You are here:
Towards a High-Throughput New Approach Method for Acute Fish Toxicity: Painting the Rainbow (Trout)
Citation:
Nyffeler, J., F. Harris, S. Lasee, B. Blackwell, J. Haselman, J. Nichols, G. Patlewicz, Dan Villeneuve, AND J. Harrill. Towards a High-Throughput New Approach Method for Acute Fish Toxicity: Painting the Rainbow (Trout). Society of Environmental Toxicology and Chemistry SETAC Europe 33rd Annual Meeting 2023, Dublin, N/A, IRELAND, April 30 - May 04, 2023. https://doi.org/10.23645/epacomptox.22798634
Impact/Purpose:
Poster presented to the Society of Environmental Toxicology and Chemistry SETAC Europe 33rd Annual Meeting May 2023
Description:
An in vitro method (OECD TG 249) using rainbow trout (Oncorhynchus mykiss) gill cells (RTgill-W1) in a plate-reader based viability assay (CV-PR) has been developed as a potential alternative to acute fish toxicity tests. We miniaturized this existing New Approach Method (NAM) from 24-well to 384-well format and added two previously established image-based assays: image-based cell viability (CV-IB) and high-throughput phenotypic profiling (HTPP, ‘Cell Painting’). We then tested 233 unique chemicals with the aim to compare the results from this NAM to existing in vitro and in vivo data. Cells were seeded at 22,500 cells/well and exposed to the test chemicals for 24 h in a minimal medium. Chemicals were tested at 8 concentrations (0.05 – 100 µM) in four independent cultures. The 3 assays resulted in 6 endpoints. The lowest potency value was considered as the in vitro potency. Existing in vitro data from RTgill-W1 cells were collected for 51 chemicals. The fraction of chemical freely available in the culture wells was estimated using 3 different models (Armitage 2019, Armitage 2021, Fischer 2017). In vivo biotransformation was accounted for with the Arnot 2008 model. Existing in vivo mortality data for rainbow trout and fathead minnow (Pimephales promelas) was gathered from the ECOTOX Knowledgebase (~11,000 records). Only 26/51 chemicals were active in literature in vitro studies and in the present study, with 14 and 24 chemicals having potency estimates within one and two orders of magnitude, respectively. The overall root mean squared error (RMSE) was 1.2 (on a log10 mg/L scale). Overall, 78 chemicals had all the required information for modelling and at least three ECOTOX records. Of those, 45 were active in our NAM. The Armitage 2021 model resulted in the best prediction relative to the existing in vivo effects data, with a RMSE of 1.4, and 24 and 39 chemicals had potency estimates within one and two orders of magnitude of the in vivo value, respectively. By miniaturizing the existing assay, we were able to generate in vitro data for >200 chemicals, compared to data for ~100 chemicals obtained previously. Ongoing efforts include the analytical measurement of the free chemical concentration for 12 exemplary chemicals to better understand which in vitro disposition model is the most accurate compared to empirical measurements. In the future, we aim to leverage the HTPP data to identify putative modes-of-action. This work does not reflect US EPA policy.
URLs/Downloads:
DOI: Towards a High-Throughput New Approach Method for Acute Fish Toxicity: Painting the Rainbow (Trout)
NYFFELER_SETAC2023_POSTER_V3.PDF (PDF, NA pp, 990.218 KB, about PDF)