Both CLO and IMD, individually, were found to be toxic to zebrafish embryos at high concentrations, namely 500 and 1,000 μg/mL (Supplementary Figure S1). Potential mixture toxicity tests for zebrafish embryos were designed with ECx values estimated from a single toxicity test (Table 1; Supplementary Figure S1), which involved a range from the lowest effective concentration (EC5) to the half effective concentration (EC50) for each target chemical. Potential mixture toxicity tests in the embryonic stages of zebrafish were performed under the same conditions as the single toxicity tests, and the mixing ratios of CLO and IMD were fixed at 1:1, reflecting the detected mixing ratio in the natural environment (Bonmatin et al., 2015; Sánchez-Bayo et al., 2016).

To verify the developmental effects in zebrafish embryos caused by exposure to realistic mixtures, mixture toxicity tests with zebrafish embryos were performed at a range of environmentally relevant concentrations, and three exposure concentrations at a 1:1 mixing ratio (i.e., 1, 10, and 100 ng/mL of each chemical). The exposure concentrations and mixing ratios of CLO and IMD in the mixtures reflected actual measured environmental concentrations (Bonmatin et al., 2015; Sánchez-Bayo et al., 2016). In addition, to examine the toxicity pathway related to embryonic development, zebrafish embryos exposed to mixtures of CLO and IMD were sampled at 6 days after exposure and placed in a sterilized 1.5 mL microtube with 100 μL of RNAlater™ (Thermo Fisher Scientific, Massachusetts, United States) and were stored at −40°C until use in the expression change analysis of target genes. Realistic mixture toxicity tests using zebrafish embryos were performed under the same conditions as the single-toxicity tests.

The concentrations of CLO and IMD in the zebrafish test media were analyzed using high performance liquid chromatography–tandem mass spectrometry (HPLC-MS/MS) with an Agilent 6,420 Triple Quadrupole MS instrument connected to an Agilent 1,260 Infinity HPLC system containing a binary pump, autosampler, and degasser (Agilent, Santa Clara, CA, United States). Chromatographic separation was performed using a Poroshell 120 EC-C18 column (2.1 × 100 mm, 2.7 µm particle size; Agilent, Santa Clara, CA, United States) connected to a C18 guard column (Phenomenex, Torrance, CA, United States). Mobile phase A was 0.1% formic acid in water, and mobile phase B was acetonitrile. HPLC was run using a linear gradient as follows: 0–5 min, 20%–40% B; 5–7 min, 40%–70% B. The column was then equilibrated with 20% B for 4.9 min. The flow rate was set at 0.4 mL/min, and the injection volume was set at 10 μL. MS analysis was conducted in the positive ion mode using electrospray ionization. The MS conditions were optimized as follows: capillary voltage, 4000 V; sheath gas temperature, 370°C; sheath gas flow, 12 L/min; gas temperature, 350°C; gas flow, 8 L/min; nebulizer gas pressure, 40 psi; fragmentor voltage, 75 V for CLO and 80 V for IMD. On the selected reaction monitoring mode (SRM), the collision energy (CE) and selected reaction monitoring channels for the analytes were as follows: CLO, CE 10 V and m/z 250.0→169.0; CLO-d3, CE 10 V and m/z 253.0→172.0; IMD, CE 12 V and m/z 256.0→209.0; IMD-d4, CE 10 V and m/z 260.1→213.1. HPLC-MS/MS data were collected and processed using MassHunter Workstation Software Qualitative Analysis (ver. B.06.00, Agilent, Santa Clara, CA, United States). The relative error (RE) values for the residual concentration of each target chemical were calculated as: RE (%) = (mean concentration of media samples–nominal concentration)/nominal concentration) × 100).

To examine the transcriptional alteration of target genes linked to zebrafish embryonic development, The total RNA from 20 zebrafish embryos was extracted using a RNeasy Mini kit (Qiagen, Hilden, Germany), following the manufacturer’s instructions. The total RNA concentration was measured spectrophoto-metrically at 260/280 nm using a Gen5™ spectrophotometer (BioTek^®, Winooski, VT, United States). First-strand cDNA was synthesized with 0.5 µg of total RNA using a Superscript III First-Strand Synthesis System kit (Thermo Fisher Scientific, Vilnius, Lithuania) according to the manufacturer’s manual.

The primer sequences for target genes were designed to use for quantitative real-time polymerase chain reaction (qRT–PCR): namely aryl hydrocarbon receptor 2 (ahr2), cytochrome P450 monooxygenases (i.e., cyp1a1 and cyp1b1), nuclear factor erythroid 2-related factors (i.e., nrf1a and nrf2a), tumor suppressor gene (p53), thyroid stimulating hormone beta subunit (tsh-β), and thyroid hormone receptors (i.e., thraa and thrb) (Supplementary Table S1) (Fan et al., 2018; Liu et al., 2011; Ren et al., 2020; Williams et al., 2013). The qRT–PCR was performed using a Thermal Cycler Dice^® Real Time System III (Takara, Tokyo, Japan) using the Go Taq^® qPCR Master Mix (Promega, Madison, WI, United States). The qRT–PCR reactions were conducted as follows: an initial hot-start activation step at 95°C for 2 min, followed by 40 cycles at 95°C for 15 s and 60°C for 60 s. Melting curve analyses were performed to optimize primers for qRT–PCR performance. These transcript abundances for the target genes were normalized to a housekeeping gene (i.e., β-actin). The relative quantification of target genes expression was calculated using the 2^−ΔΔCT method (Park et al., 2022).

The stability and persistence of CLO and IMD in binary mixtures were assessed through two different types of mixing combinations, high and low concentrations (i.e., mixtures containing 10 and 100 ng/mL of each chemical, respectively), at 6 days after exposure. CLO and IMD were not detected in the control solution. Residual concentrations (compared to initial concentrations) of CLO and IMD in the mixture solutions ranged from 92.1% to 103.0% in the case of CLO and from 91.6% to 101.7% in the case of IMD, which indicates that both of these neonics were stable in the mixtures; that is, no or negligible volatilization or denaturation (Figure 2). Furthermore, the half-life of CLO reaches 56 days in sediment and up to 1,386 days in soil, and the half-life of IMD reaches 129 days in sediment and up to 228 days in soil (Mason et al., 2013). These results indicate that mixtures of CLO and IMD can steadily accumulate and persist in aquatic environments after agriculture or soil applications. These phenomena are not independent of the continuous increase in the detected concentrations of CLO and IMD in surface waters over the past 15 years and may lead to acute toxicity in aquatic organisms, as well as chronic toxicity (Sánchez-Bayo et al., 2016; Simon-Delso et al., 2015). Therefore, to better understand the toxic effects of CLO and IMD mixtures, considering their detected environmental concentrations, further studies on the phenotypic toxicity caused by short-term exposure and the toxic mechanisms for predicting chronic toxicity are required.

Some studies reported that CLO and IMD might be remained in high concentrations in surface waters, through leaching and runoff under worst-case environmental scenarios (e.g., incorrect handing or improper disposal): up to IMD 320 ng/mL in Netherlands (Van Dijk et al., 2013) and CLO up to 170 ng/mL in Unite States (Miles et al., 2017). Therefore, we hypothesized that single and mixture of ≤100 ng/mL CLO and IMD can reflect as the detectable concentrations in the aquatic environment, including worst-case scenarios. In the present study, phenotypic toxicity was not observed in binary mixtures of CLO and IMD that reflected the detectable concentrations in the aquatic environment. There was no significant difference between all exposure groups compared with solvent control group, indicating that embryonic toxicity on zebrafish embryos, i.e., lethality and abnormality, were less than 15% after binary mixture exposure (Table 2). Thus, when exposed to detectable concentrations in the aquatic environment (i.e., ≤100 ng/mL), CLO and IMD, whether singly or in mixtures, are little to cause short-term developmental toxicity in zebrafish embryos. However, a strong acute toxicity to zebrafish was shown in the neonics mixtures-exposure groups and this phenomenon was exerted by synergistic effects with mixing ratio and concentration of components in the mixtures. These results imply that neonics mixture toxicity was associated with co-existing components and concentrations in the mixtures (Shukla et al., 2017; Wang et al., 2017). Therefore, although there was no phenotypic toxicity in zebrafish embryos, developmental defects in zebrafish embryos may appear upon co-exposure at low concentrations by mixing ration and concentrations of CLO and IMD (Wu et al., 2018).

In this study, because ≤10 ng/mL CLO and IMD were simultaneously detected in runoff samples after spraying on agricultural environment (Woodward et al., 2022), the combined effects of CLO and IMD on zebrafish embryonic development were compared with 10 ng/mL single chemical-exposure group as a positive control. The mixture effects of CLO and IMD at realistic environmental concentration and mixing ratio, including worst-case scenarios (i.e., ≤100 ng/mL), were assessed with the toxic alteration of two different pathways which can affect the development of zebrafish embryos: the expression levels of target genes that are induced to cell damage by oxidative stress response and involved in thyroid hormone biosynthesis.

Aryl hydrocarbon receptor-mediated oxidative stress can induce toxic signaling pathways linked to cell damage, such as irritation of the inflammatory reactions or the induction of peroxidases by cytochrome P450-dependent oxygenase, during the early stages of embryonic development (Jin et al., 2020; Ren et al., 2020; Shankar et al., 2020). Several studies reported that the activation to oxidative stress is regulated by the redox-sensitive nuclear factor erythroid 2-related factors (Rousseau et al., 2015; Ulin et al., 2019; Williams et al., 2013). Because oxidative stress response also may affect the expression of proto-oncogene (p53), a tumor suppressor gene that regulates the cell cycle, we targeted genes which are involved aryl hydrocarbon receptor-mediated oxidative stress response, including p53, to examine the mixture toxicity of CLO and IMD at environmental concentrations (Chen, 2016; Fan et al., 2018; Mugoni et al., 2014). The adverse effects caused by the form of realistic mixtures of CLO and IMD in zebrafish embryos were evaluated with the expression levels of targeted genes after binary mixtures exposure.

Interestingly, the expression levels of target genes linked to the cell damage pathway [aryl hydrocarbon receptor (ahr2), cytochrome P450 subunits (cyp1a1 and cyp1b1), nuclear factor erythroid 2-related factors (nrf1a and nrf2a), and p53] were markedly up- or downregulated in the mixture exposure groups compared to the corresponding gene expression levels in the control group (P < 0.05, Figure 3; Supplementary Table S2). These results indicate that exposure to binary mixtures of CLO and IMD may cause developmental defects in zebrafish embryos through cell damage mediated by chemical stimulation, even at low concentrations. Previously other studies were reported that the transcriptional expression of ahr2 mRNA, which reacts to chemicals with aromatic carbocyclic rings (e.g., neonics), induces the transcriptional activation of cytochrome P450 monooxygenases (cyp1a1 and cyp1b1) and nuclear factor erythroid 2-related factors (nrf1a and nrf2a), indicating that their transcriptional alterations involve cellular oxidative stress response to xenobiotics (Jin et al., 2020; Ren et al., 2020; Rousseau et al., 2015; Shankar et al., 2020; Ulin et al., 2019; Williams et al., 2013). Excessive oxidative stress downregulated the transcriptional level of p53, leading to cell cycle arrest and/or cytotoxicity (Chen, 2016; Fan et al., 2018; Mugoni et al., 2014). Consequently, although abnormalities were not observed in short-term toxicity tests, in the present study, alterations in the transcriptional levels of target genes for cell damage implied that exposure to binary mixtures of CLO and IMD may induce developmental defects in zebrafish embryos through DNA damage linked to acute oxidative stress responses (P < 0.05, Figure 3; Supplementary Table S2).

Thyroid hormones are involved in early development and growth-related biological processes and functions in zebrafish embryos (Liu and Chan, 2002). In particular, to determine the zebrafish embryonic toxicity for potentially harmful substances, thyroid stimulating hormone beta (tsh-β) which promotes thyroid hormone secretion and thyroid hormone receptor alpha and beta (thraa and thrb) have been used as toxicological indicators.

In this study, the transcriptional levels of tsh-β, thraa, and thrb were significantly decreased in the single- and binary mixture groups of CLO and IMD compared to those in the control group (P < 0.05, Figure 4; Supplementary Table S2). Our findings indicate that CLO and IMD, either singly or combined in mixtures, can disturb thyroid hormone biosynthesis in the embryo-larva development of zebrafish. Similar to our findings, some studies reported that hazardous chemicals distributed in natural environment, including neonics, can interfere with the activation of thyroid hormones through receptor-mediated signal transduction (Boas et al., 2012; Liu et al., 2011; Ma et al., 2019; Walter et al., 2019). Thus, these results imply that the disturbance of thyroid hormone biosynthesis may be due to the intrinsic toxic characteristics of single chemicals and can be promoted by binary mixture exposure (P < 0.05, Figure 4; Supplementary Table S2). However, synergistic effects on thyroid hormonal activity were not confirmed in the mixtures. To more clearly elucidate the combined effects on developmental endocrine system in zebrafish, such as thyroid hormone biosynthesis, further studies examining the developmental impairment of zebrafish embryos through chemical interactions in the mixtures are required.

Taken together, a novel finding is that exposure to binary mixtures of CLO and IMD in the aquatic environment (i.e., concentration and ratio) can cause the promotion of cell damage by oxidative stress response and the suppression of thyroid hormone synthesis which is involved embryonic development. This phenomenon indicate that the expression alteration of target genes linked to cell damage and thyroid hormonal synthesis could impair the embryonic development in zebrafish, by the components and concentrations of chemical mixtures (Wang et al., 2020; Wu et al., 2018). Our results will be useful in examining the combined effects of mixtures following short-term exposure, to predict the developmental toxic effects of chronic exposure to CLO and IMD. However, to clearly confirm the developmental toxicity caused by significant changes of target genes linked to cell damage and thyroid hormonal synthesis, it is necessary to investigate the effects of long-term exposure as well as the expressions of corresponding genes.