Science Inventory

High-throughput Identification of Androgen Disrupting ToxCast Chemicals and CYP450 Metabolites

Citation:

Simmons, S., D. Hallinger, AND E. Brown. High-throughput Identification of Androgen Disrupting ToxCast Chemicals and CYP450 Metabolites. SOT Conference 2024: A New Approach Method (NAM) to Screen for the Impact of Endogenous Stress on Chemical Toxicity, Salt Lake City, UT, March 10 - 14, 2024. https://doi.org/10.23645/epacomptox.25545850

Impact/Purpose:

Presentation to SOT Conference 2024: A New Approach Method (NAM) to Screen for the Impact of Endogenous Stress on Chemical Toxicity

Description:

Background and Purpose The identification of chemicals with the potential to disrupt androgen signaling is a key component of the EPA’s Endocrine Disruption Screening Program (EDSP). EPA researchers have worked to devise in vitro assays and computational models to serve as alternatives to animal based EDSP tests. One of the major limitations of these assays is the inability to recapitulate the effects of in vivo xenobiotic metabolism. To address in vitro metabolic deficiency, the latest ToxCast strategic plan proposed transfecting chemically modified mRNAs encoding human cytochrome P450 (CYP) enzymes into cells to induce intracellular CYP expression and functional metabolic activity. In this study, we screened 2,165 ToxCast chemicals in an AR homodimerization assay in parallel across 10 human liver CYP enzymes to identify anti-androgenic parent chemicals and their metabolites. Methods The human androgen receptor dimerization assay (AR2) was executed in antagonist mode to conduct an initial single-concentration screening of 2,165 ToxCast chemical samples in cells transfected with mRNAs encoding one of 10 human CYP450 enzymes (CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, CYP2J2, CYP3A4), beta-galactosidase (Bgal; no metabolism control mRNA), or without mRNA (No RNA). The loss-of-signal format facilitated the identification of inactive samples with no potential anti-androgen activity in any transfected biogroup. Active samples were re-screened in the AR2 antagonist mode assay across the same 12 biogroups in an 8-point concentration-response format. AR inhibition responses were normalized 0 to 100 using the responses of DMSO (vehicle) controls and 100 µM bicalutamide, respectively. Active responses were delineated using 3x baseline median absolute deviation of DMSO controls as the threshold. The effect size of CYP activity compared to no metabolism controls (Bgal + No RNA) was calculated using the strictly standardized mean difference (SSMD) for each chemical sample by concentration by CYP enzyme combination (117,200 total). CYP metabolism effects were defined as those chemical samples with at least one tested concentration in which: (a) either the median CYP or no metabolism response exceed the activity threshold, (b) the difference between the median CYP and no metabolism responses was > 20%, and (c) the SSMD was > 0.5 (bioactivation) or < -0.5 (deactivation). Results The initial single concentration testing identified 700 (32%) chemical samples as inactive across all biogroups. Multi-concentration testing of the remaining 1,465 samples revealed 960 with potential anti-androgen activity in at least one biogroup. Metabolic shifts in AR bioactivity were identified for 373 samples or 17% of the original 2,165 tested. Most of the shifted bioactivity resulted from CYP-induced deactivation (i.e., reduced AR inhibition), observed in 245 samples while only 141 exhibited CYP-mediated bioactivation (i.e., increased AR inhibition). Of the deactivated samples, 167 (68%) had a weak response (-0.5 > SSMD > -1), 68 (28%) a moderate response (-1 > SSMD > -2), and 10 (4%) a strong response (-2 > SSMD). Among the bioactivated samples, 77 (55%) had a weak response (0.5 < SSMD < 1), 47 (33%) a moderate response (1 < SSMD < 2), and 17 (12%) a strong response (2 < SSMD). In total, 427 unique CYP enzyme by chemical sample combinations resulted in deactivation and 210 unique combinations resulted in bioactivation. Most of the metabolically shifted bioactivity was CYP enzyme specific as 65% of deactivated and 74% of bioactivated samples were affected by a single CYP enzyme. Only 9.8% of deactivated and 6.4% of bioactivated samples were affected by more than three CYP enzymes. Among unique . . .