Understanding how cells respond differently to perturbation is crucial in cell biology, but existing methods often fail to accurately quantify and interpret heterogeneous single-cell responses. Here we introduce the perturbation-response score (PS), a method to quantify diverse perturbation responses at a single-cell level. Applied to single-cell perturbation datasets such as Perturb-seq, PS outperforms existing methods in quantifying partial gene perturbations. PS further enables single-cell dosage analysis without needing to titrate perturbations, and identifies 'buffered' and 'sensitive' response patterns of essential genes, depending on whether their moderate perturbations lead to strong downstream effects. PS reveals differential cellular responses on perturbing key genes in contexts such as T cell stimulation, latent HIV-1 expression and pancreatic differentiation. Notably, we identified a previously unknown role for the coiled-coil domain containing 6 (CCDC6) in regulating liver and pancreatic cell fate decisions. PS provides a powerful method for dose-to-function analysis, offering deeper insights from single-cell perturbation data.

Abstract CRISPR screens have become the primary discovery engine in modern biology and are widely used to uncover novel targets in immuno-oncology. However, these genetic screens are usually coupled to rather simplistic read-outs such as cell fitness. In contrast, CROP-Seq (or “CRISPR Droplet” Sequencing) screens combine CRISPR perturbation with single-cell transcriptomics, enabling high-content phenotyping at single-cell resolution. Briefly, cells are perturbed with a pooled sgRNA library and transcriptomic profiles of each cell are recorded using conventional single-cell RNA sequencing platforms. Currently, the scale of these screens is limited to the perturbation of a couple of hundred genes, possibly up to 1.000 genes, because of increased costs for single-cell library preparation and next generation sequencing (NGS). At Myllia, we have built a platform allowing high-content CROP-Seq screens in both CRISPR-ready cancer cell lines as well as primary immune cells, e.g., human T cells. Using CRISPR KO, CRISPR interference or CRISPR activation workflows, we can help accelerate the identification of target genes and regulatory networks involved in immuno-oncology. Primary T cells are of great interest in the immunotherapy community as they are key players in autoimmune and inflammatory disease. Engineered T-lymphocytes such as CAR-T cells are currently developed as novel cellular medicines, yet many T cell-intrinsic features involved in CAR-T potency remain elusive. To enable the discovery of novel targets, we have performed multiple customized CROP-Seq CRISPR KO screens in primary human T cells investigating T cell stemness and effector phenotypes as well as CD4+ T helper cell differentiation towards Th1, Th2, or Th17 subsets. Here, we present screens for regulators of functional T cell plasticity and Th2 commitment in which we aimed to identify genes determining critical T cell fate decisions, potentially linking gene (dys)function to either autoimmune/inflammatory disease or functional persistence of cytotoxic T cells for cancer immunotherapy. Overall, the advanced CROP-Seq platform combining large-scale perturbations and targeted sequencing (TA) read-outs will catalyse a paradigm shift for CROP-Seq enabling high-throughput functional genomic screens that support the validation of drug targets in autoimmunity, inflammation and immuno-oncology. Citation Format: Nicole Untermoser, Johanna Irnstorfer, Anatoly Vasilyev, Nikola Vinko, Sumit Pawar, Anke Loregger, Adam Krejci, Henrik Schmidt, Tilmann Bürckstümmer. Single-cell CRISPR screens at scale to understand T cell biology [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 LB289.

SREBP transcription factors are central regulators of lipid metabolism. Their proteolytic activation requires ER to the Golgi translocation and subsequent cleavage by site-1-protease (S1P). Produced as a proprotein, S1P undergoes autocatalytic cleavage from its precursor S1PA to mature S1PC form. Here, we report that SPRING (previously C12ORF29) and S1P interact through their ectodomains, and that this facilitates the autocatalytic cleavage of S1PA into its mature S1PC form. Reciprocally, we identified a S1P recognition-motif in SPRING and demonstrate that S1P-mediated cleavage leads to secretion of the SPRING ectodomain in cells, and in liver-specific Spring knockout (LKO) mice transduced with AAV-mSpring. By reconstituting SPRING variants into SPRINGKO cells we show that the SPRING ectodomain supports proteolytic maturation of S1P and SREBP signaling, but that S1P-mediated SPRING cleavage is not essential for these processes. Absence of SPRING modestly diminishes proteolytic maturation of S1PA -> C and trafficking of S1PC to the Golgi. However, despite reaching the Golgi in SPRINGKO cells, S1PC fails to rescue SREBP signaling. Remarkably, whereas SREBP signaling was severely attenuated in SPRINGKO cells and LKO mice, that of ATF6, another S1P substrate, was unaffected in these models. Collectively, our study positions SPRING as a dedicated licensing factor for SREBP-specific activation by S1P.

The SREBP transcription factors are central regulators of fatty acid and cholesterol metabolism. Produced as membrane-resident precursor proteins in the ER, their transcriptional activation requires the cholesterol-dependent translocation to the Golgi, and subsequent proteolytic cleavage by S1P, a type-I transmembrane protein. S1P is produced as a proprotein convertase that needs to undergo autocatalytic cleavage to attain its mature form in the Golgi, in a process that is not fully elucidated. We have recently identified SPRING (C12ORF49) as a novel regulator of the SREBP pathway and reported that S1P activity and retrograde recycling of the SREBP chaperone SCAP are affected. Here, we demonstrate that SPRING and S1P interact and that in co-transfection experiments in mammalian cells this facilitates the autocatalytic activation of S1P A→C form. Accordingly, S1P A→C processing of stably overexpressed S1P in SPRING KO cells is attenuated, but not abolished, and does not rescue SREBP signaling. Reciprocally, we identified a conserved S1P cleavage site in SPRING, and demonstrate that cleavage of SPRING results in secretion of the SPRING ectodomain. SPRING cleavage is S1P-specific and can be pharmacologically inhibited by S1P inhibitors or by mutating the S1P cleavage site. Functional analysis revealed that the SPRING ectodomain was sufficient to support S1P A→C processing and SREBP signaling, but that SPRING cleavage is not a prerequisite for this. In conclusion, our study reveals a complex interplay between the proteolytic activation of S1P and SPRING yet suggests that this is not the primary mechanism underlying the role of SPRING in SREBP signaling.

CRISPR screens have become the primary discovery engine in modern biology. However, many screening workflows are still performed in cancer cell lines and coupled to simplistic read-outs such as cellular fitness. At Myllia Biotechnology, we combine CRISPR screening with single-cell RNA sequencing, leveraging two transformative technologies to enable genetic screening for complex phenotypes. We utilize the CRISPR screening workflow to map the impact of thousands of genetic perturbations on the global transcriptome at single-cell resolution. Our powerful approach has broad applications in identifying novel drug targets or elucidating unknown mechanisms of actions of drugs. Primary human T cells are currently of great interest in the scientific community. They are not only key players in autoimmunity and other inflammatory diseases, but also represent attractive targets for immunotherapy of cancer. To enable the discovery of novel targets, we built a workflow that utilizes CD4+ T cells from peripheral blood and allows functional genomic screens in these cells. Upon activation, naïve CD4+ T cells proliferate and differentiate into specific T helper cell subsets, such as Th1, Th2, or Th17 cells. Here, we present data of an experiment in which we screened for regulators of T helper cell differentiation and skewed cells towards the Th2 subset. We aimed to identify genes whose knockout boosts or attenuates the ability of primary naïve CD4+ T cells to become Th2 cells. Th2 cells support the humoral immune response, and their dysfunction has been linked to inflammatory diseases, including asthma. In our screen, the different T cell subsets could be captured using curated transcriptomic signatures. Importantly, several gene KOs introduced in a pooled fashion using CRISPR/Cas9 accumulated in distinct subpopulations, suggesting that these genes regulate the differentiation of naïve T cells into the various T helper cell subsets. Overall, our pooled screening approach in primary human T cells allows for novel insights in the plasticity of T cells and identifies genes that could serve as drug targets in autoimmunity, inflammation and immuno-oncology. Citation Format: Anke Loregger, Johanna Irnstorfer, Nicole Untermoser, Nikola Vinko, Adam Krejci, Henrik Schmidt, Tilmann Bürckstümmer. Single-cell CRISPR screens in primary human T cells identify regulators of Th2 cell skewing [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 LB341.

CROP-seq/Perturb-seq screens combine CRISPR perturbation with single-cell RNA sequencing. In brief, cells are perturbed with a pooled sgRNA library and transcriptomic profiles of each cell are recorded using conventional single-cell RNA sequencing platforms. Currently, the scale of these screens is limited to the perturbation of a couple of hundred genes, possibly up to 1.000 genes, because costs for single-cell library preparation and next generation sequencing are high. Here, we aimed to conduct a genome-scale CRISPR screen within a reasonable budget. We chose Jurkat T cells for this experiment and focused on T cell activation because signaling pathways in these cells are well understood and we could utilize the published knowledge to benchmark the performance of our platform. To accomplish this, we utilized a CRISPR interference setup in which multiple sgRNAs can be delivered to the same cell, thus “squeezing” more perturbations into one cell. We targeted 18.595 human genes with four sgRNAs per gene and delivered this genome-scale library to Jurkat cells harboring dCas9-KRAB. Cells were stimulated with anti-TCR and anti-CD28 antibodies for 24 hours to activate T cell signaling. Following that, we processed 1.000.000 cells in one go, using the Chromium X platform that has recently been launched by 10X Genomics. Following single-cell library preparation, we amplified a selected set of 374 transcripts and submitted the corresponding library for NGS on one NovaSeq S4 flowcell. First, we confirmed that the perturbation of 374 marker genes by CRISPR interference led to the downregulation of the cognate targets, suggesting that the CRISPR perturbation workflow is functional. Then, we assessed whether T cell activation could be recapitulated from the chosen markers and found that activated Jurkat T cells can be distinguished from their unactivated counterparts using the signature in question. Finally, we assessed the phenotypes of gene knockouts introduced at genome-scale. Of the 18.595 genes perturbed, a set of 70 genes affected T cell activation, partitioning to 55 activators (whose knockout led to diminished signaling) and 15 inhibitors (whose knockout led to enhanced signaling). Of note, our screen recovered key signaling nodes that are proximal to the TCR and have been well described in the literature, including LAT, LCK, ZAP70, CD3E, ITK, RASGRP1 and VAV1. Overall, the screen presented here will catalyse a paradigm shift for CROP-seq/Perturb-seq type CRISPR screens towards genome scale. Citation Format: Anke Loregger, Nicole Untermoser, Anatoly Vasilyev, Adam Krejci, Henrik Schmidt, Tilmann Bürckstümmer. A genome-scale CROP-seq screen reveals mediators of T cell signaling [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 LB249.

The sterol regulatory element binding proteins (SREBPs) are transcription factors that govern cholesterol and fatty acid metabolism. We recently identified SPRING as a post-transcriptional regulator of SREBP activation. Constitutive or inducible global ablation of Spring in mice is not tolerated, and we therefore develop liver-specific Spring knockout mice (LKO). Transcriptomics and proteomics analysis reveal attenuated SREBP signaling in livers and hepatocytes of LKO mice. Total plasma cholesterol is reduced in male and female LKO mice in both the low-density lipoprotein and high-density lipoprotein fractions, while triglycerides are unaffected. Loss of Spring decreases hepatic cholesterol and triglyceride content due to diminished biosynthesis, which coincides with reduced very-low-density lipoprotein secretion. Accordingly, LKO mice are protected from fructose diet-induced hepatosteatosis. In humans, we find common genetic SPRING variants that associate with circulating high-density lipoprotein cholesterol and ApoA1 levels. This study positions SPRING as a core component of hepatic SREBP signaling and systemic lipid metabolism in mice and humans.

The sterol-regulatory element binding proteins (SREBP) are central transcriptional regulators of lipid metabolism. Using haploid genetic screens we identify the S REB P R egulat in g G ene ( SPRING/C12ORF49 ) as a determinant of the SREBP pathway. SPRING is a glycosylated Golgi-resident membrane protein and its ablation in Hap1 cells, Hepa1-6 hepatoma cells, and primary murine hepatocytes reduces SREBP signaling. In mice, Spring deletion is embryonic lethal yet silencing of hepatic Spring expression also attenuates the SREBP response. Mechanistically, attenuated SREBP signaling in SPRING KO cells results from reduced SREBP cleavage-activating protein (SCAP) and its mislocalization to the Golgi irrespective of the cellular sterol status. Consistent with limited functional SCAP in SPRING KO cells, reintroducing SCAP restores SREBP-dependent signaling and function. Moreover, in line with the role of SREBP in tumor growth, a wide range of tumor cell lines display dependency on SPRING expression. In conclusion, we identify SPRING as a previously unrecognized modulator of SREBP signaling.

Scope The mechanisms underlying the deleterious effects of trans fatty acids on plasma cholesterol and non‐alcoholic fatty liver disease (NAFLD) are unclear. Here, the aim is to investigate the molecular mechanisms of action of industrial trans fatty acids. Methods and results Hepa1‐6 hepatoma cells were incubated with elaidate, oleate, or palmitate. C57Bl/6 mice were fed diets rich in trans‐unsaturated, cis‐unsaturated, or saturated fatty acids. Transcriptomics analysis of Hepa1‐6 cells shows that elaidate but not oleate or palmitate induces expression of genes involved in cholesterol biosynthesis. Induction of cholesterogenesis by elaidate is mediated by increased sterol regulatory element‐binding protein 2 (SREBP2) activity and is dependent on SREBP cleavage–activating protein (SCAP), yet independent of liver‐X receptor and ubiquitin regulatory X domain‐containing protein 8. Elaidate decreases intracellular free cholesterol levels and represses the anticholesterogenic effect of exogenous cholesterol. In mice, the trans‐unsaturated diet increases the ratio of liver to gonadal fat mass, steatosis, hepatic cholesterol levels, alanine aminotransferase activity, and fibrosis markers, suggesting enhanced NAFLD, compared to the cis‐unsaturated and saturated diets. Conclusion Elaidate induces cholesterogenesis in vitro by activating the SCAP–SREBP2 axis, likely by lowering intracellular free cholesterol and attenuating cholesterol‐dependent repression of SCAP. This pathway potentially underlies the increase in liver cholesterol and NAFLD by industrial trans fatty acids.

Rho GTPases control both the actin cytoskeleton and adherens junction stability and are recognized as essential regulators of endothelial barrier function. They act as molecular switches and are primarily regulated by the exchange of GDP and GTP. However, posttranslational modifications such as phosphorylation, prenylation, and ubiquitination can additionally alter their localization, stability, and activity. F-box proteins are involved in the recognition of substrate proteins predestined for ubiquitination and subsequent degradation. Given the importance of ubiquitination, we studied the effect of the loss of 62 members of the F-box protein family on endothelial barrier function in human umbilical vein endothelial cells. Endothelial barrier function was quantified by electrical cell impedance sensing and macromolecule passage assay. Our RNA interference–based screen identified FBXW7 as a key regulator of endothelial barrier function. Mechanistically, loss of FBXW7 induced the accumulation of the RhoB GTPase in endothelial cells, resulting in their increased contractility and permeability. FBXW7 knockdown induced activation of the cholesterol biosynthesis pathway and changed the prenylation of RhoB. This effect was reversed by farnesyl transferase inhibitors and by the addition of geranylgeranyl pyrophosphate. In summary, this study identifies FBXW7 as a novel regulator of endothelial barrier function in vitro. Loss of FBXW7 indirectly modulates RhoB activity via alteration of the cholesterol biosynthesis pathway and, consequently, of the prenylation status and activity of RhoB, resulting in increased contractility and disruption of the endothelial barrier.

Background: The importance of protein glycosylation in regulating lipid metabolism is becoming increasingly apparent. We set out to further investigate this by studying patients with type I congenital disorders of glycosylation (CDGs) with defective N-glycosylation. Methods: We studied 29 patients with the 2 most prevalent types of type I CDG, ALG6 (asparagine-linked glycosylation protein 6)-deficiency CDG and PMM2 (phosphomannomutase 2)-deficiency CDG, and 23 first- and second-degree relatives with a heterozygous mutation and measured plasma cholesterol levels. Low-density lipoprotein (LDL) metabolism was studied in 3 cell models-gene silencing in HepG2 cells, patient fibroblasts, and patient hepatocyte-like cells derived from induced pluripotent stem cells-by measuring apolipoprotein B production and secretion, LDL receptor expression and membrane abundance, and LDL particle uptake. Furthermore, SREBP2 (sterol regulatory element-binding protein 2) protein expression and activation and endoplasmic reticulum stress markers were studied. Results: We report hypobetalipoproteinemia (LDL cholesterol [LDL-C] and apolipoprotein B below the fifth percentile) in a large cohort of patients with type I CDG (mean age, 9 years), together with reduced LDL-C and apolipoprotein B in clinically unaffected heterozygous relatives (mean age, 46 years), compared with 2 separate sets of age- and sex-matched control subjects. ALG6 and PMM2 deficiency led to markedly increased LDL uptake as a result of increased cell surface LDL receptor abundance. Mechanistically, this outcome was driven by increased SREBP2 protein expression accompanied by amplified target gene expression, resulting in higher LDL receptor protein levels. Endoplasmic reticulum stress was not found to be a major mediator. Conclusions: Our study establishes N-glycosylation as an important regulator of LDL metabolism. Given that LDL-C was also reduced in a group of clinically unaffected heterozygotes, we propose that increasing LDL receptor-mediated cholesterol clearance by targeting N-glycosylation in the LDL pathway may represent a novel therapeutic strategy to reduce LDL-C and cardiovascular disease.

In this Article, we reported that liver X receptors constrain metastatic development of prostate cancer in Pten-null mice. However, following institutional investigations by Universite Clermont Auvergne, it has come to our attention that much of the data reported in the paper were a result of manipulation or fabrication. Specifically, differences in protein expression in western blots presented in Fig. 1e, 3b, g, i, m, and 6b, and Supplementary Figures 2d, 6d, 6h, 7h, 8c, 8d, and 14 were generated by unequal loading of samples. In qPCR experiments presented in Figs. 1f, 3f, 4g, 5h, and 6a, d and Supplementary Figures 2c, 5d, 6g, 8d, 11a-c, and 12b, differences in expression level were manipulated through adjustment of cycle numbers, selection of samples, and data fabrication. Differences in immunostaining in Figs. 2g, k and 7e were obtained by selection of images or manipulation of exposure levels. Observed differences in relative luminescence units in Figs. 3m, 4e, and Supplementary Figures 6f, 7a, 7b, 10, and 11f were due to experiment selection. In light of these findings, we have no confidence in the accuracy of the reported data and the conclusions of the paper. We therefore wish to retract the paper. We deeply regret these circumstances and apologize to the scientific community.

This paper has been retracted at the request of the authors.

Background and aims: Cholesterol is an essential lipid for cellular function and membrane integrity, and hence its cellular levels and distribution must be tightly regulated. Biosynthesis of cholesterol is ramped when its cellular levels are low. Herein, the ER-resident and rate-limiting enzymes 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR) and squalene monooxygenase (SQLE) play a prominent role. We have recently reported that MARCH6, an E3 ubiquitin ligase, specifically promotes cholesterol-stimulated ubiquitylation and subsequent proteasomal degradation of SQLE, but not of HMGCR. To further delineate how post-translational regulation of SQLE and HMGCR is differentially achieved, we hypothesized that their sterol-dependent degradation machinery makes use of distinct E2 ubiquitin conjugating enzymes. Methods: To study this possibility, we therefore used a CRISPR/Cas9-based approach to screen for ER-associated degradation (ERAD)-associated E2 enzymes that are essential for MARCH6-dependent degradation of SQLE. Results: We report here the identification of UBE2J2 as the primary E2 ubiquitin conjugating enzyme essential for this process in mammalian cells, in contrast to UBE2G2, which is essential for sterol-stimulated degradation of HMGCR. We demonstrate that ablating UBE2J2 disturbs cholesterol-accelerated SQLE degradation in multiple human cell types, including cells of hepatic origin, and that the ability of UBE2J2 to support SQLE degradation critically depends on its enzymatic activity. Conclusions: Our findings establish UBE2J2 as an important partner of MARCH6 in cholesterol-stimulated degradation of SQLE, thereby contributing to the complex regulation of cellular cholesterol homeostasis.

Determination of LDL particle uptake into cells is a valuable technique in the field of cholesterol metabolism. This allows assessment of LDL uptake capacity in different adherent and non-adherent cells types, as well as the effect of cellular, genetic, or pharmacological perturbations on this process. Here, we detail a general procedure that describes the production of fluorescently-labeled LDL particles and quantitative and non-quantitative assays for determining cellular LDL uptake.

Objective— The cellular demand for cholesterol requires control of its biosynthesis by the mevalonate pathway. Regulation of HMGCR (3-hydroxy-3-methylglutaryl coenzyme A reductase), a rate-limiting enzyme in this pathway and the target of statins, is a key control point herein. Accordingly, HMGCR is subject to negative and positive regulation. In particular, the ability of oxysterols and intermediates of the mevalonate pathway to stimulate its proteasomal degradation is an exquisite example of metabolically controlled feedback regulation. To define the genetic determinants that govern this process, we conducted an unbiased haploid mammalian genetic screen. Approach and Results— We generated human haploid cells with mNeon fused to endogenous HMGCR using CRISPR/Cas9 and used these cells to interrogate regulation of HMGCR abundance in live cells. This resulted in identification of known and new regulators of HMGCR, and among the latter, UBXD8 (ubiquitin regulatory X domain-containing protein 8), a gene that has not been previously implicated in this process. We demonstrate that UBXD8 is an essential determinant of metabolically stimulated degradation of HMGCR and of cholesterol biosynthesis in multiple cell types. Accordingly, UBXD8 ablation leads to aberrant cholesterol synthesis due to loss of feedback control. Mechanistically, we show that UBXD8 is necessary for sterol-stimulated dislocation of ubiquitylated HMGCR from the endoplasmic reticulum membrane en route to proteasomal degradation, a function dependent on its UBX domain. Conclusions— We establish UBXD8 as a previously unrecognized determinant that couples flux across the mevalonate pathway to control of cholesterol synthesis and demonstrate the feasibility of applying mammalian haploid genetics to study metabolic traits.

Cellular cholesterol metabolism is subject to tight regulation to maintain adequate levels of this central lipid molecule. Herein, the sterol-responsive Liver X Receptors (LXRs) play an important role owing to their ability to reduce cellular cholesterol load. In this context, identifying the full set of LXR-regulated genes will contribute to our understanding of their role in cholesterol metabolism. Using global transcriptional analysis we report here the identification of RNF145 as an LXR-regulated target gene. We demonstrate that RNF145 is regulated by LXRs in both human and mouse primary cells and cell lines, and in vivo in mice. Regulation of RNF145 by LXR depends on a functional LXR-element in its proximal promotor. Consistent with LXR-dependent regulation of Rnf145 we show that regulation is lost in macrophages and fibroblasts from Lxrαβ(-/-) mice, and also in vivo in livers of Lxrα(-/-) mice treated with the LXR synthetic ligand T0901317. RNF145 is closely related to RNF139/TRC8, an E3 ligase implicated in control of SREBP processing. However, silencing of RNF145 in HepG2 or HeLa cells does not impair SREBP1/2 processing and sterol-responsive gene expression in these cells. Similar to TRC8, we demonstrate that RNF145 is localized to the ER and that it possesses intrinsic E3 ubiquitin ligase activity. In summary, we report the identification of RNF145 as an ER-resident E3 ubiquitin ligase that is transcriptionally controlled by LXR.