the generation of mitochondrial and cytosolic ROS (12).

PPARγ co-activator-1 (PGC-1) family of transcriptional co-activators are key regulators of mitochondrial biogenesis. They play a role in regulating cellular energy metabolism (13). Of this family, PGC1α is an important regulator of ROS scavenging by controlling antioxidant expression during mitochondrial respiration (14-16). Although recent studies have investigated the link between PGC1α and ferroptosis, the specific role of PGC1α in erastin-induced mitochondrial dysfunction during ferroptotic cell death remains unclear. This study demonstrates that inhibiting the expression or activity of PGC1α in HT1080 fibrosarcoma cells could downregulate lipid peroxidation, interfere with mitochondrial dysfunction, and protect against erastin-induced ferroptosis.

Down-regulation of PGC1α attenuates erastin-induced death of HT1080 cells

Mitochondria play an important role in metabolism including catabolism and anabolism, iron metabolism, and energy metabolism (17). Additionally, PGC1α plays an important role in cells by contributing to mitochondrial biogenesis (18). To examine the role of PGC1α, ferroptotic cell death was induced with erastin. Total proteins and RNAs were extracted from HT1080 cells treated with erastin (10 μM) at designated time points (0, 2, 4, 6, 8, and 10 hours). Comopared to vehicle treatment, erastin (10 μM) administration resulted in higher PGC1α protein and mRNA levels (Fig. 1A, B). Similarly, RSL3 (100 nM), an inhibitor of GPX4, also increased PGC1α protein and mRNA levels (Supplementary Fig. 1A, B). Subsequently, To elucidate the role of PGC1α in ferroptotic cell death, we induced ferroptosis with erastin and evaluated viability of HT1080 cells treated with SR18292, an inhibitor of PGC1α. As shown in Fig. 1 CPGC1α inhibitor SR18292 (20 μM) restored cell viability decreased by erastin treatment. Lactate dehydrogenase (LDH) release assay was used to determine cytotoxicity of erastin (10 μM) or vehicle in the presence or absence of SR18292, an inhibitor of PGC1α (20 μM), to HT1080 cells. SR18292 (20 μM) treatment reduced cytotoxicity of erastin (10 μM) (Fig. 1D), confirming the protective effect of SR18292 (an inhibitor of PGC1α). Similarly, treatment with RSL3 (100 nM) in the presence of SR18292 (20 μM) significantly restored cell viability and reduced cytotoxicity of erastin as evidenced by decreased LDH release, confirming the protective effect of PGC1α inhibition on ferroptosis (Supplementary Fig. 1D). Additionally, to investigate the effect of PGC1α expression under erastin treatment conditions, HT1080 cells were transfected with PGC1α shRNA. Whether PGC1α expression was down-regulated in HT1080 cells transfected with PGC1α shRNA compared to cells transfected with control shRNA was then determined. We confirmed that PGC1α protein and mRNA levels were reduced in HT1080 cells transfected with PGC1α shRNA (Fig. 1E, F). To further investigate the critical function of PGC1α, the viability of HT1080 cells transfected with PGC1α shRNA after treatment with erastin (10 μM) was examined and compared with that of HT1080 cells transfected with control shRNA. When erastin was present, viability of HT1080 cells transfected with PGC1α shRNA was increased compared to that of HT1080 cells transfected with control shRNA (Fig. 1G). In addition, cytotoxicity was verified by lactate dehydrogenase (LDH) release assay. When transfected HT1080 cells were stimulated with erastin (10 μM), LDH release levels were lower in HT1080 cells transfected with PGC1α shRNA than in control shRNA transfected cells (Fig. 1H). Consistently, when ferroptosis was induced by RSL3 (100 nM), PGC1α shRNA-transfected cells showed significantly improved cell viability and reduced LDH release levels compared to control shRNA-transfected cells, further confirming the protective effect of PGC1α down-regulation on ferroptosis (Supplementary Fig. 1E, F). These findings suggest that PGC1α down-regulation can effectively shield HT1080 cells against erastin-induced cell death. Thus, PGC1α expression level is an important mediator regulating erastin-induced ferroptosis.

Down-regulated PGC1α blocks erastin-induced mitochondrial ROS

PGC1α is well known as a master regulator of mitochondria. It plays a role in regulating ROS. ROS can induce lipid peroxidation and affect ferroptotic cell death (19). To determine whether ROS generated by erastin treatment could affect mitochondrial ROS, we investigated the effect of mito-TEMPO (10 μM), a mitochondrial-specific antioxidant, on erastin-induced ferroptosis in HT1080 cells by examining cell viability. Erastin-induced cell death was restored by mito-TEMPO (Fig. 2A). MitoSOX (a fluorescent probe for mitochondrial ROS) was used to measure mitochondrial ROS by flow cytometry to assess the involvement of PGC1α in erastin-induced mitochondrial ROS. MitoSOX fluorescence was increased at 12 hours after treating HT1080 cells with 10 μM erastin. Erastin-induced mitochondrial ROS was reduced by SR18292 (20 μM), a PGC1α inhibitor (Fig. 2B). In addition, to determine whether PGC1α downregulation could alter mitochondrial ROS, we treated cells with erastin (10 μM) and measured changes in mitochondrial ROS by flow cytometry using a fluorescent probe mitoSOX. After erastin treatment, HT1080 cells transfected with PGC1α shRNA showed lower levels of mitochondrial ROS than HT1080 cells transfected with control shRNA (Fig. 2C). To determine cytoplasmic ROS, a CellROX^Ⓡ Deep Red fluorescent probe was used to measure cytoplasmic ROS levels. Cytoplasmic ROS levels remained unchanged regardless of the presence or absence of erastin or PGC1α down-regulation (Fig. 2D, E). These results suggest that ROS generated by erastin is specific to mitochondria.

Down-regulation of PGC1α blocks erastin-induced mitochondria-dependent lipid peroxidation

We hypothesized that PGC1α regulation might contribute to erastin-induced mitochondria-dependent lipid peroxidation. Using a fluorescent probe MitoPeDPP, confocal imaging was used to investigate mitochondrial lipid peroxidation. In the presence of erastin (10 μM), the green fluorescence signal showed an increase (Fig. 3A, middle panels) in comparison with that in the vehicle control (Fig. 3A, top panels). Compared to erastin treatment alone, green fluorescence signals enhanced by erastin were nearly completely prevented with PGC1α suppression (Fig. 3A, lower panels). Following treatment with erastin (10 μM), mitochondrial lipid peroxidation levels were significantly increased in HT1080 cells. Conversely, SR18292 (20 μM), a PGC1α inhibitor, markedly reduced mitochondrial lipid peroxidation levels. Treatment with erastin can increase generation of mitochondrial ROS, which in turn can lead to ATP depletion, ∆Ψm dissipation, and opening of the mitochondrial permeability transition pore (mPTP) (8, 9). A fluorescence probe JC-1 was used to evaluate mitochondria membrane potential to ascertain whether PGC1α altered erastin-induced mitochondria dysfunction. In HT1080 cells, the mitochondrial membrane potential level decreased at 12 hours after erastin administration. In the presence of SR18292, a PGC1α inhibitor, the loss of mitochondrial membrane potential caused by erastin was dramatically restored (Fig. 3B). Additionally, mitochondrial membrane potential in HT1080 cells transfected with PGC1α shRNA or control shRNA transfected cells was measured. Erastin-induced mitochondrial membrane potential reduction was restored in HT1080 cells transfected with PGC1α shRNA compared to HT1080 cells transfected with control shRNA transfected cells (Fig. 3C). Taken together, these results suggest that PGC1α down-regulation can guard against mitochondrial-dependent lipid peroxidation and mitochondria dysfunction caused by erastin.

Down-regulation of PGC1α reduces erastin-induced lipid peroxidation

Since down-regulation of PGC1α prevented erastin-induced mitochondria-dependent lipid peroxidation and mitochondrial dysfunction, an experiment was conducted to determine whether mitochondria-dependent lipid peroxidation regulated by PGC1 α could affect cellular lipid peroxidation. Confocal imaging with SR18292 (20 μM) and BODIPY-C11, a fluorescent probe for detecting lipid peroxidation, was used to analyze HT1080 cells treated with vehicle or erastin (10 μM) in the presence or absence of PGC1α inhibitors in order to examine the involvement of PGC1α in erastin-induced lipid peroxidation. Under vehicle conditions, an extremely weak signal of oxidized BODIPY-C11 was found, whereas a red fluorescence distribution, mainly in a reduced state, was observed in the intracellular compartment (Fig. 4A, top panel). The oxidative signal was rapidly increased after erastin (10 μM) treatment for 12 hours in HT1080 cells (Fig. 4A, middle panel). In contrast, compared to treatment with erastin alone, the enhanced oxidative signal was nearly abolished by suppression of PGC1α (Fig. 4A, bottom panel). Flow cytometry was used to investigate lipid peroxidation. In HT1080 cells, erastin administration increased lipid peroxidation relative to the vehicle control treatment (Fig. 4B). Similarly, transfected HT1080 cells were treated with erastin (10 μM) and the fluorescent probe C11-BODIPY was used to assess lipid peroxidation by flow cytometry. Erastin administration promoted lipid peroxidation relative to the vehicle control in HT1080 cells transfected with control shRNA, but not in HT1080 cells transfected with PGC1α shRNA (Fig. 4C). Lipid peroxidation induced by erastin was reduced by down-regulation of PGC1α. These results demonstrate that the level of erastin-induced lipid peroxidation is regulated by the expression of PGC1α.

Ferroptosis is a unique cell death that has been associated with a number of disorders. It is a programmed cell death defined by iron-dependent lipid peroxidation (20). Ferroptotic cell death has morphological, biochemical, and genetic characteristics that distinguish it from other forms of cell death. Ferroptosis is characterized by changes in mitochondrial membrane density, cristae reduction, or membrane rupture when the balance between mitochondrial fission and fusion is disrupted. PGC1α is a protein that is regulated in response to mitochondrial stimulation. It plays a key role in regulating mitochondrial biogenesis and function (21). However, the specific role of PGC1α in erastin-induced mitochondrial dysfunction during ferroptotic cell death has not been fully elucidated yet. The purpose of this study was to investigate the involvement of PGC1α in erastin-induced ferroptosis in HT1080 cells and to identify potential therapeutic targets. Here, we demonstrated the importance of PGC1α in ferroptotic cell death by showing that PGC1α inhibition could protect HT1080 cells from erastin-induced cell death. In Fig. 1, erastin-induced ferroptotic cell death was prevented by downregulation of PGC1α. These data suggest that PGC1α expression might be an important mediator of erastin-induced ferroptotic cell death. Fig. 2 demonstrated that suppression of mitochondrial ROS by downregulation of PGC1α protected HT1080 cells from erastin-induced ferroptotic cell death. PGC1α down-regulation prevented erastin-induced mitochondrial-dependent lipid peroxidation. In addition, PGC1α down-regulation prevented mitochondrial dysfunction by preventing erastin-induced loss of mitochondrial membrane potential (Fig. 3). In addition, cellular lipid peroxidation was prevented by PGC1α down-regulation (Fig. 4). These findings raise interesting questions about how PGC1α might influence mitochondrial lipid peroxidation. When PGC1α activity is reduced, mitochondrial content might decrease, resulting in lower overall ROS production and consequently leading to reduced lipid peroxidation. This reduction in mitochondrial mass might underlie protective effects observed during ferroptosis. Additionally, cells may activate compensatory pathways to adapt to the decrease of PGC1α activity. For example, antioxidant systems such as GPX4 could be upregulated to counteract the accumulation of ROS and lipid peroxidation (22). Our data supported this notion, as GPX4 expression was restored in cells where PGC1α was inhibited or knocked down (Supplementary Fig. 2). This adaptive response likely contributes to enhanced resistance to ferroptosis and highlights the interplay between PGC1α activity and key regulators of ferroptosis. By influencing both mitochondrial biogenesis and antioxidant defenses, PGC1α appears to play a multifaceted role in maintaining cellular homeostasis during ferroptosis. Further studies are needed to explore how these mechanisms interact and whether additional ferroptosis regulators such as FSP1 and xCT are similarly affected by changes in PGC1α activity. These insights into PGC1α’s role in ferroptosis not only can advance our understanding of this cell death pathway, but also suggest new avenues for therapeutic interventions targeting mitochondrial dysfunction and ROS regulation. Taken together, these data suggest that PGC1α plays a critical role in mitochondrial dysfunction during erastin-induced ferroptosis of HT1080 cells.

Reagents

Erastin (329600) (EMD Millipore Corporation, USA), RSL3 (HY-100218A) (MedChemExpress, NJ 08852, USA), mito-TEMPO (SML0737) (Sigma-Aldrich, St. Louis, MO, USA), SR18292 (HY-101491) (MedChemExpress, NJ 08852, USA), anti-PGC1α antibody (SC-517380) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were purchased for reagents.

Cell culture

HT1080 fibrosarcoma cells were obtained from ATCC (Manassas, VA, USA) and cultivated in RPMI 1640 media (Life Technologies, Grand Island, NY) supplemented with 1X penicillin-streptomycin (100 U/ml) and 10% fetal bovine serum (Gibco, Waltham, MA, USA). All cells were incubated at 37°C in a humidified atmosphere of 5% CO₂ and 95% air.

Cell viability assay

Using the CellTiter 96^Ⓡ AQueous One Solution Cell Proliferation Assay kit (Promega, Madison, WI), the MTS assay was used to assess the viability of the cells. In 96-well plates, cells were seeded at a density of 0.7 × 10⁴ cells per well. Each well received 20 μl of MTS solution after the reagent treatment. The plates were incubated at 37°C for an extra 2 hours. The absorbance at 490 nm was then measured using a microplate reader to determine the cell survival percentages.

Construction of shRNA expressing HT1080 cells

Using transfection reagents (#E2691) (Promega, Madison, WI), HT1080 cells were transfected with PGC1α shRNA and nonspecific control shRNA (Sigma-Aldrich, St. Louis, MO, USA). The sequences of human PGC1α shRNA were as follow: 5’-CCG GCC TCC TCA TAA AGC CAA CCA ACT CGA GTT GGT TGG CTT TAT GAG GAG GTT TTT-3’. In stable cells, PGC1α and β-actin were examined.

Western blot analysis

The cells were harvested using RIPA buffer containing 1X Complete^TM Protease Inhibitor Cocktail (#39922700) (Roche Applied Science, Mannheim, Germany). Thermo Scientific, Rockford, IL’s Pierce BCA protein assay kit (#23225) was used to measure the protein contents in cell lysates. Following resolution on PAGE gels containing 12% sodium dodecyl sulfate, the materials were transferred to PVDF (polyvinylidene difluoride) membranes for an overnight period at a current of 120 mA. Using a 5% nonfat milk solution in TBST buffer (20 mM Tris–HCl, pH 7.4, 500 mM NaCl, 0.1% Tween 20), membranes were blocked for 2 hours at room temperature. The blots were then incubated for a continuously at room temperature with different antibodies (diluted 1:1,000) in TBST. Following three TBST washes, the blots were incubated with anti-rabbit or anti-mouse secondary antibodies (diluted 1:5,000) in TBST for 1 hour at room temperature. Ultimately, immunoblots were observed upon exposure to X-ray film using SuperSignal^Ⓡ West Pico Chemiluminescent Substrate (#34580) (Thermo Fisher Scientific, Inc., Waltham, MA).

Quantitative real-time reverse transcription-polymerase chain reaction (qRT-PCR)

With the aid of the TRIzol reagent (Invitrogen, Life Technologies, Carlsbad, CA), total RNA was extracted from cultured cells. Reverse transcription was performed using the SuperScript^TM III First-Strand Synthesis System (#18080-044) (Thermo Fisher Scientific, Inc., Waltham, MA). iQ SYBR Green Supermix (#170-8882AP) (Bio-Rad, Hercules, CA) was used for real-time quantitative PCR. An internal control, β-actin (a loading control frequently used for gene knockdown in PCR), was employed. The StepOnePlus Real-Time PCR System’s absolute quantitation option was used for the analysis. The primers sequences were as follows: human PGC1α (forward: 5’- GGACATGTGCAGCCAAGACTCT-3’ and reverse: 5’- CACTTCAATCCACCCAGAAAGCT-3’), human β-actin (forward: 5’-ATCGTGCGTGACATTAAGGAGAAG-3’ and reverse: 5’-AGGAAGGAAGGCTGGAAGAGTG-3’).

Flow cytometry for cytosolic ROS and lipid peroxidation

In 6-well plates, cells were seeded at a density of 3 × 10⁵ cells per well. The following day, cells received a 12-hour reagent treatment. After 12 hours, the cells were treated for 30 minutes at 37°C in the dark with 2 μM C11-BODIPY581/591 or 2 mM CellROX^Ⓡ Deep Red (Invitrogen, Life Technologies, Grand Island, NY). The dye was removed after 30 minutes by washing with 2% FBS in PBS. The samples were centrifuged at 1,000 rpm for 3 minutes, and the pellet was resuspended in 500 μl of 2% FBS in PBS. Measurements were performed on a FACSCalibur (Becton Dickinson, San Jose, CA, USA) flow cytometer. At least 10,000 cells provided data for the collection.

Confocal microscopy

In 6-well plates, cells were seeded at a density of 1 × 10⁵ cells per well on glass coverslips and exposed to the reagents for 12 hours. Following the reagent treatment, the cells were exposed to 2 μM C11-BODIPY581/591 (lipid peroxidase) for 30 minutes at 37°C in the dark after being rinsed with PBS to remove the media. Following a fixation in 4% formaldehyde for 20 minutes. After nuclear counterstaining using Hoechst 33258 staining solution (1 μg/ml), they were mounted on slides using Fluorescence Mounting Medium (DAKO North America Inc., Carpinteria, CA).

Flow cytometry for mitochondrial ROS

HT1080 cells were seeded in 6-well plates in order to identify the formation of ROS by the mitochondria. Mitochondrial reactive oxygen species (ROS) production was monitored using MitoSOX red label (Invitrogen, Life Technologies, Grand Island, NY). After reagent treatment, cells were stained with MitoSOX red for 30 minutes at 37°C. Red fluorescence was identified by FACS analysis after cells were resuspended in PBS following collection and one PBS wash. At least 10,000 cells provided data for the collection.

Mitochondrial membrane potential

Using the MitoProbeTM JC-1 (Invitrogen, Life Technologies, Grand Island, NY), the MMP in HT1080 cells was measured after the reagents were exposed. 6-well plates were seeded with HT1080 cells in order to identify variations in MMP. Measurements were performed on a FACSCalibur (Becton Dickinson, San Jose, CA, USA) flow cytometer. At least 10,000 cells provided data for the collection.

LDH release assay

LDH release was measured using the Cytotoxicity Detection Kit (LDH) (Roche Applied Science, Mannheim, Germany). In 96-well plates, cells were seeded at a density of 0.7 × 10⁴ cells per well. Following reagent treatment, 100 μl of LDH assay buffer was used to reconstitute the positive control. Quantitative analysis was performed on cell culture supernatant (5 μl/well). LDH release percentages were then computed by measuring absorbance at 450 nm.

Statistical analysis

All results were confirmed in at least three independent experiments; data from one representative experiment are shown. Quantitative data are shown as means ± standard deviation and significance of statistical analysis was determined with two-tailed, unpaired Student’s t-test. P-values < 0.05 were considered significant.

This work was supported by the Research Fund of University of Ulsan and the KRIBB Research Initiative Program (KGM5322523).

The authors have no conflicting interests.