SUMMARY Compelling evidence demonstrates a functional link between neuronal activity and myelination, highlighting the vital importance of axon-oligodendrocyte crosstalk in myelin physiology and function. However, how neuronal activity is relayed to oligodendroglia to regulate myelin formation remains not fully understood. Here, we aimed to characterize how that myelination is regulated by glutamate vesicular release in zebrafish spinal cord. We compared oligodendrocyte precursor cells (OPCs) and myelinating oligodendrocytes (mOLs) for their close apposition with pre-synaptic boutons and found that these are increased in number on mOLs during myelin internode elongation. Consistently, mOLs show more pre-synaptic boutons during myelin internode elongation compared to OPCs. In addition, we also found that oligodendroglial cells express the post-synaptic density protein 95 (PSD-95) along punctated domains, regardless of their differentiation stage. Genetically targeted PSD-95-GFP expression in oligodendroglia revealed post-synaptic-like domains along their processes and sheaths, which are contacted by axonal pre-synaptic varicosities. These contacts are increased in mOLs. Importantly, CRISPR-Cas9 mediated deletion of dlg4 in oligodendroglia impairs myelin sheath growth , in vivo . Overall, our data indicate that PSD-95 is a key component of axons to oligodendrocytes neurotransmission that regulates myelin sheath growth. HIGHLIGHTS Glutamate vesicular release is required for myelination Axon-oligodendroglia connectivity increases with oligodendrocyte maturation Oligodendrocytes express the post-synaptic density protein 95 Dlg4 loss-of-function in oligodendroglia impedes myelin sheath growth GRAPHICAL ABSTRACT

Induced pluripotent stem cell (iPSC) models are powerful tools for neurodegenerative disease modelling, as they allow mechanistic studies in a human genetic environment and they can be differentiated into a range of neuronal and non-neuronal cells. However, these models come with inherent challenges due to line-to-line and clonal variability. To combat this issue, the iPSC Neurodegenerative Disease Initiative (iNDI) has generated an iPSC repository using a single clonal reference line, KOLF2.1J, into which disease-causing mutations and revertants are introduced via gene editing. Here we describe the generation and validation of lines carrying the most common causative mutation for amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD), a repeat expansion in the C9orf72 gene, for the iNDI collection of neurodegenerative iPSC models. We demonstrate that these C9orf72 knock-in lines differentiate efficiently into neurons and display characteristic C9orf72 -associated pathologies, including reduced C9orf72 levels and the presence of dipeptide repeat proteins (DPRs) and RNA foci, which increase in abundance over time in culture. These pathologies are not present in revertant cells lacking the repeat expansion. These repeat expansion and revertant cell lines are now available to academic and for-profit institutions through the JAX iPS cell repository and will help to facilitate and standardise iPSC-based ALS/FTD research.

Objective Candida auris has emerged as a fungal pathogen of particular concern owing in part to its propensity to exhibit antifungal resistance, especially to the commonly prescribed antifungal fluconazole. In this work we aimed to determine how mutations in the transcription factor gene TAC1B , which are common among resistant isolates and confer fluconazole resistance, exert this effect. Methods Selected TAC1B mutations from clinical isolates were introduced into a susceptible isolate and reverted to the wild-type sequence in select clinical isolates using CRISPR Cas9 gene editing. Disruption mutants were likewise generated for select genes of interest. TAC1B mutants were subjected to transcriptional profiling by RNA-seq, and relative expression of specific genes of interest was determined by qRT-PCR. Antifungal susceptibilities were determined by modified CLSI broth microdilution. Results TAC1B mutations leading to A640V, A657V, and F862_N866del conferred fluconazole resistance, as well as increased resistance to other triazoles, when introduced into a susceptible isolate. RNA-seq revealed that the ATP-Binding Cassette (ABC) transporter gene CDR1 as well as the Major Facilitator Superfamily (MFS) transporter gene MDR1 were both upregulated by these TAC1B mutations. Disruption of CDR1 greatly abrogated resistance in strains with TAC1B mutations whereas disruption of MDR1 had little to no effect. However, disruption of both CDR1 and MDR1 resulted in an additional reduction in resistance as compared to disruption of either gene alone. Conclusion TAC1B mutations leading to A640V, A657V, and F862_N866del all result in increased resistance to fluconazole and other triazole antifungals, and increased expression of both CDR1 and MDR1 in C. auris . CDR1 is the primary driver of resistance conferred by these TAC1B mutations.

Kaposi Sarcoma-associated herpesvirus (KSHV) persists as a latent episome in infected cells. While the virus efficiently infects established cell lines and primary cells in vitro , the early events guiding establishment of latent infection and the dynamic interplay between viral episomes and host factors remain incompletely understood. Here, we describe the development and application of a CRISPR/Cas9-based 3D live cell imaging system capable of tracking single KSHV episomes in real-time. Our approach exploits the SunTag technology, wherein deactivated Cas9 (dCas9) molecules are fused to repetitive epitope arrays recognized by superfolder GFP-fused single-chain antibodies. By targeting these complexes to terminal repeat units of KSHV, we achieve high level signal amplification, allowing us not only to detect newly incoming viral genomes within the first hours of de novo infection, but also to follow their spatiotemporal trajectories through different stages of the viral lifecycle. Furthermore, to facilitate efficient generation of stable reporter cell lines, we adapted the transposon-based piggyBac system to combine all SunTag components into a single-vector targeting system (SunSeT). Using these systems, we demonstrate the ability to observe both transient and stable interactions between KSHV episomes and key cellular regulators, including the variant polycomb-repressive complex 1 (vPRC) component KDM2B and the innate immune sensor IFI16. Furthermore, our platform allows detailed visualization of episodic changes in episome localization, abundance and distribution during de novo and long-term infection, providing critical insights into how viral genome positioning and dynamics correlate with host subnuclear environments. Overall, our study introduces a robust and adaptable imaging platform to dissect the earliest events of KSHV infection. The ability to track viral episomes in living cells offers a powerful tool to advance our understanding of the spatial and temporal regulation of individual KSHV genomes, shedding light on fundamental mechanisms of herpesvirus latency and persistence.

Phenotypic outcomes can be heavily affected by environmental factors. In this study, we exploited the previously observed nutrient-dependency of cell biological phenotypic features, captured by a cross-condition image-based profiling assay of Escherichia coli deletion strains, to examine this in more detail. We identified several general principles, including the existence of a spectrum of deviating phenotypes across nutrient conditions (i.e., from nutrient- or feature-specific to pleiotropic phenotypic deviations), limited conservation of phenotypic deviations across nutrient conditions (i.e., limited phenotypic robustness), and a subset of nutrient-independent phenotypic deviations (indicative of consistent genetic determinants of specific phenotypic features). In a subsequent step, we used this cross-condition dataset to identify five genes of unknown function of which the deletion displayed either nutrient-independent phenotypic deviations or phenotypic similarities to genes of known function: yibN , yaaY , yfaQ , ybiJ , and yijD . These genes showed different levels of phylogenetic conservation, ranging from conserved across the tree of life ( yibN ) to only present in some genera of the Enterobacterales ( yaaY ). Analysis of the structural properties of the proteins encoded by these y-genes, identification of structural similarities to other proteins, and the examination of their subcellular localization yielded new insights into their contribution to E. coli cell morphogenesis, cell cycle progression and cell growth. Together, our approach showcases how bacterial image-based profiling assays and datasets can serve as a gateway to reveal the function of uncharacterized proteins. Importance Despite unprecedented access to genomic information, predicting phenotypes based on genotypes remains notoriously difficult. One major confounding factor is the environment and its ability to modulate phenotypic outcomes. Another is the fact that a large fraction of protein-coding genes in bacterial genomes remains uncharacterized and have no known function. In this work, we use a large-scale cross-condition image-based profiling dataset to characterize nutrient-dependent phenotypic variability of E. coli deletion strains and exploit it to provide insight into the cellular role of genes of unknown function. Through our analysis, we identified five genes of unknown function that we subsequently further characterized by examining their phylogenetic conservation, predicted structural properties and similarities, and their intracellular localization. Combined, this approach highlights the potential of cross-condition image-based profiling, which extracts many cell biological phenotypic readouts across multiple conditions, to better understand nutrient-dependent phenotypic variability and uncover protein function.

ABSTRACT Oncogenic gene fusions are key drivers of cancer, yet most remain untargetable by current therapies. Here, we establish CRISPR- Psp Cas13b as a personalizable platform for systematic silencing of various fusion transcripts. We reveal that recognition and cleavage of the breakpoint sequence by PspCas13b disrupts the fusion transcript, resulting in unexpected RNA nicking and ligation near the cleavage site, which generates out-of-frame, translation-incompetent transcripts. This approach efficiently degrades canonical and drug-resistant BCR::ABL1 mutants (e.g., T315I), a primary cause of resistance to tyrosine kinase inhibitors (TKIs) and relapse in chronic myeloid leukemia (CML). Silencing T315I BCR::ABL1 mRNA in drug-resistant CML cells triggers extensive transcriptomic, proteomic, and phosphoproteomic remodelling, causing erythroid differentiation and apoptosis. Beyond BCR-ABL1 mutants, personalized design of Psp Cas13b effectively silences other undruggable fusions, including RUNX1::RUNX1T1 and EWSR1::FLI1, key drivers in acute myeloid leukemia and in Ewing sarcoma, respectively. Collectively, this study establishes a framework for systematic, precise, and personalizable targeting of otherwise undruggable or drug-resistant oncogenic transcripts.

ABSTRACT Regulation of protein synthesis is central to maintaining skeletal muscle integrity and its understanding is important for the treatment of muscular and neuromuscular pathologies. The eIF3f subunit of the translation initiation factor eIF3 has a key role, as it stands at the crossroad between protein-synthesis-associated hypertrophy and MAFbx/atrogin-1-dependent. To decipher the molecular mechanisms underpinning the role of eIF3f in regulating muscle mass, we established a cellular model that enables interrogation of eIF3f functionality via identification of proximal interactors. Using CRISPR-Cas9 molecular scissors, we generated single cell clones of immortalised human muscle cells expressing eIF3f fused to the BirA biotin ligase (eIF3f-BioID1 chimera) from the endogenous EIF3F locus. Biotinylated proteins, representing interactors of eIF3f in nanometer range distance, were identified by streptavidin pull-downs and mass spectrometry. In both proliferating and differentiated muscle cells, the eIF3f-BioID1 chimera co-sedimented with ribosomal complexes in polysome profiles and interacted mainly with components of the eIF3 complex, and with the eIF4E, eIF4G, and eIF5 initiation factors. Surprisingly, we identified several nucleus-localised interactors of eIF3f, and the immunofluorescence analyses revealed a previously unknown nuclear localization of eIF3f in both myoblasts and myotubes. We also identified novel cytoplasmic partners of eIF3f, responsible for the maintenance of skeletal muscle ultrastructure (sarcomeric/Z-disc (SYNPO2) bound proteins) and proteins of the lysosomal compartment (LAMP1). The established tagging system should be useful to further advance studies of eIF3f function in hypertrophic and atrophic conditions in skeletal muscle.

Abstract Endocrine therapy in combination with CDK4/6 inhibition doubles the progression-free survival of patients with advanced ER + breast cancer, but resistance is inevitable, leaving patients with limited treatment options. Here, we performed unbiased genome-wide CRISPR/Cas9 knockout screens using ER + breast cancer cells to identify novel drivers of resistance to combination endocrine therapy (tamoxifen) and CDK4/6 inhibitor (palbociclib) treatment. Our screens identified the inactivation of JNK signalling, including loss of the kinase MAP2K7 , as a key driver of combination resistance. We developed multiple CRISPR/Cas9 knockout ER + breast cancer cell lines (MCF-7 and T-47D) to investigate the effects of MAP2K7 and downstream MAPK8 and MAPK9 loss. MAP2K7 knockout increased metastatic burden in vivo and led to impaired JNK-mediated stress responses, as well as promoting cell survival and reducing senescence entry following endocrine therapy and CDK4/6 inhibitor treatment. Mechanistically, this occurred via loss of the AP-1 transcription factor c-JUN, leading to an attenuated response to combination endocrine therapy plus CDK4/6 inhibition. Furthermore, we analysed ER + advanced breast cancer patient cohorts and found that inactivation of the JNK pathway was associated with increased metastatic burden, and low pJNK T183/Y185 activity correlated with a poorer response to systemic endocrine and CDK4/6 inhibitor therapies. Overall, we demonstrate that suppression of JNK signalling enables persistent growth during combined endocrine therapy and CDK4/6 inhibition. Our data provide a pre-clinical rationale to screen patients’ tumours for JNK signalling deficiency prior to receiving combined endocrine therapy and CDK4/6 inhibition.

Parasitic plants initiate rapid de novo organogenesis of a specialized feeding structure called a haustorium upon contact with their hosts. Currently, little is known about the internal signals regulating haustorium development. Here, we identify root meristem growth factor (RGF) peptides in Phtheirospermum japonicum as endogenous inducers of prehaustorium formation. Treatment with specific RGF peptides in the absence of hosts triggered prehaustoria and induced expression of PjYUC3 , a gene required for auxin biosynthesis and prehaustorium formation. CRISPR-mediated knockouts showed that PjRGFR1 and PjRGFR3, receptors activated by the haustorium specific RGF peptides PjRGF2 and PjRGF5, are essential for prehaustorium formation, revealing functional redundancy. Phylogenic analyses indicate that PjRGF2 is broadly conserved among Orobanchaceae, whereas PjRGF5 appears to have recently evolved through tandem multiplication and neofunctionalization. Our findings establish RGF peptides and their corresponding receptors as critical components of haustorium developmental signaling and provide insights into the evolutionary trajectories that shape plant parasitism. Teaser Plant peptide hormones regulate and induce the parasitic plant specialized organ for connecting to and feeding from the host.

ABSTRACT Coronaviruses, including SARS-CoV-2, rely on host factors for their replication and pathogenesis, while hosts deploy defense mechanisms to counteract viral infections. Although numerous host proviral factors have been identified, the landscape of host restriction factors and their underlying mechanisms remain less explored. Here, we conducted genome-wide CRISPR knockout screens using three distinct coronaviruses—SARS-CoV-2, HCoV-OC43 (a common cold human virus from the genus Betacoronavirus ) and porcine epidemic diarrhea virus ( Alphacoronavirus ) to identify conserved host restriction factors. We identified glycosylphosphatidylinositol (GPI) biosynthesis as the pan-coronavirus host factor that restrict viral entry by disrupting spike protein-mediated membrane fusion at both endosomal and plasma membranes. GPI biosynthesis generates GPI moieties that covalently anchor proteins (GPI-anchored proteins [GPI-APs]) to the cell membrane, playing essential roles in various cellular processes. Through focused CRISPR knockout screens targeting 193 GPI-APs, we identified LY6E as the key downstream effector mediating the antiviral activity of the GPI biosynthesis pathway. These findings reveal a novel role for GPI biosynthesis as a conserved host defense mechanism against coronaviruses and highlight LY6E as a critical antiviral effector. This study provides new insights into virus-host interactions and the development of host-directed antiviral therapies.

The regulation of mRNA decay is important for numerous cellular and developmental processes. Here, we use the patterning gene even-skipped ( eve ) in the early Drosophila embryo to investigate the contribution of mRNA decay to shaping mature expression patterns. Through P-body colocalisation analysis and mathematical modelling of live and fixed imaging data, we present evidence that eve mRNA stability is regulated across stripe 2, with enhanced mRNA decay at the edges of the stripe. To manipulate mRNA stability, we perturbed mRNA decay in the embryo by optogenetic degradation of the 5’ to 3’ exoribonuclease Pacman (Pcm). Depleting Pcm results in larger P-bodies, which accumulate eve mRNAs, and disrupted eve expression patterns. Overall, these data show how eve mRNA instability can function with transcriptional regulation to define sharp expression domain borders. We discuss how spatially regulated mRNA stability may be widely used to sculpt expression patterns during development.

The ability to quantitatively study mRNA translation using SunTag imaging is transforming our understanding of the translation process. Here, we expand the SunTag method to study new aspects of translation regulation in Drosophila . Repression of the maternal hunchback ( hb ) mRNA in the posterior of the Drosophila embryo is a textbook example of translational control. Using SunTag imaging to quantitate translation of maternal SunTag-hb mRNAs, we show that repression in the posterior is leaky as ∼5% of SunTag-hb mRNAs are translated. In the anterior of the embryo, the maternal and zygotic SunTag-hb mRNAs show similar translation efficiency despite having different UTRs. We demonstrate that the SunTag-hb mRNA can be used as a reporter to study ribosome pausing at single-mRNA resolution, by exploiting the conserved xbp1 mRNA and A60 pausing sequences. Finally, we adapt the detector component of the SunTag system to visualise and quantitate translation of the short gastrulation ( sog ) mRNA, encoding an essential secreted extracellular BMP regulator, at the endoplasmic reticulum in fixed and live embryos. Together, these tools will facilitate the future dissection of translation regulatory mechanisms during development.

Macrophage phagocytosis is an essential immune response that eliminates pathogens, antibody-opsonized cancer cells and debris. Macrophages can also trogocytose, or nibble, targets. Trogocytosis and phagocytosis are often activated by the same signal, including IgG antibodies. What makes a macrophage trogocytose instead of phagocytose is not clear. Using both CD47 antibodies and a Her2 Chimeric Antigen Receptor (CAR) to induce phagocytosis, we found that macrophages preferentially trogocytose adherent target cells instead of phagocytose in both 2D cell monolayers and 3D cancer spheroid models. Disrupting target cell integrin using an RGD peptide or through CRISPR-Cas9 knockout of the αV integrin subunit in target cells increased macrophage phagocytosis. Conversely, increasing cell adhesion by ectopically expressing E-Cadherin in Raji B cell targets reduced phagocytosis. Finally, we examined phagocytosis of mitotic cells, a naturally occurring example of cells with reduced adhesion. Arresting target cells in mitosis significantly increased phagocytosis. Together, our data show that target cell adhesion limits phagocytosis and promotes trogocytosis.

Dosage compensation (DC) in C. elegans utilizes a condensin complex that resembles mitotic condensins, but differs by one subunit, DPY-27. DPY-27 replaces SMC-4, one of the Structural Maintenance of Chromosome (SMC) proteins that is responsible for hydrolyzing ATP, required for condensation of DNA and other mitotic condensin functions. To understand if the ATPase function is required in DC, we first demonstrated that DPY-27 is capable of hydrolyzing ATP in vitro . Then, we used CRISPR/Cas9-mediated genome editing to generate an ATPase mutation in dpy-27 and demonstrated that this mutation results in a loss of DC. Specifically, we found that without ATPase function, DPY-27 containing condensin I DC has reduced capacity to bind DNA, condense the X chromosomes, and facilitate H4K20me1 enrichment on the X-chromosomes. Our results suggest that condensin I DC , like mitotic condensins, uses ATP hydrolysis to perform its functions, making C. elegans DC a model for how activities attributed to mitotic condensins can be used to regulate gene expression.

Fate decisions of T helper (Th) cells are tightly linked to their metabolic states, but precise mechanistic links remain unknown, especially in humans. Using in vitro stimulation in combination with gene editing we studied how metabolic regulation shapes human Th1 cell identity and effector function. Differentiated Th1 cells displayed elevated STAT1 phosphorylation at Tyr701 and Ser727 as well as heightened T-bet and IFNγ expression, which were dampened by CRISPR/Cas9-mediated STAT1 deletion. Metabolic profiling revealed enhanced glycolytic activity in Th1 in comparison to Act.T cells, evidenced by increased extracellular acidification rate, ATP production via glycolysis, glucose uptake, lactate secretion and NADH abundance. SCENITH analysis demonstrated elevated glycolysis-dependent anabolic activity of Th1 cells. Inhibition of glycolysis reduced IFNγ production and STAT1 phosphorylation independent of JAK1/2 activity, STAT1 abundance or SHP-2 activity, implicating glycolysis directly in sustaining STAT1-mediated Th1 functionality. Mechanistically, O-Glycosylation, facilitated by O-Glycosyltransferase, emerged as pivotal in modulating STAT1 activity, as evident through immunoprecipitation and Western blot analysis. Pharmaceutical O-Glycosyltransferase inhibition prevented Th1 differentiation as well as STAT1 O-glycosylation. CRISPR/Cas9 mediated mutation of the O-glycosylation sites Ser499 and Thr510 sites diminished STAT1 Ser727 phosphorylation and IFNγ synthesis. Together, this study highlights glycolysis as key regulator of human Th1 cell identity and effector function, with STAT1 O-Glycosylation selectively maintaining Th1 effector capacity. This mechanism could be explored to safeguard Th1 cells in antiviral immunity and autoimmunity. Graphical abstract

H3K4me3 is a fundamental and highly conserved chromatin mark across eukaryotes, playing a central role in many genome-related processes, including transcription, maintenance of cell identity, DNA damage repair, and meiotic recombination. However, identifying the causal function of H3K4me3 in these diverse pathways remains a challenge, and we lack the tools to manipulate it for agricultural benefit. Here we use the CRISPR-based SunTag system to direct H3K4me3 methyltransferases in the model plant, Arabidopsis thaliana . Targeting of SunTag-SDG2 activates the expression of the endogenous reporter gene, FWA . We show that SunTag-SDG2 can be employed to increase pathogen resistance by targeting the H3K4me3-dependent disease resistance gene, SNC1 . Meiotic crossover recombination rates impose a limit on the speed with which new traits can be transferred to elite crop varieties. We demonstrate that targeting of SunTag-SDG2 to low recombining centromeric regions can significantly stimulate crossover formation. Finally, we reveal that the effect is not specific to SDG2 and is likely dependent on the H3K4me3 mark itself, as the orthogonal mammalian-derived H3K4me3 methyltransferase, PRDM9, produces a similar effect on gene expression with reduced off-target potential. Overall, our study supports an instructive role for H3K4me3 in transcription and meiotic recombination and opens the door to precise modulation of important agricultural traits.

Genetic diagnosis is fast and cheap, challenging our capacity to evaluate the functional impact of novel disease-causing variants or identify potential therapeutics. Model organisms including C. elegans present the possibility of systematically modelling genetic diseases, yet robust, high-throughput methods have been lacking. Here we show that automated multi-dimensional behaviour tracking can detect phenotypes in 25 new C. elegans disease models spanning homozygous loss-of-function alleles and patient-specific single-amino-acid substitutions. We find that homozygous loss-of-function (LoF) mutants across diverse genetic pathways (including BORC, FLCN, and FNIP-2) exhibit strong, readily detectable abnormalities in posture, locomotion, and stimulus responses compared to wild-type animals. An smc-3 mutant strain—modelled by introducing a patient-identified missense change—causes developmental anomalies and distinct behavioural profiles even though complete loss of SMC-3 is lethal. In contrast, patient-derived missense mutations in another essential gene, tnpo-2 , did not show a strong phenotype initially but it could be “sensitized” chemically (e.g., with aldicarb), potentially facilitating future drug screens. Our findings show that scalable behavioural phenotyping can capture a wide range of mutant effects—from strong to subtle—in patient-avatar worm lines. We anticipate that this standardized approach will enable systematic drug repurposing for rare genetic disorders as new disease variants are discovered.

A critical phase of mammalian brain development takes place after birth. Neurons of the mouse neocortex undergo dramatic changes in their morphology, physiology, and synaptic connections during the first postnatal month, while properties of immature neurons, such as the capacity for robust axon outgrowth, are lost. The genetic and epigenetic programs controlling prenatal development are well studied, but our understanding of the transcriptional mechanisms that regulate postnatal neuronal maturation is comparatively lacking. By integrating chromatin accessibility and gene expression data from two subtypes of neocortical pyramidal neurons in the neonatal and maturing brain, we predicted a role for the Krüppel-Like Factor (KLF) family of Transcription Factors in the developmental regulation of neonatally expressed genes. Using a multiplexed CRISPR Interference (CRISPRi) knockdown strategy, we found that a shift in expression from KLF activators (Klf6, Klf7) to repressors (Klf9, Klf13) during early postnatal development functions as a transcriptional ‘switch’ to first activate, then repress a set of shared targets with cytoskeletal functions including Tubb2b and Dpysl3 . We demonstrate that this switch is buffered by redundancy between KLF paralogs, which our multiplexed CRISPRi strategy is equipped to overcome and study. Our results indicate that competition between activators and repressors within the KLF family regulates a conserved component of the postnatal maturation program that may underlie the loss of intrinsic axon growth in maturing neurons. This could facilitate the transition from axon growth to synaptic refinement required to stabilize mature circuits.

ABSTRACT Hexokinase (HK) catalyzes the synthesis of glucose-6-phosphate, marking the first committed step of glucose metabolism. Most cancer cells express two homologous isoforms (HK1 and HK2) that can each bind to the outer mitochondrial membrane (OMM). CRISPR screens across hundreds of cancer cell lines indicate that both are dispensable for cell growth in traditional culture media. By contrast, HK2 deletion impairs cell growth in Human Plasma-Like Medium (HPLM). Here, we find that HK2 is required to maintain sufficient cytosolic (OMM-detached) HK activity under conditions that enhance HK1 binding to the OMM. Notably, OMM-detached rather than OMM-docked HK promotes “aerobic glycolysis” (Warburg effect), an enigmatic phenotype displayed by most proliferating cells. We show that several proposed theories for this phenotype cannot explain the HK2 dependence and instead find that HK2 deletion severely impairs glycolytic ATP production with little impact on total ATP yield for cells in HPLM. Our results reveal a basis for conditional HK2 essentiality and suggest that demand for compartmentalized ATP synthesis underlies the Warburg effect.

Promoter DNA methylation is a key epigenetic mark, commonly associated with gene silencing. However, we noticed that a positive association between promoter DNA methylation and expression is surprisingly common in cancer. Here, we use hit-and-run CRISPR/dCas9 epigenome editing to evaluate how deposition of DNA methylation can regulate gene expression dependent on pre-existing chromatin environment. While the predominant effect of DNA methylation in non-bivalent promoters is gene repression, we show that in bivalent promoters this often leads to gene activation. We demonstrate that gain of DNA methylation leads to reduced MTF2 binding and eviction of H3K27me3, a repressive mark that guards bivalent genes against activation. Our cancer patient data analyses reveal that in cancer, this mechanism likely leads to activation of a large group of transcription factors regulating pluripotency, apoptosis, and senescence signalling. In conclusion, our study uncovers an activating role of DNA methylation in bivalent promoters, with broad implications for cancer and development.

Motivation Controlling the outcomes of CRISPR editing is crucial for the success of gene therapy. Since donor template-based editing is often inefficient, alternative strategies have emerged that leverage mutagenic end-joining repair instead. Existing machine learning models can accurately predict end-joining repair outcomes, however: generalisability beyond the specific cell line used for training remains a challenge, and interpretability is typically limited by suboptimal feature representation and model architecture. Results We propose X-CRISP, a flexible and interpretable neural network for predicting repair outcome frequencies based on a minimal set of outcome and sequence features, including microhomologies (MH). Outperforming prior models on detailed and aggregate outcome predictions, X-CRISP prioritised MH location over MH sequence properties such as GC content for deletion outcomes. Through transfer learning, we adapted X-CRISP pre-trained on wild-type mESC data to target human cell lines K562, HAP1, U2OS, and mESC lines with altered DNA repair function. Adapted X-CRISP models improved over direct training on target data from as few as 50 samples, suggesting that this strategy could be leveraged to build models for new domains using a fraction of the data required to train models from scratch. Availability An implementation of X-CRISP is available at github.com/joanagoncalveslab/xcrisp .

Autosomal dominant polycystic kidney disease (ADPKD) is the most prevalent genetic kidney disorder, affecting over 10 million individuals worldwide. Cystic expansion typically progresses to kidney failure and also involves the liver with limited treatment options. Pathogenic variants in PKD1 or PKD2 account for 85-90% of cases. Genetic re-expression of Pkd1 or Pkd2 has been shown to partially reverse key characteristics of the disease phenotype in mice. Despite advancements in the understanding of the genetic basis, it remains unclear whether the correction of underlying pathogenic variants can effectively prevent, modify, or reverse the disease. Additionally, the feasibility of extrinsically delivered genome editing as a treatment option for ADPKD remains largely unexplored. In this study, we employed CRISPR base editing to correct a spectrum of representative pathogenic PKD1 variants selected from a patient cohort achieving precise and efficient editing in vitro . Correction of a representative murine missense variant (c.6646C>T (R2216W)) in primary renal epithelial cells successfully increased polycystin-1 expression and reduced levels of the endoplasmic reticulum stress marker sXBP1. In vivo , base editor delivery to the c.6646C>T (R2216W) knock-in mouse enabled correction of the pathogenic variant, resulting in a significant reduction in liver cysts. These findings provide the first evidence of ADPKD reversibility through genome editing, opening promising novel therapeutic perspectives for affected patients and their families.

ABSTRACT Basal-like breast cancer (BLBC) is characterized by an aggressive clinical course, high genomic instability, and limited therapeutic options. The Inhibitor of Differentiation 4 (ID4) protein has been identified as a critical regulator of BLBC, where its overexpression correlates with poor prognosis. However, the mechanistic contributions of ID4 to BLBC tumorigenesis remain incompletely understood. In this study, we employed an integrative approach combining CRISPR-Cas9-mediated ID4 knockout, small-molecule inhibition, in vivo tumor modeling, and in silico transcriptional analyses to investigate the functional role of ID4 in BLBC. CRISPR-Cas9-mediated knockout of ID4 in MDA-MB-231 cells resulted in significant reductions in proliferation, colony formation, and Ki67 expression, indicating a loss of aggressive phenotypic traits. In vivo xenograft studies further revealed that ID4-silenced cells exhibited markedly delayed tumor formation and a significant reduction in metastatic potential compared to controls. Kaplan-Meier survival analysis of basal-like tumors from The Cancer Genome Atlas (TCGA) dataset demonstrated that patients with low ID4 expression had improved relapse-free survival. Gene set enrichment analysis (GSEA) of BLBC tumors stratified by ID4 expression revealed a shift toward luminal-like transcriptional programs in the ID4-low subgroup, including increased estrogen response and inflammatory signaling pathways. Furthermore, transcription factor activity analysis identified the activation of MYC, JUN, and STAT in ID4-low tumors, suggesting a transition toward a more differentiated phenotype. Finally, pharmacological inhibition of ID4 using the small-molecule degrader AGX51 effectively reduced proliferation in TNBC cells, highlighting ID4 as a potential therapeutic target. Together, these findings establish ID4 as a key driver of BLBC aggressiveness and suggest that its inhibition may represent a viable therapeutic strategy. This study provides compelling evidence supporting the development of ID4-targeted therapies for TNBC patients, with the potential to improve clinical outcomes in this challenging disease subset.

A longstanding barrier in genome engineering with CRISPR-Cas9 has been the inability to measure Cas9 edit outcomes and their functional effects at single-cell resolution. Here we present Superb-seq, a new technology that leverages T7 in situ transcription and single-cell RNA sequencing to jointly measure on- and off-target Cas9 edits and their effects on gene expression. We performed Superb-seq on 10,000 K562 cells, targeting four chromatin remodeler genes with seven guide RNAs. Superb-seq identified 11,891 edit events in 6,230 edited cells at all seven on-target sites and at an additional 36 off-target sites. Although the seven guides were selected for high specificity, six of them caused off-target edits, ranging in frequency from 0.03% to 18.6% of cells. A notable off-target edit within the first intron of USP9X disrupted the expression of this gene and over 150 downstream genes. In summary, Cas9 off-targeting is pervasive due to a combination of rare and common edit events, occurs primarily within introns of off-target genes, and can exert widespread effects on gene expression. Superb-seq uses off-the-shelf kits, standard equipment, and requires no virus, which will enable genome-wide CRISPR screens in diverse cell types as well as functional characterization of clinically-relevant guides.

SUMMARY Splicing factors shape the isoform pool of most transcribed genes, playing a critical role in cellular physiology. Their dysregulation is a hallmark of diseases like cancer, where aberrant splicing contributes to progression. While exon inclusion signatures accurately assess changes in splicing factor activity, systematically mapping disease-driver regulatory interactions at scale remains challenging. Perturb-seq, which combines CRISPR-based perturbations with single-cell RNA sequencing, enables high-throughput measurement of perturbed gene expression signatures but lacks exon-level resolution, limiting its application for splicing factor activity analysis. Here, we show that shallow artificial neural networks (ANNs) can estimate splicing factor activity from gene expression signatures, bypassing the need for exon-level data. As a case study, we map the genetic interactions regulating splicing factors during carcinogenesis, using the shift in splicing program activity –where oncogenic-like splicing factors become more active than tumor suppressor-like factors– as a molecular reporter of a Perturb-seq screen. Our analysis reveals a cross-regulatory loop among splicing factors, involving protein-protein and splicing-mediated interactions, with MYC linking cancer driver mutations to splicing regulation. This network recapitulates splicing factor modulation during development. Altogether, we establish a versatile framework for studying splicing factor regulation and demonstrate its utility for uncovering disease mechanisms. GRAPHICAL ABSTRACT