Genome-wide CRISPR-Cas9 Knockout Research Articles

Targeting NRAS via miR-1304-5p or farnesyltransferase inhibition confers sensitivity to ALK inhibitors in ALK-mutant neuroblastoma

Targeting Anaplastic lymphoma kinase (ALK) is a promising therapeutic strategy for aberrant ALK-expressing malignancies including neuroblastoma, but resistance to ALK tyrosine kinase inhibitors (ALK TKI) is a distinct possibility necessitating drug combination therapeutic approaches. Using high-throughput, genome-wide CRISPR-Cas9 knockout screens, we identify miR-1304-5p loss as a desensitizer to ALK TKIs in aberrant ALK-expressing neuroblastoma; inhibition of miR-1304-5p decreases, while mimics of this miRNA increase the sensitivity of neuroblastoma cells to ALK TKIs. We show that miR-1304-5p targets NRAS, decreasing cell viability via induction of apoptosis. It follows that the farnesyltransferase inhibitor (FTI) lonafarnib in addition to ALK TKIs act synergistically in neuroblastoma, inducing apoptosis in vitro. In particular, on combined treatment of neuroblastoma patient derived xenografts with an FTI and an ALK TKI complete regression of tumour growth is observed although tumours rapidly regrow on cessation of therapy. Overall, our data suggests that combined use of ALK TKIs and FTIs, constitutes a therapeutic approach to treat high risk neuroblastoma although prolonged therapy is likely required to prevent relapse.

Abstract LB191: A genome-wide CRISPR-based functional screen uncovers the KLF4-Survivin/BIRC5 signaling pathways as a key regulator of cell fate following endoplasmic reticulum stress: Implications for tumor growth and therapeutic targeting

Abstract The growth and spread of tumors involve intricate interactions with the surrounding tissue microenvironment. As a tumor develops, it faces various stresses from its microenvironment, including hypoxia, nutrient deficiency, and acidosis. In response to these challenges, cancer cells can hijack existing cytoprotective mechanisms, such as the Unfolded Protein Response (UPR). The UPR activates signaling pathways at the translational and later transcriptional levels to alleviate cellular stress and prevent cell death. However, under unresolved or chronic endoplasmic reticulum (ER) stress, the UPR may paradoxically promote cell death. The precise mechanisms through which the UPR influences cellular fate during ER stress remain poorly understood. To address this knowledge gap, we employed a functional CRISPR-based genetic knockout screen to identify novel regulators of the UPR. Using a lentiviral genome-wide CRISPR-Cas9 knockout library, we identified over-represented sgRNAs (associated with pro-apoptotic genes) and underrepresented sgRNAs (associated with pro-survival genes) following ER stress induced by thapsigargin and tunicamycin. One of the top candidates identified was Survivin/BIRC5, a protein with anti-apoptotic and cell cycle regulatory activities and which is overexpressed in tumor cells compared to healthy tissues. Functional studies revealed that genetic or pharmacological inhibition of Survivin expression enhanced ER stress-induced apoptosis, but not apoptosis in response to other forms of stress. These findings were validated in multiple cancer cell lines, ruling out off-target effects. Mechanistically, ER stress led to PERK-dependent downregulation of Survivin, and overexpression of Survivin blocked Thapsigargin-induced apoptosis. In vivo xenograft models showed that combining Survivin ablation with PERK inhibition significantly delayed tumor growth and improved overall survival. Through a combination of bioinformatics analysis and loss-of-function studies, we identified the transcription factor KLF4 as a mediator of PERK-dependent effects on Survivin repression. Notably, CRISPR-mediated KLF4 knockout successfully rescued Survivin levels post-ER stress. Additionally, an analysis of TCGA data across various tumor types provided supporting evidence of a negative correlation between KLF4 and Survivin expression. In summary, this study unveils a novel regulatory axis, PERK-KLF4-Survivin, as a key determinant of cell fate in the cellular responses to ER stress. Targeting Survivin in conjunction with PERK inhibition showed promising antitumor effects in vivo. Citation Format: Nektaria M. Leli, Hengxi Liu, Souvik Dey, Lauren Brady, Carlo Salas Salinas, Victoria Wu, Spiros Tastsoglou, Giorgos Skoufos, Ioannis Verginadis, Ilias Karagounis, Artemis G. Hatzigeorgiou, Constantinos Koumenis. A genome-wide CRISPR-based functional screen uncovers the KLF4-Survivin/BIRC5 signaling pathways as a key regulator of cell fate following endoplasmic reticulum stress: Implications for tumor growth and therapeutic targeting [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB191.

Abstract 5841: A genome wide CRISPR-Cas9 screen identifies mediators of resistance to dual PI3K/mTOR inhibition in glioblastoma multiforme

Abstract Glioblastoma multiforme (GBM) is a highly aggressive primary brain tumor with a devastating 5-year survival rate of 6.8%. The current standard of care, consisting of surgery, radiation, and temozolomide (TMZ), has remained unchanged for over the past 15 years despite its limited efficacy and the serious therapy-related adverse events associated to this regimen. Although some frequent genetic alterations have been identified in GBM, no targeted therapies have shown efficacy in these patients to date. Among these, the phosphoinositide 3-kinase (PI3K) pathway is a signaling network that regulates cell growth, proliferation, migration, and invasion. Extensive molecular characterization of GBM has identified alterations in the components of the PI3K pathway in 80% of primary GBMs, including amplification of receptor tyrosine kinases (RTKs), activating mutations in the PI3K catalytic subunit (PIK3CA), loss-of-function mutations in the PI3K regulatory subunit (PIK3R1), and loss of the major tumor suppressor of the pathway, phosphatase and tensin homologue (PTEN). Despite the universal upregulation of PI3K pathway signaling associated with these genetic alterations, clinical trials investigating PI3K inhibitors in GBM have failed due to insufficient antitumor activity and dose-limiting toxicities. To elucidate mechanisms of resistance to PI3K inhibition in GBM, we used a genome-wide functional CRISPR-Cas9 knockout screen to identify genes that mediate resistance to the dual PI3K/mTOR inhibitor XL765 in PTEN-null SF295 GBM cells. We identified reactivation of the PI3K pathway and inhibition of autophagy as potential resistance mechanisms to dual PI3K/mTOR inhibition. Using orthogonal approaches, we will functionally validate our candidate hits through mechanistic studies, with patient-derived models, and with in vivo orthotopic xenografts. Our findings will elucidate novel mechanisms of resistance to PI3K/mTOR inhibition in GBM and will streamline rational drug combinations aimed to overcome resistance, thus eventually maximizing therapeutic responses to PI3K inhibitors in GBM. Citation Format: Annalisa V. Ferrotta, Arianna Izawa-Ishiguro, Meaghan Grogan, Ingo K. Mellinghoff. A genome wide CRISPR-Cas9 screen identifies mediators of resistance to dual PI3K/mTOR inhibition in glioblastoma multiforme [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5841.

Abstract 1223: KDM6A orchestrates NK cell response via CD38/48 regulation in multiple myeloma

Abstract Daratumumab (Dara), a first-in-class humanized monoclonal antibody targeting CD38, demonstrates notable efficacy in both newly diagnosed and relapsed/refractory multiple myeloma (MM) when administrated alone or in combination with other agents. However, the development of Dara resistance and subsequent relapse is common.To identify the tumor-intrinsic genes involved in resistance to NK-mediated cytotoxicity in the presence of Dara in MM cells, we conducted an in vitro genome-wide CRISPR-Cas9 knockout (KO) screen and found 433 genes exhibiting a positive correlation with MM cytotoxicity elicited by Dara and primary NK cells. Remarkably, 16 of these genes demonstrated a consistent correlation with cytolytic activity across the majority of the 36 cancer types in the TCGA database.The efficacy of Dara is closely linked to the expression of CD38, with nonresponders exhibiting low CD38 expression. Therefore, we conducted another CRISPR screen to identify the genes whose KO resulted in diminished CD38 expression. The genes that overlapped in these two screens are implicated in regulating both Dara-mediated ADCC and CD38 expression. Our focus turned to KDM6A among the overlapping genes. We validated that KDM6A KO significantly decreased CD38 expression and that reintroducing KDM6A into KDM6A KO cells restored the expression of CD38 and Dara-induced cytotoxicity.KDM6A demethylates lysine 27 of histone H3 (H3K27) to promote gene expression. Our ChIP-seq data unveiled that KDM6A-KO cells exhibited an increased H3K27me3 level at the CD38 promoter area than control cells. Remarkably, this pattern was consistent with Dara-resistant cell lines, which displayed reduced CD38 expression alongside elevated levels of H3K27me3 at the CD38 promoter. Interestingly, we found that CD38 overexpression only partially restored the sensitivity of KDM6A KO cells to Dara treatment, implying the existence of an additional mechanism. We performed RNA-seq in KDM6A KO cells and found CD48, an NK-activating ligand, was significantly repressed. Overexpression of CD48 in KDM6A KO cells resulted in a substantial increase in the secretion of Granzyme B and Perforin by NK cells and restored the activity of NK cells and re-sensitized KDM6A KO cells to Dara treatment. These data suggest that KDM6A mediates Dara-mediated ADCC sensitivity through the dual regulation of CD38 and CD48. KDM6A acts to oppose the EZH2/PRC2 complex by demethylating H3K27me3, and we found that Tazemetostat (Taze), an FDA-approved EZH2 inhibitor, increased the expression of CD38 and CD48, especially in KDM6A-KO cells. Importantly, Taze also restored the sensitivity of KDM6A KO cells to Dara-mediated NK cell cytotoxicity. In summary, our findings illuminate the dual role of KDM6A in enhancing the efficacy of Dara in MM. Imitating KDM6A function using an FDA-approved inhibitor holds promise in overcoming Dara resistance and improving patient outcomes in MM. Citation Format: Jiye Liu, Lijie Xing, Jiang Li, Kenneth Wen, Ning Liu, Yuntong Liu, Annamaria Gulla, Nikhil Munshi, Teru Hideshima, Kenneth Carl Anderson. KDM6A orchestrates NK cell response via CD38/48 regulation in multiple myeloma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1223.

Abstract 5858: Systematic identification of pathways associated with antibody drug conjugate sensitivity in breast cancer

Abstract Antibody drug conjugates (ADCs) are revolutionizing precision therapeutics in oncology. Sacituzumab Govitecan (SG), combining a Topoisomerase I (TOP1) inhibitor payload (SN38) and the tumor-selective antigen TROP2-targeting antibody hRS7, has demonstrated efficacy in challenging cancers including triple negative breast cancer (TNBC). Despite its promising clinical efficacy, the intricacies of sensitivity and resistance mechanisms to SG remain elusive. To systematically interrogate ADC sensitivity mechanisms, we employed genome-wide CRISPR-Cas9 knockout screening to discover novel regulators influencing SG sensitivity in human TNBC cells. Interactome analysis of top SG sensitizing hits showed clustered pathways of interest, including DNA repair/replication, mTORC1/metabolism, ubiquitin/proteolysis and endosome trafficking/sorting. A top druggable hit sensitizing to SG was PARP1, encoding the poly (ADP-ribose) polymerase 1 that is a key enzyme required for single-strand break repair pathways and for DNA replication fork stability. Furthermore, the synergistic lethality between SG and PARP1 was confirmed in experiments combining SG with PARP inhibitor treatment of TNBC cell lines. To selectively identify SG/ADC-specific genes and pathways, we performed a secondary round of CRISPR knockout and activation screens using a custom library comprising top hits discovered in the primary screen, treating cells in parallel with either SG or cytotoxic payload SN38 alone. Pathway analysis of genes selectively modulating sensitivity to SG but not SN38 revealed novel mediators involved in TROP2 turnover and recycling. We then performed validation and mechanistic studies to elucidate how specific genetic permutations regulate antibody delivery and sensitivity to SG. In summary, employing a systematic CRISPR screening method, we uncovered numerous novel pathways associated with both resistance and sensitization, playing a role in ADC delivery and target protein turnover. Our current focus involves prioritizing druggable genes and pathways, laying the foundation for proof-of-concept studies that will combine ADCs with specific targeted therapeutics, aiming to enhance tumor-killing efficacy. Citation Format: Bogang Wu, Sheng Sun, Nayana Thimmiah, Aiko Nagayama, James T. Coates, Ting Liu, Win Thant, Elena Bitman, David Li, John Doench, Aditya Bardia, Leif W. Ellisen. Systematic identification of pathways associated with antibody drug conjugate sensitivity in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5858.

Abstract 5839: Identification of low-dose ligand dependent drug-resistance in ALK/ROS1 positive NSCLC via CRISPR library screening

Abstract Tyrosine kinase inhibitor (TKI) therapy has displayed significant effectiveness in treating patients with anaplastic lymphoma kinase (ALK) and ROS proto-oncogene 1 (ROS1)-positive non-small cell lung cancer (NSCLC). Nonetheless, the emergence of acquired resistance in most patients presents a major challenge. Drug-tolerant persister (DTP) cells have recently gained attention as potential sources of resistance, highlighting the need for a deeper understanding of drug resistance mechanisms. To address this issue, a comprehensive genome-wide CRISPR-Cas9 knockout screening was conducted using an ALK-positive NSCLC cell line derived from pleural effusion in a patient who had not been exposed to ALK-TKI treatment. Following a 9-day ALK-TKI treatment, sequencing analysis was performed to identify sgRNAs in DTP cells. Notably, this analysis detected the involvement of tumor suppressor genes, including NF2, BAX, and ERRFI1.NF2 depletion was found to induce a substantial resistance to ALK-TKIs, and a combination therapy involving mTOR inhibitors showed promise in partially overcoming this resistance. Moreover, knockout of ERRFI1 or BAX led to an increase in DTP cells, drawing particular attention to ERRFI1, also known as MIG6, which is known for negatively regulating EGFR signaling. One intriguing discovery was that the loss of MIG6 induced resistance to ALK-TKIs when exposed to a relatively low dose of EGF, comparable to concentrations in plasma of healthy adults. This resistance was attributed to the upregulation of the MAPK and PI3K/Akt/mTOR pathways. Similar to ALK-rearranged NSCLC, ROS1 fusion-positive NSCLC cell lines also showed resistance by MIG6 knockout. Interestingly, some cell lines with reduced MIG6 expression tended to develop resistance with low-dose EGF, even without prior treatment, and overexpression of MIG6 restored sensitivity to TKIs. We further explored the potential of combination therapy, combining ALK-TKIs with anti-EGFR antibodies, and this approach effectively addressed acquired resistance in both in vivo and in vitro models. Furthermore, an analysis of clinical samples revealed that individuals resistant to ALK-TKIs displayed decreased mRNA MIG6 levels in comparison to their sensitive counterparts. Investigating the mechanisms behind MIG6 expression, we confirmed that ALK-TKI therapy suppressed MIG6 expression levels, while factors such as corticosteroid usage or exposure to hypoxia induced the upregulation of MIG6, potentially influencing the formation of DTP cells. In conclusion, this research has unveiled a novel factor contributing to resistance in ALK and ROS1-TKI therapies by activating the EGFR pathway through exposure to low-dose ligands, shedding light on potential therapeutic strategies to combat acquired resistance. Citation Format: Nobuyuki Kondo, Takahiro Utsumi, Ken Uchibori, Yasunari Miyazaki, Ryohei Katayama. Identification of low-dose ligand dependent drug-resistance in ALK/ROS1 positive NSCLC via CRISPR library screening [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5839.

Abstract 4921: A user-friendly R Shiny web app for predicting cancer genetic dependencies using deep learning

Abstract Understanding the genetic vulnerabilities (i.e., genetic dependencies) of cancer cells is crucial for developing novel anti-cancer treatments. The Cancer Dependency Map (DepMap) projects, conducted by the Broad Institute and Wellcome Sanger Institute, have performed comprehensive genome-wide CRISPR-Cas9 knockout screens across diverse cancer cell lines. These data resources have enabled computational biologists to glean insights into cancer genetic dependencies using sophisticated computational models. As different genomic mechanisms govern tumorigenesis and genetic vulnerability, the relationship between cancer genomics and genetic dependencies is nonlinear, making the prediction of genetic dependencies challenging. Addressing this challenge, we previously developed a deep learning model, namely DeepDEP, to predict cancer genetic dependencies using multi-omic profiles (Chiu et al. DOI:10.1126/sciadv.abh1275). DeepDEP has a transfer learning scheme that integrates genomic representations captured from unlabeled tumor samples of the Cancer Genome Atlas (TCGA) and genomic features that are predictive of genetic dependencies in cell lines. Such a design enables the application of DeepDEP to both cell lines and tumors. The model demonstrated superior performance compared to various conventional machine learning methods. Prediction results were validated using independent cell-line datasets and patient clinical data. Here, we developed an intuitive web application for DeepDEP which allows users to easily utilize the model on their in-house data. Users can simply indicate whether the query sample is derived from a cancer cell line or a tumor, upload any combination of mutation, gene expression, DNA methylation, and copy number alteration datasets, and view the predicted dependencies of 1,298 cancer-relevant genes. Along with the genetic dependency predictions, users are provided with model performance metrics, detailed gene annotations, and links to useful resources such as Ensembl and COSMIC. Our web application features a variety of interactive visualizations to guide users to interpret the results at the levels of individual genes and pathways. The tool has two supplemental modules to enhance its functionality. One allows users to search for similar cell lines to the query sample within the DepMap project or for similar tumors within the TCGA project. The other supplemental module provides data exploration tools for pre-calculated predictions across all TCGA tumors. In summary, this user-friendly web application provides researchers with access to deep learning predictions of cancer genetic dependencies with various interactive visualization and analysis tools. It is anticipated that this application will facilitate the study of cancer genetic dependencies and enhance the efficiency of anti-cancer drug discovery. Citation Format: Michael J. Kasper, Li-Ju Wang, Michael Ning, Yufei Huang, Yu-Chiao Chiu. A user-friendly R Shiny web app for predicting cancer genetic dependencies using deep learning [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4921.

PTEN Mediates the Silencing of Unintegrated HIV-1 DNA.

The integration of viral DNA into a host genome is an important step in HIV-1 replication. However, due to the high failure rate of integration, the majority of viral DNA exists in an unintegrated state during HIV-1 infection. In contrast to the robust expression from integrated viral DNA, unintegrated HIV-1 DNA is very poorly transcribed in infected cells, but the molecular machinery responsible for the silencing of unintegrated HIV-1 DNA remains poorly characterized. In this study, we sought to characterize new host factors for the inhibition of expression from unintegrated HIV-1 DNA. A genome-wide CRISPR-Cas9 knockout screening revealed the essential role of phosphatase and tensin homolog (PTEN) in the silencing of unintegrated HIV-1 DNA. PTEN's phosphatase activity negatively regulates the PI3K-Akt pathway to inhibit the transcription from unintegrated HIV-1 DNA. The knockout (KO) of PTEN or inhibition of PTEN's phosphatase activity by point mutagenesis activates Akt by phosphorylation and enhances the transcription from unintegrated HIV-1 DNA. Inhibition of the PI3K-Akt pathway by Akt inhibitor in PTEN-KO cells restores the silencing of unintegrated HIV-1 DNA. Transcriptional factors (NF-κB, Sp1, and AP-1) are important for the activation of unintegrated HIV-1 DNA in PTEN-KO cells. Finally, the knockout of PTEN increases the levels of active epigenetic marks (H3ac and H3K4me3) and the recruitment of PolII on unintegrated HIV-1 DNA chromatin. Our experiments reveal that PTEN targets transcription factors (NF-κB, Sp1, and AP-1) by negatively regulating the PI3K-Akt pathway to promote the silencing of unintegrated HIV-1 DNA.

Radiogenomic landscape: Assessment of specific phagocytosis regulators in lower-grade gliomas.

Genome-wide CRISPR-Cas9 knockout screens have emerged as a powerful method for identifying key genes driving tumor growth. The aim of this study was to explore the phagocytosis regulators (PRs) specifically associated with lower-grade glioma (LGG) using the CRISPR-Cas9 screening database. Identifying these core PRs could lead to novel therapeutic targets and pave the way for a non-invasive radiogenomics approach to assess LGG patients' prognosis and treatment response. We selected 24 PRs that were overexpressed and lethal in LGG for analysis. The identified PR subtypes (PRsClusters, geneClusters, and PRs-score models) effectively predicted clinical outcomes in LGG patients. Immune response markers, such as CTLA4, were found to be significantly associated with PR-score. Nine radiogenomics models using various machine learning classifiers were constructed to uncover survival risk. The area under the curve (AUC) values for these models in the test and training datasets were 0.686 and 0.868, respectively. The CRISPR-Cas9 screen identified novel prognostic radiogenomics biomarkers that correlated well with the expression status of specific PR-related genes in LGG patients. These biomarkers successfully stratified patient survival outcomes and treatment response using The Cancer Genome Atlas (TCGA) database. This study has important implications for the development of precise clinical treatment strategies and holds promise for more accurate therapeutic approaches for LGG patients in the future.

Abstract A053: The dual BRG1/BRM (SMARCA4/2) inhibitor FHD-286 induces functional differentiation and splicing defects in preclinical models of acute myeloid leukemia (AML)

Abstract The BRG/BRM associated factor (BAF) complex is a chromatin remodeler critical for maintenance of cell viability in many cancers. This includes acute myeloid leukemia (AML), where BAF maintains the stem-like transcriptional state of blast cells. FHD-286 is a dual inhibitor of the BAF ATPase subunits BRG1 and BRM and is currently being investigated in relapsed or refractory AML and myelodysplastic syndrome. We have shown previously that FHD-286 induces the expression of the myeloid differentiation marker CD11b in preclinical models after treatment with sub-cytoreductive doses at long timepoints. Here we further characterized the differentiation phenotype induced by low-dose FHD-286 in AML cell lines through RNA-seq and a genome-wide CRISPR-Cas9 knockout screen. These datasets and mechanistic follow-up studies suggested AML cell lines treated with FHD-286 can functionally differentiate to gain the ability to produce superoxide anion and perform phagocytosis. We also found that FHD-286 treatment disrupts mRNA splicing and that this is a likely contributor to AML cell growth defects induced by BRG1/BRM inhibition. This study suggests multiple mechanisms by which FHD-286 is able to disrupt the stem-like transcriptional state of AML blasts to cause differentiation and ultimately cell death. Citation Format: Ashley K. Gartin, David N. Mayhew, David L. Lahr, GiNell Elliott, Heena Gandevia, Molly M. Wilson, Richard Centore, Mike Collins, Astrid Thomsen, Jessica Piel, Gabriel J. Sandoval, Martin Hentemann. The dual BRG1/BRM (SMARCA4/2) inhibitor FHD-286 induces functional differentiation and splicing defects in preclinical models of acute myeloid leukemia (AML) [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr A053.

Abstract A176: Loss of SMAD4 unleashes mTOR and increases dependency on cap-dependent translation in esophageal tumorigenesis

Abstract Application of molecular targeted therapies for esophageal adenocarcinoma (EAC) has been limited by a lack of druggable oncogenic drivers. We propose that synthetic lethal interactions may provide new opportunities for targeted therapies in EAC. We have taken an integrated multi-omics approach incorporating Perturb-Seq (CRISPR editing combined with single cell RNA sequencing) and in vivo tumorigenesis assays to perform high-throughput characterisation of >70 high-confidence EAC driver genes, and genome-wide CRISPR-Cas9 knockout screens in isogenic models of EAC tumorigenesis to identify disease relevant synthetic lethal genetic interactions. MS-based proteomics, reverse phase protein arrays, polysome profiling and bulk RNA-sequencing of isogenic models were utilised to interrogate the biology of bona fide EAC drivers and associated driver specific gene dependencies. The overall goals were to (i) enhance our understanding of EAC tumorigenesis, (ii) identify potential opportunities for therapeutic interventions targeting EAC drivers via synthetic lethal-like approaches, and (iii) reduce the complexity of genetic heterogeneity by categorising EAC drivers with similar phenotypic outcomes. Through our approach we have identified complex crosstalk between the tumor suppressor SMAD4 and regulation of mTOR signaling, with specific downstream effects on translational reprogramming in EAC. Mutation or loss of SMAD4 occurs in up to 20% of EAC, but not pre-malignant tissue (Barrett’s esophagus), and is sufficient to promote transformation of pre-malignant cells in our in vivo tumorigenesis model. In this model, xenotransplanted SMAD4-deficient (via CRISPR-Cas9 knockout or shRNA knockdown) Barrett’s metaplasia cells formed invasive, metastatic tumors after a period of latency. SMAD4 deficient cells had downregulated expression of 4E-BP1, which inhibits EIF4E, the cap-dependent translation initiation factor. This was accompanied by increased mTOR activity, including phosphorylation and inactivation of 4EBP1. Moreover, we found that SMAD4-deficient cells preferentially upregulate cap-dependent translation and become addicted to translation of oncogenic mRNAs. Furthermore, perturbation of additional negative regulators of mTOR signaling in combination with SMAD4 knockout exacerbated these effects and accelerated tumorigenesis in vivo. We have extended these findings to a model of Barrett’s esophagus patient-derived organoids (PDOs) and observed increased proliferative potential of our genetically modified PDOs. Finally, analysing gene ontologies for differentially expressed genes from Perturb-seq revealed that driver-dependent transcriptional changes can be categorized into a smaller number of functional pathways allowing us to potentially consider groups of drivers as functional units. This work advances our understanding of EAC tumorigenesis, provides new mechanistic insights into SMAD4-driven transformation as well as novel potential therapeutic avenues for SMAD4-deficient EAC. Citation Format: Julia V Milne, Ebtihal Mustafa, Kenji Fujihara, Eric Kusnadi, Niko Thio, Maree Pechlivanis, Carlos Cabalag, Twishi Gulati, Kaylene Simpson, Cuong Duong, Luc Furic, Wayne Phillips, Nicholas J Clemons. Loss of SMAD4 unleashes mTOR and increases dependency on cap-dependent translation in esophageal tumorigenesis [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr A176.

Inhibition of lysosome-tethered Ragulator-Rag-3D complex restricts the replication of Enterovirus 71 and Coxsackie A16.

Enterovirus 71 (EV71) and Coxsackie A16 (CVA16) are two major causative agents of hand, foot, and mouth disease (HFMD) in young children. However, the mechanisms regulating the replication and pathogenesis of EV71/CVA16 remain incompletely understood. We performed a genome-wide CRISPR-Cas9 knockout screen and identified Ragulator as a mediator of EV71-induced apoptosis and pyroptosis. The Ragulator-Rag complex is required for EV71 and CVA16 replication. Upon infection, the Ragulator-Rag complex recruits viral 3D protein to the lysosomal surface through the interaction between 3D and RagB. Disruption of the lysosome-tethered Ragulator-Rag-3D complex significantly impairs the replication of EV71/CVA16. We discovered a novel EV71 inhibitor, ZHSI-1, which interacts with 3D and significantly reduces the lysosomal tethering of 3D. ZHSI-1 treatment significantly represses replication of EV71/CVA16 as well as virus-induced pyroptosis associated with viral pathogenesis. Importantly, ZHSI-1 treatment effectively protects against EV71 infection in neonatal and young mice. Thus, our study indicates that targeting lysosome-tethered Ragulator-Rag-3D may be an effective therapeutic strategy for HFMD.

The ligation between ERMAP, galectin-9 and dectin-2 promotes Kupffer cell phagocytosis and antitumor immunity.

Kupffer cells, the liver tissue resident macrophages, are critical in the detection and clearance of cancer cells. However, the molecular mechanisms underlying their detection and phagocytosis of cancer cells are still unclear. Using in vivo genome-wide CRISPR-Cas9 knockout screening, we found that the cell-surface transmembrane protein ERMAP expressed on various cancer cells signaled to activate phagocytosis in Kupffer cells and to control of liver metastasis. ERMAP interacted with β-galactoside binding lectin galectin-9 expressed on the surface of Kupffer cells in a manner dependent on glycosylation. Galectin-9 formed a bridging complex with ERMAP and the transmembrane receptor dectin-2, expressed on Kupffer cells, to induce the detection and phagocytosis of cancer cells by Kupffer cells. Patients with low expression of ERMAP on tumors had more liver metastases. Thus, our study identified the ERMAP-galectin-9-dectin-2 axis as an 'eat me' signal for Kupffer cells.

DMT1 differentially regulates mitochondrial complex activities to reduce glutathione loss and mitigate ferroptosis.

Mitochondria are vital for energy production and redox homeostasis, yet knowledge of relevant mechanisms remains limited. Here, through a genome-wide CRISPR-Cas9 knockout screening, we have identified DMT1 as a major regulator of mitochondria membrane potential. Our findings demonstrate that DMT1 deficiency increases the activity of mitochondrial complex I and reduces that of complex III. Enhanced complex I activity leads to increased NAD+ production, which activates IDH2 by promoting its deacetylation via SIRT3. This results in higher levels of NADPH and GSH, which improve antioxidant capacity during Erastin-induced ferroptosis. Meanwhile, loss of complex III activity impairs mitochondrial biogenesis and promotes mitophagy, contributing to suppression of ferroptosis. Thus, DMT1 differentially regulates activities of mitochondrial complex I and III to cooperatly suppress Erastin-induced ferroptosis. Furthermore, NMN, an alternative method of increasing mitochondrial NAD+, exhibits similar protective effects against ferroptosis by boosting GSH in a manner similar to DMT1 deficiency, shedding a light on potential therapeutic strategy for ferroptosis-related pathologies.

Genome-wide CRISPR screens identify noncanonical translation factor eIF2A as an enhancer of SARS-CoV-2 programmed −1 ribosomal frameshifting

Many positive-strand RNA viruses, including all known coronaviruses, employ programmed -1 ribosomal frameshifting (-1 PRF) to regulate the translation of polycistronic viral RNAs. However, only a few host factors have been shown to regulate -1 PRF. Through a genome-wide CRISPR-Cas9 knockout screen, we have identified host factors that either suppress or enhance severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-1 PRF. Among them, eukaryotic translation initiation factor 2A (eIF2A) specifically and directly enhances -1 PRF independent of changes in initiation. Consistent with the crucial role of efficient -1 PRF in transcriptase/replicase expression, loss of eIF2A reduces SARS-CoV-2 replication in cells. Furthermore, transcriptome-wide analysis shows that eIF2A preferentially binds CG-rich RNA motifs, including a region within 18S ribosomal RNA near the contacts between the SARS-CoV-2 frameshift-stimulatory element (FSE) and the ribosome. Thus, our results indicate a role for eIF2A in modulating the translation of specific RNAs independent of its role during initiation.

Genome-wide CRISPR-Cas9 screening to identify CDS2 as an essential gene in uveal melanoma.

e21585 Background: Despite low rates of local recurrence in primary uveal melanomas following radical treatment, approximately half of all patients will develop metastatic disease. There are limited effective systemic therapeutic options and survival outcomes remain poor. Genome-wide CRISPR-Cas9 knockout screening enables appropriate molecular targets and pathways that underlie cancer development to be identified. Methods: Genome-wide single gRNA CRISPR knockout screens were performed using Human Improved Genome-wide Knockout CRISPR Library v1.1. Eight human uveal melanoma cell lines were screened for 14 days to identify gene dependencies. Prior to screening, all cell lines underwent RNA-Seq to quantify gene expression. CRISPR screen hits were validated using single gRNA competitive growth assays. In vitro functional assessment of CDS2 knockout was performed using inducible CDS2 knockout isogenic cell lines in colony forming assays and Annexin V apoptosis assays. To determine the effects of CDS2 loss on phosphoinositide synthesis, quantification of phosphatidic acid and phosphoinositides was performed using HPLC-MS. To test the association between CDS2 essentiality and CDS1 expression across all cancer types, we analysed gene essentiality data from Project Score and Project Achilles with gene expression data obtained from Cell Model Passports and the DepMap download portal (22Q4). Cell lines with CDS1 expression log2(TPM+1) < 1 were annotated as low-expressors, and the rest were annotated as high-expressors. Single group data were assessed using Student's t test. Results: We identified CDS2, which encodes enzymes that maintain phosphoinositide levels during PLC signalling, to be an essential gene in 7 out of 8 screened cell lines. All cell lines were found to have low expression of its paralogous gene, CDS1. Doxycycline-inducible CDS2 knockout clones were generated from cell lines MP41 and OMM2.5. Reduced colony formation, with a significant decrease in surviving fraction was observed after doxycycline compared with treatment with a DMSO control in both cell lines (P < 0.0001). Apoptotic and dead cells increased significantly when doxycycline was added in MP41(13.6 ± 0.9% to 47.4 ± 3.3%; P= 0.0006) and OMM2.5 (3.7 ± 0.4% to 60.4± 0.7%, P< 0.0001). CDS2 knockout resulted in an increase in the precursor phosphatidic acid, and a significant reduction of the phosphoinositides PI (MP41, P= 0.018; OMM2.5, P= 0.0005) and PIP (MP41, P= 0.033; OMM2.5, P= 0.006). CDS2 was more essential in CDS1 low-expressors in both Project Score ( P< 0.0001) and Project Achilles ( P< 0.0001) datasets, compared with CDS1 high-expressors. Conclusions: Disruption of CDS2 results in reduced cellular fitness in uveal melanoma. Given the synthetic lethal relationship between CDS1 and CDS2, in the context of low CDS1 gene expression as observed in uveal melanomas and in other tumour types, CDS2 is a potential novel therapeutic target.

HNRNPH1 regulates the neuroprotective cold-shock protein RBM3 expression through poison exon exclusion.

Enhanced expression of the cold-shock protein RNA binding motif 3 (RBM3) is highly neuroprotective both in vitro and in vivo. Whilst upstream signalling pathways leading to RBM3 expression have been described, the precise molecular mechanism of RBM3 cold induction remains elusive. To identify temperature-dependent modulators of RBM3, we performed a genome-wide CRISPR-Cas9 knockout screen using RBM3-reporter human iPSC-derived neurons. We found that RBM3 mRNA and protein levels are robustly regulated by several splicing factors, with heterogeneous nuclear ribonucleoprotein H1 (HNRNPH1) being the strongest positive regulator. Splicing analysis revealed that moderate hypothermia significantly represses the inclusion of a poison exon, which, when retained, targets the mRNA for nonsense-mediated decay. Importantly, we show that HNRNPH1 mediates this cold-dependent exon skipping via its thermosensitive interaction with a G-rich motif within the poison exon. Our study provides novel mechanistic insights into the regulation of RBM3 and provides further targets for neuroprotective therapeutic strategies.

SUMO1 degrader induces ER stress and ROS accumulation through deSUMOylation of TCF4 and inhibition of its transcription of StarD7 in colon cancer.

Small molecule degraders of small ubiquitin-related modifier 1 (SUMO1) induce SUMO1 degradation in colon cancer cells and inhibits the cancer cell growth; however, it is unclear how SUMO1 degradation leads to the anticancer activity of the degraders. Genome-wide CRISPR-Cas9 knockout screen has identified StAR-related lipid transfer domain containing 7 (StarD7) as a critical gene for the degrader's anticancer activity. Here, we show that both StarD7 mRNA and protein are overexpressed in human colon cancer and its knockout significantly reduces colon cancer cell growth and xenograft progression. The treatment with the SUMO1 degrader lead compound HB007 reduces StarD7 mRNA and protein levels and increases endoplasmic reticulum (ER) stress and reactive oxygen species (ROS) production in colon cancer cells and three-dimensional (3D) organoids. The study further provides a novel mechanism of the compound anticancer activity that SUMO1 degrader-induced decrease of StarD7 occur through degradation of SUMO1, deSUMOylation and degradation of T cell-specific transcription 4 (TCF4) and thereby inhibition of its transcription of StarD7 in colon cancer cells, 3D organoids and patient-derived xenografts (PDX).

CRISPR-Cas9 screen reveals a role of purine synthesis for estrogen receptor α activity and tamoxifen resistance of breast cancer cells

In breast cancer, resistance to endocrine therapies that target estrogen receptor α (ERα), such as tamoxifen and fulvestrant, remains a major clinical problem. Whether and how ERα+ breast cancers switch from being estrogen-dependent to estrogen-independent remains unclear. With a genome-wide CRISPR-Cas9 knockout screen, we identified previously unknown biomarkers and potential therapeutic targets of endocrine resistance. We demonstrate that high levels of PAICS, an enzyme involved in the de novo biosynthesis of purines, can shift the balance of ERα activity to be more estrogen-independent and tamoxifen-resistant. We find that this may be due to elevated activities of cAMP-activated protein kinase A and mTOR, kinases known to phosphorylate ERα specifically and to stimulate its activity. Genetic or pharmacological targeting of PAICS sensitizes tamoxifen-resistant cells to tamoxifen. Addition of purines renders them more resistant. On the basis of these findings, we propose the combined targeting of PAICS and ERα as a new, effective, and potentially safe therapeutic regimen.

Abstract LB069: Predictive modeling of drug sensitivity based on the genetic dependency

Abstract Functional genetic screens have been frequently used to repurpose existing drugs or to identify new drug targets. However, the associations between hits on screens and candidate drugs have been unclear, and a manual approach to directly infer drug candidates from “druggable” candidates suffers from its laborious, ad hoc investigation and low accuracy. Here, we describe a systematic approach to explore the associations between genetic dependencies and drug sensitivity of cancer cell lines and to identify potential drug repositioning opportunities. We first developed drug sensitivity predictive models trained on genome-wide CRISPR-Cas9 knockout viability profiles and large-scale drug response screens of 4,518 drugs from DepMap. We obtained the highest predictive power from the decision-tree-based models and observed previously reported associations between gene perturbation and drug response. For example, cancer cells that are more sensitive to TP53 or MDM2 perturbation were more sensitive to MDM inhibitors, and EGFR or ERBB2 dependency strongly predicted the sensitivity of cells to EGFR inhibitors. Interestingly, for 3,260 drugs with known targets, only 1.5% of these genes were selected as top features to predict drug sensitivity, indicating that genes that are not the direct targets of drugs should be considered. In addition, we applied our models to the CRISPR-Cas9 knockout screens that target 1,805 “druggable” genes and predicted the sensitivity of cancer cells to known and novel drugs. Furthermore, we were able to validate the candidate drugs using cell viability assays. Finally, a multi-class classification deep learning model, based on the TorchDrug framework, reaches higher recall values than a decision-tree-based method but with much lower precision values. Our study provided the first systematic evaluation between large-scale gene dependency and drug responses. The machine learning predictive model provides an in-silico approach to perform drug screens from functional genetic screens. Citation Format: Bicna Song, Lumen Chao, Xiaolong Cheng, Yuan Gao, Ruocheng Shan, Wei Li. Predictive modeling of drug sensitivity based on the genetic dependency [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr LB069.