This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zhang, J. Articles by Matusik, R. J. Search for Related Content PubMed PubMed Citation Articles by Zhang, J. Articles by Matusik, R. J. Pubmed/NCBI databases Medline Plus Health Information Steroids

A Small Composite Probasin Promoter Confers High Levels of Prostate-Specific Gene Expression through Regulation by Androgens and Glucocorticoids in Vitro and in Vivo¹

Departments of Urologic Surgery (J.Z., T.Z.T., S.K., R.J.M.) and Cell Biology (J.Z., S.K., R.J.M.), The Vanderbilt Prostate Cancer Center (J.Z., T.Z.T., S.K., R.J.M.), Center for Reproductive Biology Research (R.J.M.), and The Vanderbilt-Ingram Cancer Center (R.J.M.), Nashville, Tennessee 37232

Address all correspondence and requests for reprints to: Dr. Robert J. Matusik, Department of Urologic Surgery, A-1302 Medical Center North, Vanderbilt University Medical Center, Nashville, Tennessee 37232-2765. E-mail: robert.matusik{at}mcmail.vanderbilt.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Transient transfection studies have shown that the probasin (PB) promoter confers androgen selectivity over other steroid hormones, and transgenic animal studies have demonstrated that the PB promoter will target androgen, but not glucocorticoid, regulation in a prostate-specific manner. Previous PB promoters either targeted low levels of transgene expression or became too large to be conveniently used. The goal was to design a PB promoter that would be small, yet target high levels of prostate-specific transgene expression. Thus, a composite probasin promoter (ARR₂PB) coupled to the bacterial chloramphenicol acetyltransferase reporter (ARR₂PBCAT) was generated and tested in prostatic and nonprostatic cell lines and in a transgenic mouse model. In PC-3, LNCaP, and DU145 prostate cancer cell lines, the ARR₂PB promoter gave basal expression and was induced in response to androgen and glucocorticoid treatment after cotransfection with the respective steroid receptor. Basal expression of ARR₂PBCAT in the nonprostatic COS-1, MCF-7, ZR-75–1, and PANC-1 cell lines was very low; however, CAT activity could be induced in response to androgens and glucocorticoids when cells were cotransfected with either the AR or GR. In contrast to the transfection studies, ARR₂PBCAT transgene expression remained highly specific for prostatic epithelium in transgenic mice. CAT activity decreased after castration, and could be induced by androgens and, in addition, glucocorticoids. This demonstrates that the necessary sequences required to target prostate-specific epithelial expression are contained within the composite ARR₂PB minimal promoter, and that high transgene expression can now be regulated by both androgens and glucocorticoids. The ARR₂PB promoter represents a novel glucocorticoid inducible promoter that can be used for the generation of transgenic mouse models and in viral gene therapy vectors for the treatment of prostate cancer in humans.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
THE ANDROGEN RECEPTOR (AR) is a potent regulator of gene transcription (1, 2, 3, 4). One prostate-specific gene, probasin (PB), was isolated as an abundant protein from rat dorsolateral prostate nuclei (5, 6, 7, 8, 9). Amino acid analysis revealed that PB belonged to the lipocalin superfamily, a family of ligand carrier proteins (8). Immunohistochemical studies demonstrated that PB is localized to the nucleus of prostatic epithelial cells and in prostatic secretions (8). The nuclear and secreted forms of the protein are translated from a bifunctional messenger RNA (mRNA), where initiation at the first AUG codon produces a protein with a signal peptide for secretion and initiation at the second AUG codon produces an intracellular protein that is truncated by the first 17 amino acids and transported into the nucleus (8).

Androgens regulate the developmental expression of PB gene expression such that PB mRNA levels rise significantly in the prostatic lobes in parallel with increasing serum levels of androgens during the onset of sexual maturity (5, 10, 11). Maximal levels of PB are maintained by androgens in the adult, and castration studies have demonstrated that within 7 days postcastration, PB mRNA levels decrease as the gland regresses (11). In the castrated rat, PB mRNA levels rebound to intact levels between 12–15 days postcastration, suggesting that other factors may also regulate PB gene expression (11).

In previous transient transfection and deoxyribonuclease I footprinting studies, two distinct AR-binding sites, ARBS-1 (located at position -236 to -223) and ARBS-2 (at position -140 to -117), were identified as being required for maximal androgen induction of chloramphenicol acetyltransferase (CAT) gene expression in the human PC-3 prostate cell line (6). Neither binding site alone could induce CAT gene expression in response to androgen treatment. Also, androgen-specific transcriptional activation could not be reconstituted by linking two or three copies of either ARBS-1 or ARBS-2 alone (12). A single point mutation in either ARBS prevented the AR from binding to both sites and eliminated androgen-induced biological activity (13). Therefore, PB was androgen regulated when both AR binding sites were present in the wild-type (wt) configuration.

Androgen-selective activity was further characterized by comparing the relative binding affinities (K_d) of a purified synthetic peptide for the AR (AR2) or a synthetic peptide for the glucocorticoid receptor (GR; GR-DBD) on oligonucleotides for the consensus glucocorticoid response element (GRE) and the PB ARBS-1 or ARBS-2. As indicated in the band shift assays, the DNA binding affinity of the GR for either ARBS-1 or ARBS-2 was so low that it could not be determined, whereas the AR bound with high affinities for ARBS-1 and ARBS-2 (12). Furthermore, the substitution of the C(3)1 gene GRE with the PB ARBS-2 element converted glucocorticoid regulation of C(3)1 gene transcription into an androgen-specific response (14). As well, the PB ARBS-2 DNA element only recognized AR and not the glucocorticoid, mineralocorticoid, or progesterone receptors (15). The determinant for receptor selectivity appeared to reside in the N-terminal part of the second zinc finger as well as in a four-residue C-terminal extension of the DNA-binding domain (15). In PB gene expression, AR selectively binds ARBS-1 and ARBS-2 to regulate gene transcription, and together these two androgen-binding sites constitute the androgen response region (ARR) located between -244 to -96 bp of the PB promoter (12, 13).

The androgen-specific regulation of PB gene transcription and the prostatic epithelial cell specificity of PB gene transcription makes the promoter an ideal candidate to target transgenes to the prostate. Initial transgenic experiments used a small fragment of the PB promoter (-426 to +28 bp) linked to the bacterial CAT reporter gene. Both developmental and hormonal regulation of CAT gene expression was restricted to the transgenic prostatic epithelium; however, the levels of transgene expression were modest (7). Castration studies also showed that CAT gene expression could be restored to precastration levels with androgen treatment, but that glucocorticoid treatment had no effect (7). These data suggested that this small PB promoter lacked important enhancers that regulated PB gene expression. Therefore, a second transgenic model, using a large 12-kb PB promoter fragment (LPB) linked to the CAT gene was developed, which conferred high levels of androgen-regulated transgene expression to the transgenic prostate (16). Immunohistochemical studies showed that LPB-CAT gene expression was epithelial cell specific and that transgene expression was increased dramatically over the -426/+28PB promoter construct (16). Again, castration studies demonstrated that CAT gene expression could be restored to precastration levels with androgen treatment, but that glucocorticoid treatment had no effect, indicating that no functional GRE was present within 12 kb of the PB promoter (16).

In the current study we have developed a composite ARR₂PB promoter that not only maintains prostate-specific transgene expression, but also provides a small promoter that reliably expresses high to very high transgene expression in transgenic mice. Indeed, the ARR₂PB promoter resulted in high levels of transgene expression in 80% of the founders. This new construct offers a further advantage over the other PB promoter constructs in that it now exhibits glucocorticoid inducibility, and its small size makes it ideal to regulate any transgene even in viral vectors where size becomes a limiting problem.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Plasmid construction
The first step in the construction of the minimal PB promoters was the removal of sequences between -426 to -287. Two resulting PB promoter-CAT constructs were generated, one that spanned the endogenous PB promoter sequence from -244 to +28 bp (-244PBCAT) and the other from -286 +28 bp (-286PBCAT). Next, one or two copies of the PB ARR (-244 to -96 bp of the PB promoter, which encompasses one copy of ARBS-1 and one copy of ARBS-2 in the wild-type configuration) (12) were placed 5' to the -286PBCAT or to the shorter -244PBCAT construct.

Cell lines, media, and transient transfection assays
The human prostate carcinoma cell lines PC-3, LNCaP, and DU145; the monkey kidney cell line COS-1; the human breast carcinoma cell line MCF-7; the human pancreas carcinoma cell line PANC-1; and the human mammary gland carcinoma cell line ZR-75–1 were obtained from American Type Culture Collection (Manassas, VA). PC-3 cells were maintained in RPMI medium, 10% FBS, and 250 nM dexamethasone, and the other cell lines were cultured as recommended by American Type Culture Collection. For transient transfection assays, cells were plated at an initial density of 3–5 x 10⁵/100-mm dish or 2–3 x 10⁴/well in 24-well tissue culture plates and transfected by the calcium phosphate method as previously described (6). Briefly, the cells were exposed to the DNA precipitate for 6 h, followed by removal of the medium and addition of steroids for 18 h. The rat AR (rAR) expression vector and the rat glucocorticoid receptor (rGR) expression vector were used in the transfection experiments. Both the rAR and rGR expression vectors use the simian virus 40 promoter to direct expression of these transgenes. The absolute level of protein expression from each vector may vary from cell line to cell line due to the concentrations of transcription factors that may regulate the simian virus 40 promoter and changes in stability of the mRNA and the stability of the receptor proteins. This can ultimately influence the level of induction of the reporter gene. The transfection conditions were initially optimized by cotransfecting mouse mammary tumor virus (MMTV)-CAT with increasing concentrations of rat AR and rat GR expression vectors (12). These experiments determined that the optimal amounts of each steroid receptor expression vector were similar; therefore, all cotransfection experiments were performed using an equal concentration of plasmid (5 µg/ml), which was the optimal rAR and rGR concentration for the maximal response of MMTV-CAT to dihydrotestosterone (DHT) or dexamethasone (DEX) treatment, respectively (12). The cells transfected with PBCAT plasmids were harvested after 24 h of incubation from the beginning of the transfection, and CAT activity was determined by the two-phase fluor diffusion assay using [³H]acetyl-coenzyme (5.9 Ci/mmol; Amersham Pharmacia Biotech, Piscataway, NJ) as the substrate (6). The transfection efficiency was determined by cotransfecting pRL-CMV containing the Renilla luciferase reporter gene (Promega Corp., Madison, WI). The luciferase activity was determined using the Promega Corp. luciferase assay system (NOREF>E1980). The values plotted represent the mean of at least three individual samples ± SD.

Establishment of ARR₂PBCAT transgenic mouse lines
All animal studies were conducted in accordance with the principles and procedures outlined by Vanderbilt University council on animal care. From the experimental cell culture results, the ARR₂PB construct was chosen for further study. The ARR₂PBCAT transgenic mice were generated by microinjection of the DNA into the male pronucleus of a fertilized oocyte by the Transgenic Core/ES Cell Shared Resource of the Vanderbilt-Ingram Cancer Center. Four transgenic founders, three males and one female, were generated. Animals were housed in standard lighting (12 h of light, 12 h of darkness) and allowed food and water ad libitum. Transgenic animals were identified by PCR-based screening using isolated mouse tail DNA as described previously (16), and the lines were maintained as the B6D2F1 mouse strain. Transgene copy number was determined by Southern blot and DNA slot blot hybridization (17). Mouse genomic DNA from transgenic animals containing a single copy of the transgene, mouse epididymal retinoic acid-binding protein (RABP) promoter linked to the CAT transgene (mE-RABPCAT), served as an internal control (18). A 1200-bp complementary DNA fragment of the CAT gene was used to detect the both the ARR₂PBCAT and the mE-RABPCAT (control) transgene. In addition, as a further internal control, a 600-bp mE-RABP DNA fragment was used to detect the endogenous RABP gene. The DNA probes were labeled with [³²P]deoxy-CTP (3000 Ci/mmol; Amersham Pharmacia Biotech) using the Rediprimer II kit (Amersham Pharmacia Biotech).

Prostate-specific and developmental expression of the ARR₂PBCAT transgene
The levels of ARR₂PBCAT gene expression in the four lines were compared in males and females ranging from 8–9 weeks of age (at least five mice per group). Tissue samples from the anterior (AP), dorsolateral (DLP), and ventral prostate (VP) were microdissected for the determination of CAT activity. To determine tissue specificity, other tissues, including brain, spleen, thymus, heart, lung, liver, kidney, testes, muscle, adrenals, epididymis, and seminal vesicle from transgenic males, and brain, spleen, thymus, heart, lung, liver, kidney, muscle, adrenals, uterus, ovary, and oviduct from transgenic females, were analyzed for CAT activity. Proteins were extracted by homogenization in buffer containing 0.1 M Tris (pH 7.8) and 0.1% Triton X-100 buffer, and protein concentrations were determined by protein assay (Bio-Rad Laboratories, Inc., Richmond, CA). The CAT assay was performed as previously described (6), and results were expressed as disintegrations per min/mg protein ± SD or, in cases of extremely high levels of CAT activity, per µg protein.

To determine the developmental pattern of ARR₂PBCAT gene expression before, during, and after sexual maturation, four to seven transgenic males per time point were killed at 1–14 weeks and 36 weeks of age. The AP, DLP, and VP were microdissected, and CAT activity for the individual lobes was determined as described above.

Immunohistochemical studies
Animals from line 3699b, aged 17 weeks (n = 2) or 36 weeks (n = 2), were killed, and the prostates were excised. The individual prostate lobes (AP, DP, LP, and VP) were dissected and fixed in 4% buffered formalin. After processing and embedding in paraffin, 5-µm sections were cut for immunohistochemical detection of CAT protein.

Sections were deparaffinized, rehydrated, and placed in 0.05 M citrate buffer, pH 6.0. Antigenic sites were exposed by microwaving the sections for 30 min at 95–99 C before the removal of endogenous peroxidase activity with 3% H₂O₂. Nonspecific binding was blocked by incubating the sections with 1% DIG Blocking Reagent (Roche Molecular Biochemicals, Mannheim, Germany) in 100 mM Tris-HCl and 50 mM NaCl for 30 min. The sections were incubated with anti-CAT antibody (5'-3', Inc., Boulder, CO) overnight at 4 C (1:400 dilution), washed in PBS (14 mM NaCl, 0.3 mM KCl, 1 mM Na₂PO₄, and 0.2 mM KH₂PO₄, pH 7.4), and incubated with biotinylated antirabbit IgG (1:300; DAKO Corp., Carpinteria, CA) at room temperature for 2 h. The sections were washed with PBS and incubated for 1 h with streptavidin (1:50; Vectastain Elite ABC Kit, Vector Laboratories, Inc., Burlingame, CA). After additional washes with PBS and 50 mM Tris-HCl (pH 7.4), peroxidase activity was detected using 3',3'-diaminobenzidine tetrahydrochlorate (Liquid DAB Substrate-Chromagen System, DAKO Corp.). The reaction was terminated in distilled water, and the sections were counterstained with Harris hematoxylin (Surgipath, Richmond, VA), dehydrated, and permanently mounted with cytoseal XYL (Stephens Scientific, Kalamazoo, MI).

Hormonal regulation of ARR₂PBCAT gene expression in the prostate
To determine the effects of androgen ablation and hormone treatment on ARR₂PBCAT gene expression, transgenic males were castrated at 10–11 weeks of age by the scrotal route. Sham-operated age-matched males served as controls (n = 5). At 14 days postcastration, five castrated males per group received one of the following pellets sc. Steroid pellets (Innovative Research of America, Sarasota, FL) consisted of DHT (1.5 mg/pellet), DEX (1.5 mg/pellet), flutamide (25 mg/pellet), RU486 (5 mg/pellet), or placebo (5-mg pellet containing no hormone). After 7 days of treatment, the animals were killed, and CAT activity was measured in protein extracts obtained from the anterior, dorsolateral, and ventral lobes of the prostate. The effects of long-term castration on transgene expression in the absence of hormone treatment were determined by measuring the levels of CAT activity in four animals per group at each of the following time points: 14, 28, and 42 days postcastration.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Multiple copies of the ARR produce enhanced steroid hormone regulation to the PB promoter
The present study used a number of minimal PB promoter constructs to determine the region required for prostate-specific gene expression and to determine whether glucocorticoid regulation could be added to androgenic regulation in transgenic mice. The first step in the construction of the minimal PB promoters was the removal of the sequences between -426 to -287, which are inhibitory to reporter gene expression in tissue culture (6). Two resulting PB promoter-CAT constructs were generated, one that spanned the endogenous PB promoter sequence from -244 to +28 bp (-244PBCAT), and the other from -286 to +28 bp (-286PBCAT). Next, one or two copies of the PB ARR were placed 5' to the -286PBCAT or to the shorter -244PBCAT construct. To determine whether both androgens and glucocorticoids had differential effects on the transcriptional regulation of these constructs, the induction of CAT activity was measured in the presence of AR and DHT or in the presence of GR and DEX, respectively, and compared with the basal level of CAT activity in the absence of hormone treatment in LNCaP and PC-3 cells.

Initially, CAT gene expression was determined in LNCaP cells in response to increasing concentrations of either DHT (Fig. 1A) or DEX (Fig. 1B). As shown in Fig. 1, the optimal concentration for steroid hormone-induced CAT gene expression was 10^-8 M DHT and 10^-8 M DEX for any construct tested. At these optimal steroid conditions, the -426/+28PBCAT construct was preferentially induced by DHT (11-fold) compared with DEX (1.8-fold) treatment, consistent with our previous observations about the biological activity of the wt PB promoter, which contains one copy of the ARR (6, 13). When a second ARR DNA sequence was added upstream of the -244PB promoter [(ARR)-244PBCAT], CAT activity increased 88-fold in the presence of DHT, and a 285-fold increase in CAT activity was seen when a third copy of ARR was added [(ARR)(ARR)-244PBCAT; Fig. 1A]. Surprisingly, when LNCaP cells were transfected with (ARR)-244PBCAT plus the GR expression vector and treated with DEX, the levels of CAT gene activity increased dramatically to 137-fold when ARR copy number was doubled and 301-fold when ARR copy number was tripled (Fig. 1B), compared with the 1.8-fold induction seen with -426/+28PBCAT. Therefore, the addition of one or two ARR DNA elements appears to alter biological activity in response to glucocorticoid hormone treatment in vitro.

View larger version (27K): [in this window] [in a new window]   Figure 1. Comparison of androgen- and glucocorticoid-induced CAT activities in LNCaP cells. One or two copies of ARR were subcloned into -244/+28 PBCAT and -286/+28 PBCAT. -244/+28 PBCAT and -286/+28 PBCAT, which contain one copy of ARR between -244 to -96 bp, served as wild-type controls for both androgen- and glucocorticoid-induced activity. These constructs were cotransfected either with an AR expression vector into LNCaP cells and treated with 10-11, 10-10, 10-9, or 10-8 M DHT (A) or with a GR expression vector and treated with 10-11, 10-10, 10-9, or 10-8 M DEX (B). The fold induction in response to hormone treatment was determined by comparing receptor gene activity induced by DHT or DEX to the corresponding baseline values in the absence of hormone. Activities and fold induction were measured in at least three determinations ± SD, and all experiments were adjusted for transfection efficiency.

The (ARR)-286PBCAT and (ARR)(ARR)-286PBCAT constructs were also tested for glucocorticoid induction by transfecting LNCaP cells with either construct together with the GR expression vector plus DEX treatment (Fig. 1B). The CAT activity increased to 203-fold with (ARR)-286PBCAT and to 238-fold with (ARR)(ARR)-286PBCAT in the presence of glucocorticoid treatment. Again, these promoters were able to greatly increase responsiveness to glucocorticoid treatment compared with that seen with the parent construct -426/+28PBCAT. Indeed, the activity of (ARR)-286PBCAT induced by glucocorticoids was virtually equivalent (203-fold) to that induced by DHT (229-fold).

As the largest rise in CAT gene expression in the LNCaP assay occurred when one or more ARR DNA fragments were linked to the basic PB promoter region, these constructs were further evaluated in a second human prostatic cell line. The results obtained in the PC-3 bioassay showed that the response to 10^-8 M DHT treatment of (ARR)-286PBCAT gene expression was 480-fold, and that of (ARR)(ARR)-286PBCAT gene expression was 324-fold (Fig. 2A). Treatment with DEX at 10^-8 M produced a 496-fold induction, similar to that observed with DHT treatment (Fig. 2B). When a second ARR was added upstream of the -244PB promoter [(ARR)-244PBCAT], CAT activity increased 79-fold in the presence of DHT; however, only a 30-fold increase in CAT activity was seen when a third copy of ARR was added [(ARR)(ARR)-244PBCAT] (Fig. 2A). When PC-3 cells were transfected with (ARR)-244PBCAT plus the GR expression vector and treated with DEX, the levels of CAT gene activity rose from 5- to 66-fold when ARR copy number was doubled and 74-fold when ARR copy number was tripled (Fig. 2B). Again, in a second prostatic cell line a similar trend was seen, and the data suggest that the close proximity of multiple AR-binding regions inhibits transgene expression in the presence of androgens and glucocorticoids.

View larger version (18K): [in this window] [in a new window]   Figure 2. Evaluation of androgen- and glucocorticoid-induced activities in PC-3 cells. One or two copies of ARR were subcloned into -244/+28 PBCAT or -286/+28 PBCAT. -244/+28 PBCAT and -286/+28 PBCAT, which contain one copy of ARR between -244 to -96 bp, served as wild-type controls for both androgen- and glucocorticoid-induced activities. These constructs were cotransfected either with an AR expression vector into PC-3 cells and treated with 10-8 M DHT (A) or with a GR expression vector and treated with 10-8 M DEX (B). The fold induction in response to hormone treatment was determined by comparing receptor gene activity induced by DHT or DEX to the corresponding baseline value in the absence of hormone. The fold induction is shown in numbers above the histogram bars. Activities and fold induction were measured in at least three determinations ± SD, and all experiments were adjusted for transfection efficiency.

ARR₂PB promotes basal levels of prostate-specific CAT expression in vitro
The specificity of the ARR₂PB promoter to regulate transgene expression was determined in both prostatic and nonprostatic cell lines. The ARR₂PBCAT construct was cotransfected with the pRL-CMV vector (transfection efficiency control; Promega Corp.) into several prostatic (PC-3, LNCaP, and DU145) and nonprostatic (COS-1, MCF-7, PANC-1, and ZR-75–1) cell lines. Basal levels of CAT activity were detected in all three prostate cancer cell lines, whereas low to negligible levels of CAT activity were detected in the nonprostatic cell lines (Fig. 3). Compared with PC-3 cells, the basal levels of CAT activity in ZR-75–1, MCF-7, COS-1, and PANC-1 cells were only 23%, 6.8%, 3.7%, and 1%, respectively. These results indicate that the ARR₂PB promoter maintains a high degree of specificity for preferential expression in human prostate cancer cell lines.

View larger version (33K): [in this window] [in a new window]   Figure 3. Relative ARR2PBCAT activities in different cell lines. Prostatic and nonprostatic cell lines were transfected with the ARR2PBCAT construct together with the pRL-CMV construct to control for transfection efficiency. All assays measured triplicate samples in at least three independent experiments. The CAT activity was normalized for luciferase activity and protein content. The relative CAT activities of samples are show as the mean ± SD.

View larger version (21K): [in this window] [in a new window]   Figure 4. Androgen and glucocorticoid induction of ARR2PBCAT in different cell lines. Prostatic and nonprostatic cell lines were transfected with ARR2PBCAT together with pRL-CMV and cotransfected with either rAR or rGR expression vectors. The CAT activities were determined in the presence or absence of 10-8 M DHT or 10-8 M DEX. CAT activity was normalized for luciferase activity and protein content. All assays measured triplicate samples in at least two independent experiments. The relative CAT activities in different cell lines are shown as the mean ± SD.

The ARR₂PBCAT line 3699b and line 3706 showed similar levels of CAT activity to that seen in the -426/+28PBCAT line. In most of the transgenic lines, the highest levels of PB-directed transgene expression occurred in the VP, and the lowest levels occurred in the AP (Table 2). The exception was line 3699b, in which transgene expression was highest in the DLP, similar to that observed in LPBCAT line VI (16). These results suggest that the site of transgene integration and copy number not only affect the rate of transgene expression but may also influence the level of transgene expression in the individual lobes of the prostate.

Developmental regulation of ARR₂PBCAT expression
PB gene expression is regulated developmentally by androgens, such that accumulation of PB mRNA increases in parallel with the rise in serum androgens and plateaus with the onset of sexual maturity. In line 3706, a moderately expressing line, at least four males were killed per time point at weekly intervals from 1–14 weeks and at 36 weeks of age. The prostates of the 1-week-old animals were too small to microdissect into individual lobes; therefore, CAT gene expression was measured as the total CAT activity per gland. In all other age groups, the lobes of the prostate were separated and analyzed individually (Fig. 5A). As early as 1 week of age, CAT activity in the prostate of line 3706 mice was detectable at very low levels. Thereafter, CAT activity increased rapidly by 2- to 3-fold each week, reaching a peak activity of 35,550 ± 3,233 dpm/min·mg protein in the AP by the seventh week, 54,210 ± 3,348 dpm/min·mg protein in the DLP by the eighth week, and 80,770 ± 6,053 dpm/min·mg protein in the VP by the eighth week. A small decrease in CAT activity was seen thereafter, but CAT activity was still maintained at this high level. Therefore, the increase in ARR₂PBCAT gene expression followed the rise and plateau in serum androgen levels that occur during sexual maturation.

View larger version (23K): [in this window] [in a new window]   Figure 5. Developmental expression of the ARR2PBCAT transgene in lines 3706 and 3699b. The ARR2PBCAT transgenic animals of line 3706 (A) and 3699b (B) were killed, and the prostate lobes were dissected and frozen at -70 C. Individual lobes of 1-week-old animals could not be dissected; therefore, the values represent the whole prostate. Each time point represents the mean CAT activity of four or five animals. In A, time points were taken weekly starting at 1 week of age to 16 weeks of age. Twenty weeks later the last time point was added. In B, time points included 2, 3, 5, 6, 9, 12, and 36 weeks of age. Values are presented as the mean ± SD. Prostatic lobes are as follows: , anterior lobe; , dorsolateral lobe; , ventral lobe.

ARR₂PB targets transgene expression to prostatic epithelial cells
In the construction of the composite ARR₂PB promoter, the sequences between -426 and -286 were removed. This deletion did not affect prostate-specific expression in vivo (Table 1); however, the removal of these sequences could have influenced the prostatic epithelial cell-specific CAT expression observed previously in the -426/+28PBCAT transgenic mice (7). Therefore, the individual lobes of the prostate from male mice that were 17 and 36 weeks old (line 3699b) were microdissected, fixed in formalin, and processed for immunohistochemical studies using an anti-CAT antibody. The prostates showed specific positive immunoreactivity for the CAT protein only in the epithelial cells (Fig. 6). There were no differences in staining pattern or staining intensity for CAT protein between the two age groups. Control sections incubated with normal rabbit serum showed no positive immunoreactivity in either the epithelium or the stroma. Very faint staining for CAT protein was observed in the anterior prostate (Fig. 6A), consistent with the low levels of CAT activity determined for this lobe in line 3699b (Table 2). Figure 6, B and C, shows positive CAT immunostaining in the dorsal and lateral prostate lobes, respectively. Although the amount of CAT activity was not individually determined for the dorsal and lateral lobes, the intensity of staining would suggest that the amount of CAT protein is greater in the lateral lobe (Fig. 6C) than in the dorsal lobe (Fig. 6B). This observation is consistent with the expression and staining pattern for the endogenous PB protein in the rodent prostate (8). Figure 6D shows strong positive CAT immunostaining in the ventral prostate. In contrast, no positive immunoreactivity for CAT protein was detectable in the seminal vesicles (Fig. 6E) or in controls treated with normal rabbit serum (Fig. 6F). Positive immunoreactivity in the nucleus of some sections was observed in both control and CAT antibody-treated sections. This background immunoreactivity was not alleviated by different antigen retrieval techniques, new antibody batches, or changes in the staining method. Similar nuclear staining with this CAT antibody has been previously reported (18).

View larger version (126K): [in this window] [in a new window]   Figure 6. Immunohistochemical analysis of the localization of CAT protein in ARR2PBCAT line 3699b transgenic mice. Using an anti-CAT protein antibody, CAT protein was localized to the secretory epithelium of anterior (A), dorsal (B), lateral (C), and ventral (D) lobes of the prostate of mice from line 3699b. No positive immunoreactivity was observed in the seminal vesicles (E) or in control sections incubated with normal rabbit serum (F). All photographs were taken at a magnification of x400.

View larger version (24K): [in this window] [in a new window]   Figure 7. Long-term castration effects on the ARR2PBCAT transgene expression. Nine-week-old male mice of line 3699b were castrated. Animals (n = 4/group) were killed at 1, 28, and 42 days postcastration, tissues were collected, and the CAT activity was determined. The relative CAT activity in the different tissues are presented as the mean ± SD.

View larger version (33K): [in this window] [in a new window]   Figure 8. Hormonal regulation of ARR2PBCAT gene expression in transgenic mice. Nine-week-old transgenic male mice of line 3696 were castrated. Fourteen days postcastration, the animals were implanted with placebo, DHT, DEX, flutamide, or RU486 pellets (n = 5 animals/group) as described in Materials and Methods. After 7 days of treatment, these animals were killed, and the CAT activities were measured. Intact animals served as controls. To determine the P values, the CAT activities in DHT-, DEX-, flutamide-, and RU486-treated animals were compared with that in placebo-treated animals. P values are shown by the numbers above the histogram bars.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Several androgen-regulated prostate-specific genes have been identified, and these could become the stepping stones for identifying other prostate-specific regulatory proteins that may be important in prostate morphogenesis. Although the presence of a functional AR is a critical event in the development and differentiation of the prostate, its presence in many tissues in both males and females indicates that it is not sufficient to explain prostate-specific gene expression. We have shown that the rat PB promoter targets the prostate of transgenic mice (7, 16). Similarly, the human prostate-specific antigen (PSA) gene is expressed in a highly prostate-specific manner in man. A large 6-kb 5'-flanking region (19) and a 14-kb genomic PSA fragment (20) target the prostate of transgenic mice, but a smaller 600-bp PSA promoter fragment is not prostate specific (21).

Transgenic animal studies have determined that bases -426 to +28 of the endogenous PB promoter are sufficient to target prostate-specific transgene expression (7). As a result, researchers have been analyzing this promoter fragment to define potential prostate-specific elements. Analysis of this promoter fragment has centered around transient transfection assays in prostatic and nonprostatic cell lines. Within the -426/+28PB promoter resides the ARR (-244/-96 bp) of the PB gene (12), which contains two AR-binding sites, ARBS-1 at -236 to -223 bp and ARBS-2 at -140 to -117 bp (7). The size of this ARR regulatory element suggests that transcription factors other than androgens may interact with the AR to contribute to prostate-specific gene expression. Analysis of the PB promoter by Patrikainen et al. revealed that the PB sequences between positions -278 to -240 bp were critical for both androgen- and prostate-specific gene regulation in cell culture, and they concluded that an important transcription factor-binding site at position -251 to -240 bp was integral for determining tissue specificity (22). Caution must be exercised in interpreting transient transfection into prostatic and nonprostatic cell lines as a measure of tissue specificity, as expression of the PB promoter in nonprostatic cells can be induced by the cotransfection of steroid receptors (23). Interpretation of data generated in vitro assumes that 1) the critical regulatory factors that confer prostate-specific gene expression are present in prostate cancer cell lines, and 2) that transformed nonprostatic cells lines have not acquired the expression of regulatory factors critical for prostate-specific gene expression. Unfortunately, neither of these assumptions can be confirmed because the critical regulatory factors that govern prostate-specific gene expression have not been identified. Furthermore, normal prostatic cells cannot be maintained in cell culture, and therefore, we have to rely upon established prostate cancer cell lines to perform these studies. Therefore, the definitive test of tissue specificity is the transgenic mouse model.

The ARR₂PB promoter conferred prostate-specific transgene expression that was developmentally and hormonally regulated in transgenic mice. These results were consistent in four separate transgenic lines. Indeed, the new ARR₂PB promoter was highly effective, resulting in high levels of transgene expression in four of the five established lines (80%). No CAT activity was measured in any other tissue tested, thus confirming that ARR₂PB contained the elements required for prostate-specific gene expression. In the -426/+28PBCAT transgenic mice, one of three founders (33%) had relatively high levels of transgene expression (7), whereas in LPBCAT transgenic mice, two of six founders (33%) showed high levels of transgene expression (16). In vivo, the new small composite ARR₂PB promoter could increase CAT activity in the prostate up to 1436–2435% greater than that measured with the highest -426/+28PBCAT-expressing transgenic line. Similarly, expression was comparable to that observed in the highest expressing LPBCAT line. Thus, the ARR₂PB construct further narrows down the prostate-specific cis-acting element as residing between -286 to +28 bp of the PB promoter and indicates that the strong hormone-responsive region of the ARR₂PB, as demonstrated in vitro, is also functioning in vivo.

In most transgenic animal models, the PB constructs target higher transgene expression to the ventral lobe, where the endogenous PB protein is expressed at lower levels. The dorsolateral prostate, which actually expresses the endogenous PB protein at the highest level, generally expresses lower levels of the transgene. These observations suggest that the lobe-specific DNA sequences are missing from these PB constructs, but prostate-specific sequences are distinct and included. However, the ARR₂PBCAT line 3699b does show lobe-specific expression of the transgene, suggesting that the sequences required for lobe-specific expression are present in our constructs; however, the site of transgene integration into the genome may override these signals. Similarly, the levels of transgene expression, which are not consistent with transgene copy number, may be affected by the site of transgene integration through silencing. We have previously reported this silencing effect on the -426/+28PB and LPB promoters (7, 16). Furthermore, numerous transgenes in tandem may condense into euchromatin, thereby silencing transcription through interactions between the repeats (24, 25). In addition, line 3699 split at the F₁ generation to establish 3699a (12 gene copies) and 3699b (4 gene copies). Subsequent studies with line 3699b suggest that this line appears to have undergone further changes in regulation since the F₁ males were first analyzed, as future generations now show increased transgene expression. This may reflect DNA imprinting, which would alter transgene expression if the methylation pattern at the site of integration changed.

Transient transfection and transgenic mouse studies have previously shown that the PB promoter in the wild-type conformation confers androgen selectivity over other steroid hormones (6, 7, 16, 26, 27, 28). In this study we have characterized the probasin ARR by combining two or three probasin ARRs. Surprisingly, this combination altered steroid hormone specificity such that transgene expression was now regulated by glucocorticoids in addition to androgens in both cell culture and transgenic mice. This novel characteristic was unexpected, because a single ARR sequence confers strong androgenic regulation of transgenes (13).

An androgen response element (ARE) was previously defined as having the identical canonical sequence and activity as the MMTV GRE. Specificity for AR and GR binding was believed to result from the interactions of the AR or GR with other coactivators. Probasin ARBSs can bind the AR, but each site alone functions poorly as an ARE (12, 13). The combination of ARBS-1 and ARBS-2 forms the ARR, which now regulates androgen-induced gene transcription in a cooperative manner (12, 13). Notably, these binding sites can differentiate between the AR and GR when they are in the wt configuration (12, 13). Furthermore, ARBS-2 is the first DNA-binding element that has been shown to be selective for transcriptional regulation by the AR (6, 12, 15, 29). Recently, AREs have been redefined as low affinity AR-binding sites, and multiple AREs are now believed to function cooperatively to regulate transcription in response to androgen treatment (3, 12, 30).

DNA binding studies have shown that ARBS-1 and ARBS-2 bind the AR selectively (12) and that the -426/+28PB promoter is transcriptionally regulated by the AR with weak activity for the GR in vitro (6). Transgenic mouse studies using the -426/+28PB and LPB have also determined that no functional GRE resides within these PB promoter constructs (7, 16, 26, 27, 28). Comparison of the -426/+28PB and ARR₂PB promoters in LNCaP cells in a dose-response curve to androgens and glucocorticoids confirmed that the -426/+28PB promoter is preferentially induced by androgens (11-fold) over glucocorticoids (1.8-fold), and that overall this promoter has very weak induction of transcription. In contrast, the ARR₂PB promoter in LNCaP cells showed strong induction by androgens (229-fold) and near equal induction by glucocorticoids (203-fold). Clearly, the discriminating nature between AR and GR does involve precise cis-DNA elements (12, 14, 15); however, changing the arrangement of these elements alters their activity, most likely by changing the cooperative effects between receptors and other corepressors and/or coactivators. Thus, the increased sensitivity to glucocorticoids does not appear to be promoter specific, but, rather, is due to the interactions between the multiple ARRs. It is possible that the weak individual interactions of the GR for one ARR are further stabilized by having multiple copies present in tandem, thereby allowing the cooperative interaction between these enhancer regions. The strong induction seen with the ARR₂PB promoter in vitro suggests that transcriptional regulation by this promoter in vivo would be distinctly different from that seen with the -426/+28PB promoter. It is important to note that the cell culture experiments use transformed prostate cells cotransfected with a GR expression vector; this would result in of thousands of copies of GR DNA per cell. This would most likely achieve higher levels of the steroid receptor in vitro than existing levels in vivo in normal prostatic cells. In vivo, castration induces prostatic apoptosis by which the VP regresses to about 1/10th, and the DLP regresses to about 1/5th of its original size, suggesting that epithelial cells from these two lobes respond differently to androgen. The remaining atrophic epithelial cells may have lost androgen-regulated cofactors that are important for regulation of the probasin promoter. Thus, the administration of DEX to castrated mice may not be able to induce the necessary androgen regulated cofactors that are required for ARR₂PB transgene expression. Therefore, either or both explanations could account for the strong glucocorticoid effects observed in vitro that are not duplicated in vivo.

Previous castration studies in the -426/+28PBCAT or LPBCAT transgenic mice showed that prostatic CAT activity dropped to undetectable levels by 3 weeks postcastration (7, 16). Similarly, the current castration study using ARR₂PBCAT mice demonstrates that removal of androgens decreases CAT activity, whereas treatment with DHT restores prostatic CAT activity to precastration levels. In contrast to the -426/+28PBCAT or LPBCAT transgenic mice, relatively high levels of CAT activity were still measurable in the prostate of ARR₂PBCAT mice 6 weeks postcastration, suggesting that other factors could maintain levels of transgene expression. Treatment of castrated ARR₂PBCAT mice with flutamide further reduced prostatic CAT activity compared to the placebo group, suggesting that flutamide could inhibit low levels of adrenal androgens. However, the adrenal androgens alone were not sufficient to explain the continued transgene expression after castration.

In vitro experiments indicated that glucocorticoids regulated transgene expression; therefore, it is possible that the continued CAT expression after castration could be due to adrenal glucocorticoids. To test this hypothesis, pellets containing either DEX or the antiglucocorticoid, RU486, were administered to castrated mice. DEX treatment increased CAT activity (P < 0.05) in the prostate of castrated mice, whereas RU486 decreased CAT activity (P < 0.05) over and above that seen with flutamide. Thus, ARR₂PB gene expression can be regulated by both androgens and glucocorticoids in vivo.

Gene therapy using prostate-specific promoters to target viral vectors to prostate cancer cells has been proposed as a new treatment for prostate cancer patients that have failed androgen deprivation therapy. Even though greater than 80% of patients show a favorable response to androgen deprivation as measured by tumor regression and a decrease in serum PSA levels, the duration of response is usually 2–3 yr (31, 32). A number of potential mechanisms involving the AR have been proposed as reasons for the failure of androgen deprivation therapy (33, 34, 35, 36, 37, 38, 39, 40). These observations indicate that the progression of prostate cancer may still involve an active AR. In gene therapy constructs that use an androgen-driven promoter, such as the PB and PSA promoters, it may be possible to take advantage of this active AR in relapsed patients. However, it would be unacceptable to treat these men with androgens to induce an androgen-dependant viral vector, because androgen treatment would also drive prostate cancer growth. This presents a major problem in using an androgen-regulated prostate-specific promoter for gene therapy.

However, men who fail androgen deprivation therapy are often treated with glucocorticoids because it improves the quality of life, and a decrease in serum PSA levels has also been reported (41, 42). As glucocorticoid treatment is currently an acceptable therapy, it could potentially be used to induce transcription of an ARR₂PB therapeutic gene in a prostate-specific manner for the treatment of prostate cancer. Thus, the glucocorticoid-inducible ARR₂PB promoter may provide a novel strategy for regulating transgene expression in the development of gene therapy vectors for the treatment of prostate cancer.

In conclusion, the ARR₂PB promoter maintains prostate, epithelial cell specificity in vivo, suggesting that the sequence from -286 to +28 bp of the endogenous PB promoter retains the DNA sequences necessary for prostate-specific transgene expression. This new ARR₂PB composite promoter provides major advantages over previous constructs. First, the composite promoter is highly consistent in generating transgenic lines with high levels of transgene expression (80%) compared with the -426/+28PB and LPB promoters (33%). This is advantageous because if only a single transgenic line is generated, it is impossible to determine the effects of the site of integration on tissue specificity or, if no transgene expression occurs, to determine whether the site of integration or the actual transgene construct resulted in inactivity. Further, if the transgenic line has low levels of transgene expression, it may be difficult to obtain a prostatic phenotype. Second, the ARR₂PB promoter is small enough to link to any transgene in viral vectors for the development of therapeutic gene therapy in the treatment of prostate cancer. The added glucocorticoid regulation via this composite promoter makes it a valuable tool for the treatment of prostate cancer patients that have failed androgen deprivation therapy.

	Acknowledgments

 
The authors thank Mr. Tom Case and Mr. Manik Paul for their technical assistance, and Miss Lisa Howell for the preparation of this manuscript.

	Footnotes

 
¹ This work was supported by R01-DK-55748 from the NIH and the Frances Williams Preston Laboratories of the T. J. Martell Foundation. Transgenic mice were bred by the Transgenic Core/ES Cell Shared Resource of the Vanderbilt-Ingram Cancer Center (NCI Grant 2P30-CA-68485–05).

Received March 2, 2000.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Brinkmann AO, Trapman J 1992 Androgen receptor mutants that affect normal growth and development. Cancer Surv 14:95–111[Medline]
2. Kokontis J, Takakura K, Hay N, Liao S 1994 Increased androgen receptor activity and altered c-myc expression in prostate cancer cells after long-term androgen deprivation. Cancer Res 54:1566–1573[Abstract/Free Full Text]
3. Shrahle U, Klock G, Schutz G 1987 A DNA sequence of 15 base pairs is sufficient to mediate both glucocorticoid and progesterone induction of gene expression. Proc Natl Acad Sci USA 84:7871–7875[Abstract/Free Full Text]
4. Ham J, Thomson A, Needham M, Webb P, Parker M 1988 Characterization of response elements for androgens, glucocorticoids and progestins in mouse mammary tumour virus. Nucleic Acids Res 16:5263–5276[Abstract/Free Full Text]
5. Dodd JG, Sheppard PC, Matusik RJ 1983 Characterization and cloning of rat dorsal prostate mRNAs. Androgen regulation of two closely related abundant mRNAs. J Biol Chem 258:10731–10737[Abstract/Free Full Text]
6. Rennie PS, Bruchovsky N, Leco KJ, Sheppard PC, McQueen SA, Cheng H, Block ME, MacDonald BS, Nickel BE, Chang C, Liao S, Cattini PA, Matusik RJ 1993 Characterization of two cis-acting elements involved in the androgen regulation of the probasin gene. Mol Endocrinol 7:23–36[Abstract]
7. Greenberg NM, DeMayo FJ, Sheppard PC, Barrios R, Lebovitz M, Finegold M, Angelopoulou R, Dodd JG, Duckworth ML, Rosen JM, Matusik RJ 1994 The rat probasin gene promoter directs hormonally and developmentally regulated expression of a heterologous gene specifically to the prostate in transgenic mice. Mol Endocrinol 8:230–239[Abstract]
8. Spence AM, Sheppard PC, Davie JR, Matuo Y, Nishi N, McKeehan WL, Dodd JG, Matusik RJ 1989 Regulation of a bifunctional mRNA results in synthesis of secreted and nuclear probasin. Proc Natl Acad Sci USA 86:7843–7847[Abstract/Free Full Text]
9. Matuo Y, Nishi N, Negi T, Tanaka Y, Wada F 1982 Isolation and characterization of androgen-dependent non-histone chromosomal protein from dorsolateral prostate of rats. Biochem Biophys Res Commun 109:334–340[CrossRef][Medline]
10. Matusik RJ, Kreis C, McNicol P, Sweetland R, Mullin C, Fleming WH, Dodd JG 1986 Regulation of prostatic genes: role of androgens and zinc in gene expression. Biochem Cell Biol 64:601–607[Medline]
11. Sweetland R, Sheppard PC, Dodd JG, Matusik RJ 1988 Post-castration rebound of an androgen regulated prostatic gene. Mol Cell Biochem 84:3–15[Medline]
12. Kasper S, Rennie PS, Bruchovsky N, Lin L, Cheng H, Snoek R, Dahlman-Wright K, Gustafsson JA, Shui R, Sheppard PC, Matusik RJ 1999 Selective activation of the probasin androgen responsive region by steroid hormones. J Mol Endocrinol 22:313–325[Abstract]
13. Kasper S, Rennie PS, Bruchovsky N, Sheppard PC, Cheng H, Lin L, Shiu RPC, Snoek R, Matusik RJ 1994 Cooperative binding of androgen receptors to two DNA sequences is required for androgen induction of the probasin gene. J Biol Chem 269:31763–31769[Abstract/Free Full Text]
14. Claessens F, Alen P, Devos A, Peeters B, Verhoeven G, Rombauts W 1996 The androgen-specific probasin response element 2 interacts differentially with androgen and glucocorticoid receptors. J Biol Chem 271:19013–19016[Abstract/Free Full Text]
15. Schoenmakers E, Alen P, Verrijdt G, Peeters B, Verhoeven G, Rombauts W, Claessens F 1999 Differential DNA binding by the androgen and glucocorticoid receptors involves the second Zn-finger and a C-terminal extension of the DNA- binding domains. Biochem J 341:515–521
16. Yan Y, Sheppard PC, Kasper S, Lin L, Hoare S, Kapoor A, Dodd JG, Duckworth ML, Matusik RJ 1997 A large fragment of the probasin promoter targets high levels of transgene expression to the prostate of transgenic mice. Prostate 32:129–139[CrossRef][Medline]
17. Sambrook JF, Fritsch EF, Maniatis T 1989 Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor
18. Lareyre JJ, Thomas TZ, Zheng WL, Kasper S, Ong DE, Orgebin-Crist MC, Matusik RJ 1999 A 5-kilobase pair promoter fragment of the murine epididymal retinoic acid-binding protein gene drives the tissue-specific, cell-specific, and androgen-regulated expression of a foreign gene in the epididymis of transgenic mice. J Biol Chem 274:8282–8290[Abstract/Free Full Text]
19. Cleutjens KB, van der Korput HA, Ehren-van Eekelen CC, Sikes RA, Fasciana C, Chung LW, Trapman J 1997 A 6-kb promoter fragment mimics in transgenic mice the prostate-specific and androgen-regulated expression of the endogenous prostate- specific antigen gene in humans. Mol Endocrinol 11:1256–1265
20. Wei C, Willis RA, Tilton BR, Looney RJ, Lord EM, Barth RK, Frelinger JG 1997 Tissue-specific expression of the human prostate-specific antigen gene in transgenic mice: implications for tolerance and immunotherapy. Proc Natl Acad Sci USA 94:6369–6374[Abstract/Free Full Text]
21. Schaffner DL, Barrios R, Shaker M, Rajagopalan S, Huang S, Tindall DJ, Young CYF, Overbeek PA, Lebovitz RM, Lieberman MW 1995 Transgenic mice carrying a PSArasT24 hybrid gene develop salivary gland and GI tract neoplasms. Lab Invest 72:283–390[Medline]
22. Patrikainen L, Shan J, Porvari K, Vihko P 1999 Identification of the deoxyribonucleic acid-binding site of a regulatory protein involved in prostate-specific and androgen receptor-dependent gene expression. Endocrinology 140:2063–70:[Abstract/Free Full Text]
23. Brookes DE, Zandvliet D, Watt F, Russell PJ, Molloy PL 1998 Relative activity and specificity of promoters from prostate-expressed genes. Prostate 35:18–26[CrossRef][Medline]
24. Garrick D, Fiering S, Martin DI, Whitelaw E 1998 Repeat-induced gene silencing in mammals. Nat Genet 18:56–59[CrossRef][Medline]
25. Henikoff S 1998 Conspiracy of silence among repeated transgenes. BioEssays 20:532–535[CrossRef][Medline]
26. Greenberg NM, DeMayo FJ, Finegold MJ, Medina D, Tilley WD, Aspinall JO, Cunha GR, Donjacour AA, Matusik RJ, Rosen JM 1995 Prostate cancer in a transgenic mouse. Proc Natl Acad Sci USA 92:3439–3443[Abstract/Free Full Text]
27. Barrios R, Lebovitz RM, Wiseman AL, Weisoly DL, Matusik RJ, DeMayo F, Lieberman MW 1996 RasT24 driven by a probasin promoter induces prostatic hyperplasia in transgenic mice. Transgenics 2:23–28
28. Kasper S, Sheppard PC, Yan Y, Pettigrew N, Borowsky AD, Prins GS, Dodd JG, Duckworth ML, Matusik RJ 1998 Development, progression and androgen-dependence of prostate tumors in transgenic: a model for prostate cancer. Lab Invest 78:319–334[Medline]
29. Schoenmakers E, Verrijdt G, Peeters B, Verhoeven G, Rombauts W, Claessens F 2000 Differences in DNA binding characteristics of the androgen and glucocorticoid receptors can determine hormone-specific responses. J Biol Chem 275:12290–12297[Abstract/Free Full Text]
30. Huang W, Shostak Y, Tarr P, Sawyers C, Carey M 1999 Cooperative assembly of androgen receptor into a nucleoprotein complex that regulates the prostate-specific antigen enhancer. J Biol Chem 274:25756–25768[Abstract/Free Full Text]
31. Taplin ME, Bubley GJ, Shuster TD, Frantz ME, Spooner AE, Ogata GK, Keer HN, Balk SP 1995 Mutation of the androgen-receptor gene in metastatic androgen- independent prostate cancer. N Engl J Med 332:1393–1398[Abstract/Free Full Text]
32. Kelly WK, Scher HI 1993 Prostate specific antigen decline after antiandrogen withdrawal: the flutamide withdrawal syndrome. J Urol 149:607–609[Medline]
33. Culig Z, Hobisch A, Cronauer MV, Cato AC, Hittmair A, Radmayr C, Eberle J, Bartsch G, Klocker H 1993 Mutant androgen receptor detected in an advanced-stage prostatic carcinoma is activated by adrenal androgens and progesterone. Mol Endocrinol 12:1541–1550
34. Newmark JR, Hardy DO, Tonb DC, Carter BS, Epstein JI, Isaacs WB, Brown TR, Barrack ER 1992 Androgen receptor gene mutations in human prostate cancer. Proc Natl Acad Sci USA 89:6319–6323[Abstract/Free Full Text]
35. Veldscholte J, Berrevoets CA, Ris-Stalpers C, Ku