You are here:
Benchmark Dose (BMD) Modeling of Image-based Phenotypic Profiling Data Yields More Potent Estimates of Chemical Bioactivity Compared to Cell Viability and Apoptosis Assays
Citation:
Willis, C., J. Nyffeler, AND J. Harrill. Benchmark Dose (BMD) Modeling of Image-based Phenotypic Profiling Data Yields More Potent Estimates of Chemical Bioactivity Compared to Cell Viability and Apoptosis Assays. Presented at Society of Toxicology annual meeting, Baltimore, MD, March 10 - 14, 2019. https://doi.org/10.23645/epacomptox.7874132
Impact/Purpose:
Poster for presentation at Society of Toxicology annual meeting March 2019. Here we describe workflows for concentration-response screening and image analysis using an HTPP assay that quantitatively evaluates changes in organelle morphology (i.e. Cell Painting), as well as calculation of biological pathway altering concentrations (BPACs) using high-throughput concentration-response modeling.
Description:
High-throughput imaging-based phenotypic profiling (HTPP) is a chemical screening method that combines automated microscopy and image analysis to measure a large variety of morphological features at the single cell level. Here we describe workflows for concentration-response screening and image analysis using an HTPP assay that quantitatively evaluates changes in organelle morphology (i.e. Cell Painting), as well as calculation of biological pathway altering concentrations (BPACs) using high-throughput concentration-response modeling (BMDExpress 2.2). A set of 16 reference chemicals were tested in six human cell lines (U-2 OS, MCF7, HTP-9, A549, ARPE-19, HepG2). Cells were plated in 384-well plates and after 24 h treated with 7 concentrations (semi-log spacing, n = 3/plate, 3 cultures) in a randomized pattern. After 48 h, cells were live labeled with MitoTracker (mitochondria), fixed, permeabilized and labeled with Hoechst-33342 (nuclei), SYTO14 (nucleoli) and fluorescent conjugates of concanavalin A (ER), phalloidin (cytoskeleton), and wheat germ agglutinin (Golgi/plasma membrane). A multiplexed cell viability (CV) and apoptosis (AP) assay was run in parallel. Confocal images were acquired using an Opera Phenix HCS system and analyzed using Harmony software, yielding ~1200 features per cell. Cell-level data were median absolute deviation (MAD) normalized to DMSO controls. BMD modeling was performed on well-level median values. Most chemicals (n=14) affected cell morphology in a concentration-dependent manner. Distinct patterns of affected cellular features were observed across the chemical set and, in most cases, were consistent with observations from the literature. In general, the chemicals produced similar patterns, with highly correlated potency estimates, across the six different cell lines. For all compounds, HTPP BMDs were at least as sensitive as CV or AP BMDs. In some cell lines, profiling BMDs were > 10x lower that CV or AP BMDs. Screening of a larger set of chemicals (n=480) also demonstrated marked differences in HTPP and CV or AP potency estimates. In summary, testing of diverse compounds yielded distinct patterns of affected features below the threshold for cytotoxicity, indicating that this profiling method could be used to derive in vitro potency estimates for screening level risk assessments. This abstract does not necessarily reflect USEPA policy.
URLs/Downloads:
DOI: Benchmark Dose (BMD) Modeling of Image-based Phenotypic Profiling Data Yields More Potent Estimates of Chemical Bioactivity Compared to Cell Viability and Apoptosis Assays
SOT 2019 POSTER WILLIS_V5.PDF (PDF, NA pp, 3455.262 KB, about PDF)