Development of Two Androgen Receptor Assays Using Adenoviral Transduction of MMTV-Luc Reporter and/or hAR for Endocrine Screening

P. C. Hartig,1, K. L. Bobseine, B. H. Britt, M. C. Cardon, C. R. Lambright, V. S. Wilson and L. E. Gray, Jr.

Reproductive Toxicology Division, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711

Received August 17, 2001; accepted November 19, 2001


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The discovery of xenobiotics that interfere with androgen activity has highlighted the need to assess chemicals for their ability to modulate dihydrotestosterone (DHT)-receptor binding. Previous test systems have used cells transfected with plasmid containing a reporter gene. Here we report the use of transduction for gene delivery and assessment of the modulation of DHT-induced gene activation. Transduction, the ability of replication-defective viruses to deliver biologically competent genes, is a well understood biological process, which has been utilized to repair defective genes in humans as well as to express exogenous genes in rodent models. Human breast carcinoma cells (MDA-MB-453) containing endogenous copies of the androgen (hAR) and glucocorticoid (GR) receptors were transduced with replication-defective human adenovirus type 5 containing the luciferase (Luc) reporter gene driven by the AR- and GR-responsive glucocorticoid-inducible hormone response element found with the mammary tumor virus LTR (Ad/MLUC7). In a second set of experiments, CV-1 cells were transduced as above with MMTV-luc and also hAR. Cells were subcultured in 96-well plates, transduced with virus, exposed to chemicals, incubated for 48 h, lysed, and assayed for luciferase. Luc gene expression was induced in a dose-dependent manner by DHT, estradiol, and dexamethasone (MDA only) and inhibited by AR antagonist hydroxyflutamide (OHF), hydroxy-DDE, HPTE (2,2-bis(p-hydroxyphenyl)-1,1, 1-trichloroethane), a methoxychlor metabolite, and M1 and M2 (vinclozolin metabolites). The transduced cells responded to AR agonists and antagonists as predicted from our other studies, with a very robust and reproducible response. Over all replicates, 0.1 nM DHT induced luc expression by about 45-fold in CV-1 cells (intra-assay CV = 20%) and 1micromolar OHF inhibited DHT by about 80%. In the transduced MDA cells, 0.1 nM DHT induced luc by about 24-fold (intra-assay CV = 33%), which was inhibited by OHF by about 85%. DHT-induced luciferase activity peaked in both cell lines between 1 and 100 nM, displaying about 64- and 115-fold maximal induction in the CV-1 and MDA 453 cells, respectively. For agonists, a two-fold induction of luc over media control was statistically significant. For AR antagonists, a 25–30% inhibition of DHT-induced luc expression was typically statistically significant. Comparing the two assays, the transduced CV-1 cells were slightly more sensitive to AR-mediated responses, but the transduced MDA 453 cells were more responsive to GR agonists. In summary, these assays correctly identified the endocrine activity of all chemicals examined and displayed sensitivity with a relatively low variability and a high-fold induction over background. Adenovirus transduction for EDC screening has the potential to be employed in a high-throughput mode, and could easily be applied to other cell lines and utilized to deliver other receptors and reporter genes.

Key Words: adenovirus; transduction; screening; assay; androgen receptor; human; MDA, CV-1.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
In 1996, the Safe Drinking Water Act and the Food Quality Protection Act required that the U.S. Environmental Protection Agency (EPA) develop screening and testing programs for endocrine-disrupting chemicals (EDCs). To address these issues, the EPA formed a multi-stakeholder Endocrine Disrupter Screening and Testing Advisory Committee (EDSTAC) to make recommendations on how to address this legislative mandate. In 1998, the EDSTAC completed a final report recommending a 2-tiered screening and testing program designed to identify and characterize chemicals that had anti-estrogenic, anti-androgenic and anti-thyroidal activities. The recommended EDSTAC Tier-1 screening program includes in vivo and in vitro assays to detect androgenic and anti-androgenic activity. EDSTAC recommended that an androgen receptor (AR)-binding assay and/or an AR-dependent gene expression assay that examined AR function in vitro, be used to detect chemicals that acted as AR agonists or antagonists. In the final report, EDSTAC provided a protocol for AR binding using receptors isolated one day after castration from the ventral prostate of the young adult male rat. The screening and testing work group of EDSTAC elected to present this assay as an example of how such in vitro screening might be done, because it has had extensive use over several decades and would most likely be the easiest to standardize and validate for interlaboratory use, as mandated by FQPA. However, EDSTAC acknowledged that this assay had several limitations, and newer assays on the horizon might eventually be preferable, since new methods for the assessment of EDC action is an exciting and rapidly developing field. For example, the proposed AR-binding assay requires the use of animals as a source of AR, uses radioactivity, is not amenable to high-throughput, suffers from poor yield of intact receptor, and does not discriminate AR agonists from antagonists. In this regard, our research program has developed, or is developing, in vitro assays to address mechanisms of action of EDCs on reproductive development that will eliminate some or all of the above limitations. Hence, upon further standardization and validation, some of these newer in vitro AR assays may be appropriate for inclusion in the agency's Tier-1 screening battery as a replacement for the described AR binding assay.

In this endeavor we have developed several in vitro approaches to screen for AR activity using gene expression assays. Historically, this type of in vitro assay involves the transient transfection of a tester cell line with a plasmid base receptor and the reporter, followed by chemical exposure and measurement of the modulation of gene expression. These types of assays require relatively large numbers of cells and costly transfection material, with each transfection event introducing yet another source of experimental variation. The KB2 assay, a significant improvement on this type of assay system, was presented recently (Wilson et al., 2002Go). In the aforementioned assay, we describe the effects of androgens and antiandrogens in a stably transformed cell line that we developed (MDA-KB2), which expresses the human AR (hAR) and an AR-responsive promoter linked to a luciferase reporter gene (MMTV-luc). The main advantage of the KB2 assay is that it employs a genetically modified cell line, which eliminates the effort and inherent variability associated with repeated transient transfections. The approach taken in this study was to attempt to deliver the genes via replication-defective adenovirus (transduction). Viral transduction is a precise and reproducible way of delivering genes in a cost-effective manner that should increase the robustness of the response to androgens and antiandrogens as compared to the transformed MDA-KB2 cell line and should reduce the variability as compared to transient transfected AR assays.

Herein, we describe the responses of several androgens and antiandrogens in two AR-responsive in vitro assays. These assays used transduced MDA-453 and CV-1 cells, cell lines employed extensively in our laboratory. The chemicals and concentrations used in the current study are some that we have extensive experience with and that were utilized in the phenotypically transformed, AR-responsive KB2 cells (Wilson, et al., 2002Go). In this study we have assessed the ability of an adenovirus to transduce the human breast cancer line MDA-MB-453 (AR+, GR+, PR-, ERa-, weak ERb+) with a luciferase gene regulated by the androgen- and glucocorticoid-inducible hormone response element found in the mouse mammary tumor virus (MMTV) LTR. Utilization of MDA cells, which already contained an endogenous AR, simplified the development of this assay by reducing the transduction requirements to a single reporter gene. In addition, we also assessed the ability of an adenovirus to transduce the CV-1 (African green monkey kidney cells: AR-, GR-, ER-, PR-) (Coutte et al., 1994Go; Fenton et al., 1997Go; Vamvakopoulous et al., 1993) with both the MMTV-luc reporter and the human androgen receptor (hAR) genes. The responses of each assay to AR agonists, AR antagonists, and GR agonists are described (Table 1Go).


View this table:
[in this window]
[in a new window]
 
TABLE 1 Expected Effects of the Chemicals Used in the Current Study on Gene Expression in Our Transduced CV-1 and MDA-453 Cell Lines
 

    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Virus.
Ad/mLuc7, which contains the luciferase gene regulated by the glucocorticoid-inducible hormone response element found in the mouse mammary tumor virus (MTV) LTR (Shih et al., 1991Go), was a gift from Cary Weinberger (NIEHS). This virus had a necessary early gene (adenoviral gene E1) replaced with the reporter gene. In addition, the virus is replication-deficient and can only be propagated in a complementary cell line such as the transformed human embryonal kidney cell line 293, which contains a copy of the E1 gene. Plasmid pCMVhAR, containing an expression cassette of the human androgen receptor gene under the control of the CMV promoter (Lubahn et al., 1988Go), was a gift of Dr. E. Wilson. p{Delta}E1sp1A and pBHG 11were purchased as part of a virus construction kit from Microbix Biosystems, Inc. (Ontario, Canada).

Ad5hAR was generated by subcloning the hAR expression cassette of pCMVhAR into pE1sp1A (see Fig. 1Go). pCMVhAR was digested with BrsB1 and Fsp1. Digestion liberated a blunt-ended, 5124-bp fragment (bp 22 to 5145) containing the hAR cDNA flanked by promoter and polyadenylation signals. The fragment was blunt-end cloned into the Eco RV site of pE1sp1A in an E1-antiparallel orientation yielding plasmid p{Delta}E1sp1AhAR. pBHG 11(7 µg) and p{Delta}E1sp1AhAR (9 µg) were mixed and applied to a 35-mm dish containing a monolayer of 293 using the reduced-CO2 transfection method (Sambrook, et al., 1989Go). After overnight transfection, the cells were washed twice in medium, then incubated in 5% CO2 in medium containing reduced (5%) serum, and fed when medium became acidic. After 6 weeks, some cultures presented with cytopathic effects consistent with adenovirus infection. Fluids from infected cultures were saved and subjected to plaque assays for purification. Isolated clones were collected, replaqued twice, and the viral DNA was isolated, and the insert identity and orientation confirmed by restriction analysis (Sambrook et al., 1989Go).



View larger version (33K):
[in this window]
[in a new window]
 
FIG. 1. Schematic diagram of the generation of Ad5 hAR. The human AR gene was removed from plasmid pCMVhAR and moved into plasmid p{Delta}E1spIA, which contains Ad5 sequences flanking the insertion site. The product p{Delta}E1sp1AhAR contained the pCMV promoter, hAR and Ad5 flanking sequences. This plasmid was combined with pBHG 11, which contains the Ad5 genome devoid of the E1 gene and packaging sequences. p{Delta}E1sp1AhAR and pBGH 11 were cotransfected into 298 cells, which contain copies of the E1 gene. Spontaneous homologous recombination between Ad5 sequences yield Ad5 hAR. Drawing is not to scale.

 
Cell and viral culture.
Low passage level 293 cells were purchased from Microbix Biosystems Inc. The human breast carcinoma cell line MDA-MB-453 [(MDA) cat # ATCC HTB 131] and African green monkey cells [(CV-1) cat.# ATCC CCL 70] were obtained from the American Type Culture Collection (Rockville, MD). Serum was purchased from Hyclone (Logan, UT). All other tissue culture and molecular biology reagents were purchased from GIBCO BRL (Grand Island, NY). Monolayer 293 cell cultures were maintained at 37°C, 100% humidity, and in 5% CO2 in growth medium (MEM [Cat.# 11435–039] supplemented with amphotericin B [2.5 µg/ml], penicillin, and streptomycin [100 U and 100 µg/ml, respectively], glucose [3.5 mg/ml], glutamine [300 µg/ml], HEPES pH 7.3 [2.4 mg/ml], NaHCO3 [2.2 mg/ml] and 10% heat inactivated [56°C, 30 min] fetal bovine serum [FBS]). Cells were suspended with citric saline (KCL 10 mg/ml, Na3C6H5O7 * 2H2O, 4.4mg/ml) and subcultured 1:3 when 80% confluent. MDA cells were maintained at 37°C, 100% humidity with no supplemental CO2 in Leibovitz's L-15 medium. CV-1 cells were maintained in DMEM with 5% CO2. Both were supplemented with 100 U penicillin, 100 µg/ml streptomycin and 0.25 µg/ml amphotericin B and 10% FBS (not heat-inactivated). When confluent, MDA and CV-1 cells were suspended by standard trypsinization techniques and subcultured 1: 4.

Initial viral stocks contained about 1 x 109 plaque-forming units (pfu)/ml. Virus was propagated by inoculating 150-mm dishes containing 293 cells which were 80% confluent. Medium was removed, inocula instilled, and dishes rocked every 15 min for 1 h. Then 20 ml of fresh media was added. The cells were incubated 2–3 days until 95% of the cells exhibited cytopathic effect. Progeny virus remains associated with cell debris. Cells were scraped off the dish and the cells and media centrifuged at 700 x g for 5 min. The cell pellet was suspended in 1/20th of its original volume in spent media containing 10% glycerol, suspended, and subjected to 3 cycles of freeze-thawing to release the virions from the cell debris. Virus was dispensed in 50-µl aliquots and frozen. Virus was assayed on 293 cells by standard plaque assay (Graham and Prevec, 1991Go). Briefly, 35-mm cluster dishes were seeded with 293 cells at least 24 h prior to the assay. When monolayers were 80% confluent, medium was removed and 200 µl of serial 10-fold dilutions of the virus added. Dishes were rocked every 15 min for 1 h. Inoculum was removed and agarose overlay added. Dishes were placed at 4°C for 10 min., then incubated for 6 days at 37°C at 100% humidity in 5% CO2. Overlay was prepared by making 2x growth medium, warming it to 37°C, and mixing it with 37°C, 2% aqueous SeaPlaque low gelling temp agarose (FMC Corp., Rockland, ME), mixing, incubating at 37°C for 10 min, and adding 3 ml overlay/well. After incubating for 6 days, 2 ml additional overlay containing 50 µg/ml neutral red was added and dishes incubated overnight. Plaques appeared as light areas on a red background. Values were reported as plaque-forming units (pfu)/ml. Each value was the mean of at least 2 dishes.

Transduction assay.
Transduction assays were performed in 96-well plates. Twenty-four h prior to transduction, 5 x 104 cells were plated per well. Medium was removed and replaced with 20 µl of control medium or medium with diluted virus. MDA cells (which contain endogenous AR) were transduced with Ad/mLuc7 reporter virus at a multiplicity of infection (MOI) of 50 (i.e., 50 virions per cell). CV-1 cells (which lack native AR) were transduced with Ad/mLuc7 at a MOI of 50 and Ad5hAR at a MOI of one. Dishes were rocked every 15 min for 1 h, incubated 3 additional h, then 200 µl of medium or medium and test chemicals were added to each well, followed by a 48-h incubation. Plates were washed 2x with PBS, pH 7.4, decanted, and 25 µl cell culture lysis reagent (25 mM Tris-phosphate, pH 7.8, 2 mM DTT, 2 mM 1,2-diaminocyclohexane-N,N,N`,N` tetra-acetic acid, 10% glycerol, 1% Triton X-100 (Promega Corp., Madison WI)] added, and incubated 30 min., or until cells lysed. Plates were either frozen at –80°C, or assayed immediately for luciferase activity. Each well received 25 µl reaction buffer (25 mM glycylglycine, 15 mM MgCl2, 5 mM ATP, 0.1 mg/ml BSA, pH 7.8), followed by 25 µl 1-mM D-luciferin 5 s later. Luciferase activity was quantitated in an MLX microtiter plate luminometer (Dynex Tech, Chantilly, VA) and data expressed in relative light units (RLU).

Chemicals.
All chemicals were purchased from Sigma (purity >99%; St. Louis, MO) unless stated otherwise. The antiestrogen, ICI 182780 (ICI) was supplied by ICI Pharmaceuticals (Macclesfield, England). Hydroxyflutamide (OHF) was provided by R.O. Neri at Schering Corp. (Bloomfield, NJ). (4-[2,2-Dichloro-1-(4-hydroxyphenyl)vinyl]phenol), (OH-DDE: lot c1f03065, purity = 100%) was purchased from SPECS and BioSPECS B.V. (Rijswijk, Netherlands). Vinclozolin metabolite, M2, was obtained from BASF Ag and metabolite, M1, was synthesized from vinclozolin and purified as previously described (Kelce et al., 1994Go). Synthesis of the methoxychlor metabolite 2,2-bis(p-hydroxyphenyl)-1,1, 1-trichloroethane (HPTE), was previously described (Waller et al., 1996Go). The chemicals and dosage levels selected for these 2 assays were chosen because they produce agonist and/or antagonist responses in the MDA-KB2 and CV-1 cell lines in our laboratory (Wilson et al., 2002Go).

Data collection and analysis.
The data were collected from several independent experiments, with 3–4 replicates/plates per experiment. For each cell line, the individual experiments were (1) 5 {alpha}-dihydrotestosterone (DHT) dose response; (2) medroxyprogesterone acetate (MPA) dose response; (3) 17-ß-estradiol (E2) dose response with and without the antiandrogen hydroxyflutamide (OHF) or the antiestrogen ICI; (4) dose response with dexamethasone, a synthetic corticosteroid (DEX); (5) different doses of the vinclozolin metabolites M1 and M2, with and without 0.1 nM DHT; (6) dose response with OH-DDE with and without 0.1 nM DHT; and (7) dose-response effects of HPTE, the estrogenic and antiandrogenic metabolite of the insecticide methoxychlor (Gaido et al., 2000Go) with and without 0.1 nM DHT. A replicate was a 96-well plate, which included 4–8 independent observations of the media control (plus Et-OH, the dosing solution) and all other treatment groups. Hence, the design is a randomized, complete block design (the term block being equivalent to a plate, referred to herein as a replicate).

Data were analyzed by two-way ANOVA using PROC GLM available with SAS version 6.08 on the U.S. EPA`s IBM mainframe. Relative light units (RLU), fold, and log10 fold data were analyzed in a GLM model, which included the concentrations and replicates (most chemicals being run in 3 replicate assays). Using "replicates" as a blocking factor in the analysis has the effect of "normalizng" the data for overall differences from plate to plate, on average.

Luciferase response.
Statistically significant effects (p < 0.01, F statistic) were examined using the LSMEANS procedure (t-test). Means and standard errors (SE) were calculated using PROC means. In this regard, the SE in the tables are not corrected for replicate variation. For androgen agonists, which stimulate luciferase expression, treatments were compared to the media/ethanol control group, while androgen antagonists, which block DHT-induced luciferase expression, were compared to the 0.1 nM DHT group. Relative light units were converted to fold induction above the media value for each replicate, which in turn were log10 transformed (to correct for heterogeneity of variance, the SD being proportional to the means) for statistical analysis.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
CV-1 cells transduced with hAR and MMTV-luc.
DHT, MPA (Table 2Go), and E2 (Table 3Go) displayed AR agonist activity. E2-induced expression was blocked with OHF, but not the antiestrogen ICI, indicating that these were AR-mediated responses (Fig. 2Go). M2, M1, OH-DDE, and HPTE attenuated the effects of 0.1 nM DHT (Fig. 3Go, Table 4Go). In general, a 25–30% reduction in DHT-induced luc activity was statistically significant, an effect that was achieved with the above chemicals at concentrations of 0.05–0.2 µM. Across all replicates, 0.1 nM DHT induced luciferase activity 45 fold (log10 fold = 1.55 versus 0 for media control) with an intra-assay CV of 20%, an inter-assay (rep effect) CV of 63% for fold induction, an intraassay CV of 5%, and an inter-assay CV equal to 19% for log10 fold data. When 0.1 nM DHT was coadministered with 1 µM hydroxyflutamide, luciferase activity was reduced to 8.2-fold (log10 = 0.84), which is still significantly above the medium control value. Maximal luc induction of about 63.8-fold was attained at 1 nM DHT.


View this table:
[in this window]
[in a new window]
 
TABLE 2 Effects of Medroxyprogesterone Acetate (MPA) and Dexamethasone (DEX) on AR/GR-dependent Gene Expression in CV-1 and MDA-453 Cells
 

View this table:
[in this window]
[in a new window]
 
TABLE 3 Effects of Chemicals Known to Act As AR Agonists at Relatively High Concentrations on AR/GR-Dependent Gene Expression in CV-1 and MDA-453 Cells
 


View larger version (38K):
[in this window]
[in a new window]
 
FIG. 2. Both cell lines were transduced with the MMTV-luciferase reporter gene. CV-1 cells also received Ad5 hAR virus. E2 concentrations are listed. OHF and ICI were both 1 µM. Fold induction over Et-OH.

 


View larger version (36K):
[in this window]
[in a new window]
 
FIG. 3. Both cell lines were transduced with the MMTV-luciferase reporter gene. CV-1 cells also received Ad5 hAR virus. Cells received antagonists and 0.1 nM DHT. Values are relative 0.1 nM DHT induction alone.

 

View this table:
[in this window]
[in a new window]
 
TABLE 4 Effects of Known Androgen Receptor Antagonists on 0.1nM DHT-Induced Luciferase Induction in CV-1 and MDA-453 Cells
 
Several AR antagonists including M2, M1, and OH-DDE were agonists at higher concentrations (Table 3Go). The relative potency of these chemicals as agonists at 10 µM was M2 (17-fold) > M1 (9.1-fold), OH-DDE = (8.4-fold) > HPTE (which displayed statistically significant but negligible agonist activity of 2.5-fold).

MPA and DEX were less effective in the transduced CV-1 cells than in the MDA-453 cells. MPA induced luc expression by about 10-fold at 0.01–0.05 µM, which was the maximum induction attained (Fig. 4Go, Table 2Go). Dexamethasone induced luc by 3–20-fold, a level that was only about 5% of the value seen in the MDA cells at 100 nM (Table 2Go).



View larger version (26K):
[in this window]
[in a new window]
 
FIG. 4. Both cell lines were transduced with the MMTV-luciferase reporter gene. CV-1 cells also received Ad5 hAR virus. Values are relative 0.1 nM DHT induction.

 
MDA 453 cells transduced with MMTV-luc.
Among all replicates, 0.1 nM DHT induced luc by 23.7-fold (intra-assay CV of 34%; inter-assay/rep CV = 85%) while log10 fold induction was 1.198 (intra-assay CV of 10%; inter-assay/rep CV = 34%). Ten to 100 nM DHT induced maximal luc expression of 100- to 116-fold, respectively. DHT, MPA (Table 2Go), and E2 (Table 3Go) displayed AR agonist activity. MPA induced luc expression by 64-fold at 0.01 µM, with a maximum induction of 174-fold at 0.5 µM (Table 2Go). Dexamethasone induced luc by 248-fold in these cells at 1 µM (Table 2Go). E2-induced expression was blocked with OHF, but the effects of the antiestrogen ICI were less pronounced and not statistically significant (Fig. 2Go). M2, M1, OH-DDE, and HPTE attenuated the effects of 0.1 nM DHT (Table 4Go) at concentrations of 0.05–0.2 µM. In the MDA cells, only M2 displayed mixed antagonist/agonist activity, stimulating luc by almost 5-fold at 10 µM.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
In the current study, we developed user-friendly in vitro assays for screening androgens and antiandrogens, and we evaluated the performance of these assays using known AR agonists and antagonists. CV-1 cells (transduced using adenovirus with hAR and MMTV-luc) and MDA-453 cells (transduced using MMTV-luc) responded as expected (Table 1Go) in a robust manner to each of the chemicals examined. DHT, MPA (Table 2Go), and E2 (Table 3Go) (Wilson et al., 2002Go) displayed AR agonist activity. In experiments with agonists (each of which contained at least 3 replicates with 4 observations per group per replicate), a 2-fold induction of luciferase over media control was a statistically significant effect in either cell line (Table 3Go), albeit a relatively small response when compared to the maximum level attained by potent AR and GR agonists. E2-induced expression was blocked with OHF, but not the antiestrogen ICI, indicating that these were AR-mediated responses (Fig. 2Go). M2, M1 (Wong et al., 1995Go), OH-DDE (Waller et al., 1996Go), and HPTE (Gaido et al., 2000Go) attenuated the effects of 0.1-nM DHT in both cell lines (Table 4Go). In general, a 25–30% reduction in DHT-induced luc activity was required for statistical significance, an effect that was achieved with the above chemicals at concentrations of 0.05–0.2 µM.

In the CV-1 cells (over all replicates), 0.1 nM DHT induced luciferase activity 45-fold (log10 fold = 1.55 vs. 0 for media control) and displayed an intra-assay CV of 20% for fold induction and a CV of 5% for log10 fold data. When 0.1 nM DHT was coadministered with 1 µM hydroxyflutamide in these replicates, luciferase activity was reduced by about 80% to 8.2-fold (log10 = 0.84), which is still significantly above the media control value. Maximal luc induction of about 63.8-fold was attained at 1 nM DHT.

Across replicates using the MDA 453 transduced cells, 0.1 nM DHT induced luc by 23.7-fold (intra-assay CV of 34%) while log10 fold induction was 1.198 (intra-assay CV of 10%). Ten to 100 nM DHT induced maximal luc expression of 100- to 116-fold, respectively.

Consistent with published studies (Table 1Go), several chemicals displayed mixed activity: they exhibited behavior of AR antagonists at lower concentrations and agonists at higher concentrations (Table 3Go). M2, M1 (reported by Wong et al., 1995Go), and OH-DDE displayed weak agonist activity at higher concentrations, effects that typically were more apparent in the CV-1 cell line. In this cell line, the relative potency of these chemicals as agonists was M2 (17-fold) > M1 (9.1-fold) = OH-DDE (8.4-fold) > HPTE (which displayed statistically significant but negligible agonist activity of 2.5-fold). In the MDA cells, M2 induced luc by almost 5-fold.

There did not appear to be a major advantage of one cell line over the other. E2 induced luciferase in transduced CV-1 cells to a greater degree at lower concentrations than it did in MDA cells. Furthermore, the effectiveness of HPTE as an AR antagonist was more apparent in the CV cells than in the MDA cells. In addition, the mixed agonist/antagonist activity of the compounds was more evident in the CV-1 cell assay than in the MDA 453 cell line. On the other hand, MPA, which is both an AR and GR agonist, was more potent in inducing luciferase in the MDA 453 cells, which have GR as well as AR activity, than in the CV-1 cell line, which has little or no functional levels of GR. Similarly, dexamethasone, a potent GR agonist, induced luc by 248-fold in the MDA cells but only by 9.3-fold in the CV-1 cell (Table 2Go). These results suggest that MPA and DEX are acting, at least in part, via GR in the MDA cells. If the objective were to screen for both AR and GR activities, then the MDA cell assay would be more useful, while the transduced CV-1 cells provide greater specificity in displaying AR-mediated responses.

Transduction, the delivery of genes to cells and tissues by replication-deficient viruses, has been studied extensively (Graham, 2000Go; Haddada et al., 1995Go; Hitt et al., 1997Go; Hartig and Hunter, 1998aGo; Hunter and Hartig, 2000Go). While the technology has only recently been applied to toxicology studies (Hartig and Hunter, 1998bGo), it is a basic and well understood technique of molecular biology. Like transfection, it allows for the delivery of genetic material to the target cells. Unlike transfection, in which the DNA is delivered in a dynamic chemical solution or suspension, in transduction the DNA is packaged within the protein coat of a virus particle that has lost the capacity to replicate. Owing to the virus`s stable particulate nature, the dose of a gene (i.e., virus) to a cell can be easily quantified and duplicated with precision. Because the virus is replication-defective, it presents no hazard of infection. In fact during this study, the only exposure risk that required special management was to the chemicals and pesticides being tested. Transduction and transfection techniques require similar facilities and both can be accomplished fairly simply, requiring only basic laboratory equipment (e.g., a tissue culture hood and a CO2 incubator).

Cell lines stably transformed with steroid hormone receptor and/or reporters have recently become useful for evaluating xenobiotic perturbations of transcription activation (Terouanne et al., 2000; Wilson et al., 2002Go). These cell lines were derived via transfection, followed by antibiotic selection and clonal expansion. Ideally, this methodology produces a stable cell population that will respond uniformly to exogenous stimuli. However, a considerable investment of resources is required to produce each cell line, and if it becomes desirable to change the parental cell line, then another round of transfection and antibiotic selection is necessary. In contrast, any cell permissive to transduction can be substituted for any other cell, modifying the assay only minimally as required for the optimum culture condition of the new cell line (i.e., medium formulation, etc.).

Conclusions.
Adenovirus transduction provides a valuable method for delivering exogenous genes. The behavior of the transduced genes can be utilized to assess endocrine-disrupting chemicals. The androgen- and glucocorticoid-regulated luciferase gene (MMTV promoter) responds similarly to chemical stimulus whether it is delivered by transfection or transduction, or is stably integrated into the cellular genome. Transduction utilizes the innate cellular entry mechanisms of the parental virus. Because adenovirus can enter a wide variety of cells, this method should allow the efficient and cost-effective delivery of genes to various cell lines with different compliments of endogenous receptors and co factors. The ability to easily transduce numerous cell lines should facilitate the studies of chemical/receptor interaction.


    ACKNOWLEDGMENTS
 
The authors wish to thanks Drs. T. Stoker, M. Selgrade, J. Rogers, and R. Kavlock for their reviews of this manuscript.


    NOTES
 
The research described in this article has been reviewed by the National Health and Environmental Effects Research Laboratory, U. S. Environmental Protection Agency, and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

1 To whom correspondence should be addressed at MD 67,ORD, NHEERL, RTP, NC 27711. Fax: (919) 541-4017. E-mail: hartig.phillip{at}epa.gov. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bentel, J. M., Birrell, S. N., Pickering, M. A., Holds, D. J., Horsfall, D. J., and Tilley, W. D. (1999). Androgen receptor agonist activity of the synthetic progestin, medroxyprogesterone acetate, in human breast cancer cells. Mol. Cell. Endocrinol. 154, 11–20.[ISI][Medline]

Coutte, B., Le Ricousse, C., Fortin, D., Rafestin-Oblin, M. E., and Richard Foy, H. (1994). The establishment of the long terminal repeat of the mouse mammary tumor virus into CV-1 cells allows a functional analysis of steroid receptors. Biochem. Biophys. Acta 1219, 607–612.[ISI][Medline]

Fenton, M. A., Shuster, T. D., Fertig, A. M., Taplin, M. E., Kolvenbag, G., Bubley, G. J., and Balk, S. P. (1997). Functional characterization of mutant androgen receptors from androgen-independent prostate cancer. Clin. Cancer Res 3, 1383–1388.[Abstract]

Gaido, K. W., Maness, S. C., McDonnell, D. P., Dehal, S. S., Kupfer, D., and Safe, S. (2000). Interaction of methoxychlor and related compounds with estrogen receptor alpha and beta, and androgen receptor: Structure-activity studies. Mol. Pharmacol. 58, 852–858.[Abstract/Free Full Text]

Graham, F. L. 2000. Adenovirus vectors for high-efficiency gene transfer into mammalian cells. Immunol. Today 21, 426–428.[ISI][Medline]

Graham, F. L., and Prevec, L. (1991). Manipulation of adenovirus vectors. Methods Mol. Biol. 7, 109–128.

Haddada, H., Cordier, L., and Perricaudet, M. (1995). Gene therapy using adenovirus vectors. Curr. Top. Microbiol. Immunol. 199, 297–306.[ISI][Medline]

Hartig, P. C., and Hunter, E. S., III (1998a). Gene delivery to the neurulating embryo during culture. Teratology 58, 103–112.[ISI][Medline]

Hartig, P. C., and Hunter, E. S., III (1998b). Effects of over-expression of human p53 on arsenite-induced dysmorphology on mouse embryos in culture. Teratology 75, 215.

Hitt, M. M., Addison, C. L., and Graham, F. L. (1997). Human adenovirus vectors for gene transfer into mammalian cells. Gene Ther. 40, 137–206.

Hunter, E. S., III, and Hartig, P. (2000). Transient modulation of gene expression in the neurulation-staged mouse embryo. Ann. N.Y. Acad. Sci. 919, 278–283.[Abstract/Free Full Text]

Kelce, W. R., Monosson, E., Gamcsik, M. P., Laws, S. C., and Gray, L. E., Jr. (1994). Environmental hormone disruptors: Evidence that vinclozolin developmental toxicity is mediated by antiandrogenic metabolites. Toxicol. Appl. Pharmacol. 126, 276–285.[ISI][Medline]

Kemppainen, J. A., Langley, E., Wong, C. I., Bobseine, K., Kelce, W. R., and Wilson, E. M. (1999). Distinguishing androgen receptor agonists and antagonists: Distinct mechanisms of activation by medroxyprogesterone acetate and dihydrotestosterone. Mol. Endocrinol. 13, 440–454.[Abstract/Free Full Text]

Kontula, K., Paavonen, T., Luukkainen, T., and Andersson, L. C. (1983). Binding of progestins to the glucocorticoid receptor. Correlation to their glucocorticoid-like effects on in vitro functions of human mononuclear leukocytes. Biochem. Pharmacol. 32, 1511–1518.[ISI][Medline]

Lubahn, D. B., Joseph, D. R., Sar, M., Tan, J., Higgs, H. N., Larson, R. E., French, F. S., and Wilson, E. M. (1988). The human androgen receptor: Complementary deoxyribonucleic acid cloning, sequence analysis and gene expression in prostate. Mol. Endocrinol. 2, 1265–1275.[Abstract]

Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Shih, W., Mears, T., Bradley, D. J., Parandoosh, Z., and Weinberger, C. (1991). An adenoviral vector system for functional identification of nuclear receptor ligands. Mol. Endocrinol. 5, 300–309.[Abstract]

Vamvakopoulous, N. C., and Chrousos, G. P. (1993). Evidence of direct estrogenic regulation of human corticotropin-releasing hormone gene expression. Potential implications for the sexual dimophism of the stress response and immune/inflammatory reaction. J. Clin. Invest. 92, 1896–1902.[ISI][Medline]

Waller, C. L., Juma, B. W., Gray, L.E., Jr., and Kelce, W. R., (1996). Three-dimensional quantitative structure-activity relationships for androgen receptor ligands. Toxicol. Appl. Pharmacol. 37, 219–227.

Wilson, V. S., Bobseine, K. L., Lambright C. R., and Gray, L. E., Jr. (2002). A novel cell line, MDA-kb2, that stably expresses an androgen and glucocorticoid responsive reporter for the detection of hormone receptor agonists and antagonists. Toxicol. Sci. 66, 69–81.[Abstract/Free Full Text]

Wong, C., Kelce, W. R., Sar, M., and Wilson, E. M. (1995). Androgen receptor antagonist versus agonist activities of the fungicide vinclozolin relative to hydroxyflutamide. J. Biol. Chem. 270, 19998–20003.[Abstract/Free Full Text]

Zhang, Z., Lundeen, S. G., Zhu, Y., Carver, J. M., and Winneker R. C. (2000). In vitro characterization of trimegestone: A new potent and selective progestin. Steroids 65, 637–643.[ISI][Medline]