Experimental Exposure Assessment of Ionizable Organic Chemicals in In Vitro Cell-Based Bioassays

Effects of Chemicals in Reporter Gene Bioassays with Different Metabolic Activities Compared to Baseline Toxicity

High-throughput cell-based bioassays are used for chemical screening and risk assessment. Chemical transformation processes caused by abiotic degradation or metabolization can reduce the chemical concentration or, in some cases, lead to the formation of more toxic transformation products. Unaccounted loss processes may falsify the bioassay results. Capturing the formation and effects of transformation products is important for relating the in vitro effects to in vivo. Reporter gene cell lines are believed to have low metabolic activity, but inducibility of cytochrome P450 (CYP) enzymes has been reported. Baseline toxicity is the minimal toxicity a chemical can have and is caused by the incorporation of the chemical into cell membranes. In the present study, we improved an existing baseline toxicity model based on a newly defined critical membrane burden derived from freely dissolved effect concentrations, which are directly related to the membrane concentration. Experimental effect concentrations of 94 chemicals in three bioassays (AREc32, ARE-bla and GR-bla) were compared with baseline toxicity by calculating the toxic ratio (TR). CYP activities of all cell lines were determined by using fluorescence-based assays. Only ARE-bla showed a low basal CYP activity and inducibility and AREc32 showed a low inducibility. Overall cytotoxicity was similar in all three assays despite the different metabolic activities indicating that chemical metabolism is not relevant for the cytotoxicity of the tested chemicals in these assays. Up to 28 chemicals showed specific cytotoxicity with TR > 10 in the bioassays, but baseline toxicity could explain the effects of the majority of the remaining chemicals. Seven chemicals showed TR < 0.1 indicating inaccurate physicochemical properties or experimental artifacts like chemical precipitation, volatilization, degradation, or other loss processes during the in vitro bioassay. The new baseline model can be used not only to identify specific cytotoxicity mechanisms but also to identify potential problems in the experimental performance or evaluation of the bioassay and thus improve the quality of the bioassay data.

Experimental exposure assessment for in vitro cell-based bioassays in 96- and 384-well plates

In vitro cell-based bioassays have great potential for applications in the human health risk assessment of chemicals. The quantification of freely dissolved concentrations (C free) in in vitro assays is essential to generate reliable data for in vitro-to-in vivo extrapolation. Existing methods for the quantification of C free are limited to low-throughput microtiter plates. The present study is a proof of principle for the applicability of a solid-phase microextraction (SPME) method for the determination of C free in the peroxisome proliferator-activated receptor gamma (PPARγ) bioassay run in 384-well plates with 80 µL medium per well. The effect concentrations obtained from 384-well plates were compared with those obtained from 96-well plates in a previous study. Nominal effect concentrations obtained using 96- and 384-well plates agreed with each other within a factor of three, and freely dissolved effect concentrations agreed within a factor of 6.5. The good degree of agreement in the results from both plate formats proves the general applicability of the SPME method for the determination of C free for bioassays in 384-well plates, making the present study a first step toward exposure assessment in high-throughput bioassays.

Role of bioavailability and protein binding of four anionic perfluoroalkyl substances in cell-based bioassays for quantitative in vitro to in vivo extrapolations

Perfluoroalkyl substances (PFAS) are persistent and pose a risk to human health. High throughput screening (HTS) cell-based bioassays may inform risk assessment of PFAS provided that quantitative in vitro to in vivo extrapolation (QIVIVE) can be developed. The QIVIVE ratio is the ratio of nominal (C_nom) or freely dissolved concentration (C_free) in human blood to C_nom or C_free in the bioassays. Considering that the concentrations of PFAS in human plasma and in vitro bioassays may vary by orders of magnitude, we tested the hypothesis that anionic PFAS bind to proteins concentration-dependently and therefore the binding differs substantially between human plasma and bioassays, which has an impact on QIVIVE. Solid phase microextraction (SPME) with C18-coated fibers served to quantify the C_free of four anionic PFAS (perfluorobutanoate (PFBA), perfluorooctanoate (PFOA), perfluorohexane sulfonate (PFHxS) and perfluorooctane sulfonate (PFOS)) in the presence of proteins and lipid, medium components, cells and human plasma over five orders of magnitude in concentrations. The C18-SPME method was used to quantify the non-linear binding to proteins, human plasma and medium, and the partition constants to cells. These binding parameters were used to predict C_free of PFAS in cell bioassays and human plasma by a concentration-dependent mass balance model (MBM). The approach was illustrated with a reporter gene assay indicating activation of the peroxisome proliferator-activated receptor gamma (PPARγ-GeneBLAzer). Blood plasma levels were collected from literature for occupational exposure and the general population. The QIVIVE_nom ratios were higher than the QIVIVE_free ratios due to the strong affinity to proteins and large differences in protein contents between human blood and bioassays. For human health risk assessment, the QIVIVE_free ratios of many in vitro assays need to be combined to cover all health relevant endpoints. If C_free cannot be measured, they can be estimated with the MBM and concentration-dependent distribution ratios.

Validation of an SH-SY5Y Cell-Based Acetylcholinesterase Inhibition Assay for Water Quality Assessment

The acetylcholinesterase (AChE) inhibition assay has been frequently applied for environmental monitoring to capture insecticides such as organothiophosphates (OTPs) and carbamates. However, natural organic matter such as dissolved organic carbon (DOC) co-extracted with solid-phase extraction from environmental samples can produce false-negative AChE inhibition in free enzyme-based AChE assays. We evaluated whether disturbance by DOC can be alleviated in a cell-based AChE assay using differentiated human neuroblastoma SH-SY5Y cells. The exposure duration was set at an optimum of 3 h considering the effects of OTPs and carbamates. Because loss to the airspace was expected for the more volatile OTPs (chlorpyrifos, diazinon, and parathion), the chemical loss in this bioassay setup was investigated using solid-phase microextraction followed by chemical analysis. The three OTPs were relatively well retained (loss <34%) during 3 h of exposure in the 384-well plate, but higher losses occurred on prolonged exposure, accompanied by slight cross-contamination of adjacent wells. Inhibition of AChE by paraoxon-ethyl was not altered in the presence of up to 68 mg_c /L Aldrich humic acid used as surrogate for DOC. Binary mixtures of paraoxon-ethyl and water extracts showed concentration-additive effects. These experiments confirmed that the matrix in water extracts does not disturb the assay, unlike purified enzyme-based AChE assays. The cell-based AChE assay proved to be suitable for testing water samples with effect concentrations causing 50% inhibition of AChE at relative enrichments of 0.5-10 in river water samples, which were distinctly lower than corresponding cytotoxicity, confirming the high sensitivity of the cell-based AChE inhibition assay and its relevance for water quality monitoring. Environ Toxicol Chem 2022;41:3046-3057. © 2022 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Toward Realistic Dosimetry In Vitro: Determining Effective Concentrations of Test Substances in Cell Culture and Their Prediction by an In Silico Mass Balance Model

Nominal concentrations (C_Nom) in cell culture media are routinely used to define concentration-effect relationships in the in vitro toxicology. The actual concentration in the medium (C_Medium) can be affected by adsorption processes, evaporation, or degradation of chemicals. Therefore, we measured the total and free concentration of 12 chemicals, covering a wide range of lipophilicity (log K_OW -0.07-6.84), in the culture medium (C_Medium) and cells (C_Cell) after incubation with Balb/c 3T3 cells for up to 48 h. Measured values were compared to predictions using an as yet unpublished in silico mass balance model that combined relevant equations from similar models published by others. The total C_Medium for all chemicals except tamoxifen (TAM) were similar to the C_Nom. This was attributed to the cellular uptake of TAM and accumulation into lysosomes. The free (i.e., unbound) C_Medium for the low/no protein binding chemicals were similar to the C_Nom, whereas values of all moderately to highly protein-bound chemicals were less than 30% of the C_Nom. Of the 12 chemicals, the two most hydrophilic chemicals, acetaminophen (APAP) and caffeine (CAF), were the only ones for which the C_Cell was the same as the C_Nom. The C_Cell for all other chemicals tended to increase over time and were all 2- to 274-fold higher than C_Nom. Measurements of C_Cytosol, using a digitonin method to release cytosol, compared well with C_Cell (using a freeze-thaw method) for four chemicals (CAF, APAP, FLU, and KET), indicating that both methods could be used. The mass balance model predicted the total C_Medium within 30% of the measured values for 11 chemicals. The free C_Medium of all 12 chemicals were predicted within 3-fold of the measured values. There was a poorer prediction of C_Cell values, with a median overprediction of 3- to 4-fold. In conclusion, while the number of chemicals in the study is limited, it demonstrates the large differences between C_Nom and total and free C_Medium and C_Cell, which were also relatively well predicted by the mass balance model.

Update and Evaluation of a High-Throughput In Vitro Mass Balance Distribution Model: IV-MBM EQP v2.0

This study demonstrates the utility of an updated mass balance model for predicting the distribution of organic chemicals in in vitro test systems (IV-MBM EQP v2.0) and evaluates its performance with empirical data. The IV-MBM EQP v2.0 tool was parameterized and applied to four independent data sets with measured ratios of bulk medium or freely-dissolved to initial nominal concentrations (e.g., C24/C0 where C24 is the measured concentration after 24 h of exposure and C0 is the initial nominal concentration). Model performance varied depending on the data set, chemical properties (e.g., "volatiles" vs. "non-volatiles", neutral vs. ionizable organics), and model assumptions but overall is deemed acceptable. For example, the r² was greater than 0.8 and the mean absolute error (MAE) in the predictions was less than a factor of two for most neutral organics included. Model performance was not as good for the ionizable organic chemicals included but the r² was still greater than 0.7 and the MAE less than a factor of three. The IV-MBM EQP v2.0 model was subsequently applied to several hundred chemicals on Canada's Domestic Substances List (DSL) with nominal effects data (AC50s) reported for two in vitro assays. We report the frequency of chemicals with AC50s corresponding to predicted cell membrane concentrations in the baseline toxicity range (i.e., >20-60 mM) and tabulate the number of chemicals with "volatility issues" (majority of chemical in headspace) and "solubility issues" (freely-dissolved concentration greater than water solubility after distribution). In addition, the predicted "equivalent EQP blood concentrations" (i.e., blood concentration at equilibrium with predicted cellular concentration) were compared to the AC50s as a function of hydrophobicity (log octanol-water partition or distribution ratio). The predicted equivalent EQP blood concentrations exceed the AC50 by up to a factor of 100 depending on hydrophobicity and assay conditions. The implications of using AC50s as direct surrogates for human blood concentrations when estimating the oral equivalent doses using a toxicokinetic model (i.e., reverse dosimetry) are then briefly discussed.

Critical Membrane Concentration and Mass-Balance Model to Identify Baseline Cytotoxicity of Hydrophobic and Ionizable Organic Chemicals in Mammalian Cell Lines

All chemicals can interfere with cellular membranes and this leads to baseline toxicity, which is the minimal toxicity any chemical elicits. The critical membrane burden is constant for all chemicals; that is, the dosing concentrations to trigger baseline toxicity decrease with increasing hydrophobicity of the chemicals. Quantitative structure-activity relationships, based on hydrophobicity of chemicals, have been established to predict nominal concentrations causing baseline toxicity in human and mammalian cell lines. However, their applicability is limited to hydrophilic neutral compounds. To develop a prediction model that includes more hydrophobic and charged organic chemicals, a mass balance model was applied for mammalian cells (AREc32, AhR-CALUX, PPARγ-BLA, and SH-SY5Y) considering different bioassay conditions. The critical membrane burden for baseline toxicity was converted into nominal concentration causing 10% cytotoxicity by baseline toxicity (IC_10,baseline) using a mass balance model whose main chemical input parameter was the liposome-water partition constants (K_lip/w) for neutral chemicals or the speciation-corrected D_lip/w(pH 7.4) for ionizable chemicals plus the bioassay-specific protein, lipid, and water contents of cells and media. In these bioassay-specific models, log(1/IC_10,baseline) increased with increasing hydrophobicity, and the relationship started to level off at log D_lip/w around 2. The bioassay-specific models were applied to 392 chemicals covering a broad range of hydrophobicity and speciation. Comparing the predicted IC_10,baseline and experimental cytotoxicity IC₁₀, known baseline toxicants and many additional chemicals were identified as baseline toxicants, while the others were classified based on specificity of their modes of action in the four cell lines, confirming excess toxicity of some fungicides, antibiotics, and uncouplers. Given the similarity of the bioassay-specific models, we propose a generalized baseline-model for adherent human cell lines: log[1/IC_10,baseline (M)] = 1.23 + 4.97 × (1 - e^{-0.236 log D_lip/w}). The derived models for baseline toxicity may serve for specificity analysis in reporter gene and neurotoxicity assays as well as for planning the dosing for cell-based assays.

Quantitative In Vitro-to-In Vivo Extrapolation: Nominal versus Freely Dissolved Concentration

Discussions are ongoing on which dose metric should be used for quantitative in vitro-to-in vivo extrapolation (QIVIVE) of in vitro bioassay data. The nominal concentration of the test chemicals is most commonly used and easily accessible, while the concentration freely dissolved in the assay medium is considered to better reflect the bioavailable concentration but is tedious to measure. The aim of this study was to elucidate how much QIVIVE results will differ when using either nominal or freely dissolved concentrations. QIVIVE_nom and QIVIVE_free ratios, that is, the ratios of plasma concentrations divided by in vitro effect concentrations, were calculated for 10 pharmaceuticals using previously published nominal and freely dissolved effect concentrations for the activation of the peroxisome proliferator-activated receptor gamma (PPARγ) and the activation of oxidative stress response. The QIVIVE_nom ratios were higher than QIVIVE_free ratios by up to a factor of 60. The risk of in vivo effects was classified as being high or low for four chemicals using the QIVIVE_nom and for three chemicals using QIVIVE_free ratios. Unambiguous classification was possible for nine chemicals by combining the QIVIVE_nom or QIVIVE_free ratios with the respective specificity ratios (SR_nom or SR_free) of the in vitro effect data, which helps to identify whether the specific effect was influenced by cytotoxicity. QIVIVE_free models should be preferred as they account for differences in bioavailability between in vitro and in vivo, but QIVIVE_nom may still be useful for screening the effects of large numbers of chemicals because it is generally more conservative. The use of SR of the in vitro effect data as a second classification factor is recommended for QIVIVE_nom and QIVIVE_free models because a clearer picture can be obtained with respect to the likelihood that a biological effect will occur and that it is not caused by nonspecific cytotoxicity.

Effect-Based Trigger Values for Mixtures of Chemicals in Surface Water Detected with In Vitro Bioassays

Effect-based trigger (EBT) values for in vitro bioassays are important for surface water quality monitoring because they define the threshold between acceptable and poor water quality. They have been derived for highly specific bioassays, such as hormone-receptor activation in reporter gene bioassays, by reading across from existing chemical guideline values. This read-across method is not easily applicable to bioassays indicative of adaptive stress responses, which are triggered by many different chemicals, and activation of nuclear receptors for xenobiotic metabolism, to which many chemicals bind with rather low specificity. We propose an alternative approach to define the EBT from the distribution of specificity ratios of all active chemicals. The specificity ratio is the ratio between the predicted baseline toxicity of a chemical in a given bioassay and its measured specific endpoint. Unlike many previous read-across methods to derive EBTs, the proposed method accounts for mixture effects and includes all chemicals, not only high-potency chemicals. The EBTs were derived from a cytotoxicity EBT that was defined as equivalent to 1% of cytotoxicity in a native surface water sample. The cytotoxicity EBT was scaled by the median of the log-normal distribution of specificity ratios to derive the EBT for effects specific for each bioassay. We illustrate the new approach using the example of the AREc32 assay, indicative of the oxidative stress response, and 2 nuclear receptor assays targeting the peroxisome proliferator-activated receptor gamma and the arylhydrocarbon receptor. The EBTs were less conservative than previously proposed but were able to differentiate untreated and insufficiently treated wastewater from wastewater treatment plant effluent with secondary or tertiary treatment and surface water. Environ Toxicol Chem 2021;40:487-499. © 2020 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.