Grantee Research Project Results

2024 Progress Report: Skeletal Teratogenicity of Industrial and Environmental Chemicals Predicted with Human Pluripotent Stem Cells in Vitro

EPA Grant Number: R839502
Title: Skeletal Teratogenicity of Industrial and Environmental Chemicals Predicted with Human Pluripotent Stem Cells in Vitro
Investigators: zur Nieden, Nicole I , Volz, David C.
Institution: University of California - Riverside
EPA Project Officer: Packard, Benjamin H
Project Period: August 1, 2019 through July 31, 2021 (Extended to July 31, 2025)
Project Period Covered by this Report: August 1, 2023 through July 31,2024
Project Amount: $849,811
RFA: Advancing Actionable Alternatives to Vertebrate Animal Testing for Chemical Safety Assessment (2018) RFA Text | Recipients Lists
Research Category: Chemical Safety for Sustainability

Objective:

We hypothesize that cultures of human embryonic stem cells (hESCs) can be utilized to predict the skeletal embryotoxicity of industrial and environmental chemicals in vitro. The objectives of this study are to differentiate hESCs into bone-forming osteoblasts with concomitant chemical exposure (the test bank includes selected chemicals from the ToxCast I library with known effects on the skeleton). Detrimental effects of the chemical are evaluated based on a reduction in matrix mineralization (as a surrogate of skeletal maturation). To identify suitable assays for matrix mineralization that adequately predict human risk, we compare inexpensive absorbance-based assays and image analysis tools recently developed by us. Hierarchies of chemical potency and efficacy will be established by setting the half-maximal inhibitory doses of differentiation inhibition in relation to the half-maximal cytotoxic dose using a biostatistical model and canonical linear discriminants to create a skeletal embryotoxicity score. A case study will determine whether the scoring system is sufficiently sensitive to distinguish between closely related chemical derivatives and will consequently establish whether the assay can be used to identify safer chemical alternatives. Lastly, testing whether the chemicals affect neural crest or mesodermal osteoblasts coupled with next-generation sequencing will identify potential modes-of-action (disruption of organogenesis) and mechanisms of toxicity (molecularlevel initiating event).

Progress Summary:

At the end of this reporting period, we have successfully tested four reference chemicals, 18 training set chemicals and 20 test set chemicals. To develop a predictive in vitro assay, we have obtained concentration-response curves for three endpoints: 1) the osteogenic differentiation efficiency (measured via a calcium assay to test for the matrix calcification specific to bone tissue), 2) the cytotoxicity elicited by the chemical (via an MTT assay) – both using embryonic stem cells and 3) the cytotoxicity elicited by the same test chemical, but on fully differentiated fibroblast cells. We tested all endpoints in human as well as mouse cells.

Using a previously established biostatistical prediction model, we determined the predictivity of the human cell-based assay to be 89.5%. Using linear discriminant analysis, the predictivity of the human cell-based assay increased to 97.4%. However, when using mouse cells, the predictivity dropped to 63.1% overall (biostatistical model), but increased to 84.2% when we applied linear discriminant functions. Thus, our overall results so far suggest that the human assay seems to much better suited to predict risk for human exposure than the mouse assay.

Next, we have applied this assay to the prediction of skeletal embryotoxicity associated with exposure to 10 TSCA priority chemicals including six phthalates and identified di-isobutyl- and butylbenzyl phthalate as weakly embryotoxic, whereas dibuthyl phthalate and di-2-ethylhexyl phthalate exerted strongly embrytoxic effects in osteogenically differentiating human embryonic stem cells. Moreover, 1,4 dioxane, bromopropane and N-methylpyrrolidone were classified as weakly embryotoxic to developing skeletal cells.

With the linear discriminant analysis in place, we also finalized the screening of thiourea and six of its derivatives to determine whether the hESTo developed here can distinguish between derivatives with differential toxicity. Indeed, the canonical plot clustering allowed for the grading of the severity of the toxic effects associated with thiourea exposure. The parent compound as well as 4 additional thiourea derivatives (1,3-dibutylthiourea, 1,3-diphenyl thiourea, trimethylthiourea, and 1,3-ethylene thiourea) were classified as weakly embryotoxic with the latter being the most toxic.

Future Activities:

We plan to continue to define a novel alternative prediction model using linear discriminant functions and integrate the fibroblast data in doing so. We will utilize a image analysis pipeline to measure toxicity from images using a set of selected chemicals. We will also use the additional time to interrogate RNA sequencing data for selected chemicals to identify a toxicity gene signature, the potential target cells and potential modes of toxicity action.

Journal Articles:

No journal articles submitted with this report: View all 8 publications for this project

Supplemental Keywords:

Exposure, risk assessment, sensitive populations, dose-response, teratogen, develop-mental toxicity, pluripotent stem cells, adverse outcome pathway, imaging, chemical derivatives

Progress and Final Reports:

Original Abstract

The perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.