Morphological profiling in human neural progenitor cells classifies hits in a pilot drug screen for Alzheimer's disease.

Alzheimer's disease accounts for 60-70% of dementia cases. Current treatments are inadequate and there is a need to develop new approaches to drug discovery. Recently, in cancer, morphological profiling has been used in combination with high-throughput screening of small-molecule libraries in human cells in vitro. To test feasibility of this approach for Alzheimer's disease, we developed a cell morphology-based drug screen centred on the risk gene, SORL1 (which encodes the protein SORLA). Increased Alzheimer's disease risk has been repeatedly linked to variants in SORL1, particularly those conferring loss or decreased expression of SORLA, and lower SORL1 levels are observed in post-mortem brain samples from individuals with Alzheimer's disease. Consistent with its role in the endolysosomal pathway, SORL1 deletion is associated with enlarged endosomes in neural progenitor cells and neurons. We, therefore, hypothesized that multi-parametric, image-based cell phenotyping would identify features characteristic of SORL1 deletion. An automated morphological profiling method (Cell Painting) was adapted to neural progenitor cells and used to determine the phenotypic response of SORL1-/- neural progenitor cells to treatment with compounds from a small internationally approved drug library (TargetMol, 330 compounds). We detected distinct phenotypic signatures for SORL1-/- neural progenitor cells compared to isogenic wild-type controls. Furthermore, we identified 16 compounds (representing 14 drugs) that reversed the mutant morphological signatures in neural progenitor cells derived from three SORL1-/- induced pluripotent stem cell sub-clones. Network pharmacology analysis revealed the 16 compounds belonged to five mechanistic groups: 20S proteasome, aldehyde dehydrogenase, topoisomerase I and II, and DNA synthesis inhibitors. Enrichment analysis identified DNA synthesis/damage/repair, proteases/proteasome and metabolism as key pathways/biological processes. Prediction of novel targets revealed enrichment in pathways associated with neural cell function and Alzheimer's disease. Overall, this work suggests that (i) a quantitative phenotypic metric can distinguish induced pluripotent stem cell-derived SORL1-/- neural progenitor cells from isogenic wild-type controls and (ii) phenotypic screening combined with multi-parametric high-content image analysis is a viable option for drug repurposing and discovery in this human neural cell model of Alzheimer's disease.

High throughput screening identifies dasatinib as synergistic with trametinib in low grade serous ovarian carcinoma.

BACKGROUND: Low grade serous ovarian carcinoma (LGSOC) is a distinct histotype of ovarian cancer characterised high levels of intrinsic chemoresistance, highlighting the urgent need for new treatments. High throughput screening in clinically-informative cell-based models represents an attractive strategy for identifying candidate treatment options for prioritisation in clinical studies. METHODS: We performed a high throughput drug screen of 1610 agents across a panel of 6 LGSOC cell lines (3 RAS/RAF-mutant, 3 RAS/RAF-wildtype) to identify novel candidate therapeutic approaches. Validation comprised dose-response analysis across 9 LGSOC models and 5 high grade serous comparator lines. RESULTS: 16 hits of 1610 screened compounds were prioritised for validation based on >50% reduction in nuclei counts in over half of screened cell lines at 1000 nM concentration. 11 compounds passed validation, and the four agents of greatest interest (dasatinib, tyrosine kinase inhibitor; disulfiram, aldehyde dehydrogenase inhibitor; carfilzomib, proteasome inhibitor; romidepsin, histone deacetylase inhibitor) underwent synergy profiling with the recently approved MEK inhibitor trametinib. Disulfiram demonstrated excellent selectivity for LGSOC versus high grade serous ovarian carcinoma comparator lines (P = 0.003 for IC50 comparison), while the tyrosine kinase inhibitor dasatinib demonstrated favourable synergy with trametinib across multiple LGSOC models (maximum zero interaction potency synergy score 46.9). The novel, highly selective Src family kinase (SFK) inhibitor NXP900 demonstrated a similar trametinib synergy profile to dasatinib, suggesting that SFK inhibition is the likely driver of synergy. CONCLUSION: Dasatinib and other SFK inhibitors represent novel candidate treatments for LGSOC and demonstrate synergy with trametinib. Disulfiram represents an additional treatment strategy worthy of investigation.

Chemogenomic library design strategies for precision oncology, applied to phenotypic profiling of glioblastoma patient cells.

Designing a targeted screening library of bioactive small molecules is a challenging task since most compounds modulate their effects through multiple protein targets with varying degrees of potency and selectivity. We implemented analytic procedures for designing anticancer compound libraries adjusted for library size, cellular activity, chemical diversity and availability, and target selectivity. The resulting compound collections cover a wide range of protein targets and biological pathways implicated in various cancers, making them widely applicable to precision oncology. We characterized the compound and target spaces of the virtual libraries, in comparison with a minimal screening library of 1,211 compounds for targeting 1,386 anticancer proteins. In a pilot screening study, we identified patient-specific vulnerabilities by imaging glioma stem cells from patients with glioblastoma (GBM), using a physical library of 789 compounds that cover 1,320 of the anticancer targets. The cell survival profiling revealed highly heterogeneous phenotypic responses across the patients and GBM subtypes.

Discovery of senolytics using machine learning.

Cellular senescence is a stress response involved in ageing and diverse disease processes including cancer, type-2 diabetes, osteoarthritis and viral infection. Despite growing interest in targeted elimination of senescent cells, only few senolytics are known due to the lack of well-characterised molecular targets. Here, we report the discovery of three senolytics using cost-effective machine learning algorithms trained solely on published data. We computationally screened various chemical libraries and validated the senolytic action of ginkgetin, periplocin and oleandrin in human cell lines under various modalities of senescence. The compounds have potency comparable to known senolytics, and we show that oleandrin has improved potency over its target as compared to best-in-class alternatives. Our approach led to several hundred-fold reduction in drug screening costs and demonstrates that artificial intelligence can take maximum advantage of small and heterogeneous drug screening data, paving the way for new open science approaches to early-stage drug discovery.

Phenonaut: multiomics data integration for phenotypic space exploration.

SUMMARY: Data integration workflows for multiomics data take many forms across academia and industry. Efforts with limited resources often encountered in academia can easily fall short of data integration best practices for processing and combining high-content imaging, proteomics, metabolomics, and other omics data. We present Phenonaut, a Python software package designed to address the data workflow needs of migration, control, integration, and auditability in the application of literature and proprietary techniques for data source and structure agnostic workflow creation. AVAILABILITY AND IMPLEMENTATION: Source code: https://github.com/CarragherLab/phenonaut, Documentation: https://carragherlab.github.io/phenonaut, PyPI package: https://pypi.org/project/phenonaut/.

11ß-HSD1 inhibition does not affect murine tumour angiogenesis but may exert a selective effect on tumour growth by modulating inflammation and fibrosis.

Glucocorticoids inhibit angiogenesis by activating the glucocorticoid receptor. Inhibition of the glucocorticoid-activating enzyme 11ß-hydroxysteroid dehydrogenase type 1 (11ß-HSD1) reduces tissue-specific glucocorticoid action and promotes angiogenesis in murine models of myocardial infarction. Angiogenesis is important in the growth of some solid tumours. This study used murine models of squamous cell carcinoma (SCC) and pancreatic ductal adenocarcinoma (PDAC) to test the hypothesis that 11ß-HSD1 inhibition promotes angiogenesis and subsequent tumour growth. SCC or PDAC cells were injected into female FVB/N or C57BL6/J mice fed either standard diet, or diet containing the 11ß-HSD1 inhibitor UE2316. SCC tumours grew more rapidly in UE2316-treated mice, reaching a larger (P<0.01) final volume (0.158 ± 0.037 cm3) than in control mice (0.051 ± 0.007 cm3). However, PDAC tumour growth was unaffected. Immunofluorescent analysis of SCC tumours did not show differences in vessel density (CD31/alpha-smooth muscle actin) or cell proliferation (Ki67) after 11ß-HSD1 inhibition, and immunohistochemistry of SCC tumours did not show changes in inflammatory cell (CD3- or F4/80-positive) infiltration. In culture, the growth/viability (assessed by live cell imaging) of SCC cells was not affected by UE2316 or corticosterone. Second Harmonic Generation microscopy showed that UE2316 reduced Type I collagen (P<0.001), whilst RNA-sequencing revealed that multiple factors involved in the innate immune/inflammatory response were reduced in UE2316-treated SCC tumours. 11ß-HSD1 inhibition increases SCC tumour growth, likely via suppression of inflammatory/immune cell signalling and extracellular matrix deposition, but does not promote tumour angiogenesis or growth of all solid tumours.

Patient derived glioma stem cell spheroid reporter assays for live cell high content analysis.

Three dimensional models of cell culture enables researchers to recreate aspects of tumour biology not replicated by traditional two dimensional techniques. Here we describe a protocol to enable automated high throughput phenotypic profiling across panels of patient derived glioma stem cell spheroid models. We demonstrate the use of both live/dead cell end-points and monitor the dynamic changes in the cell cycle using cell lines expressing the FUCCI cell cycle reporter. Together, these assays provide additional insight into the mechanism of action of compound treatments over traditional cell viability assay endpoints.

Characterisation of a nucleo-adhesome.

In addition to central functions in cell adhesion signalling, integrin-associated proteins have wider roles at sites distal to adhesion receptors. In experimentally defined adhesomes, we noticed that there is clear enrichment of proteins that localise to the nucleus, and conversely, we now report that nuclear proteomes contain a class of adhesome components that localise to the nucleus. We here define a nucleo-adhesome, providing experimental evidence for a remarkable scale of nuclear localisation of adhesion proteins, establishing a framework for interrogating nuclear adhesion protein functions. Adding to nuclear FAK's known roles in regulating transcription, we now show that nuclear FAK regulates expression of many adhesion-related proteins that localise to the nucleus and that nuclear FAK binds to the adhesome component and nuclear protein Hic-5. FAK and Hic-5 work together in the nucleus, co-regulating a subset of genes transcriptionally. We demonstrate the principle that there are subcomplexes of nuclear adhesion proteins that cooperate to control transcription.

Pathway profiling of a novel SRC inhibitor, AZD0424, in combination with MEK inhibitors for cancer treatment.

A more comprehensive understanding of how cells respond to drug intervention, the likely immediate signalling responses and how resistance may develop within different microenvironments will help inform treatment regimes. The nonreceptor tyrosine kinase SRC regulates many cellular signalling processes, and pharmacological inhibition has long been a target of cancer drug discovery projects. Here, we describe the in vitro and in vivo characterisation of the small-molecule SRC inhibitor AZD0424. We show that AZD0424 potently inhibits the phosphorylation of tyrosine-419 of SRC (IC50 ~ 100 nm) in many cancer cell lines; however, inhibition of cell viability, via a G1 cell cycle arrest, was observed only in a subset of cancer cell lines in the low (on target) micromolar range. We profiled the changes in intracellular pathway signalling in cancer cells following exposure to AZD0424 and other targeted therapies using reverse-phase protein array (RPPA) analysis. We demonstrate that SRC is activated in response to treatment of KRAS-mutant colorectal cell lines with MEK inhibitors (trametinib or AZD6244) and that AZD0424 abrogates this. Cell lines treated with trametinib or AZD6244 in combination with AZD0424 had reduced EGFR, FAK and SRC compensatory activation, and cell viability was synergistically inhibited. In vivo, trametinib treatment of mice-bearing HCT116 tumours increased phosphorylation of SRC on Tyr419, and, when combined with AZD0424, inhibition of tumour growth was greater than with trametinib alone. We also demonstrate that drug-induced resistance to trametinib is not re-sensitised by AZD0424 treatment in vitro, likely as a result of multiple compensatory signalling mechanisms; however, inhibition of SRC remains an effective way to block invasion of trametinib-resistant tumour cells. These data imply that SRC inhibition may offer a useful addition to MEK inhibitor combination strategies.

A Novel High-Content Phenotypic Screen To Identify Inhibitors of Mitochondrial DNA Maintenance in Trypanosomes.

Kinetoplastid parasites cause diverse neglected diseases in humans and livestock, with an urgent need for new treatments. The survival of kinetoplastids depends on their uniquely structured mitochondrial genome (kDNA), the eponymous kinetoplast. Here, we report the development of a high-content screen for pharmacologically induced kDNA loss, based on specific staining of parasites and automated image analysis. As proof of concept, we screened a diverse set of ∼14,000 small molecules and exemplify a validated hit as a novel kDNA-targeting compound.

A Conformation Selective Mode of Inhibiting SRC Improves Drug Efficacy and Tolerability.

Despite the approval of several multikinase inhibitors that target SRC and the overwhelming evidence of the role of SRC in the progression and resistance mechanisms of many solid malignancies, inhibition of its kinase activity has thus far failed to improve patient outcomes. Here we report the small molecule eCF506 locks SRC in its native inactive conformation, thereby inhibiting both enzymatic and scaffolding functions that prevent phosphorylation and complex formation with its partner FAK. This mechanism of action resulted in highly potent and selective pathway inhibition in culture and in vivo. Treatment with eCF506 resulted in increased antitumor efficacy and tolerability in syngeneic murine cancer models, demonstrating significant therapeutic advantages over existing SRC/ABL inhibitors. Therefore, this mode of inhibiting SRC could lead to improved treatment of SRC-associated disorders. SIGNIFICANCE: Small molecule-mediated inhibition of SRC impairing both catalytic and scaffolding functions confers increased anticancer properties and tolerability compared with other SRC/ABL inhibitors.

Application of a High-Content Screening Assay Utilizing Primary Human Lung Fibroblasts to Identify Antifibrotic Drugs for Rapid Repurposing in COVID-19 Patients.

Lung imaging and autopsy reports among COVID-19 patients show elevated lung scarring (fibrosis). Early data from COVID-19 patients as well as previous studies from severe acute respiratory syndrome, Middle East respiratory syndrome, and other respiratory disorders show that the extent of lung fibrosis is associated with a higher mortality, prolonged ventilator dependence, and poorer long-term health prognosis. Current treatments to halt or reverse lung fibrosis are limited; thus, the rapid development of effective antifibrotic therapies is a major global medical need that will continue far beyond the current COVID-19 pandemic. Reproducible fibrosis screening assays with high signal-to-noise ratios and disease-relevant readouts such as extracellular matrix (ECM) deposition (the hallmark of fibrosis) are integral to any antifibrotic therapeutic development. Therefore, we have established an automated high-throughput and high-content primary screening assay measuring transforming growth factor-ß (TGFß)-induced ECM deposition from primary human lung fibroblasts in a 384-well format. This assay combines longitudinal live cell imaging with multiparametric high-content analysis of ECM deposition. Using this assay, we have screened a library of 2743 small molecules representing approved drugs and late-stage clinical candidates. Confirmed hits were subsequently profiled through a suite of secondary lung fibroblast phenotypic screening assays quantifying cell differentiation, proliferation, migration, and apoptosis. In silico target prediction and pathway network analysis were applied to the confirmed hits. We anticipate this suite of assays and data analysis tools will aid the identification of new treatments to mitigate against lung fibrosis associated with COVID-19 and other fibrotic diseases.

Targeting FAK in anticancer combination therapies.

Focal adhesion kinase (FAK) is both a non-receptor tyrosine kinase and an adaptor protein that primarily regulates adhesion signalling and cell migration, but FAK can also promote cell survival in response to stress. FAK is commonly overexpressed in cancer and is considered a high-value druggable target, with multiple FAK inhibitors currently in development. Evidence suggests that in the clinical setting, FAK targeting will be most effective in combination with other agents so as to reverse failure of chemotherapies or targeted therapies and enhance efficacy of immune-based treatments of solid tumours. Here, we discuss the recent preclinical evidence that implicates FAK in anticancer therapeutic resistance, leading to the view that FAK inhibitors will have their greatest utility as combination therapies in selected patient populations.

High-content phenotypic and pathway profiling to advance drug discovery in diseases of unmet need.

Conventional thinking in modern drug discovery postulates that the design of highly selective molecules which act on a single disease-associated target will yield safer and more effective drugs. However, high clinical attrition rates and the lack of progress in developing new effective treatments for many important diseases of unmet therapeutic need challenge this hypothesis. This assumption also impinges upon the efficiency of target agnostic phenotypic drug discovery strategies, where early target deconvolution is seen as a critical step to progress phenotypic hits. In this review we provide an overview of how emerging phenotypic and pathway-profiling technologies integrate to deconvolute the mechanism-of-action of phenotypic hits. We propose that such in-depth mechanistic profiling may support more efficient phenotypic drug discovery strategies that are designed to more appropriately address complex heterogeneous diseases of unmet need.

Trichoplein binds PCM1 and controls endothelial cell function by regulating autophagy.

Autophagy is an essential cellular quality control process that has emerged as a critical one for vascular homeostasis. Here, we show that trichoplein (TCHP) links autophagy with endothelial cell (EC) function. TCHP localizes to centriolar satellites, where it binds and stabilizes PCM1. Loss of TCHP leads to delocalization and proteasome-dependent degradation of PCM1, further resulting in degradation of PCM1's binding partner GABARAP. Autophagic flux under basal conditions is impaired in THCP-depleted ECs, and SQSTM1/p62 (p62) accumulates. We further show that TCHP promotes autophagosome maturation and efficient clearance of p62 within lysosomes, without affecting their degradative capacity. Reduced TCHP and high p62 levels are detected in primary ECs from patients with coronary artery disease. This phenotype correlates with impaired EC function and can be ameliorated by NF-κB inhibition. Moreover, Tchp knock-out mice accumulate of p62 in the heart and cardiac vessels correlating with reduced cardiac vascularization. Taken together, our data reveal that TCHP regulates endothelial cell function via an autophagy-mediated mechanism.

High content phenotypic screening identifies serotonin receptor modulators with selective activity upon breast cancer cell cycle and cytokine signaling pathways.

Heterogeneity in disease mechanisms between genetically distinct patients contributes to high attrition rates in late stage clinical drug development. New personalized medicine strategies aim to identify predictive biomarkers which stratify patients most likely to respond to a particular therapy. However, for complex multifactorial diseases not characterized by a single genetic driver, empirical approaches to identifying predictive biomarkers and the most promising therapies for personalized medicine are required. In vitro pharmacogenomics seeks to correlate in vitro drug sensitivity testing across panels of genetically distinct cell models with genomic, gene expression or proteomic data to identify predictive biomarkers of drug response. However, the vast majority of in vitro pharmacogenomic studies performed to date are limited to dose-response screening upon a single viability assay endpoint. In this article we describe the application of multiparametric high content phenotypic screening and the theta comparative cell scoring method to quantify and rank compound hits, screened at a single concentration, which induce a broad variety of divergent phenotypic responses between distinct breast cancer cell lines. High content screening followed by transcriptomic pathway analysis identified serotonin receptor modulators which display selective activity upon breast cancer cell cycle and cytokine signaling pathways correlating with inhibition of cell growth and survival. These methods describe a new evidence-led approach to rapidly identify compounds which display distinct response between different cell types. The results presented also warrant further investigation of the selective activity of serotonin receptor modulators upon breast cancer cell growth and survival as a potential drug repurposing opportunity.

A Synergistic Anticancer FAK and HDAC Inhibitor Combination Discovered by a Novel Chemical-Genetic High-Content Phenotypic Screen.

We mutated the focal adhesion kinase (FAK) catalytic domain to inhibit binding of the chaperone Cdc37 and ATP, mimicking the actions of a FAK kinase inhibitor. We reexpressed mutant and wild-type FAK in squamous cell carcinoma (SCC) cells from which endogenous FAK had been deleted, genetically fixing one axis of a FAK inhibitor combination high-content phenotypic screen to discover drugs that may synergize with FAK inhibitors. Histone deacetylase (HDAC) inhibitors represented the major class of compounds that potently induced multiparametric phenotypic changes when FAK was rendered kinase-defective or inhibited pharmacologically in SCC cells. Combined FAK and HDAC inhibitors arrest proliferation and induce apoptosis in a subset of cancer cell lines in vitro and efficiently inhibit their growth as tumors in vivo Mechanistically, HDAC inhibitors potentiate inhibitor-induced FAK inactivation and impair FAK-associated nuclear YAP in sensitive cancer cell lines. Here, we report the discovery of a new, clinically actionable, synergistic combination between FAK and HDAC inhibitors.

Drug Screening Platforms and RPPA.

Since its inception as a scalable and cost-effective method for precise quantification of the abundance of multiple protein analytes and post-translational epitopes across large sample sets, reverse phase protein array (RPPA) has been utilized as a drug discovery tool. Key RPPA drug discovery applications include primary screening of abundance or activation state of nominated protein targets, secondary screening for toxicity and selectivity, mechanism-of-action profiling, biomarker discovery, and drug combination discovery. In recent decades, drug discovery strategies have evolved dramatically in response to continual advances in technology platforms supporting high-throughput screening, structure-based drug design, new therapeutic modalities, and increasingly more complex and disease-relevant cell-based and in vivo preclinical models of disease. Advances in biological laboratory capabilities in drug discovery are complemented by significant developments in bioinformatics and computational approaches for integrating large complex datasets. Bioinformatic and computational analysis of integrated molecular, pathway network and phenotypic datasets enhance multiple stages of the drug discovery process and support more informative drug target hypothesis generation and testing. In this chapter we discuss and present examples demonstrating how the latest advances in RPPA complement and integrate with other emerging drug screening platforms to support a new era of more informative and evidence-led drug discovery strategies.

Ovarian cancer cell lines derived from non-serous carcinomas migrate and invade more aggressively than those derived from high-grade serous carcinomas.

The term ovarian cancer describes a heterogeneous group of tumours that grow in the ovary but are not necessarily of ovarian origin. Recent genomic analysis has shown that many of the most commonly used ovarian cancer cell lines have been mischaracterised, leading to erroneous conclusions and a gap in the translation of laboratory research into novel treatments for patients. Here, we use 10 epithelial ovarian cancer cell lines to investigate 2D migration, cell cycle parameters and 3D invasion behaviour into different substrates and find significant differences between the behaviours of cell lines from different origins. Cell lines derived from non-serous carcinomas migrated more quickly and were more likely to invade into Matrigel and collagen I substrates than cell lines derived from high-grade serous carcinomas. However not all cell lines derived from non-serous carcinomas exhibited similar invasive behaviour. These findings may reflect differences in the behaviour of the primary tumour types from which the cell lines were derived, given that high-grade serous carcinomas typically expand and spread over peritoneal surfaces. These results provide the basis of an in vitro model for identifying differences between ovarian cancer tumour types.

Evaluation of Machine Learning Classifiers to Predict Compound Mechanism of Action When Transferred across Distinct Cell Lines.

Multiparametric high-content imaging assays have become established to classify cell phenotypes from functional genomic and small-molecule library screening assays. Several groups have implemented machine learning classifiers to predict the mechanism of action of phenotypic hit compounds by comparing the similarity of their high-content phenotypic profiles with a reference library of well-annotated compounds. However, the majority of such examples are restricted to a single cell type often selected because of its suitability for simple image analysis and intuitive segmentation of morphological features. The aim of the current study was to evaluate and compare the performance of a classic ensemble-based tree classifier trained on extracted morphological features and a deep learning classifier using convolutional neural networks (CNNs) trained directly on images from the same dataset to predict compound mechanism of action across a morphologically and genetically distinct cell panel. Our results demonstrate that application of a CNN classifier delivers equivalent accuracy compared with an ensemble-based tree classifier at compound mechanism of action prediction within cell lines. However, our CNN analysis performs worse than an ensemble-based tree classifier when trained on multiple cell lines at predicting compound mechanism of action on an unseen cell line.