Ozone responsive gene expression as a model for describing repeat exposure response trajectories and inter-individual toxicodynamic variability in vitro

Inhaled chemical/material exposures are a ubiquitous part of daily life around the world. There is a need to evaluate potential adverse effects of both single and repeat exposures for thousands of chemicals and an exponentially larger number of exposure scenarios (e.g., repeated exposures).  Meeting this challenge will require the development and use of in vitro new approach methodologies (NAMs); however, two major challenges face the deployment of NAMs in risk assessment are (1) characterizing what apical outcome(s) acute assays inform regarding the trajectory to long term events, especially under repeated exposure conditions, and (2) capturing inter-individual variability as it informs considerations of potentially susceptible and/or vulnerable populations.  To address these questions, we used a primary human bronchial epithelial cell air-liquid interface model exposed to ozone (O3), a model oxidant and ubiquitous environmental chemical.  Here we report that the induction of the O3-responsive pro-inflammatory genes IL-8, IL-6, HMOX1, and COX2 is attenuated in repeated exposures, despite similar effects of single and repeated exposures on reductions in epithelial barrier integrity.  Thus, demonstrating that single acute exposure outcomes do not reliably represent the trajectory of responses after repeated or chronic exposures and that differences between single and repeated exposure outcomes in general functional endpoints (i.e., TEER) are not necessarily reflective of all exposure-responsive and adverse outcome relevant endpoints.  Further, we observed 10.1-, 10.3-, 14.2-, and 7-fold ranges of induction of IL-8, IL-6, HMOX1, and COX2 transcripts, respectively, within in our population of 25 unique donors.  Calculation of sample size estimates that indicated that 27, 24, 299, and 13 donors would be required to significantly power similar in vitro studies to identify a two-fold change in IL-8, IL-6, HMOX1, and COX2 transcript induction, respectively, to inform considerations of the uncertainty factors to reflect variability within the human population (UFH) for in vitro studies.  Overall, this study identified key considerations that need to be addressed to support the successful integration and sustainability of the use of NAMs for chemical decision making. Does not reflect EPA policy.

Impact/Purpose

Presentation on a recently published study describing outcomes of a study examining the inter-individual variability observed in in vitro primary cell-derived cell culture models of the human tracheobronchial region of the respiratory tract.  First, this study suggests that effects of a single exposure may not reliably represent the trajectory of repeated exposure study outcomes.  Second, the use of current in vitro paradigms, such as the routine use of small donor sample sizes (e.g., n=3 or 5), is not a reliable method for identifying the presence and/or representing the range of responses within the “normal healthy” population using primary cell-based in vitro systems.  Thus, the use of traditionally small sample sizes may bias in vitro study results to over or underestimate effects due to the random chance that the small number of donors used exhibit a more, or less, robust response, respectively.  Alternatively, it is also possible that the effects observed in a small sample size would represent the mean of a larger population, but not reach statistical significance due to a high variation within a small sample size. 

Citation

Bowers, E., E. Martin, A. Jarabek, D. Morgan, H. Smith, L. Dailey, E. Aungst, D. Diaz-Sanchez, AND S. McCullough. Ozone responsive gene expression as a model for describing repeat exposure response trajectories and inter-individual toxicodynamic variability in vitro. Society of Toxicology Annual Meeting, San Diego, California, March 27 - 31, 2022.