CCNE1 amplification is synthetic lethal with PKMYT1 kinase inhibition – Nature

Cell lines and cell culture

All cell lines were grown at 37 °C and 5% CO₂. RPE1-hTERT TP53^−/− Cas9 (ref.⁹) and RPE1-hTERT TP53^−/− Cas9 PKMYT1^−/− cells were grown in DMEM (Life Technologies catalogue (cat.) no. 11965-092) with 10% FBS (Wisent cat. no. 080150) and 1% penicillin-streptomycin (Wisent cat. no. 450-201-EL). RPE1-hTERT Cas9 TP53^-/- PKMYT1^−/− cells were constructed by nucleofection of the parental cell line with PKMYT1-7 sgRNA targeting exon 4 and single cell clones were generated by limiting dilution. Two clones were confirmed to be PKMYT1^−/− using western blot (clone J3.38 and J3.43). RPE1-hTERT TP53^−/− Cas9 CCNE1-high cell lines were constructed by piggyBac transposition of CCNE1-2A-GFP into the parental cell line and selection of clones with mid-range GFP expression. FT282-hTERT TP53^R175H wild-type (empty vector) and CCNE1-high cell lines were obtained from R. Drapkin²⁴ and cultured in DMEM: F-12(1:1) (Life Technologies cat. no. 11330-032) with 5% FBS, 1% UltroserG (Pall Life Sciences cat. no.15950-017) and 1% penicillin-streptomycin. FT282-hTERT TP53^R175H CCNE2, MYBL2 and CCNB1 overexpressing cell lines were also constructed by piggyBac transposition of CCNE2-2A-GFP, MYBL2-2A-GFP or CCNB1-2A-GFP into the parental cell line and selection of clones with high GFP expression. FT282-hTERT TP53^R175H PCNA-cb-TagRFP expressing cell lines (wild-type and CCNE1-high) were transduced with PCNA-cb-TagRFP lentiviral particles and high RFP-expressing cells were selected. 293T cells (ATCC) were cultured in DMEM with 10% FBS and 1% penicillin-streptomycin. HEK293T cells (ATCC) were cultured in DMEM with 10% FBS and 1% penicillin-streptomycin. HCC1569 cells (ATCC) were cultured in RPMI 1640 (Life Technologies cat. no. 118575-093) with 10% FBS and 1% penicillin-streptomycin. SNU8 cells (KCLB) were cultured in RPMI 1640 with 10% FBS, 1% penicillin-streptomycin, 25 mM HEPES. OVCAR3 cells (ATCC) were cultured in RPMI 1640 with 20% FBS, 1% penicillin-streptomycin and 0.01 mg ml⁻¹ insulin. A2780 cells (Sigma) were cultured in RPMI 1640 with 10% FBS and 1% penicillin-streptomycin. SUM149PT cells (Asterand Bioscience) were cultured in Ham’s F12 (Life Technologies cat. no. 11765-054) with 5% FBS, 10 mM HEPES, 1% penicillin-streptomycin, 1 μg ml⁻¹ hydrocortisone and 5 μg ml⁻¹ insulin. KYSE30 cells (DSMZ) were cultured in 45% RPMI 1640 with 45% Ham’s F12, 10% heat-inactivated FBS and 1% penicillin-streptomycin. TOV112D cells (ATCC) were cultured in 42.5% MCDB 105, 42.5% Medium 199 (Life Technologies cat. no. 11150-059), 15% FBS and 1% penicillin-streptomycin. NUGC3 cells (JCRB) were cultured in RPMI 1640 with 10% FBS and 1% penicillin-streptomycin. COV362 cells (Sigma) were cultured in DMEM with 10% FBS and 1% penicillin-streptomycin. DOTC24510 cells (ATCC) were cultured in DMEM with 10% FBS and 1% penicillin-streptomycin. None of the cell lines used were authenticated after reception. All cell lines used tested negative for mycoplasm contamination using MycoAlert. The OVCAR3 and HCC1569 cells have been shown to have amplified CCNE1^49,50, whereas SNU8 has been shown to have CCNE1 copy number gain (CCLE database (https://portals.broadinstitute.org/ccle)). SUM149PT cells are reported to have high cyclin E levels due to an FBXW7 mutation⁵¹ but the clone we use does not display this cyclin E increase (Extended Data Fig. 9a).

Plasmids

For CRISPR–Cas9 genome editing, sgRNAs were cloned either in lentiCRISPRv2 or in lentiguide NLS–GFP as described⁵². For PKMYT1 overexpression in cells, an N-terminally 3×Flag-tagged PKMYT1 open reading frame (CCDS10486.1) was cloned into the pDONR221 Gateway entry vector (Thermo Fisher Scientific, 12536017). Mutagenesis was performed by PCR to generate a PKMYT1 sgRNA-resistant version carrying silent mutations between nucleotides 966 and 981 (tgagttcactgccggt to cgaatttaccgctggc) and the kinase-dead mutant N238A. PKMYT1 coding sequences were transferred by Gateway technology to the destination vector pCW57.1 (Addgene #41393) used for transduction in cells. For CDK1 mutant expression in cells the coding sequence for CDK1(T14A/Y15AF)–GFP was synthesized and cloned into the pHIV-NAT-hCD52 vector (a gift from R. Scully) using EcoRI and BamHI restriction enzymes. Mutagenesis was then performed to revert each phosphosite back to the wild type amino acids to create CDK1–GFP, CDK1(T14A)–GFP and CDK1(Y15F)–GFP. For time-lapse cell cycle microscopy, a PCNA-chromobody-TagRFP insert was amplified from pCCC-TagRFP (Chromotek) with EcoRI and BamHI restriction site sequence extensions and then cloned into pHIV-NAT-hCD52 vector. The sgRNA sequences used in this study are included in Supplementary Table 5.

Lentiviral transduction

Lentiviral particles were produced in 293T cells in 10-cm plates by co-transfection of 10 μg of targeting vector with 3 μg VSV-G, 5 μg pMDLg/RRE and 2.5 μg pRSV-REV (Addgene #14888, #12251 and #12253) using calcium phosphate. Medium was refreshed 12–16 h later. Virus-containing supernatant was collected 36–40 h after transfection and cleared through a 0.2-μm filter. Viral transductions were performed in the presence of polybrene (Sigma-Aldrich, 4 μg ml⁻¹ RPE1-hTERT TP53^−/− Cas9 and 16 μg ml⁻¹ FT282-hTERT TP53^R175H) at a multiplicity of infection (MOI) < 1.

Antibodies

Primary antibodies used in this study include: histone H2A.X (phospho-S139, Cell Signalling Technologies cat. no. 2577, 1:500 for immunofluorescence), histone H2A.X (phospho-S139, Millipore Sigma cat. no. 05-636, 1:500 for immunofluorescence), CDK1 (Thermo Fisher Scientific cat. no. 33-1800, 1:1,000 for immunoblot and ELISA), CDK1-phosphoT14 (Abcam cat. no. ab58509, 1:1,000 for immunoblot and ELISA), CDK1-phoshoY15 (Cell Signaling Technology cat. no. 9111, 1:1,000 for immunoblot), PKMYT1 (Bethyl A302-424A, 1:1,000 for immunoblot), Histone H3-phosphoS10 (Cell Signaling Technology cat. no. 9706, 1:500 flow cytometry), lamin A/C (Cell Signaling Technology 4C11 cat. no. 4777, 1:500 for immunofluorescence), lamin A/C-phosphoS22 (Cell Signaling Technology D2B2E cat. no. 13448, 1:500 flow cytometry and for immunofluorescence), cyclin B1 (Cell Signalling Technologies cat. no. 2577, 1:500 for immunofluorescence, 1:1,000 for immunoblotting), α-tubulin (Millipore DM1A CP06, 1:4,000 for iimunoblotting), CDK2 (Upstate 05-596, 1:1,000 for immunoblotting), cyclin B1-phosphoS126 (Abcam ab55184, 1:500 for immunofluorescence), MCM2 (BD Biosciences 610700, 1:250 for immunofluorescence), MCM4 (Novus Biologicals H0004137-B01P, 1:500 for immunofluorescence), CHK1-phosphoS345 (Bethyl 2348, 1:1,000 for immunoblotting), cyclin E1 (Abcam ab3927, 1:1,000 for immunoblotting or Cell Marque cat. no. AC0120RUO 1:1,000 for immunohistochemistry), α-actinin (Millipore Sigma 05-384, 1:1,000 for immunoblotting), vinculin (Cell Signaling 13901S, 1:1,000 for immunoblotting), MYBL2 (Millipore MABE886, 1:1,000 for immunoblotting), MYBL2-pT487 (Abcam ab76009, 1:500 for immunoblotting). The following agarose-coupled antibodies were used for immunoprecipitation in kinase assays: CDK1 (Santa Cruz sc-54 AC) and CDK2 (Santa Cruz sc-6248 AC). The following secondary antibodies were used for immunoblotting: anti-mouse Irdye 800CW, anti-rabbit IRdye 680RD (926-32210 and 926-68071; LiCOR, 1:5,000), anti-mouse IgG–horseradish peroxidase (HRP) (Cedarlane cat. no. NA931-1ML, 1:4,000), anti-rabbit IgG–HRP (Cedarlane cat. no. 111-035-144, 1:4,000), anti-rabbit IgG–HRP (abcam 97051, 1:10,000). The secondary antibody used for ELISA was anti-rabbit IgG–HRP (Jackson Immunoresearch cat. no. 111-035-144). The following secondary antibodies were used for immunofluorescence and flow cytometry: AlexaFluor 488 donkey anti-rat IgG (Thermo Fisher Scientific A21208, 1:1,000), AlexaFluor 647 donkey anti-mouse IgG (Thermo Fisher Scientific A31571, 1:1,000), AlexaFluor 488 goat anti-mouse IgG (Thermo Fisher Scientific A11029, 1:1,000), AlexaFluor 647 goat anti-rabbit IgG (Thermo Fisher Scientific A21244, 1:1,000). Finally, the following secondary antibodies were used for AlphaLISA assays: AlphaLISA anti-rabbit IgG Acceptor beads (Perkin Elmer cat. no. AL104C) and AlphaLISA anti-mouse IgG Donor beads (Perkin Elmer cat. no. AS104D).

Short interfering RNAs

Short interfering RNA (siRNA) oligonucleotides (siCTRL ON-TARGET Plus D-001210-03-50 and siCCNB1 ON-TARGET Plus L-003206-00-0005; Dharmacon) were transfected in Opti-MEM reduced-serum medium using Lipofectamine RNAiMAX agent (Thermo Fisher Scientific cat. no. 13778-075) following the manufacturer’s recommended protocol. Fresh medium was added to cells 16 h after transfection. Cells were used for high content imaging and immunoblotting 48 h after transfection.

Fine chemicals

The following drugs were used in the course of the study: RP-6306 (this study), RP-6421 (this study) AZD1775 (Selleckchem, S1525), dinaciclib (MedChemExpress, HY-10492), PF-06873600 (MedChemExpress, HY-114177), RO-3306 (Selleckchem, S7747), gemcitabine (Cayman Chemicals, 9003096) and hydroxyurea (Sigma-Aldrich cat. no. H8627). Synthesis of RP-6306 and RP-6421 is described in the Supplementary Information. Concentration and duration of treatment is indicated in the legends of the corresponding figures.

CRISPR screens

CCNE1-overexpression synthetic lethality screens were conducted as three parallel screens with a parental cell line and two isogenic clones overexpressing CCNE1 (C2 and C21). For the screens, RPE1-hTERT Cas9 TP53^−/− parental and RPE1-hTERT Cas9 TP53^−/− CCNE1-overexpressing clones were transduced with the lentiviral TKOv2 sgRNA library at a low MOI (about 0.3) and medium containing 20 μg ml⁻¹ puromycin (Life Technologies) was added the next day to select for transductants. The following day, cells were trypsinized and replated in the same plates while maintaining puromycin selection. Three days after infection, which was considered the initial time point (t₀), cells were pooled together and divided into two sets of technical replicates. Cells were grown for a period of 18 d and cell pellets were collected every 3 d. Each screen was performed as a technical duplicate with a theoretical library coverage of ≥400 cells per sgRNA maintained at every step. Genomic DNA was isolated using the QIAamp Blood Maxi Kit (Qiagen) and genome-integrated sgRNA sequences were amplified by PCR using NEBNext Ultra II Q5 Master Mix (New England Biolabs). i5 and i7 multiplexing barcodes were added in a second round of PCR and final gel-purified products were sequenced on an Illumina NextSeq500 system at the LTRI NBCC facility (https://nbcc.lunenfeld.ca/) to determine sgRNA representation in each sample. Later, another screen was conducting using the next-generation TKOv3 library in RPE1-hTERT Cas9 TP53^−/− parental and RPE1-CCNE1 (C2) cells using the same procedure outlined above.

The RP-6306 resistance screen was performed in two FT282-hTERT TP53^R175H CCNE1-high clones (C3 and C4) using TKOv3 sgRNA library at a MOI about 0.3. The screen was conducted in technical duplicates, and library coverage of >100 cells per sgRNA was maintained at every step. Puromycin-containing medium (2 µg ml⁻¹) was added 2 days after infection to select for transductants. Selection was continued until 96 h after infection, which was considered the initial time point (t₀). RP-6306 was added to the cells starting from time at day 6 (t₆) at doses corresponding to individual LD80 (40 nM and 80 nM for clones C3 and C4, respectively). From t₁₀ onwards, RP-6306 dose was adjusted to 60 nM for both clones and drug-containing medium was subsequently refreshed at t₁₂, t₁₆ and t₁₈. The screen was terminated at t₂₁. To identify genes whose deletion caused resistance to RP-6306, genomic DNA was isolated from surviving cells and processed as described above. Sample data analysis was performed using DrugZ algorithm previously described https://github.com/hart-lab/drugz.

DepMap data mining

CRISPR dependency data^14,53 (CERES scores) and gene-level copy number data⁵⁴ were downloaded from the 2021 Q1 DepMap release using the Broad Institute’s DepMap portal. Cell lines were characterized as being ‘CCNE1-amplified’ if they had a copy number value that was greater than 1.58 (approximately equal to 2× total copy number relative to ploidy), or ‘WT’ if they had a copy number value that was less than or equal to 1.58; cell lines with no copy number data for CCNE1 were removed from the analysis. From a total of 808 cell lines in the dependency dataset, 6 were removed, 20 were classified as CCNE1-amplified, and 782 were classified as WT. The Wilcoxon rank-sum test was used to compare dependency scores for each gene between the 2 groups. In Fig. 1b the difference in median gene depletion was plotted on the x-axis versus the nominal P value of the difference on the y-axis. Nominal P values are provided. Results of the analysis can be found in a tabular format in the source data.

Clonogenic survival assays

Cells were seeded in 6-well plates, 300 cells per well for RPE1 and 400 for FT282. Single cells were grown out until distinct colonies formed with greater than 50 cells per colony. Colonies were rinsed with PBS and stained with 0.4% (w/v) crystal violet in 20% (v/v) methanol for 30 min. The stain was aspirated, and plates were rinsed twice in double-distilled H₂O and air-dried. Colonies were counted using a GelCount instrument (Oxford Optronix, GelCount).

Cell proliferation assays

RPE1-hTERT Cas9 TP53^−/−, FT282-hTERT TP53^R175H and their respective CCNE1-high isogenic pairs were seeded in 96-well plates (Corning Costar cat. no. 5595) at a density of 150 cells per well for RPE1-hTERT Cas9 TP53^−/− CCNE1 (C2) or 100 cells per well for all others. After 24 h, cells were treated using an automated D300e digital dispenser (Tecan) at drug concentrations ranging from 0.15 nM to 3 µM. Medium and drugs were refreshed every 3–4 days and cellular confluency was monitored up to 6 population doublings using an IncuCyte S3 Live-Cell Imager (Sartorius). Per cent confluence relative to a non-treated control was used to evaluate growth inhibition induced by test compounds. Synergy between RP-6306 and hydroxyurea or gemcitabine was analysed using the online SynergyFinder v2.0 tool⁵⁵ using the ZIP model⁵⁶ (https://synergyfinder.fimm.fi).

Immunofluorescence

Cells were seeded onto glass coverslips and treated as indicated in the figure legends. Before collection, cells were pulsed with 20 μM EdU (5-ethynyl-2-deoxyuridine, Life Technologies cat. no. A10044) for 30 min and then washed with PBS and fixed with 4% paraformaldehyde (PFA) in PBS for 15 min at room temperature. Cells were then rinsed with PBS and permeabilized using 0.3% Triton X-100/PBS for 30 min. For chromatin-bound MCM measurements, cells were pre-extracted for 15 min on ice with CSK buffer (300 mM sucrose, 100 mM NaCl, 3 mM MgCl₂, 10 mM PIPES pH 7.0, 0.5% v/v Triton-X 100) before PFA fixation. Cells were washed with PBS and incubated in blocking buffer (10% goat serum (Sigma cat. no. G6767), 0.5% NP-40 (Sigma-Aldrich, cat. no. I3021), 5% w/v saponin (Sigma-Aldrich, cat. no. 84510), diluted in PBS) for 30 min. Fresh blocking buffer containing primary antibodies was added for 2 h. Cells were rinsed three times with PBS and then blocking buffer, with secondary antibodies and 0.4 μg ml⁻¹ DAPI (4,6-diamidino-2-phenylindole, Sigma-Aldrich, cat. no. D9542) was added for 1 h. After rinsing with PBS, immunocomplexes were fixed again using 4% PFA/PBS for 5 min. Cells were rinsed with PBS and incubated with EdU staining buffer (150 mM Tris-Cl pH 8.8, 1 mM CuSO₄, 100 mM ascorbic acid and 10 μM AlexaFluor 555 azide (Life Technologies, cat. no. A20012) for 30 min. After rinsing with PBS coverslips were mounted onto glass slides with ProLong Gold mounting reagent (Invitrogen, cat. no. P36930). Images were acquired using a Zeiss LSM780 laser-scanning microscope (Oberkochen) with ZEN 2.3 SP1 software. Image analysis was performed using ImageJ v2.0.0.

High content imaging and QIBC

For high-throughput analysis of nuclear γH2AX, 3,000 cells per well were seeded in 96-well plates and cultured for up to 72 h depending on the experiment. Cells were fixed, permeabilized and stained in the same manner as immunofluorescence described above. Wells were filled with 200 μl PBS and images were acquired at the Network Biology Collaborative Centre (LTRI) on an InCell Analyzer 6000 automated microscope (GE Life Sciences) with a 20× objective. Image analysis was performed using Cellprofiler 3.1.9 and RStudio v1.2.5019⁵⁷ (Supplementary Fig. 4).

Time-lapse microscopy

PCNA-cb-TagRFP expressing cells were maintained at 37 °C and 5% CO2 while deconvolution wide-field microscopy was performed using the DeltaVision Elite system equipped with an NA 0.75 20× UPlanSApo objective (Olympus) and an sCMOS 2,048 × 2,048 camera (Leica Microsystems). Each field was acquired every 10 min over 23 h with a z-step of 2 μm through the entire cell (7 sections) and deconvolved using softWoRx (v6.0, Leica Microsystems). Maximum intensity projections are shown (0.330 μm per pixel).

Immunoblotting

Cell pellets were extracted by incubation in NP-40 lysis buffer (50 mM Tris-Cl pH 7.4, 250 mM NaCl, 5 mM EDTA, 1% NP-40, 0.02% NaN₃, 1× protease inhibitor cocktail (Roche cat. no. 11836170001) for 30 min on ice. Extracts were cleared by centrifugation at 13,000g for 10 min at 4 °C. Cleared extracts were diluted in 2× sample buffer (20% glycerol, 2% SDS, 0.01% bromophenol blue, 167 mM Tris-Cl pH 6.8, 20 mM DTT) and boiled prior to separation by SDS–PAGE on Novex Tris–glycine gradient gels (Invitrogen, cat. no. XV0412PK20). Alternatively, cell pellets were boiled directly in 2× sample buffer before separation by SDS–PAGE. Proteins were transferred to nitrocellulose membranes (VWR, cat. no. CA10061-152), then blocked in 5% milk TBST and probed overnight with primary antibodies. Membranes were washed three times for five minutes with TBST, then probed with appropriate secondary antibodies for one hour, and washed again with TBST, three times for five minutes. Secondary antibody detection was achieved using an Odyssey Scanner (LiCOR) and analysed using Image Studio Lite v5.2.5 or enhanced chemiluminescence (ECL SuperSignal West Pico, Thermo Fisher Scientific cat. no. 34579).

Flow cytometry

Cells were pulsed with 20 μM EdU (Life Technologies cat. no. A10044) for 30 min, collected by trypsinization, resuspended as single cells, washed once in PBS and pelleted at 600g for 3 min at 4 °C. All subsequent centrifugations were performed in this manner. Cells were fixed in 4% PFA/PBS for 15 min at room temperature, excess ice cold PBSB (1% BSA in PBS, 0.2 μM filtered) was added before pelleting. Cells were resuspended in permeabilization buffer (PBSB, 0.5% Triton-X 100) and incubated at room temperature for 15 min. Excess blocking buffer (PBSB, 0.1% NP-40) was added, cells were pelleted, resuspended in blocking buffer containing primary antibodies and incubated at room temperature for 1 h. Excess blocking buffer containing secondary antibodies was added, cells were pelleted, resuspended in blocking buffer and incubated at room temperature for 30 min. Excess blocking buffer was added, cells were pelleted and washed one additional time in PBSB. Cells were resuspended in EdU staining buffer (150 mM Tris-Cl pH 8.8, 1 mM CuSO₄, 100 mM ascorbic acid and 10 μM AlexaFluor 555 azide (Life Technologies, cat. no. A20012)) and incubated at room temperature for 30 min. Excess PBSB was added, cells were pelleted and washed one additional time in PBSB. Cells were resuspended in analysis buffer (PBSB, 0.5 µg ml⁻¹ DAPI, 250 µg µl⁻¹ RNase A (Sigma-Aldrich, cat. no. R4875)) and incubated at 37 °C for 30 min or left at 4 °C overnight. Cells were analysed at the LTRI flow cytometry facility on a Fortessa X-20 (Becton Dickinson) using FACSDIVA v8.0.1 with at least 9,000 events collected and analysed using FlowJo v10.

Immune complex histone H1 kinase assays

Cell pellets were resuspended in 250 μl EBN buffer (150 mM NaCl, 0.5% NP-40, 80 mM β-glycerol phosphate (Sigma-Aldrich, cat. no. 50020), 15 mM MgCl₂, 20 mM EGTA, 1 mg ml⁻¹ ovalbumin (Sigma-Aldrich, cat. no. 5503), 1× protease inhibitor cocktail (Roche, cat. no. 11836170001) pH 7.3) and incubated on ice for 5 min. Cell lysis was induced by two freeze–thaw cycles of incubation in liquid nitrogen and a 37 °C water bath, and lysates were cleared by centrifugation at 13,000g at 4 °C for 10 min. Protein concentration was determined by Bradford assay (Thermo Fisher Scientific cat. no. 1856209). For immunoprecipitation of kinases, 200 μg of extract was diluted in 750 μl EBN buffer and 10 μg of CDK1 or CDK2 primary antibody agarose bead conjugates were added to the extract and rotated at 4 °C overnight. Immunoprecipitates were pelleted by centrifugation at 2,500g at 4 °C for 5 min and washed 2× in 750 μl EBN followed by 1 ml EB (80 mM β-glycerol phosphate, 15 mM MgCl₂, 20 mM EGTA, 1 mg ml⁻¹ ovalbumin). After the final wash, the immunoprecipitates were resuspended in 500 μl EB and split into two samples. One sample was used for immunoblot analysis and the other used for kinase assays. Following removal of the final wash, immunoprecipitates were resuspended in 11 μl histone H1 kinase assay buffer (80 mM β-glycerol phosphate, 15 mM MgCl₂, 20 mM EGTA, 1 mg ml⁻¹ ovalbumin, 10 mM DTT, 0.15 μg μl⁻¹ histone H1 (Sigma-Aldrich, cat. no. H1917), 22 μM ATP, 0.05 μCi μl⁻¹ γ³²P-ATP (Perkin Elmer NEG502A250UC), pH 7.3) and incubated at room temperature for 30 min. Reactions were quenched by addition of 5 μl 6× sample buffer (60% glycerol, 6% SDS, 0.03% bromophenol blue, 1,500 mM Tris-Cl pH 6.8, 60 mM DTT) and resolved by SDS–PAGE. Gels were exposed to a phosphor imaging screen for 1–2 d and imaged using a Typhoon FLA 9500 (GE Healthcare Life Sciences). 32P-H1 band intensity was quantified using ImageJ v2.0.0.

Cytogenetic analyses

A total of 1.5 × 10⁶ FT282-hTERT TP53^R175H or HCC1569 cells was seeded in 10-cm dishes. Twenty-four hours later, RP-6306 was added at a final concentration of 500 nM for 24 h. KaryoMAX colcemid (100 ng ml⁻¹) (Thermo Fisher Scientific cat. no. 15212-012) was added to the medium in the last 2 h of treatment and cells were collected as follows. Growth medium was stored in a conical tube. Cells were treated twice for 5 min with 1 ml trypsin. The growth medium and the 2 ml of trypsinization incubations were centrifuged (250g, 5 min, 4 °C). Cells were then washed with PBS and resuspended in 75 mM KCl for 15 min at 37 °C. Cells were centrifuged again, the supernatant was removed, and cells were fixed by dropwise addition of 1 ml fixative (ice-cold methanol:acetic acid, 3:1) with gentle vortexing. An additional 9 ml fixative was then added, and cells were fixed at 4 °C for at least 16 h. Once fixed, metaphases were dropped on glass slides and air-dried overnight. To visualize mitotic cells, slides were mounted in DAPI-containing ProLong Gold mounting medium (Invitrogen, cat. no. P36930). Images were captured on a Zeiss LSM780 laser-scanning confocal microscope with ZEN 2.3 SP1 software.

MMB–FOXM1 transcriptional signature

Promoters of 114 protein-coding genes bound by both MYBL2 and FOXM1⁵⁸ were used to create a MMB–FOXM1 transcriptional signature. To eliminate genes whose expression correlated poorly with other gene set members in TCGA samples the log₂ fragments per kilobase of exon per million mapped fragments (FPKM) gene expression values were used to calculate pairwise Spearman correlations across the 11 genes in the signature. Genes with a mean correlation value below 0.4 were eliminated resulting in the 60 gene refined MMB–FOXM1 signature. The refined signature score for each TCGA sample was calculated by taking the median log₂ FPKM value of all genes in the signature.

RNA-seq sample preparation, sequencing and analysis

Cells were seeded in 10-cm dishes (2.5 × 10⁶ FT282-hTERT TP53^R175H wild type or 2 × 10⁶ CCNE1-high clone cells (C3 and C4)). The next day, cells were collected by trypsinization, washed once in PBS, and then pelleted. Pellets were snap-frozen in liquid nitrogen. RNA extraction and sequencing of the full transcriptome was performed using NovaSeq at BGI Hong Kong. Raw FASTQ files from a paired-end library were assessed using the FastQC v0.11.9 software (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) to determine the quality of the reads; read length was 150 bp. The FASTQ files were then aligned to the GENCODE GRCh38 v36 primary assembly of the human genome and quantified using Salmon v1.4.0⁵⁹ with the command line flags “–validateMappings” and “–gcBias” to obtain read counts. Raw counts were processed using the bioconductor package edgeR v3.30.3 in R⁶⁰. Genes expressed with counts per million (CPM) > 0.1 in at least two samples were considered and normalized using trimmed mean of M-values (TMM) to remove the library-specific artefacts. For subsequent analyses, voomY transformation was applied to RNA-seq count data to obtain normalized expression values on the log₂ scale. Raw counts of sequencing reads with quality scores in FASTQ format and normalized transcript abundance measurements have been deposited in NCBI’s Gene Expression Omnibus⁶¹ and are accessible through GEO Series accession number GSE171453.

Heat maps were generated using the package heatmap3 v1.1.9 in R. Unsupervised hierarchical clustering was performed by calculating distances using the Pearson correlation metric and clustering using the complete method. Gene expression values were averaged and scaled across the row to indicate the number of standard deviations above (red) or below (blue) the mean, denoted as row z-score. GSEA⁶² was performed to identify the enrichment of genes co-regulated by MMB–FOXM1 in the FT282-hTERT TP53^R175H CCNE1 C3 and C4 clones compared to parental wild-type cells. Analysis was performed using 1,000 permutations of the gene set, and normalized enrichment scores (NES) were obtained to reflect the degree to which the gene set is overrepresented in the FT282-hTERT TP53^R175H CCNE1-high C3 and C4 clones.

ADP-Glo assay

For the ADP-Glo assay (Promega cat. no. V9103) human recombinant PKMYT1 (full-length human GST–PKMYT1 recombinant protein; Thermo Fisher Scientific A33387, lot 1938686), was diluted in enzyme assay buffer (70 mM HEPES, 3 mM MgCl₂, 3 mM MnCl₂, 50 μg ml⁻¹ PEG20000, 3 μM sodium orthovanadate, 1.2 mM DTT) in a 5 μl volume and plated in 384-well plates (to a final concentration of 18.5 nM) followed by addition of 5 μl enzyme assay buffer. The enzyme–compound mix was incubated at room temperature for 15 min before addition of 5 μl of 30 μM ATP (diluted in enzyme assay buffer) so that the final ATP concentration was 10 μM. After incubation at 30 °C for 1 h, 15 μl of ADP-Glo reagent was added and incubated at room temperature for 40 min. Finally, 30 μl of the kinase detection reagent was added, the plate was incubated at room temperature for 30 min and luminescence was measured using an EnVision plate reader (Perkin-Elmer). Luminescence is measured for 0.25 s per well and rate per second was obtained by multiplying the luminescence value by 4.

NanoBRET assay

To determine the affinity of RP-6306 in the PKMYT1 or WEE1 NanoBRET target engagement assay (Promega), HEK293T cells were transfected with a NanoLuc fusion vector for PKMYT1 (Promega NV1871) or WEE1 (Promega NV2231) with transfection carrier DNA (Promega E4881) using Fugene HD Transfection reagent (Promega E2311) in Opti-MEM without phenol red (Thermo Fisher Scientific, 11058021) and incubated overnight. Cells were trypsinized, counted and 17,000 cells per well were plated in 96-well plates with K-5 cell-permeable kinase NanoBRET TE tracer (Promega N2482) and RP-6306 and incubated for 2 h at 37 °C. Intracellular TE Nano-Glo Substrate/Inhibitor (Promega N2160) was added, and the intensity of the acceptor emission (610 nm) and the donor emission (450 nm) were measured using an EnVison plate reader (Perkin-Elmer).

AlphaLISA assay

HCC1569 cells were plated into a 96-well TC-treated culture plate (30,000 cells per well) and grown overnight. The next day, RP-6306 was dispensed using a Tecan D300e digital dispenser with threefold dilutions. After compound addition, cell plates were centrifuged at 300g for 10 s, and then placed in the incubator for 2 h. Cells were lysed in AlphaLISA lysis buffer supplemented with 1× protease (Roche cat. no. 11836170001), and phosphatase inhibitors (Sigma-Aldrich cat. no. 4906837001) and 1 mM PMSF. Plates were rotated at 500g for 20 min to facilitate lysis. Plates were then sealed with aluminium foil and frozen at −80 °C for at least 1 h. Lysates were thawed at 37 °C for 10 min and 10 μl of each lysate was transferred in duplicate to 384 well assay plates. Antibodies were added at a final concentration of 5 nM or 1 nM for CDK1-pT14 and total CDK1 or CDK1-pY15 and total CDK1, respectively, sealed and stored at 4 °C overnight. Anti-rabbit IgG Acceptor and anti-mouse IgG donor beads were each added at a final concentration of 20 μg ml⁻¹ and the reactions were incubated in the dark for 2 h at room temperature. Luminescence was measured using a Perkin Elmer EnVision Multimode plate reader with excitation at 680 nm and emission at 615 nm.

Animal studies

Mice were housed and experiments were performed at Repare Therapeutics (NEOMED site, Montreal, Canada), which is a Canadian Council on Animal Care (CCAC)-accredited vivarium. Studies were conducted under a protocol approved by the NEOMED Institutional Animal Care Committee (NIACC). Mice were inspected upon arrival and group-housed (3–5 per cage) in individual HEPA ventilated autoclaved cages (Innocage IVC, Innovive) in a temperature-controlled environment (22 ± 1.5 °C, 30–80 % relative humidity, 12 h:12 h light:dark). Mice were provided with autoclaved corncob bedding, irradiated food (Harlan Teklad) and filtered water ad libitum. They were also provided with nesting material (Ketchum cat. no. 087) and a plastic shelter as enrichment. Fresh bedding, nesting material and water was replenished and replaced on a weekly basis. Mice were acclimatized in the animal facility for at least 5 days prior to use and were identified with indelible ink. Experiments were performed during the light phase of the cycle.

Cell line-derived and patient-derived xenografts

HCC1569, OVCAR3 and SUM149PT cells were implanted at 5 × 10⁶ cells per mouse into the right flanks of female CB17-SCID, SCID-beige and NOD-SCID mice respectively (5–7 weeks old; Charles River), in 1:1 Matrigel:medium (Matrigel Corning, cat. no. CB35248). When tumours reached the target size of 100–150 mm³, mice (n = 8) were randomized to treatment groups according to tumour volume and body weight using the ‘stratified’ method in Studylogv4.4 software and treatment with RP-6306 was initiated.

In vivo studies using PDX were conducted at Crown Biosciences. Fresh primary human tumour tissue was collected and cut into small pieces (approximately 2–3 mm in diameter). These tumour fragments were inoculated subcutaneously into the right flank of female BALB/c nude mice (5–7 weeks old) for tumour development and subsequently passaged by implantation into the cohort of mice enrolled in the efficacy study. Mice were randomized according to growth rate into treatment groups (n = 6) when the mean tumour size reached approximately 150 (100–200) mm³ using the stratified method in Studylogv4.4 software. The procedures involving the care and use of animals in for the PDX model were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of CrownBio prior to execution. During the study, the care and use of animals were conducted in accordance with the regulations of the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC).

RP-6306 was formulated in 0.5% methylcellulose and orally administered twice daily (BID, 0–8 h) for a maximum of 21 days. Gemcitabine was administered once weekly intraperitoneally in saline. Animals were monitored for tumour volume, clinical signs and body weight three times per week. Tumour volume was measured using a digital calliper and calculated using the formula 0.52 × L × W², where L is length and W is width. Response to treatment was evaluated for tumour growth inhibition (% TGI). Tumour growth inhibition (TGI) was defined as: TGI = ((TV_vehicle/last − TV_vehicle/day0) − (TV_treated/last − TV_treated/day0))/(TV_vehicle/last − TV_vehicle/day0) × 100% calculated based on the means of the treatment groups at day 0 and last day of measurement. TV is tumour volume and subscripts indicate treatment group and time of sampling. According to NIACC and IACUC approved animal protocols, mice were euthanized as soon as their tumour volume exceeded 2,000 mm³. Change in body weight (BW) was calculated using the formula: %BW change = (BW_last − BW_day0/BW_day0) × 100. BW change was calculated based on individual body weight changes relative to day 0. Statistical significance relative to vehicle control or other test groups was established by one-way ANOVA followed by Fisher’s least significant difference test for multiple groups and unpaired t-test for two group comparisons (GraphPad Prism v9.0). Investigators were not blinded during data collection and analysis.

Blood and tumour tissue collection

Under isoflurane anaesthesia, whole blood was collected by cardiac puncture and transferred to tubes containing 0.1 M citric acid (3:1 citric acid:blood) and stored at −20 °C for LC–MS/MS analysis. Tumours were removed from mice flanks and cleared of surrounding mouse stroma. Tumour pieces between 50 mg and 100 mg were collected in a pre-weighed pre-filled bead mill tube (Fisher Scientific, cat. no. 15-340-154) and then flash-frozen in liquid nitrogen. Other tumour fragments from vehicle- and compound-treated mice were placed in 10% neutral buffered formalin (NBF) within 2–5 min of surgical excision, fixed in NBF for 18–24 h at room temperature and embedded in paraffin.

RP-6306 quantification by LC–MS/MS

The extraction of whole blood samples was performed by protein precipitation using four volumes of acetonitrile. The sample extracts were analysed using a Transcend LX2 Ultimate 3000 liquid chromatography system coupled to a Thermo Altis triple quadrupole electrospray mass spectrometer (Thermo Fisher Scientific) operated in positive mode. Separations were performed using a 2 × 50 mm, 2.8 µm Pursuit XRS C8 HPLC column (Agilent). A reversed-phase linear gradient of water + 0.1% formic acid and 1:1 acetonitrile:MeOH was used to elute RP-6306 and the internal standard. Samples were quantified against a ten-point linear standard curve and three levels of quality control samples. Whole blood concentrations of RP-6306 were converted to free unbound plasma concentrations using an in vitro derived blood/plasma ratio = 1.2 and fraction unbound (f_u) plasma = 0.185 from the CD-1 mouse strain.

Immunohistochemistry

Histology in Extended Data Fig. 9 was performed by HistoWiz. In brief, the formalin-fixed tissues were dehydrated through a 20%, 80%, 95% and 100 % ethanol series, cleaned in xylene, embedded in paraffin then sectioned at 4 μm. Immunohistochemistry for γH2AX, CDK1-pT14 and cyclin B1-pS126 were performed on a Bond Rx autostainer (Leica Biosystems) with heat antigen retrieval. Bond polymer refine detection (Leica Biosystems) was used according to manufacturer’s protocol. After staining, sections were dehydrated and film coverslipped using a TissueTek-Prisma and Coverslipper (Sakura). Whole-slide scanning (40×) was performed on an Aperio AT2 (Leica Biosystems). Image quantification analysis was performed using HALO. H-score is given by the formula: H-score = (1 × percentage of weakly positive cells) + (2 × percentage of moderately positive cells) + (3 × percentage of strongly positive cells). Histology in Extended Data Fig. 10c was performed by NeoGenomics. In brief, formalin-fixed, paraffin-embedded tumour samples were sectioned at 4 μm, mounted on charged glass slides and baked at 60 °C for 1 h. Immunohistochemistry for cyclin E1 was performed on a Bond-III autostainer (Leica Biosystems). Bond polymer refine detection (Leica Biosystems) was used according to the manufacturer’s protocol. Slides were then removed from the instrument dehydrated, cleared and coverslipped. Bright-field images (20×) were acquired on an Aperio AT2 (Leica Biosystems).

ELISA

Tumour samples were homogenized in MSD Tris lysis buffer (Meso Scale Discovery cat. no. R60TX-2) supplemented with 1× Halt Protease (Thermo Fisher Scientific cat. no. 78429) and phosphatase inhibitors (Thermo Fisher Scientific cat. no. 78426) using a Beadruptor tissue homogenizer (OMNI International. After homogenization, samples were centrifuged at 14,000g for 5 min at 4 °C. ELISA plates were coated with the capture antibody (CDK1) incubated overnight at 4 °C, washed and then blocked for 1 h at room temperature. Tissue samples (60 μg) were added to the plates to incubate for 2.5 h at room temperature. After washing, the detector antibody (CDK1-pT14) was added for 1 h at room temperature. After washing and plate drying, detection occurred using a secondary anti-rabbit HRP conjugate incubation for 1 h followed by a 10-min incubation with TMB peroxidase substrate stop solution (Thermo Fisher Scientific cat. no. N600). The absorbance was measured in 96-well plate format on an EnVision2105 at 450 nm. Samples were quantified relative to a standard protein extract and an MSD lysis buffer used as a blank to control for inter-day variability.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this paper.