An In Vivo Bioassay for Detecting Antiandrogens Using Humanized Transgenic Mice Coexpressing the Tetracycline-Controlled Transactivator and Human CYP1B1 Gene

Abstract

The typical strategy used in analysis of antiandrogens involves the morphological changes of a marker in castrated rats Hershberger assay for the prostate, seminal vesicle, levator ani plus bulbocavernosus muscles (LABC), Cowper’s gland, and glans penis. However, there are disadvantages to this approach, such as the time required, and the results may not correspond to those in actual human exposure. To evaluate its ability for detecting antiandrogens, in vivo the dose effect of di-(2-ethylhexyl) phthalate (DEHP) and time effect of five antiandrogens, DEHP, di-n-butyl phthalate (DBP), diethyl phthalate (DEP), linuron (3-(4-dichlorophenyl)-methoxy-1-methylurea), and 2,4′-DDE (1,1-dichloro-2-(p-chlorophenyl)-2-(o-chlorophenyl)ethylene), were investigated using humanized transgenic mice coexpressing tetracycline-controlled transactivator (tTA) and the human cytochrome P450 (CYP) enzyme CYP1B1 (hCYP1B1). Adult transgenic males were treated with each of the five antiandrogens, and their tTA-driven hCYP1B1 expressions analyzed by real-time polymerase chain reaction (PCR) and/or Western blot and for O-debenzylation activity. Herein, the treatments of adult males with the five antiandrogens were shown to affect the increased levels of tTA-driven hCYP1B1 expression in both dose-dependent and repeated experiments. Thus, this novel in vivo bioassay, using humanized transgenic mice, is useful for measuring antiandrogens, and is a means to a more relevant bioas-say relating to actual human exposure.

Exposures to natural or synthetic androgens and antiandrogens are critical for the development of male phenotypes. Androgens and antiandrogens are involved in endocrine disruptors (EDs), which are known to produce adverse effects in humans and wildlife by mimicking or interfering with the action of endogenous hormones. The Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC) have completed a charter for evaluating compounds for their potential to act as agonists and antagonists to the estrogen or androgen receptor (ER and AR, respectively) and steroid biosynthesis inhibitors, or their ability to alter thyroid functions (EDSTAC 1998; Stoker et al. 2002).

Antiandrogens can also have considerable environmental and public health consequences. Di-(2-ethylhexyl) phthalate (DEHP), di-n-butyl phthalate (DBP), and diethyl phthalate (DEP) are well known antiandrogens. DEHP is converted to mono-(2-ethylhexyl) phthalate (MEHP) when administered orally, and then absorbed in the small intestine (Anbro et al. 1989), which disrupts both the prenatal development and postnatal maturation of male and female reproductive organs (Merkle, Klimisch, and Jackh 1988; Dostal et al. 1988). DBP appears to down-regulate the steroidogenic enzymes within the testis (Shultz et al. 2001). Linuron (3-(4-dichlorophenyl)-methoxy-1-methylurea) (Cook et al. 1993; Lambright et al. 2000), and 2,4-DDE (1,1-dichloro-2-(p-chlorophenyl)-2-(o-chlorophenyl)ethylene) (Kelce et al. 1995; O’Connor et al. 1999) have been identified as AR antagonists that alter the androgen status.

Castrated male rats, with blocked testosterone production, were utilized to monitor the possible androgenic activity of chemicals, which is named the Hershberger assay (Hersh-berger, Shipley, and Meyer 1953). In the Hershberger assay, the androgen-responsive tissue of the male reproductive tract in surgically castrated male rat is allowed to regress over a period of about 10 days, stimulating androgen. Therefore, the effects on the weights of the sex accessory tissues can be used to test for androgens. This Hershberger assay is also used for detecting the antiandrogens, using the modified testosterone propionate-stimulated weanling male rat assay (Ashby et al. 2004). Simpler chemicals, such as flutamide and 2,4-DDE, substitute for this surgery, producing antiandrogen effects. However, these typical strategies are time consuming and have very little in common with actual human exposure.

Humanized transgenic mice expressing human genes for the cytochrome P450 (CYP) enzymes provide a novel strategy for the assessment of xenobiotics, including androgens and antiandrogens. Therefore, humanized double-transgenic mice have previously been generated that coexpress the tetracycline-controlled transactivator (tTA) and human CYP1B1 (hCYP1B1) (Hwang et al. 2001a). The tTA-driven hCYP1B1 was expressed at high levels in neonates and at low levels in adults (Hwang et al. 2003). Upon castration, the patterns of tTA-driven hCYP1B1 expression were converted to those of female mice, and treatment of castrated mice with testosterone resulted in restoration to adult male patterns. Also, the tTA-driven hCYP1B1 in adult males was increased by treatment with flutamide, a representative antiandrogen (Hwang et al. 2003). These results provide us with an opportunity for evaluating the ability to detect androgen and antiandrogens, in vivo, using humanized transgenic mice.

In this study, an in vivo bioassay, using humanized transgenic mice, coexpressing tTA-driven human CYP1B1, was applied for evaluating the activities of five antiandrogens by real-time polymerase chain reaction (real-time PCR) and/or Western blot analysis and measurement of the benzyloxy resorufin O-debensylase activity. Five antiandrogens, DEHP, DBP, DEP, linuron, and 2,4-DDE, were examined. The results showed that treatment of adult males with the five antiandrogens affected the levels of tTA-driven hCYP1B1 expression at the transcript and/or proteins and activity levels.

MATERIALS AND METHODS

Chemicals

Linuron was purchased from Sigma (St. Louis, MO, USA). The DEHP, DBP, and DEP were obtained from Aldrich Chemicals (Miluwaukee, WI, USA), and 2,4-DDE from Wako (Osaka, Japan).

Humanized Transgenic Mice

By mating the first lineage of the Tet-tTA with the second lineage of the Tet-hCYP1B1 transgenic lines, the double-transgenic, single-transgenic, and nontransgenic littermate offspring were segregated (Hwang et al. 2001a). The double trans-genic mice were then crossed with age-matched control mice for subsequent quantitative production. All the transgenic mice transmitted the introduced Tet/tTA and Tet/hCYP1B1 genes into the genomes of about 50% of the hemizygotes, in a Mendelian fashion (Hwang et al. 2001a). All the transgenic and nontransgenic mice used in these experiments were handled in an accredited Korea Food and Drug Administration (KFDA) animal facility, in accordance with the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International Animal Care Policies (Accredited Unit-Korea Food and Drug Administration: Unit Number 000936), and maintained in a specified pathogen-free state. All the mice were housed in cages under a strict light cycle (lights on at 08:00 h and off at 20:00 h), and given a standard irradiated chow diet (Purina Mills) ad libitum.

Treatments

Ten-week-old double-transgenic males (5 mice/group) were used in this experiment. Here, nontransgenic control mice were not included, because the endogenous mouse CYP1B1 gene is almost undetectable, or only at low levels, in their livers (Hwang et al. 2001a). For dose-dependent experiment, transgenic adult males were given a single subcutaneous injection of DEHP at dose level 0.1 g/kg, 0.5 g/kg, or 1 g/kg. In addition, the control mice were injected with just the corn oil vehicle. The five antiandrogens (DEHP [1 g/kg], DBP [1 g/kg], DEP [1 g/kg], linuron [100 mg/kg], and 2,4-DDE [100 mg/kg]) were consecutively injected subcutaneous into the mice in 0.2 ml of corn oil.

RNA Isolation and cDNA Amplification

Total RNA was isolated from the livers using RNAzol (Tel-Test, CS104), according to the manufacturer’s instructions. The RNA pellets were suspended in diethylpyrocarbonate (DEPC)-treated distilled water and stored at −70°C until subsequent analyses. The quantity and quality of the RNA were checked by the ratio of the absorbance at 260 nm to that at 280 nm. The reverse transcriptase–polymerase chain reaction (RT-PCR) was performed using 5 μg of total RNA from the tissue of each treatment group, with 0.5 ng of oligo-dT primer (Gibco BRL, 18064-012) and annealed at 42°C for 50 min. Complementary DNA, serving as template, for further amplification was synthesized by addition of dATP, dCTP, dGTP, dTTP, and 200 unit of reverse transcriptase (Gibco BRL, 18064-014). Thereafter, 3.2 units of RNase H were added to remove the RNA hybridized with the cDNA, and incubated for 30 min at 37°C.

PCR Analysis

The hCYP1B1 was synthesized using the sense (5^′-GGTCT CCCCG TCTGT GCCTT CTCA-3^′) and antisense (5^′-TTGCA TGGTG CTGGT GCGCA GACC-3) primers, with a sequence complementary to the hCYP1B1, ranging from 1310 to 1900 nucleotides, as the DNA template. Because these primer nucleosides were not homologized with a mouse CYP1B1 primer nucleotide, they were not able to detect the endogenous mouse CYP1B1 at the RNA level. In addition, the tTA-specific primers (5^′-CAGAA GCTAG GTGTA GAGCA G-3^′ [nucleotides + 611 to + 631] and 5^′-GCTCT GCACC TTGGT GATCA-3^′ (nucleotides + 1048 to + 1067)) were selected for their amplification by PCR from the coding region of the tTA gene. RT-PCR, using primers specific for ß-actin, was also performed to ensure the integrity of the RNA. The PCR primers used for the mouse CYP1B1 were as follows: sense primer, 5^′-CTTAG TGCAG ACAGT CCACA G-3^′ and antisense primer, 5^′-GAAAG CACGC ATCGT GCTAT AG-3^′, corresponding to the 192 to 212 and 622 to 644 nucleotides of the mouse CYP1B1 gene, respectively. The sequences of the ß-actin sense and antisense primers were 5^′-TGGAA TCCTG TGGCA TCCAT GAAAC-3^′ and 5^′-TAAAA CGCAG CTCAG TAACA GTCCG-3^′, respectively. A quantitative analysis of the RNA was performed from the results of the RT-PCR.

Real-Time PCR Analysis

Real-time PCR analysis was performed using TaqMan Universal PCR Master Mix (AB4034437) on an ABI PRISM Sequence Detection System (ABI-SDS) (Applied Biosystems). Following reverse transcription, the PCR was carried out, in triplicate, for each mixture with synthesized cDNA added to the Taq-Man Universal Master Mix and CYP AdB product (mixture of sense, antisense, and FAM primer). PCR for glyceraldehyde-3-phosphate dehydrogenase (GAPDH), as an endogenous control, was simultaneously performed, with the exception of that for the rodent GAPDH control reagent (AB4308313). TaqMan probes have two fluorescent dyes, one at the 5^′-terminal (reporter, R) for displacing the strand, with the other at the 3^′-terminal (quencher, Q) for blocking the extension. The reaction mixture was then subjected to amplification, with the following sequence: 1 cycle at 50^°C for 2 min (annealing stage), 1 cycle at 95^°C for 10 min (denaturation stage), and 40 cycles at 95^°C for 15 s and at 60^°C for 2 min (extension stage). In all cases, an initial denaturation was carried out at 95^°C for 5 min to decrease the primer-dimer formation under these PCR conditions. The ABI-SDS was programmed to take R and Q fluorescent dye readings after each cycle at a temperature several degrees (60^°C) lower than the melting temperature of the target amplicon. This step, at 60^°C, was taken to avoid, or minimize any potential contribution to the overall fluorescent dye signals due to primer-dimer formation. A calibration curve was constructed by plotting the R/Q ratio against the amounts of hCYP1B1 cDNA synthesized, on the basis of the RNA isolated from the livers of the transgenic mice treated with the androgens, with antiandrogens, and those of the nontransgenic mice were used as the control.

Microsomal Protein Preparation and Western Blot Analysis

The microsomal proteins were prepared as previously described (Hwang et al. 2001a; Kupfer and Levine 1972). Briefly, liver tissue was homogenized in ice-cold 0.25 M sucrose solution, and the homogenate centrifuged at 10,000 × g and 4^°C for 15 min, followed by 15,000 × g for 15 min at 4°C. The resultant microsomal pellet suspended in 0.25 M sucrose solution. The protein content was determined by the bicinchroninic acid (BCA) method with a bovine serum albumin (BSA) standard using a BCA Protein Assay Kit (PIERLE, Rockford, IL). The microsomal proteins (50 μg) were separated on a 10% polyacrylamide gel for 3 h, and then transferred to a nitrocellulose membrane, using an electroblot at 40 V for 4 h. A solution of 5% powdered nonfat milk, 25 mM Tris (pH 7.5) and 150 mM NaCl was used to block the nonspecific binding sites for 1 h at room temperature. The membranes were incubated separately with primary anti-human CYP1B1 (BD Bioscience, San Jose, CA), at a 1:1000 dilution, in the blocking buffer for 3 h at room temperature, and then washed in washing buffer, containing 10 mM Tris (pH 8.0), 150 mM NaCl, and 0.05% Tween-20. This was followed by incubation with a secondary antibody, horseradish peroxidase–conjugated goat anti-rabbit immunoglobulin G (IgG) (BD Bioscience, San Jose, CA), at a 1:1000 dilution for 1 h at room temperature. The human CYP1B1-specific proteins were detected by an enhanced chemiluminescent substrate (Amersham Bioscience, RPN2132). The anti-human CYP1B1 was able to specifically detect the hCYP1B1 from Western blot analyses when used as recommended by the manufacturer. The protein levels were quantified using an Imazing Densitometer Model GS-690 (Hercules, CA).

Benzyloxyresorufin O-Dealkylase Activity

The benzyloxyresorufin O-debenzylase activity (BROD) activity was determined by measuring the dealkylation of benzyloxyresorufin O-debenzylase (Prough, Burke, and Mayer 1978), using 0.5 mM benzyloxyresorufin as the substrate. In the metabolism of the CYP enzyme substrate, BROD is relatively specific to hCYP1B1 (Hwang et al. 2001a, 2003), although it is also a substrate for CYP2B (Nerurkar et al. 1993). The resorufin fluorescence was recorded with excitation and emission wavelengths of 532 and 586 nm, respectively. The fluorescence was corrected using a blank (incubation mixture) prior to interpretation, and then compare with a resorufin calibration curve. The rate of resorufin formation was defined as the CYP1B1 activity, and expressed in pmol/min/mg protein.

Statistical Analysis

Tests for significance were performed using a one-way analysis of variance (ANOVA) (SPSS for Windows, Release 10.01, Standard Version; Chicago, IL). All values are reported as the standard deviation for n =5 from duplicate experiments. Statistical significance was set at p < .05.

RESULTS

Humanized Double-Transgenic Mice

The humanized double-transgenic mice expressing the tTA and hCYP1B1 were previously produced in our laboratory (Hwang et al. 2001a). In the livers of this double transgenic line, the minimum base activity of the hCMV promoter resulted in very low levels of the tTA protein in the absence of doxycycline. However, the tTA protein produced was blocked from binding to the tet promoter in the presence of doxycycline (Figure 1A ), and the expression of the hCYP1B1 gene were consequently maintained at low levels. When the doxycycline was removed, small amounts of the tTA proteins were bound to the tetracycline-regulated promoter, stimulating expression of the tTA gene. Higher levels of the tTA protein then stimulated higher levels of hCYP1B1 expression. The tetracycline-regulated promoter sequence consists of seven copies of the tetracycline operator sequence upstream of the minimal hCMV promoter region, which contains the TATA box and transcription start site (Figure 1B ). It was shown that endogenous levels of mouse CYP1B1 was not presented in the nontransgenic brains, whereas exogenous levels of human CYP1B1 were highly expressed in the doubly transgenic mice upon doxycycline removals (Hwang et al. 2001a) (Figure 1C ).

Effect of the DEHP on hCYP1B1 Expression in Transgenic Adult Males

Transgenic adult males were treated with different doses of DEHP (0.1 g/kg, 0.5 g/kg, and 1.0 g/kg) and with corn oil (control). These doses were under the range of dose (2 g/kg) that had been used for the study of testicular toxicity in Sprague-Dawley rats (Park, Habeenbu, and Klassen 2002). And then 3 day later, the RNA and microsomal proteins were prepared for measurements of the transcript, protein, and activity levels. Gradual increases in transcript, protein, and activity were observed after DEHP treatment, but treatment of transgenic adult male with corn oil resulted in no changes in their measurements (Figure 2A, B , and C).

Effect of the Five Antiandrogens on hCYP1B1 Expression in Transgenic Adult Males

Flutamide, a representative antiandrogen, led to an increase in the expression of the hCYP1B1 transgene in the livers of adult males (Hwang et al. 2003), and this effect was almost identical to that due to castration (Hwang et al. 2003). To test the three antiandrogens, DEHP, DBP, and DEP, adult males were injected with each antiandrogen on days 1, 3, and 9, and the RNA and microsomal proteins prepared for measurement of the transcript, protein, and activity levels of the tTA-driven hCYP1B1. The tTA-driven hCYP1B1 expressions began to increase with increasing amounts of DEHP, DBP, and DEP treatments, at the transcript, protein, and activity levels (Figure 3A, B, and C ). The treatments with linuron and 2,4-DDE also caused increases in the tTA-driven hCYP1B1 at the transcript level (Figure 4A and B ).

DISCUSSION

The five antiandrogens used in this experiment were mostly androgen receptor (AR) antagonists, which serve as key regulators for development of male phenotypes, such as external genitalis and urethra. DEHP, and its metabolite MEHP, which are AR antagonists, activate the peroxisome proliferator-activated receptors, PPARα and PPARγ, leading to the induction of CYP1B1 expression (Lovecamp-Swan, Jetten, and Davis 2003). The CYPs are a superfamily of hemoproteins that catalyze the mono-oxygenation of a wide range of endocrine disruptors, including endobiotics, such as androgens, antiandrogens, and xenobiotic substrates (Spink et al. 1990; Otto et al. 1991; Walker et al. 1995). Of the members of the CYP superfamily, CYP1B1 is constitutively expressed in the adrenal gland, ovaries, and testis, and can be induced by polycyclic aromatic hydrocarbons (PAHs), adrenocorticotropins, and peptide hormones (Otto et al. 1991; Otto, Bhataacharrya, and Jefcoate 1992). Thus, hCYP1B1 is vital for an effective and relevant in vivo screening strategy for antiandrogens.

An in vivo transgenic mouse model is a novel approach, and should ultimately have an advanced benefit to public health. An increase in the tTA-driven hCYP1B1 expression is found in the livers of transgenic adult males treated with antiandrogens, and these changes are increased by repeated treatments and different doses of DEHP treatment. In addition, the doubly transgenic line used in this experiment transmitted the introduced Tet/tTA and Tet/CYP1B1 into the genomes of about 50% of the hemizygotes, in a Mendelian fashion. Thus, it is neither the manner of impact solely by its location in the genome or the loci in a gene that is impacted by gestational or neonatal imprinting. In general, the offspring of a litter show the transgenes have been inserted with a constant copy number, as in the case of transgenic mice expressing the lacZ gene under the control of a neuron-specific enolase promoter (Forss-Petter et al. 1990).

There are two utilities of these transgenic bioassays. Firstly, adult transgenic males could be useful in the testing of antiandrogens. The typical strategy employed in the analysis of androgen and antiandrogen actions involves the morphological changes of a marker in the castrated rat Hershberger assay for the prostate, but there are some disadvantages to this approach, such as the time required. Secondly, determination of the toxico- and pharmacokinetics of antiandrogens are possible using this in vivo transgenic bioassay, because the action of antiandrogens is almost impossible to measure in an in vitro bioassay system. Indeed, liver microsomal CYP enzymes are crucial in studying the determination of the intensities and durations, the detoxification and activation, the tumorigenic metabolites, and the biosynthesis or catabolism of the chemical actions of androgens, antiandrogens, and toxic chemicals. Thirdly, the assessment of antiandrogens and the detoxification processes did not correspond to those in actual humans, because these rat were not similar to humans in relation to their metabolisms of antiandrogens. Here, the hCYP1B1 gene was chosen because the endogenous mouse CYP1B1 gene expression is almost undetectable, or only at low levels in the livers of normal mice, but expressed in the livers of doubly transgenic mice (Hwang et al. 2001b).

The five antiandrogens used in this experiment are AR antagonists, and have been identified as compounds that alter the androgen status (O’Connor, Frame, and Ladics 2002). Here, the tTA-driven hCYP1B1 expressions in the livers of adult transgenic mice were increased by the antiandrogens treatments. This result has raised two possibilities: (i) antiandrogens bind to the tTA-binding site on the tet-promoter in the absence of doxycycline, and thus induce the expression of human CYP1B1, or (ii) an as yet unidentified factor, regulated by antiandrogens, is induced in the livers of adult transgenic mice, which binds to the tTA binding site on the tet-promoter sequence, and thus activates hCYP1B1 transcription. In either case, the tet-promoter sequence might play an important role in an enhanced expression of the tTA-driven hCYP1B1, and this presents a unique chance to analyze the molecular mechanism of anti-androgenic actions. The tet promoter consists of seven copies of the tetO sequence upstream of the minimal hCMV promoter region, and contains a TATA box and the transcription start site (Figure 1B ). The application of humanized transgenic mice, expressing the tTA-controlled hCYP1B1, should lead to a much greater understanding of the potential risks associated with exposure to specific antiandrogen components contained in foods and herbal medicines, and will provide strategies for their assessment. Here, the human CYP1B1 was chosen because endogenous mouse CYP1B1 is almost undetectable in the liver, and is not expressed in the endogenous mouse CYP1B1 in the livers of normal mice (Hwang et al. 2001a). Thus, humanized transgenic mice, expressing the tTA-controlled hCYP1B1, can be used as the basis for a proposed strategy for the detection of antiandrogens, and these mice will stimulate the further development of a predictive model for the identification and assessment of potential antiandrogens.