Protective effects of genistein and estradiol on PAHs-induced developmental toxicity in zebrafish embryos

Abstract

The toxicity of exposure to polycyclic aromatic hydrocarbons (PAHs) or phytoestrogen is relatively well characterized. However, the toxicity of combined exposure to PAHs and phytoestrogen is not well investigated. In the present study, benzo(a)pyrene (B(a)P) and benzo(k)fluorathene (B(k)F), genistein, along with 17β-estradiol (E2), were investigated for their single and combined developmental toxicity using zebrafish embryos as model system. We demonstrated that two representative PAHs, both B(a)P (≥1 μM) and B(k)F (≥10 μM), can cause significant malformation and mortality in developing zebrafish embryos. The toxicity effect of B(a)P was in general higher than that of B(k)F. Developmental exposure to high level of genistein (>20 μM) or E2 (>10 μM), also caused significant malformation and mortality in zebrafish larvae at 120 hours post fertilization (hpf). However, different toxic effects were observed for the combined exposure to PAHs and phytoestrogen in zebrafish. Lower doses of genistein (1 and 10 μM) and E2 (0.1 and 1 μM), when used in combination with high concentration of B(a)P (1 μM) or B(k)F (20 μM), can significantly suppress the toxicity effect of B(a)P and B(k)F in developing zebrafish embryos. The beneficial effect of genistein may be due to the inhibition of cytochrome P450 enzymes via directly interacting with aryl-hydrocarbon receptor (AhR) pathway, or disturbing the AhR pathway through interacting with estrogen receptor pathway.

Introduction

Endocrine disrupting chemicals (EDCs) are widely distributed in the environment and derive from many different human activities (pesticides and industrial products) and are often found in many natural products (phytoestrogens). Currently, there are considerable health concerns over the environmental occurrence of EDCs, since these molecules have the potential to disrupt the body homeostasis of endogenous hormones and consequently affect development, reproduction and behavior in many organisms. ¹

Polycyclic aromatic hydrocarbons (PAHs), one major class of EDCs, are ubiquitous environmental contaminants as a result of incomplete combustion of fossil fuels and other organic matter. ² PAHs are resistant to environmental degradation due to their highly hydrophobic nature, causing detrimental biological effects, toxicity, mutagenecity and carcinogenicity. ^3,4 Benzo(a)pyrene (B(a)P) and benzo(k)fluorathene (B(k)F) are two typical representatives of the PAHs. The PAHs are bioactivated by cytochrome P450 (CYP) 1A1/CYP1B1 enzymes to reactive intermediates that bind to DNA, a critical step in the initiation of carcinogenesis. ⁵ Aryl-hydrocarbon receptor (AhR) plays a critical role in the induction of CYP1 enzymes (i.e. CYP1A1, CYP1A2 and CYP1B1) by PAHs, many of which are AhR ligands. ⁶ In addition to AhR-dependent toxicological pathway, several studies also suggested that PAHs or their metabolites may interact with estrogen receptors (ERs) and exhibited either weak estrogenic or antiestrogenic responses. ⁷

Phytoestrogens, including isoflavones, coumestans and lignans, are nonsteroidal and plant-derived compounds and have been found in many fruits, vegetables and whole grains commonly consumed by humans. Soybeans, clover, alfalfa sprouts and oilseeds are the most significant dietary sources of isoflavones, coumestans and lignans. ⁸ Phytoestrogens are structurally very similar to the estrogen 17β-estradiol and can exhibit selective ER modulating activities. ⁹ Some phytoestrogens also alter CYPs through binding to AhR, acting as either AhR agonists or antagonists. ¹⁰ Genistein (4′,5,7-trihydroxyisoflavone) belongs to the family of isoflavones found in high levels in soybean, its prophylactic use has been studied with regards to cancer, osteoporosis, cardiovascular disease and for the treatment of menopausal symptoms. ¹¹ However, the beneficial effects of long-term exposure to phytoestrogen genistein are controversial and both tumorigenic and antitumorigenic effects have been reported, especially in hormone-dependent breast and prostate cancer. ^12,13

Humans are exposed to two major classes of EDCs, PAHs (environmental pollutants) and phytoestrogens (dietary component), throughout their daily lives. Both the classes of chemicals have estrogenic or antiestrogenic activity via binding either to ER or AhR. The two classes of EDCs are ubiquitous and coexist under many situations in real human life. When coexposure occurs, phytoestrogens and PAHs can exert their unique toxicity independently, but they may also interfere with each other’s activity in some aspects of biological event. For example, the soya isoflavone genistein has been shown previously to be an inhibitor of PAH metabolite binding to DNA, and the oxidative DNA damage induced by PAHs in noncancerous breast cells MCF-10A could be removed by cotreating with 10 μm genistein. ¹⁴

The toxicity of exposure to phytoestrogens or PAHs has been studied intensively on various cell lines and animal models. However, most of these experiments examined the toxic effects following single chemical exposure. The combined toxic effects of exposure to phytoestrogens and PAHs have not been fully investigated yet using an in vivo animal model, while there are only a few in vitro studies. Recently, zebrafish embryos have become one of the most popular and well-established models for assessing the toxicity of chemicals and drugs. ¹⁵ This model is exceptional because of its high reproductive capacity and the transparency of embryos as well as its rapid development that allows morphological observations of the organogenesis following chemical treatment. In this study, we chose B(a)P and B(k)F as two representatives of the PAHs and genistein as a representative of phytoestrogen. The aim is to explore the complex biological or toxic effect following coexposure to phytoestrogens and PAHs using zebrafish as a model system.

Materials and methods

Test chemicals

B(a)P (purity >98.7% w/w) and B(k)F (purity >99.6% w/w) were purchased from AccuStandard (New Haven, Connecticut, USA). Genistein (purity >97% w/w), 17β-estradiol (E2; purity >98% w/w) and dimethyl sulfoxide (DMSO) were purchased from Sigma (St Louis, Missouri, USA). All tested chemicals were dissolved in DMSO and primary stock solutions were stored at −20°C per manufacture’s instruction. A serial dilution was made by 100% (v/v) DMSO. The final DMSO concentration in the treatment solution is 0.1% (v/v). All other chemicals used in this study were of analytical grade.

Fish husbandry and embryo collection

Adult zebrafish of wild-type strain (AB line) were raised and kept at standard laboratory conditions of 28°C with a 14:10 dark/light photoperiod (lights on at 08:00 a.m.) in a recirculation system according to standard zebrafish breeding protocols. ¹⁶ Water supplied to the system was filtered by reverse osmosis (pH 7.0–7.5), and Instant Ocean^® salt was added to the water to raise the conductivity to 450–1000 µS/cm. The adult fish were fed twice daily with live Artemia (Jiahong Feed Co., Tianjin, China) and dry flake diet (Zeigler, Aquatic Habitats, Apopka, Florida, USA).

Zebrafish embryos used for chemical exposure were obtained from spawning adults in tanks overnight with the sex ratio of 1:1. Embryos were collected within 1 hour after the light was switched on in the morning and rinsed in fish water. ¹⁶ The fertilized and normal healthy embryos were collected and staged for the following experiment under dissection microscope (Nikon, Japan) according to the previous descriptions. ¹⁷ Fish care was in accordance with the approved Institutional Animal Care and Use Committee protocols at Wenzhou Medical College, China.

Exposure of zebrafish embryos

Chemical exposure was initiated at 50% epiboly stage (6 hours post fertilization (hpf)). Briefly, chemical stock solutions were diluted in fresh fish water to create a series of working solutions with the concentration gradient as followings: genistein: 0.01, 0.1, 1, 10, 20 and 50 μM; E2: 0.1, 1 and 10 μM; B(a)P: 0.01, 0.1, 0.5, 1 and 10 μM; and B(k)F: 0.01, 0.1, 1, 10 and 50 μM. Experiments were conducted in six-well plates with 10 embryos per well in 3 mL working solution. The control group was treated with fish water containing 0.1% DMSO. The exposure started at 6 hpf and ended at 120 hpf. No fish water was changed during the 5-day exposure. Plates were covered with lids to avoid evaporation. For coexposure experiment, embryos were treated with PAHs (B(a)P or B(k)F) and phytoestrogen (genistein or E2) at the same time, which had the following concentrations: genistein (1 or 10 μM), E2 (0.1 or 1 μM), B(a)P (0.1 or 1 μM) and B(k)F (10 or 20 μM). To further confirm the protective effect of genistein and E2 against the PAHs, zebrafish embryos were pretreated with genistein (1 or 10 μM) or E2 (1, 2, 5 μM) for 24 hours beginning 6 hpf. At 30 hpf, fish water containing genistein or E2 was removed and embryos were exposed to 1 μM B(a)P only until 96 hpf. In this study, the concentrations of the chemicals were chosen based on the published literatures ^18,19 and our preliminary dose-range finding study, since there is very limited information on zebrafish model with these chemicals.

For all experiments, embryos were examined daily for developmental defect. Hatch, mortality and malformation of embryos were recorded and dead embryos were removed to avoid possible adverse effects to other living embryos. The percentage mortality was calculated as the number of dead embryos during the course of experiment divided by the total number of embryos. Cessation of heartbeat and blood circulation was used as an indicator for death. The percent malformation was calculated as the number of embryos having any deformities (including dead animals resulted from malformation) divided by the total number of embryos. For each experimental run, all treatments were performed in triplicate. All experiments were replicated three times, thus yielding 90 embryos per treatment group in total.

Statistical analysis

One-way analysis of variance followed by Tukey’s honestly significant difference procedure was used to compare differences among various treatment groups. All statistical analyses were run using SAS 9.0 software (SAS Institute Inc., Cary, North Carolina, USA) and p < 0.05 was considered a statistically significant difference. The data were reported as mean ± SD.

Results

Developmental toxicity of exposure to B(a)P or B(k)F on zebrafish embryos

In this study, exposure to B(a)P and B(k)F significantly increased the rate of malformation and mortality at the concentration of 1 and 10 μM, respectively, at 120 hpf (Figure 1(a) and (b)). When exposed to 1 μM B(a)P, the rates of malformation and mortality of embryos were nearly 100% at 120 hpf and the hatch rate was below 80%. When the exposure level increased to 10 μM, B(a)P treatment killed all the embryos prior to 48 hpf before hatching (Figure 1(a)). Compared with B(a)P, the teratogenic and lethal effects of B(k)F were relatively low. When exposed to 10 μM of B(k)F, the malformation and mortality of embryos were 69% and 53%, respectively (Figure 1(b)). When exposed to 50 μM B(k)F, the malformation and mortality of embryos were increased to 85% and 69%, with the hatch rate below 77% (Figure 1(b)). As we observed, developmental exposure to higher concentration of B(a)P (1 μM) and B(k)F (10 or 50 μM) induced similar types of malformations on zebrafish larvae at 120 hpf, for example, pericardial edema, somite deformity, yolk sac edema and weak pigmentation (Figure 1(c) to (f)). However, no significant differences were observed on the type and incidence of malformations between two chemical treatments. In general, hatched embryos had very slow heart beat and difficulty in movement, and usually they will not survive longer than a few days beyond 120 hpf.

Figure 1.

The rate of hatch, malformation and mortality in the zebrafish larvae at 120 hpf following exposure to PAH. Wild-type zebrafish embryos (AB line) were exposed to varying concentration of B(a)P (a) or B(k)F (b) beginning 6 hpf and the rate of hatch, malformation and mortality were monitored from 6 hpf until 120 hpf. Exposure to both B(a)P and B(k)F significantly increased the rate of malformation and mortality at the concentration of 1 and 10 µM and decreased the hatch rate at 1 and 50 µM, respectively (a and b). (c–f) Representative microphotographs of various types of malformations in zebrafish larvae (120 hpf) caused by B(a)P and B(k)F developmental exposure. The experiment was repeated three times, each with a new batch of embryos from different parental stocks. N = 20–30 per treatment group in each replicate experiment. hpf: hours post fertilization; B(a)P: benzo(a)pyrene; B(k)F: benzo(k)fluorathene; PAH: polycyclic aromatic hydrocarbon.

Developmental toxicity of genistein or E2 exposure on zebrafish embryos

We found that either genistein or E2 exposure significantly increased the rate of malformation and mortality at the concentration of 20 and 10 μM, respectively, at 120 hpf (Figure 2(a) and (b)). When exposed to 20 μM genistein, the malformation and mortality of embryos were 100% and 84%, respectively, at 120 hpf and the hatch rate was below 75%. When the exposure level increased to 50 μM, genistein treatment further increased mortality to 98% while the hatch rate and malformation maintain same as 20 μM treatment group (Figure 2(a)). Compared with genistein, E2 developmental exposure caused more severe teratogenic and lethal effects. When exposed to 10 μM of E2, the malformation and mortality of embryos were 98% and 51%, respectively, with no effect on hatching (Figure 2(b)). As we observed over the course of zebrafish early development, exposure to higher concentration of genistein (20 or 50 μM) and E2 (10 μM) had a similar teratogenic effect. Various types of malformations on zebrafish larvae were observed including pericardial edema, somite deformity, yolk sac edema, defects in submaxilla and ocular region, enlarged head, weak pigmentation and tail curvature (Figure 2(c) to (f)). No significant differences were observed on the type and incidence of malformations between genistein and E2 exposure.

Figure 2.

The rate of hatch, malformation and mortality in zebrafish larvae at 120 hpf following genistein and E2 exposure. Wild-type zebrafish embryos (AB line) were exposed to varying concentration of genistein (a) or E2 (b) beginning 6 hpf, and the rate of hatch, malformation and mortality were monitored from 6 hpf until 120 hpf. Exposure to both genistein (20 µM) and E2 (10 µM) significantly increased the rate of malformation and mortality (a and b). Genistein exposure at 20 µM also significantly decreased hatch rate. However, developmental exposure to E2 at 10 µM did not affect the hatch rate (b). (c–f) Representative microphotographs of various types of malformations in zebrafish larvae at 120 hpf following developmental exposure to genistein and E2. The experiment was repeated three times, each with a new batch of embryos from different parental stocks. N = 20–30 per treatment group in each replicate experiment. E2: 17β-estradiol; hpf: hours post fertilization.

Developmental toxicity of combined B(a)P and genistein or E2 exposure on zebrafish embryos

Our data demonstrated that embryos from all coexposure treatment groups had significant lower mortality and malformation rate in comparison with B(a)P (1 μM) single exposure, suggesting that phytoestrogen, genistein or E2, when used at lower dose, may protect embryos against the teratogenic and lethal effect of B(a)P in zebrafish larvae at 120 hpf (Figures 3 and 4). We also observed similar protective effect of genistein (1 or 10 μM) and E2 (0.1 or 1 μM), when embryos were coexposed to B(k)F (20 μM) with genistein or E2 (data not shown).

Figure 3.

The rate of malformation and mortality in zebrafish larvae at 120 hpf following coexposure to B(a)P and phytoestrogen genistein. Wild-type zebrafish embryos (AB line) were exposed to B(a)P (0.1 or 1 µM) or genistein (1 or 10 µM), or to B(a)P and genistein in combination beginning from 6 hpf. Mortality and malformation rate were monitored and calculated until 120 hpf. Genistein both at 1 or 10 µM can significantly antagonize the teratogenic and lethal effect of B(a)P in zebrafish larvae at 120 hpf when embryos were coexposed to B(a)P and genistein (a and b). This experiment was repeated three times, each with a new batch of embryos from different parental stocks. N = 20–30 per treatment group in each replicate experiment. *Significantly different from vehicle control at p < 0.05. ^#Significantly different from B(a)P (1 µM) treatment group at p < 0.05. hpf: hours post fertilization; B(a)P: benzo(a)pyrene.

Figure 4.

The rate of malformation and mortality in zebrafish larvae at 120 hpf following coexposure to B(a)P and estradiol (E2). Wild-type zebrafish embryos (AB line) were exposed to B(a)P (0.1 or 1 µM), E2 (0.1 or 1 µM), or to B(a)P and E2 in combination beginning from 6 hpf. Mortality and malformation rate were monitored and calculated until 120 hpf. E2 both at 0.1 or 1 µM can significantly antagonize the teratogenic and lethal effect of B(a)P in zebrafish larvae at 120 hpf when embryos were coexposed to B(a)P and E2 (a and b). This experiment was repeated three times, each with a new batch of embryos from different parental stocks. N = 20–30 per treatment group in each replicate. *Significantly different from vehicle control at p < 0.05. ^#Significantly different from B(a)P (1 µM) treatment group at p < 0.05. E2: 17β-estradiol; hpf: hours post fertilization; B(a)P: benzo(a)pyrene.

To further confirm the protective effect of genistein and E2 against the PAHs, zebrafish embryos were pretreated with genistein (1 or 10 μM) or E2 (1, 2 and 5 μM) for 24 hours beginning 6 hpf. At 30 hpf, medium containing genistein or E2 was removed and embryos were exposed to 1 μM B(a)P only until 96 hpf. In relation to the pretreatment with genistein or E2, our result demonstrated that the malformation rate and mortality of embryos at 96 hpf were both 78% when treated with 1 μM B(a)P only. However, the malformation rate was dropped to 12% and mortality dropped to 5%–8%, when embryos were pretreated with genistein at 1 and 10 μM (data not shown). Pretreatment with E2 at 1, 2 and 5 μM similarly blocked the toxic effects of subsequent 1 μM B(a)P exposure. The malformation and mortality of pretreated fish were both significantly lower than that of 1 μM B(a)P treatment group at 96 hpf (p < 0.05).

Discussion

In this study, we demonstrated that two representative PAHs, B(a)P (≥1 μM) and B(k)F (≥10 μM), can cause significant malformation and mortality in developing zebrafish embryos. The toxicity effect of B(a)P was in general much pronounced than that of B(k)F. The observed phenotypes of malformation in fish embryos exposed to B(a)P or B(k)F included somite deformities, pericardial edema, yolk sac edema and weak pigmentation, which were consistent with previous reports. ^20,21 Developmental exposure to high level of genistein (≥20 μM) or E2 (≥10 μM) also caused significant malformation and mortality in zebrafish larvae at 120 hpf. Our study also demonstrated for the first time the different toxic effects for combined exposure to PAHs and phytoestrogen in zebrafish model. Interestingly, lower doses of genistein (1 and 10 μM) and E2 (0.1 and 1 μM), when used in combination with high concentration of B(a)P (1 μM) or B(k)F (20 μM), can significantly suppress the toxicity effect of B(a)P and B(k)F in developing zebrafish embryos.

PAHs and phytoestrogens are two major classes of EDCs in the environment. Numerous studies have discovered that EDCs (i.e. phytoestrogen from diet and PAHs from synthesized compound), like endogenous estrogen E2, are capable of binding with ER and exert the estrogenic and antiestrogenic activity. ⁹ The ERs, ERα and ERβ, play a central role in mediating the biological effects of estrogen, which is a key regulatory hormone that affects numerous physiological processes. ²² The expression of ERα and ERβ vary dramatically among different organs or cell types. The two receptors regulate different sets of biological functions and induce dissimilar responses within the same cell type or tissue. Previous reports have shown that phytoestrogens (genistein and coumestrol) are ERβ stimulants and preferentially binds human ERβ over ERα; compared with ERα, ERβ exhibits a 7- to 70-fold greater binding affinity for genistein, whereas E2 binds ERα and ERβ with equal affinity. ^18,23 In this study, zebrafish embryos were developmentally coexposed to the PAHs with genistein or E2 during early developmental stage. Our results demonstrated that genistein as strong ERβ stimulant is capable of suppressing the toxicity of B(a)P with even higher efficiency than E2, suggesting that ERβ, rather than ERα, may play a central role in mediating the beneficial effects of phytoestrogen genistein against toxicity of PAH, if phytoestrogen exert protective effects via ER-mediated pathway.

How does phytoestrogen suppress the toxicity of PAHs in zebrafish embryos? Based on the published studies, we speculate that the underlying mechanisms may involve ER-mediated pathways or AhR-mediated pathways. First, it has been shown that activation of ERα may affect the AhR gene expression. ²⁴ As genistein can still bind to ERα with low affinity, ^18,23 genistein may counteract the toxicity of PAHs by regulating the expression level of AhR via ERα-mediated pathway. Second, CYP genes are sensitive to estrogen treatment. ²⁵ It has been shown that some phytoestrogens can modulate the CYP system including activation or inhibition of these enzymes. ^26,27 For instance, previous studies demonstrated that genistein inhibits CYP1A1, CYP1B1 and CYP3A4 activity in vitro ^28,29 and inhibits CTP1A2 and CTP2E1 activity in vivo. ³⁰ This inhibited CYPs activity may interfere with the formation of reactive intermediates of PAHs and therefore slow down the initiation of carcinogenesis or other cytotoxcity. Third, some phytoestrogens alter CYPs via binding to AhR directly, acting as either AhR agonists or antagonist. ²⁶

In summary, the present study utilized zebrafish embryos as a new model to evaluate the developmental toxicity of single or combined exposure to PAHs and phytoestrogens. Our results demonstrate that genistein and E2, when used at lower dose, can suppress the toxic effect of B(a)P and B(k)F on zebrafish embryos. The beneficial effect of genistein may be due to the inhibition of CYP enzymes (CYP1A1, CYP1A2 and CYP1B1) via directly interacting with AhR pathway, or disturbing the AhR pathway through interacting with ER pathway. However, further studies on elucidating the molecular mechanism and the cross talk between ER and AhR pathway will be needed.