Simplified models to analyse time- and dose-dependent responses of populations to toxicants

The basis of ecotoxicology lies currently in the dose–response of organisms to toxicants, as typically described by probit and logistic models. While recognising its merits, standard endpoints ignore the process of toxicity with time, and consequently our ability to predict direct toxic effects in environmental risk assessments is seriously curtailed. Although the response of toxicants with time has been studied before, its application in ecotoxicology remains underutilised. One reason is that no convincing mechanism has been proposed to explain the hyperbolic curves of such responses, whereas a variety of models have been used to describe them. The explanation of both time- and dose-dependent responses is found ultimately in the natural variability of receptor sites among individuals of populations exposed to a toxicant inhibitor with time. The process can be explained by the kinetics of inhibition, and is appropriately described by a simple mathematical expression like the Michaelis–Menten equation, though other asymptotic models (e.g. logistic model) can also be used. The advantage of the hyperbolic model is that median effect values (e.g. LC₅₀ for dose- and ET₅₀ for time-dependent responses) enable calculation of toxicity effects at any concentration level and/or time of exposure, thus making it especially attractive for risk assessment.     

Automated swimming activity monitor for examining temporal patterns of toxicant effects on individual Daphnia magna

Aquatic pollutants are often biologically active at low concentrations and impact on biota in combination with other abiotic stressors. Traditional toxicity tests may not detect these effects, and there is a need for sensitive high‐throughput methods for detecting sublethal effects. We have evaluated an automated infra‐red (IR) light‐based monitor for recording the swimming activity of Daphnia magna to establish temporal patterns of toxicant effects on an individual level. Activity was recorded for 48 h and the sensitivity of the monitor was evaluated by exposing D. magna to the reference chemicals K₂Cr₂O₇ at 15, 20 and 25 °C and 2,4‐dichlorophenol at 20 °C. Significant effects (P < 0.001) of toxicant concentrations, exposure time and incubation temperatures were observed. At 15 °C, the swimming activity remained unchanged for 48 h at sublethal concentrations of K₂Cr₂O₇ whereas activity at 20 and 25 °C was more biphasic with decreases in activity occurring after 12–18 h. A similar biphasic pattern was observed after 2,4‐dichlorophenol exposure at 20 °C. EC₅₀ values for 2,4‐dichlorophenol and K₂Cr₂O₇ determined from automated recording of swimming activity showed increasing toxicity with time corresponding to decreases in EC₅₀ of 0.03–0.07 mg l^–1 h^–1. EC₅₀ values determined after 48 h were comparable or lower than EC₅₀ values based on visual inspection according to ISO 6341. The results demonstrated that the swimming activity monitor is capable of detecting sublethal behavioural effects that are toxicant and temperature dependent. The method allows EC values to be established at different time points and can serve as a high‐throughput screening tool in toxicity testing. Copyright © 2015 John Wiley & Sons, Ltd.     

A high-throughput method for assessing chemical toxicity using a Caenorhabditis elegans reproduction assay

The National Research Council has outlined the need for non-mammalian toxicological models to test the potential health effects of a large number of chemicals while also reducing the use of traditional animal models. The nematode Caenorhabditis elegans is an attractive alternative model because of its well-characterized and evolutionarily conserved biology, low cost, and ability to be used in high-throughput screening. A high-throughput method is described for quantifying the reproductive capacity of C. elegans exposed to chemicals for 48 h from the last larval stage (L4) to adulthood using a COPAS Biosort. Initially, the effects of exposure conditions that could influence reproduction were defined. Concentrations of DMSO vehicle ≤ 1% did not affect reproduction. Previous studies indicated that C. elegans may be influenced by exposure to low pH conditions. At pHs greater than 4.5, C. elegans reproduction was not affected; however below this pH there was a significant decrease in the number of offspring. Cadmium chloride was chosen as a model toxicant to verify that automated measurements were comparable to those of traditional observational studies. EC₅₀ values for cadmium for automated measurements (176-192 µM) were comparable to those previously reported for a 72-h exposure using manual counting (151 µM). The toxicity of seven test toxicants on C. elegans reproduction was highly correlative with rodent lethality suggesting that this assay may be useful in predicting the potential toxicity of chemicals in other organisms.     

Cadmium and chlorpyrifos inhibit cellular immune response in spleen of rats

Cadmium (Cd) and chlorpyrifos (CPF) are common pollutants coexisting in the environment, and both of them have been reported to have immunotoxicity to organisms. However, the joint effects of these two chemicals on the immune system are still unknown. In this study, we used CdCl₂ and CPF to study their combined effects on immune functions in the spleen of rats. In in vivo experiments, SD rats were exposed to different doses of CdCl₂ (0.7 and 6 mg kg^?1 body weight/day) and CPF (1.7 and 15 mg kg^?1 body weight/day) or their combinations for consecutive 28 days. The proliferation and cytokine production ability of the splenocytes isolated from the treated animals were assessed. In in vitro experiments, we used different concentrations of CdCl₂ and CPF to treat concanavalin A (Con A)‐induced splenocytes isolated from untreated rats. We found that the combination of CPF and high dose of CdCl₂ had a synergistic inhibitory effect on production of IFN‐γ by spleen cells induced by Con A. The in vitro results showed that two chemicals had different effects on the cell proliferation and cytokine production depending on the exposure doses and time. This result suggests that exposure to both CdCl₂ and CPF at the environmentally‐relevant low dose may be potentially more hazardous than exposure to each individual toxicant.     

Acute lethal toxicity of hydrocarbons and chlorinated hydrocarbons to two planktonic crustaceans: The key role of organism-water partitioning

The acute toxicities of 38 hydrocarbons and chlorinated hydrocarbons were determined for two planktonic crustaceans, freshwater Daphnia magna and saltwater Artemia. In both cases median lethal concentrations (LC₅₀ values) at 48 h and 24 h, respectively, were strongly correlated with aqueous solubility, there being little direct dependence on chemical structure. In the case of solid chemicals, it is suggested that the appropriate correlating solubility is that of the subcooled liquid chemical, not that of the solid. Many solids constrained by low water solubility are thus unable to achieve dissolved concentrations sufficient to cause acute toxicity to the test organisms. It is hypothesised that acute toxicity is non-selective and is controlled by organism-water partitioning so that each hydrocarbon group may contribute equally to toxicity and no single group of hydrocarbons is the dominant toxicant.The partitioning characteristics are expressed in two mathematical models which relate the LC₅₀ to the chemical properties of solubility and molar volume.     

Development of quantitative structure-activity relationship (QSAR) models to predict the carcinogenic potency of chemicals: I. Alternative toxicity measures as an estimator of carcinogenic potency

Determining the carcinogenicity and carcinogenic potency of new chemicals is both a labor-intensive and time-consuming process. In order to expedite the screening process, there is a need to identify alternative toxicity measures that may be used as surrogates for carcinogenic potency. Alternative toxicity measures for carcinogenic potency currently being used in the literature include lethal dose (dose that kills 50% of a study population [LD₅₀]), lowest-observed-adverse-effect-level (LOAEL) and maximum tolerated dose (MTD). The purpose of this study was to investigate the correlation between tumor dose (TD₅₀) and three alternative toxicity measures as an estimator of carcinogenic potency. A second aim of this study was to develop a Classification and Regression Tree (CART) between TD₅₀ and estimated/experimental predictor variables to predict the carcinogenic potency of new chemicals. Rat TD₅₀s of 590 structurally diverse chemicals were obtained from the Cancer Potency Database, and the three alternative toxicity measures considered in this study were estimated using TOPKAT^®, a toxicity estimation software. Though poor correlations were obtained between carcinogenic potency and the three alternative toxicity (both experimental and TOPKAT) measures for the CPDB chemicals, a CART developed using experimental data with no missing values as predictor variables provided reasonable estimates of TD₅₀ for nine chemicals that were part of an external validation set. However, if experimental values for the three alternative measures, mutagenicity and logP are not available in the literature, then either the CART developed using missing experimental values or estimated values may be used for making a prediction.     

Physiologically based pharmacokinetic model for developmental exposures to TCDD in the rat.

2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is a potent developmental toxicant in rodents, and these effects occur at exposures similar to background human body burdens. A physiologically based pharmacokinetic (PBPK) model can aid in quantitatively describing the relationship between exposure, dose, and response. The aim of this work was the development a PBPK model to describe the relationship between maternal TCDD exposure and fetal TCDD concentrations during critical windows of susceptibility in the rat. This PBPK model is a modification of an eight-compartment model that describes the adult female rat. The modified model reduces the compartments from eight to four maternal compartments (liver, fat, placenta and rest of the body). Activation of the placental compartment and a separate fetal compartment occurs during gestation. The systemic circulation connects the maternal compartments. The physiological and biochemical parameters were obtained from the literature. The model validation used experimental data from acute and subchronic exposures prior to and during gestation. The simulations predict the TCDD tissue concentrations of the maternal compartments within the standard deviation of the experimental data. The model overestimates the fetal concentrations by approximately a factor of two at low subchronic exposures, but does predict the fetal tissue concentrations within the range of the experimental data at the higher exposures. This model may provide a framework for the development of a human PBPK model to estimate fetal TCDD concentrations in human health risk assessments.     

Toxicokinetic-toxicodynamic modelling of survival of Gammarus pulex in multiple pulse exposures to propiconazole: model assumptions, calibration data requirements and predictive power

Toxicokinetic-toxicodynamic (TKTD) models quantify the time-course of internal concentration, which is defined by uptake, elimination and biotransformation (TK), and the processes which lead to the toxic effects (TD). TKTD models show potential in predicting pesticide effects in fluctuating concentrations, but the data requirements and validity of underlying model assumptions are not known. We calibrated TKTD models to predict survival of Gammarus pulex in propiconazole exposure and investigated the data requirements. In order to assess the need of TK in survival models, we included or excluded simulated internal concentrations based on pre-calibrated TK. Adding TK did not improve goodness of fits. Moreover, different types of calibration data could be used to model survival, which might affect model parameterization. We used two types of data for calibration: acute toxicity (standard LC50, 4 d) or pulsed toxicity data (total length 10?d). The calibration data set influenced how well the survival in the other exposure scenario was predicted (acute to pulsed scenario or vice versa). We also tested two contrasting assumptions in ecotoxicology: stochastic death and individual tolerance distribution. Neither assumption fitted to data better than the other. We observed in 10-d toxicity experiments that pulsed treatments killed more organisms than treatments with constant concentration. All treatments received the same dose, i.e. the time-weighted average concentration was equal. We studied mode of toxic action of propiconazole and it likely acts as a baseline toxicant in G. pulex during 10-days of exposure for the endpoint survival.     

Acute toxicity testing in cultures of mouse neuroblastoma cells.

Abstract Cultured mouse neuroblastoma cells (C1300) may be used as models for nerve cells since they have a number of properties in common with their normal counterparts in vivo. In order to test the possibility of using C1300 cells as alternativeto experimental animals when testing for acute toxicity, cells (clone 41A₃) were exposed to a number of common chemicals (CH₃HgCl, CdCl₂HgCl₂ ppDDT, n-butanol, benzene, dioxan, n-propanol, aceton and t-butanol). The toxic effect was quantified by measuring the degree of cell detachment in the cultures. The concentrations of chemicals that caused 25% of the total cell number to detach (TD₂₅) were used for comparison with LD₅₀ values. In spite of the very simplified situation in culture, where the toxicity of a substance is little or not at all influenced by factors like penetration, storage, metabolism and excretion a good correlation (corr. coeff. 0,98) was obtained between TD₂₅ values and LD₅₀ values. Good correlations between in vitro and in vivo tests have also been reported by others. One possible explanation to these findings could be simplified in vivo toxicokinetics of these substances when tested in high doses for general effects like animal death. If so, simple in vitro tests may be used for predicting acute toxicity of certain groups of substances.     

Comparison of models to analyze mortality data and derive concentration–time response relationship of inhaled chemicals

The derivation of thresholds for lethal effects for inhaled chemicals is a key issue in accidental risk management because they largely determine the outcome of land use planning, among which localization of habitations in the vicinity of a factory. This derivation is generally performed on the basis of rodent lethality data analyzed by statistical models able to extrapolate effects for different times and concentrations of exposure. A model commonly used in France is the standard probit model. In this model, effects is related to exposure concentration and duration according to the Haber’s law and considers that individual thresholds, corresponding to the maximum tolerated effects before dying, are log-normally distributed among the population. This approach has been criticized for its lack of biological parameters and its inability to treat data characterized by only one time of exposure. In order to improve the current state of modeling, we proposed three alternative models. Two of them (DEBtox and Haber-TKTD models) incorporate the kinetics of the chemicals. The third one (Loguniform model) is a linearization of the standard probit model. We evaluated their performance by analyzing real data and simulated data generated with each model. For data characterized by several times of exposure, the standard probit model outperformed all other models in terms of goodness of fits and estimation of parameters. For data characterized by only one time of exposure, only DEBtox model was able to fit the data and estimate parameters, provided we dispose of several observation times, typically just after exposure and a long period afterwards.     

Sublethal hydrocarbon phytotoxicity in the marine unicellular alga Pavlova lutheri Droop

The unicellular marine alga Pavlova lutheri Droop responds to petroleum hydrocarbons by changing its normal pattern of motility. This toxic response occurs at very low concentrations of hydrocarbons (0.1 ppm naphthalene, 0.5 ppm Kuwait crude oil, 0.01 ppm No. 2 diesel fuel), and is diagnostic and consistent. The regularity of this phytotoxic response enabled us to analyze it mathematically by fitting toxicokinetic models to the various responses at different toxicant concentrations and exposure times. The data were fitted to a class of generalized compartmental models of which a cumulative (time-dependent) and a compartmental (concentration-dependent) model are specialized cases. Observations using glucose and tryptophan (10^?3 M) were included in order to distinguish between generalized nonspecific and phytotoxic responses. Contrary to expectations, the phytotoxic response exhibited by this alga to the petroleum hydrocarbons was found to be dependent only on concentration of the toxicant, and not on time. Thus, the cell's vulnerability remains constant in time, and its vulnerability is not influenced by the length of exposure to these toxicants.     

The endothelin system and its role in acute myocardial infarction

Endothelin (ET)-1 is a potent coronary vasoconstrictor. On the heart, ET-1 is a potent positive inotrope and may be pro-arrhythmic. Plasma ET-1 levels are raised after acute myocardial infarction (AMI) and recanilisation in humans. This probably contributes to the coronary vasoconstriction that underlies the myocardial ischaemia and ventricular dysfunction at this time. During occlusion of the rat coronary artery, ventricular arrhythmias are reduced by ET_A receptor blockade. Short-term ET_A receptor blockade also reduces infarct size in animal models of AMI (coronary occlusion followed by reperfusion). Blockade of the endothelin-converting enzyme with SM-19712 reduced the infarct size in the rabbit model of AMI. ET_A receptor blockade is associated with coronary artery dilation in humans. As there are indications that ET_A receptor antagonists are protective in animal models of AMI, short-term ET_A receptor blockade should be considered for trial in human AMI.     

The Acute Exposure Guideline Level (AEGL) program: applications of physiologically based pharmacokinetic modeling

The primary aim of the Acute Exposure Guideline Level (AEGL) program is to develop scientifically credible limits for once-in-a-lifetime or rare acute inhalation exposures to high-priority, hazardous chemicals. The program was developed because of the need of communities for information on hazardous chemicals to assist in emergency planning, notification, and response, as well as the training of emergency response personnel. AEGLs are applicable to the general population, including children, the elderly, and other potentially susceptible subpopulations. AEGLs are the airborne concentrations of chemicals above which a person could experience notable discomfort or irritation (AEGL-1); serious, long-lasting health effects (AEGL-2); and life-threatening effects or death (AEGL-3). AEGLs are determined for five exposure periods (10 and 30 min and 1, 4, and 8 h). Physiologically based pharmacokinetic (PBPK) models can be very useful in the interspecies and time scaling often required here. PBPK models are used for the current article to predict AEGLs for trichlorethylene (TCE), based on the time course of TCE in the blood and/or brain of rats and humans. These AEGLs are compared to values obtained by standard time-scaling methods. Comprehensive toxicity assessment documents for each chemical under consideration are prepared by the National Advisory Committee for AEGLs, a panel comprised of representatives of federal, state, and local governmental agencies, as well as industry and private-sector organizations. The documents are developed according to National Research Council (NRC) guidelines and must be reviewed by the NRC Subcommittee on Acute Exposure Guideline Levels before becoming final. AEGLs for 18 chemicals have been published, and it is anticipated that 40 to 50 chemicals will be evaluated annually.     

ZD4054: a specific endothelin A receptor antagonist with promising activity in metastatic castration-resistant prostate cancer

Background: Overexpression of the endothelin A (ET_A) receptor has been found in a number of human cancer cell lines. Activation of the ET_A receptor by endothelin-1 (ET-1) promotes cell proliferation and survival in these tumors, whereas activation of the endothelin B (ET_B) receptor results in an opposing effect. Therefore, blockade of ET_A may have antitumor effects, while sparing ET_B-mediated effects such as induction of apoptosis and clearance of ET-1. Objective: ZD4054 is an orally bioavailable, specific ET_A antagonist currently being investigated in prostate cancer. In receptor-binding studies, ZD4054 only bound to ET_A, with no binding detected towards ET_B. Prostate cancer cell lines are known to produce ET-1 and there is a relative increase in expression of ET_A to ET_B in these cancers. There is also an association of greater ET_A expression in higher grade versus lower grade tumors, suggesting that ET_A may be involved in the malignant transformation process. As ET-1 may also mediate nociceptive effects and osteoblastic effects, there is much interest in clinically assessing ZD4054 in prostate cancer. Methods: The data describing the endothelin axis, the role of ET_A and ET_B in malignancy, and the effects of ET_A antagonist ZD4054 in prostate cancer, as demonstrated in preclinical and clinical studies, are reviewed. Results: Further investigation of ZD4054 in prostate cancer is warranted, and Phase III trials are already planned in patients with non-metastatic castrate-resistant prostate cancer (CRPC) with rising prostate specific antigen values, metastatic (asymptomatic) CRPC, and in metastatic CRPC in combination with docetaxel, assessing either differences in progression-free survival and overall survival or overall survival alone.     

Heterogeneity of toxicant response: sources of human variability.

While risk assessment models attempt to predict human risk to toxicant exposure, in many cases these models cannot account for the wide variety of human responses. This review addresses several primary sources of heterogeneity that may affect individual responses to drug or toxicant exposure. Consideration was given to genetic polymorphisms, age-related factors during development and senescence, gender differences associated with hormonal function, and preexisting diseases influenced by toxicant exposure. These selected examples demonstrate the need for additional steps in risk assessment that are needed to more accurately predict human responses to toxicants and drugs.     

Integration of dosimetry, exposure, and high-throughput screening data in chemical toxicity assessment

High-throughput in vitro toxicity screening can provide an efficient way to identify potential biological targets for chemicals. However, relying on nominal assay concentrations may misrepresent potential in vivo effects of these chemicals due to differences in bioavailability, clearance, and exposure. Hepatic metabolic clearance and plasma protein binding were experimentally measured for 239 ToxCast Phase I chemicals. The experimental data were used in a population-based in vitro-to-in vivo extrapolation model to estimate the daily human oral dose, called the oral equivalent dose, necessary to produce steady-state in vivo blood concentrations equivalent to in vitro AC(50) (concentration at 50% of maximum activity) or lowest effective concentration values across more than 500 in vitro assays. The estimated steady-state oral equivalent doses associated with the in vitro assays were compared with chronic aggregate human oral exposure estimates to assess whether in vitro bioactivity would be expected at the dose-equivalent level of human exposure. A total of 18 (9.9%) chemicals for which human oral exposure estimates were available had oral equivalent doses at levels equal to or less than the highest estimated U.S. population exposures. Ranking the chemicals by nominal assay concentrations would have resulted in different chemicals being prioritized. The in vitro assay endpoints with oral equivalent doses lower than the human exposure estimates included cell growth kinetics, cytokine and cytochrome P450 expression, and cytochrome P450 inhibition. The incorporation of dosimetry and exposure provide necessary context for interpretation of in vitro toxicity screening data and are important considerations in determining chemical testing priorities.     

A Critical Evaluation of Predicting Ocular lrritancy Potential from an in Vitro Cytotoxicity Assay

A Critical Evaluation of Predicting Ocular Irritancy Potentialfrom an in Vitro Cytotoxicity Assay. KENNAH, H. E., II, ALBULESCU,D., HIGNET, S., AND BARROW. C. S. (1989) Fundam Appl. Toxicol12,281-290. Numerous in vitro cytotoxicity assays have beenproposed as potential alternatives to the Draize eye irritancytest. The results reported, based upon the rank correlationof ocular irritancy with cytotoxicity, have been encouraging.However. direct calibration of in vivo to in vitro data utilizingseveral categories of chemicals has not been reported. Thisstudy evaluated the use of in vitro cytotoxicity data for predictingthe ocular irritancy potential of 24 chemicals(six surfactants,seven alcohols four ketones, four acetates and three aromatics).BALB/c 3T3 cells were grown overnight, then exposed for 30 minto at least four different concentrations of each chemical (expressedas volume percentage). Linear regression analysis of the logconcentration versus percentage of control growth was used tocalculate the concentration of toxicant that inhibited the normalgrowth rate by 50% (G150). The rank ordering of cytotoxicitybased upon the GI50s was surfactants > aromatics > alcohols> ketones or acetates. The larger molecular weight representativeof each senes (i.e., 2-ethyl-I-hexanol for alcohols) had lowerGI50 values than those of the lower molecular weight substances.The GI50 values were then directly calibrated against in vivoocular irritancy quantitated as percentage corneal swellingfollowing exposure of rabbits to the same test chemicals. Asignificant linear correlation between cytotoxicity and ocularirritancy was established only for surfactants and alcohols.For acetates, ketones, and aromatics there was little correlation.The overall poor correlation between cytotoxicity and ocularirritancy was attributed to differences in mechanisms of irritancy.The lack of correlation illustrates that in vitro cytotoxicitydata cannot be used to predict the ocular irritancy potentialof a broad spectrum of chemicals.     

An assessment of the potential of the microbial assay for risk assessment (MARA) for ecotoxicological testing

Rapid microscale toxicity tests make it possible to screen large numbers of compounds and greatly simplify toxicity identification evaluation and other effect directed chemical analyses of effluents or environmental samples. Tests using Vibrio fischeri (such as Microtox?) detect toxicants that cause non-specific narcosis, but are insensitive to other important classes of contaminants. The microbial assay for risk assessment (MARA) is a 24 h multi-species test that seeks to address this problem by using a battery of ten bacteria and a fungus. But there has been little independent evaluation of this test, and there is no published information on its sensitivity to pesticides. Here, we assess the performance of MARA using a range of toxicants including reference chemicals, fungicides and environmental samples. Mean MARA microbial toxic concentrations and IC₂₀s (20% Inhibitory concentrations) indicate the toxicant concentrations affecting the more sensitive micro-organisms, while the mean IC₅₀ (50% Inhibitory concentration) was found to be the concentration that was toxic to most MARA species. For the two fungicides tested, the yeast (Pichia anomalia) was the most sensitive of the ten MARA species, and was more sensitive than the nine other yeasts tested. The test may be particularly valuable for work with fungicides. Mean MARA IC₅₀s were comparable to values for nine other yeast species and the lowest individual IC₅₀s for each toxicant were comparable to reported IC₅₀s for Daphnia magna, Selenastrum capricornutum and Microtox? bioassays. MARA organisms exhibited more variable sensitivities, with the most sensitive organism being different for different samples, enhancing the likelihood of toxicity detection and giving a toxicity “fingerprint” that may help identify toxicants. The test, therefore, has great potential and would be valuable for ecotoxicological testing of pollutants.