Cell lines were maintained at 37°C in a 5% CO₂ atmosphere. The A673, TC71, and EW8 cell lines were provided by Dr. Kimberly Stegmaier (Dana-Farber Cancer Institute, Boston, MA), the CB-AGPN cell line was obtained from the Childhood Cancer Repository (Children’s Oncology Group), the HT1080 (RRID: CVCL_0317) and U2OS (RRID: CVCL_0042) cell lines were obtained from ATCC, and the RD cell line was provided by Dr. Munir Tanas (University of Iowa, Iowa City, IA). The non-transformed BJ-tert (RRID: CVCL_6573) and RPE-tert (RRID: CVCL_4388) cell lines were obtained from ATCC. Cells were grown in Dulbecco’s Modified Eagle’s Media (DMEM) supplemented with 10% FBS, 100 IU ml^-1 penicillin and 100 µg ml^-1 streptomycin. DNA fingerprinting confirmation of cell lines was performed using the short tandem repeat method and cell lines were used within 8–10 passages of thawing.

Auranofin and TAK-228 (sapanisertib) were obtained from MedChemExpress. N-acetylcysteine, hydrogen peroxide, and doxycycline were obtained from Sigma. Puromycin was obtained from ThermoFisher Scientific. Auranofin was dissolved in DMSO and then used at a 1:1000 dilution in assays. TAK-228 was also prepared in DMSO and used at a 1:1000 dilution in assays. N-acetylcysteine was prepared fresh for each experiment in DMEM media.

Cell proliferation was measured using the AlamarBlue (resazurin; Sigma) fluorescence assay, as previously described (11, 12, 28, 29). Briefly, approximately 5 x 10⁴ cells were plated per well of a 96-well plate. Cells were then treated the next day with a range of drug concentrations for 72 hours. Fluorescence measurements were obtained, after adding the AlamarBlue reagent, using a FLUOstar Omega microplate reader (BMG Labtech). IC50 values were calculated using log-transformed and normalized data (GraphPad Prism 10.1.0). Cell viability was quantified using 0.4% Trypan Blue (ThermoFisher), which stains dead cells, with a DeNovix Cell Drop BF automated cell counter.

Protein synthesis was assessed using puromycin labeling (SUnSET technique), as described (11, 12, 30, 31). For labeling of newly synthesized proteins, puromycin (2 μg/mL) was added to cells at a 1:400 dilution. The cells were then incubated with the puromycin for one hour before cell lysates were collected, as described in the Immunoblotting section. Protein loading for the immunoblots was normalized using cell number.

Protein synthesis was assessed using O-propargyl-puromycin Click-iT labeling according to the manufacturer’s instructions (ThermoFisher Scientific) (32, 33). Flow cytometry was performed using a Becton Dickinson LSR II instrument. Briefly, cells were incubated with O-propargyl-puromycin (1:1000, 10 μM final concentration) for one hour. Cells were then collected using trypsin and fixed in a 4% paraformaldehyde fixative solution. Cells were washed by centrifugation and resuspended in Click-iT saponin-based permeabilization and wash reagent. Click-iT Plus reaction cocktail was added to the cell suspension for thirty minutes, after which the cells were washed by centrifugation. Flow cytometry was performed on a Becton Dickinson LSR II flow cytometer.

Protein extracts for immunoblotting were prepared by incubating cells in RIPA buffer (Boston BioProducts), supplemented with protease and phosphatase inhibitors (Halt Protease & Phosphatase Inhibitor Cocktail, EDTA-free; ThermoFisher Scientific), for 20 minutes. Supernatants were collected following a 15-minute centrifugation at 17,000 r.c.f. at 4°C. SDS-PAGE was used to separate proteins, which were then transferred to polyvinylidene difluoride membranes (Millipore). Antibodies to the following proteins were used in the immunoblots: puromycin (Millipore, #AF488, 1:2000; RRID: AB_2737590), 4E-BP1 (Cell Signaling, #9644, 1:1000; RRID: AB_2097841), p-4E-BP1–37/46 (Cell Signaling, #2855, 1:1000; RRID: AB_560835), RRM2 (Santa Cruz Biotechnology, #398294, 1:500; RRID: AB_2894824), PLK1 (Cell Signaling, #4513, 1:1000; RRID: AB_2167409), tubulin (Proteintech, #66031, 1:2000; RRID: AB_28834830), and vinculin (Proteintech, #26520, 1:5000; RRID: AB_2919877). Protein loading for the immunoblots was normalized using cell number.

ROS was quantified using a DCFDA (2’,7’-dichlorofluorescin diacetate) assay kit according to the manufacturer’s instructions (Abcam). Briefly, 1 x 10⁴ cells were plated per well of a 96-well plate in phenol red-free media. Drugs were then added the next day and incubated for the amount of time indicated in the figure legends. The DCDFA reagent, which was dissolved in the kit buffer at a 20 mM concentration, was added to the wells at a 1:800 dilution and then incubated for 45 minutes at 37°C. Fluorescence measurements (excitation 485 nM/emission 535 nM) were obtained using a FLUOstar Omega microplate reader (BMG Labtech).

Detection of DNA replication was performed in duplicate using a Click-iT EdU-488 kit (ThermoFisher Scientific) (34). Briefly, cells were labeled with 10 μM EdU for one hour and then harvested using trypsin and fixed using the Click-iT fixative buffer. Cells were then washed by centrifugation and resuspended in Click-iT saponin-based permeabilization and wash reagent. Click-iT Plus reaction cocktail was added to the cell suspension for thirty minutes, after which the cells were washed by centrifugation. Flow cytometry was performed on a Becton Dickinson LSR II flow cytometer.

The full-length 4E-BP1 cDNA, with alanine substitutions at Thr37, Thr46, Ser65, and Thr70 and a FLAG-tag, was obtained as a gene block (IDT; Coralville, IA) and cloned into a lentiviral vector as described in previous publications (12, 34).

RPPA analysis of cell lines were performed by the RPPA Core Facility at the MD Anderson Cancer Center. Cells were provided to the core facility as frozen pellets and the protein extraction, data normalization, and analysis were performed according to facility protocols (https://www.mdanderson.org/research/research-resources/core-facilities/functional-proteomics-rppa-core/education-and-references.html) (RRID: SCR_016649).

Student’s t-test two-tailed was used to calculate P-values for the comparison of two groups. Statistical analyses were conducted using GraphPad Prism 10.1.0 (RRID: SCR_002798).

To directly test the hypothesis that ROS inhibits protein synthesis in Ewing sarcoma cells we treated two cell lines, EW8 and TC71, with hydrogen peroxide for three hours and then measured nascent protein synthesis using puromycin labeling (SUnSET technique) and immunoblotting (30, 31). Figure 1A shows that hydrogen peroxide decreases protein synthesis in Ewing sarcoma cells and that this effect is attenuated by co-treatment of the cells with the ROS-scavenger n-acetylcysteine (NAC). We also used a flow cytometry approach (O-propargyl-puromycin) to quantify, in single cells, the effect of hydrogen peroxide on protein synthesis (32, 33). Notably, hydrogen peroxide reduced protein synthesis to the same extent as cycloheximide, a well-established inhibitor of protein synthesis that blocks translation elongation, and this effect was blocked by co-treatment with NAC ( Figures 1B, C ). We then tested additional sarcoma cell lines, HT1080 (fibrosarcoma), RD (rhabdomyosarcoma), and U2OS (osteosarcoma), that represent other sarcoma subtypes and found that hydrogen peroxide also inhibited protein synthesis, similar to the results obtained with the Ewing sarcoma cell lines ( Figure 1D ).

Next, we treated the Ewing sarcoma cells lines with hydrogen peroxide for one hour, in contrast to the three hours used in the previous experiments, and identified a significant decrease in protein synthesis at this earlier time point, albeit to a lesser extent than at the later time point ( Figure 1E ). We also treated Ewing sarcoma cells lines with hydrogen peroxide for three hours and then removed the hydrogen peroxide and allowed the cells to recover for twenty-one hours before measuring nascent protein synthesis. Figure 1F shows that a three-hour treatment with hydrogen peroxide is sufficient to cause prolonged inhibition of protein synthesis, that extends after the hydrogen peroxide is removed. In addition, this more prolonged inhibition of protein synthesis also reduced the level of the protein loading control, tubulin. Furthermore, staining total protein with Ponceau S showed a significant reduction in total protein levels at the later time point. Of note, the loading for these and subsequent immunoblots were normalized using cell number and not total protein due to the effects of the drugs on protein synthesis.

Auranofin is an inhibitor of thioredoxin reductase and regulator of oxidative stress that is used for the treatment of rheumatoid arthritis (35, 36). Auranofin was previously identified, using a multipronged drug screening approach, as a drug that inhibits the growth of Ewing sarcoma cells in vitro and in vivo (23). Of note, this sensitivity of Ewing sarcoma cell lines to auranofin was dependent on expression of the EWS-FLI1 oncoprotein (23). Consistent with this result, analysis of Cancer Dependency Map data (Broad Institute) identifies that Ewing sarcoma cell lines are significantly more sensitive to CRISPR-mediate knockout of thioredoxin reductase than other cancer cell lines ( Figure 2A ) (37, 38). Figure 2B demonstrates that treatment of Ewing sarcoma cells with auranofin increases reactive oxygen species, as detected using the fluorescent ROS-detector DCFDA (2’,7’-dichlorofluorescin diacetate). Auranofin, as previously reported, also decreases the growth of Ewing sarcoma cell lines in a 72 hour growth assay ( Figure 2C ) (23). Next, we assessed the effect of auranofin dose on protein synthesis using a 24 hour drug treatment and identified that inhibition of protein synthesis occurred in the low (2–5) micromolar dose range ( Supplementary Figure 1 ). To confirm that the effect of auranofin on protein synthesis is mediated by ROS we performed a co-treatment experiment with NAC and found that the ROS scavenger was able to block the effects of auranofin on nascent protein synthesis ( Figure 2D ). We also used flow cytometry (O-propargyl-puromycin) to investigate the effect of auranofin on protein synthesis in single cells and confirmed that the drug reduces protein synthesis ( Figures 2E, F ). The effect of auranofin on protein synthesis was then assessed at an earlier time point, 6 hours after drug addition, compared to the timepoint of 24 hours used in the previous experiments. Figure 2G demonstrates that auranofin inhibits protein synthesis in Ewing sarcoma cells as early as 6 hours after drug addition, although the effect is larger with a longer duration of drug treatment. We also used flow cytometry and O-propargyl-puromycin labeling to validate that auranofin inhibits protein synthesis at this earlier (6 hour) time point ( Supplementary Figures 1B, C ). Cell viability was not significantly compromised at either the 6 hour or 24 hour timepoints ( Supplementary Figures 1D, E ). We then tested additional sarcoma cell lines that represent other sarcoma subtypes and found that auranofin also inhibited protein synthesis, similar to the results obtained with the Ewing sarcoma cell lines ( Figure 2H ). Finally, we treated two non-transformed cell lines, BJ-tert and RPE-tert, with auranofin and then labeled newly synthesized proteins with puromycin. Supplementary Figure 2A shows that auranofin inhibits protein synthesis in these non-transformed cell lines, and the effect is rescued by treatment with NAC. However, in contrast to the Ewing sarcoma cell lines, auranofin had minimal effect on the growth of the BJ-tert and RPE-tert cell lines in dose response assays, suggesting that the differential toxicity of the drug with the Ewing sarcoma and non-transformed cells is due to an increased reliance on active protein synthesis in the cancer cells ( Supplementary Figures 2B, C ).

In previous work, we identified that Ewing sarcoma cell lines require active protein synthesis to maintain a stable level of the ribonucleotide reductase M2 (RRM2) protein, which is a potential therapeutic target in Ewing sarcoma tumors (12, 34, 39, 40). Consequently, we next evaluated whether auranofin and ROS downregulate RRM2 levels in Ewing sarcoma cells. Figure 3A demonstrates that auranofin depletes RRM2 in Ewing sarcoma cell lines and that this effect is blocked by co-treatment of the cells with the ROS-scavenger NAC. Knockdown or inhibition of RRM2 in Ewing sarcoma is lethal and Figure 3B shows that auranofin significantly impairs the growth of four different Ewing sarcoma cell lines and that this effect is abrogated by co-treatment with NAC. RRM2 is a subunit of ribonucleotide reductase, which is the rate limiting complex in the synthesis of deoxyribonucleotides, and required for progression through S-phase (41–43). Cell cycle analysis demonstrates that auranofin blocks DNA replication, as assessed using EdU labeling, and arrests cells throughout S-phase, as expected for a drug that inhibits ribonucleotide reductase activity ( Figure 3C ).

Reactive oxygen species are reported to inhibit protein synthesis via multiple mechanisms, including modification of cysteines in ribosomal proteins and activation of the translational repressor 4E-BP1 (14–16). Specifically, 4E-BP1 binds to the eukaryotic translation initiation factor eIF4E and prevents the formation of the translation initiation complex (26, 27). Notably, in previous work, we identified a role for 4E-BP1, which is regulated by phosphorylation, in blocking protein synthesis in Ewing sarcoma cells (11, 12). Consequently, we tested whether ROS activates 4E-BP1 in Ewing sarcoma cells. Figure 4A shows that hydrogen peroxide decreases the phosphorylation of 4E-BP1, which reflects activation of the translational repressor, in Ewing sarcoma cell lines and this effect is blocked by co-treatment with NAC. Similarly, auranofin also inhibits the phosphorylation of 4E-BP1 and, thereby, activates the repressor of cap-dependent protein translation ( Figure 4B ). To investigate the effect of active 4E-BP1 we previously generated multiple sarcoma cell lines that express a doxycycline-inducible, constitutively-active 4E-BP1 protein with alanine substitutions at the phosphorylation sites Thr37, Thr46, Ser65, and Thr70 (TO-4E-BP1-Ala) (12, 44). Figure 4C shows that the expression of constitutively-active 4E-BP1-Ala in Ewing sarcoma cells inhibits protein synthesis and reduces the level of the RRM2 protein, which we previously showed is regulated by 4E-BP1 (12). To identify additional proteins that are regulated by 4E-BP1 we evaluated the effect of the inducible, constitutively-active 4E-BP1-Ala on the proteome of multiple sarcoma cell lines, described in an earlier publication, using reverse phase protein arrays (RPPA) ( Figure 4D ; Supplementary Table 1 ) (12). Overlap of the RPPA data for the five cell lines identified that the expression of active 4E-BP1-Ala in all of the sarcoma cell lines results in the downregulation of the level of the PLK1 protein ( Figure 4E ; Supplementary Table 2 ).

PLK1 is a well-described therapeutic target in multiple types of cancer, including Ewing sarcoma (45–49). To validate PLK1 as a target of 4E-BP1 we treated the 4E-BP1-Ala cells with doxycycline and performed immunoblotting for PLK1. Figure 5A shows that the expression constitutively-active 4E-BP1 reduces the level of the PLK1 protein, consistent with the RPPA results. Auranofin also reduced the level of the PLK1 protein in Ewing ( Figure 5B ) and other ( Figure 5C ) sarcoma cell lines. In previous work, we identified that inhibition of mTORC1/2 activates 4E-BP1 and inhibits protein synthesis in sarcoma cell lines (12). Figure 5D shows that treatment of sarcoma cell lines with the mTORC1/2 inhibitor TAK-228 activates 4E-BP1 and downregulates the level of the PLK1 protein ( Figure 5D ).