Umeå University's logo

Innovative tools suitable for chemical risk assessment are being developed in numerous domains, such as non-target chemical analysis, omics, and computational approaches. These methods will also be critical components in an efficient early warning system (EWS) for the identification of potentially hazardous chemicals. Much knowledge is missing for current use chemicals and thus computational methodologies complemented with fast screening techniques will be critical. This paper reviews current computational tools, emphasizing those that are accessible and suitable for the screening of new and emerging risk chemicals (NERCs). The initial step in a computational EWS is an automatic and systematic search for NERCs in literature and database sources including grey literature, patents, experimental data, and various inventories. This step aims at reaching curated molecular structure data along with existing exposure and hazard data. Next, a parallel assessment of exposure and effects will be performed, which will input information into the weighting of an overall hazard score and, finally, the identification of a potential NERC. Several challenges are identified and discussed, such as the integration and scoring of several types of hazard data, ranging from chemical fate and distribution to subtle impacts in specific species and tissues. To conclude, there are many computational systems, and these can be used as a basis for an integrated computational EWS workflow that identifies NERCs automatically.

Per- and polyfluoroalkyl substances (PFAS) are ubiquitously distributed in the aquatic environment. They include persistent, mobile, bioaccumulative, and toxic chemicals and it is therefore critical to increase our understanding on their adsorption, distribution, metabolism, excretion (ADME). The current study focused on uptake of seven emerging PFAS in zebrafish (Danio rerio) and their potential maternal transfer. In addition, we aimed at increasing our understanding on mixture effects on ADME by developing a physiologically based kinetic (PBK) model capable of handling co-exposure scenarios of any number of chemicals. All studied chemicals were taken up in the fish to varying degrees, whereas only perfluorononanoate (PFNA) and perfluorooctanoate (PFOA) were quantified in all analysed tissues. Perfluorooctane sulfonamide (FOSA) was measured at concerningly high concentrations in the brain (C_max over 15 μg/g) but also in the liver and ovaries. All studied PFAS were maternally transferred to the eggs, with FOSA and 6:2 perfluorooctane sulfonate (6,2 FTSA) showing significant (p < 0.02) signs of elimination from the embryos during the first 6 days of development, while perfluorobutane sulfonate (PFBS), PFNA, and perfluorohexane sulfonate (PFHxS) were not eliminated in embryos during this time-frame. The mixture PBK model resulted in >85 % of predictions within a 10-fold error and 60 % of predictions within a 3-fold error. At studied levels of PFAS exposure, competitive binding was not a critical factor for PFAS kinetics. Gill surface pH influenced uptake for some carboxylates but not the sulfonates. The developed PBK model provides an important tool in understanding kinetics under complex mixture scenarios and this use of New Approach Methodologies (NAMs) is critical in future risk assessment of chemicals and early warning systems.

New chemicals are constantly produced and large data gaps exist on hazards of currently used industrial chemicals, stressing the need for rapid, ethically sound and cost-efficient hazard assessment methods. Traditional methods for effect assessment based on animal testing, do not meet these requirements and thus the toxicology field has been moving towards the development of new approach methodologies which include in vitro approaches but also computational methods. The current work has mainly focused on computational tools but also employed in vitro and in vivo methodologies for the development and validation of the in silico approaches.

We firstly explored chemical variation of emerging chemicals as a basis for selecting sub-groups of per- and polyfluoroalkyl substances (PFASs) and bisphenols for Papers I and II. These compounds can be used for future testing and as case study compounds for in silico tools development. The PFASs selection showed compounds with large differences in structure and highlighted the lack of knowledge for large parts of the PFASs chemical domain. This likely is the main driver of the low predictive accuracy of some current fate models and the need for expanding their applicability domains. 

In Paper II we investigated the toxicokinetics of selected bisphenols in a commonly studied model organism, the zebrafish (Danio rerio), and developed a physiologically-based toxicokinetic model. Novel data for fish biotransformation was derived and showed lower rates than those measured in humans, providing valuable insight for both model parameterization and for chemical safety assessment using fish. The model also demonstrated the ability to predict and rank hazard of these bisphenols in terms of organ-specific bioaccumulation making it a useful tool for chemical screening and prioritization efforts. The results indicate that bisphenols AP, C and Z as well as tetrabromo bisphenol A may have larger potential for bioaccumulation than the widely used bisphenol A (BPA), indicating that these compounds do not constitute safer industrial substitutions.  

Lastly, we present in Paper III the development of a toxicokinetic model for the zebrafish embryo life-stage. Since the zebrafish embryo test is widely applied in toxicology research, the developed model provides a tool to better understand how varying testing conditions may affect dose at target thus providing a means to compare internal effect concentrations. Additionally, we applied the model in combination with data on estrogenic activity in order to rank the relative hazard of investigated bisphenols, which showed that bisphenols AF, C, B and Z may be more hazardous than BPA.

Overall the developed computational tools showed good predictive performance and improvements in parameterization, thus providing tools for understanding dose at target and toxicokinetic variation of emerging substances. Furthermore, the thesis presents novel data and findings for per- and polyfluoroalkyl substances and bisphenols, which are environmental pollutants of emerging concern of relevance for future hazard assessments and substitution processes.

Zebrafish embryos (ZFE) is a widely used model organism, employed in various research fields including toxicology to assess e.g., developmental toxicity and endocrine disruption. Variation in effects between chemicals are difficult to compare using nominal dose as toxicokinetic properties may vary. Toxicokinetic (TK) modeling is a means to estimate internal exposure concentration or dose at target and to enable extrapolation between experimental conditions and species, thereby improving hazard assessment of potential pollutants. In this study we advance currently existing TK models for ZFE with physiological ZFE parameters and novel experimental bisphenol data, a class of chemicals with suspected endocrine activity. We developed a five-compartment model consisting of water, plastic, chorion, yolk sack and embryo in which surface area and volume changes as well as the processes of biotransformation and blood circulation influence mass fluxes. For model training and validation, we measured internal concentrations in ZFE exposed individually to BPA, bisphenol AF (BPAF) and Z (BPZ). Bayesian inference was applied for parameter calibration based on the training data set of BPZ. The calibrated TK model predicted internal ZFE concentrations of the majority of external test data within a 5-fold error and half of the data within a 2-fold error for bisphenols A, AF, F, and tetrabromo bisphenol A (TBBPA). We used the developed model to rank the hazard of seven bisphenols based on predicted internal concentrations and measured in vitro estrogenicity. This ranking indicated a higher hazard for BPAF, BPZ, bisphenol B and C (BPB, BPC) than for BPA.

Bisphenol A (BPA) is an industrial chemical, which has raised human health and environmental concerns due to its endocrine-disrupting properties. BPA analogues are less well-studied despite their wide use in consumer products. These analogues have been detected in water and aquatic organisms around the world, with some analogues showing toxic effects in various species including fish. Here, we present novel organ-specific time-course distribution data of bisphenol Z (BPZ) in female zebrafish (Danio rerio), including concentrations in the ovaries, liver, and brain, a rarely sampled organ with high toxicological relevance. Furthermore, fish-specific in vitro biotransformation rates were determined for 11 selected bisphenols. A physiologically based toxicokinetic (PBTK) model was adapted for four of these bisphenols, which was able to predict levels in the gonads, liver, and brain as well as the whole body within a 2-5-fold error with respect to experimental data, covering several important target organs of toxicity. In particular, predicted liver concentrations improved compared to currently available PBTK models. Predicted data indicate that studied bisphenols mainly distribute to the carcass and gonads and less to the brain. Our model provides a tool to increase our understanding on the distribution and kinetics of a group of emerging pollutants.

Environmental context A diverse range of materials contain organofluorine chemicals, some of which are hazardous and widely distributed in the environment. We investigated an inventory of over 4700 organofluorine compounds, characterised their chemical diversity and selected representatives for future testing to fill knowledge gaps about their environmental fate and effects. Fate and property models were examined and concluded to be valid for only a fraction of studied organofluorines. Many per- and polyfluoroalkyl substances (PFASs) have been identified in the environment, and some have been shown to be extremely persistent and even toxic, thus raising concerns about their effects on human health and the environment. Despite this, little is known about most PFASs. In this study, the comprehensive database of over 4700 PFAS entries recently compiled by the OECD was curated and the chemical variation was analysed in detail. The analysis revealed 3363 individual PFASs with a huge variation in chemical functionalities and a wide range of mixtures and polymers. A hierarchical clustering methodology was employed on the curated database, which resulted in 12 groups, where only half were populated by well-studied compounds thus indicating the large knowledge gaps. We selected both a theoretical and a procurable training set that covered a substantial part of the chemical domain based on these clusters. Several computational models to predict physicochemical and environmental fate related properties were assessed, which indicated their lack of applicability for PFASs and the urgent need for experimental data for training and validating these models. Our findings indicate reasonable predictions of the octanol-water partition coefficient for a small chemical domain of PFASs but large data gaps and uncertainties for water solubility, bioconcentration factor, and acid dissociation factor predictions. Improved computational tools are necessary for assessing risks of PFASs and for including suggested training set compounds in future testing of both physicochemical and effect-related data. This should provide a solid basis for better chemical understanding and future model development purposes.