Residual molecular and behavioral effects of the phenylpyrazole pesticide fipronil in larval zebrafish (Danio rerio) following a pulse embryonic exposure

Highlights

• After a 48-hour exposure, gene expression patterns remained altered in 9-day old larvae after being in clean water for 7 days.

• Fipronil suppressed glycogen and glycerate metabolism at the gene network level.

• Early pulse exposure altered behavioral patterns related to anxiety after 7 days depuration.

Abstract

Pesticides are typically applied to crops as acute applications, and residual effects of such intermittent exposures are not often characterized in developing fish. Fipronil is an agricultural pesticide that inhibits γ-amino-butyric acid (GABA) gated chloride channels. In this study, zebrafish (Danio rerio) embryos were exposed for 48 h (starting at ~3 h post fertilization, hpf) to various concentrations of fipronil (0.02 μg/L up to 4000 μg/L). Following this acute exposure, a subset of fish was transferred to clean water for a 7-day depuration phase. We hypothesized that a pulse exposure to fipronil during critical periods of central nervous system development would adversely affect fish later in life. After a 48 hour pulse exposure, survival was reduced in embryos exposed to 2 μg fipronil/L or greater. However, there was no further mortality during the depuration phase, nor were there changes in body length nor notochord length in larvae 9 dpf (days post-fertilization) compared to controls. Additional experiments were carried out at higher concentrations over 96 h (up to 4 dpf) to also elucidate developmental effects and teratogenicity of fipronil (43.7 μg/L up to 4370 μg/L). Fipronil at these higher concentrations significantly impacted the development of zebrafish, and the following morphometric and teratogenic effects were observed in 4 dpf fish; reduced body length, yolk sac and pericardial edema, reduced midbrain length, reduced optic and otic diameter, and truncation of the lower jaw. In depurated fish, we hypothesized that there would exist residual effects of exposure at the molecular level. Transcriptome profiling was therefore conducted on 9 dpf depurated larvae exposed initially for 48 h to one dose of either 0.2 μg/L, 200 μg/L or 2000 μg/L fipronil. The expression of gene networks associated with glycogen and omega-3-fatty acid metabolism were decreased in larvae exposed to each of the three concentrations of fipronil, suggesting metabolic disruption. Moreover, transcriptomics revealed that fipronil suppressed gene networks related to light-dark adaptation, photoperiod sensing, and circadian rhythm. Based on these data, we tested fish for altered behavioral responses in a Light-Dark preference test. Larvae exposed to >200 μg fipronil/L as embryos showed fewer number of visits (20–30% less) to the dark zone compared to controls. Larvae also spent a lower amount of time in the dark zone compared to controls, suggesting that fipronil strengthened dark avoidance behavior which is indicative of anxiety. This study demonstrates that a short pulse exposure to fipronil can affect transcriptome networks for metabolism, circadian rhythm, and response to light in fish after depuration, and these molecular responses are hypothesized to be related to aberrant behavioral effects observed in the light-dark preference test.

Introduction

The phenylpyrazole fipronil is a broad-spectrum chiral insecticide used for agricultural pest control and domestic lawn care, as well as veterinary treatment and prevention of ectoparasite infestations in pets and livestock. Fipronil is a γ -amino-butyric acid (GABA) receptor antagonist, acting to block GABA_A receptors in the central nervous system (CNS), which prevents GABAergic synaptic transmission (Gant et al., 1998). This leads to excessive neuronal stimulation that results in death. Fipronil (and its major metabolite fipronil sulphone) has strong affinity for invertebrate GABA receptors compared with mammals (Cole et al., 1993), and animals can show differences in sensitivity to the pesticide. In zebrafish for example, fipronil has been shown to be relatively toxic (Zheng et al., 2014). Thus, fipronil and its degradation products/metabolites can bind to and modulate GABA_A receptor modulators in aquatic organisms.

Fipronil can be problematic in indoor and aquatic environments due to its widespread use in residential areas and agriculture (Pitton et al., 2016; Glinski et al., 2018; Testa et al., 2019). Fipronil has been detected in the soil and in the water of metropolitan areas, in addition to agricultural areas (Sprague and Nowell, 2008; Cryder et al., 2019). The concentration of fipronil in surface waters that are found in close proximity to agricultural areas is typically below 0.01 μg/L (Gunasekara et al., 2007); however the pesticide has been detected at levels as high as 9.0 μg/L in regions where it has recently been applied to crops (Schlenk et al., 2001). In South Georgia, samples of water were taken from ponds and streams at an agricultural research site and maximum concentration of fipronil in surface water was reported at 0.19 μg/L (Glinski et al., 2018). In residential areas in Sacramento County and Orange County (CA, USA), concentrations of fipronil and its degradation products ranged from 0.20 to 0.44 μg/L, and could be detected in some cases as high as 1 μg/L (Gan et al., 2012). In animals, fipronil can be metabolized into fipronil sulphone, which can exacerbate sub-lethal effects due to a longer half-life (0.6 days for fipronil in comparison with 1.8 days for fipronil sulphone); fipronil sulphone also appears to have a higher binding affinity for vertebrate GABA_A receptors compared to fipronil (Hainzl et al., 1998). Degradation products of fipronil also show a relatively high degree of bioaccumulation in animals and this can be problematic over time (Zhao et al., 2005; Konwick et al., 2006; Gunasekara et al., 2007). Another point of concern for environmental exposure risk is that, depending upon the soil conditions, fipronil can persist and accumulate for months or even years after initial applications (Bonmatin et al., 2015). As such, continued monitoring and assessments of this agricultural and residential pesticide are warranted.

Fipronil can induce sub-lethal adverse effects in non-target aquatic organisms. In juvenile zebrafish for example, exposure to the pesticide can alter glutathione transferase enzyme activity in tissues (Wu et al., 2014). The same study also reported that ethoxyresorufin-O-deethylase (EROD), an indicator of cytochrome P450 1A1, is induced by fipronil in the brain, liver, gill, and muscle at water concentrations ranging from 2 to 20 μg/L (Wu et al., 2014), suggesting active metabolism of the contaminant. Uncontrolled neural excitation, neural degeneration, and scoliosis have been observed in larval fathead minnow (Pimephales promelas) after 24 h of exposure to fipronil at concentrations ranging from 31 μg/L to 153 μg/L (fractions of the LC₁₀ value) (Beggel et al., 2012). In another study investigating embryonic/larval zebrafish, neurotoxic effects of fipronil were documented following a 30 hour exposure to fipronil at concentrations ranging 33 μg/L to 5000 μg/L; this was accompanied by scoliosis, notochord degeneration, and an up-regulation in the expression of genes involved in detoxification (Stehr et al., 2006). Wang and colleagues (Wang et al., 2016) reported that 10 μg/L fipronil increased locomotor speed in larval zebrafish, and that fish showed abnormal photoperiod accommodation. Behavioral responses were also associated with altered metabolites that included fatty acids and glycerol, in addition to reduced levels of glycine and branched amino acids. More recently, there are reports that female zebrafish exposed to fipronil (1–10 μg/L) for 28 days can transfer fipronil and its major metabolite fipronil sulphone to the F1 generation, leading to dysregulation in the expression of the thyroid hormone axis (Xu et al., 2019). Although these findings demonstrate adverse effects in fish at multiple levels of biological organization, less is known about the global transcriptome response and how molecular changes associate with higher level effects.

The objectives of this study were to determine whether a short-term, environmentally relevant pulse exposure to fipronil during embryogenesis resulted in any residual morphometric or behavioral deficits in larval zebrafish. This experimental design was chosen because agricultural and residential applications of pesticides typically occur as pulse exposures (i.e. single application of the pesticide at multiple times) followed by run-off into aquatic environments. While a number of studies utilize a continuous dosing regime, fewer studies investigate whether short pulse exposures to pesticides during critical periods of development induce significant adverse outcomes later in life (Gormley and Teather, 2003; Weis, 2014). These effects can be persistent, and there is now strong evidence for latent effects in fish over multiple generations for aquatic contaminants including fungicides (Cao et al., 2019), industrial chemicals (Sadoul et al., 2017), and pharmaceuticals (Vera-Chang et al., 2018). As pointed out above, fipronil has also been shown to affect F1 zebrafish after a parental exposure (Xu et al., 2019). Here, we measured morphology, transcriptome profiles, and behavioral responses in 9 dpf larvae after a 48-hour exposure to fipronil (0.02 μg/L–2000 μg/L) early in development. We also characterized teratogenicity of fipronil in a continuous exposure regime. Zebrafish are widely used vertebrate models for toxicity testing (Nagel, 2002; Hill et al., 2005); data such as this inform on safe water quality levels and health assessments in aquatic organisms. Transcriptomic profiles were hypothesized to relate to the known mode of action of fipronil toxicity, that being neurotransmitter disruption, GABA receptor antagonism, oxidative damage, and metabolic imbalances.