FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3136205387> ?p ?o ?g. }

• W3136205387 endingPage "319" @default.
• W3136205387 startingPage "300" @default.
• W3136205387 abstract "Profiling approaches such as gene expression or proteome profiling generate small-molecule bioactivity profiles that describe a perturbed cellular state in a rather unbiased manner and have become indispensable tools in the evaluation of bioactive small molecules. Automated imaging and image analysis can record morphological alterations that are induced by small molecules through the detection of hundreds of morphological features in high-throughput experiments. Thus, morphological profiling has gained growing attention in academia and the pharmaceutical industry as it enables detection of bioactivity in compound collections in a broader biological context in the early stages of compound development and the drug-discovery process. Profiling may be used successfully to predict mode of action or targets of unexplored compounds and to uncover unanticipated activity for already characterized small molecules. Here, we review the reported approaches to morphological profiling and the kind of bioactivity that can be detected so far and, thus, predicted. Profiling approaches such as gene expression or proteome profiling generate small-molecule bioactivity profiles that describe a perturbed cellular state in a rather unbiased manner and have become indispensable tools in the evaluation of bioactive small molecules. Automated imaging and image analysis can record morphological alterations that are induced by small molecules through the detection of hundreds of morphological features in high-throughput experiments. Thus, morphological profiling has gained growing attention in academia and the pharmaceutical industry as it enables detection of bioactivity in compound collections in a broader biological context in the early stages of compound development and the drug-discovery process. Profiling may be used successfully to predict mode of action or targets of unexplored compounds and to uncover unanticipated activity for already characterized small molecules. Here, we review the reported approaches to morphological profiling and the kind of bioactivity that can be detected so far and, thus, predicted. A phenotype unites the observable characteristics of an organism or a cell, such as gene and protein expression, morphology, and biochemical properties, and results from the interaction of genotype and environment (Nussinov et al., 2019Nussinov R. Tsai C.J. Jang H. Protein ensembles link genotype to phenotype.PLoS Comput. Biol. 2019; 15https://doi.org/10.1371/journal.pcbi.1006648Crossref Scopus (18) Google Scholar). Cell morphology has been linked to specific cellular states or cellular processes and, thus, has predictive value in the analysis of genetic, chemical, or disease-related perturbations. However, morphological alterations often are not obvious to the human eye, which may not be able to discern subtle changes independent of visualization tools, thus calling for the development of reliable and unbiased analysis methods. There has been a high demand for detailed mapping of the bioactivity space, i.e., targets, off-targets, and mode of action (MoA) of small molecules in general and, more importantly, drug candidates in particular. Whereas available approaches to detect bioactivity mostly address the already known drug-target space, e.g., G-protein-coupled receptors (GPCRs), kinases, and enzymes, in general, “omics” approaches such as transcriptomics, proteomics, epigenomics, and metabolomics enable profiling by collecting all measurable parameters to obtain a holistic view of a given cellular state. Although omics studies rarely provide direct proof for small-molecule targets, the inherently rich data they deliver may inform about numerous altered cellular traits between two states, in particular when different omics approaches are combined. Moreover, to date, generic methods for target identification of small molecules are not available, and target identification often is labor- and time-intensive (Saxena, 2016Saxena C. Identification of protein binding partners of small molecules using label-free methods.Expert Opin. Drug Dis. 2016; 11: 1017-1025Crossref PubMed Scopus (12) Google Scholar; Wilkinson et al., 2020Wilkinson I.V.L. Terstappen G.C. Russell A.J. Combining experimental strategies for successful target deconvolution.Drug Discov. Today. 2020; 25: 1998-2005Crossref Scopus (8) Google Scholar; Ziegler et al., 2013Ziegler S. Pries V. Hedberg C. Waldmann H. Target identification for small bioactive molecules: finding the needle in the haystack.Angew. Chem. Int. Ed. Engl. 2013; 52: 2744-2792Crossref PubMed Scopus (325) Google Scholar). Complementary strategies employ comparison of structural similarity (Awale and Reymond, 2019Awale M. Reymond J.L. Web-based tools for polypharmacology prediction.Methods Mol. Biol. 2019; 1888: 255-272Crossref PubMed Scopus (12) Google Scholar; Byrne and Schneider, 2019Byrne R. Schneider G. In silico target prediction for small molecules.Methods Mol. Biol. 2019; 1888: 273-309Crossref Scopus (20) Google Scholar), similarity in gene expression profiles (Lamb et al., 2006Lamb J. Crawford E.D. Peck D. Modell J.W. Blat I.C. Wrobel M.J. Lerner J. Brunet J.P. Subramanian A. Ross K.N. et al.The connectivity map: using gene-expression signatures to connect small molecules, genes, and disease.Science. 2006; 313: 1929-1935Crossref PubMed Scopus (3178) Google Scholar; Subramanian et al., 2017Subramanian A. Narayan R. Corsello S.M. Peck D.D. Natoli T.E. Lu X. Gould J. Davis J.F. Tubelli A.A. Asiedu J.K. et al.A next generation connectivity map: L1000 platform and the first 1,000,000 profiles.Cell. 2017; 171: 1437-1452 e1417Abstract Full Text Full Text PDF PubMed Scopus (878) Google Scholar), or selective cell toxicity profiles to deduce target hypotheses (Rees et al., 2016Rees M.G. Seashore-Ludlow B. Cheah J.H. Adams D.J. Price E.V. Gill S. Javaid S. Coletti M.E. Jones V.L. Bodycombe N.E. et al.Correlating chemical sensitivity and basal gene expression reveals mechanism of action.Nat. Chem. Biol. 2016; 12: 109-116Crossref PubMed Scopus (299) Google Scholar; Seashore-Ludlow et al., 2015Seashore-Ludlow B. Rees M.G. Cheah J.H. Cokol M. Price E.V. Coletti M.E. Jones V. Bodycombe N.E. Soule C.K. Gould J. et al.Harnessing connectivity in a large-scale small-molecule sensitivity dataset.Cancer Discov. 2015; 5: 1210-1223Crossref PubMed Scopus (314) Google Scholar). Similar compounds have been linked to the modulation of similar targets (Keiser et al., 2007Keiser M.J. Roth B.L. Armbruster B.N. Ernsberger P. Irwin J.J. Shoichet B.K. Relating protein pharmacology by ligand chemistry.Nat. Biotechnol. 2007; 25: 197-206Crossref PubMed Scopus (1208) Google Scholar, Keiser et al., 2009Keiser M.J. Setola V. Irwin J.J. Laggner C. Abbas A.I. Hufeisen S.J. Jensen N.H. Kuijer M.B. Matos R.C. Tran T.B. et al.Predicting new molecular targets for known drugs.Nature. 2009; 462: 175-U148Crossref PubMed Scopus (1190) Google Scholar) and similar gene expression patterns are provoked by compounds with common targets or targeted pathways (Lamb et al., 2006Lamb J. Crawford E.D. Peck D. Modell J.W. Blat I.C. Wrobel M.J. Lerner J. Brunet J.P. Subramanian A. Ross K.N. et al.The connectivity map: using gene-expression signatures to connect small molecules, genes, and disease.Science. 2006; 313: 1929-1935Crossref PubMed Scopus (3178) Google Scholar; Subramanian et al., 2017Subramanian A. Narayan R. Corsello S.M. Peck D.D. Natoli T.E. Lu X. Gould J. Davis J.F. Tubelli A.A. Asiedu J.K. et al.A next generation connectivity map: L1000 platform and the first 1,000,000 profiles.Cell. 2017; 171: 1437-1452 e1417Abstract Full Text Full Text PDF PubMed Scopus (878) Google Scholar). By analogy, selective growth inhibition in a broad cell-line panel has been successfully employed for target/MoA studies (Rees et al., 2016Rees M.G. Seashore-Ludlow B. Cheah J.H. Adams D.J. Price E.V. Gill S. Javaid S. Coletti M.E. Jones V.L. Bodycombe N.E. et al.Correlating chemical sensitivity and basal gene expression reveals mechanism of action.Nat. Chem. Biol. 2016; 12: 109-116Crossref PubMed Scopus (299) Google Scholar; Seashore-Ludlow et al., 2015Seashore-Ludlow B. Rees M.G. Cheah J.H. Cokol M. Price E.V. Coletti M.E. Jones V. Bodycombe N.E. Soule C.K. Gould J. et al.Harnessing connectivity in a large-scale small-molecule sensitivity dataset.Cancer Discov. 2015; 5: 1210-1223Crossref PubMed Scopus (314) Google Scholar). A general challenge at the heart of chemical biology is how to detect bioactivity in (synthesized) compound collections as early as possible and how to address various biological processes simultaneously that may inform compound optimization. Omics approaches per se are well suited for this purpose; however, they are not amenable to high-throughput screening. The advent of automated high-content imaging and the development of algorithms for pattern recognition, feature extraction, and, thus, image data analysis paved the way for an additional pillar in the suite of profiling approaches, i.e., morphological profiling, for chemical biology research and drug discovery. High-content assays usually are designed to monitor a given biological process and detect a limited number of parameters. However, the extraction of hundreds of parameters independent of a particular process enables the exploration of small-molecule-related bioactivity in a more unbiased manner. For morphological profiling (Figure 1), cellular components are stained using fluorescent dyes or antibodies to detect pattern changes upon perturbation by small molecules. Alternatively, fluorescent tags can be genetically introduced into cells to directly monitor morphology markers. High-content imaging is employed to capture images for each cellular component, and cell segmentation identifies cellular and subcellular regions (Bougen-Zhukov et al., 2017Bougen-Zhukov N. Loh S.Y. Lee H.K. Loo L.H. Large-scale image-based screening and profiling of cellular phenotypes.Cytometry A. 2017; 91: 115-125Crossref PubMed Scopus (32) Google Scholar). Hundreds of numerical phenotypic descriptors are then extracted to generate morphological profiles that describe phenotypes (as compared with the non-perturbed state) (Bougen-Zhukov et al., 2017Bougen-Zhukov N. Loh S.Y. Lee H.K. Loo L.H. Large-scale image-based screening and profiling of cellular phenotypes.Cytometry A. 2017; 91: 115-125Crossref PubMed Scopus (32) Google Scholar; Boutros et al., 2015Boutros M. Heigwer F. Laufer C. Microscopy-based high-content screening.Cell. 2015; 163: 1314-1325Abstract Full Text Full Text PDF PubMed Scopus (178) Google Scholar; Caicedo et al., 2017Caicedo J.C. Cooper S. Heigwer F. Warchal S. Qiu P. Molnar C. Vasilevich A.S. Barry J.D. Bansal H.S. Kraus O. et al.Data-analysis strategies for image-based cell profiling.Nat. Methods. 2017; 14: 849-863Crossref PubMed Scopus (234) Google Scholar; Grys et al., 2017Grys B.T. Lo D.S. Sahin N. Kraus O.Z. Morris Q. Boone C. Andrews B.J. Machine learning and computer vision approaches for phenotypic profiling.J. Cell Biol. 2017; 216: 65-71Crossref PubMed Scopus (79) Google Scholar). These features are related to size and shape of cells and organelles, intensity, and texture, among others. Profiles of unexplored compounds can then be compared with the profiles of agents with known targets and mode of action, where profile similarity implies common MoA or target. Dimensionality reduction is usually performed to convert high-dimensional into low-dimensional profiles by removing redundant or irrelevant features while retaining the variance of the dataset (Bougen-Zhukov et al., 2017Bougen-Zhukov N. Loh S.Y. Lee H.K. Loo L.H. Large-scale image-based screening and profiling of cellular phenotypes.Cytometry A. 2017; 91: 115-125Crossref PubMed Scopus (32) Google Scholar; Grys et al., 2017Grys B.T. Lo D.S. Sahin N. Kraus O.Z. Morris Q. Boone C. Andrews B.J. Machine learning and computer vision approaches for phenotypic profiling.J. Cell Biol. 2017; 216: 65-71Crossref PubMed Scopus (79) Google Scholar). Clusters of compounds with similar profiles and, thus, MoA, can be mapped using unsupervised machine learning, as the expected phenotypic output is unknown (Grys et al., 2017Grys B.T. Lo D.S. Sahin N. Kraus O.Z. Morris Q. Boone C. Andrews B.J. Machine learning and computer vision approaches for phenotypic profiling.J. Cell Biol. 2017; 216: 65-71Crossref PubMed Scopus (79) Google Scholar). Phenotypic categories that can be defined prior to the analysis can be used in supervised approaches to train models that will then assign new profiles to existing categories (Bougen-Zhukov et al., 2017Bougen-Zhukov N. Loh S.Y. Lee H.K. Loo L.H. Large-scale image-based screening and profiling of cellular phenotypes.Cytometry A. 2017; 91: 115-125Crossref PubMed Scopus (32) Google Scholar; Grys et al., 2017Grys B.T. Lo D.S. Sahin N. Kraus O.Z. Morris Q. Boone C. Andrews B.J. Machine learning and computer vision approaches for phenotypic profiling.J. Cell Biol. 2017; 216: 65-71Crossref PubMed Scopus (79) Google Scholar). Thus, morphological profiling has the power to assign already known and unanticipated MoAs to annotated compounds, to detect bioactivity, and to predict MoAs for biologically uncharacterized small molecules. This review aims to give an overview of the currently reported approaches to morphological profiling of small molecules according to the employed biological system and strategies to detect cellular components, the kind of bioactivity that can be assessed by these methods (see Table S1), and the lessons learned by their application. It does not focus on the computational approaches for extraction and analysis of morphological features (Bougen-Zhukov et al., 2017Bougen-Zhukov N. Loh S.Y. Lee H.K. Loo L.H. Large-scale image-based screening and profiling of cellular phenotypes.Cytometry A. 2017; 91: 115-125Crossref PubMed Scopus (32) Google Scholar; Grys et al., 2017Grys B.T. Lo D.S. Sahin N. Kraus O.Z. Morris Q. Boone C. Andrews B.J. Machine learning and computer vision approaches for phenotypic profiling.J. Cell Biol. 2017; 216: 65-71Crossref PubMed Scopus (79) Google Scholar; Scheeder et al., 2018Scheeder C. Heigwer F. Boutros M. Machine learning and image-based profiling in drug discovery.Curr. Opin. Syst. Biol. 2018; 10: 43-52Crossref PubMed Scopus (70) Google Scholar). We lay emphasis on the Cell Painting Assay (CPA) (Bray et al., 2016Bray M.A. Singh S. Han H. Davis C.T. Borgeson B. Hartland C. Kost-Alimova M. Gustafsdottir S.M. Gibson C.C. Carpenter A.E. Cell Painting, a high-content image-based assay for morphological profiling using multiplexed fluorescent dyes.Nat. Protoc. 2016; 11: 1757-1774Crossref PubMed Scopus (211) Google Scholar; Gustafsdottir et al., 2013Gustafsdottir S.M. Ljosa V. Sokolnicki K.L. Anthony Wilson J. Walpita D. Kemp M.M. Petri Seiler K. Carrel H.A. Golub T.R. Schreiber S.L. et al.Multiplex cytological profiling assay to measure diverse cellular states.PLoS One. 2013; 8: e80999Crossref PubMed Scopus (105) Google Scholar), as it does not require genetic manipulation, can be used with a variety of cell lines, and has found broad acceptance and application in the chemical biology community and increasingly in the pharmaceutical industry. Automated detection of the major cellular compartments (endoplasmic reticulum [ER], Golgi, mitochondria, lysosomes, endosomes, the actin and tubulin cytoskeleton, nucleoli, and nucleus) was first reported by Boland and Murphy, 2001Boland M.V. Murphy R.F. A neural network classifier capable of recognizing the patterns of all major subcellular structures in fluorescence microscope images of HeLa cells.Bioinformatics. 2001; 17: 1213-1223Crossref PubMed Scopus (375) Google Scholar) in HeLa cells. Upon staining of cells for one of the compartments and for DNA, 84 features per image were extracted which, for example, monitored image texture, patterns, object distance measures with respect to the cell center, object size, stain overlap with the nucleus, and others. The DNA pattern was employed as a standardization between cells as it is consistent among cells, and protein localization patterns were referred to the DNA stain as a common landmark. This pioneering work paved the way for the high-content analysis and image-based profiling techniques for small molecules that differ in the number and type of detected cellular components and the type of visualization, which is addressed in the following sections. The Morphobase, a cell morphology database that employs DNA staining and bright-field images of srcts-NRK and HeLa cells upon compound treatment (Futamura et al., 2012Futamura Y. Kawatani M. Kazami S. Tanaka K. Muroi M. Shimizu T. Tomita K. Watanabe N. Osada H. Morphobase, an encyclopedic cell morphology database, and its use for drug target identification.Chem. Biol. 2012; 19: 1620-1630Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar), detected clusters for several target classes such as tubulin, actin, DNA synthesis, histone deacetylase (HDAC), and heat-shock proteins (HSPs) (Table S1). Importantly, the clusters of HDAC and proteasome inhibitors could not be distinguished in srcts-NRK, whereas RNA and protein synthesis inhibitors could not be separated in HeLa cells. However, when the perturbation profiles in both cell lines were considered, the activities could be clearly differentiated. The Morphobase linked the mitochondrial complex I inhibitor rotenone to tubulin, the cyclin-dependent kinase (CDK) inhibitor 3-ATA, and the polypharmacological compound resveratrol to DNA synthesis. As the observed activities had been previously associated with these compounds, these findings emphasize the need of complete annotation for reference compounds as mostly only the nominal target, i.e., the target most commonly associated with the compound (Moret et al., 2019Moret N. Clark N.A. Hafner M. Wang Y. Lounkine E. Medvedovic M. Wang J. Gray N. Jenkins J. Sorger P.K. Cheminformatics tools for analyzing and designing optimized small-molecule collections and libraries.Cell Chem. Biol. 2019; 26: 765-777Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar), is annotated in public databases or vendors’ websites. Further studies using the Morphobase identified inhibitors of tubulin and the proteasome also in combination with proteome profiling using the ChemProteobase (Futamura et al., 2012Futamura Y. Kawatani M. Kazami S. Tanaka K. Muroi M. Shimizu T. Tomita K. Watanabe N. Osada H. Morphobase, an encyclopedic cell morphology database, and its use for drug target identification.Chem. Biol. 2012; 19: 1620-1630Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar, Futamura et al., 2013Futamura Y. Kawatani M. Muroi M. Aono H. Nogawa T. Osada H. Identification of a molecular target of a novel fungal metabolite, pyrrolizilactone, by phenotypic profiling systems.Chembiochem. 2013; 14: 2456-2463Crossref PubMed Scopus (23) Google Scholar; Minegishi et al., 2015Minegishi H. Futamura Y. Fukashiro S. Muroi M. Kawatani M. Osada H. Nakamura H. Methyl 3-((6-methoxy-1,4-dihydroindeno[1,2-c]pyrazol-3-yl)amino)benzoate (GN39482) as a tubulin polymerization inhibitor identified by MorphoBase and ChemProteoBase profiling methods.J. Med. Chem. 2015; 58: 4230-4241Crossref Scopus (28) Google Scholar). Markers for DNA synthesis (5-ethynyl-3-deoxyuridine [EdU]) and mitosis (phospho-histone H3) along with a DNA stain were employed to profile compounds in HeLa cells (Young et al., 2008Young D.W. Bender A. Hoyt J. McWhinnie E. Chirn G.W. Tao C.Y. Tallarico J.A. Labow M. Jenkins J.L. Mitchison T.J. et al.Integrating high-content screening and ligand-target prediction to identify mechanism of action.Nat. Chem. Biol. 2008; 4: 59-68Crossref PubMed Scopus (268) Google Scholar) (Table S1). The most information-rich characteristics stemmed from the nuclear stain, i.e., size of the nucleus and DNA quantity. Profile analysis revealed bioactivity clusters around the vacuolar ATPase, of antimitotics, corticosteroids, and progesterone-related compounds. Interestingly, a cell-cycle-arrest-related cluster featured compounds with different MoAs such as cardiac glycosides, which target Na+/K+-ATPases, the protein translation inhibitors emetine and cycloheximide, and steroid hormones such as progesterone and danatrol. The authors compared the predictive nature of the phenotypic profiling with target prediction based on structural similarity. Overall, phenotypes were better correlated with the known or predicted targets than with compound structures. In addition to visualizing DNA, microtubules and Golgi stainings were employed for compound profiling in five different cell lines (SKOV3, A549, SF268, DU145, and HUVEC [human umbilical vein endothelial cell], see Table S1) (Adams et al., 2006Adams C.L. Kutsyy V. Coleman D.A. Cong G. Crompton A.M. Elias K.A. Oestreicher D.R. Trautman J.K. Vaisberg E. Compound classification using image-based cellular phenotypes.Methods Enzymol. 2006; 14: 440-468Crossref Scopus (42) Google Scholar; Tanaka et al., 2005Tanaka M. Bateman R. Rauh D. Vaisberg E. Ramachandani S. Zhang C. Hansen K.C. Burlingame A.L. Trautman J.K. Shokat K.M. et al.An unbiased cell morphology-based screen for new, biologically active small molecules.PLoS Biol. 2005; 3: e128https://doi.org/10.1371/journal.pbio.0030128Crossref PubMed Scopus (197) Google Scholar). Subpopulation analysis was performed based on the nuclear stain to group cells in populations depending on the cell-cycle phase by employing the information on DNA content, morphology, and condensation. Different profiles were observed for two structurally related compounds, the Src inhibitor PP2 and hydroxy-PP, in A549 cells and HUVECs, whereas the effects of both compounds on DU145, SF268, and SKOV3 cells were similar. Like PP2, hydroxy-PP inhibited the kinases Fyn, p38-α, and protein kinases A and B, which may explain the profile similarity in DU145, SF268, and SKOV3 cells. The lack of similarity in A549 cells and HUVECs may stem from different expression levels of targeted proteins. Thus, a cell line set that represents diverse tissues and genetic backgrounds may be beneficial for the detection of subtle morphological changes. Carragher's group employed DNA-, actin-, and tubulin-staining in four cell lines (Ovcar3, MiaPaCa2, MCF7, and MCF7 with truncated, dominant negative TP53 mutant) (Caie et al., 2010Caie P.D. Walls R.E. Ingleston-Orme A. Daya S. Houslay T. Eagle R. Roberts M.E. Carragher N.O. High-content phenotypic profiling of drug response signatures across distinct cancer cells.Mol. Cancer Ther. 2010; 9: 1913-1926Crossref PubMed Scopus (94) Google Scholar). They even defined subpopulations based on cell-shape descriptors and distinguished an epithelial from a mesenchymal phenotype in MCF7 cells. The assay differentiated clusters of inhibitors of protein synthesis, proteases, DNA synthesis, and tubulin or actin modulators, as well as Aurora kinase and Eg5 inhibitors (Table S1) (Caie et al., 2010Caie P.D. Walls R.E. Ingleston-Orme A. Daya S. Houslay T. Eagle R. Roberts M.E. Carragher N.O. High-content phenotypic profiling of drug response signatures across distinct cancer cells.Mol. Cancer Ther. 2010; 9: 1913-1926Crossref PubMed Scopus (94) Google Scholar; Ljosa et al., 2013Ljosa V. Caie P.D. Ter Horst R. Sokolnicki K.L. Jenkins E.L. Daya S. Roberts M.E. Jones T.R. Singh S. Genovesio A. et al.Comparison of methods for image-based profiling of cellular morphological responses to small-molecule treatment.J. Biomol. Screen. 2013; 18: 1321-1329Crossref PubMed Scopus (84) Google Scholar). Of note, tubulin-targeting compounds were separated in stabilizing (Taxol, epothilone B) and destabilizing (colchicine, nocodazole) agents. Further analysis revealed resistance and selectivity across the cell-line panel for some compounds, while other compounds induced similar phenotypes in all four cell lines. Overall, Ovcar3 cells were more resistant to several compound classes, whereas MiaPaCa2 cells were often most sensitive to compound perturbations (Caie et al., 2010Caie P.D. Walls R.E. Ingleston-Orme A. Daya S. Houslay T. Eagle R. Roberts M.E. Carragher N.O. High-content phenotypic profiling of drug response signatures across distinct cancer cells.Mol. Cancer Ther. 2010; 9: 1913-1926Crossref PubMed Scopus (94) Google Scholar). DNA, tubulin, and phospho-histone H2A.X were used to classify compound activity in HeLa cells (Twarog et al., 2016Twarog N.R. Low J.A. Currier D.G. Miller G. Chen T. Shelat A.A. Robust classification of small-molecule mechanism of action using a minimalist high-content microscopy screen and multidimensional phenotypic trajectory analysis.PLoS One. 2016; 11: e0149439https://doi.org/10.1371/journal.pone.0149439Crossref PubMed Scopus (9) Google Scholar). This approach identified clusters related to Aurora kinase, tubulin, proteasome, topoisomerases, HDAC, antimetabolites, and HSP90. However, mammalian target of rapamycin (mTOR) and phosphatidylinositol 3-kinase (PI3K) inhibitors were not classified due to low activity in this minimalist assay, in contrast to approaches that employ additional staining (Reisen et al., 2015Reisen F. Sauty de Chalon A. Pfeifer M. Zhang X. Gabriel D. Selzer P. Linking phenotypes and modes of action through high-content screen fingerprints.Assay Drug Dev. Technol. 2015; 13: 415-427Crossref PubMed Scopus (40) Google Scholar). Genotype-dependent profiling was performed upon DNA and actin staining in 12 isogenic cell lines based on the parental cell line HCT116 (Breinig et al., 2015Breinig M. Klein F.A. Huber W. Boutros M. A chemical-genetic interaction map of small molecules using high-throughput imaging in cancer cells.Mol. Syst. Biol. 2015; 11: 846https://doi.org/10.15252/msb.20156400Crossref PubMed Scopus (43) Google Scholar). The cell lines harbored deletions in oncogenic mutations or knockouts (see Table S1). Both genotype-dependent and genotype-independent phenotypes were observed in compound-treated cells, and the number of drug-gene interactions varied across the cell-line panel. Drug-gene-phenotype interactions were associated with 15% of the compounds and provided information on crosstalk of signaling pathways and potential drug synergy. The analysis clustered compounds that target tubulin, MEK, p38, glucocorticoid receptor, DNA alkylation, or mitochondrial proton gradient. Clusters of connected biological processes were observed for an iron chelator, antifolates, and DNA methyltransferase inhibition. Uncoupling of the mitochondrial proton gradient was observed as unanticipated bioactivity for the protein kinase Cδ (PKCδ) inhibitor rottlerin (Figure 2A), which already had been reported in 2001 (Soltoff, 2001Soltoff S.P. Rottlerin is a mitochondrial uncoupler that decreases cellular ATP levels and indirectly blocks protein kinase C delta tyrosine phosphorylation.J. Biol. Chem. 2001; 276: 37986-37992Abstract Full Text Full Text PDF PubMed Google Scholar). In addition, proteasome inhibition was predicted by the chemical-genetic matrix for the aldehyde dehydrogenase (ALDH) inhibitor disulfiram, endothelial growth factor receptor (EGFR) inhibitor tyrphostin AG555, and the nuclear factor κB (NF-κB) inhibitor caffeic acid phenyl ester (CAPE), which was experimentally validated (see Figure 2B). A more physiological setup for morphological profiling employed staining of DNA, actin, and membrane integrity to profile compounds in patient-derived organoids (PDOs) from 19 patients with colorectal cancer at different clinical stages (Betge et al., 2019Betge J. Rindtorff N. Sauer J. Rauscher B. Dingert C. Gaitantzi H. Herweck F. Miersch T. Valentini E. Hauber V. et al.Multiparametric phenotyping of compound effects on patient derived organoids.bioRxiv. 2019; : 660993https://doi.org/10.1101/660993Crossref Scopus (0) Google Scholar). Analysis of compound-induced phenotypes revealed clusters of MEK, PI3K/AKT/mTOR, glycogen synthase kinase 3 (GSK3), EGFR, and CDK inhibitors (Table S1). Whereas phenotypic activity of MEK and CDK inhibitors was observed in all PDO lines, AKT and GSK3 activity clusters were detected only in a subset of PDOs. In addition, morphological responses provided mechanistic insights for MEK and GSK3 inhibitor classes and linked phenotypes to cancer mutations. The use of multiple staining sets enables detection of various cellular components in parallel processed samples and is thus not limited by the number of spectrally separable dyes (Table S1). A seminal work on morphological profiling of small molecules by the Altschuler and Wu groups explored 100 drugs mostly with known MoA in HeLa cells using a DNA stain and five different staining sets: (1) splicing factor SC35 and the cytokinesis protein anillin; (2) β-tubulin and actin; (3) phospho-p38 and phospho-extracellular signal-regulated kinase (ERK); (4) p53, c-Fos; and (5) phosphoadenosine 3,5-monophosphate response element-binding protein (CREB) and calmodulin (Perlman et al., 2004Perlman Z.E. Slack M.D. Feng Y. Mitchison T.J. Wu L.F. Altschuler S.J. Multidimensional drug profiling by automated microscopy.Science. 2004; 306: 1194-1198Crossref PubMed Scopus (513) Google Scholar). For compounds that provoked strong responses, some descriptors changed differently at the different concentrations, implying differences of high and low concentrations in target engagement or interaction with several targets with different affinities. For example, camptothecin inhibits at low concentration topoisomerase I and causes an S-phase arrest, whereas at high concentration it inhibits transcription along with further cellular processes. The authors predicted the MoA of the poorly characterized compound austocystin that clustered with transcription and translation inhibitors. Already this first report made some essential observations: (1) compounds with common targets displayed similar profiles irrespective" @default.
• W3136205387 created "2021-03-29" @default.
• W3136205387 creator A5013934686 @default.
• W3136205387 creator A5048674803 @default.
• W3136205387 creator A5069927005 @default.
• W3136205387 date "2021-03-01" @default.
• W3136205387 modified "2023-09-30" @default.
• W3136205387 title "Morphological profiling of small molecules" @default.
• W3136205387 cites W1147709937 @default.
• W3136205387 cites W1185588434 @default.
• W3136205387 cites W1522234506 @default.
• W3136205387 cites W1591129275 @default.
• W3136205387 cites W1752612204 @default.
• W3136205387 cites W1963836342 @default.
• W3136205387 cites W1964915719 @default.
• W3136205387 cites W1967660686 @default.
• W3136205387 cites W1993669821 @default.
• W3136205387 cites W1998202992 @default.
• W3136205387 cites W2035164474 @default.
• W3136205387 cites W2058711722 @default.
• W3136205387 cites W2076317736 @default.
• W3136205387 cites W2082969252 @default.
• W3136205387 cites W2104036010 @default.
• W3136205387 cites W2107554012 @default.
• W3136205387 cites W2113221184 @default.
• W3136205387 cites W2118812289 @default.
• W3136205387 cites W2121604817 @default.
• W3136205387 cites W2123059506 @default.
• W3136205387 cites W2133422502 @default.
• W3136205387 cites W2138512826 @default.
• W3136205387 cites W2144963196 @default.
• W3136205387 cites W2145594780 @default.
• W3136205387 cites W2145752049 @default.
• W3136205387 cites W2147443169 @default.
• W3136205387 cites W2160676466 @default.
• W3136205387 cites W2162011385 @default.
• W3136205387 cites W2173816878 @default.
• W3136205387 cites W2175970000 @default.
• W3136205387 cites W2183389629 @default.
• W3136205387 cites W2211797036 @default.
• W3136205387 cites W2220553089 @default.
• W3136205387 cites W2272856212 @default.
• W3136205387 cites W2276949837 @default.
• W3136205387 cites W2406928327 @default.
• W3136205387 cites W2409683822 @default.
• W3136205387 cites W2472153835 @default.
• W3136205387 cites W2480007569 @default.
• W3136205387 cites W2507552372 @default.
• W3136205387 cites W2509141893 @default.
• W3136205387 cites W2560056529 @default.
• W3136205387 cites W2560220895 @default.
• W3136205387 cites W2560741024 @default.
• W3136205387 cites W2586869611 @default.
• W3136205387 cites W2594024619 @default.
• W3136205387 cites W2596519089 @default.
• W3136205387 cites W2601884717 @default.
• W3136205387 cites W2612467560 @default.
• W3136205387 cites W2612655602 @default.
• W3136205387 cites W2627654851 @default.
• W3136205387 cites W2751625431 @default.
• W3136205387 cites W2766601509 @default.
• W3136205387 cites W2803317672 @default.
• W3136205387 cites W2888318454 @default.
• W3136205387 cites W2888436565 @default.
• W3136205387 cites W2902713094 @default.
• W3136205387 cites W2903435531 @default.
• W3136205387 cites W2912235843 @default.
• W3136205387 cites W2922305141 @default.
• W3136205387 cites W2922760757 @default.
• W3136205387 cites W2924013657 @default.
• W3136205387 cites W2934863540 @default.
• W3136205387 cites W2951276412 @default.
• W3136205387 cites W2951576022 @default.
• W3136205387 cites W2952361150 @default.
• W3136205387 cites W2955182975 @default.
• W3136205387 cites W2963713805 @default.
• W3136205387 cites W2973052723 @default.
• W3136205387 cites W2983562173 @default.
• W3136205387 cites W2990166790 @default.
• W3136205387 cites W2998120965 @default.
• W3136205387 cites W3010841569 @default.
• W3136205387 cites W3013805333 @default.
• W3136205387 cites W3026775944 @default.
• W3136205387 cites W3034781784 @default.
• W3136205387 cites W3036278624 @default.
• W3136205387 cites W3040220019 @default.
• W3136205387 cites W3047346587 @default.
• W3136205387 cites W3089034220 @default.
• W3136205387 cites W4254733843 @default.
• W3136205387 doi "https://doi.org/10.1016/j.chembiol.2021.02.012" @default.
• W3136205387 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/33740434" @default.
• W3136205387 hasPublicationYear "2021" @default.
• W3136205387 type Work @default.
• W3136205387 sameAs 3136205387 @default.
• W3136205387 citedByCount "27" @default.
• W3136205387 countsByYear W31362053872021 @default.
• W3136205387 countsByYear W31362053872022 @default.
• W3136205387 countsByYear W31362053872023 @default.

• next