Sex as a biological variable. Our study exclusively examined female mice because the disease modeled is only relevant in females.

Cell culturing. The human OC cell lines OVCAR4, Kuramochi, and A2780 were cultured in RPMI-1640 (11875093, Thermo Fisher Scientific) supplemented with 10% FBS (35011CV, Corning). HEK 293 cells were cultured in DMEM (11965118, Thermo Fisher Scientific) containing 10% FBS. COV362 cells (human OC cell line) were cultured in DMEM supplemented with 10% FBS and 1% GlutaMAX (35050061, Thermo Fisher Scientific), while the ID8 mouse ovarian surface epithelial cell line was cultured in DMEM containing 5% FBS. FT-190, a human telomerase reverse transcriptase (hTERT)-immortalized human fallopian tube secretory epithelial cell line, cells were cultured in DMEM/F12 50:50 Mix [–] l-glutamine (15-090-CV, Corning) supplemented with 10% FBS and 2 mM l-glutamine (25030081, Thermo Fisher Scientific). All media were supplemented with 1% penicillin/streptomycin (15140122, Thermo Fisher Scientific). The cells were maintained in a humidified incubator at 37°C. FT-190, Kuramochi, and ID8 cells were provided by Daniela Matei (Northwestern University, Evanston, Illinois, USA). A2780, OV81, COV362, and OVCAR4 cells were provided by Charles Landen (University of Virginia, Charlottesville, Virginia, USA). To generate Pt-resistant cell lines, A2780, OV81, and COV362 cells were treated with IC50 doses of cisplatin at regular intervals (6). Meanwhile, ID8 and OVCAR4 cells were continuously exposed to carboplatin and cisplatin, respectively. All cell lines developed at least 3- to 4-fold resistance to Pt treatment.

Western blotting. Cells were lysed using RIPA buffer (BP-115, Boston BioProducts), and protein concentrations were determined using a bicinchoninic acid (BCA) assay (23225, Thermo Fisher Scientific). Proteins (1 μg/μL) were mixed with 4× sample buffer (BP-110R, Boston BioProducts) and boiled at 95°C for 10 minutes. For Western blot analysis, 20 μg protein was loaded onto a NuPAGE 4% to 12% Bis-Tris gradient gel (NP0335, Thermo Fisher Scientific) and run at 140 V for approximately 1 hour. Proteins were transferred onto a nitrocellulose membrane using the iBlot dry transfer system. Next, membranes were blocked with 5% milk dissolved in TBS-T (20 mM Tris, 150 mM NaCl, 0.1% Tween 20; pH 7.6) for 1 hour at room temperature (RT). After blocking, membranes were incubated overnight at 4°C with primary antibodies prepared in the blocking buffer. The following primary antibodies were used in this study: anti-PRMT5 (1:2,000, ab210437, Abcam), anti-SDMA (1:1,000, 13222, Cell Signaling Technology [CST]), anti-Actin (1:10,000, a2228, MilliporeSigma), anti-P53 (1:1,000, 2524, CST), anti-KEAP1 (1:1,000, ab227828, Abcam), anti-ubiquitin (1:1,000, 3936S, CST), anti-FLAG (1:1,000, F3165, MilliporeSigma), anti-γ-H2AX (1:1,000, 9718, CST), and anti-FER (1:500, 4268, CST). The next day, membranes were washed 3 times with TBS-T for 5 minutes each wash. Next, they were incubated for 1 hour at RT with either anti–mouse HRP conjugate (1:10,000, W402B, Promega) or anti–rabbit HRP conjugate (1:10,000, W401B, Promega). After the incubation, membranes were washed 3 times for 10 minutes each. Finally, membranes were covered with Western blot detection reagents (37074, Thermo Fisher Scientific) and visualized using the Invitrogen iBright imaging system (Thermo Fisher Scientific). To isolate cytoplasmic and nuclear fractions, cells were lysed using a Cell Fractionation Kit following the manufacturer’s instructions (9038, CST). Proteins from each fraction were then analyzed by Western blotting as described above.

qPCR. RNA extraction was performed using the Quick-RNA MiniPrep kit (R1054, Zymo Research). After isolation, 1 μg RNA was converted to cDNA using the High-Capacity RNA-to-cDNA Kit (4387406, Thermo Fisher Scientific). The cDNA was then diluted at a 1:5 ratio using nuclease-free water. qPCR was performed using Fast SYBR Green Master Mix (4385616, Thermo Fisher Scientific) with 15 ng cDNA per reaction, and each reaction was performed in triplicate. qPCR results were obtained using the QuantStudio 3 system, and Ct values were analyzed using the 2–ΔΔCt method (61). GAPDH expression was used as an internal control for normalization.

Immunofluorescence. Cells (1.5 × 105) were seeded onto coverslips in 6-well plates. The next day, cells were treated with the 1 μM EPZ015666 (CS7748, Selleckchem) and 10 μM carboplatin (S1215, Selleckchem) for 3 days. After treatment, cells were washed with PBS and fixed using 4% paraformaldehyde in PBS for 10 minutes at RT. Following fixation, cells were washed 3 times with ice-cold PBS and incubated with 0.25% Triton X-100 in PBS for permeabilization. Next, cells were washed 3 times for 5 minutes each with PBS and blocked using 1% BSA and 22.52 μg/mL glycine in PBS plus 0.1% Tween 20 (PBS-T) for 1 hour at RT. After blocking, cells were incubated with anti-γ-H2AX (1:500, 9718, CST) antibody in 1% BSA in PBS-T overnight at 4°C in a humidified chamber. The next day, cells were washed 3 times for 5 minutes each with PBS-T and then incubated with anti–rabbit Alexa Fluor 594 (1:500, A-11012, Invitrogen, Thermo Fisher Scientific) in 1% BSA in PBS-T for 1 hour at RT. Subsequently, cells were washed 3 times for 5 minutes each with PBS-T. Coverslips were then mounted onto microscopy slides using mounting medium with DAPI (S36939, Thermo Fisher Scientific). Finally, slides were visualized using the EVOS cell imaging system (Thermo Fisher Scientific), and the images were analyzed using ImageJ software (NIH).

Comet assay. Microscopy slides were coated with 1.5% normal melting agarose to create a base layer. Cells were then counted and resuspended at a concentration of 1 × 105 cells/mL. The cells were combined with low-melting agarose provided by the OxiSelect Comet Assay kit (STA-351, Cell Biolabs) at a 1:10 ratio, and 75 μL of this mixture was immediately transferred to the precoated slides and covered with a coverslip. After incubation at 4°C in the dark, the coverslips were carefully removed, and the slides were transferred to a container containing prechilled lysis buffer (provided by the kit). Following a 1-hour incubation with the lysis buffer at 4°C in the dark, the lysis buffer was replaced with prechilled alkaline solution (300 mM NaOH, 1 mM EDTA) and incubated for 30 minutes at 4°C in the dark. The slides were then transferred to an electrophoresis chamber filled with cold alkaline electrophoresis solution (prepared in-house) (300 mM NaOH, 1 mM EDTA, pH 13). Electrophoresis was performed at 17 V (1 volt/cm electrode distance) for 30 minutes at 4°C. After electrophoresis, the slides were washed 3 times for 2 minutes each with prechilled, distilled water. Subsequently, the slides were incubated once with cold 70% ethanol for 5 minutes and then air-dried at RT. Finally, the slides were stained with diluted Vista Green Dye (1:10,000 in Tris-EDTA buffer; STA-351, Cell Biolabs) for 15 minutes and visualized using the EVOS cell imaging system. Comet tails were measured using OpenComet software.

Cell-cycle analysis. Cells (1.5 × 105 cells/well) were seeded in 6-well plates. The following day, cells were either mock treated or treated with 250 μM or 1,000 μM EPZ015666 for 72 hours. After the treatment, cells were harvested via trypsinization and fixed using 90% methanol with gentle vortexing. The fixed cells were incubated at 4°C for 30 minutes, followed by 2 washes with PBS. The cells were then stained with DAPI (1 mg/mL) for 30 minutes at RT. Finally, a minimum of 50,000 events were analyzed using a BD FACSMelody Cell Sorter, and the data were analyzed with FlowJo software.

Lentivirus production and transduction. To generate virus, 4 × 106 HEK293 cells were seeded in a 10 cm dish. The next day, transgene (4 μg), psPAX2 (2 μg), and pmD2.G (1 μg) plasmids were transferred to an Eppendorf tube containing Opti-MEM (31985070, Thermo Fisher Scientific). Then, 21 μL of 1 mg/mL polyethylenimine (PEI) (24765, Polysciences) was added to the mixture (600 μL final volume), which was incubated for 15 minutes at RT. The mixture was then added dropwise to the cells. The following day, the media were replenished with fresh media, and the cells were incubated overnight. The next day, the culture media were collected and filtered using 0.45 μm filters (431225, Corning) and aliquoted for storage at –80°C.

For transduction, 1.25 × 105 cells were seeded in a 6-well plate. The next day, the virus was thawed, and polybrene (TR-1003-G, MilliporeSigma) was added at a final concentration of 10 μg/mL. The cell media were replaced with viral media, and the cells were incubated overnight. The following day, the media were replenished with fresh media, and the cells were incubated for 48 hours. Finally, selection was performed using the appropriate selection marker until all uninfected cells were dead.

CRISPR/Cas9-mediated gene manipulation. For gene KO, at least 2 different sgRNAs were designed using CRISPOR software, selecting the lenti-guide-puro protocol. Each oligonucleotide pair (10 μM final concentration) was mixed in annealing buffer (10 mM Tris, pH 8, 50 mM NaCl, 1 mM EDTA) in a total volume of 50 μL and incubated at 95°C for 5 minutes. The mixture was then allowed to slowly cool to RT. The annealed oligonucleotides were diluted 1:200 with sterile water. A ligation reaction was performed using 50 ng BsmBI-v2 (R0739S, New England Biolabs [NEB]) digested backbone, 1 μL of the diluted oligonucleotides, and T4 ligase (M0202S, NEB), with incubation overnight at 16°C. The next day, 2.5 μL of the ligation reaction was transformed into stable, competent E. coli (C3040H, NEB) and incubated overnight in the presence of ampicillin selection. Several colonies were selected and grown overnight. The following day, plasmid DNA was isolated using the QIAprep Spin Miniprep Kit (27106, Qiagen) and sent for Sanger sequencing to validate successful insertion. After subcloning the sgRNAs, lentivirus was generated using HEK 293 cells as explained above. Cell lines were then infected with these viruses and selected with 1 μg/mL puromycin (ant-pr-1, InvivoGen) until all uninfected cells were killed. After selection, cells were maintained for an additional 3 days, and then proteins were extracted using RIPA buffer. Finally, Western blotting was performed to determine the level of protein downregulation.

To achieve epigenetic downregulation of PRMT5 expression, a parental cell line expressing the dCas9-KRAB protein was first generated by infecting cells with a virus encoding the dCas9-KRAB protein. Simultaneously, sgRNAs targeting PRMT5 promoter regions were designed and subcloned using the same strategy described above. After delivering the sgRNAs, qPCR and Western blot analyses were performed to confirm PRMT5 downregulation.

Cell viability assays and synergy calculations. To determine cell viability, 1 × 103 cells/well were seeded in 96-well plates (3610, Corning). The next day, the cells were treated with different doses of EPZ015666 and carboplatin and incubated for 6 days. After incubation, the media were aspirated, and 44 μL CellTiter-Glo 2.0 Reagent (G9242, Promega) diluted 1:10 with colorless media (11058021, Thermo Fisher Scientific) was added to each well (40 μL media, 4 μL reagent). The luminescence signal was then recorded using a plate reader, and the results were analyzed with GraphPad Prism software. To assess the synergistic response, the SynergyFinder package in RStudio was used (62), along with the Bliss method (48).

Apoptosis assay. Cells were seeded in a 96-well plate at a density of 4 × 103 cells/well. The following day, treatments were administered using a drug of interest mixed with a 1:1,000 dilution of caspase 3/7 dye (10403, Biotum) and a 1:2,000 dilution of SiRDNA nuclei stain (CY-SC007, Spirochrome). Subsequently, cells were monitored with the IncuCyte (Sartorius) live imaging system, using phase, green, and red channels. The apoptosis rate was determined using the green integrated intensity/red object count, and the results were plotted in GraphPad Prism.

Analysis of public datasets. PRMT5 correlation values in HGSOC were obtained from the cBioPortal. The data were ranked according to correlation values and plotted using ggplot2. Positively correlated genes were identified on the basis of a Spearman correlation value of greater than 0.5 and a FDR of less than 0.001. These identified genes were then analyzed using DAVID GO term analysis, and the resulting terms were plotted in RStudio.

TCGA dataset was used to illustrate PRMT5 expression across various tumors, with expression values plotted using the ggplot2 package in RStudio. Comparison of PRMT5 expression levels between the fallopian tube, ovary, and ovarian serous cystadenocarcinoma (OSC) was conducted using the UCSC Xena Browser. Additionally, progression-free survival plots were generated with the Kaplan-Meier plotter online tool (https://kmplot.com/analysis/), using the following criteria: split patients by – median, histology - serous, stage - all, grade - 3, TP53 mutation - mutated.

TMA staining. TMAs were stained using anti-PRMT5 antibody (ab210437, Abcam) following the manufacturer’s instructions. The TMA was analyzed with QuPath software (38). Initially, TMA dearray was used to correctly assign annotations with the metadata. Subsequently, PRMT5 staining intensities were detected in individual cells. Following this, QuPath’s machine learning algorithm was utilized to distinguish between stroma and tumor cells. Finally, the staining intensities were converted to H-score values and plotted with ggplot2 in RStudio.

CRISPR screening using the druggable genome library. Cells were initially transduced with a Cas9-expressing construct. Following the generation of Cas9-expressing cells, the druggable genome library was transduced at approximately 0.3 MOI. Approximately 15 × 106 cells were transduced to achieve 500× coverage. The next day, the media were replenished, and puromycin (1 μg/mL) was added 48 hours later for selection. After selection, cells were split into 4 replicates and passaged every 4–5 days until confluent. After 2 weeks, they were trypsinized and resuspended in 200 μL PBS. Subsequently, they were fixed using 200 μL of a 1% formaldehyde solution while vortexing and then incubated for 10 minutes at 37°C. For permeabilization, ice-cold absolute methanol was added dropwise, and the cells were incubated for 30 minutes at 4°C, after which they were washed twice with PBS and resuspended in 400 μL PBS containing 2% BSA (PBS-BSA). Then, 1 μL anti–PE-PRMT5 (ab210437, Abcam) antibody was added, and the cells were incubated on ice for 30 minutes. After incubation, the cells were washed twice with PBS-BSA and resuspended in 4 mL PBS-BSA solution. The cells were then sorted to obtain populations with high (top 20%) and low (bottom 20%) PRMT5 levels using the BD FACSMelody Cell Sorter. The sorted cells were resuspended in lysis buffer (10 mM Tris-HCl, 150 mM NaCl, 10 mM EDTA, 0.1% SDS) supplemented with proteinase K (500 μg/mL, EO0491, Thermo Fisher Scientific) and RNase A (100 μg/mL, EN0531, Thermo Fisher Scientific) and incubated for 24 hours at 55°C. The following day, DNA was extracted using the standard phenol-chloroform extraction protocol. Libraries were prepared using the NEBNext High-Fidelity 2× PCR Master Mix (M0541L, NEB) PCR protocol with custom-barcoded oligonucleotides. In total, 11 PCR reactions per replicate were performed using 1 μg DNA per reaction. DNA was amplified for 25 cycles and loaded onto an agarose gel. Bands corresponding to the libraries were excised and purified using the QIAquick gel extraction kit (28704, Qiagen). Finally, the libraries were pooled and sequenced on the Illumina NextSeq 500 platform to obtain single-end reads. To analyze the reads, the CB2 package was utilized according to the package instructions in RStudio (40). Briefly, reads were mapped to a FASTA file containing sgRNA library sequences. The aligned reads were then used to calculate the average read count per sgRNA. After ensuring that each sgRNA had enough coverage (<500 reads per sgRNA), the data were normalized using counts per million (CPM) normalization. Subsequently, CPM-normalized reads were statistically compared using the measure_sgrna_stats function in CB2. Finally, sgRNA statistics were converted to gene-level enrichments using the “measure_gene_stats function.” Genes were assigned significance for PRMT5hi on the basis of a log2 fold change (FC) of greater than 0.25 and a P value of less than 0.01, and for PRMT5lo, a log2 FC of less than –0.25 and a P value of less than 0.001. Plots were generated using ggplot2.

Co-IP assay. HEK 293 cells (5 × 106) were seeded in 10 cm dishes and transfected with 5 μg FLAG-KEAP1 plasmid. MG132 (1 μM, M7449, MilliporeSigma) was added after 24 hours. At 48 hours, cells were washed once with PBS and collected by scraping. Following centrifugation, the cell pellet was lysed in 1 mL co-IP buffer (50 mM Tris, pH 8.0, 0.5% NP-40, 200 mM NaCl, 0.1 mM EDTA, 10% glycerol, and protease inhibitors). The samples were then incubated on ice for 30 minutes and centrifuged at 15,000g for 20 minutes at 4°C. Subsequently, 75 μL lysate was transferred into a new Eppendorf tube and mixed with 25 μL sample buffer to obtain whole-cell extract. To the remaining lysate, 2 μg Flag antibody (F3165, MilliporeSigma) was added and incubated overnight at 4°C with rotation. The next day, 30 μL protein A/G magnetic beads (88802, Thermo Fisher Scientific) per sample were equilibrated with co-IP buffer (by washing 3 times with 1 mL buffer) and added to the samples. After a 2-hour incubation at 4°C, the beads were immobilized and washed 3 times with 1 mL co-IP buffer and once with PBS. Finally, the beads were resuspended in 60 μL 2× sample buffer and boiled at 95°C for 10 minutes, and the samples were analyzed by SDS-PAGE.

IP analysis. HEK 293 cells were seeded in 10 cm dishes and transfected the next day with PRMT5 (5 μg), KEAP1 (5 μg), and ubiquitin plasmids (5 μg). Vector plasmid was also used to equalize transfection amounts. The following day, cells were treated overnight with either 1 μM MG132 or DMSO. Next, cells were scraped in ice-cold PBS and subsequently lysed using 2% SDS and 50 mM Tris, pH 8. Afterwards, they were briefly sonicated and then diluted 1:10 using IP buffer (50 mM Tris, pH 8, 200 mM NaCl, 0.1 mM EDTA, 0.5% NP-40, 10% glycerol, and protease inhibitors). Lysates were incubated with 5 μg anti-PRMT5 antibody (07-405, MilliporeSigma) at 4°C overnight with rotation. The next day, protein A/G magnetic beads were equilibrated in IP buffer and added to the lysates (30 μL/sample). After incubating the lysate-bead complex at 4°C for 2 hours, the beads were washed 4 times using IP buffer and once with PBS. Finally, the beads were resuspended in 2× sample buffer and boiled at 95°C for 10 minutes. The samples were subjected to Western blotting as explained above.

In vivo xenograft experiment. For the in vivo experiments, we used a PDX model developed by Dong et al. (63). The primary clinical and pathological characteristics of this model include high-grade serous carcinoma (HGSC), stage T3C, with pathologic lymph node status cannot be assessed (pNX) following total abdominal hysterectomy and bilateral salpingo-oophorectomy. The primary tumor measured 7 cm in diameter, and it harbored a p53 mutation. The patient’s age at diagnosis was not recorded. More detailed histological and molecular analyses of the PDX model are provided in the article. To create a carboplatin-resistant PDX model for this study, PDX tumor–engrafted mice were treated with carboplatin for approximately 4 weeks, after which tumor regrowth was allowed. The tumors were then passaged into new recipient mice and treated with carboplatin again. This cycle of carboplatin treatment was repeated at least 4 times to ensure the development of resistance. Six- to 7-week-old female NSG mice were obtained from The Jackson Laboratory (strain no. 005557). All mice were housed in a pathogen-free animal barrier facility. PDX tumors were s.c. engrafted into NSG mice. When the tumors reached a size of 100–250 mm3, the animals were randomly assigned to 4 groups. Carboplatin (HY-100235, MedChemExpress) was administered at 10 mg/kg once a week, while GSK591 (S1215, Selleckchem) was administered at 50 mg/kg 5 days a week. Both treatments were given via i.p. injection. Tumor growth was measured every 2–3 days using calipers. Animals were sacrificed when tumors reached approximately 1,500 mm3 in size.

ChIP with high-throughput sequencing. Cells (5 × 106) were seeded onto 15 cm dishes and grown until approximately 90% confluence. Proteins were crosslinked to DNA by adding formaldehyde (1% final concentration), followed by a 15-minute incubation at RT. The crosslinking reaction was then stopped by adding glycine (125 mM final concentration) to each sample and incubating for 5 minutes at RT. Media were aspirated, and cells were washed twice using 5 mL ice-cold PBS containing protease inhibitors. Cells were then scraped into 5 mL PBS and centrifuged at 4°C for 10 minutes at 1650g. Next, cells were resuspended in 1 mL SDS lysis buffer (1% SDS, 10 mM EDTA, 50 mM Tris-HCl, pH 8.1) and incubated on ice for 10 minutes. Following incubation, DNA was sonicated to obtain 100–400 bp fragments using a Diagenode Bioruptor sonicator (7 cycles, 30 seconds on, 30 seconds off). Sonicated samples were centrifuged at 24°C for 10 minutes at maximum speed. Samples were then diluted 1:5 using ChIP dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl, pH 8.1, 167 mM NaCl, protease inhibitors). Whole-cell extract (250 μL) was transferred to an Eppendorf tube and stored at 4°C for DNA extraction. Anti-PRMT5 (5 μg) (07-405, MilliporeSigma) and anti-H3K27ac (2 μg) (ab4729, Abcam) antibodies were added to the samples, which were incubated at 4°C overnight on a rotator. The next day, 30 μL protein A or G Dynabeads (10001D-10003D, Thermo Fisher Scientific) per sample were equilibrated with ChIP dilution buffer and added to the samples. After a 2-hour incubation at 4°C, Dynabeads were immobilized using a magnetic stand and washed twice with 1 mL Low Salt Immune Complex Wash Buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 150 mM NaCl). Beads were then washed twice with 1 mL LiCl Immune Complex Wash Buffer (0.25 M LiCl, 1% NP-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl, pH 8.1). Next, beads were washed twice with 1 mL TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). Beads were then resuspended in 125 μL elution buffer freshly supplemented with DTT (1× TE, pH 8.0, 1% SDS, 150 mM NaCl, 5 mM DTT) and incubated at 65°C for 10 minutes. Beads were immobilized, and eluted samples were transferred to a new tube. The elution step was repeated to obtain a total of 250 μL eluted samples. Ten microliters of 5 M NaCl, 25 μL of 10% SDS, and 1.25 μL of 1 M DTT were added to the whole-cell extract samples. Samples were incubated at 65°C overnight for reverse cross-linking. The next day, proteinase K (500 μg/mL, EO0491, Thermo Fisher Scientific) was added, and samples were incubated at 55°C for 2 hours. Finally, DNA was extracted from the samples using phenol-chloroform extraction. To make libraries, the NEBNext Ultra II DNA Library Prep Kit for Illumina (E7645S, NEB) was used following the manufacturer’s instructions. Libraries were sequenced using the Illumina HiSeq 4000 to obtain 50 bp single-end reads.

ATAC-Seq. Cells were trypsinized and counted to obtain 50,000 cells/sample. Cells were washed once with ice-cold PBS and centrifuged at 500g for 5 minutes at 4°C. Nuclei were extracted by resuspending the cells in 50 μL ATAC-Seq lysis buffer (10 mM Tris-HCl, 10 mM NaCl, 3 mM MgCl2, 0.1% Tween 20, 0.1% NP-40, 0.01% digitonin) and incubating on ice for 3 minutes. Next, 1 mL wash buffer (10 mM Tris-HCl, 10 mM NaCl, 3 mM MgCl2, 0.1% Tween 20) was added, and the tubes were inverted 3 times. Samples were centrifuged at 500g for 10 minutes at 4°C. The supernatant was discarded, and nuclei were dissolved in 50 μL transposition mix (25 μL 2× TDE buffer, 2.5 μL TDEI Tn5 enzyme, 16.5 μL PBS, 0.5 μL 1% Digitonin, 0.5 μL 10% Tween 20, 5 μL nuclease-free water), with pipetting done 6 times. Transposition was carried out for 30 minutes at 37°C. Next, transposed DNA was eluted using the Qiagen MinElute kit (28004, Qiagen) according to the manufacturer’s instructions. To prepare libraries, 13 cycles of PCR were performed using custom oligonucleotides as described before (64). Libraries were purified by double-sided bead purification using AMPure beads (A63881, Beckman Coulter). Libraries were pooled and sequenced on an Illumina NextSeq 500 to obtain 50 million paired-end reads per sample.

RNA-Seq. Cells (2 × 105) were seeded into 6-well plates in triplicate. The next day, the cells were treated with either DMSO or 5 μM EPZ015666 for 5 days. RNA was extracted using the Quick-RNA Miniprep Kit (R1054, Zymo Research). For library preparation, 1 μg input RNA was used following the NEBNext Ultra II Directional RNA Library Prep Kit for Illumina (E7760S, NEB) protocol. The prepared libraries were pooled and sequenced on an Illumina HiSeq 4000.

Data analysis. All sequencing data were first quality checked using FastQC. ATAC-Seq reads were aligned to the T2T reference genome using Bowtie2 software (settings: --local --very-sensitive) (65). Reads aligned to the mitochondrial genome were then removed using Samtools (66). PCR duplicates and blacklisted regions were removed using Picard and Bedtools, respectively. Peak calling was performed using MACS2 (settings: -f BAMPE --call-summits --keep-dup all) (67). BigWig files were generated using deeptools (settings: --binSize 1 --normalizeUsing RPGC). Genomic tracks were visualized on the UCSC Genome Browser. Heatmaps were generated using deeptools bamCoverage (settings: -a 2000 -b 2000 --skipZeros --missingDataAsZero --referencePoint center) (68). For ChIP-Seq reads, a similar pipeline was used, as with ATAC-Seq, except mitochondrial reads were kept. To identify peaks, tag directories were created using the HOMER makeTagDirectory function (69). PRMT5 binding regions were then identified with the findPeaks function, applying a FDR threshold of less than 0.001. Peaks were annotated using HOMER.

For RNA-Seq analysis, reads were aligned using STAR (settings: --quantMode GeneCounts --outSAMtype BAM SortedByCoordinate --outSAMunmapped Within --outSAMattributes Standard) (70). Subsequently, a count matrix file was generated from the STAR output. To determine differentially expressed genes, the DESeq2 package was used with default settings (71). Resistant cells were used as a reference sample, and genes with a FDR of less than 0.05 were assigned as differentially expressed. For visualization, BigWig files were generated using deeptools bamCoverage (settings: --binSize 1 --normalizeUsing RPKM) (68) and tracks were displayed on the UCSC Genome Browser.

Statistics. Statistical analyses were performed using GraphPad Prism 10 (GraphPad Software). Normality was assessed through the Shapiro-Wilk test. Differences between 2 groups were evaluated using 2-tailed Student’s t test or the Wilcoxon rank-sum test where appropriate. For comparisons of 3 or more groups, 1-way ANOVA with Dunnett’s multiple-comparison test was used. The tumor growth data were analyzed by 2-way ANOVA. Statistical significance was defined as a P value of less than 0.05. Data represent the mean ± SEM or SD, as indicated in the figure legends.

Study approval. All animal experiments and procedures complied with ethics regulations of the IACUC of Northwestern University under approved protocol no. IS000007992.

Data availability. All data used to generate the graphs are provided in the Supporting Data Values file. All genomic data generated in this manuscript have been deposited in the NCBI’s Gene Expression Omnibus (GEO) database (GEO GSE282674, GSE282675, GSE282676, and GSE282677).