Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Nov;148(1):137-54.
doi: 10.1093/toxsci/kfv168. Epub 2015 Aug 13.

Integrated Model of Chemical Perturbations of a Biological Pathway Using 18 In Vitro High-Throughput Screening Assays for the Estrogen Receptor

Richard S Judson  1 , Felicia Maria Magpantay  2 , Vijay Chickarmane  3 , Cymra Haskell  4 , Nessy Tania  5 , Jean Taylor  6 , Menghang Xia  7 , Ruili Huang  7 , Daniel M Rotroff  8 , Dayne L Filer  9 , Keith A Houck  10 , Matthew T Martin  10 , Nisha Sipes  11 , Ann M Richard  10 , Kamel Mansouri  9 , R Woodrow Setzer  10 , Thomas B Knudsen  10 , Kevin M Crofton  10 , Russell S Thomas  10
Affiliations

Integrated Model of Chemical Perturbations of a Biological Pathway Using 18 In Vitro High-Throughput Screening Assays for the Estrogen Receptor

Richard S Judson et al. Toxicol Sci. 2015 Nov.

Abstract

We demonstrate a computational network model that integrates 18 in vitro, high-throughput screening assays measuring estrogen receptor (ER) binding, dimerization, chromatin binding, transcriptional activation, and ER-dependent cell proliferation. The network model uses activity patterns across the in vitro assays to predict whether a chemical is an ER agonist or antagonist, or is otherwise influencing the assays through a manner dependent on the physics and chemistry of the technology platform ("assay interference"). The method is applied to a library of 1812 commercial and environmental chemicals, including 45 ER positive and negative reference chemicals. Among the reference chemicals, the network model correctly identified the agonists and antagonists with the exception of very weak compounds whose activity was outside the concentration range tested. The model agonist score also correlated with the expected potency class of the active reference chemicals. Of the 1812 chemicals evaluated, 111 (6.1%) were predicted to be strongly ER active in agonist or antagonist mode. This dataset and model were also used to begin a systematic investigation of assay interference. The most prominent cause of false-positive activity (activity in an assay that is likely not due to interaction of the chemical with ER) is cytotoxicity. The model provides the ability to prioritize a large set of important environmental chemicals with human exposure potential for additional in vivo endocrine testing. Finally, this model is generalizable to any molecular pathway for which there are multiple upstream and downstream assays available.

Keywords: EDSP; In vitro; biological modeling; estrogen receptor; high-throughput screening; prioritization.

PubMed Disclaimer

Figures

FIG. 1.
FIG. 1.
(A) Graphical representation of the computational network used in the in vitro analysis of the ER pathway across assays and technology platforms. Colored arrow nodes represent “receptors” with which a chemical can directly interact. Colored circles represent intermediate biological processes that are not directly observable. White stars represent the in vitro assays that measure activity at the biological nodes. Arrows represent transfer of information. Gray arrow nodes are the pseudoreceptors. Each in vitro assay (with the exception of A16) has an assay-specific pseudoreceptor, but only a single example is explicitly shown, for assay A1. (B) Patterns of assays that would be activated when specific receptors are activated by the chemical, in particular R1, R2 and R6. The activating chemical in its receptor are circled in pink, and the activated assays and the pathways to them are also highlighted in pink.
FIG. 2.
FIG. 2.
Two-way hierarchical clustering of chemical activity across the 18 in vitro assays used to test for ER activity. Assays and technologies are aligned across the x-axis, where the “A” and “R” values refer to the assay and receptors/pseudoreceptors from Figure 1 and Table 1.Chemicals are aligned along the y-axis. The heatmap shows –log10(AC50) values for all assays and all chemicals with at least one assay hit. Darker red indicates more potent activity (lower AC50), while white represents inactive chemical-assay pairs. Note that the assays cluster by technology/pseudoreceptor.
FIG. 3.
FIG. 3.
Results of the model for 3 prototype chemicals. For each chemical, the left-hand panel shows the synthetic concentration-response data for the 18 assays, colored by assay groups defined in the legend. The right-hand panel shows the corresponding magnitude of the modeled receptor responses. The agonist receptor (R1) is designated by blue, the antagonist receptor (R2) by red and the other pseudoreceptors are colored as indicated in the legend. AUC values for the agonist (R1) and antagonist (R2) receptors are provided below the chemical name. For chemicals with cell-stress/cytotoxicity activity (2 or more cytotoxicity hits, see Methods), the cell-stress/cytotoxicity center is indicated by a vertical red line, and the cell-stress/cytotoxicity region (starting 3 cell-stress/cytotoxicity MAD below the cell-stress/cytotoxicity center) is indicated by the gray shaded region. A green horizontal bar indicates the median-AC50 of the active assays. Similar plots for all chemicals are given in Supplemental File S3.
FIG. 3.
FIG. 3.
Results of the model for 3 prototype chemicals. For each chemical, the left-hand panel shows the synthetic concentration-response data for the 18 assays, colored by assay groups defined in the legend. The right-hand panel shows the corresponding magnitude of the modeled receptor responses. The agonist receptor (R1) is designated by blue, the antagonist receptor (R2) by red and the other pseudoreceptors are colored as indicated in the legend. AUC values for the agonist (R1) and antagonist (R2) receptors are provided below the chemical name. For chemicals with cell-stress/cytotoxicity activity (2 or more cytotoxicity hits, see Methods), the cell-stress/cytotoxicity center is indicated by a vertical red line, and the cell-stress/cytotoxicity region (starting 3 cell-stress/cytotoxicity MAD below the cell-stress/cytotoxicity center) is indicated by the gray shaded region. A green horizontal bar indicates the median-AC50 of the active assays. Similar plots for all chemicals are given in Supplemental File S3.
FIG. 4.
FIG. 4.
Plots showing activity of the agonist (top) and antagonist (bottom) reference chemicals. Chemicals that are intended to be positive are indicated by green circles, while those intended to be inactive are indicated by red circles. For the agonists, the expected potency range is also indicated (middle column). For chemicals with one or more pseudoreceptor AUC values greater than zero, the value is indicated by an X, and the pseudoreceptor name is indicated. The inset shows the assay curves for dibutyl phthalate, as described in the text (colored based on Fig. 3).
FIG. 5.
FIG. 5.
Plot of the maximum AUC vs. minimum-AC50 values. Each point is a single chemical that was active in at least 1 assay. The AUC value given is the maximum of the AUC (agonist) and AUC (antagonist) values for the chemical. The dashed line is the best-fit for AUC(agonist) values >0.1. Chemicals are labeled in order: black circle, at least 1 AUC > 0.1; green up-triangle, positive agonist reference chemical; green down-triangle, positive antagonist reference chemical; red diamond, negative reference chemical; cyan circle, example chemicals with AUC significantly below the fitted line but above 0.1. The vertical line at 100 µM indicates the highest concentration tested, while the horizontal line at AUC = 0.1 indicates an approximate threshold between chemicals with clear agonist/antagonist activity and those that are potentially active through interference processes. The inset shows graphs of assay activity for 4-androstene-3,17-dione (colored based on Fig. 3).
FIG. 6.
FIG. 6.
Histogram of Z-scores for the assay ATG_ERa_TRANS_up. The Z-score is defined as the distance between the median cytotoxicity concentration and a chemical’s AC50 in this assay, in units of global cytotoxicity MAD, for all chemicals active in this assay. One can see a bimodal distribution with one peak at zero (marked with a heavy line) and another with a peak near 6. We hypothesize that chemicals active in the low-Z region are more likely to be false positives and less likely to be estrogenic than those active in the high-Z region.
FIG. 7.
FIG. 7.
Bar chart showing the fraction of chemicals remaining for each of the multi-assay receptors/pseudoreceptors after filtering for efficacy (T) and cytotoxicity (Z-score). The receptors were limited to those with 5 or more chemicals with AUC > 0.1 from Table 2. If there were no chemicals for a pseudoreceptor in a given AUC bin, a small negative bar is shown. The legend indicates AUC ranges corresponding to Table 2.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Akaike H. (1998). Information Theory and an Extension of the Maximum Liklihood Principle. Springer, New York, NY.
    1. Auld D. S., Thorne N., Nguyen D. T., Inglese J. (2008). A specific mechanism for nonspecific activation in reporter-gene assays. ACS Chem. Biol. 3, 463–470. - PMC - PubMed
    1. Baell J. B., Holloway G. A. (2010). New substructure filters for removal of pan assay interference compounds (PAINS) from screening libraries and for their exclusion in bioassays. J. Med. Chem. 53, 2719–2740. - PubMed
    1. Browne P., Judson R. S., Casey W. M., Kleinstreuer N. C., Thomas R. S. (2015). Screening chemicals for estrogen receptor bioactivity using a computational model. Environ Sci. Technol. 49, 8804–8814. - PubMed
    1. Bruns R. F., Watson I. A. (2012). Rules for identifying potentially reactive or promiscuous compounds. J. Med. Chem., 55, 9763–9772. - PubMed

Publication types

MeSH terms