Using Tox21 High-Throughput Screening Assays for the Evaluation of Botanical and Dietary Supplements

Abstract Introduction: Recent nationwide surveys found that natural products, including botanical dietary supplements, are used by ∼18% of adults. In many cases, there is a paucity of toxicological data available for these substances to allow for confident evaluations of product safety. The National Toxicology Program (NTP) has received numerous nominations from the public and federal agencies to study the toxicological effects of botanical dietary supplements. The NTP sought to evaluate the utility of in vitro quantitative high-throughput screening (qHTS) assays for toxicological assessment of botanical and dietary supplements. Materials and Methods: In brief, concentration–response assessments of 90 test substances, including 13 distinct botanical species, and individual purported active constituents were evaluated using a subset of the Tox21 qHTS testing panel. The screen included 20 different endpoints that covered a broad range of biologically relevant signaling pathways to detect test article effects upon endocrine activity, nuclear receptor signaling, stress response signaling, genotoxicity, and cell death signaling. Results and Discussion: Botanical dietary supplement extracts induced measurable and diverse activity. Elevated biological activity profiles were observed following treatments with individual chemical constituents relative to their associated botanical extract. The overall distribution of activity was comparable to activities exhibited by compounds present in the Tox21 10K chemical library. Conclusion: Botanical supplements did not exhibit minimal or idiosyncratic activities that would preclude the use of qHTS platforms as a feasible method to screen this class of compounds. However, there are still many considerations and further development required when attempting to use in vitro qHTS methods to characterize the safety profile of botanical/dietary supplements.


MCF-7 aro-ERE and ER-vMCF-7-Luc luciferase reporter gene assays: MCF-7 aro-ERE and
VM7Luc4E2 cells were dispensed at 1500 and 4000 respectively in 5 µL/well (agonist mode) and (Greiner Bio-One) using a Multidrop Combi dispenser. After the assay plates were incubated at 37°C for 5 h (MCF-7 aro-ERE) and 24 h (VM7Luc4E2), 23 nL of compounds dissolved in DMSO, positive and negative controls were transferred to the assay plates using a Pintool station. For antagonist modes, 1 µL/well agonist (0.2 and 0.5 nM β-Estradiol for MCF-7 aro-ERE and VM7Luc4E2 respectively and 0.5 nM Testosterone for MCF-7 aro-ERE testing aromatase activity) or assay medium was added to each well using an FRD. The assay plates were incubated for 22 h (VM7Luc4E2) and 24 h (MCF-7 aro-ERE) at 37°C. For cell viability assay, 1 µL/well CellTiter-Fluor™ Cell Viabillity reagent (Promega Corporation, Madison, WI) was added to each well using an FRD. After 0.5 h incubation at 37°C, the fluorescence intensity was measured using a ViewLux plate reader. For luciferase reporter gene assay, 4 µL/well ONE-Glo™ Luciferase Assay reagent was added to each well using an FRD. After 0.5 h incubation at room temperature, the luminescence intensity was measured using a ViewLux plate reader. Data were represented as relative fluorescence units (cell viability assay) and relative luminescence units (luciferase reporter assay).

Real Time Cell Viability Assay:
To monitor cytotoxicity and cell viability in real time of continues cell culture, a multiplex assay by combining RealTime-Glo™ MT Cell Viability Assay (Promega Corporation) and CellTox™ Green Cytotoxicity Assay were used. 3 The mixture of HepG2 cell or

Data Analyses:
The new wAUC metric has more compact scale of values due to its normalization relative to the infinite dilution concentration, to enable comparisons at alternate exposure concentration ranges/units. Decreasing effects were flagged for cytotoxicity if the POD value of the effect was not more potent than the POD of cytotoxicity from an assay's counter-screen. Also, the effects identified in the β-lactamase assays were flagged if the responses in the reporter gene channel readout were not matched with the responses after normalization to background. Analysis of compound concentration-response data was performed as previously described Huang 2016. 5 Briefly, raw plate reads for each titration point were first normalized relative to the positive control compound (-100% for antagonist mode and 100% for agonist mode) and DMSO-only wells (0%) as follows: % Activity = ((Vcompound -VDMSO)/(Vpos -VDMSO)) ×100, where Vcompound denotes the compound well values, Vpos denotes the median value of the positive control wells, and VDMSO denotes the median values of the DMSO-only wells, and then corrected by applying a NCATS in-house pattern correction algorithm. 6 Concentration-response titration points for each compound were fitted to a four-parameter Hill equation yielding concentrations of half-maximal inhibitory activity (IC50) or half-maximal stimulatory activity (EC50) and maximal response (efficacy) values. 7 Compounds were designated as Class 1-4 according to the type of concentration-response curve observed. 5,8 Noise level (threshold, THR) as the Curvep input: Curvep, a response noise filtering algorithm, was used to process the curves. Curvep relies on user-defined thresholds such as the baseline noise threshold (THR) and the maximum curve deviation (MXDV) to filter the response noise. Among thresholds, it is known that the THR has direct and significant impact on defining activity of testing chemicals.
As the THR is a user-defined parameter, we think that the optimal THR should reflect the intrinsic response variation in the screening data of the substances and has the meaning of minimum response threshold. To achieve this, we iteratively applied Curvep using various THRs on the simulated curves derived from the screening data of the substances, and the optimal THR was identified as the lowest THR at which the variance in potency estimation was sufficiently reduced. 9 The specific details are explained as follows: Curve simulation: The response of a curve for each substance was calculated using the Equation 1, where # and # were estimated from the linear regression with response as dependent variable and concentration as independent variance from the screening data and is the randomly generated noise = * # + , # , + The noise was generated from bootstrapping responses from a normal distribution with mean = 0 and SD = standard deviation of responses from vehicle control-only wells. In total, 100 curves were created for each substance in every readout. For each readout, to identify the optimal THR for either increased or decreased effect, THR from 5% to 95% with an increment of 5% was applied on the simulated curves. Then, the potency/concentration at x threshold was reported (x is 5% to 95% with an increment of 5%) for each simulated curve. For inactive curve (i.e., all responses = 0 after Curvep), the potency was fixed to the maximum tested concentration. The pooled variance of potency of all substances for each THR was calculated. The optimal THR was considered as the lowest THR at which the variance in potency estimation was sufficiently reduced and stabilized.
The implementation for identification of the optimal THR is available in the R package, Rcurvep