Performance of the BG1Luc ER TA Method in a qHTS Format
In 2012, the BG1Luc4E2 estrogen receptor (ER) transactivation (TA) method (BG1Luc ER TA) was accepted by U.S. regulatory agencies and the Organisation for Economic Co-operation and Development to detect substances with ER agonist activity. The method is now part of the Tier 1 testing battery in the Environmental Protection Agency’s Endocrine Disruptor Screening Program. The BG1Luc ER TA method uses the BG1 ovarian cell line that endogenously expresses full-length ER (α and β) and is stably transfected with a plasmid containing four estrogen responsive elements upstream of a luciferase reporter gene. To allow increased throughput and testing efficiency, the BG1Luc ER TA (“BG1 manual”) method was adapted for quantitative high-throughput screening (BG1 qHTS) in the U.S. Tox21 testing program. The BG1 qHTS test method was used to test approximately 10,000 chemicals three times each, and concentration-response data (n = 15) were analyzed to evaluate test method performance. The balanced accuracy of the BG1 qHTS test method (97%, i.e., 32/33) was determined by comparing results to ER TA performance standards for the BG1 manual method. Concordance between the BG1 manual and qHTS methods was 92% (57/62) when calculated for a larger set of non-reference chemicals tested in both methods. These data demonstrate that the performance of the BG1 qHTS is similar to the currently accepted BG1 manual method, thereby establishing the utility of the BG1 qHTS method for identifying ER active environmental chemicals.
Figure 1. Linear regression analysis of BG1 manual and qHTS EC50 /AC50 values.
A linear regression analysis was conducted of EC50/AC50 values for 33 substances that tested positive in the BG1 manual and HTS methods. A list of the chemicals used to create Figure 1 is included in Table 2. The slope of the linear regression is 0.48 with r2 of 0.69.
- Figure 1 (42 KB)
Figure 2. Discordant response for dicofol in BG1 manual and qHTS methods.
Concentration-response curves for dicofol tested with the agonist protocols for the BG1 manual and qHTS methods. The data plotted for BG1 manual represent results from a single laboratory, and each point is the mean of three within-experiment replicates (+/- standard deviation). The qHTS data represents the mean of three curves (one each in each experiment, +/- standard deviation).
- Figure 2 (71 KB)
Table 1. Summary of differences between the BG1Luc ER TA manual and qHTS methods.
- Table 1 (74 KB)
Table 2. EC50 and AC50 values for the 33 positive substances in both the BG1 manual and qHTS methods.
Table contains median EC50 and AC50 values for the 33 substances that were positive in both the BG1 manual and qHTS methods. Comparison of chemicals on an individual bases using a paired t-test indicated that there were no significant differences in EC50/AC50 values. Evaluation of the chemicals on a population basis using a two-tailed Mann-Whitney U test indicated no quantitative difference. CASRN, Chemical Abstracts Service Registry Number.
- Table 2 (109 KB)
Table 3. BG1 qHTS classifications of performance standards substances.
Table contains positive/negative classiﬁcations for the list of 34 reference substances for assessing the sensitivity and speciﬁcity of proposed published test methods. CASRN: Chemical Abstracts Service Registry Number; NEG: negative; POS: positive.
- Table 3 (70 KB)
Table 4. Concordance of the agonist protocols for the BG1 manual and qHTS methods.
Concordance was evaluated for 64 substances (38 positive, 26 negative) tested using both the BG1 manual and qHTS methods, omitting substances that yielded inconclusive results in the BG1 manual method. Overall concordance between t he two methods for the 64 substances was 92% (59/64).
- Table 4 (34 KB)
- Supplementary Data (302 KB)