The Convergence of Stem Cell Technologies and Phenotypic Drug Discovery

Phenotype-based drug screening: An in vivo strategy to classify and identify the chemical compounds modulating zebrafish M-cell regeneration

Several disease-modulatory FDA-approved drugs are being used in patients with neurodegenerative diseases. However, information on their toxicity-related profiles is very limited. Therefore, measurement of drug toxicity is essential to increase the knowledge of their side effects. This study aimed to identify compounds that can modulate M-cell regeneration by causing neuro-protection and -toxicity. Here, we developed a simple and efficient in vivo assay using Tg (hsp: Gal4FF62A; UAS: nfsB-mCherry) transgenic zebrafish larvae. Interestingly, via the phenotype-based drug screening approach, we rapidly investigated 1,260 compounds from the United States drug collection and validated these in large numbers, including 14 compounds, that were obstructing this regeneration process. Next, 4 FDA-approved drugs out of 14 compounds were selected as the lead hits for in silico analysis to clarify their binding patterns with PTEN and SOCS3 signaling due to their significant potential in the inhibition of axon regeneration. Molecular docking studies indicated good binding affinity of all 4 drugs with the respective signaling molecules. This may point to PTEN and SOCS3 as the signaling molecules responsible for reducing axon regeneration. Moreover, the acute effect of compounds in reducing M-cell regeneration delineated their toxic effect. In conclusion, our in vivo along with in silico screening strategy will promote the rapid translation of new therapeutics to improve knowledge of the toxicity profile of approved/non-approved drugs efficiently.

Reproducibility of developmental neuroplasticity in in vitro brain tissue models

The current prevalence of neurodevelopmental, neurodegenerative diseases, stroke and brain injury stimulates studies aimed to identify new molecular targets, to select the drug candidates, to complete the whole set of preclinical and clinical trials, and to implement new drugs into routine neurological practice. Establishment of protocols based on microfluidics, blood-brain barrier- or neurovascular unit-on-chip, and microphysiological systems allowed improving the barrier characteristics and analyzing the regulation of local microcirculation, angiogenesis, and neurogenesis. Reconstruction of key mechanisms of brain development and even some aspects of experience-driven brain plasticity would be helpful in the establishment of brain in vitro models with the highest degree of reliability. Activity, metabolic status and expression pattern of cells within the models can be effectively assessed with the protocols of system biology, cell imaging, and functional cell analysis. The next generation of in vitro models should demonstrate high scalability, 3D or 4D complexity, possibility to be combined with other tissues or cell types within the microphysiological systems, compatibility with bio-inks or extracellular matrix-like materials, achievement of adequate vascularization, patient-specific characteristics, and opportunity to provide high-content screening. In this review, we will focus on currently available and prospective brain tissue in vitro models suitable for experimental and preclinical studies with the special focus on models enabling 4D reconstruction of brain tissue for the assessment of brain development, brain plasticity, and drug kinetics.

A chalcone derivative suppresses TSLP induction in mice and human keratinocytes through binding to BET family proteins

Although treatments for allergic diseases have improved, side effects and treatment resistance remain as challenges. New therapeutic drugs for allergic diseases are urgently required. Thymic stromal lymphopoietin (TSLP) is a cytokine target for prevention and treatment of allergic diseases. Since TSLP is produced from epithelial cells in allergic diseases, TSLP inhibitors may be new anti-allergic drugs. We previously identified a new inhibitor of TSLP production, named 16D10. However, its target of action remained unclarified. In this study, we found proteins binding to 16D10 from 24,000 human protein arrays by AlphaScreen-based high-throughput screening and identified bromodomain and extra-terminal (BET) family proteins as targets. We also clarified the detailed mode of interaction between 16D10 and a BET family protein using X-ray crystallography. Furthermore, we confirmed that inhibitors of BET family proteins suppressed TSLP induction and IL-33 and IL-36γ expression in both mouse and human keratinocyte cell lines. Taken together, our findings suggest that BET family proteins are involved in the suppression of TSLP production by 16D10. These proteins can contribute to the pathology of atopic dermatitis via TSLP regulation in keratinocytes and have potential as therapeutic targets in allergic diseases.

Combining Automated Organoid Workflows With Artificial Intelligence-Based Analyses: Opportunities to Build a New Generation of Interdisciplinary High-Throughput Screens for Parkinson's Disease and Beyond

Parkinson's disease (PD) is the second most common neurodegenerative disease and primarily characterized by the loss of dopaminergic neurons in the substantia nigra pars compacta of the midbrain. Despite decades of research and the development of various disease model systems, there is no curative treatment. This could be due to current model systems, including cell culture and animal models, not adequately recapitulating human PD etiology. More complex human disease models, including human midbrain organoids, are maturing technologies that increasingly enable the strategic incorporation of the missing components needed to model PD in vitro. The resulting organoid-based biological complexity provides new opportunities and challenges in data analysis of rich multimodal data sets. Emerging artificial intelligence (AI) capabilities can take advantage of large, broad data sets and even correlate results across disciplines. Current organoid technologies no longer lack the prerequisites for large-scale high-throughput screening (HTS) and can generate complex yet reproducible data suitable for AI-based data mining. We have recently developed a fully scalable and HTS-compatible workflow for the generation, maintenance, and analysis of three-dimensional (3D) microtissues mimicking key characteristics of the human midbrain (called "automated midbrain organoids," AMOs). AMOs build a reproducible, scalable foundation for creating next-generation 3D models of human neural disease that can fuel mechanism-agnostic phenotypic drug discovery in human in vitro PD models and beyond. Here, we explore the opportunities and challenges resulting from the convergence of organoid HTS and AI-driven data analytics and outline potential future avenues toward the discovery of novel mechanisms and drugs in PD research. © 2021 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Cell-Type-Specific High Throughput Toxicity Testing in Human Midbrain Organoids

Toxicity testing is a crucial step in the development and approval of chemical compounds for human contact and consumption. However, existing model systems often fall short in their prediction of human toxicity in vivo because they may not sufficiently recapitulate human physiology. The complexity of three-dimensional (3D) human organ-like cell culture systems ("organoids") can generate potentially more relevant models of human physiology and disease, including toxicity predictions. However, so far, the inherent biological heterogeneity and cumbersome generation and analysis of organoids has rendered efficient, unbiased, high throughput evaluation of toxic effects in these systems challenging. Recent advances in both standardization and quantitative fluorescent imaging enabled us to dissect the toxicities of compound exposure to separate cellular subpopulations within human organoids at the single-cell level in a framework that is compatible with high throughput approaches. Screening a library of 84 compounds in standardized human automated midbrain organoids (AMOs) generated from two independent cell lines correctly recognized known nigrostriatal toxicants. This approach further identified the flame retardant 3,3',5,5'-tetrabromobisphenol A (TBBPA) as a selective toxicant for dopaminergic neurons in the context of human midbrain-like tissues for the first time. Results were verified with high reproducibility in more detailed dose-response experiments. Further, we demonstrate higher sensitivity in 3D AMOs than in 2D cultures to the known neurotoxic effects of the pesticide lindane. Overall, the automated nature of our workflow is freely scalable and demonstrates the feasibility of quantitatively assessing cell-type-specific toxicity in human organoids in vitro.

The right tools for the job: the central role for next generation chemical probes and chemistry-based target deconvolution methods in phenotypic drug discovery

The reconnection of the scientific community with phenotypic drug discovery has created exciting new possibilities to develop therapies for diseases with highly complex biology. It promises to revolutionise fields such as neurodegenerative disease and regenerative medicine, where the development of new drugs has consistently proved elusive. Arguably, the greatest challenge in readopting the phenotypic drug discovery approach exists in establishing a crucial chain of translatability between phenotype and benefit to patients in the clinic. This remains a key stumbling block for the field which needs to be overcome in order to fully realise the potential of phenotypic drug discovery. Excellent quality chemical probes and chemistry-based target deconvolution techniques will be a crucial part of this process. In this review, we discuss the current capabilities of chemical probes and chemistry-based target deconvolution methods and evaluate the next advances necessary in order to fully support phenotypic screening approaches in drug discovery.

Phenotypic technologies in stem cell biology

The high-throughput phenotypic screen (HTPS) has become an emerging technology to discover synthetic small molecules that regulate stem cell fates. Here, we review the application of HTPS to identify small molecules controlling stem cell renewal, reprogramming, differentiation, and lineage conversion. Moreover, we discuss the use of HTPS to discover small molecules/polymers mimicking the stem cell extracellular niche. Furthermore, HTPSs have been applied on whole-animal models to identify small molecules regulating stem cell renewal or differentiation in vivo. Finally, we discuss the examples of the utilization of HTPS in stem cell-based disease modeling, as well as in the discovery of novel drug candidates for cancer, diabetes, and infectious diseases. Overall, HTPSs have provided many powerful tools for the stem cell field, which not only facilitate the generation of functional cells/tissues for replacement therapy, disease modeling, and drug screening, but also help dissect molecular mechanisms regulating physiological and pathological processes.

The future of phenotypic drug discovery

Phenotypic drug discovery (PDD) uses biological systems directly for new drug screening. While PDD has proved effective in the discovery of drugs with novel mechanisms, for broader adoption, key challenges need resolution: progression of poorly qualified leads and overloaded pipelines due to lack of effective tools to process and prioritize hits; and advancement of leads with undesirable mechanisms that fail at more expensive stages of discovery. Here I discuss how human-based phenotypic platforms are being applied throughout the discovery process for hit triage and prioritization, for elimination of hits with unsuitable mechanisms, and for supporting clinical strategies through pathway-based decision frameworks. Harnessing the data generated in these platforms can also fuel a deeper understanding of drug efficacy and toxicity mechanisms. As these approaches increase in use, they will gain in power for driving better decisions, generating better leads faster and in turn promoting greater adoption of PDD.

Use of artificial intelligence to enhance phenotypic drug discovery

Research and development (R&D) productivity across the pharmaceutical industry has received close scrutiny over the past two decades, especially taking into consideration reports of attrition rates and the colossal cost for drug development. The respective merits of the two main drug discovery approaches, phenotypic and target based, have divided opinion across the research community, because each hold different advantages for identifying novel molecular entities with a successful path to the market. Nevertheless, both have low translatability in the clinic. Artificial intelligence (AI) and adoption of machine learning (ML) tools offer the promise of revolutionising drug development, and overcoming obstacles in the drug discovery pipeline. Here, we assess the potential of target-driven and phenotypic-based approaches and offer a holistic description of the current state of the field, from both a scientific and industry perspective. With the emerging partnerships between AI/ML and pharma still in their relative infancy, we investigate the potential and current limitations with a particular focus on phenotypic drug discovery. Finally, we emphasise the value of public-private partnerships (PPPs) and cross-disciplinary collaborations to foster innovation and facilitate efficient drug discovery programmes.

Robust parameter design of human induced pluripotent stem cell differentiation protocols defines lineage-specific induction of anterior-posterior gut tube endodermal cells

Tissues and cells derived from pluripotent stem cells (PSC) are likely to become widely used in disease modeling, drug screening, and regenerative medicine. For these applications, the in vitro PSC differentiation process must be elaborately investigated and controlled to reliably obtain the desired end products. However, because traditional experimental methods, such as one factor at a time or brute-force approaches, are impractical for detailed screening of complex PSC cultivation conditions, more strategic and effective screening based on statistical design of experiments (DOE) ought to be indispensable. Among various DOE approaches, we regard robust parameter design (RPD) as particularly suited for differentiation protocol optimization due to its suitability for multifactorial screening. We confirmed the adaptability of RPD for investigating human induced PSC lineage specification toward anterior-posterior gut tube endodermal cells and clarified both the contribution of each cell signaling pathway and the effect of cell signaling condition alteration on marker RNA expression levels, while increasing the efficiency of the screening in 243-fold (18 vs 4374) compared with that of a brute-force approach. Specific induction of anterior foregut, hepatic, pancreatic, or mid-hindgut cells was achieved using seven iPSC strains with the optimal culture protocols established on the basis of RPD analysis. RPD has the potential to enable efficient construction and optimization of PSC differentiation protocols, and its use is recommended from fundamental research to mass production of PSC-derived products.

Integrated, automated maintenance, expansion and differentiation of 2D and 3D patient-derived cellular models for high throughput drug screening

Patient-derived cellular models become an increasingly powerful tool to model human diseases for precision medicine approaches. The identification of robust cellular disease phenotypes in these models paved the way towards high throughput screenings (HTS) including the implementation of laboratory advanced automation. However, maintenance and expansion of cells for HTS remains largely manual work. Here, we describe an integrated, complex automated platform for HTS in a translational research setting also designed for maintenance and expansion of different cell types. The comprehensive design allows automation of all cultivation steps and is flexible for development of methods for variable cell types. We demonstrate protocols for controlled cell seeding, splitting and expansion of human fibroblasts, induced pluripotent stem cells (iPSC), and neural progenitor cells (NPC) that allow for subsequent differentiation into different cell types and image-based multiparametric screening. Furthermore, we provide automated protocols for neuronal differentiation of NPC in 2D culture and 3D midbrain organoids for HTS. The flexibility of this multitask platform makes it an ideal solution for translational research settings involving experiments on different patient-derived cellular models for precision medicine.

Development and Validation of Disease Assays for Phenotypic Screening

Screening in phenotypic assays is an important strategy for the discovery of innovative drugs and novel targets. Here we present key strategies for developing successful phenotypic screens and prosecuting phenotypic drug discovery (PDD) programs. Successful screens incorporate physiological relevance through the use of human cell types and assay designs that have (1) strong mechanistic connection to clinical outcomes and (2) strong biological justification for both efficacy and safety. In addition to guidance for designing successful screens, we also propose incorporation of specific counterscreens at an early point in the program. The suggested counterscreens are based on analysis of 1000s of drugs and drug candidates profiled through a large set of human-based phenotypic assays. These assays include cytotoxicity in human primary vascular endothelial cells, proliferation of endothelial cells, and proliferation of lymphocytes, all under specific activation conditions. These counterscreens form a generic screening funnel to triage a large fraction of early-stage hits, binning compounds into those with undesirable mechanisms (associated with acute toxicity), mechanisms with utility for oncology indications, and mechanisms useful for autoimmune indications. The application of this screening funnel offers a standardized and more predictable path for prosecuting PDD programs, reducing the risk of failure, and improving program timelines.

A fully automated high-throughput workflow for 3D-based chemical screening in human midbrain organoids

Three-dimensional (3D) culture systems have fueled hopes to bring about the next generation of more physiologically relevant high-throughput screens (HTS). However, current protocols yield either complex but highly heterogeneous aggregates (‘organoids’) or 3D structures with less physiological relevance (‘spheroids’). Here, we present a scalable, HTS-compatible workflow for the automated generation, maintenance, and optical analysis of human midbrain organoids in standard 96-well-plates. The resulting organoids possess a highly homogeneous morphology, size, global gene expression, cellular composition, and structure. They present significant features of the human midbrain and display spontaneous aggregate-wide synchronized neural activity. By automating the entire workflow from generation to analysis, we enhance the intra- and inter-batch reproducibility as demonstrated via RNA sequencing and quantitative whole mount high-content imaging. This allows assessing drug effects at the single-cell level within a complex 3D cell environment in a fully automated HTS workflow.

In 1907, the American zoologist Ross Granville Harrison developed the first technique to artificially grow animal cells outside the body in a liquid medium. Cells are still grown in much the same way in modern laboratories: a single layer of cells is placed in a warm incubator with nutrient-rich broth. These cell layers are often used to test new drugs, but they cannot recapitulate the complexity of a real organ made from multiple cell types within a living, breathing human body.

Growing three-dimensional miniature organs or 'organoids' that behave in a similar way to real organs is the next step towards creating better platforms for drug screening, but there are several difficulties inherent to this process. For one thing, it is hard to recreate the multitude of cell types that make up an organ. For another, the cells that do grow often fail to connect and communicate with each other in biologically realistic ways. It is also tough to grow a large number of organoids that all behave in the same way, making it hard to know whether a particular drug works or whether it is just being tested on a 'good' organoid.

Renner et al. have been able to overcome these issues by using robotic technology to create thousands of identical, mid-brain organoids from human cells in the lab. The robots perform a series of precisely controlled tasks – including dispensing the initial cells into wells, feeding organoids as they grow and testing them at different stages of development. These mini-brains, which are the size of the head of a pin, mimic the part of the brain where Parkinson's disease first manifests. They can be used to test new drugs for Parkinson's, and to better understand the biology of the brain. Perhaps more importantly, other types of organoids can be created using the same technique to model diseases that affect other areas of the brain, or other organs altogether. For example, Renner et al. also generated forebrain organoids using an automated approach for both generation and analysis.

This research, which shows that organoids can be grown and tested in a fully automated, reproducible and scalable way, creates a platform to quickly, cheaply and easily test thousands of drugs for Parkinson's and other difficult-to-treat diseases in a human setting. This approach has the potential to reduce research waste by increasing the chances that a drug that works in the lab will also ultimately work in a patient; and reduce animal experiments, as drugs that do not work in human tissues will not proceed to animal testing.

BRCA1 and BRCA2 associated breast cancer and the roles of current modelling systems in drug discovery

For a drug candidate to be fully developed takes years and investment of hundreds of millions of dollars. There is no doubt that drug development is difficult and risky, but vital to protecting against devastating disease. This difficulty is clearly evident in BRCA1 and BRCA2 related breast cancer, with current treatment options largely confined to invasive surgical procedures, as well as chemotherapy and radiotherapy regimes which damage healthy tissue and can leave remnant disease. Consequently, patient survival and relapse rates are far from ideal, and new candidate treatments are needed. The preclinical stages of drug discovery are crucial to get right for translation to hospital beds. Disease models must take advantage of current technologies and be accurate for rapid and translatable treatments. Careful selection of cell lines must be coupled with high throughput techniques, with promising results trialled further in highly accurate humanised patient derived xenograft models. Traditional adherent drug screening should transition to 3D culture systems amenable to high throughput techniques if the gap between in vitro and in vivo studies is to be partially bridged. The possibility of organoid, induced pluripotent stem cell, and conditionally reprogrammed in vitro models is tantalising, however protocols are yet to be fully established. This review of BRCA1 and BRCA2 cancer biology and current modelling systems will hopefully guide the design of future drug discovery endeavours and highlight areas requiring improvement.

Advancements in Assay Technologies and Strategies to Enable Drug Discovery

Assays drive drug discovery from the exploratory phases to the clinical testing of drug candidates. As such, numerous assay technologies and methodologies have arisen to support drug discovery efforts. Robust identification and characterization of tractable chemical matter requires biochemical, biophysical, and cellular approaches and often benefits from high-throughput methods. To increase throughput, efforts have been made to provide assays in miniaturized volumes which can be arrayed in microtiter plates to support the testing of as many as 100,000 samples/day. Alongside these efforts has been the growth of microtiter plate-free formats with encoded libraries that can support the screening of billions of compounds, a hunt for new drug modalities, as well as emphasis on more disease relevant formats using complex cell models of disease states. This review will focus on recent developments in high-throughput assay technologies applied to identify starting points for drug discovery. We also provide recommendations on strategies for implementing various assay types to select high quality leads for drug development.

Gini Coefficients as a Single Value Metric to Define Chemical Probe Selectivity

Selectivity is a key requirement of high-quality chemical probes and lead medicines; however, methods to quantify and compare the selectivity of small molecules have not been standardized across the field. Herein, we discuss the origins and use of a comprehensive, single value term to quantify selectivity, the Gini coefficient. Case studies presented include compounds that target protein kinases, small molecules that bind RNA structures, and small molecule chimeras that bind to and degrade the target RNA. With an increasing number of transcriptome- and proteome-wide studies, we submit that reporting Gini coefficients as a quantitative descriptor of selectivity should be used broadly.

Recent advances in phenotypic drug discovery

There is a great need for innovative new medicines to treat unmet medical needs. The discovery and development of innovative new medicines is extremely difficult, costly, and inefficient. In the last decade, phenotypic drug discovery (PDD) was reintroduced as a strategy to provide first-in-class medicines. PDD uses empirical, target-agnostic lead generation to identify pharmacologically active molecules and novel therapeutics which work through unprecedented drug mechanisms. The economic and scientific value of PDD is exemplified through game-changing medicines for hepatitis C virus, spinal muscular atrophy, and cystic fibrosis. In this short review, recent advances are noted for the implementation and de-risking of PDD (for compound library selection, biomarker development, mechanism identification, and safety studies) and the potential for artificial intelligence. A significant barrier in the decision to implement PDD is balancing the potential impact of a novel mechanism of drug action with an under-defined scientific path forward, with the desire to provide infrastructure and metrics to optimize return on investment, which a known mechanism provides. A means to address this knowledge gap in the future is to empower precompetitive research utilizing the empirical concepts of PDD to identify new mechanisms and pharmacologically active compounds.

Phenotyping Reveals Targets of a Pseudo-Natural-Product Autophagy Inhibitor

Pseudo-natural-product (NP) design combines natural product fragments to provide unprecedented NP-inspired compounds not accessible by biosynthesis, but endowed with biological relevance. Since the bioactivity of pseudo-NPs may be unprecedented or unexpected, they are best evaluated in target agnostic cell-based assays monitoring entire cellular programs or complex phenotypes. Here, the Cinchona alkaloid scaffold was merged with the indole ring system to synthesize indocinchona alkaloids by Pd-catalyzed annulation. Exploration of indocinchona alkaloid bioactivities in phenotypic assays revealed a novel class of azaindole-containing autophagy inhibitors, the azaquindoles. Subsequent characterization of the most potent compound, azaquindole-1, in the morphological cell painting assay, guided target identification efforts. In contrast to the parent Cinchona alkaloids, azaquindoles selectively inhibit starvation- and rapamycin-induced autophagy by targeting the lipid kinase VPS34.

Intensive care for human hearts in pluripotent stem cell models

Successful drug discovery is ultimately contingent on the availability of workable, relevant, predictive model systems. Conversely, for cardiac muscle, the lack of human preclinical models to inform target validation and compound development has likely contributed to the perennial problem of clinical trial failures, despite encouraging non-human results. By contrast, human cardiomyocytes produced from pluripotent stem cell models have recently been applied to safety pharmacology, phenotypic screening, target validation and high-throughput assays, facilitating cardiac drug discovery. Here, we review the impact of human pluripotent stem cell models in cardiac drug discovery, discussing the range of applications, readouts, and disease models employed, along with the challenges and prospects to advance this fruitful mode of research further.

Analysis of calcium signaling in live human Tongue cell 3D-Cultures upon tastant perfusion

Bridging the gap between two-dimensional cell cultures and complex in vivo tissues, three-dimensional cell culture models are of increasing interest in the fields of cell biology and pharmacology. However, present challenges hamper live cell imaging of three-dimensional cell cultures. These include (i) the stabilization of these structures under perfusion conditions, (ii) the recording of many z-planes at high spatio-temporal resolution, (iii) and the data analysis that ranges in complexity from whole specimens to single cells. Here, we addressed these issues for the time-lapse analysis of Ca²⁺ signaling in spheroids composed of human tongue-derived HTC-8 cells upon perfusion of gustatory substances. Live cell imaging setups for confocal and light sheet microscopy were developed that allow simple and robust spheroid stabilization and high-resolution microscopy with perfusion. Visualization of spheroids made of HTC-8 cells expressing the G-GECO fluorescent Ca²⁺ sensor revealed Ca²⁺ transients that showed similar kinetics but different amplitudes upon perfusion of bitter compounds Salicine and Saccharin. Dose-dependent responses to Saccharin required extracellular Ca²⁺. From the border towards the center of spheroids, compound-induced Ca²⁺ signals were progressively delayed and decreased in amplitude. Stimulation with ATP led to strong Ca²⁺ transients that were faster than those evoked by the bitter compounds and blockade of purinergic receptors with Suramin abutted the response to Saccharin, suggesting that ATP mediates a positive autocrine and paracrine feedback. Imaging of ATP-induced Ca²⁺ transients with light sheet microscopy allowed acquisition over a z-depth of 100 μm without losing spatial and temporal resolution. In summary, the presented approaches permit the study of fast cellular signaling in three-dimensional cultures upon compound perfusion.