QSAR MODELING OF BIOCONCENTRATION FACTORS IN FISH BASED ON FRAGMENT CONSTANTS AND STRUCTURAL CORRECTION FACTORS

The quantitative structure-activity relationship between the BCF and fragment constant of organic chemicals was studied using a database containing 337 experimental BCF values. The database covered a large variety of chemicals ranging from the very hydrophobic to the very hydrophilic with logKow values between 0.39 and 8.60. The structural features affecting the BCFs were identified and evaluated during a preliminary modeling. A final linear multivariate regression model was derived that was able to account for as much as 98.0% of the variation in the experimental BCF values. The mean absolute error for the final model was 0.315 log-units. In addition, the predictability and robustness of the model was also evaluated.     

Fragment constant method for prediction of fish bioconcentration factors of non-polar chemicals

A fragment constant method for prediction of fish bioconcentration factor (BCF) was established based on experimental BCF values for 80 non-polar chemicals from nine classes. The model was evaluated using coefficients of determination and mean residuals, which are 0.995 and 0.1836, respectively. Jackknife tests were applied to examine the robustness of the prediction model on a class-by-class basis.     

Approach for extrapolating in vitro metabolism data to refine bioconcentration factor estimates

National and international chemical management programs are assessing thousands of chemicals for their persistence, bioaccumulative and environmental toxic properties; however, data for evaluating the bioaccumulation potential for fish are limited. Computer based models that account for the uptake and elimination processes that contribute to bioaccumulation may help to meet the need for reliable estimates. One critical elimination process of chemicals is metabolic transformation. It has been suggested that in vitro metabolic transformation tests using fish liver hepatocytes or S9 fractions can provide rapid and cost-effective measurements of fish metabolic potential, which could be used to refine bioconcentration factor (BCF) computer model estimates. Therefore, recent activity has focused on developing in vitro methods to measure metabolic transformation in cellular and subcellular fish liver fractions. A method to extrapolate in vitro test data to the whole body metabolic transformation rates is presented that could be used to refine BCF computer model estimates. This extrapolation approach is based on concepts used to determine the fate and distribution of drugs within the human body which have successfully supported the development of new pharmaceuticals for years. In addition, this approach has already been applied in physiologically-based toxicokinetic models for fish. The validity of the in vitro to in vivo extrapolation is illustrated using the rate of loss of parent chemical measured in two independent in vitro test systems: (1) subcellular enzymatic test using the trout liver S9 fraction, and (2) primary hepatocytes isolated from the common carp. The test chemicals evaluated have high quality in vivo BCF values and a range of logK(ow) from 3.5 to 6.7. The results show very good agreement between the measured BCF and estimated BCF values when the extrapolated whole body metabolism rates are included, thus suggesting that in vitro biotransformation data could effectively be used to reduce in vivo BCF testing and refine BCF model estimates. However, additional fish physiological data for parameterization and validation for a wider range of chemicals are needed.     

Calculation of bioconcentration factors from kinetic data by non-linear iterative least-squares regression analysis using a programmable minicalculator

A program to fit theoretically modelled uptake and clearance periods of the bioconcentration of chemicals to experimental data using a programmable pocket calculator with thermal printer is reported. The bioconcentration factor (BCF) is calculated from the ratio of uptake and clearance rate constants. Because of the impossibility to linearize the kinetic equations an iterative nonlinear GAUSS-NEWTON least-squares fit has been applied for the pocket calculator programm ACCUTI-59, which calculates from a set of start parameters the best values for uptake and clearance rate constants, BCF, and the standard errors as well as the sum of squared deviations. A listing of the program is given.     

Bioconcentration factor of relatively low concentrations of chlorophenols in Japanese medaka

Bioconcentration factors (BCF) for pentachlorophenol (PCP) and 2,4-dichlorophenol (2,4-DCP) in Japanese medaka (Oryzias latipes) were determined at five different concentrations of the chemicals, between 0.1 and 10 microg/l (PCP), 0.3 and 30 microg/l (2,4-DCP), in the ambient water. Medaka were exposed to each chemicals in a continuous-flow system during the embryonic development period and 60 days after hatching from eggs collected in the laboratory. Both the exposure time and the aqueous concentrations are much more realistic and closer to natural aquatic environments than those used in conventional BCF studies. The BCF values of PCP were from (4.9+/-2.8)x10(3) at the aqueous concentration of 0.074+/-0.028 microg/l to (2.1+/-1.4)x10(3) at 9.70+/-0.56 microg/l. The BCF value of 2,4-DCP were from (3.4+/-3.0)x10(2) at 0.235+/-0.060 microg/l to 92+/-27 at 27.3+/-1.6 microg/l. Generally, BCF values increased as the aqueous concentrations of PCP or 2,4-DCP decreased. This finding suggests that a relatively low and realistic aqueous concentration of these compounds is necessary to more accurately determine their BCF values in natural aquatic environments. Conventional BCF experiments at higher aqueous concentrations may underestimate the BCF values.     

Uptake of polychlorinated biphenyls and organochlorine pesticides from soil and air into radishes (Raphanus sativus)

Uptake of organochlorine pesticides and polychlorinated biphenyls from soil and air into radishes was measured at a heavily contaminated field site. The highest contaminant concentrations were found for DDT and its metabolites, and for β-hexachlorocyclohexane. Bioconcentration factor (BCF, defined as a ratio between the contaminant concentration in the plant tissue and concentration in soil) was determined for roots, edible bulbs and shoots. Root BCF values were constant and not correlated to log K_OW. A negative correlation between BCF and log K_OW was found for edible bulbs. Shoot BCF values were rather constant and varied between 0.01 and 0.22. Resuspended soil particles may facilitate the transport of chemicals from soil to shoots. Elevated POP concentrations found in shoots of radishes grown in the control plot support the hypothesis that the uptake from air was more significant for shoots than the one from soil. The uptake of POPs from air was within the range of theoretical values predicted from log K_OA.     

A fragment constant QSAR model for evaluating the EC50 values of organic chemicals to Daphnia magna.

The quantitative relationship between the median effective concentration (EC50) of organic chemicals to Daphnia magna and the number of molecular fragments was investigated based on experimental EC50 values for 217 chemicals derived from the literature. A fragment constant model was developed based on a multivariate linear regression between the number of fragments and the logarithmically transformed reciprocal values of EC50. Functional correction factors were introduced into the model. The model was verified using an independent set of randomly selected data. The mean residual of the final model was 0.4 log-units. The robustness of the model was discussed based on the results of three jackknife tests.     

Bioconcentration of superlipophilic persistent chemicals

According to present understanding, persistent superlipophilic chemicals — such as octachlorodibenzo-p-dioxin, octachlorodibenzofuran, Mirex etc — with log K_ow > 6 and cross sections > 9.5 Å, bioconcentrate in aquatic organisms only little from ambient water. The most convincing argument against it is that in bioconcentration experiments with superlipophilic chemicals amounts applied exceeded water solubility by several orders of magnitude. This paper describes various methods for determining bioconcentration factors (BCF) of superlipophilic compounds. As exemplified with octachlorodibenzo-p-dioxin, BCF values evaluated by these methods match well with those calculated by QSARs for fish and mussels based on log K_ow and water solubility. As expected, these BCF values exceed previous values by several orders of magnitude. For BCF evaluation of superlipophilic chemicals in aquatic organisms we recommend:

1. flow-through systems, kinetic method (OECD guideline No. 305 E)
2. ambient concentrations < water solubility
3. during the uptake and especially during the elimination phase no toxic effects of the test organisms should occur.

     

Bioavailability of atrazine, pyrene and benzo[a]pyrene in European river waters.

Thirteen river waters and one humic lake water were characterized. The effects of dissolved organic matter (DOM) on the bioavailability of atrazine, pyrene and benzo[a]pyrene (B[a]P) was evaluated. Binding of the chemicals by DOM was analyzed with the equilibrium dialysis technique. For each of the water samples, 24 h bioconcentration factors (BCFs) of the chemicals were measured in Daphnia magna. The relationship between DOM and other water characteristics (including conductivity, water hardness and pH), and bioavailability of the chemicals was studied by performing several statistical analyses, including multiple regression analyses, to determine how much of the variation of BCF values could be explained by the quantity and quality of DOM. The bioavailability of atrazine was not affected by DOM or any other water characteristics. Although equilibrium dialysis showed binding of pyrene to DOM, the bioavailability of pyrene was not significantly affected by DOM. The bioavailability of B[a]P was significantly affected by both the quality and quantity of DOM. Multiple regression analyses, using the quality (ABS270 and HbA%) and quantity of DOM as variables, explained up to 70% of the variation in BCF of B[a]P in the waters studied.     

A new hybrid system of QSAR models for predicting bioconcentration factors (BCF)

The aim was to develop a reliable and practical quantitative structure–activity relationship (QSAR) model validated by strict conditions for predicting bioconcentration factors (BCF). We built up several QSAR models starting from a large data set of 473 heterogeneous chemicals, based on multiple linear regression (MLR), radial basis function neural network (RBFNN) and support vector machine (SVM) methods. To improve the results, we also applied a hybrid model, which gave better prediction than single models. All models were statistically analysed using strict criteria, including an external test set. The outliers were also examined to understand better in which cases large errors were to be expected and to improve the predictive models. The models offer more robust tools for regulatory purposes, on the basis of the statistical results and the quality check on the input data.     

A new hybrid system of QSAR models for predicting bioconcentration factors (BCF)

The aim was to develop a reliable and practical quantitative structure-activity relationship (QSAR) model validated by strict conditions for predicting bioconcentration factors (BCF). We built up several QSAR models starting from a large data set of 473 heterogeneous chemicals, based on multiple linear regression (MLR), radial basis function neural network (RBFNN) and support vector machine (SVM) methods. To improve the results, we also applied a hybrid model, which gave better prediction than single models. All models were statistically analysed using strict criteria, including an external test set. The outliers were also examined to understand better in which cases large errors were to be expected and to improve the predictive models. The models offer more robust tools for regulatory purposes, on the basis of the statistical results and the quality check on the input data.     

The influence of gill and liver metabolism on the predicted bioconcentration of three pharmaceuticals in fish

The potential for xenobiotic compounds to bioconcentrate is typically expressed through the bioconcentration factor (BCF), which has gained increased regulatory significance over the past decade. Due to the expense of in vivo bioconcentration studies and the growing regulatory need to assess bioconcentration potential, BCF is often calculated via single-compartment models, using K(OW) as the primary input. Recent efforts to refine BCF models have focused on physiological factors, including the ability of the organism to eliminate the compound through metabolic transformation. This study looks at the ability of in vitro biotransformation assays using S9 fractions to provide an indication of metabolic potential. Given the importance of the fish gill and liver in metabolic transformation, the metabolic loss of ibuprofen, norethindrone and propranolol was measured using rainbow trout (Oncorhynchus mykiss) and channel catfish (Ictalurus punctatus) gill and liver S9 fractions. Metabolic transformation rates (k(M)) were calculated and integrated into a refined BCF model. A significant difference was noted between BCF solely based on K(OW) and BCF including k(M). These studies indicate that the inclusion of k(M) in BCF models can bring predicted bioconcentration estimates closer to in vivo values.     

Influence of dispersants on bioconcentration factors of seven organic compounds with different lipophilicities and structures

Seven compounds with different lipophilicities and structures—1,3,5-trichlorobenzene, pentachlorobenzene, acenaphthylene, 1,4-dimethyl-2-(1-methylphenyl)benzene, 4-ethylbiphenyl, 4,4′-dibromobiphenyl, and 1,1,1-trichloro-2,2-bis(4-chlorophenyl)ethane—were subjected to bioconcentration tests in carp at concentrations below the water solubilities of the compounds in the presence or absence of a dispersant (either an organic solvent or a surfactant). The bioconcentration factors (BCFs) of the compounds were on the order of 10²–10⁴. The BCF values remained in the range of 15–49% for all the compounds, whether or not a dispersant was present, i.e., the BCF values in the presence of an organic solvent or a surfactant at a concentration below the critical micelle concentration were not significantly smaller than the BCF values in the absence of the solvent or surfactant. This result indicates that the dispersants had no influence on the evaluation of the bioconcentration potential of these test substances.     

Pesticide bioconcentration modelling for fruit trees

The model presented allows simulating the pesticide concentration evolution in fruit trees and estimating the pesticide bioconcentration factor in fruits. Pesticides are non-ionic organic compounds that are degraded in soils cropped with woody species, fruit trees and other perennials. The model allows estimating the pesticide uptake by plants through the water transpiration stream and also the time in which maximum pesticide concentration occur in the fruits. The equation proposed presents the relationships between bioconcentration factor (BCF) and the following variables: plant water transpiration volume (Q), pesticide transpiration stream concentration factor (TSCF), pesticide stem-water partition coefficient (K(Wood,W)), stem dry biomass (M) and pesticide dissipation rate in the soil-plant system (k(EGS)). The modeling started and was developed from a previous model "Fruit Tree Model" (FTM), reported by Trapp and collaborators in 2003, to which was added the hypothesis that the pesticide degradation in the soil follows a first order kinetic equation. The FTM model for pesticides (FTM-p) was applied to a hypothetic mango plant cropping (Mangifera indica) treated with paclobutrazol (growth regulator) added to the soil. The model fitness was evaluated through the sensitivity analysis of the pesticide BCF values in fruits with respect to the model entry data variability.     

Bioaccumulation kinetics and bioconcentration factor of chlorinated pesticides in tissues of Puntius ticto (Ham.)

Bioaccumulation kinetics and bioconcentration factor (BCF) of chlorinated pesticides like Aldrin, Dieldrin, Benzene hexachloride (BHC), and Dichlorodiphenyl-dichloro-ethane (DDT) in fish tissues of Puntius ticto was studied in detail in a continuous fed system. The bioconcentration process is summarized by using a first order uptake model and the steady-state BCF is calculated based on the 30 days exposure. Rate of bioaccumulation of DDT was maximum of 4.6432 microg g(-1) wet weight per day in liver tissue whereas it was minimum of 0.0002 microg g(-1) wet weight per day in case of Dieldrin in the muscle tissue among the pesticides. It was observed that DDT showed maximum BCF of 89.010 in case of liver tissue of the fish exposed to 30 days. The regression coefficient (r2) between pesticide concentration and exposure time varied between 0.6212 and 0.9817 indicating high correlation. Based on actual calculated BCF values, the octanol water partition coefficient (Kow) values were predicted. In order to prove the hydrophobic property of chlorinated compounds and its affinity towards lipid, the Kow is predicted. Results showed that pesticide burden differ from tissue to tissue and can be correlated to the lipid content, size, exposure time, and species.