A unifying concept for assessing toxicological interactions: changes in slope

Contaminant-by-environment interactive effects on animal behavior in the context of global change: Evidence from avian behavioral ecotoxicology

The potential for chemical contaminant exposure to interact with other stressors to affect animal behavioral responses to environmental variability is of mounting concern in the context of anthropogenic environmental change. We systematically reviewed the avian literature to evaluate evidence for contaminant-by-environment interactive effects on animal behavior, as birds are prominent models in behavioral ecotoxicology and global change research. We found that only 17 of 156 (10.9 %) avian behavioral ecotoxicological studies have explored contaminant-by-environment interactions. However, 13 (76.5 %) have found evidence for interactive effects, suggesting that contaminant-by-environment interactive effects on behavior are understudied but important. We draw on our review to develop a conceptual framework to understand such interactive effects from a behavioral reaction norm perspective. Our framework highlights four patterns in reaction norm shapes that can underlie contaminant-by-environment interactive effects on behavior, termed exacerbation, inhibition, mitigation and convergence. First, contamination can render individuals unable to maintain critical behaviors across gradients in additional stressors, exacerbating behavioral change (reaction norms steeper) and generating synergy. Second, contamination can inhibit behavioral adjustment to other stressors, antagonizing behavioral plasticity (reaction norms shallower). Third, a second stressor can mitigate (antagonize) toxicological effects of contamination, causing steeper reaction norms in highly contaminated individuals, with improvement of performance upon exposure to additional stress. Fourth, contamination can limit behavioral plasticity in response to permissive conditions, such that performance of more and less contaminated individuals converges under more stressful conditions. Diverse mechanisms might underlie such shape differences in reaction norms, including combined effects of contaminants and other stressors on endocrinology, energy balance, sensory systems, and physiological and cognitive limits. To encourage more research, we outline how the types of contaminant-by-environment interactive effects proposed in our framework might operate across multiple behavioral domains. We conclude by leveraging our review and framework to suggest priorities for future research.

Acute toxicity of pesticide mixtures to honey bees is generally additive, and well predicted by Concentration Addition

Understanding the frequency of non-additive effects of pesticides (synergism and antagonism) is important in the context of risk assessment. The goal of this study was to investigate the prevalence of non-additive effects of pesticides to honey bees (Apis mellifera). We investigated a large set of mixtures including insecticides and fungicides of different chemical modes of action and classes. The mixtures included represent a relevant sample of pesticides that are currently used globally. We investigated whether the experimental toxicity of the mixtures could be predicted based on the Concentration Addition (CA) model for acute contact and oral adult bee toxicity tests. We measured the degree of deviation from the additivity predictions of the experimental toxicity based on the well-known Mixture Deviation Ratio (MDR). Further, we investigated the appropriate MDR thresholds that should be used for the identification of non-additive effects based on acceptable rates for false positive (alpha) and true positive (beta) findings. We found that a deviation factor of MDR = 5 is a sound reference for labeling potential non-additive effects in acute adult bee experimental designs when assuming a typical Coefficient of Variation (CV%) = 100 in the determination of the LD₅₀ of a pesticide (a factor of 2× deviation in the LD ₅₀ resulting from inter-experimental variability). We found that only 2.4 % and 9 % of the mixtures evaluated had an MDR > 5 and MDR < 0.2, respectively. The frequency and magnitude of deviation from additivity found for bees in this study are consistent with those of other terrestrial and aquatic taxa. Our findings suggest that additivity is a good baseline for predicting the toxicity of pesticide mixtures to bees, and that the rare cases of synergy of pesticide mixtures to bees are not random but have a mechanistic basis.

Machine learning approach for hemorrhagic transformation prediction: Capturing predictors' interaction

Background and purpose

Patients with ischemic stroke frequently develop hemorrhagic transformation (HT), which could potentially worsen the prognosis. The objectives of the current study were to determine the incidence and predictors of HT, to evaluate predictor interaction, and to identify the optimal predicting models.

Methods

A prospective study included 360 patients with ischemic stroke, of whom 354 successfully continued the study. Patients were subjected to thorough general and neurological examination and T2 diffusion-weighted MRI, at admission and 1 week later to determine the incidence of HT. HT predictors were selected by a filter-based minimum redundancy maximum relevance (mRMR) algorithm independent of model performance. Several machine learning algorithms including multivariable logistic regression classifier (LRC), support vector classifier (SVC), random forest classifier (RFC), gradient boosting classifier (GBC), and multilayer perceptron classifier (MLPC) were optimized for HT prediction in a randomly selected half of the sample (training set) and tested in the other half of the sample (testing set). The model predictive performance was evaluated using receiver operator characteristic (ROC) and visualized by observing case distribution relative to the models' predicted three-dimensional (3D) hypothesis spaces within the testing dataset true feature space. The interaction between predictors was investigated using generalized additive modeling (GAM).

Results

The incidence of HT in patients with ischemic stroke was 19.8%. Infarction size, cerebral microbleeds (CMB), and the National Institute of Health stroke scale (NIHSS) were identified as the best HT predictors. RFC (AUC: 0.91, 95% CI: 0.85-0.95) and GBC (AUC: 0.91, 95% CI: 0.86-0.95) demonstrated significantly superior performance compared to LRC (AUC: 0.85, 95% CI: 0.79-0.91) and MLPC (AUC: 0.85, 95% CI: 0.78-0.92). SVC (AUC: 0.90, 95% CI: 0.85-0.94) outperformed LRC and MLPC but did not reach statistical significance. LRC and MLPC did not show significant differences. The best models' 3D hypothesis spaces demonstrated non-linear decision boundaries suggesting an interaction between predictor variables. GAM analysis demonstrated a linear and non-linear significant interaction between NIHSS and CMB and between NIHSS and infarction size, respectively.

Conclusion

Cerebral microbleeds, NIHSS, and infarction size were identified as HT predictors. The best predicting models were RFC and GBC capable of capturing nonlinear interaction between predictors. Predictor interaction suggests a dynamic, rather than, fixed cutoff risk value for any of these predictors.

Understanding the combined effects of multiple stressors: A new perspective on a longstanding challenge

Wildlife populations and their habitats are exposed to an expanding diversity and intensity of stressors caused by human activities, within the broader context of natural processes and increasing pressure from climate change. Estimating how these multiple stressors affect individuals, populations, and ecosystems is thus of growing importance. However, their combined effects often cannot be predicted reliably from the individual effects of each stressor, and we lack the mechanistic understanding and analytical tools to predict their joint outcomes. We review the science of multiple stressors and present a conceptual framework that captures and reconciles the variety of existing approaches for assessing combined effects. Specifically, we show that all approaches lie along a spectrum, reflecting increasing assumptions about the mechanisms that regulate the action of single stressors and their combined effects. An emphasis on mechanisms improves analytical precision and predictive power but could introduce bias if the underlying assumptions are incorrect. A purely empirical approach has less risk of bias but requires adequate data on the effects of the full range of anticipated combinations of stressor types and magnitudes. We illustrate how this spectrum can be formalised into specific analytical methods, using an example of North Atlantic right whales feeding on limited prey resources while simultaneously being affected by entanglement in fishing gear. In practice, case-specific management needs and data availability will guide the exploration of the stressor combinations of interest and the selection of a suitable trade-off between precision and bias. We argue that the primary goal for adaptive management should be to identify the most practical and effective ways to remove or reduce specific combinations of stressors, bringing the risk of adverse impacts on populations and ecosystems below acceptable thresholds.

Varying impact of neonicotinoid insecticide and acute bee paralysis virus across castes and colonies of black garden ants, Lasius niger (Hymenoptera: Formicidae)

Pesticides and pathogens are known drivers of declines in global entomofauna. However, interactions between pesticides and viruses, which could range from antagonistic, over additive to synergistic, are poorly understood in ants. Here, we show that in ants the impact of single and combined pesticide and virus stressors can vary across castes and at the colony level. A fully-crossed laboratory assay was used to evaluate interactions between a sublethal dose of the neonicotinoid thiamethoxam and Acute bee paralysis virus (ABPV) in black garden ants, Lasius niger. After monitoring colonies over 64 weeks, body mass, neonicotinoid residues and virus titres of workers and queens, as well as worker behavioural activity were measured. ABPV, but not thiamethoxam, reduced activity of workers. Neonicotinoid exposure resulted in reduced body mass of workers, but not of queens. Further, thiamethoxam facilitated ABPV infections in queens, but not in workers. Overall, virus exposure did not compromise detoxification and body mass, but one colony showed high virus titres and worker mortality. Although the data suggest additive effects at the level of individuals and castes, co-exposure with both stressors elicited antagonistic effects on colony size. Our results create demand for long-term holistic risk assessment of individual stressors and their interactions to protect biodiversity.

A Novel Approach to Chemical Mixture Risk Assessment-Linking Data from Population-Based Epidemiology and Experimental Animal Tests

Humans are continuously exposed to chemicals with suspected or proven endocrine disrupting chemicals (EDCs). Risk management of EDCs presents a major unmet challenge because the available data for adverse health effects are generated by examining one compound at a time, whereas real-life exposures are to mixtures of chemicals. In this work, we integrate epidemiological and experimental evidence toward a whole mixture strategy for risk assessment. To illustrate, we conduct the following four steps in a case study: (1) identification of single EDCs ("bad actors")-measured in prenatal blood/urine in the SELMA study-that are associated with a shorter anogenital distance (AGD) in baby boys; (2) definition and construction of a "typical" mixture consisting of the "bad actors" identified in Step 1; (3) experimentally testing this mixture in an in vivo animal model to estimate a dose-response relationship and determine a point of departure (i.e., reference dose [RfD]) associated with an adverse health outcome; and (4) use a statistical measure of "sufficient similarity" to compare the experimental RfD (from Step 3) to the exposure measured in the human population and generate a "similar mixture risk indicator" (SMRI). The objective of this exercise is to generate a proof of concept for the systematic integration of epidemiological and experimental evidence with mixture risk assessment strategies. Using a whole mixture approach, we could find a higher rate of pregnant women under risk (13%) when comparing with the data from more traditional models of additivity (3%), or a compound-by-compound strategy (1.6%).

Using Various Nonlinear Response Surfaces for Mathematical Description of the Type of Combined Toxicity

The article considers the problem of characterizing the type of combined action produced by a mixture of toxic substances with the help of nonlinear response functions. Most attention is given to second-order models: the linear model with a cross-term and the quadratic model. General propositions are formulated based on the data on combined toxicity patterns previously obtained by the Ekaterinburg nanotoxicology team in animal experiments and analyzed with the help of the linear model with a cross-term. It is shown now that the quadratic model features these general characteristics in full measure, but interpretation of combined toxicity types based on isobolograms obtained by the quadratic model is more difficult. This suggests that where both models ensure a comparable quality of combined toxicity type identification, it would be enough to use the linear model with a cross-term.

When the Whole is Not Greater than the Sum of the Parts: A Critical Review of Laboratory Bioassay Effects Testing for Insecticidal Protein Interactions

Many studies have been conducted to investigate synergism among insecticidal proteins; however, a consensus on minimal data requirements and interpretation is lacking. While some have concluded that all additive predictive-type models should be abandoned, we advocate that additivity models can remain useful as assessment tools and that an appropriately designed interaction study will never systematically underestimate the existence of synergism, irrespective of which additivity model (or none at all) may be used. To generate the most meaningful synergy assessment datasets in support of safety assessments, we highlight two beneficial steps to follow: (i) select a testing model which is the most consistent with current knowledge regarding the action of the insecticidal proteins and (ii) avoid using bioassay methods which may result in excess response heterogeneity. We also outline other experimental design elements to consider, which may be crucial for conducting future studies of this type. A contrast of underlying testing assumptions associated with the additivity models is provided, along with a comprehensive review of interaction data for Cry1, Cry2, Cry3, Cry9, and Vip3A insecticidal proteins. Our review captures four recurrent findings: i) experiments reporting synergistic interactions are a minority, ii) the degree of synergism reported is low in magnitude, iii) reported interactions are sometimes equivocal/inconclusive due to unconfirmed model assumptions or other bioassay challenges, and iv) due to biological response variation many of the reported interactions may be artefactual. A brief overview of the positioning of interaction testing data in safety assessments of GM food crops is also provided.

Mixture Concentration-Response Modeling Reveals Antagonistic Effects of Estradiol and Genistein in Combination on Brain Aromatase Gene (cyp19a1b) in Zebrafish

Comprehension of compound interactions in mixtures is of increasing interest to scientists, especially from a perspective of mixture risk assessment. However, most of conducted studies have been dedicated to the effects on gonads, while only few of them were. interested in the effects on the central nervous system which is a known target for estrogenic compounds. In the present study, the effects of estradiol (E2), a natural estrogen, and genistein (GEN), a phyto-estrogen, on the brain ER-regulated cyp19a1b gene in radial glial cells were investigated alone and in mixtures. For that, zebrafish-specific in vitro and in vivo bioassays were used. In U251-MG transactivation assays, E2 and GEN produced antagonistic effects at low mixture concentrations. In the cyp19a1b-GFP transgenic zebrafish, this antagonism was observed at all ratios and all concentrations of mixtures, confirming the in vitro effects. In the present study, we confirm (i) that our in vitro and in vivo biological models are valuable complementary tools to assess the estrogenic potency of chemicals both alone and in mixtures; (ii) the usefulness of the ray design approach combined with the concentration-addition modeling to highlight interactions between mixture components.

Additive Dose Response Models: Explicit Formulation and the Loewe Additivity Consistency Condition

High-throughput techniques allow for massive screening of drug combinations. To find combinations that exhibit an interaction effect, one filters for promising compound combinations by comparing to a response without interaction. A common principle for no interaction is Loewe Additivity which is based on the assumption that no compound interacts with itself and that two doses from different compounds having the same effect are equivalent. It then should not matter whether a component is replaced by the other or vice versa. We call this assumption the Loewe Additivity Consistency Condition (LACC). We derive explicit and implicit null reference models from the Loewe Additivity principle that are equivalent when the LACC holds. Of these two formulations, the implicit formulation is the known General Isobole Equation (Loewe, 1928), whereas the explicit one is the novel contribution. The LACC is violated in a significant number of cases. In this scenario the models make different predictions. We analyze two data sets of drug screening that are non-interactive (Cokol et al., 2011; Yadav et al., 2015) and show that the LACC is mostly violated and Loewe Additivity not defined. Further, we compare the measurements of the non-interactive cases of both data sets to the theoretical null reference models in terms of bias and mean squared error. We demonstrate that the explicit formulation of the null reference model leads to smaller mean squared errors than the implicit one and is much faster to compute.

Individual and combined cytotoxicity assessment of zearalenone with ochratoxin A or α-zearalenol by full factorial design

The combined mycotoxins zearalenone (ZEA) with ochratoxin A (OTA) or α-zearalenol (α-ZOL) are frequently found together in milk. Toxicological data concerning the combined effects of these mycotoxins are sparse. In present study, individual and combined ZEA, OTA and α-ZOL caused cytotoxicity and oxidative damage, including reductions in intracellular superoxide dismutase and glutathione peroxidase activities and glutathione content, along with increases in malonaldehyde content on human Hep G2 cells after 48 h of exposure. Among individual mycotoxins, OTA had the greatest cytotoxic effect followed by α-ZOL. Compared with individual mycotoxins, combinations produced more serious negative effects, more importantly, ZEA + OTA was antagonistic for these effects, whereas ZEA + α-ZOL was antagonistic at low concentrations, but synergistic at high concentrations of ZEA, which were evaluated by 3 × 3 full factorial analysis and estimated marginal means plots. Our results also demonstrated a significant correlation between cytotoxicity and oxidative damage in response to these combinations.

Determining lower threshold concentrations for synergistic effects

Though only occurring rarely, synergistic interactions between chemicals in mixtures have long been a point of focus. Most studies analyzing synergistic interactions used unrealistically high chemical concentrations. The aim of the present study is to determine the threshold concentration below which proven synergists cease to act as synergists towards the aquatic crustacean Daphnia magna. To do this, we compared several approaches and test-setups to evaluate which approach gives the most conservative estimate for the lower threshold for synergy for three known azole synergists. We focus on synergistic interactions between the pyrethroid insecticide, alpha-cypermethrin, and one of the three azole fungicides prochloraz, propiconazole or epoxiconazole measured on Daphnia magna immobilization. Three different experimental setups were applied: A standard 48h acute toxicity test, an adapted 48h test using passive dosing for constant chemical exposure concentrations, and a 14-day test. Synergy was defined as occuring in mixtures where either EC₅₀ values decreased more than two-fold below what was predicted by concentration addition (horizontal assessment) or as mixtures where the fraction of immobile organisms increased more than two-fold above what was predicted by independent action (vertical assessment). All three tests confirmed the hypothesis of the existence of a lower azole threshold concentration below which no synergistic interaction was observed. The lower threshold concentration, however, decreased with increasing test duration from 0.026±0.013μM (9.794±4.897μgL^-1), 0.425±0.089μM (145.435±30.46μgL^-1) and 0.757±0.253μM (249.659±83.44μgL^-1) for prochloraz, propiconazole and epoxiconazole in standard 48h toxicity tests to 0.015±0.004μM (5.651±1.507μgL^-1), 0.145±0.025μM (49.619±8.555μgL^-1) and 0.122±0.0417μM (40.236±13.75μgL^-1), respectively, in the 14-days tests. Testing synergy in relation to concentration addition provided the most conservative values. The threshold values for the vertical assessments in tests where the two could be compared were in general 1.2 to 4.7 fold higher than the horizontal assessments. Using passive dosing rather than dilution series or spiking did not lower the threshold significantly. Below the threshold for synergy, slight antagony could often be observed. This is most likely due to induction of enzymes active in metabolization of alpha-cypermethrin. The results emphasize the importance of test duration when assessing synergy, but also show that azole concentrations within the typically monitored range of up to 0.5μgL^-1 are not likely to cause severe synergy concerning Daphnia magna immobilization.

Current Trends in Multidrug Optimization: An Alley of Future Successful Treatment of Complex Disorders

The identification of effective and long-lasting cancer therapies still remains elusive, partially due to patient and tumor heterogeneity, acquired drug resistance, and single-drug dose-limiting toxicities. The use of drug combinations may help to overcome some limitations of current cancer therapies by challenging the robustness and redundancy of biological processes. However, effective drug combination optimization requires the careful consideration of numerous parameters. The complexity of this optimization problem is clearly nontrivial and likely requires the assistance of advanced heuristic optimization techniques. In the current review, we discuss the application of optimization techniques for the identification of optimal drug combinations. More specifically, we focus on the application of phenotype-based screening approaches in the field of cancer therapy. These methods are divided into three categories: (1) modeling methods, (2) model-free approaches based on biological search algorithms, and (3) merged approaches, particularly phenotypically driven network biology methods and computation network models relying on phenotypic data. In addition to a brief description of each approach, we include a critical discussion of the advantages and disadvantages of each method, with a strong focus on the limitations and considerations needed to successfully apply such methods in biological research.

Current Trends in Multidrug Optimization

The identification of effective and long-lasting cancer therapies still remains elusive, partially due to patient and tumor heterogeneity, acquired drug resistance, and single-drug dose-limiting toxicities. The use of drug combinations may help to overcome some limitations of current cancer therapies by challenging the robustness and redundancy of biological processes. However, effective drug combination optimization requires the careful consideration of numerous parameters. The complexity of this optimization problem is clearly nontrivial and likely requires the assistance of advanced heuristic optimization techniques. In the current review, we discuss the application of optimization techniques for the identification of optimal drug combinations. More specifically, we focus on the application of phenotype-based screening approaches in the field of cancer therapy. These methods are divided into three categories: (1) modeling methods, (2) model-free approaches based on biological search algorithms, and (3) merged approaches, particularly phenotypically driven network biology methods and computation network models relying on phenotypic data. In addition to a brief description of each approach, we include a critical discussion of the advantages and disadvantages of each method, with a strong focus on the limitations and considerations needed to successfully apply such methods in biological research.

Interactions between cadmium and decabrominated diphenyl ether on blood cells count in rats-Multiple factorial regression analysis

The objective of this study was to assess toxicity of Cd and BDE-209 mixture on haematological parameters in subacutely exposed rats and to determine the presence and type of interactions between these two chemicals using multiple factorial regression analysis. Furthermore, for the assessment of interaction type, an isobologram based methodology was applied and compared with multiple factorial regression analysis. Chemicals were given by oral gavage to the male Wistar rats weighing 200-240g for 28days. Animals were divided in 16 groups (8/group): control vehiculum group, three groups of rats were treated with 2.5, 7.5 or 15mg Cd/kg/day. These doses were chosen on the bases of literature data and reflect relatively high Cd environmental exposure, three groups of rats were treated with 1000, 2000 or 4000mg BDE-209/kg/bw/day, doses proved to induce toxic effects in rats. Furthermore, nine groups of animals were treated with different mixtures of Cd and BDE-209 containing doses of Cd and BDE-209 stated above. Blood samples were taken at the end of experiment and red blood cells, white blood cells and platelets counts were determined. For interaction assessment multiple factorial regression analysis and fitted isobologram approach were used. In this study, we focused on multiple factorial regression analysis as a method for interaction assessment. We also investigated the interactions between Cd and BDE-209 by the derived model for the description of the obtained fitted isobologram curves. Current study indicated that co-exposure to Cd and BDE-209 can result in significant decrease in RBC count, increase in WBC count and decrease in PLT count, when compared with controls. Multiple factorial regression analysis used for the assessment of interactions type between Cd and BDE-209 indicated synergism for the effect on RBC count and no interactions i.e. additivity for the effects on WBC and PLT counts. On the other hand, isobologram based approach showed slight antagonism for the effects on RBC and WBC while no interactions were proved for the joint effect on PLT count. These results confirm that the assessment of interactions between chemicals in the mixture greatly depends on the concept or method used for this evaluation.

Impact of co-exposure with lead and decabromodiphenyl ether (BDE-209) on thyroid function in zebrafish larvae

Polybrominated diphenyl ethers (PBDEs) and metals are the main contaminants at waste electrical and electronic equipment ("e-waste") recycling sites. However, the potential environmental health effects of mixtures of PBDEs and metals are not known. We investigated co-exposure of lead (Pb) with decabromodiphenyl ether (BDE-209) on thyroid function in zebrafish larvae. Seven groups of embryos/larvae of zebrafish were treated with Pb (0, 2, 5, 10, 15, 20, and 30 μg/L), six groups were exposed to BDE-209 (0, 50, 100, 200, 400, and 800 μg/L), and nine groups of zebrafish larvae were treated with Pb and BDE-209 (5, 10, and 20 μg/L Pb; 50, 100, and 200 μg/L BDE-209). Embryos/larvae were exposed from 2h post-fertilization (hpf) until 144 hpf, and thyroid hormone (TH) content measured. Pb exposure significantly decreased whole-body TH contents (triiodothyroxine (T3) and thyroxine (T4)) but BDE-209 exposure significantly increased T3 and T4 levels. Pb or BDE-209 treatment alone caused a predicted downregulation of TH transport (i.e., expression of the mRNA or proteins of transthyretin). Chemical analyses showed Pb uptake to be increased by BDE-209, but BDE-209 bioconcentration was decreased and the ability to metabolize BDE-209 was reduced in the presence of Pb. We also found that a mixture of the two chemicals had a synergistic effect on TH levels in zebrafish.

Cadmium and decabrominated diphenyl ether mixture: In vitro evaluation of cytotoxic, prooxidative and genotoxic effects

In order to look into the combined effects of Cd and BDE-209 in vitro, this study was aimed at examining cytotoxic and genotoxic effects using the human colon carcinoma cell line (SW 480) as a biological test system as well as to determine if ROS production was one of the possible mechanisms of their mixture action. This cell line was chosen since ingestion of contaminated food/water represents an important route of exposure to both Cd and BDE-209, which is why intestinal cells are a common target for the contaminants present in food and water. Cells were treated with single Cd in concentrations of 2.5, 7.5 or 15μg Cd/mL (corresponding to 22, 67 or 134μM), single BDE-209 in concentrations of 2.5, 5 or 10μg BDE209/mL (corresponding to 2.5, 5 or 10μM), and their mixtures (design 3×3). Mixture of Cd and BDE-209 has shown clear potential to reduce the viability of SW 480 cells, as evidenced by cytotoxicity associated with ROS generation. Factorial regression models used to identify type of interaction revealed synergism related to mixture citotoxicity and additive interaction for the effect on ROS production. The results from this introductory study could contribute to the issue of possible adverse effects associated with co-exposure and body burden with two persistent environmental pollutants, Cd and BDE-209.

The identification of complex interactions in epidemiology and toxicology: a simulation study of boosted regression trees

BACKGROUND

There is a need to evaluate complex interaction effects on human health, such as those induced by mixtures of environmental contaminants. The usual approach is to formulate an additive statistical model and check for departures using product terms between the variables of interest. In this paper, we present an approach to search for interaction effects among several variables using boosted regression trees.

METHODS

We simulate a continuous outcome from real data on 27 environmental contaminants, some of which are correlated, and test the method's ability to uncover the simulated interactions. The simulated outcome contains one four-way interaction, one non-linear effect and one interaction between a continuous variable and a binary variable. Four scenarios reflecting different strengths of association are simulated. We illustrate the method using real data.

RESULTS

The method succeeded in identifying the true interactions in all scenarios except where the association was weakest. Some spurious interactions were also found, however. The method was also capable to identify interactions in the real data set.

CONCLUSIONS

We conclude that boosted regression trees can be used to uncover complex interaction effects in epidemiological studies.

Evaluation of the ecotoxicity of pollutants with bioluminescent microorganisms

This chapter deals with the use of bioluminescent microorganisms in environmental monitoring, particularly in the assessment of the ecotoxicity of pollutants. Toxicity bioassays based on bioluminescent microorganisms are an interesting complement to classical toxicity assays, providing easiness of use, rapid response, mass production, and cost effectiveness. A description of the characteristics and main environmental applications in ecotoxicity testing of naturally bioluminescent microorganisms, covering bacteria and eukaryotes such as fungi and dinoglagellates, is reported in this chapter. The main features and applications of a wide variety of recombinant bioluminescent microorganisms, both prokaryotic and eukaryotic, are also summarized and critically considered. Quantitative structure-activity relationship models and hormesis are two important concepts in ecotoxicology; bioluminescent microorganisms have played a pivotal role in their development. As pollutants usually occur in complex mixtures in the environment, the use of both natural and recombinant bioluminescent microorganisms to assess mixture toxicity has been discussed. The main information has been summarized in tables, allowing quick consultation of the variety of luminescent organisms, bioluminescence gene systems, commercially available bioluminescent tests, environmental applications, and relevant references.

Optimal design for the precise estimation of an interaction threshold: the impact of exposure to a mixture of 18 polyhalogenated aromatic hydrocarbons

Traditional additivity models provide little flexibility in modeling the dose-response relationships of the single agents in a mixture. While the flexible single chemical required (FSCR) methods allow greater flexibility, its implicit nature is an obstacle in the formation of the parameter covariance matrix, which forms the basis for many statistical optimality design criteria. The goal of this effort is to develop a method for constructing the parameter covariance matrix for the FSCR models, so that (local) alphabetic optimality criteria can be applied. Data from Crofton et al. are provided as motivation; in an experiment designed to determine the effect of 18 polyhalogenated aromatic hydrocarbons on serum total thyroxine (T(4)), the interaction among the chemicals was statistically significant. Gennings et al. fit the FSCR interaction threshold model to the data. The resulting estimate of the interaction threshold was positive and within the observed dose region, providing evidence of a dose-dependent interaction. However, the corresponding likelihood-ratio-based confidence interval was wide and included zero. In order to more precisely estimate the location of the interaction threshold, supplemental data are required. Using the available data as the first stage, the Ds-optimal second-stage design criterion was applied to minimize the variance of the hypothesized interaction threshold. Practical concerns associated with the resulting design are discussed and addressed using the penalized optimality criterion. Results demonstrate that the penalized Ds-optimal second-stage design can be used to more precisely define the interaction threshold while maintaining the characteristics deemed important in practice.

Biogeographical analysis of chemical co-occurrence data to identify priorities for mixtures research

A challenge with multiple chemical risk assessment is the need to consider the joint behavior of chemicals in mixtures. To address this need, pharmacologists and toxicologists have developed methods over the years to evaluate and test chemical interaction. In practice, however, testing of chemical interaction more often comprises ad hoc binary combinations and rarely examines higher order combinations. One explanation for this practice is the belief that there are simply too many possible combinations of chemicals to consider. Indeed, under stochastic conditions the possible number of chemical combinations scales geometrically as the pool of chemicals increases. However, the occurrence of chemicals in the environment is determined by factors, economic in part, which favor some chemicals over others. We investigate methods from the field of biogeography, originally developed to study avian species co-occurrence patterns, and adapt these approaches to examine chemical co-occurrence. These methods were applied to a national survey of pesticide residues in 168 child care centers from across the country. Our findings show that pesticide co-occurrence in the child care center was not random but highly structured, leading to the co-occurrence of specific pesticide combinations. Thus, ecological studies of species co-occurrence parallel the issue of chemical co-occurrence at specific locations. Both are driven by processes that introduce structure in the pattern of co-occurrence. We conclude that the biogeographical tools used to determine when this structure occurs in ecological studies are relevant to evaluations of pesticide mixtures for exposure and risk assessment.

Dose-response modelling with two agents: application to the bioassay of oil and shoreline cleaning agents

Single and joint effects of hydrocarbons and a shoreline cleaning agent (SCA) were studied by measuring the inhibition of the larval growth of sea urchin. Different dosage methods of hydrophobic compounds were compared. The results obtained in the evaluation of CytoSol toxicity revealed that the method of variable dilution of water accommodated fraction (WAF) led to the more conservative toxicological approach. Regarding to Libyan oil, the use of DMSO as carrier allowed us the evaluation of its potential toxicity in comparison with the limitations imposed to the use of WAF method. A reparametrised form of the Weibull equation was slightly modified to be useful for dose-response analysis. This was the basis for modelling single sigmoid responses, which were used to simulate biphasic profiles with addition of effects and to describe both the concentration addition (CA) and independent action (IA) hypotheses. In all cases, its descriptive ability was graphically and statistically satisfactory. The IA model was the best option to explain the combined experimental responses obtained.

Mixtures of environmental pollutants: effects on microorganisms and their activities in soils

Soil is the ultimate sink for most contaminants and rarely has only a single contaminant. More than is generally acknowledge, environmental pollutants exist as mixtures (organic-organic, inorganic-inorganic, and organic-inorganic). It is much more difficult to study chemical mixtures than individual chemicals, especially in the complex soil environment. Similarly, understanding the toxicity of a chemical mixture on different microbial species is much more complex, time consuming and expensive, because multiple testing designs are needed for an increased array of variables. Therefore, until now, scientific enquiries worldwide have extensively addressed the effects of only individual pollutants toward nontarget microorganisms. In this review, we emphasize the present status of research on (i) the environmental occurrence of pollutant mixtures; (ii) the interactions between pollutant mixtures and ecologically beneficial microorganisms; and (iii) the impact of such interactions on environmental quality. We also address the limitations of traditional cultivation based methods for monitoring the effects of pollutant mixtures on microorganisms. Long-term monitoring of the effects of pollutant mixtures on microorganisms, particularly in soil and aquatic ecosystems, has received little attention. Microbial communities that can degrade or can degrade or can develop tolerance to, or are inhibited by chemical mixtures greatly contribute to resilience and resistance in soil environments. We also stress in this review the important emerging trend associated with the employment of molecular methods for establishing the effects of pollutant mixtures on microbial communities. There is currently a lack of sufficient cogent toxicological data on chemical mixtures for making informed decision making in risk assessment by regulators. Therefore, not only more toxicology information on mixtures is needed but also there is an urgent need to generate sufficient, suitable, and long-term modeling data that have higher predictability when assessing pollutant mixture effects on microorganisms. Such data would improve risk assessment at contaminated sites and would help devise more effective bioremediation strategies.

Predictive Modeling of a Mixture of Thyroid Hormone Disrupting Chemicals That Affect Production and Clearance of Thyroxine

Thyroid hormone (TH) disrupting compounds interfere with both thyroidal and extrathyroidal mechanisms to decrease circulating thyroxine (T4). This research tested the hypothesis that serum T4 concentrations of rodents exposed to a mixture of both TH synthesis inhibitors (pesticides) and stimulators of T4 clearance in the liver (polyhalogenated aromatic hydrocarbons, PHAHs) could be best predicted by an integrated addition model. Female Long-Evans rats, 23 days of age, were dosed with dilutions of a mixture of 18 PHAHs (2 dioxins, 4 dibenzofurans, and 12 PCBs, including dioxin-like and non-dioxin like PCBs) and a mixture of 3 pesticides (thiram, pronamide, and mancozeb) for four consecutive days. Serum was collected 24 hours after the last exposure and T4 concentrations were measured by radioimmunoassay. Animals exposed to the highest dose of the mixture experienced a 45% decrease in serum T4. Three additivity model predictions (dose addition, effect addition, and integrated addition) were generated based on single chemical data, and the results were compared. Effect addition overestimated the effect produced by the combination of all 21 chemicals. The results of the dose- and integrated-addition models were similar, and both provided better predictions than the effect-addition model. These results support the use of dose- and integrated additivity models in predicting the effects of complex mixtures.

The possible role of ammonia toxicity on the exposure, deposition, retention, and the bioavailability of nicotine during smoking

A complete and rigorous review is presented of the possible effect(s) of ammonia on the exposure, deposition and retention of nicotine during smoking and the bioavailability of nicotine to the smoker. There are no toxicological data in humans regarding ammonia exposure within the context of tobacco smoke. Extrapolation from occupational exposure of ammonia to smoking in humans suggests minimal, non-toxicological effects, if any. No direct study has examined the effect of the ammonia on the total rate or amount of nicotine reaching the arterial bloodstream or brains of smokers. Machine-smoking methods have been reported which accurately quantify >99% of the nicotine in mainstream (MS) smoke for a wide variety of commercial and test cigarettes, including a series of experimental cigarettes having a range in MS smoke ammonia yields using the US Federal Trade Commission (FTC) protocol. However, the actual exposure of nicotine to smokers depends on their own smoking behavior. The nicotine ring system is relatively thermally stable. Protonated nicotine forms nicotine which evaporates before the nicotine ring system decomposes. The experimental data indicate that neither nicotine transfer from tobacco to MS smoke nor nicotine bioavailability to the smoker increases with an increase in any of the following properties: tobacco soluble ammonia, MS smoke ammonia, "tobacco pH" or "smoke pH" at levels found in commercial cigarettes. Gas phase nicotine deposits primarily in the mouth and upper respiratory tract. To the extent that ammonia increases the deposition of nicotine in the buccal cavity and upper respiratory tract during smoking, the total rate and amount of nicotine into the arterial bloodstream and to the central nervous system will decrease. Charged nicotine analogues are actively transported in a number of tissues. This active transport system appears to be insensitive to pH and the form of nicotine in the biological milieu, suggesting that protonated nicotine may be a substrate for active transport. Neither "smoke pH" of commercial cigarettes nor "smoke pHeff" nor the fraction of non-protonated nicotine in tobacco smoke particulate matter are useful, practical smoke parameters for providing understanding or predictability of nicotine bioavailability to smokers. Greater than 95% of both ammonia and nicotine are in the gas phase of environmental tobacco, and both are likely to deposit in the buccal cavity and upper respiratory tract following exposure.

A mathematical approach to study combined effects of toxicants in vitro: Evaluation of the Bliss independence criterion and the Loewe additivity model

The study of interactions among toxicants is of fundamental interest and practical importance in toxicological sciences. However, a final agreement on the definition of agent interaction is lacking, and therefore, particular care should be adopted when using the terms additivity, synergism and antagonism, unless the exact toxicological pathways of the compounds studied are known. Two main different approaches, the Bliss independence criterion and the Loewe additivity model, have been generally used in co-exposure experiments. In some cases, they can present dramatically different results, depending on the slope of the pure dose-response curves of single substances. Here, we consider both models in in vitro experiments, where the dose-response curves can be extrapolated. Advantages and limitations of both approaches are discussed, using several mathematical simulations to better explain them, and applying the Hill function for the dose-response model curve. Overall we conclude that the Loewe additivity model is slightly preferable because of a general higher biological plausibility. However, which model to use must be determined case by case and the evaluation can be aided by experimental approaches, such as the study of multiple biomarkers and asynchronous exposures.

Analysis of the interaction of phytoestrogens and synthetic chemicals: an in vitro/in vivo comparison

In the evaluation of chemical mixture toxicity, it is desirable to develop an evaluation paradigm which incorporates some critical attributes of real world exposures, particularly low dose levels, larger numbers of chemicals, and chemicals from synthetic and natural sources. This study evaluated the impact of low level exposure to a mixture of six synthetic chemicals (SC) under conditions of co-exposure to various levels of plant-derived phytoestrogen (PE) compounds. Estrogenic activity was evaluated using an in vitro human estrogen receptor (ER) transcriptional activation assay and an in vivo immature rat uterotrophic assay. Initially, dose-response curves were characterized for each of the six SCs (methoxyclor, o,p-DDT, octylphenol, bisphenol A, beta-hexachlorocyclohexane, 2,3-bis(4-hydroxyphenyl)-propionitrile) in each of the assays. The six SCs were then combined at equipotent ratios and tested at 5-6 dose levels spanning from very low, sub-threshold levels, to a dose in which every chemical in the mixture was at its individual estrogenic response threshold. The SC mixtures also were tested in the absence or presence of 5-6 different levels of PEs, for a total of 36 (in vitro) or 25 (in vivo) treatment groups. Both in vitro and in vivo, low concentrations of the SC mixture failed to increase estrogenic responses relative to those induced by PEs alone. However, significant increases in response occurred when each chemical in the SC mixture was near or above its individual response threshold. In vitro, interactions between high-doses of SCs and PEs were greater than additive, whereas mixtures of SCs in the absence of PEs interacted in a less than additive fashion. In vivo, the SC and PE mixture responses were consistent with additivity. These data illustrate a novel approach for incorporating key attributes of real world exposures in chemical mixture toxicity assessments, and suggest that chemical mixture toxicity is likely to be of concern only when the mixture components are near or above their individual response thresholds. However, these data suggest that extrapolation from in vitro assays to in vivo mixture effects should be approached with caution.

Testing for additivity at select mixture groups of interest based on statistical equivalence testing methods

Several assumptions, defined and undefined, are used in the toxicity assessment of chemical mixtures. In scientific practice mixture components in the low-dose region, particularly subthreshold doses, are often assumed to behave additively (i.e., zero interaction) based on heuristic arguments. This assumption has important implications in the practice of risk assessment, but has not been experimentally tested. We have developed methodology to test for additivity in the sense of Berenbaum (Advances in Cancer Research, 1981), based on the statistical equivalence testing literature where the null hypothesis of interaction is rejected for the alternative hypothesis of additivity when data support the claim. The implication of this approach is that conclusions of additivity are made with a false positive rate controlled by the experimenter. The claim of additivity is based on prespecified additivity margins, which are chosen using expert biological judgment such that small deviations from additivity, which are not considered to be biologically important, are not statistically significant. This approach is in contrast to the usual hypothesis-testing framework that assumes additivity in the null hypothesis and rejects when there is significant evidence of interaction. In this scenario, failure to reject may be due to lack of statistical power making the claim of additivity problematic. The proposed method is illustrated in a mixture of five organophosphorus pesticides that were experimentally evaluated alone and at relevant mixing ratios. Motor activity was assessed in adult male rats following acute exposure. Four low-dose mixture groups were evaluated. Evidence of additivity is found in three of the four low-dose mixture groups. The proposed method tests for additivity of the whole mixture and does not take into account subset interactions (e.g., synergistic, antagonistic) that may have occurred and cancelled each other out.

Review of the toxicity of chemical mixtures: Theory, policy, and regulatory practice

An analysis of current mixture theory, policy, and practice was conducted by examining standard reference texts, regulatory guidance documents, and journal articles. Although this literature contains useful theoretical concepts, clear definitions of most terminology, and well developed protocols for study design and statistical analysis, no general theoretical basis for the mechanisms and interactions of mixture toxicity could be discerned. There is also a poor understanding of the relationship between exposure-based and internal received dose metrics. This confounds data interpretation and limits reliable determinations of the nature and extent of additivity. The absence of any generally accepted classification scheme for either modes/mechanisms of toxic action or of mechanisms of toxicity interactions is problematic as it produces a cycle in which research and policy are interdependent and mutually limiting. Current regulatory guidance depends heavily on determination of toxicological similarity concluded from the presence of a few prominent constituents, assumed from a common toxicological effect, or presumed from an alleged similar toxic mode/mechanism. Additivity, or the lack of it, is largely based on extrapolation of existing knowledge for single chemicals in this context. Thus, regulatory risk assessment protocols lack authoritative theoretical underpinnings, creating substantial uncertainty. Development of comprehensive classification schemes for modes/mechanisms of toxic action and mechanisms of interaction is needed to ensure a sound theoretical foundation for mixture-related regulatory activity and provide a firm basis for iterative hypothesis development and experimental testing.

D-Optimal Experimental Designs to Test for Departure from Additivity in a Fixed-Ratio Mixture Ray

Traditional factorial designs for evaluating interactions among chemicals in a mixture may be prohibitive when the number of chemicals is large. Using a mixture of chemicals with a fixed ratio (mixture ray) results in an economical design that allows estimation of additivity or nonadditive interaction for a mixture of interest. This methodology is extended easily to a mixture with a large number of chemicals. Optimal experimental conditions can be chosen that result in increased power to detect departures from additivity. Although these designs are used widely for linear models, optimal designs for nonlinear threshold models are less well known. In the present work, the use of D-optimal designs is demonstrated for nonlinear threshold models applied to a fixed-ratio mixture ray. For a fixed sample size, this design criterion selects the experimental doses and number of subjects per dose level that result in minimum variance of the model parameters and thus increased power to detect departures from additivity. An optimal design is illustrated for a 2:1 ratio (chlorpyrifos:carbaryl) mixture experiment. For this example, and in general, the optimal designs for the nonlinear threshold model depend on prior specification of the slope and dose threshold parameters. Use of a D-optimal criterion produces experimental designs with increased power, whereas standard nonoptimal designs with equally spaced dose groups may result in low power if the active range or threshold is missed.