biostudies-arrayexpressEleanor WigmoreHomo sapienshttps://www.ebi.ac.uk/biostudies/studies/E-MTAB-16438A genome-wide CRISPR-Cas9 knockout screen was performed in the breast cancer cell lines T47D, MCF7, and CAMA-1 to identify genes modulating sensitivity and resistance to capivasertib, a selective AKT inhibitor. Cells were transduced with a CRISPR library, followed by treatment with capivasertib or vehicle control (DMSO). Guide RNA (sgRNA) enrichment and depletion were assessed via next-generation sequencing to determine gene-level effects on cell viability and drug response. Gene-level data are provided for the initial plasmid library, baseline, DMSO-treated controls, and post-capivasertib treatment across replicates.biostudies-arrayexpressLibrary Construction - NGS libraries were prepared in a two-step PCR process starting with amplification of the integrated lentiviral gRNA cassette from 5 µg gDNA per reaction using specific primers (NGS_v3_for and NGS_v3_rev) and Q5 Hot Start High-Fidelity 2× Master Mix, achieving at least 200-fold coverage. PCR products were pooled and purified using the QIAquick PCR Purification Kit (Qiagen). Final libraries were generated from 5 ng of purified first PCR product using a dual-indexing Illumina-compatible kit, followed by purification with AMPure XP beads at a 0.7 ratio and quantification with the Qubit dsDNA Assay Kit.Sequencing - Purified second-step libraries were sequenced on the HiSeq4000 platform using paired-end 50 bp reads with a 30% PhiX spike-in to ensure sequencing quality. Sequencing reads were then counted against the Yusa_V3_human library using an in-house Python-based string-matching algorithm to quantify sgRNA abundances. This approach provided high-resolution data for downstream analysis of gene perturbations.Sample Collection - For each cell line (MCF7, T47D and CAMA-1 ), 2 x 108 cells were each transduced with Yusa_V3_human gRNA lentiviral library (37) (pKLV2-U6gRNA5(BbsI)-PGKpuro2ABFP-W V3 gRNA) which targets 18,365 of genes using 113,526 sgRNAs. The library was transduced with 8µg/ml polybrene to achieve a transduction MOI of 0.3, at which most cells receive only one genetic perturbation and therefore the gRNA library has a coverage of >500 expressing each gRNA. Three days post transduction, the percentage of BFP+ cells was checked by flow cytometry and puromycin added to media to kill non-library transduced cells. Cells were maintained in culture for at least 14 days to allow for complete depletion of protein products of targeted genes. Cells were expanded and baseline pellets were collected at the point of first compound treatment. Cells were split into the dimethyl sulfoxide (DMSO) and the treatment arm (capivasertib) with two technical replicates in each and treated for 23-29 days. The exact length of drug treatment corresponded to at least 6 cell doublings of DMSO-treated cells. Fresh media and compound were replaced every 4-5 days. Cells were maintained at a library coverage of at least 750 cells per gRNA throughout the expansion and drug treatment phases. Concentrations of compounds used in screens were 0.75µM capivasertib for T47D and MCF7 and 0.4µM for CAMA-1. At the end of drug selection, at least 100 million cells from each of the different treatment arms were pelleted and used for genomic DNA extraction for PCR and Illumina sequencing. Pellets were stored at -20°C before processing. Genomic DNA isolation and Next Generation Sequencing (NGS) Genomic DNA (gDNA) was isolated from 100 million cells from each treatment arm/replicate using the QIAamp DNA Blood Maxi Kit (Qiagen) according to manufacturer’s instruction. NGS libraries were prepared in a two-step process. First, the integrated lentiviral cassette containing the gRNA was amplified from gDNA using primer NGS_v3_for5’-ACACTCTTTCCCTACACGACGCTCTTCCGATCTCTTGTGGAAAGGACGAAACA-3’ and reverse primer NGS_v3_rev 5’- GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTACCCAGACTGCTCATCGTC-3’. PCR reactions with 5µg gDNA per well were set up using the Q5 Hot Start High-Fidelity 2× Master Mix (NEB #M0494) in a total volume of 50µl, scaled to amplify the gRNAs at a coverage of at least 200-fold. The PCR products were then pooled in each group and purified using QIAquick PCR Purification Kit (Qiagen). Final NGS libraries were generated using 5ng of the purified 1st PCR product using the dual-indexing Illumina-compatible DNA HT Dual Index kit (Takara #R4000660, R400661). 2nd PCR products were purified with AMPure XP beads (Beckman Coulter) at an 0.7 ratio. Purified 2nd step libraries were quantified using the Qubit dsDNA Quantification Assay Kit (ThermoFisher) and sequenced on either HiSeq4000 by PE50bp p with a 30% PhiX spike-in. Analysis of CRISPR screening data CRISPR screening data was analysed using the MAGeCK RRA (version 0.5.8) (16) R package (R version 3.6.3) and BAGEL (38) as described in (39). Illumina sequencing reads were counted against the Yusa_V3_human library using an in-house string-matching algorithm operating in Python (version 3.8.15) to quantify sgRNA counts. Sensitisation and resistance genes were identified by comparing capivasertib treated vs DMSO treated counts at equivalent timepoints (2 replicates each) for each of the cell lines CAMA-1, T47D and MCF7. Genes with a MAGeCK-RRA FDR value of <0.1 were considered as hits. For quality control purposes, BAGEL was used to compare the DMSO treated counts against the guide counts in the original sequencing of the plasmid library. Genes were considered essential if their Bayes Factor exceeded 3.0. The ROC curves in Supplementary Figure 2 assess the calling of essential genes in this study using BAGEL against 2 published lists of common-essential genes: the Hart lab’s 218 “Constitutive Core Essentials” (40) and 552 genes identified as Pan-essential in publicly available Sanger CRISPR screens (41). The high AUC values against these lists of >0.9 indicate that the screens were successful in identifying known essential genes, giving confidence in the capivasertib-specific sensitisation hits called in this study. True Positive and False Positive rates and AUC were computed using the ROCR package version 1.0.11 (42) in R version 4.3.2. Cell culture and reagents Cell lines were authenticated using short tandem repeat (STR) profiling. Cells were cultured in a humidified incubator with 5% CO2 at 37°C. All MCF7 cells (PIK3CA_E545K), T47D cells (PIK3CA_H1047R) and CAMA-1 (PTEN null) were cultivated with RPMI supplemented with 1% Glutamax and 10% FBS. All the capivasertib resistant cells (capiR) were cultivated with RPMI supplemented with 1% Glutamax, 10% FBS and 10μM capivasertib. All the fulvestrant resistant cells (FulvR) and the MCF7 ESR1mut were cultivated with RPMI supplemented with 1% Glutamax, 10% FBS and 100 nM fulvestrant. Compounds Capivasertib and fulvestrant were synthesised by AZ. Capivasertib was synthesized according to (43). Pan-KDM5i, KAT6A/Bi were purchased. Compounds were dissolved in DMSO at a concentration of 10 mmol/L and administered at the following concentrations: Capivasertib: 1μM in MCF7, 750nM in T47D and 400nM in CAMA-1. Pan KDM5i: 50μM in all cell lines used in this study. KAT6Ai: 3μM in MCF7, 1μM in T47D and 1μM in CAMA-1. Fulvestrant 10nM in all cell lines used in this study. Cas9 expressing cells and KO generations gRNAs are summarized in Table S1 Generation and validation of pooled Cas9-expressing BC cell lines. Cell lines were transduced with a lentivirus produced from the Cas9 expression vector pKLV2-EF1a-Cas9Bsd-W vector. 3 days after transduction, cells were selected with blasticidin (10 µg/ml for MCF7 & T47D; 25 µg/ml for CAMA-1). Pooled selected cells were further expanded and analysed for Cas9 cutting activity using a Cas9 reporter assay as previously described in (37). Briefly, cells were transduced separately with lentivirus produced with control vector, pKLV2-U6gRNA5(Empty)-PGKBFPGFP-W and Cas9 activity vector, pKLV2-U6gRNA5(GFP gRNA)-PGKBFPGFP-W. The Cas9 activity vector contains both a BFP and GFP expression cassette as well as a gRNA targeting GFP – efficient Cas9 activity would therefore be expected to result in silencing of GFP signal. 4 days post-transduction, the ratio of BFP and GFP-BFP double-positive cells analysed using flow cytometry using LSRFortessa instrument (BD) and resulting data analysed using FlowJo. Cas9 activity in cells (%) was calculated as (BFP-single positive cells) / (total number of BFP+ cells). All Cas9-cell lines used in this study had genome-editing Cas9-activity >90%. Generation of KO cells Cells expressing Cas9 were used to generate KDM5C and KAT6A KO cells. gRNA-expressing and Cas9-expressing lentivirus were produced as previously described (8). KDM5C and KAT6A gRNAs were designed using Yusa Human CRISPR library V1 (Addgene #67989) and cloned into the lentiviral plasmid pKLV-U6gRNA (BbsI)-PGKpuro2ABFP (Addgene #50946). Three days after transduction, puromycin (2μg/ml) (Sigma–Aldrich) was added to the media for 7 days to kill non-transduced cells. Control cells (CTRL) were infected with the pLentiV2 lacking the guide.Nucleic Acid Extraction - Genomic DNA (gDNA) was isolated from pellets of 100 million cells per treatment arm and replicate using the QIAamp DNA Blood Maxi Kit (Qiagen) according to the manufacturer's instructions. This process ensured high-quality extraction suitable for downstream PCR amplification of integrated lentiviral cassettes containing gRNAs. The extraction was scaled to maintain gRNA coverage during subsequent steps.OrganizationMINSEQE ScoreAssays and DataProcessed DataMAGE-TAB FilesSequence Alignment - Illumina sequencing reads were aligned and counted against the Yusa_V3_human library using an in-house string-matching algorithm implemented in Python (version 3.8.15) to quantify sgRNA abundances. This high-throughput approach efficiently matched reads to the 113,526 sgRNAs targeting 18,365 genes, ensuring accurate representation of library coverage. The resulting sgRNA count data was used for comparing treatment conditions in downstream analyses like MAGeCK RRA.Data Transformation - CRISPR screening data was analyzed using MAGeCK RRA (version 0.5.8), which inherently normalizes sgRNA read counts across samples by comparing treated (capivasertib) versus control (DMSO) conditions at equivalent timepoints, identifying hits with FDR <0.1. For quality control, BAGEL compared DMSO-treated counts against plasmid library counts, defining essential genes with Bayes Factor >3.0. ROC curves were computed using the ROCR package to validate essential gene calling, with high AUC values (>0.9) confirming robust data transformation and normalization.UnknownTranscriptomicsGenomicsProteomicsIllumina HiSeq 4000genotyping by high throughput sequencingHomo sapiensTargeting the epigenetic regulators KDM5C enhances AKT inhibitor response in ER+ breast cancer.Eleanor WigmoreSimon BarryValentina Cutano, Shanade Dunn, Sungmi Park-Chouinard, Eleanor M. Wigmore, Qing Wu, Lambert Montava Garriga, Lorna Hopcroft, David Esopi, Michele Chirichella, Siyu Liu, Cath Eberlein, Ming Tang, Hyung Joo Lee, Wenhao Zhang, Megan Callahan, Barrett Nuttall, Daniel Barrell, Toby Gurran, Jennifer Hillis, Adam Spruce, Ramy Elgendy, Jenna Bradley, Chang Kim, Sara Talbot, Laura Rosenberg, Susana Ros, Huayang Liu, Neeraj Aryal, Functional Genomics Centre, Jerome T. Mettetal, Ho Man Chan, Ultan McDermott and Simon T. BarryfalseCRISPR screen exploring resistance to AKTi (Capivasertib) in ER+ Breast Cancer in MCF7, T47D and CAMA-1 cell linesA genome-wide CRISPR-Cas9 knockout screen was performed in the breast cancer cell lines T47D, MCF7, and CAMA-1 to identify genes modulating sensitivity and resistance to capivasertib, a selective AKT inhibitor. Cells were transduced with a CRISPR library, followed by treatment with capivasertib or vehicle control (DMSO). Guide RNA (sgRNA) enrichment and depletion were assessed via next-generation sequencing to determine gene-level effects on cell viability and drug response. Gene-level data are provided for the initial plasmid library, baseline, DMSO-treated controls, and post-capivasertib treatment across replicates.2026-05-25T00:00:00Z2026-05-26T17:05:07.759Z2025-12-19T17:14:07.934ZE-MTAB-16438ERP186854E-MTAB-16414EFO_0002944EFO_0004170EFO_0002771EFO_0004917EFO_0005518EFO_0003816EFO_0004184