CoChat

Eprenetapopt triggers ferroptosis, inhibits NFS1 cysteine desulfurase, and synergizes with serine and glycine dietary restriction

Open access

The DOI matches its Crossref record; 58 of the 58 references it cites were found in a registry or an index; 52 of the 76 sources its claims cite were checked for support.

Abstract

The mechanism of action of eprenetapopt (APR-246, PRIMA-1 MET ) as an anticancer agent remains unresolved, although the clinical development of eprenetapopt focuses on its reported mechanism of action as a mutant-p53 reactivator. Using unbiased approaches, this study demonstrates that eprenetapopt depletes cellular antioxidant glutathione levels by increasing its turnover, triggering a nonapoptotic, iron-dependent form of cell death known as ferroptosis. Deficiency in genes responsible for supplying cancer cells with the substrates for de novo glutathione synthesis ( SLC7A11 , SHMT2 , and MTHFD1L ), as well as the enzymes required to synthesize glutathione ( GCLC and GCLM ), augments the activity of eprenetapopt. Eprenetapopt also inhibits iron-sulfur cluster biogenesis by limiting the cysteine desulfurase activity of NFS1, which potentiates ferroptosis and may restrict cellular proliferation. The combination of eprenetapopt with dietary serine and glycine restriction synergizes to inhibit esophageal xenograft tumor growth. These findings reframe the canonical view of eprenetapopt from a mutant-p53 reactivator to a ferroptosis inducer.

Overview

CoChat Summary

TL;DR

Eprenetapopt (APR-246) depletes glutathione by increasing its turnover, triggers iron-dependent nonapoptotic ferroptosis, and inhibits NFS1 cysteine desulfurase rather than acting primarily as a mutant-p53 reactivator. Combining eprenetapopt with dietary serine and glycine restriction synergizes to inhibit esophageal tumor growth in vivo.

Key findings

  • Genome-wide CRISPR screens in OACM5.1 cells identified GSH synthesis genes as central modulators: SLC7A11 and GCLM dropped out in CRISPRko, GCLC was top enriched in CRISPRa, and mito-1C genes SHMT2 and MTHFD1L plus ESD also sensitized cells to eprenetapopt; TP53 and apoptotic machinery were not identified.
  • Eprenetapopt decreased reduced GSH without increasing GSSG, and 13C2-cystine tracing showed increased incorporation into glutamylcysteine, GSH, GSSG, and cysteinylglycine, with MQ-GSH and MQ-cysteine adducts detected, indicating increased GSH turnover/degradation rather than blocked de novo synthesis.
  • Eprenetapopt-induced cell death was rescued by ferrostatin-1 and ciclopirox olamine but not zVAD-FMK in H1299 and FLO-1 cells, and apoptosis-deficient MC38 cells lacking Bax/Bak/Bid/Casp3/Casp7 were still killed, supporting ferroptosis.
  • MQ inhibited NFS1 cysteine desulfurase activity in cell-free and 293T assays and reduced aconitase activity; dietary serine/glycine restriction plus eprenetapopt (100 mg/kg daily) inhibited primary tumor growth, delayed metastasis, and prolonged survival in FLO-1 LM mice, and limited esophageal adenocarcinoma PDX growth.

Methods

  • Multiomics approach: genome-wide CRISPRko (Brunello, 76,441 sgRNAs) and CRISPRa (Calabrese, 56,762 sgRNAs) screens in OACM5.1 esophageal cancer cells; untargeted LC-MS metabolomics and label-free quantitative proteomics after 50 μM eprenetapopt for 12 hours; DepMap/CTRPv2 correlation analyses.
  • Cell death and mechanism assays: PI uptake/IncuCyte and alamarBlue viability; ferroptosis rescue with Fer-1, CPX, zVAD-FMK, or NAC; apoptosis-resistant MC38 cells; NFS1 cysteine desulfurase assays measuring H2S release or 35S-cysteine persulfide formation; aconitase activity assays.
  • In vivo models: FLO-1 LM cell line xenografts and esophageal adenocarcinoma PDXs in NSG mice, randomized to normal or serine/glycine-free diet and eprenetapopt (100 mg/kg daily) or vehicle; tumor growth, bioluminescence metastasis, and survival analyzed with t tests, ANOVA, and log-rank tests.

Limitations

  • The paper states that the precise mechanism by which eprenetapopt depletes GSH and inhibits NFS1 remains to be determined.
  • Evident from the text: the CRISPRa screen was performed with n = 1, some assays had n = 1 or n = 2, and no power analyses were performed to determine sample size before experimentation.
  • Evident from the text: eprenetapopt monotherapy failed to inhibit primary tumor growth or metastatic spread in the FLO-1 LM model, and efficacy in culture was limited by high cell confluency and low O2 levels; in vivo synergy was shown only in esophageal cancer models.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1126/sciadv.abm9427

Cites

58 of 58 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1. Multiomics strategy to determine the MoA of eprenetapopt. (A) Schematic diagram showing a strategy to determine the MoA of eprenetapopt. (B) Heatmap of sensitivity to eprenetapopt analog (APR-017). Cancer lineages are ordered by sensitivity determined by delta median, area under the curve (AUC) of compound activity. Percentages denote the frequency of TP53 mutations in each lineage. CNS, central nervous system; PNS, peripheral nervous system. (C) Schematic showing the workflow for the CRISPRko and CRISPRa screens in OACM5.1 cells. (D) MAGeCK scores (negative indicating dropout and positive indicating “enrichment”) from CRISPR screens, plotted in order of magnitude. (E) Cellular metabolic activity measured by alamarBlue as a surrogate readout for cell viability following 72-hour exposure with eprenetapopt at indicated doses in cells transduced with individual sgRNA targeting identified hits. Heatmap represents the change in GI50 (dose where 50% growth inhibition is achieved) relative to control. (F) Fisher’s transformed z-scored Pearson correlation strength of gene dependency from the DepMap database with eprenetapopt analog (APR-017) sensitivity data from CTRPv2, plotted in order of magnitude. (G) Comparison of CRISPR screens and DepMap gene dependency data shows overlay of glutamate-cysteine ligase units (GCLC and GCLM). Plot is representing the rank of top hits (ordered by dropout in CRISPRko, by enrichment in CRISPRa, and by positive correlations in DepMap). Dashed lines indicate overlap of top 50 ranked genes. (H) Changes in polar metabolites determined by untargeted liquid chromatography–mass spectrometry (LC-MS) metabolomics and (I) proteins determined by label-free quantitative proteomics in OACM5.1 cells following treatment with 50 μM eprenetapopt for 12 hours compared to vehicle. Dotted lines indicate significance cutoffs {P < 0.05, |log2[fold change (FC)]| > 0.5}. Two-tailed unpaired t test (H and I). Error bars = SEM. (D) n = 2 for CRISPRko, n = 1 for CRISPRa, (E) n = 3 to 4, (H) n = 6, and (I) n = 4. See also fig. S1.
Fig. 1. Multiomics strategy to determine the MoA of eprenetapopt. (A) Schematic diagram showing a strategy to determine the MoA of eprenetapopt. (B) Heatmap of sensitivity to eprenetapopt analog (APR-017). Cancer lineages are ordered by sensitivity determined by delta median, area under the curve (AUC) of compound activity. Percentages denote the frequency of TP53 mutations in each lineage. CNS, central nervous system; PNS, peripheral nervous system. (C) Schematic showing the workflow for the CRISPRko and CRISPRa screens in OACM5.1 cells. (D) MAGeCK scores (negative indicating dropout and positive indicating “enrichment”) from CRISPR screens, plotted in order of magnitude. (E) Cellular metabolic activity measured by alamarBlue as a surrogate readout for cell viability following 72-hour exposure with eprenetapopt at indicated doses in cells transduced with individual sgRNA targeting identified hits. Heatmap represents the change in GI50 (dose where 50% growth inhibition is achieved) relative to control. (F) Fisher’s transformed z-scored Pearson correlation strength of gene dependency from the DepMap database with eprenetapopt analog (APR-017) sensitivity data from CTRPv2, plotted in order of magnitude. (G) Comparison of CRISPR screens and DepMap gene dependency data shows overlay of glutamate-cysteine ligase units (GCLC and GCLM). Plot is representing the rank of top hits (ordered by dropout in CRISPRko, by enrichment in CRISPRa, and by positive correlations in DepMap). Dashed lines indicate overlap of top 50 ranked genes. (H) Changes in polar metabolites determined by untargeted liquid chromatography–mass spectrometry (LC-MS) metabolomics and (I) proteins determined by label-free quantitative proteomics in OACM5.1 cells following treatment with 50 μM eprenetapopt for 12 hours compared to vehicle. Dotted lines indicate significance cutoffs {P < 0.05, |log2[fold change (FC)]| > 0.5}. Two-tailed unpaired t test (H and I). Error bars = SEM. (D) n = 2 for CRISPRko, n = 1 for CRISPRa, (E) n = 3 to 4, (H) n = 6, and (I) n = 4. See also fig. S1.CC BY · PubMed Central
Fig. 2. Eprenetapopt and mito-1C metabolism. (A) Schematic diagram illustrating the connections between de novo GSH synthesis and mito-1C metabolism. THF, tetrahydrofolate; GSH, reduced GSH; MQ-Cys, MQ-conjugated cysteine; MQ-GSH; MQ-conjugated GSH. (B) Cellular metabolic activity measured by alamarBlue as a surrogate readout for cell viability compared to complete medium (CM) following 72 hours of serine, glycine, or SG deprivation and glycine deprivation rescued with 1 mM GSH monoethyl ester (GSH) in OACM5.1 cells. (C) Relative cell number following treatment with 10 μM eprenetapopt ± 1 mM glycine supplementation for 4 days. + indicates glycine in RPMI 1640, and ++ indicates supplementation with additional 1 mM glycine. (D) Top: Growth curves and overall survival (time to reach a tumor volume of ≥1400 mm3) of nonobese diabetic–severe combined immunodeficient interleukin-2RγKO (NSG) mice inoculated with FLO-1 LM tumors treated with eprenetapopt (EPR; 100 mg/kg, daily) or vehicle on either normal or SG-free diets for 35 days. Bottom: Representative bioluminescence images illustrating the metastatic burden in mice after 35 days of treatment. (E) Growth curves of esophageal adenocarcinoma PDXs in NSG mice treated with eprenetapopt (100 mg/kg, daily) or vehicle on either normal or SG-free diet for 21 days. (F) Fractional labeling of de novo GSH synthesis intermediates from 13C2-cystine following 12-hour exposure to 50 μM eprenetapopt compared to vehicle in OACM5.1 cells. Two-tailed t test (B and C), one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test (D and E), and log-rank (Mantel-Cox) test (D). *P < 0.05, **P < 0.01, and ***P < 0.001. NS, not significant. Error bars = SEM. (A) n = 1, (B and F) n = 3, (C) n = 5, (D) n = 8, and (E) n = 4 to 5. See also fg. S2.
Fig. 2. Eprenetapopt and mito-1C metabolism. (A) Schematic diagram illustrating the connections between de novo GSH synthesis and mito-1C metabolism. THF, tetrahydrofolate; GSH, reduced GSH; MQ-Cys, MQ-conjugated cysteine; MQ-GSH; MQ-conjugated GSH. (B) Cellular metabolic activity measured by alamarBlue as a surrogate readout for cell viability compared to complete medium (CM) following 72 hours of serine, glycine, or SG deprivation and glycine deprivation rescued with 1 mM GSH monoethyl ester (GSH) in OACM5.1 cells. (C) Relative cell number following treatment with 10 μM eprenetapopt ± 1 mM glycine supplementation for 4 days. + indicates glycine in RPMI 1640, and ++ indicates supplementation with additional 1 mM glycine. (D) Top: Growth curves and overall survival (time to reach a tumor volume of ≥1400 mm3) of nonobese diabetic–severe combined immunodeficient interleukin-2RγKO (NSG) mice inoculated with FLO-1 LM tumors treated with eprenetapopt (EPR; 100 mg/kg, daily) or vehicle on either normal or SG-free diets for 35 days. Bottom: Representative bioluminescence images illustrating the metastatic burden in mice after 35 days of treatment. (E) Growth curves of esophageal adenocarcinoma PDXs in NSG mice treated with eprenetapopt (100 mg/kg, daily) or vehicle on either normal or SG-free diet for 21 days. (F) Fractional labeling of de novo GSH synthesis intermediates from 13C2-cystine following 12-hour exposure to 50 μM eprenetapopt compared to vehicle in OACM5.1 cells. Two-tailed t test (B and C), one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test (D and E), and log-rank (Mantel-Cox) test (D). *P < 0.05, **P < 0.01, and ***P < 0.001. NS, not significant. Error bars = SEM. (A) n = 1, (B and F) n = 3, (C) n = 5, (D) n = 8, and (E) n = 4 to 5. See also fg. S2.CC BY · PubMed Central
Fig. 3. Eprenetapopt triggers ferroptosis. (A) Percentage (%) of dead cells [as determined by % PI+ cells] and percentage of cell confluency following treatment for 24 hours with 50 μM eprenetapopt with or without 50 μM zVAD-FMK (pan-caspase inhibitor), 12.5 μM Fer-1 (lipophilic antioxidant), 6.25 μM CPX (iron chelator), or 2.5 mM NAC (cysteine supplement) in H1299 (left) and FLO-1 cells (right). (B) Representative merged phase and red channel images of results from (A) acquired on IncuCyte. (C) Percentage of cell death determined by PI uptake following 24-hour treatment with 50 μM eprenetapopt in H1299 cells at indicated cell densities. (D) Immunoblot of Bax, Bak, Bid, Casp3, and Casp7 illustrating the efficiency of CRISPR editing in MC38 cells. (E) Left: Cell death in MC38 wild-type (WT) and apoptosis-deficient (Apop KO) cells following 24-hour treatment with 50 μM eprenetapopt or 12.5 μM STS. Right: Cellular metabolic activity measured by alamarBlue as a surrogate readout for cell viability following 72-hour exposure with eprenetapopt at indicated doses in wild-type and apoptosis-deficient MC38 cells. Two-way t test (A and E). *P < 0.05, **P < 0.01, and ****P < 0.0001. (A) n = 2 to 3, (D) n = 2, representative blots shown, and (E) n = 3. See also fig. S3 and movie S1.
Fig. 3. Eprenetapopt triggers ferroptosis. (A) Percentage (%) of dead cells [as determined by % PI+ cells] and percentage of cell confluency following treatment for 24 hours with 50 μM eprenetapopt with or without 50 μM zVAD-FMK (pan-caspase inhibitor), 12.5 μM Fer-1 (lipophilic antioxidant), 6.25 μM CPX (iron chelator), or 2.5 mM NAC (cysteine supplement) in H1299 (left) and FLO-1 cells (right). (B) Representative merged phase and red channel images of results from (A) acquired on IncuCyte. (C) Percentage of cell death determined by PI uptake following 24-hour treatment with 50 μM eprenetapopt in H1299 cells at indicated cell densities. (D) Immunoblot of Bax, Bak, Bid, Casp3, and Casp7 illustrating the efficiency of CRISPR editing in MC38 cells. (E) Left: Cell death in MC38 wild-type (WT) and apoptosis-deficient (Apop KO) cells following 24-hour treatment with 50 μM eprenetapopt or 12.5 μM STS. Right: Cellular metabolic activity measured by alamarBlue as a surrogate readout for cell viability following 72-hour exposure with eprenetapopt at indicated doses in wild-type and apoptosis-deficient MC38 cells. Two-way t test (A and E). *P < 0.05, **P < 0.01, and ****P < 0.0001. (A) n = 2 to 3, (D) n = 2, representative blots shown, and (E) n = 3. See also fig. S3 and movie S1.CC BY · PubMed Central
Fig. 4. Eprenetapopt inhibits NFS1 cysteine desulfurase activity. (A) Schematic diagram summarizing the role of NFS1 cysteine desulfurase and important iron-sulfur cluster containing enzymes (TCA, tricarboxylic acid; ETC, electron transport chain; POLε, DNA polymerase epsilon). mAco, mitochondrial aconitase; cAco, cytosolic aconitase. (B) Fisher’s transformed z-scored Pearson correlation strength of NFS1 gene dependency and the compound activity of the 481 CTRPv2 compounds. (C) Heatmap (top) and bar chart (bottom) demonstrating cell viability following 72-hour eprenetapopt and elesclomol (ELE) cotreatment in OACM5.1 cells. (D) In-gel aconitase assay and Coomassie stain (C. stain) (for protein loading) in OACM5.1 cells following treatment with 50 μM eprenetapopt for indicated times. (E) Cell-free NFS1 cysteine desulfurase assay at indicated doses of MQ and cysteine following 1-hour incubation measured by the release of hydrogen sulfide (H2S) by dithiothreitol (DTT) coupled to N,N-dimethyl-p-phenylenediamine (DMPD) hydrochloride to produce methylene blue. (F) Top: Two micromolars of purified NFS1 was incubated with indicated concentrations of MQ, and reaction was started with 35S-cysteine for 1 hour at 30°C, followed by separation on 12% nonreducing SDS–polyacrylamide gel electrophoresis (PAGE). Bottom: ImageJ quantification of the resulting bands. (G) Top: NFS1 cysteine desulfurase activity in human embryonic kidney 293T (293T) cells following 24-hour exposure with MQ at indicated doses measured by the release of H2S mediated by DTT and NFS1 cofactor pyridoxal 5′-phosphate. Bottom: Aconitase activity following 24-hour exposure with MQ at indicated doses in 293T cells. (H) Schematic diagram summarizing the MoA of eprenetapopt. Two-way unpaired t test (C and G), **P < 0.01, ***P < 0.001, and ****P < 0.0001. Error bars = SEM. (C, E, and G) n = 3. (D) n = 2, representative gel shown, and (F) n = 1.
Fig. 4. Eprenetapopt inhibits NFS1 cysteine desulfurase activity. (A) Schematic diagram summarizing the role of NFS1 cysteine desulfurase and important iron-sulfur cluster containing enzymes (TCA, tricarboxylic acid; ETC, electron transport chain; POLε, DNA polymerase epsilon). mAco, mitochondrial aconitase; cAco, cytosolic aconitase. (B) Fisher’s transformed z-scored Pearson correlation strength of NFS1 gene dependency and the compound activity of the 481 CTRPv2 compounds. (C) Heatmap (top) and bar chart (bottom) demonstrating cell viability following 72-hour eprenetapopt and elesclomol (ELE) cotreatment in OACM5.1 cells. (D) In-gel aconitase assay and Coomassie stain (C. stain) (for protein loading) in OACM5.1 cells following treatment with 50 μM eprenetapopt for indicated times. (E) Cell-free NFS1 cysteine desulfurase assay at indicated doses of MQ and cysteine following 1-hour incubation measured by the release of hydrogen sulfide (H2S) by dithiothreitol (DTT) coupled to N,N-dimethyl-p-phenylenediamine (DMPD) hydrochloride to produce methylene blue. (F) Top: Two micromolars of purified NFS1 was incubated with indicated concentrations of MQ, and reaction was started with 35S-cysteine for 1 hour at 30°C, followed by separation on 12% nonreducing SDS–polyacrylamide gel electrophoresis (PAGE). Bottom: ImageJ quantification of the resulting bands. (G) Top: NFS1 cysteine desulfurase activity in human embryonic kidney 293T (293T) cells following 24-hour exposure with MQ at indicated doses measured by the release of H2S mediated by DTT and NFS1 cofactor pyridoxal 5′-phosphate. Bottom: Aconitase activity following 24-hour exposure with MQ at indicated doses in 293T cells. (H) Schematic diagram summarizing the MoA of eprenetapopt. Two-way unpaired t test (C and G), **P < 0.01, ***P < 0.001, and ****P < 0.0001. Error bars = SEM. (C, E, and G) n = 3. (D) n = 2, representative gel shown, and (F) n = 1.CC BY · PubMed Central

Verification

Claims the paper makes

35 Valid · 1 Not valid · 14 Not checked · 87 Own results and interpretations

74 more claims · 17 with their cited sources checked

Create a free account to see all 137 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 30
  • It may state two separate claims as one. 33
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 58 references

Cited by

Metadata

Authors
Fujihara, K. M., Zhang, B. Z., Jackson, T. D., Ogunkola, M. O., Nijagal, B., Milne, J. V., Sallman, D. A., Ang, C.-S., Nikolic, I., Kearney, C. J., Hogg, S. J., Cabalag, C. S., Sutton, V. R., Watt, S., Fujihara, A. T., Trapani, J. A., Simpson, K. J., Stojanovski, D., Leimkühler, S., Haupt, S., Phillips, W. A., & Clemons, N. J.
Affiliations
  • The University of Melbourne (AU)
  • University of Potsdam (DE)
  • Moffitt Cancer Center (US)
  • Victorian Centre for Functional Genomics, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia.
  • Memorial Sloan Kettering Cancer Center (US)
  • Peter MacCallum Cancer Centre (AU)
Corresponding author
  • Fujihara, Kenji M.
  • Clemons, Nicholas J.
Venue
Science Advances
Published
2022-09-16
Volume
8
Issue
37
Pages
eabm9427-eabm9427
ISSN
2375-2548 (ISSN-L)
OpenAlex
W4295900094
Cited by
102
References
58
Citations by year
2026: 30 · 2025: 30 · 2024: 24 · 2023: 15 · 2022: 3
FWCI
6.34
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • Science Advances (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Minerva Access (University of Melbourne) (submittedVersion, cc-by)
  • publish.UP (University of Potsdam) (submittedVersion, cc-by)
Funders
National Cancer Institute (P30CA008748)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Cancer Research › Ferroptosis and cancer prognosis
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Biochemistry › Sulfur Compounds in Biology
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Cell death mechanisms and regulation
Keywords
APR-246, ferroptosis, glutathione depletion, glutathione synthesis, iron-sulfur cluster assembly, NFS1, cysteine desulfurase, dietary restriction, esophageal cancer, SLC7A11, SHMT2, MTHFD1L, GCLC, GCLM, tumor growth inhibition

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.