The herbicide atrazine activates endocrine gene networks via non-steroidal NR5A nuclear receptors in fish and mammalian cells

Abstract

Atrazine (ATR) remains a widely used broadleaf herbicide in the United States despite the fact that this s-chlorotriazine has been linked to reproductive abnormalities in fish and amphibians. Here, using zebrafish we report that environmentally relevant ATR concentrations elevated zcyp19a1 expression encoding aromatase (2.2 microg/L), and increased the ratio of female to male fish (22 microg/L). ATR selectively increased zcyp19a1, a known gene target of the nuclear receptor SF-1 (NR5A1), whereas zcyp19a2, which is estrogen responsive, remained unchanged. Remarkably, in mammalian cells ATR functions in a cell-specific manner to upregulate SF-1 targets and other genes critical for steroid synthesis and reproduction, including Cyp19A1, StAR, Cyp11A1, hCG, FSTL3, LHss, INHalpha, alphaGSU, and 11ss-HSD2. Our data appear to eliminate the possibility that ATR directly affects SF-1 DNA- or ligand-binding. Instead, we suggest that the stimulatory effects of ATR on the NR5A receptor subfamily (SF-1, LRH-1, and zff1d) are likely mediated by receptor phosphorylation, amplification of cAMP and PI3K signaling, and possibly an increase in the cAMP-responsive cellular kinase SGK-1, which is known to be upregulated in infertile women. Taken together, we propose that this pervasive and persistent environmental chemical alters hormone networks via convergence of NR5A activity and cAMP signaling, to potentially disrupt normal endocrine development and function in lower and higher vertebrates.

Figure 1. ATR stimulates expression of gonadal zcyp19A1 encoding aromatase, but not zcypA2 in zebrafish.

Schematic of zCyp19a1 and zCyp19a2 zebrafish promoters with binding sites and start site indicated (arrow) as previously shown by . A. Relative expression of zCyp19a1 and zCyp19a2 transcripts determined by RT-qPCR in juvenile zebrafish (20 dpf) following exposure (72 hrs) to endocrine disruptors including 0.1 µM of 17βE2, 1 µM of genistein (Gen), 10 µM of 4-nonylphenol (4-NP), 10 µM of bisphenol A (Bis A) and 10 µM of atrazine (ATR) beginning at 17dpf. B. Relative expression of zCyp19a1 and zCypA2 transcripts in dissected 20 dpf zebrafish bodies and heads after 72 hrs ATR treatment (10 µM). C. Relative zCyp19a1 and zCyp19a2 transcript levels are shown with ATR stimulation at doses ranging from 0.01 to 10 µM for 72 hrs (left panel), or at different time points (hrs) with 10 µM ATR (right panel). For all panels error bars represent the S.E.M. obtained from analysis of three independent groups of fish (n = 5) using validated primers, with reactions carried out three times each. T-test analysis reveal statistical significance with **p<0.01, *p<0.05.

Figure 2. Chronic exposure to ATR increases the percentage of female zebrafish.

A. The percentage of female zebrafish is shown after chronic exposure to DMSO and increasing ATR concentrations (µM). The number of female fish was assessed visually at three, four (data not shown) and at six months, as shown. Visual inspection for sex-specific landmarks (body shape, body color, fin shape) at both stages suggested an increase in the ratio of female to male fish. Following visual inspection of six month old fish, unambiguous assignment of sex was determined by morphological inspection of gonads. ATR treatment began at 17dpf post-hatching. B. Representative pictures of fish treated with ATR and corresponding gonadal sections stained with hematoxylin and eosin. Oocytes within the perinuclear follicles of the ovary are indicated (black arrowhead), and spermatozoa in the testis are indicated (white arrowhead).

Figure 3. ATR activates NR5A receptors, but does not activate ERα.

A. Luciferase activity is shown after transfection of human placental JEG3 cells with ARO-Luc reporter (200 ng) and the parent reporter (pGL3-Luc) with mSF-1, hLRH-1 or zff1d expression vectors (25 ng). Drug treatments with EDCs at the dose indicated was for 24 hrs. B. JEG3 were transfected with the ERE-Luc (50 ng) and hERα (5 ng) and treated with 17βE2 (0.1 µM) with or without ATR (0.1–10 µM). Cells were treated with drug for 24 hr. C. Luciferase activities in JEG3 cells following treatment with EDCs, concentrations are indicated (ranging from 1 nM to 10 µM). All cells were transfected with mSF-1 and Aro-Luc with concentrations as described for panel A. T-test analysis reveal statistical significance with **p<0.01, *p<0.05.

Figure 4. ATR activates the MAPK, PI3K and modestly stimulates cAMP production.

A. Pharmacological inhibitors of MAPK (U0126), PI3K (LY294002) or Gi (PTX), and PLC (U73122) signaling were added, with ATR (10 µM) or without (DMSO), and with mSF-1 (25 ng), as indicated (+). All inhibitors were added 60 min prior to ATR treatment. B. Western blotting was performed using antibodies against phospho-ERK1/2, total ERK, phospho-SF-1, Flag, phospho-AKT and total AKT using JEG3 cellular extracts transfected with Flag-tagged mSF-1 (1 µg). Twenty-four hours after transfection, cells were starved for 3 hrs and treated with 10 µM of ATR for the indicated times (5 min to 24 hrs). C. Levels of total cellular cAMP (pmol/ml) were determined in lysed JEG3 cellular extracts according to Materials and Methods. Cells were treated for 30 min with increasing concentrations of drug (estradiol, ATR, or forskolin) as indicated (10 nM to 10 µM) following serum starvation for 24 hrs as described in Materials and Methods. T-test analysis reveal statistical significance with **p<0.01, *p<0.05.

Figure 5. ATR is cell-specific and induces a cluster of genes involved in hormonal responses.

A. Relative luciferase activities for the ARO-Luc reporter are shown in responsive (HepG2, human liver) and non-responsive (Ishikawa, uterus, HEK293, embryonic kidney) cell lines following transfection of mSF-1 (25 ng) with increasing doses of ATR. B. Relative levels of endogenous Cyp19A1 are shown for different cell lines after ATR treatment (10 µM, 24 hrs). Cell lines that showed a statistically significant increase in Cyp19A1 expression when compared to DMSO treatment indicated. C. List of endocrine/reproductive related genes with their relative rank order as defined by their fold enrichment following ATR treatment. Known SF-1 target genes and those that are cAMP responsive are indicated as (+). D. Relative expression levels of transcripts in JEG3 cells (without transfection of SF-1) are shown after DMSO (-) or treatment with ATR (1 to 10 µM). JEG3 cells were treated with the indicated doses of ATR for 24 hrs and RT-qPCR analysis was carried out using validated primers as indicated in Table S2. T-test analysis reveal statistical significance with **p<0.01, *p<0.05.

Figure 6. Overexpression of SF-1 enhances ATR effects, while knock-down of SF-1 and SGK-1 diminish or attenuate ATR effects on selective genes.

A. Relative endogenous transcript levels in JEG3 cells are shown with and without mSF-1 and with or without ATR treatment (24 hrs, 10 µM). B. Relative expression of transcripts in human JEG3 cells after transfection with si-RNAs directed to human SF-1 (si-SF-1, 50 nM) and human SGK-1 (si-Sgk1, 30 nM) in JEG3 cells, with DMSO (black bars) or with ATR treatment (gray bars, as described above). The fold induction with ATR treatment is indicated above bars. Levels of endogenous human SF-1 and human SGK-1 are shown after si-RNA treatment.