High-content Image-based Screen Research Articles

A deep learning dataset for sample preparation artefacts detection in multispectral high-content microscopy

High-content image-based screening is widely used in Drug Discovery and Systems Biology. However, sample preparation artefacts may significantly deteriorate the quality of image-based screening assays. While detection and circumvention of such artefacts could be addressed using modern-day machine learning and deep learning algorithms, this is widely impeded by the lack of suitable datasets. To address this, here we present a purpose-created open dataset of high-content microscopy sample preparation artefact. It consists of high-content microscopy of laboratory dust titrated on fixed cell culture specimens imaged with fluorescence filters covering the complete spectral range. To ensure this dataset is suitable for supervised machine learning tasks like image classification or segmentation we propose rule-based annotation strategies on categorical and pixel levels. We demonstrate the applicability of our dataset for deep learning by training a convolutional-neural-network-based classifier.

High-Content Image-Based Screening and Deep Learning for the Detection of Anti-Inflammatory Drug Leads.

We developed a high-content image-based screen that utilizes the pro-inflammatory stimulus lipopolysaccharide (LPS) and murine macrophages (RAW264.7) with the goal of enabling the identification of novel anti-inflammatory lead compounds. We screened 2,259 bioactive compounds with annotated mechanisms of action (MOA) to identify compounds that block the LPS-induced phenotype in macrophages. We utilized a set of seven fluorescence microscopy probes to generate images that were used to train and optimize a deep neural network classifier to distinguish between unstimulated and LPS-stimulated macrophages. The top hits from the deep learning classifier were validated using a linear classifier trained on individual cells and subsequently investigated in a multiplexed cytokine secretion assay. All 12 hits significantly modulated the expression of at least one cytokine upon LPS stimulation. Seven of these were allosteric inhibitors of the mitogen-activated protein kinase kinase (MEK1/2) and showed similar effects on cytokine expression. This deep learning morphological assay identified compounds that modulate the innate immune response to LPS and may aid in identifying new anti-inflammatory drug leads.

High-throughput and high-content bioassay enables tuning of polyester nanoparticles for cellular uptake, endosomal escape, and systemic in vivo delivery of mRNA.

Nanoparticle-based mRNA therapeutics hold great promise, but cellular internalization and endosomal escape remain key barriers for cytosolic delivery. We developed a dual nanoparticle uptake and endosomal disruption assay using high-throughput and high-content image-based screening. Using a genetically encoded Galectin 8 fluorescent fusion protein sensor, endosomal disruption could be detected via sensor clustering on damaged endosomal membranes. Simultaneously, nucleic acid endocytosis was quantified using fluorescently tagged mRNA. We used an array of biodegradable poly(beta-amino ester)s as well as Lipofectamine and PEI to demonstrate that this assay has higher predictive capacity for mRNA delivery compared to conventional polymer and nanoparticle physiochemical characteristics. Top nanoparticle formulations enabled safe and efficacious mRNA expression in multiple tissues following intravenous injection, demonstrating that the in vitro screening method is also predictive of in vivo performance. Efficacious nonviral systemic delivery of mRNA with biodegradable particles opens up new avenues for genetic medicine and human health.

DMN-Tre Labeling for Detection and High-Content Screening of Compounds against Intracellular Mycobacteria.

4-N,N-Dimethylamino-1,8-naphthalimide conjugate of trehalose (DMN-Tre) is a fluorogenic dye recently developed as a diagnostic tool for tuberculosis. DMN-Tre selectively labels the mycobacterial cell wall through the Ag85 enzymes. In this work, we disclose a protocol describing the total synthesis of DMN-Tre with more than 99% purity. We further developed a protocol for in vitro and intercellular labeling of various mycobacterial strains. DMN-Tre labeling was found to be a useful tool to study in vitro and intracellular Mycobacterium tuberculosis (Mtb) physiology and as an end-point readout system in high-content image-based screening (HCS) of drug molecules. Such uses of DMN-Tre labeling provide a simple, fast, and cheap alternative to the existing, time-consuming approach that requires Mtb strains to be genetically transformed with fluorescent reporter genes.

Myeloid Cell-Targeted Nanocarriers Efficiently Inhibit Cellular Inhibitor of Apoptosis for Cancer Immunotherapy

Immune-checkpoint blockers can promote sustained clinical responses in a subset of cancer patients. Recent research has shown that a subpopulation of tumor-infiltrating dendritic cells functions as gatekeepers, sensitizing tumors to anti-PD-1 treatment via production of interleukin-12 (IL-12). Hypothesizing that myeloid cell-targeted nanomaterials could be used to deliver small-molecule IL-12 inducers, we performed high-content image-based screening to identify the most efficacious small-molecule compounds. Using one lead candidate, LCL161, we created a myeloid-targeted nanoformulation that induced IL-12 production in intratumoral myeloid cells invivo, slowed tumor growth as a monotherapy, and had no significant systemic toxicity. These results pave the way for developing combination immunotherapeutics by harnessing IL-12 production for immunostimulation.

A novel inhibitor of inducible NOS dimerization protects against cytokine-induced rat beta cell dysfunction.

Beta cell apoptosis is a major feature of type 1 diabetes, and pro-inflammatory cytokines are key drivers of the deterioration of beta cell mass through induction of apoptosis. Mitochondrial stress plays a critical role in mediating apoptosis by releasing cytochrome C into the cytoplasm, directly activating caspase-9 and its downstream signalling cascade. We aimed to identify new compounds that protect beta cells from cytokine-induced activation of the intrinsic (mitochondrial) pathway of apoptosis. Diabetogenic media, composed of IL-1β, IFN-γ and high glucose, were used to induce mitochondrial stress in rat insulin-producing INS1E cells, and a high-content image-based screen of small molecule modulators of Casp9 pathway was performed. A novel small molecule, ATV399, was identified from a high-content image-based screen for compounds that inhibit cleaved caspase-9 activation and subsequent beta cell apoptosis induced by a combination of IL-1β, IFN-γ and high glucose, which together mimic the pathogenic diabetic milieu. Through medicinal chemistry optimization, potency was markedly improved (6-30 fold), with reduced inhibitory effects on CYP3A4. Improved analogues, such as CAT639, improved beta cell viability and insulin secretion in cytokine-treated rat insulin-producing INS1E cells and primary dispersed islet cells. Mechanistically, CAT639 reduced the production of NO by allosterically inhibiting dimerization of inducible NOS (iNOS) without affecting its mRNA levels. Taken together, these studies demonstrate a successful phenotypic screening campaign resulting in identification of an inhibitor of iNOS dimerization that protects beta cell viability and function through modulation of mitochondrial stress induced by cytokines.

A seeding based cellular assay of tauopathy.

BackgroundTauopathy is characterized by neurofibrillary tangles composed of insoluble hyperphosphorylated tau protein. Currently, cellular models that mimic neurofibrillary tangles in vitro are lacking. Previous studies indicate that neurofibrillary tangles form via a prion replication mechanism. In the present work, we establish a seeding based cellular model according to the prion hypothesis.ResultsWe show that cellular soluble tau can be converted to insoluble tau by seeds from the brain lysate of rTg4510 mice or synthetically generated preformed tau fibrils (PFFs). The cellular insoluble tau exhibits classic features of neurofibrillary tangles. Using genetic and pharmacological methods, we demonstrate that inhibition of autophagy increases whereas enhancement of autophagy reduces insoluble tau in our seeding based cellular model. The insoluble tau can be detected and quantified by thioflavin-S staining, thus allowing us to adapt our cellular model to a high-content image-based screening platform.ConclusionsOur seeding based cellular model reproduces neurofibrillary tangle pathology in vitro and serves as a useful tool for studying tauopathy and identifying tau modulators.

Concomitant BET and MAPK blockade for effective treatment of ovarian cancer

Ovarian cancer is the most lethal gynecologic malignancy, and it is imperative to develop new treatments to ameliorate patient survival. Using an anti-cancer drug library containing 180 small molecule inhibitors, we performed a high-content image-based screen and found that BET and MEK inhibitors are among the candidates which were able to effectively inhibit ovarian cancer cell growth. However, BET inhibition alone was largely cytostatic, possibly due to feedback activation of the MAPK pathway. Consequently, the combination of MEK and BET inhibitors suppressed both cell proliferation and survival, and was more efficacious than single agent. Mechanistically, BET and MEK inhibitors exerted synergistic effects on apoptosis regulators including BIM and BAD. Our findings support concomitant BET and MAPK blockade as an effective therapeutic strategy in ovarian cancer.

“Function-First” Lead Discovery: Mode of Action Profiling of Natural Product Libraries Using Image-Based Screening

Cytological profiling is a high-content image-based screening technology that provides insight into the mode of action (MOA) for test compounds by directly measuring hundreds of phenotypic cellular features. We have extended this recently reported technology to the mechanistic characterization of unknown natural products libraries for the direct prediction of compound MOAs at the primary screening stage. By analyzing a training set of commercial compounds of known mechanism and comparing these profiles to those obtained from natural product library members, we have successfully annotated extracts based on MOA, dereplicated known compounds based on biological similarity to the training set, and identified and predicted the MOA of a unique family of iron siderophores. Coupled with traditional analytical techniques, cytological profiling provides an avenue for the creation of "function-first" approaches to natural products discovery.

Identification of Regulators of Polyploidization Presents Therapeutic Targets for Treatment of AMKL

The mechanism by which cells decide to skip mitosis to become polyploid is largely undefined. Here we used a high-content image-based screen to identify small-molecule probes that induce polyploidization of megakaryocytic leukemia cells and serve as perturbagens to help understand this process. Our study implicates five networks of kinases that regulate the switch to polyploidy. Moreover, we find that dimethylfasudil (diMF, H-1152P) selectively increased polyploidization, mature cell-surface marker expression, and apoptosis of malignant megakaryocytes. An integrated target identification approach employing proteomic and shRNA screening revealed that a major target of diMF is Aurora kinase A (AURKA). We further find that MLN8237 (Alisertib), a selective inhibitor of AURKA, induced polyploidization and expression of mature megakaryocyte markers in acute megakaryocytic leukemia (AMKL) blasts and displayed potent anti-AMKL activity in vivo. Our findings provide a rationale to support clinical trials of MLN8237 and other inducers of polyploidization and differentiation in AMKL.

Cell cycle phase classification in 3D in vivo microscopy of Drosophila embryogenesis.

BackgroundCell divisions play critical roles in disease and development. The analysis of cell division phenotypes in high content image-based screening and time-lapse microscopy relies on automated nuclear segmentation and classification of cell cycle phases. Automated identification of the cell cycle phase helps biologists quantify the effect of genetic perturbations and drug treatments. Most existing studies have dealt with 2D images of cultured cells. Few, if any, studies have addressed the problem of cell cycle classification in 3D image stacks of intact tissues.ResultsWe developed a workflow for the automated cell cycle phase classification in 3D time-series image datasets of live Drosophila embryos expressing the chromatin marker histone-GFP. Upon image acquisition by laser scanning confocal microscopy and 3D nuclear segmentation, we extracted 3D intensity, shape and texture features from interphase nuclei and mitotic chromosomes. We trained different classifiers, including support vector machines (SVM) and neural networks, to distinguish between 5 cell cycles phases (Interphase and 4 mitotic phases) and achieved over 90% accuracy. As the different phases occur at different frequencies (58% of samples correspond to interphase), we devised a strategy to improve the identification of classes with low representation. To investigate which features are required for accurate classification, we performed feature reduction and selection. We were able to reduce the feature set from 42 to 9 without affecting classifier performance. We observed a dramatic decrease of classification performance when the training and testing samples were derived from two different developmental stages, the nuclear divisions of the syncytial blastoderm and the cell divisions during gastrulation. Combining samples from both developmental stages produced a more robust and accurate classifier.ConclusionsOur study demonstrates that automated cell cycle phase classification, besides 2D images of cultured cells, can also be applied to 3D images of live tissues. We could reduce the initial 3D feature set from 42 to 9 without compromising performance. Robust classifiers of intact animals need to be trained with samples from different developmental stages and cell types. Cell cycle classification in live animals can be used for automated phenotyping and to improve the performance of automated cell tracking.

High-Content Image-Based Screening for Small-Molecule Chaperone Amplifiers in Heat Shock

Heat shock proteins represent the major elements of the cellular stress response that protects cells from diseases caused by protein misfolding. Small-molecule amplifiers of heat shock proteins have shown promising results in several animal models, demonstrating the potential importance of such compounds for therapeutics. The expression of many heat shock proteins is controlled by HSF1, which forms stress granules in the nucleus when transcriptionally activated. Activation of the cellular stress also correlates with the translocation of HSP70 into nucleoli. The authors have developed an image-based, multiparametric assay to simultaneously monitor the effects of compounds on HSF1/HSP70 stress granule formation in heat-shocked Hela cells. High-content screening of the compound library was performed with a Z' of 0.62, demonstrating a highly robust assay for large-scale screening. The resulting hits showed prolonged amplification of HSP70 induction in heat-stressed cells but no effects in cells without stress. Treatment of cells with selected hits exhibited significant cytoprotection from both oxygen glucose deprivation and rotenone-induced stresses. Thus, high-content screening of HSF1/HSP70 amplifiers provides a practical opportunity for clinical therapeutics targeting protein misfolding diseases.