henhaobio
Sign in

bioRxivCancer BiologyPreprint

Targeting MAT2A-S-adenosylmethionine (SAM) Axis Attenuates DNA Damage Response and Cancer Stemness to Increase Platinum Sensitivity in Ovarian Cancer

Shu ZhangiD, Zhen FuiD, Yanchi Zhou, Tara X. MetcalfeiD, Truc T. VuongiD, Lindsay W Brubaker, Benjamin Guy BitleriD, Heather M. O’HaganiD, Kenneth P. NephewiD

PDFDOI

Abstract

Abstract Metabolic and epigenetic reprogramming drives development of platinum resistance and disease recurrence in high-grade serous ovarian cancer (HGSC), a major clinical challenge in the field. S-adenosylmethionine (SAM) is the universal methyl group donor, synthesized by methionine adenosyltransferase 2A (MAT2A) from methionine. Prior studies have linked altered SAM-dependent DNA methylation to acquired platinum resistance in HGSC, connecting metabolism and epigenetic regulation. However, how MAT2A-driven SAM synthesis coordinates the metabolic-epigenetic remodeling axis to affect platinum sensitivity remains unclear. Here, we report that MAT2A-driven SAM synthesis is required for platinum-induced alterations in DNA methylation and the enrichment of ovarian cancer stem cells (OCSCs). We found that MAT2A upregulation correlated with poor progression-free survival in OC patients, and both pharmacological inhibition and genetic knockdown of MAT2A increased sensitivity to cisplatin. To investigate the underlying epigenetic mechanism, we profiled genome-wide changes in DNA methylation using OVCAR3 treated with cisplatin (15M, 16hr) and/or MAT2A siRNA (48hr), which showed that MAT2A knockdown reversed platinum-induced DNA methylation dynamics. Subsequent analysis revealed enrichment of pathways associated with platinum resistance, DNA repair, and stemness. Mechanistically, inhibiting MAT2A abrogated both the platinum-induced hypermethylation at promoter regions and the SAMTOR-mTOR-S6K-FANCD2 signaling axis, resulting in accumulated R-loops, attenuated DNA repair activation in response to platinum, and enhanced platinum-induced DNA damage and cell death. Using a functional DNA damage reporter assay, we directly showed that MAT2A knockdown reduced DNA repair through homologous recombination (HR) and non-homologous end joining (NHEJ) pathways. By detecting key DNA damage response kinases governing HR and NHEJ signaling, we further demonstrated that MAT2A inhibition abrogated DNA repair activation in response to platinum. Furthermore, single-sample Gene Set Enrichment Analysis of paired primary and recurrent tumors from HGSC patients revealed an association between MAT2A expression and increased OCSC features in recurrent tumors. In vitro, MAT2A inhibition prevented cisplatin-induced enrichment of OCSCs, reduced stemness markers, and inhibited spheroid-forming ability, all of which were rescued by SAM supplementation. Together, our study reveals a previously unrecognized mechanism linking metabolism to epigenetic regulation through platinum-induced remodeling and establishes the MAT2A-SAM axis as a promising therapeutic target to enhance platinum sensitivity by abrogating DNA damage response and OCSC enrichment and ultimately reduce OCSC-driven disease recurrence in HGSC.

In-depth summary

Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (13 Oct 2026) · CC BY 4.0

Blocking MAT2A-driven SAM synthesis in ovarian cancer cells prevents cisplatin-induced DNA hypermethylation, impairs DNA repair signaling, and blocks stem cell enrichment, sensitizing MTAP-proficient and MTAP-deficient lines.

Most high grade serous ovarian cancers relapse with platinum resistance, partly through surviving cancer stem cells, yet how methionine-derived SAM coordinates platinum-induced methylation remodeling and stemness was unexplored in this disease. This work links MAT2A-SAM to methylome change, DNA damage response and stem cell enrichment, and shows effects independent of MTAP status.

Key results

  1. ShownCisplatin remodels the OVCAR3 methylome toward promoter-focused hypermethylation, and MAT2A knockdown reverses much of this, giving a less promoter-focused landscape.

    8-hour cisplatin raised the SAM/SAH index and global 5-mC rose by 16 hours. Hypermethylated CpGs: approx 46% at TSS200/TSS1500, 15.4% in CpG islands. siMAT2A+cisplatin vs cisplatin: hypomethylated sites 24.3% TSS200-1500, 41.8% intergenic, 86% of DMRs open sea. · Inhibiting MAT2A abrogates platinum-induced…

  2. ShownMAT2A knockdown combined with cisplatin raises apoptosis, DNA damage and R-loops while reducing HR and NHEJ repair activity and checkpoint kinase phosphorylation.

    siMAT2A+cisplatin increased comet tail moments vs either alone and apoptosis vs cisplatin alone. MAT2A knockdown lowered %GFP (NHEJ) and %mCherry (HR) vs vector. Cisplatin-induced phosphorylation of ATR, CHK1, ATM, CHK2 and DNA-PKcs was abrogated. · MAT2A inhibition increases platinum sensitivity…

  3. SuggestedSAM sensing through SAMTOR-GATOR1 connects MAT2A to mTOR-S6K-FANCD2 signaling, and exogenous SAM rescues repair signaling and limits DNA damage after MAT2A loss.

    siMAT2A increased Flag-Depdc5/BMT2 co-IP, reduced by SAM. 50 μM SAM restored FANCD2 and partly restored p-S6K, p-ATM, p-ATR, p-CHK1, p-CHK2, and lowered tail moments to a level comparable to or below cisplatin alone. · SAM synthesis mediates DDR through a…; Figure legends

  4. SuggestedPlatinum-induced cancer stem cell enrichment needs methionine and MAT2A-derived SAM, in both MTAP-proficient OVCAR3 and MTAP-null OVCAR5, and tracks with MAT2A in recurrent patient tumors.

    Cisplatin raised %ALDH+ cells 1.5-2 fold; methionine depletion and FIDAS-5 blocked this and spheroid formation. SAM restored ALDH+ increase under siMAT2A. In 24 paired primary/recurrent tumors, ΔMAT2A vs CSC score Pearson r=+0.556, p=0.005. · SAM synthesis is essential for platinum-enriched…

  5. SuggestedMAT2A is elevated in HGSC tumors, its post-chemotherapy rise associates with shorter progression-free survival, and its change tracks with DNMT changes in recurrent tumors.

    CPTAC tumor n=84 vs normal n=19. Post-NACT MAT2A vs PFS pooled Spearman ρ=-0.32, p=0.023 (n=57, three cohorts). ΔMAT2A vs ΔDNMT1 r=+0.564, p=0.0041; vs ΔDNMT3A r=+0.621, p=0.0012 (24 pairs). · MAT2A analysis; Altered MAT2A-driven SAM synthesis is associated…; Inhibiting MAT2A abrogates platinum-induced…

Methods, data and reproducibility

HGSC lines OVCAR3, OVCAR5 and Kuramochi were exposed to cisplatin, MAT2A knockdown or FIDAS-5, methionine depletion and exogenous SAM, then profiled by LC-MS, EPIC v2.0 methylation arrays, comet, repair reporters, western blots, ALDH and spheroid assays; patient datasets were analyzed by ssGSEA.

  • HGSC cell lines OVCAR3 (MTAP-high), Kuramochi (MTAP moderate) and OVCAR5 (MTAP-null), used across the in vitro assays in different combinations · human · n = three independent biological replicates
  • EPIC v2.0 methylome of OVCAR3 with siScramble, cisplatin, siMAT2A and combination treatments · human · GSE335264
  • Matched primary and recurrent HGSC tumor expression pairs for ssGSEA · human · n = 24 matched pairs · GSE295041
  • CPTAC ovarian cancer proteomics, tumor vs normal ovarian tissue (10 matched pairs) · human · n = tumor n=84, normal n=19
  • Three HGSC cohorts (Adzib, EGA, Jav) with post-NACT MAT2A expression and PFS; the listed accession is for the EGA cohort only · human · n = n=57 total · EGAD00001006456
  • Untargeted metabolomics of platinum-resistant vs platinum-sensitive CaOV3 HGSC cells · human · ST002010
  • OVCAR3 subcutaneous xenografts, vehicle vs cisplatin, MAT2A/ALDH1A1/SIRT1 IHC · mouse (human OVCAR3 cells in NSG mice)
Methods & tools
gseapy v1.1.12, SeSAMe v1.24.0, Infinium MethylationEPIC v2.0, Ingenuity Pathway Analysis, Probabilistic Genomic Concordance, TRRUST v2, Agilent MassHunter Quantitative Analysis v9.0, QuPath, CometScore 2.0, ImageJ/Fiji, FIDAS-5 (MAT2A inhibitor), MethPrimer
Reagents & environment
gseapy v1.1.12; SeSAMe R package v1.24.0; Agilent MassHunter Quantitative Analysis v9.0; CometScore 2.0; GraphPad Prism and FlowJo (versions not given).
Reproducibility gaps
No code repository stated. Methylation array deposited under GSE335264; CSC gene panel, primers and reagents are in supplemental tables (not in the text); some cohorts are named only (Adzib, Jav).
Limitations (6)
  • Clinical relevance needs confirmation in xenografts or patient-derived organoids; the functional work is mainly in vitro cell line models. · stated by the authors
  • Predicted-TFBS pathway enrichment cannot show direction (activated or suppressed), and downstream transcriptional consequences of platinum-induced epigenetic reprogramming remain undefined. · stated by the authors
  • SAM loss also alters protein and histone methylation, which may contribute to R-loop increase; lineage tracing is needed to tell reprogramming of survivors from support of existing stem cells. · stated by the authors
  • Linking methylome changes causally to resistance phenotypes would require additional approaches such as multi-omics. · stated by the authors
  • The FANCD2 axis rests on western blots and SAM rescue; no FANCD2 re-expression or depletion test, and most mechanistic assays use OVCAR3 or a few lines. · noted in this summary
  • Patient associations are correlative and modest (pooled ρ=-0.32, n=57 across heterogeneous cohorts), so they cannot show MAT2A causes poor outcome or stemness. · noted in this summary

Read this if you study platinum resistance, metabolic-epigenetic crosstalk, or MAT2A inhibition beyond MTAP-deleted cancers in high grade serous ovarian cancer.

Citations
0

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.