Sub-lethal toxicity of five disinfection by-products on microalgae determined by flow cytometry - Lines of evidence for adverse outcome pathways.

Standardised tests are often used to determine the ecotoxicity of chemicals and focus mainly on one or a few generic endpoints (e.g. mortality, growth), but information on the sub-cellular processes leading to these effects remain usually partial or missing. Flow cytometry (FCM) can be a practical tool to study the physiological responses of individual cells (such as microalgae) exposed to a stress via the use of fluorochromes and their morphology and natural autofluorescence. This work aimed to assess the effects of five chlorine-based disinfection by-products (DBPs) taken individually on growth and sub-cellular endpoints of the green microalgae Raphidocelis subcapitata. These five DBPs, characteristic of a chlorinated effluent, are the following monochloroacetic acid (MCAA), dichloroacetic acid (DCAA), trichloroacetic acid (TCAA), bromochloroacetic acid (BCAA) and 1,1-dichloropropan-2-one (1,1-DCP). Results showed that 1,1-DCP had the strongest effect on growth inhibition (EC50 = 1.8 mg.L-1), followed by MCAA, TCAA, BCAA and DCAA (EC50 of 10.1, 15.7, 27.3 and 64.5 mg.L-1 respectively). Neutral lipid content, reactive oxygen species (ROS) formation, red autofluorescence, green autofluorescence, size and intracellular complexity were significantly affected by the exposure to the five DBPs. Only mitochondrial membrane potential did not show any variation. Important cellular damages (>10%) were observed for only two of the chemicals (BCAA and 1,1-DCP) and were probably due to ROS formation. The most sensitive and informative sub-lethal parameter studied was metabolic activity (esterase activity), for which three types of response were observed. Combining all this information, an adverse outcome pathways framework was proposed to explain the effect of the targeted chemicals on R. subcapitata. Based on these results, both FCM sub-cellular analysis and conventional endpoint of algal toxicity were found to be complementary approaches.

36Cl, a new tool to assess soil carbon dynamics.

Soil organic carbon is one of the largest surface pools of carbon that humans can manage in order to partially mitigate annual anthropogenic CO2 emissions. A significant element to assess soil sequestration potential is the carbon age, which is evaluated by modelling or experimentally using carbon isotopes. Results, however, are not consistent. The 14C derived approach seems to overestimate by a factor of 6-10 the average carbon age in soils estimated by modeling and 13C approaches and thus the sequestration potential. A fully independent method is needed. The cosmogenic chlorine nuclide, 36Cl, is a potential alternative. 36Cl is a naturally occurring cosmogenic radionuclide with a production that increased by three orders of magnitude during nuclear bomb tests. Part of this production is retained by soil organic matter in organochloride form and hence acts as a tracer of the fate of soil organic carbon. We here quantify the fraction and the duration of 36Cl retained in the soil and we show that retention time increases with depth from 20 to 322 years, in agreement with both modelling and 13C-derived estimates. This work demonstrates that 36Cl retention duration can be a proxy for the age of soil organic carbon.

Modelling geochemical and kinetic processes involved in lead (Pb) remobilization during resuspension events of contaminated sediments.

The objective of this paper is to present a model simulating and predicting the exchange kinetics of lead (Pb) between contaminated sediments and water during resuspension events potentially occurring in reservoirs. We developed an innovative model that combines thermodynamic speciation of particulate surfaces (oxides and Particulate Organic Carbon (POC)), thermodynamic Pb speciation in water, and kinetic modelling of exchanges between free Pb and particulate sites (i.e., dissolution of a carbonate carrier phase, adsorption/coprecipitation and desorption/dissolution to/from oxides, and adsorption and desorption/degradation to/from particulate organic particles). We used results from laboratory resuspension experiments performed on sediments from three contaminated dam reservoirs to calibrate a new chemical speciation model. Uptake and release processes to/from sediments were found to be controlled by at least two successive reactions that are associated with two particulate pools (here oxides and POC). Kinetic adsorption and desorption rates were calibrated for seven experimental conditions. Variability in kinetic rates allowed evaluation of the effect of the solid-to-liquid ratio and sediment origin on exchange kinetics at the water-particle interface. The kinetic release of dissolved Pb by desorption or dissolution from the oxides was reproduced almost identically between the experiments, regardless of the solid-to-liquid ratio or sediment origin. Long-term readsorption on POC sites is more variable, even if ranges of variation in the adsorption and desorption kinetic rates related to POC remain limited, considering that tested sediments vary significantly. CAPSULE: A kinetic model simulating the dynamics of lead (Pb) during sediment resuspension was developed and calibrated to laboratory experiments performed on three contaminated sediments.

"AMORE" Decision Support System for probabilistic Ecological Risk Assessment - Part II: Effect assessment of the case study on cyanide.

Ecotoxicological data are highly important for risk assessment processes and are used for deriving environmental quality criteria, which are enacted for assuring the good quality of waters, soils or sediments and achieving desirable environmental quality objectives. Therefore, it is of significant importance the evaluation of the reliability and relevance of available data for analysing their possible use in the aforementioned processes. In this context, a new methodology which has been developed based on Multi-Criteria Decision Analysis (MCDA) techniques, is being used, demonstrated and tested for analysing the reliability and relevance of ecotoxicological data of cyanide (which are produced through laboratory biotests for individual effects). The proposed methodology is also used for the production of Weighted by Data Quality Species Sensitivity Distributions (SSD-WDQ), as a component of the Ecological Risk Assessment of chemicals in aquatic systems. The SSD-WDQ production resulted in the estimation of environmental quality criteria (hazard concentration affecting 5% and 50% of the species). The proposed work is part of the development of the AMORE Decision Support System (DSS) for the application of probabilistic Ecological Risk Assessment (ERA), presented in the companion paper (Isigonis et al., 2019). The DSS has been tested through a case study on ERA of cyanide in the watershed of river Selune in France. The paper presents the 'Effect Assessment' of cyanide, based on the aforementioned methodologies. The main results presented in the paper are the probabilistic analysis of the estimated species sensitivity on cyanide (Effect Assessment) and the calculation of Hazardous Concentration (HCx) of the same contaminant in the Selune river area, based on the functionalities of the DSS. The results are described and discussed in detail, with the use of various graphs and indices. The indices are calculated for all the available ecotoxicological data, as well as for the data on trophic levels or taxonomic groups separately. An effect comparison is presented between the innovative methodologies included in the DSS and the currently existing methodologies.

"AMORE" Decision Support System for probabilistic Ecological Risk Assessment - Part I: Exposure and risk assessment of the case study on cyanide.

Ecological Risk Assessment of chemicals in fluvial systems is a highly researched topic, but its importance for the environmental protection of our planet is vital. Thus, new developments and improvements to existing methodologies are proposed constantly, for providing more advanced tools and more accurate results to researchers and other interested parties. In the field of probabilistic Ecological Risk Assessment, a new Decision Support System is proposed, developed, tested and evaluated. The AMORE DSS is a modular DSS, which incorporates a series of new methodologies, and is built upon the notions of 'Exposure Assessment', 'Effect Assessment' and 'Risk Assessment'. The AMORE Decision Support System has been developed as part of the AMORE research project (French National Research Agency project). The DSS provides a set of tools for analysing and integrating both exposure and effect information in order to evaluate the risk for species living on a given contaminated aquatic system in terms of the Potentially Affected Fraction. The DSS has been tested through a case study on ERA of cyanide in the watershed of river Selune in France. The paper presents the 'Exposure Assessment' and 'Risk Assessment' of the cyanide case study, as well as the complete functionalities of the AMORE DSS. The main results presented in the paper are the statistical analysis of the measured environmental concentrations of cyanide (Exposure Assessment) and the probabilistic 'Risk assessment' of the same contaminant in the area of interest, based on the functionalities of the DSS. The results are described and discussed in detail with the use of various graphs and risk indices. The risk indices are calculated for all the available ecotoxicological data, as well as for the data on trophic levels or taxonomic groups separately. A risk comparison is presented between the innovative methodologies included in the DSS and the currently existing methodologies.

Modelling ecological and human exposure to POPs in Venice lagoon - Part II: Quantitative uncertainty and sensitivity analysis in coupled exposure models.

The study is focused on applying uncertainty and sensitivity analysis to support the application and evaluation of large exposure models where a significant number of parameters and complex exposure scenarios might be involved. The recently developed MERLIN-Expo exposure modelling tool was applied to probabilistically assess the ecological and human exposure to PCB 126 and 2,3,7,8-TCDD in the Venice lagoon (Italy). The 'Phytoplankton', 'Aquatic Invertebrate', 'Fish', 'Human intake' and PBPK models available in MERLIN-Expo library were integrated to create a specific food web to dynamically simulate bioaccumulation in various aquatic species and in the human body over individual lifetimes from 1932 until 1998. MERLIN-Expo is a high tier exposure modelling tool allowing propagation of uncertainty on the model predictions through Monte Carlo simulation. Uncertainty in model output can be further apportioned between parameters by applying built-in sensitivity analysis tools. In this study, uncertainty has been extensively addressed in the distribution functions to describe the data input and the effect on model results by applying sensitivity analysis techniques (screening Morris method, regression analysis, and variance-based method EFAST). In the exposure scenario developed for the Lagoon of Venice, the concentrations of 2,3,7,8-TCDD and PCB 126 in human blood turned out to be mainly influenced by a combination of parameters (half-lives of the chemicals, body weight variability, lipid fraction, food assimilation efficiency), physiological processes (uptake/elimination rates), environmental exposure concentrations (sediment, water, food) and eating behaviours (amount of food eaten). In conclusion, this case study demonstrated feasibility of MERLIN-Expo to be successfully employed in integrated, high tier exposure assessment.

Potential for MERLIN-Expo, an advanced tool for higher tier exposure assessment, within the EU chemical legislative frameworks.

MERLIN-Expo merges and integrates advanced exposure assessment methodologies, allowing the building of complex scenarios involving several pollution sources and targets. The assessment of exposure and risks to human health from chemicals is of major concern for policy and ultimately benefits all citizens. The development and operational fusion of the advanced exposure assessment methodologies envisaged in the MERLIN-Expo tool will have a significant impact in the long term on several policies dealing with chemical safety management. There are more than 30 agencies in Europe related to exposure and risk evaluation of chemicals, which have an important role in implementing EU policies, having especially tasks of technical, scientific, operational and/or regulatory nature. The main purpose of the present paper is to introduce MERLIN-Expo and to highlight its potential for being effectively integrated within the group of tools available to assess the risk and exposure of chemicals for EU policy. The main results show that the tool is highly suitable for use in site-specific or local impact assessment, with minor modifications it can also be used for Plant Protection Products (PPPs), biocides and REACH, while major additions would be required for a comprehensive application in the field of consumer and worker exposure assessment.

Multimedia & PBPK modelling with MERLIN-Expo versus biomonitoring for assessing Pb exposure of pre-school children in a residential setting.

This paper reports on a case study - conducted within the European FP7 project "4FUN" - focusing on exposure of pre-school children to lead resulting from past emissions by non-ferrous smelters in Belgium (Northern Campine area). Exposure scenarios were constructed and simulated with the MERLIN-Expo tool to estimate external Pb exposure as well as the Pb body burden in children living in the vicinity of the former industrial sites as compared to children living in adjacent areas and a reference area. Simulations were run for several scenarios ranging from very simple to rather complex in order to study the effect of different simulation approaches (e.g., deterministic vs. probabilistic, individual vs. aggregated population exposure) and different exposure scenarios (e.g., with vs. without considering local food consumption or time activity patterns) on the model outcomes (predicted concentrations of Pb in environmental and human matrices). This paper discusses the two most complex scenarios, namely exposure at the aggregated population level and at the individual level for a random sub-sample of subjects, respectively. In the final and most realistic exposure scenario, simulating individual lead exposure, model predictions were shown to be higher than the biomonitoring data. Blood Pb levels in children, irrespective of the area they lived in, were overpredicted by MERLIN-Expo with a factor of about 2 on average. The model predictions for individual children overlap with the prediction interval calculated by MERLIN-Expo based on population averages, demonstrating the use of probabilistic approaches in risk assessment. While these results constitute a first verification of the model performance of MERLIN-Expo dealing with inorganic pollutants in a complex real-world exposure scenario and a demonstration of the robustness of the modelling tool, further validation and benchmarking efforts are required for a larger number of inorganic pollutants and different exposure settings.

Assessing multimedia/multipathway exposures to inorganic arsenic at population and individual level using MERLIN-Expo.

In this study, we report on model simulations performed using the newly developed exposure tool, MERLIN-Expo, in order to assess inorganic arsenic (iAs) exposure to adults resulting from past emissions by non-ferrous smelters in Belgium (Northern Campine area). Exposure scenarios were constructed to estimate external iAs exposure as well as the toxicologically relevant As (tAs, i.e., iAs, MMA and DMA) body burden in adults living in the vicinity of the former industrial sites as compared to adults living in adjacent areas and a reference area. Two scenarios are discussed: a first scenario studying exposure to iAs at the aggregated population level and a second scenario studying exposure at the individual level for a random sub-sample of subjects in each of the three different study areas. These two scenarios only differ in the type of human related input data (i.e., time-activity data, ingestion rates and consumption patterns) that were used, namely averages (incl. probability density functions, PDFs) in the simulation at population level and subject-specific values in the simulation at individual level. The model predictions are shown to be lower than the corresponding biomonitoring data from the monitoring campaign. Urinary tAs levels in adults, irrespective of the area they lived in, were under-predicted by MERLIN-Expo by 40% on average. The model predictions for individual adults, by contrast, under-predict the biomonitoring data by 7% on average, but with more important under-predictions for subjects at the upper end of exposure. Still, average predicted urinary tAs levels from the simulations at population level and at individual level overlap, and, at least for the current case, lead to similar conclusions. These results constitute a first and partial verification of the model performance of MERLIN-Expo when dealing with iAs in a complex site-specific exposure scenario, and demonstrate the robustness of the modelling tool for these situations.

A Multi-Criteria Decision Analysis based methodology for quantitatively scoring the reliability and relevance of ecotoxicological data.

Ecotoxicological data are highly important for risk assessment processes and are used for deriving environmental quality criteria, which are enacted for assuring the good quality of waters, soils or sediments and achieving desirable environmental quality objectives. Therefore, it is of significant importance the evaluation of the reliability of available data for analysing their possible use in the aforementioned processes. The thorough analysis of currently available frameworks for the assessment of ecotoxicological data has led to the identification of significant flaws but at the same time various opportunities for improvement. In this context, a new methodology, based on Multi-Criteria Decision Analysis (MCDA) techniques, has been developed with the aim of analysing the reliability and relevance of ecotoxicological data (which are produced through laboratory biotests for individual effects), in a transparent quantitative way, through the use of expert knowledge, multiple criteria and fuzzy logic. The proposed methodology can be used for the production of weighted Species Sensitivity Weighted Distributions (SSWD), as a component of the ecological risk assessment of chemicals in aquatic systems. The MCDA aggregation methodology is described in detail and demonstrated through examples in the article and the hierarchically structured framework that is used for the evaluation and classification of ecotoxicological data is shortly discussed. The methodology is demonstrated for the aquatic compartment but it can be easily tailored to other environmental compartments (soil, air, sediments).

Identification of sensitive parameters in the modeling of SVOC reemission processes from soil to atmosphere.

Semi-volatile organic compounds (SVOCs) are subject to Long-Range Atmospheric Transport because of transport-deposition-reemission successive processes. Several experimental data available in the literature suggest that soil is a non-negligible contributor of SVOCs to atmosphere. Then coupling soil and atmosphere in integrated coupled models and simulating reemission processes can be essential for estimating atmospheric concentration of several pollutants. However, the sources of uncertainty and variability are multiple (soil properties, meteorological conditions, chemical-specific parameters) and can significantly influence the determination of reemissions. In order to identify the key parameters in reemission modeling and their effect on global modeling uncertainty, we conducted a sensitivity analysis targeted on the 'reemission' output variable. Different parameters were tested, including soil properties, partition coefficients and meteorological conditions. We performed EFAST sensitivity analysis for four chemicals (benzo-a-pyrene, hexachlorobenzene, PCB-28 and lindane) and different spatial scenari (regional and continental scales). Partition coefficients between air, solid and water phases are influent, depending on the precision of data and global behavior of the chemical. Reemissions showed a lower variability to soil parameters (soil organic matter and water contents at field capacity and wilting point). A mapping of these parameters at a regional scale is sufficient to correctly estimate reemissions when compared to other sources of uncertainty.

Using an Ecosystem Approach to complement protection schemes based on organism-level endpoints.

Radiation protection goals for ecological resources are focussed on ecological structures and functions at population-, community-, and ecosystem-levels. The current approach to radiation safety for non-human biota relies on organism-level endpoints, and as such is not aligned with the stated overarching protection goals of international agencies. Exposure to stressors can trigger non-linear changes in ecosystem structure and function that cannot be predicted from effects on individual organisms. From the ecological sciences, we know that important interactive dynamics related to such emergent properties determine the flows of goods and services in ecological systems that human societies rely upon. A previous Task Group of the IUR (International Union of Radioecology) has presented the rationale for adding an Ecosystem Approach to the suite of tools available to manage radiation safety. In this paper, we summarize the arguments for an Ecosystem Approach and identify next steps and challenges ahead pertaining to developing and implementing a practical Ecosystem Approach to complement organism-level endpoints currently used in radiation safety.

Bioavailability of copper in contaminated sediments assessed by a DGT approach and the uptake of copper by the aquatic plant Myriophyllum aquaticum.

The assessment of the potentially harmful effects of metals on biota depends on the speciation and bioavailability of the metals. In the present study, the authors investigated Cu accumulation and toxicity in the aquatic plant Myriophyllum aquaticum after exposure to artificial sediments varying in peat or ferric hydroxide content and spiked with Cu (5-200 mg kg(-1)). Modeling of the kinetic diffusive gradient in thin film (DGT) measurements revealed fast and slow Cu resupply from the solid phase for sediment formulated with and without peat, respectively. Myriophyllum aquaticum proved to be sensitive to Cu, as the Cu accumulation and growth differed depending on the sediment composition and Cu concentration. Comparing the Cu accumulation in M. aquaticum with total dissolved concentration, free concentration, and concentration in solution derived from DGT measurements (CDGT), Cu concentrations revealed that CDGT concentrations were a better predictor of accumulation than the others. However, the relatively weak correlation observed (r(2) = 0.6) and the fact that plant uptake does not increase proportionally to DGT fluxes suggest that Cu uptake in plants was not diffusion limited. Thus, the free Cu concentrations near the root surface were sufficient to meet the plant's demand during the experiment. Furthermore, labile complexes that continuously resupply the Cu(2+) pool may also contribute to the concentrations available for plant uptake. In the range of Cu concentrations investigated in the present study, saturation of uptake processes as well as toxicity are considered responsible for the poor DGT prediction of plant uptake.

Probabilistic approaches to accounting for data variability in the practical application of bioavailability in predicting aquatic risks from metals.

The biotic ligand model (BLM) theoretically enables the derivation of environmental quality standards that are based on true bioavailable fractions of metals. Several physicochemical variables (especially pH, major cations, dissolved organic carbon, and dissolved metal concentrations) must, however, be assigned to run the BLM, but they are highly variable in time and space in natural systems. This article describes probabilistic approaches for integrating such variability during the derivation of risk indexes. To describe each variable using a probability density function (PDF), several methods were combined to 1) treat censored data (i.e., data below the limit of detection), 2) incorporate the uncertainty of the solid-to-liquid partitioning of metals, and 3) detect outliers. From a probabilistic perspective, 2 alternative approaches that are based on log-normal and Γ distributions were tested to estimate the probability of the predicted environmental concentration (PEC) exceeding the predicted non-effect concentration (PNEC), i.e., p(PEC/PNEC>1). The probabilistic approach was tested on 4 real-case studies based on Cu-related data collected from stations on the Loire and Moselle rivers. The approach described in this article is based on BLM tools that are freely available for end-users (i.e., the Bio-Met software) and on accessible statistical data treatments. This approach could be used by stakeholders who are involved in risk assessments of metals for improving site-specific studies.

Speciation and bioavailability of dissolved copper in different freshwaters: comparison of modelling, biological and chemical responses in aquatic mosses and gammarids.

Biological and chemical measurements were performed in mesocosms to investigate the bioavailability of copper, with a greater emphasis on the effects of competing ions and copper speciation. Measurements were achieved in three different natural waters for two aquatic species (Gammarus pulex and Fontinalis antipyretica) along a copper gradient concentration: natural concentration, spiked at 5 and 15 µg L(-1). Aquatic mosses exhibited high enrichment rates that were above the background levels compared to gammarids. The accumulation of copper in F. antipyretica is better correlated to the weakly complexed copper concentrations measured using differential pulse anodic stripping voltammetry (DPASV) and diffusive gradient in thin film (DGT) than to the free copper concentration measured using an ion selective electrode (ISE). In unspiked natural waters, the presence of dissolved organic ligands strongly controls the metal speciation and consequently largely minimised the impact of competing cations on the accumulation of Cu in mosses. Furthermore, the BioMet Biotic Ligand Model (BLM) successfully describes the site-specific copper bioaccumulation for the freshwater mosses studied. However, the comparison of the results with a previous study appears to indicate that the adsorption/desorption of Cu in mosses is impacted by seasons. This highlights a limit of the BioMet model in which the physiological state of aquatic organisms is not considered. No toxic effect of Cu exposure on lipid peroxidation was observed in the mosses and gammarids regardless of the site and the concentration considered. However, the oxidative stress measured in the mosses via their guaiacol peroxidase (GPX) activity increased in the case where internalised Cu reached maximal values, which suggests a threshold effect on the GPX activity.

DGT as surrogate of biomonitors for predicting the bioavailability of copper in freshwaters: an ex situ validation study.

The present report is the companion study of our previous study in which we investigated the impact of the dissolved organic matter, water cationic composition and pH on the bioavailability and the bioaccumulation of copper (Cu) in aquatic mosses (Fontinalis antipyretica). The impact had been assessed under laboratory controlled conditions and modelled using a two-compartment model calibrated under a wide range of water compositions (Ferreira et al., 2008, 2009). Herein are reported the validation stage of the abovementioned approach for contrasted geochemical field conditions. Experiments were performed with aquatic mosses that were exposed for 7d to two nominal Cu concentrations (5 and 15µgL(-1)) in a flow-through field microcosm supplied with four contrasting natural waters. At the end of the exposure period, a 6-fold difference in the bioaccumulated Cu contamination levels was found among the four deployment sites, suggesting a significant control of the water quality on the metal bioaccumulation by aquatic mosses. In parallel, the so-called 'labile' Cu concentration for the same four field conditions was determined using a DGT device (Diffusive Gradient in Thin film). By coupling these DGT measurements and a cation competition model involving Ca(2+), Mg(2+), Na(+) and H(+), the time-dependent Cu concentrations in aquatic mosses were predicted; these simulation results were compared to the actual bioaccumulation of Cu in mosses. We found that any bioaccumulation model that ignores water characteristics is not suitable to predict the Cu accumulation by aquatic mosses under various water quality conditions. Instead, we found that our approach integrating DGT measurements and cationic composition was able to reproduce the Cu bioaccumulation kinetics by aquatic mosses for a wide range of water quality conditions. In conclusion, the DGT approach was demonstrated to be a dynamic in situ measuring technique that can be used as a surrogate of bioindicators if the cationic correction is taken into account.

Estimating hazardous concentrations by an informative Bayesian approach.

The species sensitivity distribution (SSD) approach is recommended for assessing chemical risk. In practice, however, it can be used only for the few substances for which large-scale ecotoxicological results are available. Indeed, the statistical frequentist approaches used for building SSDs and for deriving hazardous concentrations (HC5) inherently require extensive data to guarantee goodness-of-fit. An alternative Bayesian approach to estimating HC5 from small data sets was developed. In contrast to the noninformative Bayesian approaches that have been tested to date, the authors' method used informative priors related to the expected species sensitivity variance. This method was tested on actual ecotoxicological data for 21 well-informed substances. A cross-validation compared the HC5 values calculated using frequentist approaches with the results of our Bayesian approach, using both complete and truncated data samples. The authors' informative Bayesian approach was compared with noninformative Bayesian methods published in the past, including those incorporating loss functions. The authors found that even for the truncated sample the HC5 values derived from the informative Bayesian approach were generally close to those obtained using the frequentist approach, which requires more data. In addition, the probability of overestimating an HC5 is rather limited. More robust HC5 estimates can be practically obtained from additional data without impairing regulatory protection levels, which will encourage collecting new ecotoxicological data. In conclusion, the Bayesian informative approach was shown to be relatively robust and could be a good surrogate approach for deriving HC5 values from small data sets.

Interpreting PCB levels in breast milk using a physiologically based pharmacokinetic model to reconstruct the dynamic exposure of Italian women.

Polychlorinated biphenyls (PCBs) are persistent contaminants suspected to cause adverse health effects in humans. As PCBs levels in food have not been monitored frequently in the past, modeling approaches based on environmental data have been proposed to predict the human dietary intake. In this work, we propose to improve these approaches by taking into account internal levels of PCBs in humans. This methodology is based on the analysis of biomonitoring data using exposure and physiologically based pharmacokinetic (PBPK) modeling to determine the most probable scenario of exposure. Breast milk concentrations were measured in Italian women for PCB-138, PCB-153 and PCB-180. For each congener, three exposure scenarios were derived and a PBPK model was used to relate the lifetime exposure to the breast milk levels. For the three PCBs, we determined the most probable scenario of exposure. Our results support the adequacy of the exposure and the PBPK models for PCB-180 and PCB-153, whereas we observed discrepancies between the models and the biomonitoring data for PCB-138. Our intake estimates are in good agreement with previous exposure assessments based solely on food contamination demonstrating the relevance of our approach to reconstruct accurately the exposure and to fill in data gaps on exposure.

Linking fate model in freshwater and PBPK model to assess human internal dosimetry of B(a)P associated with drinking water.

In the present study, we demonstrate an integrated modeling approach for predicting internal tissue concentrations of chemicals by coupling a multimedia environmental model and a generic physiologically based pharmacokinetic (PBPK) model. A case study was designed for a region situated on the Seine river watershed, downstream of the Paris megacity, and for benzo(a)pyrene emitted from industrial zones in the region. In this case study, these two models are linked only by water intake from riverine system for the multimedia model into human body for the PBPK model. The limited monitoring data sets of B(a)P concentrations in bottom sediment and in raw river water, obtained at the downstream of Paris, were used to re-construct long-term daily concentrations of B(a)P in river water. The re-construction of long-term series of B(a)P level played a key role for the intermediate model calibration (conducted in multimedia model) and thus for improving model input to PBPK model. In order to take into account the parametric uncertainty in the model inputs, some input parameters relevant for the multimedia model were given by probability density functions (PDFs); some generic PDFs were updated with site-specific measurements by a Bayesian approach. The results of this study showed that the multimedia model fits well with actual annual measurements in sediments over one decade. No accumulation of B(a)P in the organs was observed. In conclusion, this case study demonstrated the feasibility of a full-chain assessment combining multimedia environmental predictions and PBPK modeling, including uncertainty and sensitivity analyses.