Improving drug discovery using image-based multiparametric analysis of the epigenetic landscape

With the advent of automatic cell imaging and machine learning, high-content phenotypic screening has become the approach of choice for drug discovery due to its ability to extract drug-specific multi-layered data, which could be compared to known profiles. In the field of epigenetics, such screening methods have suffered from a lack of tools sensitive to selective epigenetic perturbations. Here we describe a novel approach, Microscopic Imaging of Epigenetic Landscapes (MIEL), which captures the nuclear staining patterns of epigenetic marks (e.g., acetylated and methylated histones) and employs machine learning to accurately distinguish between such patterns. We validated the fidelity and robustness of the MIEL platform across multiple cells lines and using dose-response curves, to insure the robustness of this approach for high content high throughput drug discovery. Focusing on alternative, non-cytotoxic, glioblastoma treatments, we demonstrated that the MIEL assay can identify epigenetically active drugs and classify them by molecular function. Furthermore, we show MIEL was able to accurately rank candidate drugs by their ability to produce a set of desired epigenetic alterations consistent with increased sensitivity to chemotherapeutic agents or with induction of glioblastoma differentiation. Glioblastoma tumor propagating cells untreated or treated with either fetal bovine serum (10%) or Bmp4 (10ng/ml); 3 replicates each. Glioblastoma tumor propagating cells treated with DMSO or one of the following compounds: Digoxin, Digitoxigenin, Fenbendazole, Mebendazole, Etoposide, Irinoteca, Cytarabine, Trifluridine, SAHA, TSA, Valproic-Acid; one replicate each.