Machine learning on large scale perturbation screens for SARS-CoV-2 host factors identifies ß-catenin/CBP inhibitor PRI-724 as a potent antiviral.

Expanding antiviral treatment options against SARS-CoV-2 remains crucial as the virus evolves under selection pressure which already led to the emergence of several drug resistant strains. Broad spectrum host-directed antivirals (HDA) are promising therapeutic options, however the robust identification of relevant host factors by CRISPR/Cas9 or RNA interference screens remains challenging due to low consistency in the resulting hits. To address this issue, we employed machine learning, based on experimental data from several knockout screens and a drug screen. We trained classifiers using genes essential for virus life cycle obtained from the knockout screens. The machines based their predictions on features describing cellular localization, protein domains, annotated gene sets from Gene Ontology, gene and protein sequences, and experimental data from proteomics, phospho-proteomics, protein interaction and transcriptomic profiles of SARS-CoV-2 infected cells. The models reached a remarkable performance suggesting patterns of intrinsic data consistency. The predicted HDF were enriched in sets of genes particularly encoding development, morphogenesis, and neural processes. Focusing on development and morphogenesis-associated gene sets, we found ß-catenin to be central and selected PRI-724, a canonical ß-catenin/CBP disruptor, as a potential HDA. PRI-724 limited infection with SARS-CoV-2 variants, SARS-CoV-1, MERS-CoV and IAV in different cell line models. We detected a concentration-dependent reduction in cytopathic effects, viral RNA replication, and infectious virus production in SARS-CoV-2 and SARS-CoV-1-infected cells. Independent of virus infection, PRI-724 treatment caused cell cycle deregulation which substantiates its potential as a broad spectrum antiviral. Our proposed machine learning concept supports focusing and accelerating the discovery of host dependency factors and identification of potential host-directed antivirals.

Dual role of neddylation in transcription of hepatitis B virus RNAs from cccDNA and production of viral surface antigen.

Background & Aims: HBV persistence is maintained by both an episomal covalently closed circular (ccc)DNA reservoir and genomic integration of HBV DNA fragments. While cccDNA transcription is regulated by Cullin4A-DDB1-HBx-mediated degradation of the SMC5/6 complex, HBsAg expression from integrants is largely SMC5/6 independent. Inhibiting neddylation of Cullin-RING ubiquitin ligases impairs degradation of substrates. Herein, we show that targeting neddylation pathway components by small-interfering (si)RNAs or the drug MLN4924 (pevonedistat) suppresses expression of HBV proteins from both cccDNA and integrants. Methods: An siRNA screen targeting secretory pathway regulators and neddylation genes was performed. Activity of MLN4924 was assessed in infection and integration models. Trans-complementation assays were used to study HBx function in cccDNA-driven expression. Results: siRNA screening uncovered neddylation pathway components (Nedd8, Ube2m) that promote HBsAg production post-transcriptionally. Likewise, MLN4924 inhibited production of HBsAg encoded by integrants and reduced intracellular HBsAg levels, independent of HBx. MLN4924 also profoundly inhibited cccDNA transcription in three infection models. Using the HBV inducible cell line HepAD38 as a model, we verified the dual action of MLN4924 on both cccDNA and integrants with sustained suppression of HBV markers during 42 days of treatment. Conclusions: Neddylation is required both for transcription of a cccDNA reservoir and for the genomic integration of viral DNA. Therefore, blocking neddylation might offer an attractive approach towards functional cure of chronic hepatitis B. Lay summary: Current treatments for chronic hepatitis B are rarely able to induce a functional cure. This is partly because of the presence of a pool of circular viral DNA in the host nucleus, as well as viral DNA fragments that are integrated into the host genome. Herein, we show that a host biological pathway called neddylation could play a key role in infection and viral DNA integration. Inhibiting this pathway could hold therapeutic promise for patients with chronic hepatitis B.

Neuronal miR-29a protects from obesity in adult mice.

OBJECTIVE: Obesity, a growing threat to the modern society, represents an imbalance of metabolic queues that normally signal to the arcuate hypothalamic nucleus, a critical brain region sensing and regulating energy homeostasis. This is achieved by various neurons many of which developmentally originate from the proopiomelanocortin (POMC)-expressing lineage. Within the mature neurons originating from this lineage, we aimed to identify non-coding genes in control of metabolic function in the adulthood. METHODS: In this work, we used microRNA mimic delivery and POMCCre-dependent CRISPR-Cas9 knock-out strategies in young or aged mice. Importantly, we also used CRISPR guides directing suicide cleavage of Cas9 to limit the off-target effects. RESULTS: Here we found that mature neurons originating from the POMC lineage employ miR-29a to protect against insulin resistance obesity, hyperphagia, decreased energy expenditure and obesity. Moreover, we validated the miR-29 family as a prominent regulator of the PI3K-Akt-mTOR pathway. Within the latter, we identified a direct target of miR-29a-3p, Nras, which was up-regulated in those and only those mature POMCCreCas9 neurons that were effectively transduced by anti-miR-29 CRISPR-equipped construct. Moreover, POMCCre-dependent co-deletion of Nras in mature neurons attenuated miR-29 depletion-induced obesity. CONCLUSIONS: Thus, the first to our knowledge case of in situ Cre-dependent CRISPR-Cas9-mediated knock-out of microRNAs in a specific hypothalamic neuronal population helped us to decipher a critical metabolic circuit in adult mice. This work significantly extends our understanding about the involvement of neuronal microRNAs in homeostatic regulation.

MicroRNAs Influence the Migratory Ability of Human Umbilical Vein Endothelial Cells.

To identify miRNAs that are involved in cell migration in human umbilical vein endothelial cells (HUVECs), we employed RNA sequencing under high glucose incubation and text mining within the databases miRWalk and TargetScanHuman using 83 genes that regulate HUVECs migration. From both databases, 307 predicted miRNAs were retrieved. Differentially expressed miRNAs were determined by exposing HUVECs to high glucose stimulation, which significantly inhibited the migratory ability of HUVECs as compared to cells cultured in normal glucose. A total of 35 miRNAs were found as differently expressed miRNAs in miRNA sequencing, and 4 miRNAs, namely miR-21-3p, miR-107, miR-143-3p, and miR-106b-5p, were identified as overlapping hits. These were subjected to hub gene analysis and pathway analysis using the Kyoto Encyclopedia of Genes and Genomes (KEGG), identifing 71 pathways which were influenced by all four miRNAs. The influence of all four miRNAs on HUVEC migration was phenomorphologically confirmed. miR21 and miR107 promoted migration in HUVECs while miR106b and miR143 inhibited migration. Pathway analysis also revealed eight shared pathways between the four miRNAs. Protein-protein interaction (PPI) network analysis was then performed to predict the functionality of interacting genes or proteins. This revealed six hub genes which could firstly be predicted to be related to HUVEC migration.

High-content analysis of microRNAs involved in the phenotype regulation of vascular smooth muscle cells.

In response to vascular injury vascular smooth muscle cells (VSMCs) alternate between a differentiated (contractile) and a dedifferentiated (synthetic) state or phenotype. Although parts of the signaling cascade regulating the phenotypic switch have been described, the role of miRNAs is still incompletely understood. To systematically address this issue, we have established a microscopy-based quantitative assay and identified 23 miRNAs that induced contractile phenotypes when over-expressed. These were then correlated to miRNAs identified from RNA-sequencing when comparing cells in the contractile and synthetic states. Using both approaches, six miRNAs (miR-132-3p, miR-138-5p, miR-141-3p, miR-145-5p, miR-150-5p, and miR-22-3p) were filtered as candidates that induce the phenotypic switch from synthetic to contractile. To identify potentially common regulatory mechanisms of these six miRNAs, their predicted targets were compared with five miRNAs sharing ZBTB20, ZNF704, and EIF4EBP2 as common potential targets and four miRNAs sharing 16 common potential targets. The interaction network consisting of these 19 targets and additional 18 hub targets were created to facilitate validation of miRNA-mRNA interactions by suggesting the most plausible pairs. Furthermore, the information on drug candidates was integrated into the network to predict novel combinatorial therapies that encompass the complexity of miRNAs-mediated regulation. This is the first study that combines a phenotypic screening approach with RNA sequencing and bioinformatics to systematically identify miRNA-mediated pathways and to detect potential drug candidates to positively influence the phenotypic switch of VSMCs.

PITX1 Is a Regulator of TERT Expression in Prostate Cancer with Prognostic Power.

The current risk stratification in prostate cancer (PCa) is frequently insufficient to adequately predict disease development and outcome. One hallmark of cancer is telomere maintenance. For telomere maintenance, PCa cells exclusively employ telomerase, making it essential for this cancer entity. However, TERT, the catalytic protein component of the reverse transcriptase telomerase, itself does not suit as a prognostic marker for prostate cancer as it is rather low expressed. We investigated if, instead of TERT, transcription factors regulating TERT may suit as prognostic markers. To identify transcription factors regulating TERT, we developed and applied a new gene regulatory modeling strategy to a comprehensive transcriptome dataset of 445 primary PCa. Six transcription factors were predicted as TERT regulators, and most prominently, the developmental morphogenic factor PITX1. PITX1 expression positively correlated with telomere staining intensity in PCa tumor samples. Functional assays and chromatin immune-precipitation showed that PITX1 activates TERT expression in PCa cells. Clinically, we observed that PITX1 is an excellent prognostic marker, as concluded from an analysis of more than 15,000 PCa samples. PITX1 expression in tumor samples associated with (i) increased Ki67 expression indicating increased tumor growth, (ii) a worse prognosis, and (iii) correlated with telomere length.

Image-Based Screening for Stress Granule Regulators.

Stress granule (SG)-based RNA interference (RNAi) screening is a powerful method to discover factors that control protein synthesis and aggregation, as well as regulators of SG assembly and disassembly. Here, we describe how to set up and optimize a large-scale siRNA screen, and give a detailed outline for the automated quantification of SGs as a visual readout. Hit evaluation via calculated Z scores provides a list of candidates for further in-depth studies.

Convergent use of phosphatidic acid for hepatitis C virus and SARS-CoV-2 replication organelle formation.

Double membrane vesicles (DMVs) serve as replication organelles of plus-strand RNA viruses such as hepatitis C virus (HCV) and SARS-CoV-2. Viral DMVs are morphologically analogous to DMVs formed during autophagy, but lipids driving their biogenesis are largely unknown. Here we show that production of the lipid phosphatidic acid (PA) by acylglycerolphosphate acyltransferase (AGPAT) 1 and 2 in the ER is important for DMV biogenesis in viral replication and autophagy. Using DMVs in HCV-replicating cells as model, we found that AGPATs are recruited to and critically contribute to HCV and SARS-CoV-2 replication and proper DMV formation. An intracellular PA sensor accumulated at viral DMV formation sites, consistent with elevated levels of PA in fractions of purified DMVs analyzed by lipidomics. Apart from AGPATs, PA is generated by alternative pathways and their pharmacological inhibition also impaired HCV and SARS-CoV-2 replication as well as formation of autophagosome-like DMVs. These data identify PA as host cell lipid involved in proper replication organelle formation by HCV and SARS-CoV-2, two phylogenetically disparate viruses causing very different diseases, i.e. chronic liver disease and COVID-19, respectively. Host-targeting therapy aiming at PA synthesis pathways might be suitable to attenuate replication of these viruses.

TOR signaling regulates liquid phase separation of the SMN complex governing snRNP biogenesis.

The activity of the SMN complex in promoting the assembly of pre-mRNA processing UsnRNPs correlates with condensation of the complex in nuclear Cajal bodies. While mechanistic details of its activity have been elucidated, the molecular basis for condensation remains unclear. High SMN complex phosphorylation suggests extensive regulation. Here, we report on systematic siRNA-based screening for modulators of the capacity of SMN to condense in Cajal bodies and identify mTOR and ribosomal protein S6 kinase ß-1 as key regulators. Proteomic analysis reveals TOR-dependent phosphorylations in SMN complex subunits. Using stably expressed or optogenetically controlled phospho mutants, we demonstrate that serine 49 and 63 phosphorylation of human SMN controls the capacity of the complex to condense in Cajal bodies via liquid-liquid phase separation. Our findings link SMN complex condensation and UsnRNP biogenesis to cellular energy levels and suggest modulation of TOR signaling as a rational concept for therapy of the SMN-linked neuromuscular disorder spinal muscular atrophy.

Transcriptome profiling reveals Silibinin dose-dependent response network in non-small lung cancer cells.

Silibinin (SIL), a natural flavonolignan from the milk thistle (Silybum marianum), is known to exhibit remarkable hepatoprotective, antineoplastic and EMT inhibiting effects in different cancer cells by targeting multiple molecular targets and pathways. However, the predominant majority of previous studies investigated effects of this phytocompound in a one particular cell line. Here, we carry out a systematic analysis of dose-dependent viability response to SIL in five non-small cell lung cancer (NSCLC) lines that gradually differ with respect to their intrinsic EMT stage. By correlating gene expression profiles of NSCLC cell lines with the pattern of their SIL IC50 response, a group of cell cycle, survival and stress responsive genes, including some prominent targets of STAT3 (BIRC5, FOXM1, BRCA1), was identified. The relevancy of these computationally selected genes to SIL viability response of NSCLC cells was confirmed by the transient knockdown test. In contrast to other EMT-inhibiting compounds, no correlation between the SIL IC50 and the intrinsic EMT stage of NSCLC cells was observed. Our experimental results show that SIL viability response of differently constituted NSCLC cells is linked to a subnetwork of tightly interconnected genes whose transcriptomic pattern can be used as a benchmark for assessment of individual SIL sensitivity instead of the conventional EMT signature. Insights gained in this study pave the way for optimization of customized adjuvant therapy of malignancies using Silibinin.

Automated 3D light-sheet screening with high spatiotemporal resolution reveals mitotic phenotypes.

3D cell cultures enable the in vitro study of dynamic biological processes such as the cell cycle, but their use in high-throughput screens remains impractical with conventional fluorescent microscopy. Here, we present a screening workflow for the automated evaluation of mitotic phenotypes in 3D cell cultures by light-sheet microscopy. After sample preparation by a liquid handling robot, cell spheroids are imaged for 24 h in toto with a dual-view inverted selective plane illumination microscope (diSPIM) with a much improved signal-to-noise ratio, higher imaging speed, isotropic resolution and reduced light exposure compared to a spinning disc confocal microscope. A dedicated high-content image processing pipeline implements convolutional neural network-based phenotype classification. We illustrate the potential of our approach using siRNA knockdown and epigenetic modification of 28 mitotic target genes for assessing their phenotypic role in mitosis. By rendering light-sheet microscopy operational for high-throughput screening applications, this workflow enables target gene characterization or drug candidate evaluation in tissue-like 3D cell culture models.

Pre-arrayed Pan-AAV Peptide Display Libraries for Rapid Single-Round Screening.

Display of short peptides on the surface of adeno-associated viruses (AAVs) is a powerful technology for the generation of gene therapy vectors with altered cell specificities and/or transduction efficiencies. Following its extensive prior use in the best characterized AAV serotype 2 (AAV2), recent reports also indicate the potential of other AAV isolates as scaffolds for peptide display. In this study, we systematically explored the respective capacities of 13 different AAV capsid variants to tolerate 27 peptides inserted on the surface followed by production of reporter-encoding vectors. Single-round screening in pre-arrayed 96-well plates permitted rapid and simple identification of superior vectors in >90 cell types, including T cells and primary cells. Notably, vector performance depended not only on the combination of capsid, peptide, and cell type, but also on the position of the inserted peptide and the nature of flanking residues. For optimal data availability and accessibility, all results were assembled in a searchable online database offering multiple output styles. Finally, we established a reverse-transduction pipeline based on vector pre-spotting in 96- or 384-well plates that facilitates high-throughput library panning. Our comprehensive illustration of the vast potential of alternative AAV capsids for peptide display should accelerate their in vivo screening and application as unique gene therapy vectors.

Hyperparameter optimization for image analysis: application to prostate tissue images and live cell data of virus-infected cells.

PURPOSE: Automated analysis of microscopy image data typically requires complex pipelines that involve multiple methods for different image analysis tasks. To achieve best results of the analysis pipelines, method-dependent hyperparameters need to be optimized. However, complex pipelines often suffer from the fact that calculation of the gradient of the loss function is analytically or computationally infeasible. Therefore, first- or higher-order optimization methods cannot be applied. METHODS: We developed a new framework for zero-order black-box hyperparameter optimization called HyperHyper, which has a modular architecture that separates hyperparameter sampling and optimization. We also developed a visualization of the loss function based on infimum projection to obtain further insights into the optimization problem. RESULTS: We applied HyperHyper in three different experiments with different imaging modalities, and evaluated in total more than 400.000 hyperparameter combinations. HyperHyper was used for optimizing two pipelines for cell nuclei segmentation in prostate tissue microscopy images and two pipelines for detection of hepatitis C virus proteins in live cell microscopy data. We evaluated the impact of separating the sampling and optimization strategy using different optimizers and employed an infimum projection for visualizing the hyperparameter space. CONCLUSIONS: The separation of sampling and optimization strategy of the proposed HyperHyper optimization framework improves the result of the investigated image analysis pipelines. Visualization of the loss function based on infimum projection enables gaining further insights on the optimization process.

Reciprocal Effects of Fibroblast Growth Factor Receptor Signaling on Dengue Virus Replication and Virion Production.

Dengue virus (DENV) is a human arboviral pathogen accounting for 390 million infections every year. The available vaccine has limited efficacy, and DENV-specific drugs have not been generated. To better understand DENV-host cell interaction, we employed RNA interference-based screening of the human kinome and identified fibroblast growth factor receptor 4 (FGFR4) to control the DENV replication cycle. Pharmacological inhibition of FGFR exerts a reciprocal effect by reducing DENV RNA replication and promoting the production of infectious virus particles. Addressing the latter effect, we found that the FGFR signaling pathway modulates intracellular distribution of DENV particles in a PI3K-dependent manner. Upon FGFR inhibition, virions accumulate in the trans-Golgi network compartment, where they undergo enhanced maturation cleavage of the envelope protein precursor membrane (prM), rendering virus particles more infectious. This study reveals an unexpected reciprocal role of a cellular receptor tyrosine kinase regulating DENV RNA replication and the production of infectious virions.

High-Density Cell Arrays for Genome-Scale Phenotypic Screening.

Due to high associated costs and considerable time investments of cell-based screening, there is a strong demand for new technologies that enable preclinical development and tests of diverse biologicals in a cost-saving and time-efficient manner. For those reasons we developed the high-density cell array (HD-CA) platform, which miniaturizes cell-based screening in the form of preprinted and ready-to-run screening arrays. With the HD-CA technology, up to 24,576 samples can be tested in a single experiment, thereby saving costs and time for microscopy-based screening by 75%. Experiments on the scale of the entire human genome can be addressed in a real parallel manner, with screening campaigns becoming more comfortable and devoid of robotics infrastructure on the user side. The high degree of miniaturization enables working with expensive reagents and rare and difficult-to-obtain cell lines. We have also optimized an automated imaging procedure for HD-CA and demonstrate the applicability of HD-CA to CRISPR-Cas9- and RNAi-mediated phenotypic assessment of the gene function.

Systematic RNA-interference in primary human monocyte-derived macrophages: A high-throughput platform to study foam cell formation.

Macrophage-derived foam cells are key regulators of atherogenesis. They accumulate in atherosclerotic plaques and support inflammatory processes by producing cytokines and chemokines. Identifying factors that regulate macrophage lipid uptake may reveal therapeutic targets for coronary artery disease (CAD). Here, we establish a high-throughput screening workflow to systematically identify genes that impact the uptake of DiI-labeled low-density lipoprotein (LDL) into monocyte-derived primary human macrophages. For this, monocytes isolated from peripheral blood were seeded onto 384-well plates, solid-phase transfected with siRNAs, differentiated in vitro into macrophages, and LDL-uptake per cell was measured by automated microscopy and quantitative image analysis. We applied this workflow to study how silencing of 89 genes impacts LDL-uptake into cells from 16 patients with CAD and 16 age-matched controls. Silencing of four novel genes (APOC1, CMTM6, FABP4, WBP5) reduced macrophage LDL-uptake. Additionally, knockdown of the chemokine receptor CXCR4 reduced LDL-uptake, most likely through a G-protein coupled mechanism that involves the CXCR4 ligand macrophage-induced factor (MIF), but is independent of CXCL12. We introduce a high-throughput strategy to systematically study gene function directly in primary CAD-patient cells. Our results propose a function for the MIF/CXCR4 signaling pathway, as well as several novel candidate genes impacting lipid uptake into human macrophages.

Comparison of Cell Arrays and Multi-Well Plates in Microscopy-Based Screening.

Multi-well plates and cell arrays enable microscopy-based screening assays in which many samples can be analysed in parallel. Each of the formats possesses its own strengths and weaknesses, but reference comparisons between these platforms and their application rationale is lacking. We aim to fill this gap by comparing two RNA interference (RNAi)-mediated fluorescence microscopy-based assays, namely epidermal growth factor (EGF) internalization and cell cycle progression, on both platforms. Quantitative analysis revealed that both platforms enabled the generation of data with the appearance of the expected phenotypes significantly distinct from the negative controls. The measurements of cell cycle progression were less variable in multi-well plates. The result can largely be attributed to higher cell numbers resulting in less data variability when dealing with the assay generating phenotypic cell subpopulations. The EGF internalization assay with a uniform phenotype over nearly the whole cell population performed better on cell arrays than in multi-well plates. The result was achieved by scoring five times less cells on cell arrays than in multi-well plates, indicating the efficiency of the cell array format. Our data indicate that the choice of the screening platform primarily depends on the type of the cellular assay to achieve a maximum data quality and screen efficiency.

Solid-phase reverse transfection for intracellular delivery of functionally active proteins.

Delivery of large and functionally active biomolecules across cell membranes presents a challenge in cell biological experimentation. For this purpose, we developed a novel solid-phase reverse transfection method that is suitable for the intracellular delivery of proteins into mammalian cells with preservation of their function. We show results for diverse application areas of the method, ranging from antibody-mediated inhibition of protein function to CRISPR/Cas9-based gene editing in living cells. Our method enables prefabrication of "ready to transfect" substrates carrying diverse proteins. This allows their easy distribution and standardization of biological assays across different laboratories.

Super-Resolution Microscopy Techniques and Their Potential for Applications in Radiation Biophysics.

Fluorescence microscopy is an essential tool for imaging tagged biological structures. Due to the wave nature of light, the resolution of a conventional fluorescence microscope is limited laterally to about 200 nm and axially to about 600 nm, which is often referred to as the Abbe limit. This hampers the observation of important biological structures and dynamics in the nano-scaled range ~10 nm to ~100 nm. Consequentially, various methods have been developed circumventing this limit of resolution. Super-resolution microscopy comprises several of those methods employing physical and/or chemical properties, such as optical/instrumental modifications and specific labeling of samples. In this article, we will give a brief insight into a variety of selected optical microscopy methods reaching super-resolution beyond the Abbe limit. We will survey three different concepts in connection to biological applications in radiation research without making a claim to be complete.

Fully Automated Targeted Confocal and Single-Molecule Localization Microscopy.

Single-molecule localization microscopy (SMLM) enables imaging of biological structures in the nanometre range. Long measurement times are the consequence of this kind of microscopy due to the need of acquiring thousands of images. We built a setup that automatically detects target structures using confocal microscopy and images them with SMLM. Utilizing the Konstanz Information Miner (KNIME), we were able to connect a confocal microscope with an SMLM unit for targeted screening. In this process, we developed KNIME plugins to communicate with the microscope components and combined them to a workflow. Thus, measuring biological nanometre-sized structures in a sufficient number to get statistical significance becomes feasible. For proof of principle HIV-1 assembly complexes in HeLa cells derived from transfection of replication deficient viral construct were imaged by a fully automated screen.