Development of a generic zebrafish embryo PBPK model and application to the developmental toxicity assessment of valproic acid analogs
Introduction
Prediction of chemicals’ developmental and reproductive toxicity is a complex challenge. Toxicity assays are in majority conducted in mammals, due to their recognized efficacy at predicting toxicity in humans. However, mammalian assays are expensive, strictly regulated by law, and time-consuming. Given the morphological and developmental similarities among vertebrates, fish are a relevant test alternative to mammals [1,2]. For various reasons, the zebrafish embryo is particularly attractive in toxicology and pharmacology [3]. First, its transparency allows the visual detection of malformations, without interrupting development or invasive interventions [4]. Second, the number of eggs laid is high, and its development time is short [5]. Third, it is easy to maintain in the laboratory [6]. Finally, there are considerable gene homologies and neurophysiological similarities between zebrafish, mammals and humans [7,8]. Therefore, evidence obtained with zebrafish embryos should at least partially be translatable to humans [9]. From a regulatory point of view, until the age of 120 h, zebrafish embryos are an alternative to experiments with adult vertebrate species, because they are not protected by European animal welfare legislation until five days post-fertilization [10,11].
Yet, extrapolation of toxicity from zebrafish to humans requires, at least, accounting for differences in pharmacokinetics between the two species [12]. Such differences might translate into differences in target organ concentrations for the same systemic exposure dose. In addition, knowing chemical concentrations in organs is fundamental to understand dose-response relationships [13,14]. Internal concentrations are often difficult to measure, but physiologically-based pharmacokinetic (PBPK) models can estimate them [15,16]. PBPK models connect anatomy, physiology, and biochemical processes to understand and compute a chemical’s fate in the body. They allow approximate predictions of chemicals’ concentration-time profiles in experimentally inaccessible organs from minimal data. Thereby, they provide mechanistic insight into toxicity and help reduce time, cost and need for animal experiments [17].
PBPK models have been published for adult zebrafish, mostly for ecotoxicological risk assessment [[18], [19], [20]]. Recently, Brox et al. [21] used a one-compartment two-parameter model developed by Gobas and Zhang [22] to explore the impact of physicochemical properties of polar compounds and of biological processes on embryo concentrations. This model, however, cannot describe decreases in concentrations due to metabolism [23] or dilution by organ growth, and Brox et al. concluded the necessity of more sophisticated toxicokinetic models for the zebrafish embryo.
In order to better explain, predict, and extrapolate developmental toxicity observed in zebrafish embryos, we developed a generic PBPK model integrating organ growth and hepatic metabolism. The model assumes quasi steady-state distribution between zebrafish cells, lysosomes, and mitochondria in different tissues (yolk, liver, gut, muscle, skeleton, eye, brain, heart, skin, and lumped other tissues). The model is generic in that it can simulate the distribution of many chemicals in zebrafish embryos on the basis of their physicochemical properties: chemicals’ partition coefficients between cells or organelles and culture medium are calculated with the Simcyp® virtual in vitro intracellular distribution (VIVD) model [24]. The model can therefore be used for high-throughput predictions of internal concentrations in zebrafish.
We applied our model to the analysis of developmental toxicity data on valproic acid (VPA) and nine of its analogs: 2,2-dimethylvaleric acid, 2-ethylbutyric acid, 2-ethylhexanoic acid, 2-methylhexanoic acid, 2-methylpentanoic acid, 2-propylheptanoic acid, 4-eneVPA, 4-pentenoic acid and hexanoic acid. VPA is a notorious teratogenic antiepileptic and thymoregulator, inducing neural tube defects in mammalian embryos, probably by inhibition of histone deacetylase, interference with folate metabolism and inducing oxidative stress. However, the mechanism of action remains not well known [[25], [26], [27]]. VPA exerts its pharmacological effects mainly in the central nervous system by inhibition of the citric acid cycle and elevation of γ-aminobutyric acid (GABA) level [26]. Chemicals with similar structure can have similar properties, but for toxicological properties this should be backed-up by in silico and in vitro evidence. We demonstrate how the model can be used to base the toxicity ranking of VPA and the above analogs on internal concentration estimates rather than on nominal water medium concentrations, thereby helping transferability of zebrafish embryo test results to human risk assessment. In addition, we use the data published by Brox et al. [21] on 16 other chemicals to discuss the model performance for a larger class of chemicals.
Section snippets
Test chemicals
Except for 4-eneVPA (Santa Cruz Biotechnology, Dallas, Texas, USA; 98 % purity), all other chemicals (valproic acid, 2,2-dimethylvaleric acid, 2-ethylbutyric acid, 2-ethylhexanoic acid, 2-methylhexanoic acid, 2-methylpentanoic acid, 2-propylheptanoic acid, 4-pentenoic acid and hexanoic acid) were purchased at the highest purity available from Sigma (Deisenhofen, Germany). After initial range-finding tests, the ten chemicals were tested at three to eight different concentrations prepared from a
Physiological parameters’ calibration
The model accounts for the embryo’s organ growth over time. We estimated the yolk consumption rate constant, the embryo’s volume at 120 hpf, the fractional volume of muscle at 120 hpf and the “other organs” fractional volume at 120 hpf on the basis of our experimental embryo volume data. Figure S1 (in Supporting Information) shows the observed and predicted organ growth over time. Predicted total embryo volume and embryo volume without yolk fit the data rather well: the median relative error
Conclusions
We developed the structure and equations of the first PBPK model for the zebrafish embryo. Our model integrates previously developed predictive models to account for the impact of physicochemical properties of the chemicals of interest on partition between test system components, cells of different types, and sub-cellular organelles. It also describes metabolism and anatomical volume changes of the embryo during growth. Its structure and parameter values integrate a large amount of biological
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgment
This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No. 681002 (Eu-ToxRisk).
References (49)
- et al.
In vivo study of teratogenic and anticonvulsant effects of antiepileptics drugs in zebrafish embryo and larvae
Neurotoxicol. Teratol.
(2018) - et al.
Modification and quantification of in vivo EROD live-imaging with zebrafish (Danio rerio) embryos to detect both induction and inhibition of CYP1A
Sci. Total Environ.
(2018) - et al.
Zebrafish embryos as an alternative to animal experiments—a commentary on the definition of the onset of protected life stages in animal welfare regulations
Reprod. Toxicol.
(2012) - et al.
Variation in predicted internal concentrations in relation to PBPK model complexity for rainbow trout
Sci. Total Environ.
(2016) - et al.
Predicting cadmium and lead toxicities in zebrafish (Danio rerio) larvae by using a toxicokinetic–toxicodynamic model that considers the effects of cations
Sci. Total Environ.
(2018) - et al.
Generic physiologically-based toxicokinetic modelling for fish: integration of environmental factors and species variability
Sci. Total Environ.
(2019) - et al.
Measuring bioconcentration factors and rate constants of chemicals in aquatic organisms under conditions of variable water concentrations and short exposure time
Chemosphere
(1992) - et al.
VIVD: Virtual in vitro distribution model for the mechanistic prediction of intracellular concentrations of chemicals in in vitro toxicity assays
Toxicol. Vitr.
(2019) - et al.
Histone deacetylase is a direct target of valproic acid, a potent anticonvulsant, mood stabilizer, and teratogen
J. Biol. Chem.
(2001) - et al.
Is the fish embryo toxicity test (FET) with the zebrafish (Danio rerio) a potential alternative for the fish acute toxicity test?
Comp. Biochem. Physiol. Part C Toxicol. Pharmacol.
(2009)
The fish embryo toxicity test as an animal alternative method in hazard and risk assessment and scientific research
Aquat. Toxicol.
A physiologically based toxicokinetic model for the uptake and disposition of waterborne organic chemicals in fish
Toxicol. Appl. Pharmacol.
Theoretical and mathematical foundation of the Virtual Cell based Assay – a review
Toxicol. Vitr.
Metabolic biotransformation half-lives in fish: QSAR modeling and consensus analysis
Sci. Total Environ.
Zebrafish as a model vertebrate for investigating chemical toxicity
Toxicol. Sci.
Modelling oral up-take of hydrophobic and super-hydrophobic chemicals in fish
Environ. Sci. Process. Impacts
A toxicokinetic model for fish including multiphase sorption features: a high-detailed, physiologically based toxicokinetic model
Environ. Toxicol. Chem.
Regard à travers le danio pour mieux comprendre les interactions hôte/pathogène
MdecineSciences
Think Small: Zebrafish as a Model System of Human Pathology
J. Biomed. Biotechnol.
DarT: The embryo test with the Zebrafish Danio rerio--a general model in ecotoxicology and toxicology
ALTEX.
The zebrafish reference genome sequence and its relationship to the human genome
Nature.
Le poisson zèbre (danio rerio) un modèle en biologie du développement
MdecineSciences
Extrapolating in vitro results to predict human toxicity
Toxicokinetics in aquatic systems: model comparisons and use in hazard assessment
Environ. Toxicol. Chem.
Cited by (0)
- 1
Present Address: Alderley Analytical Ltd. Alderley Park, Macclesfield, Cheshire, SK10 4TG, United Kingdom.
- 2
Present Address: Appleyard Lees IP LLP, The Lexicon Mount Street, Manchester, Greater Manchester, M2 5NT, United Kingdom.