1
|
Krentzel D, Shorte SL, Zimmer C. Deep learning in image-based phenotypic drug discovery. Trends Cell Biol 2023:S0962-8924(22)00262-8. [PMID: 36623998 DOI: 10.1016/j.tcb.2022.11.011] [Citation(s) in RCA: 8] [Impact Index Per Article: 8.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/25/2022] [Revised: 11/26/2022] [Accepted: 11/29/2022] [Indexed: 01/08/2023]
Abstract
Modern drug discovery approaches often use high-content imaging to systematically study the effect on cells of large libraries of chemical compounds. By automatically screening thousands or millions of images to identify specific drug-induced cellular phenotypes, for example, altered cellular morphology, these approaches can reveal 'hit' compounds offering therapeutic promise. In the past few years, artificial intelligence (AI) methods based on deep learning (DL) [a family of machine learning (ML) techniques] have disrupted virtually all image analysis tasks, from image classification to segmentation. These powerful methods also promise to impact drug discovery by accelerating the identification of effective drugs and their modes of action. In this review, we highlight applications and adaptations of ML, especially DL methods for cell-based phenotypic drug discovery (PDD).
Collapse
Affiliation(s)
- Daniel Krentzel
- Institut Pasteur, Université Paris Cité, Imaging and Modeling Unit, F-75015 Paris, France; Institut Pasteur, Joint International Unit Artificial Intelligence for Image-based Drug Discovery & Development (PIU-Ai3D), F-75015 Paris, France.
| | - Spencer L Shorte
- Institut Pasteur, Joint International Unit Artificial Intelligence for Image-based Drug Discovery & Development (PIU-Ai3D), F-75015 Paris, France; Institut Pasteur, Université Paris Cité, Photonic Bio-Imaging, Centre de Ressources et Recherches Technologiques (UTechS-PBI, C2RT), F-75015 Paris, France
| | - Christophe Zimmer
- Institut Pasteur, Université Paris Cité, Imaging and Modeling Unit, F-75015 Paris, France; Institut Pasteur, Joint International Unit Artificial Intelligence for Image-based Drug Discovery & Development (PIU-Ai3D), F-75015 Paris, France.
| |
Collapse
|
2
|
Andersson CR, Selvin T, Blom K, Rubin J, Berglund M, Jarvius M, Lenhammar L, Parrow V, Loskog A, Fryknäs M, Nygren P, Larsson R. Mebendazole is unique among tubulin-active drugs in activating the MEK-ERK pathway. Sci Rep 2020; 10:13124. [PMID: 32753665 PMCID: PMC7403428 DOI: 10.1038/s41598-020-68986-0] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/03/2019] [Accepted: 06/19/2020] [Indexed: 11/09/2022] Open
Abstract
We recently showed that the anti-helminthic compound mebendazole (MBZ) has immunomodulating activity in monocyte/macrophage models and induces ERK signalling. In the present study we investigated whether MBZ induced ERK activation is shared by other tubulin binding agents (TBAs) and if it is observable also in other human cell types. Curated gene signatures for a panel of TBAs in the LINCS Connectivity Map (CMap) database showed a unique strong negative correlation of MBZ with MEK/ERK inhibitors indicating ERK activation also in non-haematological cell lines. L1000 gene expression signatures for MBZ treated THP-1 monocytes also connected negatively to MEK inhibitors. MEK/ERK phosphoprotein activity testing of a number of TBAs showed that only MBZ increased the activity in both THP-1 monocytes and PMA differentiated macrophages. Distal effects on ERK phosphorylation of the substrate P90RSK and release of IL1B followed the same pattern. The effect of MBZ on MEK/ERK phosphorylation was inhibited by RAF/MEK/ERK inhibitors in THP-1 models, CD3/IL2 stimulated PBMCs and a MAPK reporter HEK-293 cell line. MBZ was also shown to increase ERK activity in CD4+ T-cells from lupus patients with known defective ERK signalling. Given these mechanistic features MBZ is suggested suitable for treatment of diseases characterized by defective ERK signalling, notably difficult to treat autoimmune diseases.
Collapse
Affiliation(s)
- Claes R Andersson
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden.
| | - Tove Selvin
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden
| | - Kristin Blom
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden
| | - Jenny Rubin
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden
| | - Malin Berglund
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden
| | - Malin Jarvius
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden
| | - Lena Lenhammar
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden
| | - Vendela Parrow
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden
| | - Angelica Loskog
- Department of Immunology, Genetics and Pathology, Section of Oncology, Uppsala University, 75185, Uppsala, Sweden
| | - Mårten Fryknäs
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden
| | - Peter Nygren
- Department of Immunology, Genetics and Pathology, Section of Oncology, Uppsala University, 75185, Uppsala, Sweden
| | - Rolf Larsson
- Division of Cancer Pharmacology and Computational Medicine, Department of Medical Sciences, Uppsala University, 75185, Uppsala, Sweden.
| |
Collapse
|